UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of genetic variation in the human 5HT Type 2A G-protein coupled receptor (GPCR) Harvey, Layne Joseph 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2003-0082.pdf [ 4.17MB ]
Metadata
JSON: 831-1.0090781.json
JSON-LD: 831-1.0090781-ld.json
RDF/XML (Pretty): 831-1.0090781-rdf.xml
RDF/JSON: 831-1.0090781-rdf.json
Turtle: 831-1.0090781-turtle.txt
N-Triples: 831-1.0090781-rdf-ntriples.txt
Original Record: 831-1.0090781-source.json
Full Text
831-1.0090781-fulltext.txt
Citation
831-1.0090781.ris

Full Text

Characterization of genetic variation in the human 5HT Type 2A G-protein coupled receptor (GPCR) by Layne Joseph Harvey B.Sc., The University of British Columbia 1994 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF T H E REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Zoology) We accept this th^sisas conforming tiy(h^eqy(re$ standard THE UNIVERSITY OF BRITISH C O L U M B I A December 2002 © Layne Joseph Harvey i 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract Human serotonin type 2A (5HT2A) receptors are members of the large class of proteins known as G-protein coupled receptors (GPCRs). These receptors are implicated in a variety of human disorders including schizophrenia, anxiety, depression, migraine and * obesity. Recently several allelic variations in the 5HT2A receptor have been identified and attempts have been made to implicate these mutations in the etiology of schizophrenia as well as the variability in patient response to neuroleptic drugs. The objective of this work was to determine i f these allelic variants in the 5HT2 receptors affect the potency of the agonist serotonin, the typical antagonist loxapine, and the atypical antagonist clozapine. In this study four naturally occurring polymorphisms, (T25N, I197V, A447V, and H452Y) were generated by site directed mutagenesis in the human 5HT2A gene. The activity of the 5HT2A wild type and mutant receptors was investigated using an in-vitro functional GPCR assay. This assay measures G-protein receptor activity when challenged with agonists or antagonists. The results indicated that the I197V missense variant required a two fold higher concentration of clozapine to inhibit serotonin receptor activation when compared to the wildtype receptor. The I197V variant had no effect on serotonin potency or on the potency of loxapine. This study also indicated that there was no change in potency for the T25N, A447V and H452Y variants K when challenged with the agonist serotonin or the antagonists loxapine and clozapine. ii Table of contents CHARACTERIZATION OF HUMAN GENETIC VARIATION OF 5HT TYPE 2A GPCRs IN INSECT CELLS Abstract ii Table of Contents ii List of figures v List of tables viii Acknowledgments ix Chapter 1 INTRODUCTION 1 1.1 Pharmacogenomics and single nucleotide polymorphisms (SNP) 1.2 GPCRs and G protein interactions 3 1.3 Sertonin Receptors ; 1.4 Sertonin Type 2 (5HT2) Receptors 1.5 5HT2A receptors 1.6 Mutational analysis of the 5HT2A receptor 10 1.7 Single nucleotide polymorphism of the 5HT2A receptor 12 1.8 Schizophrenia and neuroleptic drugs 16 1.9 GPCR functional assay li Chapter 2 Materials and Methods 24 2.1 Cloning wildtype 5HT2A human cDNA into an expression vector 2< (a) cDNA and expression vector 2^  (b) Cloning strategy 24 ii i 2.2 Engineering of naturally occurring 5HT2A variants 25 2.3 DNA preparations 27 2.4 Sequencing of all constructs 29 2.5 Maintenance of insect cell line 30 2.6 Counting cells 31 2.7 Transient expression of the human 5HT2A receptor in Sf-9 cells 32 2.8 Selection of stable lines 33 2.9 Protein preparation and quantification for Western Blot Analysis 34 2.10 Polyacrylamide gel separation and Western blot analysis 35 2.11 GPCR functional assay of transiently transformed cells 37 2.12 Agonist Dose Response Experiments 39 2.13 Antagonist Dose Response Experiments 40 2.14 Competition Studies 41 2.15 Standardization of fractional luminescence 42 2.16 Graph and Curve fitting 43 2.17 Coelenterazine leaching experiments 44 2.18 Effects of Solvents on stable cell lines 45 2.19 Effects of extended incubation of stable cell lines with antagonists 46 Chapter 3 Results 47 3.1 Transient expression of the human 5HT2A receptor in Sf-9 cells 47 3.2 The human 5HT2A receptor functions in Sf-9 insect cells 49 3.3 Creation of stable Sf-9 cell line expressing 5HT2A receptor, Ga 16 and aequorin 52 3.4 Determination of the effects of coelenterazine leaching from charged cells 52 iv 3.5 Serotonin Dose Response Experiments 56 3.6 Serotonin Dose Response Curves 57 3.7 Comparison of LogEC50 values between wild type and allelic variants 60 3.8 Effects of ethanol and DMSO on the Sf-9 insect cell line 62 3.9 Loxapine and clozapine dose response experiments 64 3.10 Loxapine Dose Response Curves 65 3.11 Comparison of the LogIC50 values between the wildtype and allelic variants for loxapine 68 3.12 Clozapine Dose Response Curves 70 3.13 Comparison of the LogIC50 values between the wildtype and allelic variants for clozapine 72 3.15 Does long exposure to antagonist affect the response of GPCRs in stable cell lines? 74 3.16 Competitive versus not-competitive assays 76 Chapter 4 Discussion 78 4.1 Justification for using and insect cell based assay 78 4.2 Effects of missense variants on drug potencies 80 4.3 Antagonist mode of action 83 4.4 Physiological relevance of the I197V variant 84 4.5 Future directions 87 Bibliography 90 v List of Figures Figure 1-1 Schematic representation of the signal transduction mechanism of G protein regulated receptors 5 Figure 1-2. Schematic diagram of the 5HT2A receptor and location of allelic variations 15 Figure 1-3 Chemical structure of the typical drug loxapine and atypical drug clozapine 17 Figure 1-4 GPCR functional assay system 21 Figure 2-1 The Insect Select expression vector p2Zop2F 25 Figure 2-2 Graph demonstrating the calculation of fractional luminescence (A/(A+L)) 39 Figure 2-3 Typical agonist layout of the first row from a 96 well plate 43 Figure 3-1. Western blot analysis of Sf9 cells transiently expressing the human 5HT2A. 48 Figure 3-2 Bar graph depicting fractional luminescence of transiently transformed Sf-9 cells 51 Figure 3-3 Experimental procedures for agonist dose response curves 53 Figure 3-4a Comparison of fractional luminescent values between stable Sf-9 cells with a change of media and without a change of media 55 Figure 3-4b Comparison of total luminescent values between stable Sf-9 cells with a change of media and without a change of media 55 Figure 3-5 Serotonin concentration layout into a 96 well plate 56 Figure 3-6 Comparison of maximal responses (ImM 5HT) achieved at the beginning and at the end of a dose response experiment 57 Figure 3-7 Serotonin dose response curves 59 Figure 3-8 Comparison of serotonin LogEC 5 0 values 61 Figure 3-9a The fractional luminescence response of stable Sf-9 cells challenged with varying concentrations of DMSO 63 VI Figure 3-9b The fractional luminescence response of stable Sf-9 cells challenged with varying concentrations of ethanol 63 Figure 3-10 Clozapine concentration layout into a 96 well plate for dose response experiments 65 Figure 3.11 Loxapine dose response curves 67 Figure 3-12 Comparison of loxapine LogIC 5 0 values 69 Figure 3-13 Clozapine dose response curves 71 Figure 3-14 Comparison of clozapine LogIC 5 0 values 73 Figure 3-15a Fractional luminescent values for the human wildtype 5HT2A receptor in the presence of 10 nM loxapine 75 Figure 3-15b Fractional luminescent values for the human wildtype 5HT2A receptor in the presence of 73 n M clozapine 75 Figure 3-16a Competition studies with serotonin and loxapine 77 Figure 3-16b Competition studies with serotonin and clozapine 77 Figure 4-1 Scheme for the reaction between drug A and receptor R 82 vii List of Tables Table 1-1 Classification of 5-HT Receptors 7 Table 1-LT Allelic variations in 5HT2A receptor 14 Table 3-1 Transient transformations performed in Sf-9 cells 50 Table 3-II Summary of Log E C 5 0 values obtained for 5HT2A receptors with Serotonin 61 Table 3-III Summary of Log IC 5 0 values obtained for 5HT2A receptors with loxapine 69 Table 3-IV Summary of Log IC 5 0 values obtained for 5HT2A receptors with clozapine 73 viii Acknowledgements This study could not have been undertaken without the aid and guidance provided by a great many people. Firstly, I would like to thank my supervisor and mentor Dr. Tom Grigliatti, for the countless hours of listening time. Dr. Ron Reid has been the driving force behind this project as a whole and his knowledge of pharmacology has been an essential asset. I would like to thank all the members of the Grigliatti lab, especially Dr. Tom Pfeifer and Dr. Peter Knight. Their ability to impart their knowledge and skills onto others has allowed me to succeed. Thank-you to Caixia Ma for her help in obtaining the 5HT2A polymorphs. The companionship of my friends and fellow classmates Jason and Natasha McNamee, provided comfort when all seemed lost. To my wide circle of family and friends, thank you for the unwavering support. Finally, I would like to acknowledge and thank my wife Kate; without her love, understanding and support I would never have been able to succeed. ix Chapter 1 Introduction 1.1 Pharmacogenomics and single nucleotide polymorphisms (SNP) Pharmacogenomics examines the influence of genetic variation on patient response to specific drug therapies. The correlation of genotype with phenotype in drug therapy requires an understanding of a complex array of gene products involved in drug metabolism, drug transport and drug target interactions (Meyer, 2000; Liggett, 2000). Psychiatric or behavioral disorders are of particular interest because the pharmacotherapy of these disorders exhibits a wide variability in therapeutic response with little scientific guidance for choice of treatment on a patient-by-patient basis (Pickar and Rubinow, 2001). The ultimate goal of pharmacogenomics is to identify genetic variations that affect drug function, thereby enhancing drug discovery and providing a scientific basis for drug therapies based on a patient's genetic variability (Kalow, 2001). The aim of this study is to determine if genetic variants in drug targets, in particular the human 5HT Type 2A receptor, have an effect on drug potency in an attempt to explain variable patient response seen with neuroleptics used to treat schizophrenia. To date, only the effects of allelic variants on drug metabolism have been extensively studied; whereas, the allelic variations affecting drug transport and drug-target protein interactions are poorly understood in terms of their importance in drug potency and efficacy (Pickar and Rubinow, 2001). The paucity of knowledge about the effects of drug transport and drug-target variants is limited due to lack of scientific techniques to 1 properly study them (Cichon, et al. 2000). In order to develop a better understanding of the effects amino acid variants in drug targets have on variable patient response, a precise molecular assay is required to provide a consistent definition of the drug response phenotype. In conjunction with this assay a method to identify genetic variations that occur within these drug targets is required. The Human Genome Project has begun to reveal a pattern of gene sequence variation in human genes that appears largely in the form of single nucleotide polymorphisms (SNPs) (Cargill, et al. 1999, Cravchik, et al. 2000). These single nucleotide differences in D N A code exist throughout the human genome at a frequency of 1 in 1000 to 2000 basepairs (Kawanishi, et al. 2000). Many nucleotide exchanges have been identified in non-coding regions of the genome; however, numerous SNPs have been found to be located in important regulatory regions of genes or within their coding regions (Cichon, et al. 2000). These nucleotide changes can potentially impact gene and protein function. Presently, the SNP Consortium is assembling a single nucleotide polymorphism map of the human genome, which provides a starting point for variation studies (Kawanishi, et al. 2000). This map together with advances in molecular genetic assay techniques for protein function provides the foundation for examining variants in drug uptake, transport metabolism and drug target response. One group of drug targets that have been screened for the presence of polymorphisms are the dopamine and serotonin receptors (Cargill, et al. 1999). The dopamine and serotonin receptors are members of the superfamily of receptors known as G-protein coupled 2 receptors (GPCRs). Both families of receptors are involved in a variety of behaviors including feeding, mood, and cognitive behaviors, and by corollary are associated with eating disorders, mood effects disorders, depression, and schizophrenia (Cowen, 1991; Meltzer and Nash, 1991; Petronis, 2000; Roth, 1998; Saxena, 1995; Schmidt, et al.1995). Due to their involvement in many disorders, these receptors provide specific targets for drug action and polymorphisms identified in these receptors may play a role in the observed variable patient response to drugs. 1.2 GPCRs and G protein interactions G protein coupled receptors (GPCRs) are likely the largest single family of proteins in vertebrate genomes (Leurs, et al. 1998). They are a functionally diverse group of receptors that are involved in many physiological responses. A l l GPCR's span the cell membrane and are characterized by seven membrane spanning domains connected by three extracellular and three intracellular loops, an extracellular N terminus and an intracellular C terminus (Hamm, 1998). Comparing the sequences of GPCR's reveals the existence of several different receptor families with little or no sequence similarity (Bockaert, et al. 1999). There are three main families: Type A , (the largest group), the rhodopsin related receptors; Type B , the calcitonin related receptors; and Type C, the metabotropic related receptors (Gether, et al. 1998). There are also three lesser families: Family 4 comprising the pheromone receptors, Family 5 including the frizzled and smoothened receptors involved with embryonic development; and, finally, the c A M P receptors which, to date, have only been found in D. discoideum (Bockaert, et al 1999). 3 Each family is comprised of its own grouping of receptor subtypes, encoded by a different gene with characteristic biological activities and ligand specificities. GPCRs regulate GTP-hydrolases called G proteins. These heterotrimeric G-proteins are responsible for transducing receptor activation via ligand binding, into intracellular responses. G-proteins consist of three separate subunits, G a , G(3 and Gy, which remains heterotrimeric and bound to GDP when a GPCR is in an inactive state. Once a GPCR is activated through ligand binding it associates with an appropriate G-protein (Figure 1-1) and facilitates the exchange of GDP for GTP causing the G-protein complex to separate into the two subunits, G a and Gpy. Separation of the heterotrimeric G protein into G a and GPy subunits, in turn, activates intracellular effector molecules and initiates a cascade of biochemical events, which results in the observed pharmacological activity of the receptor agonist (Clapham, and Neer, 1997, Raymond, 1995). Within the three subunits of the heterotrimeric G proteins there are many isoforms. To date, eighteen a subunits, five P subunits, and eleven y subunits have been identified in the human genome (Hildebrandt, 1997). Both the a monomer and the (3y subunit have regulatory activity (Figure 1-1); however, the activity of G a proteins is best characterized (Clapham and Neer, 1997, Raymond, 1995, Hildebrant, 1997). The G a proteins are divided into four classes based upon the downstream proteins that they regulate: the G ^ which activate adenyl cyclase, G a j which inhibit adenyl cyclase, G,^ which activates phospholipase C, and G a l 2 and G a l 3 which have been reported to couple with the thrombin, thromboxane and angiotensin receptors, but whose actions are less well characterized (Hamm, 1998). 4 GTP GDP Figure 1-1 Schematic representation of the signal transduction mechanism of G protein regulated receptors. A = receptor agonist; R = receptor; cc,p\ and y = G protein subunits; E; = hypothetical enzyme regulated by the (3y subunit dimer of the G protein; E 2 = the hypothetical enzyme regulated by the a subunit of the G protein. 1.3 Sertonin Receptors Serotonin (5-hydroxytryptamine; 5HT) is a major neurotransmitter involved in a large number of CNS processes. Serotonin appears to play a role in the pathophysiology of a 5 range of neuropsychiatric disorders, including among others, schizophrenia, depression, eating and mood effecting disorders (Cowen, 1991; Golimbet, et al. 2002; Roth, 1998; Schmidt, et al. 1995). Peripherally, 5HT is important for the regulation of vascular and non-vascular smooth muscle contraction, platelet aggregation, uterine smooth muscle growth, and gastrointestinal functioning. The serotonin receptor family is large and diverse consisting of thirteen G-protein coupled receptors and one ligand-gated ion channel (Cravchik, et al. 2000). These receptors have been classified by recognitory, transductional, and structural characteristics and have been divided into seven groups (Hoyer and Martin, 1997; see Table 1-1). Agents that interact with serotonergic receptors are of central importance in neuropharmacology (Veenstra-VanderWeele, et al. 2000). The 5HT Type 2 class of receptors appears to be important drug targets in the treatment of psychological disorders (Porter, et al. 1999) and one member (5HT2A) is the focus of this study. 6 Tab! e 1-1 Classification of 5-HT Receptors Selective Selective Radioligands Effector Gene Structural Receptor Agonists Antagonists Information 5HT,A 8-OH- (±)WAY [23]-8-OH-DPAT Gi/o 5ht,A h421aa; 7TM DPAT 100635 [23]-WAY P8908 100635 5HT1B SB GR 55562 [23] sumatriptan Gi/o 5ht1D() h 390aa; 7TM 216641 [24J-GTI P28222 5HT 1 D - - [23] sumatriptan Gi/o 5ht IDa h 377aa; 7TM [24]-GTI P11614 5HT1 E - - [23] 5-HT G i / 0 5ht1E h 365aa; 7TM P28566 5HT1 F - - [23] sumatriptan Gi/„ 5ht1F h 366aa; 7TM [24] LSD P30939 5HT2 A . a-Me-5- kcunserm [23]-ketanserin •G<Vfi ' ' ' -5ht2, h 477aa; 7TM : HT ' M M . 100907 5HT 2 B: a-Me-5- SB200646 |'H]5-HT 'Gq,n 5ht,B h479aa;7TM HT. SB204741 BW723C8 6 5HT2 C ' " (x-Me-5- Mesulergine [3H]-mesulergine * 5ht2C h 458aa: 7TM HT. ., ' SB200646/.; '' ' ' " . 5HT3 2-Me-5- granisetron [3H]-zacopride int. 5ht3A m 487aa; 4TM HT ondansetron [125I]-zacopride cation m- tropisetron channel chlorophe nyl-biguanide 5HT4 BIMU8 GR 113808 [3H]-GR113808 G s 5ht4 r387aa; 7TM RS67506 SB204070 [125I]-SB2047710 5HT 5 a - - [3H]5-CT G s? 5ht5A h 357aa; 7TM [125I]LSD [3H]5-CT Unknown 5HT5P — — 5ht5B m 370aa; 7TM [125I]LSD 5HT6 - - [3H]5-CT G s 5ht6 h 440aa; 7TM [125I]LSD 5HT7 - - [3H]5-CT G s 5ht7 H 445aa; 7TM [125I]LSD [3H]5-HT 7 1.4 Sertonin Type 2 (5HT2) Receptors The 5HT type 2 group consists of three separate receptors, 5HT2A, 5HT2B and 5HT2C (Humphrey, et al. 1993). A l l three members regulate type G proteins and are linked to phosphoinisitide metabolism through the activation of phospholipase C (Hoyer, 1997). The 5HT type 2 receptors demonstrate similar specificities towards various ligands, a further indicator of their shared function (Bonhaus, et al. 1995). Despite these commonalities, the amino acid sequence identities between these receptors are low. Overall only 40% amino acid sequence identity is shared. (Bonhaus, et al. 1995). Comparing the transmembrane domains of 5HT2A and 5HT2C shows a 78% amino acid sequence identity (Stam, et al. 1992). However, when comparing the extracellular N -terminal portion of the receptors, 5HT2A has only about 27% amino acid sequence identity with the 5HT2B and 5HT2C receptors (Wu, et al. 2000). These receptors are expressed in a wide range of tissues throughout the body. In the brain the 5HT2A and 5HT2C receptors have similar expression patterns. The 5HT2A receptors are expressed in pyramidal cells of the brain cortex with lower levels of expression in the basal ganglia and hippocampus; the 5HT2C receptors are found in the choroid plexus, as well as many other regions of the brain including the cortex, basal ganglia, hippocampus and hypothalamus (Roth, et al. 1998). Both the 5HT2A and 5HT2C receptors are found at lower levels in the liver, spleen, pancreas and kidney ((Bonhaus, et al. 1995). Conversely, the 5HT2B receptor is expressed at lower levels in 8 the cerebral cortex and in the brain and at higher levels in the liver, spleen, pancreas and kidney (Bonhaus, et al. 1995). 1.5 5HT2A receptors The human gene responsible for encoding the 5HT2A receptor has been localized to chromosome 13 ql4-q21 (Inayama, et al. 1996). The entire gene spans over 20 kb and analysis of the genomic structure of this gene indicates the presence of three exons separated by two introns (Chen, et al. 1992). It is one of a few serotonin receptors that possess introns (Hoyer and Martin, 1997). When compared with the mouse homolog, the genomic structure of the human 5HT2 receptor is highly conserved and shows an identical intron-exon organization and similar intron sizes (Chen, et al. 1992). The amino acid sequence identity between human, mouse and rat homologues is 91% (Saltzman, et al.1991, Stam, et al. 1992, Chen, et al. 1992). Comparing the sequence identity breakdown between human and rat homologues indicates 98% amino acid identity in the transmembrane domain, 75% amino acid identity in the N-terminal and 67% in the C-terminal of the receptor (Slatzman, et al. 1991). The high sequence identity between the human, mouse, and rat 5HT2A homologues suggests a common ancestral origin (Hurley, et al. 1999). 9 1.6 Mutational analysis of the 5HT2A receptor A high degree of amino acid sequence identity exists in the transmembrane domains of GPCRs. Investigations of these regions have uncovered several amino acids that are conserved in many receptors both in terms of the residue type as well as their location within the transmembrane cc helices (Sealfon, et al. 1995). Attempts to unravel the complexities of 5HT2A receptor activation have resulted in a variety of mutation analyses and biophysical modeling of the 5HT2A receptor. Porter and Perez (1999) postulated that a salt bridge or hydrogen bonding interaction between a highly conserved polar residue in transmembrane domain two (D or Q) and a positively charged polar residue in transmembrane seven (K or N) could form, and was responsible for holding the receptor in an inactive form. Site directed mutagenesis was employed to create several different single amino acid mutations at D155 and N363 (the residues believed to be required for salt bridge formation) in the 5HT2A receptor. Interestingly, the resulting mutant receptors had reduced basal activity instead of constitutive activity as predicted (Kristiansen, et al. 2000). Studies of other 5HT receptors had revealed that the D155 residue may be involved in anchoring indole ligands (Ho, et al. 1992). Analysis of the mutation effects using two tryptamine analogs revealed that D155 in 5HT2A is essential for high affinity binding and further investigation demonstrated involvement of D155 in transport of the receptor to the cell surface (Kristiansen, et al. 2000). These results support the many models of the 5HT2A receptor structure that postulate that a negatively charged carboxylate anion in the third transmembrane domain is involved in the anchoring of the positively charge amine moiety of serotonin (Roth, et al. 1997) 10 Later studies investigated salt bridge formation between other conserved transmembrane amino acids. Shapiro (2002) demonstrated strong ionic interactions between R173, (which is found in the third transmembrane a helix), and E318, (located in the sixth transmembrane a helix), within the rat 5HT2A receptor. Such an interaction is thought to help stabilize the receptor in its inactive form. Alanine scanning mutagenesis studies of the residues surrounding E318 demonstrated that many of these amino acids (N317, Q319, K320, C322, V324 and L325) were involved in agonist activation of the 5HT2A receptor (Shapiro, et al. 2002). Models of agonist binding to the serotonin receptors have alluded to the potential importance of aromatic residues for agonist binding and efficacy. Studies have revealed that F340, a highly conserved aromatic residue is involved in the stabilization of the aromatic ring of the serotonin molecule (Choudhary, et al. 1995). Furthermore, studies of both the bacteriorhodopsin and the vertebrate rhodopsin receptors have highlighted the importance of a number of aromatic residues, located on adjacent transmembrane helixes, for ligand binding (Roth, et al. 1997). A systematic study that made mutations in both conserved and non-conserved (to control for non specific mutagenic effects) transmembrane aromatic residues of the 5HT2A receptor was undertaken. These studies revealed that the three tryptophans W200, W336 and W367, which are believed to compose part of the ligand binding pocket, when mutated greatly reduce agonist binding affinity and efficacy (Roth, et al. 1997). Similar results were seen with the other highly conserved aromatic residues F340 and Y370; however, minimal effects on agonist binding were observed with mutations of the neighboring phenylalanine residues F339 11 and F365 (Roth, et al. 1997). Other key aromatic residues have also been examined in the fifth transmembrane a helix. From binding affinity and phosphoinositide hydrolysis studies it has been demonstrated that F243 and F244 interact specifically with agonists and antagonists (Shapiro, et al. 2000). These results suggest that some but not all of the conserved aromatic residues are involved in agonist binding affinity and efficacy. Although most mutational analysis and molecular modeling have been done with the rat 5HT2A, the information gathered should be directly applicable to the human 5HT2A receptor due to its extremely high level of amino acid sequence similarity. 1.7 Single nucleotide polymorphism of the 5HT2A receptor There is no doubt that mutational analysis combined with functional assays can elucidate the regions or motifs essential for receptor function. Of the mutations created to study receptor activation, none have been found to occur naturally in the human population. The aim of this study is to try to determine if single nucleotide polymorphisms that exist in the human population have an effect on the activity of a series of ligands that target the 5HT2A receptor. Table l-II shows a total of nine SNPs for the 5HT2A receptor, which are documented to exist within the human population. Of those nine, two are silent mutations and two exist in the 3' and 5' non-coding regions of the gene. From these nine, I chose to look at the four SNPs (of the five) that resulted in amino acid substitutions (missense alleles). The fifth missense mutation S421F was only recently added to the SNP database and therefore, it was not included in this study. Curiously, these four missense variants occur in, or in close proximity to, structurally important 12 regions of the 5HT2A receptor. Figure 1-2 is a schematic representation of the 5HT2A receptor and shows the approximate locations of the four variants that were examined. The T25N allelic variant exists in the N-terminal region of the protein. Both A447V and H452Y are located at the carboxyl terminal of the protein, which is thought to be one of the regions for G protein interaction (Morris and Malbon, 1999). The I197V polymorphism is in the fourth transmembrane a helix of the receptor, closest to the intracellular environment near intracellular loop two and in close proximity to the W200, an amino acid thought to be involved in agonist binding (Roth, et al. 1997). The I197V variant may alter the ability of the receptor to undergo the appropriate shifts in conformation that are associated with proper function. 13 Table l-II Allelic variations in 5HT2A receptor NCBI SNP Nucleotide ID Variation Amino Average Average Acid estimated Allele Variation heterozygosity Frequency Type of Mutation rs1805055 74C/A T25N unknown C=0.98 A=0.02 nonsynonymous rs6313 102T/C S34S 0.377 C= 0.676 T= 0.324 synonymous rs6305 516C/T D172D 0.047 C = 0.975 T = 0.025 synonymous rs6304 589A/G 1197V 0.047 A = 0.975 G = 0.025 nonsynonymous rs6308 1340C/T A447V 0.047 C = 0.972 T = 0.028 nonsynonymous rs6314 1354C/T H452Y 0.175 C = 0.895 T = 0.105 nonsynonymous rs1058576 1262C/T S421F unknown unknown nonsynonymous rs3125 1536G/C none unknown unknown 3' non coding region -1438G/A none unknown unknown 5' non-coding promoter region The polymorphism frequencies were obtained from the N C B I SNP Databank (http://www.ncbi.nlm.nih.gov/SNP/index.html), 14 5HT2A Receptor T25N Extracellular environment Intracellular environment Figure 1-2. Schematic diagram of the 5HT2A receptor and location of allelic variations examined in this study Many attempts have been made to implicate some of these polymorphisms with psychological disorders including schizophrenia. Investigators have been unable to find a link between T25N or H452Y and schizophrenia (Erdmann, et al. 1996). With the silent mutation T102C there are conflicting reports. One group demonstrated an association of 15 schizophrenia with the polymorphism (Inayama, et al. 1996), whereas Chen, et al in 2001 demonstrated that there was no association. Clinical studies attempting to link the G/A polymorphism in the 5' promoter region (-1438 bp) flanking the 5HT2A receptor gene to mood disorders were also not successful (Ohara, et al. 1998). These studies have produced somewhat equivocal results and have not led to any significant conclusions regarding the causes of schizophrenia; however these studies have led investigators to postulate an involvement of the polymorphisms of the 5HT2A receptor in variable patient response to drugs used to treat schizophrenia. 1.8 Schizophrenia and neuroleptic drugs Schizophrenia is a major psychological disorder that affects approximately 1% of the world's population (Chen, et al. 2001). Clinically, schizophrenia appears to be a complex trait, however, it is characterized by several predictable positive and/or negative symptoms. Positive symptoms are behaviors that are present in schizophrenics, but not in the general population and include thought disorders, delusions and hallucinations (King, 1998). Negative symptoms are abnormal behaviors found among schizophrenics but not healthy individuals and include apathy, affective flattening, social withdraw and poverty of speech (King, 1998). The age of onset of the disorder typically peaks in the early twenties with fewer cases developing beyond the age of 30 (Andreoli, et al. 2002). Fortunately, a number of neuroleptic drug therapies provide relief to many of those affected with schizophrenia. However, there are still many people who respond only moderately to these drug therapies and still require extensive support and many who do not respond at all (Mancama, et al. 2002). 16 The neuroleptic drugs used to treat schizophrenia fall into two classes, typical and atypical (Singh, et al. 1996). This study looked at two antagonists of the 5HT2A receptor, clozapine and H Figure 1-3 Chemical structure of the typical drug loxapine and atypical drug clozapine loxapine (Figure 1-3), which are two of the many drugs that are commonly used to treat schizophrenia. The atypical antagonist clozapine has been the main focus of many variable response studies and therefore is included in this study (Arranz, 1995; Arranz, 1998; Arranz 2000; Malhotra, 1996; Nothen 1995). The typical neuroleptic loxapine is close in molecular structure to clozapine, but interestingly, this drug has a different range of receptor targets and drug activities. Typical neuroleptics, such as loxapine, affect only the positive symptoms of schizophrenia. This class of neuroleptics also produces strong extra pyramidal side effects such as parkinsonism and tardive dyskinesia (King, 1998). Clozapine is classified as an atypical neuroleptic agent. The atypical drugs treat both the positive and negative symptoms of schizophrenia and because of the ability of atypical 17 antipsychotics to target a narrower range of different receptor types, they generally produce fewer adverse side effects (Remington and Kapur, 2000). Disappointingly, the effectiveness of these drugs to treat schizophrenia varies considerably from patient to patient, and leads to a rather protracted 'trial and error' period searching for a drug that ameliorates some of the clinical manifestations of the disease while still producing tolerable side effects. Hence many studies have attempted to ascertain a genetic cause for these variable patient responses in order to reduce unnecessary delays in the appropriate patient treatment (Arranz, 1995; Arranz, 1998; Arranz 2000; Malhotra, 1996; Nothen 1995). Because the neuroleptics used to treat schizophrenia target the 5HT2A receptor, (among others), the polymorphisms that are found in this receptor have been investigated as a genetic cause of variable patient response. The effect of the H452Y variation in the 5HT2A receptor on intracellular calcium mobilization was investigated in the blood platelets of S A D (Seasonal Affective Disorder) patients homozygous H / H and heterozygous H / Y for the H452Y allele (Ozaki, et al. 1997). The heterozygotes demonstrated smaller peak amplitude in calcium mobilization after stimulation with 5HT, as well as longer peak latency and longer half time compared with the H/H homozygotes. Neither the Y / Y homozygote nor the less frequent A447V 5HT2A allelic variant were investigated due to the lack of available samples. The possibility of the H452Y allelic variant being modestly related to clozapine therapy response has been investigated (Masellis, et al.1998, Arranz, et al. 1996) although not all studies support the results (Malhotra, et al. 1996, Nothen, et al. 1995). Clinical investigations have found no association between the allelic variants and the lack of response to clozapine for the T25N polymorphism (Nothen, et al. 1995). To date no studies have looked at the effects 18 of either the A447V or the 1197V variants on neuroleptic activity. The clinical studies are often done on small samples sizes and the results are subject to variations in the genetic background and life style choices (environment) of the different individuals comprising the study group (Arranz, et al. 1998). Hence, there is a clear need for a detailed, controlled examination of the action of different antagonists on the 5HT2A receptor and its alleles. 1.9 GPCR functional assay In an attempt determine the effects of the four missense mutations (T25N, 1197V, A447V, and H452Y) of the 5HT2A receptor on the activity of the neuroleptic drugs clozapine and loxapine, an insect cell based GPCR functional assay system was employed. Using the insect expression vectors from Insect Select™ (Invitrogen) an initial portion of the 5HT2A signal transduction pathway is expressed in the Spodoptera frugiperda derived cell line Sf9. The assay requires Sf9 cells to be transformed with three heterologous genes encoding the following proteins: 1) human G-protein coupled receptor (GPCR) 2) human G a l 6 protein 3) Ca 2 + sensitive bioluminescent protein aequorin. Each of these proteins are expressed simultaneously in Sf9 cells from separate expression vectors. The G a l 6 protein is a member of the G ^ family and links the 5HT2A receptor to the phospholipase C P system. It has been demonstrated that the G a l 6 protein is able to interact with the insect G P y subunits to form a functional G protein (Grigliatti, et al. 2000). This functional G protein is able to interact with the endogenous insect phospholipase CP 19 and, therefore, able to link the human 5HT2A GPCR to the insect's transduction pathway. The G a l 6 protein was used for this assay due to its ability to link many human GPCRs to the phospholipase CP pathway (Offermanns, 1995). The promiscuity of the G a l 6 protein has been further confirmed in our lab by using this assay to measure the activation of the dopamine type 1 receptor, which normally regulates the G s type G protein, when challenged with the agonist dopamine (Grigliatti, 2000). The components and the signal transduction pathway activated by the GPCR functional assay are depicted in Figure 1-4. One of the ultimate consequences of activating the phospholipase C P pathway is the production of a calcium flux. This calcium flux can be detected with the bioluminescent reporter protein aequorin. Prior to challenging the cells with an agonist the cells must be 'charged' with the aequorin co-factor coelenterazine. Upon activation of a GPCR with an agonist, G a l 6 becomes phosphorylated and activates phospholipase C P . Activated phospholipase CP cleaves phosphoinositol diphosphate (PIP2) to release diacylglycerol and inositol triphosphate. This transduction pathway in turn releases the second messenger Ca 2 + from the endoplasmic reticulum (Kiselyov and Muallem, 1999). The coelenterazine cofactor and the C a 2 + sensitive photoprotein aequorin form a complex and the presence of C a 2 + mediates the oxidation of bound coelenterazine leading to a flash of light (Jones, et al. 1999). The intensity and duration of this transient flash of light is measured with a luminometer (Kendall and Badminton, 1998). The luminescence produced by this system has the great advantage of low background since cells do not normally produce flashes of light making this assay very sensitive (Jones, et al. 1999). 20 Figure 1-4 GPCR functional assay system. Adapted from patent submission (Grigliatti, et al. 2000.) PLC0 phospholipase C beta, PIP2 phosphoinositol diphosphate, D A G diacyl glycerol, IP3 inositol triphosphate (Grigliatti, 2000). The insect based functional human GPCR assay was used over other similar systems for several reasons. Firstly insect cells are easier to maintain than mammalian cell lines because they do not require incubation at 37°C or the presence of C 0 2 , a benefit since the assay is run at room temperature (Pfeifer, et al. 1998). Secondly, their media for growth is both serum and protein free which reduces the chances of receptor desensitization and eliminates other protein interaction with the ligands used to challenge the target GPCR (Grigliatti, 2000). Thirdly, and perhaps most importantly, insect cells appear to have 21 fewer endogenous GPCRs; therefore, less background noise and fewer possibilities of producing false positives (Grigliatti, 2000). Finally, insect cells appear to have similar expression machinery compared to mammalian cells that enables insect cells to express functionally active human GPCRs and human G proteins required for this assay. The fact that this assay system detects receptor function rather than simply ligand binding is also very important. Changes in ligand binding affinities do not always reflect receptor activity. For instance, when 4-iodo-2,5 dimethylphenylisopropylamine (DOI), an agonist of the 5HT2A receptor with hallucinogenic properties, was used to challenge the mutation F243A there was a 40 fold increase in binding affinity (Shapiro, et al. 2000). However, when function was measured, there was a 216 fold decrease in receptor activity. Interestingly, with the same mutation there was no change in binding affinity for serotonin, however, there was a 200 fold decrease in receptor efficacy (Shapiro, et al. 2000). Hence there is often little correlation between binding assays and receptor function. The key question is whether a particular polymorphism in the receptor structure is associated with receptor function. Studies that look at only binding affinity may not give the correct predication therefore, it is imperative to have an assay that measures the functionality of a receptor. Although there are many different receptors expressed in the human brain, this study focused on the G-protein coupled (GPCR) serotonin type 2A receptor (5HT2A). The 5HT2A receptor is thought to play a significant role in the treatment of schizophrenia (Ohara, et al. 1999). Indeed, it is one of the targets of many of the drugs used to treat this 22 disease. Variants in this receptor and its signal transduction system may be related to the high number of controversial findings and the varied response of patients to psychiatric drugs used in the treatment of this disease. The goal of my thesis is to determine if four SNPs in the 5HT2A receptor have an effect on the potency of neuroleptic drugs. In an effort to clarify some of these problems, this study used an insect cell based GPCR functional assay to quantitate the relative activity of serotonin (5HT), the natural ligand. With this as a foundation, the action of two neuroleptics, namely clozapine and loxapine, on the wild type and four variant forms of the 5HT2A gene product was examined. 23 Chapter 2 Materials and Methods 2.1 Cloning wildtype 5HT2A human cDNA into an expression vector a) cDNA and expression vector The cDNA for the human 5HT2A receptor was cloned into an insect cell line specific expression vector. The human 5HT2A cDNA clone was received as a gift from Dr. Alan Salzman via Dr. Stuart C. Sealfon of the Mount Sinai Medical Centre in New York. The human 5HT2A cDNA was cloned into the insect expression vector, p2Zop2F (figure 2-1: commercially available from Invitrogen as pl2V5H6). The insect expression vector was chosen for three reasons. Firstly, the promoter driving protein expression is from an immediate early gene in the bacoluvirus Orgyia pseudotsugata and it has been found to be transcriptionally active in many different insect species (Hegedus, et al. 1998). Secondly, the selection agent Zeocin, is effective in both bacterial and insect tissue culture cells; therefore, a single vector functions as both a shuttle vector and an expression vector (Pfeifer, et al. 1997). Finally, the vector contains a resistance gene, which is small (374bp) thus allowing the vector to accommodate either a large gene or several smaller genes (Pfeifer, 1998). b) Cloning strategy The human 5HT2A cDNA was received in the mammalian expression vector pcDNA3.1. This plasmid was cleaved using the restriction enzyme Nco I. A fill-in reaction using Klenow and dNTPs was performed in order to blunt the 5' overhanging end. A subsequent Xba I digest released the 1.6 kb cDNA fragment from the pcDNA3.1 vector. 24 This cDNA fragment was cloned into the Hind III (blunted with Klenow and dNTPs) and Xba I sites of the expression vector p2Zop2F. 500 1000 1500 2000 2500 kp p2Zop2F OpIE-2 promoter OpIE-2 Zeocin promoter Resistance Gene Figure 2-1 The Insect Select expression vector p2Zop2F (Hegedus, et al. 1998) 2.2 Engineering of naturally occurring 5HT2A variants The four different single nucleotide polymorphisms (SNPs) were engineered by in-vitro site directed mutagenesis. These SNPs comprise four of the five naturally occurring missense alleles found in the human population. The c D N A clone encoding the human 5-HT2A receptor was subcloned from the pCDNA3.1 plasmid into pBluescript II KS/+(Stratagene) at a Bam HI site. In vitro site-directed mutagenesis was used to introduce each polymorphism (T25N, I197V, A447V, 25 H452Y) into p5HT2A based on the procedure described in the Transformer Site-Directed Mutagenesis Kit from Clontech. The mutagenic primers utilized were: T25N 5' - C A T T A C T G T A G A G C C T G T T G T C A T C A T T T - 3 ' I197V 5 ' - G T C C A A A C A G C A A C G A T T T T C A G A A A T G -3' A447V 5' - G C T T T C C T A G A A C A A C C A T T G A G C A GTCATTATC-3 ' H452Y 5 ' -CCTCTTCAGAATACTGCTTTCCTA-3 ' and the selection primer was, 5 ' -CTTTTGCTCA G_AT£TTCTTTCC-3', which eliminated the unique restriction enzyme Afl III site. One mutagenic primer and the selection primer were simultaneously annealed to the same strand of denatured p5HT2A plasmid D N A . Elongation and ligation was achieved with T4 D N A polymerase and ligase resulted in the incorporation of both mutations into the same newly synthesized strand. The D N A was digested by Afl HI, which cuts the parental plasmids and transformed into E. coli mismatch repair deficient strain to propagate the mutant plasmid. The pool of plasmids was subjected to a second round of digestion by Afl III and the resulting plasmids were transformed into competent D H 5 a E. coli cells. The transformants were isolated and sequenced to confirm the existence of the desired mutations. The newly created variants were subcloned into the insect expression vectors. The A447V variant was subloned into p2Zop2F using the same cloning procedure as for the wildtype 5HT2A c D N A (See Section 2.1b). The remaining three variants were subcloned into the insect expression vector p2Zop2SK. This expression vector is 26 identical to p2Zop2F with the exception of the multiple cloning region. Each variant was separately cloned into p2Zop2SK via the Xba I and Cla I restriction enzyme sites. 2.3 DNA preparations The large quantity of plasmid D N A required for cellular transformation was purified using a CsCl (Merck) gradient. Each construct was separately transformed into competent DH5cc E. coli cells by heat shocking the cells at 42.5°C for 90 seconds. The cells were grown for 1 hr at 37°C in L B (5 g yeast extract, 5 g.NaCl 10 g bacto-tryptone made up to 1 liter pH 7.0) media, to allow for the expression of the Zeocin resistance gene located on the plasmid, and then plated onto L B agar plates containing Zeocin (Invitrogen) at a concentration of 25 u.g / ml. The plates were grown at 37°C overnight and a single colony for each construct was used to inoculate a separate 200 ml aliquot of L B media with Zeocin (Invitrogen; 25 ug / ml) contained in a sterile 2 litre Erlenmeyer flask. The flasks were placed on a orbital shaker (250 rpm) and incubated at 37°C for 16 hours. After the incubation the cells were harvested by centrifugation at 4000 rpm for fifteen minutes at 4°C. The resulting pellet was resuspended in 4 ml of ice cold Solution 1 (50 m M Glucose, 25 m M Tris pH8.0, 10 m M E D T A pH 8.0). Once the cells were fully resuspended, a 400 [il aliquot of freshly made lysozyme (Roche) was added (10 mg/ml in 10 mM Tris pH 8.0). To this mixture a 8 ml aliquot of Solution 2 was added (0.2 N NaOH 1% SDS). The tube was gently inverted three times and incubated at room 27 temperature for ten minutes. A 6 ml ice cold aliquot of Solution 3 (60 ml of 5M potassium acetate, 11.5 ml glacial acetic acid, 28.5 ml distilled H 2 0) was added and the tube was inverted three times then placed on ice for ten minutes. After the incubation, the solution was centrifuged at 4000 rpm for fifteen minutes at 4°C. The resulting supernatant was removed and placed into a sterile centrifuge tube. Isopropanol (0.6 volumes) was added, the tube was mixed by inversion and incubated at room temperature for ten minutes. The tube was then spun at 12 000 rpm for 30 minutes at 15°C. The supernatant was removed and discarded and the clear pellet was washed with 70% ethanol. The pellet was air dried and dissolved in 3ml of TE pH 8.0 (lOmM Tris pH 8.0,1 m M E D T A ) . For each 1 ml volume of D N A solution generated above, exactly 1.00 g of CsCl were added and the tube was mixed gently until the salt dissolved. The resulting solution was pipetted into a Beckman Quick Seal centrifuge tube. A 600 | i l aliquot of ethidium bromide (10 mg/ml in distilled H 2 0) was added to the tube, which was then inverted. The remaining tube volume was filled with light mineral oil (Fisher). The tubes were heat sealed, balanced and placed into a Ti70 rotor (Beckman). The samples were spun at 45 000 rpm for 48 hours at 20°C. Closed circular plasmid D N A was visualized by long wave U V light and removed from the resulting gradient using a syringe and a large bore needle. The ethidium bromide was removed from the sample by washing the sample five times with butanol. The resulting mixture was diluted with three volumes of sterile distilled H 2 0 and the plasmid D N A precipitated from the solution with two volumes of 95% ethanol. The sample was spun at 10 000 rpm for 30 minutes at 4°C. The pellet was 28 washed with 70% ethanol, air dried and dissolved in 1 ml of T E pH 8.0. The concentration of the sample was determined using a spectophotometer. 2.4 Sequencing of all constructs A l l the constructs expressing the wildtype 5HT2A receptor, or any of the four missense variants were sequenced to ensure that their sequence's were still intact. A l l sequencing reactions used one of the following two primers: Forward primer denoted ie2-S 5' A AT A C A G C C C G C A A C G A T CTG G 3' Reverse primer denoted ie2 3' 5' C G G G T G C G C A C G C G C TTG A A A G G A The forward primer annealed to the OpJJE 2 promoter just upstream of the transcriptional start site, whereas the reverse primer annealed to the poly A signal located downstream from the inserted cDNA. Since there are two OpEE-2 promoters in the insect expression vectors (See Figure 2-1) the vector was digested with Apa I prior to sequencing with the forward primer. This digestion cleaved the plasmid into two fragments. The two fragments were separated on 0.8 % Agarose gels and the appropriate band was cut from the gel and recovered. When sequencing with the forward primer, between 30 ng - 90 ng of purified fragment D N A was aliquoted into a 500 \il eppendorf tube along with 4 J L L I of the terminator premix 29 including, polymerase, dye labeled primers and terminators, and buffer (Nucleic Acid -Protein Service Unit, (NAPS) UBC). To this mixture a 0.5 | i l aliquot of the appropriate primer at a concentration of 10 u M was added and the volume was topped up to 20 u,l with sterile deionized H 2 0 . The same protocol was used for the reverse primer except that 250 ng - 500 ng of the plasmid D N A was added to the reaction. The sequencing reactions were placed in a PCR machine and the following protocol was utilized: Rapid thermal ramp to 96°C 96°C for 30 seconds Rapid thermal ramp to 50°C 50°C for 15 seconds Rapid thermal ramp to 60°C 60°C for 4 minutes 25 cycles total Rapid thermal ramp to 4°C and hold To precipitate the D N A a 2 ui aliquot of 3M sodium acetate pH 4.6 and a 50 u.1 aliquot of 95% ethanol were added to the sequencing reaction. The D N A was pelleted with centrifugation, resuspended and sequenced using a 373 automated D N A sequencer (NAPS unit UBC). 2.5 Maintenance of insect cell line The lepidopteran cell line Sf9 was chosen for use in this study. Previous studies, which looked at expression and function of the dopamine receptor in various insect cell lines, indicated that the Sf9 cell line performed well (Grigliatti, 2000). The apparent strong performance of Sf9 insect cells with the functional expression of the dopamine receptor, made for a good starting point to investigate the expression of the 5HT2A receptor. Indeed my studies on serotonin receptors, and subsequent studies with other receptors in 30 our lab, indicate that the engineered Sf9 cell system provides a high signal to low background noise ratio. A l l the Sf9 insect cell lines were maintained at 26° C in T-25 tissue culture flasks (VWR) and grown in the serum free media ESF 921 (Expression Systems, Woodlands, CA). Henceforth, it will be referred to as ESF. Once the cells had grown to form a confluent layer on the bottom of the flask they were seeded into a new flask. To the new flask a 4 ml of ESF media was aliquoted. The confluent layer of cells was flushed off the bottom of the original flask using a 5 ml sterile pipette. A 1 ml aliquot of the resuspended cells was then transferred to the new T-25 flask. When large quantities of cells were required for experimental purposes the cell lines were scaled up into larger T-75 tissue culture flasks (VWR). A 3 ml aliquot of resuspended cells from a confluent T-25 flask was transferred into a T-75 flask containing 12 ml of ESF media. 2.6 Counting cells A confluent layer of cells was flushed with ESF growth medium using a sterile pipette. Once the cells were in suspension, a 100 u,l aliquot was removed and placed into a 1.5 ml eppendorf tube. From that 100 u.1 aliquot, a 10 u,l sample was removed and added to a 10 ul aliquot of 0.4% Trypan Blue in I X phosphate buffered saline (137rnM NaCl, 2.7 m M KC1, lOmM Na 2 HP0 4 , 1.7mM K H 2 P 0 4 pH 7.4). This mixture was incubated at room temperature for three minutes. A 10 uJ aliquot of this mixture was placed into a Spotlite™ hemacytometer (American Scientific Products) and counted through an 31 observation microscope. The number of cells that appeared in the central grid of the hemacytometer, which was broken up into twenty-five smaller squares, is multiplied by a factor of two to account for the dilution factor due to the addition of 10 |xl of trypan blue. This resulting number was multiplied by 104 and represented the number of cells found per ml of media. 2.7 Transient expression of the human 5HT2A receptor in Sf-9 cells While all of the final experiments were done in permanently transformed Sf9 cells, the initial experiments with the 5HT2A receptor involved transient expression of the receptor. Many have shown that a greater amount of protein product is obtained from transient expression than from a stable cell line transformed with the identical plasmid construct. This is expected since the number of vector copies within a newly transfected cell-line (transient expression) is much greater than the number of inserts (integrated vectors) in stable cell lines. Thus transient expression provides the maximum opportunity to detect a protein product. A l l transformations took place in six well tissue culture plates (VWR). A 10 ul aliquot of CellFectin (Gibco B R L ) , a liposomal transfection agent, was added to a sterile 1.7 ml eppendorf tube containing 1ml of Grace's media (Gibco BRL) . Grace's media was made with deionized H 2 0 , containing 4.5 m M NaHC0 3 and was adjusted to a pH of 6.2 with 5 M K O H . To this mixture a 1 ng aliquot of plasmid D N A was added. The tube was vortexed for five seconds to ensure thorough mixing and allowed to stand for 30 minutes. Sf-9 cells were resuspended and counted following the protocol outline in section 2.4 and 32 2.5. These cells were diluted in Grace's media to 0.75 X 106 cells / ml and 2.0 ml was placed into each well of a six well tissue culture plate. The cells were allowed to adhere to the plate for 30 minutes. After the 30 minute incubation the media covering the cells was removed and discarded. The 1 ml aliquot of Grace's containing the D N A CellFectin mixture was added to each well of the plate and the plate was incubated at room temperature for four hours. After this incubation a 1 ml aliquot of ESF growth media was added to each well. The plate was incubated at 26°C for 48 hours. After the 48 hour incubation the cells could either be used for immediate experimental procedures (transient GCPR assays, or protein extraction, see below) or be placed on selection in order to create a stable transformed insect cell line. 2.8 Selection of stable lines In order to create stable insect cell lines the transiently transformed cells, after the 48 hours incubation outlined above, were subjected to 500 ng of Zeocin / ml of ESF. Initially, the transient transformations were in six well tissue culture plates. When the wells of the six well plate grew to form a confluent layer fixed to the bottom the cells were transferred to a T-25 flask. Once the transformed cells were placed in a T-25 flask they were maintained as described in Section 2.5. After one week at a Zeocin concentration of 500 u,g / ml the cells were moved to a Zeocin concentration of 750 \ig I ml of ESF. After the second week the final concentration of Zeocin was increased to 1 mg / ml of ESF. The resulting polyclonal stable cell lines were passaged a minimum of 33 three times at 1 mg Zeocin / ml ESF, prior to their use in agonist or antagonist dose response studies. 2.9 Protein preparation and quantification for Western Blot Analysis A transient transformation using the wildtype 5HT2A receptor, along with a negative control transformation, following the same transformation protocol except for the exclusion of D N A , were performed simultaneously. Both sets of transformations, test and control, were done in triplicate and the cells were harvested 48 hours later, by resuspending the cells in the growth medium using a sterile 1 ml pipette. The harvested cells were removed from the growth medium by centrifugation, and treated with the Pierce mem-Per® eukaryotic membrane protein extraction reagent kit (Biolynx Inc.), which separates the hydrophobic proteins from the hydrophilic proteins. The supernatant was saved in order for it to be run out on a gel as a control. Following the protocol of the kit, the cells were lysed and then the proteins were extracted from the cell membrane. Once the membrane proteins were in solution, the mixture was spun at 10 000 X g for three minutes at 4°C to remove the remaining cell membrane fragments. Since the human 5HT2A receptor is a transmembrane protein, the cellular membrane debris generated in this step was collected and resuspended in 5X SDS loading buffer (62.5 mM Tris pH 6.8, 2% SDS, 10% Glycerol, 715 m M P-mercpatoethanol, 0.001% Bromophenol blue). The supernatant previously removed from the cellular debris was further treated to separate hydrophobic and hydrophilic proteins. As a result three fractions were 34 recovered for each sample: cell membrane fragments, hydrophobic proteins and hydrophilic proteins. In an attempt to equalize loading onto the polyacryamide gel the total protein was measured for each sample. A 3 ul aliquot of the sample was pipetted into a 1 ml cuvette. To that sample a 500 | i l aliquot of Protein Assay dye (BioRad), diluted 1:4 in distilled H 20, was added. The absorbance of the resulting solution was read in a Beckman DU-64 spectophotometer at a wavelength of 595nm. The total protein concentration of the sample was ascertained using a standardized curve generated with known concentrations of BSA (New England Biolabs) 2.10 Polyacrylamide gel separation and Western blot analysis The appropriate volume of protein sample (based on the above protein quantifications) was aliquoted into 500 iul Eppendorf tubes and brought up to a total volume of 20 \i\ with sterile distilled H 2 0 . To these samples 5 ul of 5X SDS loading buffer was added (with the exception of the cell membrane fragments). The tubes were incubated at 95°C for four minutes and then spun at 14 000 rpm for three seconds to settle all the liquid to the bottom of the tube. The entire volume from each sample was loaded onto a 10% polyacrylamide gel (10 % polyacrylamide gel: 7.25 ml dH 2 0, 3.75 ml 1.5 M Tris pH 8.8, 150 jxl 10% SDS, 3.75 ml 40 % bis/acrylamide (BioRad), 75 ul 10% ammonium persulphate (Fisher), 7.5 ul T E M E D , stacking gel: 3.05 ml distilled H 2 0 , 1.25 ml 0.5 M Tris pH 6.8, 50 u.1 10% SDS, 0.65 ml 40% bis/acrylamide, 25 fil ammonium persulphate, 5 |xl TEMED) . A 10 (ill aliquot of the Benchmark™ protein ladder (Invitrogen) was 35 loaded alongside the samples. 60 volts were applied for three hours using I X Tris Glycine pH 8.3. After separation the proteins were transferred overnight onto Hybond C Extra membrane (Amersham Pharmacia Biotech) at 22 volts in Towbins buffer (25 m M Tris, 192mM Glycine, 20% Methanol). After the transfer, the membrane was allowed to air dry. The membrane was placed onto an orbital shaker (60 rpm) and a blocking agent containing 5% skim milk dissolved in TBST (10 m M Tris, 150 m M NaCl, 0.1% Tween 20) was added to it. After a one hour incubation the blocking agent was discarded and the membrane was washed once with TBST. The monoclonal antibody for the 5HT2A receptor (Research Diagnostics, Inc. #RDI-5HT2Aabm) was made up in a 10 ml volume of TBST at a concentration of 2ug of mAb / ml of TBST. The membrane was incubated with the primary antibody for one hour and then the membrane was washed three times with TBST. Each wash cycle lasted for a five minute period. The secondary antibody, goat antimouse IgG (H+L) HRP conjugate (BioRad) was diluted 1:10 000 in a 10 ml volume of TBST. The membrane was incubated with the secondary antibody for 1 hour then washed three times with TBST. The E C L system (Amersham Pharmacia Biotech) was utilized to detect the secondary antibody signal. A 500 u.1 aliquot of detection reagent 1 was added to a 500 | i l aliquot of detection reagent 2. The resulting solution was overlayed on the membrane and allowed to stand for one minute. The excess E C L detection reagent was drained from the 36 membrane. The membrane was wrapped and placed in a film cartridge with a piece of autoradiography film (LabScientific, Inc) for 30 minutes. The film was then developed. 2.11 GPCR functional assay of transiently transformed cells Cells transiently expressing the appropriate proteins (5HT2A receptor, G a l 6 , and/or aequorin) were harvested at the end of the 48 hour incubation in growth media (see Section 2.7). A 1 ml aliquot of these cells was placed into a 1.7 ml Eppendorf tube. To this tube a 5 \il aliquot of a ImM coelenterazine solution was added and the cells were placed on a Labquake tube rotating shaker (VWR) to prevent the cells from settling out. A lightproof container was placed over the shaker and the cells were incubated at room temperature for two hours. After the two hour incubation, 100 |xl of cell suspension was aliquoted into three separate wells of a 96 well plate. The 96 well plate was inserted into the reading chamber of the luminometer. The functional GPCR assay utilized a luminometer (Fluorskan Ascent F L , Labsystems) to measure light output as a result of receptor activation. Transiently transformed cells were tested to demonstrate that 5HT2A receptors were functional in Sf-9 cells by challenging the cells with the agonist serotonin (Sigma). These experiments were done with small volumes of cells. The cells were manually injected into a 96 well plate. The luminometer was equipped with three separate injectors that allowed for the addition of agonist and TX-100 directly into the 96 well plate. A 50 [d injection of 3mM serotonin (Sigma) made up in ESF growth media (final concentration of ImM) was added to each of the three wells that contained the 37 aliquoted cells. Control experiments that injected 50 u,l of ESF growth media without serotonin were also performed to determine background response. After the initial injection of serotonin in ESF the plate then moved to allow the luminometer measuring head to scan and monitor the three wells for a total of 90 seconds. The luminescent output was plotted against time (Curve A as seen in Figure 2-2 below). After the agonist induced light output was measured, a 50 ul aliquot of 0.05% TX-100 (final concentration 0.0125%) in I X phosphate buffered saline was injected into the same three wells. The TX-100 solution lysed the cells and exposed the totality of the aequorin present. The light output was monitored for 60 seconds in the same manner as described above and produced a second curve (Curve L). The fractional luminescence was calculated by integrating the area under both curve A and L separately, and dividing integral A by the total of integral A plus integral L (Figure 2-2). Representing the response of the receptor to its ligand as fractional luminescence controls for any variation in cell number between samples (the response is calculated as a proportion of total Ca 2 + available). The error bars on the graphs presenting the fractional luminescence for GPCR functional assays with transiently transformed cells are 95% confidence intervals. The confidence intervals were calculated using the student t-test. 38 Figure 2-2 Graph demonstrating the calculation of fractional luminescence (A/(A+L)) Adapted from patent submission (Grigliatti, et al. 2000) 2.12 Agonist Dose Response Experiments Transiently expressing cell lines were used only for initial experiments. For all the remaining experiments, permanently transformed cells (See Section 2.8 for protocol), were used with the functional GPCR assay and all experiments followed the same protocol indicated below. Any deviations from the following standard protocol have been identified. For dose response curves, the agonist was arrayed in triplicate into a 96 well plate in varying concentrations. Serial dilutions of serotonin were made from a 30 m M stock solution of serotonin. An aliquot of cells (between 5 ml and 7 ml), containing 2.24 x 106 39 cells/ ml of media, was harvested and charged with coelenterazine (5 \il of ImM solution / ml of media) for two hours (see Section 2.11). After the incubation, the cells were spun at 2000 rpm for four minutes and the supernatant was discarded. A volume of ESF equal to that removed was used to resuspend the cells and a 100 \il aliquot of cells was removed for cell counts. The cells were placed into a low speed stir vessel that kept the cells in suspension. The stir vessel was internalized in the luminometer chamber in order to connect it to the luminometer's third injector and to allow the cells to be kept in a dark environment. A 100 [il aliquot of cells was injected into three wells of the 96 well plate, each containing the same concentration of agonist. Directly after the cells were injected into the three wells the luminometer cycled through the three wells and measured light output (as previously described). A 50 | i l aliquot of 0.05% TX-100 (final concentration 0.012%) was injected into those three same wells and light output was monitored again. The luminometer then cycled through the next three wells containing a different concentration of agonist and performed the same series of injections and measurements. This cycling continued until all the agonist concentrations had been tested. Fractional luminescence (A/(A+L)) for each concentration was calculated as described in section 2.11. 2.13 Antagonist Dose Response Experiments Stable insect cell lines expressing the wildtype 5HT2A receptor or one of its four missense variants, along with Ga}6 and aequorin were harvested and charged with coelenterazine. The cells were incubated as described in Section 2.11 for two hours. The 40 antagonist loxapine (Sigma) was dissolved in ESF and clozapine (Sigma) was dissolved in DMSO. New stock solutions of both antagonists were made up to a concentration of 10 mM. Serial dilutions of each antagonist were made in ESF. A 10 ul aliquot of antagonist was arrayed in the same triplicate fashion as described in Section 2.12 into a 96 well plate. After the two hour incubation the cells were spun at 2 000 rpm for four minutes and resuspended in the same volume of ESF media. The cells were loaded into the low speed stir vessel and the arrayed plate was loaded into the luminometer. A 90 ul aliquot of the charged assay cell line was injected from the stir vessel into three wells of the antagonist arrayed 96 well plate. The light output was monitored for a 30 second period to ensure that the antagonist did not produce any agonist effect. Then, a 50 ul aliquot of serotonin at lOOuM (final concentration) was injected into the wells containing cells, and the light output was measured for 90 seconds. Finally a 50 ul aliquot of T X -100 was injected and the light output was measured for 45 seconds. The luminometer cycled the plate through to the next set of three wells containing a different concentration of antagonist. The same series of injections and measurements as outlined above were performed on the next three wells. The luminometer continued to cycle through sets of three wells until all the antagonist concentrations had been tested. The fractional luminescence was calculated for each well measured. 2.14 Competition Studies Competition study experiments were performed to ascertain whether a specific antagonist competed for the same binding site on the human 5HT2A receptor as serotonin (the 41 natural ligand). These dose response studies were set up in the same manner as the agonist dose response experiments. However, when the cells were poured into the stir vessel an aliquot of antagonist was added to the cells. The cells were incubated with the antagonist for five minutes prior to performing the agonist dose response experiments. The fractional luminescence was calculated for each agonist at a constant antagonist concentration (See section 2.11) 2.15 Standardization of fractional luminescence In an effort to compare the values from different permanently transformed cell lines the fractional luminescence values were standardized. Prior to standardization, background fractional luminescence (wells containing ESF without agonist) was subtracted from the fractional luminescence obtained from test doses of agonist. (Figure 2-3). A 1 m M concentration of serotonin, which is far greater then physiological concentrations, always produced a maximum response. This fractional luminescence measurement was assigned a value of 1. A l l other values were normalized to the maximal response and became a fraction of the total response. For each dose response experiment three replicas for maximal response were measured; they occurred in order (first-second-third well of a 96 well plate). The first value achieved for the maximal response (A4 from Figure 2-3) was used as the denominator to calculate the fractional response of lower, more physiologically relevant concentrations. For example, to calculate the fractional response for A7 the fractional luminescence value from A7 was the numerator and the fractional luminescence value from A4 was the denominator. This analysis was repeated 42 with all the first values for all the agonist concentrations i.e. A10/A4 and so on throughout the entire plate. The same procedure was used for the second (A5) and third (A6) maximal response values (Figure 2-3). 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 A 1 m M 1 m M 1 m M 1 0 0 u M 1 0 0 u M 1 0 0 u M 3 0 u M 3 0 u M 3 0 u M ESF ESF ESF 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T Figure 2-3 Typical agonist layout of the first row from a 96well plate. 2.16 Graph and Curve fitting The standardized fractional luminescence values obtained in the dose response experiments were plotted against the Log 1 0 of the corresponding concentration of drug tested. For each dose response experiment one mean value was calculated from the three replicas and was plotted with its associated standard error of the mean (SEM) value. The graphs were prepared in the statistical program Graph Pad Prism (GraphPad Software Inc. San Diego, CA). Once graphed this software package used the three parameter logistic equation for non-linear regression to fit the sigmoidal curve. Y=Bottom + ((Top -Bottom)/(l+10 ( L o g E C 5°- X ) ))) X is the logarithm of concentration and Y is the response. The three parameter logistic equation determined the maximal value (top) and the minimal value (bottom) from the 43 input data. To fit the curve, the Hi l l Slope for the curve was held constant at 1 and a LogECjo (effective concentration) value was derived from this fit curve. The LogEC 5 0 value is the log concentration of agonist required to produce a response that is half of the total response, and this value provides an indication of potency of the agonist for the receptor. In the case of the antagonist dose response curves the Hillslope was held constant at -1 and the LogIC 5 0 value was reported. The inhibitory concentration (IC) represents the dose of antagonist required to reduce the response to that which is half of the maximal response. 2.17 Coelenterazine leaching experiments A 5ml aliquot of cells from a polyclonal cell line expressing the 5HT2A wildtype receptor, G a l 6 and aequorin was harvested and charged with coelenterazine (5 ju.1 of a ImM solution/ ml of media) in a 15 ml Falcon tube. The cells were incubated in the same fashion as described in Section 2.11. After the two hour incubation period the cells were centrifuged and resuspended in 5 ml of fresh ESF growth media. The cells were poured into the stir vessel to prevent cells from settling out. At 15 minute intervals a 100 jul aliquot of cells was injected into three wells of a 96 well plate and tested for response to 100 uM of serotonin (final concentration), as outlined in Section 2.12. Following the measurement of agonist activation a 50 j i l aliquot of TX-100 (final concentration 0.0125%) was injected into the same three wells thereby exposing the totality of the coelenterazine. A second 5 ml aliquot of the same polyclonal cell line was tested in the identical manner with the exception that the cells remained in the media containing 44 coelenterazine for the entire experiment. The fractional luminescence A/(A+L) and the total luminescence (A+L) was calculated for both experiments. 2.18 Effects of solvents on stable cell lines A 7 ml aliquot of permanently transformed Sf9 cells expressing the human wildtype 5HT2A receptor along with G a l 6 and aequorin were harvested and charged for two hours with coelenterazine (5 ul of a l m M solution/ ml of media). The cells were centrifuged at 2000 rpm for two minutes, resuspended in fresh ESF media and poured into the low speed stir vessel connected to the third injector of the luminometer. A 10 \il aliquot of solvent was arrayed into a 96 well plate in the same triplicate fashion as used with the dose response experiments. The two solvents tested were dimethylsulfoxide (DMSO) and ethanol and both were diluted into ESF. Once all the solvent concentrations were arrayed the plate was inserted into the luminometer. A 90ul aliquot of cells was injected into three wells containing the same concentration of solvent. The light output was measured from those three wells for 90 seconds. A 50ul aliquot of 0.05% TX-100 (final concentration 0.0125%) was injected into the same three wells and light output was measured for an additional 60 seconds. Control experiments that included response to ImM serotonin and ESF only were also included. The fractional luminescence was calculated for each category. 45 2.19 Effects of extended incubation of stable cell lines with antagonists A stable cell line expressing the wildtype 5HT2A receptor, G a l 6 , and aequorin proteins was harvested and divided into two 7 ml aliquots each with 2.2 X 106 cells / ml. These cells were incubated with coelenterazine as described in Section 2.11. A 50 uJ aliquot of 3 mM serotonin was added to 36 wells of a 96 well plate. One batch of cells was spun down after the two hour incubation and resuspended in fresh ESF see Section 2.11. Clozapine was added to these cells to achieve a final concentration of 73 nM. These cells were then placed into the stir vessel. At three minute intervals 100 ul of cells were injected into three wells of the 96 well plate. Light output was measured as described in Section 2.11. A 50 J L I I aliquot of 0.05% TX-100 (final concentration 0.0125%) was added and light output again was measured. The other batch of cells was placed into fresh media after its two hour incubation with coelenterazine (Section 2.12) and no antagonist was added. The same series of injections and measurements were made as described above. The same experimental approach was used to test loxapine; however, the final concentration of loxapine was 10 nM. Fractional luminescence was calculated for each set of experiments as outlined in Section 2.11. 46 Chapter 3 Results 3.1 Transient expression of the human 5HT2A receptor in Sf-9 cells Prior to initiating dose response experiments to determine drug potencies, it had to be ascertained whether Sf9 cells could produce the human 5HT2A receptor protein. Sf9 cells were transformed, in triplicate, with the human 5HT2A gene or no D N A at all (negative control). The cells were harvested after 48 hours and these cells were fractioned according to the Biolynx membrane protein protocol. Western blot analysis was used to monitor production of the human 5HT2A receptor (See Section 2.7 for protocol). Three protein containing fractions were recovered for each sample: cell membrane fragments, hydrophobic proteins and hydrophilic proteins. The total protein concentration for each preparation was determined in an attempt to equalize protein loading onto a gel. The proteins from each sample were separated by electrophoresis on a 10% Tris-Glycine SDS polyacrylamide gel. To ensure that the 5HT2A receptor was not secreted from the cell, an aliquot of the spent growth medium (see Section 2.9) was also included on the gel. The separated proteins were transferred to a membrane and probed with the 5HT2A receptor monoclonal antibody developed by Wu, et al. (1998). Using this antibody a band of the predicted size (55kDa) was detected in the cellular membrane fraction for each of the three test samples (Figure 3-1 Lanes 4-6). None of the hydrophobic and hydrophilic protein fractions for the transformed Sf-9 cell line contained the human 5HT2A receptor protein as indicated by the lack of bands on the western blot (Figure 3-1 Lanes 8-13). Similarly, no bands (i.e. e. no 5HT2A receptor protein) were detected in the spent medium sample (Figure 3-1 Lane 7). Finally, no bands were 47 detected in any of the three protein fractions obtained from the untransformed Sf-9 cells (Figure 3-1 Lanes 1-3). Hence, these data indicate that the human 5HT2A receptor is produced in Sf9 cells transformed with the human gene and the 5HT2A receptor is found in the membrane fraction. Figure 3-1. Western blot analysis of Sf9 cells transiently expressing the human 5HT2A wildtype receptor probed with monoclonal antibody for 5HT2A. Legend Lane 1 Sf9 control cellular debris Lane 2 Sf9 control hydrophobic fraction Lane 3 Sf9 control hydrophilic fraction Lane 4 Sf9 transformation #1 cellular membrane fraction Lane 5 Sf9 transformation #2 cellular membrane fraction Lane 6 Sf9 transformation #3 cellular membrane fraction Lane 7 Sf9 transformation #1 spent media Lane 8 Sf9 transformation #1 hydrophobic fraction Lane 9 Sf9 transformation #2 hydrophobic fraction Lane 10 Sf9 transformation #3 hydrophobic fraction Lane 11 Sf9 transformation #1 hydrophilic fraction Lane 12 Sf9 transformation #2 hydrophilic fraction Lane 13 Sf9 transformation #3 hydrophilic fraction Lane 14 10 ul Benchmark Protein ladder (Invitrogen) 48 3.2 The human 5HT2A receptor functions in Sf-9 insect cells The human 5HT2A receptors were expressed in Sf-9 cells, and assayed for function using a GPCR functional assay system developed in our lab (Grigliatti, et al. 2000), the general components of which were described in Chapter 1. Transient transformations were employed to determine whether the human proteins function appropriately in insect cells. Three separate expression plasmids carrying the 5HT2A receptor, the human G a l 6 and aequorin, were transformed in various combinations into Sf9 cells. These plasmids were added in equal quantities (Table 3-1). One set of transformations included G a l 6 and aequorin only (Table 3-1, transformation B), and another set included the 5HT2A receptor and aequorin only (Table 3-1, transformation C). These experiments were included as controls to test for endogenous activity of 5HT2 receptors in insect cells and constitutive activity of the human receptor in this ex vivo system. In order to maintain the same amount of D N A in all experiments 330ng of a non-expression plasmid (pBluescript) was added to each of the control transformations. A minimum of three independent transformations were done for each transformation group (Table 3-1 transformations A-C). After a 48 hour period the cells were harvested, counted and tested for their ability to respond to 1 m M serotonin, the natural ligand of the 5HT2A receptor. 49 Table 3-1 Transient transformations performed in Sf-9 cells to test for 5HT2A function. The three different transformations were each set up in triplicate. Transformation Constructs Rationale A 330 ng 5HT2A 330 ng G a l 6 330 ng Aq Tests for 5HT2A Function B 330 ng G a l 6 330 ng Aq 330ng pBluescript Tests for presence of endogenous serotonin receptors C 330 ng 5HT2A 330 ng Aq 330 ng pBluescript Tests for presence of Gocq type Proteins To test for response to serotonin, the harvested cells were incubated with coelenterazine for a two hour period. Each of the three transformation groups (A-C) were tested independently using the protocol for assaying transient GPCR activity (See Section 2.11). Each transformation was tested for its ability to respond to I m M serotonin (final concentration). In order to determine background response each transformation was also challenged with ESF growth medium without serotonin. For each replicate of a transformation set, three fractional luminescent values were calculated. The mean of the three values was calculated for each replicate. The three replicate means were used to calculate the overall mean for the transformation set. The students' t-test was employed to determine the 95% confidence intervals. When 50 transformation A was challenged with 1 m M serotonin a large luminescent response was measured as indicated by the large fractional luminescent value calculated (Figure 3-2 bar #1). However when the 5HT2A receptor was left out of the transformation mixture and the cells were challenged with ImM serotonin no response was observed (Figure 3-2 bar #3). The transformation that left G a , 6 out of the transformation mixture demonstrated a response that was equivalent to that observed for transformation A ( G a l 6 included) when challenged with ImM serotonin (Figure 3-2 bar #4). Finally, when ESF growth media alone was used to challenge any of the transformation no response was observed. This is indicated by the lack of response to ESF by transformation A (Figure 3-2 Bar #2). Serotonin 1mM a> o c CD O (A CO _ .E - i (0 c o o 0.900 0.800 0.700 0.600 0.500 0.400 0.300 0.200 0.100 0.000 i l l lPlw m B Fractional luminescence jr J> * 4? Figure 3-2 Bar graph depicting fractional luminescence indicating that the human 5HT2A receptor functions in transiently transformed Sf-9 cells only when challenged with 1 m M serotonin. A '+' above the bar graph indicates the addition of ImM serotonin a '- ' indicates the addition of ESF. Error bars represent 95% C.I. from three independent experiments performed in triplicate. 51 Hence, it appeared that the insect G a protein was capable of coupling with the human 5HT2A receptor. This was an interesting finding; other human GPCRs tested with this system required the addition of the human G a l 6 protein to initiate the signal transduction cascade (Grigliatti, et al. 2000). However, the data shown here also indicate that expression of the human G a l 6 subunit of the trimeric G protein in no way lessens the serotonin response of the receptor. Therefore, all subsequent stable cell lines and experiments in this thesis included the human G a l 6 protein. 3.3 Creation of stable Sf-9 cell line expressing 5HT2A receptor, Ga 16 and aequorin Knowing that the functional GPCR assay was able to detect the human 5HT2A receptor activation in Sf9 cells, permanently transformed cell lines expressing either the wild type or the four variants of the 5HT2A receptor as well as the human G a l 6 and aequorin constructs were created. The stable cell lines provided a uniform platform for establishing dose response curves. To begin these studies separate stable cell lines expressing the receptor, human G a l 6 and aequorin were created for each of the five alleles of the human 5HT2A receptor. Transformations and selection were performed as described in Materials and Methods. 3.4 Determination of the effects of coelenterazine leaching from charged cells The procedure to establish dose response curves differs from the initial agonist studies and is shown in Figure 3-3. As described in section 2.12 the cells were charged with 52 coelenterazine and then the excess coelenterazine was removed prior to placing the charged cells into the stir chamber. Since the dose response experiment typically took up to 45 minutes, an experiment was designed to assess whether coelenterazine leached from the cells into the fresh media during the dose response experiment. A stable polyclonal cell line expressing the human 5HT2A wildtype receptor, G a i 6 and aequorin was charged with coelenterazine as described in Section 2.12. After the incubation the ESF media was removed and fresh ESF media (lacking coelenterazine) was added to the cells. These cells were challenged with 100 u M of serotonin (final concentration) at fifteen minute intervals over a period of three hours. Step 1 1) Array Agonist 2) Inject Cells 3) Monitor light I Output Step 2 1) Inject Tx-100 2) Monlor light Output Figure 3-3 Experimental procedures for agonist dose response curves 53 A typical dose response experiment as described in this thesis, ranges in time from 30 -40 minutes. To test for leaching of coelenterazine from the charged cells samples were taken at 15 minute intervals from time 0 minutes (point at which fresh media was added to the cells) through to 180 minutes. The experiment was repeated a minimum of three times. An identical group of cells was tested in the same manner with the exception that the cells remained in the media containing coelenterazine for the duration of the experiment. Both the fractional luminescence and total luminescence for each set of stable cell lines was analyzed. The series of bar graphs in figure 3-4 depicts the overall mean for each time point. The error bars that appear were calculated using the students' distribution to determine the 95% confidence interval. For both test groups the fractional luminescence remained relatively constant throughout the three hour time span (Figure 3-4a). The total luminescence values for the samples, in which the medium was exchanged for ESF media lacking coelenterazine, remained constant for the first four time points (up to 45 minutes Figure 3-4b). After 45 minutes the total luminescence decreased slowly, and at 180 minutes had dropped to about 70% of its initial value. In the group in which the ESF media was not changed, a slight increase in total luminescence was observed and the total luminescence never dropped below the initial value (Figure 3-4b). This set of experiments demonstrated that the removal of excess coelenterazine from the system had no effect on the outcome of a typical dose response study, that is the fractional luminescence was not affected by the removal of coelenterazine (Figure 3-4a). 54 Q) O C a> u w o> c E 3 ra c o u ra 0.200 0 .150 < 0 .100 0 .050 0 .000 1 l i * \ ° »> *r <b A q> N * N<v N«b N<5 N<© N<b Time (minutes) Figure 3-4a Comparison of fractional luminescent values between stable Sf-9 cells with a change of media (solid bars) and without a change of media (hatched bars) when challenged with ImM serotonin. Error bars represent 95% C.I. + < 0) o c o u w o c E 3 ra *•> o 100.000 90.000 80.000 70.000 60.000 50.000 40.000 30.000 20.000 10.000 0.000 *5> V * A * N * N * N«b N<5 N<b N<6 Time (minutes) Figure 3-4b Comparison of total luminescent values between stable Sf-9 cells with a change of media (solid bars) and without a change of media (hatched bars) when challenged with ImM serotonin. Error bars represent 95% C.I. 55 3.5 Serotonin Dose Response Experiments Having established the parameters for the assay and its uniformity over time, it was possible to test the permanently transformed cell lines for their response to varying concentrations of serotonin. ESF media was used to prepare a dilution series from a 30 mM stock solution of serotonin. Into 36 wells of a 96 well plate a 50 \il aliquot of each dilution was arrayed in triplicate resulting in eleven different concentrations of serotonin ranging from zero to ImM. Figure 3-5 shows a typical layout for a dose response experiments. 1 2 3 4 5 6 7 8 9 10 1 1 12 A I m M 1 m M 1 m M 1 0 0 u M - IOOUM I O O u M 3 0 u M 3 0 u M 3 0 u M ESF ESF ESF 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T B 1 0 u M 1 0 u M 1 0 u M 3 u M 3 u M 3 u M 1 u M 1 u M 1 u M 3 0 0 n M 3 0 0 n M 3 0 0 n M 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T C 1 0 0 n M 1 0 0 n M 1 0 0 n M 3 0 n M 3 0 n M 3 0 n M 1 0 n M 1 0 n M 1 0 n M 1 m M 1 m M 1 m M 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T 5 H T Figure 3-5 Serotonin concentration layout into a 96 well plate. The concentrations shown represent final concentrations The samples were tested in groups of three with each group corresponding to a particular serotonin concentration. To ensure that no drastic decline in cell numbers occurred within the stir vessel during the experiment, cell counts were done at the start and end of the experiment. A full response concentration (1 mM) of serotonin was included, group 2, (wells A4-A6), and group 12, (wells C10-C12, figure 3-5), to provide an internal check for decline in response. The fractional luminescence values at these two test points were 56 compared to ensure no change of maximal activity occurred during the experiment (Figure 3-6). 2 0 . 2 0 0 c 0) o <D 0 . 1 5 0 c 2 0 . 1 0 0 "5 § 0 . 0 5 0 "J o (0 £ o.ooo "' -{vViV/f*'". -G r o u p 2 G r o u p 1 2 Group number Figure 3-6 Comparison of maximal responses (ImM 5HT) achieved at the beginning and at the end of a dose response experiment. The error bars represent the S E M of a single experiment performed in triplicate. 3.6 Serotonin Dose Response Curves For any given dose response experiment the percent of total or maximal response was plotted on the Y-axis against the corresponding L o g 1 0 of the serotonin molar concentration on the X-axis. The software package GraphPad Prism (GraphPad Software Inc.) was utilized for curve fitting. For each serotonin concentration, in a dose response experiment, three independent values for percent of maximal response were obtained. GraphPad Prism took the mean of these three values and plotted that value on a graph against the Log 1 0 of the molar concentration for the agonist. The standard error of 57 the mean was calculated from the three values and was idsplayed as error bars on the graph. Dose response curves were performed for the five different alleles at the 5HT2A receptor. These included the wildtype allele and the four missense variants (T25N, 1197V, A447V and H452Y). In this series of experiments each cell line expressed one of the five alleles of the human 5HT2A receptor along with the human G a l 6 gene and the aequorin gene. A minimum of three independent agonist dose response experiments, and a minimum of three replicates per experiment were performed on these stable Sf-9 cell lines. Each dose response experiment was plotted and fitted separately using the protocol outlined in Section 2.15. The three dose response curves generated for a given 5HT2A allele were overlayed onto a single graph (Figure 3-7). The error bars on the data points represent the standard error of the mean. The graphs for each of the variants were aligned and examined for shifts in dose response. Visually there appeared to be no dramatic difference in potency for any of the missense variants when compared to the wildtype 5HT2A receptor (Figure 3-7). A dramatic difference in functional response was not expected between the wildtype and any of the missense variants. However, there may be more subtle, yet functionally significant differences, and to examine this, E C 5 0 values were determined and examined 58 WildType 1.0 0.5 0.0 J — - 1 I iiiirfTiTnq i in^ imiq 11 0 10"9 Iff" i f f 7 i f f 6 i f f 5 1 0 1 0 " Replicate LogECr, SEM 1 -5.947 0.045 2 -5.886 0.016 3 -5.909 0.019 I197V Replicate LogEC5r) SEM 1 -5.858 0.049 2 -5.896 0.051 3 -5.842 0.063 T25N Replicate LogEC s n SEM 1 -5.860 0.030 2 -5.826 0.025 3 -5.766 0.025 A447V Replicate LogEC 5 0 SEM 1 -6.035 0.038 2 -5.701 0.050 3 -5.949 0.046 4 -5.910 0.031 H452Y Replicate LogEC 5 0 SEM 1 -5.778 0.042 2 -5.812 0.032 3 -5.820 0.038 Log [5HT] M Figure 3-7 Serotonin dose response curves for wildtype 5HT2Areceptors and the four allelic variants. Error bars represent SEM from three independent experiments performed in triplicate. 59 3.7 Comparison of LogEC 5 0 values between wild type and allelic variants To determine whether any of the mutant alleles had an altered response to serotonin, the mean of the three LogEC 5 0 values was calculated for the wildtype and each of the four allelic variant alleles using Graph Pad Prism (GraphPad Software Inc.). The LogEC 5 0 value is the log concentration of agonist required to produce a response that is half of the total response, and these values provide an indication of potency of the agonist for the receptor. A comparison of the LogEC 5 0 values along with their associated S E M , for each of the five alleles of 5HT2A is shown in Figure 3-8. The error bars for I197V and A447V overlapped with the E C 5 0 value of the wildtype allele indicating no difference (Figure 3-8). However there appeared to be a difference with the T25N and H452Y variants when compared to the wild type 5HT2A receptor (Figure 3-8). To determine if these apparent differences had any statistical significance the Mann-Whitney test, a non-parametric test, was performed. The LogEC 5 0 values for each of the four variants were separately compared to the LogEC 5 0 values obtained for the wild type receptor. The P value obtained from each comparison was greater than 0.05 for each of the variant alleles tested (Table 3-II Column five). While not statistically significant in this set of experiments, we note that the T25N variant requires nearly 25% higher serotonin levels to evoke 50% activation (EC 5 0). 60 -5.50-1 o u> o HI G) O •5.75 -6.00H -6.25 Wildtype T25N 1197V A447V H452Y 5HT2A receptor variants Figure 3-8 Comparison of serotonin LogEC 5 0 values for wildtype 5HT2A receptor and the four allelic variants. Error bars represent S E M from three independent experiments performed in triplicate. Table 3-II Summary of Log E C 5 0 values obtained for 5HT2A receptors with serotonin Variant LogEC 5 0 S E M E C 5 0 uM P value from Mann Whitney test Wild type -5.912 0.033 1.23 I197V -5.861 0.047 1.38 0.2000 T25N -5.818 0.028 1.52 0.1000 A447V -5.896 0.046 1.27 0.2000 H452Y -5.804 0.04 1.57 0.2000 61 3.8 Effects of ethanol and DMSO on the Sf-9 insect cell line Serotonin is a highly soluble chemical and is easily dissolved in ESF growth medium. However, many drugs have low solubility in aqueous medium and must first be dissolved in an organic solvent. Of the two antagonists investigated in this study, clozapine was not directly soluble in ESF whereas loxapine was. Two compounds, D M S O and ethanol (EtOH), were both considered as solvents for clozapine. An experiment was designed to determine the maximum concentration of solvent that the Sf-9 cells could withstand without causing damage to the cells or producing a high background response. Four different final DMSO concentrations were tested (5%, 1% 0.5% and 0.3%). DMSO concentrations of 1% 0.5% and 0.3% produced fractional luminescent values that were similar to the fractional luminescent values obtained when the cells were challenged with ESF growth media alone (Figure 3.9a). In contrast the fractional luminescence response of cells challenged with D M S O at a final concentration of 5% was equal to the response achieved when the cells were challenged with 1 m M serotonin (Figure 3-9a). Clearly DMSO at final concentrations of 5% or greater were not useful. The effect of ethanol, at final concentrations of 1%, 0.5%, 0.3%, and 0.1%, on fractional luminescence was also tested. Ethanol at all four concentrations produced fractional luminescent values that were less than or comparable to background response (Figure 3-9b). 62 0) o c 0) o at a> ^ .E - i E + - ^ _ < ra ~ — o ra 0.200 0.150 0.100 0.050 0.000 F N 5% DMSO 1% DMSO 0.5% 0.3% 1mM 5HT DMSO DMSO ESF reagents Figure 3-9a The fractional luminescence response of stable Sf-9 cells challenged with varying concentrations of DMSO, ImM serotonin or ESF growth media. The error bars represent 95% confidence intervals from three independent experiments performed in triplicate. o 0.200 T <u o % 0.150 1 1 c _i § < 0.100 _ < g 0.050 o o o.OOO - I — = ' • = ' ' = — ' — — <—' ' ' — * — 1%EtOH 0.5% 0.3% 0.1% 1mM ESF EIOH EtOH EtOH 5HT reagents Figure 3-9b The fractional luminescence response of stable Sf-9 cells challenged with varying concentrations of ethanol, ImM serotonin or ESF growth media. The error bars represent 95% confidence intervals from three independent experiments performed in triplicate. 63 These data demonstrate that up to a 1% solution of either DMSO or EtOH can be used as a solvent with no significant alteration of the background response in our assay. Clearly the final concentration of either solvent in any of the antagonist dose response experiments must not exceed 1 % of the total liquid media. Clozapine is made up in DMSO (the recommended solvent) at a stock concentration of 10 m M ; therefore, the maximum concentration that can be tested without producing increased background is 100 uM. 3.9 Loxapine and clozapine dose response experiments Dose response experiments were performed separately with loxapine and clozapine to determine each antagonist's ability to block the activation of the 5HT2A receptor. Serial dilutions were made of both antagonists in ESF, starting with stock solutions at 10 mM. A IOJJ-1 aliquot of antagonist was arrayed in triplicate into a 96 well plate with the final concentrations of antagonist ranging from a maximum of 100 u,M down to 1 nM. Figure 3-10 shows the 96 well plate layout for a clozapine dose response experiment; an identical format was used for loxapine dose response experiments. Permanently transformed Sf-9 cells expressing the 5HT2A receptor (wildtype or one of the variants), G a l 6 and aequorin were charged with coelenterazine and handled in the same manner as those cells used in the serotonin dose response experiments (see Section 3.8). 1 2 3 4 5 6 7 8 9 10 11 64 A 100um 100um 100um 10uM 10uM 10uM 1uM 1uM 1uM ESF ESF ESF clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine B 300nM 300nM 300nM 100nM 100nM 100nM 30nM 30nM 30nM 10nM 10nM 10nM clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine C 3nM 3nM 3nM 1nM 1nM 1nM 100pM 100pM 100pM 10pM 10pM 10pM clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine clozapine D blank blank blank blank blank blank blank blank blank ESF ESF ESF Figure 3-10 Clozapine concentration layout into a 96 well plate for dose response experiments A minimum of three separate dose response experiments were performed on stable Sf-9 cell lines for the wildtype 5HT2A receptor and each of the four variants. These experiments were performed with both loxapine and clozapine and always in the presence of 100 uM serotonin (final concentration). The fractional luminescence data collected for these antagonists was handled in the same manner as for the agonist serotonin (Section 3.8). 3.10 Loxapine Dose Response Curves The fractional luminescence data was plotted against the L o g 1 0 corresponding molar concentration of loxapine. The sigmoidal curves were fitted using the three parameter nonlinear regression equation (Section 2.15). The error bars represent the standard error of the mean. For antagonists dose response curves, the Hi l l Slope value was held constant at - 1 . The LogIC 5 0 values were determined; the LogIC 5 0 value represents the concentration of antagonist required to inhibit a response that is half of the total response. 65 The alignment of the five graphs, each representing a minimum of three dose response experiments for each variant, displayed no apparent shift in potency for loxapine (Figure 3-11). 66 Wild Type 1.0-1 i3 0.5 0.0. 1.0. 0.5. 0.0. -0.5. 1.0. 0.5. 0.0. -0.5. I I I ^ mi I I I ^ nm I I M 1 Replicate LOGIC™ SEM 1 -7.222 0.051 2 -7.144 0.084 3 -7.246 0.057 I197V Replicate LOGIC™ SEM 1 -7.128 0.034 2 -7.134 0.042 3 -7.142 0.029 T25N Replicate LOGIC™ SEM 1 -7.287 0.041 2 -7.327 0.036 3 -7.327 0.036 A447V Replicate LOGICfi0 SEM 1 -7.365 0.052 2 -7.310 0.034 3 -7.508 0.042 H452Y Replicate LOGIC™ SEM 1 -7.177 0.066 2 -7.227 0.049 3 -7.263 0.050 -1 I 1 1 1 ^ 1 1 1 ^ 1 1 1 ^ IIIM I l l l l l l ^ [I'm , 10" 10-10 10-8 10"" lO-7 10-6 10-6 10"4 10-Log[bxapine]M Figure 3-11 Loxapine dose response curves for the wildtype receptor and the four allelic variants. Error bars represent S E M from three independent experiments performed in triplicate. 67 3.11 Comparison of the LogIC5 0 values between the wildtype and allelic variants for loxapine The mean LogIC 5 0 values for loxapine were calculated for the wildtype 5HT2A receptor and each of the four missense variants as described in Section 3.10 and plotted onto a bar graph (Figure 3.12). The error bars for the T25N, I197V and H452Y variants overlapped those of the wildtype 5HT2A receptor (Figure 3-12). There appeared to be a difference for the A447V variant with an IC 5 0 value of 40nM compared to the wildtype 5HT2A receptor with an I C 5 0 value of 62.4 n M for loxapine (Table 3-III column four). The Mann-Whitney non-parametric t-test showed no significant difference in potency of loxapine between the A447V variant and the wildtype 5HT2A receptor (P value of 0.1000; Table 3-III column five). The same statistical test was used to compare the other three receptor variants and confirmed that there was no difference in potency for loxapine when compared to the wildtype 5HT2A receptor. 68 -6.75-1 Figure 3-12 Comparison of loxapine LogIC 5 0 values for wildtype 5HT2A receptor and the four allelic variants. Error bars represent S E M from three independent experiments performed in triplicate. Table 3-III Summary of Log IC 5 0 values obtained for 5HT2A receptors with loxapine and the results from Mann Whitney test comparing the wildtype 5HT2A receptor's LogIC 5 0 values for loxapine to the LogIC 5 0 values obtained for the four 5HT2A variants. Variant LogIC 5 0 S E M i c 5 0 n M P value from Mann Whitney test Wild type -7.205 0.064 62.4 I197V -7.14 0.029 72.5 0.1000 T25N -7.313 0.036 48.7 0.1000 A447V -7.398 0.043 40 0.1000 H452Y -7.224 0.051 59.7 0.7000 69 3.12 Clozapine Dose Response Curves The fractional luminescent data collected from the clozapine dose response were analyzed and plotted in the same manner as those for loxapine (Section 2.15). Each graph represents three independent dose response experiments for a single allele of the 5HT2A receptor. The five graphs obtained were aligned. Like the loxapine data no dramatic shift in clozapine potency was observed (Figure 3-13). However, the LogIC 5 0 values for the 1197V variant appeared to differ from that of the wildtype receptor. Therefore, two additional dose response experiments were performed to confirm those results. The first set of three dose response experiments, for the I197V variant, were done on separate days within the same week. The two additional experiments were performed two months after the initial set of three. 70 1.5-1.0 ¥-i—* 0.5-Wild Type Replicate L O G I C * S E M 1 -6.937 0.031 2 -6.970 0.037 3 -6.980 0.042 I197V 1.0-0.5' 0.0 1-0-0.5J 0.0, -0.5. ^iJ-iVi7-iVi^* , ,Tf^"'^-"io-e'i^5'i^^'irjj Replicate LOGIC™ S E M 1 -6.637 0.045 2 -6.672 0.038 3 -6.681 0.037 4 -6.634 0.047 5 -6.629 0.042 T25N Replicate L O G I C 5 0 S E M 1 -6.817 0.048 2 -6.951 0.033 3 -6.849 0.036 A447V Replicate LOGIC™ S E M 1 -7.139 0.084 2 -7.074 0.074 3 -7.038 0.049 4 -7.086 0.033 H452Y Replicate L O G I C 5 0 S E M 1 -6.806 0.072 2 -6.787 0.068 3 -6.919 0.044 Log [dozaphe], M Figure 3-13 Clozapine dose response curves for the wildtype 5HT2A and four allelic variants. Error bars represent S E M from three independent experiments performed in triplicate. 71 3.13 Comparison of the LogIC5 0 values between the wildtype and allelic variants for clozapine The mean LogIC 5 0 values ± S E M obtained for the five 5HT2A receptor alleles challenged with clozapine were plotted onto a bar graph. None of the error bars from any of the receptor variants overlapped with the error bars calculated for the wildtype 5HT2A receptor. Indeed a Mann-Whitney test comparing the potency of clozapine for the allelic variants T25N, A447V and H452Y to the clozapine potency against the wildtype 5HT2A receptor showed no statistically significant difference (P values of 0.2000, 0.1000, and 0.1000 respectively; Table IV, column five). However, the two fold difference in the potency of clozapine against the I197V variant, an IC 5 0 value of 223.2 nM, versus the wildtype, an IC 5 0 value of 109.1 nM, was statistically different with a P value of 0.0375 (Table IV, column five). 72 -6.50-1 Wildtype T25N 1197V A447V H452Y 5HT2A receptor variants Figure 3-14 Comparison of clozapine LogIC 5 0 values for wildtype 5HT2A receptor and the four allelic variants. Error bars represent S E M from three independent experiments performed in triplicate. Table 3-IV Summary of Log IC 5 0 values obtained for 5HT2A receptors with clozapine and the results from Mann Whitney test comparing the wildtype 5HT2A receptor's LogIC 5 0 values for clozapine to the LogIC 5 0 values obtained for the four 5HT2A variants. Variant LogIC50 S E M IC50 nM P value from Mann Whitney test Wild type -6.962 0.029 109.1 I197V -6.651 0.034 223.2 0.0357 T25N -6.873 0.052 134 0.2000 A447V -7.083 0.03 82.7 0.1000 H452Y -6.836 0.051 145.7 0.1000 73 3.15 Does long exposure to antagonist affect the response of GPCRs in stable cell lines? Competitive studies were performed to determine the nature of the interaction between the natural agonist serotonin, and the antagonists, loxapine and clozapine. Prior to undertaking the competition studies, we had to determine whether long-term incubation with either antagonist had any adverse effects on stable cell lines. Stable cell lines expressing the wildtype 5HT2A receptor, G a l 6 and aequorin were exposed to a constant concentration of antagonist for 33 minutes. The cells along with antagonist were injected into the wells of a microtitre plate and were challenged with 1 m M serotonin at three minute intervals for 33 minutes. The results from this group were compared to another aliquot of the same cells treated similarly except they were incubated in the absence of an antagonist. The fractional luminescence was calculated for each time point and the data points for the two experiments were overlayed onto the same graph (Figure 3-15a and 3-15b). The error bars represent 95% confidence intervals from three experiments. The fractional luminescence declined slightly over time for both clozapine and loxapine (Figure 3-15a and Figure 3-15b). However, a similar decline was observed in both the control experiments. When the graphs of the control and experimental groups were overlayed, it was apparent that the decline in response to serotonin was not due to the addition of the antagonist (Figure 3-15a and Figure 3-15b). 74 0.000 12 15 18 21 Time (minutes) 27 30 33 Figure 3-15a Fractional luminescent values for the human wildtype 5HT2A receptor (•) and the wildtype 5HT2A receptor in the presence of 10 n M loxapine ( • ) when challenged with 1 m M serotonin. Error bars represent 95% C.I. from three independent experiments performed in triplicate. 0.000 12 15 18 21 Time (minutes) 24 27 Figure 3-15b Fractional luminescent values for the human wildtype 5HT2A receptor (•) and the wildtype 5HT2A receptor in the presence of 73 n M clozapine ( • ) when challenged with 1 m M serotonin. Error bars represent 95% C.I. from three independent experiments performed in triplicate. 75 3.16 Competitive versus not-competitive Finally, the issue of whether loxapine and clozapine act as competitive or non-competitive inhibitors of the natural agonist serotonin was addressed. A stable cell line expressing the wildtype human 5HT2A receptor, G a i 6 , and aequorin was incubated with several different concentrations of loxapine and clozapine both prior to and during a serotonin dose response experiment (See Section 2.14). Four loxapine concentrations were examined: 0.3 nM, 1.0 nM, 3.0nM, and 10.0 nM. Each dose response experiment was run as described in Section 2.11. Fractional luminescent values were calculated, standardized and plotted as outline in Section 2.15. Each curve represents one experiment performed in triplicate and was plotted along with its S.E. M . The addition of even a relatively low concentration of loxapine causes a reduction in the maximal response achieved compared to the experiment in which loxapine was not added (Figure 3-16a). As the concentration of loxapine was increased the maximal response decreased and a shift in the LogEC 5 0 value for serotonin occurred (Figure 3-16a). Four concentrations of clozapine (O.lnM, 1.0 nM, 3.0nM, and 10.0 nM) were examined. A similar reduction in maximal response and shift in the LogEC 5 0 value was observed when clozapine was added (Figure 3-16b). The effect on reduction of maximal response for any given clozapine concentration was less severe that the effect achieved for the same concentration of loxapine (Figure 3-16b and Figure 3-16a). These results again point to a difference in the potency of these two agonists. 76 1.5-1 Log [5HT], M Figure 3-16a Competition studies with serotonin and loxapine. The symbols indicate • no loxapine,T 0.3 n M loxapine, • 1.0 n M loxapine,* 3.0 n M loxapine, • 10 n M loxapine. Error bars represent S E M of one experiment performed in triplicate. 15-1 Log [5HT], M Figure 3-16b Competition studies with serotonin and clozapine. The symbols indicate • no clozapine,Y 0.1 n M clozapine, • 1.0 n M clozapine,* 3.0 n M clozapine, • 10 n M clozapine. Error bars represent S E M of one experiment performed in triplicate. 77 Chapter 4 Discussion 4.1 Justification for using an insect cell based assay This study used an insect cell based GPCR functional assay to reconstruct the initial portion of the human 5HT2A signal transduction pathway. To ensure that this system was a suitable platform to build a portion of the human proteome, several experiments were initiated prior to examining the response of GPCR variants alleles to agonists and then establishing drug response profiles. First, I asked whether the insect cells would produce the human protein and, if so, whether it was trafficked to the cell membrane. Western blot analysis, using a commercially available monoclonal 5HT2A antibody, of transiently transformed Sf9 cells expressing the wild type human 5HT2A receptor depicted a band of 55kDa, which is of the expected size. When the transformed cells were fractionated the human 5HT2A protein was found only in the membrane fragment sample. Together these results demonstrate that protein synthesis of the human 5HT2A receptor within the Sf9 cell generally parallels that of human cells. Once it was established that the Sf9 cells could produce the human 5HT2A protein, it was important to determine whether the 5HT2A receptor functioned properly. A series of functional GPCR assays done on transiently transformed cells, as outlined in Section 3.2 The human 5HT2A receptor responded to 1 m M of serotonin; and serotonin dose response curves in the Sf9 cells expressing the human 5HT2A receptor were analogous to responses measured in mammalian cells. Therefore, these data demonstrate that the 78 human 5HT2A receptor, produced in the Sf9 cell line and is functional. Sf9 cells transiently expressing all the components of the assay minus the target receptor did not produce any response when challenged with 1 m M serotonin. This result indicates that there are no endogenous 5HT type 2 receptors expressed in the Sf9 cells; or that if endogenous 5HT like receptors are expressed in these cell lines, which is not likely, their activation is so low that the assay is unable to detect them. Next it was demonstrated that the activation of the 5HT2A receptor is due to the presence of the agonist since challenging the cells with only ESF produced no response. The lack of response in the absence of an agonist indicates that the human 5HT2A receptor expressed in insect cells is not constitutively active. Finally, the lack of response to ESF alone (absence of agonist) and the lack of endogenous insect 5HT receptors ensure that the background signal for this functional assay was extremely low. Collectively, these data indicate that the initial part of the human GPCR signaling pathway (receptor and G protein) can be functionally reconstituted in Sf-9 insect cells, and that this system produced a high signal to noise ratio. Interestingly, the human 5HT2A receptor did not require the addition of the human G „ i 6 protein in order to become activated. The human G a l 6 protein was included in the assay due to its ability to couple to many different GPCR subclasses and link them to the phospholipase C(3 pathway. The 5HT2 receptors in mammalian cells couple to the G ^ type G protein, of which G a l 6 is a member. The ability of the 5HT2A receptor to activate the phospholipase C P pathway without the addition of the human G a l 6 protein suggests that the Sf9 cells possesses an endogenous G q type G protein, which is not altogether 79 surprising. However, the endogenous insect G a protein did not appear to have the promiscuity of the human G a l 6 protein. The insect G a subunit did not interact with the dopamine 1 receptor, which regulates G a s type proteins, and required the addition of the human G a l 6 protein in order for the activation of the receptor to be detected by the functional GPCR assay (Grigliatti, 2000). The inclusion of the human G a l 6 protein in this assay did not impede the function of the human 5HT2A receptor nor did adding it improve the response. This result suggests that the plateau of maximal response achieved with the 5HT2A receptor is not due to a shortage of G a proteins to which the receptor must couple. Lastly, a series of antagonist dose response curves were generated using loxapine, an example of the 'typical' class of neuroleptics drugs and clozapine and example of the 'atypical class of neuroleptic drugs. The addition of the antagonist loxapine or clozapine blocked the activation of the 5HT2A receptor in a dose dependent manner. As expected of a strong antagonist the IC 5 0 value of the antagonists were lower that the E C 5 0 value of the natural agonist. 4.2 Effects of missense variants on drug potencies Dose response experiments were performed in stably transformed cell lines for the wildtype and four missense variants with the natural ligand, serotonin. The potency of serotonin for each of the alleles was determined from the dose response curves generated. There was no significant alteration in the potency of serotonin for any of the four missense variants tested. Though serotonin was the only agonist tested in this study, it 80 can be predicted that, synthetic agonists of the 5HT2A receptor would also show no dramatic shifts in potency with these four allelic variants. The four missense variants did not differ in their response to the antagonist loxapine. This lack of change in potency suggests that none of these SNPs play a significant role in the activity of loxapine antagonism. In contrast, the 1197V variant resulted in a greater than two-fold decrease in the potency of clozapine, an atypical neuroleptic 5HT2A antagonist, when challenged with 100 u M of serotonin, compared to the wildtype receptor. This difference in clozapine potency is statistically significant using the Mann-Whitney non-parametric t-test. The remaining three missense alleles showed no change in potency for clozapine when compared to the wildtype receptor. These results suggest that T25N, A447V and H452Y are not involved in clozapine activity. The effectiveness of the two antagonists to block 5HT2A receptor activation can be compared within a single stable cell line. In this system, loxapine was found to be a\ more potent than clozapine when targeting the wildtype 5HT2A receptor. The difference in potency was almost two fold; the IC 5 0 value for loxapine of 62.4 n M whereas clozapine had an IC 5 0 value of 109.1 n M for the same cell line. Therefore, compared to clozapine, less loxapine is required to elicit the same degree of antagonism. 81 A drug does not bind to a receptor in a static manner, but instead is in constant flux between the bound and unbound state. The equilibrium dissociation constant of the complex formed between the drug and receptor determines the affinity a drug has for a particular receptor and is a function of the ratio between the rate of offset (k2) divided by the rate on onset (kl) (Figure 4-1). k1 A + R ^ A R k2 Figure 4-1 Scheme for the reaction between drug A and receptor R. A R indicates drug receptor complex. An increase in the drug concentration thereby causes an increase in the number of receptor drug complexes formed. When an agonist and an antagonist target the same receptor and are present within the same environment they can interact in two general manners. The two ligands can both compete for the same binding site on the receptor. This mechanism is referred to as competitive antagonism. The mechanics of this interaction is such that increasing the concentration of the agonist, compared to the antagonist eventually allows for the agonist to overcome the action of the antagonist. Alternatively, the two ligands may bind to different regions of the receptor (non-competitive antagonism). In this circumstance, increasing the concentration of the agonist is unable to overcome the activity of the antagonist. To determine the nature of the interactions between serotonin and the two antagonists, loxapine and clozapine, competition experiments were performed. Firstly, it should be noted that long term exposure of the stable insect cell lines expressing the human 5HT2A 82 receptor to either of the antagonist did not cause a reduction in the maximal signal over time (Section 3.15). From this experiment it can be determined that the binding of the antagonists loxapine and clozapine to the human 5HT2A receptor is reversible. If either loxapine or clozapine were competitive antagonist then the act of increasing the concentration of serotonin should allow for a maximal response to be attained by the 5HT2A receptor in the presence of a constant concentration of either antagonist. Serotonin was unable to overcome the effects of even very small doses of both loxapine and clozapine (see figures 3-16a and 3-16b). Therefore, in this system, both antagonists are behaving in a non-competitive fashion. The fact that clozapine was the only drug which was affected by the 1197V missense variant is further evidence for a non-competitive interaction. 4.3 Antagonist mode of action The crystal structure of the bovine GPCR rhodopsin has been determined to a resolution of 2.8 A, and these data match the general models that have been developed for GPCR structure (Palczewski, 2000). It is postulated that the receptor undergoes allosteric changes in response to the binding of an agonist, which allows for the G protein to associate with the receptor. However, the nature of the conformational changes that accompany the activation of a GPCR still remain unclear (Choi, 2002). The manner in which a non-competitive antagonist blocks the activation of a GPCR by its agonist is poorly understood. When a non-competitive antagonist binds to a receptor it may physically impede the conformational change that occurs when an agonist binds to the 83 receptor. Without the conformational change the receptor may be unable to bind and activate the G proteins and thus unable to transmit a response. Alternatively, antagonist binding may change the receptor conformation so that the agonist is unable to recognize or interact with the receptor and thus no response is initiated. The I197V variant reduces the ability of clozapine to block receptor activation. Based on the above hypothesis clozapine must invoke a conformational change in the 5HT2A receptor that includes the isoleucine residue. In the wildtype situation clozapine induces a conformational change that causes a reduction in the affinity serotonin has for the 5HT2A receptor thereby blocking receptor activation. However, when valine is replaced for isoleucine these conformational changes are different and as a result there is a reduction in clozapine's potency. Since the potency of loxapine is unaffected by the 1197V variant it suggests that the conformational changes that occur when loxapine is bound to the 5HT2A receptor are not affected by the I197V variant. It can be concluded that the changes in receptor structure induced by loxapine binding do not involve the 1197 residue. 4.4 Physiological relevance of the 1197V variant This was an introductory study that investigated the molecular consequences of missense variants in the human 5HT2A receptor when challenged with the natural ligand serotonin and two antagonists clozapine and loxapine. The ultimate goal of this project is to determine if the functional GCPR assay employed in this study can be used as a tool to 84 predict the in vivo effects SNPs within receptors will have on drug potency. This assay has demonstrated a two fold difference in clozapine potency for the II97V variant when compared to the wildtype 5HT2A receptor. This ability to detect a difference in potency between allelic variants provides evidence of the sensitivity of the assay. This two fold difference was observed in all assays even replicates that were performed at very different times (months apart) and all transformed cell-lines behaved similarly, demonstrating the high reproducibility of the assay. However, what remains unclear is whether this difference in potency has a significant physiological effect in humans. Clozapine is administered to patients in doses ranging from 200-500 mg / day and therapeutic levels of clozapine in blood serum are between 350 ng - 510 ng / ml (Spina, 2000). The IC 5 0 value obtained with this assay for clozapine at the wildtype receptor was 109.1 n M which translates into 35.6 ng of clozapine / ml. Though the values obtained with this assay do differ from therapeutic levels, they are within a reasonable range. Also the concentration of clozapine crossing the blood-brain barrier remains unknown as does the concentration at the target site. The elimination half-life of clozapine was found to 14.2 hours and is such that a two fold difference in receptor potency may play a factor in the pharmacological activity of clozapine (Choc, 1990). Though the physiological composition of the Sf-9 insect cells is not identical to human cells, this system was utilized for several reasons. Firstly, there were no endogenous 5HT2A receptors expressed in these cells which provided a low noise to signal ratio. Furthermore, the Sf-9 cell line appears to have few detectable endogenous GPCRs 85 (Grigliatti, 2000). Secondly, the insect cells were grown in a serum free environment, not only making them easier to maintain but the testing of the cells occurred in the same environment which they were grown in. Therefore, the effects that serum has on the cells and the receptors prior to challenging them with an agonist is not of a concern as it may be with other systems. Two systems have recently been employed to investigate the pharmacology of the 5HT type 2 receptors (Porter, 1999; Jerman, 2001). Both systems utilized Fluo-3 to measure changes in concentration for intracellular calcium with the aid of a fluorometric imaging plate reader. The first study expressed the human 5HT2A receptor in CHO (Chinese hamster ovary) cells and determined the potency of several agonists (Porter, 1999). The second study performed similar experiments however the S H - S Y 5 Y (human neuroblastoma) cell line was utilized to express the 5HT2A receptor (Jerman, 2001). Though the LogEC 5 0 values obtained from these two systems differed, the ranked order of potency for the four most potent agonists of the 5HT2A receptor were identical producing the following ranked list: 2,5-dimethoxy-4-iodoamphetamine hydrobromide (DOI) > SCH-23390 > serotonin (5HT) > a-methyl-5-hydroxytryptamine; the remaining agonists tested were not identical in ranked order but showed many similarities (Porter, 1999; Jerman, 2001). Reproducing the above results with the functional GPCR assay in insect cells would further validate the change in potency observed with the II97V variant in the presence of clozapine. 86 4.5 Future directions From this study it is evident that the I197V polymorphism changes the potency of clozapine; however, it is unclear just how the I197V change causes this effect. To determine whether or not the 1197 is involved in clozapine binding, the affinity clozapine has for both the wildtype and 1197V alleles of the 5HT2A receptor must first be ascertained. A comparison of these two values would then indicate the role affinity plays in the potency reduction observed with the I197V variant. Binding affinity studies together with this study could also be used to investigate the mechanism for non-competitive antagonism. The affinity serotonin has for the 5HT2A receptor, in the absence and in the presence of clozapine, should be investigated. If a difference in affinity between the two conditions is found, it supports the theory that the antagonist causing a conformational change to the receptor upon binding, in turn, changes the agonist's affinity. The 1197V variant should be included in these affinity studies to determine the role the isoleucine residue plays in the two fold decrease in potency observed for clozapine. The functional GPCR assay used in this study is a good tool for determining drug potency. Using a single stable cell line expressing the 5HT2A receptor, G a l 6 and aequorin it is possible to compare the activities of several different drugs. To further strengthen the theory that similar drug activity occurs between insect cells and mammalian cells the potency of more agonists should be determined. Utilizing this assay 87 for this type of study would provide a list of LogEC 5 0 values for the agonists, which could be compared to the ranked order of potency achieved with the other two systems previously mentioned (Section 4.4). Though the potency values may differ from one system to the next, the ranked order of potency derived for the agonists should be the same when compared to the CHO and SH-SY5Y cell systems. Demonstrating such a similarity between these three systems would provide further evidence to suggest that the change in potency with the I197V variants has physiological significance. In conjunction with the creation of a ranked order of potency for agonists of the 5HT2A receptor, other antagonists of this receptor need to be investigated. A change in potency has been demonstrated for one SNP of the 5HT2A receptor with one atypical antagonist. It would be of interest to determine if the four missense variants of the 5HT2A receptor show any change in potency for other antipsychotics used to treat schizophrenia. The three main antipsychotics that should be considered are risperidone, olanzapine and quetiapine, which are all atypical in nature. The typical neuropletics haloperidol and chloropromazine, though not used as often due to their side effects, are of equal importance. If a potency difference for another agonist could be linked to one the 5HT2A missense variants, it would further validate the merits of the functional GPCR assay system. Also the determination of a change in potency of these antagonists for any of the four missense variants of the 5HT2A receptor would provide a larger group of potential candidates to investigate in clinical studies. 88 The long term goal of this study is to demonstrate that this functional GPCR assay system can be used as a tool to link variable response to neuroleptic drugs to genetic variations in drug receptors that exist in the population. Once a similarity between the mammalian cell systems and this assay system is established it would seem prudent to begin clinical studies of schizophrenic patients to assess whether the I197V variant has a phenotypic consequence in schizophrenic patients. If other SNPs of the 5HT2A receptor have been shown to have an effect on potency with other antagonists they would also be include in these clinical trials. Clinical trials should be of two types. Samples from schizophrenic patients who failed to respond or responded poorly to clozapine drug therapies should be typed to determine which SNP of the 5HT2A (if any) they possessed. These results would be compared to the genotypes for the 5HT2A loci of schizophrenic patients that responded to clozapine drug therapies in order to determine if a lack of clozapine response is linked to the I197V variant. This type of study provides beneficial information, however, there are no controls over drug levels administered or the criteria used to classify a patient as a non-responder. Therefore, studies in which schizophrenic patients are genotyped for the SNP, placed onto a clozapine drug regime and monitored for a response to clozapine should be undertaken. Linking the results from the above studies with studies that have control of clozapine dose levels and non-responder criteria would strengthen the hypothesis that could be drawn into whether variable patient response to clozapine is linked to the I197V variant. 89 Bibliography Andreoli, S. et al. Age at first registration in schizophrenia: a comparison of mental health registers from Australia and Brazil Schizophrenia Research, 2002. 54: p. 277-279. Arranz, M . , et al., Association between clozapine response and allelic variation in 5-HT2A receptor gene [see comments]. Lancet, 1995. 346: p. 281-2. Arranz, M.J. , et al., Analysis of a structural polymorphism in the 5-HT2A receptor and clinical response to clozapine. Neuroscience Letters, 1996. 217: p. 177-8. Arranz, M.J. , et al., Meta-analysis of studies on genetic variation in 5-HT2A receptors and clozapine response. Schizophrenia Research, 1998. 32: p. 93-9. Arranz, M.J. , et al., Pharmaco genetic prediction of clozapine response [letter]. Lancet, 2000. 355(9215): p. 1615-6. Bockaert, J. et al. Molecular tinkering ofG protein -coupled receptors: an evolutionary success. The E M B O Journal, 1999. 18: p. 1723-1729. Bonhaus, D. et al. The pharmacology and distribution of human 5-hydoxytryptamine2B(5HT2B) receptor gene products comparison with 5HT2A and 5HT2C receptors. British Journal of Pharmacology, 1995. 115: p. 622-628. Cargill, M . et al. Characterization of single nucleotide polymorphisms in coding regions of human genes. Nature Genetics, 1999. 22: p. 231-238. Cichon, S et al. Pharmacogenetics of Schizophrenia. American Journal of Medical Genetics, 2000 97: p. 98-106. Chen, K. et al. The human 5-HT2 receptor is encoded by a multiple intron-exon gene. Molecular Brain Research, 1992. 14: p. 20-26. Chen, R. et al. No association between T102Cpolymorphism of serotonin-2A receptor and clinical phenotypes of Chinese schizophrenic patients. Psychiatry Research, 2001. 105: p. 175-185. Choc, M . et al. Single vs. multiple dose pharmacokinetics of clozapine in psychiatric patients. Pharmaceutical Research. 1990. 7: p. 347-351. Choi, G. et al. Structural studies of metarhodopsin II, the activated form of the G-protein coupled receptor, rhodopsin. Biochemistry, 2002. 41: p. 7318-7324. Choudhary, M . et al. Differential ergoline and ergopeptine binding to 5-hydroxytryptamine 2A receptors: ergolines require an aromatic residue at 90 position 340for high affinity binding. Molecular Pharmacology, 1995. 47: p. 450-457. Clapham, D. and E. Neer, G protein beta/gamma subunits. Annu Rev Pharmacol Toxicol, 1997. 37: p. 167-203. Cowen, P. Serotonin Receptor Subtypes: Implications for Psychopharmacology. British Journal of Psychiatry. 1991 Supp: p. 7-14. Cravchik, A . et al. Neurochemical individuality. Arch Gen Psychiatry, 2000. 57: p. 1105-1114. Erdmann, J., et al., Systematic screening for mutations in the human serotonin-2A (5-HT2A) receptor gene: identification of two naturally occurring receptor variants and association analysis in schizophrenia. Hum Genet, 1996. 97(5): p. 614-9. Gether, U . et al. G-protein coupled receptors. Journal of Biological Chemistry, 1998. 273: p. 17979-17982. Golimbet, V . E. et al. 5HT2A gene polymorphisms and personality traits in patients with major psychoses. European Psychiatry. 2002. 17: p. 24-28. Grigliatti, T.A., P.J.K. Knight, and T.A. Pfeifer, Functional Assay for G-Protein Coupled Receptors Based on Insect Cells. Provisional Application for Patent, 2000. Docket number UBC 98-055. Hamm, H. The many faces ofG protein signaling. Journal of Biological Chemistry, 1998. 273: p. 669-672. Hegedus, D.D., et al., A series of broad host range shuttle vectors for constitutive and inducible expression of heterologous proteins in insect cell lines. Gene, 1998. 207: p. 241-249. Hildebrandt, J.D., Role of subunit diversity in signaling by heterotrimeric G proteins. Biochem Pharmacol, 1997. 54(3): p. 325-39. Ho, B. et al. The role of conserved aspartate and serine residues in ligand binding and in function of the 5-HT1A receptor: a site directed mutation study. FEBS Letters, 1992 312: p. 259-262. Hoyer, D. and G. Martin, 5-HT receptor classification and nomenclature: Towards a harmonization with the human genome. Neuropharmacology, 1997. 36(4/5): p. 419-428. Humphrey, P. P. A . et al. A proposed new nomenclature for 5HT receptors. Trends Pharmacol Sci, 1993.14: p. 233-236. 91 Hurley, J. et al. Structure function studies of the eighth hydrophobic domain of a serotonin receptor. Journal of Neurochemistry, 1999. 72: p. 413-421. Inayama, Y. , et al., Positive association between a DNA sequence variant in the serotonin 2A receptor gene and schizophrenia. Am J Med Genet, 1996. 67(1): p. 103-5. Jerman, J. et al. Pharmacological characterisation of human 5-HT s receptor subtypes. European Journal of Pharmacology. 2001. 414: p. 23-30. Jones, K. , F. Hibbert, and M . Keenan, Glowing jellyfish, luminescence and a molecule called coelenterazine. Trends in Biotechnology, 1999. 17(12): p. 477-81. Kalow, W. Pharmacogenetics in perspective. Drug Metabolism and Disposition, 2001. 29: p. 468-470. Kawanishi, Y . et al. Pharmacogenomics and schizophrenia. European Journal of Pharmacology, 2000. 410: p. 227-241. Kendall, J .M. and M . N . Badminton, Aequorea victoria bioluminescence moves into an exciting new era. Trends in Biotechnology, 1998. 16: p. 216-224. King, D. Drug treatment of negative symptoms of schizophrenia. European Journal of Neuropsychopharmacology, 1998. 8: p. 33-42. Kiselyov, K . and S. Muallem, Fatty acids, diacylglycerol, Ins( 1,4,5)P3 receptors and Ca2+ influx. Trends in Neurosciences, 1999. 22(8): p. 334-7. Kristiansen, K . et al. A highly conserved aspartic acid (Asp-155) anchors the terminal amine moiety of tryptamines and is involved in membrane targeting if the 5-HT2A serotonin receptor but does not participate in activation via a slat bridge disruption mechanism. The Journal of Pharmacology and Experimental Therapeutics, 2000. 293: p. 735-746. Leurs, R. et al. Agonist-independent regulation of constitutively active G-protein-coupled receptors. TIBS, 1998. 23: p. 418-422. Liggett, S. Pharmacogenetics ofbeta-1- and beta-2-adrenergic receptors. Pharmacology, 2000. 61: p. 167-73. Malhotra, A . K . , et al., Lack of association between polymorphisms in the 5-HT2A receptor gene and the antipsychotic response to clozapine. A m J Psychiatry, 1996.153(8): p. 1092-4. Mancama, D. et al., Genetics Predictors of Therapeutic Response to Clozapine. CNS Drugs, 2002. 16: p. 317-324. 92 Masellis, M . , et al., Serotonin subtype 2 receptor genes and clinical response to clozapine in schizophrenia patients. Neuropsychopharmacology, 1998. 19(2): p. 123-32. Meyer, U . Pharmacogenetics and adverse drug reactions. The Lancet, 2000 356: p. 1667-1671. Morris, A . and Malbon, C. Physiological Regulation of G protein linked signaling. Physiological Reviews, 1999. 74: p. 1373-1430. Nothen, M . M . , et al., Genetic variation of the 5-HT2A receptor and response to clozapine [letter; comment]. Lancet, 1995. 346(8979): p. 908-9. Offermanns, S. and M.I. Simon, Galpha 15 and Galpha 16 couple a wide variety of receptors to phospholipase C. The Journal of Biological Chemistry, 1995. 270(25): p. 15175-15180. Ozaki, N . , et al., A naturally occurring amino acid substitution of the human serotonin 5-HT2A receptor influences amplitude and timing of intracellular calcium mobilization. J Neurochem, 1997. 68(5): p. 2186-93. Ohara, K. , et al., 5HT2A receptor gene promoter polymorphism -1438 G/A and mood disorders. Clinical Neuroscience and Neuropathology, 1998. 9: p. 1139-1141. Ohara, K . et al., Schizophrenia and the serotonin-2A receptor promoter polymorphism. Psychiatry Research, 1999. 85: p. 221-224. Palczewski, K. et al. Crystal structure of rhodopsin: a G protein coupled receptor. Science, 2000. 289: p. 739-745. Petronis, A . The genes for major psychosis: aberrant sequence of regulation. Neuropsychopharmacology, 2000. 23: p. 1-12. Pfeifer, T.A., et al., Baculovirus immediate-early promoter-mediated expression of the Zeocin resistance gene for use as a dominant selectable marker in Dipteran and Lepidopteran insect cell lines. Gene, 1997.188: p. 183-190. Pfeifer, T.A., Expression of heterologous proteins in stable insect cell culture. Current Opinion in Biotechnology, 1998. 9(5): p. 518-521. Pickar, D. and Rubinow, K . Pharmacogenomics of psychiatric disorders. TRENDS in Pharmacological Sciences, 2001. 22: p. 75-83. Porter, J, and Perez, D. Characteristics for a Salt bridge switch mutation ofthealb adrenergic receptor. The Journal of Biological Chemistry, 1999. 274: p. 34535-93 34538. Porter, R.H.P. et al. Functional characterization of agonists at recombinant human 5-HT2A, 5-HT2B, and 5-HT2C receptors in CH0-K1 cells. British Journal of Pharmacology, 1999. 128: p. 13-20. Raymond, J.R., Multiple mechanisms of receptor-G protein signaling specificity. American Journal of Physiology, 1995. 269: p. F141-F158. Remington, G. and Kapur, S. Atypical antipsychotics: are some more atypical than others. Psychopharmacology, 2000. 148: p. 3-15. Roth, B. et al. Identification of conserved aromatic residues essential for agonist binding and second messenger production at 5-hydroxytryptamine2A receptors. Molecular Pharmacology. 1997 52: p. 259-266. Roth, B . et al. 5-hydroxytryptamine2-family receptors (5-hydroxytryptaine2A, 5-hydroxytryptamine2B, 5-hydroxytrptamine2C): where structure meets function. Pharmacology Therapeutics, 1998. 79: p. 231-257 Saltzman, A. et al. Cloning of the human serotonin 5-HT2 and 5-HT1C receptor subtypes. Biochemical and Biophysical Research Communications. 1991. 181: p. 1469-1478. Saxena, P. Serotonin Receptors: Subtypes, functional responses and therapeutic relevance. Pharmacology Therapeutics, 1995. 66: p. 339-368. Schmidt, C. et al. The role of 5HT2A receptors in antipsychotic activity. Life Sciences, 1995. 56: p. 2209-2222. Sealfon, S, et al. Related Contribution of Specific Helix 2 and 7 residues to conformational activation of the serotonin 5-HT2A receptor. The Journal of Biological Chemistry, 1995. 270: p. 16683-16688. Shapiro, D. et al Differential modes of agonist binding to 5-hydroxytryptamine 2A receptors revealed by mutation and molecular modeling of conserved residues in transmembrane region 5. Molecular Pharmacology, 2000. 58: p. 877-886. Shapiro, D. et al. Evidence for a model of agonist-induced activation of 5-hydroxytryptamine 2A serotonin receptors that involves the disruption of a strong ionic interaction between helices 3 and 6. The Journal of Biological Chemistry, 2002. 277: p. 11441-11449. Singh, A . N . et al. A neurochemical basis for the antipsychotic activity of Loxapine:interactions with dopamine DI, D2, D4 and Serotonin 5-HT2 receptor subtypes. Journal of Psychiatry and Neuroscience,1996. 21: p. 29-35. 94 Spina, E et al. Relationship between plasma concentrations of clozapine and norclozapine and therapeutic response in patients with schizophrenia resistant to conventional neuroleptics. Psychopharmacology. 2000. 148: p. 83-89. Stam, N . et al. Genomic organization, coding sequence and functional expression of human 5-HT2 and 5-HT1A receptors. European Journal of Pharmacology, 1992. 227: p. 153-162. Veenstra-VanderWeele, J. et al. Pharmacogenetics and the serotonin system: initial studies on future directions. European Journal of Pharmacology. 2000. 410: p. 165-181. Wu, C. et al. Development and characterization of monoclonal antibodies specific to the serotonin 5-HT2a receptor. The Journal of Histochemistry and Cytochemistry, 2000. 46(7): p. 811-824. 95 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090781/manifest

Comment

Related Items