Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Factors that may impact drug disposition and metabolism
(USC Thesis Other)
Factors that may impact drug disposition and metabolism
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
FACTORS THAT MAY IMPACT DRUG DISPOSITION AND METABOLISM by SIYU LIU A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY CLINICAL AND EXPERIMENTAL THERAPEUTICS August 2018 Copyright 2018 Siyu Liu Page 2 of 140 Dedication I would like to dedicate this dissertation to my parents, friends and other family members, whose love and enduring support have allowed me to accomplish my dreams. This dissertation was only possible because of your inspiration. Page 3 of 140 Acknowledgement I would like to express my great appreciation for my mentor Dr. Stan Louie, who patiently guided me throughout my graduate studies. You not only encouraged me to investigate my scientific hypotheses but also helped me conquer the various obstacles that came up in the projects. Your trust in my ability gave me the confidence to march forward in my scientific career. I am very grateful to my committee members: Drs. Daryl Davies, Wei-Chiang Shen, and Curtis Okamoto. Your warmth and precious tutorage lead me to fulfill my doctoral tasks. Your knowledge and suggestions helped me to refine my thoughts and expand the scope of my scientific investigation. I would also thank Dr. Neely for entrusting lead role in the voriconazole study to me. I have earned invaluable clinical experience from this project. I would also like to acknowledge my laboratory colleagues. I am grateful to Dr. Isaac Asante, Dr. Eunjoo (Julie) Yoo, Hua Pei, Eugene Zhou, Darryl Chui, Tracey Lin and Tiange Dong. You created an environment where I could thrive. You always cheered me on when I made accomplishments. You called me “leader,” and inspired me to become one. I could not have done all this without your undying support. Thank you all. Page 4 of 140 Table of Contents DEDICATION .................................................................................................................. 2 ACKNOWLEDGEMENT .................................................................................................. 3 LIST OF TABLES ............................................................................................................ 9 LIST OF FIGURES ........................................................................................................ 11 ABBREVIATIONS ......................................................................................................... 13 ABSTRACT ................................................................................................................... 15 CHAPTER 1.OVERVIEW: IMPACTS OF INDIVIDUAL VARIATION AND FACTORS ON DRUG METABOLISM AND DISPOSITION ............................................................ 17 ABSTRACT ................................................................................................................... 17 1.1 DRUG METABOLIZING ENZYMES (DMES) AND DRUG METABOLISM .............................. 18 1.1.1 Phase I reactions through CYP450 enzymes .............................................. 20 1.1.2 Phase I reactions through Flavin-containing monooxygenase (FMO) ......... 22 1.1.3 Phase I reactions through non-CYP enzymes ............................................. 24 1.1.4 Phase II reactions through UDP-glucuronosyltransferase (UGT) ................ 24 1.1.5 Phase II reactions through other enzymes .................................................. 26 1.2 DRUG DISPOSITION VIA CELLULAR TRANSPORTERS ................................................... 27 1.2.1 Influx transporters ........................................................................................ 27 1.2.1.1 Organic anion transporters (OAT) ........................................................ 28 1.2.1.2 Organic cation transporters (OCT) ....................................................... 29 1.2.1.3 Organic anion transporter polypeptide (OATPs) .................................. 30 1.2.1.4 Equilibrative nucleoside transporter (ENTs, SLC29) and concentrative nucleoside transporters (CNTs, SLC28) ............................................................. 31 1.2.2 Efflux transporters ....................................................................................... 32 1.2.2.1 P-glycoprotein (P-gp) ........................................................................... 32 1.2.2.2 Multidrug resistance-associated proteins (MRPs) ................................ 33 1.2.2.3 Breast cancer resistance protein (BCRP) ............................................. 34 Page 5 of 140 1.3 MOLECULAR REGULATION OF METABOLIC ENZYMES AND MEMBRANE TRANSPORTERS EXPRESSION AND FUNCTION ......................................................................................... 35 1.3.1 Constitutive androstane receptor (CAR) ...................................................... 36 1.3.2 Pregnane X receptor (PXR) ......................................................................... 36 1.3.3 Aryl hydrocarbon receptor (AhR) ................................................................. 38 1.3.4 Nuclear transcription factor E2-related factor 2 (Nrf2) ................................. 39 1.3.5 Peroxisome proliferator activated receptor (PPAR) ..................................... 40 1.3.6 Other Orphan Nuclear Transcriptional Factors ............................................ 40 1.4 INDIVIDUAL FACTORS ON DRUG METABOLISM AND DISPOSITION ................................ 41 1.4.1 Genetic Polymorphisms............................................................................... 41 1.4.2 Gender and age .......................................................................................... 42 1.4.3 Other factors ................................................................................................ 44 1.5 INFLUENCES OF DRUG METABOLISM AND DISPOSITION ON DRUG EFFICACY ................. 44 1.5.1 Induction and inhibition ................................................................................ 44 1.5.2 Drug-drug interactions (DDIs) ...................................................................... 45 1.5.3 The interplay between drug metabolizing enzymes (DMEs) and transporters ............................................................................................................ 46 1.6 HYPOTHESIS AND SPECIFIC AIMS ............................................................................ 46 1.7 OUTLINE OF THE DISSERTATION .............................................................................. 47 CHAPTER 2.PEDIATRIC INDIVIDUAL FACTORS AND METABOLIC CAPACITY IN RELATION TO VORICONAZOLE THERAPEUTIC EFFECTS ..................................... 49 ABSTRACT ................................................................................................................... 49 2.1 INTRODUCTION ..................................................................................................... 49 2.1.1 Pharmacotherapy for pediatrics ................................................................... 49 2.1.2 Voriconazole ................................................................................................ 51 2.1.3 Metabolism probes ...................................................................................... 51 2.2 MATERIAL AND METHODS ....................................................................................... 53 2.2.1 Chemicals and Materials ............................................................................. 53 2.2.2 Study design ................................................................................................ 53 2.2.3 Subjects and procedures ............................................................................. 53 Page 6 of 140 2.2.4 Determination of probes and voriconazole .................................................. 54 2.2.5 Statistical Analysis and Software modeling ................................................. 55 2.3 RESULTS .............................................................................................................. 56 2.3.1 Demographics ............................................................................................. 56 2.3.2 LC-MS/MS Assay ........................................................................................ 56 2.3.3 Pharmacokinetics parameters of voriconazole treatment ............................ 62 2.3.4 Enzyme activity and individual factors in relation to the metabolic ratio of voriconazole ........................................................................................................... 63 2.4 DISCUSSION ......................................................................................................... 67 2.5 CONCLUSION ........................................................................................................ 71 CHAPTER 3.DRUG DISPOSITION OF PHARMACOENHANCER BOOSTED PROTEASE INHIBITORS IN CD4+ CELLS CEM AND U937 ....................................... 72 ABSTRACT ................................................................................................................... 72 3.1 INTRODUCTION ..................................................................................................... 73 3.1.1 AIDS and HIV infections .............................................................................. 73 3.1.2 HAART ........................................................................................................ 73 3.1.3 CD4+ Cell lines............................................................................................ 76 3.1.4 Clinical findings ........................................................................................... 77 3.2 MATERIALS AND METHODS ..................................................................................... 78 3.2.1 Materials and chemicals .............................................................................. 78 3.2.2 Determination of cellular concentration of PIs and PEs ............................... 78 3.2.3 Study design ................................................................................................ 79 3.2.4 Cell culture .................................................................................................. 79 3.2.5 Drug treatment ............................................................................................ 79 3.2.6 Sample preparation and analysis ................................................................ 80 3.2.7 Statistical analysis and pharmacokinetics ................................................... 80 3.3 RESULTS .............................................................................................................. 80 3.3.1 LC-MS assay ............................................................................................... 80 3.3.2 Cellular disposition of PI with and without pharmacoenhancers .................. 83 3.3.3 Elimination rate constant ............................................................................. 86 Page 7 of 140 3.3.4 Cellular PI AUC after washout ..................................................................... 87 3.4 DISCUSSION ......................................................................................................... 89 3.5 CONCLUSION ........................................................................................................ 91 CHAPTER 4.RELATIONSHIP OF PHARMACOENHANCERS (PES) BOOSTED ATAZANAVIR (ATV) AND DARUNAVIR (DRV) TREATMENTS WITH EFFLUX TRANSPORTERS ......................................................................................................... 92 ABSTRACT ................................................................................................................... 92 4.1 INTRODUCTION ..................................................................................................... 93 4.1.1 Role of Efflux transporters on antiretroviral therapies .................................. 93 4.2 MATERIALS AND METHODS ..................................................................................... 96 4.2.1 Chemicals .................................................................................................... 96 4.2.2 Study Design ............................................................................................... 97 4.2.3 Transfected cell lines’ culture conditions ..................................................... 98 4.2.4 qRT-PCR ..................................................................................................... 99 4.2.5 Western blot ................................................................................................ 99 4.2.6 Flow cytometry .......................................................................................... 100 4.2.7 Data analysis ............................................................................................. 101 4.3 RESULTS ............................................................................................................ 101 4.3.1 Transporter transfected HEK and MDCK cell variants .............................. 101 4.3.2 Cellular disposition in the presence of transporter inhibitors ..................... 104 4.3.2.1 P-gp .................................................................................................... 104 4.3.2.2 MRP1/MRP4 ...................................................................................... 106 4.3.2.3 Breast Cancer Resistance Protein (BCRP) ........................................ 107 4.3.2.4 Multi-drug resistant associated protein-2 (MRP2) .............................. 108 4.3.3 Transporter gene expression in CD4+ cells .............................................. 110 4.3.4 Protein level expression of transporters in CD4+ cells .............................. 113 4.3.5 Transporter functional tests ....................................................................... 116 4.4 DISCUSSION ....................................................................................................... 117 4.5 CONCLUSION ...................................................................................................... 121 CHAPTER 5.CONCLUSIONS AND FUTURE DIRECTIONS ...................................... 123 Page 8 of 140 5.1 SUMMARY .......................................................................................................... 123 5.1.1 Summary of voriconazole study ................................................................ 123 5.1.2 Summary of Pharmacoenhancer (PE) study ............................................. 124 5.2 CONCLUSION AND SIGNIFICANCE .......................................................................... 126 5.3 FUTURE DIRECTION ............................................................................................. 127 5.3.1 Voriconazole .............................................................................................. 127 5.3.2 PI/PE ......................................................................................................... 127 BIBLIOGRAPHY AND REFERENCES ........................................................................ 130 Page 9 of 140 List of Tables Table 1.1 Overview of the types of metabolic reactions for xenobiotics. ....................... 19 Table 1.2 Chemical reactions mediated by the five major CYP isoforms . .................... 21 Table 1.3 Types of biosensor/nuclear receptors. .......................................................... 36 Table 2.1 Patient demographics and demographic parameters……………………. …...56 Table 2.2 Intraday accuracy (n=3) of the probes quality controls. ................................. 57 Table 2.3 Interday accuracy (n=3) of the probes quality controls. ................................. 58 Table 2.4 Interday accuracy (n=3) of voriconazole and voriconazole N-oxide quality controls. ......................................................................................................................... 58 Table 2.5 Mean and median pharmacokinetic parameters of voriconazole and its N- oxide voriconazole......................................................................................................... 62 Table 2.6 Voriconazole metabolic ratio and relative activity of CYP3A4, CYP2C19, and FMO3 (Mean ± SD). ............................................................................................... 63 Table 3.1 Summary of clinical outcome using various PE…………………………... …...77 Table 3.2 Cellular disposition study design. .................................................................. 79 Table 3.3 Interday QC analysis of (a) CEM and (b) U937 standard curves................... 81 Table 3.4 Elimination rate constant (Ke) of (a) ATV and (b) DRV when treated alone or in combination with PEs in U937 and CEM.. ............................................................. 86 Table 3.5 Cellular PI AUC (based on median concentration at each time point, n=3) during washout period after 24hr treatment in CEM and U937. .................................... 88 Table 4.1 Impact of various PI and PEs on cellular transporters…………………..…… 94 Table 4.2 Overall Study Design to determine the effects of transporter expression and function of efflux transporters. ....................................................................................... 97 Table 4.3 Efflux Transporter primers ............................................................................. 99 Table 4.4 Signal intensity ratio of drug treatments to non-treated control based on normalized intensity. (a) U937 ATV group. (b) U937 DRV group. (c) CEM ATV group. (d). CEM DRV group. Cells were treated with 4000 ng/mL of PI and 1000 ng/mL for 24 hours. ..................................................................................................................... 114 Table 4.5 The protein expression of each of the efflux transporter in relations to untreated cells. (a) CEM and (b) U937 cells. ............................................................... 115 Page 10 of 140 Table 4.6 Rhodamine efflux after 24-hour COBI or RTV treatments in CEM and U937. .................................................................................................................................... 116 Table 4.7 MRP1 functions after 24-hour PE treatments in CEM and U937. ................ 117 Page 11 of 140 List of Figures Figure 1.1 The catalytic cycle of FMOs ........................................................................ 23 Figure 1.2 General role of transporters. ........................................................................ 27 Figure 1.3 PXR and CAR activation pathway.. .............................................................. 37 Figure 1.4 AhR and Nrf2 activation pathway. ................................................................ 39 Figure 2.1 Chromatograms of (a) probes and (b) voriconazole LC-MS/MS assays…... …………………………………………………………………………………………...60 Figure 2.2 Interday standard curves (n=3) for voriconazole and pharmacologic probes and their respective metabolites.. .................................................................................. 61 Figure 2.3 Age in relation to voriconazole metabolic ratio and enzyme activity.. ........... 64 Figure 2.4 Relationship between enzyme activity and voriconazole metabolic ratio in oral groups.. .................................................................................................................. 65 Figure 2.5 Relationship between enzyme activity and voriconazole metabolic ratio in oral groups.. .................................................................................................................. 66 Figure 2.6 Relationship between sex and enzyme activities on voriconazole metabolic ratio. .............................................................................................................................. 67 Figure 2.7 Voriconazole metabolic ratio in relation to age and FMO3. .......................... 70 Figure 3.1 Standard curves of interday tests in CEM and U937……………………. …...82 Figure 3.2 Chromatographs of various analytes that include COBI, RTV, ATV, DRV, and LPV. ....................................................................................................................... 83 Figure 3.3 Effect of COBI and RTV on ATV and DRV cellular disposition in CD4+ cells. .............................................................................................................................. 85 Figure 3.4 ATV (left) and DRV (right) Ke in relation to AUC. ......................................... 90 Figure 4.1 ATV versus DRV cellular concentration in HEK (a) and MDCK (b) transporter overexpressed variants after 24-hour treatment.. ..................................... 101 Figure 4.2 ATV versus DRV cellular concentration in MDCK transporter overexpressed variants after 24-hour PE boosted treatment.. .................................... 103 Figure 4.3 ATV versus DRV cellular concentration in HEK overexpressing variants after 24- hour PE boosted treatment. .......................................................................... 104 Figure 4.4 Impact of P-gp inhibition on PI CC in CEM and U937. ............................... 105 Figure 4.5 Impact of MRP1/MRP4 on PI CC in CEM and U937.. ................................ 106 Page 12 of 140 Figure 4.6 Impacts of BCRP on PI in CEM and U937. 107 Figure 4.7 Impacts of MRP2 on PI CC in CEM and U937. .......................................... 109 Figure 4.8 Transporter gene expression in U937 when treated with PIs alone or in combination with PEs. (a) ABCC1 (b) ABCC2 (c) ABCB1 (d) ABCG2 ........................ 111 Figure 4.9 Transporter gene expression in CEM when treated with PIs alone or in combination with PEs. (a) ABCC1 (b) ABCC2 (c) ABCB1 (d) ABCG2 ........................ 112 Figure 4.10 Transporter expression in relation to transporter function. (a) P-gp and (b) MRP1 expression in relation to transporter function indicator fluorescence intensity in CEM. (c) P-gp and (d) MRP1 expression in relation to transporter function indicator fluorescence. ................................................................................................ 121 Page 13 of 140 Abbreviations ABC ATP-binding cassettes AhR Aryl hydrocarbon receptor AIDS Acquired immunodeficiency syndrome ARV Antiretrovirals ATV Atazanavir AUC Area under the curve BCRP Breast cancer resistant protein CAR Constitutive androstane receptor CC Cellular concentration CCRP CAR retention protein Cmax Maximum concentration CNT Concentrative nucleoside transporter COBI Cobicistat CYP Cytochrome P450 DBD DNA binding domain DDI Drug-drug interaction DME Drug metabolizing enzyme DRV Darunavir ENT Equilibrative nucleoside transporter FMO Flavin-containing monooxygenase FXR Farnesoid X receptor HAART Highly active antiretroviral therapy HEK Human embryo kidney 293 HIV Human immunodeficiency virus HSP Heat shock protein IV Intravenous Ke Elimination rate constant LBD ligand binding domain LPV Lopinavir LXR Liver X receptor MDCK II Madin-Darby Canine Kidney II MRP Multidrug resistant proteins Nrf2 Nuclear transcription factor E2-related factor 2 OAT Organ anionic transporter OATP Organ anionic transporter proteins OCT Organ cationic transporters PE Pharmacoenhancer P-gp P-glycoprotein PI Protease inhibitor Page 14 of 140 PK Pharmacokinetics PO Per os (by mouth) PPAR Peroxisome proliferator activated receptor PXR Pregnane xenobiotic receptor RTV Ritonavir RXR Retinoid xenobiotic receptor SLC Solute carriers TDM Therapeutic drug monitoring TI Therapeutic Index XRE Xenobiotic response element Page 15 of 140 Abstract Background: Drug disposition describes the processes of drug absorption, distribution, metabolism, and elimination, collectively known as ADME. Drug disposition in the target tissues determines the overall drug exposure, and thus treatment efficacy. Essential factors in these processes include drug metabolizing enzymes (DMEs), membrane transporters, as well as protein binding, patient gender, and patient age. Clinical practices require the balance of efficacy and safety, especially among the pediatric population. Understanding the pharmacological mechanisms improves the clinical impact of drug therapies, by giving insight on how to improve and optimize treatment efficacy and safety. In this dissertation, I studied the age-related impacts of DMEs on the pharmacokinetics (PK) of the drug voriconazole and the roles of transporters in anti-HIV combination therapies. Aims: 1) To understand the ontogeny of Cytochrome P 450 (CYP) 2C19 (CYP2C19), CYP3A4/5, and flavin-containing monooxygenase 3 (FMO3) in relation to age-related voriconazole PK. 2) To evaluate CD4+ cellular disposition of HIV protease inhibitors (PIs) atazanavir (ATV) and darunavir (DRV) when treated alone or in combination with pharmacoenhancers (PEs). 3) To assess the impacts of PI/PE treatments on efflux transporters expression and function in CD4+ cells. Results: For the first aim, In accordance with my initial hypothesis, CYP2C19, FMO3, and CYP3A4/5 activities decline with increasing age, consistent with age-related voriconazole metabolism. Unlike in adult patients, both FMO3 and CYP2C19 exhibit a significant correlation between age and voriconazole metabolism in pediatric patients. Page 16 of 140 With regards to the specific aims 2 and 3, I found that cellular concentrations of atazanavir (ATV) is more responsive to PE treatments when compared to darunavir (DRV). The cellular exposure AUC of ATV is about two times that of DRV. I also found that cobicistat (COBI) is inferior to ritonavir (RTV) in prolonging cellular retention when co-administer with a PI. COBI is a weaker inhibitor of efflux transporters when compared to RTV. Moreover, ATV itself is an inhibitor of efflux transporter when compared to DRV. Also, CD4+ cell types impact the treatment efficacy, as lymphocytes are more responsive to HIV treatments than HIV infected monocytes. These findings are consistent with current data suggesting that monocytes may be a site of persistent HIV infections, due to its ability to eliminate antiretroviral (ARV) agents. Conclusion: Evaluating DME activities among pediatric patients can optimize drug treatment, especially for narrow therapeutic window drugs like voriconazole. The DMEs FMO3 and CYP2C19 have age-sensitive expression, and thus explain the extra dosage difference in voriconazole between children and adults. DRV and ATV cellular retention were different when given alone. This may be attributed to ATV’s ability to inhibit various efflux transporters. When DRV or ATV was combined with either COBI or RTV, RTV was a more potent efflux transporter inhibitor than COBI. This may explain the higher incidence of clinical virologic failure of DRV versus ATV when used in combination with either RTV or COBI containing treatments, particularly in HIV infected CD4+ monocytic/macrophagic cells. Page 17 of 140 Chapter 1. Overview: Impacts of individual variation and factors on drug metabolism and disposition Abstract Drug disposition is a key element in determining treatment success versus failure. Drug disposition is regulated by several factors: absorption, distribution, metabolism, and excretion (ADME). The molecular basis of drug distribution is reliant on metabolic enzymes and cellular transporter expression and their functional activities. Metabolic enzymes such as the cytochrome P450 (CYP) family, flavin monooxygenase (FMO), and esterases have natural substrates which include steroids and neurotransmitters. However, these metabolic enzymes can also affect xenobiotics: drugs, nutrients, and metabolic byproducts. The expression and function(s) of influx and efflux transporters dictate the intracellular concentrations of various xenobiotics. Influx transporters promote the influx of the xenobiotics or nutrients across the cellular membrane. In contrast, the efflux transporters regulate the levels of xenobiotics and byproducts by active elimination. The expression of both metabolic enzymes and cellular transporters are regulated by a family of orphan xenobiotic receptors, which were later found to be nuclear transcriptional factors. These transcriptional factors are biosensors that are activated through displacement of their natural repressors. Once activated, these xenobiotic receptors will translocate from the cytoplasm and into the nucleus, where they heterodimerize with the retinoid xenobiotic receptor (RXR) and function as the nuclear transcriptional factors regulating expression of metabolic enzymes and cellular transporters. In this thesis, I will investigate the impact of age on the expression of metabolic enzymes, FMO isoenzymes in particular. To further dissect the impact of cellular transporters in Page 18 of 140 regulating intracellular drug levels, I will interrogate the molecular mechanism of using inhibitors of cellular transporters to characterize the impact of pharmacoenhancers (PEs) on human immunodeficiency virus (HIV) protease inhibitors elimination. 1.1 Drug metabolizing enzymes (DMEs) and drug metabolism The ADME properties of a drug compound dictate the concentration found in various compartments (e.g. tissues, fluids) at various time points after administration. Initially, drug disposition is primarily dependent on the physicochemical properties of the compound, e.g. lipophilicity, molecular weight, and the ability to partition in and out of tissues[1]. Rational drug design aims to maximize substrate interaction with the targeted protein(s) and the ability to cross various tissue barriers and fluids, i.e. lipophilicity. Lipophilicity is a key component of elimination; lipophilic compounds are more likely to be metabolized to a more hydrophilic version that can undergo glomerular filtration or further metabolism [2]. Currently, metabolic reactions are classified into Phase I and Phase II biotransformation reactions. Phase I reactions include hydrolysis, hydroxylation, reduction, dealkylation, and oxidation where the ultimate aim is to enhance polarity [3]. In contrast, Phase II reactions are the addition of polar substrates such as glucuronic acid or sulfates. The primary Phase I reactions are summarized in Table 1.1. Enhanced polarity of a molecule will increase its solubility in water and increase free fraction, thus increase the ability for glomerular filtration mediated elimination. Most Phase I reactions are mediated by CYP enzymes, which primarily promote the oxidation and hydration of the xenobiotics [4]. This family of enzyme complexes acts to incorporate an atom of oxygen into non-activated hydrocarbons that introduce OH-, N-, Page 19 of 140 O- and S-dealkylation of the substrate. The most common Phase I oxidation reaction involves the conversion of a C-H bond to a C-OH. There are other non-CYP-mediated Phase I enzymes such as FMOs, esterases, monoamine oxidases (MAOs), and amidases. Similar to CYP-catalyzed reactions, FMO mediated reactions promote the formation of mono-oxygen, which facilitate additional metabolism via N-oxidation or S- oxidation reactions. Esterases metabolize esters into a carboxylic acid and alcohol moiety. MAOs catalyze the oxidation of monoamines. Table 1.1 Overview of the types of metabolic reactions for xenobiotics. Phase II reactions are typically derivatization reactions which conjugate substrates with highly charged substituents, such as glucuronic acid and sulfate [5]. Phase II metabolism can also conjugate glutathione, γ-glutamate, and glycine to xenobiotics. These conjugated metabolites are highly charged and easily excreted. Phase II reactions are mediated by the enzymes UDP-glucuronosyltransferases (UGTs), N-acetyltransferases (NATs), glutathione S-transferases (GST), and sulfotransferases (SULTs) [5]. Enzyme Reaction PHASE I CYP R-H + XOOH (O2, 2e-, 2H+) →ROH + XOH FMO N-oxidation & S-oxidation Esterases R1OOR2 →R1OH + R2OOH MAO RCH2NR1R2→ RCHO + NHR1R2 PHASE II UGT Uridine-5´-diphospho-α-D-glucuronic acid + R-X-H → β -D- Glucuronide + Uridine diphosphate NAT R-NH2 →R-NHCOCH3 GST Glutathione + RX → Glutathione-S-conjugate SULT SO4 2 -+ 2ATP + R-OH → R-OSO3H + PAP Page 20 of 140 1.1.1 Phase I reactions through CYP450 enzymes CYP P450 enzymes constitute the predominant DME family and are found most abundantly in the liver and intestinal tissues [2]. CYP P450 is a superfamily of heme- containing metabolic enzymes. CYP-mediated enzymatic reactions are responsible for the metabolic elimination of most small hydrophobic molecule substrates [2]. Currently, there are 57 human CYP450 enzymes that have been identified, most of which are engaged in the biotransformation of lipophilic compounds such as steroids and oil soluble vitamins [6]. These 57 CYPs are classified into 10 sub-families, where amino acid homology among the various families is well recognized [7]. Five major isoforms have been identified as the most important xenobiotic metabolic enzymes in adult humans: CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4. The reactions mediated by these isoenzymes are summarized in Table 1.2 [8]. CYP1A2 is an oxidase and constitutes approximately 13% of all CYP450 enzymes found in the adult liver [9]. It can metabolize a structurally diverse spectrum of xenobiotics, such as clozapine, flutamide, lidocaine, melatonin, tacrine, and mexiletine [10]. Other compounds susceptible to CYP1A2-mediated metabolism include clomipramine, acetaminophen, verapamil, warfarin, and zotepine [10]. Enzymes Types of Reactions CYP1A2 Demethylation Hydroxylation Oxidation Dealkylation CYP2C9 Hydroxylation Demethylation Oxidation Dealkylation CYP2C19 Hydroxylation Demethylation CYP2D6 Hydroxylation Page 21 of 140 Demethylation Oxidation CYP3A4 Hydroxylation Demethylation Oxidation Dealkylation Table 1.2 Chemical reactions mediated by the five major CYP isoforms [11]. CYP2C9, a member of the CYP2C subfamily, catalyzes the catabolism of weak acid and lipophilic molecules. Approximately 20% of the drugs used clinically are susceptible to CYP2C9-mediated metabolism including agents tolbutamide, ibuprofen, phenytoin, warfarin, and losartan [12]. The ability to metabolize this large spectrum of compounds may be due to the six substrate recognition sites have been identified on this isoenzyme [13]. CYP2C19, another important member of the CYP2C subfamily, shares over 80% structure homology with CYP2C9 [14]. CYP2C19 is mainly expressed in the liver and intestines but is less abundant than CYP2C9. This isoenzyme is known to mediate metabolism of proton inhibitors (e.g. omeprazole, lansoprazole), anticonvulsant agents (e.g. mephenytoin, phenobarbital) and CNS agents (citalopram, aripiprazole) [14]. CYP2D6 catalyzes hydroxylation, demethylation and dealkylation reactions [11]. The gene encoding for CYP2D6 is located on chromosome 22, and is expressed in high levels in the liver and central nervous system (CNS), particularly in the substantia nigra [15]. The presence of this enzyme in the CNS suggests that it is critical in the metabolism of endogenous substrates such as neurosteroids, tyramine isoforms, and dopamine. In addition, CYP2D6 promotes the metabolism of opioid pain relievers, antidepressants, antipsychotics, β-adrenergic blockers, and anticancer drugs (e.g. tamoxifen) [15]. There is high genetic variability for these enzymes thus explaining the large interpatient Page 22 of 140 variability in drug levels. The catalytic activity of this isoenzyme can be classified as poor, rapid, and ultra-rapid metabolizers. Although all previously mentioned metabolic enzymes are important, none is more important than CYP3A4. This isoenzyme plays a vital role in the metabolism of almost half of all clinically used drugs. Like many CYPs, CYP3A4 is found most abundantly in the intestines and liver. Despite its importance, CYP3A4 is not highly expressed in fetal liver tissue. Instead the predominant CYP3A isoenzyme in fetuses is CYP3A7. Expression of CYP3A7 is reduced at birth, while CYP3A5 is predominant till puberty. Despite the presence of CYP3A7 and CYP3A5, CYP3A4 is expressed at a very young age and increases to 40% and 72% of adult levels after four and 12 months of life [16]. Over 28 single nucleotide polymorphisms (SNPs) has been identified for the gene encoding CYP3A4. These SNPs are associated with ethnicity [14]. Compared to the CYP2 family, CYP3A4 contains an additional helical structure, thus the active site of CYP3A4 might be capable of binding three molecules simultaneously [17]. CYP3A4 is capable of metabolizing antibiotics, antiarrhythmics, antivirals, sedatives, antihistamines, calcium channel blockers, and HMG CoA reductase inhibitors [14]. 1.1.2 Phase I reactions through Flavin-containing monooxygenase (FMO) FMO is a flavin adenine dinucleotide- (FAD-) and nicotinamide adenine dinucleotide- (NADPH-) dependent drug metabolic enzyme system and acts as a complementary system to the CYP450 enzymes for clearing N- and S- containing endogenous molecules [18]. In contrast to CYPs, the FMO oxygenation-retro-reduction uses two electrons instead of one to form polar N-oxide and S-oxide metabolites. Since these metabolites are readily eliminated from the body, FMO has a lower potential to be induced or inhibited Page 23 of 140 by drugs [19]. Gene clusters encoding for FMOs are located in chromosome 1. The FMO protein is composed of two structural domains with a channel between them [20]. FMOs are divided into five families according to its sequence homology [18, 21]. FMO1 is highly expressed in fetal livers, and is closely associated with CYP3A7. After birth, FMO1 is mainly expressed in the kidney and has a low expression level in the adult human liver. FMO2 is most prevalent in the lung in adults. FMO4 is expressed in the adult liver and kidney. Despite the abundant expression, FMO2 and FMO4 are considered to have limited function in humans [22]. FMO3 and FMO5 are most abundant in adult human livers, and are implicated in drug metabolism. FMO3, the major FMO, is known to oxygenate highly polarizable small nucleophilic heteroatom-containing molecules, while FMO5 substrates have been poorly identified at this point [22, 23]. FMO3 drug substrates include H2-receptor antagonists (e.g. cimetidine), antifungal drugs (e.g. voriconazole), and anticancer agents (e.g. tamoxifen) [20, 23]. In this thesis, I will discuss the impact of aging on the expression of FMO3 in chapter 2. Figure 1.1 The catalytic cycle of FMOs [20]. Page 24 of 140 1.1.3 Phase I reactions through non-CYP enzymes Esterases are Phase I enzymes catalyzing the hydrolysis of esters to form corresponding carboxylic acids and alcohols [24]. However, esterases can also metabolize a variety of other molecules including peptides and amides [24]. Esterases are categorized into three classes, A, B, and C based on substrate specificity [25]. Class B esterases, including carboxylesterases, cholinesterases and acetylcholinesterases, are important for metabolism of endogenous substrates and xenobiotic compounds [26]. The endogenous substrates of esterases include neurotransmitters acetylcholine and secondary messengers cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) [26]. Drug substrates affected by esterases include ciclesonide, irinotecan, rifampin, and rifabutin [26, 27]. Monoamine oxidases (MAOs) are the other major class of non-CYP phase I enzymes. MAOs catalyze oxidative deamination, which is required to breakdown catecholamines [28]. MAOs have been classified as either MAO A or MAO B; both have been found to be important in nerve tissues and the brain. Dysregulation of MAO has been implicated in several neurologic diseases due to its importance in the metabolism of dopamine and serotonin [29]. 1.1.4 Phase II reactions through UDP-glucuronosyltransferase (UGT) UGT is one of the major enzymes mediating phase II reactions. A superfamily of 22 enzymes, UGTs are classified into 5 families and 6 subfamilies based on amino acid sequence [30]. UGT is expressed in the cytosol and is critical for the metabolism of many lipophilic substrates including steroidal hormones, vitamins, cytotoxic chemotherapeutic agents, and nucleoside analogs [31]. UGT transfers the glucuronic acid moiety onto the Page 25 of 140 hydroxyl group found on the substrate [31]. This reaction occurs predominately on compounds that have oxygen, nitrogen, sulfur, or carboxyl functional groups. Following glucuronidation, the metabolized substrates have increased water solubility and are easily excreted by the kidney. Additionally, the glucuronidated molecules are targeted for cellular efflux, as glucuronic acid “tags” the compound for elimination by various membrane transporters such as multidrug resistant proteins (MRPs) [32]. The major UGT isoenzymes include UGT1A1, UGT1A4, UGT1A6, UGT1A9, UGT2B7 and UGT2B15 [33]. UGT is expressed in great abundance in the liver, kidney and gastrointestinal tract [34, 35]. UGT1 consists of thirteen variable exons linked with common exon 2 to exon 5 to generate UGT1 isoenzymes. The genes encoding the UGT2 family of enzymes is found on chromosome 4 with 6 exons different from UGT1 [36, 37]. The UGT2B subfamily is more abundant than the UGT1A subfamily. UGT2B4, 2B15, and 2B10 are highly expressed in the liver, while UGT1A10, 2B17, and 2B7 are most abundant in the gastrointestinal tract. In the kidney, expression of UGT1A9, 2B7 and 1A6 combined is over 95% [30]. UGT is also found in the endoplasmic reticulum and nuclear compartments in cells [38]. UGT1A1 is mainly expressed in the liver, and interacts with bilirubin in particular. Drug substrates that undergo UGT1A1 metabolism include anticancer agents (e.g. irinotecan), antivirals (e.g. dolutegravir), buprenorphine, as well as flavonoids [39]. UGT1A4 interacts with amines such as clozapine, benzidine, nicotine, and imipramine. UGT1A6 is found to metabolize drugs with phenol moieties such as acetaminophen, serotonin, and salicylic acids. UGT1A9 has a similar function as UGT1A6, but instead glucuronidates phenolic molecules such as acetaminophen, retinoic acid, and flavonoids. [3, 36, 40]. Page 26 of 140 The metabolically important UGT2 family of enzymes include UGT2B4 which catalyzes glucuronidation of estrogen and phenols. UGT2B4 is highly expressed in the liver, kidney, placenta, and lung. Substrates for UGT2B4 may overlap with UGT2B7. UGT2B7 catalyzes glucuronidation of a broad spectrum of drugs including NSAIDs, anesthetics (e.g. morphine), and antivirals (e.g. efavirenz) [41, 42]. UGT2B15 is expressed in liver, kidney, lung, skin, adipose, and prostate tissues. UGT2B15 glucuronidates various steroids and benzodiazepines (e.g. lorazepam and oxazepam) [43, 44]. 1.1.5 Phase II reactions through other enzymes Non-UGT detoxifying enzymes include NATs, GSTs, and SULTs. NATs catalyze N- acetylation, a reaction that metabolizes arylamines (e.g. procainamide and dapsone) and hydrazine drugs (e.g. clonazepam) [45]. Major isoforms of NATs include NAT1 and NAT2 which are localized on chromosome 8, where polymorphisms have been identified [46] [47]. GST is an essential phase II enzyme that catalyzes the formation of thioether conjugates and is classified into several subfamilies [48]. GSTs metabolize a variety of compounds, with the majority being electrophilic xenobiotics and phase I intermediates [3]. GST activity is found to be influenced by age as well as polymorphisms [49]. SULTs comprise 4 subfamilies and at least 13 members with wide tissue distribution [48]. Their major function is to conjugate sulfur to various endogenous and xenobiotic substrates. Major substrates include quinolones and amino drugs [48]. The activity of SULTs is affected by age, pregnancy and genetic polymorphisms [49]. Page 27 of 140 1.2 Drug disposition via cellular transporters Cellular membrane transporters play an important role in tissue drug disposition. Transporters facilitate movement of nutrients and xenobiotics across cellular membranes. Thus the varying expression of these transporters can impact the pharmacokinetics of drug compounds. Membrane transporters can be classified into two major families, ATP- binding cassettes (ABCs) and solute carriers (SLCs) [50]. According to their functions, transporters are also categorized as either influx or efflux transporters. Figure 1.2 shows the general roles of transporters. Figure 1.2 General role of transporters. Influx transporters: organic anion transporters (OATs), organic cation transporters (OCTs), organic anion transporter polypeptides (OATPs), equilibrative nucleoside transporters (ENTs, SLC29) and concentrative nucleoside transporters (CNTs, SLC28). Efflux transporters: P-glycoprotein (P-gp), multidrug resistance-associated protein (MRP) and breast cancer-resistance protein (BCRP). 1.2.1 Influx transporters Most influx transporters are primarily classified as solute carriers (SLC), which are further divided into organic anion transporters (OATs), organic cation transporters (OCTs), Page 28 of 140 organic anion transporter polypeptides (OATPs), equilibrative nucleoside transporters (ENTs, SLC29) and concentrative nucleoside transporters (CNTs, SLC28), depending on the type of substrate they influx into the cells. Nucleoside and nucleoside analogs are key components for biosynthesis and bioenergetics. SLCs uptake of a wide spectrum of drug substrates, significantly affecting their absorption and disposition. 1.2.1.1 Organic anion transporters (OAT) Currently, ten isoforms of OAT have been identified, most of which are generally expressed at the basolateral membrane of hepatocytes and kidney proximal tubular cells. OATs have a wide range of endogenous and xenobiotic substrates with overlapping substrates. OAT1 is composed of 563 amino acids, forming twelve transmembrane domains with an intracellularly located N- and C-terminus, a large extracellular loop between helices 1 and 2, and a large intracellular loop between helices 6 and 7 [51]. OAT1 is normally expressed on the basolateral membrane of proximal tubule cells and is able to transport drug substrates from the blood into the proximal tubules [51]. Endogenous substrates utilizing OAT1 include cAMP, cGMP, bile salts, and hormones. Drug substrates include HMG- CoA reductase inhibitors or “statins” (e.g. Fluvastatin), angiotensin-converting enzyme inhibitors (ACEIs, e.g. temocaprilat), antibiotics (e.g. Ceftibuten), NSAIDs, and antivirals (e.g. Acyclovir) [51]. OAT2 is structurally similar to OAT1, also characterized by twelve transmembrane domains and an intracellularly located N- and C-terminus. OAT2 generally transports substrates similar to that of OAT1 such as statins, antibiotics, NSAIDs and antivirals [51, 52]. However, in contrast to OAT1, OAT2 is highly expressed in liver. Page 29 of 140 OAT3 is a typical OAT with a similar secondary structure. Like OAT1, OAT3 is expressed in the basolateral membrane of renal proximal tubule cells. However, OAT3 prefers lipophilic organic anions. Endogenous substrates using OAT3 include vitamins, cAMP, bile salts, hormones, and urate. OAT3 also transports a broad group of drugs including statins, antibiotics, angiotensin II receptor blockers (ARBs) and angiotensin converting enzyme inhibitors (ACEIs) [53, 54]. OAT4 is expressed in kidney, liver, placenta, and adrenal gland. OAT4 shares similar functions and binds similar substrates as OAT1-3 [51]. 1.2.1.2 Organic cation transporters (OCT) Organic cation transporters (OCTs) transport cationic molecules including various toxins and endogenous compounds. The mechanism of OCTs is driven by the electrochemical potential rather than sodium (Na + ) or proton (H + ) gradient alone. OCTs are from SLC family 22 (organic cation transporters, OCTs) and SLC family 47 (multidrug and toxin extrusion, MATEs) [55]. The OCT general structure contains twelve transmembrane domains, intracellular amino and carboxyl terminus, a large glycosylated extracellular loop between the first two transmembrane helices, and a large intracellular phosphorylation-purpose loop [56]. The major OCTs that have drug substrates are OCT1 (SLC22A1), OCT2 (SLC22A2), OCT3 (SLC22A3) and MATE1. OCT1 is expressed in the liver, localizing on the sinusoidal membrane of the hepatocytes. Downregulation of OCT1 has been linked to liver cancer [57]. OCT2 is mostly expressed in the basolateral membrane of proximal kidney tubule cells but is also found in multiple tissues such as the lung and intestine [58]. Compared to OCT1 and OCT2, OCT3 has a much wider tissue distribution, expressed in skeletal muscle, the brain, the heart, liver Page 30 of 140 hepatocytes, kidney proximal tubule cells, and intestinal epithelium [59]. Similar to OATs, OCT1-3 also share an overlapping spectrum of substrates. Endogenous molecules include neurotransmitters and hormones. Drug substrates include histamine H2 receptor antagonists (e.g. cimetidine), anesthetic drugs, antiviral drugs (e.g. lamivudine), antidiabetic drugs (e.g. metformin), and β-adrenergic blockers [56]. Multi-antimicrobial extrusion protein (MATEs) include MATE1 and MATE2-K. MATEs are highly expressed in the liver, kidney, heart and skeletal muscle tissues. MATEs are structured similarly to OCTs, but with thirteen transmembrane domains instead of twelve and an extracellular carboxy terminus. The C-terminus is also without glycosylation sites or intracellularly located phosphorylation sites [60]. Dozens of drug substrates are found to utilize MATEs, such as antibiotics (e.g. levofloxacin), anti-allergic drugs (e.g. fexofenadine), and antiviral drugs (e.g. tenofovir) [56]. 1.2.1.3 Organic anion transporter polypeptide (OATPs) OATPs mediate sodium-independent transport of amphiphilic organic molecules as well as some cationic and anionic compounds, including steroids, bile salts, and multiple drugs [61]. OATPs are grouped into six subfamilies with eleven isoforms and are localized in the apical and basolateral membranes of polarized cells in a variety of tissues. The structure of OATPs is comprised of twelve transmembrane domains surrounding a positively charged pore [62]. Among them, OATP1A2 (OATP-A), OATP1B1(OATP-C), OATP1B3 and OATP2B1 (OATP-B) are the most vital in drug disposition [62]. OATP1A2 is distributed in the liver as well as the epithelium of the intestines, kidney, and brain. OATP1A2 is responsible for the transport of bile salts and hormones [63]. Drug substrates for OATP1A2 include antihistamines (e.g. fexofenadine), antibiotics (e.g. Page 31 of 140 levofloxacin), statins (e.g. rosuvastatin, pravastatin), neuromuscular-junction blocking agents (e.g. pancuronium, rocuronium), and HIV protease inhibitors (e.g. saquinavir, atazanavir) [64]. OATP1B1 and OATP1B3 are expressed on the sinusoidal membrane of hepatocytes. Both transporters can uptake a wide variety of substrates, such as statins (e.g. rosuvastatin, pravastatin), antifungal drugs (e.g. caspofungin), ACEIs (e.g. enalapril), antibiotics (e.g. rifampicin, rifapentine), ARBs (valsartan, losartan), anticancer drugs (methotrexate), and some antiviral drugs [64]. OATP2B1 is widely expressed in a variety of tissues such as liver, intestine, heart, and skin. Its transport activity is pH-dependent [65]. Substrates for OATP2B1 include statins, endothelin receptor antagonists, and antidiabetic drugs [66]. The OATPs share some overlapping drug substrates like antibiotics, chemotherapeutic agents, statins and antihistaminic drugs [59]. For example, rosuvastatin is a substrate for OATP2B1, OATP1B3, and OATP1B1, while fexofenadine is a substrate for OATP1A2 and OATP2B1 [67]. 1.2.1.4 Equilibrative nucleoside transporter (ENTs, SLC29) and concentrative nucleoside transporters (CNTs, SLC28) ENTs and CNTs are responsible for transporting nucleosides and relevant analogs. Multiple anticancer and antiviral agents are impacted by these transporters [68]. There are currently four isoforms for ENTs (ENT1-4) and three isoforms for CNTs (CNT1-3) [68]. All ENTs have eleven transmembrane domains [69]. Substrates for ENT1 and ENT2 include purine, pyrimidine nucleosides, and nucleobases. Both ENTs have a wide tissue distribution [70]. ENT3 is more efficient in transporting purine and pyrimidine nucleosides Page 32 of 140 compared to ENT1 and ENT2 [70]. ENT4 transports adenosine and monoamines and is found in the brain and the heart [71]. CNTs are Na+ dependent transporters and have thirteen transmembrane domains [72]. CNT1 transports pyrimidine nucleosides, CNT2 purine nucleosides, and CNT3 both nucleoside types [68]. CNT3 also has a wider tissue distribution than CNT1 and CNT2 [73]. 1.2.2 Efflux transporters ABC transporters are most identified efflux transporters. These transporters are found in an array of tissues including liver, kidney, intestine and blood-brain barrier. The ABC family contains about fifty transporters, classified into seven subfamilies, ABCA to ABCG. The most vital in regulating drug disposition are P-glycoprotein (P-gp), multidrug resistance-associated proteins (MRPs) and breast cancer resistance protein (BCRP). 1.2.2.1 P-glycoprotein (P-gp) P-gp is the most intensely studied efflux transporter and belongs to ABC-B subfamily. P- gp is also known as MDR1 or ABCB1. It possesses the characteristic ABC transporter structure: two membrane-spanning domains with six transmembrane regions each and two nucleotide-binding domains (NBDs) with a size of 170 Daltons [74]. P-gp can transport extremely varied substrates including lipophilic and amphipathic molecules. Due to its wide array of substrates, P-gp has been implicated in the exportation of drugs from GI tract to the lumen, from the liver to bile ducts, as well as from kidney to urinary ducts [75]. P-gp is highly expressed in kidney, adrenal gland, placenta, liver, intestine, blood- brain barrier, and lung tissues, as well as in lymphocytes [76]. Furthermore, P-gp has Page 33 of 140 high expression in solid tumor epithelial membranes [77]. The key property of P-gp is the extremely wide array of substrates. P-gp drug substrates include cardiac drugs, statins, antibiotics, anticancer drugs, anti-HIV, immunosuppressive agents and more. [76]. It has been reported that P-gp is functionally interactive with CYP3A4. The induction and inhibition of P-gp are closely related to the occurrence of drug-drug interactions in clinical practice [78]. 1.2.2.2 Multidrug resistance-associated proteins (MRPs) MRPs, also known as ABCCs, are transmembrane proteins with nine types: MRP1 to MRP9. MRPs functions similarly as P-gp, expelling lipophilic and amphipathic molecules. Unlike P-gp, MRPs are localized in both apical and basolateral membranes. The most drug-related MRPs are MRP1, MRP2, MRP3, and MRP4. MRP1 is primarily localized in the basolateral membrane of cells and tissues, including lung, kidney, and skeletal as well as in macrophages [79]. MRP1 structure has a large core segment similar to P-gp, contains an N-terminal membrane spanning domain (MSD), two membrane-spanning domains and a nucleotide-binding domain following each transmembrane domain [80, 81]. MRP1 shows a broad range of substrates, but prefers amphiphilic organic anions and glutathione or glucuronate conjugates. Endogenous substrates include cysteinyl leukotriene LTC4, glutathione disulfide (GSSG), folate, steroids, and more. [82]. Drug substrates for MRP1 include chemotherapeutic agents (e.g. daunorubicin), HMG-CoA reductase inhibitors (e.g. atorvastatin), and HIV protease inhibitors (e.g. saquinavir) [83, 84]. MRP2 is exclusively expressed in the apical membrane of the liver, kidney, and intestine, demonstrating a similar distribution to UGT [59]. MRP2 consists of 1545 amino acids, Page 34 of 140 forming two nucleotide-binding domains, two membrane-spanning domains, a linker segment, plus the common membrane-spanning domain (MSD0) of MRP1 [80, 85]. Human MRP2 helices 6, 9, 16, and 17 bind substrates [86]. MRP2 also affects drug absorption, predominately lipophilic molecules conjugated with glutathione, glucuronate, and sulfate [87]. Substrates for MRP2 are generally same as that of MRP1. Endogenous substrates of MRP2 include glutathione, bile salts, bilirubin, estradiol, and leukotrienes [82]. Relevant drug substrates include chemotherapeutic agents (e.g. cisplatin), antibiotics (e.g. ampicillin), statins (e.g. pravastatin), and anti-HIV agents (e.g. saquinavir) [87, 88]. MRP3 is similar to MRP1 in terms of amino acid sequences. MRP3 transports molecules conjugated to GSH, sulfate, and glucuronate [89]. This isoform is highly expressed in the basolateral membrane of liver, intestine, and colon tissues. MRP3 shares overlapping substrates with MRP1 [77]. MRP4 is another MRP transporter covering a variety of substrates, both endogenous and xenobiotic. MRP4 is encoded by 1325 amino acids, forms two membrane-spanning domains (MSD) with two ATP-binding domains but does not have an N-terminal MSD [90]. MRP4 is unique among MRPs due to its dual membrane localization, found in both apical and basolateral sides of cells [91]. Endogenous substrates of MRP4 are cyclic and ADP nucleotides, eicosanoids and urate [92]. MRP4 targets antivirals (e.g. tenofovir), antibiotics, cardiovascular drugs and cytotoxic agents (e.g. methotrexate) [90]. 1.2.2.3 Breast cancer resistance protein (BCRP) BCRP is a member of the ABC-G family and is also classified as an MDR protein. Other names of BCRP include MXR, ABCP or ABCG2. It was originally identified in MCF-7 Page 35 of 140 breast cancer cells, where increased expression correlated with resistance towards chemotherapeutic agents [93]. BCRP is described as a “half transporter” as it only has one MSD and one NBD, instead of two of each [77]. BCRP is expressed on the apical membrane of the liver and kidney tissues and is highly expressed in the placenta and jejunum as well as in breast, colon, and stomach cancer cells [94, 95]. Endogenous substances of BCRP include folic acid, vitamin K3, uric acid and more. Drug substrates include statins (e.g. pitavastatin), cardiac drugs (e.g. azidopine), anticancer drugs, antibiotics (e.g. ciprofloxacin), as well as antiviral drugs (e.g. zidovudine) [60]. 1.3 Molecular Regulation of Metabolic Enzymes and Membrane Transporters Expression and Function Expression of CYP450 enzymes and transporters is regulated by a number of orphan nuclear hormone receptors, which was later to found to be nuclear transcriptional factors. Currently identified nuclear receptors are aryl hydrocarbon receptor (AhR), peroxisome proliferator activated receptor (PPAR), constitutive androstane receptor (CAR), pregnane xenobiotic receptor (PXR), farnesoid X receptor (FXR) and liver X receptor (LXR). The general structure of nuclear hormone receptors encompasses a ligand binding domain (LBD) to bind to xenobiotics and a conserved DNA binding domain (DBD) to bind to the DNA response elements in the promoter region of target genes. The target gene mediates associated protein products and thus alters the disposition of lipophilic xenobiotics through metabolism or cellular efflux elimination [96]. Nuclear Receptors AhR Aryl hydrocarbon receptor CAR Constitutive androstane receptor FXR Farnesoid X receptor Page 36 of 140 LXR Liver X receptor PPAR Peroxisome proliferator activated receptor PXR Pregnane xenobiotic receptor Table 1.3 Types of biosensor/nuclear receptors. 1.3.1 Constitutive androstane receptor (CAR) CAR is another essential nuclear receptor with transcriptional factor properties which belongs to nuclear receptor subfamily 1 group 1 (NR1I) [97]. Unlike other transcriptional factors, the base transcriptional activity of CAR is relatively high compared to other nuclear receptors since it can be active without a ligand [98]. CAR localizes in the cytoplasm in a phosphorylated form and can be activated by binding to specific endogenous molecules and xenobiotics, or in the presence of an activator compound, such as phenobarbital. After activation, CAR displaces the co-chaperone proteins, heat shock protein 90 (HSP90) and CAR retention protein (CCRP). CAR then translocates to the nucleus with the assistance of protein phosphatase PP2A, and binds to the retinoid xenobiotic receptor (RXR), leading to transcriptional activation [75, 99]. Expression of CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP3A4, UGT1As, UGT2B1, UGT2B5 and P-gp, MRPs BCRP and OATP2 appear to be regulated through CAR [90, 96, 100]. 1.3.2 Pregnane X receptor (PXR) PXR can be regarded as the most dominant nuclear receptor, mediating various vital DMEs as well as transporter proteins. PXR is also part of the NR1I family, sharing some similarity to CAR, but the PXR structure contains a large insert between helix 1 and helix 3 to form a more flexible ligand-binding pocket [97]. Many CAR ligands are also PXR Page 37 of 140 ligands. PXR is composed of 434 amino acids and is predominately expressed in liver and intestine tissues. Inactive PXR is found in the cytoplasm of cells, complexed with HSP90 and CCRP. PXR is primarily phosphorylated by protein kinase A and protein kinase C. Various xenobiotics trigger activation of PXR by the displacement of the repressor protein. The activated PXR then translocates into the nucleus and heterodimerizes with RXR to form a transcriptional factor. The heterodimer complex will then bind to the hormone response elements and xenobiotic responsive elements in the target promoter region to modulate the expression of DMEs along with transporter proteins [99]. PXR mediates the expression of a myriad of enzymes and transporters, namely CYP1A2, CYP2B6, CYP2B8, CYP2C9, CYP2C19, the CYP3A family with particular importance in CYP3A4, UGT1 family, P-gp, MRPs (e.g. MRP2, MRP5), BCRP, OATPs, and OCTs [99, 101]. CYP-mediated metabolism, Phase II metabolism, and efflux transporters are all largely dependent on PXR regulation [98]. Therefore, PXR is a key inducible system that affects drug response. Figure 1.3 PXR and CAR activation pathway [99]. Page 38 of 140 1.3.3 Aryl hydrocarbon receptor (AhR) AhR localizes in the cytoplasm of in barrier and mucocutaneous tissues such as skin, gut, and lung [31]. AhR, like other biosensors, remains in the inactivated state when its cognate co-repressors heat shock protein 90 (HSP90), aryl hydrocarbon receptor interacting protein (AIP), p23, and hepatitis virus X protein associated protein (XAP-2) are interacting with AhR. However AhR is activated when the co-repressors are displaced, which triggers translocation of AhR into the nucleus. In the nucleus, AhR will be complex onto AhR nuclear translocator (ARNT) to form a heterodimer. This AhR-ARNT complex binds to dioxin-responsive elements (DREs) and xenobiotic response elements (XREs) in the promoter region of target genes, upregulating the synthesis of relevant drug- metabolizing enzymes and transporters [98]. Additionally, AhR can regulate UGT expression. The bound AhR-ARNT complex also concomitantly activates additional transcription factors, such as nuclear transcription factor E2-related factor 2 (Nrf2). Together they regulate expression of CYP1As, CYP1Bs, GSTs, and some UGTs (e.g. UGT1A1, UGT1A6). In addition, AhR regulates BCRP transcription and expression. [96, 100] [102] Page 39 of 140 Figure 1.4 AhR and Nrf2 activation pathway [102]. 1.3.4 Nuclear transcription factor E2-related factor 2 (Nrf2) Distinct from above nuclear receptors, Nrf2 activation is generally not through a ligand- binding mechanism because Nrf2 has no LBD. Instead, Nrf2 is activated by the presence of activator molecules, which tend to be electrophilic compounds. Inactive Nrf2 associates with Keap1 protein and is retained in the cytosol. When activated by certain molecules or an oxidative stress condition, Nrf2 releases Keap1 and translocates to the nucleus. Nrf2 then forms heterodimers with Maf proteins and other transcriptional factors, and also complexes with co-activators such as CEP, all of which leads to transcription of the target genes. Interestingly, Nrf2 can also be regulated by the kinases PI3K, PKC, and MAPK. Drug-metabolizing enzymes and transporters modulated by Nrf2 include GSTs, UGTs (e.g. UGT1A6, UGT2B1), MRPs, and OATP2B1. [102] Page 40 of 140 1.3.5 Peroxisome proliferator activated receptor (PPAR) PPARs is classified into three isoforms, PPARα, PPARβ, and PPARγ. PPARα localizes in tissues of the intestines, heart, liver, and kidney. PPARβ is mainly found in the intestine, brain, and kidney tissues. PPARγ is mostly localized in liver and adipose tissues. Similar to CAR and PXR, inactive PPAR is associated with co-repressors, in this case nuclear receptor co-repressors and SMRTs. PPARs’ ligands are mostly fatty acids and eicosanoids, such as 20-HETE and LTB4. When activated, PPAR displaces the co- repressors while recruiting co-activators, such as CREB-binding protein and mediator complex subunit 1, and forms a heterodimer. The PPAR complex then binds to the peroxisome proliferator response elements (PPRE) in the promoter region of target genes, stimulating transcription of drug-metabolizing enzymes and transporters [98]. PPARs regulate the CYP4 family, UGTs (e.g. UGT1A1, UGT1A4, UGT2B4 and UGT2B7) and MRPs (e.g. MRP4) [31, 90]. 1.3.6 Other Orphan Nuclear Transcriptional Factors Liver X receptor (LXR) and farnesoid X receptor (FXR) are two lesser nuclear receptors that also participate in the regulation of drug-metabolizing enzymes and transporters. LXRα is found mostly in the liver while LXRβ is found in most tissues. LXR forms a heterodimer with RXR after activation by ligands, which will bind to LXR response elements in the promoter region of target genes. It is known that LXR can partially mediate CYP3A4 and CYP2B6 expression. FXR is found in the liver and intestine. Similarly, FXR binds to RXR, which will recognize FXR response elements in the xenobiotic responsive enhancer modules of target genes. FXR can regulate bile salt export pump and CYP7A. [98] Page 41 of 140 Kinases may also be capable of regulating the expression of enzymes and transporters. There is some evidence that P-gp expression is regulated by PKC. Specifically, PKC is able to phosphorylate the linker structure of P-gp, altering its expression and function [77]. Interaction of kinases with AhR, CAR, and PXR has been reported. AhR is found to be a target gene of PXR while CAR is somewhat impacted by AhR [103, 104]. CYP1A1 and CYP1A2 are found to be indirectly governed by PXR due to induction of AhR by PXR [101]. Expression and modulation of drug-metabolizing enzymes and transporters involve multiple mechanisms. However, nuclear receptors play the key role in their molecular regulation. The importance of these nuclear receptors indicates that evaluation of a drug’s impact on these receptors needs to be thorough. Currently, PXR is the major factor tested for potential drug-related induction. Yet other nuclear receptors can also affect drug disposition, which should also be taken into consideration. 1.4 Individual Factors on Drug Metabolism and Disposition Factors that further complicate drug disposition include variation in individuals and genetic polymorphisms. These differences affect the precision use of drug therapies by also potentially impacting DMEs and transporters. 1.4.1 Genetic Polymorphisms Allelic variants, or more commonly referred to as genetic polymorphisms, of drug- metabolizing enzymes and transporters genes have gained greater attention in recent clinical practice. Ethnicity is one determining factor of enzyme polymorphisms, often associated with the incidence of distinct single nucleotide polymorphism (SNP) frequencies. These variations can result in wildly different drug response among patients. Page 42 of 140 For example, CYP2D6 polymorphisms affect venlafaxine and tricyclic antidepressants, resulting in poor efficacy and side effects in poor metabolizing phenotypes [2]. Another example is the effect of CYP2C9 on warfarin use. Patients with CYP2C9*2/*3 should be given a reduced starting dose of warfarin due to reduced metabolism [2]. Additionally, UGT1A1 G71R, found in higher frequency among East Asians, is connected with Gilbert syndrome due to reduced glucuronidation ability while the UGT1A1 TATA box polymorphism is related to the increased side effects of irinotecan [36]. Transporter polymorphisms also exert influence on drug disposition and occurrence of certain diseases. For example, the SNP C3435T at exon 26 of P-gp (MDR1 gene) occurs with high frequency among European Americans. Presence of the SNP results in lower digoxin plasma levels [77]. The presence of MRP2 (ABCC2) SNP 2302 C>T A768W is an indicator for Dubin Johnson Syndrome. OATP1B1 (SLCO1B1) SNPs are linked with lower drug efficacy in pravastatin due to lowered pravastatin plasma levels [105]. Polymorphisms of other transporters like BCRP and OCT2 have also been reported to alter drug metabolism and dispositions. 1.4.2 Gender and age Gender and age impact drug metabolism and disposition primarily through hormones. During adolescent development, CYPs expression alters with age. In general, CYPs have limited functions among neonates and increases dramatically in the first year, along with the development of individual variation [5]. Specifically, CYP3A7 is highly expressed in the fetus and diminishes to very low levels over the first several years of life [106]. Puberty is another important developmental period, as levels of drug-metabolizing enzymes fluctuate dramatically due to the constantly changing levels of endogenous hormones. Page 43 of 140 CYP3A4, ubiquitously expressed in adults, is low in children until post-puberty [107]. These differences in CYP expression is one of the many challenges in treating the pediatric population [106]. Another condition that impacts drug metabolism and disposition is pregnancy. Pregnant women experience multiple physiological changes, particularly increasing levels of estrogen, progesterone, prolactin and placental growth hormones. Nuclear receptors like PXR and CAR are regulated by these hormones, thus the downstream expression of drug-metabolizing enzyme and transporters will change accordingly. For example, elevated drug metabolism through CYP2D6 and CYP3A4 has been reported in pregnant women. [107] The physiological differences in gender also influence the metabolism and elimination of endobiotics and xenobiotics. Of note, results from a study of 400 adult subjects claim that substantial difference of brain region and relevant glucose metabolism were found, specifically that females displayed a higher metabolism than males [108]. Aging is another factor impacting drug metabolism and disposition. The elderly population often have reduced hepatic and renal clearance, resulting in increased plasma levels. Body composition also changes dramatically (e.g. blood flow, volume of distribution and organ functions) plus the number and intensity of chronic diseases are often higher in the elderly population versus the normal adult; both contribute to the alternation of drug metabolism and disposition [109]. Additionally, the function and activity of drug-metabolizing enzymes and transporters are affected by the aging process, further altering drug disposition [110]. Page 44 of 140 1.4.3 Other factors Patient diet and lifestyle habits can affect drug metabolism and disposition. For example, grapefruit juice can reduce circulating levels of OATP substrates and P-gp substrates by inhibiting both transporters [59]. Additionally, coffee and broccoli can lead to induction of CYP1A2 [100]. Smoking can modulate drug metabolism through the effects of nicotine. Nicotine induces CYP1A1 and CYP1A2, impacting the metabolism of theophylline, tacrine, imipramine, propranolol, flecainide, and estradiol. It is also reported that smoking will lead to faster clearance of heparin [111]. Lastly, diseases such as liver and renal diseases will also alter the expression and functions of enzymes and transporters, resulting in significant changes in drug disposition. Often, dosage adjustment is required in patients with these diseases. 1.5 Influences of drug metabolism and disposition on drug efficacy Drug metabolism and disposition determine the empirical drug concentration in a defined tissue or cellular type. Drug efficacy, on the other hand, is decided by the effective drug concentration at target tissues and cells. In clinical practice, drug-drug interactions are often caused by induction and inhibition of metabolizing enzymes or transporters. The interplay between enzymes and transporters affect the intracellular drug concentrations, thereby impacting drug efficacy. 1.5.1 Induction and inhibition Induction and inhibition of drug metabolism can significantly impact drug efficacy. They are typically aroused by the augmented activity of DMEs and transporters. Drugs are xenobiotics and can activate nuclear receptors, facilitating the expression and function of Page 45 of 140 the enzymes and transporters. For instance, P-gp induction will diminish the bioavailability of digoxin by more than 20% while increasing the clearance of talinolol and carvedilol [59]. As mentioned above, polymorphism of DMEs will result in different drug metabolism extents, such as poor metabolism and extensive metabolism. In practice, warfarin is an example under this circumstance. It is known that warfarin has a narrow therapeutic window with high individual variation partially due to polymorphism of CYP2C9. In detail, CYP2C9*2 reduces warfarin metabolism by 30% while CYP2C9*3 reduces warfarin metabolism by 90% [112]. Thereby, the change of drug disposition and metabolism will ultimately impact its clinical efficacy. 1.5.2 Drug-drug interactions (DDIs) Drug-drug interactions (DDIs), a major source of therapeutic difficulty, often are caused by the induction or inhibition of DMEs and transporters. DDIs are particularly important in ARV therapies as most anti-HIV therapies contain multiple ARTs. For example, by impacting P-gp expression, administration of Kaletra (lopinavir/ritonavir) elevates the plasma concentration of concomitant protease inhibitors atazanavir and saquinavir but decreases the plasma concentration of darunavir [113]. Valproic acid and probenecid inhibit UGT2B7, which increases plasma levels of lorazepam and zidovudine. Concurrent administration of the UGT2B7 inducer with its substrate (ex. valproic acid with zidovudine) could result in toxic levels of the “victim” (zidovudine) [30]. Thus multi-drug therapies should be screened for possible DME and transporter interactions to avoid clinically disastrous outcomes. Page 46 of 140 1.5.3 The interplay between drug metabolizing enzymes (DMEs) and transporters Since DMEs and transporters share similar or even identical molecular regulation pathways, they often interact with each other and accordingly alter drug efficacy. CYP3A and P-gp exhibit overlapping tissue distribution and substrates, which can lower the efficacy of their substrates by repeating transport cycles to maximize access to CYP3A [75]. For example, atorvastatin is transported by OATP1B1 and metabolized by CYP3A4 and UGT1A1 in hepatocytes. However, atorvastatin also interferes in its uptake by inhibiting OATP1B1, hereby intervening its following metabolism and leaving a higher plasma level [59]. 1.6 Hypothesis and specific aims In order to understand the impact of various factors on specific drug disposition in vivo and in vitro, I used two different experimental approaches to achieve this goal. The first was to assess the impact of age on individual factors and metabolic capacity on voriconazole pharmacokinetics in pediatric patients with fungal infections. In this study, the emphasis was put on metabolic variation factors such as DMEs and their impacts on voriconazole metabolism among the special population. It was found that larger than calculated doses are required for children patients to reach therapeutic target levels. This may be a consequence of FMO3 activity in younger patients, as the expression levels decline as children mature into adolescence. To further complicate the dosing, voriconazole has nonlinear pharmacokinetics and a narrow therapeutic window. These features of voriconazole mandate precise dosing to optimize safety and efficacy. The hypothesis is that ontogeny of CYP2C19, CYP3A4 and especially FMO3 will significantly Page 47 of 140 correlate with observed age-related changes in voriconazole PK. Specific aims include (1) Describe voriconazole PK profile in pediatric patients and (2) Characterize CYP2C19, CYP3A4, and FMO3 activity in relation to age-related voriconazole metabolism in pediatric patients. To understand the impact of intracellular expression of cellular transporters, I evaluated the impact of pharmacoenhancers (PEs) ritonavir and cobicistat on drug disposition of HIV protease inhibitors (PIs). PEs are used as CYP3A4 inhibitors to enhance the pharmacokinetic profile of concomitant protease inhibitors. However, clinical data suggests that the second-generation PE cobicistat was associated with higher virologic failure among patients. Since virologic failure is usually due to low cellular concentration of drugs in infected CD4+ cells, assessing the cellular disposition of PIs after boosting by the two PEs can improve treatment efficacy. I hypothesized that COBI dissimilarly alters the expression and function of cellular efflux transporters and CYP3A enzymes compared to RTV, and these differences explain the divergence in the cellular disposition of ATV and DRV in CD4+ cells. Three specific aims were generated from this hypothesis: (1) Determine the cellular concentrations, elimination rate constant (Ke), and cellular AUC of PIs when given alone and in combination with COBI or RTV in CD4+ (monocytic and lymphocytic) cell lines; (2) Determine impact of PI/PE treatments on efflux transporters; and (3) Examine PE/PI impact on CYP3A4 and the nuclear receptors PXR and CAR. 1.7 Outline of the dissertation In Chapter 2, I detail the impact of age on the metabolism of various drug substrates, voriconazole in particular. Voriconazole not only has nonlinear pharmacokinetics (PK) profile but also requires a dose greater than allometric scaling for pediatric patients Page 48 of 140 compared to adults. In this project, I analyzed the metabolic function of three major DMEs: CYP3A4, CYP2C19, and FMO in relation to voriconazole metabolism through a probe substrates cocktail for these enzymes. Also, I further assessed individual factors, such as gender and age, and their influences on voriconazole metabolism. The results provide mechanistic data for voriconazole dosing strategy among pediatric patients. In Chapter 3 and 4, I conducted in vitro assays to evaluate drug induction and inhibition on cellular disposition. In order to fulfill the aims, I first determined the cellular retention, cellular exposure, and clearance of boosted protease inhibitors after PI/PE treatments in Chapter 3. In addition, I studied factors that affect cellular drug disposition in CD4+ cells, particularly the efflux transporters P-gp, MRPs, and BCRP in Chapter 4. I utilized various complementary methods and assays to evaluate the expression and function of these factors after treatments. The results elucidate the mechanistic differences of cobicistat and ritonavir and thus explain the clinical differences. Furthermore, these results may improve prediction of virologic failure despite administering therapeutic doses. Page 49 of 140 Chapter 2. Pediatric individual factors and metabolic capacity in relation to voriconazole therapeutic effects Abstract Voriconazole has been shown to have age-related clearance. However, the mechanism for how this occurs has not been well delineated. Nonlinear pharmacokinetics (PK) of voriconazole suggests that clearance is dependent on metabolic catabolism. It is known that CYP2C19 is a major enzyme system catalyzing voriconazole metabolism in adults. CYP3A4 and FMO3 have also been identified as being important in the metabolic process. I hypothesized that the ontogeny of CYP2C19, CYP3A4, and FMO3 will closely relate to the age-related voriconazole metabolism. To investigate this hypothesis, I utilized probe cocktails to determine the specific enzyme activity and its relationship with actual voriconazole metabolism. I evaluated other factors that may impact metabolism including routes of administration, age, and sex. Results suggest that FMO3, in addition to CYP2C19, are critical factors in age-related voriconazole metabolism. This finding offers a molecular explanation of pediatric voriconazole pharmacokinetics, which provides direction for future dosing strategy optimization. 2.1 Introduction 2.1.1 Pharmacotherapy for pediatrics Pharmacogenomics has been heralded as the seminal tool that will usher in the practice of individualized therapy [114, 115]. Although the entire human genome has been sequenced for several years, implementation of precision medicine in the clinical setting is still limited to the cutting edge of technology. Various pharmacogenomic approaches have been tested and implemented, yet the ability to phenotype individuals continues to Page 50 of 140 be the best predictor of drug absorption, distribution, metabolism and excretion (ADME) [116]. This is particularly of importance for drugs that are eliminated through enzymatic biotransformation and thus subjected to alteration of enzyme expression during the aging process [117, 118]. Precise pharmacotherapy in the pediatric population is incredibly important since small changes in dose can transform efficacy to toxic levels, particularly in drugs with already low therapeutic index. Currently, pediatric dosing is dependent on adult dosages scaled by weight, body surface area, or other size metrics [119]. Despite size-scaled dosing, reaching target concentrations in pediatric patients is not guaranteed and thus continues to be a challenge. Simply to increase dosage without definitive data increases risks for drug-related adverse effects as well as lack of efficacy. To further complicate issues, the expression of various metabolic enzymes have been previously documented to alter based on age, such as the expression of various isoforms of CYP3A. For example, at birth the expression of CYP3A7 is predominant in infants, while expression of CYP3A5 predominates until puberty [16]. At puberty, there is evidence that CYP3A4 emerges as the most abundant metabolic enzyme found in the liver. Thus compounds that are eliminated primarily through hepatic metabolism will encounter the high levels of variability throughout the aging process from birth to puberty [120]. An additional concern in the pediatric population is the elevated risk of drug-related toxicities. As a result, the capacity to phenotype drug metabolizing potential can potentially avoid unexpected drug toxicities while improving effectiveness among the younger pediatric population. Page 51 of 140 2.1.2 Voriconazole Voriconazole is an FDA approved drug to treat difficult fungal infections in the pediatric population. It is a potent antifungal agent with broad spectrum activity against mold, yeast and fungal infections, including Aspergillosis species [121]. Unlike other triazole-based antifungal agents, voriconazole exhibits a narrow therapeutic window with a range from 1.0 to 6.0 mg/L [122, 123]. Voriconazole has a saturable metabolism which explains the nonlinear pharmacokinetics observed in both adults and children [124, 125]. As previously mentioned, pediatric patients have higher clearance and thus require a higher dose when compared to adult patients to maintain steady state therapeutic concentration [126, 127]. Multiple trials suggest that the threshold for clinical efficacy correlates with the ability to attain voriconazole trough concentration (Cmin) above 1 mg/L [126]. Despite these findings, the molecular mechanism for why the need for higher dosage in the pediatric setting has not been clearly delineated. There are hints that voriconazole elimination is a key mechanism, particularly the preceding metabolism in predicting circulating level and therapeutic effects. Voriconazole is metabolized by drug metabolizing enzymes that include CYP3A4/5, CYP2C19 and FMO3 [128, 129]. To determine the impact of age on the clearance of voriconazole, I employed a probe cocktail to evaluate the enzymatic activity of these enzyme systems and their contribution to metabolic clearance. This method was used to characterize the metabolic pathways associated with the clearance of voriconazole. 2.1.3 Metabolism probes The use of approved drugs as pharmacologic probes to evaluate metabolic activities is a novel method to determine the contribution of various metabolic systems involved in the Page 52 of 140 hepatic clearance of voriconazole. The hepatic clearance was calculated by determining the ratio of the parent and metabolite levels over time. The probe cocktail is a combination of drug probes with well-characterized metabolism using specific metabolic pathways. Ideal drug probe candidates are compounds with only one metabolite, which is produced by a known enzymatic system and gives a good estimate of the contribution of that specific pathway [130]. Each probe compound in the cocktail was a specific substrate utilizing a different metabolic enzyme for clearance. The ability to determine the change of parent compound and metabolite generated over a timeframe allowed us to characterize a particular metabolizing enzyme. In this study, I used an enzyme probe cocktail that can be administered to younger patients with high therapeutic index (TI). In order to probe the CYP3A4/5, CYP2C19 and FMO3 pathways, the probe cocktail contained midazolam, esomeprazole, and ranitidine. To enhance the safety margins, the cocktail dosage was 10% of the normal therapeutic dosages. This probe cocktail was administered by intravenously or orally. To complement the probe cocktail, I developed an accurate, highly sensitive, and precise LC-MC/MS method to measure the plasma concentrations of these probes and their metabolites simultaneously. It has been reported that CYP2C19 polymorphisms play a key role in the PK of voriconazole and its primary metabolite; however, it may not be the only factor for dose adjustment in the pediatric population [131-133]. In pediatric patients, there are distinct enzymes that are absent or otherwise undetectable in adults. The presence of these enzymes has significant impact on the clearance of various compounds. To assess the various types of enzymes seen in the pediatric population, I evaluated individual factors such as gender/sex, age, and metabolic capacity in relation to voriconazole metabolism. Page 53 of 140 In this study, I hypothesized that ontogeny of CYP2C19, CYP3A4, and FMO3 significantly correlates with observed age-related changes in voriconazole PK and thus has a major impact on dosing and therapeutic drug monitoring (TDM) strategies. 2.2 Material and methods 2.2.1 Chemicals and Materials HPLC grade voriconazole, midazolam, 1-OH midazolam, diazepam, esomeprazole, and formic acid were purchased from Sigma Aldrich (St. Louis, MO). Voriconazole N-oxide, 5-OH omeprazole, ranitidine, ranitidine N-oxide, cimetidine, lansoprazole, and fluconazole were purchased from Santa Cruz Biotechnology (Dallas, TX). Acetonitrile was purchased from EMD Millipore (Billerica, MA). Other lab supplies used in this project, such as centrifuge tubes, were purchased from VWR (Radnor, PA). 2.2.2 Study design This project was registered at clinicaltrials.gov under No. NCT 01976078 with the title “Development of voriconazole pharmacokinetics and metabolism in children and adolescents.” Subjects included were pediatric patients who were already prescribed voriconazole. 2.2.3 Subjects and procedures Patients who received voriconazole under the age of 18 years were eligible for enrollment. After receiving their prescribed voriconazole, either intravenously or orally, blood samples were collected intravenously at the time point of 0 (pre-dose), 0.5 (oral only), 1, 1.5 (oral only), 2, 4, 6, 8 and 12 hours after administration. Blood samples were immediately Page 54 of 140 centrifuged at 1000 rpm for 10 minutes at 4 o C. Plasma was then isolated and transferred to a cryogenic tube and stored at -80 o C till analysis. 2.2.4 Determination of probes and voriconazole A multiplex liquid chromatography-mass spectrometry (LC-MS) assay was developed to measure the plasma concentrations of the probes and their respective metabolites. The assay was performed on a Shimadzu LC-20AD HPLC (Shimadzu, Japan) linked to an AB Sciex API4000 triple-quadrupole tandem mass spectrometer was operated in positive electrospray ionization mode (ESI) (AB Sciex LLC, USA). During the assay validation, each calibration standard and quality control sample was prepared by spiking a specified amount of midazolam, 1-hydroxyl midazolam, esomeprazole, 5-hydroxyl omeprazole, ranitidine, ranitidine N-oxide and diazepam/cimetidine/lansoprazole (internal standards), into 150 µL of clarified plasma. The isolated analytes were extracted by protein precipitating from the plasma using 600 µL of acetonitrile. The samples were centrifuged at 13,000 rpm at 4 o C for 15 minutes, where the supernatants were then collected and evaporated to dryness under a steady stream of dried filtered nitrogen. The residues were reconstituted with 50 µL of acetonitrile. After centrifuging at 13,000 rpm for 5 minutes, the clarified supernatants were transferred into an HPLC vial and 45 µL was injected to the LC-MS. Analytes were separated by a Gemini column (Phenomenex, USA) with the following dimensions, 150 x 4.6 mm, 3 µm particle size (Part #: 00F-4439- E0; Serial #: 634302-1). Analytes were detected using multiple-reaction-monitoring with their own distinctive M + →T + signature for each of the respective analytes. Mobile phases included 10mM ammonium acetate in water (component A) and acetonitrile (component B), where a gradient from 10% B to 90% B was employed. Page 55 of 140 Plasma concentration of voriconazole and its metabolite were determined by using the same LC-MS system described above. The appropriate concentration of voriconazole and voriconazole N-oxide standard was added into 50 µL of human blank plasma. Then, 50 µL of fluconazole (1000 ng/mL) was added to each sample as the internal standard. After mixing thoroughly, 300 µL of acetonitrile was added for protein precipitation. Samples were centrifuged at 13,000 rpm for 5 minutes and 50 µL of supernatant was injected into the LC-MS for quantitation. The isolated analytes were separated using reverse-phase high performance LC ACE 3 C18 column (ACE) with the following dimensions of 50 x 3.0 mm, 5μm particle size (Part #AC3-111-0503T). Voriconazole, voriconazole N-oxide, and fluconazole were detected using multiple-reaction-monitoring (MRM) specific for each of the respective analyte. Mobile phases consist of water with 0.1% formic acid (component A) and acetonitrile with 0.1% formic acid (component B). Mobile phase B was increased from 60% to 90% in 1 minute and held for another 3 minutes. The total analytical run time was 6 minutes. 2.2.5 Statistical Analysis and Software modeling Enzyme activity and voriconazole metabolic ratio were calculated as the ratios of the metabolite to parent AUC. GraphPad Prism 5 (GraphPad Software, Inc. CA, USA) and MATLAB (The MathWorks, Inc, MA, USA) were used for statistical analysis and figures. Pmetrics (www.lapk.org) was used for non-compartmental analysis [134]. Page 56 of 140 2.3 Results 2.3.1 Demographics The demographic information of the subjects in this study is summarized in Table 2.1. A total of forty-five subjects were recruited for this study with an average age of 8.6 years old and a range from 0.6 to 17 years of age. Eleven of the subjects received voriconazole as an intravenous administration, while thirty-one subjects received oral dosages. There were three subjects who received voriconazole by both routes. Characteristics Age (years) 8.6 (0.6 - 17) Route of administration IV 11 PO 31 IV&PO 3 Sex/Gender Male 24 Female 21 Race Caucasian 38 Asian 3 African American 4 Table 2.1 Patient demographics and demographic parameters. 2.3.2 LC-MS/MS Assay All the probe substrates (which include midazolam, esomeprazole, ranitidine and their respective metabolites) underwent qualification testing. The intraday and interday variation were evaluated on three separate days where the standard curves and qualify controls were determined. In this study, the lowest limit of quantification (LLQ) was identified at 0.05 ng/mL for all the analytes. Linear regression was applied to the standard curves where R 2 for all of the analytes were over 0.99. Dynamic range for probe assay Page 57 of 140 ranged from 0.05 to 200 ng/mL, with an accuracy and precision within 100% ± 15% where the coefficient of variation (CV%) did not exceed 15%. Table 2.2 and Table 2.3 summarizes the intraday and interday accuracy of the probe analytes at 0.5, 10 and 100 ng/mL. The chromatogram of the assay is shown in Figure 2.1a. Table 2.4 summarizes the interday accuracy results of the quality controls of voriconazole assay. The lowest limit of quantitation (LLOQ) was 100 ng/mL for both voriconazole and its metabolite voriconazole N-oxide. The dynamic range of this assay is from 100 to 20000 ng/mL. The chromatogram of the assay is shown in Figure 2.1b. Figure 2.2 showed the robustness of interday standard curves of both assays. Probe drugs Concentration (ng/mL) Accuracy (%) SD CV (%) Midazolam 0.05 101 1.3 1 10 99 2.6 3 100 95 4.9 5 1-OH Midazolam 0.05 100 0.6 1 10 98 7.5 8 100 98 5.3 5 Esomeprazole 0.05 100 0.3 0.3 10 102 9.3 9 100 99 7.8 8 5-OH Omeprazole 0.05 100 2.9 3 10 104 5.1 5 100 102 11.2 11 Ranitidine 0.05 101 1.0 1 10 104 4.8 5 100 97 3.6 4 Ranitidine N- Oxide 0.05 99 0.3 0.3 10 101 5.5 5 100 102 4.4 4 Table 2.2 Intraday accuracy (n=3) of the probes quality controls. Page 58 of 140 Probe drugs Concentration Accuracy SD CV (%) Midazolam 0.05 101 1.4 1 10 108 3.2 3 100 94 4.2 4 1-OH Midazolam 0.05 97 1.6 2 10 97 4.1 4 100 97 8.2 8 Esomeprazole 0.05 99 0.5 1 10 96 2.9 3 100 99 2.0 2 5-OH Omeprazole 0.05 98 1.6 2 10 96 5.2 5 100 104 2.3 2 Ranitidine 0.05 99 3.4 3 10 91 4.7 5 100 100 3.1 3 Ranitidine N- Oxide 0.05 99 3.2 3 10 99 2.7 3 100 102 13.2 13 Table 2.3 Interday accuracy (n=3) of the probes quality controls. Drugs Concentration Accuracy (%) SD CV (%) Voriconazole 250 97 13.8 14.3 1000 97 11.0 11.3 5000 101 1.5 1.5 Voriconazole N-Oxide 250 93 2.6 2.8 1000 92 4.8 5.2 5000 106 9.8 9.2 Table 2.4 Interday accuracy (n=3) of voriconazole and voriconazole N-oxide quality controls. Page 59 of 140 (a). Page 60 of 140 (b). Figure 2.1 Chromatograms of (a) probes and (b) voriconazole LC-MS/MS assays. (a) Chromatogram of the parent and their respective metabolites. Probes and metabolites from top to bottom: ranitidine, ranitidine-N-oxide, 5-OH omeprazole, omeprazole, 1-OH midazolam, and midazolam. (b) Chromatogram of voriconazole (top) and voriconazole N- oxide (bottom). Page 61 of 140 Voriconazole 0 5000 10000 15000 20000 25000 0.0 0.5 1.0 1.5 Day1 Day2 Day3 concentration (ng/ml) Analyte area/IS area Voriconazole N-oxide 0 5000 10000 15000 20000 25000 0.000 0.001 0.002 0.003 Day1 Day2 Day3 concentration (ng/ml) Analyte area/IS area Midazolam 0 50 100 150 200 250 0 1 2 3 4 5 set 1 set 2 set 3 Concentration(ng/mL) Analyte area/IS area 1-OH Midazolam 0 50 100 150 200 250 0.00 0.05 0.10 0.15 0.20 0.25 Set 1 Set 2 Set 3 Concentration(ng/mL) Analyte area/IS area Esomeprazole 0 50 100 150 200 250 0.0 0.5 1.0 1.5 2.0 2.5 Set 1 Set 2 Set 3 Concentration(ng/mL) Analyte area/IS area 5-OH Omeprazole 0 50 100 150 200 250 0.0 0.1 0.2 0.3 0.4 0.5 Set 1 Set 2 Set 3 Concentration(ng/mL) Analyte area/IS area Ranitidine 0 50 100 150 200 250 0.0 0.2 0.4 0.6 0.8 Set 1 Set 2 Set 3 Concentration(ng/mL) Analyte area/IS area Ranitidine N-oxide 0 50 100 150 200 250 0.0 0.2 0.4 0.6 Set 1 Set 2 Set 3 Concentration(ng/mL) Analyte area/IS area Figure 2.2 Interday standard curves (n=3) for voriconazole and pharmacologic probes and their respective metabolites. Top to bottom: voriconazole and voriconazole N-oxide; midazolam and 1-OH midazolam; esomeprazole and 5-OH omeprazole; ranitidine and ranitidine-N-oxide. Page 62 of 140 2.3.3 Pharmacokinetics parameters of voriconazole treatment Plasma samples from the subjects receiving voriconazole and probe cocktail were collected at designated time points. Voriconazole and its metabolite N-oxide-voriconazole levels were determined. Each of the drug probes and their respective metabolites were also measured. Non-compartmental PK analysis of voriconazole and N-oxide voriconazole was conducted for each patient with relevant PK parameters and summarized in Table 2.5. Median and mean PK parameters corresponded for both oral and intravenous administration. Oral administration was associated with a lower voriconazole Cmax and AUC. In addition, slightly higher voriconazole N-oxide values were seen with the oral dosing when compared to the IV group. The plasma half-life of voriconazole and its metabolite were similar between PO and IV dosing routes, which was approximately five hours and thirteen hours for voriconazole and voriconazole N- oxide, respectively. Route PK Parameters Voriconazole Voriconazole N-Oxide Mean ± SD Median (Range) Mean ± SD Median(Range) PO (n=31) Cmax (µg/mL) 5.7±4.4 4.6 (1.4-18.1) 9.8±5.0 8.3 (3.4-20.2) Tmax (hrs) 1.6±1.0 1.5 (0.5-4.0) 3.1±2.4 2.0 (0-12.0) t 1/2 (hrs) 5.1±2.4 4.7 (2.1-11.5) 14.2±7.9 13.4 (4.2-33.0) AUC 0-12 (µg*hr/mL) 31.4±26.5 21.8 (5.1-125) 87.8±45.1 78.0 (35.8-200) IV (n=11) Cmax (µg/mL) 14.7±13.0 11.4 (2.9-46.0) 7.7±3.4 7.4 (2.9-13.7) Tmax (hrs) 2.7±2.0 2.0 (1.0-6.0) 4.2±2.6 4.0 (0.5-8.0) t 1/2 (hrs) 5.0±2.9 4.1 (2.1-11.3) 13.6±6.7 11.7 (6.6-29.8) AUC 0-12 (µg*hr/mL) 78.0±64.6 69.6 (16.6-249) 73.7±34.3 71.4 (28.6-146) Table 2.5 Mean and median pharmacokinetic parameters of voriconazole and its N- oxide voriconazole. Page 63 of 140 2.3.4 Enzyme activity and individual factors in relation to the metabolic ratio of voriconazole In this study, I evaluated the enzymatic activity of CYP3A4, CYP2C19, and FMO3 in relationship to their role in eliminating voriconazole. Non-compartmental pharmacokinetic analysis (NCA) was applied to calculate AUC of the analytes for each subject. The mean ± SD value of voriconazole metabolic ratio and enzyme relative activity is summarized in Table 2.6. In general, enzyme relative activity was similar among the three metabolizing enzymes though the voriconazole metabolic ratio was considerably higher. Enzymatic System Parent/Metabolite NCA Calculated Ratio Voriconazole AUC(vori N-oxide)/AUC(vori) 3.74±3.38 CYP3A4/5 AUC(1-OH MDZ)/AUC(MDZ) 0.19±0.21 CYP2C19 AUC(5-OH OMP)/AUC(OMP) 0.12±0.08 FMO3 AUC(Ran N- oxide)/AUC(Ran) 0.40±0.39 Table 2.6 Voriconazole metabolic ratio and relative activity of CYP3A4, CYP2C19, and FMO3 (Mean ± SD). Individual factors can impact the metabolic ratio of voriconazole, thus I analyzed the impact of age on voriconazole metabolism and compared them across the relevant enzyme activities. To do this, I used a linear regression comparing age and the metabolic ratio between the parent and metabolite. In addition, I also evaluated the impact of dosing routes on the metabolic profile, with results summarized in Figure 2.3. The results demonstrated that the metabolic ratio was higher among younger patients, and the metabolic capacity declined with increasing age (R 2 =0.051). A decline in the metabolic activities of the target enzymatic systems CYP3A4/5, CYP2C19, and FMO3 was seen with advancement in age and had corresponding effects on voriconazole metabolism. Page 64 of 140 Reduction of CYP2C19 activity due to age was more significant than the other two enzyme systems (R 2 =0.050). Given that the sample size of this study was relatively small, I also conducted nonparametric analysis to evaluate the correlation between age and metabolism. For the Spearman correlation test, the Spearman correlation coefficient was -0.3525 and -0.3930 for voriconazole metabolism and CYP2C19 activity respectively. Furthermore, the effect of age on voriconazole metabolism (p=0.0324) as well as age and CYP2C19 activity (p=0.0161) were found to be statistically significant. Figure 2.3 Age in relation to voriconazole metabolic ratio and enzyme activity. (a) CYP3A4/CYP3A5 relative activity. (b) CYP2C19 relative activity. (c) FMO3 relative activity. (d) Voriconazole metabolic ratio. Linear regression with R 2 value is shown in each plot. Linear regression models were then established to compare relative enzyme activity to voriconazole metabolic ratio for both IV and PO administration groups (Figure 2.4 & 2.5). Page 65 of 140 For the PO group, FMO3 (p=0.02) and CYP2C19 (p=0.02) activity displayed a strong correlation to the voriconazole metabolic ratio. For the IV group, FMO3 activity exhibited a strong linear relation to the voriconazole metabolism (R 2 =0.8175). FMO3 (p=0.02) and CYP2C19 (p=0.01) correlate with voriconazole metabolism as well. Therefore, both FMO3 and CYP2C19 significantly affect voriconazole metabolism in children in both IV and PO routes of administration. PO Enzyme Activity vs Voriconazole Metabolic Ratio 0 5 10 15 0.0 0.5 1.0 1.5 FMO3 Voriconazole Metabolic Ratio Enzyme activity PO Enzyme Activity vs Voriconazole Metabolic Ratio 0 5 10 15 20 0.0 0.5 1.0 1.5 CYP3A4/5 Voriconazole Metabolic Ratio Enzyme activity PO Enzyme Activity vs Voriconazole Metabolic Ratio 0 5 10 15 20 0.0 0.1 0.2 0.3 0.4 0.5 CYP2C19 Voriconazole Metabolic Ratio Enzyme activity (a) (b) (c) R 2 =0.15 Spearman r=0.40 (P=0.02) R 2 =0.38 Spearman r=0.16 R 2 =0.02 Spearman r=0.41 (P=0.02) Figure 2.4 Relationship between enzyme activity and voriconazole metabolic ratio in oral groups. (a) FMO3. (b) CYP3A4/5. (c) CYP2C19. Linear regression with R 2 value and Spearman r value is shown in each plot. Page 66 of 140 IV Enzyme activity vs Voriconaozle Metabolic Ratio 0 2 4 6 8 0.0 0.5 1.0 1.5 FMO3 Voriconazole Metabolic Ratio Enzyme activity IV Enzyme activity vs Voriconaozle Metabolic Ratio 0 2 4 6 8 0.0 0.2 0.4 0.6 0.8 CYP3A4/5 Voriconazole Metabolic Ratio Enzyme activity IV Enzyme activity vs Voriconaozle Metabolic Ratio 0 2 4 6 8 0.0 0.1 0.2 0.3 0.4 CYP2C19 Voriconazole Metabolic Ratio Enzyme activity (a) (c) (b) R 2 =0.84 spearman r=0.81 (P=0.02) R 2 =0.03 spearman r=0.02) R 2 =0.2 spearman r=0.8 (P=0.01) Figure 2.5 Relationship between enzyme activity and voriconazole metabolic ratio in oral groups. (a) FMO3. (b) CYP3A4/5. (c) CYP2C19. Linear regression with R 2 value and Spearman r value is shown in each plot. I also evaluated the impact of individual factors and gender/sex on enzyme activities and voriconazole metabolism (Figure 2.6). Results showed that there was no statistically significant correlation between them. However, CYP3A4 activity was lower among males than females. Additionally, the metabolic ratio of CYP2C19 and FMO in males was slightly higher than in females, hinting at the potentially different activity levels of CYP2C19 and FMO in female versus male subjects. Though not proven statistically significant in this study, voriconazole metabolism might be influenced by sex among pediatric patients. Page 67 of 140 CYP3A4 MALE FEMALE 0.0 0.5 1.0 1.5 Sex Metabolic Ratio CYP2C19 MALE FEMALE 0.0 0.1 0.2 0.3 0.4 0.5 Sex Metabolic Ratio FMO3 MALE FEMALE 0.0 0.5 1.0 1.5 2.0 2.5 Sex Metabolic Ratio Voriconazole metabolic ratio MALE FEMALE 0 5 10 15 20 Sex Metabolic Ratio a b c d Figure 2.6 Relationship between sex and enzyme activities on voriconazole metabolic ratio. (a) CYP3A4/5. (b)CYP2C19. (c) FMO3. (d) Voriconazole. 2.4 Discussion Pediatric dosing relies on allometric scaling using weight, body surface area, or other size metric of adult doses [119]. Despite size-scaled dosing, the ability to accurately predict target concentrations continues to be a challenge among pediatric patients, often requiring higher than calculated doses. One possible reason is that children usually have a higher kidney clearance than adults. Another is that the expression of metabolic enzymes changes during individual growth and maturation in children, effects also seen in adults but to a much lesser extent. The inability to accurately predict therapeutic Page 68 of 140 dosages negatively affects treatment efficacy. Moreover, this variability presents increased risks for drug-related adverse effects. In this study, I generated some beneficial data using LC-MS assays. A major challenge in developing a drug dosing model for the pediatric population is the need for multiple sampling, often requiring milliliters of blood to perform these analytical. The multiplex assay developed for this study was able to measure the pharmacologic probe cocktail and their metabolites with accuracy, robustness, and precision using only 150 µL of plasma. The lowest level of quantitation was 50 pg/mL for all these targeted compounds. By using the multiplex LC-MS assay I developed, neonate and pediatric patients could be more easily phenotyped for their metabolic capacity. This analytical method could be adapted in other settings to enhance the capability of individualized dosing. Voriconazole is a potent antifungal drug but requires precise TDM, especially among the pediatric population due to its non-linear pharmacokinetic profile and narrow therapeutic window. Non-compartmental PK analysis showed lower voriconazole plasma Cmax and AUC for orally administered voriconazole when compared to patients receiving IV dosing. This difference may be due to the first-pass effect and corresponding diminished bioavailability in children [135]. The other PK parameters, such as plasma half-life and Tmax, were similar between the IV and PO groups. Additionally, PK parameters were equivalent for the metabolite voriconazole N-oxide, indicating saturated PK occurred. In these patients, three relevant DMEs were evaluated using AUC ratio of the metabolite to the parent drug. My findings support that metabolism of voriconazole was age- dependent, specifically that higher clearance was seen in younger pediatric patients when compared to adolescent patients. This correlates to the clinical findings that higher doses Page 69 of 140 are required for younger patients. The higher dose requirements correspond to the relative enzyme activities of FMO3 and CYP2C19 in younger patients which decline with advancement in age. These enzymatic systems are crucial in the metabolic elimination of voriconazole. My findings support that metabolic enzymes are impacted by age, where these enzyme activities are high after birth and decline as they approach puberty and into adolescent [16]. This explains that enzyme systems important for voriconazole clearance will decrease with age. Several studies have shown that FMO3 protein levels are age dependent from neonates to age 18. Yet in animals, FMO3 expression is influenced by hormone levels [136, 137]. Regardless, there is still a dearth of information on the expression level of FMO3 in humans, particularly during puberty, and increased expression does not always correlate to increased function. Currently, CYP2C19 is considered the major impact on voriconazole metabolism in adults [133]. In this study, I confirmed its essential influences on voriconazole metabolism among pediatric patients. Yet, CYP2C19 is not the only enzyme serving in this process. In both IV and PO administration groups, FMO3 also exhibited significant activity in voriconazole metabolism. Although a few in vitro studies have mentioned that FMO3 oxidizes voriconazole, little has been studied regarding the role of FMO3 on the saturated metabolism of voriconazole in vivo [138]. My findings revealed that FMO3 is another key enzyme for voriconazole metabolism in both oral and intravenous routes among the pediatric population. Especially in the IV group, FMO displayed a much stronger linear relation to the voriconazole metabolic ratio. Figure 2.7 exhibits the relationship between voriconazole metabolism versus age and FMO3 activity. By LOWESS analysis, R 2 was 0.6646, which indicates the potential vital correlation of age and FMO3 activity towards Page 70 of 140 voriconazole metabolism among pediatric patients. This finding may be due to avoidance of the first-pass effect in IV administration, providing a higher chance for voriconazole to react with FMO3. However, further studies need to be conducted to validate my hypothesis because this study had too a small sample size to confer statistical significance. Figure 2.7 Voriconazole metabolic ratio in relation to age and FMO3. Lowess fitting was used to smooth the data. According to my results, both CYP2C19 and FMO3 activities significantly correlated to voriconazole metabolism among the pediatric patients in both PO and IV routes. Furthermore, FMO3 may possess a more vital role than CYP2C19 in terms of voriconazole metabolism. This suggests that the reduced activity of FMO3 with age, in addition to CYP2C19 and CYP3A4/5 age-related effects, contributes to the saturated clearance of voriconazole in pediatric patients. The interplay of FMO3 with CYP2C19 and Page 71 of 140 CYP3A4/5 is an underappreciated aspect of voriconazole dosing in the pediatric population. Also, I found that voriconazole metabolism might be potentially sex-related, implied by FMO and CYP2C19 trends. Multiple studies have reported the impacts of sex on DMEs, specifically that the same drug therapy exhibited different pharmacological effects and PK/PD in males versus females [139, 140]. However, there is limited and even controversial information regarding pediatric sex-related DME activities. Hormone levels also regulate CYPs activity, including CYP3A4 and CYP2C19, which may explain the findings in this study [16]. Thus, the different metabolic activities of CYP3A4, CYP2C19, and FMO between pediatric males and females in this study, although not found to be statistically significant in this study, may still affect overall voriconazole metabolism in the pediatric population. 2.5 Conclusion Metabolism of voriconazole to voriconazole-N-oxide is higher in younger patients as a consequence of increased FMO3 and CYP2C19 activity. Although there is a trend for increased activity of CYP3A4/5, its impact is not as high FMO3 and CYP2C19 enzyme systems in voriconazole metabolism. A relationship was seen in patient sex and voriconazole metabolism; however, it needed to be verified in a larger cohort. The sex- related difference was evidenced by trends seen in the FMO3 and CYP2C19 activity. In conclusion, FMO3 and CYP2C19 significantly contribute to voriconazole metabolism in pediatric patients. Additionally, the tendency of patients to display saturated clearance of voriconazole is likely due in part to FMO3 activity. Page 72 of 140 Chapter 3. Drug disposition of pharmacoenhancer boosted protease inhibitors in CD4+ cells CEM and U937 Abstract Atazanavir (ATV) and darunavir (DRV) are two key HIV protease inhibitors (PIs) used to inhibit HIV proliferation. DRV has been shown to be the more potent PI using in vitro assays. Despite the in vitro data demonstrating DRV superiority in suppressing HIV replication, DRV administered in HIV-infected patients appears to be merely non-inferior to patients treated with ATV-containing regimens. Although ATV can be combined with two nucleosides for once a day dosing, DRV must be co-administered with a pharmacoenhancer such as ritonavir (RTV) or cobicistat (COBI). Both RTV and COBI are CYP3A4 inhibitors which increase the circulating level of concomitant antiretroviral agents. COBI was designed as a more selective CYP3A4 inhibitor with less antiviral activity compared to RTV. However, clinical studies suggest that when co-administrated with PEs, DRV has much higher virologic failure rates than ATV when they are combined with COBI- or RTV. Based on this fact, I hypothesized that RTV and COBI alter the disposition of concomitant ATV and DRV in the CD4+ cells differently. I established validated LC-MS assays to measure PI/PE cellular concentrations at multiple time points to examine the cellular retention of ATV and DRV, either treated alone or in combination with PEs. My results suggested that DRV cellular retention was significantly lower than corresponding ATV treatments. In addition, I found that COBI exhibited a poorer enhancement than RTV in CD4+ CEM and U937 cells. Moreover, PI combination with PE treatments had a higher cellular disposition in CEM cells compared to U937 cells. Page 73 of 140 These distinctive differences in CD4+ cellular cells may explain the level of clinical virologic failure seen with DRV combined with PE. 3.1 Introduction 3.1.1 AIDS and HIV infections Human immunodeficiency virus (HIV), the causative pathogen of acquired immunodeficiency syndrome (AIDS), is a retrovirus, a member of the lentivirus family, and expresses tropism for CD4 positive (CD4+) cells such as lymphocytes, monocytes and macrophages [141]. Despite its tropism, HIV penetration into CD4+ cells requires either CCR5 or CXCR4 to promote viral entry. Once viral entry has occurred, the virus unloads its genetic cargo, converting the viral RNA (vRNA) genetic template into viral complement DNA, a process catalyzed by viral reverse transcriptase. The antisense viral DNA (vDNA) is then synthesized to form double-stranded DNA. Once the vDNA has been assembled, it complexes with HIV integrase, a viral enzyme that promotes incorporation into the host genome, and is shuttled into the nucleus. vDNA integrates into the host cell DNA through DNA nicking and ligation events by viral integrase. Upon activation, proviral proteins are synthesized and activated upon viral protease liberation [142, 143]. 3.1.2 HAART Introduction of antiretroviral (ARV) combinations and Highly Active Antiretroviral Therapy (HAART) transformed how HIV infections were managed [142]. Although there is currently no cure for HIV, the use of HAART has transformed HIV infection from a short- lived fatal infection into a chronic condition, where patients can attain a similar quality of life as uninfected individuals. The number of individuals living with HIV has increased from Page 74 of 140 <1 million in 2000 to >17 million in 2015, which supports the notion that HAART can increase patient survival [144]. However, new HIV infections continue to grow with estimates of 5700 new infections daily, translating to approximately 2 million new cases annually. Patients receiving HAART no longer succumb as easily to the co-morbidities associated with HIV infection. To achieve this clinical outcome, strict adherence to the HAART is crucial. Unfortunately, high pill burden and adverse effects associated with early ARVs has led to poor patient compliance. A number of ART agents have been developed to prevent viral replication, including nucleoside reverse transcriptase inhibitors, non-nucleoside reverse transcriptase inhibitors, integrase inhibitors and PIs. HIV PIs were the first anti-HIV agents developed. The first of these agents was saquinavir and ritonavir, potent inhibitors of HIV protease. However, this new class of agents needed large doses to provide an adequate concentration for HIV suppression. Additional problems with PIs is their rapid metabolism by cytochrome P450 (CYP) enzymes and short plasma half-life. These limitations drove the development of atazanavir (ATV). ATV is a truly once-daily PI with a half-life of around 6.5 hours and can be used as part of a combination regimen [145]. Currently, the World Health Organization (WHO) classified ATV/RTV combination treatment as a second-line therapy. However, ATV is a highly protein bound drug with high interpatient and intrapatient variability. It is a substrate and inhibitor of the CYP3A4 enzyme and has a distinct resistance profile compared to other PIs [146]. Additionally, ATV inhibits P-gp and MRPs. Conversely, ATV has also been reported as an inducer of P-gp expression [147, 148]. Page 75 of 140 Darunavir (DRV) is a potent HIV PI with a wide spectrum of activity against PI-resistant HIV (EC50 ranges from <0.1 to 4.3 nM), and has a high genetic barrier against the emergence of viral resistance. Similar to ATV, DRV was approved as once-daily therapy, but only when co-administered with RTV to suppress HIV in both naïve and experienced patients. The pharmacokinetics (PK) of DRV alone did not support once a day dosing, thus the addition low dose of RTV was necessary to produce the PK parameters that are consistent for once a daily dosing [149-151]. Administration of PI ritonavir (RTV) was found to prolong drug levels of concomitantly administered PIs. The use of RTV substantially alleviated pill burden and dosing frequency while improving treatment efficacy. The PI level prolonging property was attributed to RTV’s ability to inhibit CYP3A4-mediated metabolism [152]. RTV was found to also increase the antiviral activity of co-administered nucleosides. The combination of RTV with nucleosides and PI further reduced the emergence of drug resistance mutations in patients [153]. This spawned the theory that RTV not only blocks metabolism but also alters the intracellular disposition of both nucleosides and concomitantly administered PI. Despite the importance of cellular disposition, ARV plasma concentration continues to be used to determine whether adequate ARV levels have been achieved to suppress the HIV. Little is known what effect PEs like RTV will have on drug accumulation and elimination in the targeted cells. More recently, RTV was found to also be an inhibitor of efflux transporters like P-gp and MRPs. RTV’s ability to maintain PI levels made it an indispensable part of HAART, “boosting” concomitantly administered PI. However, RTV has some disadvantages, including unfavorable side effects, poor solubility, co- Page 76 of 140 formulation issues and non-selective or unintended inhibition of CYP3A leading to drug- drug interactions [154]. COBI is an analog of RTV but with no apparent antiviral activity at the current dosage. It was developed to be a more selective CYP3A inhibitor with less off-target drug effects. COBI is also more soluble to be easier to co-formulate with other ARVs. COBI has been approved to boost ATV and DRV as Evotaz TM (ATV/COBI) and Prezcobix TM (DRV/COBI). Although RTV has been shown to inhibit P-gp, BCRP, OATP1B1, OATP1B3 and MATE1, less is known about the ability of COBI to influence the activity of these transporters, especially when used in combination therapies [155]. There is data suggesting that both RTV and COBI can inhibit creatinine clearance (Crcl) and reduce estimated glomerular filtration (eGFR), a major drawback when combining PIs with RTV and renal toxic agents like tenofovir.[155-157] 3.1.3 CD4+ Cell lines HIV infects CD4+ cells, which constitute a major part of peripheral blood mononuclear cells (PBMCs). CD4+ cells including lymphocytes, monocytes, and macrophages are susceptible to HIV infection. Since PBMCs are not immortal cell lines, a number of CD4+ cancer cell lines have been widely used as substitutes for HIV studies. In this project, I used CCRF-CEM and U937 as cell models. CEM and U937 both express CD4 antigen [141]. CEM is derived from peripheral blood tissue of a 4-year juvenile with acute lymphocytic leukemia. U937 monocytes were isolated from the pleural effusion of a 37- year male patient with histiocytic lymphoma. These two different CD4+ cell lines will allow us to evaluate the impacts of cell types on the responses to PI/PE therapies. Page 77 of 140 3.1.4 Clinical findings WEEK 48 ATV+COBI ATV+RTV DRV+COBI DRV+RTV Virologic Response 85% 87% 81% 83% Virologic Failure 5.8% 4% 11% 7% HIV-1 ≥50 copies/mL 1.7% 2% 4% 5% Table 3.1 Summary of clinical outcome using various PE [18-20]. COBI was compared with RTV in several clinical studies. In these trials, COBI was found to be non-inferior to RTV when combined with ATV or DRV. However, a closer analysis demonstrated that DRV with either RTV or COBI had a different level of viral suppression. Virologic failures in patients receiving boosted ATV or DRV are summarized in Table 3.1 [158-160]. From the data, virologic failure for HIV patients receiving boosted DRV were ~2-fold higher than boosted ATV though DRV was found to more potent than ATV in vitro [161, 162]. Additionally, virologic failure rates of COBI-mediated regimens were 1.45 and 1.57-fold higher than RTV containing regiments, but was lower when comparing patients with virologic relapse (HIV-1 RNA ≥50 copies/mL). These differences suggest that COBI affects intracellular sequestration differently than RTV, and the mechanism(s) need to be dissected to fully optimize the activity of these PEs. In this project, I hypothesized that PEs can alter the cellular disposition of ATV and DRV. Specifically, COBI can alter expression and function of cellular efflux transporters and CYP3A4. I evaluated cellular disposition of ATV and DRV by determining the cellular concentrations and elimination constant (Ke) of the PI when given alone and in combination with COBI or RTV in CD4+ (e.g. monocytic and lymphocytic) cell lines. Page 78 of 140 3.2 Materials and methods 3.2.1 Materials and chemicals CCRF-CEM cell line and U937 cell line were purchased from ATCC. RPMI 1640 cell culture media were purchased from Corning (NY, USA). Cobicistat, ritonavir, atazanavir, and darunavir were purchased from Clearsynth. Other chemicals and lab supplies such as culture flasks and plates were purchased from VWR (PA, USA) and Sigma Aldrich (MO, USA). 3.2.2 Determination of cellular concentration of PIs and PEs A qualified LC-MS method was developed for the quantitation of cellular concentrations of PIs and PEs. The LC-MS system consisted of a Shimadzu UPLC (Kyoto, Japan) linked to an AB Sciex (CA, USA) API3000 triple quadrupole mass spectrometer with ESI using positive mode. The analytes were detected by multiple reaction monitoring (MRM) specific to each, which was 705.6→335.2, 548.5→391.8, 721.7 →296.3, 776.5 → 606.1, and 629.6→183.3 for ATV, DRV, RTV, COBI and LPV (internal standard), respectively. Analytes were separated using Kinetex 5µ EVO C18 column with the dimensions of 100Å 50 X 3mm (Part #: 00B-4633-Y0) (Phenomenex, USA). Mobile phases used contained water with 0.5% formic acid (component A) and methanol with 0.5% formic acid (component B). A gradient LC program was adopted, where component B started from 20% at the beginning to 90% within 4 minutes and held for another 1 min before decreased to 20%. The total run time was 6 minutes. Page 79 of 140 3.2.3 Study design The general study design is summarized in Table 3.2. Cells are treated with ATV or DRV alone or in combination with RTV or COBI. Treatment concentration of ATV or DRV is 500, 4000 or 8000 ng/mL. Treatment concentration of RTV or COBI is 100, 500, 1000 or 10000 ng/mL. Concentrations are selected based on clinical pharmacokinetics data of maximum and minimum concentrations. Table 3.2 Cellular disposition study design. 3.2.4 Cell culture CEM and U937 are suspension cell lines. RPMI 1640 with 10% FBS, 1% non-essential amino acid and 1% antibiotic-antimycotic was used as the culture media. Cells were cultured in the incubator at 37 o C in a humid environment with 5% CO2. 3.2.5 Drug treatment Approximately 1 million cells in 1.8 mL of media were seeded into a 6-well plate and stabilized overnight. 200 μL of appropriate concentrations of PI/PE was added to each well and incubated for 24 hours. Cells were collected from each well and centrifuged at 1000 rpm for 10 min. Media was discarded and 1 mL of PBS was used to wash the cell pellet. After centrifuging the pellet again, PBS was removed and cells were re-suspended Protease Inhibitors PharmacoEnhancers Cells Time of Exposure (hrs) Washout Time (hrs) Type of Analysis ATV None CEM U937 0, 3, 24, 48 & 96 0, 2, 4, 8, 24 & 48 Cellular Concentration, Ke, CC-AUC RTV COBI DRV None CEM U937 0, 3, 24, 48 & 96 0, 2, 4, 8, 24 & 48 Cellular Concentration, Ke, CC-AUC RTV COBI Page 80 of 140 with 2 mL fresh media for incubation. At designated washout time points, cell samples were collected. 3.2.6 Sample preparation and analysis After cellular treatment and washout period, the media was aspirated and cells washed with 500 μL of PBS. To pellet cells, 50 μL of IS (Lopinavir 500 ng/mL) was added and mixed thoroughly. The entire mix was then protein precipitated using 500 μL of ice-cold 80% methanol (methanol: water, v: v, 8:2). The mixture was mixed thoroughly and incubated in -20 o C for 30 min. Samples were centrifuged at 13,000 rpm for 5 min. 60 μL of the resulting supernatant was transferred to an HPLC vial, where 55 μL was injected to the LC-MC for quantitation. The protein amount of each sample was measured by Nanodrop (Thermo Scientific, USA). The final cellular concentrations of the analytes were normalized by the protein amount. 3.2.7 Statistical analysis and pharmacokinetics Analyst software was used for quantitation of LC-MS data (Sciex, CA, USA). Prism GraphPad (GraphPad, CA, USA) was used to generate the statistical analysis. Non- compartmental PK analysis was conducted using PK Functions add-on to Microsoft Excel (Allergan, CA, USA). 3.3 Results 3.3.1 LC-MS assay The LC-MS assay was validated for accuracy, precision, and robustness. Qualification tests performed intraday and interday with three groups of standard curves and quality controls. The accuracy and precision were acceptable within a range of 100 ± 15% along Page 81 of 140 with a CV% of no more than 15%. Linear regression (R 2 >0.99) with 1/x 2 weighing factor was adopted and used for assessing method robustness. The lower limit of quantitation was 1 ng/mL for all the analytes. Dynamic range for all the analytes was from 1 to 2500 ng/mL for both cell lines. Quality controls were at concentrations of 10, 100 and 500 ng/mL. Results of interday QC tests for both CEM and U937 were summarized in Table 3.3. Robustness of the interday tests was shown in Figure 3.1. Chromatogram of this method was shown in Figure 3.2. (a). CEM (n=3) QC (ng/mL) Mean SD Recovery (%) CV (%) ATV 10 9.6 0.3 -4.2 3.5 100 104 5.6 3.5 5.5 500 491 36.1 -1.8 7.4 DRV 10 9.3 0.9 -7.1 10.1 100 103 3.5 3.3 3.4 500 478 17.6 -4.4 3.7 COBI 10 10.4 0.3 4.0 2.5 100 102 4.9 2.2 4.8 500 483 15.4 -3.5 3.2 RTV 10 10.0 1.1 0.4 10.8 100 105 4.0 5.0 3.8 500 465 18.2 -7.0 3.9 (b). U937 (n=3) QC (ng/mL) Mean SD Recovery (%) CV (%) ATV 10 9.9 0.4 -1.2 3.7 100 93 1.1 -6.7 1.2 500 513 10.4 2.7 2.0 DRV 10 9.9 1.2 -0.8 12.0 100 96 8.7 -4.0 9.1 500 538 12.6 7.7 2.3 COBI 10 9.7 0.8 -3.1 8.2 100 95 9.4 -5.4 9.9 500 527 15.3 5.3 2.9 RTV 10 10.3 0.1 3.3 0.6 100 91 5.4 -8.6 5.9 500 537 14.4 7.3 2.7 Table 3.3 Interday QC analysis of (a) CEM and (b) U937 standard curves. Page 82 of 140 CEM ATV 0 1000 2000 3000 0 2 4 6 8 10 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area CEM DRV 0 1000 2000 3000 0 5 10 15 20 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area CEM COBI 0 1000 2000 3000 0 10 20 30 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area CEM RTV 0 1000 2000 3000 0 2 4 6 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area U937 ATV 0 1000 2000 3000 0 2 4 6 8 10 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area U937 DRV 0 1000 2000 3000 0 5 10 15 20 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area U937 COBI 0 1000 2000 3000 0 10 20 30 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area U937 RTV 0 1000 2000 3000 0 2 4 6 Set 1 Set 2 Set 3 Concentration (ng/mL) Analyte area/IS area Figure 3.1 Standard curves of interday tests in CEM and U937. Page 83 of 140 Figure 3.2 Chromatographs of various analytes that include COBI, RTV, ATV, DRV, and LPV (as internal standard) (from Top to bottom). 3.3.2 Cellular disposition of PI with and without pharmacoenhancers CD4+ cells U937 (monocytes) and CEM (lymphocytes) were utilized in this project. Treatment incubation time was evaluated at 0, 3, 24, 48 and 96 hours. Both PIs and PEs were tested in a dose-escalation manner where concentrations were selected based on plasma Cmin to Cmax in human subjects. An extensive concentration of PEs of 10 µg/mL was also included. After treatment, cells were washed out and incubated in new media for different time points before sample collection. Generally, cellular concentrations of COBI ATV Page 84 of 140 ATV and DRV reached maximal concentrations after 24-hour treatments in both U937 and CEM. Figure 3.3 summarizes the results of cellular concentrations after 24-hour PI treatment using concentrations equivalent to the maximum plasma concentration at steady state, where the targeted concentration was 4000 ng/mL for both ATV and DRV [163, 164]. I hypothesized that cellular levels of ATV and DRV were different between the two types of CD4+ cells, and the addition of PE would further impact the cellular concentrations of the PIs. When treated with PI alone, the cellular concentration of ATV and DRV was similar in both cell lines. However, CC of DRV was lower than ATV when co-administrated with PEs in most situations. The addition of COBI or RTV increased the cellular concentration of ATV in CEM and U937 (p-value<0.05). However, when the concentration of COBI or RTV reached 1000 ng/mL, the cellular concentration of ATV in CEM began to decline (Figure 3.3a). Additional escalation to 10,000 ng/mL of COBI further reduced cellular ATV in both CEM and U937 cells. In contrast, the addition of RTV did not induce a reduction in ATV cellular concentration until 10,000 ng/mL was used. Unlike ATV, the addition of COBI or RTV with DRV in U937 did not significantly affect the cellular concentration of DRV. In U937, the additions of COBI or RTV over a dosage range was similar to DRV treatment alone (Figure 3.3b). In CEM cells, a moderate increase in cellular DRV was seen with escalating concentration of RTV, where at 10,000 ng/mL of dramatically reduced cellular concentrations of DRV to levels below DRV alone treatment. Page 85 of 140 (a) (b) Figure 3.3 Effect of COBI and RTV on ATV and DRV cellular disposition in CD4+ cells. (a) ATV and (b) DRV. *p-value<0.05, **p-value<0.01. Page 86 of 140 3.3.3 Elimination rate constant In this study, cells treated with PI/PE combination had their intracellular concentration of DRV and ATV examined over various time points. Elimination rate constant (Ke) of each treatment was determined 0 to 48 hours after washout of both PIs and PE. The cellular concentrations were calculated using non-compartmental analysis. Ke is a reflection of PIs clearance from the CD4+ cells after treatments. The median concentration of each time point was selected for calculation. Results are summarized in Table 3.4. (a). ATV K e (h -1 ) ATV ATV+Cobicistat (ng/mL) ATV+Ritonavir (ng/mL) 100 500 1000 100 500 1000 U937 0.0484 0.0631 0.0570 0.0520 0.0350 0.0617 0.0644 Ratio 1.00 1.30 1.18 1.07 0.72 1.27 1.33 CEM 0.0466 0.0494 0.0693 0.0251 0.0481 0.0362 0.0502 Ratio 1.00 1.06 1.49 0.54 1.03 0.78 1.08 (b). DRV K e (h -1 ) DRV DRV+Cobicistat (ng/mL) DRV+Ritonavir (ng/mL) 100 500 1000 100 500 1000 U937 0.0489 0.0408 0.0488 0.0535 0.0391 0.0509 0.0449 Ratio 1.00 0.83 1.00 1.09 0.80 1.04 0.92 CEM 0.0687 0.0561 0.0411 0.0628 0.0693 0.0601 0.0513 Ratio 1.00 0.82 0.60 0.91 1.01 0.87 0.75 Table 3.4 Elimination rate constant (Ke) of (a) ATV and (b) DRV when treated alone or in combination with PEs in U937 and CEM. Non-compartmental pharmacokinetic analysis was used for calculation. When cells were treated with PI alone, cellular Ke of ATV was similar for both CEM and U937. In this study, the calculated Ke for DRV was slightly higher in CEM than in U937. For ATV containing treatments, the Ke was higher when co-administered with a PE than Page 87 of 140 ATV alone. This trend was seen in both cell types. COBI followed a concentration dependent effect on Ke. For DRV containing treatments, Ke was lower when co-administered with a PE than DRV treatment alone. However, when at the highest concentration (1000 ng/mL) of COBI tested, DRV Ke was similar to DRV alone for both CEM and U937. In contrast, the highest RTV concentration reduces DRV Ke. When ATV was used as the substrate, COBI and RTV had a different cellular disposition. Similar to the ATV group, the Ke ratio of PE containing treatments to DRV alone was higher in U937 than in CEM. DRV in combination with COBI showed higher Ke value in most cases compared with DRV+RTV treatments in both cell lines. 3.3.4 Cellular PI AUC after washout In addition to Ke, cellular AUC during washout period was also calculated through non- compartmental pharmacokinetics analysis to determine the cellular exposure of PI treatments. Results are summarized in Table 3.5. Page 88 of 140 (a). (b). Table 3.5 Cellular PI AUC (based on median concentration at each time point, n=3) during washout period after 24hr treatment in CEM and U937. (a) ATV and (b) DRV cellular AUC was calculated through non-compartmental analysis. Cellular exposure of PIs may reflect anti-HIV efficacy. When treated with PI alone, both ATV AUC and DRV AUC were almost two-fold higher in CEM (lymphocytes) than in U937 (monocytes). When cells were treated with ATV plus PEs, a high concentration of PE effectively enhanced PI cellular ATV AUC. Comparing RTV with COBI, COBI was associated with a lower ATV AUC than RTV in all concentrations. For DRV treatment, 1000 µg/mL of either COBI or RTV led to increases in DRV AUC. In CEM, DRV combined with COBI had lower DRV AUC compared to DRV alone. When treated with DRV+RTV, 1000 ng/mL of RTV resulted in a lower AUC than 100 ng/mL or 500 ng/mL of RTV in CEM, hinting at induction of transporters. Similar to the ATV group, RTV containing treatments had higher cellular DRV AUC than COBI containing treatments. In addition, ATV AUC was significantly higher than corresponding DRV AUC (p<0.05). Page 89 of 140 3.4 Discussion HIV suppression is dependent on the effective cellular retention of antiretroviral agents. PEs, RTV and COBI are used as boosters to enhance co-administrated HIV PIs. The clinical data of ATV and DRV in combination with RTV or COBI have revealed differences in virologic failure rates between the two combinations. These findings led me to hypothesize that the cellular disposition of ATV and DRV when combined with either COBI or RTV may not increase the substrate PI’s intracellular concentration. Cellular concentration, Ke, and AUC were examined after PI alone or in combination with either COBI or RTV. . In order to measure drug cellular concentrations, a validated LC-MS/MS method was established with an LLOQ of 0.5 ng/mL for simultaneous detection of each of the analytes: ATV, DRV, COBI, and RTV. This method was adapted to the cell lines utilized in this study. PI cellular concentrations at each time point were determined by this LCMS method followed by PK analysis. In all these experiments, differences were found between CEM lymphocytic cells and U937 monocytic cells. Based on non-compartmental PK analysis, lower Ke and higher cellular AUC of PIs in CEM drive higher PI efficacy in these cells. Accordingly, CD4+ cell types played a role on PI suppression of HIV infection. Cellular concentrations of PI in U937 were significantly different when compared to lymphoblastic cell line CEM. One explanation is the larger cellular volume of U937 compared to CEM. Another potential reason is that expression and function of membrane transporters are different between these two cell lines. Membrane transporters adjust the cellular accumulation and elimination of PIs, which determines the cellular therapeutic drug level. Drug metabolizing Page 90 of 140 enzymes CYP3A4 may also vary in these two cell lines, contributing to their drug disposition. Although DRV demonstrated more potent anti-HIV activity in vitro, the clinical dosage of DRV is much higher than ATV because of its low bioavailability. PEs were found to increase DRV bioavailability and prolong its plasma half-life efficiently. However the intracellular concentration of CD4+ lymphocytes and monocytes has not been previously evaluated. The virologic failure rate of DRV containing treatments was almost twice that of ATV containing treatment in clinical trials. In this study, although the cellular concentration of DRV was slightly higher than ATV when administered alone, ATV intracellular concentration was significantly higher when combined with either COBI or RTV. This was particularly seen in U937. The cellular ATV AUC was significantly higher than DRV AUC when these values were converted into molar scale. Lower PI AUC was linked to COBI co-administration rather than RTV, demonstrating lower inhibitory and/or higher induction ability of COBI on cellular efflux of PIs. Figure 3.4 ATV (left) and DRV (right) Ke in relation to AUC. Nonlinear fitting analysis was used to evaluate the trend and R 2 is shown in the figure. ATV Ke vs AUC 25 50 75 100 125 150 0.00 0.02 0.04 0.06 0.08 R 2 =0.04 AUC 0-48 (nmol/mg*h) Ke(h -1 ) DRV Ke vs AUC 20 40 60 80 0.03 0.04 0.05 0.06 0.07 0.08 R 2 =0.26 AUC 0-48 (nmol/mg*h) Ke(h -1 ) Page 91 of 140 Interestingly, the relationship between Ke and AUC differs for ATV and DRV (Figure 3.4). The ATV Ke decline correlated with an increase in ATV intracellular AUC. In contrast, DRV Ke increased with the increase in intracellular ATV AUC. These findings suggest that both COBI and RTV alter cellular distribution of PIs like ATV and DRV. In the next chapter, I will determine the impact of RTV and COBI on expression and function of cellular transporters. COBI is an analog of RTV and a second generation pharmacoenhancer. The co- administration of either ATV or DRV COBI was associated with a higher virologic failure rate when compared to RTV-containing regimens. In this project, I found that in CEM lymphocytes RTV was a more potent inhibitor than COBI for its potential to increase intracellular PI concentrations. Regardless of cell types used, COBI was less potent than RTV. These findings correlate with clinical virologic failure data. 3.5 Conclusion In this chapter, I found that COBI and RTV can enhance PI accumulation, with effects stronger in lymphocytes than monocytes. When this concentration-dependent effect reached a threshold, the level of intracellular PI can induce nuclear transcription factors mediated activities, which can reduce the cellular concentration of PI. In this study, I found cellular ATV can reach a higher cellular concentration in both U937 and CEM when compared to DRV. I also found that RTV was a more potent inhibitor, leading to increasing cellular concentration of the concomitant PI. These factors contribute to the higher clinical virologic failure rate in DRV as well as COBI containing therapies. Page 92 of 140 Chapter 4. Relationship of pharmacoenhancers (PEs) boosted atazanavir (ATV) and darunavir (DRV) treatments with efflux transporters Abstract HIV can infect CD4+ cells, but the expression of the CD4+ marker is found in both lymphocytic and monocytic lineages. I evaluated the impact of COBI and RTV on efflux transporter in these two cell types. Cellular retention of ATV and DRV was found to be different between these two lineages when comparing the cellular concentration of PIs (CC PIs). Efflux transporters are key factors that regulate elimination of PI. I hypothesized that COBI and RTV inhibit efflux transporters where these differences correlated with PI disposition. I used specific transporter inhibitors in HEK and MDCK variants that overexpress various efflux transporters to evaluate the PE’s impacts on the cellular disposition of ATV and DRV. Additional PE impact on gene and protein expression of major efflux transporters were reported. The transporter expression levels were correlated with transporter functions and activities. Results suggested that expression of P-gp, MRP1, and BCRP were important factors determining CC PIs. COBI has weaker inhibitory activity compared to RTV, which was attributed with COBI’s higher ability to induce the expression of efflux transporters. Therefore, these data could provide mechanistic insights to explain the higher failure rate of COBI-containing treatments observed in clinical usage. Page 93 of 140 4.1 Introduction 4.1.1 Role of Efflux transporters on antiretroviral therapies Intracellular transport of nutrients and xenobiotics are important processes that are highly regulated by influx and efflux transporters. Once in the cells, nutrients and xenobiotics are metabolized to form active metabolites required for cellular function or catabolized into non-active forms for elimination. Antiretroviral (ARV) agents are xenobiotics that mimic important components for cellular survival, thus their intracellular levels of these components are controlled by the same mechanisms. The cellular level of drug is dependent on the transport of the drug across the cellular membrane and the ability of the cellular adaptive mechanisms to metabolize and clear the presence of the xenobiotic. As stated earlier (Chapter 1), this process can be through passive diffusion or active transport mechanisms. Once in the cells and the level have reached a certain threshold, xenobiotics can activate a set of biosensors, which also function as transcriptional factors. Once translocated into the nucleus, these transcription factors either alone or in complex with other co-transcriptional factors trigger the expression of transporters (e.g. influx and efflux transporters) and metabolic enzymes (e.g. CYP450 enzymes). The expression of these cellular proteins regulates the drug absorption and metabolism of xenobiotics, in the cells and the body as a whole [165, 166]. As previously discussed, influx and efflux transporters are either ATP-binding cassette (ABC) transporters or solute carriers (SLC). Efflux transporters include P-gp, BCRP, and MRP1-4. Solute carrier influx transporters include organ anionic transporter proteins (OATPs), organ anionic transporter (OATs) and organ cationic transporters (OCTs). Both influx and efflux transporters are important for cellular uptake of nutrients and xenobiotics. Page 94 of 140 ARVs and protease inhibitors in particular, have been described to modulate cellular transporter expression. Their pharmacologic properties are summarized in Table 4.1 [167-170]. Ritonavir and cobicistat are protease inhibitors analogs that have been subclassified as pharmacoenhancers (PE), due to their potent CYP inhibitory properties that prolong systemic half-life of PIs. However, the entire spectrum of pharmacologic activities of both of RTV and COBI has not been fully characterized. Transporters Inhibitor Inducer Substrate P-gp ATV, DRV, RTV, COBI ATV, RTV, COBI ATV, DRV, RTV MRP1 ATV, RTV, COBI RTV ATV, RTV MRP2 RTV ATV, RTV BCRP ATV, RTV, COBI RTV MATE1 RTV, COBI OATPs ATV, DRV, RTV, COBI DRV OCTs RTV,COBI Table 4. 1 Impact of various PI and PEs on cellular transporters Currently, there is a lack of available preclinical evidence to identify all the consequences of combining various ARV together. Although most of the ARV combination regimens have led to improved clinical outcomes, several combinations not only did not improve antiviral activity, but promoted pharmacologic antagonism. Such was the case when tenofovir difumerate (TDF) was combined with abacavir (ABC) or didanosine (ddI) [171]. These unfortunate outcomes highlight the need to understand the molecular impact of combining various ARVs in a therapeutic regimen. As mentioned in the previous chapter, the discovery of the first PE, ritonavir (RTV), was fortuitous when it was combined with saquinavir [152]. It was found that RTV can prolong the systemic or plasma half-life of saquinavir. This expanded the use of RTV as a “booster” for other HIV PI. As expected, some PIs benefited from the combination with Page 95 of 140 RTV. However, not all PIs benefitted. Nelfinavir and indinavir showed little to no improved effect. The PI-prolonging effects of RTV were through to be a consequence of inhibition of CYP3A alone [172]. However, additional pharmacologic properties were identified. These pharmacologic activities spurred the pharmaceutical industry to develop the second generation of PEs. Candidate PEs were screened for their ability to maximize prolongation of systemic half-life while reducing adverse effects. From this process, cobicistat (COBI) was identified and clinically evaluated leading to FDA approval in 2014. Unlike RTV, at the concentration used for pharmacokinetic enhancement, COBI has minimal to no antiviral activity and thus classified as a better PE [168]. Both RTV and COBI are pharmacokinetic enhancers due to their potent inhibition of CYP3A isoforms. What is less appreciated is their ability to modulate efflux transporter expression and functions. As summarized in Table 4.1, RTV is a potent inhibitor of efflux transporters P-gp, MRP1-2, and MRP4, which can potentially alter cellular PI disposition [173]. Similarly, the RTV analog COBI also inhibits the function of efflux transporters, though COBI has been purported to be a more selective inhibitor of CYP3A isoenzyme when compared to RTV [174]. Efflux transporters are also responsible for the distribution and elimination of the ARVs. According to the biopharmaceutics drug disposition classification system, PIs belong to Class 2, where efflux transporters are predominant factors determining drug disposition [175]. Efflux transporter activity can alter the need for CYP450 mediated metabolism, and is often referred to as transporter-enzyme interplay. For instance, P-gp inhibition will Page 96 of 140 reduce cellular levels of the P-gp and CYP3A4 substrates but also reduce the metabolism of the substrates, tying level of efflux transporters to CYP3A4 activity [176]. P-gp and MRP1 are important efflux transporters regulating the disposition of PI found in the cells. The expression of P-gp and MRP-1 are determined by intracellular biosensors such as pregnane xenobiotic receptor (PXR) which also regulate the expression of CYP3A and efflux transporters. Currently, there is no data on ARVs highlighting the differences among cells expressing CD4+, such as monocytic and lymphocytic cells. I hypothesized that there are significant differences in PIs disposition in CD4+ (e.g. monocytic and lymphocytic) when co-administered with a PE. Additionally, I hypothesized that intracellular difference is dependent on the expression of specific transporters. To fully understand the impact of PEs, changes in gene and protein expression were correlated with transporter function. This study provides insights as to the mechanism(s) that may impact cellular disposition of PIs. 4.2 Materials and methods 4.2.1 Chemicals G418 and MK571 were purchased from Santa Cruz Biotechnology. Colchicine, rhodamine 123, KO143, tariquidar, and furosemide were purchased from Sigma Aldrich. Antibodies of β-actin, MDR/P-gp, MRP1, and BCRP were purchased from Cell Signaling Technology. Antibodies of MRP2, MRP4, and OATP1B1 were purchased from Santa Cruz Biotechnology. Primers were purchased from Invitrogen. Page 97 of 140 4.2.2 Study Design Cells Inhibitors Tests Transfected lines expressing efflux transporters* PI/PE Cellular disposition of ATV or DRV CEM U937 P-gp & MRP inhibitors** Cellular disposition of ATV or DRV CEM U937 PI/PE RNA and protein expression of transporters; Functional data using flow cytometry. * Cell lines: MDCK-MDR, -MRP2, HEK-MRP4 and -5 **Inhibitors: MK571, tariquidar, KO143, and furosemide Table 4.2 Overall Study Design to determine the effects of transporter expression and function of efflux transporters. To overall study design is determine whether the co-administration of either COBI or RTV can alter the transporter expression and function alone and in combination with DRV or ATV. After incubating the cells under these conditions, the expression and function of targeted transporters were evaluated and summarized in Table 4.2. In this study design, DRV alone and ATV alone served as reference controls. Cellular concentrations were determined after the addition of specific efflux inhibitors (e.g. MK571, tariquidar, KO143, and furosemide) and compared with RTV or COBI. In addition, using inhibitors of specific efflux transporters, I can dissect the impact of PEs expression and function of transporters over a range of concentrations. Each inhibitor can selectively inhibit specific transporters at appropriate concentrations. Some examples include: MK571 (30 μM) for MRP1/MRP4, furosemide (50 μM) for MRP2, tariquidar (1 μM) for P-gp, and KO143 (20 μM) for BCRP, where selected concentrations have been previously reported to inhibit these specific transporters [177, 178]. Cells were treated with regimens shown in Table 4.2. The ability to efflux ARVs may be one mechanism for how viral failure may occur, I determined the cellular elimination rate comparing COBI, RTV, and specific efflux transporter inhibitors. Page 98 of 140 After treatment for 24 hours, total RNA and protein were harvested and the mRNA and protein expression were determined using qRT-PCR and western blot analysis, respectively. Although expression data provided information on the activity of the cellular response, flow cytometric assays will be necessary to further support the function of the transporters. I used Rhodamine 123 (P-gp), DCFDA (MRP2-4) and Calcein AM to determine the ability of specific transporters to efflux these substrates [148, 179]. 4.2.3 Transfected cell lines’ culture conditions I also evaluated the intracellular concentration of ATV and DRV in MDCK II and HEK293 wild type and stably transfected variants overexpressing specific transporters. The transporter expression was affirmed using Western blot and/or GFP fluorescence. Culture media for both wild type and transfected cell lines was high glucose DMEM supplemented with 10% FBS, 1% antibiotic-antimycotic and 1% non-essential amino acids. For MDCK- MDR and MDCK-MRP2, to enrich for cells expressing these transporters, I used selection media containing either 80 ng/mL colchicine or 0.4 mg/mL G418 (Geneticin), respectively. Cells were passaged until approximately 75% confluence, at which time cells were detached from culture dishes, medium aspirated and plates washed twice with calcium- free PBS. After aspiration of remaining PBS, cells were detached using 5 mL of 10X trypsin incubated for 5 minutes. The trypsin was neutralized using culture medium and the detached cells were collected and centrifuged at 1500 rpm at 4 o C for 5 minutes. The supernatant was removed after centrifugation, the pellet was then resuspended in 9 mL of culture medium, and split into three different flasks to expand. The medium was changed every two to three days until confluence. Page 99 of 140 4.2.4 qRT-PCR Cells were collected after treatments and RNA extracted using Direct-zol miniprep plus kit (Zymo Research, USA) according to manufacturer instructions. RNA concentrations were determined using Nanodrop One (Thermo Scientific, USA). Sample concentration was adjusted to approximately 200 ng/mL. To convert the RNA template into cDNA, iScript cDNA Synthesis Kit (Bio-Rad, USA) used 1 μg of RNA using manufacturers recommended procedures to generate the cDNA templates. The impact of treatment on gene expression was determined using qRT-PCR for each transporter gene with 10 μM of primers (Invitrogen, USA) (Table 4.3) .SYBR Green (Thermo Scientific, USA) was used for detection, where this mixture underwent 40 cycles of amplification. Primers were selected and examined through NCBI BLAST. Detailed primers information is listed in Table 4.3: Table 4.3 Efflux Transporter primers 4.2.5 Western blot Western blot analyses to assess the transporter protein expression used the following procedures, 5X10 6 cells were seeded onto a 10 cm culture dish and incubated in the 37 o C for 24 hours. After drug treatments, the cells were harvested and the cellular pellet was lysed using M-PER buffer (Thermo Scientific, USA) according to the manufacturer's Gene Primer ABCC1 Forward 5’-ACTTCGTTCTCAGGCACATC-3’ Reverse 5’-TGATCCGAAATAAGCCCAGG-3’ ABCC2 Forward 5’-TCATCGTCATTCCTCTTGGC-3’ Reverse 5’-ACGGATAACTGGCAAACCTG-3’ ABCB1 Forward 5’-CTTCAGGGTTTCACATTTGGC-3’ Reverse 5’-GGTAGTCAATGCTCCAGTGG-3’ ABCG2 Forward 5’- CAGGGTCATTCAAGAGTTAGGTC-3’ Reverse 5’- AGAACAAGATGGAAGGATCAGTG-3’ Page 100 of 140 instructions. Protein concentrations were measured by Bradford assay (Bio-Rad Inc., USA). 50 μg of protein of each sample was loaded onto a 4-20% gradient gel (Bio-Rad Inc., USA) and run at 100 volts for 1.25 hours. Subsequently, proteins were transferred onto PVDF membrane at 70 volts for 2 hrs. Antibodies of transporter and housekeeping proteins were purchased from Cell Signaling Technology (Massachusetts, USA), and a dilution of 1:500 was used for staining. Clarity ECL kit (Bio-Rad Inc., USA) was used for chemiluminescent imaging detection. 4.2.6 Flow cytometry Functional assays used dyes such as Calcein AM or Rhodamine 123 to assess specific transporter activities such as MRP1 and P-gp, respectively. To this end, 1X10 6 cells were pretreated with 200 ng/mL Rhodamine 123 or 100 nM calcein am in RPMI media for 30 min at 37 o C. After treatments, cells were washed twice with ice-cold PBS and resuspended in dye-free medium and allow dye efflux for 24 hr, with or without PI/PE treatments which were washed in PBS with 10% FBS after incubation. The intracellular fluorescence was determined by Flow Cytometer (Applied Biosystems, USA) equipped with an air-cooled 15 mW argon laser emitting at a fixed wavelength of 488 nm. The fluorescent filters and detectors used were all standard with green fluorescence collected in the FL1 channel (530±30 nm). Each sample was analyzed using 15000 events. The cells were analyzed using a logarithmic amplifier to determine the percentage of stained cells and mean fluorescence intensity. Page 101 of 140 4.2.7 Data analysis Image J (NIH, USA) was used to quantify western blot signal intensity. GraphPad Prism (La Jolla, CA, USA) was used to generate figures and conduct statistical analysis. FlowJo was used to analyze flow cytometry data. 4.3 Results 4.3.1 Transporter transfected HEK and MDCK cell variants (a) (b) Figure 4.1 ATV versus DRV cellular concentration in HEK (a) and MDCK (b) transporter overexpressed variants after 24-hour treatment. HEK 463 are HEK293T (HEK293 WT) overexpressing MRP4, while HEK5I overexpresses MRP5. MDCK MDR cells are MDCK II (MDCK WT) cells that overexpress P-gp while MDCK MRP2 overexpresses MRP2. **p- value<0.01. Wild type and variants expressing a high level of transporters were used to characterize the types of transporters which ATV and DRV used for cellular elimination. In Figure 4.1a, ATV and DRV were incubated in HEK wild type, MRP-4 and MRP-5 overexpressed transporters. In Figure 4.1b, were PI treatment in MDCK II MDR and MRP-2 overexpressing variants. ATV levels increased in HEK 463 and HEK5I, suggesting that ATV may be an inhibitor of MRP4 and MRP5. In contrast, intracellular DRV concentration HEK ATV vs DRV HEK293 WT HEK463 HEK5I HEK293 WT HEK463 HEK5I 0 500 1000 1500 2000 2500 3000 ATV DRV ** ** ** Concentration (ng/mg protein) MDCK ATV vs DRV MDCK WT MDCK MDR MDCK MRP2 MDCK WT MDCK MDR MDCK MRP2 0 100 200 300 400 ATV DRV ** ** Concentration (ng/mg protein) Page 102 of 140 was elevated in HEK463 (p-value<0.01), while there was no statistical difference in DRV concentration in HEK5I. When ATV was incubated in MDCK II, MDCK-MDR and MDCK-MRP2, the cellular concentration of ATV in both P-gp and MRP-2 overexpressing cells suggest that ATV can be eliminated via these transporters. In contrast, the intracellular concentration of DRV was no differences between all three MDCK cell lines. The effects of PEs on intracellular ATV or DRV concentrations (4.0 µg/mL) were evaluated at physiologically achievable concentration. In Figure 4.2a, ATV was combined with either COBI or RTV (1.0 µg/mL) in each HEK variant. The addition of COBI and RTV with ATV significantly reduced the ATV concentration in MDCK II. However, when COBI was added with ATV in MDCK-MDR, there was no statistical difference when compared to ATV alone. Unexpectedly, the co-administration of RTV with ATV in MDCK-MDR lead to a significant decrease in cellular ATV (Figure 4.2a). No difference in cellular ATV concentration was noted when either COBI or RTV was co- administered in MDCK-MRP2. Page 103 of 140 (a) (b) Figure 4.2 ATV versus DRV cellular concentration in MDCK transporter overexpressed variants after 24-hour PE boosted treatment. (a) ATV in MDCK variants (b) DRV in MDCK variants. *p-value<0.05. The addition of either COBI or RTV in MDCK II treated with DRV reduced the cellular levels of DRV. However, when COBI or RTV was added into P-gp-overexpressing MDCK-MDR treated with DRV, COBI reduced cellular DRV concentration (p-value<0.01). In contrast, the addition of RTV increased cellular DRV concentration (p-value <0.01). No effect on DRV concentration was seen when either COBI or RTV was added into MDCK-MRP2. Either COBI or RTV was added to HEK293, and no cellular differences were detected (Figure 4.3a). When COBI was co-administered with ATV in HEK463, no changes in cellular ATV were detected. However, at 1.0 µM RTV significantly increased ATV concentration. The addition of RTV into HEK5I significantly decrease ATV concentration, suggesting that RTV can induce MRP5. In contrast, the addition of COBI or RTV enhanced CC DRV in HEK and HEK463. However only RTV statistically increased DRV concentration in HEK5I. MDCK ATV MDCK WT MDCK MDR MDCK MRP2 0 50 100 150 200 250 300 350 400 ATV ATV+COBI ATV+RTV * * * Concentration (ng/mg protein) MDCK DRV MDCK WT MDCK MDR MDCK MRP2 0 25 50 75 100 125 150 DRV DRV+COBI DRV+RTV * * * * Concentration (ng/mg protein) Page 104 of 140 (a) (b) Figure 4.3 ATV versus DRV cellular concentration in HEK overexpressing variants after 24- hour PE boosted treatment. (a) ATV in HEK variants (b) DRV in HEK variants. *p- value<0.05. **p-value<0.01. ***p-value<0.005. 4.3.2 Cellular disposition in the presence of transporter inhibitors In order to evaluate the impact of transporter inhibitors on the cellular disposition of ATV and DRV in CD4+ cells (CEM and U937). Cellular concentrations of ATV and DRV were tested at low (L), medium (M) and high (H) concentrations, which corresponded to 500, 4000 and 8000 ng/mL respectively. Based on the results, I can determine the transporter expression CD4+ cell types in relation to efflux transporters. 4.3.2.1 P-gp The disposition of ATV or DRV in CEM and U937 were assessed by pretreating cells with P-gp specific inhibitor tariquidar using a concentration escalation design (100, 500 and 1000 nM) (Figure 4.4). This study design also evaluated the impact of ATV or DRV concentrations when they were escalated from 0.5 to 8.0 µg/mL for these PI, where the concentrations used are clinically achievable. HEK ATV HEK293 WT HEK463 HEK5I 0 1000 2000 3000 4000 ATV ATV+COBI ATV+RTV *** * Concentration (ng/mg protein) HEK DRV HEK293 WT HEK463 HEK5I 0 1000 2000 3000 4000 DRV DRV+COBI DRV+RTV * ** * Concentration (ng/mg protein) Page 105 of 140 Figure 4.4 Impact of P-gp inhibition on PI CC in CEM and U937. (a) CEM CC ATV (b) U937 CC ATV (c) CEM CC DRV (d) U937 CC DRV when treated with escalating concentration of P-gp inhibitor tariquidar. *p-value<0.05. Consistent with data from Figure 4.2, both DRV and ATV are P-gp substrates. The cellular concentration of ATV and DRV increased in a concentration-dependent manner. When Tariquidar was added to ATV or DRV, where the cellular increase was seen when Tariquidar was added. This was seen in both CEM and U937. However at the highest Tariquidar (1.0 µM) concentration when combined with either moderate to high concentration of ATV or DRV, there was a reduction in U937 cellular concentration. This suggested that U937 has mechanisms that are higher P-gp and thus reduce the level of ATV and DRV. This was not seen when CEM were used. This may be due to the difference between the two cells lines in terms of biosensor activation which regulates P- gp expression. This finding mandates further dissection as the difference for how ATV and DRV disposition is correlated with expression of P-gp. CEM Tariquidar ATV ATV(L) ATV(M) ATV(H) 0 5000 10000 15000 20000 25000 30000 35000 40000 +100 nM Tariquidar +500 nM Tariquidar Control +1000 nM Tariquidar * * * * * * * * * Concentration (ng/mg protein) U937 Tariquidar ATV ATV(L) ATV(M) ATV(H) 0 10000 20000 30000 40000 50000 60000 70000 80000 +100 nM Tariquidar +500 nM Tariquidar Control +1000 nM Tariquidar * * * * * * * * * Concentration (ng/mg protein) CEM Tariquidar DRV DRV(L) DRV(M) DRV(H) 0 5000 10000 15000 20000 25000 30000 35000 40000 +100 nM Tariquidar +500 nM Tariquidar Control +1000 nM Tariquidar * * * * Concentration (ng/mg protein) U937 Tariquidar DRV DRV(L) DRV(M) DRV(H) 0 10000 20000 30000 40000 50000 60000 70000 +100 nM Tariquidar +500 nM Tariquidar Control +1000 nM Tariquidar * * * * * * * Concentration (ng/mg protein) (a) (b) (d) (c) Page 106 of 140 4.3.2.2 MRP1/MRP4 Figure 4.5 Impact of MRP1/MRP4 on PI CC in CEM and U937. (a) CC ATV in CEM (b) CC ATV in U937 (c) CC DRV in CEM (d) CC DRV in U937 in the presence of MRP inhibitor MK571. *p-value<0.05. MK571 is a well-known inhibitor of MRP1 and MRP4. CEM and U937 have low expression of MRP4 as established by Western analysis. Using a concentration escalation design for PIs and PEs, the addition of MK571 enhanced ATV and DRV cellular concentrations in CEM (Figure 4.5 a & c). However, in the U937 cell, the addition of MK571 increased PI cellular concentrations at 3 and 15 µM. At 30 µM MK571, a reduction of ATV of cellular concentration was seen. U937 treated with ATV, MK571 escalation did not result in a concentration-dependent increase in cellular disposition. Rather at moderate or high MK571 concentrations in U937, reduced ATV concentration suggesting upregulation of elimination pathways [84]. I have previously shown that ATV and DRV use MRP2 for cellular elimination. In addition, these PIs are inhibitors of MRP4 and MRP5. This CEM MK571 ATV ATV(L) ATV(M) ATV(H) 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 Control +3 uM MK571 +15 uM MK571 +30 uM MK571 * * * * * * Concentration (ng/mg protein) U937 MK571 ATV ATV(L) ATV(M) ATV(H) 0 10000 20000 30000 40000 50000 60000 70000 80000 Control +3 uM MK571 +15 uM MK571 +30 uM MK571 * * * * * * * * * Concentration (ng/mg protein) CEM MK571 DRV DRV(L) DRV(M) DRV(H) 0 5000 10000 15000 20000 25000 30000 35000 40000 Control +30 uM MK571 +3 uM MK571 +15 uM MK571 * * * * * * Concentration (ng/mg protein) U937 MK571 DRV DRV(L) DRV(M) DRV(H) 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 Control +3 uM MK571 +15 uM MK571 +30 uM MK571 * * * * * * * * * Concentration (ng/mg protein) (a) (b) (d) (c) Page 107 of 140 suggests that U937 may have the mechanisms capable of eliminating the PI. It appeared that DRV is more likely to induce biosensors to express efflux transporters. This set of data further support that U937 is more responsive than CEM. Further exploration to understand these cellular adaptive processes in these two cells is necessary to delineate the current findings. 4.3.2.3 Breast Cancer Resistance Protein (BCRP) Figure 4.6 Impacts of BCRP on PI in CEM and U937. (a) CC ATV in CEM (b) CC ATV in U937 (c) CC DRV in CEM (d) CC DRV in U937 in the presence of BCRP inhibitor KO 143. *p-value<0.05. BCRP is an efflux transporter first identified in relations to mitoxantrone-resistant breast cancers, thus giving rise to its scientific acronym. This constitutive transporter is highly expressed in the gastrointestinal tract, renal epithelial, and blood-brain barrier cells. To determine its role in the cellular disposition, I used KO143, BCRP inhibitor, where CEM Ko143 ATV ATV(L) ATV(M) ATV(H) 0 5000 10000 15000 20000 25000 30000 35000 40000 Control +100 nM Ko143 +500 nM Ko143 +1000 nM Ko143 * * * * * * Concentration (ng/mg protein) U937 Ko143 ATV ATV(L) ATV(M) ATV(H) 0 10000 20000 30000 40000 50000 60000 70000 80000 Control +100 nM Ko143 +500 nM Ko143 +1000 nM Ko143 * * * * * * * * Concentration (ng/mg protein) CEM Ko143 DRV DRV(L) DRV(M) DRV(H) 0 5000 10000 15000 20000 25000 Control +100 nM Ko143 +500 nM Ko143 +1000 nM Ko143 * * * * Concentration (ng/mg protein) U937 Ko143 DRV DRV(L) DRV(M) DRV(H) 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 Control +100 nM Ko143 +500 nM Ko143 +1000 nM Ko143 * * * * * * * * * Concentration (ng/mg protein) (a) (c) (b) (d) Page 108 of 140 continue to use a concentration range of KO143 from 100, 500 and 1000 nM to assess what effect BCRP will have on PI cellular disposition [180]. The summary of data is reviewed in Figure 4.6. When KO143 was co-administered with ATV or DRV, low dose of ATV or DRV did not show significant changes in cellular concentration in CEM. I found that moderate to high DRV cellular concentration was not significantly increased until KO143 concentration reached 1.0 µM in CEM. Cellular concentrations of ATV and DRV are elevated with the addition of KO143 in U937. In U937, 1.0 µM KO143 combination with moderate or high concentration ATV treatment did not increase the cellular levels and even reduction. This suggests that there may be a threshold which can activate cellular adaptive mechanisms. This set of data further supports that inhibition of elimination pathways will increase intracellular concentrations of PI. However, inhibition of BCRP may increase the intracellular levels to activate elimination mechanisms. The inhibition of BCRP using KO143 is consistent when ATV in combined with inhibitors of MRPs and P-gp. Other important insights are that ATV appears to be more sensitive to activating elimination pathways where the inhibition of BCRP did not lead to cellular concentration reduction. 4.3.2.4 Multi-drug resistant associated protein-2 (MRP2) To further dissect the impact of inhibiting transporters such as resistance-associated protein 2 (MRP2) has various names including canalicular multispecific organic anion transporter 1 (cMOAT) or ATP-binding cassette sub-family C member 2 (ABCC2). This transporter has been previously reported in the efflux of HIV PIs, nucleosides and acyclic nucleoside phosphonates. Furosemide, a loop diuretic, is a substrate and inhibitor of Page 109 of 140 MRP2 when used at concentrations ranging from 5 to 50 µM, this compound is a potent inhibitor of MRP2. Figure 4.7 Impacts of MRP2 on PI CC in CEM and U937. (a) CC ATV in CEM (b) CC ATV in U937 (c) CC DRV in CEM (d) CC DRV in U937 in the presence of BCRP inhibitor furosemide. *p-value<0.05. The impact of furosemide addition with either ATV or DRV is summarized in Figure 4.7. ATV cellular concentration is sensitive to furosemide treatment, particularly higher concentrations of the inhibitor. Furosemide inhibition can cause a significant reduction in cellular ATV (L) in CEM which was not seen with other inhibitors. When furosemide and ATV were co-administered in U937, 25 µM furosemide did not increase cellular ATV but a consistent drop, where a high concentration of furosemide significantly increases ATV concentrations. When furosemide was used with DRV in CEM, there was a concentration-dependent increase in cellular DRV concentration. In U937, DRV concentration was more than two- fold higher than that seen in CEM (Figure 4.7c & d). At moderate concentration CEM Furosemide ATV ATV(L) ATV(M) ATV(H) 0 5000 10000 15000 20000 25000 30000 35000 40000 60000 90000 Control +5 uM Furosemide +25 uM Furosemide +50 uM Furosemide * * * * * * * * * Concentration (ng/mg protein) CEM Furosemide DRV DRV(L) DRV(M) DRV(H) 0 5000 10000 15000 20000 25000 30000 Control +5 uM Furosemide +25 uM Furosemide +50 uM Furosemide * * * Concentration (ng/mg protein) U937 Furosemide DRV DRV(L) DRV(M) DRV(H) 0 10000 20000 30000 40000 50000 60000 70000 Control +5 uM Furosemide +25 uM Furosemide +50 uM Furosemide * * * * * * * * * Concentration (ng/mg protein) U937 Furosemide ATV ATV(L) ATV(M) ATV(H) 0 10000 20000 30000 40000 50000 60000 70000 80000 Control +5 uM Furosemide +25 uM Furosemide +50 uM Furosemide * * * * * * * * * Concentration (ng/mg protein) (a) (c) (b) (d) Page 110 of 140 furosemide (25 µM), both cellular DRV and ATV concentration were lower than low concentrations of furosemide. Nevertheless, when the concentration of furosemide is increased to 50 µM, both cellular ATV and DRV concentration was elevated. At a moderate concentration of furosemide, an induction of efflux transporters may lead to a reduction of cellular ATV and DRV concentration. The induction of cellular adaptive processes appears to be cellular concentration-dependent where ATV appears to be significantly higher in all the tests when compared to DRV. 4.3.3 Transporter gene expression in CD4+ cells The cellular disposition data suggested that molecular interrogation will be critical to dissect the cellular adaptive mechanism(s) in response to accumulating levels of PI. I evaluated gene expression of various efflux transporter in response to PI found in CEM and U937 after PI alone or in combination with PE treatments. Targeted genes that were analyzed include ABCC1, ABCC2, ABCB1 and ABCG2 which corresponds to MRP1, MRP2, P-gp, and BCRP, respectively. Gene expression of the targeted gene in U937 and CEM were evaluated after treatments in comparison to untreated controls. The overall data was summarized in Figure 4.8 & 4.9. In U937, treatment with ATV alone increased expression of ABCB1 and ABCG2, while decreased expression of MRP2. The addition of COBI or RTV had no dramatic gene expression change for ABCC1, ABCB1 or ABCC2 as compared to ATV alone (Figure 4.8.a & 4.8.c). However, ATV in combination with COBI (Figure 4.8.b) showed a decreased expression of ABCG2, while the addition of RTV led to an increase in expression. Page 111 of 140 DRV treatment alone increased expression of ABCB1 and ABCG2, with no changes in ABCC1 and ABCB1. DRV alone also reduced expression of ABCC2. When DRV was combined with COBI, there were no changes in expression of ABCC2, ABCB1, and ABCG2 when compare to DRV alone. However, decreased ABCC1 expression was noted when compared to DRV alone. But when DRV was combined with RTV, increase in ABCC1 and ABCC2 expression when compared cells treated with DRV alone (Figure 4.8.a). U937 ABCC1 Expression 0 1 2 3 4 Control ATV ATV+COBI ATV+RTV DRV DRV+COBI DRV+RTV Treatment Fold Change U937 ABCC2 Expression 0.0 0.5 1.0 1.5 2.0 Control ATV ATV+COBI DRV DRV+COBI DRV+RTV Treatment Fold Change U937 ABCB1 Expression 0 1 2 3 4 Control ATV ATV+COBI ATV+RTV DRV DRV+COBI DRV+RTV Treatment Fold Change U937 ABCG2 Expression 0 1 2 3 4 Control ATV ATV+COBI ATV+RTV DRV DRV+COBI DRV+RTV Treatment Fold Change (a) (b) (d) (c) Figure 4.8 Transporter gene expression in U937 when treated with PIs alone or in combination with PEs. (a) ABCC1 (b) ABCC2 (c) ABCB1 (d) ABCG2 In Figure 4.9, the expression of efflux transporters in CEM cells is summarized. The addition of ATV or DRV increased the expression of ABCC1, where the addition of either COBI or RTV did not significantly alter the expression of this transporters when compared Page 112 of 140 to ATV or DRV alone (Figure 4.9.a). ATV alone did not significantly alter the expression of ABCC2, ABCB1, or ABCG2. However, the combination of ATV with COBI did increase ABCC1 and ABCB1 expression by about two-fold for both of these transporters. ATV combination with RTV did not alter the expression of ABCC2, ABCB1, or ABCG2 in CEM cells (Figure 4.9 b, c &d). In contrast, DRV induced expression of ABCC2, ABCB1, and ABCG2, while the combination with COBI did not further alter the expression of these genes (Figure 4.9 b, c, & d). When DRV was combined with RTV, additional increases in the expression of these genes were detected. CEM ABCC1 Expression 0 2 4 6 8 10 12 Control ATV ATV+COBI ATV+RTV DRV DRV+COBI DRV+RTV Treatment Fold Change CEM ABCC2 Expression 0 2 4 6 8 10 12 Control ATV ATV+COBI ATV+RTV DRV DRV+COBI DRV+RTV Treatment Fold Change CEM ABCB1 Expression 0 2 4 6 8 10 12 Control ATV ATV+COBI ATV+RTV DRV DRV+COBI DRV+RTV Treatment Fold Change CEM ABCG2 Expression 0 2 4 6 8 10 12 Control ATV ATV+COBI ATV+RTV DRV DRV+COBI DRV+RTV Treatment Fold Change (a) (b) (d) (c) Figure 4.9 Transporter gene expression in CEM when treated with PIs alone or in combination with PEs. (a) ABCC1 (b) ABCC2 (c) ABCB1 (d) ABCG2 Page 113 of 140 4.3.4 Protein level expression of transporters in CD4+ cells To verify that gene expression was correlated with protein expression, I assessed the protein level expression of these transporters using western blotting analysis in CEM and U937 (Table 4.4). To ensure that antibodies were working effectively, I used cells lines that overexpress these specific transporters as the positive control. Here I presented the treatment of 4000 ng/mL of PIs and 1000 ng/mL of PEs. The specific band signal intensity was quantitated and normalized to the corresponding β-actin intensity, where the ratios of which to the non-treatment group were summarized in Table 4.4. (a). U937 ATV Ratio to NT ATV A+C A+R COBI RTV MRP1 1.03 1.20 1.11 1.01 1.17 MRP2 0.72 0.57 0.20 0.42 0.95 MRP4 1.12 0.73 0.19 0.91 0.90 P-gp 0.97 1.62 1.38 1.01 1.04 BCRP 1.16 1.06 0.62 0.43 0.44 OATP1B1 1.16 1.17 1.62 0.84 1.26 (b). U937 DRV Ratio to NT DRV D+C D+R COBI RTV MRP1 1.33 1.36 0.94 1.16 1.22 MRP2 1.58 0.00 0.00 0.85 1.43 MRP4 1.49 1.33 0.95 0.84 1.15 P-gp 1.31 1.22 1.03 1.11 1.22 BCRP 1.44 1.24 1.19 1.40 1.49 OATP1B1 1.56 4.94 2.04 3.40 2.21 (c). CEM ATV Ratio to NT ATV A+C A+R COBI RTV MRP1 1.1 1.3 0.9 0.7 1.1 MRP2 0.7 1.2 0.6 0.4 0.6 MRP4 0.0 0.0 0.0 0.0 0.0 P-gp 1.0 0.7 1.0 1.0 0.8 BCRP 0.9 1.2 0.6 0.7 1.1 OATP1B1 1.0 1.3 0.9 0.8 1.0 (d). CEM DRV Ratio to NT DRV D+C D+R COBI RTV MRP1 1.7 1.1 1.1 0.8 1.1 MRP2 0.6 0.4 0.4 1.3 2.6 Page 114 of 140 MRP4 0.0 0.0 0.0 0.0 0.0 P-gp 0.7 1.0 0.9 0.8 1.0 BCRP 1.3 1.1 1.2 1.0 1.3 OATP1B1 2.1 1.8 1.7 1.1 1.4 NT: Non-treatment control Table 4.4 Signal intensity ratio of drug treatments to non-treated control based on normalized intensity. (a) U937 ATV group. (b) U937 DRV group. (c) CEM ATV group. (d). CEM DRV group. Cells were treated with 4000 ng/mL of PI and 1000 ng/mL for 24 hours. OATP1B1 is the only vital influx transporter involved in ATV and DRV uptake. Data suggested that PI/PE treatments induced expression of OATP1B1, which supposed to enhance cellular PI intake. Thus, efflux transporters are more important factors to determine the different PIs cellular retention (Ke and AUC) discussed in Chapter 3. Both PIs did not generate significant changes in expression of efflux transporters in CEM cells, except for DRV related induction of MRP1 and BCRP. In general, ATV alone and in combination with PEs did not significantly alter expression of the transporters (Table 4.4 a & c). The only exception was an increased expression of P-gp and MRP1 in U937 treated ATV in combination with COBI and RTV. In contrast, U937 cells treated with DRV alone had a slight increase in expression of transporters by 1.3 to 1.5-fold. However, the combination with either COBI or RTV did not further stimulate the protein expression; rather the combination of DRV with PEs reduced transporter expression. Due to the relatively low expression of MRP2 and MRP4 in CEM and U937, I hypothesized that MRP1, P-gp, and BCRP are the most involved efflux transporters for PI cellular disposition. Also, PI in combination with COBI was associated with a higher protein expression when compared to PI and RTV combination. Page 115 of 140 (a). Ratio to Untreated CEM Cobicistat (ng/mL) Ritonavir (ng/mL) 100 500 1000 10000 100 500 1000 10000 MRP1 0.85 0.82 1.05 0.65 0.79 0.90 1.03 0.88 P-gp 1.11 0.89 1.15 0.41 0.65 0.92 0.49 0.73 BCRP 0.88 0.78 0.82 0.68 0.86 0.83 1.25 0.42 Ratio to untreated U937 Cobicistat (ng/mL) Ritonavir (ng/mL) 100 500 1000 10000 100 500 1000 10000 MRP1 0.76 1.31 1.40 1.64 0.95 0.92 1.26 0.86 P-gp 0.68 2.05 0.98 2.60 0.97 2.21 0.51 2.08 BCRP 0.63 1.00 1.37 1.37 0.61 0.91 0.88 0.88 Table 4.5 The protein expression of each of the efflux transporter in relations to untreated cells. (a) CEM and (b) U937 cells. To further evaluate impacts of PEs on the protein expression of specific efflux transporters concentrations 100, 500, 1000 and 10000 ng/mL COBI or RTV were examined (Table 4.5). In this table, the protein expressions were correlated with the housekeeping protein β-actin, and the expression of untreated cells was 1.0. Expression of these efflux transporters was not significantly altered by the addition of either COBI or RTV, as the protein fold change ranged from 0.41 to 2.6. In CEM, although 1000 ng/mL of PEs induced expression of transporter proteins, 10,000 ng/mL of either COBI or RTV caused downregulation of the transporter expression. The data is summarized in Table 4.5a. A similar evaluation was performed in U397, summarized in Table 4.5b. In U937, 500 or 1000 ng/mL of COBI or RTV addition caused induced efflux transporters expression. Interestingly 1000 ng/mL of COBI and RTV reduced the expression but increased nearly two-fold at 10,000 ng/mL. Page 116 of 140 4.3.5 Transporter functional tests To confirm that protein expression corresponded to the transporter’s functional activity, I examined the function of the transporters based on fluorescent dye retention obtained through flow cytometric assays. Efflux function and activity of the transporters was determined by the inhibition of fluorescent dye efflux, which was reflected as the ratio of mean fluorescence intensity (MFI) between treated sample and non-treated control, calculated using the following formula: Inhibition on dye efflux = (MFI drug - MFI no dye) /( MFIdye only - MFI no dye). Thus, a higher ratio indicates higher cellular retention of the dye and inhibition of the transporter. In this study, I utilized and measured the cellular fluorescence intensity of P-gp (partially MRP1) substrate rhodamine 123 (Rh123) and MRP1 substrate calcein AM after 24-hour PE treatments. P-gp Function COBI (ng/mL) RTV (ng/mL) 100 500 1000 10000 100 500 1000 10000 CEM 1.03 1.05 1.32 1.87 1.21 1.13 1.11 1.63 U937 1.09 1.01 0.97 1.03 1.17 1.06 1.04 1.21 Table 4. 6 Rhodamine efflux after 24-hour COBI or RTV treatments in CEM and U937. The impact of COBI or RTV in concentration escalation on P-gp function is summarized in Table 4.6. In CEM, concentration escalation of COBI corresponded with the decrease of P-gp transport. Similarly, concentration escalation of RTV in CEM also decrease P-gp function. In U937, across the concentrations evaluated, COBI and RTV had little impact on P-gp function. However, when RTV was escalated to 10,000 ng/mL there was a 20% increase in P-gp function. Calcein AM was used to determine the impact of COBI and RTV effect on MRP1 activity, with the results summarized in Table 4.7. Page 117 of 140 MRP1 Function COBI (ng/mL) RTV (ng/mL) 100 500 1000 10000 100 500 1000 10000 CEM 1.94 1.46 1.78 5.65 1.05 1.60 1.46 5.04 U937 5.62 2.51 5.53 24.98 2.78 4.28 4.46 17.71 Table 4.7 MRP1 functions after 24-hour PE treatments in CEM and U937. In CEM, COBI exhibited reduced inhibition on MRP1 at 500 and 1000 ng/mL when compared to 100 ng/mL, while 10,000 ng/mL of COBI recovered its inhibition on MRP1. As for RTV, 500 and 1000 ng/mL showed reduced inhibition on MRP1, but 10000 ng/mL significantly enhanced its inhibition of P-gp. In U937, PEs showed extensive induction on MRP1 compared to baseline. COBI also showed reduced inhibition on MRP1 at 500 and 1000 ng/mL compared to 100 ng/mL, while 10,000 ng/mL can significantly recover its inhibition on MRP1. In contrast, RTV’s inhibition on MRP1 is increased with the increase of its concentration in U937. Therefore, for both PEs, there was reduced inhibition of P-gp and MRP1 activity at concentrations between 500 to 1000 ng/mL. This may be due to PE induction of PXR/CAR. However, 10,000 ng/mL of PEs recovered PE inhibition ability on MRP1 and P-gp functions, probably because of the high dosage. This fact is consistent with the previous western blotting expression data. 4.4 Discussion I found that COBI and RTV had a significant difference in PIs cellular disposition, which ultimately impacts the treatment efficacy. In the previous chapter, I found that COBI’s impact on PI retention is lower than RTV when using clinically achievable levels. When DRV and ATV were given alone, their cellular concentrations were similar, though PI Page 118 of 140 efflux was a determinant of overall cellular retention. I showed that PEs inhibit not only drug metabolizing enzymes but also membrane transporters. The interplay between DME and transporters has been observed to mediate a variety of drug PK profiles. It has been realized that transporters are key factors in eliminating anti-HIV agents. PEs are known CYP3A4 inhibitors but their effects on transporters are less known. As a result, I examined the roles of multiple transporters on ATV and DRV cellular disposition along with further evaluating PE impacts on transporters and alteration of PI disposition. In this study, I found that ATV was an inhibitor of MRP4 and MRP5, as ATV accumulated in HEK463 and HEK5I. DRV was also an inhibitor of MRP4 but not for MRP5 (Figure 4.1a). When both PIs were tested in MDCK wild type and overexpressing P-gp and MRP2 variants, the cellular concentration of ATV was significantly lower in the variants suggesting that MRP2 and P-gp were both potential routes for cellular ATV elimination. In contrast, DRV cellular levels were not different between the various cellular variants (Figure 4.1b). COBI and RTV were added to determine the cellular disposition of ATV or DRV, summarized in Figure 4.2. The addition of COBI or RTV reduced cellular ATV level in MDCK II, suggesting that at this concentration of PE, there is induction of elimination pathways. In MDCK-MDR, COBI had no changes in ATV concentration, while RTV induced a significant reduction of ATV cellular levels. No changes in ATV cellular levels were seen when either COBI or RTV was added to MDCK-MRP2 treated with ATV, suggesting that at this concentration of COBI or RTV, there is no effect. When DRV was evaluated as the substrate, the addition of COBI and RTV were able to reduce intracellular DRV concentration, similarly to what was seen with ATV in MDCK II. Page 119 of 140 However, when COBI or RTV was added into MDCK overexpressing MDR, the addition of COBI suggested that there was the induction of P-gp, which lowered the cellular concentration of DRV. In contrast, the addition of RTV increases MDCK-MDR levels of DRV. This suggests that RTV blocked P-gp function in this experiment. The addition of COBI or RTV did not affect MRP2 mediated elimination. This set of data suggests that both COBI and RTV induce elimination of ATV and DRV. The addition of COBI or RTV did not affect the disposition of ATV in HEK293. In MRP4- and MRP5-overexpressing HEK cells, the addition of COBI did not affect ATV cellular disposition. In contrast, RTV appeared to block MRP4 while inducing MRP5 expression in these experiments. Additionally, the cellular concentration of DRV increased with the addition of either COBI or RTV in both MRP4- and MRP5- overexpressing cells. These experiments support the data suggesting that ATV and DRV are both substrates for P-gp. This was affirmed when ATV and DRV accumulation was seen when co- administered with tariquidar in CEM. There was a concentration-dependent effect seen in CEM for both PI. I then conducted these experiments using U937, and found that increase in tariquidar levels increased cellular disposition of ATV and DRV in a dose- dependent manner. However, at high tariquidar concentrations, there was a reduction of cellular DRV and ATV. This suggests that cellular PI concentrations reached thresholds that activate additional cellular adaptive mechanisms. These findings were consistent when MK571, KO143, and furosemide was used to inhibit MRPs, BCRP and MRP2, respectively. At the highest concentration of MK571 and KO143, there was a reduction of cellular ATV and DRV concentration in U937, but not with furosemide. The reduction in cellular ATV was more significant after its cellular concentration reaches a maximal point. Page 120 of 140 I analyzed the gene expression of COBI versus RTV and found that the addition of COBI and RTV reduced the expression of ABCC1, ABCC2, but no changes in ABCB1 or ABCG2, as compared to the ATV or DRV alone. This was found in both U937 and CEM. To verify the gene expression data, I found that ATV can reduce expression of MRP2, where the addition of either COBI or RTV further reduced the protein expression. The combination of ATV and RTV had protein expression that was far more than RTV alone, suggesting a synergistic effect. This effect was seen in both CEM and U937. Finally, the overall function of these transporters impacted by PEs was evaluated through flow cytometry. Rhodamine 123 and calcein AM flow cytometric data suggested that COBI and RTV inhibit P-gp and MRP1 function in a dose-related fashion Functional data were consistent with the expression data, specifically that reduced inhibition of transporter function occurred at 500 or 1000 ng/mL while extremely high dosage restored PE inhibition ability. It was also noted that 1000 ng/mL of COBI slightly induced P-gp function in U937. COBI was a weaker P-gp inhibitor than RTV in both CD4+ cells. As MRP1 inhibitors, COBI tends to be more effective than RTV only at higher concentrations. To further evaluate the transporter function (reflected as dye fluorescence intensity) in relation to transporter expression, linear regression analysis was performed in both cell lines (Figure 4.10). Compared with protein level expression, gene level expression of MRP1 and P-gp correlated better with functional assays using fluorescence markers. The higher gene expression level correlated with lower fluorescence intensity in the cells, with p<0.05 in U937. Consider the relatively low expression of test transporters found in both CD4+ cells, gene level expression may be a better indicator to predict transporter function for future studies. Page 121 of 140 P-gp Expression VS Fluorescence Intensity in CEM 7000 7500 8000 8500 9000 0 2 4 6 8 Protein Expression Gene Expression Dye Fluorescence Expression MRP1 Expression VS Fluorescence Intensity in CEM 0 50000 100000 150000 0 1 2 3 4 5 6 Protein Expression Gene Expression Dye Fluorescence Expression P-gp Expression VS Fluorescence Intensity in U937 60000 70000 80000 90000 100000 0.0 0.5 1.0 1.5 2.0 2.5 Protein Expression Gene Expression Dye Fluorescence Expression MRP1 Expression VS Fluorescence Intensity in U937 20000 21000 22000 23000 24000 25000 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Gene Expression Protein Expression Dye Fluorescence Expression (a) (b) (d) (c) Figure 4.10 Transporter expression in relation to transporter function. (a) P-gp and (b) MRP1 expression in relation to transporter function indicator fluorescence intensity in CEM. (c) P-gp and (d) MRP1 expression in relation to transporter function indicator fluorescence. A major limitation of this study was that CD4+ cells used were not from HIV patients. However, the different properties of these two CD4+ cell models still offered important pharmacological information to illustrate clinical issues. This study provides information on dissecting the pharmacological mechanism behind the clinical virologic failure. 4.5 Conclusion The addition of COBI or RTV in combination with HIV-PI altered the expression and function of major efflux transporters. The level of alteration was dependent on the level of HIV-PI found in the cells. When HIV-PI reaches a certain threshold, PXR-mediated Page 122 of 140 activities increased. I hypothesize there is a threshold of HIV-PI required to displace the repressors of PXR. In this study, I have confirmed that the addition of COBI or RTV can induce PXR-activation and thus induce efflux transporters expression in CD4+ cells at the clinically relevant concentrations. Consistent with previous cellular retention data, COBI boosted PI treatments were inferior to RTV in terms of inhibition on the transporter. Transporters expression that were most affected were P-gp, MRP1, and BCRP. COBI is less potent inhibitor when compared to RTV. This may explain the higher incidence of clinical virologic failure in the patients receiving COBI compared to RTV. Page 123 of 140 Chapter 5. Conclusions and Future Directions 5.1 Summary Suppression of infectious pathogens is dependent on the ability to establish intracellular drug levels above the effective threshold (IC50 or IC90). Therefore, the critical factor determining treatment efficacy is the level achieved at the target site. Drug exposure at the target site is dictated by metabolism and drug disposition processes. It is well recognized that drug efficacy can vary widely among different patient populations such as children versus adults. A major factor is the differences in enzyme and transporter functions. DME and transporter expressions and functions are influenced by age, sex, ethnicity and additional individual factors. To understand the mechanisms and the roles these factors have on overall exposure can be of great importance for overall drug dosing strategy. In this thesis, I first evaluated the role of DMEs on age-related changes of the drug PK. I then dissected pharmacological mechanisms associated with drug disposition and their impact on clinical efficacy. These findings help to optimize dosing strategy, screen patients for appropriate treatments, and improve therapeutic outcomes. 5.1.1 Summary of voriconazole study In Chapter 2 I found that enzyme metabolic activities were dependent on the aging process for three enzyme systems studied. The age-dependent activity explains why pediatric patients required higher dosages compared to their adult counterpart, despite adjusting dose using size or body weight. The data suggested that the pediatric patients had a difference in expression of metabolic enzymes. To further understand why this occurred, I developed an enzyme probe cocktail capable of specifically characterizing the enzymatic activities of CYP2C19, CYP3A4, and FMO3 in pediatric patients receiving Page 124 of 140 voriconazole. The findings suggested that voriconazole metabolism reduces with increasing age in pediatric patients. At puberty, these metabolic changes reach a threshold with all three enzymes. In particular, FMO3 and CYP2C19 expression correlated with (p<0.05) with voriconazole metabolic clearance. The important role of CYP2C19 in this process in children is consistent with that seen in adults. In this study, the critical role of FMO3 in voriconazole metabolism among pediatric patients was unexpected. It is necessary to assess individual FMO3 activity when dosing voriconazole in pediatric patients. Although I did not detect a difference between the sexes in relation to voriconazole metabolism or enzyme activities, I did see a higher metabolic ratio in FMO3 and CYP2C19 in males when compared to females. A larger study to determine the impact of sex and age on the expression of these enzymes may further improve the understanding as to the impact of these two factors. 5.1.2 Summary of Pharmacoenhancer (PE) study Apart from drug metabolizing enzymes, transporters constitute another important factor regulating drug disposition. In Chapter 3 and 4, I delineated how transporter expression and function were impacted by co-administration of either COBI or RTV. I showed that COBI or RTV increased the cellular disposition of HIV protease inhibitors in a concentration dependent manner. However, when PI levels reached a threshold, there was activation of PXR-mediated activities. This led to increased expression of efflux transporters like P-gp and MRPs. Although COBI and RTV are structural analogs, COBI was found to be a significantly weaker inhibitor than RTV. Both ATV and DRV had lower cellular concentrations when co-administered with COBI when compared to RTV. These findings may explain why COBI-containing regimens had a higher virologic failure than Page 125 of 140 RTV. I also examined PI disposition in two different types of CD4+ cell types that were tropic for HIV infection, and lower cellular concentrations of both ATV and DRV were found in monocytes. This finding further supported clinical evidence of persistent HIV infection in monocytic cells, and highlights the need for further understanding of this issue to eradicate the virus. Cellular retention data presented in Chapter 3 shows that ATV was a more responsive substrate than DRV when-co-administered with either COBI or RTV. ATV containing treatments achieved higher cellular AUC than DRV. In addition, I detected a difference in cellular disposition between U937 and CEM. The cellular AUC was found to be lower in U937 when compared to CEM. This may be due to the size difference between the two cell types and/or the expression of the efflux transporters. A wide spectrum of tools was used to interrogate the molecular mechanism as detailed in Chapter 4. I have confirmed PIs themselves have efflux transporter inhibitory activities. However, increased cellular levels of these PIs also induced the expression of efflux transporters. The addition of PEs like COBI and RTV can enhance intracellular PI levels, displace repressors found on PXR, and thus induce the expression of metabolic enzymes and efflux transporters. Efflux transporters like P-gp, MRP1, and BCRP are the predominant efflux transporters regulating PI cellular disposition. The effect of PEs on these efflux transporters was found to be dose-dependent. Transporter induction and function when at concentrations of 500 and/or 1000 ng/mL of COBI or RTV was used. When COBI or RTV was used at 10,000 ng/mL there was induction of PXR-mediated activity. RTV was more potent than COBI as an inhibitor of transporters like P-gp. These findings further support previous cellular retention data in chapter 3. Page 126 of 140 To sum up this project, I found that CD4+ cell types affect PI/PE treatment efficacy. CEM lymphocytes were generally more responsive to PI/PE treatments than U937 monocytes, probably owing to high transporter-mediated elimination in U937. This correlates to HIV tropism as macrophage infection is more persistent, generating obstacles for HIV treatments [181, 182]. Thus, failure of PI/PE treatments may be due to the individual variation on the dominant infected CD4+ cell type. COBI was a weaker transporter inhibitor than RTV. These mechanistic differences could be the reasons for higher virologic failure seen in COBI and DRV containing treatments. 5.2 Conclusion and Significance In this dissertation, I conducted research on individual variation and factors that impact drug metabolism and disposition. For the voriconazole study, I found that ontogeny of drug metabolizing enzymes alters their roles on age-related voriconazole metabolism. Thus, the study recognized and confirmed that importance of FMO3 in addition to CYP2C19 for pediatric voriconazole metabolism. By understanding the mechanistic factors for voriconazole non-linear PK, science will be able to optimize drug dosing strategy more precisely. My probing of the impact of inhibitors of transporters provides crucial pharmacological evidence as to the cause of clinical virologic failure. It is important to verify PI cellular exposure, as it correlates to the virologic failure rate in the clinical setting. Although both PEs are potent CYP3A4 inhibitors, the effect on efflux transporters and nuclear receptors, especially in CD4+ cells, are less known. My data provide possible mechanistic clues to explain their distinct treatment outcomes. Moreover, it is noted that different types of Page 127 of 140 CD4+ cells react differently to PI/PE treatments. Thereby, the dominant HIV tropisms may be used as an indicator to predict potential failure of PI/PE treatments. In summary, individual variation and factors can impact drug metabolism and disposition by influencing drug metabolizing enzymes and transporters and thus mediating ultimate efficacy of the drug therapy. 5.3 Future Direction 5.3.1 Voriconazole The future plan for voriconazole project is to recruit a larger number of subjects across the entire spectrum of age. The ability to establish age, sex, and the onset of puberty may allow us to develop a rich dataset. Also, genotypes will be taken into account. This set of data can be used to develop a more precise simulation model that will enable more accurate dosing in the pediatric population. The model will help to optimize the administration dosage for the individual pediatric subject. Moreover, this method could be applied to other drug agents with similar properties for predicting and optimizing treatment outcomes among the special population. The approach of using drug probes to phenotype patients can be adapted to other age-dependent expression of various metabolic enzymes. In addition, this may be applied to chemotherapy for various pediatric cancers. This model sets the stage for additional exploration using this type of approach. 5.3.2 PI/PE My thesis found that when either COBI or RTV was co-administered with HIV-PI, this results in an increase cellular concentration of these substrates. COBI and RTV were Page 128 of 140 both able to block efflux transporter activity, thus transporter-mediated cellular elimination was inhibited. In my hands, RTV was more potent than COBI with regards to blocking transporter activity. These findings are consistent with the ability to inhibit HIV proliferation both in vitro and in human studies. COBI and RTV can both increase cellular levels of HIV-PI, however the levels in monocytes are significantly lower than that of lymphocytes. The lower intracellular concentration may be a consequence of transporter mediated elimination where monocytes express a high level of membrane transporter. Moreover, the finding that monocyte/macrophage have the cellular mechanism(s) to reduce intracellular concentration of HIV-PI needs to be further evaluated. There are studies suggesting that monocytes may be a site of persistent HIV proliferation, where they can then localize into lymph nodes. The ability to eradicate HIV in monocytes may lead to curative modality. From the current findings, I hope that others will be able to further my findings into HIV in the sanctuary and ultimately eradicate this scourge from infected patients. An additional aspect that was found in these studies showed that when HIV-PI levels reach a certain threshold, this can activate cellular adaptive mechanisms. Namely the activation of PXR-mediated transcription of transporters. The methods that I have deployed can be adapted to human ex vivo study. One way to validate the finding is to use primary human polymorphonuclear cells (PBMCs) isolated from fresh blood. PBMCs can be isolated using density gradient centrifugation, and enriched using CD4+ separation methods. To determine the impact of HIV infection on Page 129 of 140 the expression and function of efflux transporters can be easily determined using flow cytometry. To further confirm the effect of pharmacoenhancers, a similar study can be conducted using humanized mice that has been chronically infected with HIV. Alternatively, one can use HIV-infected patients who are treated with the same HIV-PI regimen alone and in combination with a pharmacoenhancer. Blood can be collected to determine the HIV-PI found in their PBMCs. Additionally, the expression and function of efflux transporter can use flow cytometric methods. I hope that others will be able to further my findings into HIV in the sanctuary and ultimately eradicate this scourge from the infected patients. Page 130 of 140 Bibliography and References 1. Lipinski, C.A., Lead-and drug-like compounds: the rule-of-five revolution. Drug Discovery Today: Technologies, 2004. 1(4): p. 337-341. 2. Wilkinson, G.R., Drug metabolism and variability among patients in drug response. New England Journal of Medicine, 2005. 352(21): p. 2211-2221. 3. Kumar, G.N. and S. Surapaneni, Role of drug metabolism in drug discovery and development. Medicinal research reviews, 2001. 21(5): p. 397-411. 4. Guengerich, F.P., Cytochrome P450s and other enzymes in drug metabolism and toxicity. The AAPS journal, 2006. 8(1): p. E101-E111. 5. Rogers, A.S., The role of cytochrome P450 in developmental pharmacology. Journal of adolescent health, 1994. 15(8): p. 635-640. 6. Guengerich, F.P., Human cytochrome P450 enzymes, in Cytochrome P450. 2015, Springer. p. 523-785. 7. Nelson, D.R., et al., P450 superfamily: update on new sequences, gene mapping, accession numbers and nomenclature. Pharmacogenetics and Genomics, 1996. 6(1): p. 1-42. 8. Shimada, T., et al., Interindividual variations in human liver cytochrome P-450 enzymes involved in the oxidation of drugs, carcinogens and toxic chemicals: studies with liver microsomes of 30 Japanese and 30 Caucasians. Journal of Pharmacology and Experimental Therapeutics, 1994. 270(1): p. 414-423. 9. Smith, D.A., M.J. Ackland, and B.C. Jones, Properties of cytochrome P450 isoenzymes and their substrates Part 1: active site characteristics. Drug Discovery Today, 1997. 2(10): p. 406-414. 10. Faber, M.S., A. Jetter, and U. Fuhr, Assessment of CYP1A2 activity in clinical practice: why, how, and when? Basic & clinical pharmacology & toxicology, 2005. 97(3): p. 125- 134. 11. Rendic, S., Summary of information on human CYP enzymes: human P450 metabolism data. Drug Metab Rev, 2002. 34(1-2): p. 83-448. 12. Miners, J.O. and D.J. Birkett, Cytochrome P4502C9: an enzyme of major importance in human drug metabolism. British journal of clinical pharmacology, 1998. 45(6): p. 525- 538. 13. Gotoh, O., Substrate recognition sites in cytochrome P450 family 2 (CYP2) proteins inferred from comparative analyses of amino acid and coding nucleotide sequences. Journal of Biological Chemistry, 1992. 267(1): p. 83-90. 14. P.B. Danielson, B.S.P., The Cytochrome P450 Superfamily: Biochemistry, Evolution and Drug Metabolism in Humans. Current Drug Metabolism, 2002. 3(6): p. 561-597. 15. Gaedigk, A., Complexities of CYP2D6 gene analysis and interpretation. Int Rev Psychiatry, 2013. 25(5): p. 534-53. 16. Kennedy, M., Hormonal regulation of hepatic drug-metabolizing enzyme activity during adolescence. Clin Pharmacol Ther, 2008. 84(6): p. 662-73. 17. Scott, E.E. and J.R. Halpert, Structures of cytochrome P450 3A4. Trends in biochemical sciences, 2005. 30(1): p. 5-7. Page 131 of 140 18. Cashman, J.R., Structural and catalytic properties of the mammalian flavin-containing monooxygenase. Chemical research in toxicology, 1995. 8(2): p. 165-181. 19. Cashman, J.R., Some distinctions between flavin-containing and cytochrome P450 monooxygenases. Biochemical and biophysical research communications, 2005. 338(1): p. 599-604. 20. Phillips, I.R. and E.A. Shephard, Flavin-containing monooxygenases: mutations, disease and drug response. Trends Pharmacol Sci, 2008. 29(6): p. 294-301. 21. Krueger, S.K. and D.E. Williams, Mammalian flavin-containing monooxygenases: structure/function, genetic polymorphisms and role in drug metabolism. Pharmacology & therapeutics, 2005. 106(3): p. 357-387. 22. Cashman, J.R., Role of flavin-containing monooxgenase in drug development. Expert opinion on drug metabolism & toxicology, 2008. 4(12): p. 1507-1521. 23. Cashman, J.R. and J. Zhang, Human flavin-containing monooxygenases. Annu. Rev. Pharmacol. Toxicol., 2006. 46: p. 65-100. 24. Walker, C.H. and M.I. Mackness, Esterases: problems of identification and classification. Biochemical pharmacology, 1983. 32(22): p. 3265-3269. 25. Aldridge, W., Serum esterases. 1. Two types of esterase (A and B) hydrolysing p- nitrophenyl acetate, propionate and butyrate, and a method for their determination. Biochemical Journal, 1953. 53(1): p. 110. 26. Mutch, E., et al., The role of esterases in the metabolism of ciclesonide to desisobutyryl- ciclesonide in human tissue. Biochemical pharmacology, 2007. 73(10): p. 1657-1664. 27. Jamis-Dow, C., et al., Rifampin and rifabutin and their metabolism by human liver esterases. Xenobiotica, 1997. 27(10): p. 1015-1024. 28. Benedetti, M.S. and K.F. Tipton, Monoamine oxidases and related amine oxidases as phase I enzymes in the metabolism of xenobiotics, in MAO — The Mother of all Amine Oxidases, J.P.M. Finberg, et al., Editors. 1998, Springer Vienna: Vienna. p. 149-171. 29. Youdim, M.B. and Y.S. Bakhle, Monoamine oxidase: isoforms and inhibitors in Parkinson's disease and depressive illness. Br J Pharmacol, 2006. 147 Suppl 1: p. S287- 96. 30. Rowland, A., J.O. Miners, and P.I. Mackenzie, The UDP-glucuronosyltransferases: their role in drug metabolism and detoxification. The international journal of biochemistry & cell biology, 2013. 45(6): p. 1121-1132. 31. Bock, K.W., Roles of human UDP-glucuronosyltransferases in clearance and homeostasis of endogenous substrates, and functional implications. Biochemical pharmacology, 2015. 96(2): p. 77-82. 32. Jedlitschky, G., et al., Transport of glutathione, glucuronate, and sulfate conjugates by the MRP gene-encoded conjugate export pump. Cancer research, 1996. 56(5): p. 988- 994. 33. Mackenzie, P.I., et al., The UDP glycosyltransferase gene superfamily: recommended nomenclature update based on evolutionary divergence. Pharmacogenetics and Genomics, 1997. 7(4): p. 255-269. 34. Knights, K.M. and J.O. Miners, Renal UDP-glucuronosyltransferases and the glucuronidation of xenobiotics and endogenous mediators. Drug metabolism reviews, 2010. 42(1): p. 63-73. Page 132 of 140 35. Izukawa, T., et al., Quantitative analysis of UDP-glucuronosyltransferase (UGT) 1A and UGT2B expression levels in human livers. Drug Metabolism and Disposition, 2009. 37(8): p. 1759-1768. 36. Maruo, Y., et al., Polymorphism of UDP-glucuronosyltransferase and drug metabolism. Current drug metabolism, 2005. 6(2): p. 91-99. 37. Mackenzie, P.I., et al., Nomenclature update for the mammalian UDP glycosyltransferase (UGT) gene superfamily. Pharmacogenetics and genomics, 2005. 15(10): p. 677-685. 38. Burchell, B. and M.W. Coughtrie, UDP-glucuronosyltransferases. Pharmacology & therapeutics, 1989. 43(2): p. 261-289. 39. Ping Goon, C., et al., UGT1A1 Mediated Drug Interactions and its Clinical Relevance. Current drug metabolism, 2016. 17(2): p. 100-106. 40. Ethell, B.T., et al., Quantitative structure activity relationships for the glucuronidation of simple phenols by expressed human UGT1A6 and UGT1A9. Drug metabolism and disposition, 2002. 30(6): p. 734-738. 41. Coffman, B.L., et al., Human UGT2B7 catalyzes morphine glucuronidation. Drug Metabolism and Disposition, 1997. 25(1): p. 1-4. 42. Bélanger, A.-S., et al., Glucuronidation of the antiretroviral drug efavirenz by UGT2B7 and an in vitro investigation of drug-drug interaction with zidovudine. Drug Metabolism and Disposition, 2009. 37(9): p. 1793-1796. 43. Green, M.D., E.M. Oturu, and T.R. Tephly, Stable expression of a human liver UDP- glucuronosyltransferase (UGT2B15) with activity toward steroid and xenobiotic substrates. Drug Metabolism and Disposition, 1994. 22(5): p. 799-805. 44. Duan, S.X., et al., Stereoselective conjugation of oxazepam by human UDP- glucuronosyltransferases (UGTs): S-oxazepam is glucuronidated by UGT2B15, while R- oxazepam is glucuronidated by UGT2B7 and UGT1A9. Drug Metabolism and Disposition, 2002. 30(11): p. 1257-1265. 45. Weber, W.W. and D.W. Hein, N-acetylation pharmacogenetics. Pharmacological Reviews, 1985. 37(1): p. 25-79. 46. Blum, M., et al., Human arylamine N-acetyltransferase genes: isolation, chromosomal localization, and functional expression. DNA and cell biology, 1990. 9(3): p. 193-203. 47. Gross, M., et al., Distribution and concordance of N-acetyltransferase genotype and phenotype in an American population. Cancer Epidemiology and Prevention Biomarkers, 1999. 8(8): p. 683-692. 48. Jancova, P., P. Anzenbacher, and E. Anzenbacherova, Phase II drug metabolizing enzymes. Biomedical Papers, 2010. 154(2): p. 103-116. 49. McCarver, D.G. and R.N. Hines, The ontogeny of human drug-metabolizing enzymes: phase II conjugation enzymes and regulatory mechanisms. Journal of Pharmacology and Experimental Therapeutics, 2002. 300(2): p. 361-366. 50. Vasiliou, V., K. Vasiliou, and D.W. Nebert, Human ATP-binding cassette (ABC) transporter family. Human genomics, 2009. 3(3): p. 281. 51. Burckhardt, G. and B.C. Burckhardt, In vitro and in vivo evidence of the importance of organic anion transporters (OATs) in drug therapy, in Drug Transporters. 2011, Springer. p. 29-104. Page 133 of 140 52. Khamdang, S., et al., Interactions of human-and rat-organic anion transporters with pravastatin and cimetidine. Journal of pharmacological sciences, 2004. 94(2): p. 197- 202. 53. Yamada, A., et al., Multiple human isoforms of drug transporters contribute to the hepatic and renal transport of olmesartan, a selective antagonist of the angiotensin II AT1-receptor. Drug metabolism and disposition, 2007. 35(12): p. 2166-2176. 54. Yuan, H., et al., Renal organic anion transporter-mediated drug-drug interaction between gemcabene and quinapril. Journal of Pharmacology and Experimental Therapeutics, 2009. 330(1): p. 191-197. 55. Koepsell, H., K. Lips, and C. Volk, Polyspecific organic cation transporters: structure, function, physiological roles, and biopharmaceutical implications. Pharmaceutical research, 2007. 24(7): p. 1227-1251. 56. Nies, A.T., et al., Organic cation transporters (OCTs, MATEs), in vitro and in vivo evidence for the importance in drug therapy, in Drug Transporters. 2011, Springer. p. 105-167. 57. Schaeffeler, E., et al., DNA methylation is associated with downregulation of the organic cation transporter OCT1 (SLC22A1) in human hepatocellular carcinoma. Genome medicine, 2011. 3(12): p. 82. 58. GORBOULEV, V., et al., Cloning and characterization of two human polyspecific organic cation transporters. DNA and cell biology, 1997. 16(7): p. 871-881. 59. Konig, J., F. Muller, and M.F. Fromm, Transporters and drug-drug interactions: important determinants of drug disposition and effects. Pharmacol Rev, 2013. 65(3): p. 944-66. 60. Ciarimboli, G., Introduction to the Cellular Transport of Organic Cations, in Organic Cation Transporters. 2016, Springer. p. 1-47. 61. Kim, R., Organic anion-transporting polypeptide (OATP) transporter family and drug disposition. European journal of clinical investigation, 2003. 33(s2): p. 1-5. 62. Svoboda, M., et al., Organic anion transporting polypeptides (OATPs): regulation of expression and function. Current drug metabolism, 2011. 12(2): p. 139-153. 63. Bossuyt, X., M. Müller, and P.J. Meier, Multispecific amphipathic substrate transport by an organic anion transporter of human liver. Journal of hepatology, 1996. 25(5): p. 733- 738. 64. Niemi, M., Role of OATP transporters in the disposition of drugs. 2007. 65. Tamai, I., Oral drug delivery utilizing intestinal OATP transporters. Advanced drug delivery reviews, 2012. 64(6): p. 508-514. 66. König, J., Uptake transporters of the human OATP family, in Drug Transporters. 2011, Springer. p. 1-28. 67. Ho, R.H. and R.B. Kim, Transporters and drug therapy: implications for drug disposition and disease. Clin Pharmacol Ther, 2005. 78(3): p. 260-77. 68. Young, J.D., The SLC28 (CNT) and SLC29 (ENT) nucleoside transporter families: a 30-year collaborative odyssey. Biochem Soc Trans, 2016. 44(3): p. 869-76. 69. Young, J.D., et al., Human equilibrative nucleoside transporter (ENT) family of nucleoside and nucleobase transporter proteins. Xenobiotica, 2008. 38(7-8): p. 995-1021. 70. Baldwin, S.A., et al., The equilibrative nucleoside transporter family, SLC29. Pflugers Arch, 2004. 447(5): p. 735-43. Page 134 of 140 71. Barnes, K., et al., Distribution and functional characterization of equilibrative nucleoside transporter-4, a novel cardiac adenosine transporter activated at acidic pH. Circulation research, 2006. 99(5): p. 510-519. 72. Pastor-Anglada, M., et al., SLC28 genes and concentrative nucleoside transporter (CNT) proteins. Xenobiotica, 2008. 38(7-8): p. 972-94. 73. Ritzel, M.W., et al., Molecular identification and characterization of novel human and mouse concentrative Na+-nucleoside cotransporter proteins (hCNT3 and mCNT3) broadly selective for purine and pyrimidine nucleosides (system cib). Journal of Biological Chemistry, 2001. 276(4): p. 2914-2927. 74. Higgins, C.F., et al., A family of related ATP-binding subunits coupled to many distinct biological processes in bacteria. Nature, 1986. 323(6087): p. 448-450. 75. Mandal, A., et al., Transporter effects on cell permeability in drug delivery. Expert Opin Drug Deliv, 2017. 14(3): p. 385-401. 76. Cascorbi, I., P-glycoprotein: tissue distribution, substrates, and functional consequences of genetic variations, in Drug Transporters. 2011, Springer. p. 261-283. 77. Chan, L.M., S. Lowes, and B.H. Hirst, The ABCs of drug transport in intestine and liver: efflux proteins limiting drug absorption and bioavailability. European journal of pharmaceutical sciences, 2004. 21(1): p. 25-51. 78. Glaeser, H., Importance of P-glycoprotein for drug–drug interactions, in Drug Transporters. 2011, Springer. p. 285-297. 79. Deeley, R.G., C. Westlake, and S.P. Cole, Transmembrane transport of endo-and xenobiotics by mammalian ATP-binding cassette multidrug resistance proteins. Physiological reviews, 2006. 86(3): p. 849-899. 80. Kruh, G.D. and M.G. Belinsky, The MRP family of drug efflux pumps. Oncogene, 2003. 22(47): p. 7537. 81. Rosenberg, M.F., et al., The structure of the multidrug resistance protein 1 (MRP1/ABCC1) crystallization and single-particle analysis. Journal of Biological Chemistry, 2001. 276(19): p. 16076-16082. 82. Keppler, D., Multidrug resistance proteins (MRPs, ABCCs): importance for pathophysiology and drug therapy, in Drug Transporters. 2011, Springer. p. 299-323. 83. Knauer, M.J., et al., Human skeletal muscle drug transporters determine local exposure and toxicity of statins. Circulation research, 2010. 106(2): p. 297-306. 84. Williams, G.C., et al., Direct evidence that saquinavir is transported by multidrug resistance-associated protein (MRP1) and canalicular multispecific organic anion transporter (MRP2). Antimicrobial agents and chemotherapy, 2002. 46(11): p. 3456- 3462. 85. König, J., et al., MRP2, the apical export pump for anionic conjugates, in ABC proteins: from bacteria to man. 2003, Elsevier. p. 423-443. 86. Zelcer, N., et al., Evidence for two interacting ligand binding sites in human multidrug resistance protein 2 (ATP binding cassette C2). Journal of Biological Chemistry, 2003. 278(26): p. 23538-23544. 87. Jedlitschky, G., U. Hoffmann, and H.K. Kroemer, Structure and function of the MRP2 (ABCC2) protein and its role in drug disposition. Expert Opin Drug Metab Toxicol, 2006. 2(3): p. 351-66. Page 135 of 140 88. Nies, A.T. and D. Keppler, The apical conjugate efflux pump ABCC2 (MRP2). Pflugers Arch, 2007. 453(5): p. 643-59. 89. Belinsky, M. and G. Kruh, MOAT-E (ARA) is a full-length MRP/cMOAT subfamily transporter expressed in kidney and liver. British journal of cancer, 1999. 80(9): p. 1342. 90. Russel, F.G., J.B. Koenderink, and R. Masereeuw, Multidrug resistance protein 4 (MRP4/ABCC4): a versatile efflux transporter for drugs and signalling molecules. Trends in pharmacological sciences, 2008. 29(4): p. 200-207. 91. Borst, P., C. de Wolf, and K. van de Wetering, Multidrug resistance-associated proteins 3, 4, and 5. Pflugers Arch, 2007. 453(5): p. 661-73. 92. Ritter, C.A., et al., Cellular export of drugs and signaling molecules by the ATP-binding cassette transporters MRP4 (ABCC4) and MRP5 (ABCC5). Drug metabolism reviews, 2005. 37(1): p. 253-278. 93. Doyle, L.A., et al., A multidrug resistance transporter from human MCF-7 breast cancer cells. Proceedings of the National Academy of Sciences, 1998. 95(26): p. 15665-15670. 94. Zu Schwabedissen, H.E.M. and H.K. Kroemer, In vitro and in vivo evidence for the importance of breast cancer resistance protein transporters (BCRP/MXR/ABCP/ABCG2), in Drug Transporters. 2011, Springer. p. 325-371. 95. Ross, D.D., et al., Atypical multidrug resistance: breast cancer resistance protein messenger RNA expression in mitoxantrone-selected cell lines. Journal of the National Cancer Institute, 1999. 91(5): p. 429-433. 96. Sinz, M., G. Wallace, and J. Sahi, Current industrial practices in assessing CYP450 enzyme induction: preclinical and clinical. AAPS J, 2008. 10(2): p. 391-400. 97. Willson, T.M. and S.A. Kliewer, PXR, CAR and drug metabolism. Nat Rev Drug Discov, 2002. 1(4): p. 259-66. 98. Chen, Y., et al., Nuclear receptors in the multidrug resistance through the regulation of drug-metabolizing enzymes and drug transporters. Biochem Pharmacol, 2012. 83(8): p. 1112-26. 99. Tolson, A.H. and H. Wang, Regulation of drug-metabolizing enzymes by xenobiotic receptors: PXR and CAR. Adv Drug Deliv Rev, 2010. 62(13): p. 1238-49. 100. Fuhr, U., Induction of drug metabolising enzymes. Clinical pharmacokinetics, 2000. 38(6): p. 493-504. 101. Kohle, C. and K.W. Bock, Coordinate regulation of human drug-metabolizing enzymes, and conjugate transporters by the Ah receptor, pregnane X receptor and constitutive androstane receptor. Biochem Pharmacol, 2009. 77(4): p. 689-99. 102. Shen, G. and A.N. Kong, Nrf2 plays an important role in coordinated regulation of Phase II drug metabolism enzymes and Phase III drug transporters. Biopharm Drug Dispos, 2009. 30(7): p. 345-55. 103. Maglich JM, S.C., Goodwin B, Hawkins-Brown D, Moore JT, Kliewer SA., Nuclear pregnane X receptor and constitutive androstane receptor regulate overlapping but distinct sets of genes involved in xenobiotic detoxification. Mol Pharmacol, 2002(62): p. 638-646. 104. Patel, R.D., et al., Aryl-hydrocarbon receptor activation regulates constitutive androstane receptor levels in murine and human liver. Hepatology, 2007. 46(1): p. 209-218. Page 136 of 140 105. Sissung, T.M., et al., Pharmacogenetics of membrane transporters: an update on current approaches. Molecular biotechnology, 2010. 44(2): p. 152-167. 106. Hines, R.N., Developmental expression of drug metabolizing enzymes: impact on disposition in neonates and young children. Int J Pharm, 2013. 452(1-2): p. 3-7. 107. Jeong, H., Altered drug metabolism during pregnancy: hormonal regulation of drug- metabolizing enzymes. Expert Opin Drug Metab Toxicol, 2010. 6(6): p. 689-99. 108. Hu, Y., et al., Gender differences of brain glucose metabolic networks revealed by FDG- PET: evidence from a large cohort of 400 young adults. PLoS One, 2013. 8(12): p. e83821. 109. Klotz, U., Pharmacokinetics and drug metabolism in the elderly. Drug Metab Rev, 2009. 41(2): p. 67-76. 110. Kinirons, M.T. and M.S. O'Mahony, Drug metabolism and ageing. Br J Clin Pharmacol, 2004. 57(5): p. 540-4. 111. Zevin, S. and N.L. Benowitz, Drug interactions with tobacco smoking. Clinical pharmacokinetics, 1999. 36(6): p. 425-438. 112. Yin, T. and T. Miyata, Warfarin dose and the pharmacogenomics of CYP2C9 and VKORC1—rationale and perspectives. Thrombosis research, 2007. 120(1): p. 1-10. 113. Kis, O., et al., The complexities of antiretroviral drug–drug interactions: role of ABC and SLC transporters. Trends in pharmacological sciences, 2010. 31(1): p. 22-35. 114. Chambliss, A.B. and D.W. Chan, Precision medicine: from pharmacogenomics to pharmacoproteomics. Clin Proteomics, 2016. 13: p. 25. 115. Empey, P.E., Pharmacogenomics to achieve precision medicine. Am J Health Syst Pharm, 2016. 73(23): p. 1906-1907. 116. Pirmohamed, M., Personalized pharmacogenomics: predicting efficacy and adverse drug reactions. Annu Rev Genomics Hum Genet, 2014. 15: p. 349-70. 117. Opdam, F.L., H. Gelderblom, and H.J. Guchelaar, Phenotyping drug disposition in oncology. Cancer Treat Rev, 2012. 38(6): p. 715-25. 118. de Wildt, S.N., S. Ito, and G. Koren, Challenges for drug studies in children: CYP3A phenotyping as example. Drug Discov Today, 2009. 14(1-2): p. 6-15. 119. Anderson, B.J. and N.H. Holford, Mechanism-based concepts of size and maturity in pharmacokinetics. Annu Rev Pharmacol Toxicol, 2008. 48: p. 303-32. 120. Kaushal, R., et al., Medication errors and adverse drug events in pediatric inpatients. Jama, 2001. 285(16): p. 2114-2120. 121. Prasad, P.A., et al., Pediatric antifungal utilization: new drugs, new trends. The Pediatric infectious disease journal, 2008. 27(12): p. 1083-1088. 122. Smith, J., et al., Voriconazole therapeutic drug monitoring. Antimicrob Agents Chemother, 2006. 50(4): p. 1570-2. 123. Brüggemann, R.J., et al., Therapeutic drug monitoring of voriconazole. Therapeutic drug monitoring, 2008. 30(4): p. 403-411. 124. Yamada, T., et al., Saturated metabolism of voriconazole N-oxidation resulting in nonlinearity of pharmacokinetics of voriconazole at clinical doses. Biological and Pharmaceutical Bulletin, 2015. 38(10): p. 1496-1503. 125. Neely MN, F.X., Lu S, Margol A, Louie SG, Hope WW. and Voriconazole pharmacokinetics and metabolism to voriconazole-N-oxide in children is age Page 137 of 140 independent with more aggressive dosing, abstr A-723., in 54th Intersci Conf Antimicrob Agents Chemother. 2014.: American Society for Microbiology, Washington, DC. 126. Neely, M., et al., Achieving target voriconazole concentrations more accurately in children and adolescents. Antimicrob Agents Chemother, 2015. 59(6): p. 3090-7. 127. FDA. VFEND® (voriconazole) Label. 2017 Jul.28, 2107 [cited 2017 Sep.7]; Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/021266s043,021267s054, 021630s032lbl.pdf. 128. Yanni, S.B., et al., Role of flavin-containing monooxygenase in oxidative metabolism of voriconazole by human liver microsomes. Drug metabolism and disposition, 2008. 36(6): p. 1119-1125. 129. Denning, D.W., et al., Efficacy and safety of voriconazole in the treatment of acute invasive aspergillosis. Clinical infectious diseases, 2002. 34(5): p. 563-571. 130. Spaggiari, D., et al., A cocktail approach for assessing the in vitro activity of human cytochrome P450s: an overview of current methodologies. J Pharm Biomed Anal, 2014. 101: p. 221-37. 131. Karlsson, M.O., I. Lutsar, and P.A. Milligan, Population pharmacokinetic analysis of voriconazole plasma concentration data from pediatric studies. Antimicrobial agents and chemotherapy, 2009. 53(3): p. 935-944. 132. Hyland, R., B. Jones, and D. Smith, Identification of the cytochrome P450 enzymes involved in the N-oxidation of voriconazole. Drug Metabolism and Disposition, 2003. 31(5): p. 540-547. 133. Lamoureux, F., et al., Impact of CYP2C19 genetic polymorphisms on voriconazole dosing and exposure in adult patients with invasive fungal infections. Int J Antimicrob Agents, 2016. 47(2): p. 124-31. 134. Neely, M.N., et al., Accurate detection of outliers and subpopulations with Pmetrics, a nonparametric and parametric pharmacometric modeling and simulation package for R. Ther Drug Monit, 2012. 34(4): p. 467-76. 135. Zane, N.R. and D.R. Thakker, A physiologically based pharmacokinetic model for voriconazole disposition predicts intestinal first-pass metabolism in children. Clinical pharmacokinetics, 2014. 53(12): p. 1171-1182. 136. Koukouritaki, S.B., et al., Human hepatic flavin-containing monooxygenases 1 (FMO1) and 3 (FMO3) developmental expression. Pediatric Research, 2002. 51(2): p. 236-243. 137. Falls, J.G., et al., Regulation of mouse liver flavin-containing monooxygenases 1 and 3 by sex steroids. Archives of biochemistry and biophysics, 1997. 342(2): p. 212-223. 138. Yanni, S.B., et al., In vitro hepatic metabolism explains higher clearance of voriconazole in children versus adults: role of CYP2C19 and flavin-containing monooxygenase 3. Drug Metab Dispos, 2010. 38(1): p. 25-31. 139. Soldin, O.P., S.H. Chung, and D.R. Mattison, Sex differences in drug disposition. J Biomed Biotechnol, 2011. 2011: p. 187103. 140. Spoletini, I., et al., Sex Differences in Drug Effects: Interaction with Sex Hormones in Adult Life, in Sex and Gender Differences in Pharmacology, V. Regitz-Zagrosek, Editor. 2012, Springer Berlin Heidelberg: Berlin, Heidelberg. p. 91-105. Page 138 of 140 141. Moriuchi H, M.M., Arthos J, Hoxie J, Fauci AS., Promonocytic U937 subclones expressing CD4 and CXCR4 are resistant to infection with and cell-to-cell fusion by T-cell-tropic human immunodeficiency virus type 1. J Virol., 1997. 71(12): p. 8. 142. Arts, E.J. and D.J. Hazuda, HIV-1 antiretroviral drug therapy. Cold Spring Harb Perspect Med, 2012. 2(4): p. a007161. 143. Naif, H.M., Pathogenesis of HIV Infection. Infect Dis Rep, 2013. 5(Suppl 1): p. e6. 144. UNAIDS, <AIDS-by-the-numbers-2016_en.pdf>. 145. Le Tiec C, B.A., Goujard C, Taburet AM., Clinical pharmacokinetics and summary of efficacy and tolerability of atazanavir. Clin Pharmacokinet., 2005. 44(10): p. 16. 146. Wood, R., Atazanavir: its role in HIV treatment. Expert Rev Anti Infect Ther, 2008. 6(6): p. 785-96. 147. Janneh, O., et al., P-glycoprotein, multidrug resistance-associated proteins and human organic anion transporting polypeptide influence the intracellular accumulation of atazanavir. Antivir Ther, 2009. 14(7): p. 965-74. 148. Lucia MB, G.C., Rutella S, Rastrelli E, Savarino A, Cauda R, Atazanavir inhibits P- glycoprotein and multidrug resistance-associated protein efflux activity. J Acquir Immune Defic Syndr, 2005. 39(5): p. 2. 149. Deeks, E.D., Darunavir: a review of its use in the management of HIV-1 infection. Drugs, 2014. 74(1): p. 99-125. 150. Phung, B.C. and P. Yeni, Darunavir: an effective protease inhibitor for HIV-infected patients. Expert Rev Anti Infect Ther, 2011. 9(6): p. 631-43. 151. Back D, Sekar V, and H. RM., Darunavir pharmacokinetics and drug interactions. Antivir Ther. , 2008. 13(1): p. 13. 152. Khoo, S.H., et al., Intracellular Accumulation of Human Immunodeficiency Virus Protease Inhibitors. Antimicrobial Agents and Chemotherapy, 2002. 46(10): p. 3228-3235. 153. MacManus S, et al., GW433908 ritonavir once daily in antiretroviral therapy-naive hiv- infected patients absence of protease resistance at 48 weeks. AIDS, 2004. 18(4): p. 5. 154. Crutchley, R.D., R.C. Guduru, and A.M. Cheng, Evaluating the role of atazanavir/cobicistat and darunavir/cobicistat fixed-dose combinations for the treatment of HIV-1 infection. HIV AIDS (Auckl), 2016. 8: p. 47-65. 155. Renjifo B, v.W.J., Salem AH, Bow D, Ng J, Norton M, Pharmacokinetic enhancement in HIV antiretroviral therapy: a comparison of ritonavir and cobicistat. AIDS REV, 2015. 17(37): p. 10. 156. Nathan, B., et al., Cobicistat: a Novel Pharmacoenhancer for Co-Formulation with HIV Protease and Integrase Inhibitors. Infect Dis Ther, 2013. 2(2): p. 111-22. 157. Deeks, E.D., Cobicistat: a review of its use as a pharmacokinetic enhancer of atazanavir and darunavir in patients with HIV-1 infection. Drugs, 2014. 74(2): p. 195-206. 158. Clotet, B., et al., Once-daily dolutegravir versus darunavir plus ritonavir in antiretroviral- naive adults with HIV-1 infection (FLAMINGO): 48 week results from the randomised open-label phase 3b study. The Lancet, 2014. 383(9936): p. 2222-2231. 159. Tashima K, C.G., Tomaka FL, Kakuda TN, Brochot A, Van de Casteele T, Opsomer M, Garner W, Margot N, Custodio JM, Fordyce MW, Szwarcberg J., Cobicistat-boosted darunavir in HIV-1-infected adults week 48 results of a Phase IIIb, open-label single-arm trial. AIDS Research and Therapy, 2014. 11. Page 139 of 140 160. Gallant, J.E., et al., Cobicistat versus ritonavir as a pharmacoenhancer of atazanavir plus emtricitabine/tenofovir disoproxil fumarate in treatment-naive HIV type 1-infected patients: week 48 results. J Infect Dis, 2013. 208(1): p. 32-9. 161. De Meyer, S., et al., TMC114, a novel human immunodeficiency virus type 1 protease inhibitor active against protease inhibitor-resistant viruses, including a broad range of clinical isolates. Antimicrob Agents Chemother, 2005. 49(6): p. 2314-21. 162. Robinson BS, et al., BMS-232632, a highly potent human immunodeficiency virus protease inhibitor that can be used in combination with other available antiretroviral agents. Antimicrob Agents Chemother., 2000. 44(8): p. 7. 163. Boffito, M., et al., Pharmacokinetics of once-daily darunavir-ritonavir and atazanavir- ritonavir over 72 hours following drug cessation. Antimicrob Agents Chemother, 2011. 55(9): p. 4218-23. 164. Sekar, V., et al., The effect of different meal types on the pharmacokinetics of darunavir (TMC114)/ritonavir in HIV-negative healthy volunteers. J Clin Pharmacol, 2007. 47(4): p. 479-84. 165. Benet, L.Z., C.L. Cummins, and C.Y. Wu, Unmasking the dynamic interplay between efflux transporters and metabolic enzymes. Int J Pharm, 2004. 277(1-2): p. 3-9. 166. International Transporter, C., et al., Membrane transporters in drug development. Nat Rev Drug Discov, 2010. 9(3): p. 215-36. 167. Lepist, E.I., et al., Cobicistat boosts the intestinal absorption of transport substrates, including HIV protease inhibitors and GS-7340, in vitro. Antimicrob Agents Chemother, 2012. 56(10): p. 5409-13. 168. Xu, L., et al., Cobicistat (GS-9350): A Potent and Selective Inhibitor of Human CYP3A as a Novel Pharmacoenhancer. ACS Med Chem Lett, 2010. 1(5): p. 209-13. 169. Alam, C., et al., Role and modulation of drug transporters in HIV-1 therapy. Adv Drug Deliv Rev, 2016. 103: p. 121-43. 170. Mitema D and A. MG., The role of organic transporters in pharmacokinetics and nephrotoxicity of newer antiviral therapies for HIV and hepatitis C. Curr Drug Metab, 2015. 16(4): p. 10. 171. Ray, A.S., et al., Lack of a metabolic and antiviral drug interaction between tenofovir, abacavir and lamivudine. Antivir Ther, 2005. 10(3): p. 451-457. 172. Eagling, V.A., D. Back, and M. Barry, Differential inhibition of cytochrome P450 isoforms by the protease inhibitors, ritonavir, saquinavir and indinavir. British journal of clinical pharmacology, 1997. 44(2): p. 190-194. 173. Jones K, et al., Differences in the intracellular accumulation of HIV protease inhibitros in vitro and the effect of active transport. AIDS 2001. 14(5): p. 7. 174. Marzolini, C., et al., Cobicistat versus ritonavir boosting and differences in the drug-drug interaction profiles with co-medications. J Antimicrob Chemother, 2016. 71(7): p. 1755- 8. 175. Minuesa, G., et al., Drug uptake transporters in antiretroviral therapy. Pharmacol Ther, 2011. 132(3): p. 268-79. 176. Benet, L.Z., The Drug Transporter-Metabolism Alliance: Uncovering and defining the interplay. MOLECULAR PHARMACEUTICS, 2009. 6(6): p. 13. Page 140 of 140 177. Janneh, O., et al., Inhibition of P-glycoprotein and multidrug resistance-associated proteins modulates the intracellular concentration of lopinavir in cultured CD4 T cells and primary human lymphocytes. J Antimicrob Chemother, 2007. 60(5): p. 987-93. 178. Bousquet, L., et al., Comparison of ABC transporter modulation by atazanavir in lymphocytes and human brain endothelial cells: ABC transporters are involved in the atazanavir-limited passage across an in vitro human model of the blood-brain barrier. AIDS Res Hum Retroviruses, 2008. 24(9): p. 1147-54. 179. Olson DP, Taylor BJ, and I. SP., Detection of MRP functional activity: calcein AM but not BCECF AM as a Multidrug Resistance-related Protein (MRP1) substrate. Cytometry, 2001. 46(2): p. 9. 180. Allen, J.D., et al., Potent and Specific Inhibition of the Breast Cancer Resistance Protein Multidrug Transporter in Vitro and in Mouse Intestine by a Novel Analogue of Fumitremorgin C 1 This work was supported in part by grant NKI 97-1433 from the Dutch Cancer Society (to AHS). Synthesis investigations by A. v. L. and GJ. K. were supported by the Netherlands Research Council for Chemical Sciences (NWO/CW) and the Netherlands Technology Foundation (STW). 1. Molecular cancer therapeutics, 2002. 1(6): p. 417-425. 181. Stevenson, M., HIV persistence in macrophages. Nature medicine, 2017. 23(5): p. 538. 182. Sharova, N., et al., Macrophages archive HIV-1 virions for dissemination in trans. The EMBO journal, 2005. 24(13): p. 2481-2489.
Abstract (if available)
Abstract
Background: Drug disposition describes the processes of drug absorption, distribution, metabolism, and elimination, collectively known as ADME. Drug disposition in the target tissues determines the overall drug exposure, and thus treatment efficacy. Essential factors in these processes include drug metabolizing enzymes (DMEs), membrane transporters, as well as protein binding, patient gender, and patient age. Clinical practices require the balance of efficacy and safety, especially among the pediatric population. Understanding the pharmacological mechanisms improves the clinical impact of drug therapies, by giving insight on how to improve and optimize treatment efficacy and safety. In this dissertation, I studied the age-related impacts of DMEs on the pharmacokinetics (PK) of the drug voriconazole and the roles of transporters in anti-HIV combination therapies. ❧ Aims: 1) To understand the ontogeny of Cytochrome P 450 (CYP) 2C19 (CYP2C19), CYP3A4/5, and flavin-containing monooxygenase 3 (FMO3) in relation to age-related voriconazole PK. 2) To evaluate CD4+ cellular disposition of HIV protease inhibitors (PIs) atazanavir (ATV) and darunavir (DRV) when treated alone or in combination with pharmacoenhancers (PEs). 3) To assess the impacts of PI/PE treatments on efflux transporters expression and function in CD4+ cells. ❧ Results: For the first aim, In accordance with my initial hypothesis, CYP2C19, FMO3, and CYP3A4/5 activities decline with increasing age, consistent with age-related voriconazole metabolism. Unlike in adult patients, both FMO3 and CYP2C19 exhibit a significant correlation between age and voriconazole metabolism in pediatric patients. ❧ With regards to the specific aims 2 and 3, I found that cellular concentrations of atazanavir (ATV) is more responsive to PE treatments when compared to darunavir (DRV). The cellular exposure AUC of ATV is about two times that of DRV. I also found that cobicistat (COBI) is inferior to ritonavir (RTV) in prolonging cellular retention when co-administer with a PI. COBI is a weaker inhibitor of efflux transporters when compared to RTV. Moreover, ATV itself is an inhibitor of efflux transporter when compared to DRV. Also, CD4+ cell types impact the treatment efficacy, as lymphocytes are more responsive to HIV treatments than HIV infected monocytes. These findings are consistent with current data suggesting that monocytes may be a site of persistent HIV infections, due to its ability to eliminate antiretroviral (ARV) agents. ❧ Conclusion: Evaluating DME activities among pediatric patients can optimize drug treatment, especially for narrow therapeutic window drugs like voriconazole. The DMEs FMO3 and CYP2C19 have age-sensitive expression, and thus explain the extra dosage difference in voriconazole between children and adults. ❧ DRV and ATV cellular retention were different when given alone. This may be attributed to ATV’s ability to inhibit various efflux transporters. When DRV or ATV was combined with either COBI or RTV, RTV was a more potent efflux transporter inhibitor than COBI. This may explain the higher incidence of clinical virologic failure of DRV versus ATV when used in combination with either RTV or COBI containing treatments, particularly in HIV infected CD4+ monocytic/macrophagic cells.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Pharmacodynamics impact of etravirine disposition into breast milk in relation to HIV replication in this milk
PDF
Factors impacting drug disposition and clinical outcomes: age, hepatic metabolism, renal elimination and pharmacogentetics
PDF
Pharmacokinetic and pharmacodynamic optimization of CFTR modulator therapy to mitigate potential drug interactions and adverse events in people with cystic fibrosis
PDF
Renin-angiotensin system modulation for the prevention and treatment of metabolic dysfunction
PDF
The role of altered folate one-carbon metabolism (FOCM) in the development of colorectal cancer
PDF
The preclinical evaluation of moxidectin as a platform for drug development for alcohol use disorder
PDF
Intracellular drug-drug interaction between nucleoside analogs leads to early virologic failure in HIV patients receiving triple nucleoside combinations of tenofovir, lamivudine and abacavir or d...
PDF
Inhibition of tumor cell growth by mefloquine via multimechanistic effects involving increased cellular stress, inhibition of autophagy, and impairment of cellular energy metabolism
PDF
Optimization of nanomedicine based drug delivery systems for the treatment of solid tumors
PDF
Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs
PDF
Characterization of the transferrin oligomer and its potential application in drug delivery
PDF
Therapeutic potential of rhesus theta (θ) defensin-1 in cystic fibrosis airway infection and inflammation
PDF
Characterization and identification of endogenous factor(s) that enhance insulin transport across primary rat alveolar epithelial cell monolayers
PDF
Averting dementia: renin-angiotensin system and angiogenic cells in cognitive decline
PDF
Investigating sodium butyrate as a potential treatment for alcohol liver disease through the gut-liver axis
PDF
Engineering chimeric antigen receptor (CAR) -modified T cells for enhanced cancer immunotherapy
PDF
Protein ADP-ribosylation: from biochemical characterization to therapeutic applications
PDF
Investigation of in vitro activity and in vivo safety of imipenem-relebactam alone and in combination with other anti-microbial agents against clinical isolates of M. abscessus from CF patients
PDF
Therapeutic potential of Rhesus theta defensin-1 for the treatment of COVID-19 pneumonia
PDF
Modulation of immune response in Systemic Lupus Erythematosus by synthetic lipoxin analogue, NAP1051
Asset Metadata
Creator
Liu, Siyu
(author)
Core Title
Factors that may impact drug disposition and metabolism
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Clinical and Experimental Therapeutics
Publication Date
07/30/2020
Defense Date
06/15/2018
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
drug disposition,drug metabolism,drug metabolizing enzymes,OAI-PMH Harvest,transporters
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Louie, Stan (
committee chair
), Davies, Daryl (
committee member
), Okamoto, Curtis (
committee member
), Shen, Wei-Chiang (
committee member
)
Creator Email
lindsay.liusiyu@gmail.com,siyu.lindsay.liu@outlook.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-42708
Unique identifier
UC11671916
Identifier
etd-LiuSiyu-6570.pdf (filename),usctheses-c89-42708 (legacy record id)
Legacy Identifier
etd-LiuSiyu-6570.pdf
Dmrecord
42708
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Liu, Siyu
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
drug disposition
drug metabolism
drug metabolizing enzymes
transporters