! "  #$# #  %###&      

             

'        !" #$ "%    " %  

#( )*+,-. ("+

" " +./-'. " . 0"1.-+. 00"1" 2332 4

Dissertation for the Degree of Doctor of Philosophy (Faculty of Medicine) in Psychiatry presented at Uppsala University in 2002

ABSTRACT

Yan, H. 2002. Stereoselective Transport of Drugs Across the Blood-Brain Barrier (BBB) In Vivo and In Vitro. Pharmacokinetic and pharmacodynamic studies of the (S)- and (R)-enantiomers of different 5-HT1A receptor agonists and antagonists. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1138. 60 pp. Uppsala. ISBN 91-554-5280-9.

Delivery of drugs to the brain requires to passage across the blood-brain barrier (BBB), which include the specialized brain endothelial cell layer with complex tight intercellular junctions. Both for drugs already on the market and for new drugs under development, it is important to know to what extent a drug enters the CNS. Many drugs used clinically are racemic mixtures (equal parts of the (S)- and (R)-enantiomers). The present studies focus on the enantiomers and racemates of a number of 5-HT1A receptor agonists and antagonists (pindolol, propranolol, 8-OH-DPAT and other 8-substituted-2-(di-n- propylamino)tetralin derivatives) and BBB transport in vitro and distribution to the brain in vivo. The relationships between drug concentrations and pharmacodynamic effects were also studied. Assays (HPLC-based) were set up or developed for determination of the racemates and the pure enantiomers (chiral column) of drugs in plasma and brain tissue. BBB transport was assessed in vitro using bovine brain endothelial cells cocultured with rat astrocytes. The physicochemical constants (log P, pKa) were determined and plasma protein binding was assessed by equilibrium dialysis. The two enantiomers of pindolol and propranolol (having the same log P value) distributed stereoselectively to the brain. Pretreatment with verapamil (a P-glycoprotein blocker) differentially affected the brain/plasma ratios of the enantiomers of pindolol and propranolol, indicating that transport mechanisms are involved. In vitro results with propranolol and the tetralins showed differences in BBB transport between their enantiomers. A more rapid apical to basolateral transport (influx) vs. the basolateral to apical (efflux) transport may be attributed to active transport mechanisms. In conclusion, some but not all chiral drugs are stereoselectively distributed to the brain. This may be due to stereoselective plasma protein binding or stereoselective transport across the BBB, and not related to lipophilicity. Apical to basolateral transport was more rapid than basolateral to apical transport in vitro for propranolol (not for pindolol) and several tetralin derivatives. The stereochemical configuration of compounds contributes to their uniqueness both with regard to pharmacokinetic and pharmacodynamic properties.

Key words: pindolol, propranolol, 2-(di-n-propylamino)tetralin, racemate, enantiomer, drug distribution, brain, rat, blood-brain barrier

Hongmei Yan, Department of Neuroscience, Psychiatry, Ulleråker, Uppsala University, SE-750 17, Uppsala, Sweden

 Hongmei Yan 2002 ISSN 0282-7476 ISBN 91-554-5280-9 Printed in Sweden by Uppsala University, Tryck & medier, Uppsala 2002 5

Contents

Papers included in the thesis 7

Abbreviations 8

1. Introduction 9 1.1. Physiology and anatomy of the BBB 9 1.2. BBB transport mechanisms 10 1.2.1. Lipid-mediated transport (passive diffusion) and plasma protein binding 12 1.2.2. Catalyzed transport through the BBB 13

1.2.3. Active efflux pumps 14 1.2.4. Methods to study BBB transport of drugs in vivo and in vitro 15 1.3. 5-HT neurons and receptors 16 1.3.1. 5-HT neurons and pathways 16 1.3.2. 5-HT receptors 17

1.3.3. 5-HT1A receptor distribution 18

1.3.4. 5-HT1A receptor pharmacology 18 1.4. Stereoselctive drugs 19 1.4.1. Definition of racemate and enantiomer 19 1.4.2. Importance of chiral drug research 21

2. Aims of the thesis 22

3. Methods and Materials 23 3.1. Compounds used in the thesis 23 3.2. Animals 23 3.3. Chromatographic systems 23 3.4. Bioanalysis of drug (racemates and enantiomers) concentrations in brain tissue and plasma 25 6

3.5. Biosynthesis of 5-HT and DA (Paper II) 27 3.6. Pharmacodynamic studies (Paper III) 28 3.6.1. Behavioral experiments 28 3.6.2. Body temperature 28 3.7. Blood-brain barrier experiments in vitro 28 3.7.1. Cell culture 29 3.7.2. Transport experiments 30 3.7.3. Data analysis and calculation 31 3.8. Plasma protein binding assay 32 3.9. Statistical methods and calculation of pharmacokinetic parameters 33

4. Results and Discussion 34 4.1. The first stereoselective study of racemic pindolol to rat brain tissue (Paper I ) 34 4.2. Differential distribution of enantiomers of racemic pindolol and propranolol to the rat brain (Paper III) 34 4.3. Relationship between pharmacokinetic and pharmacodynamics of (S)-UH-301 (Paper II) 41 4.4. Passage of tetralins across the BBB in vivo and in vitro (Paper IV) 43

5. Conclusions 48

Acknowledgements 51

References 53 7

Papers included in the thesis

This thesis is based on the following papers, which are referred to in the text by their Roman numerals:

I Yan H. and Lewander T. (1999) Differential tissue distribution of the enantiomers of racemic pindolol in the rat. European Neuropsychopharmacology 10: 59-62.

II Yan H., Yu, H. and Lewander T. (2002) An integrative pharmacokinetic and

pharmacodynamic study of the 5-HT1A receptor antagonist (S)-UH-301 in the rat. European Neuropsychopharmacology 12: 101-110.

III Yan H., Lundquist S., Renftel M. and Lewander T. (2002) Differential distribution of drug enantiomers to the rat brain: A comparative study of (R,S)-pindolol and (R,S)- propranolol. Submitted for publication.

IV Yan H., Renftel M, Lundquist S., Sundholm G., Liu Y. and Lewander T. (2002) Studies of the passage across the blood-brain barrier in vivo and in vitro of the racemates and the pure enantiomers of a series of 8-substituted 2-(di-n- propylamino)tetralins. Submitted for publication.

Reprints were made with permission from the publisher Elsevier Science. 8

Abbreviations

α1- AGP α1-acid glycoprotein ANOVA Analysis of variance ATP Adenosine triphosphate AUC Area under the curve BBB Blood-brain barrier CNS Central nervous system CSF Cerebrospinal fluid

Cmax Maximal concentration DRN Dorsal raphe nucleus DA Dopamine EC Endothelial cell HPLC-EC High performance liquid chromatography with electrochemical detection 5-HT 5-hydroxytryptamine 5-HTP 5-hydroxytryptophan 8-OH-DPAT 8-hydroxy-2-(di-n-propylamino)tetralin LY-39 2-(di-n-propylamino)tetralin (DPAT) LY-41 8-acetyl-2-(di-n-propylamino)tetralin (8-acetyl-DPAT) LY-49 8-phenyl-2-(di-n-propylamino)tetralin (8-phenyl-DPAT) UH-301 5-fluoro-8-hydroxy-2-(di-n-propylamino)tetralin (5-fluoro-8- OH-DPAT) log P Partition coefficient MAOIs Monoamine oxidase inhibitors MDR Multidrug resistance MRN Median raphe nucleus MRP Multidrug resistance protein MCT Monocarboxylic acid transporter Pe Permeability coefficient PS Permeability surface area product P-gp P-glycoprotein pKa Dissociation constant SSRIs Selective serotonin reuptake inhibitors SD Standard deviation s.c. Subcutaneous

t1/2 Half life 9

1. Introduction

The blood-brain barrier (BBB) is a set of mechanisms that protects the brain from endogenous or exogenous substances that might be toxic to the function of the central nervous system (CNS). In order to exert its action, a drug delivered to the brain requires to passage across the BBB with its specialized endothelial cell layer, enters the extracellular space, and reaches its cellular and molecular target by diffusion through the extracellular matrix. Both for drugs already on the market and for new drugs under development, it is important to know to what extent they enter the CNS.

1.1. Physiology and anatomy of the BBB

The concept of a barrier between the blood and the brain arose in the end of the nineteenth century when the German pharmacologist and physiologist Paul Ehrlich observed that certain dyes, e.g., a series of aniline derivatives, administered intravenously to small animals stained all the organs except the brain. Later, Ehrlich’s experiments were repeated by Goldmann, who injected the dye trypan blue directly into the cerebrospinal fluid of rabbits and dogs. The dye readily stained the entire brain but did not enter the bloodstream to stain the other internal organs. This experiment showed that the central nervous system is separated from the blood by a barrier of some kind. He concluded that the cerebrospinal fluid (CSF) and the choroids plexus were responsible for nutrient transport into the CNS and that the choroid plexus must possess a barrier function for substances which would be toxic to the CNS (Bradbury, 1979). In the late 1960’s, the ultrastructural basis for the barrier was established and electron microscopic studies showed that the BBB is an endothelial barrier present in capillaries that course through the brain (Reese and Karnovsky, 1967; Brightman and Reese, 1969). Ultrastructural studies also pointed out the main characteristics by which endothelial cells in brain differ fundamentally from those in most peripheral tissues: (a) the presence of complex tight junctions, which severely restricts transport through the intercellular space of the endothelial cells, (b) the very few endocytotic vesicles, which limits the transcellular transport of substances, and (c) an absence of fenestrations. The brain 10 capillary endothelial cell contains five times more mitochondria than other systemic capillaries (Oldendorf et al., 1977). The study of the cell biology of the BBB phenomena showed that astrocytes (Janzer and Raff, 1987), which are the cells closest to brain capillary endothelial cells and whose endfeet cover much of capillary’s basal surface, influence the tightness of the tight junctions when co-cultured with cerebrovascular cells (Minakawa et al., 1991).

The BBB is composed of two membranes in series: the luminal and abluminal membranes of the brain capillary endothelial cell, which are separated by approximately 300 nm of endothelial cytoplasm. Therefore, transport systems must exist on both luminal and abluminal membranes of the endothelail cell if solute transcytosis from blood to brain is to occur (Fig.1).

Beside the BBB, drug transport into the CNS is influenced by other barriers in the brain e.g. blood-CSF barrier (Brightman and Reese, 1969). However, the surface area of BBB cerebral capillaries is 5000 times larger than that of the blood-CSF barrier (Pardridge, 1986). In the human brain, there are approximately 400 miles of brain capillaries. The BBB’s large surface area makes it an important barrier for drug delivery to the brain.

1.2. BBB transport mechanisms

Anatomically the BBB is not a static, homogeneous, impermeable barrier. Its permeability is dynamically regulated with special features (physical and metabolic) arising from the absence of fenestrations, the relative lack of pinocytotic vesicles, the presence of tight junctions, abundance of mitochondria, and from the influence of astrocytes, pericytes, neurons, and blood constituents such as hormones (Van Bree et al., 1992a). These regulatory systems may have some influence on permeability and so affect drug delivery to the CNS. The tight junctions eliminate a paracellular pathway of drug movement through the BBB and the virtual absence of pinocytosis across the brain capillary endothelium eliminates transcellular bulk flow of circulating solute through the BBB. Under these conditions, a solute may gain access to brain via two pathways: lipid 11 Endothelium Astrocyte

WBC RBC Alb Tc Astrocyte P α1−AGP

D

Neuron terminal

Blood D-glucose transporter Receptor P-gp Tight Passive mediated junction transporter diffusion L-amino acid Luminal side

Endothelial cell layer Abluminal side Basal membrane

Postsynaptic Brain receptors Astrocyte endfoot

Presynaptic receptors

Figure 1. Diagram of the Blood-Brain Barrier and transport mechanisms associated with the endothelial cell layer. [ P ]- Protein, [ D]- Drug. Alb= Albumin, Tc = Thrombocyte, WBC = White blood cell, RBC = Red blood corpuscles. 12 mediation (passive diffusion) or catalyzed transport. Catalyzed transport includes carrier- mediated or receptor-mediated processes (Pardridge, 1999). Drugs can be transported out of the brain by active efflux. P-glycoprotein is one such efflux transporter (Fig. 1.).

1.2.1. Lipid – mediated transport (passive diffusion) and plasma protein binding

Most drugs are transported into the brain by passive diffusion across the endothelium through the cells themselves or through tight junctions between the cells. Three major factors are involved in this process: drug lipophilicity (log P) and ionization (pH, pKa), molecular size and plasma protein binding (Manners et al., 1988; Lombardo et al., 1996; Tsuji and Tamai, 1998).

The relative affinity between water and lipid, for example, as expressed by its octanol/water partition coefficient (P), is an important factor, because it determines the ease by which the molecule is able to permeate the lipid membranes of the endothelial cells. The penetration rate into the brain is generally, but not always, proportional to the lipid solubility of a molecule. The use of the log P as a reliable index of BBB transport of lipid-soluble small molecules has been shown to be valid only when the molecular weight of a compound is less than a threshold of approximately 400-600 Da (Levin, 1980). The less the degree of hydrogen bonding with water, the greater the possibility of partitioning of the drugs to the membrane (Abraham et al., 1994; van de Waterbeemd et al., 1998).

For a long time plasma protein binding was considered a non-complicating factor in BBB transport: only the free drug was thought to be available for transport, since the BBB was impermeable to the drug-protein complex (Oldendorf, 1974). Plasma protein binding is generally measured with such in vitro techniques as equilibrium dialysis or centrifugal filtration. However, a number of protein-bound hormones and drugs showed a higher brain uptake than expected from their free plasma concentration (Cornford et al., 1979; Pardridge and Landaw, 1985; Greig, 1992). This is attributed to predicted conformational changes in the plasma protein drug binding site during its transcapillary passage, which in turn result in enhanced dissociation of the drug from the plasma protein binding site within the brain microcirculation (Pardridge, 1987). Most CNS active drugs are lipophilic amines and these compounds avidly bind to a plasma globulin called α1-acid 13

glycoprotein (α1-AGP, also referred to as orosomucoid (Lin et al., 1987)). In the case of propranolol, the albumin-bound drug is not available for transport through the BBB, whereas α1-AGP-bound propranolol is available for transport into the brain (Pardridge, 1987). This tissue-mediated, enhanced dissociation of compounds from plasma proteins has been termed protein-mediated transport (Pardridge, 1987).

1.2.2. Catalyzed transport through the BBB

Carrier-mediated transport

Carrier-mediated transport can be facilitated diffusion and active transport. Facilitated diffusion is non-energy dependent and the compound is transported along a concentration gradient, while active transport is energy dependent and is directed against a concentration gradient (Pardridge, 1999).

The brain capillary endothelial cells have several transport systems regulating entry into the brain across the BBB for nutrients and endogenous compounds such as amino acids, monocarboxylic acids, amines, hexoses, thyroid hormones, purine bases, and nucleosides (Pardridge, 1995; Bergley, 1996).

At least ten different transport systems have been identified, e.g. GLUT-1, system-L, system-A (Van Bree et al., 1992a). Each of the carriers transports a group of different nutrients of the same structure. These transporters are also responsible for the absorption of hydrophilic drugs (Tsuji and Tamai, 1996). It also transports neutral amino acid-type drugs, such as L-DOPA (Rapoport and Pettigrew, 1979), which is used to augment brain dopamine levels. Interestingly, some transporters show stereoselectivity, e.g. D- vs. L- glucose and L- vs. D- amino acids.

Receptor-mediated transport

A number of specific receptors are expressed both on the luminal and abluminal surfaces of the endothelial cells, such as insulin- and transferrin receptors. These receptors may play a role as transport vectors or modulate the activity of channels or transporters at the 14

BBB (Pardridge, 1995). Thus, macromolecules can be transported into the brain by this mechanism.

1.2.3. Active efflux pumps

Multidrug resistance (MDR) describes the simultaneous expression of cellular resistance to a variety of unrelated drugs primarily of natural origin (Cole et al., 1992). Proteins involved in MDR mechanisms are P-glycoprotein (P-gp), Multidrug Resistance- associated Proteins (MRPs), Major Vault Protein (MVP)/Lung Resistance-related Protein (LRP) and Breast Cancer Resistance Protein (BCRP) (van den Heuvel-Eibrink et al., 2000). P-gp and MRP belong to the ATP-binding cassette (ABC) superfamily of membrane transporter proteins, which excludes from cells a wide variety of anticancer drugs and hydrophobic molecules and decreases their intracellular concentration (Abe et al., 1997; Cole and Deeley, 1998).

P-glycoprotein (P-gp), the product of the MDR1 gene, is a 170kDa membrane protein. Except for tumor cells, P-gp is present in normal tissues such as the apical membrane of intestinal epithelial cells, adrenal gland of mice and humans (not rats), kidney and brain (Schinkel, 1999). In brain, it is expressed at the luminal membrane of the brain capillary endothelial cells, which accounts for the apparently poor BBB permeability for certain drugs or limit the entry of potentially toxic compounds from blood into the brain by pumping them actively back into the blood (Schinkel et al., 1994). P-gp interacts with a range of hydrophobic substrates, either endogenous (hydrophobic peptides and steroids) or exogenous (cyclosporin A (CsA), vinblastin, rhodamine-123) (Demeule et al., 2000). P-gp (mdr1a) knock-out mice studies also showed that P-gp mediates the efflux of some drugs, e.g. quinidine, across the BBB (Kusuhara et al., 1997).

Multidrug resistance protein (MRP) is also an ATP-dependent efflux transporter, which has seven family members that are present in different tissues, such as liver, kindey and gut (Borst et al., 1999). MRP1 is ubiquitous in the body. Expression of the MRP gene (mainly MRP1) was demonstrated in bovine brain capillary endothelial cells by both functional and biochemical analyses (Huai-Yun et al., 1998). Various anionic drugs have 15 been demonstrated to be effluxed from the brain by MRP family members other than P- gp. These drugs include tolbutamide, valproic acid, p-aminohippuric acid and others (Tamai and Tsuji, 2000).

In addition to P-gp and MRP, the monocarboxylic acid transporter MCT1 may be responsible for the transport of some organic anions from brain to endothelial cell or from endothelial cells to blood because it is expressed on both the luminal and abluminal membranes of brain capillary endothelial cells. It seems to transport substrates bidirectionally (Tamai et al., 1999). Brain to blood efflux transport of valproic acid and probenecid might involve the participation of MCT1.

Clarification of the transport mechanism of drugs across the BBB is important to improve the efficacy of CNS-active drugs or to reduce the CNS toxicity of drugs that are active in peripheral tissues. In addition to utilization of influx transport systems, prevention of efflux transport by the use of specific inhibitors (such as verapamil, cyclosporin A and flufenazine as P-gp blockers, probenecid, sulfobromophthalein and indomethacin as MRP inhibitors) is also an attractive approach to improve the brain distribution of drugs (Ecker and Chiba, 1995; Tsuji and Tamai, 1996; Tamai and Tsuji, 2000).

1.2.4. Methods to study BBB transport of drugs in vivo and in vitro

During the past decades a number of methods for studying brain uptake of drugs and other substances have been developed (Van Bree et al., 1992b; Pardridge, 1998). Influx of drugs can be studied by the brain uptake index (BUI) (Fenstermacher et al., 1981) or carotid artery single injection technique (Cornford, 1998). Efflux of drugs (brain efflux index, BEI) can be studied by intracerebral injection of 0.5 – 1 µl of a drug solution and measurement of its disappearance over 20 minutes (Terasaki, 1998). Microdialysis is an increasingly popular method since it allows determination of the extracellular concentrations of a drug over time (Hammarlund-Udenaes et al., 1997; Hammarlund- Udenaes, 2000). In situ brain perfusion and intravenous injection/pharmacokinetics provide additional aspects of drug distribution to the brain tissue (Begley, 1998; Bickel, 1998). Determination of the total concentration (extracellular and intracellular) of a drug 16 in a piece of homogenized brain tissue at different time points after s.c. drug administration is used in the present study. Another method is to study BBB transport in vitro in tissue culture. Endothelial cells from the bovine brain have been co-cultured with astrocytes in order to get a monolayer of endothelial cells with the brain vascular characteristics (Dehouck et al., 1990; Van Bree et al., 1992b; Cecchelli et al., 1999). Using this method, it is possible to study the transport from the apical surface to the basolateral surface (the cell surface attached to a collagen coated filter, resembling attachment to the basal membrane of capillaries), or vice versa. There is still limited experience with the in vitro methodology in the published literature (van Bree et al., 1988a; van Bree et al., 1988b; Dehouck et al., 1995; Letrent et al., 1999) and functional characterization of each BBB model is mandatory (de Boer et al., 1999). There are few previous published studies of stereoselective transport of chiral drugs across brain endothelial cells in vitro according to our knowledge (Rochat et al., 1999). The possible formation of metabolites of compounds under study in brain endothelial cells (Ghersi-Egea et al., 1994; Ghersi-Egea et al., 1995) and their potential effects on interference with BBB transport mechanisms remain to be investigated.

1.3. 5-HT neurons and receptors

1.3.1. 5-HT neurons and pathways

5-HT (serotonin) was discovered in the 1940s by Erspamer in Italy. It was isolated from blood serum and shown to have a tonic or constrictive effect on blood vessels, which is why it was named serotonin. Later on it was found that 5-HT was a neurotransmitter. Although only about 1% of the total amount of 5-HT in the body is present in the brain, serotonin has been shown to play an important role in the regulation of many physiological functions and behaviors, e.g. temperature control, circadian rhythms, sleep, cardiovascular regulation, pain, mood and it is involved in psychiatric disorders, such as depression, anxiety, panic and obsessive compulsive disorder (Zifa and Fillion, 1992).

There are two main 5-HT system projections to the forebrain originating from the dorsal raphe nucleus (DRN) and the median raphe nucleus (MRN) in the midbrain. Many 17 forebrain 5-HT terminals arise from the DRN, which contains about half of all the brain’s 5-HT containing cells. Striatum receives serotonergic afferents mostly from the DRN, while hippocampus, hypothalamus and cortex are innervated by both the DRN and MRN, predominantly by serotonergic neurons originating in the MRN (Azmitia and Segal, 1978; Molliver, 1987; Vertes et al., 1994). Moreover, it has been reported that there are two types of 5-HT neurons, which exhibit not only a differential distribution related to the raphe nuclei, but also a differential reaction to some drugs. The axons from the DRN cell bodies appear thin with minute granular varicosities and are vulnerable to certain neurotoxic amphetamine derivatives. The median raphe axons are coarse and have large varicosities, and are resisitant to neurotoxic agents (Molliver, 1987; Brown and Molliver, 2000).

1.3.2. 5-HT receptors

Considering the widespread distribution of serotonergic neurons and diverse physiological functions of 5-HT, it is not surprising that as many as 14 different receptor subtypes have been cloned (Pauwels, 2000). All except one (5-HT3, which is a ligand gated ion channel) of the 5-HT receptors couple to guanine nucleotide-binding proteins (G proteins), producing second messengers that regulate cellular functions via phosphorylation/dephosphorylation of intracellular proteins. Five families of G protein- coupled 5-HT receptors (5-HT1, 5-HT2, 5-HT4, 5-HT6 and 5-HT7) regulate two major intracellular second messenger pathways, adenylate cyclase (AC) and phospholipase C. It has become clear that the 5-HT1 (5-HT1A and 5-HT1B/1D) autoreceptors are central to the therapeutic actions of the 5-HT uptake inhibitors (SSRIs) and might be used as drug targets for the design of new CNS drugs.

There are currently five known subtypes of 5-HT1 receptor denoted by suffixes A, B, D,

E and F. Although the roles of the 5-HT1E and 5-HT1F receptors are uncertain, the remaining three are known to act as autoreceptors on 5-HT nerve terminals and /or on cell bodies. Extensive animal experimental data (Gozlan et al., 1983; Hoyer, 1988; Radja et al., 1992; Blier and de Montigny, 1994; Barnes and Sharp, 1999) showed that 5-HT1A receptors are importantly involved in the actions of antidepressant medication because 18

5-HT1A autoreceptors are invoved in regulating 5-HT cell firing and 5-HT release, while postsynaptic 5-HT1A receptors are located in cortical and limbic regions.

1.3.3. 5-HT1A receptor distribution

5-HT1A receptors are located both postsynaptic to 5-HT neurons (in forebrain regions), and also on the 5-HT neurons themselves at the level of the soma and dendrites in the DRN and MRN, which is named as somatodendritic autoreceptors (Pazos and Palacios,

1985; Kia et al., 1996). The density of 5-HT1A binding sites is high in limbic brain areas, notably hippocampus, lateral septum, cortical areas, and also the mesencephalic raphe nuclei. In contrast, levels of 5-HT1A binding sites in the basal ganglia and cerebellum are barely detectable (Barnes and Sharp, 1999), and low in the striatum as determined by autoradiographic studies (Pazos and Palacios, 1985). The hippocampal and cortical binding sites represent postsynaptic 5-HT1A receptor whereas the binding sites in raphe neuclei are somatodendritic 5-HT1A receptors (Radja et al., 1992).

1.3.4. 5-HT1A receptor pharmacology

The pharmacological characteristics of the 5-HT1A receptor clearly set it apart from other members of the 5-HT1 family and other 5-HT receptors (Hoyer et al., 1994). Selective 5-

HT1A receptor agonists include 8-OH-DPAT, dipropyl-5-CT and gepirone. Non-selective

5-HT1A receptor antagonists include spiperone and propranolol (Tricklebank et al., 1985;

Maura et al., 1986). A number of silent 5-HT1A receptor antagonists have now been developed, such as (S)-UH-301 (Hillver et al., 1990; Björk et al., 1991a) and WAY- 100635 (Fletcher et al., 1996). WAY-100635 is the most potent although NAD-299 appears to be somewhat more selective (Johansson et al., 1997). Some 5-HT1 receptor/β- adrenoceptor antagonists such as pindolol and penbutolol have varying degrees of efficacy at 5-HT1A receptors (Sanchez et al., 1996; Clifford et al., 1998). Microdialysis studies have established that 5-HT1A receptor antagonists facilitate the effect of SSRIs, monoamine oxidase inhibitors (MAOIs) and certain tricyclic antidepressants on 5-HT release (Hjorth, 1993; Artigas et al., 1996; Sharp et al., 1997). Clinical studies have reported that the therapeutic effect of antidepressants can be accelerated and augmented 19 by the adjunctive treatment with pindolol (Artigas et al., 1994; Bordet et al., 1998; Martinez et al., 2000; Artigas et al., 2001).

5-HT1A receptor activation causes neuronal hyperpolarisation, an effect mediated through the G-protein-coupled opening of K+ channels without the involvement of diffusible intracellular messengers such as cAMP, induces behavioural and physiological effects including the induction of the 5-HT syndrome, hypothermia, hyperphagia, altered sexual behavior (Hjorth, 1985; Middlemiss et al., 1985; Lucki, 1992) and leads to neurochemical responses such as inhibition of 5-HT release and increase of noradrenaline and acetylcholine release (Sharp et al., 1989; Bianchi et al., 1990; Done and Sharp, 1994).

Neuroendocrine studies in rats have shown that 5-HT1A receptor agonists cause an elevation of plasma ACTH, corticosteroids and prolactin (Gilbert et al., 1988), and in man there is also increased secretion of growth hormone (Cowen et al., 1990).

1.4. Stereoselective drugs

Since 1994, the β-adrenoceptor and 5-HT1A/1B receptor ligand pindolol has been used to accelerate or enhance the clinical effects of antidepressant drugs, such as the selective serotonin reuptake inhibitors (SSRIs) (Artigas et al., 1994; Zanardi et al., 1997; Bordet et al., 1998; Zanardi et al., 1998; Martinez et al., 2000; Artigas et al., 2001). Pindolol was initially thought to act by preventing the inhibition of 5-HT release elicited by SSRIs. However, pindolol is used clinically as a racemic mixture (equal parts of the (S)- and (R)- enantiomers). The pharmacodynamic profiles of the two enantiomers present stereoselectivity. Thus, (S)-pindolol is more potent than the (R)-enantiomers as β-receptor

(Clark et al., 1982) as well as 5-HT1A receptor antagonists (Hjorth and Carlsson, 1985).

1.4.1. Definition of racemate and enantiomer

Stereoselectivity has become a well-recognized consideration in clinical pharmacology. A carbon (or any other) atom bonded to four dissimilar atoms or substituents is said to be asymmetric. Drugs that have an asymmetric center within their molecular structure are 20 said to be chiral. ‘Chirality’ comes from the Greek cheir, means hand. The bonds formed by an asymmetric carbon atom can be arranged in two different ways in the three dimensional space, producing molecules that are mirror images of each other, called enantiomers. Enantiomers are non-superimposable, mirror image, ‘left-handed’ and ‘right-handed’ forms (Fox and White, 1994). A racemate is a 50:50 mixture of the two enantiomers. The pindolol molecular structure is used as an example (Fig.2.).

* O N

OH H

*- chiral center N H

(R,S)-pindolol

H H

O N O N

OH H OH H

N N H H

(R)-pindolol (S)-pindolol

Figure 2. Chemical structures of (R,S)-, (R)- and (S)-pindolol

The naming of optical isomers can be confusing. One method differentiates according to the direction of rotation of polarised light. Hence (+) or d for dextrorotatory and (-) or l for laevorotatory. A second, independent method specifies the absolute chemical configuration: (S)- and (R)-isomers. Although the latter designation derives from the Latin sinister and rectus, meaning left and right, this refers to the spatial orientation of groups at the chiral center and not to the direction of rotation of polarised light. Therefore, it is possible for an isomer to be S (+), S (-), R (+) or R (-). A third naming 21 system, also based on absolute configuration, uses the letters D (D-glucose) and L (L- Dopa), but is now rarely used.

1.4.2. Importance of chiral drug research

Many commonly used drugs with a chiral center are administered as racemates, 50:50 mixtures of the enantiomers. Although the two enantiomers have identical physicochemical properties, such as log P and pKa, the drug target macromolecules (e.g. receptors, enzymes, transport proteins) are often selective for the spatial arrangement of the small drug molecule. Therefore, drug enantiomers may differ markedly with regard to their pharmacodynamic or pharmacokinetic properties. For instance (S)-propranolol is more potent than the (R)-enantiomers as a β-receptor (Howe and Shanks, 1966) as well as a 5-HT1A receptor antagonist (Middlemiss, 1984). There are also differences between the propranolol enantiomers with regard to renal clearance, hepatic metabolism and plasma protein binding (Walle et al., 1988; Takahashi et al., 1990; Egginger et al., 1994). For the

5-HT1A receptor agonist 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT), (R)-8-

OH-DPAT is a 5-HT1A receptor full agonist, whereas (S)-8-OH-DPAT is a partial agonist

(Cornfield et al., 1991). (R)-UH-301 is a 5-HT1A receptor agonist whereas (S)-UH-301 is a 5-HT1A antagonist, making racemic UH-301 (5-fluro-8-hydroxy-2-(di-n- propylamino)tetralin) inactive in vivo (Hillver et al., 1990; Hillver et al., 1996).

For many racemic drugs used clinically at present there exist few items of knowedge concerning the pharmacological, toxicological and pharmacokinetic properties of their individual enantiomers (Eichelbaum, 1992; Slovakova and Hutt, 1999). Additional testing of the enantiomers of the racemates used in clinical practice can lead to the discovery of new indications of the original drug, improve its clinical use and result in increasing its safety and efficacy. Recent examples are escitalopram ((S)-citalopram), an SSRI developed from citalopram (Montgomery et al., 2001) and esomeprazole ((S)- omeprazole), from the proton pump inhibitor omeprazole (Tonini et al., 2001). 22

2. Aims of the Thesis

The overall objective of the present thesis was to study pharmacokinetics and pharmacodynamic/pharmacokinetic interrelationships for two 5-HT1A receptor antagonists, (S)-UH-301 and racemic pindolol, and a series of structurally related 2-(di-n- propylamino)tetralins and 5-HT1A receptor ligands previously investigated with regard to pharmacodynamic stereolectivity, with emphasis on drug concentrations in brain tissue including drug transport across the blood-brain barrier.

Specific studies focused on: - the pharmacokinetics and integrative pharmacodynamic/pharmacokinetic investigations of (S)-UH-301 after subcutaneous administration in the rat; - comparative investigations of the distribution to the brain and transport across the blood-brain barrier in vitro of the structurally related 2-(di-n-propylamino)tetralins; - the pharmacokinetics and integrative pharmacodynamic/pharmacokinetic

investigations of racemic pindolol, a β-adrenoceptor antagonist and partial 5-HT1A agonist used clinically to augment and accelerate the efficacy of antidepressant drugs; - comparative investigations of the distribution to the brain and transport across the blood-brain barrier in vitro of pindolol and propranolol and their enantiomers, which were prompted by an unexpected finding of stereoselective distribution to the brain of the (S)-enantiomer vs. the (R)-enantiomer of pindolol. 23

3. Methods and Materials

3.1. Compounds used in the thesis

The compounds studied in the present thesis are racemates and enantiomers of pindolol, propranolol, 8-hydroxy-(di-n-propylamino)tetralin (8-OH-DPAT), 5-fluro-8-OH-DPAT (UH-301), 8-acetyl-DPAT (LY-41), 8-phenyl-DPAT (LY-49) and DPAT (LY-39) (Fig.3.).

3.2. Animals

Male Sprague-Dawley rats (B&K UNIVERSAL AB, Sweden), weighing 240-270 g, were kept under constant temperature (23±1°C) and lighting (6:00 - 18:00) conditions and were allowed free access to food and water. Four or five rats were kept in each cage (55×35×20 cm) and were habituated to the environment for at least a week prior to the experiments. Rats were decapitated without anaesthesia at predetermined intervals after subcutaneous injection of the test compounds. Injection volumes were 2 ml/kg. The studies were approved by the Local Ethics Committee (Uppsala) of the Swedish National Board for Laboratory Animals.

3.3. Chromatographic systems

Non-stereoselective HPLC analysis of racemic drugs was performed using a LKB 2150 HPLC pump, a Kratos spectroflow 783 variable absorbance detector set at 215 nm wavelength (for pindolol and propranolol; 200 nm for tetralins) and a reverse phase C18- column (YMC-pack, ODS-A, 100×4.6 mm I.D., S-5µm, 120A with S5 ODS precolumn). The mobile phase was 0.1 M phosphate buffer (pH 2.0): acetonitrile = 80:20 (for pindolol), 70:30 (for propranolol) or 64:36 (for tetralins); the flow rate was 0.7 ml/min. Stereoselective analysis of racemates was performed using a Chiral-AGP column (5-µm particles, 100×4.0 mm; guard column: 10×3 mm) for separation of the enantiomers of 24

* * O N O N OH OH H H

Mol.Wt. 248.3 Mol. Wt.259.4 N H

(R,S)-pindolol, pKa = 9.50, log P = 1.75 (R,S)-propranolol, pKa = 9.47, log P = 3.65

OH C3H7 * N C3H7

Mol. Wt. 247.3

(R,S)-8-OH-DPAT, pKa = 9.43, log P = 3.50 (4.29)

OH C3H7 C3H7 * N C3H7 * N C3H7

F Mol. Wt. 265.3 Mol. Wt. 307.4 (R,S)-UH-301, pKa = 9.47, log P = 4.16 (4.46) (R,S)-LY-49, pKa = 10.02, log P = 6.29 (6.64)

COCH3 C3H7 C3H7 N * N * C3H7 C3H7

Mol. Wt. 273.3 Mol. Wt. 231.3 (R,S)-LY-41, pKa = 9.99, log P = 4.10 (4.29) (R,S)-LY-39, pKa = 10.08, log P = 4.60 (4.82)

Figure 3. Chemical structures of racemates and enantiomers of pindolol, propranolol and a series of 8-substituted 2-(di-n-propylamino)tetralins being studied with code number, molecular weight, pKa, measured and (calculated) log P values. *- chiral center. pindolol and propranolol. The mobile phases were 10% of acetonitrile in 0.01M phosphate buffer (pH 7.0) for pindolol and 0.5% 2-propanol in 0.02 M ammonium acetate buffer (pH 4.1) for propranolol. The flow rate was 0.9 ml/min (Enquist and Hermansson, 1990). The enantiomers of the tetralin derivatives were separated by an 25

ULTRON ES-OVM chiral column (150×4.6 mm I.D.; mobile phase: 20 mM KH2PO4 buffer (pH 4.6): acetonitrile = 100: 2-6; Flow rate: 0.7 ml/min). The chromatograms of chiral separation of compounds are shown in Fig. 4.

3.4. Bioanalysis of drug (racemates and enantiomers) concentrations in brain tissue and plasma

Determination of tissue drug concentrations was made in rats sacrificed at predetermined intervals after subcutaneous administration of the test compounds. Blood samples were collected and plasma was separated from formed elements within 5 min after sampling. The brains were quickly taken out and cut sagittally in the midline. Both brain tissue and plasma were frozen at -20°C until assayed. The frozen tissues were weighed, thawed and homogenized in 1.1 ml of 0.1 M perchloric acid and added with internal standard (see Table 1.). After centrifugation (12000 rpm, 18600g, 4°C, 10 min.), 1 ml of the brain tissue supernatant or 1 ml plasma was added with 2 M NaOH and 4 ml of dichloromethane. The samples were vortexed for 8 s (pindolol and propranolol) or shaken for 10 min (tetralins) and centrifuged for 10 min at 3500 rpm. Three ml of the organic phases were evaporated to dryness under nitrogen and reconstituted in 100 µl of mobile phase. Injections of 20 µl of the reconstituted material were made into the HPLC system. Standard curves were prepared, using 5-6 concentrations of the analyte and internal standard added to tissue extracts and plasma, respectively, from drug naive animals. The peak height ratio increased linearly with the concentration over the range of concentrations investigated with a correlation coefficient above 0.99 (Fig. 5). The coefficient of variation of the slope was always less than 5%. The other related data are shown in Table 1. Another twenty µl aliquot of the reconstituted material was used for separation of the enantiomers on the chiral columns. Peak areas were used for calculation of the enantiomeric ((S)-/(R)-) ratios. The concentrations of the enantiomers after adminitration of the racemates were calculated from the total concentration of racemate multiplied by the enantiomeric (S)-/(R)-ratio. pindolol propranolol LY-49

AB A BA B 1 2 1 1 1 2 2 2 1 2 21

S/R = 1.7 S/R = 1.6 S/R = 1.1

UH-301 8-OH-DPAT LY-39 LY-41 26

AB AB1 AB AB1 AB 2 2 2 1 1 2 1 2 2 1 1 1 2 1 2 2 2 1

S/R = 1.1 S/R = 1.02 S/R = 0.80 S/R = 0.48 Figure 4. Chromatograms of the enantiomers of compounds in stardard solution (A) and in brain tissue from treated rats (B). Peaks: 1 = (S)-enantiomer; 2 = (R)-enantiomer. The peak area ratio (S/R) is approximately 1 in standard solutions (A). 27

2 50 r = 0.9996

40

30

20

Peak height ratio height Peak 10 (pindolol/metoprolol)

0 0 500 1000 1500 2000 2500 Pindolol concentration (ng/100µl) Figure 5. Standard curve for pindolol in brain extract (shown as an example).

Table 1. HPLC-UV system for determination of drug concentration in brain and plasma

Compound Internal Limit of detection Accuracy ± precision (%) standard (signal/noise = 5) Brain Plasma (nmol/g) (nmol/ml) Brain Plasma pindolol metoprolol 0.052 0.012 99.9 ± 0.35 99.7 ± 1.0 propranolol metoprolol 0.019 0.02 98.9 ± 2.5 97.1 ± 1.3 8-OH-DPAT LY-41 0.05 0.01 99.2 ± 5.0 98.4 ± 5.0 LY-41 8-OH-DPAT 0.04 0.04 97.4 ± 5.0 98.5 ± 5.0 UH-301 8-OH-DPAT 0.05 0.02 97.5 ± 3.0 99.3 ± 3.0 LY-49 LY-55 0.12 0.02 97.3 ± 5.0 98.3 ± 5.0 LY-39 LY-55 0.04 0.02 98.3 ± 3.0 98.2± 3.0 All values are mean ± SD, n = 5-10.

3.5. Biosynthesis of 5-HT and DA (Paper II)

The syntheses of 5-HT and DA were estimated by measuring the accumulation of 5- HTP (precursor of 5-HT) and L-DOPA (precursor of DA) after administration of NSD1015 (m-hydroxy benzylhydrazine), an inhibitor of aromatic L-amino acid decarboxylase (Carlsson et al., 1972). The animals were injected with (S)-UH-301, 28 saline or (R)-8-OH-DPAT (used as reference compound) at 30 min before injection of NSD1015 (287 µmol/kg; 60 mg/kg s.c.) and were decapitated 30 min after NSD1015. The hippocampus and the striatum were rapidly dissected out and frozen until assayed. The frozen samples were weighed and homogenized in 1 ml of 0.1 M perchloric acid and α-methyl-5-hydroxytryptophan was added as an internal standard. After centrifugation (18,600 g, 4 °C, 10 min.) and filtration, 20 µl of the supernatant was injected into HPLC-EC (electrochemical detection) to analyze 5-HTP and L-DOPA as described previously (Yu et al., 1996b).

3.6. Pharmacodynamic studies (Paper III)

3.6.1. Behavioral experiments

Observations and ratings of three behaviors were performed at 10 min (peak intensity of drug induced behavior: forepaw treading, flat body posture and hind limb abduction) after injection of (R)-8-OH-DPAT (3 µmol/kg s.c.) or saline injected at predetermined intervals (5, 30, 60, 120 and 240 min) after the administration of pindolol (8 mg/kg s.c.) or saline. The scoring was made as follows: no behaviour (0), weak behaviour (1), intermittent behaviour (2), intense or continuous behaviour (3). The maximum score observed for each symptom was used for calculation of the total 5-HT syndrome score (maximum total score = 9) for each animal.

3.6.2. Body temperature

Body temperature was determined by insertion of a thermistoprobe (Ellab Instruments, Copenhagen) into the colon, 2.5-3.0 cm from the anal orifice. (R)-8-OH-DPAT (3 µmol/kg s.c.) was given at predetermined intervals (0, 30, 60, 120, 240 min) after the pindolol or saline injection. The body temperature was measured at 30 minutes (peak hypothermic response) after the injection of (R)-8-OH-DPAT.

3.7. Blood-brain barrier experiments in vitro

The method to study the BBB in vitro developed by Dehouck et al. (Dehouck et al., 1990) was used in the present investigations and is briefly described below. 29

3.7.1. Cell culture

Bovine brain endothelial cells isolated from capillary fragments were co-cultured with primary astrocytes from newborn rats. By using mechanical homogenization and filtration a preparation enriched with capillaries was obtained without the use of collagenase digestion. The capillaries were separated from the great majority of arterioles and venules and were seeded onto extracellular matrix coated dishes. Endothelial cell islands emerging from identified capillaries was then microtrypsinized to expand the number of cells. After the first passage, the endothelial cells were harvested and seeded onto a 6 cm gelatin-coated dish. After 6-8 days, confluent cells were sub-cultured at a split ratio of 1:20. Cells from the third passage were stored in liquid nitrogen. The astrocytes were isolated from the cerebral cortex of newborn rats. After removing the meninges, the brain tissue was gently forced through a nylon sieve. The astrocytes were plated on the bottom of cell culture clusters, containing six wells each, at a concentration of 1.2 x 105 cells/ml in 2 ml Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% foetal calf serum. The medium was changed twice a week. Three weeks after seeding, cultures of astrocytes were stabilized and can be used for the co- culture. The astrocytes were characterised with glial fibrillary acidic protein (GFAP) showing typically, that more than 95% of the population was GFAP positive.

To establish the co-culture, the endothelial cells were seeded onto collagen coated Corning Costar TranswellTM inserts (0.4 µm pore), which were placed in the wells containing astrocytes. The medium used for the co-culture was Dulbecco’s modified Eagle’s medium supplemented with 10% calf serum, 10% horse serum, 2mM glutamine, 50 µg/ml gentamicin and 1 ng/ml basic fibroblast growth factor. The medium was changed every second day. Under these conditions the endothelial cells retain all the usual endothelial cell markers such as factor VIII related antigen, a non-thrombogenic surface, angiotensin converting enzyme and prostacyclin synthesis as well as the characteristics of the barrier which include e.g. complex tight junctions, low rate of pinocytosis, gamma glutamyl transpeptidase and monoamine oxidase. Experiments were usually initiated after 12 days of co-culture (Fig. 6.). 30

A.

Collagen coated insert with endothelial cells

Well Astrocytes

B.

Apical side (Luminal)

A→B Ringer-HEPES buffer

Basolateral side (Abluminal) B→A

Figure 6. In vitro model of the blood-brain barrier. A. Cell culture; B. Transport experiment.

3.7.2. Transport experiments

Compounds were dissolved at a concentration of 10, 50 or 100µM in buffered Ringer’s solution with 0.25% dimethyl sulfoxide (DMSO). To monitor the integrity of the endothelial cell monolayer, the marker molecule [14C] sucrose was added to the solution at a concentration of 0.05 µCi/ml. The concentration of DMSO used in the experiments had previously been shown not to affect the integrity of the cell monolayer. The experiments were carried out at pH = 7.4 and 37°C. At the initiation of the transport experiments in the apical to basolateral direction (AÆB), buffered Ringer’s solution (without test compound) was added to wells without astrocytes. Inserts, containing 31 confluent monolayers of brain capillary endothelial cells, were subsequently placed in these wells. The solution of the test compound was then added to the cell monolayer and at 10, 20, 30, 40, 50 and 60 minutes after the addition, the inserts were moved to new wells in order to minimize back diffusion of compound to the upper compartment. Three inserts with cells and three collagen coated inserts without cells were assayed for each test compound. For transport experiments in the basolateral to apical direction (BÆA), the solution of the test compound was added to the wells and samples were withdrawn at 10, 20, 30, 40, 50 and 60 minutes from the filter inserts with cells (upper compartment). The withdrawn volume was immediately replaced with the same volume of Ringer’s solution. After the experiments, samples were stored at -20°C until the drug concentrations were analyzed by HPLC (see Section 3.3). For chiral separation, 20 µl of sample at each time point were directly injected onto the chiral column.

3.7.3. Data analysis and calculation

The cleared volume was calculated by dividing the amount of compound in the receiver compartment by the drug concentration in the donor compartment at each time point. The average cumulative volume cleared was plotted versus time and the slope was estimated by linear regression analysis (EXCEL 5.0) to give the mean and standard deviation of the estimate.

The slope of the clearance curve with inserts alone and inserts with cells is equal to PSf and PSt respectively, where PS = the permeability surface area product (Fig.7). The units of PS and S are in microliters/minute and square centimeters, respectively. The PS-value for the endothelial monolayer (PSe) was computed as follows: 1/PSe = 1/PSt - 1/PSf. To generate the endothelial permeability coefficient, Pe (cm/min), the PSe-value was divided by the surface area (4.7 cm2) of the insert. 32

3000

2500

l) inserts without EC µ R2=0.9660 2000 PSf = 43.27 ± 0.02 1500

1000 inserts with EC Cleared volume ( R2=0.9816 500 PSt = 30.21 ± 0.03 0

0 10 20 30 40 50 60 70 Ti me (mi n)

Figure 7. (R,S)-8-OH-DPAT (50 µM) volume cleared across inserts alone or across the inserts with endothelial cells (EC). Slopes (PSf or PSt) were determined by linear regression analysis. The Pe value in this experiment was 21.31 ×10-3 cm/min.

3.8. Plasma protein binding assay

Protein binding of drugs was determined by equilibrium dialysis according to (Ekblom et al., 1992). Chambers of 1 ml were used. The dialysis membrane (Technicon chemicals) had a pore size between 20 and 60 Å. The test compound was added to rat plasma and the final pH was adjusted to 7.4 by addition of 0.1 M phosphate buffer solution. Triplicates of the test compound in 500 µl of plasma were dialyzed against exactly 500 µl of 0.1 M phosphate buffer (pH 7.4) for 5 hours (tetralins) or 6 hours (pindolol and propranolol) at 37°C. After dialysis, samples were collected from the buffer compartment of the dialysis chambers and 20 µl aliquots were taken for analysis by HPLC (see Section 3.3) with volume as external standardization. Another 20 µl of sample was directly injected onto the chiral column to measure the enantiomeric ratio. The free drug fraction was calculated according to the equation: Fu = measured drug concentration in buffer/ known drug plasma concentration. 33

3.9. Statistical methods and calculation of pharmacokinetic parameters

GraphPad InStat and GraphPad Prism (version 2.0) (GraphPad Software Inc., San Diego, USA) were used for the statistical analyses and constructions of graphs. One-way ANOVA followed by Tukey-Kramer multiple comparison tests or unpaired two-tailed t- tests were used for statistical analysis. A non-parametric method (Kruskal-Wallis test) was used in tests for statistical significance between groups in the behavioral study. A value of p< 0.05 was considered statistically significant. The trapezoidal rule was used for calculation of AUCs (0 - 4 hours). Determination of t1/2 in brain tissue and plasma was done by graphing the concentrations over time in a semilogarithmic plot, calculation of the slope (k) and determine the t1/2 according to the formula 0.693/k. 34

4. Results and Discussion

4.1. The first stereoselective study of racemic pindolol to rat brain tissue (Paper I)

Racemic pindolol has been shown to augment and accelerate the effects of antidepressant drugs in man (Artigas et al., 2001). Most experimental studies of the mechanism of action of pindolol in rats have been performed with the more pharmacologically active (S)(-)- enantiomer (Hjorth and Carlsson, 1985; Hjorth and Pettersson, 1993; Sharp et al., 1993; Monti et al., 1995). As a preliminary to pharmacodynamic studies of racemic pindolol in rats, the study in Paper I was conducted in order to confirm the assumption that the concentrations of the two enantiomers were similar in brain and plasma in rats. The chromatographic system clearly separated the two enantiomers, (S)-pindolol having a shorter retention time than (R)-enantiomers. The peak area ratio ((S)-/(R)-) was approximately 1 in standard solution. It was unexpectedly found that the (S)-/(R)-ratio was 1.74 in brain and 0.82 in plasma, indicating that the concentration of (S)-pindolol was higher than the (R)-enantiomer in brain and lower than the (R)-pindolol in plasma. The enantiomeric ratios also varied between other body tissues, ranging from 0.85 (liver) to approximately 1 (kidney and fat), and higher (muscle < heart < spleen < lung) up to 1.64. A stereoselective transport of (S)-pindolol across endothelial cells in brain and other capillaries or into tissue cells might explain the results. This initial finding prompted further investigations of pindolol and, for comparison, propranolol (Paper III) and a series of chiral aminotetralins with demonstrated stereoselective pharmacodynamic properties (Paper IV).

4.2. Differential distribution of enantiomers of racemic pindolol and propranolol to the rat brain (Paper III)

Time-courses in vivo In the present in vivo studies, pindolol or propranolol was given subcutaneously (s.c.) in equimolar doses to rats, and groups of 5 animals were sacrificed at predetermined time- points. Analysis of the time courses of pindolol concentrations in brain and plasma 35 showed that both enantiomers accumulated in brain over 60 minutes after the s.c. adminitration of the racemate, whereas plasma concentrations started to decline after 30 minutes (Fig. 8). This finding seems to indicate that both (R)- and (S)-pindolol are actively transported from plasma into brain tissue. The brain concentration of (S)- pindolol was higher than (R)-pindolol whereas (S)-pindolol in plasma was lower than (R)- pindolol. Similar findings were obtained for brain and plasma propranolol concentrations, i.e. the (S)-propranolol was higher than the (R)-enantiomer in brain but lower in plasma. Both enantiomers of propranolol accumulated in brain over 30-60 minutes, while their plasma concentrations started to decline after 15 min.

A. (R,S)-pindolol in brain 12.5 (S)- in brain (R)- in brain 10.0 (R,S)-pindolol in pla sma (S)- in plasma 7.5 (R)- in plasma 5.0 Concentration 2.5 (nmol/g or nmol/ml) (nmol/g 0.0

0 60 120 180 240 Time (min) B. (R,S)-propranolol in brain (S)- in brain 12.5 (R)- in brain (R,S)-propranolol in plasma 10.0 (S)- in plasma (R)- in plasma 7.5

5.0 Concentration 2.5 (nmol/g or nmol/ml) (nmol/g 0.0

0 60 120 180 240 Time (min) Figure 8. The concentration-time curves of the racemates, and the (S)- and (R)- enantiomers in brain and plasma after chiral separation. A. pindolol; B. propranolol. 36

However, propranolol differed from pindolol in that the brain/plasma ratio was similar over the entire time course, whereas it increased over time for pindolol. Other comparisons between pindolol and propranolol are presented in Fig.9. Interestingly, pindolol and both of its enantiomers had longer half-lives in brain (55 min) than in plasma (30 min), which indicate that both pindolol enantiomers are sequestered in a brain tissue compartment that is not in immediate equilibrium with plasma. In contrast to pindolol, the half-lives of the two enantiomers of propranolol in brain were 8-12 minutes shorter than in plasma, indicating that propranolol readily leaves brain tissue.

pindolol 10.0 B/PAUC (S)-/(R)- 7.5 AUC

5.0 Ratio

2.5 (1.74) (0.82) 0.0 (R,S)- (S)- (R)- Brain Plasma propranolol 10.0 B/PAUC (S)-/(R)-AUC 7.5

5.0 Ratio

2.5 (1.65) (0.50) 0.0 (R,S)- (S)- (R)- Brain Plasma

Figure 9. The brain/plasma (B/P) concentration ratios and (S)-/(R)-ratios based on the AUC (area under the concentration-time curve, 0-240 min). The (S)-/(R)-ratios in brain and plasma are given within brackets. 37

The half-lives for (S)-propranolol in brain and plasma was longer than for (R)- propranolol. This indicates stereoselective, most probably metabolic elimination of propranolol with a faster elimination of the (R)-enantiomer than the (S)-enantiomer. There were no statistically significant differences in pindolol concentrations or (S)-/(R)- ratios between six different brain regions.

Plasma protein binding in vitro

Stereoselective binding to plasma proteins has been suggested as an explanation for the differential tissue distribution of the enantiomers of propranolol in rats (Takahashi et al., 1990). Plasma protein binding of pindolol as determined by equilibrium dialysis showed a free fraction of 43-48%, which is in agreement with previous literature (Ylitalo et al., 1986). The (S)-/(R)- ratio in the rat plasma dialysate was close to 1, indicating that there was no difference in plasma protein binding between the pindolol enantiomers.

Stereoselectivity in pindolol binding to purified α1-acid glycoprotein has been repeated (Murai-Kushiya et al., 1993), but does not seem to be the case for rat plasma according to the present results. It needs to be emphasized, however, that the free fraction of drugs in plasma as determined in vitro might not adequately reflect the situation in vivo. Their extraction from blood into brain is often greater than and independent of their free plasma concentration (cf. Greig, 1992). Therefore, stereoselective binding of pindolol enantiomers to rat plasma does not seem to explain the unequal distribution of the two enantiomers to brain tissue.

For propranolol, the free fraction in plasma was 0.19 - 0.24, which is higher than the value of 0.05 - 0.10 reported previously (Ylitalo et al., 1986). Propranolol showed enantioselective protein binding with an (S)-/(R)-ratio of 0.75 in the present study. This is in agreement with studies in dogs and man (Bai et al., 1983; Walle et al., 1983). Whether or not this is due to more binding sites or to a higher affinity of (S)-propranolol for the binding sites on plasma proteins remain to be investigated. In conclusion, the difference in protein binding between the (S)- and (R)-enantiomers of propranolol may be part of the explanation for the difference in their distribution to the brain. 38

Brain endothelial cell permeabiblity

The in vitro model of the BBB was used to test the hypothesis that stereoselective transport mechanisms with bovine endothelial cells might explain the observations in vivo. The studies of the transcellular transport across bovine brain endothelial cells showed that pindolol has a relatively low permeability coefficient in comparison with propranolol and there was difference between the enantiomers. Thus, transcellular transport across brain endothelial cells seemed to explain the stereoselective distribution of racemic pindolol to the brain. Propranolol differed from pindolol in two aspects: (1) the permeability coefficient for propranolol was higher than for pindolol, probably due to its higher lipophilicity (log P = 3.65 vs. 1.75, respectively; Ylitalo et al., 1986), and (2) there was a consistent difference in permeability coefficient for the apical to basolateral vs. the basolateral to apical direction of transport. The difference between the directions seemed to be larger for the (R)- vs. the (S)-enantiomer. There was a small difference (18 – 30 %) in permeability coefficients between the enantiomers when studied separately, but chiral separation showed that there was no difference between the two enantiomers after the passage of racemic propranolol across the endothelial cells in either direction. Thus, stereoselective transcellular transport across bovine brain endothelial cells does not seem to explain the difference between the two enantiomers in brain in vivo after administration of racemic propranolol (Fig.10).

Eeffects of verapamil pretreatment on brain pindolol or propranolol concentrations

One hypothesis for the differences in brain concentration between the (R)- and (S)- enantiomers of pindolol and propranolol is that an efflux pump such as P-glycoprotein that might stereoselectively prevent the (R)-enantiomers from entering the brain (Letrent et al., 1999). Thus, verapamil, a P-gp inhibitor, might differentially affect the brain/plasma concentration ratios of the enantiomers of pindolol and propranolol, respectively. Surprisingly, the study of verapamil showed that the brain/plasma concentration ratio for pindolol decreased by almost 50% rather than increased as expected. Similarly, there was a decrease of the brain/plasma ratio of approximately 40% 39 for propranolol. It seems that verapamil may stereoselectively inhibit the influx of pindolol and propranolol across the BBB or its accumulation in brain tissue, which may be caused by inhibition of a transporter in the brain endothelial cells or other brain cells or inhibition of the binding of pindolol to brain protein or other cell constituents.

pindolol

7.5 *** A - B B - A ***

5.0 ***

, cm/min) , * -3 10 × 2.5 Pe ( Pe

0.0 (R,S)- (S)- (R)- (R,S)- (R,S)- 100µM 50µM 10µM propranolol

40 A - B *** B - A 30 *** *** , cm/min) ,

-3 20 *** 10 ×

Pe ( Pe 10 ***

0 (R,S)- (S)- (R)- (R,S)- (R,S)- 100µM50µM 10µM Figure 10. Permeability coefficients of pindolol and propranolol at three different concentrations (100, 50 and 10 µM) in the in vitro BBB model. A = Apical; B = Basolateral. ***p<0.001,*p<0.05. 40

Pharmacodynamic effects of pindolol

The (R)-8-OH-DPAT induced 5-HT behavioural syndrome and hypothermia was used to monitor the time course of the pharmacodynamic effect of pindolol on 5-HT1A receptors and by inference the concentration of (S)-pindolol in the biophase. The duration of action of pindolol corresponded well to its presence both in brain and plasma. The pharmacodynamic effects of pindolol seemed to follow the plasma concentration more closely than the brain concentration (Fig. 11).

pindolol + (R)-8-OH-DPAT saline + (R)-8-OH-DPAT

100 (S)-pindolol in brain 10.0 (nmol/g or nmol/ml)

(S)-pindolol in plasma Concentration 7.5

50 5.0

Antagonism (%) Antagonism 2.5

0 0.0

0 60 120 180 240 Time(min)

Figure 11. Effect of racemic pindolol (8 mg/kg s.c.) on the (R)-8-OH-DPAT (3 µmol/kg s.c.) induced 5-HT syndrome in relation to the time courses for brain and plasma concentrations of (S)-pindolol. (R)-8-OH-DPAT was given at different time points after the pindolol or saline injection.

Thus, maximal 5-HT1A antagonism was observed when pindolol and (R)-8-OH-DPAT were given simultanously, i.e. when the plasma concentration peaked but the brain concentration was still relatively low. After 30 min, 5-HT syndromes started to reappear while (S)-pindolol in plasma steadily decreased, and the total concentration of (S)- pindolol in brain increased, indicating that pindolol is located intracellularly to a large extent rather than extracellularly. 41

4.3. Relationship between pharmacokinetics and pharmacodynamics of (S)-UH-301 (Paper II)

(S)-UH-301 has been reported to be a selective and potent 5-HT1A receptor antagonist (Hillver et al., 1990; Björk et al., 1991a; Hillver et al., 1996). Thus, (S)-UH-301 is able to completely antagonise several effects induced by the prototype 5-HT1A receptor agonist (R)-8-OH-DPAT (Arvidsson et al., 1981; Hjorth et al., 1982; Middlemiss et al., 1985; Björk et al., 1991a; Björk et al., 1991b; Johansson et al., 1991; Björk et al., 1992). It is noteworthy that (R)-UH-301, in contrast to the (S)-isomer, behaves as a 5-HT1A receptor agonist (Björk et al., 1992; Hillver et al., 1996). Thus, the enantiomers of UH-301 have opposite pharmacological actions which may explain why racemic UH-301 does not produce any effects in vivo.

In order to understand the pharmacology of a compound, it is important to have knowledge of its pharmacokinetics. The integrative pharmacokinetic-/pharmacodynamic investigation of (S)-UH-301 in Paper II was performed with subcutaneous adminitration, the route commonly used in animal studies. The time- and dose-response experiments showed several interesting characteristics. There was a rapid increase up to peak brain concentrations of the drug between 5 and 15 min after injection, with only minor changes in plasma concentrations. Similar finding have been observed for (R)-8-OH-DPAT (Yu and Lewander, 1997), and for two other dipropylaminotetralin (DPAT) derivatives, (S)-8- (2-furyl)-DPAT and (R)-8-phenyl-DPAT (Yu et al., 1997). An active transport of (S)- UH-301 across the blood-brain barrier, trapping of the drug in brain cells, e.g. via an active uptake process, seems more likely to explain this than passive diffusion. There was an approximately three-fold increase in brain and plasma concentrations when dose was increased from 10 to 32 µmol/kg. The peak concentrations in brain and plasma did not increase proportionally to dose at 100 µmol/kg. However, there was an approximate doubling of the brain and plasma half-lives of (S)-UH-301 at dose of 100 µmol/kg (82 min in brain and 76 min in plasma), in contrast to the half-lives at 10 and 32 µmol/kg (31-36 min in brain and 46-49 min in plasma). Based on AUCs, the brain/plasma ratios varied between 5.4 and 11.5 as a result of the disproportionate changes of the AUCs in 42 plasma. There were statistically significant variations in (S)-UH-301concentrations between rat brain regions with the highest concentrations in hippocampus were the highest and the lowest in brain stem and cortex. This similar to (R)-8-phenyl-DPAT (Yu et al., 1997) but unlike (R)-8-OH-DPAT (Yu and Lewander, 1997). Regional differences in drug concentrations within the brain are commonly ascribed to regional differences in blood flow (Scremin et al., 1990). An alternative explanation might be that the uneven distribution of receptors to which the drug binds and that the drug-receptor complex becomes internalized by endocytosis (Tsao et al., 2001). For example, hippocampus is a region with a high density of 5-HT1A receptors as compared with other brain regions (Hall et al., 1985; Pazos et al., 1985; Gozlan et al., 1988; Wright et al., 1995; Kia et al., 1996). Taken together the present findings show that regional differences in drug concentrations within the brain varies between drugs even though they are similar in chemical structure and physicochemical characteristics.

With the present pharmacokinetic data at hand, it became possible to study pharmacokinetic/pharmacodynamic interrelationships. The 5-HT1A syndrome induced by administration of a dose of 1.0 µmol/kg of (R)-8-OH-DPAT was completely antagonized at brain concentrations above 5 nmol/g, and the time course seemed to follow total brain concentrations of (S)-UH-301 more closely than the plasma concentrations of the drug (Fig. 12). In a dose response comparison, the effects of (S)-UH-301 on brain 5-HT and dopamine biosynthesis were investigated. The (S)-UH-301 induced decrease in 5-HT and dopamine synthesis seemed to correlate with increasing brain rather than plasma concentrations of the drug. 5-HT synthesis was not affected by (S)-UH-301 at doses below 10 µmol/kg whereas dopamine synthesis started to decrease at this dose level in agreement with Björk et al. (1991a) and Ahlenius et al. (1999), probably due to an agonistic effect on dopaminergic autoreceptors. The present study showed that the magnitude of the biochemical effects is related to the brain exposure of (S)-UH-301, hippocampal 5-HT synthesis being decreased at higher tissue concentrations of (S)-UH- 301 than striatal dopamine synthesis. Based on our present knowledge, (S)-UH-301 at doses above 10 µmol/kg s.c., corresponding to brain concentration above 10 nmol/g and plasma concentration above 1 nmol/ml, is not selective for 5-HT1A receptors. 43

Antagonism by (S)-UH-301 saline + (R)-8-OH-DPAT (S)-UH-301 in brain (S)-UH-301 in plasma

100 10.0 (nmol/g ornmol/ml) Concentration 7.5

50 5.0

Antagonism (%) 2.5

0 0.0

0 60 120 180 240 Ti me (mi n) Figure 12. Time-dependent antagonism by (S)-UH-301 (10 µmol/kg s.c.) of the (R)-8-OH-DPAT (1.0 µmol/kg s.c.) induced 5-HT syndrome in rats (behavioral data from Björk et al., 1991) in relation to the time courses for brain and plasma concentrations of (S)-UH-301.

4.4. Transport of tetralins across the BBB in vivo and in vitro (Paper IV)

The 5-HT1A receptor agonist, 8-hydroxy-2- (di-n-propylamino)tetralin (8-OH-DPAT) has become the prototype pharmacological tool for studies of 5-HT1A receptor-mediated functions in the CNS. However, because of its poor oral bioavailability (Wikström et al., 1988; Yu et al., 1996a), a number of 8-OH-DPAT analogues with C8-substituents other than hydroxyl have been prepared, hoping that these structural modifications would improve the pharmacokinetic properties without changing the pharmacodynamic profile (Liu et al., 1993; Liu et al., 1995). It has been found that the enantiomers of many 2-(di- n-propylamin)tetralin (DPAT) derivatives have different pharmacodynamic characteristics (Yu et al., 1997). The time-dependent brain and plasma concentrations of 8-OH-DPAT after s.c. administration showed that brain concentrations increased rapidly towards Cmax while the plasma concentration decrease (Yu and Lewander, 1997), which 44 suggest that 8-OH-DPAT may be transported into the CNS via a carrier mediated transport mechanism.

Subsequent studies showed that LY-49, a potent 5-HT1A ligand in vitro, is poorly distributed to the brain compared to some other DPAT derivatives, such as 8-(2-furyl)- DPAT (Yu et al., 1997), 8-OH-DPAT (Yu and Lewander, 1997). Despite a 3.2 times higher dose than 8-OH-DPAT the brain and plasma of LY-49 concentrations were surprisingly low. As shown in a previous study, the time course of (R)-LY-49 in brain after s.c. administration increased slowly to a maximum at 120 minutes, whereas the plasma concentrations remained almost constant between 1 and 6 hours (Yu et al., 1997). The time courses of UH-301 in brain and plasma after s.c. administration in rats showed that the in brain concentration increased rapidly over 30-60 minutes, while the concentration in plasma decreased, suggesting that also (S)-UH-301 may be transported into the brain (Paper II).

The objectives of the investigations in Paper IV were to compare the distribution of five tetralin compounds to the rat brain. The brain/plasma (total and free) ratios at peak brain concentrations after subcutaneous administration in rats were used as a measure of brain uptake in vivo. Bovine brain endothelial cells co-cultured with rat brain astrocytes was used to assess transcellular transport across the brain capillary endothelium in vitro. Physicochemical parameters (log P and pKa) and plasma protein binding were also determined for each racemate. Since we had found differences in the distribution of the enantiomers of pindolol and propranolol to the brain in rats (Paper I and III), it was decided to study both the racemates and the separate enantiomers of the structurely related tetralins. This approach seemed to be all the more important, because of the limited knowledge of stereoselective transport of drugs across the BBB in general.

A comparison across the five racemic compounds (Fig. 13) shows that the brain /plasma concentration ratios at peak brain concentrations vary from 2.3 to 19.9. After correction for plasma protein binding the brain/free plasma concentration ratios increased approximately 3-fold without changing the order between the compounds except for LY- 49, for which the brain/free plasma concentration ratio increased from 2.4 to 23, i.e. a 45 value similar to 8-OH-DPAT. There was no correlation between the log P values and brain/free plasma concentration ratios in vivo. For example, UH-301 with a log P of 4.16 had a 3-fold higher brain/free plasma concentration ratio than LY-49 with a log P of 6.29.

A. 10.0 brain plasma 7.5

5.0

Concentration 2.5 (nmol/g or nmol/ml) or (nmol/g

0.0 8-OH-DPAT LY-41 UH-301 LY-49 LY-39 10 µmol/kg s.c. 32 µmol/kg s.c.

B. 100 B/Ptotal B/P 75 free

50 B/P ratio B/P

25

0 8-OH-DPAT LY-41 UH-301 LY-49 LY-39

Figure 13. A. Brain and plasma concentrations of the racemates of the five 8-substituted 2-(di-n-propylamino)tetralins after s.c. administration of the racemates. B. The brain/plasma (B/P, total and free) ratios at peak brain concentrations after s.c. administration in rats.

There was no indication of stereoselective distribution to the brain in vivo for the tetralin derivatives, neither when given as the racemates nor as the pure enantiomers separately with one exception, namely LY-41. The brain and plasma concentrations and the brain/plasma ratios for LY-41 were lower than for 8-OH-DPAT in spite of a higher log P 46

(4.10). (R)-LY-41 had a statistically significantly higher brain/plasma ratio than (S)-LY- 41. The results with regard to transcellular transport across bovine brain endothelial cells in vitro (Fig.14) showed several interesting features. All the tetraline derivatives had high to very high permeability coefficients (Pe). The most conspicuous finding was the difference in Pe between the apical to basolateral (A-B) vs. the basolateral to apical (B-A) directions. This difference, expressed as the A-B/B-A ratio, varied from 1-2 for the UH- 301 enantiomers up to 53.8 for (S)-LY-49. This result indicates that carrier mediated transport is involved in at least one direction. The difference in Pe-values for the enantiomers of 8-OH-DPAT was small and the chiral separation of the enantiomers after passage of the racemate across the endothelial cells showed no difference in (S) -/(R) - ratio from a standard solution. Thus, there was no indication of stereoselective transport of 8-OH-DPAT across the endothelial cells. The A-B/B-A ratio of UH-301 was low (1-2) suggesting approximately similar influx and efflux rates across the BBB. The in vitro results with LY-49 seem contradictory. There was a clear difference in Pe between the pure (S) - and (R) - enantiomers ((S)-/(R)-ratio of 2.1), whereas there was no difference between the enantiomers after passage of the racemate across the endothelial cells in either direction. There is a possibility that the enantiomers when present together might interact in such as way that one enantiomer inhibits the transport of the other and, thus, that a true difference between them is concealed. Further investigations are needed to test this hypothesis. LY-49 also had the largest difference between the Pe values for the A-B vs. the B-A direction with a ratio of 17-53 for the racemate and the pure enatiomers. Taken together, these findings strongly support that a carrier mediated transport of LY-49 across the endothelial cells is involved in addition to passive diffusion. In summary, comparing the pharmacokinetic charateristics of a series of tetralins in vivo with transcellular transport across bovine brain endothelial cells in vitro, there does not seem to be a clear relationship between lipophilicity, plasma protein binding, transport across endothelail cells in vitro and distribution in brain in vivo. It seems that influx and /or efflux, or both, across the BBB are dependent on active transport rather than passive diffusion. It seems that the different C8-substituted 2-(di-n-propylamino) tetralin analogues contribute to the uniqueness with regard to their pharmacokinetic/pharmacodynamic characteristics. 47

50 A - B *** B - A 40 *** 30 , cm/min) *** -3 * 10 20 ×

Pe ( Pe 10

0 (R,S)-8-OH- (S)-LY-41 (R)-LY-41 (S)-UH-301 (R)-UH-301 DPAT

50 *** A - B *** B - A 40

30 *** , cm/min) , -3 *** *** *** 10 20 ×

Pe ( Pe 10

0 (R,S)-LY-49 (S)-LY-49 (R)-LY-49 (S)-LY-39 (R)-LY-39

Figure 14. The permeability coefficients (Pe, cm/min) of tetralins in the apical to basolateral (A-B) and in the basolateral to apical (B-A) direction in the in vitro BBB model. *p<0.05, ***p<0.001. 48

5. Conclusions

The study of (S)-UH-301 showed accumulation in brain in spite of stable or decreasing plasma concentrations. This indicates that this drug, similarly to 8-OH-DPAT and other tetralins, may be actively transported into brain tissue. Lack of dose proportionality and dose-dependent kinetics was observed at high doses. The 5-HT1A antagonistic action closely followed the time course of both brain and plasma concentrations of the drug.

The investigations of the series of structurally related tetralins showed marked differences in peak brain concentrations and brain/plasma concentration ratios between the compounds in vivo with no apparent relationship with lipophilicity, protein binding or transport across bovine brain endothelial cells in vitro. With one exception (LY-41) there was no indication of stereoselective distribution to the brain for the tetralins. There were marked differences between the permeability coefficients for the tetralins regarding transport across bovine brain endothelial cells in vitro with no apparent relation to their lipophilicity constants. For some tetralins there was a difference in transport rate between the enantiomers when studied individually, but not when studied as the racemates followed by chiral separation of the enantiomers, discrepancies that remain to be explained. The most important finding in vitro was the marked difference in transport rate between the apical to basolateral and the basolateral to apical directions across the brain endothelial cells, which seemed to be a common finding for the tetralin enantiomers with one exception ((S)-UH-301). This finding clearly indicates the presence of active transport mechanisms for the tetralins across the bovine brain endothelial cells. These data are also consistent with previous findings in vivo, namely the accumulation of brain concentrations of the tetralins during stable or decreasing plasma concentrations (see (S)- UH-301 above). The nature of such a potential influx transporter remains to be investigated.

The pharmacokinetic studies of racemic pindolol in rats showed that pindolol, similarly to the tetralins, accumulates in brain during stable or decreasing plasma concentrations, which again indicate that pindolol is actively transported across the blood-brain barrier in 49 vivo. The half-life of pindolol in brain tissue was longer than in plasma. Two pharmacodynamic effects of pindolol (antagonism of 5-HT1A receptor mediated and (R)- 8-OH-DPAT induced behaviors and hypothermia) reflecting the pindolol concentrations in the biophase seemed to follow the plasma concentrations over time rather than the total brain concentrations. Taken together, it seems that pindolol is partly sequestered in a brain tissue compartment that is not in immediate equilibrium with the biophase or plasma. Future studies may address the nature and importance of this compartment for a full understanding of the neuropharmacological effects of pindolol.

The most important new finding was the stereoselective accumulation of the pharmacologically active (S)- over the (R)-enantiomer of pindolol in brain after admininstration of the racemate. This differential distribution of the two pindolol enantiomers could not be explained by stereoselective plasma protein binding or stereoselective transport across bovine brain endothelial cells in vitro and needs to be investigated further.

The comparative studies with racemic propranolol showed that propranolol accumulated in brain during stable or decreasing plasma concentrations and had a stereoselective distribution in brain and plasma similarly to pindolol. Stereoselective protein binding of propranolol, in contrast to pindolol, may be part of the explanation for the unequal distribution of the propranolol enantiomers. Stereoselective transport of propranolol across brain endothelial cells in vitro, however, did not seem to explain the in vivo findings. In contrast to pindolol, but similarly to the tetralins (cf. above), propranolol had a higher apical to basolateral vs. basolateral to apical transport rate in the blood-brain barrier model in vitro. An additional new finding was the stereoselective inhibitory effect of verapamil on the transport of pindolol and propranolol into the rat brain in vivo. This finding cannot be explained by inhibition of the drug efflux transporter P-glycoprotein, but indicates the presence of a verapamil sensitive drug influx transporter that remains to be characterized. 50

A general conclusion can be drawn based on the present new findings. The pharmacokinetic characteristics, including transport across the blood-brain barrier, as well as the pharmacodynamic characteristics of a compound and its enantiomers has to be determined empirically rather than based on generalizations from structural or physicochemical information. The present studies with emphasis on stereoselective properties of the enantiomers of chiral compounds have provided new information of preclinical and clinical relevance and importance. Increased knowledge of blood-brain barrier transport of drugs will help understand basic mechanisms of distribution of drugs to the brain in general, and will hopefully facilitate the discovery and development of new neuropsychopharmacological agents. 51

Acknowledgements

The present studies were performed at the Department of Neuroscience, Psychiatry, Ulleråker, Faculty of Medicine, Uppsala University.

I would like to take the opportunity to express my deepest gratitude to all the people who have contributed to this thesis and in particular to:

My supervisor, Associate Professor Tommy Lewander, for introducing me to work in the serotonin field and sharing his scientific knowledge and experience, for his caring and understanding throughout these years. His positive attitude, excellent scientific guidance, continuous encouragement and support are gratefully acknowledged. It’s truly a great pleasure to work with him!

Professor Frits-Axel Wiesel, Head of Psychiatry, Ulleråker, Department of Neuroscience, for his all support, scientific discussions and for providing excellent working facilities.

Ms. Anne-Maj Gustafsson, for her skillful and priceless guidance and assistance with laboratory analysis work, for all the laughs and good times we shared.

My co-authors, Miluse Renftel, Stefan Lundquist, Göran Sundholm and Ye Liu for much appreciated collaboration.

Dr. Hong Yu, for generous and valuable practical help and interesting discussions on pharmacology.

Associate Professor Anders Fredriksson, for all his advice and comments on my manuscripts and always finding time to solve my computer problems.

Associate Professor Karin Sonnander, for advice and comment on the manuscripts and organising research seminars in the department.

Dr. Rujia Xie, for valuable discussions within the field of pharmacokinetics. 52

Ms. Lottie Söder, secretary of the department, for her skillful assistance, her efficiency and her helpfulness.

Ms. Marianne Andersson, for her helpful assistance in the library.

All the present and former people working at the department, Marie-Louise Aarflot, Lillemor Källström, Anders Hjorth, Helena Billander, Anita Liberg, Flor Rezaei, Astrid Strömgren, for relaxing conversations, funny moments and for creating a friendly atmosphere.

My Chinese friends in Sweden now and before, Jinghong, Zhang jing and Shouting, Lao Li and Lijun, Haiyan, Xuxia and Liu liang, Yuanyue and Gao lan, Jiayang, Pingsheng, Lu li and Wang chen, Chenjie and Lihong, Guozhen and Dong jiajun, Xiaobing and Jiang su, Xiaojun, Zhao ying and Ruisheng, Zhouqin, Zou xiang, Siwei, Lai’s family, for their friendship and for giving me an enjoyable and pleasant time outside the department.

I would like to thank my parents-in-law, for their great supports at the beginning of my Ph.D. work with helping me take care of my 4-month-old daughter. And also, Fang Kun, Yu Yang, Fangin and Tianhao, for caring and good time when we were together.

My warmest thanks to my dear parents, for their loves, their endless supports and encouragement always putting my needs ahead of their own. Actually, I cannot thank them enough for all their care throughout my life. This work is dedicated to them!

My dear sister, Yiping and brother-in-law, Chen he, as well as their lovely son, Mengqing, for their always supporting me, understanding me, and sharing my happiness and sadness.

Finally, Jiang, my love and best friend, for his never failing support and encouragement. Without his support, this work has not been possible. My beloved daughter, Julie, for her good drawings with creative ideas making my life pictures more colorful and meaningful.

TACK! 53

References

Abe T, Mori T, Hori S, Koike K and Kuwano M (1997) [Multidrug resistance protein (MRP)]. Nippon Rinsho 55:1077-1082. Abraham MH, Chadha HS and Mitchell RC (1994) Hydrogen bonding. 33. Factors that influence the distribution of solutes between blood and brain. J Pharm Sci 83:1257-1268. Ahlenius S, Henriksson I, Magnusson O and Salmi P (1999) In vivo intrinsic efficacy of the 5- HT1A receptor antagonists NAD-299, WAY-100,635 and (S)-(-)-UH-301 at rat brain monoamine receptors. Eur Neuropsychopharmacol 9:15-19. Artigas F, Celada P, Laruelle M and Adell A (2001) How does pindolol improve antidepressant action? Trends Pharmacol Sci 22:224-228. Artigas F, Perez V and Alvarez E (1994) Pindolol induces a rapid improvement of depressed patients treated with serotonin reuptake inhibitors. Arch Gen Psychiatry 51:248-251. Artigas F, Romero L, de Montigny C and Blier P (1996) Acceleration of the effect of selected antidepressant drugs in major depression by 5-HT1A antagonists. Trends Neurosci 19:378-383. Arvidsson LE, Hacksell U, Nilsson JL, Hjorth S, Carlsson A, Lindberg P, Sanchez D and Wikstrom H (1981) 8-Hydroxy-2-(di-n-propylamino)tetralin, a new centrally acting 5- hydroxytryptamine receptor agonist. J Med Chem 24:921-923. Azmitia EC and Segal M (1978) An autoradiographic analysis of the differential ascending projections of the dorsal and median raphe nuclei in the rat. J Comp Neurol 179:641-667. Bai SA, Walle UK, Wilson MJ and Walle T (1983) Stereoselective binding of the (-)-enantiomer of propranolol to plasma and extravascular binding sites in the dog. Drug Metab Dispos 11:394-395. Barnes NM and Sharp T (1999) A review of central 5-HT receptors and their function. Neuropharmacology 38:1083-1152. Begley DJ (1998) In situ brain perfusion, in Introduction to the Blood-Brain Barrier:methodology, biology and pathology (Pardridge WM ed) pp 32-40, Cambridge University Press, Cambridge. Bergley DJ (1996) The blood-brain barrier: principles for targeting peptides and drugs to the central nervous system. J. Pharm. Pharmacol. 48:136-146.

Bianchi C, Siniscalchi A and Beani L (1990) 5-HT1A agonists increase and 5-HT3 agonists decrease acetylcholine efflux from the cerebral cortex of freely-moving guinea-pigs. Br J Pharmacol 101:448-452. Bickel U (1998) Intravenous injection/pharmacokinetics, in Introduction to the Blood-Brain Barrier:methodology, biology and pathology (Pardridge WM ed) pp 41-48, Cambridge University Press, Cambridge. Björk L, Cornfield LJ, Nelson DL, Hillver SE, Anden NE, Lewander T and Hacksell U (1991a) Pharmacology of the novel 5-hydroxytryptamine1A receptor antagonist (S)- 5-fluoro-8- hydroxy-2-(dipropylamino)tetralin: inhibition of (R)-8- hydroxy-2- (dipropylamino)tetralin-induced effects. J Pharmacol Exp Ther 258:58-65. Björk L, Fredriksson A, Hacksell U and Lewander T (1992) Effects of (R)-8-OH-DPAT and the enantiomers of UH-301 on motor activities in the rat: antagonism of (R)-8-OH-DPAT- induced effects. Eur Neuropsychopharmacol 2:141-147. Björk L, Lindgren S, Hacksell U and Lewander T (1991b) (S)-UH-301 antagonizes (R)-8-OH- DPAT-induced cardiovascular effects in the rat. Eur J Pharmacol 199:367-370. 54

Blier P and de Montigny C (1994) Current advances and trends in the treatment of depression [see comments]. Trends Pharmacol Sci 15:220-226. Bordet R, Thomas P and Dupuis B (1998) Effect of pindolol on onset of action of paroxetine in the treatment of major depression: intermediate analysis of a double-blind, placebo- controlled trial. Reseau de Recherche et d'Experimentation Psychopharmacologique. Am J Psychiatry 155:1346-1351. Borst P, Evers R, Kool M and Wijnholds J (1999) The multidrug resistance protein family. Biochim Biophys Acta 1461:347-357. Bradbury M (1979) The concepts of the blood-brain barrier. John Wiley & Sons, London. Brightman MW and Reese TS (1969) Junctions between intimately apposed cell membranes in the vertebrate brain. J Cell Biol 40:648-677. Brown P and Molliver ME (2000) Dual serotonin (5-HT) projections to the nucleus accumbens core and shell: relation of the 5-HT transporter to amphetamine-induced neurotoxicity. J Neurosci 20:1952-1963. Carlsson A, Davis JN, Kehr W, Lindqvist M and Atack CV (1972) Simultaneous measurement of tyrosine and tryptophan hydroxylase activities in brain in vivo using an inhibitor of the aromatic amino acid decarboxylase. Naunyn Schmiedebergs Arch Pharmacol 275:153- 168. Cecchelli R, Dehouck B, Descamps L, Fenart L, Buee-Scherrer VV, Duhem C, Lundquist S, Rentfel M, Torpier G and Dehouck MP (1999) In vitro model for evaluating drug transport across the blood-brain barrier. Adv Drug Deliv Rev 36:165-178. Clark BJ, Menninger K and Bertholet A (1982) Pindolol--the pharmacology of a partial agonist. Br J Clin Pharmacol 13:149S-158S. Clifford EM, Gartside SE, Umbers V, Cowen PJ, Hajos M and Sharp T (1998) Electrophysiological and neurochemical evidence that pindolol has agonist properties at the 5-HT1A autoreceptor in vivo. Br J Pharmacol 124:206-212. Cole SP, Bhardwaj G, Gerlach JH, Mackie JE, Grant CE, Almquist KC, Stewart AJ, Kurz EU, Duncan AM and Deeley RG (1992) Overexpression of a transporter gene in a multidrug- resistant human lung cancer cell line. Science 258:1650-1654. Cole SP and Deeley RG (1998) Multidrug resistance mediated by the ATP-binding cassette transporter protein MRP. Bioessays 20:931-940. Cornfield LJ, Lambert G, Arvidsson LE, Mellin C, Vallgarda J, Hacksell U and Nelson DL (1991) Intrinsic activity of enantiomers of 8-hydroxy-2-(di-n- propylamino)tetralin and its analogs at 5-hydroxytryptamine1A receptors that are negatively coupled to adenylate cyclase. Mol Pharmacol 39:780-787. Cornford EM (1998) The carotid artery single injection technique, in Introduction to the Blood- Brain Barrier:methodology, biology and pathology (Pardridge WM ed) pp 11-23, Cambridge University Press, Cambridge. Cornford EM, Bocash WD, Braun LD, Crane PD, Oldendorf WH and MacInnis AJ (1979) Rapid distribution of tryptophol (3-indole ethanol) to the brain and other tissues. J Clin Invest 63:1241-1248. Cowen PJ, Cohen PR, McCance SL and Friston KJ (1990) 5-HT neuroendocrine responses during psychotropic drug treatment: an investigation of the effects of lithium. J Neurosci Methods 34:201-205. de Boer AG, Gaillard PJ and Breimer DD (1999) The transference of results between blood-brain barrier cell culture systems. Eur J Pharm Sci 8:1-4. 55

Dehouck MP, Dehouck B, Sanchez C, Lemaire M and Cecchelli R (1995) Drug transport to the brain: comparison between in vitro and in vivo models of the blood-brain barrier. European Journal of Pharmaceutical Sciences 3:357-365. Dehouck MP, Meresse S, Delorme P, Fruchart JC and Cecchelli R (1990) An easier, reproducible, and mass-production method to study the blood- brain barrier in vitro. J Neurochem 54:1798-1801. Demeule M, Jodoin J, Gingras D and Beliveau R (2000) P-glycoprotein is localized in caveolae in resistant cells and in brain capillaries. FEBS Lett 466:219-224. Done CJ and Sharp T (1994) Biochemical evidence for the regulation of central noradrenergic activity by 5-HT1A and 5-HT2 receptors: microdialysis studies in the awake and anaesthetized rat. Neuropharmacology 33:411-421. Ecker G and Chiba P (1995) Structure-activity-relationship studies on modulators of the multidrug transporter P-glycoprotein - an overview. Wien Klin Wochenschr 107:681-686. Egginger G, Lindner W, Brunner G and Stoschitzky K (1994) Direct enantioselective determination of (R)- and (S)-propranolol in human plasma. Application to pharmacokinetic studies. J Pharm Biomed Anal 12:1537-1545. Eichelbaum M (1992) Enantiomers: implications and complications in developmental pharmacology. Dev Pharmacol Ther 18:131-134. Ekblom M, Hammarlund-Udenaes M, Lundqvist T and Sjoberg P (1992) Potential use of microdialysis in pharmacokinetics: a protein binding study. Pharm Res 9:155-158. Enquist M and Hermansson J (1990) Separation of the enantiomers of beta-receptor blocking agents and other cationic drugs using a CHIRAL-AGP column. Binding properties and characterization of immobilized alpha 1-acid glycoprotein. J Chromatogr 519:285-298. Fenstermacher JD, Blasberg RG and Patlak CS (1981) Methods for Quantifying the transport of drugs across brain barrier systems. Pharmacol Ther 14:217-248. Fletcher A, Forster EA, Bill DJ, Brown G, Cliffe IA, Hartley JE, Jones DE, McLenachan A, Stanhope KJ, Critchley DJ, Childs KJ, Middlefell VC, Lanfumey L, Corradetti R, Laporte AM, Gozlan H, Hamon M and Dourish CT (1996) Electrophysiological, biochemical, neurohormonal and behavioural studies with WAY-100635, a potent, selective and silent 5-HT1A receptor antagonist. Behav Brain Res 73:337-353. Fox MA and White JK (1994) Stereochemistry, in Organic Chemistry pp 155-192, Jones and Bartlett Publishers, London. Ghersi-Egea JF, Leininger-Muller B, Cecchelli R and Fenstermacher JD (1995) Blood-brain interfaces: relevance to cerebral drug metabolism. Toxicol Lett 82-83:645-653. Ghersi-Egea JF, Leninger-Muller B, Suleman G, Siest G and Minn A (1994) Localization of drug-metabolizing enzyme activities to blood-brain interfaces and circumventricular organs. J Neurochem 62:1089-1096.

Gilbert F, Brazell C, Tricklebank MD and Stahl SM (1988) Activation of the 5-HT1A receptor subtype increases rat plasma ACTH concentration. Eur J Pharmacol 147:431-439. Gozlan H, El Mestikawy S, Pichat L, Glowinski J and Hamon M (1983) Identification of presynaptic serotonin autoreceptors using a new ligand: 3H-PAT. Nature 305:140-142. Gozlan H, Ponchant M, Daval G, Verge D, Menard F, Vanhove A, Beaucourt JP and Hamon M (1988) 125I-Bolton-Hunter-8-methoxy-2-[N-propyl-N-propylamino]tetralin as a new selective radioligand of 5-HT1A sites in the rat brain. In vitro binding and autoradiographic studies. J Pharmacol Exp Ther 244:751-759. Greig NH (1992) Drug entry into the brain and its pharmacologic manipulation, in Physiology and pharmacology of the blood-brain-barrier, handbook of experimental pharmacology. (Bradbury MWB ed) pp 131-523, Springer verlag, Berlin. 56

Hall MD, el Mestikawy S, Emerit MB, Pichat L, Hamon M and Gozlan H (1985) [3H]8-hydroxy- 2-(di-n-propylamino)tetralin binding to pre- and postsynaptic 5-hydroxytryptamine sites in various regions of the rat brain. J Neurochem 44:1685-1696. Hammarlund-Udenaes M (2000) The use of microdialysis in CNS drug delivery studies. Pharmacokinetic perspectives and results with analgesics and antiepileptics. Adv Drug Deliv Rev 45:283-294. Hammarlund-Udenaes M, Paalzow LK and de Lange EC (1997) Drug equilibration across the blood-brain barrier--pharmacokinetic considerations based on the microdialysis method. Pharm Res 14:128-134. Hillver SE, Björk L, Hook BB, Cortizo L, Nordvall G, Johansson AM, Ertan A, Csoregh I, Johansson L, Lewander T and Hacksell U (1996) Synthesis and pharmacology of the enantiomers of UH301: opposing interactions with 5-HT1A receptors. Chirality 8:531- 544. Hillver SE, Björk L, Li YL, Svensson B, Ross S, Anden NE and Hacksell U (1990) (S)-5-fluoro- 8-hydroxy-2-(dipropylamino)tetralin: a putative 5-HT1A- receptor antagonist [letter]. J Med Chem 33:1541-1544. Hjorth S (1985) Hypothermia in the rat induced by the potent serotoninergic agent 8-OH- DPAT. J Neural Transm 61:131-135.

Hjorth S (1993) Serotonin 5-HT1A autoreceptor blockade potentiates the ability of the 5- HT reuptake inhibitor citalopram to increase nerve terminal output of 5- HT in vivo: a microdialysis study. J Neurochem 60:776-779. Hjorth S and Carlsson A (1985) (-)-Pindolol stereospecifically inhibits rat brain serotonin (5-HT) synthesis. Neuropharmacology 24:1143-1146. Hjorth S, Carlsson A, Lindberg P, Sanchez C, Wikström H, Arvidsson LE, Hacksell U and Nilsson JL (1982) 8-Hydroxy-2-di-n-propylaminotetralin, 8-OH-DPAT, a potent and selective simple ergot congener with 5-HT-receptor stimulating activity. J. Neural Transm. 55:168-188.

Hjorth S and Pettersson G (1993) 5-HT1A autoreceptor-mediated effects of the amperozide congeners, FG5865 and FG5893, on rat brain 5-hydroxytryptamine neurochemistry in vivo. Eur J Pharmacol 238:357-367. Howe R and Shanks RG (1966) Optical isomers of propranolol. Nature 210:1336-1338.

Hoyer D (1988) Functional correlates of serotonin 5-HT1 recognition sites. J Recept Res 8:59-81. Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylecharane EJ, Saxena PR and Humphrey PP (1994) International Union of Pharmacology classification of receptors for 5- hydroxytryptamine (Serotonin). Pharmacol Rev 46:157-203. Huai-Yun H, Secrest DT, Mark KS, Carney D, Brandquist C, Elmquist WF and Miller DW (1998) Expression of multidrug resistance-associated protein (MRP) in brain microvessel endothelial cells. Biochem Biophys Res Commun 243:816-820. Janzer RC and Raff MC (1987) Astrocytes induce blood-brain barrier properties in endothelial cells. Nature 325:253-257.

Johansson CE, Meyerson BJ and Hacksell U (1991) The novel 5-HT1A receptor antagonist (S)- UH-301 antagonizes 8-OH-DPAT- induced effects on male as well as female rat copulatory behaviour. Eur J Pharmacol 202:81-87. Johansson L, Sohn D, Thorberg SO, Jackson DM, Kelder D, Larsson LG, Renyi L, Ross SB, Wallsten C, Eriksson H, Hu PS, Jerning E, Mohell N and Westlind-Danielsson A (1997) The pharmacological characterization of a novel selective 5- hydroxytryptamine1A receptor antagonist, NAD-299. J Pharmacol Exp Ther 283:216-225. 57

Kia HK, Miquel MC, Brisorgueil MJ, Daval G, Riad M, El Mestikawy S, Hamon M and Verge D (1996) Immunocytochemical localization of serotonin1A receptors in the rat central nervous system. J Comp Neurol 365:289-305. Kusuhara H, Suzuki H, Terasaki T, Kakee A, Lemaire M and Sugiyama Y (1997) P-Glycoprotein mediates the efflux of quinidine across the blood-brain barrier. J Pharmacol Exp Ther 283:574-580. Letrent SP, Polli JW, Humphreys JE, Pollack GM, Brouwer KR and Brouwer KL (1999) P- glycoprotein-mediated transport of morphine in brain capillary endothelial cells. Biochem Pharmacol 58:951-957. Levin VA (1980) Relationship of octanol/water partition coefficient and molecular weight to rat brain capillary permeability. J Med Chem 23:682-684. Lin TH, Sawada Y, Sugiyama Y, Iga T and Hanano M (1987) Effects of albumin and alpha 1- acid glycoprotein on the transport of imipramine and desipramine through the blood- brain barrier in rats. Chem Pharm Bull (Tokyo) 35:294-301. Liu Y, Cortizo L, Yu H, Svensson B, Lewander T and Hacksell U (1995) C8-Substituted derivatives of 2-(dipropylamino)tetralin: exploration of the effect of C8-aryl and heteroaryl substituents on the interaction with 5-HT1A-receptors. Eur J Med Chem 30:277-286. Liu Y, Yu H, Svensson BE, Cortizo L, Lewander T and Hacksell U (1993) Derivatives of 2- (dipropylamino)tetralin: effect of the C8-substituent on the interaction with 5-HT1A receptors. J Med Chem 36:4221-4229. Lombardo F, Blake JF and Curatolo WJ (1996) Computation of brain-blood partitioning of organic solutes via free energy calculations. J Med Chem 39:4750-4755.

Lucki I (1992) 5-HT1 receptors and behavior. Neurosci Biobehav Rev 16:83-93. Manners CN, Payling DW and Smith DA (1988) Distribution coefficient, a convenient term for the relation of predictable physico-chemical properties to metabolic processes. Xenobiotica 18:331-350. Martinez D, Broft A and Laruelle M (2000) Pindolol augmentation of antidepressant treatment: recent contributions from brain imaging studies. Biol Psychiatry 48:844-853. Maura G, Roccatagliata E and Raiteri M (1986) Serotonin autoreceptor in rat hippocampus: pharmacological characterization as a subtype of the 5-HT1 receptor. Naunyn Schmiedebergs Arch Pharmacol 334:323-326. Middlemiss DN (1984) Stereoselective blockade at [3H]5-HT binding sites and at the 5-HT autoreceptor by propranolol. Eur J Pharmacol 101:289-293. Middlemiss DN, Neill J and Tricklebank MD (1985) Subtypes of the 5-HT receptor involved in hypothermia and forepaw treading induced by 8-OH-DPAT. Br J Pharmacol 85:251p. Minakawa T, Bready J, Berliner J, Fisher M and Cancilla PA (1991) In vitro interaction of astrocytes and pericytes with capillary-like structures of brain microvessel endothelium. Lab Invest 65:32-40. Molliver ME (1987) Serotonergic neuronal systems: what their anatomic organization tells us about function. J Clin Psychopharmacol 7:3S-23S. Montgomery SA, Loft H, Sanchez C, Reines EH and Papp M (2001) Escitalopram (S-enantiomer of citalopram): clinical efficacy and onset of action predicted from a rat model. Pharmacol Toxicol 88:282-286. Monti JM, Jantos H, Silveira R, Reyes-Parada M and Scorza C (1995) Sleep and waking in 5,7- DHT-lesioned or (-)-pindolol-pretreated rats after administration of buspirone, ipsapirone, or gepirone. Pharmacol Biochem Behav 52:305-312. 58

Murai-Kushiya M, Okada S, Kimura T and Hasegawa R (1993) Stereoselective binding of beta- blockers to purified rat alpha 1-acid glycoprotein. J Pharm Pharmacol 45:225-228. Oldendorf WH (1974) Lipid solubility and drug penetration of the blood brain barrier. Proc Soc Exp Biol Med 147:813-815. Oldendorf WH, Cornford ME and Brown WJ (1977) The large apparent work capability of the blood-brain barrier: a study of the mitochondrial content of capillary endothelial cells in brain and other tissues of the rat. Ann Neurol 1:409-417. Pardridge WM (1986) Blood-brain barrier transport of nutrients. Nutr Rev 44 Suppl:15-25. Pardridge WM (1987) Plasma protein-mediated transport of steroid and thyroid hormones. Am J Physiol 252:E157-164. Pardridge WM (1995) Transport of small molecules through the blood-brain barrier: biology and methodology. Adv. Drug Delivery Rev. 15:5-36. Pardridge WM (1998) Methodology, biology and pathology, in Introduction to the Blood-Brain Barrier, Cambridge University Press. Pardridge WM (1999) Blood-brain barrier biology and methodology. J Neurovirol 5:556-569. Pardridge WM and Landaw EM (1985) Testosterone transport in brain: primary role of plasma protein-bound hormone. Am J Physiol 249:E534-542. Pauwels PJ (2000) Diverse signalling by 5-hydroxytryptamine (5-HT) receptors. Biochem Pharmacol 60:1743-1750. Pazos A, Cortes R and Palacios JM (1985) Quantitative autoradiographic mapping of serotonin receptors in the rat brain. II. Serotonin-2 receptors. Brain Res 346:231-249. Pazos A and Palacios JM (1985) Quantitative autoradiographic mapping of serotonin receptors in the rat brain. I. Serotonin-1 receptors. Brain Res 346:205-230. Radja F, Daval G, Hamon M and Verge D (1992) Pharmacological and physicochemical properties of pre-versus postsynaptic 5-hydroxytryptamine1A receptor binding sites in the rat brain: a quantitative autoradiographic study. J Neurochem 58:1338-1346. Rapoport SI and Pettigrew KD (1979) A heterogenous, pore-vesicle membrane model for protein transfer from blood to cerebrospinal fluid at the choroid plexus. Microvasc Res 18:105- 119. Reese TS and Karnovsky MJ (1967) Fine structural localization of a blood-brain barrier to exogenous peroxidase. J Cell Biol 34:207-217. Rochat B, Baumann P and Audus KL (1999) Transport mechanisms for the antidepressant citalopram in brain microvessel endothelium. Brain Res 831:229-236. Sanchez C, Arnt J and Moltzen EK (1996) The antiaggressive potency of (-)-penbutolol involves both 5-HT1A and 5- HT1B receptors and beta-adrenoceptors. Eur J Pharmacol 297:1-8. Schinkel AH (1999) P-Glycoprotein, a gatekeeper in the blood-brain barrier. Adv Drug Deliv Rev 36:179-194. Schinkel AH, Smit JJ, van Tellingen O, Beijnen JH, Wagenaar E, van Deemter L, Mol CA, van der Valk MA, Robanus-Maandag EC, te Riele HP and et al. (1994) Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood-brain barrier and to increased sensitivity to drugs. Cell 77:491-502. Scremin OU, Scremin AM, Somani SM and Giacobini E (1990) Brain regional distribution of physostigmine and its relation to cerebral blood flow following intravenous administration in rats. J Neurosci Res 26:188-195. Sharp T, Bramwell SR, Hjorth S and Grahame-Smith DG (1989) Pharmacological characterization of 8-OH-DPAT-induced inhibition of rat hippocampal 5-HT release in vivo as measured by microdialysis. Br J Pharmacol 98:989-997. 59

Sharp T, McQuade R, Fozard JR and Hoyer D (1993) The novel 5-HT1A receptor antagonist, SDZ 216-525, decreases 5-HT release in rat hippocampus in vivo. Br J Pharmacol 109:699- 702. Sharp T, Umbers V and Gartside SE (1997) Effect of a selective 5-HT reuptake inhibitor in combination with 5- HT1A and 5-HT1B receptor antagonists on extracellular 5-HT in rat frontal cortex in vivo. Br J Pharmacol 121:941-946. Slovakova A and Hutt AJ (1999) [Chiral compounds and their pharmacologic effects]. Ceska Slov Farm 48:107-112. Takahashi H, Ogata H, Kanno S and Takeuchi H (1990) Plasma protein binding of propranolol enantiomers as a major determinant of their stereoselective tissue distribution in rats. J Pharmacol Exp Ther 252:272-278. Tamai I, Sai Y, Ono A, Kido Y, Yabuuchi H, Takanaga H, Satoh E, Ogihara T, Amano O, Izeki S and Tsuji A (1999) Immunohistochemical and functional characterization of pH- dependent intestinal absorption of weak organic acids by the monocarboxylic acid transporter MCT1. J Pharm Pharmacol 51:1113-1121. Tamai I and Tsuji A (2000) Transporter-mediated permeation of drugs across the blood-brain barrier. J Pharm Sci 89:1371-1388. Terasaki T (1998) Development of brain efflux index (BEI) method and its application to the blood-brain barrier efflux transport study, in Introduction to the blood-brain barrier: methodology, biology and pathology (Pardridge WM ed) pp 24-31, Cambridge University Press, Cambridge. Tonini M, Vigneri S, Savarino V and Scarpignato C (2001) Clinical pharmacology and safety profile of esomeprazole, the first enantiomerically pure proton pump inhibitor. Dig Liver Dis 33:600-606. Tricklebank MD, Forler C, Middlemiss DN and Fozard JR (1985) Subtypes of the 5-HT receptor mediating the behavioural responses to 5- methoxy-N,N-dimethyltryptamine in the rat. Eur J Pharmacol 117:15-24. Tsuji A and Tamai I (1996) Carrier-mediated intestinal transport of drugs. Pharm Res 13:963- 977. Tsuji A and Tamai I (1998) Blood-brain barrier transport of drugs, in Introduction to the Blood- Brain Barrier:methodology, biology and pathology (Pardridge WM ed) pp 238-247, Cambridge University Press, Cambridge, UK. Walle T, Webb JG, Bagwell EE, Walle UK, Daniell HB and Gaffney TE (1988) Stereoselective delivery and actions of beta receptor antagonists. Biochem Pharmacol 37:115-124. Walle UK, Walle T, Bai SA and Olanoff LS (1983) Stereoselective binding of propranolol to human plasma, alpha 1-acid glycoprotein, and albumin. Clin Pharmacol Ther 34:718- 723. van Bree JB, Audus KL and Borchardt RT (1988a) Carrier-mediated transport of baclofen across monolayers of bovine brain endothelial cells in primary culture. Pharm Res 5:369-371. Van Bree JB, De Boer AG, Danhof M and Breimer DD (1992a) Drug transport across the blood-- brain barrier. I. Anatomical and physiological aspects. Pharm Weekbl Sci 14:305-310. Van Bree JB, De Boer AG, Danhof M and Breimer DD (1992b) Drug transport across the blood- brain barrier. II. Experimental techniques to study drug transport. Pharm Weekbl Sci 14:338-348. van Bree JB, de Boer AG, Danhof M, Ginsel LA and Breimer DD (1988b) Characterization of an "in vitro" blood-brain barrier: effects of molecular size and lipophilicity on cerebrovascular endothelial transport rates of drugs. J Pharmacol Exp Ther 247:1233- 1239. 60 van de Waterbeemd H, Camenisch G, Folkers G, Chretien JR and Raevsky OA (1998) Estimation of blood-brain barrier crossing of drugs using molecular size and shape, and H-bonding descriptors. J Drug Target 6:151-165. van den Heuvel-Eibrink MM, Sonneveld P and Pieters R (2000) The prognostic significance of membrane transport-associated multidrug resistance (MDR) proteins in leukemia. Int J Clin Pharmacol Ther 38:94-110. Vertes RP, Kinney GG, Kocsis B and Fortin WJ (1994) Pharmacological suppression of the median raphe nucleus with serotonin1A agonists, 8-OH-DPAT and buspirone, produces hippocampal theta rhythm in the rat. Neuroscience 60:441-451. Wikström H, Elebring T, Hallnemo G, Andersson B, Svensson K, Carlsson A and Rollema H (1988) Occurrence and pharmacological significance of metabolic ortho- hydroxylation of 5- and 8-hydroxy-2-(di-n-propylamino)tetralin. J Med Chem 31:1080-1084. Wright DE, Seroogy KB, Lundgren KH, Davis BM and Jennes L (1995) Comparative localization of serotonin1A, 1C, and 2 receptor subtype mRNAs in rat brain. J Comp Neurol 351:357-373. Ylitalo P, Linden IB, Penttila A and Vapaatalo H (1986) Distribution of beta-adrenoceptor blocking drugs into cerebrospinal fluid in rats. Acta Pharmacol Toxicol (Copenh) 58:84- 87. Yu H and Lewander T (1997) Pharmacokinetic and pharmacodynamic studies of (R)-8-hydroxy- 2-(di-n- propylamino)tetralin in the rat. Eur Neuropsychopharmacol 7:165-172. Yu H, Liu Y, Hacksell U and Lewander T (1996a) Oral cavity absorption of (R)-8-hydroxy-2-(di- n-propylamino)tetralin and (S)-8-acetyl-2-(di-n-propylamino)tetralin in the rat. J Pharm Pharmacol 48:41-45. Yu H, Liu Y, Li HB, Martin AR, Hacksell U and Lewander T (1997) Pharmacodynamic and pharmacokinetic studies in rats of S-8-(2-Furyl)- and R-8-phenyl-2-(di-n-propylamino) tetralin, two novel 5-HT1A receptor agonists in-vitro with different properties in-vivo. J Pharm Pharmacol 49:169-177. Yu H, Liu Y, Malmberg A, Mohell N, Hacksell U and Lewander T (1996b) Differential serotoninergic and dopaminergic activities of the (R)- and the (S)-enantiomers of 2-(di-n- propylamino)tetralin. Eur J Pharmacol 303:151-162. Zanardi R, Artigas F, Franchini L, Sforzini L, Gasperini M, Smeraldi E and Perez J (1997) How long should pindolol be associated with paroxetine to improve the antidepressant response? J Clin Psychopharmacol 17:446-450. Zanardi R, Franchini L, Gasperini M, Lucca A, Smeraldi E and Perez J (1998) Faster onset of action of fluvoxamine in combination with pindolol in the treatment of delusional depression: a controlled study. J Clin Psychopharmacol 18:441-446. Zifa E and Fillion G (1992) 5-Hydroxytryptamine receptors. Pharmacol Rev 44:401-458.