- Email: [email protected]
abcb1ab P-glycoprotein is involved in the uptake of citalopram and trimipramine into the brain of mice
Journal of Psychiatric Research 37 (2003) 179–185 www.elsevier.com/locate/jpsychires
abcb1ab P-glycoprotein is involved in the uptake of citalopram and trimipramine into the brain of mice Manfred Uhr*, Markus T. Grauer Max Planck Institute of Psychiatry, Kraepelinstrasse 10, D-80804 Munich, Germany Received 13 September 2002; received in revised form 10 October 2002; accepted 13 January 2003
Abstract The phenomenon of a heterogeneous response to the same drug in different patients is well-known. An important reason is that, even at equal concentrations, the bioavailability of a drug depends on the interaction of the drug with the blood–brain barrier (BBB). In part, this is due to the drug-transporting P-glycoprotein (P-gp), a product of the multiple drug resistance gene (ABCB1), which can transport drugs against a concentration gradient across the BBB back into the plasma and thereby reduce the bioavailability in the brain. In the present study, we have examined the uptake of the antidepressants citalopram and trimipramine into the brain of abcb1ab knockout mice compared with controls. One hour after s.c. injection of the drugs, concentrations of the two drugs and of the metabolite d-trimipramine in brain, spleen, kidney, liver and plasma were measured with HPLC. Significantly higher brain concentrations in knockout mice, showing that these drugs are substrates of P-gp and that the presence of P-gp reduces the effective bioavailability of these substances in the brain. The results of our study contradict an earlier report that citalopram is not actively transported from endothelial cells. These results were derived from an in vitro study, showing that due to the complexity of the BBB–drug interaction, it is difficult to transfer results from in vitro studies to the in vivo situation. We hypothesize that interindividual differences in the activity of the ABCB1 gene can account in part for the great variation in clinical response to antidepressants in psychiatric patients, even at comparable plasma levels. # 2003 Elsevier Science Ltd. All rights reserved. Keywords: Multiple drug resistance; Knock-out mice; P-glycoprotein; Blood–brain barrier; Citalopram; Trimipramine
1. Introduction The blood–brain barrier (BBB) regulates the uptake of substances from the plasma into the brain. It is formed by endothelial cells of the brain capillaries which are connected via tight junctions and the transport of substances across the BBB is regulated through endocytosis and/or transcytosis. The BBB has been studied intensively throughout the past few decades and the basic mechanisms are comparatively well-understood, although many details remain unclear. Any substance trying to cross into the brain has to interact with the BBB, and these interactions have been the subject of great scientific interest in the past, because they determine whether and to what extent a drug can enter the brain and subsequently exert a therapeutic effect. A complex interactive * Corresponding author. Tel.: +49-89-30622-651; fax: +49-8930622-310. E-mail address: [email protected] (M. Uhr).
framework exists between the BBB and a drug, depending on many factors, among them molecular weight, hydrophobicity, degree of ionization, protein and tissue binding (Saunders et al., 1999). One cannot easily predict the ability of a certain drug to cross the BBB simply from the chemical structure. Strong hydrophobicity, for example, is considered a prerequisite for a substance to enter the brain (Habgood et al., 2000), however, various hydrophobic substances cannot easily penetrate the BBB. One reason why it is difficult to anticipate the uptake of a certain drug into the brain is the existence of extrusion pumps, transporting substances against a concentration gradient from the brain back into the plasma. One of these extrusion pumps is P-glycoprotein (P-gp) which is expressed in the cells forming the BBB by the ABCB1 gene (Schinkel et al., 1996; Schinkel, 1999; von Moltke and Greenblatt, 2000). P-gp, a 170-kDa glycoprotein, is encoded by the ABCB1 gene in humans, shares many features with numerous bacterial and eucaryotic ATPbinding cassette (ABC) transport proteins and is a member
0022-3956/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0022-3956(03)00022-0
180
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
of a phylogenetically highly conserved superfamily of transport proteins (Gottesman et al., 1995; Fromm, 2000; Kerb et al., 2001b). In mice, P-gp is encoded by the abcb1a (also called mdr1a and mdr3) and abcb1b (also called mdr1b and mdr1) gene (Devault and Gros, 1990) and although abcb1a and abcb1b are not always expressed in the same organs, the overall distribution in mice tissue overlaps well with the single ABCB1 gene in humans, suggesting that these two genes work in the same manner as the single ABCB1 (van de Vrie et al., 1998). ABCB1 P-glycoprotein is a 1280 amino acid, glycosylated plasma membrane protein. It can actively transport its substrates against a concentration gradient, utilizing ATP hydrolysis as an energy source. The transported substrates for P-glycoprotein include anticancer drugs, such as anthracyclines, alkaloids, and immunosuppressive agent cyclosporin A, the glycoside digoxin, the synthetic glucocorticoid dexamethasone (Meijer et al., 1998), physiological steroids such as cortisol (Karssen et al., 2001), corticosterone, aldosterone and progesterone (Uhr et al., 2002), the antidepressant drug amitriptyline (Uhr et al., 2000), local anesthetics, and the anthelmintic drug ivermectin (Seelig, 1998). ABCB1-type P-glycoprotein has been found in the apical membrane of intestinal epithelial cells (Mukhopadhyay et al., 1988), in the biliary canalicular membrane of hepatocytes, and in the lumenal membrane of proximal tubular epithelial cells in the kidney (Thiebaut et al., 1987, 1989). High levels of ABCB1 P-glycoprotein have been found in the lumenal membrane of the endothelial cells that line small blood capillaries and form the blood– brain and blood–testis barrier (Cordon-Cardo et al., 1990; Tamai and Tsuji, 2000; Wijnholds et al., 2000). P-glycoprotein is therefore an important component of the BBB and has an impact on the actual bioavailability of a substance in the brain. For this reason, interactions between central nervous system (CNS) drugs and P-gp are of clinical relevance. Any physician who administers antidepressants has experienced that it is impossible to predict the wide range of possible outcomes when therapy is initiated. The resistance rate of antidepressant therapy can vary between 15 and 30%. We hypothesize that P-gp can be an obstacle to the successful treatment of CNS diseases. Since the expression of the ABCB1 gene varies inter-individually, drug concentrations in the CNS can vary greatly, depending on P-glycoprotein levels. A high activity of the ABCB1 gene could lead to a considerably reduced bioavailability of a drug in the CNS and contribute to a reduced therapeutic effect or non-response. In these cases, administration of drugs that are not a substrate of P-gp would be warranted. To test our hypothesis, we studied the brain uptake of two structurally different antidepressants, trimipramine and citalopram (Fig. 1) in knockout mice that lack the abcb1ab gene needed for P-glycoprotein production. Our goal was to determine whether the absence or pre-
sence of P-glycoprotein affects the concentrations of these substances in the brain.
2. Materials and methods 2.1. Materials Citalopram was used as a racemic mixture of s- and rcitalopram and obtained from Lundbeck (Copenhagen, Denmark), trimipramine was obtained from RhonePoulen-Rorer (Cologne, Germany). 2.2. Animals Male abcb1ab( / ) mice and FVB/N wildtype mice were housed individually and maintained on a 12:12 h light/dark cycle (lights on at 07:00), with food and water ad libitum. For age and body weight see Table 1. abcb1ab double knockout mice, originally created by A. Schinkel by sequential gene targeting in 129/Ola E14 embryonic stem cells (Schinkel et al., 1997), and backcrossed seven times (N7) to FVB/N from the C57BL/ 6129 chimera, and FVB/N wildtype mice were received from Taconic (Germantown, USA; FVB/Tac-[KO]Pgy2 N7). A homozygous colony is maintained at the Max Planck Institute of Psychiatry on the N7 FVB/N background through intercrossing of homozygous mice. 2.3. Experimental procedures Citalopram and trimipramine dissolved in 0.9% sodium chloride and 0.5% ethanol were administered
Fig. 1. Structural formula of citalopram and trimipramine.
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185 Table 1 Method details of the animals, experimental and extraction procedure and the high-performance liquid chromatography Trimipramine
Citalopram
Animals gender Group size [n] Age [weeks] Weight abcb1ab( / ) [g] SEM Weight abcb1ab(+/+)[g] SEM
Female 10 13–15 26.10.3 24.00.6
Female 10 12–15 26.40.4 27.80.9
Experimental procedures Subcutaneous administered substance
10 mg/g b.w.
1 mg/g b.w.
High-performance liquid chromatography Mobile phase gradient [% B] Isocratic 35 Detection UV [nm] 254 Detection fluorescence ex/em [nm] –
5-25 – 230/300
subcutaneous (s.c.) in the area of the nape of the neck. The entire volume injected was 10 ml/g mouse, for concentrations see Table 1. One hour after injection, the mice were anesthetized with halothane and decapitated. Trunk blood was collected in EDTA-coated tubes and centrifuged at 3000 g for 5 min for determination of the plasma concentration of the substances. Cerebrum, liver, spleen and kidney were dissected and weighed and then homogenized in the five-fold volume of an HBSS, 0.02 M HEPES buffer, pH 7.2 with a polytron PT 1200 Kinematic AG (Luzern, Switzerland). The homogenates were frozen at 80 C until further use. 2.4. Extraction procedure After thawing, the samples were homogenized in a Branson sonifier; 0.05 ml of an internal standard solution (protriptyline 10 mg/ml) and 0.4 ml of 2 M sodiumcarbonate buffer (adjusted to pH=10.5 with NaOH) were added to 0.4 ml of plasma or homogenate and vortexed. Five millilitres of n-hexane was added and the samples were mixed for 30 min at room temperature. After centrifugation for 15 min at 3000 g at room temperature, the organic layer was transferred to a tube containing 0.3 ml of 0.18 M phosphoric acid, mixed for 30 min and centrifuged at 3000 g for 10 min. The organic layer was then discarded, and an aliquot of the aqueous phase was injected for chromatographic separation. The extraction recoveries were > 90% for trimipramine, d-trimipramine and for citalopram.
181
Separations were made on a Luna 5m C18(2) 2504.6 mm column (Phenomenex, Torrance, USA). The mobile phase A: acetonitrile–water, 43.6 mM orthophosphoric acid, 35.9 mM triethylamine (15:85) and B: acetonitrile were degased for 15 min in an ultrasonic bath immediately before use. The column temperature was 60 C and the flow of the mobile phase was 1.0 ml/min. A mobile phase gradient were used for citalopram chromatographic analysis, for trimipramine and d-trimipramine we used isocratic chromatography (Table 1). The substances and its metabolites were determined by UV absorption or fluorescence at the described wavelength (Table 1). 2.6. Quantification Plasma and organ samples were calibrated by using spiked samples at 10 different concentrations, ranging from 1 to 1000 ng/ml. The concentrations were in the measurement range to the respective substances. Quantification was performed by calculating the analyte/ internal-standard peak-area ratio, and a regression model was fitted to the peak-area ratio of each compound to internal standard versus concentration. 2.7. Statistical analysis Statistical analysis was carried out by the department of biostatistics of the Max-Planck Institute for Psychiatry. Differences in the plasma and organ concentrations of the examined substances and its metabolites between the mutants and the wild-type mice were tested for significance by one-factorial multivariate analyses of variance (MANOVAs). The plasma and organ concentrations were the dependent variables and group, a between-subjects factor with two levels (mutants versus wild-type mice), was the independent variable. When a significant group effect was found for an organ sample or plasma, univariate F-tests followed to identify the variables, whose differences between the two groups contributed significantly to the global group effect. As a nominal level of significance a=0.05 was accepted and corrected (reduced according to the Bonferroni procedure) for all a posteriori tests (univariate F-tests) in order to keep the type I error less than or equal to 0.05.
3. Results
2.5. High-performance liquid chromatography
3.1. Trimipramine
A Beckman 166 variable-wavelength UV detector and a Merck L-7480 fluorescence detector, a Beckman gradient pump 126 Solvent Module equipped with a Beckman autoinjector 508 autosampler were used for high-performance liquid chromatography analysis.
One hour after s.c. injection of 10 mg trimipramine/g bodyweight the trimipramine concentrations in the cerebrum were different between the abcb1ab( / ) mutant and the wildtype controls. Analysis of variance revealed a significant group effect on the trimipramine con-
182
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
centrations [Wilks multivariate test of significance; effect of group: F(5, 14)=7.9; significance of F=0.001], to which reasonably only the concentration in the cerebrum contributed (univariate F-test; P-values < 0.05). The cerebrum concentration for trimipramine was 1.2 times higher in the knockout mice than in controls (Table 2 and Fig. 2). There were no significant differences in the trimipramine concentrations in plasma, spleen, kidney and liver. 3.2. Desmethyltrimipramine (d-trimipramine) One hour after s.c. injection of 10 mg trimipramine/g bodyweight analysis of variance revealed a significant group effect on the concentrations of the metabolite d-trimipramine [Wilks multivariate test of significance; effect of group: F(5, 14)=3.2; significance of F=0.04 (univariate F-test; P-values < 0.05), to which reasonably only the concentrations in the cerebrum contributed (Table 2 and Fig. 2). 3.3. Citalopram One hour after s.c. injection of 1 mg citalopram/g bodyweight analysis of variance revealed a significant group effect on the citalopram concentrations [Wilks multivariate test of significance; effect of group: F(5, 14)=97.7; significance of F < 0.001], to which reasonably only the concentrations in the cerebrum contributed (univariate F-test; P-values < 0.05). The cerebrum citalopram concentration of the abcb1ab(+/+) control
mice was less than 1/3 of the concentration in the cerebrum of the abcb1ab( / ) mutant mice (Table 2 and Fig. 2).
4. Discussion Our study shows considerable differences in the cerebral concentrations of citalopram between knock-out mice and controls. The differences were smaller for trimipramine and d-trimipramine, however they were still significant. This suggest that citalopram and possibly trimipramine and d-trimipramine are substrates of P-gp. Any drug treatment requires sufficient bioavailability of the drug in the affected tissue. For cerebral disorders, this poses a particular problem: Substances that are administered systemically have to cross the BBB to penetrate from the circulation into the brain tissue. P-glycoprotein is an important component of the BBB and determines in part how much of a substance can cross into the brain and determines subsequently the actual cerebral bioavailability of a drug (van Asperen et al., 1997). Located at the apical membrane of the small capillary endothelial cells, P-glycoprotein is able to transport its substrates against the concentration gradient back into the blood. Using abcb1a knock-out mice, Schinkel et al. (1994) showed that various substances, which are structurally entirely unrelated, are substrates of P-glycoprotein. These substances are exported against a concentration gradient out of the brain into systemic circulation. Our group showed for
Table 2 Organ concentrations of antidepressant drugs and the metabolite 1 h after subcutaneous injection (values are meanSEM) Substance
Organ
abcb1ab / [ng/ml]
abcb1ab+/+ [ng/ml]
Ratio
P-value
meanSEM
meanSEM
plasma cerebrum spleen kidney liver
170.99.4 1666.275.3 1926.2114.2 2142.9165.2 358.229.8
198.49.0 1388.889.8 2291.3245.8 2498.7131.1 410.130.5
0.9 1.2 0.8 0.9 0.9
n.sa * n.s n.s n.s
plasma cerebrum spleen kidney liver
14.5 1.2 48.4 3.7 91.4 5.3 119.98.6 76.6 6.7
12.01.4 33.23.7 89.16.3 125.011.3 83.86.2
1.2 1.5 1.0 1.0 0.9
n.s * n.s n.s n.s
plasma cerebrum spleen kidney liver
28.9 1.9 455.115.5 555.942.5 426.830.0 128.16.5
33.21.5 153.58.3 610.948.4 500.644.3 128.58.8
0.9 3.0 0.9 0.9 1.0
n.s * n.s n.s n.s
Trimipramine
D-Trimipramine
Citalopram
a
n.s not significant. * P <0.05.
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
183
Fig. 2. Per cent change in antidepressant concentration. Black columns are abcb1ab ( / ) knock-out mice, grey column wild-type mice, generally set to 100%. pla=plasma, cer=cerebrum, spl=spleen, kid=kidney and liv=liver.
the first time in vivo that some antidepressants are substrates of P-gp and that the absence or presence of P-gp affects the cerebral bioavailability of antidepressants. Other antidepressants, such as fluoxetine, are not substrates of P-glycoprotein. Also, P-glycoprotein is not present in all organs and in certain organs, for example in, kidney or liver, P-gp is not located in endothelial cells. In our study, concentrations of all three substances did not differ significantly in spleen, kidney or liver. Plasma concentrations also showed no significant differences. This supports the hypothesis that the antidepressants are exported against the concentration gradient from the intracerebral into the extracerebral space via the BBB and that the observed effect is a specific effect of transport across the BBB and not due to passive mechanisms. One has to take into account that in this study, the substances have been administered in a dose that is close to the upper limit of what can be used in therapy in humans. In our study, the coefficient of variation of the cerebral drug concentrations was approximately 20% in both groups. This surprisingly low coefficient of variation can be attributed to measuring methods and the
characteristics of the laboratory animals. In both groups, the mice share the same genetic background and live under standardized conditions. Presumably, this creates a relatively homogenous group, in which expression and function of the P-glycoprotein is similar in all individuals. In general, only certain substances are substrates of Pglycoprotein, and not all substrates show the same quantitative interactions. When concentration in the brain tissue is compared between both, the abcb1ab( / ) and the wild type mouse group, the ratio is quite low for trimipramine with 1.2 and d-trimipramine with 1.5, i.e. if P-gp is absent, the concentration of the drug in the brain is 1.2 and 1.5 times higher, respectively. The ratio is much higher for citalopram (3.0). We showed recently for the antidepressant amitriptyline and its metabolites that the ratio of concentration between knock-outs and wild-type ranges between 1.8 and 4.5 (Uhr et al., 2000). This ratio is not very high when compared with substrates with a high affinity for P-glycoprotein, such as cytostatic agents, digoxine or invermectine. However, one would easily expect that a 100–300% change in cerebral bioavailability of antidepressants could be of clinical relevance in humans.
184
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
Moreover, the therapeutic picture in humans in general is more complex, among the complicating factors are newly discovered polymorphisms and co-medications that affect P-gp activity. Frequently, several CNS drugs are co-administered, and often additional, nonCNS drugs are added. In recent studies, polymorphisms of ABCB1 in humans have been described. In human studies, these polymorphisms have been shown to affect intestinal uptake of the P-gp substrate digoxin (Hoffmeyer et al., 2000; Cascorbi et al., 2001; Kerb et al., 2001b), to affect phenytoin plasma levels in healthy volunteers (Kerb et al., 2001a), and to be associated with orthostatic hypotension in patients with depression treated with nortriptyline (Roberts et al., 2002). Co-medications, such as the popular non-prescription antidepressant St. John’s Wort can increase expression of P-gp (Hennessy et al., 2002) in humans and the P-gp inhibitor quinidine reverses the lack of central respiratory depression of loperamide (Sadeque et al., 2000). It can therefore be suspected that co-medications counteract therapeutic efforts in humans, explaining inter-individual differences in the response to drugs that are substrates of P-glycoprotein. Besides this human data, many more modulating effects have been inferred from cell culture and animal experimental studies (Drewe et al., 2000; Ibrahim et al., 2000; Lentz et al., 2000; Dagenais et al., 2001; van der Sandt et al., 2001). Here, as well in cancer therapy, a number of agents have successfully been used to suppress the activity of the P-glycoprotein (Ford and Hait, 1990) or to increase the expression of the P-gp (Seelig, 1998). Pretreatment of mice with the P-gp inhibitor cyclosporine enhances and prolongs opiate-induced analgesia in mice (Thompson et al., 2000). For this reason, plasma concentrations of antidepressants alone do not allow for a definitive statement about the concentration in the brain. A patient with normal plasma concentrations of citalopram may in fact be receiving an insufficient dose because the penetration into the brain is reduced by an active export of the antidepressant drug through the P-gp located at the blood–brain barrier. In addition to varying pharmacobiology of the patients, this is another reason why therapy resistance could arise. Plasma levels and therapeutic drug monitoring are therefore important tools for any physician administering CNS drugs, but additional information about possible interactions with P-gp would be helpful for the clinician. It is important to note that so far, it is not possible to predict from the structure of the molecule which substance will be a substrate of P-glycoprotein and which one will not be (Fig. 1). Therefore, it is currently necessary to test all substances used as antidepressants in order to gather more data about the interaction between P-glycoprotein and antidepressants. Such a survey is currently in progress.
Because of the complexity of the blood–brain barrier, it is difficult to investigate the interactions between the blood–brain barrier and drugs. An earlier in vitro study, using monolayers of bovine brain microvessel endothelial cells (BMECs), had reported that citalopram is not a substrate for active efflux mechanisms (Rochat et al., 1999).This was based on the observation that cyclosporin, a P-glycoprotein inhibitor, did not modify the transport of citalopram. Our results, however, show that citalopram is exported actively, similar to amitriptyline. This shows that for the investigation of the blood–brain barrier, in vivo studies are indispensable and results of in vitro studies cannot easily be transferred to the in vivo situation, due to the complexity of the blood–brain barrier. In sum, the present study shows that CNS drugs, such as antidepressants, can be substrates of P-gp at the blood–brain barrier and are actively exported out of the brain. Together with the new human, animal and cell culture data about P-gp, its activity and substrates, polymorphisms and effect of co-medications, it becomes increasingly clear that the complexity of the blood– brain barrier has to be taken into account in the administration of CNS drugs more than before, even if the CNS drugs used have been shown to enter the brain and to have a therapeutic effect. Future research in this field and more data about these interactions will probably allow in the future to tailor CNS pharmacotherapy more individually to each patient, to optimize therapeutic efficacy and to minimize undesired side effects. The rapid progress in pharmacogenomics will accelerate this development towards an individual, pharmacogenomically oriented therapy.
Acknowledgements We would like to thank Ms. A. Rippl and Ms. M. Ha¨usler for their very helpful technical assistance.
References Cascorbi I, Gerloff T, Johne A, Meise C, Hoffmeyer S, Schwab M, et al. Frequency of single nucleotide polymorphisms in the P-glycoprotein drug transporter MDR1 gene in white subjects. Clinical Pharmacology and Therapeutics 2001;69:169–74. Cordon-Cardo C, O’Brien JP, Boccia J, Casals D, Bertino JR, Melamed MR. Expression of the multidrug resistance gene product (Pglycoprotein) in human normal and tumor tissues. Journal of Histochemistry and Cytochemistry 1990;38:1277–87. Dagenais C, Zong J, Ducharme J, Pollack GM. Effect of mdr1a P-glycoprotein gene disruption, gender, and substrate concentration on brain uptake of selected compounds. Pharmaceutical Research 2001;18:957–63. Devault A, Gros P. Two members of the mouse mdr gene family confer multidrug resistance with overlapping but distinct drug specificities. Molecular and Cellular Biology 1990;10:1652–63.
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185 Drewe J, Ball HA, Beglinger C, Peng B, Kemmler A, Schachinger H, et al. Effect of P-glycoprotein modulation on the clinical pharmacokinetics and adverse effects of morphine. British Journal of Clinical Pharmacology 2000;50:237–46. Ford JM, Hait WN. Pharmacology of drugs that alter multidrug resistance in cancer. Pharmacological Reviews 1990;42:155–99. Fromm MF. P-glycoprotein: a defense mechanism limiting oral bioavailability and CNS accumulation of drugs. International Journal of Clinical Pharmacology and Therapeutics 2000;38:69–74. Gottesman MM, Hrycyna CA, Schoenlein PV, Germann UA, Pastan I. Genetic analysis of the multidrug transporter. Annual Review of Genetics 1995;29:607–49. Habgood MD, Begley DJ, Abbott NJ. Determinants of passive drug entry into the central nervous system. Cellular and Molecular Neurobiology 2000;20:231–53. Hennessy M, Kelleher D, Spiers JP, Barry M, Kavanagh P, Back D, et al. St John’s wort increases expression of P-glycoprotein: implications for drug interactions. British Journal of Clinical Pharmacology 2002;53:75–82. Hoffmeyer S, Burk O, von Richter O, Arnold HP, Brockmoller J, Johne A, et al. Functional polymorphisms of the human multidrugresistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo. Proceedings of the National Academy of Sciences of the United States of America 2000;97:3473–8. Ibrahim S, Peggins J, Knapton A, Licht T, Aszalos A. Influence of antipsychotic, antiemetic, and Ca(2+) channel blocker drugs on the cellular accumulation of the anticancer drug daunorubicin: P-glycoprotein modulation. Journal of Pharmacology and Experimental Therapeutics 2000;295:1276–83. Karssen AM, Meijer OC, van der Sandt I, Lucassen PJ, de Lange ECM, de Boer AG, et al. Multidrug resistance P-glycoprotein hampers the access of cortisol but not of corticosterone to mouse and human brain. Endocrinology 2001;142:2686–94. Kerb R, Aynacioglu AS, Brockmoller J, Schlagenhaufer R, Bauer S, Szekeres T, et al. The predictive value of MDR1, CYP2C9, and CYP2C19 polymorphisms for phenytoin plasma levels. The Pharmacogenomics Journal 2001a;1:204–10. Kerb R, Hoffmeyer S, Brinkmann U. ABC drug transporters: hereditary polymorphisms and pharmacological impact in MDR1, MRP1 and MRP2. The Pharmacogenomics Journal 2001b;2:51–64. Lentz KA, Polli JW, Wring SA, Humphreys JE, Polli JE. Influence of passive permeability on apparent P-glycoprotein kinetics. Pharmaceutical Research 2000;17:1456–60. Meijer OC, de Lange EC, Breimer DD, de Boer AG, Workel JO, de Kloet ER. Penetration of dexamethasone into brain glucocorticoid targets is enhanced in mdr1A P-glycoprotein knockout mice. Endocrinology 1998;139:1789–93. Mukhopadhyay T, Batsakis JG, Kuo MT. Expression of the mdr (Pglycoprotein) gene in Chinese hamster digestive tracts. Journal of the National Cancer Institute 1988;80:269–75. Roberts RL, Joyce PR, Mulder RT, Begg EJ, Kennedy MA. A common P-glycoprotein polymorphism is associated with nortriptylineinduced postural hypotension in patients treated for major depression. The Pharmacogenomics Journal 2002;2:191–6. Rochat B, Baumann P, Audus KL. Transport mechanisms for the antidepressant citalopram in brain microvessel endothelium. Brain Research 1999;831:229–36. Sadeque AJ, Wandel C, He H, Shah S, Wood AJ. Increased drug delivery to the brain by P-glycoprotein inhibition. Clinical Pharmacology and Therapeutics 2000;68:231–7. Saunders NR, Habgood MD, Dziegielewska KM. Barrier mechanisms
185
in the brain, I. Adult brain. Clinical and Experimental Pharmacology and Physiology 1999;26:11–19. Schinkel AH. P-glycoprotein, a gatekeeper in the blood–brain barrier. Advanced Drug Delivery Reviews 1999;36:179–94. Schinkel AH, Mayer U, Wagenaar E, Mol CA, van Deemter L, Smit JJ, et al. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proceedings of the National Academy of Sciences of the United States of America 1997;94:4028–33. Schinkel AH, Smit JJ, van Tellingen O, Beijnen JH, Wagenaar E, van Deemter L, et al. Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood–brain barrier and to increased sensitivity to drugs. Cell 1994;77:491–502. Schinkel AH, Wagenaar E, Mol CA, van Deemter L. P-glycoprotein in the blood–brain barrier of mice influences the brain penetration and pharmacological activity of many drugs. Journal of Clinical Investigation 1996;97:2517–24. Seelig A. A general pattern for substrate recognition by P-glycoprotein. European Journal of Biochemistry 1998;251:252–61. Tamai I, Tsuji A. Transporter-mediated permeation of drugs across the blood–brain barrier. Journal of Pharmaceutical Sciences 2000; 89:1371–88. Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, Willingham MC. Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proceedings of the National Academy of Sciences of the United States of America 1987;84:7735–8. Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, Willingham MC. Immunohistochemical localization in normal tissues of different epitopes in the multidrug transport protein P170: evidence for localization in brain capillaries and crossreactivity of one antibody with a muscle protein. Journal of Histochemistry and Cytochemistry 1989;37:159–64. Thompson SJ, Koszdin K, Bernards CM. Opiate-induced analgesia is increased and prolonged in mice lacking P-glycoprotein. Anesthesiology 2000;92:1392–9. Uhr M, Holsboer F, Mu¨ller MB. Penetration of endogenous steroid hormones corticosterone, cortisol, aldosterone and progesterone into the brain is enhanced in mice deficient for both mdr1a and mdr1b P-glycoproteins. Journal of Neuroendocrinology 2002;14: 753–9. Uhr M, Steckler T, Yassouridis A, Holsboer F. Penetration of amitriptyline, but not of fluoxetine, into brain is enhanced in mice with blood–brain-barrier deficiency due to Mdr1a P-glycoprotein gene disruption. Neuropsychopharmacology 2000;22:380–7. van Asperen J, Mayer U, van Tellingen O, Beijnen JH. The functional role of P-glycoprotein in the blood–brain barrier. Journal of Pharmaceutical Sciences 1997;86:881–4. van de Vrie W, Marquet RL, Stoter G, de Bruijn EA, Eggermont AM. In vivo model systems in P-glycoprotein-mediated multidrug resistance. Critical Reviews in Clinical Laboratory Sciences 1998;35:1–57. van der Sandt I, Vos CM, Nabulsi L, Blom-Roosemalen MC, Voorwinden HH, de Boer AG, et al. Assessment of active transport of HIV protease inhibitors in various cell lines and the in vitro blood– brain barrier. AIDS 2001;15:483–91. von Moltke LL, Greenblatt DJ. Drug transporters in psychopharmacology—are they important? Journal of Clinical Pharmacology 2000;20:291–4. Wijnholds J, de Lange EC, Scheffer GL, van den Berg DJ, Mol CA, van der Valk MA, et al. Multidrug resistance protein 1 protects the choroid plexus epithelium and contributes to the blood–cerebrospinal fluid barrier. Journal of Clinical Investigation 2000;105:279–85.
abcb1ab P-glycoprotein is involved in the uptake of citalopram and trimipramine into the brain of mice Manfred Uhr*, Markus T. Grauer Max Planck Institute of Psychiatry, Kraepelinstrasse 10, D-80804 Munich, Germany Received 13 September 2002; received in revised form 10 October 2002; accepted 13 January 2003
Abstract The phenomenon of a heterogeneous response to the same drug in different patients is well-known. An important reason is that, even at equal concentrations, the bioavailability of a drug depends on the interaction of the drug with the blood–brain barrier (BBB). In part, this is due to the drug-transporting P-glycoprotein (P-gp), a product of the multiple drug resistance gene (ABCB1), which can transport drugs against a concentration gradient across the BBB back into the plasma and thereby reduce the bioavailability in the brain. In the present study, we have examined the uptake of the antidepressants citalopram and trimipramine into the brain of abcb1ab knockout mice compared with controls. One hour after s.c. injection of the drugs, concentrations of the two drugs and of the metabolite d-trimipramine in brain, spleen, kidney, liver and plasma were measured with HPLC. Significantly higher brain concentrations in knockout mice, showing that these drugs are substrates of P-gp and that the presence of P-gp reduces the effective bioavailability of these substances in the brain. The results of our study contradict an earlier report that citalopram is not actively transported from endothelial cells. These results were derived from an in vitro study, showing that due to the complexity of the BBB–drug interaction, it is difficult to transfer results from in vitro studies to the in vivo situation. We hypothesize that interindividual differences in the activity of the ABCB1 gene can account in part for the great variation in clinical response to antidepressants in psychiatric patients, even at comparable plasma levels. # 2003 Elsevier Science Ltd. All rights reserved. Keywords: Multiple drug resistance; Knock-out mice; P-glycoprotein; Blood–brain barrier; Citalopram; Trimipramine
1. Introduction The blood–brain barrier (BBB) regulates the uptake of substances from the plasma into the brain. It is formed by endothelial cells of the brain capillaries which are connected via tight junctions and the transport of substances across the BBB is regulated through endocytosis and/or transcytosis. The BBB has been studied intensively throughout the past few decades and the basic mechanisms are comparatively well-understood, although many details remain unclear. Any substance trying to cross into the brain has to interact with the BBB, and these interactions have been the subject of great scientific interest in the past, because they determine whether and to what extent a drug can enter the brain and subsequently exert a therapeutic effect. A complex interactive * Corresponding author. Tel.: +49-89-30622-651; fax: +49-8930622-310. E-mail address: [email protected] (M. Uhr).
framework exists between the BBB and a drug, depending on many factors, among them molecular weight, hydrophobicity, degree of ionization, protein and tissue binding (Saunders et al., 1999). One cannot easily predict the ability of a certain drug to cross the BBB simply from the chemical structure. Strong hydrophobicity, for example, is considered a prerequisite for a substance to enter the brain (Habgood et al., 2000), however, various hydrophobic substances cannot easily penetrate the BBB. One reason why it is difficult to anticipate the uptake of a certain drug into the brain is the existence of extrusion pumps, transporting substances against a concentration gradient from the brain back into the plasma. One of these extrusion pumps is P-glycoprotein (P-gp) which is expressed in the cells forming the BBB by the ABCB1 gene (Schinkel et al., 1996; Schinkel, 1999; von Moltke and Greenblatt, 2000). P-gp, a 170-kDa glycoprotein, is encoded by the ABCB1 gene in humans, shares many features with numerous bacterial and eucaryotic ATPbinding cassette (ABC) transport proteins and is a member
0022-3956/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0022-3956(03)00022-0
180
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
of a phylogenetically highly conserved superfamily of transport proteins (Gottesman et al., 1995; Fromm, 2000; Kerb et al., 2001b). In mice, P-gp is encoded by the abcb1a (also called mdr1a and mdr3) and abcb1b (also called mdr1b and mdr1) gene (Devault and Gros, 1990) and although abcb1a and abcb1b are not always expressed in the same organs, the overall distribution in mice tissue overlaps well with the single ABCB1 gene in humans, suggesting that these two genes work in the same manner as the single ABCB1 (van de Vrie et al., 1998). ABCB1 P-glycoprotein is a 1280 amino acid, glycosylated plasma membrane protein. It can actively transport its substrates against a concentration gradient, utilizing ATP hydrolysis as an energy source. The transported substrates for P-glycoprotein include anticancer drugs, such as anthracyclines, alkaloids, and immunosuppressive agent cyclosporin A, the glycoside digoxin, the synthetic glucocorticoid dexamethasone (Meijer et al., 1998), physiological steroids such as cortisol (Karssen et al., 2001), corticosterone, aldosterone and progesterone (Uhr et al., 2002), the antidepressant drug amitriptyline (Uhr et al., 2000), local anesthetics, and the anthelmintic drug ivermectin (Seelig, 1998). ABCB1-type P-glycoprotein has been found in the apical membrane of intestinal epithelial cells (Mukhopadhyay et al., 1988), in the biliary canalicular membrane of hepatocytes, and in the lumenal membrane of proximal tubular epithelial cells in the kidney (Thiebaut et al., 1987, 1989). High levels of ABCB1 P-glycoprotein have been found in the lumenal membrane of the endothelial cells that line small blood capillaries and form the blood– brain and blood–testis barrier (Cordon-Cardo et al., 1990; Tamai and Tsuji, 2000; Wijnholds et al., 2000). P-glycoprotein is therefore an important component of the BBB and has an impact on the actual bioavailability of a substance in the brain. For this reason, interactions between central nervous system (CNS) drugs and P-gp are of clinical relevance. Any physician who administers antidepressants has experienced that it is impossible to predict the wide range of possible outcomes when therapy is initiated. The resistance rate of antidepressant therapy can vary between 15 and 30%. We hypothesize that P-gp can be an obstacle to the successful treatment of CNS diseases. Since the expression of the ABCB1 gene varies inter-individually, drug concentrations in the CNS can vary greatly, depending on P-glycoprotein levels. A high activity of the ABCB1 gene could lead to a considerably reduced bioavailability of a drug in the CNS and contribute to a reduced therapeutic effect or non-response. In these cases, administration of drugs that are not a substrate of P-gp would be warranted. To test our hypothesis, we studied the brain uptake of two structurally different antidepressants, trimipramine and citalopram (Fig. 1) in knockout mice that lack the abcb1ab gene needed for P-glycoprotein production. Our goal was to determine whether the absence or pre-
sence of P-glycoprotein affects the concentrations of these substances in the brain.
2. Materials and methods 2.1. Materials Citalopram was used as a racemic mixture of s- and rcitalopram and obtained from Lundbeck (Copenhagen, Denmark), trimipramine was obtained from RhonePoulen-Rorer (Cologne, Germany). 2.2. Animals Male abcb1ab( / ) mice and FVB/N wildtype mice were housed individually and maintained on a 12:12 h light/dark cycle (lights on at 07:00), with food and water ad libitum. For age and body weight see Table 1. abcb1ab double knockout mice, originally created by A. Schinkel by sequential gene targeting in 129/Ola E14 embryonic stem cells (Schinkel et al., 1997), and backcrossed seven times (N7) to FVB/N from the C57BL/ 6129 chimera, and FVB/N wildtype mice were received from Taconic (Germantown, USA; FVB/Tac-[KO]Pgy2 N7). A homozygous colony is maintained at the Max Planck Institute of Psychiatry on the N7 FVB/N background through intercrossing of homozygous mice. 2.3. Experimental procedures Citalopram and trimipramine dissolved in 0.9% sodium chloride and 0.5% ethanol were administered
Fig. 1. Structural formula of citalopram and trimipramine.
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185 Table 1 Method details of the animals, experimental and extraction procedure and the high-performance liquid chromatography Trimipramine
Citalopram
Animals gender Group size [n] Age [weeks] Weight abcb1ab( / ) [g] SEM Weight abcb1ab(+/+)[g] SEM
Female 10 13–15 26.10.3 24.00.6
Female 10 12–15 26.40.4 27.80.9
Experimental procedures Subcutaneous administered substance
10 mg/g b.w.
1 mg/g b.w.
High-performance liquid chromatography Mobile phase gradient [% B] Isocratic 35 Detection UV [nm] 254 Detection fluorescence ex/em [nm] –
5-25 – 230/300
subcutaneous (s.c.) in the area of the nape of the neck. The entire volume injected was 10 ml/g mouse, for concentrations see Table 1. One hour after injection, the mice were anesthetized with halothane and decapitated. Trunk blood was collected in EDTA-coated tubes and centrifuged at 3000 g for 5 min for determination of the plasma concentration of the substances. Cerebrum, liver, spleen and kidney were dissected and weighed and then homogenized in the five-fold volume of an HBSS, 0.02 M HEPES buffer, pH 7.2 with a polytron PT 1200 Kinematic AG (Luzern, Switzerland). The homogenates were frozen at 80 C until further use. 2.4. Extraction procedure After thawing, the samples were homogenized in a Branson sonifier; 0.05 ml of an internal standard solution (protriptyline 10 mg/ml) and 0.4 ml of 2 M sodiumcarbonate buffer (adjusted to pH=10.5 with NaOH) were added to 0.4 ml of plasma or homogenate and vortexed. Five millilitres of n-hexane was added and the samples were mixed for 30 min at room temperature. After centrifugation for 15 min at 3000 g at room temperature, the organic layer was transferred to a tube containing 0.3 ml of 0.18 M phosphoric acid, mixed for 30 min and centrifuged at 3000 g for 10 min. The organic layer was then discarded, and an aliquot of the aqueous phase was injected for chromatographic separation. The extraction recoveries were > 90% for trimipramine, d-trimipramine and for citalopram.
181
Separations were made on a Luna 5m C18(2) 2504.6 mm column (Phenomenex, Torrance, USA). The mobile phase A: acetonitrile–water, 43.6 mM orthophosphoric acid, 35.9 mM triethylamine (15:85) and B: acetonitrile were degased for 15 min in an ultrasonic bath immediately before use. The column temperature was 60 C and the flow of the mobile phase was 1.0 ml/min. A mobile phase gradient were used for citalopram chromatographic analysis, for trimipramine and d-trimipramine we used isocratic chromatography (Table 1). The substances and its metabolites were determined by UV absorption or fluorescence at the described wavelength (Table 1). 2.6. Quantification Plasma and organ samples were calibrated by using spiked samples at 10 different concentrations, ranging from 1 to 1000 ng/ml. The concentrations were in the measurement range to the respective substances. Quantification was performed by calculating the analyte/ internal-standard peak-area ratio, and a regression model was fitted to the peak-area ratio of each compound to internal standard versus concentration. 2.7. Statistical analysis Statistical analysis was carried out by the department of biostatistics of the Max-Planck Institute for Psychiatry. Differences in the plasma and organ concentrations of the examined substances and its metabolites between the mutants and the wild-type mice were tested for significance by one-factorial multivariate analyses of variance (MANOVAs). The plasma and organ concentrations were the dependent variables and group, a between-subjects factor with two levels (mutants versus wild-type mice), was the independent variable. When a significant group effect was found for an organ sample or plasma, univariate F-tests followed to identify the variables, whose differences between the two groups contributed significantly to the global group effect. As a nominal level of significance a=0.05 was accepted and corrected (reduced according to the Bonferroni procedure) for all a posteriori tests (univariate F-tests) in order to keep the type I error less than or equal to 0.05.
3. Results
2.5. High-performance liquid chromatography
3.1. Trimipramine
A Beckman 166 variable-wavelength UV detector and a Merck L-7480 fluorescence detector, a Beckman gradient pump 126 Solvent Module equipped with a Beckman autoinjector 508 autosampler were used for high-performance liquid chromatography analysis.
One hour after s.c. injection of 10 mg trimipramine/g bodyweight the trimipramine concentrations in the cerebrum were different between the abcb1ab( / ) mutant and the wildtype controls. Analysis of variance revealed a significant group effect on the trimipramine con-
182
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
centrations [Wilks multivariate test of significance; effect of group: F(5, 14)=7.9; significance of F=0.001], to which reasonably only the concentration in the cerebrum contributed (univariate F-test; P-values < 0.05). The cerebrum concentration for trimipramine was 1.2 times higher in the knockout mice than in controls (Table 2 and Fig. 2). There were no significant differences in the trimipramine concentrations in plasma, spleen, kidney and liver. 3.2. Desmethyltrimipramine (d-trimipramine) One hour after s.c. injection of 10 mg trimipramine/g bodyweight analysis of variance revealed a significant group effect on the concentrations of the metabolite d-trimipramine [Wilks multivariate test of significance; effect of group: F(5, 14)=3.2; significance of F=0.04 (univariate F-test; P-values < 0.05), to which reasonably only the concentrations in the cerebrum contributed (Table 2 and Fig. 2). 3.3. Citalopram One hour after s.c. injection of 1 mg citalopram/g bodyweight analysis of variance revealed a significant group effect on the citalopram concentrations [Wilks multivariate test of significance; effect of group: F(5, 14)=97.7; significance of F < 0.001], to which reasonably only the concentrations in the cerebrum contributed (univariate F-test; P-values < 0.05). The cerebrum citalopram concentration of the abcb1ab(+/+) control
mice was less than 1/3 of the concentration in the cerebrum of the abcb1ab( / ) mutant mice (Table 2 and Fig. 2).
4. Discussion Our study shows considerable differences in the cerebral concentrations of citalopram between knock-out mice and controls. The differences were smaller for trimipramine and d-trimipramine, however they were still significant. This suggest that citalopram and possibly trimipramine and d-trimipramine are substrates of P-gp. Any drug treatment requires sufficient bioavailability of the drug in the affected tissue. For cerebral disorders, this poses a particular problem: Substances that are administered systemically have to cross the BBB to penetrate from the circulation into the brain tissue. P-glycoprotein is an important component of the BBB and determines in part how much of a substance can cross into the brain and determines subsequently the actual cerebral bioavailability of a drug (van Asperen et al., 1997). Located at the apical membrane of the small capillary endothelial cells, P-glycoprotein is able to transport its substrates against the concentration gradient back into the blood. Using abcb1a knock-out mice, Schinkel et al. (1994) showed that various substances, which are structurally entirely unrelated, are substrates of P-glycoprotein. These substances are exported against a concentration gradient out of the brain into systemic circulation. Our group showed for
Table 2 Organ concentrations of antidepressant drugs and the metabolite 1 h after subcutaneous injection (values are meanSEM) Substance
Organ
abcb1ab / [ng/ml]
abcb1ab+/+ [ng/ml]
Ratio
P-value
meanSEM
meanSEM
plasma cerebrum spleen kidney liver
170.99.4 1666.275.3 1926.2114.2 2142.9165.2 358.229.8
198.49.0 1388.889.8 2291.3245.8 2498.7131.1 410.130.5
0.9 1.2 0.8 0.9 0.9
n.sa * n.s n.s n.s
plasma cerebrum spleen kidney liver
14.5 1.2 48.4 3.7 91.4 5.3 119.98.6 76.6 6.7
12.01.4 33.23.7 89.16.3 125.011.3 83.86.2
1.2 1.5 1.0 1.0 0.9
n.s * n.s n.s n.s
plasma cerebrum spleen kidney liver
28.9 1.9 455.115.5 555.942.5 426.830.0 128.16.5
33.21.5 153.58.3 610.948.4 500.644.3 128.58.8
0.9 3.0 0.9 0.9 1.0
n.s * n.s n.s n.s
Trimipramine
D-Trimipramine
Citalopram
a
n.s not significant. * P <0.05.
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
183
Fig. 2. Per cent change in antidepressant concentration. Black columns are abcb1ab ( / ) knock-out mice, grey column wild-type mice, generally set to 100%. pla=plasma, cer=cerebrum, spl=spleen, kid=kidney and liv=liver.
the first time in vivo that some antidepressants are substrates of P-gp and that the absence or presence of P-gp affects the cerebral bioavailability of antidepressants. Other antidepressants, such as fluoxetine, are not substrates of P-glycoprotein. Also, P-glycoprotein is not present in all organs and in certain organs, for example in, kidney or liver, P-gp is not located in endothelial cells. In our study, concentrations of all three substances did not differ significantly in spleen, kidney or liver. Plasma concentrations also showed no significant differences. This supports the hypothesis that the antidepressants are exported against the concentration gradient from the intracerebral into the extracerebral space via the BBB and that the observed effect is a specific effect of transport across the BBB and not due to passive mechanisms. One has to take into account that in this study, the substances have been administered in a dose that is close to the upper limit of what can be used in therapy in humans. In our study, the coefficient of variation of the cerebral drug concentrations was approximately 20% in both groups. This surprisingly low coefficient of variation can be attributed to measuring methods and the
characteristics of the laboratory animals. In both groups, the mice share the same genetic background and live under standardized conditions. Presumably, this creates a relatively homogenous group, in which expression and function of the P-glycoprotein is similar in all individuals. In general, only certain substances are substrates of Pglycoprotein, and not all substrates show the same quantitative interactions. When concentration in the brain tissue is compared between both, the abcb1ab( / ) and the wild type mouse group, the ratio is quite low for trimipramine with 1.2 and d-trimipramine with 1.5, i.e. if P-gp is absent, the concentration of the drug in the brain is 1.2 and 1.5 times higher, respectively. The ratio is much higher for citalopram (3.0). We showed recently for the antidepressant amitriptyline and its metabolites that the ratio of concentration between knock-outs and wild-type ranges between 1.8 and 4.5 (Uhr et al., 2000). This ratio is not very high when compared with substrates with a high affinity for P-glycoprotein, such as cytostatic agents, digoxine or invermectine. However, one would easily expect that a 100–300% change in cerebral bioavailability of antidepressants could be of clinical relevance in humans.
184
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185
Moreover, the therapeutic picture in humans in general is more complex, among the complicating factors are newly discovered polymorphisms and co-medications that affect P-gp activity. Frequently, several CNS drugs are co-administered, and often additional, nonCNS drugs are added. In recent studies, polymorphisms of ABCB1 in humans have been described. In human studies, these polymorphisms have been shown to affect intestinal uptake of the P-gp substrate digoxin (Hoffmeyer et al., 2000; Cascorbi et al., 2001; Kerb et al., 2001b), to affect phenytoin plasma levels in healthy volunteers (Kerb et al., 2001a), and to be associated with orthostatic hypotension in patients with depression treated with nortriptyline (Roberts et al., 2002). Co-medications, such as the popular non-prescription antidepressant St. John’s Wort can increase expression of P-gp (Hennessy et al., 2002) in humans and the P-gp inhibitor quinidine reverses the lack of central respiratory depression of loperamide (Sadeque et al., 2000). It can therefore be suspected that co-medications counteract therapeutic efforts in humans, explaining inter-individual differences in the response to drugs that are substrates of P-glycoprotein. Besides this human data, many more modulating effects have been inferred from cell culture and animal experimental studies (Drewe et al., 2000; Ibrahim et al., 2000; Lentz et al., 2000; Dagenais et al., 2001; van der Sandt et al., 2001). Here, as well in cancer therapy, a number of agents have successfully been used to suppress the activity of the P-glycoprotein (Ford and Hait, 1990) or to increase the expression of the P-gp (Seelig, 1998). Pretreatment of mice with the P-gp inhibitor cyclosporine enhances and prolongs opiate-induced analgesia in mice (Thompson et al., 2000). For this reason, plasma concentrations of antidepressants alone do not allow for a definitive statement about the concentration in the brain. A patient with normal plasma concentrations of citalopram may in fact be receiving an insufficient dose because the penetration into the brain is reduced by an active export of the antidepressant drug through the P-gp located at the blood–brain barrier. In addition to varying pharmacobiology of the patients, this is another reason why therapy resistance could arise. Plasma levels and therapeutic drug monitoring are therefore important tools for any physician administering CNS drugs, but additional information about possible interactions with P-gp would be helpful for the clinician. It is important to note that so far, it is not possible to predict from the structure of the molecule which substance will be a substrate of P-glycoprotein and which one will not be (Fig. 1). Therefore, it is currently necessary to test all substances used as antidepressants in order to gather more data about the interaction between P-glycoprotein and antidepressants. Such a survey is currently in progress.
Because of the complexity of the blood–brain barrier, it is difficult to investigate the interactions between the blood–brain barrier and drugs. An earlier in vitro study, using monolayers of bovine brain microvessel endothelial cells (BMECs), had reported that citalopram is not a substrate for active efflux mechanisms (Rochat et al., 1999).This was based on the observation that cyclosporin, a P-glycoprotein inhibitor, did not modify the transport of citalopram. Our results, however, show that citalopram is exported actively, similar to amitriptyline. This shows that for the investigation of the blood–brain barrier, in vivo studies are indispensable and results of in vitro studies cannot easily be transferred to the in vivo situation, due to the complexity of the blood–brain barrier. In sum, the present study shows that CNS drugs, such as antidepressants, can be substrates of P-gp at the blood–brain barrier and are actively exported out of the brain. Together with the new human, animal and cell culture data about P-gp, its activity and substrates, polymorphisms and effect of co-medications, it becomes increasingly clear that the complexity of the blood– brain barrier has to be taken into account in the administration of CNS drugs more than before, even if the CNS drugs used have been shown to enter the brain and to have a therapeutic effect. Future research in this field and more data about these interactions will probably allow in the future to tailor CNS pharmacotherapy more individually to each patient, to optimize therapeutic efficacy and to minimize undesired side effects. The rapid progress in pharmacogenomics will accelerate this development towards an individual, pharmacogenomically oriented therapy.
Acknowledgements We would like to thank Ms. A. Rippl and Ms. M. Ha¨usler for their very helpful technical assistance.
References Cascorbi I, Gerloff T, Johne A, Meise C, Hoffmeyer S, Schwab M, et al. Frequency of single nucleotide polymorphisms in the P-glycoprotein drug transporter MDR1 gene in white subjects. Clinical Pharmacology and Therapeutics 2001;69:169–74. Cordon-Cardo C, O’Brien JP, Boccia J, Casals D, Bertino JR, Melamed MR. Expression of the multidrug resistance gene product (Pglycoprotein) in human normal and tumor tissues. Journal of Histochemistry and Cytochemistry 1990;38:1277–87. Dagenais C, Zong J, Ducharme J, Pollack GM. Effect of mdr1a P-glycoprotein gene disruption, gender, and substrate concentration on brain uptake of selected compounds. Pharmaceutical Research 2001;18:957–63. Devault A, Gros P. Two members of the mouse mdr gene family confer multidrug resistance with overlapping but distinct drug specificities. Molecular and Cellular Biology 1990;10:1652–63.
M. Uhr, M.T. Grauer / Journal of Psychiatric Research 37 (2003) 179–185 Drewe J, Ball HA, Beglinger C, Peng B, Kemmler A, Schachinger H, et al. Effect of P-glycoprotein modulation on the clinical pharmacokinetics and adverse effects of morphine. British Journal of Clinical Pharmacology 2000;50:237–46. Ford JM, Hait WN. Pharmacology of drugs that alter multidrug resistance in cancer. Pharmacological Reviews 1990;42:155–99. Fromm MF. P-glycoprotein: a defense mechanism limiting oral bioavailability and CNS accumulation of drugs. International Journal of Clinical Pharmacology and Therapeutics 2000;38:69–74. Gottesman MM, Hrycyna CA, Schoenlein PV, Germann UA, Pastan I. Genetic analysis of the multidrug transporter. Annual Review of Genetics 1995;29:607–49. Habgood MD, Begley DJ, Abbott NJ. Determinants of passive drug entry into the central nervous system. Cellular and Molecular Neurobiology 2000;20:231–53. Hennessy M, Kelleher D, Spiers JP, Barry M, Kavanagh P, Back D, et al. St John’s wort increases expression of P-glycoprotein: implications for drug interactions. British Journal of Clinical Pharmacology 2002;53:75–82. Hoffmeyer S, Burk O, von Richter O, Arnold HP, Brockmoller J, Johne A, et al. Functional polymorphisms of the human multidrugresistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo. Proceedings of the National Academy of Sciences of the United States of America 2000;97:3473–8. Ibrahim S, Peggins J, Knapton A, Licht T, Aszalos A. Influence of antipsychotic, antiemetic, and Ca(2+) channel blocker drugs on the cellular accumulation of the anticancer drug daunorubicin: P-glycoprotein modulation. Journal of Pharmacology and Experimental Therapeutics 2000;295:1276–83. Karssen AM, Meijer OC, van der Sandt I, Lucassen PJ, de Lange ECM, de Boer AG, et al. Multidrug resistance P-glycoprotein hampers the access of cortisol but not of corticosterone to mouse and human brain. Endocrinology 2001;142:2686–94. Kerb R, Aynacioglu AS, Brockmoller J, Schlagenhaufer R, Bauer S, Szekeres T, et al. The predictive value of MDR1, CYP2C9, and CYP2C19 polymorphisms for phenytoin plasma levels. The Pharmacogenomics Journal 2001a;1:204–10. Kerb R, Hoffmeyer S, Brinkmann U. ABC drug transporters: hereditary polymorphisms and pharmacological impact in MDR1, MRP1 and MRP2. The Pharmacogenomics Journal 2001b;2:51–64. Lentz KA, Polli JW, Wring SA, Humphreys JE, Polli JE. Influence of passive permeability on apparent P-glycoprotein kinetics. Pharmaceutical Research 2000;17:1456–60. Meijer OC, de Lange EC, Breimer DD, de Boer AG, Workel JO, de Kloet ER. Penetration of dexamethasone into brain glucocorticoid targets is enhanced in mdr1A P-glycoprotein knockout mice. Endocrinology 1998;139:1789–93. Mukhopadhyay T, Batsakis JG, Kuo MT. Expression of the mdr (Pglycoprotein) gene in Chinese hamster digestive tracts. Journal of the National Cancer Institute 1988;80:269–75. Roberts RL, Joyce PR, Mulder RT, Begg EJ, Kennedy MA. A common P-glycoprotein polymorphism is associated with nortriptylineinduced postural hypotension in patients treated for major depression. The Pharmacogenomics Journal 2002;2:191–6. Rochat B, Baumann P, Audus KL. Transport mechanisms for the antidepressant citalopram in brain microvessel endothelium. Brain Research 1999;831:229–36. Sadeque AJ, Wandel C, He H, Shah S, Wood AJ. Increased drug delivery to the brain by P-glycoprotein inhibition. Clinical Pharmacology and Therapeutics 2000;68:231–7. Saunders NR, Habgood MD, Dziegielewska KM. Barrier mechanisms
185
in the brain, I. Adult brain. Clinical and Experimental Pharmacology and Physiology 1999;26:11–19. Schinkel AH. P-glycoprotein, a gatekeeper in the blood–brain barrier. Advanced Drug Delivery Reviews 1999;36:179–94. Schinkel AH, Mayer U, Wagenaar E, Mol CA, van Deemter L, Smit JJ, et al. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proceedings of the National Academy of Sciences of the United States of America 1997;94:4028–33. Schinkel AH, Smit JJ, van Tellingen O, Beijnen JH, Wagenaar E, van Deemter L, et al. Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood–brain barrier and to increased sensitivity to drugs. Cell 1994;77:491–502. Schinkel AH, Wagenaar E, Mol CA, van Deemter L. P-glycoprotein in the blood–brain barrier of mice influences the brain penetration and pharmacological activity of many drugs. Journal of Clinical Investigation 1996;97:2517–24. Seelig A. A general pattern for substrate recognition by P-glycoprotein. European Journal of Biochemistry 1998;251:252–61. Tamai I, Tsuji A. Transporter-mediated permeation of drugs across the blood–brain barrier. Journal of Pharmaceutical Sciences 2000; 89:1371–88. Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, Willingham MC. Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proceedings of the National Academy of Sciences of the United States of America 1987;84:7735–8. Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, Willingham MC. Immunohistochemical localization in normal tissues of different epitopes in the multidrug transport protein P170: evidence for localization in brain capillaries and crossreactivity of one antibody with a muscle protein. Journal of Histochemistry and Cytochemistry 1989;37:159–64. Thompson SJ, Koszdin K, Bernards CM. Opiate-induced analgesia is increased and prolonged in mice lacking P-glycoprotein. Anesthesiology 2000;92:1392–9. Uhr M, Holsboer F, Mu¨ller MB. Penetration of endogenous steroid hormones corticosterone, cortisol, aldosterone and progesterone into the brain is enhanced in mice deficient for both mdr1a and mdr1b P-glycoproteins. Journal of Neuroendocrinology 2002;14: 753–9. Uhr M, Steckler T, Yassouridis A, Holsboer F. Penetration of amitriptyline, but not of fluoxetine, into brain is enhanced in mice with blood–brain-barrier deficiency due to Mdr1a P-glycoprotein gene disruption. Neuropsychopharmacology 2000;22:380–7. van Asperen J, Mayer U, van Tellingen O, Beijnen JH. The functional role of P-glycoprotein in the blood–brain barrier. Journal of Pharmaceutical Sciences 1997;86:881–4. van de Vrie W, Marquet RL, Stoter G, de Bruijn EA, Eggermont AM. In vivo model systems in P-glycoprotein-mediated multidrug resistance. Critical Reviews in Clinical Laboratory Sciences 1998;35:1–57. van der Sandt I, Vos CM, Nabulsi L, Blom-Roosemalen MC, Voorwinden HH, de Boer AG, et al. Assessment of active transport of HIV protease inhibitors in various cell lines and the in vitro blood– brain barrier. AIDS 2001;15:483–91. von Moltke LL, Greenblatt DJ. Drug transporters in psychopharmacology—are they important? Journal of Clinical Pharmacology 2000;20:291–4. Wijnholds J, de Lange EC, Scheffer GL, van den Berg DJ, Mol CA, van der Valk MA, et al. Multidrug resistance protein 1 protects the choroid plexus epithelium and contributes to the blood–cerebrospinal fluid barrier. Journal of Clinical Investigation 2000;105:279–85.