• Application of radiotelemetry in pharmacokinetic-pharmacodynamic modeling of procainamide and N-acetylprocainamide in sedentary and exercise trained rats

      Kharidia, Jahnavi Sanjay; Eddington, Natalie D. (1996)
      Understanding the pharmacokinetics (PK) and pharmacodynamics (PD) of a new drug in an appropriate animal model provides a scientific framework for efficient and rational drug development. Cardiovascular measurements from small animals are difficult or impossible to obtain. The conventional methods, chemical restraint, physical restraint, cuff blood pressure and tethering used to obtain cardiovascular PD have limitations. These methods affect physiological function of animals and introduce stress artifact into data. Recently, radiotelemetry has been proposed as a new method to obtain cardiovascular PD from small animals. The objective of this research was to investigate the application of radiotelemetry in the pharmacokinetic/pharmacodynamic (PK/PD) modeling of cardiovascular agents in small animals. PD of various cardioactive agents such as procainamide (PA) and its active metabolite, N-acetylaprocainamide (NAPA) were evaluated. Additionally, one of the objectives of this dissertation was to evaluate the effects of exercise training on the pharmacokinetics of a model substrate of N-acetylation reactions (phase II enzymes). PA was chosen as a model drug for these studies. Radiotelemetry avoided the stress associated with animal handling and the use of anesthesia during PD studies, which allowed for the attainment of higher quality and more accurate data. PA displayed a significant increase in QT, QR and QS intervals from the baseline. The QT prolongation has been used as a surrogate marker in PA clinical studies. The PK of PA and NAPA were best described by a one compartment model. An Emax model was able to accurately describe the relationship between PA concentration and the QT interval. Effects of exercise training on the PK as well as PD of PA and NAPA were evaluated using radiotelemetry. Rats were exercised for four weeks by treadmill running. These studies noted that four weeks of exercise training did not alter the PK or PD of PA and NAPA suggesting that exercise training may not have significantly affected N-acetyltransferase. In conclusion, radiotelemetry allows the prediction of the PK/PD relationship observed in clinical studies by conducting such studies in small animals.
    • Biopharmaceutic and pharmacokinetic properties that influence the development and evaluation of the product quality performance tool, in vitro in vivo correlation, and the product performance

      Sirisuth, Nattee; Eddington, Natalie D. (2000)
      In vitro in vivo correlation (IVIVC) is a mathematical model describing the relationship between in vitro drug release and its in vivo drug absorption. The benefit of establishing this correlation is to enable the dissolution to serve as a surrogate of the in vivo drug behavior. This should minimize the number of bioavailability studies performed during the initial approval process, scale-up and postapproval change. IVIVC development may be significantly influenced by the biopharmaceutic and pharmacokinetic properties of a drug since they are the determinants of in vitro drug dissolution and the in vivo drug absorption. The objectives of this dissertation research were to: (1) investigate the influence of biopharmaceutic and pharmacokinetic properties (i.e. biopharmaceutical classification system, stereoisomerisms, first pass effect, inter- and intra-subject variability) on the IVIVC development and validation for metoprolol extended release (ER) product, (2) develop and evaluate a deconvolution-based IVIVC for a diltiazem ER product, and (3) elucidate the effect of variability associated with the pharmacokinetic of the drug products on the in vitro and in vivo release of metoprolol and diltiazem ER products. Dissolution studies were performed for metoprolol (100 mg) and diltiazem (90 mg) ER bio-batch using various dissolution testing conditions (USP Apparatus I, II, and III). Two clinical studies on both products were conducted in a randomized single dose, fasting, and crossover study design. Metoprolol studies were conducted in normal healthy extensive metabolizers, whereas subjects participating on diltiazem studies were randomly selected from the population. The study of stereoselectivity on the IVIVC development for metoprolol demonstrated an inability of using racemate data to predict R-enantiomer concentrations, while they were predictive of the active S-stereoisomer. Enantiomer specific IVIVCs were predictive of their in vivo performance. In the evaluation of first pass metabolism with metoprolol and its metabolites, the developed IVIVC was not able to directly predict the bioavailability of metoprolol, suggesting that first pass metabolism will play a role in IVIVC development. Additionally, deconvolution-based IVIVC for diltiazem was not predictive due to an extensive pharmacokinetic variation in its first pass elimination. In addition, inter- and intra-subject variability was found to minimize the predictive performance of the average and individual IVIVCs for S-metoprolol and diltiazem.
    • Characterization of pharmacokinetic and pharmacodynamic drug interactions of 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy)

      Upreti, Vijay V.; Eddington, Natalie D. (2007)
      Use of the illicit recreational drug 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy) is one of the fastest growing drug abuse problems. MDMA has serious and potentially fatal toxic effects with increasing numbers of MDMA abuse related emergency room visits and deaths. MDMA undergoes extensive multiple cytochrome P450 (CYP) isoform-mediated metabolic elimination that generates active metabolites. Moreover, MDMA has a narrow recreational window with little margin between recreational and toxic blood levels. Hence drug-drug interactions could be a contributing factor to MDMA toxicity. Users often co-administer fluoxetine, an antidepressant and potent inhibitor of CYP2D6 with MDMA, in order to alleviate post-use depression. We investigated the possible pharmacokinetic drug interaction between MDMA and fluoxetine. Co-administration of MDMA with fluoxetine resulted in a significant increase in the exposure (1.5 fold) and prolonged the elimination half-life (2-4 fold) of MDMA and its active metabolite 3,4-methylenedioxyamphetamine (MDA) in the blood and the brain of rats. The persistent-elevated levels of MDMA and MDA suggested an enhanced risk of MDMA toxic effects upon combining MDMA with fluoxetine. However, we observed an attenuation of the MDMA pharmacodynamic effects (adrenergic-noradrenergic stimulation responsible for cardiovascular and temperature toxic effects of MDMA) upon co-administration with fluoxetine in rats. The observed paradox was due to an overwhelming antagonism of MDMA effects at its molecular site of action by fluoxetine. Over recent years, membrane transporters have been realized as important determinants of pharmacokinetics, drug response and drug interactions. However, little is known about interaction of MDMA with membrane transporters. We evaluated substrate activity of MDMA for the drug efflux transporter P-glycoprotein and the drug influx organic cation transporters using various in vitro (bidirectional transport and uptake studies) and in vivo (knock out mice) models. MDMA did not display substrate activity for these evaluated transporters. Furthermore, transport studies indicated that MDMA is a highly permeable agent and hence drug transporters will not play a physiologically relevant role in determining drug interactions of MDMA. Liver damage is one of the most serious and life threatening tissue-specific toxic effect of MDMA. However, the underlying mechanism is not known. Our mechanistic studies using a proteomics approach revealed oxidative modification and inactivation of mitochondrial proteins involved in energy metabolism as a mechanism for MDMA-mediated liver damage. In addition, these studies revealed that MDMA inactivated mitochondrial aldehyde dehydrogenase, an enzyme that catalyzes the elimination of acetaldehyde, a toxic metabolite of ethanol. Upon co-administration of ethanol with MDMA, a trend towards an increase in blood acetaldehyde levels was observed in rats, which suggests a clinically relevant and potentially toxic drug interaction between MDMA and ethanol.
    • Characterization of the pharmacokinetics and pharmacodynamics of the stereoisomers of mivacurium in humans and dogs

      Lugo, Sonia I.; Eddington, Natalie D. (1995)
      The issue of stereoisomers in drug development and therapy is becoming increasingly important as a more complete understanding of differences between drug isomers is developed with respect to pharmacokinetics and pharmacodynamics. Mivacurium chloride is a short-acting, non-depolarizing neuromuscular blocking agent. Mivacurium is a mixture of three stereoisomers: the trans-trans, the cis-trans, and the cis-cis diester. The pharmacokinetics and pharmacodynamics of these isomers were investigated in humans and dogs to develop a mathematical model. This model may be used in a closed-loop system to optimize drug administration and patient monitoring and, henceforth, to improve clinical outcomes. A bolus dose (0.01-0.02 mg/kg) and two infusion rates (1.0-1.5 {dollar}\mu{dollar}g/kg/min) of mivacurium were administered to four beagle dogs. Dogs were anesthetized with 5% isofluorane. NMB was assessed by train-of four (TOF) and single twitch stimulation. Each dog received two treatments with a washout period of ten days. Blood samples were collected up to 6 min after the bolus dose, for 4 min after the administration of each infusion and during recovery. Eleven subjects agreed to participate in a pharmacodynamics study. Patients received a bolus dose (0.10-0.20 mg/kg) and sequential infusions of mivacurium as needed to maintain twitch height (TH) 10% baseline. NMB was assessed by TH depression after TOF stimulation. Another group of five subjects, agreed to participate in a pharmacokinetics-pharmacodynamics study. The dosing schedule and monitoring ofthe paralysis was as previously discussed. Blood samples were collected up to 10 min after the bolus dose, and up to 10 min after the administration of the last infusion. Samples were analyzed using a stereospecific KPLC-fluorescence method. Modeling of the data was accomplished using ADAPT II. The pharmacokinetics of the two major isomers of mivacurium in dogs were best described by a two-compartment model. TOF was a more sensitive measure of onset and offset of NMB than TH. Onset of action showed dose-dependence. The pharmacokinetics of isomers of mivacurium were best described by a two-compartment model in three of the five patients. The model developed with the first eleven patients was predictive of the concentration-effect (CE) relationship. The CE relationship was best described by a sigmoid E{dollar}\rm\sb{lcub}max{rcub}{dollar} model in humans and dogs. The value of pharmacokinetics-pharmacodynamics models should be to allow better insight into the principles governing the time course of drug action and to get a better predictive estimate of the patients response to dosing and to improve drug therapy outcomes.
    • The drug delivery factors that influence the tissue distribution and pharmacokinetics of a series of active antiepileptic agents, the enaminones

      Cox, Donna S.; Eddington, Natalie D. (2000)
      Early Central Nervous System (CNS) discovery generated new pharmaceuticals based on two properties of the compound, lipophilicity and molecular size. Almost all CNS pharmaceuticals presently in clinical trials possess both of these criteria and are capable of crossing the blood-brain barrier (BBB). Currently, modern methods of drug discovery utilize high throughput screening methods that may select molecules that lack both properties and will not undergo transport across the BBB. With the advent of in vitro cell culture models that mimic the BBB in vivo, researchers are now able to evaluate drug interactions at the BBB and to elucidate mechanisms at both the cellular and molecular levels pertinent for drug delivery. The utilization of both in vitro models for relatively rapid screening of permeability and related transport mechanisms, and in vivo models to assess drug pharmacokinetic distribution to the CNS provides a powerful assessment of drug delivery across the BBB to the CNS. Enaminone esters in the carbomethoxy series have previously been evaluated and shown to possess potent oral anticonvulsant activity in the mouse and rat. However, preliminary studies assessing enaminone analogs did not show expected correlations between relevant physiochemical parameters such as lipophilicity and BBB permeability. Therefore, in vitro models were utilized to assess factors associated with drug transport. An in vitro model of the BBB, Bovine Brain Microvessel Endothelial cells (BBMECs) were isolated and used to evaluated permeability and cellular mechanisms influencing enaminone transport. Results demonstrated that a multidrug resistant protein (MDR) influences enaminone permeability at the BBB. Further elucidation of possible mechanisms influencing enaminone distribution and pharmacokinetics were performed in a genetically altered mouse model [mdr1a/ b (--/--)] deficient in the expression of P-glycoprotein. Results comparing the brain distribution and partition coefficients in knockout mdr1a/b (--/--) versus wild type (+/+) counterparts demonstrated a higher accumulation of enaminones in brain tissue of knockout mice. Pharmacokinetic analysis of the tissue disposition of enaminones additionally demonstrated that the lack of P-glycoprotein in the lung and liver influence drug disposition in knockout animals. Lastly, a physiological based pharmacokinetic model was developed and found to be predictive of DM5 [methyl 4-[(4-chlorophenyl) amino]-6-methyl-2-oxocyclohex-3-en-1-oate] tissue distribution in mdr1a/b (--/--) knockout and wild type mice.
    • Effect of the tight junction modulator Zonula occludens toxin (Zot) and the P-glycoprotein inhibitor itraconazole on brain transport enhancement and their potential for renal drug interactions

      Karyekar, Chetan S.; Eddington, Natalie D.; Dowling, Thomas C. (2002)
      Drug delivery to the brain presents many challenges to the pharmaceutical scientist in part due to limited drug transport across the blood brain barrier (BBB). This dissertation focused on evaluation of a novel protein, Zonula occludens toxin (Zot) and a P-glycoprotien (P-gp) modulator, itraconazole, to enhance brain drug delivery, and to investigate their potential for drug interactions in the kidney. Zot, a protein elaborated by vibrio cholerae, is known to modulate tight junctions in intestinal epithelial cell models and enhance oral bioavailability. We initially evaluated the ability of Zot to modulate tight junctions in a bovine brain microvessel endothelial cell (BBMEC) model. The results from this study indicate that Zot transiently and reversibly enhances paracellular transport of selected marker and chemotherapeutic compounds. Subsequent in vivo studies in rats were consistent with these findings, where Zot caused a four-fold enhancement in brain uptake of paclitaxel, a poorly permeable compound. Truncation studies on Zot isolated its active fragment (DeltaG). Preliminary studies in rats indicated that DeltaG was susceptible to metabolism by proteases. Use of the protease inhibitors (leupeptin and captopril) with DeltaG improved its activity as measured by brain uptake of sucrose. However, DeltaG did not significantly enhance brain uptake of paclitaxel in rats. Itraconazole, a known P-gp inhibitor has been safely used for many years to treat mycotic infections. We assessed the effect of itraconazole on the brain uptake of paclitaxel in rats. The dosing strategy applied in this study did not significantly enhance brain uptake of paclitaxel suggesting that either higher concentrations (>1 mug/ml) are required or itraconazole may not effectively inhibit P-gp at the BBB. Next, we evaluated the effect of DeltaG and itraconazole on the renal handling of a P-gp substrate, cimetidine, in vitro. In the kidney, P-gp is known to be located in high concentrations in mesangium and proximal tubule where active secretion of drugs occurs. Studies in the renal tubular epithelial cell line MDR1-MDCK indicated that itraconazole decreased P-gp mediated cimetidine efflux. However, DeltaG did not alter cimetidine transport in this model. To further evaluate the effect of itraconazole on renal P-gP in vivo, a Phase I pharmacokinetic drug interaction study was conducted in healthy volunteers. Cimetidine and iothalamate (GFR) clearances were measured at baseline and following itraconazole dosing (200 mg bid x 4 days). Here, itraconazole reduced the tubular secretory clearance of cimetidine by 25%, resulting in a 21% increase in the plasma cimetidine AUC0--4 hrs. This data suggests that the observed renal P-gp interaction may be clinically significant. In conclusion, we anticipate that this research will provide a basis for newer strategies to effectively deliver drugs to the brain and increase awareness of potential drug interactions when using such approaches.
    • Evaluation of a possible drug interaction between remifentanil and esmolol using pharmacokinetic and pharmacodynamic modeling

      Haidar, Sam H.; Eddington, Natalie D. (1997)
      Remifentanil (ULTIVA) is an ultra short-acting opioid which has recently been approved for use during surgical procedures requiring opioid analgesia. After i.v. administration, it is rapidly metabolized by non-specific esterases in the blood and other tissues to less active metabolites. Total body clearance of remifentanil in man is 41.2 mL/min/kg, the Vd is 390 mL/kg and the elimination half-life is 10-48 minutes. In dogs, the clearance is 63 mL/min/kg, Vdss is 222 mL/kg and the elimination half-life is 5.7 minutes. Esmolol is an ultra short-acting beta-blocker administered intravenously during surgical procedures to decrease heart rate and blood pressure in patients at risk of cardiovascular complications. It is also metabolized by non-specific esterases in the blood and other tissues to form an inactive acid metabolite and methanol. It is likely both drugs will be administered together in patients with cardiac disease during surgical procedures. The goal of this study was to evaluate a possible drug interaction between remifentanil and esmolol following concomitant administration. This was accomplished by comparing pharmacokinetic and pharmacodynamic parameters of remifentanil after administration alone and in combination with esmolol. In the first study, remifentanil (treatment I, 25 mug/kg/min) and remifentanil plus esmolol (treatment II, (25 mug/kg/min and 200 mug/kg/min, respectively)) were infused for 20 minutes into male Sprague-Dawley rats in a random parallel design. Arterial blood samples were collected over the course of the study, extracted with methylene chloride, and analyzed (off-site) by a validated GC-MS assay. Additionally, cardiovascular measurements were continuously collected from 15 minutes pre-dose until about 20 minutes after end of infusion. Compartmental modeling was performed to determine the pharmacokinetic parameters. In a follow-up study, remifentanil (treatment I, 15 mug/kg/mL) and remifentanil plus esmolol (treatment II, (15 mug/kg/min and 600 mug/kg/min, respectively)) were infused into male Sprague-Dawley rats in a random two-way cross-over design. Blood samples (using limited sampling strategy) were collected and analyzed as before. EEG data were collected continuously over the course of the study. Bayesian estimation was used to determine the pharmacokinetic parameters. Pharmacodynamic parameters were obtained by modeling EEG spectral edge, using a pharmacokinetic-pharmacodynamic link model. Comparison of the pharmacokinetic and pharmacodynamic parameters between treatments I and II in both studies did not detect any significant (p < 0.05) differences. This indicates that at the doses tested, the co-administration of esmolol did not cause a significant change in the pharmacokinetics or pharmacodynamics of remifentanil no drug-drug interaction.
    • Identification of Drugs of Abuse as Modulators of Drug-Metabolizing Enzymes through Nuclear Receptor-Mediated Mechanisms

      Baaqee, Antonia Tolson; Eddington, Natalie D. (2012)
      To date, the majority of reports discussing opioid-drug interactions focus intensively on characterizing how other drugs affect the metabolic and pharmacokinetic (MPK) profile of opioids, however little has been published regarding the potential for opioids to modulate MPK-based drug-drug interactions (DDIs) involving other commonly co-administered or co-abused drugs. Moreover, virtually no mechanistic evidence has been explored. Thus, the objective of this work was to elucidate how opioids affect the MPK of other drugs, thereby undertaking research from a perspective that has been historically overlooked. Accordingly, the specific aims of this study were to: 1) Screen several different drugs of abuse for nuclear receptor (NR) activation potential, 2) Determine the expression profiles of key drug-metabolizing enzymes (DMEs) or drug transporters for selected drugs in human primary hepatocytes (HPHs), and 3) Characterize the mechanistic roles played by xenoreceptors Pregnane X Receptor (PXR) and Constitutive Androstane Receptor (CAR) underlying observed DME modulation. Results: Here we show that several opioids were identified as potential NR activators, and selected drugs of abuse exhibited differential induction profiles at the mRNA level for target genes CYP2B6 and CYP3A4. Overall, for opioid therapies MD and BUP: 1) MD induced the hepatic expression of multiple key DMEs by activating PXR- and CAR-mediated pathways; 2) More specifically, MD treatment resulted in significant nuclear accumulation of adenovirus/enhanced yellow fluorescent protein tagged-hCAR in HPHs, which has been regarded as the initial step of CAR activation, and additional analysis of the two enantiomers of racemic MD, R-(-)-MD (active) and S- (+)-MD (inactive), indicated a lack of stereoselectivity pertaining to MD-mediated DME induction; 3) For BUP, although hPXR-mediated CYP2B6 and CYP3A4 reporter activities were significantly increased in HepG2 cells, treatment with identical concentrations of buprenorphine in HPHs resulted in literally no induction of target gene expression. Taken together, these results provide much-needed mechanistic evidence which demonstrates that MD may be more likely than BUP to modulate CAR- and PXR- mediated DME perturbation during opioid-drug interactions. This research is of great importance to the overall public health industry, particularly to those clinicians and research scientists whom administer MD or BUP as part of opioid maintenance pharmacotherapy.
    • Improving the transport and oral bioavailability of therapeutic agents via modulation of the paracellular absorption by Delta G and inhibition of P-glycoprotein mediated drug efflux

      Salama Mohamed Ahmed Salama, Noha Nabil; Eddington, Natalie D. (2004)
      This research examined the effect of the active fragment of the absorption enhancer, Zonula occludens toxin (Zot), DeltaG, and potential P-glycoprotein (P-gp) inhibitors, the enaminones, on transepithelial transport and oral bioavailability of paracellular markers and/or low bioavailable drugs. The first approach focused on studies performed with the newly isolated biologically active fragment of Zot, DeltaG. The permeability of paracellular markers and low oral bioavailable compounds was studied in Caco-2 cells with/without DeltaG accompanied with cytotoxicity studies. In addition, DeltaG effect on the oral absorption of paracellular markers and low bioavailable drug candidates was investigated in jugular vein cannulated Sprague Dawley rats via intraduodenal administration. The second approach focused on a novel class of chemical compounds, the enaminones, as potential P-gp inhibitors. In vitro screening studies by efflux inhibition identified potential P-gp inhibitors, DM40 and DM27. The in vitro transport, in vivo absorption, and brain distribution were studied for P-gp substrates with DM27 and DM40. The results of this research displayed the effectiveness of AG in improving the in vitro and in vivo transport of paracellular markers and low bioavailable drugs. The in vitro transport enhancement reached 300% without any toxicity. Despite susceptibility to metabolism, DeltaG produced significant increases in Cmax and AUC, for all the investigated drugs with protease inhibitors (PI), up to 57 and 50 fold, respectively, for cyclosporin A. For certain agents, the increase in drug absorption observed with DeltaG/PI was a composite effect of DeltaG modulation of tight junctions and PI inhibition of metabolism. The screening of the enaminones revealed two P-gp inhibitors, DM27 and DM40. DM27 was more potent than DM40. Despite the distinctive enhancement observed with paclitaxel/DM27, further in vitro evaluation of the effect of DM27 with P-gp substrates displayed significant transport increases within ∼23 to 51% without any toxicity or effects on the paracellular/transcellular routes. In vivo studies found that both enaminones increased the brain levels of paclitaxel with a maximum of 227%. In conclusion, this research displayed the potential usefulness of paracellular absorption modulation by DeltaG and efflux inhibition by enaminones to increase the transport and/or oral bioavailability of therapeutic drugs.
    • Modulation of tight junctions and efflux transporters at the blood-brain barrier to increase the delivery of chemotherapeutic and antiretroviral agents to the brain

      Menon, Divya; Eddington, Natalie D. (2005)
      Effective drug delivery to the brain continues to be a challenge to successful therapy of disease states of the central nervous system. Chemotherapeutic molecules such as methotrexate that are used extensively in the treatment of systemic cancers have been found to be ineffective in the management of brain tumors, mainly due to subtherapeutic levels. Similarly, anitretrovirals like zidovudine and saquinavir are effective in significantly reducing systemic viral loads, but fail to reduce viral loads in the central nervous system. After systemic administration, therapeutic agents must permeate the blood brain barrier to reach the brain. The blood brain barrier functions to protect the brain from substances in the systemic circulation. The inter endothelial tight junctions restrict the paracellular diffusion of molecules and the efflux transporters expressed at the blood brain barrier actively transport substrates out of the brain. Thus, the blood brain barrier regulates transport of therapeutic molecules into the central nervous system and by preventing therapeutic levels of drugs from reaching the brain, it hinders effective therapy of these life threatening diseases. This work considered two different approaches in drug delivery to the brain via the transvascular route, namely, tight junction modulation and efflux transport inhibition, so as to enhance the brain distribution of commonly used antiretroviral and anticancer agents. DeltaG is the 12 kD active fragment of Zot, that has been shown to reversibily open tight junctions and enhance the systemic availability of a number of therapeutic agents. Administration of methotrexate in male Sprague Dawley rats along with DeltaG (200 to 600 mug/kg) resulted in an eight to twenty five fold increase in the brain to plasma partitioning of methotrexate. Our results show that DeltaG is capable of tight junctions modulation at the blood brain barrier, similar to its effect on the tight junctions at the intestinal epithelium. Both zidovudine and saquinavir are antiretroviral agents that exhibit limited distribution into the brain and are known to be substrates to efflux transporters expressed at the blood brain barrier. A three fold increase in the brain distribution of zidovudine in vivo in male FVB mice was achieved on treatment with a combination of the MRP1/2 selective inhibitor MK-571 and the non specific MRP inhibitor indomethacin. Similarly, a ten fold increase in the brain concentrations of saquinavir was achieved in vivo in male FVB mice on treatment with a combination of the P-gp inhibitor cyclosporin A and the CYP 3A4 inhibitor ketoconazole. Hence, both tight junction and efflux transport modulation hold considerable potential in enhancing the brain distribution of therapeutic agents.
    • Pregnancy-Induced Changes in Blood-Brain-Barrier and Placental Transporters and Blood-Brain and Maternal-Fetal Drug Distribution

      Coles, Lisa D.; Eddington, Natalie D. (2009)
      Pregnancy is a dynamic process in which both the mother's and fetus' physiological systems change continuously throughout gestation. Moreover, efflux transporters such as P-glycoprotein (P-gp) play a critical role in the maternal-to-fetal and blood-to-brain transfer of many drugs. Previous studies indicate that placental P-gp expression is altered during pregnancy. Furthermore, some central nervous system (CNS) acting drugs exhibit altered efficacy in the mother and fetal toxicity profiles when taken during pregnancy. Thus, it is important to characterize the distribution of agents routinely taken by pregnant women in this population. The objective of this work is to evaluate the effect of pregnancy on transporter expression and maternal blood-brain and maternal-fetal drug transport using in vitro cell models and an in vivo pregnant mouse model. To that end, the effect of progesterone (P4) and estradiol (E2) treatment on the expression and function of P-gp was evaluated in the placental JAR cells and P-gp over-expressing cells, NCI-ADR-RES. Using a mouse model, the effects of gestational age on P-gp and MRP expression in the placenta and brain were evaluated along with the brain, placental and fetal distribution of various P-gp and non P-gp substrates. Treatment with either E2 or P4 resulted in a significant increase in P-gp protein levels in both JAR and NCI-ADR-RES cells with the placental JAR cells more sensitive to E2, requiring 100-fold lower concentrations for P-gp induction. Furthermore, E2 or P4 treatment resulted in significant decreases in cellular uptake of select P-gp substrates indicating that the induced P-gp is functional. In vivo, P-gp protein levels in the placenta and brain were greater at mid- (gd 13) than late-gestation (gd 18). Likewise, brain MRP1 levels were greater in mice at gd 13, whereas, placental levels were greater at gd 18. To evaluate the effect of altered P-gp expression on drug disposition, the placenta, fetal and brain uptake of the P-gp substrates, saquinavir and methadone, and the non P-gp substrates; buprenorphine and paracellular markers, mannitol and sucrose was evaluated in mice. Following i.v. dosing, [3H]saquinavir placenta-to-plasma and fetal-to-plasma ratios were significantly greater in late-gestation mice versus mid-gestation. Furthermore, late-gestation mice experienced significant increases in the [3H]saquinavir and [3H]methadone brain-to-plasma ratios 60 min after dosing relative to mid-gestation. No significant differences were observed in tissue-to-plasma ratios for buprenorphine, mannitol or sucrose. Repeated dosing (3 doses, once daily) decreased the differential uptake of [3H]saquinavir in brain but potentiated it in the fetus. In conclusion, P4 or E2 increase P-gp expression and function in NCI-ADR-RES and JAR cells with the placental JAR more sensitive to E2. Furthermore, differential expression of P-gp in the brain and placenta likely contributes to the gestational-induced changes in brain and fetal uptake of saquinavir. These observed gestational-induced differences in brain and fetal uptake of saquinavir may have important implications for clinical dosing and efficacy of this compound as well as other P-gp substrates.