Dalwadi, Dhwanil, A, PREVENTION and TREATMENT of DISEASES: a SMALL MOLECULE DISCOVERY and DEVELOPMENT. Doctor of Philosophy (Bio

Home , Chlorophenylbiguanide, Dextrallorphan, Dichloroisoprenaline

Dalwadi, Dhwanil, A, PREVENTION AND TREATMENT OF DISEASES: A

SMALL MOLECULE DISCOVERY AND DEVELOPMENT. Doctor of Philosophy

(Biomedical Sciences), August, 2016, 227 pp., 6 tables, 32 illustrations, 256 references.

This work examined the structure-activity relationship, and molecular mechanisms of different structural classes of small molecules at their target receptors.

Three different systems were explored and each chapter is devoted to a single system. All three systems utilized similar experimental approaches, and practical application of the same core pharmacological principles. The first system involved the evaluation of the structure-activity space of small molecules acting on the α-like octopamine receptors from the barnacle Balanus improvisus (BiOctR) and the fruit fly Drosophila melanogaster (DmOctR). A number of molecules belonging to the imidazole and imidazole structural class were determined to have high potency for the BiOctR and the

DmOctR. This information will be useful in designing new OctR ligands that are highly selective for the OctRs over their mammalian off-targets. Similarly, for the second system, the structure-activity space of different structural classes of sigma-1 receptor

(S1R) ligands were evaluated. Four novel EPGN compounds with more than 100-fold selectivity for the S1R over the sigma-2 receptor were identified which were able to stimulate S1R-mediated BDNF secretion. Potential therapeutic applications of these compounds include the treatment of neurodegenerative diseases like Alzheimer’s disease,

Parkinson’s disease, and amyotrophic lateral sclerosis. The third system involved the identification of receptor off-targets of efavirenz that may be responsible for efavirenz’s

neuropsychiatric adverse events (NPAEs). In this study, multiple receptor targets of efavirenz belonging to the serotonin receptor family and the muscarinic receptor family of G protein-coupled receptors (GPCR) were identified, and its mechanism of action at these targets was established. The most prominent finding of this study was that efavirenz functioned as an inverse agonist, antagonist and an allosteric modulator, depending on off-the target receptor. Knowing which off-target receptors efavirenz interacts with may help to understand the molecular mechanisms responsible for efavirenz’s NPAEs.

Overall, the insights gained regarding the mechanisms of action of small molecules will aid in the discovery and development of novel compounds, or an improved understanding of known compounds with established or potential therapeutic value.

PREVENTION AND TREATMENT OF DISEASES:

A SMALL MOLECULE DISCOVERY

AND DEVELOPMENT APPROACH

Dhwanil A Dalwadi, BSc hon

APPROVED:

______

Major Professor

______

Committee Member

______

Committee Member

______

Committee Member

______

University Member

______

Chair, Department of Pharmacology and Neuroscience

______

Dean, Graduate School of Biomedical Sciences

______

PREVENTION AND TREATMENT OF DISEASES:

A SMALL MOLECULE DISCOVERY

AND DEVELOPMENT APPROACH

DISSERTATION

Presented to the Graduate Council of the

Graduate School of Biomedical Sciences

University of North Texas

Health Science Center at Fort Wroth

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

Dhwanil A. Dalwadi, BSc hon

Fort Wroth, Texas

August 2016

Dhwanil A. Dalwadi

2016

ii ACKNOWLEDGEMENTS

I wish to thank my mentor Dr. John A Schetz for his guidance and the opportunity to work on exciting projects. I would also like to thank my committee members for their support and valuable input that helped to make this project a success.

iii TABLE OF CONTENTS

LIST OF TABLES……………………………………………………………………………...viii

LIST OF ILLUSTRATIONS…………………………………..…………………………………ix

I. INTRODUCTION…………………………………………………………………………….1

Classical versus reverse pharmacology……………………………………………………….2

Application of pharmacological principles for evaluating small molecules at receptor targets………………………………………………………………………………………….4

In vitro experimental approaches to studying receptor-ligand interactions…………………...7

Estimating agonist affinity from functional assays using the operational model of agonism………………………………………………………………………………………..9

Estimating antagonists affinity from functional assays using Schild analysis………………11

Application of theory………………………………………………………………………...13

References……………………………………………………………………………………28

II. EXPLORATION OF THE STRUCTURE-ACTIVITY SPACE OF CLONED Α-LIKE OCTOPAMINE RECEPTORS FROM A MARINE BARNACLE AND A TERRESTRIAL FLY………………………………………………………………………..33

Abstract………………………………………………………………………………………33

Introduction…………………………………………………………………………………..34

Materials and methods……………………………………………………………………….40

Chemicals………………………………………………………………………………...40

Establishment of a stable cell line expressing the Barnacle Balanus Improvisus and D. melanogaster octopamine receptors (BiOctR and DmOctR, respectively)……………...40

Profiling of BiOctR by radioligand binding……………………………………………..41

iv Assessment of sodium sensitivity of BiOctR……………………………………………43

Protein Assay…………………………………………………………………………….43

Intracellular Calcium assay: measurement of Gq-coupled response……………………..43

Operational model of agonism and Schild-shift analysis………………………………..44

Cyprid motility assay…………………………………………………………………….45

Results………………………………………………………………………………………..46

Discussion……………………………………………………………………………………56

Conclusion…………………………………………………………………………………...61

References……………………………………………………………………………………79

III. NOVEL SELECTIVE SIGMA-1 RECEPTOR LIGANDS FACILITATE BDNF RELEASE FROM A NEURONAL CELL………………………………………………………………86

Abstract………………………………………………………………………………………86

Introduction…………………………………………………………………………………..87

Materials and methods……………………………………………………………………….92

Chemical and reagents…………………………………………………………………...92

Cell culture……………………………………………………………………………….92

Measurement of BDNF secretion via in situ ELISA…………………………………….93

Measuring receptor density with radioligand binding…………………………………...94

Measuring the ligand affinities at S1R and S2R receptors by radioligand binding……...95

Measurement of the effect of sigma ligands on IP3-induced change in intracellular calcium…………………………………………………………………………………...95

Effect of sigma ligands on NGF induced neurite sprouting……………………………..96

Results………………………………………………………………………………………..97

Discussion…………………………………………………………………………………..104

v Conclusion………………………………………………………………………………….111

References…………………………………………………………………………………..126

IV. MOLECULAR MECHANISMS OF SEROTONERGIC ACTION OF THE HIV-1 ANTIRETROVIRAL EFAVIRENZ……………………………………………………….135

Abstract……………………………………………………………………………………..135

Introduction…………………………………………………………………………………138

Materials and methods……………………………………………………………………...141

Chemicals……………………………………………………………………………….141

Profiling serotonin receptors by radioligand binding…………………………………..141

Measuring the affinity of efavirenz for selected serotonin receptor subtypes using competition radioligand binding………………………………………………………..143

Intracellular Calcium and IP-One Assay: measurement of Gq-coupled response……...144

Effect of prolonged exposure to efavirenz on receptor density and Gq-coupled activation of the 5-HT2A receptor………………………………………………………………….145

Cyclic adenosine monophosphate (cAMP) Assay: measurement of Gs-coupled responses Inhibition of monoamine oxidase A (MAO-A) activity………………………………..146

Inhibition of monoamine oxidase A (MAO-A) activity………………………………..147

Protein Assay…………………………………………………………………………...147

Measurement of 5-HT3A and GABAA receptor currents using electrophysiology……..147

Results………………………………………………………………………………………149

Discussion…………………………………………………………………………………..158

Conclusion………………………………………………………………………………….166

References…………………………………………………………………………………..195

Acknowledgements…………………………………………………………………………207

Author contribution…………………………………………………………………………208

vi Funding sources…………………………………………………………………………….208

Conflict of interest………………………………………………………………………….208

V. CONCLUSION……………………………………………………………………………..209

Exploration of the structure-activity space of cloned α-like octopamine receptors from a marine barnacle and a terrestrial fly………………………………………………………...209

Novel selective Sigma-1 receptor ligands facilitate BDNF release from a neuronal cell line……………………………………………………………………………………...212

Molecular mechanisms of serotonergic action of the HIV-1 antiretroviral efavirenz……...215

Summary statement…………………………………………………………………………217

References…………………………………………………………………………………..218

vii LIST OF TABLES

Table 2-1 BiOctR has affinity for α-adrenergic receptor antagonists, D2 dopamine receptor antagonists and a histamine H1 receptor antagonist……………………………………………...76

Table 2-2 Table 2. Imidazoles are potent agonists at the BiOctR and have higher affinity than biogenic amines………………………………………………………………………………….77

Table 3-1 PRE-084 is the most selective prototypical agonist and BD1063 is the most selective prototypical antagonist………………………………………………………………………….123

Table 3-2 Effect of specific substructural modifications on selectivity for S1R over S2R…….124

Table 3-3 Efficacy of S1R ligands that function as agonists of BDNF secretion……………....125

Table 4-1 Conditions for measuring efavirenz’s ability to displace radioligands specifically bound to the serotonin receptor subtypes……………………………………………………….142

viii LIST OF ILLUSTRATIONS

Figure 1-1. Simplified schematic representation of forward and reverse pharmacology…..……14

Figure 1-2. Pharmacological classification of ligands…………………………………………...16

Figure 1-3. Structure-activity relationships controlling functional selectivity of the MOR agonist oliceridine, a new investigational drug for the treatment of moderate to severe pain…………...18

Figure 1-4. Ligand structural modification that increases β-adrenergic receptor (β-AR) selectivity over the α-adrenergic receptor (α-AR)…………………………………………………………..20

Figure 1-5. Basic components of two common types of equilibrium radioligand binding Assays……………………………………………………………………………………………22

Figure 1-6. Estimation of agonist affinity using the operational model of agonism…………….24

Figure 1-7. Estimation of antagonist potency using Schild analysis…………………………….26

Figure 2-1. [3H]-rauwolscine binds with high affinity to BiOctR under high salt condition……62

Figure 2-2. BiOctR has high affinity for α-adrenergic antagonists and the imidazole dexmedetomidine………………………………………………………………………………...64

Figure 2-3. The α-adrenergic receptor agonists activate BiOctR-mediated Gq signaling……….66

Figure 2-4. BiOctR has higher potency for imidazolines than biogenic amines………………...68

Figure 2-5. Imidazoles induce hyperactivity in barnacle Balanus amphitrite cyprids…………..70

Figure 2-6. The drosophila α-like octopamine receptor (DmOctR) has similar pharmacology to the BiOctR……………………………………………………………………………………….72

Figure 2-7. Receptor reserve has only a small effect on the agonist potency and rauwolscine is non-competitive antagonist………………………………………………………………………74

Figure 3-1. Prototypical sigma receptor ligands PRE-084, NE-100 and BD1063 have greater selectivity for S1R over S2R, whereas 4-PPBP does not………………………………………113

ix Figure 3-2. EPGN745, 862, 863, 644 and 1276 are highly selective for the S1R…………...…115

Figure 3-3. All compounds in series A, B and C facilitate S1R mediated BDNF secretion from the neuronal MN9D cell line……………………………………………………………………117

2+ Figure 3-4. S1R does not appear to modulate Gq-PLC-IP3-[Ca ]i mediated calcium release…119

Figure 3-5. S1R activation has no effect on neurite sprouting, nor does it potentiate NGF induced neurite sprouting in PC-6-15…………………………………………………………………....121

Figure 4-1. Efavirenz has significant interactions with the cloned metabotropic 5-HT2A, 5-HT2B, 5-HT2C and 5-HT6 receptors……………………………………………………………………168

Figure 4-2. Efavirenz has low micromolar affinity for cloned metabotropic 5-HT2A, 5-HT2C and 5-HT6 receptors…………………………………………………………………………………170

Figure 4-3. The cloned metabotropic 5-HT2A, 5-HT2C and 5-HT6 receptors are potently activated by serotonin……………………………………………………………………………………..172

Figure 4-4. Efavirenz is a 5-HT2A receptor antagonist of Gq signaling………………………...174

Figure 4-5. Schild shift analysis of efavirenz at the cloned 5-HT2A receptor suggests that it competes for the same binding site as (±)-DOI and LSD……………………………………....176

Figure 4-6. Chronic exposure to efavirenz reduces the 5-HT2A receptor density and 5-HT mediated Gq activation at the 5-HT2A receptor…………………………………………………178

Figure 4-7. Efavirenz is an antagonist of the cloned 5-HT2C receptor………………………….181

Figure 4-8. Efavirenz partially inhibits the activation of the M3 muscarinic receptor, and completely inhibits the activation of the M1 muscarinic receptor……………………………...183

Figure 4-9. Efavirenz behaves as an inverse agonist of the cloned 5-HT6 receptor……………185

Figure 4-10. Efavirenz inhibits cloned 5-HT3A receptor currents………………………………187

Figure 4-11. Efavirenz modestly inhibits monoamine oxidase-A……………………………...189

Figure 4-12. Nevirapine, but not emtricitabine or zidovudine, interacts with the cloned 5-HT6 receptor…………………………………………………………………………………………191

Figure 4-13. The ARV drugs zidovudine (ZDV), emtricitabine (FTC), but not nevirapine (NVP), inhibit cloned GABAA receptor current………………………………………………………...193

x CHAPTER 1

INTRODUCTION

In pharmacology, a drug is considered a small molecule if its molecular weight is ≤ 500

Daltons (g/mole) (Craik et al. 2013; Leeson and Springthorpe 2007; Lipinski 2004). According to the FDA small molecules are chemically synthesized compounds with a well-defined structure and can be thoroughly characterized. Biologics are another class of drugs that are generally derived from living materials (e.g. human, animal, microorganisms, etc.). They are large molecules with complex structures and are usually not fully characterized. Examples of biologics include vaccines, blood or blood components, gene therapies, and recombinant proteins (FDA

2015). Between 1983 and 2010, out of all the drugs approved, only 39% were biologics, the rest were small molecules (Rask-Andersen et al. 2011). In 2015 alone, the FDA approved 45 new drugs, of which 69% were small molecules (FDA 2016). Hence, the majority of the drugs currently in the market are small molecules. One of the greatest advantages of small molecules over biologics is that they are too small to be immunogenic, thus they are not recognized by the immune system as invaders, whereas biologics have a greater potential for inducing an immune reaction due to their larger size and biological nature (Morrow and Felcone 2004). Further, biologics are also costlier to maintain due to special handling requirements (such as controlled

1 temperature, protection from jostling, sterile environment), less stability, and the possibility of microbial contamination. Another advantage small molecules have over biologics is that they are more likely to be easily absorbed, and thus can be taken orally. Biologics on the other hand are easily degraded if administered orally and most cannot penetrate cell membranes due to their large size, hence all currently available biologics are administrated via injection or other parenteral route of administration. This does not mean that biologics are not useful drugs, it’s just that the current technology for biologics is not as cost effective as it is for small molecules and requires special delivery methods which can be cumbersome from a patient’s perspective.

One of the key aspects of drug discovery is understanding the mechanisms of drug action at the receptor target site, which can help in the development of novel compounds for the treatment or prevention of diseases. The overall goal of this study was to evaluate the mechanisms of action and structure-activity relationships (SAR) of small molecules at their targets. Assessing SAR would help to identify compounds that have the desired mechanism such as agonists, antagonists and allosteric modulators, as well as identifying compounds that activate the desired signaling pathway with minimal interaction with off-target receptors. Information obtained from these studies would be useful in developing novel molecules with potential therapeutic value or other biotechnology applications.

Classical versus reverse pharmacology

Historically, drug development has been serendipitous in nature and it played a major role in the discovery of prototypical psychotropic drugs (Ban 2006). For example, meprobamate, chlorpromazine, and imipramine were discovered while looking for something else; e.g., preservative for penicillin in the case of meprobamate, development of new antihistamines in the case of chlorpromazine, and the development of a drug to treat schizophrenia in the case of

2 imipramine (Ban 2006). Similarly, other drugs like potassium bromide, chloral hydrate, and lithium were considered serendipitous discoveries because even though correct outcomes were achieved they were based upon false rationale (Ban 2006). Though serendipitous discoveries have led to the development of new medications, the majority of the drugs were discovered using scientific, hypothesis-driven approaches like classical pharmacology (forward pharmacology) or reverse pharmacology (target-based drug discovery, TDD). Generally, classical pharmacology relies on identification of compounds with a desirable therapeutic effect in vivo, followed by optimization of potency, selectivity and other pharmacological properties to produce a drug-like molecule. In contrast, TDD takes advantage of bioinformatics databases to identify targets

(usually a protein) with potential therapeutic value (Figure 1 shows a simplified schematic of forward and reverse pharmacology) (Ashburn and Thor 2004; Lazo 2008; Takenaka 2001).

The advantage of forward pharmacology is that the therapeutic effect of the compound is studied first (whether it is in animals, organs, tissues, etc.), hence the likelihood of achieving desired effect in humans is high with little or no understanding of the mechanism (Takenaka

2001). Another benefit of starting with an in vivo model is that information about duration of action, safety and metabolism is obtained early. Major disadvantages are that this approach is not possible without a good animal model, hence the translational benefit to humans is dependent upon the validity and strength of the animal model. Also, it is very difficult to identify a compound with a therapeutic effect without having a good understanding of its mechanism of action or its target(s). The TDD approach gained popularity after the completion of the human genome project, primarily due to the rapid cloning and synthesis of target proteins (Rask-

Andersen et al. 2011; Takenaka 2001). The advantage of TDD is that the investigator can utilize bioinformatics databases to identify potential therapeutic targets and evaluate large numbers of

3 compounds using high-throughput in vitro assays (Swinney and Anthony 2011). The disadvantage of TDD is that since the compounds are evaluated for activity at a specific target, compounds that do not act on the target are going to be excluded, though it is possible that those compounds might still have therapeutic effect via some other mechanism. Also, just because a compound has the desired effect at the target does not guarantee therapeutic value. Though the in vivo efficacy is not evaluated until the later phases of the drug discovery process, the advantage of evaluating large sets of compounds early in the study is that it is likely to increase the probability of finding several lead compounds with the desired therapeutic outcome. Though it takes approximately two years to identify a drug candidate by TDD, between the years 1999 to

2008, 56% of small molecules and 68% of biologics were developed using forward pharmacology, whereas only 40% of the drugs approved during this period utilized TDD (Sams- dodd, 2013). For this dissertation project a combination approach was utilized; receptor targets were selected based on either preliminary in vivo data or from reproducible literature reports.

This is likely to increase the probability of finding a lead compound in a time and cost-effective manner, as well as allowing the evaluation of a large set of compounds. Additionally, it allows for the systematic evaluation of the effects of chemical structural modifications on the receptor pharmacology of the compounds, which will help to identify a structural class with the potential to be a lead molecule.

Application of pharmacological principles for evaluating small molecules at receptor targets

Receptors are complex proteins with multiple functions which are influenced by the ligand they interact with (Andresen 2011; Gao and Jacobson 2013; Kenakin 2001; Kenakin

2011). A ligand is any molecule that interacts with the receptor, and when this interaction elicits

4 a biological response similar to the endogenous agonist, it is called an agonist (Figure 2 shows pharmacological classification of ligands). A full agonist produces a maximum, saturable response and this is generally defined by the endogenous agonist, for example, serotonin is a full agonist at the serotonin receptors. An agonist that produces less than maximal response compared to a full agonist (e.g. endogenous agonist) is called a partial agonist. In contrast to agonists, a ligand that physically interacts with the receptor but does not activate it, and blocks or dampens the agonist response is called an antagonist. Some receptors are constitutively active, which means they produce a biological response in the absence of a bound-agonist (Milligan

2003). A ligand that inhibits this constitutive activity and reduces the receptor’s response to below basal level is called an inverse agonist.

Every ligand interacts with the receptor in a unique way and can stabilize different receptor conformations (Kenakin 2011). Depending on how the ligand interacts with the receptor, different signaling pathways can be activated. This ability of a ligand to control which signaling pathways get activated is called functional selectivity or biased agonism (Andresen

2011; Gao and Jacobson 2013; Kenakin 2001; Kenakin 2011). Figure 3 shows an example of functional selectivity using the µ-opioid receptor (MOR) agonist oliceridine (TRV130) and its structural derivatives as an example (Chen et al. 2013). Oliceridine is a new drug for the treatment of moderate to severe acute pain and is entering phase 3 clinical trials. Oliceridine is a

MOR agonist and is meant to be a morphine replacement. Morphine is associated with the development of tolerance due to activation of the β-arrestin-2 (βarr2) pathway by the MOR

(Bohn et al. 1999; Bohn et al. 2000; Bohn et al. 2002; Koch and Hollt 2008; Raehal et al. 2005;

Roshanpour et al. 2009). At the molecular level, morphine functions as a full agonist of the Gs and βarr2 pathways (Chen et al. 2013), and since activation of the βarr2 pathway is linked to

5 morphine tolerance, then compounds that can selectively activate the Gs pathway over the βarr2 pathway should result in reduced tolerance. All three compounds shown in Figure 3 are able to activate the Gs and βarr2 pathways, hence are considered MOR agonists, but depending on the structural modification of the ligands, these ligands are either full or weak partial agonists of the

βarr2 pathway. All three ligands have approximately equal potency (EC50, ligand concentration required to achieve half maximal effect) and efficacy for the Gs pathway. Similarly, all three compounds have equal potency for the βarr2 pathway, however the efficacy changes for compound-B. Both oliceridine (methoxy modification) and compound-A are partial agonists of the βarr2 pathway, but the dimethyl modification (compound-B) caused it to become a full agonist, that is even stronger than the endogenous agonist morphine. This example demonstrates that structurally related ligands acting on the same receptor can activate different pathway depending on the ligand-receptor interaction. Since oliceridine is a weak partial agonist for the

βarr2 pathway, unlike morphine which is a full agonist, the probability of developing tolerance is likely reduced. This type of structure-activity relationship (SAR) focused on functional selectivity profiling can be crucial in drug discovery programs seeking a single functional outcome amongst many possible outcomes.

Target selectivity is a ligand’s preference for interacting with one receptor type over another (e.g. preference for serotonin receptor over the dopamine receptor), and is just as important as functional selectivity because ligands with off-target activity are likely to cause unwanted side effects. The same principles of SAR for functional selectivity can be applied for target selectivity. Figure 4 shows an example of how structural modifications can affect α- adrenergic versus β-adrenergic receptor selectivity. Both epinephrine (E) and norepinephrine

(NE) are structurally similar biogenic amines except E has a methyl group making it a secondary

6 amine, whereas NE is a primary amine. These biogenic amines act on the α-adrenergic receptors

(α-AR) and β-adrenergic receptors (β-AR), and in general they have similar potencies for both class of receptors (though there are exceptions, e.g., α1: NE > E; β2: E > > NE) (Green et al.

1992). Isoproterenol is an isopropyl analog of epinephrine and this has a large effect on α vs β selectivity, increasing and decreasing the potency for the β-AR and α-AR, respectively. The potency for isoproterenol is so far shifted to the right at the α-AR that it is analogous to having essentially no activity. Further, when the catechol group of isoproterenol is changed to a dichlorobenzene (dichloroisoproterenol), it becomes an antagonist (Glover et al. 1962a; Glover et al. 1962b). Though dichloroisoproterenol is not an example target selectivity, but rather functional activity, it is shown here to demonstrate that that the isopropyl group is controlling the selectivity of isoproterenol and the hydroxyl groups are controlling the functional activity of the ligand.

In vitro experimental approaches to studying receptor-ligand interactions

Establishing the affinity of compounds at target receptors is key to determining receptor target selectivity. The radioligand binding assay offers a valuable tool for identifying interactions between a novel ligand and a target receptor, and assessing affinities. Two essential components of the radioligand binding assay are the target receptor and the radioligand. In this post-genomic era, target receptors can be cloned and expressed in heterologous systems, and if cloned receptors are not available then they can be sourced from tissues or cell cultures as long as the expression system has been validated. If the compound of interest is available as a radioligand, then saturation isotherms can be performed under equilibrium conditions to obtained the dissociation constant (KD), which is the concentration at which half of the receptor sites are occupied at equilibrium , and is also a measure of affinity (Figure 5A). A requirement for

7 equilibrium binding is that the ligands bind reversibly and that equilibrium is attained. Another useful piece of information obtained from a saturation isotherm is the maximum receptor density

(Bmax), which is useful in determining the amount of target protein in a particular system.

Limitations of saturation isotherm-type assays is that they can require large quantities of the radioligand to achieve saturation which can be expensive, and the ligand of interest may not be available as a radioligand, as is almost always the case when studying novel compounds. An indirect approach to estimating the ligand affinity is by competition style inhibition curves (also performed under equilibrium conditions) (Figure 5B). In contrast to the saturation isotherm assay, which utilizes a fixed concentration of the receptor and an increasing concentration of the radioligand, competition inhibition curves utilizes a fixed concentration of the radioligand and an increasing concentration of the non-isotopic (unlabeled) competing ligand. In the absence of the competing ligand, the receptors will only be occupied by the radioligand. As the concentration of the competing ligand is increased, the amount of non-isotopic bound receptors should increase and there should be a corresponding decrease in specifically bound radioligand, until all radioligands are displaced from the receptor. From the inhibition curve, the IC50 value can be extracted from which the inhibition constant (Ki) of the non-isotopic ligand can be calculated using the Cheng-Prusoff equation (Cheng and Prusoff 1973). Ki is the concentration of the competing ligand at which 50% of the receptors would be occupied if no radioligand was present. The reason for converting the IC50 to the Ki is that the IC50 is influenced by the amount of radioligand used, hence if too much radioligand is used, the IC50 would be shifted to the right, and if too little is used it would be shifted to the left. The Cheng-Prusoff equation corrects for the concentration of the radioligand. Both Ki and KD are measures of affinity (KD = Ki), the difference is in the approach used to obtained the affinity (saturation isotherm vs competition

8 inhibition curves). Another useful information that can be obtained form an inhibition curve is a pseudo-Hill slope, where a slope of 1 suggests a single-site competition. See the Schild-analysis section below for other interpretations of the slope.

When selecting a radioligand for studying GPCRs by competition inhibition curves, antagonists are the preferred choice. This is because GPCRs have two affinity states: a high affinity state (RH) and a low affinity state (RL), and antagonists are generally considered to have similar affinities for the two states (Katritch et al. 2014; Park et al. 2008; Schetz 2005). Agonists can exhibit high or low affinity depending on the state of the receptor, hence if an agonist is used as the radioligand for competition experiments, the affinities obtained for the competing ligand are likely to be inaccurate and have high variability.

Estimating agonist affinity from functional assays using the operational model of agonism

An alternative to approach for measuring affinities for agonists is by a functional assay approach and extrapolating the affinity from the potency. According to the theory underpinning the operational model of agonism, as the receptor density approaches zero, the potency approaches the affinity (Black et al. 2010; Leff et al. 1985). In radioligand binding assays, only the receptor-ligand interaction is measured and this interaction is mostly independent of the receptor density, as long as the receptor density is not so high as to cause ligand depletion for the lower concentrations of the radioligand. In contrast, potencies are influenced by the receptor density, especially if the receptor reserve is high, meaning, a high receptor density as is often the case in heterologous expression systems (Chidiac 2016; Pineda et al. 1997). Receptor reserve occurs when the number of receptors in a system is in excess of that needed to produce a maximal response (Chidiac 2016; Pineda et al. 1997). In other words, when there is a receptor reserve, a maximum functional response is achieved at submaximal receptor occupancy, hence,

9 the EC50 value would be lower than the Ki value (i.e., potency > affinity). To estimate the affinity using functional assays, the receptor reserves must be depleted. One way to do this is by using an irreversible antagonist that covalently binds in the receptor’s binding pocket (e.g. phenoxybenzamine in the case of α-AR). This effectively reduces the Bmax (Figure 6) by chemical means rather than by decreasing the expression level. Using the irreversible antagonist paradigm, concentration response curves are generated in the absence and then in the presence of at least two concentrations of the irreversible antagonist. The concentration of the irreversible antagonist has to be high enough to cause a measurable decrease in Bmax, hence a decrease in efficacy, but not so high that all receptors are irreversibly bound to the antagonist and a concentration response cannot be performed. The concentration response curves for each of the conditions (presence and absence of irreversible antagonist) are then curve fitted according to the following relationship:

푛 푛 (퐸푚휏 [퐴] ) 푅 = 푛 푛 푛 (퐾퐴 + [퐴]) + 휏 [퐴]

The relationship between KA, [A50], and τ for the above function

퐾 [퐴 ] = 퐴 50 (2 + 휏푛)1/푛 − 1

According to this relationship (when slope equals 1), as the Bmax is reduced, the efficacy is reduced, hence τ is reduced. As τ approaches 0, the denominator becomes 1 ((2 + 01)1/1-1 = 1), hence the potency would equal the affinity.

R = response

KA = agonist dissociation constant (i.e., agonist affinity, KA = Ki)

Em = efficacy, maximum response

[A] = agonist concentration

[A50] = potency at corresponding τ

τ = transducer constant, a measure of efficacy; i.e., if τ = 10, then occupation of 10% of the receptors will lead to half maximal response. n = slope parameter

Estimating antagonists affinity from functional assays using Schild analysis

The Schild analysis is a useful approach for studying agonist / antagonist interactions and for measuring the antagonist potency (Kb). In order for the Schild model to be valid, the following assumptions have to be met (Arunlakshana and Schild 1959; Colquhoun 2007; Schild

1957):

1. The antagonist should have equal affinity for all binding sites

2. The antagonist should not change the confirmation of the receptor

3. The agonist and antagonist are competitive

4. The efficacy should remain unchanged (i.e. there should not be a reduction in the

efficacy)

5. Measurements are obtained at equilibrium and the interactions of all ligands are

reversible

Similar to how the operational model is performed, concentration response curves are generated in the absence and presence of increasing concentration of the competing antagonist. Figure 7 shows a representative example of how Schild analysis is performed. In the presence of the increasing concentration of the competing antagonist there should be a parallel rightward shift in the agonist concentration response, with no change in the efficacy. If there is a change in efficacy then it violates assumption 4 and Schild analysis cannot be performed. Further it also suggests

11 that the antagonist is unsurmountable, that the maximum effect of the agonist is reduced and no amount of agonist can overcome it, which means it meets the requirement of an allosteric modulator. If none of the assumptions are violated, then the concentration curves can be transformed into a Schild plot, which is a double logarithmic plot. The derivation of the Schild equation is as follows:

The following relationship should hold true if the magnitude of the agonist response (i.e., the efficacy) is the same in the presence and absence of the antagonist.

[퐴] [퐴′] = [퐴] + 퐾푑 ′ [퐵] [퐴 ] + 퐾푑 (1 + ) 퐾푏

[A] = Agonist EC50 in the absence of the competing antagonist

[A’] = Agonist EC50 in the presence of the competing antagonist

[B] = Antagonist concentration

KD = Equilibrium dissociation constant for the agonist

Kb = Equilibrium dissociation constant for the antagonist

The above equation rearranges to:

[퐴′] [퐵] − 1 = [퐴] [퐾푏]

Logarithm of both sides yields:

[퐴′] log ( − 1) = nlog[퐵] − 푙표푔퐾 [퐴] 푏 n = slope constant

From the Schild plot, the pA2 can be obtained from the x-intercept, which is the concentration of antagonist needed to shift the agonist dose response curve by a factor of 2

(Arunlakshana and Schild 1959; Kenakin 1984; Schild 1957). If the agonist/antagonist

12 interaction is perfectly competitive, then the Schild plot should have a slope of one, and when this is the case, the pA2 (antagonist potency) is equal to the pKb, which is the antagonist affinity

(Arunlakshana and Schild 1959; Kenakin 1984; Schild 1957). If the slope is not one, than the pKb estimate will not be accurate: if the slope is greater than one then the pKb value will be underestimated, and if it is less than one, the pKb will be overestimated. A slope > 1 also indicates positive cooperativity (binding of one molecule increases the receptor’s affinity for binding to the second molecule), hence it also suggests the presence of an allosteric interaction at the target receptor (Kenakin 1984; Wyllie and Chen 2007). A slope less than one could indicate negative cooperativity, removal of the agonist by uptake process or it could be due to the agonist acting on second receptor (Kenakin 1984).

Application of theory

The overall goal of this dissertation was to study the mechanisms of action of small molecules at their target receptor(s) by applying the theories described in this chapter to practice.

The principles of SAR, functional and target selectivity were utilized to identify substructural features that produce the desired biological effect. The broad implication is that the information obtained from these studies will provide insight into designing and discovering novel small molecules having the desired mechanisms of action at the target receptor of interest with minimal interactions at off-target receptors. In the following chapters, the properties of different small molecules acting on different receptor systems are explored and their molecular mechanisms of action are elucidated. This includes the evaluation of small molecules having potential arthropod deterrent properties, potential Alzheimer’s disease-modifying properties, and identification of receptor off targets of the HIV medication efavirenz. This strategy allowed for virtually all types of interactions and selectivities discussed in theoretical terms to be evaluated in practice.

Figure 1. Simplified schematic representation of forward and reverse pharmacology.

Classical pharmacology starts with identifying compounds with desirable therapeutic effect, followed by optimization to identify a drug candidate. The molecular mechanisms are elucidated last. In reverse pharmacology, a target is first identified and compounds are screened (usually in vitro) to identify a small set of lead compounds with desirable properties. They are then evaluated further in vivo for their biological effect. Classical pharmacology takes approximately

5 years to identify a drug candidate, while reverse pharmacology takes an average of 2 years

(Takenaka, 2001).

Figure 1. Simplified schematic representation of forward and reverse pharmacology.

Figure 2. Pharmacological classification of ligands. A full agonist has maximum efficacy

(100% response) which is defined by the endogenous agonist. A partial agonist produces a submaximal response at full receptor occupancy. Antagonists are ligands that interact with the receptor but have no effect on receptor activity on their own. They can however suppress the activity of the competing agonist. A ligand that reduces the constitutive activity of a receptor is called an inverse agonist.

Figure 2. Pharmacological classification of ligands.

Figure 3. Structure-activity relationships controlling functional selectivity of the MOR agonist oliceridine, a new investigational drug for the treatment of moderate to severe pain.

The core structure is shown in black (compound A, no modification) and structural modifications are shown in color. Compound B: dimethyl modification (shown in red); oliceridine: methoxy modification (shown in blue). All three compounds have equal potency and efficacy for the Gs pathway, and have equal potency for the βarr2 pathway. However, compound A, and oliceridine are partial agonists of the βarr2 pathway whereas compound-B is a full agonist and has greater efficacy than even the endogenous agonist morphine. Knowing the functional selectivity for the

βarr2 pathway is important because compounds that strongly activate this pathway lead to the development of tolerance. Data is replotted from Chen et al., 2013.

Figure 3. Structure-activity relationships controlling functional selectivity of the MOR agonist oliceridine, a new investigational drug for the treatment of moderate to severe pain.

Figure 4. Ligand structural modification that increases β-adrenergic receptor (β-AR) selectivity over the α-adrenergic receptor (α-AR). The black line shows the concentration response curve for epinephrine, the red dotted line shows the curve for norepinephrine, while the purple curve is for isoproterenol and orange curve is for dichloroisoproterenol. Isoproterenol, an isopropyl analog of epinephrine has greater selectivity for β-AR over the α-AR as shown by leftward shift at the β-AR and a rightward shift at the α-AR. Further, changing isoproterenol from a catechol to a dichlorobenzene causes it to become an antagonist, which suggests that the selectivity is controlled by the isopropyl group and the hydroxyls on the benzene ring is controlling the activity.

Figure 4. Ligand structural modification that increases β-adrenergic receptor (β-AR)

selectivity over the α-adrenergic receptor (α-AR).

Figure 5. Basic components of two common types of equilibrium radioligand binding assays. A, A saturation isotherm is generated by quantifying the amount of specifically bound radioligand in the presence of increasing concentration of the radioligand and fixed concentration of the receptor. As the radioligand concentration increases, total binding (radioligand bound to designated target and off-targets) and nonspecific binding (biding of the radioligand to non-target receptor sites-determined by blocking all target sites with a high concentration of a known ligand that interacts with the receptor) increase. Once all sites are radiolabeled, total and nonspecific binding will increase in a parallel linear fashion. By subtracting nonspecific binding from total binding, the specific binding curve is obtained and should appear asymptotic if maximum receptor occupancy is achieved. A rectangular hyperbola function is used for the curve fit. The maximum receptor density (Bmax) is obtained from the asymptote and the dissociation constant

(KD) is the concentration of the radioligand ligand at which half of Bmax is achieved. B, If the ligand of interest is not available as a radioligand, the affinity can be calculated indirectly by competition style inhibition curves. Inhibition binding curves are performed in the presence of a fixed concentration of the radioligand, a fixed concentration of the receptor and an increasing concentration of the competing ligand (the ligand of interest). The IC50 value can be extracted and the affinity (Ki) can be calculated using the Cheng-Prusoff equation.

Figure 5. Basic components of two common types of equilibrium radioligand binding

assays.

Figure 6. Estimation of agonist affinity using the operational model of agonism. When a receptor is treated with an irreversible antagonist, it forms a covalent bond with the receptor and is analogous to the receptor being essentially removed from the system because it can no longer interact with agonist. The figure shows representative concentration-response curves in the presence and absence of an irreversible antagonist (shown as a red X); the agonist is shown in green circles. As the receptor density is depleted, there is a reduction in the efficacy (Emax) (i.e., maximal response) and a rightward shift in the potency (indicated by the orange Xs). According to the operational model of agonism, as the receptor density approaches zero, the potency collapse on the affinity (pKA).

Figure 6. Estimation of agonist affinity using the operational model of agonism.

Figure 7. Estimation of antagonist potency using Schild analysis. Schild analysis is performed by measuring concentration responses for an agonist in the absence or presence of increasing concentration of the competing antagonist. There should be a progressive rightward shift in the potency, without loss of efficacy. A Schild transformation of the data should generate a linear plot if the antagonist-agonist interaction is perfectly competitive and should have a slope of one. Under these conditions, the antagonist potency can be accurately determined from the x- intercept. See text for interpretations of slopes different from one.

Figure 7. Estimation of antagonist potency using Schild analysis.

References

Andresen B. T. (2011) A pharmacological primer of biased agonism. Endocr Metab. Immune

Disord. Drug Targets. 11, 92-98.

Arunlakshana O. and Schild H. O. (1959) Some quantitative uses of drug antagonists. Br. J.

Pharmacol. Chemother. 14, 48-58.

Ashburn T. T. and Thor K. B. (2004) Drug repositioning: identifying and developing new uses

for existing drugs. Nat. Rev. Drug Discov. 3, 673-683.

Ban T. A. (2006) The role of serendipity in drug discovery. Dialogues Clin. Neurosci. 8, 335-

344.

Black J. W., Leff P., Shankley N. P. and Wood J. (2010) An operational model of

pharmacological agonism: the effect of E/[A] curve shape on agonist dissociation constant

estimation. 1985. Br. J. Pharmacol. 160 Suppl 1, S54-64.

Bohn L. M., Gainetdinov R. R., Lin F. T., Lefkowitz R. J. and Caron M. G. (2000) Mu-opioid

receptor desensitization by beta-arrestin-2 determines morphine tolerance but not

dependence. Nature. 408, 720-723.

Bohn L. M., Lefkowitz R. J. and Caron M. G. (2002) Differential mechanisms of morphine

antinociceptive tolerance revealed in (beta)arrestin-2 knock-out mice. J. Neurosci. 22,

10494-10500.

Bohn L. M., Lefkowitz R. J., Gainetdinov R. R., Peppel K., Caron M. G. and Lin F. T. (1999)

Enhanced morphine analgesia in mice lacking beta-arrestin 2. Science. 286, 2495-2498.

Chen X. T., Pitis P., Liu G., Yuan C., Gotchev D., Cowan C. L., Rominger D. H., Koblish M.,

Dewire S. M., Crombie A. L., Violin J. D. and Yamashita D. S. (2013) Structure-activity

relationships and discovery of a G protein biased mu opioid receptor ligand, [(3-

methoxythiophen-2-yl)methyl]({2-[(9R)-9-(pyridin-2-yl)-6-oxaspiro-[4.5]decan- 9-

yl]ethyl})amine (TRV130), for the treatment of acute severe pain. J. Med. Chem. 56, 8019-

8031.

Cheng Y. and Prusoff W. H. (1973) Relationship between the inhibition constant (K1) and the

concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic

reaction. Biochem. Pharmacol. 22, 3099-3108.

Chidiac P. (2016) RGS proteins destroy spare receptors: Effects of GPCR-interacting proteins

and signal deamplification on measurements of GPCR agonist potency. Methods. 92, 87-93.

Colquhoun D. (2007) Why the Schild method is better than Schild realised. Trends Pharmacol.

Sci. 28, 608-614.

Craik D. J., Fairlie D. P., Liras S. and Price D. (2013) The future of peptide-based drugs. Chem.

Biol. Drug Des. 81, 136-147.

FDA (2016) Novel Drug Approvals for 2015. 2016.

FDA (2015) Frequently Asked Questions About Therapeutic Biological Products. 2016.

Gao Z. G. and Jacobson K. A. (2013) Allosteric modulation and functional selectivity of G

protein-coupled receptors. Drug Discov. Today Technol. 10, e237-43.

Glover W. E., Greenfiled A. D. and Shanks R. G. (1962a) Effect of dichloroisoprenaline on the

peripheral vascular responses to adrenaline in man. Br. J. Pharmacol. Chemother. 19, 235-

244.

Glover W. E., Shanks R. G. and Stanford C. F. (1962b) Peripheral vascular effects of mixtures of

isopropylnoradrenaline and noradrenaline in man. Nature. 196, 999.

Green S. A., Holt B. D. and Liggett S. B. (1992) Beta 1- and beta 2-adrenergic receptors display

subtype-selective coupling to Gs. Mol. Pharmacol. 41, 889-893.

Katritch V., Fenalti G., Abola E. E., Roth B. L., Cherezov V. and Stevens R. C. (2014) Allosteric

sodium in class A GPCR signaling. Trends Biochem. Sci. 39, 233-244.

Kenakin T. (2011) Functional selectivity and biased receptor signaling. J. Pharmacol. Exp. Ther.

336, 296-302.

Kenakin T. (2001) Inverse, protean, and ligand-selective agonism: matters of receptor

conformation. FASEB J. 15, 598-611.

Kenakin T. P. (1984) The classification of drugs and drug receptors in isolated tissues.

Pharmacol. Rev. 36, 165-222.

Koch T. and Hollt V. (2008) Role of receptor internalization in opioid tolerance and dependence.

Pharmacol. Ther. 117, 199-206.

Lazo J. S. (2008) Rear-view mirrors and crystal balls: a brief reflection on drug discovery. Mol.

Interv. 8, 60-63.

Leeson P. D. and Springthorpe B. (2007) The influence of drug-like concepts on decision-

making in medicinal chemistry. Nat. Rev. Drug Discov. 6, 881-890.

Leff P., Martin G. R. and Morse J. M. (1985) Application of the operational model of agonism to

establish conditions when functional antagonism may be used to estimate agonist

dissociation constants. Br. J. Pharmacol. 85, 655-663.

Lipinski C. A. (2004) Lead- and drug-like compounds: the rule-of-five revolution. Drug Discov.

Today Technol. 1, 337-341.

Milligan G. (2003) Constitutive activity and inverse agonists of G protein-coupled receptors: a

current perspective. Mol. Pharmacol. 64, 1271-1276.

Morrow T. and Felcone L. H. (2004) Defining the difference: What Makes Biologics Unique.

Biotechnol. Healthc. 1, 24-29.

Park PS, Lodowski DT, Palczewski K. (2008) Activation of G protein-coupled receptors: beyond

two-state models and tertiary conformational changes. Annu Rev Pharmacol Toxicol 48,

107-141.

Pineda J., Ruiz-Ortega J. A. and Ugedo L. (1997) Receptor reserve and turnover of alpha-2

adrenoceptors that mediate the clonidine-induced inhibition of rat locus coeruleus neurons in

vivo. J. Pharmacol. Exp. Ther. 281, 690-698.

Raehal K. M., Walker J. K. and Bohn L. M. (2005) Morphine side effects in beta-arrestin 2

knockout mice. J. Pharmacol. Exp. Ther. 314, 1195-1201.

Rask-Andersen M., Almen M. S. and Schioth H. B. (2011) Trends in the exploitation of novel

drug targets. Nat. Rev. Drug Discov. 10, 579-590.

Roshanpour M., Ghasemi M., Riazi K., Rafiei-Tabatabaei N., Ghahremani M. H. and Dehpour

A. R. (2009) Tolerance to the anticonvulsant effect of morphine in mice: blockage by ultra-

low dose naltrexone. Epilepsy Res. 83, 261-264.

Sams-Dodd F. (2013) Is poor research the cause of the declining productivity of the

pharmaceutical industry? An industry in need of a paradigm shift. Drug Discov Today 18,

211-217.

Schild H. O. (1957) Drug antagonism and pAx. Pharmacol. Rev. 9, 242-246.

Schetz J. A. (2005) Allosteric modulation of dopamine receptors. Mini Rev. Med. Chem. 5, 555-

561.

Swinney D. C. and Anthony J. (2011) How were new medicines discovered? Nat. Rev. Drug

Discov. 10, 507-519.

Takenaka T. (2001) Classical vs reverse pharmacology in drug discovery. BJU Int. 88 Suppl 2,

7-10; discussion 49-50.

Wyllie D. J. and Chen P. E. (2007) Taking the time to study competitive antagonism. Br. J.

Pharmacol. 150, 541-551.

CHAPTER 2

EXPLORATION OF THE STRUCTURE-ACTIVITY SPACE OF CLONED Α-LIKE OCTOPAMINE RECEPTORS FROM A MARINE BARNACLE AND A TERRESTRIAL FLY

Abstract

Alpha adrenergic receptor agonists possess antifouling properties against barnacles by preventing barnacle cyprid settlement; in other words, the cyprids are unable to attach to surfaces coated with alpha adrenergic agonists. It was reported that medetomidine, one of the compounds known to inhibit cyprid settlement, functioned as a potent agonist at the cloned α-like barnacle

Balanus Improvisus octopamine receptor (BiOctR), and induced hyperactivity in cyprids. In this report functional responses of a series of α-adrenergic receptor agonists were evaluated at the

BiOctR and found that dexmedetomidine, clonidine, naphazoline, and tizanidine acted as full agonists while (-)-NE and (-)-epinephrine acted as partial agonists. Dexmedetomidine had subnanomolar potency, whereas the others had potency in the low nanomolar range. Further, all compounds were able to induce hyperactivity in barnacle cyprids, and this effect could be blocked by the antagonist atipamezole, supporting the hypothesis that the response is being mediated by the octopamine receptor (OctR). The pharmacology of OctR was also compared with a cloned α-like OctR from the terrestrial arthropod D. melanogaster (DmOctR). Although the rank order potency was similar between the two receptors, there were striking species differences in efficacy: compounds that were full agonists in barnacle were partial agonists in fly and vice versa. The antagonist profile of α-adrenergic, histaminergic and dopaminergic receptor

33 antagonists were also explored and all were found to function as antagonists in both species. In

[3H]-rauwolscine competition binding assays, the substituted benzamides metoclopramide and nemonapride, dopamine receptor antagonists, had the lowest affinities. The affinities of several antagonists and agonist were measured at the BiOctR and found that salt increases the affinity of rauwolscine and possibly other antagonists as well, and decreases the affinity for agonists.

Further, rauwolscine was shown to interact with an allosteric site at the BiOctR. Hence, the sodium sensitivity and allosteric nature of rauwolscine makes it challenging to measure affinities via radioligand binding. Though radioligand binding is a convenient method for obtaining affinities, there are other methods for measuring affinities (e.g. operational model of agonism for obtaining affinities of agonists or Schild analysis for obtaining affinities of antagonists). By evaluating the functional responses of different structural classes of ligands at OctRs from two different species, valuable insight was gained regarding the structure-activity space of OctR ligands, which will be useful in designing novel OctR selective ligands.

Introduction

The OctR was first discovered in the salivary gland of the Octopus in 1951 by Erspamer and Boretti, and since then it has been found in neuronal and non-neuronal tissues of many invertebrate species (Roeder 1999). The endogenous ligand for the OctR is the biogenic amine octopamine and is structurally related to the vertebrate adrenergic neurotransmitter norepinephrine (NE). Structurally, NE has a catechol group (benzene with two hydroxyl group at positions 1 and 2), whereas octopamine has a phenol group (benzene with a hydroxyl group at position 1) (See Table 2 in the Results section for comparison of the structures). Octopamine is functionally active only in invertebrates and has no known physiological role in vertebrates

(Roeder 1999). Though octopamine is found in vertebrates, it is present in nanomolar quantities

(1 to 6 nM in whole brain rat tissue), whereas biogenic amines like NE, serotonin and dopamine are present in micromolar quantities (2 to 8 µM in whole brain rat tissue) (Berry 2004). Hence, octopamine is considered a trace amine in vertebrates, and to date there are no reports of OctRs being expressed in vertebrates. However, octopamine has been reported to interact with the mammalian Trace amine-associated receptor 1 (TAAR1) with a potency in the 4 to 15 µM range

(Kleinau et al. 2011; Maguire and Davenport 2015). Since octopamine requires micromolar quantity to activate the TAAR1 and it is only present in nanomolar quantities in the brain, then it is likely to have minimal or no activity in vertebrates via the TAAR1 under normal physiological conditions. The synaptic concentrations may be high enough to activate the TAAR1 but there are no reports that measured the synaptic concentrations of octopamine, and the physiological role of octopamine in vertebrates, if any, has yet to be determined. Similarly, norepinephrine has no physiological role in invertebrates (Roeder 1999).

Both the octopaminergic and the adrenergic systems have homologous functions in invertebrates and vertebrates, respectively. The endogenous ligands for both systems function as a hormone and a neurotransmitter, and both are involved in the stress response, regulating processes like arousal, motivational states, startle behavior, and fight or flight response (Roeder

1999). Octopamine is considered a multipotent neurotransmitter that modulates activity of numerous tissues in the periphery, as well as modulating activity of sense organs and CNS functions. In the periphery it controls muscle function, endocrine organs, sting response, light organ, and the heart. Via the sense organs, octopamine also modulates the pheromone response, proprioception, taste, vision and olfaction. In the CNS octopamine modulates processes like learning and memory, rhythmic behaviors (e.g., flight motor activity, respiratory rhythm, and

35 circadian rhythm), desensitization, and motivation (Roeder 1999). All of these processes are mediated through the actions of the OctRs, which are all G-protein coupled receptors.

One of the first OctRs characterized was from the CNS of the cockroach Periplaneta

Americana (Nathanson and Greengard 1973). The receptor was identified by its positive coupling to adenylate cyclase and changes in intracellular cAMP, which became the preferred method for characterizing OctRs in numerous other invertebrates. In 1981 Peter Evans proposed the first insect OctR classification scheme utilizing pharmacological profiles of physiological responses to octopamine in the locust extension-tibiae-muscle (Evans 1981). This approach allowed for the identification of three different pharmacologically distinct subtypes of OctRs.

The OctRs responsible for the slowing of the myogenic rhythm in the myogenic bundle at the proximal part of the muscle were classified as Octopamine1 (Oct1) (Evans 1981). Receptors modulating neuromuscular transmission were designated as Octopamine2 (Oct2) receptors, which were further subdivided as 2A and 2B receptors, where the 2A were presynaptic and the 2B were postsynaptic. Subsequent studies showed that Oct1 mediated its effect via Gq-coupling, resulting

2+ in an increase in intracellular Ca , whereas Oct2 receptors are Gs-coupled, increasing the intracellular cAMP levels (Evans 1984a; Evans 1984b; Evans 1993; Evans and Robb 1993).

Pharmacologically chlorpromazine (D2 dopamine receptor (D2R) antagonist) and yohimbine (α2 adrenergic receptor antagonist) were more potent than metoclopramide (D2R antagonist) at the

Oct1 receptor (rank order potency: Chlorpromazine > yohimbine > metoclopramide), and vice versa at the Oct2 receptors (rank order potency: metoclopramide > chlorpromazine > yohimbine)

(Evans 1981). For the α-adrenergic receptor agonists clonidine and naphazoline, the rank order potency were clonidine > naphazoline at the Oct1 receptor and naphazoline > clonidine at the

Oct2 receptors. Subsequently, a third class of octopamine receptor (Octopamine3, Oct3) was

36 discovered in the locust brain (Roeder 1992) whose pharmacology differed from Oct1 and Oct2 present in the periphery (rank order potency: demethylchlordimeform >> chlordimeform)

(Roeder 1995). Oct3 was later reclassified as Oct2C due to its similar naphazoline-clonidine pharmacology (rank order affinity: naphazoline > clonidine) compared to the Oct2 receptors, and its coupling to Gs (Farooqui 2007; Roeder 1995). This classification system was based on whole tissue physiological responses, but then a new classification system was developed which was based on studies of the cloned Drosophila GPCRs, and is currently the mainstay classification system for classifying OctRs (Evans and Maqueira 2005; Farooqui 2007). According to the new classification scheme OctRs can be classified as α-adrenergic like (α-like OctR), β-adrenergic like (β-like OctR) or octopamine/tyramine (Oct/Tyr, or TyrR). Both α and β-like OctRs exhibit higher affinity for octopamine over tyramine (Oct > Tyr); however, the α-like OctRs strongly couple to Gq (and weakly to Gs), whereas the β-like OctR subtypes thus far are only known to couple to Gs. In contrast, Oct/Tyr receptors couple to both Gi (resulting in inhibition of adenylyl cyclase when activated) and Gq, depending on the preference for the agonist (tyramine vs octopamine; Gi-coupling when Tyr > Oct and Gq-coupling when Oct ≥ Tyr). Since OctRs are found only in invertebrates and modulate many processes, they are considered an attractive target for developing arthropod deterrents.

There are several OctR ligands with insecticidal activity with chlordimeform (CDM) being the prototype. CDM is effective against insects and acarine (arachnids such as mites and ticks) pests and causes excitation in the adult rice stemborers (Matsumura and Beeman 1976).

Numerous mechanisms were proposed but none could explain the invertebrate behavioral effects.

Subsequent studies on fireflies revealed that CDM could stimulate the firefly light organ to glow brightly, a response that is controlled by octopamine (endogenous agonist of the OctR)

(Hashemzadeh et al. 1985; Hollingworth and Murdock 1980; Nathanson and Hunnicutt 1979;

Swale et al. 2014). Further, two metabolites of CDM, demethylchlordimeform (DCDM) and didemethylchlordimeform (DDCDM) elicited similar responses, and were more potent than

CDM (Hollingworth and Murdock 1980; Nathanson 1985a; Nathanson 1985b). CDM and

DCDM also stimulates OctR mediated adenylate cyclase activity like the agonist octopamine, providing further confirmation that CDM functions as an OctR agonist (Hollingworth and

Murdock 1980; Nathanson and Hunnicutt 1981; Nathanson 1985b). Though CDM is an effective insecticide, its use was discontinued when the then International Agency for Research on Cancer reported that DCDM was a carcinogen in humans. Amitraz is another insecticide that functions as an octopamine receptor agonist (Casida and Durkin 2013; Gross et al. 2015; Roeder

1995) and is still in use today. Although amitraz is not carcinogenic like CDM, both amitraz and

CDM have α-2 adrenoceptor agonist activity which is responsible for vertebrate toxicity like mydriasis, bradycardia, local anesthetic like action, and inhibition of neuromuscular transmission

(Boyes and Moser 1988; Costa et al. 1988; Costa et al. 1991; Hsu and Kakuk 1984; Hsu et al.

1988). Some essential oils are also OctR agonists and have insecticidal activity. For example, cinnamyl alcohol, eugenol, and trans-anethole activate the Gq-signaling pathway at the cloned cockroach and drosophila OctRs (Pa oa1 and DmOctR, respectively), and have insecticidal activity against cockroaches (Enan 2001; Enan 2005).

The OctR is also used as a target for developing antifouling agents. Barnacles are one of the most problematic marine biofouling organisms, costing the shipping industry millions in added fuel use due to increased frictional drag (Monty et al. 2016; Winner 2013). Biofouling by sessile barnacles can be averted if settlement by vagile cyprids can be prevented. The α- adrenergic antagonists like phentolamine and idazoxan are able to prevent cyprid settlement at

38 high concentrations (EC50:1 to 100 µM, depending on the species of Barnacle) (Dahlstrom and

Elwing 2006; Dahlström et al. 2000; Dahms et al. 2004). Similarly, α-adrenergic agonists like medetomidine, clonidine, guanabenz, moxonidine, tetrahydrozoline also inhibit cyprid settlement but have higher potency (EC50: 1 to 100 nM) than the antagonists phentolamine and idazoxan

(Dahlstrom and Elwing 2006; Dahlström et al. 2000; Dahms et al. 2004). Further, in addition to preventing settlement, medetomidine also induced hyperactivity in cyprids (Lind et al. 2010a).

Recently, the α-like barnacle receptor (BiOctR) was cloned and medetomidine displayed agonist activity at the BiOctR (Lind et al. 2010a). One of the common features of OctR agonists across different species of arthropods is that at a low dose they induce hyperactivity and repel, and at high doses, hyperactivity can induce tremors leading to death (Enan 2001; Lind et al. 2010a;

Matsumura and Beeman 1976; Roeder 2005; Rosenberg et al. 2007).

There is considerable evidence in the literature to support the hypothesis that OctR agonists have arthropod deterrent property. However, the challenge with targeting the OctR is that the compounds currently available are not selective for OctR: they also have α2 adrenergic activity (e.g. CDM, amitraz, medetomidine, clonidine, Guanabenz, moxonidine, tetrahydrozoline) which can result in sedation and muscle relaxation in vertebrates. Having a high-throughput in vitro assay system that can be used to evaluate the functional activity of OctR ligands would be a valuable tool in discovering new OctR ligands. Since the BiOctR has been cloned and there is a strong correlation between BiOctR agonists and prevention of barnacle settlement, it is an excellent receptor system of evaluating OctR ligands and identifying compounds with potential antifouling / deterrent activity. In this study, the BiOctR was used to develop a robust high-throughput cloned receptor system for evaluate OctR ligands. The general approach was to develop a stable cell line expressing the BiOctR, and use it to develop high-

39 throughput functional and radioligand binding assays. The binding assay would be used for measuring the affinity of the OctR ligands and the functional assay would be used to determine the mechanism of action of the compounds (i.e. agonist, antagonist, allosteric modulator). Once established, future applications include evaluating novel compounds for OctR agonist activity and identifying OctR ligands with minimal affinity for the α-adrenergic receptors.

Materials and Methods:

Chemicals

The radioligand methyl (1S,15S,18S,19S,20S)-18-hydroxy-

1,3,11,12,14,15,16,17,18,19,20,21-dodecahydroyohimban-19-carboxylate ([3H]-rauwolscine,

NET722250UC, 80.5 Ci/mmol) was purchased from Perkin Elmer (Saint Louis, MO). Unless otherwise noted, all other drugs and reagents were purchased from Tocris Biosciences (via R&D

Systems, Inc. in Minneapolis, MN), Santa Cruz Biotechnology (Dallas, TX) or Sigma-Aldrich

(St. Louis, MO). All compounds were solubilized in DMSO at concentrations ranging from 10–

1000 mM and diluted at least 1:1000 v/v in the final assay solution.

Establishment of a stable cell line expressing the Barnacle Balanus Improvisus and D. melanogaster octopamine receptors (BiOctR and DmOctR, respectively)

HEK293 cells, which are devoid of adrenergic receptors were used for expressing the exogenous BiOctR and DmOctR. The BiOctR (Accession #: GU074418) expression vector was a gift form Dr. Ulrika Lind (Lind et al. 2010a) and DmOctR (Accession #: AF065443) vector was a gift from Dr. Ronald Davis (Han et al. 1998). Both expression vectors had the pcDNA3.1(+) backbone and were under the control of the CMV promoter. A day before the transfection, HEK293 cells were plated in a 150 cm2 dish at density of ~100,000 cells/dish in

Dulbecco’s modified eagles medium (DMEM) complete, which is DMEM supplemented with 50

40 units/mL streptomycin, 50 µg/mL penicillin, 1 mM sodium pyruvate, and 10% fetal bovine serum. The following day, media was replaced with DMEM complete and after 2 hours, cells were transfected with BiOctR or DmOctR expression vectors using the calcium phosphate transfection method (Invitrogen, CA). Briefly, 20 g of purified plasmid DNA was mixed with a final volume of 1 mL CaPO4/HEPES solution and the resulting precipitate was added dropwise to a 150 cm2 dish containing HEK293 cells. The following day, media was removed and replaced with fresh media containing 2 mg/mL geneticin (G418). Cells were grown under constant G418 selection and individual clones were selected once the colonies start to appear.

The colonies were evaluated for receptor expression by measuring octopamine mediated change in intracellular calcium. The clone that produced the strongest response was used for subsequent functional and radioligand binding experiments.

Profiling of BiOctR by radioligand binding

HEK293 cells stably expressing BiOctR were harvested from culture flask in Ca2+ and

Mg2+ free Dulbecco’s Phosphate Buffered Saline (Fisher Scientific 55-031-PB) supplemented with 5 mM EDTA (Sigma-Aldrich E6511). The cell pellet was lysed in ice-cold lysis buffer (5

o mM Tris-HCl, 5 mM MgCl2, pH 7.4 at 4 C) and homogenized using a dounce homogenizer (8 strokes). A membrane pellet was obtained by centrifugation at 35,000 x g for 60 min. The membrane pellet was re-suspended in ice cold binding buffer (50 mM Tris-HCl, pH 7.4 at 4oC) and re-centrifuged at 35,000 x g for 30 min. The final membrane pellet was re-suspended in binding buffer (50 mM Tris-HCl, pH 7.4 at 25oC).

All binding studies were performed in 1 mL volume and each assay tube contained 50 L of the non-isotopic ligand or vehicle (DMSO), 100 L of membrane suspension, 100 L of [3H]- rauwolscine, and 750 L of binding buffer. To optimize the binding reaction conditions, the

41 following conditions were tested: 50 mM Tris-HCl, pH 7.5 at 25oC, supplemented with either 15,

50, 120, 240, 360, or 480 mM NaCl. We had also tested 50 mM Tris-HCl, pH 7.4 at 25oC supplemented with 480 mM NaCl with either 10.5 mM KCl or 10.5 mM KCl + MgCl2.

Additionally, we tested 35% coral life (cat# CD-76412, Doctors Foster and Smith, Rhinelander,

WI), which is a scientific grade salt preparation that replicates natural sea water chemistry, and

50 mM potassium phosphate buffer at pH 7.4. The optimal binding reaction condition, which was determined to be 50 mM Tris-HCl, 240 mM NaCl, pH 7.4 at 25oC, was used for all subsequent experiments.

The expression level of the BiOctR in individual clone was determine by saturation isotherm binding using the rapid filtration technique as described previously (Kortagere et al.

2004). Eight concentrations of [3H]-rauwolscine ranging from 0.0625 to 8 nM were tested. Non- specific binding was defined by 10 M atipamezole. The binding reaction was incubated at room temperature for 90 minute to reach equilibrium, followed by rapid filtration through a GF/C filter pretreated with 0.3% polyethyleneimine, and three rapid washes with 3 mL of ice-cold wash buffer (10 mM Tris, pH 7.4, supplemented with 145 mM NaCl at 4°C). Dried filters were cut into individual scintillation vials, filled with 3.5 ml of scintillation fluid, mixed, and the radioactivity bound to the filters was quantified via scintillation spectroscopy. Membrane protein concentrations varied from 0.02-0.06 mg/mL. To determine the affinity (Ki) of OctR ligands, competition binding was utilized, where the concentration of the test ligand was increased and the concentration of the radioligand was held constant at 2 nM. Equilibrium inhibition constants (Ki), representing binding affinities, were calculated from IC50 values using the Cheng-Prusoff equation: Ki = IC50/(1 + [ligand]/KD). Concentration-response curves were

42 fitted with a four parameter logistics equation that included a variable slope using a 95% confidence interval for all curve-fits using Graphpad Prism version 4.0.

Assessment of sodium sensitivity of BiOctR via sodium replacement

To assess if the sodium sensitivity exhibited by [3H]-rauwolscine is due sodium and not charge, sodium in the binding buffer (50 mM Tris-HCl, pH 7.4 at 25oC, 240 mM NaCl) was replaced with 240 mM N-Methyl-D-glucamine (NMDG) (M-2004, Sigma Aldrich, St. Louis,

MO), which is a commonly used sodium substitute. To determine if the sodium site is accessible from the intracellular or extracellular site, whole cell binding was performed. The whole cell pellet was re-suspended in a modified HBSS buffer containing either NaCl or NMDG, and these intact cells were used for radioligand binding as described previously, except, HBSS with either

NaCl or NMDG was used as the binding buffer instead of the Tris-HCl buffer. The composition of the HBSS buffer was same as the commercially available Gibco® HBSS (cat# 14065,

ThermoFisher Scientific, Waltham, MA) except, sodium phosphate dibasic was replaced with potassium phosphate dibasic and supplemented with either 140 mM NaCl or 140 mM NMDG.

Protein Assay

Membrane protein concentrations were measured by bicinchoninic acid assay (Pierce, IL) according to the manufacturer’s instructions. Protein standard curves were constructed using purified bovine serum albumin.

Intracellular Calcium assay: measurement of Gq-coupled response

HEK293 cells stably expressing the BiOctR receptors were seeded into Poly-L-Lysine- coated, clear-bottomed, black-walled 96 well plates at a density of 120,000 cells per well. Cells were grown in Dulbecco’s modified eagles medium (DMEM) complete which is DMEM supplemented with 50 units/mL streptomycin, 50 µg/mL penicillin, 1 mM sodium pyruvate, 0.1

43 mg/mL G418 to maintain selection, and 10% fetal bovine serum. The following day, media was removed and cells were loaded for three hours in the dark with the FLIPR Calcium-6 QF dye

(Molecular Devices, Sunnyvale CA) dissolved in Hank’s buffered saline supplemented with 20 mM HEPES pH = 7.4. In cases where drug pretreatments were needed (as in the case of antagonist blocking an agonist response), the pretreatment drug or vehicle was added to the loading dye solution. Following dye loading and any pretreatments, rapid fluorescent signals in response to changes in intracellular calcium were measured at 485 nm excitation, 525 nm emission with a cutoff filter set at 515 nm using the Flexstation 3 (Molecular Devices,

Sunnyvale, CA). The baseline calcium signal was measured for 120 seconds followed by injection of test drug and reading of any change in the calcium signal for an additional 780 seconds. At the termination of the experiment all signals were baseline subtracted and change in intracellular calcium was quantified as the area under the curve, measured by integration (Prism version 4.0). All data were normalized to maximal functional response defined as the asymptote of the sigmoidal curve.

Operational model of agonism and Schild-shift analysis

HEK293 cells stably expressing the BiOctR receptors were plated in clear-bottomed, black-walled 96 well plates as described above. On the day of the experiment media was removed and cells were treated with increasing concentrations of the irreversible antagonist phenoxybenzamine (for operational model) or the antagonist rauwolscine (for Schild shift) prepared in 20 mM HEPES Hank’s buffer, pH 7.4. Immediately after adding the antagonists,

FLIPR Calcium-6 dye was added and incubated at 37oC for three hours in the dark. Following dye loading, fluorescent signals were measured for a total of 900 seconds. The baseline signal was measured for the first 120 seconds, followed by stimulation with either dexmedetomidine or

44 octopamine. For each concentration of the antagonist, a full concentration response curve of the agonist was generated. At the termination of the experiment, all signals were baseline subtracted and the areas under the curves were calculated by integration (Prism version 4.0). All data sets were normalized to the maximum response obtained in the absence of the antagonists. For calculating the dissociation constant Ka for the agonists dexmedetomidine and octopamine, phenoxybenzamine-agonist concentration responses were curve fitted using the operational model-depletion equation in Prism 4.0. For rauwolscine Schild-shift analysis, rauwolscine- dexmedetomidine concentration response curves were fitted with four parameter logistic equation that included a variable slope. If the results met all assumptions of Schild analysis (i.e., the antagonist is competitive and does not result in result in reduction of efficacy), the data would be transformed into a linear Schild plot, from which the antagonist potency and the slope obtained.

Cyprid motility assay

Cyprids of B. amphitrite were prepared and collected at Duke Marine Laboratory in

Beaufort, NC (Berntsson et al., 2000). Cyprid motility was assessed by immobilizing the cyprids in an agarose matrix as described by Lind et al., 2010 with following modifications. Briefly, 2% ultra-low gelling agarose (A-2576, Sigma Aldrich, St. Louis, MO) was prepared using 35% coral life (cat# CD-76412, Doctors Foster and Smith, Rhinelander, WI), which is a scientific grade salt preparation that replicates natural sea water chemistry. The agarose solution was allowed to cool room temperature just above its gelling temperature by incubating in a 28oC water bath. Once the agarose had cooled, DMSO vehicle or test compounds (dexmedetomidine, levomedetomidine, clonidine, naphazoline, tizanidine, octopamine) or test compounds + 50 µM atipamezole were added to the agarose and mixed. In a 96 well plate, 3 to 5 cyprids were added per well and

45 immobilized using the agarose mixtures containing vehicle or test compounds. The agarose was allowed to solidify at room temperature for an hour. Then the cyprids were viewed under a stereoscope at 3x magnification and individual cyprid movements were quantified as number of leg kicks per minute. Any movement of the leg was considered a kick; movements of the antennas were not counted. Each experiment was repeated 2-3 times and one-way ANOVA followed by Dunnett’s multiple comparison test was performed to assess significance at P <

0.05.

Results

The barnacle Balanus Improvisus octopamine receptor (BiOctR) was first cloned by Lind et al., 2010 and was classified as an α-like receptor because it exhibits higher potency for octopamine over tyramine and because it is Gq-coupled (Lind et al. 2010a). It was also demonstrated that medetomidine, a potent and selective α2-adrenergic agonist (Virtanen 1989), functioned as a full and potent agonist at the BiOctR and induced hyperactivity in barnacle cyprids (Lind et al. 2010a). However, the BiOctR was never fully characterized by pharmacological methods. In this report a stable cell line expressing the BiOctR was generated and thoroughly evaluated the receptor pharmacology to demonstrate that is it a robust model for evaluating OctR ligands. Further, additional compounds with BiOctR activity in vitro and in vivo were identified. The pharmacology of BiOctR was also compared to another α-like octopamine receptor from D. melanogaster called the DmOctR to further demonstrate that the OctR responses to the compound are species independent.

To quantify the expression level of BiOctR in the selected clone, [3H]-rauwolscine, a

3 selective α2-adrenergic receptor antagonist, was utilized as the radioligand. [ H]-rauwolscine was

o able to bind to the cloned human α2C-adrenergic receptor using 50 mM Tris-HCl, pH 7.4 at 25 C

46 as the binding conditions, but was unable to bind to BiOctR under the same conditions (Figure

1A). However, in the presence of increasing concentration Na+, [3H]-rauwolscine was able to bind to the BiOctR (Figure 1A). At 15 mM Na+, which is an approximation of the intracellular sodium concentration in vertebrates (Keenan and Niedergerke, 1967; Madelin et al. 2014), there was a significant increase in the specifically bound radioligand (P < 0.05, ANOVA followed by

Bonferroni’s multiple comparison test). Further increases in sodium concentration resulted in a concentration-dependent increase in specific binding which plateaued at concentrations ≥ 240 mM. Based on this finding, all subsequent experiments were performed using 50 mM Tris-HCl, pH 7.4 at 25oC, supplemented with 240 mM NaCl. To demonstrate that the increase in specific binding is due to sodium and that it is not a charge effect, a sodium substitute, N-methyl-D- glucamine (NMDG) was used. Replacing sodium with NMDG did not increase specifically bound [3H]-rauwolscine indicating that sodium is necessary for high affinity binding of [3H]- rauwolscine and that this was not a simple charge effect (Figure 1B). Using the optimized binding conditions, a saturation isotherm was performed to determine the Bmax and KD; the Bmax was determined to be 7.0 ± 2.7 pmol/mg membrane protein and the KD was determined to be

4.56 ± 0.54 nM (Figure 1C). Figure 1D also shows that there is no significant binding of the radioligand to membranes isolated from the untransfected HEK293 cells, hence the increase in specific binding is due the presence of the transfected BiOctR rather than to some non-specific signal or receptor endogenous to HEK293 cells. To determine if the sodium binding site is accessible from the intracellular or the extracellular side, whole cell radioligand binding assay was performed. The whole cell binding was performed using a modified HBSS buffer that contained either 140 mM NaCl or 140 mM NNDG. The anticipated outcomes were: if the site was accessible extracellularly, then replacing Na+ with NMDG should result in significantly

47 reduced ligand/receptor interaction, and if the site is accessible intracellularly, then the extracellular sodium concentration should have no effect on ligand/receptor interaction. The whole cell binding data demonstrated that there was no significant difference in specific binding when the extracellular sodium was replaced with NMDG, indicating that the sodium site is accessible only from the intracellular side (Figure 1E). To determine if salt is influencing the KD

3 or the Bmax, saturation isotherms were performed with [ H]-rauwolscine in the presence of high salt (240 mM NaCl) or the lowest concentration of salt that still gave some specific binding (15 mM NaCl). Scatchard plot transformation of the data revealed that increasing the Na+ concentration increased the slope, which suggests an increase in the affinity, while the x-

+ intercept remained unchanged, which suggests that Na had no effect on the Bmax (Figure 1F).

Though BiOctR was originally classified as an α-like OctR based primarily on its Gq- coupling and having a higher potency for octopamine over tyramine (Lind et al., 2010), little is known about its pharmacological properties. Based on a thorough literature survey of insect

OctR pharmacology, a pharmacological criteria were developed for classifying α-like OctRs based on their potencies for agonists and antagonists. For agonists, the expected rank order affinities and potencies should be octopamine > tyramine and clonidine > naphazoline, and for antagonists, chlorpromazine > yohimbine > metoclopramide (Evans 1981; Evans and Maqueira

2005). To test if these criteria hold true for BiOctR, the rank order affinities of several antagonists and agonists know to have high-affinity interactions with other GPCRs in mammals were evaluated. For these competition assays, high sodium condition (240 mM NaCl) was employed and [3H]-rauwolscine was used as the radioligand. The antagonists included three known dopamine D2 receptor antagonists: chlorpromazine, metoclopramide and nemonapride; three α-adrenergic antagonists: yohimbine, Rs-76648, and Rx-821002; and an H1 histamine

48 antagonists, promethazine. The agonists that were tested included adrenergic receptor agonists:

Dexmedetomidine, tizanidine, naphazoline, clonidine, and levomedetomidine. In addition, the affinities of OctR agonists octopamine and tyramine were also measured.

The BiOctR had low nanomolar affinity for the adrenergic antagonists ranging from 0.6 to 17 nM. Based on the affinities, the rank order was determined to be Rs-79948 > Rx821002 > atipamezole ≈ rauwolscine > yohimbine. The dopamine D2 receptor and Histamine H1 receptor antagonists had affinity for BiOctR ranging from 3.0 nM to 1700 nM, and the rank order was determined to be chlorpromazine > promethazine > nemonapride > Metoclopramide. In order to evaluate the pseudo Hill slope parameter, all the slopes were treated as a variable (instead of a fixed value = -1.0) for the purpose of curve fitting. The pseudo Hill slopes ranged from -0.74 to -

1.7, and for all compounds the pseudo Hill slopes did not deviate significantly from unity (P >

0.001, t-test). Hence a pseudo Hill slope equal to unity is the preferred model for all the antagonists tested. Based on the pharmacological criteria for classifying an OctR as α-like (see above paragraph), the rank order affinities of chlorpromazine, yohimbine, and metoclopramide at the BiOctR did meet the criteria. Hence, the BiOctR does fit the α-like classification for insects and may be a reliable pharmacological method classifying marine OctRs.

For the agonists, the affinities for two biogenic amines: octopamine and tyramine, three imidazolines: clonidine, naphazoline and tizanidine, and two imidazoles: dexmedetomidine (+ enantiomer of medetomidine), levomedetomidine (- enantiomer of medetomidine) were measured. Octopamine and tyramine had identical affinities, 43.4 ± 3.3 µM and 44.8 ± 11.5 µM, respectively (Figure 2B, Table 2). The imidazoles and imidazolines had affinities ranging from low to high nanomolar (8.2 to 540 nM), with the highest for dexmedetomidine and lowest for tizanidine (Figure 2B, Table 2). The rank order affinity for the agonists was determined to be

49 dexmedetomidine > naphazoline > clonidine ≈ levomedetomidine >> octopamine ≈ tyramine.

The pseudo Hill slopes ranged from -0.87 to -1.4 and none of them were significantly different from unity (P > 0.001, t-test), hence a fixed slope equal to unity was the preferred model. In the case of the agonists, the rank order affinities were not in agreement with the insect classification described above; octopamine was expected to have greater affinity than tyramine but both had similar affinity. Similarly, clonidine should have greater affinity than naphazoline according the hypothesized criteria, but they were reversed.

Once it was established that all the compounds tested in Figure 2 interacted with the

BiOctR, their functional responses were assessed by measuring change in intracellular calcium as a measure of Gq activation. Calcium signaling was measured because BiOctR was reported to be Gq-coupled GPCR (Lind et al. 2010b). At equimolar concentration (1 µM), octopamine, (-)- epinephrine, (-)-norepinephrine, tyramine, clonidine, naphazoline and tizanidine were each able to activate Gq signaling, and their responses could be antagonized by atipamezole (30 µM)

(Figure 3A, P < 0.05, one-way ANOVA, Bonferroni’s post-hoc). At 1 µM other biogenic amines: dopamine, histamine, (+)-norepinephrine ((+)-NE) and serotonin were unable to induce a measurable Gq response (data not shown). To completely rule out the possibility of very weak agonism, a higher concentration (100 µM) of these ligands was tested, and only (+)-NE induced a weak but significant Gq response which was reversed by atipamezole (Figure 3A, P < 0.05, one-way ANOVA, Bonferroni’s post-hoc). To determine the functional effects of compounds that act as antagonists at dopamine, α-adrenergic, and histamine receptors in mammals, each was first tested for possible agonist activity at BiOctR at the highest possible concentration that could be tested in our assay system. At their solubility limit, none of these mammalian biogenic amine receptor antagonists activated the BiOctR. To determine if these compounds might have other

50 actions, their ability to modify an agonist response was evaluated. The mammalian GPCR antagonists chlorpromazine, rauwolscine, Rs-79948, Rx-821002 and yohimbine (all tested at 30

µM) completely reversed the agonist response induced by 1 µM octopamine. Nemonapride at 30

µM reduced the agonist response by 66% (Figure 3B) and metoclopramide had no inhibitory effect (data not shown). However, when tested at its maximum solubility limit under our assay condition (100 µM), metoclopramide was able to reduce the octopamine agonist response by

45% (Figure 3B). In all cases, inhibition of the octopamine functional responses was statistically significant (P < 0.05 by one-way ANOVA followed by Bonferroni’s post-hoc).

Dexmedetomidine was also tested as part of the single point functional evaluation shown in figure 3A, and it did induce a robust calcium response at 1 µM. However, the dexmedetomidine response could not be reversed by 30 µM atipamezole (data not shown). Thus, a concentration-response curve was generated for dexmedetomidine in the absence and presence of 30 µM atipamezole (Figure 3C). Dexmedetomidine had a subnanomolar potency at the

BiOctR (EC50 = 0.087 ± 0.023 nM) and in the presence of 30 µM atipamezole the potency was decreased by more than 2000-fold (EC50 in presence of 30 µM atipamezole = 201.5 ± 18.1 nM).

Since at 1 µM the dexmedetomidine response is fully saturated, 30 µM atipamezole could not overcome it, hence the reversal could not be detected in the single point evaluation. In light of the finding that in comparison to (-)-NE, (+)-NE functioned as a very weak agonist at the

BiOctR, the functional response of levomedetomidine was compared to dexmedetomidine. While levomedetomidine had low nanomolar potency (EC50 = 8.3 ± 2.2 nM) for BiOctR, its potency was still 95-fold lower than the dextrorotary isomer. Atipamezole decreased the potency of levomedetomidine but a complete concentration response could not be generated due to solubility restriction of the compound. The pseudo Hill slopes for dexmedetomidine and

51 levomedetomidine were 0.9 ± 0.2 and 1.0 ± 0.2, respectively, and were not different from unity making the fixed slope model the preferred model.

To compare the efficacy of the different agonists to the endogenous full agonist octopamine, BiOctR was stimulated at saturating concentrations of the ligands. Under saturating conditions clonidine, dexmedetomidine, naphazoline, and tizanidine induced responses comparable to the full agonist octopamine and were not statistically different from one another.

Levomedetomidine, (-)-epinephrine, (-)-NE and tyramine all induced a response that was statistically lower than the maximum response generated by octopamine (Figure 3D, * P < 0.05, one-way ANOVA, Bonferroni’s post-hoc) indicating they are partial agonists.

To have a better understanding of the structure-activity space of BiOctR agonists, potencies of several agonists from different structural class (imidazolines: clonidine, naphazoline and tizanidine, and biogenic amines: octopamine, tyramine, (-)-NE, (-)-Epi) were measured. All three imidazolines had low nanomolar potencies ranging from 1.2 to 3.1 nM, and the rank order was naphazoline ≈ tizanidine > clonidine (Figure 4A, Table 2). The pseudo Hill slopes ranged from 0.65 to 0.80 and were not significantly different from unity (P > 0.001). The biogenic amines (Figure 4B) had potencies ranging from 28 to 862 nM with octopamine having the highest potency and (-)-NE the lowest. The rank order was octopamine > tyramine ≈ (-)-Epi > (-

)-NE. The pseudo Hill slopes for the biogenic amines ranged from 0.55 to 1.51, and only the slope for tyramine differed from unity. Since octopamine had higher potency than tyramine, this fits the insect classification system for an α-like OctR. However, the potencies for clonidine and naphazoline did not match the classification criteria, which suggests that these compounds are not a suitable for classifying OctRs. Hence, although the affinities for the agonist do not match

52 the classification criteria, the potencies for octopamine and tyramine do, and can be used for classifying marine OctRs.

The α2-adrenoceptor agonists like medetomidine, clonidine and tetrahydrozoline have been shown to be capable of inhibiting the cyprid settling process of B. improvisus and B. amphitrite (Dahlstrom and Elwing 2006; Dahlström et al. 2000), and medetomidine was shown to function as an agonist of the cloned BiOctR and induce hyperactivity responses in cyprids

(Lind et al. 2010a). Since dexmedetomidine, levomedetomidine, clonidine, naphazoline and tizanidine had high potencies for the BiOctR, the ability of these compounds to induce hyperactivity in barnacle cyprids was tested (Figure 5). Dexmedetomidine was able to induce hyperactivity in a concentration dependent manner, and the response to 1 µM dexmedetomidine was statistically significant compared to vehicle (P < 0.05, one-way ANOVA, Bonferroni’s post- hoc). The kicking response induced by dexmedetomidine (1 µM) was reversed by the antagonist atipamezole (30 µM). Treatment with levomedetomidine resulted in an increase in the number of cyprid kicks but the response did not reach statistical significance. Clonidine, naphazoline and tizanidine also induced significant hyperactivity in cyprids, and the response could be reversed by atipamezole, but were not as efficacious as dexmedetomidine.

Pharmacology of the D. melanogaster α-like octopamine receptor (DmOctR)

To determine if the pharmacology of the BiOctR is species independent, the same compounds were tested at the drosophila α-like octopamine receptor (DmOctR). Only the functional activity of these compounds was assessed because no specific binding could be detected for [3H]-rauwolscine. In addition to octopamine, the receptor could be activated by two other biogenic amines (-)-epinephrine and (-)-NE, and this activation could be reversed by the antagonist atipamezole. (+)-NE also exhibited weak agonist activity but was not statistically

53 significant. The receptor could also be activated by dexmedetomidine, clonidine, naphazoline, and tizanidine and their effects were also reversed by atipamezole (Figure 6A). In contrast to

BiOctR, levomedetomidine had no agonist activity at the highest concentration that could be tested under the assay conditions. Similar to the BiOctR, the antagonists chlorpromazine, metoclopramide, nemonapride, rauwolscine, promethazine, yohimbine, Rs-79948 and Rx-

821002 had no agonist activity at the highest concentration that could be tested, but they were able to antagonize the octopamine response. Metoclopramide and nemonapride only partially inhibited the octopamine response (Figure 6B). A full concentration response was performed for octopamine, clonidine, dexmedetomidine, naphazoline, tizanidine, (-)-epinephrine and (-)-NE to obtain the rank order potencies (Figure 6C). The rank order potency for the biogenic amines was: octopamine > (-)-epinephrine > (-)-NE, and for the other compounds the rank order was dexmedetomidine > Naphazoline, > clonidine > tizanidine. The pseudo Hill slopes ranged from

0.6 to 2.1, where clonidine had the steepest slope and octopamine, (-)-NE and (-)-epinephrine had the shallowest slopes. However, the slopes were not significantly different unity for any of the agonists tested (P > 0.001, t-test). Further, when the efficacy for all the agonists was compared, both octopamine and dexmedetomidine had similar efficacies, whereas clonidine, naphazoline, and tizanidine had lower efficacies, and (-)-Epi and (+)-Epi had higher efficacies than octopamine (P < 0.05, one-way ANOVA, Bonferroni’s post-hoc). In contrast, (-)-NE and (-

)-epinephrine had higher efficacies than octopamine, and were statistically different from octopamine (Figure 6D).

Effect of receptor reserve on agonist potency

For all the agonists tested at the BiOctR, when the potencies were compared to the affinities, the potencies were 12 to 450 fold higher than the affinities. To test if receptor reserve

54 might be responsible for the large differences between the potencies and affinities, the operational model of agonism was employed. According to this model, as the Bmax (receptor density) approaches zero, the potency should collapse onto the dissociation constant (Ka), which is same as the binding affinity Ki (Black et al. 2010; Leff et al. 1985). To deplete the receptor density, the irreversible antagonist phenoxybenzamine was used, and concentration response curves were generated for each of the phenoxybenzamine concentrations tested using the agonists dexmedetomidine and octopamine (Figure 7A, B). The data points were curve fitted using the operational model and the Ka was estimated. The Ka for dexmedetomidine was 0.3 ±

0.1 nM and 588 ± 226 nM for octopamine. The calculated Ka values for dexmedetomidine and octopamine were 27 and 73-fold higher than the Ki values, respectively, while the potencies for dexmedetomidine and octopamine were 3 and 20-fold higher than the Ka. This suggests that receptor reserve is not the only factor responsible for the large differences between potencies and the affinities for these agonists. To determine if rauwolscine competes for the same binding site as the agonist dexmedetomidine, a Schild shift analysis was performed; dexmedetomidine concentration curves were generated in the presence of increasing concentration of rauwolscine

(Figure 7C). Presence of rauwolscine caused a reduction in the efficacy and the potency which suggests that rauwolscine and dexmedetomidine are not competing for the same binding site rather rauwolscine is acting as a non-competitive antagonist. Since rauwolscine caused a reduction in the efficacy, it violated the Schild-shift assumptions, hence the data cannot be transformed in to a Schild plot and the antagonist potency cannot be obtained for rauwolscine using this method.

Discussion

A number of new findings are reported in this study. First, the radioligand binding studies clearly demonstrated that the BiOctR exhibits sodium sensitivity for [3H]-rauwolscine and that sodium increases the receptor’s affinity for this antagonists while having no effect on receptor density (Bmax). Second, the receptor reserve only had a small effect on the potency which suggests that the receptor also exhibits sodium sensitivity for agonists, but in contrast to rauwolscine, sodium decreases the affinity for agonists. Third, rauwolscine binds to a different binding site than the agonist dexmedetomidine. It was also demonstrated that both the BiOctR and the DmOctR have a preference for the (+) enantiomer of medetomidine and (-) enantiomer of norepinephrine. Further, all imidazoles tested here functioned as full agonists and (-)- epinephrine and (-)-NE functioned as strong partial agonists at the BiOctR whereas at the

DmOctR, clonidine, naphazoline, and tizanidine were partial agonists and (-)-epinephrine and (-

)-NE functioned as full agonists. We also show that all the imidazoles and octopamine are able to induce hyperactivity in barnacle cyprids providing further support for the hypothesis that OctR agonists are likely to have deterrent properties.

The OctR is a unique biogenic amine receptor because it is only found in invertebrates, which makes it an attractive target for developing arthropod deterrents. However, most of the

OctR compounds have mammalian off-target activity at the α-adrenergic receptor (see

Introduction for more detail). Based on a thorough literature survey, only one compound was found (the NC-5 compound) that had moderate selectivity (32-fold) for an arthropod OctR over mammalian α-adrenoceptors (Nathanson 1985a; Nathanson 1985b; Roeder 1995). There is only one published report suggesting that NC-5 has arthropod deterrence property, but no supporting data was shown (Nathanson 1985a); it has primarily been used as an experimental tool. The

56 cloning of the barnacle OctR helped to confirm that the settlement inhibition activity of medetomidine is likely due to activation of the OctR, resulting in hyperactivity, thus preventing settlement (Dahlstrom and Elwing 2006; Dahlström et al. 2000; Lind et al. 2010a). It was expected that the cloned BiOctR could be used to evaluate the affinities compounds at the OctR via radioligand binding. However, when binding studies were performed using [3H]-rauwolscine as the radioligand, no specific binding could be detected. Since the exoplasmic and hemolymph sodium concentrations in the squid (a marine invertebrate) giant axon are 50 mM and 440 mM, respectively (Kuffler and Nicholls 1976), and assuming that the concentrations are similar in barnacles, then the ideal binding conditions might require sodium. Presence of sodium resulted in an increase in specific binding in a concentration dependent manner. Significant specific binding could be detected at sodium concentration as low as 15 mM (which is approximately the concentration inside mammalian cells). Results of the whole cell binding demonstrated that the sodium site is accessible from the intracellular side. This is important to know because if the site was accessible from the extracellular side then the receptor would always be in a low affinity state and would require a very high concentration of the agonist to elicit any response. The physiological relevance of the site being accessible intracellularly could be that upon depolarization of the cell following a stimulation, the increase in Na+ would switch the receptor to the low affinity state and prevent hyperactivation of the cell (Schetz 2005).

Sodium is a well-known allosteric modulator of GPCRs and has been shown to reduce the receptor’s affinity for agonists by switching the receptor to a low affinity state (Katritch et al.

2014; Schetz 2005). Since antagonists are generally considered to be sodium-insensitive, with a few exceptions, the initial assumption was that rauwolscine would be sodium-insensitive too because it is an antagonist. Although it is unusual for antagonists to display sodium sensitivity, it

57 is not unprecedented. For example, the affinity of the substituted benzamide antagonists for the

+ dopamine receptor is increased in the presence of Na , while having no effect on Bmax (Malmberg et al. 1993; Neve 1991). In contrast, sodium increases the Bmax of raclopride (a substituted benzamide) at the D2L receptor while having no effect on the KD (Schetz et al., 1999). To determine if sodium is affecting the KD or the Bmax of rauwolscine, saturation isotherms were performed in the presence of high and low sodium, as no specific binding could be observed in the absence of salt. At a high sodium concentration (240 mM) the affinity was approximately 4- fold greater than at a low sodium concentration (15 mM) with no change in Bmax. This indicates that like the substituted benzamides acting on mammalian D2 dopamine receptors, sodium only affects rauwolscine’s affinity, but not its receptor density. We also measured the affinities for several agonists and found that aside from dexmedetomidine, all compounds tested had affinities in the high nanomolar to high micromolar range. However, the potencies were in the sub to high nanomolar range, hence the potencies were 12 to more than 1000-fold higher than the affinities.

Since an overexpression system was used, the maximum functional response can be achieved at submaximal receptor occupancy, where the unbound receptors are called spare receptors or receptor reserve. Hence, the potency can be higher than the affinity when there is an excess of receptor reserve. To test if receptor reserve was responsible for the large difference between potency and affinity, the receptor density was depleted using the irreversible antagonist phenoxybenzamine, and theoretically as the receptor density approaches zero, the potency should collapse onto the affinity. Using this approach, the Ka (agonist dissociation constant) was calculated for dexmedetomidine and octopamine. The potency was only three fold higher than the Ka for dexmedetomidine indicating that receptor reserve has a very small effect on the potency. However, for octopamine, the potency was 20 fold higher than the Ka, suggesting

58 receptor reserve has a large effect on the potency. It is unclear why receptor reserve would have a small effect for one ligand and a large effect for the other, but this could be due to the affinities of the ligand. Since dexmedetomidine had higher affinity for BiOctR than octopamine, it is possible that compounds with high affinity are less affected by spare receptors, whereas spare receptors might have a large effect on low affinity compounds. Further, the Ka for dexmedetomidine and octopamine were 27 and 73 fold higher than the Ki, which suggests that sodium is decreasing the affinity for agonists. It would then be expected that when binding inhibition curves are generated for agonists in the presence of low sodium concentration then the affinity should increase, but this was not the case; lowering the sodium concentration had no effect on the affinity (data not shown). We had also tested a different structural class of radioligand ([3H]-Rx-821002), and Na+ also enhanced its binding like it did for rauwolscine (data not shown). However, unlike [3H]-rauwolscine where we could not get any specific binding in the absence of sodium, we could get measurable binding for [3H]-Rx-821002 in the absence of sodium. This is perhaps not surprising since Rx-821002 had approximately 3 fold higher affinity for BiOctR than rauwolscine (Table 1). When we performed binding inhibition curves for dexmedetomidine using [3H]-Rx-821002 in the presence or absence of Na+, the results were similar to those obtained using [3H]-rauwolscine (in the presence of high and low Na+). Since the

+ Ki of the agonists did not increase even in the absence of Na , it is hypothesized that rauwolscine may be interacting with an allosteric site at the BiOctR. To test this hypothesis, Schild-shift type analysis was performed using rauwolscine and dexmedetomidine. Similar experiment was performed using octopamine as the agonist (data not shown) and the result was the same as dexmedetomidine: reduction in the efficacy and a rightward shift in the potency. This confirmed that rauwolscine interacts with the BiOctR at an allosteric site. Taken together, the data suggests

59 that it is not possible to obtain the true affinity for agonists using [3H]-rauwolscine as the radioligand.

In addition to radioligand binding, the functional profile of the BiOctR was also assessed by measuring the effects of ligands on Gq-activation, and compared it to another α-like octopamine receptor (drosophila octopamine receptor DmOctR). Both the BiOctR and DmOctR could be activated by dexmedetomidine, naphazoline, clonidine, tizanidine, octopamine, (-)- epinephrine and (-)-NE, but only the BiOctR could be activated by (+)-NE and tyramine.

Additionally, dexmedetomidine and all three imidazolines were full agonists, whereas levomedetomidine and the biogenic amines (-)-NE, (-)-epinephrine and tyramine were partial agonists at the BiOctR. In contrast, at the DmOctR, the imidazolines were partial agonists, whereas (-)-NE and (-)-epinephrine were full agonists. We also tested the effects of the imidazoles, imidazolines and octopamine on the activity of barnacle cyprids. Lind and colleagues only showed that medetomidine is an agonist at the BiOctR and is able to induce hyperactivity in cyprids. Since clonidine has also been reported to inhibit cyprid settlement (Dahlstrom and

Elwing 2006; Dahlström et al. 2000), it was hypothesized that all the imidazoles and imidazolines tested in this report with agonist activity will also be able to induce hyperactivity in cyprids, and this was shown to be true. In addition, octopamine at concentrations presumed to be saturating is also able to induce hyperactivity. This further supports the idea that activation of the

OctR by agonists is likely to induce motor hyperactivity in barnacle cyprids and possibly other arthropods. The concentrations of the agonists needed to induce the hyperactivity response were higher than the potency and this is likely do to a drug exposure effect. Since the ligands would have to cross physical barriers (i.e. penetrate the surface and enter the body), and if the compounds are being ingested, they are likely being metabolized as well. Hence, because of the

60 barriers that would have to be overcome to reach the receptor, high concentrations are needed to achieve the desired physiological outcome.

Conclusion

In this report it was demonstrated that the BiOctR exhibits sodium sensitivity for both antagonists and agonists, where the affinity for antagonist is increased and that for the agonist decreased. Further, rauwolscine was shown to interact with an allosteric site on the BiOctR, which is not an ideal trait for a radioligand since it is then not possible to obtain the true affinities for agonists acting at the orthosteric site. However, it can be used to obtain the rank order affinity since they were similar to the rank order potencies. It was also reported here that the imidazoles and imidazolines are potent agonists and have similar rank order potencies at the BiOctR and the

DmOctR, but differ in their efficacy. That is, clonidine, naphazoline, and tizanidine are full agonists at the BiOctR, whereas they are partial agonists at the DmOctR. Finally, it was demonstrated that all the imidazoles and octopamine were able to induce hyperactivity in cyprids and since some of these compounds have been reported to be effective as antifouling agents, the

BiOctR can be used to discover novel OctR agonists which would be predictive of the compounds having antifouling property as well.

Figure 1. [3H]-rauwolscine binds with high affinity to BiOctR under high salt condition.

HEK293 cells stably transfected with the BiOctR were used for optimizing the binding conditions. A, In the absence of salt, binding of [3H]-rauwolscine to BiOctR was undetectable, however, there is a concentration dependent increase in specifically bound radioligand in the presence of Na+. B, To rule out the possibility that the binding could be due to charge, NMDG was used to substitute for NaCl. There was no detectable binding in the absence of sodium (0 mM Na+, 240 mM NMDG, and 50 mM potassium phosphate buffer). C, Representative example of [3H]-rauwolscine saturation isotherm binding to BiOctR. The calculated affinity and receptor density values are expressed as mean ± SEM from three separate experiments (n=3): KD = 4.56 ±

0.54 nM and Bmax = 7.0 ± 2.7 pmol/mg membrane protein, respectively. D, Bmax for the untransfected HEK293 cells was estimated using the square hyperbola model Bmax = (Y •

(KD+X))/X, where Y is [specifically bound radioligand] and X = [radioligand concentration] and was estimated to be 0.02 ± 0.008 pmol/mg. Bmax for the BiOctR was estimated from 3 saturation isotherms and is shown here for completeness. E, Whole cell radioligand binding using the modified physiological HBSS buffer prepared with either NaCl or N-Methyl-D-glucamine

(NMDG, a sodium substitute) demonstrated that extracellular sodium had no effect on specifically bound [3H]-rauwolscine. F, Scatchard transformation of saturation isotherms performed in the presence of high (240 mM Na+) or low (15 mM Na+) salt. High salt concentration has no effect on the Bmax since the x-intercept remained changed, but increases the

KD as evidenced by an increase in the slope. * indicates groups significantly different than in the absence of sodium by one-way ANOVA followed by Bonferroni’s multiple comparison test.

A B

C D

E F

Figure 1. [3H]-rauwolscine binds with high affinity to BiOctR under high salt condition.

Figure 2. BiOctR has high affinity for the α-adrenergic receptor antagonists and the imidazole dexmedetomidine. Affinities were measured using [3H]-rauwolscine as the radioligand under high salt condition (240 mM NaCl). The Ki was calculated from the IC50 using the Cheng-Prusoff equation. A, Rs-79948 had the highest affinity and metoclopramide had the lowest affinity for the BiOctR. Pseudo Hill slope equal to unity is the preferred model for all the compounds. See Table 1 for the affinities. B, Dexmedetomidine had the highest affinity and octopamine and tyramine had the lowest affinity at the BiOctR. Pseudo Hill slope equal to unity is the preferred model for all compounds. See Table 2 for the affinities. Each experiment was performed in triplicate and was repeated at least two times; data represents mean ± SEM.

Figure 2. BiOctR has high affinity for the α-adrenergic receptor antagonists and the

imidazole dexmedetomidine.

Figure 3. The α-adrenergic receptor agonists activate the BiOctR-mediated Gq signaling.

Change in intracellular calcium was used as a measure of Gq-activation using a fluorescent calcium probe. A, In addition to octopamine, tyramine, (-)-epinephrine, (-)-norepinephrine, and

(+)-norepinephrine were able to activate the BiOctR. This activation was fully reversed by the antagonist atipamezole. * represents significant reversal of agonist response (P < 0.05, one-way

ANOVA, Bonferroni’s post-hoc). Other biogenic amines: histamine, serotonin, and dopamine were unable to active the BiOctR. B, The dopamine D2 receptor antagonists chlorpromazine, metoclopramide, nemonapride, and the α-adrenergic receptor antagonists rauwolscine, Rs-79948,

Rx-821002, and yohimbine had no agonist activity, but were able to antagonize Gq signaling at the BiOctR. All antagonist effects were statistically different from the 1 µM octopamine response (*represents significant reversal of 1 µM octopamine response. P < 0.05, one-way

ANOVA, Bonferroni’s post-hoc). C, Dexmedetomidine had subnanomolar potency for the

BiOctR and 30 µM atipamezole caused rightward shift, decreasing the potency by more than

1000-fold. Levomedetomidine had nanomolar potency, and 30 µM atipamezole shifts the curve to right by ~ 470 fold (EC50 = 3929 ± 1007 nM). Both Dexmedetomidine and levomedetomidine had pseudo Hills equal to unity, and a fixed slope was the preferred model. Potencies for the imidazoles can be found in Table 2. D, Comparison of maximum efficacy of select adrenoceptor agonists at the BiOctR receptor. Efficacy of clonidine, dexmedetomidine, naphazoline and tizanidine were not significantly different from octopamine, but efficacy of levomedetomidine, (-

)-epinephrine, (-)-norepinephrine and tyramine were significantly lower than octopamine (* P <

0.05, one-way ANOVA, Bonferroni’s post-hoc). Each experiment was performed in triplicate and was repeated at least two times.

A B

C D

Figure 3. The α-adrenergic receptor agonists activate the BiOctR-mediated Gq signaling.

Figure 4. BiOctR has higher potency for imidazolines than biogenic amines. A, Clonidine, naphazoline and tizanidine had potency in the low nanomolar range, where the potency for naphazoline and tizanidine was greater than clonidine. B, Biogenic amines had potency in the nanomolar to high nanomolar range. Octopamine had the highest potency and (-)-NE had the lowest potency. Each experiment was performed in triplicate and was repeated at least two times.

Figure 4. BiOctR has higher potency for imidazolines than biogenic amines.

Figure 5. Imidazoles induce hyperactivity in barnacle Balanus amphitrite cyprids. Cyprids were immobilized in 2% low melting agarose and then treated with either vehicle (1:10,000 v/v

DMSO), test compounds. At 1 µM concentrations, all compounds except levomedetomidine and octopamine induced hyperactivity, and in each case the kicking response was reversed by the antagonist atipamezole (50 µM). At high micromolar concentration, octopamine was able to induce hyperactivity and was reversed atipamezole. For each experiment, movements of 3-5 cyprids were measured per treatment and each experiment was repeated at least two times. A one-way ANOVA followed by Dunnett’s post-hoc test was performed to identify compounds that induced significant hyperactivity compared to vehicle. * denotes P < 0.05.

Figure 5. Imidazoles induce hyperactivity in barnacle Balanus amphitrite cyprids.

Figure 6. The drosophila α-like octopamine receptor (DmOctR) has similar pharmacology to the BiOctR. A, In addition to octopamine, (-)-epinephrine, (-)-NE, clonidine, dexmedetomidine, naphazoline and tizanidine are also agonists at the DmOctR and their responses could be reversed by the antagonist atipamezole. B, Chlorpromazine, metoclopramide, nemonapride, rauwolscine, Rs-79948, and Rx-821002 had no agonist activity at the DmOctR but were able to suppress the octopamine agonist response. Metoclopramide and nemonapride could only partially inhibit the agonist response. C, The dexmedetomidine and imidazolines had higher potency than the biogenic amines. The potencies are as follows (mean ± SEM nM): Octopamine:

163.9 ± 47.9; Dexmedetomidine: 0.4 ± 0.1; Clonidine: 12.7 ± 4.6; Naphazoline: 3.5 ± 0.5;

Tizanidine: 49.6 ± 6.1; (-)-NE: 1980 ± 529; (-)-Epi: 1187 ± 265. The pseudo Hill slopes ranged from 0.6 to 2.1, but were not significantly different from unity. D, Octopamine and dexmedetomidine have similar efficacy, whereas clonidine, naphazoline and tizanidine have lower efficacy, and (-)-epinephrine and (-)-NE had higher efficacy than octopamine and dexmedetomidine. Values represent mean ± SEM. Each experiment was performed in triplicates and was repeated at least two times. * P < 0.05 by one-way ANOVA followed by Bonferroni’s post-hoc.

A B

C D

Figure 6. The drosophila α-like octopamine receptor (DmOctR) has similar pharmacology

to the BiOctR.

Figure 7. Receptor reserve has only a small effect on the agonist potency and rauwolscine is non-competitive antagonist. A, B, Octopamine and dexmedetomidine concentration response curves were generated in the presence of increasing concentration of phenoxybenzamine. The operational model of agonism was used for curve fitting and estimating the dissociation constant

Ka. The Ka for octopamine and dexmedetomidine were calculated to be 588 ± 226 nM and 0.3 ±

0.1 nM, respectively. C, To determine if rauwolscine and dexmedetomidine are competing for the same binding site, Schild shift analysis was performed. For each fixed concentration of rauwolscine, dexmedetomidine concentration curves were generated. As the concentration of rauwolscine was increased, there was decrease in both efficacy and potency. Each experiment was performed in triplicates and was repeated at least two times.

Figure 7. Receptor reserve has only a small effect on the agonist potency and rauwolscine is

non-competitive antagonist.

Table 1. BiOctR has affinity for α-adrenergic receptor antagonists, D2 dopamine

3 receptor antagonists and a histamine H1 receptor antagonist. [ H]-rauwolscine was used

+ to perform inhibition curves for the different ligands in the presence of 240 mM Na . The Ki values were calculated from the IC50 values using the Cheng-Prusoff equation. Values represent mean ± SEM. Each experiment was performed in triplicate and repeated ≥ 2 times.

Mammalian Affinity, K Ligand Structure i receptor target (nM ± SEM)

Rs-79948 α2-adrenoceptors 0.6 ± 0.08

Rx-821002 α2-adrenoceptors 1.5 ±0.2

D -like and D -like Chlorpromazine 2 1 3.0 ± 0.6 dopamine receptors

Atipamezole α2-adrenoceptors 3.0 ± 0.4

Rauwolscine α2-adrenoceptors 4.6 ± 0.5

H histamine Promethazine 1 10 ± 1 receptors

Yohimbine α2-adrenoceptors 17 ± 2

D -like dopamine Nemonapride 2 66 ± 10 receptors

D -like dopamine Metoclopramide 2 1700 ± 71 receptors

Table 2. Imidazoles are potent agonists at the BiOctR and have higher affinity than biogenic amines. [3H]-rauwolscine was used to perform inhibition curves for the different

+ ligands in the presence of 240 mM Na . The Ki was calculated from the IC50 values using the

Cheng-Prusoff equation. The potencies are 12 to 450 fold higher than the affinities. Values represent mean ± SEM. Each experiment was performed in triplicate and repeated ≥ 2 times.

Efficacies different from octopamine are marked with an asterisk (P < 0.05, Bonferroni’s post- hoc).

Efficacy Potency, (normalized to Affinity, K Ligand Structure EC (nM ± i octopamine, 50 (nM ± SEM) SEM) mean ± SEM) Dexmedetomidine 1.0 ± 0.1 0.09 ± 0.02 8.2 ± 0.8

Tizanidine 1.0 ± 0.09 1.2 ± 0.3 617 ± 103

Naphazoline 0.9 ± 0.1 1.4 ± 0.3 42 ± 8

Clonidine 0.9 ± 0.03 3.1 ± 1.0 99 ± 22

Levomedetomidine 0.7 ± 0.05* 8.3 ± 2.2 101 ± 20

Octopamine 1.0 28 ± 9 43,350 ± 3260

44840 ± Tyramine 0.5 ± 0.05* 130 ± 22 11541 (-)-Epinephrine 0.6 ± 0.05* 221 ± 46 N.D.

(-)-Norepinephrine 0.7 ± 0.05* 862 ± 171 N.D.

(+)- Very weak N.D. N.D. Norepinephrine agonist

Dopamine No activity N.D. N.D.

Serotonin No activity N.D. N.D.

Histamine No activity N.D N.D

References

Berry M. D. (2004) Mammalian central nervous system trace amines. Pharmacologic

amphetamines, physiologic neuromodulators. J. Neurochem. 90, 257-271.

Black J. W., Leff P., Shankley N. P. and Wood J. (2010) An operational model of

pharmacological agonism: the effect of E/[A] curve shape on agonist dissociation constant

estimation. 1985. Br. J. Pharmacol. 160 Suppl 1, S54-64.

Boyes W. K. and Moser V. C. (1988) An alpha 2-adrenergic mode of action of chlordimeform on

rat visual function. Toxicol. Appl. Pharmacol. 92, 402-418.

Casida J. E. and Durkin K. A. (2013) Neuroactive insecticides: targets, selectivity, resistance,

and secondary effects. Annu. Rev. Entomol. 58, 99-117.

Costa L. G., Gastel J. and Murphy S. D. (1991) The formamidine pesticides chlordimeform and

amitraz decrease hepatic glutathione in mice through an interaction with alpha 2-

adrenoceptors. J. Toxicol. Environ. Health. 33, 349-358.

Costa L. G., Olibet G. and Murphy S. D. (1988) Alpha 2-adrenoceptors as a target for

formamidine pesticides: in vitro and in vivo studies in mice. Toxicol. Appl. Pharmacol. 93,

319-328.

Dahlstrom M. and Elwing H. (2006) Adrenoceptor and other pharmacoactive compounds as

putative antifoulants. Prog. Mol. Subcell. Biol. 42, 171-202.

Dahlström M., Mårtenssonb L. G., Jonssonc P. R., Arnebrantd T. and Elwinge H. (2000) Surface

active adrenoceptor compounds prevent the settlement of cyprid larvae of Balanus

improvisus. Volume 16, 191-203.

Dahms H. U., Jin T. and Qian P. Y. (2004) Adrenoceptor compounds prevent the settlement of

marine invertebrate larvae: Balanus amphitrite (Cirripedia), Bugula neritina (Bryozoa) and

Hydroides elegans (Polychaeta). Biofouling. 20, 313-321.

Enan E. (2001) Insecticidal activity of essential oils: octopaminergic sites of action. Comp.

Biochem. Physiol. C. Toxicol. Pharmacol. 130, 325-337.

Enan E. E. (2005) Molecular and pharmacological analysis of an octopamine receptor from

American cockroach and fruit fly in response to plant essential oils. Arch. Insect Biochem.

Physiol. 59, 161-171.

Evans P. D. (1993) Molecular studies on insect octopamine receptors. EXS. 63, 286-296.

Evans P. D. (1984a) A modulatory octopaminergic neurone increases cyclic nucleotide levels in

locust skeletal muscle. J. Physiol. 348, 307-324.

Evans P. D. (1984b) The role of cyclic nucleotides and calcium in the mediation of the

modulatory effects of octopamine on locust skeletal muscle. J. Physiol. 348, 325-340.

Evans P. D. (1981) Multiple receptor types for octopamine in the locust. J. Physiol. 318, 99-122.

Evans P. D. and Maqueira B. (2005) Insect octopamine receptors: a new classification scheme

based on studies of cloned Drosophila G-protein coupled receptors. Invert. Neurosci. 5, 111-

118.

Evans P. D. and Robb S. (1993) Octopamine receptor subtypes and their modes of action.

Neurochem. Res. 18, 869-874.

Farooqui T. (2007) Octopamine-mediated neuromodulation of insect senses. Neurochem. Res.

32, 1511-1529.

Gross A. D., Temeyer K. B., Day T. A., Perez de Leon A. A., Kimber M. J. and Coats J. R.

(2015) Pharmacological characterization of a tyramine receptor from the southern cattle

tick, Rhipicephalus (Boophilus) microplus. Insect Biochem. Mol. Biol. 63, 47-53.

Han K. A., Millar N. S. and Davis R. L. (1998) A novel octopamine receptor with preferential

expression in Drosophila mushroom bodies. J. Neurosci. 18, 3650-3658.

Hashemzadeh H., Hollingworth R. M. and Voliva A. (1985) Receptors for 3H-octopamine in the

adult firefly light organ. Life Sci. 37, 433-440.

Hollingworth R. M. and Murdock L. L. (1980) Formamidine pesticides: octopamine-like actions

in a firefly. Science. 208, 74-76.

Hsu W. H. and Kakuk T. J. (1984) Effect of amitraz and chlordimeform on heart rate and pupil

diameter in rats: mediated by alpha 2-adrenoreceptors. Toxicol. Appl. Pharmacol. 73, 411-

415.

Hsu W. H., Smith B. E. and Hollingworth R. M. (1988) The bradycardic and mydriatic effects of

chlordimeform and its demethylated analogs in the rat: antagonism by idazoxan but not by

prazosin. Life Sci. 43, 1897-1904.

Katritch V., Fenalti G., Abola E. E., Roth B. L., Cherezov V. and Stevens R. C. (2014) Allosteric

sodium in class A GPCR signaling. Trends Biochem. Sci. 39, 233-244.

Kleinau G., Pratzka J., Nurnberg D., Gruters A., Fuhrer-Sakel D., Krude H., Kohrle J.,

Schoneberg T. and Biebermann H. (2011) Differential modulation of Beta-adrenergic

receptor signaling by trace amine-associated receptor 1 agonists. PLoS One. 6, e27073.

Kortagere S., Gmeiner P., Weinstein H. and Schetz J. A. (2004) Certain 1,4-disubstituted

aromatic piperidines and piperazines with extreme selectivity for the dopamine D4 receptor

interact with a common receptor microdomain. Mol. Pharmacol. 66, 1491-1499.

Kuffler S. and Nicholls J. (1976) From Neuron to Brain. A cellular approach to the function of

the nervous system. Sinauer Associates, INC., Sunderland, Massachusetts.

Leff P., Martin G. R. and Morse J. M. (1985) Application of the operational model of agonism to

establish conditions when functional antagonism may be used to estimate agonist

dissociation constants. Br. J. Pharmacol. 85, 655-663.

Lind U., Alm Rosenblad M., Hasselberg Frank L., Falkbring S., Brive L., Laurila J. M.,

Pohjanoksa K., Vuorenpaa A., Kukkonen J. P., Gunnarsson L., Scheinin M., Martensson

Lindblad L. G. and Blomberg A. (2010a) Octopamine receptors from the barnacle balanus

improvisus are activated by the alpha2-adrenoceptor agonist medetomidine. Mol.

Pharmacol. 78, 237-248.

Lind U., Alm Rosenblad M., Hasselberg Frank L., Falkbring S., Brive L., Laurila J. M.,

Pohjanoksa K., Vuorenpaa A., Kukkonen J. P., Gunnarsson L., Scheinin M., Martensson

Lindblad L. G. and Blomberg A. (2010b) Octopamine receptors from the barnacle balanus

improvisus are activated by the alpha2-adrenoceptor agonist medetomidine. Mol.

Pharmacol. 78, 237-248.

Maguire J. J. and Davenport A. P. (2015) Trace amine receptor: TA1 receptor. IUPHAR/BPS

Guide to PHARMACOLOGY. 2016.

Malmberg A., Jackson D. M., Eriksson A. and Mohell N. (1993) Unique binding characteristics

of antipsychotic agents interacting with human dopamine D2A, D2B, and D3 receptors.

Mol. Pharmacol. 43, 749-754.

Matsumura F. and Beeman R. W. (1976) Biochemical and physiological effects of

chlordimeform. Environ. Health Perspect. 14, 71-82.

Monty J. P., Dogan E., Hanson R., Scardino A. J., Ganapathisubramani B. and Hutchins N.

(2016) An assessment of the ship drag penalty arising from light calcareous tubeworm

fouling. Biofouling. 32, 451-464.

Nathanson J. A. (1985a) Characterization of octopamine-sensitive adenylate cyclase: elucidation

of a class of potent and selective octopamine-2 receptor agonists with toxic effects in

insects. Proc. Natl. Acad. Sci. U. S. A. 82, 599-603.

Nathanson J. A. (1985b) Phenyliminoimidazolidines. Characterization of a class of potent

agonists of octopamine-sensitive adenylate cyclase and their use in understanding the

pharmacology of octopamine receptors. Mol. Pharmacol. 28, 254-268.

Nathanson J. A. and Greengard P. (1973) Octopamine-sensitive adenylate cyclse: evidence for a

biological role of octopamine in nervous tissue. Science. 180, 308-310.

Nathanson J. A. and Hunnicutt E. J. (1981) N-demethylchlordimeform: a potent partial agonist

of octopamine-sensitive adenylate cyclase. Mol. Pharmacol. 20, 68-75.

Nathanson J. A. and Hunnicutt E. J. (1979) Neural control of light emission in Photuris larvae:

identification of octopamine-sensitive adenylate cyclase (1). J. Exp. Zool. 208, 255-262.

Neve K. A. (1991) Regulation of dopamine D2 receptors by sodium and pH. Mol. Pharmacol.

39, 570-578.

Roeder T. (2005) Tyramine and octopamine: ruling behavior and metabolism. Annu. Rev.

Entomol. 50, 447-477.

Roeder T. (1999) Octopamine in invertebrates. Prog. Neurobiol. 59, 533-561.

Roeder T. (1995) Pharmacology of the octopamine receptor from locust central nervous tissue

(OAR3). Br. J. Pharmacol. 114, 210-216.

Roeder T. (1992) A new octopamine receptor class in locust nervous tissue, the octopamine 3

(OA3) receptor. Life Sci. 50, 21-28.

Rosenberg L. A., Glusman J. G. and Libersat F. (2007) Octopamine partially restores walking in

hypokinetic cockroaches stung by the parasitoid wasp Ampulex compressa. J. Exp. Biol.

210, 4411-4417.

Schetz J. A. (2005) Allosteric modulation of dopamine receptors. Mini Rev. Med. Chem. 5, 555-

561.

Swale D. R., Sun B., Tong F. and Bloomquist J. R. (2014) Neurotoxicity and mode of action of

N, N-diethyl-meta-toluamide (DEET). PLoS One. 9, e103713.

Winner C. (2013) To Banish Biofouling: Can a new approach help sailors prevent barnacles and

algae from coating ship hulls? Oceanus Magazine. 50, 9-11.

CHPATER 3

NOVEL SELECTIVE SIGMA-1 RECEPTOR LIGANDS FACILITATE BDNF RELEASE FROM A NEURONAL CELL LINE

Abstract

The Sigma-1 receptor (S1R) is an endoplasmic reticulum (ER) chaperone protein that has been implicated in attenuating inflammatory stress-mediated brain injuries. Selective S1R agonists represent a new class of therapeutic agent for treating neuropsychiatric and neurodegenerative disorders. The EPGN compounds used in this study are a new structural class of S1R ligands and their structure-activity relationships (SAR) were assessed by evaluating target and functional selectivity of these compounds. By evaluating series of structurally similar molecules with single modifications, substructural features that control S1R over S2R selectivity were identified. In this study four S1R selective compounds with more than 100 fold selectivity over the S2R were identified. Using an in situ ELISA approach, the SAR of S1R-mediated

BDNF secretion was explored and identified structural features that influence functional selectivity of BDNF secretion. These compounds were classified as S1R agonists because the

BDNF response was comparable to the prototypical agonist 4-PPBP and because it could be reversed by a S1R selective concentration of the antagonist BD1063. When modulation of IP3 mediated calcium response and NGF-induced neurite sprouting were used as a measure of S1R

86 activation, it was determined that they are not reliable measures for evaluating functional properties of S1R ligands.

Introduction

Alzheimer’s disease (AD) is a neurodegenerative disorder, and the disease pathology has been historically associated with deposits of amyloid beta (Aβ) plaques and tau tangles. These insoluble Aβ plaques and tau tangles provide obvious stimuli for inducing inflammatory stress which exacerbates the pathogenic process significantly contributing to AD pathogenesis. AD is the sixth leading cause of death in the US and one in three seniors die with AD or other forms of dementia (Klemas and Dowling 2015). More than 46 million people worldwide are living with some form of dementia (Prince et al. 2015), of which 60 to 70% of dementia cases are attributed to AD (WHO 2016). As the proportion of elderly population increases, by the year 2050, the global and US prevalence of dementia is projected to increase to 131.5 million and 13.8 million, respectively (Klemas and Dowling 2015; Prince et al. 2015). Treatment of dementia is also a huge global economic burden, with a cost of $818 billion in total direct costs (Prince et al. 2015).

In the US, indirect AD costs are estimated to be $217 billion, and $226 billion in direct costs

(Klemas and Dowling 2015). Of this cost, Medicare and Medicaid are expected to cover $153 billion, or 68% of the total direct cost (Klemas and Dowling 2015). By 2050, the cost is estimated to increase to $1.1 trillion in the U.S. alone (Alzheimer's Association 2015a). Even with this staggering cost, to date there are only four AD treatments approved by the US FDA

(Ehret and Chamberlin 2015).

Current treatment options include acetylcholinesterase (AChE) inhibitors donepezil, rivastigmine, and galantamine and the NMDA receptor antagonist memantine (Ehret and

Chamberlin 2015). Tacrine was the first AChE inhibitor available, but in 2013 it was

87 discontinued in the U.S. due to concerns of hepatic toxicity (Ehret and Chamberlin 2015). In AD, there is a significant decline in cholinergic neurons and the AChE inhibitors reduce the rate of acetylcholine (ACh) breakdown (Lombardo and Maskos 2015) leading to an increase in the concentration of ACh in the brain. Another common feature of AD is glutamate excitotoxicity; glutamate is an excitatory neurotransmitter but in excess it can lead to accumulation of toxic levels of intracellular calcium leading to cell death. Memantine is a noncompetitive NMDA receptor antagonist, which inhibits glutamate excitotoxicity. These drugs help to mask the symptoms of AD for a brief period of time by temporarily improving cognitive function, but do not treat the underlying cause or delay/halt the progression of the disease. The AChE inhibitors only delay the worsening of symptoms by 6 to 12 months in only half of the patients that take the medication (Alzheimer's Association 2015b). Similarly, memantine only temporarily delays the worsening of symptoms for some patients (Alzheimer's Association 2015b). Since these drugs only provide temporary relief in only fraction of the patients, there is a need for target-based drug discovery programs to discover new compounds that can slow the progression and/or prevent the disease.

The Sigma-1 receptor (S1R) is a stress and ligand-regulated, endoplasmic reticulum chaperone protein that shuttles lipids and proteins to the plasma membrane, and they are ubiquitous throughout the body, especially in the CNS (Su et al. 2010). High densities of the

S1R are found in brain tissue, including the prefrontal and parietal cortex and various limbic structures such as the olfactory bulb, the hypothalamus, and the hippocampus (Alonso et al.

2000; Hashimoto et al. 1995). Within the tissues of the nervous system, the S1R is located predominantly in the gray matter, including neurons (Alonso et al. 2000; Klette et al. 1995;

Peviani et al. 2014) and a variety of glial cell types: astrocytes, microglia, Müller cells,

88 oligodendrocytes and Schwann cells (Gekker et al. 2006; Hayashi and Su 2004; Jiang et al.

2006; Palacios et al. 2003; Palacios et al. 2004; Peviani et al. 2014; Robson et al. 2014). The

S1R modulates the actions of neurotransmitter receptors, ion channels and it is also involved in the regulation of diverse processes such as neuroprotection, neurorestoration, neuroplasticity, and the release of neurotransmitters (Kourrich et al. 2012; Ruscher and Wieloch 2015; Su et al.

2010; Zheng 2009).

The S1R has gained considerable attention as a therapeutic target for treating neurodegenerative diseases because of its involvement in regulating a variety of processes, including cell survival. There is compelling evidence for the use of S1R agonists in the treatment of cognitive dysfunction since they have been shown to be neuroprotective and protect against

Aβ toxicity (Jin et al. 2015). While the precise functions of the S1Rs are unclear, in an acute

A25-35 intracerebroventricular injection mouse model of AD, S1R-preferring compounds such as

(+)-pentazocine and PRE-084, which are categorized as S1R agonists, attenuate memory deficits and resist neurotoxicity, an effect that is prevented by the S1R antagonist NE-100 (Marrazzo et al. 2005; Maurice et al. 1998). S1R agonists also have anti-amnesic properties in amnesia models, including amyloid induced AD models (Maurice et al. 1998; Villard et al. 2009), as well as in a cholinergic deficit model (Zou et al. 2000). The S1R is also involved in the regulation of

Brain-Derived Neurotrophic Factor (BDNF), which is an important CNS neuropeptide involved in the maintenance and repair of neurons, synaptic plasticity, and learning and memory (Cunha et al. 2010; Korte et al. 1995; Lewin and Barde 1996). In animal models, chronic administration of the S1R agonist SA4503 increases the level of BDNF protein in the rat hippocampus (Kikuchi-

Utsumi and Nakaki 2008); similarly, in vitro, SA4503 is associated with enhanced secretion of

BDNF into the extracellular environment (Fujimoto et al. 2012). S1Rs have also been shown to

89 regulate neurite outgrowth. In vitro studies using PC12 cells (a cell line derived from a pheochromocytoma of the rat adrenal medulla) demonstrated that S1R ligands like (+)- pentazocine, imipramine, fluvoxamine, donepezil, and dehydroepiandrosterone-sulfate, fluoxetine, escitalopram, mirtazapine, sertraline, paroxetine, SA4503, and 4-PPBP facilitate

NGF-induced neurite sprouting, an effect that is reversed by S1R antagonists NE100 and

BD1063 (Ishima et al. 2008; Ishima and Hashimoto 2012; Ishima et al. 2014; Nishimura et al.

2008; Rossi et al. 2011; Takebayashi et al. 2002). Given the multitude of processes involving the

S1R and the ability of certain S1R ligands to promote BDNF synthesis/secretion, facilitate NGF- induced neurite outgrowth and reduce nitrosative stress, the S1R is a very attractive druggable target for potentially slowing the progression or preventing AD.

There are two known subtypes of sigma receptors, S1R and the sigma-2 receptor (S2R), and though the S1R has been extensively studied, the role of the S2R is still not well understood.

Pharmacologically, the S1R and the S2R are very similar, but differ in their stereoselectivity for benzomorphans (Hellewell and Bowen 1990). Both S1R and S2R have high affinity for haloperidol, DTG, and (+)-3-PPP, however, the S1R has high selectivity for the positive enantiomer of benzomorphans, whereas the S2R has low affinity for (+)-benzomorphans and high affinity for (-)-benzomorphans (Hellewell and Bowen 1990). The S2R has been reported to be involved in the regulation of cell survival, morphology and differentiation (Guitart et al. 2004;

Huang et al. 2014; Vilner et al. 1995a), and are highly expressed in cancer cells (Al-Nabulsi et al. 1999; Vilner et al. 1995b). Further the S2R agonists have been reported to induce cell death via activation of the apoptotic pathway (Bowen 2000; Hornick et al. 2012; Zeng et al. 2012;

Zeng et al. 2013; Zeng et al. 2014), have immunosuppressor and anti-inflammatory effects

(Iniguez et al. 2013), increase reactive oxygen species in cells, and induce lysosomal

90 permeabilization (Hornick et al. 2012; Ostenfeld et al. 2005), and modify cardiac repolarization by blocking inward rectifying K+ channel in the heart, which has the potential to cause cardiac sudden death (Monassier et al. 2007). While activation of some of these pathways may be attractive targets from a cancer therapeutic perspective, from a neurodegenerative disease perspective, activation of these pathways could exacerbate the disease pathology. Hence, in this study the goal was to discover novel S1R selective compounds, without any S2R activity and evaluate their functional profile using multiple functional assays. Since S1R agonists have neuroprotective properties, which are blocked by antagonists, the focus was to select for compounds with S1R agonist activity. The primary means of qualifying S1R ligands as agonists involve behavioral readouts: 1) memory and cognitive function improvement, 2) antidepressant- like effects and 3) psychostimulant-induced behavior modulation (Hayashi and Su 2004). This approach is not high throughput nor cost effective for evaluating large numbers of compounds.

However, there are some reports which suggest that S1R agonists potentiate the IP3 mediated calcium response (Hayashi et al. 2000; Hayashi and Su 2007; Hong et al. 2004; Wu and Bowen

2008), facilitate BDNF release (Fujimoto et al. 2012; Malik et al. 2015), and potentiate NGF- induced neurite sprouting (Ishima et al. 2008; Ishima and Hashimoto 2012; Ishima et al. 2014;

Nishimura et al. 2008; Rossi et al. 2011; Takebayashi et al. 2002). In this study, all three measures were utilized to develop robust high-throughput assays for evaluating S1R compounds.

Based on the findings of this study, only BDNF release appears to be a reproducible measure for evaluating S1R ligands, specifically, the BDNF assay utilized here can be used to identify functionally selective agonists that activate the BDNF secretion pathway.

Materials and Methods

Chemicals and reagents

Compounds were purchased from the following sources: BD1063, 4-PPBP, and (+)- igmesine from Tocris Biosciences (Minneapolis, MN). NE-100 was purchased from Santa Cruz

Biotechnology (Dallas, TX), oxeladin was from Sigma-Aldrich (St. Louis, MO) and the EPGN compounds were from Epigen Biosciences (San Diego, CA). Radioligands were purchased from

PerkinElmer: ([3H]-(+)-pentazocine ((+)-Pentazocine, [RING-1,3-3H], 33.9 Ci/mmol, NET1056),

[3H]DTG (1,3-Di-o-tolylguanidine, [p-RING-3H]-, 50 Ci/mmol, NET986) (Saint Louis, MO).

All compounds were prepared in DMSO at concentrations ranging from 10-100 mM. Stocks were then diluted 1:1000 (v/v) in the final assay solution.

Cell culture

Human MCF-7 cells (American Type Cell Culture, Manassas, VA) were grown in

Dulbecco's Modified Eagle's Medium (DMEM; Fisher Scientific, Pittsburgh, PA) supplemented with 10% Fetal Bovine Serum (FBS, Fisher Scientific, Pittsburgh, PA), 100 μg/ml nonessential amino acids (Hyclone, Logan, UT), 2 mM L-glutamine (Sigma-Aldrich, St. Louis, MO), and 10

μg/L Bovine Insulin (Sigma-Aldrich, Sigma-Aldrich, St. Louis, MO). MCF-7 cells served as the source of the human Sigma-2 receptors (S2R) as they express the S2R but lack detectable levels of the S1R (Schetz et al. 2007; Vilner et al. 1995b). MCF-7 cells stably expressing the human

S1R (MCF7-hS1R) were prepared as described previously (Schetz et al. 2007) and kept under constant selective pressure with 100 µg/mL G-418 (Invivogen, San Diego, CA). The rat PC12 cells were purchased from ATCC and PC-6-15, a variant of the PC12 cells overexpressing the trkA receptor (Hempstead et al. 1992), were obtained from Dr. Moses Chao. Both PC12 and PC-

6-15 cells were maintained in RPMI1640 media supplemented with 10% horse serum and 5%

FBS. The PC-6-15 cell medium also contained 100 µg/mL G-418 for constant selective pressure.

The neuronal MN9D cells were obtained from (American Type Cell Culture, Manassas, VA) and were maintained in DMEM complete with 10% FBS. All culture media were additionally supplemented by 100 IU/ mL of penicillin and streptomycin (Corning, Manassas, VA) and 1 mM sodium pyruvate (Sigma-Aldrich, St. Louis, MO), and grown at 37 °C under 95% air, 95% humidity, and 5% CO2.

Measurement of BDNF secretion via in situ ELISA

The in situ ELISA format utilized here has been shown to have improved sensitivity related to its rapid capture of secreted BDNF (Balkowiec and Katz 2000). The in situ ELISA improves on traditional ELISA by seeding BDNF-secreting cells directly into the wells pre- coated with an anti-BDNF primary antibody. BDNF secreted by these cells is immediately captured by the primary antibody, drastically increasing the sensitivity and reproducibility of the

ELISA assay. Intact cells are removed subsequent to the addition of the secondary and tertiary antibodies. The amount of BDNF secreted from the MN9D cells was quantified using an in situ

ELISA assay developed using the BDNF Emax ImmunoAssay kit (Cat. No. G7611, Promega,

Madison, WI). Briefly, a Nunc MaxiSorp flat-bottom, polystyrene, 96-well immunoplate was coated for 48 hrs at 4ºC with an anti-BDNF monoclonal antibody diluted 1:1000 v/v in carbonate buffer containing 25 mM sodium bicarbonate and 25 mM sodium carbonate, pH 9.7. Unbound antibody was removed by washing 5 times with 150 µL of TBST buffer (20 mM Tris-HCl, pH7.6, 150 mM NaCl, and 0.05% (v/v) Tween 20), before blocking non-specific sites first with blocking buffer for 1 hr and then with culture medium for 2 hrs. Cells were seeded at 35,000 cells per well and incubated overnight at 37°C in a humidified CO2 incubator. The following day, the wells were replaced with fresh culture media containing either experimental compounds

93 or vehicle controls, then incubated for an additional 24 hours. In addition, on the same plate, but in separate wells, BDNF standards were added ranging in concentration from 15.6–250 pg/mL.

After 24 hrs incubation with experimental compounds, the media was aspirated and 100 μL of

Dulbecco’s phosphate saline (D-PBS without Ca2+ and Mg2+ supplemented with 5 mM EDTA) was added to each well and incubated for 15 min at 37°C to promote cell lifting. Cells were then detached from the bottom of wells by triturating in the center and around the edges of the well.

After removing all cell debris, the wells were rinsed five times with 150 µL of TBST. The plate was then incubated with 1:500 v/v diluted polyclonal anti-human BDNF antibody for 2 hrs at room temperature. This antibody was removed and wells were washed five times with 150 µL of

TBST, before incubating with 1:200 v/v diluted polyclonal Anti-IgY HRP conjugate for 2 hrs at room temperature. Wells were then washed five times with 150 µL of TBST and the remaining specifically bound polyclonal antibody was detected with the 50 µL colorimetric HRP substrate

3,3’,5,5’-Tetramethylbenzidine (TMB). The reaction was terminated with 50 µL of 1 M HCl and the color intensity was quantified by measuring the absorbance at 450 nm using a Flex Station 3 plate reader (Molecular Devices, Sunnyvale, CA). Measurements from multiple experiments were normalized to maximal BDNF responses achieved by stimulating the prototypical sigma ligand 4-PPBP. A one-way ANOVA with a Bonferroni multiple comparisons post-hoc analysis

(P < 0.05) was applied to determine significant differences between groups. When converted to pg/mL averaged BDNF values ± SEM (n = 3-10 experiments) were: 75.4 ± 6.9 for baseline

(vehicle control) and 128.6 ± 12.8 for maximal stimulation by 10 µM 4-PPBP.

Measuring receptor density with radioligand binding

The density or maximum number of binding sites (Bmax) for the S1R in PC-6-15 cells was estimated by employing 10 nM of [3H]-(+)-pentazocine as the radioligand followed by

94 calculation of the Bmax at saturation using a square hyperbola model: Bmax = (Y • (KD+X))/X, where Y is [specifically bound radioligand] and X = [radioligand concentration]. The affinity

3 (KD) for [ H]-(+)-pentazocine at the cloned human S1R had been previously determined to be

3.7 nM (Schetz et al. 2007).

Measuring the ligand affinities at S1R and S2R receptors by radioligand binding

The affinity (Ki) values of compounds interacting with the S1R and S2R were determined by displacement of 0.5 nM [3H]-(+)-pentazocine from MCF-7-S1R and 2.5 nM [3H]DTG from untransfected MCF-7 cells, respectively. Binding conditions were the same as described previously for the S1R (Schetz et al. 2007): binding buffer (Tris 50 mM, pH = 8.1 at 37°C), ice- cold wash buffer (Tris 10 mM, pH = 8.1 at 0-2°C), and incubation time (3 hrs at 37°C) with shaking. Non-specific binding for the S1R and S2R was determined in the presence of 5 μM

BD1063 and 15 μM haloperidol, respectively. Following incubation, receptors were collected via rapid filtration through GF/C filters (Brandel, Gaithersburg, MD) followed by washing three times with 3 mL of ice-cold wash buffer. Dried filters were transferred to vials filled with 3.5 ml of scintillation fluid, and the radioactivity was quantified on a liquid scintillation analyzer (Tri

Carb 2800TR) from Perkin Elmer (Saint Louis, MO). Mean values from duplicate or triplicate determinations are reported along with their associated standard error of the mean (SEM). Ki values were calculated from the IC50 values using the Cheng-Prusoff equation. The concentration of membrane protein was determined using a BCA Protein Assay kit (Life Technologies, Grand

Island, NY) following the manufacturer’s protocol.

Measurement of the effect of sigma ligands on IP3 induced change in intracellular calcium

MCF7 and MCF7-hS1R cells were plated in Poly-L-Lysine coated black walled 96 well plates at a density of 120,000 cells per well in complete media and incubated overnight in a

95 humidified CO2 cell culture incubator. The following day, the media was removed and the cells were loaded with the FLIPR Calcium-6 QF dye (Molecular Devices, Sunnyvale CA) dissolved in

Hank’s buffered saline supplemented with 20 mM HEPES pH = 7.4, for three hours in the dark.

To test the effect of sigma ligands on IP3 induced calcium response, the loading dye was supplemented with the sigma ligands or vehicle (DMSO, 1:1000). Following dye loading and any pretreatments, rapid fluorescent signals in response to changes in intracellular calcium was measured at 485 nm excitation, 525 nm emission with the cutoff filter set at 515 nm using the

Flexstation 3 (Molecular Devices, Sunnyvale, CA). The baseline calcium signal was measured for 120 seconds followed by stimulation of the endogenous bradykinin receptors with bradykinin

2+ to activate the Gq-PLC-IP3-Ca pathway, and measured changes in the calcium signal for an additional 780 seconds. At the termination of the experiment, all signals were baseline subtracted and any change in intracellular calcium were quantified as the area under the curve (AUC), calculated by integration (Graphpad Prism version 4.0). For the bradykinin concentration response curves, the AUC was normalized to the maximal functional response defined by the asymptote and plotted as a sigmoidal dose response curve. AUCs for sigma ligand treated groups were normalized to the bradykinin response in the presence of vehicle. If the presence of the sigma ligand potentiated the bradykinin calcium response, it would be categorized as an agonist, and if it suppressed the response then it would categorized as an inverse agonist. Further, if the sigma ligand had no effect the bradykinin induced calcium response on its own, but suppressed the effects of an agonist sigma ligand then it would be categorized as an antagonist.

Effect of sigma ligands on NGF induced neurite sprouting

A variant of the rat PC12 cells, called PC-6-15 (gift from Dr. Moses Chao), that overexpress the trkA receptor was used to shorten the duration of the assay (Hempstead et al.

1992). Untransfected PC12 normally take 5 to 7 days to respond to NGF treatment and have a tendency to clump. The PC-6-15 cells respond to NGF within 24 hours and do not clump. Cells were maintained in RPMI1640 media supplemented with 50 units/mL streptomycin, 50 µg/mL penicillin, 1 mM sodium pyruvate, 0.1 mg/mL G418 to maintain selection, and 10% horse serum and 5% FBS. On day 1 of the experiment, cells were plated in rat type I collagen (EMD

Millipore, Massachusetts) coated 96 well plates in low serum media (0.5% serum) and treated with vehicle (DMSO), S1R ligands, NGF (Cat# 01-125, βNGF, EMD Millipore, Massachusetts)

(EC50 concentration) or S1R ligand + NGF; each treatment was performed in duplicate. Cells

o were then incubated at 37 C in a humidified incubator with 5% CO2 for 24 hrs. After the 24 hrs treatment, bright field images were obtained from three different fields of view per well. The neurite length was quantified by tracing the neurites using ImageJ (Schneider et al. 2012) and presented as total neurite length divided by the total number of cells.

Results

In this study we report on novel S1R ligands that have greater S1R selectivity over the

S2R, and evaluating their functional responses using three different measures of S1R activation.

S1R ligands were first empirically evaluated and determined their selectivity over the S2R.

Known S1R agonists include 4-PPBP, PRE-084, donepezil and (+)-igmesine and known antagonists include BD1063 and NE-100, served as reference compounds in this study. Human

MCF-7 cells were used as the source of the S2R because these cells express the endogenous S2R

(Vilner et al. 1995b; Wu and Bowen 2008) but lack detectable levels of the S1R (Schetz et al.

2007); [3H]- DTG was used as the radioligand to probe for the S2R. MCF-7 cells stably expressing the cloned human S1R were used as the source of S1R and the high affinity S1R selective radioligand [3H]-(+)-pentazocine was used to probe for the S1R (Schetz et al. 2007).

The advantage of the MCF-7 system is that since these cells lack detectable levels of

S1R, there is no need to block the S1R sites with a S1R selective ligand like (+)-pentazocine or dextrallorphan. If the receptors were sourced from tissue, then the S1R site would have to be blocked because both receptor subtypes would be present, and DTG has approximately equal affinity for S1R and S2R (Hellewell and Bowen 1990; Matsuno et al. 1996), hence without blocking the S1R site it would be impossible to accurately measure affinity at the S2R. The affinities for the prototypical agonists at the S1R ranged from 1 to 92 nM, where 4-PPBP had the highest affinity and PRE-084 had the lowest affinity (Figure 1, Table 1). In order to evaluate the pseudo Hill slope parameter, all the slopes were treated as a variable (instead of a fixed value) for the purpose of curve fitting. Using this approach, variable slope was the preferred model for

4-PPBP, with a pseudo Hill slope of -3.1 (P < 0.001, t-test). At the S2R the affinities for the same agonists ranged from 0.8 to 13,070 nM, where 4-PPBP had the highest affinity and PRE-

084 had the lowest affinity. At the S2R, both 4-PPBP and PRE-084 had pseudo Hill slopes near unity, thus a pseudo Hill slope equal to unity was the preferred model (P > 0.001, t-test).

Although PRE-084 had the greatest selectivity over the S2R, and 4-PPBP had no selectivity, in our pilot functional studies using BDNF secretion as the functional output, 4-PPBP treatment resulted in a robust BDNF secretion whereas PRE-084 had no effect (Figure 3). Hence, 4-PPBP was used as the reference agonist for all BDNF functional assay reported here. Affinities of two prototypical antagonists were also evaluated at the S1R and S2R (Figure 1, Table 1). Both NE-

100 and BD1063 had low nanomolar affinities at the S1R (1.9 and 5.6 nM, respectively), and high nanomolar affinities at the S2R (41.4 and 210.2 nM, respectively). At the S1R, a variable slope was the preferred model for NE-100 (P < 0.001, t-test), whereas a pseudo Hill slope equal to unity was the preferred model for BD1063 (P > 0.001, t-test). At the S2R, pseudo Hill slope

98 equal to unity was the preferred model for both NE-100 and BD1063 (P > 0.001, t-test). Both

NE-100 and BD1063 have greater selectivity for S1R over the S2R, however, BD1063 had slightly greater selectivity than NE100 (38 vs 21 fold). Knowing the selectivity of the antagonists was an important parameter because it allowed us to identify a S1R selective antagonist concentration for experiments involving antagonist inhibition of the agonist response. The implication of this is that if the agonist response can be reversed by a S1R selective antagonist concentration then the observed response is likely mediated by the S1R and not the S2R. The selective concentration of BD1063 was calculated to be 15 nM because it corresponds to a calculated S1R and S2R occupancy of 74% and 7%, respectively (Fractional Occupancy =

[ligand]/([ligand]+Ki)), hence BD1063 was used as the preferred antagonist in the subsequent functional assays.

In an effort to discover S1R selective agonists, more than 100 novel compounds were evaluated and Figure 2 and Table 2 show the S1R over S2R selectivity for six of the novel compounds (EPGN compounds) belonging to different structural class. Activation of the S1R has also been linked to antitussive effects and carbetapentane is a known antitussive agent that is commonly used in over-the-counter cough medication (Drugs.com 2014). Carbetapentane is also classified as an S1R agonist (Brown et al. 2004) and other antitussive agents like oxeladin and butamirate belong to the same structural class. Since all three antitussive agents only differ by a single modification it allowed for the evaluation of the structure-activity relationship (SAR) of these compounds at the S1R in the context of BDNF release. Series A (Figure 2 and Table 2) represents the antitussive agents and all three compounds had nanomolar affinities for both S1R and S2R demonstrating only weak selectivity over the S2R. Carbetapentane was only 3-fold selective over the S2R, and oxeladin and butamirate were 9 and 10 fold selective over the S2R,

99 respectively. A pseudo Hill slope equal to unity was the preferred model for all three antitussive agents at the S1R and S2R. The EPGN compounds (Series B and C) had subnanomolar to low nanomolar affinity at the S1R, and EPGN745 had the highest affinity overall. For all Series B compounds, a variable slope was the preferred model at the S1R and a pseudo Hill slope equal to unity was the preferred model at the S2R. For the Series C compounds, a variable slope was the preferred model for EPGN1276 and EPGN 863 at the S1R and at the S2R, a pseudo Hill slope equal to unity was the preferred for all the Series C compounds. With the exception of

EPGN794, which had nanomolar affinity at the S2R, all other EPGN compounds had high nanomolar to micromolar affinity at the S2R. The rank order selectivity for the EPGN compounds was EPGN1276 > EPGN862 > EPGN644 ≈ EPGN863 > 745 > EPGN 794.

Once it was established that our compounds interact with the S1R, the functional selectivity of these compounds was evaluated. The prototypical S1R agonist SA4503, and a novel S1R agonist LS-1-137 were shown to stimulate BDNF secretion from neuronal and glial cells, respectively (Fujimoto et al. 2012; Malik et al. 2015), so we tested the ability of our compounds to stimulate BDNF secretion. The neuronal MN9D cell line was selected because preliminary data generated in our lab suggested that compared to the B104 neuronal cell line, which was shown to stimulate BDNF release following stimulation with SA4503 (Fujimoto et al.

2012), the MN9D cell line produced a more robust response. The hypothesis was that if these compounds are S1R agonists, then they should be able to facilitate BDNF secretion. Since BDNF is known to play an important role in neuronal health, growth and differentiation, and intraventricular administration of BDNF stimulates neuronal growth, compounds that can facilitate the release of endogenous BDNF may have therapeutic potential in neurodegenerative disorders. An in situ ELISA approach was utilized because of its improved sensitivity related to

100 its rapid capture of secreted BDNF (Balkowiec and Katz 2000). In addition to producing robust response, the MN9D neuronal cell line also expresses BDNF, S1R (data not shown), and can be stimulated by the S1R agonist 4-PPBP to facilitate BDNF release. Further, with the MN9D cells, secreted BDNF could be detected following 24 hr treatment with the S1R ligand, whereas the

B104 cell line takes 7 days (Fujimoto et al. 2012). All compounds were tested at an equimolar concentration of 10 µM, which represents maximum receptor occupancy (Figure 1 and 2). 4-

PPBP was used as the reference agonist and represented the maximum BDNF secretion response; all compounds were normalized to the 4-PPBP response (Figure 3). Another reference agonist PRE-084 had no BDNF secretion activity. Reference compounds with BDNF secreting activity included (+)-igmesine, donepezil, and the compound with the Cas#1796909-31-3. The amount of BDNF secreted ranged from 24 to 53% of the maximum response (Table 3), where donepezil had the highest response. Further, all compounds had significantly lower efficacy than

4-PPBP (P < 0.05, one-way ANOVA, Bonferroni’s post-hoc). The responses of donepezil and

(+)-igmesine could be reversed at a selective concentration of BD1063 (15 nM). With the exception of butamirate, all other compounds in Series A, B and C were able to stimulate BDNF release, and their efficacy ranged from 54% to 144%. All Series A agonists and EPGN794 (from

Series B) were less efficacious than 4-PPBP (P < 0.05, one-way ANOVA, Bonferroni’s post- hoc), whereas all Series C compounds and EPGN745 and EPGN 862 (from Series B) were just as efficacious as 4-PPBP (P > 0.05, one-way ANOVA, Bonferroni’s post-hoc). With the exception of EPGN794, the agonist responses of all other Series A, B and C compounds could be reversed at a selective concentration of BD1063.

101

S1R ligands like (+)-pentazocine, (+)-igmesine, PRE-084, and SKF-10047 have been characterized as agonist based on their in vivo behavioral profile (see Introduction for the behavioral readouts). BD1063, NE100 and Haloperidol are also established sigma ligands but they block the effects (+)-pentazocine and other S1R agonists, hence they are considered prototypical antagonists (Hayashi and Su 2004; Matsumoto et al. 1995; Okuyama and Nakazato

1996). At the cellular level, S1R agonists have been reported to potentiate IP3-induced calcium mobilization (Hayashi et al. 2000; Hayashi and Su 2007; Hong et al. 2004; Wu and Bowen

2008), which offers another approach for evaluating the function of the S1R ligands. The MCF7 cells (which do not express detectable levels of S1R) and MCF7-hS1R (which have been stably transfected with the human S1R) cells were used to evaluate S1R mediated potentiation of IP3- induced calcium mobilization since this model was previously reported to exhibit this response when stimulated with the S1R agonist (+)-pentazocine (Wu and Bowen 2008). Since these cells express the endogenous Gq-coupled bradykinin B2 receptor (Searovic et al. 2009), bradykinin was used to stimulate the IP3 pathway. Bradykinin concentration response curve was generated

2+ by measuring the change in intracellular calcium as a measure of Gq-PLC-IP3-[Ca ]i signaling pathway activation and bradykinin potencies were obtained for each cell line (Figure 4A). The potencies (EC50) for bradykinin were 2.8 ± 0.3 nM and 0.9 ± 0.1 nM for the MCF7-hS1R and

MCF7 untransfected cell lines, respectively, and were not significantly different from each other

(P > 0.01, t-test). These bradykinin parameters were important to know because a submaximal

IP3 response would allow for the detection of small changes in calcium mobilization induced by

S1R ligands, whereas these small changes are likely to be masked if a saturating bradykinin concentration is used. It was expected that when the cells are pre-treated with the S1R agonists there would be a potentiation of the bradykinin-induced calcium mobilization, and that this

102 response could be reversed by the S1R antagonists, however, this was not the case (Figure 4A).

As an additional control, the MCF7 untransfected cells were utilized, which do not express the

S1R; there should not be any potentiation of the bradykinin-induced IP3 response if the response was truly mediated by the S1R. The response was similar to the MCF7-hS1R cell line, there was no modulation of the IP3 induced calcium mobilization. Hence, this measure is not a reliable method for evaluating the functional pharmacology of the S1R ligands, and as a result novel compounds were not evaluated using this approach.

In the PC12 cell line, S1R agonists like (+)-pentazocine, SA4503 and 4-PPBP were reported to facilitate NGF-induced neurite sprouting (Ishima et al. 2008; Ishima and Hashimoto

2012; Ishima et al. 2014; Nishimura et al. 2008; Rossi et al. 2011; Takebayashi et al. 2002). In an effort to reduce the assay duration, we utilized a modified PC12 cell line called PC-6-15 which overexpresses the trkA receptor, which is a tyrosine kinase receptor for the neurotrophin nerve growth factor (NGF), which is involved in neuronal differentiation (Hempstead et al.

1992). Unlike the PC12 cells, PC-6-15 do not clump when plated which makes neurite quantification more reproducible and reliable, and they are also potently activated by NGF within 12 hours. In contrast, it takes approximately 4-6 days for PC12 cells to produce similar levels of sprouting (Ishima et al. 2014; Rossi et al. 2011; Takebayashi et al. 2002). We first measured the density of the S1R protein in PC-6-15 cell membranes utilizing the [3H]-(+)- pentazocine radioligand, and the Bmax (maximum number of binding site) was determined to be

0.24 ± 0.01 pmoles/mg membrane protein (Figure 5A). To identify a submaximal concentration of NGF that induced neurite sprouting, cells were treated with 0.3 to 100 ng/mL NGF, and the neurite length/cell was quantified by tracing individual neurites to obtain the total neurite length, and dividing this number by the total number of cells (Figure 5B). NGF increased the total

103 neurite length in a concentration dependent manner, and since 0.3 ng/mL was the lowest concentration tested that induced neurite sprouting, it was used to test the potentiation effects of

S1R ligands. When these cells were treated with the S1R ligands alone, there was no significant effect on neurite sprouting. When co-treated with 0.3 ng/mL NGF, S1R ligands did not potentiate NGF induced sprouting. Overall, out of the three functional assays tested, in our hands, only BDNF secretion seems to be a reliable measure for assessing functionally selective

S1R agonist activity.

Discussion

In this study the structure-activity relationship (SAR) of three different structural classes of S1R ligands were evaluated, which lead to the discovery of new compounds with high S1R selectivity over the S2R. Three different in vitro functional assays were utilized to evaluate the functions of those compounds at the S1R. Most but not all of the S1R ligands tested had BDNF secreting property indicating that these ligands are activating functionally selective pathways that is responsible for the BDNF secreting activity.

Though the Ki for 4-PPBP, PRE-084, NE-100 and BD1063 at the S1R and S2R have been previously reported (Garces-Ramirez et al. 2011; Lee et al. 2008; Okuyama and Nakazato

1996; Whittemore et al. 1997), they were repeated in this report to validate our assay model and to do a parallel comparison of the affinities using the same cell background (See Results section,

1st paragraph for model detail on assay model). To evaluate the selectivity of these compounds inhibition curves were generated at the S1R and S2R. The S1R and S2R affinity values for 4-

PPBP, PRE-084, NE-100 and BD1063 reported here are comparable to those reported previously

(Garces-Ramirez et al. 2011; Lee et al. 2008; Okuyama and Nakazato 1996; Whittemore et al.

1997), thus validating the approach. It is interesting to note that though 4-PPBP had no

104 selectivity over the S2R, at the S1R it had a very steep pseudo Hill slope (-3.1) which indicates positive cooperativity, hence the S1R has more than one binding sites for 4-PPBP. In contrast,

PRE-084 had a pseudo Hill of slope of -0.98 which was not different from unity suggesting binding to a single site on the S1R. Both 4-PPBP and PRE-084 are considered S1R agonists, however, when BDNF secretion was used as a measure of S1R activation, 4-PPBP induced a robust BDNF secretion response, whereas PRE-084 had no activity. One might speculate then that 4-PPBP is likely interacting with a S1R binding site different from PRE-084 which may account for the functionally selective profile of these compounds. Further, though 4-PPBP is not a S1R-selective compound, its effect on BDNF secretion could be reversed by a S1R-selective concentration of BD1063 suggesting that the response is likely being mediated by the S1R and not the S2R.

To identify different structural features that impact selectivity and have the desired functional output, the structure-activity space of two different structural classes of ligands were evaluated. The compounds in Series A are known S1R ligands and have been classified as agonists but to our knowledge their affinity at the S2R has not been evaluated. The structures of carbetapentane, butamirate and oxeladin only differ by the type of alkyl modification at one location (Table 2). Carbetapentane has a cyclopentyl group, and is only three fold selective over the S2R, and its efficacy for BDNF secretion is 72% of the maximum response. Changing the cyclopentyl (carbetapentane) group to a diethyl (oxeladin) or an ethyl (butamirate) group increased the selectivity by 2 and 3-fold, respectively, over carbetapentane. However, these changes also reduced the BDNF secretion activity. Oxeladin had slightly reduced BDNF secretion activity (55%), thus, the presence of the diethyl group caused an 18% reduction in efficacy relative to carbetapentane. Butamirate has an ethyl group, and this completely abolished

105 the BDNF secretion activity. These modifications had very little to no effect on the affinity at the

S1R, but moderately decrease the affinity at the S2R suggesting that this region exerts only modest control on selectivity for S1R over S2R. Further, bulky alkyl groups (e.g. cyclopentyl or diethyl) at that position may be necessary for activating the BDNF secretion pathway.

Series B compounds represent a novel class of sigma ligands and they also contain substructural modifications at a single location. Out of the three compounds in this series,

EPGN794, which contained an alkane group in the substructural region of interest, was the least selective compound. When this alkane group is changed to an amide (EPGN862), the affinity at the S1R does not change but the affinity at the S2R is decreased by 98-fold, making this compound more than 600-fold selective for the S1R over the S2R. When the alkane in EPGN794 is changed to an ether group (EPGN745), there is no effect on S1R affinity, but the S2R affinity is decreased by 5 fold. The ether modification only increased the S1R over S2R selectivity by 7 fold relative to EPGN794. When these compounds were tested in the BDNF secretion assay,

EPGN794 had 54% efficacy. The amide and ether modifications of EPGN862 and EPGN745 increased the efficacy of BDNF secretion to 84% and 114%, respectively. Though the amount of

BDNF secreted by EPGN862 and EPGN745 are not statistically different from 4-PPBP or each other, there is a nice correlation between the increase in selectivity and an increase in BDNF secretion. Further, EPGN794 and EPGN745 had steep pseudo Hill slopes at the S1R in the radioligand binding assays, where EPGN794 had the steepest slope of -2.9, and EPGN745 had a slope of -1.9, which were significantly different from unity (P < 0.001, t-test). EPGN862 had a pseudo Hill slope of -1.3, which was not significantly different from unity (P > 0.001, t-test).

One might speculate then that the Series B compounds, especially EPGN745 and EPGN862 may

106 be interacting with the S1R in a similar manner as 4-PPBP, which may be why their functional response is similar to 4-PPBP in the BDNF functional assay.

In Series C, the effect of a fluorine substitution was evaluated. EPGN644, which had no fluorine, was 160-fold selective for the S1R over the S2R. Addition of fluorine at position X

(EPGN863) had no effect on selectivity, and was similar to EPGN644. Addition of fluorine at position Y (EPGN1276) had no effect on the S1R affinity but caused a large reduction in the

S2R affinity, making it the most selective compound in this study. In the BDNF functional assay,

EPGN644 had 78% efficacy, where as EPGN863 and EPGN1276 had efficacies of 1.2 and 1.4, respectively. Though the magnitude of BDNF secretion by EPGN644 is lower than 4-PPBP and

EPGN1276 and EPGN863 have higher efficacy than 4-PPBP, the efficacies for all series C compounds were not statistically different from 4-PPBP. Hence, fluorine substitution does not appear to have any impact on BDNF secretion but is controlling the selectivity.

In addition to modulating BDNF secretion, classical sigma agonists haven been reported to modulate IP3 induced calcium response (Hayashi et al. 2000; Hayashi and Su 2007; Hong et al. 2004; Wu and Bowen 2008). In NG108 (neuroblastoma-glioma hybrid cell line) and SH-

SY5Y (neuronal) cells, (+)-pentazocine and PRE-084 were reported to potentiate bradykinin- induced increases in intracellular calcium and this effect was reversed by treatment with the antagonists NE-100 and haloperidol (Hayashi et al. 2000; Hong et al. 2004). Similarly, in the

MCF7 cells transfected with the S1R, (+)-pentazocine increased bradykinin induced calcium release and this response was reduced in the presence of BD1063 (Wu and Bowen 2008).

Further, BD1063 on its own was able to reduce the bradykinin induced calcium response suggesting that it functions as an inverse agonist at the S1R (Wu and Bowen 2008). When some of the same sigma ligands were tested in the untransfected MCF7 cells, no effect on bradykinin

107 induced calcium mobilization was observed (Figure 4B). The lack of an effect makes sense if the effects are due to the S1R, since the untransfected MCF7 cells lack detectable levels of the S1R

(Schetz et al. 2007; Vilner et al. 1995b; Wu and Bowen 2008). However, when the effects of the sigma reference ligands 4-PPBP, PRE-084, (+)-SKF10047, BD1063 and NE100 were tested, in the MCF7-hS1R cell line, no significant potentiation or suppression of the bradykinin response by the reference sigma agonists or antagonists, respectively, was detected. In other words, the results from the MCF-h-S1R cell line looked similar to the untransfected MCF7 cell line. The lack of any S1R response in this system is not due to low expression of the S1R because the Bmax for S1R in the MCF7-hS1R cell line was 109 ±23.7 pmol/mg (Lee et al. 2008) whereas the S1R

Bmax in Wu and Bowen’s model was 31.4 ± 4.5 pmol/mg (Wu and Bowen 2008). The major difference between our study and Wu and Bowen study is that they used (+)-pentazocine as their reference agonist. It is possible that the reference agonists we tested do not have the same functionally selective profile as (+)-pentazocine, which might explain why the S1R mediated potentiation of the bradykinin response was not detected in this study. However, Wu and Bowen had shown that BD1063 can suppress the bradykinin response in the S1R-expressing cell line but not in the untransfected cell line. When the same experiment was performed in this study, the

BD1063 inverse agonist response in the S1R expressing cell line was not reproducible. Further, since the MCF7-hS1R had a higher Bmax than Wu and Bowen’s S1R expressing MCF7 cell line, it would be expected that the inverse agonist response would stronger than that observed by Wu ad Bowen. Also, in the NG108 cell line, PRE-084 was reported to potentiate the bradykinin- induced calcium response (Hayashi et al. 2000), but the result was not reproducible this in this report, using the MCF7-hS1R model. It is possible that the MCF7 cells (a human breast cancer cell line) do not have the same machinery as the NG108 cell line (hybrid of mouse

108 neuroblastoma and rat glioma) which is why it might not be able to produce this response. If the

S1R mediated modulation of the IP3-induced calcium response only requires the presence of BiP,

IP3R and the S1R, then the MCF7-hS1R model should have all the necessary components to elicit the same response as the NG108 cell line. The MCF7 cells express BiP (Fu et al. 2007),

IP3R as evidenced by the bradykinin induced Gq response, and S1R was overexpressed in these cells (Schetz et al. 2007),

The idea of S1R agonists potentiating IP3 mediated calcium response by stabilizing the

IP3 receptor (IP3R) originates from the work of Hayashi and Su, 2007. In that report they demonstrated that S1R agonists (+)-pentazocine, PRE-084, and SKF10047 (all previously characterized as agonists based on their in vivo behavioral profile) caused the S1R to dissociate from the chaperone protein BiP and interact with the type 3 IP3R (IP3R3); treatment with the S1R antagonists NE-100 and haloperidol had no effect BiP-S1R interaction (Hayashi and Su 2007).

Further they showed that there was a correlation between IP3R3-S1R interaction and an increase in calcium mobilization from the ER to the mitochondria in the presence of IP3. Since the IP3R3s are predominantly localized at the mitochondria-associated ER membranes (MAM) and provide direct connection for ER-mitochondria calcium signaling (Mendes et al. 2005), then it seems reasonable stabilization of the IP3R3 by the S1R would potentiate calcium mobilization into the mitochondria. Cytosolic calcium mobilization is mediated primarily via the type 1 IP3R (IP3R1)

(Mendes et al. 2005) and there is no evidence that the activated S1R interacts with and stabilizes

IP3R1. Since cytosolic calcium was measured in this study, it is possible that the S1R mediated potentiation on the IP3 response cannot be detected by measuring change in cytoplasmic calcium because the S1R is not interacting with IP3R1 but rather with the IP3R3 and mobilizing calcium from the ER to the mitochondria. Additionally, since the IP3R1-S1R interaction has not been

109 studied, it is possible that the S1R does not interact with or stabilize IP3R1. The other possibility is, S1R could be interacting with IP3R1, but the interaction alone may not be enough to cause an increase in calcium mobilization, and that additional stimulus like ER stress may be required. If a strong correlation between S1R agonists and an increase in cytosolic calcium mobilization was established here, then it would have been a useful tool for identifying S1R agonists. However, it

2+ appears that the S1R-IP3R-Ca relationship is not that simple and cannot be used as a reliable measure for evaluating S1R functional pharmacology (i.e., agonist/antagonist/inverse agonist properties) using changes cytosolic calcium as the outcome measure.

S1R agonists have been reported to modulate neurite sprouting (Ishima et al. 2008;

Ishima and Hashimoto 2012; Ishima et al. 2014; Nishimura et al. 2008; Rossi et al. 2011;

Takebayashi et al. 2002). For example, (+)-pentazocine, 4-PPBP, and PRE-084 were shown to potentiate NGF-induced neurite sprouting in the PC12 cell line (Nishimura et al. 2008; Rossi et al. 2011; Takebayashi et al. 2002). PC12 cells normally take approximately 5 days to sprout measurable neurites following chronic exposure to NGF and all studies to date using PC12 cells involved 4-5 day treatments. In an effort to reduce the duration of the assay and make it higher throughput, we utilized a variant of the PC12 cell line (PC-6-15) that overexpresses the trkA receptor (Hempstead et al. 1992). The PC-6-15 cell line produces long neurites (neurite length greater than two cell bodies) within 2 hours of NGF treatment, and after 24 hrs more than 90% of cells have formed dense neuronal processes. The expectation was that this cell line could be used to shorten the assay time for evaluating the effects of S1R ligands on neurite sprouting.

However, the reference S1R agonist 4-PPBP was unable to stimulate neurite sprouting, nor did it potentiate NGF induced sprouting. Other S1R agonists like PRE-084, carbetapentane and fluvoxamine were also tested, and none of them had any effect on sprouting on their own or on

110

NGF-induced sprouting (data not shown) in the PC-6-15 cell line. This response is not due the lack of S1R in these cells, because the radioligand binding data using the S1R selective [3H]-(+)- pentazocine as the radioligand clearly shows that the S1R protein is present in these cells. It is possible that the S1R is not a major contributor in the neurite sprouting process (i.e., it only has a minor effect on neurite sprouting), and that the neurite tracing method is not sensitive enough to detect those small changes. The other possibility is that overexpressing the trkA receptor is somehow decreasing the expression of the S1R by either decreasing gene expression, translation, and / or targeting the protein for degradation, and as a result the S1R levels may not high enough to produce a measurable change in neurite sprouting. It is also possible that the overexpression of the trkA receptor resulted in a NGF ceiling effect due to the presence of NGF in the media.

Regardless of what the explanation may be, the PC-6-15 is not a suitable model for evaluating the role of the S1R on neurite sprouting, however it is an excellent model for evaluating small

NGF-like molecules targeting the trkA receptor.

Conclusion

In this study, novel S1R ligands capable of stimulating BDNF secretion from a neuronal cell line were identified. Further, four new sigma ligands (EPGN compounds) with more than

100-fold selectivity over the S2R were identified. These compounds were categorized as functionally selective agonists of the BDNF secretion pathway. S1R agonists have also been reported to modulate IP3-induced calcium mobilization and NGF-induced neurite sprouting. To determine if the novel S1R ligands can activate additional pathways, S1R reference ligands were evaluated for their ability to modulate IP3-mediated calcium mobilization and potentiation of

NGF-induced neurite sprouting. However, when we attempted to reproduce the published S1R-

2+ IP3-Ca response using prototypical S1R agonists, we were unable to reproduce the results and

111 eventually reached the conclusion that it is not reliable measure for evaluating S1R ligands using changes in cytosolic calcium as the outcome measure. For measuring the effects of S1R ligands on neurite sprouting we utilized a modified PC12 cell model that overexpresses the trkA receptor because a 5 day assay is not a time productive option for evaluating large number of compounds.

The overexpression system we used allowed us to reduce the duration of the assay to one day.

Potentiation of NGF-induced neurite sprouting could not be detected for any of the reference

S1R agonists tested. This may be because the PC-6-15 cell model used is not appropriate for testing this parameter for reason not yet known.

112

Figure 1. Prototypical sigma receptor ligands PRE-084, NE-100 and BD1063 have greater selectivity for S1R over S2R, whereas 4-PPBP does not. A, Competition binding of prototypical sigma receptor ligands with [3H]-(+)-pentazocine at the human S1R. The rank order affinities were as follows: 4-PPBP ≈ NE-100 ≈ BD1063 > PRE-084. 4-PPBP, NE100 and

BD1063 had low nanomolar affinities for the S1R (range: 1.0 to 5.6 nM) whereas PRE-084 had high nanomolar affinity the S1R (92 nM) A variable slope was the preferred model for 4-PPBP and NE-100. B, Competition binding of prototypical sigma receptor ligands with [3H]-DTG at the human S2R. The affinities ranged from 0.8 nM to 13,070 nM, and the rank order affinities were as follows: 4-PPBP > NE-100 > BD1063 > PRE-084. A pseudo Hill slope equal to unity was the preferred model for all ligands. Affinity values are listed in Table 1. Values are averages of 2-3 experiments ± SEM, and each experiment was performed in triplicate.

113

Figure 1. Prototypical sigma receptor ligands PRE-084, NE-100 and BD1063 have greater

selectivity for S1R over S2R, whereas 4-PPBP does not.

114

Figure 2. EPGN745, 862, 863, 644 and 1276 are highly selective for the S1R. Competition binding of sigma receptor ligands with [3H]-(+)-pentazocine at the human S1R and with [3H]-

DTG at the human S2R. A, B, Carbetapentane, butamirate and oxeladin had only 3 to 10 fold greater selectivity for the S1R over the S2R. C, D, In series B, EPGN794 had only 7 fold selectivity over the S2R, whereas EPGN862 and EPGN745 were 649 and 49 fold selective over the S2R, respectively. E, F, In series C, EPGN644 and 863 were > 100 fold selective, whereas

EPGN1276 was > 1000 fold selective. Affinity values are listed in Table 2. Values are averages of 2-3 experiments ± SEM, and each experiment was performed in triplicate.

115

A B

Series A Series

C D

Series B Series

E F

Series C Series

Figure 2. EPGN745, 862, 863, 644 and 1276 are highly selective for the S1R.

116

Figure 3. All compounds in series A, B and C facilitate S1R mediated BDNF secretion from the neuronal MN9D cell line. Levels of secreted BDNF were measured using in situ ELISA.

All compounds were tested at 10 µM and were reversed with 15 nM BD1063. Data was normalized to the response produced by 10 µM 4-PPBP. EPGN863 and PRE-084 data was generated by others in the lab but shown here completeness. Values are averages of 2-10 experiments ± SEM, and each experiment was performed in triplicate. Statistical significance at

P < 0.05 was determined by ANOVA followed by a Bonferroni’s post-hoc test and are marked with an asterisk. Efficacy of S1R ligands tested here can be found in Table 3.

117

Figure 3. All compounds in series A, B and C facilitate S1R mediated BDNF secretion from

the neuronal MN9D cell line.

118

2+ Figure 4. S1R does not appear to modulate Gq-PLC-IP3-[Ca ]i mediated calcium release.

A, MCF7 cells express the endogenous bradykinin receptor B2 (Searovic et al. 2009).

Bradykinin (BDK) concentration response curves were generated for the untransfected MCF7 cells and MCF7-hS1R by measuring changes in intracellular calcium as a measure of Gq- activation. The potencies were 2.8 ± 0.3 nM for the MCF7-hS1R cell line and 0.9 ± 0.1 nM for the untransfected MCF7 cells. The potencies were not significantly from one another (P > 0.01, t-test). B, Activation of the S1R by prototypical sigma ligands did not have any effect on the Gq-

PLC-IP3 mediated calcium response. Data presented as mean ± SEM. Values are averages of 2 experiments ± SEM, and each experiment was performed in triplicate.

119

2+ Figure 4. S1R does not appear to modulate Gq-PLC-IP3-[Ca ]i mediated calcium release.

120

Figure 5. S1R activation has no effect on neurite sprouting, nor does it potentiate NGF induced neurite sprouting in PC-6-15. A, The maximum number of binding sites (Bmax) in PC-

6-15 membranes was estimated from measurements of 10 nM of specifically bound [3H]-(+)- pentazocine followed by calculation of the Bmax at saturation using a square hyperbola model as describe in the methods. B, NGF induces neurite sprouting in concentration dependent manner in the PC-6-15 cells. C, PC-6-15 cells were maintained in media with 0.5% serum and treated with either vehicle (1:1000 v/v DMSO), 0.3 ng/mL NGF, sigma ligands, or sigma ligands + 0.3 ng/mL NGF for 24 hours. Treatment with 0.3 ng/mL NGF resulted in a significant increase neurite sprouting, but sigma ligands had no effect on neurite sprouting nor did they potentiate

NGF induced neurite sprouting. Data presented as mean ± SEM. * P < 0.05 by One-way

ANOVA, Bonferroni’s post-hoc test. Values are averages of 3-4 experiments, and each experiment was performed in triplicate.

121

A B

Figure 5. S1R activation has no effect on neurite sprouting, nor does it potentiate NGF

induced neurite sprouting in PC-6-15.

122

Table 1. PRE-084 is the most selective prototypical agonist and BD1063 is the most selective prototypical antagonist. Affinity (Ki) values presented as mean ± SEM.

S1R affinity, S2R affinity, Compound Structure Ki S2R / Ki S1R Ki (nM ± SEM) Ki (nM ± SEM)

4-PPBP 1.0 ± 0.3 0.8 ± 0.2 0.8

PRE-084 92 ± 15 13070 ± 1725 142

(+)-Igmesine 9.2 ± 2.2 164 ± 29 18

BD1063 5.6 ± 0.5 210 ± 23 38

NE-100 1.9 ± 0.4 41.4 ± 4.6 21

123

Table 2. Effect of specific substructural modifications on selectivity for S1R over S2R. In series A, the modifications had a very small impact on selectivity. Changing the cyclopentane group to diethyl or an ethyl group only increased the selectivity by approximately 2 and 3-fold, respectively, over carbetapentane. In series B, the ether and carboxamide modifications had a large impact on selectivity, increasing the selectivity by 7 and 93 fold, respectively, over

EPGN794. In series C, fluorine at position Y (EPGN1276) increased the selectivity of the compound by 7 fold compared to no fluorine (EPGN644). Affinity (Ki) values presented as the mean ± SEM.

Substructural S1R affinity S2R affinity S2R/S1R ratio Compound feature (Ki, nM ± SEM) (Ki, nM ± SEM) (Ki S2R / Ki S1R) Series A Carbetapentane 19.8 ± 2.3 55.7 ± 3.8 3

Oxeladin 25 ± 4.3 148 ± 64 6

Butamirate 17.2 ± 1.7 214 ± 41 10

Series B EPGN794 3.7 ± 1.5 26.8 ± 7.5 7

EPGN745 2.6 ± 0.6 127.3 ± 18.9 49

EPGN862 4.0 ± 0.4 2627.0 ± 475.0 649

Series C EPGN644 No F 5.2 ± 0.8 831.7 ± 139.1 160 EPGN863 F at position X 4.4 ± 0.8 581.5 ± 110.1 132 EPGN1276 F at position Y 1.6 ± 0.3 1714.0 ± 731.1 1068

124

Table 3. Efficacy of S1R ligands that function as agonists of BDNF secretion. Values are normalized to the max response of 10 µM 4-PPBP. Efficacies significantly different from 4-

PPBP are marked with an asterisk (P < 0.05, one-way ANOVA, Bonferroni’s post-hoc). All other efficacies were not significantly different. Values represent mean ± SEM.

Compound Efficacy of BDNF secretion (relative to 10 µM 4-PPBP) 4-PPBP 1 Cas# 1796909-31-3 0.24 ± 0.04* (+)-igmesine 0.42 ± 0.03* Donepezil 0.53 ± 0.08* Carbetapentane 0.73 ± 0.05* Oxeladin 0.55 ± 0.04* EPGN745 1.1 ±0.1 EPGN794 0.54 ± 0.01* EPGN862 0.84 ± 0.12 EPGN644 0.78 ± 0.09 EPGN1276 1.4 ± 0.2 EPGN863 1.2 ± 0.3

125

References

Al-Nabulsi I., Mach R. H., Wang L. M., Wallen C. A., Keng P. C., Sten K., Childers S. R. and

Wheeler K. T. (1999) Effect of ploidy, recruitment, environmental factors, and tamoxifen

treatment on the expression of sigma-2 receptors in proliferating and quiescent tumor cells.

Br. J. Cancer. 81, 925-933.

Alonso G., Phan V., Guillemain I., Saunier M., Legrand A., Anoal M. and Maurice T. (2000)

Immunocytochemical localization of the sigma(1) receptor in the adult rat central nervous

system. Neuroscience. 97, 155-170.

Alzheimer's Association (2015a) 2015 Alzheimer's Disease Facts and Figures. 2015.

Alzheimer's Association (2015b) Medications for memory loss. 2015.

Balkowiec A. and Katz D. M. (2000) Activity-dependent release of endogenous brain-derived

neurotrophic factor from primary sensory neurons detected by ELISA in situ. J. Neurosci.

20, 7417-7423.

Bowen W. D. (2000) Sigma receptors: recent advances and new clinical potentials. Pharm. Acta

Helv. 74, 211-218.

Brown C., Fezoui M., Selig W. M., Schwartz C. E. and Ellis J. L. (2004) Antitussive activity of

sigma-1 receptor agonists in the guinea-pig. Br. J. Pharmacol. 141, 233-240.

Cunha C., Brambilla R. and Thomas K. L. (2010) A simple role for BDNF in learning and

memory? Front. Mol. Neurosci. 3, 1.

Drugs.com (2014) Carbetapentane. 2016.

Ehret M. J. and Chamberlin K. W. (2015) Current Practices in the Treatment of Alzheimer

Disease: Where is the Evidence After the Phase III Trials? Clin. Ther.

126

Fu Y., Li J. and Lee A. S. (2007) GRP78/BiP inhibits endoplasmic reticulum BIK and protects

human breast cancer cells against estrogen starvation-induced apoptosis. Cancer Res. 67,

3734-3740.

Fujimoto M., Hayashi T., Urfer R., Mita S. and Su T. P. (2012) Sigma-1 receptor chaperones

regulate the secretion of brain-derived neurotrophic factor. Synapse. 66, 630-639.

Garces-Ramirez L., Green J. L., Hiranita T., Kopajtic T. A., Mereu M., Thomas A. M.,

Mesangeau C., Narayanan S., McCurdy C. R., Katz J. L. and Tanda G. (2011) Sigma

receptor agonists: receptor binding and effects on mesolimbic dopamine neurotransmission

assessed by microdialysis. Biol. Psychiatry. 69, 208-217.

Gekker G., Hu S., Sheng W. S., Rock R. B., Lokensgard J. R. and Peterson P. K. (2006)

Cocaine-induced HIV-1 expression in microglia involves sigma-1 receptors and

transforming growth factor-beta1. Int. Immunopharmacol. 6, 1029-1033.

Guitart X., Codony X. and Monroy X. (2004) Sigma receptors: biology and therapeutic potential.

Psychopharmacology (Berl). 174, 301-319.

Hashimoto K., Scheffel U. and London E. D. (1995) In vivo labeling of sigma receptors in

mouse brain with [3H]4-phenyl-1-(4-phenylbutyl)piperidine. Synapse. 20, 85-90.

Hayashi T., Maurice T. and Su T. P. (2000) Ca(2+) signaling via sigma(1)-receptors: novel

regulatory mechanism affecting intracellular Ca(2+) concentration. J. Pharmacol. Exp. Ther.

293, 788-798.

Hayashi T. and Su T. P. (2007) Sigma-1 receptor chaperones at the ER-mitochondrion interface

regulate Ca(2+) signaling and cell survival. Cell. 131, 596-610.

127

Hayashi T. and Su T. P. (2004) Sigma-1 receptors at galactosylceramide-enriched lipid

microdomains regulate oligodendrocyte differentiation. Proc. Natl. Acad. Sci. U. S. A. 101,

14949-14954.

Hellewell S. B. and Bowen W. D. (1990) A sigma-like binding site in rat pheochromocytoma

(PC12) cells: decreased affinity for (+)-benzomorphans and lower molecular weight suggest

a different sigma receptor form from that of guinea pig brain. Brain Res. 527, 244-253.

Hempstead B. L., Rabin S. J., Kaplan L., Reid S., Parada L. F. and Kaplan D. R. (1992)

Overexpression of the trk tyrosine kinase rapidly accelerates nerve growth factor-induced

differentiation. Neuron. 9, 883-896.

Hong W., Nuwayhid S. J. and Werling L. L. (2004) Modulation of bradykinin-induced calcium

changes in SH-SY5Y cells by neurosteroids and sigma receptor ligands via a shared

mechanism. Synapse. 54, 102-110.

Hornick J. R., Vangveravong S., Spitzer D., Abate C., Berardi F., Goedegebuure P., Mach R. H.

and Hawkins W. G. (2012) Lysosomal membrane permeabilization is an early event in

Sigma-2 receptor ligand mediated cell death in pancreatic cancer. J. Exp. Clin. Cancer Res.

31, 41-9966-31-41.

Huang Y. S., Lu H. L., Zhang L. J. and Wu Z. (2014) Sigma-2 receptor ligands and their

perspectives in cancer diagnosis and therapy. Med. Res. Rev. 34, 532-566.

Iniguez M. A., Punzon C., Nieto R., Burgueno J., Vela J. M. and Fresno M. (2013) Inhibitory

effects of sigma-2 receptor agonists on T lymphocyte activation. Front. Pharmacol. 4, 23.

Ishima T., Fujita Y. and Hashimoto K. (2014) Interaction of new antidepressants with sigma-1

receptor chaperones and their potentiation of neurite outgrowth in PC12 cells. Eur. J.

Pharmacol. 727, 167-173.

128

Ishima T. and Hashimoto K. (2012) Potentiation of nerve growth factor-induced neurite

outgrowth in PC12 cells by ifenprodil: the role of sigma-1 and IP3 receptors. PLoS One. 7,

e37989.

Ishima T., Nishimura T., Iyo M. and Hashimoto K. (2008) Potentiation of nerve growth factor-

induced neurite outgrowth in PC12 cells by donepezil: role of sigma-1 receptors and IP3

receptors. Prog. Neuropsychopharmacol. Biol. Psychiatry. 32, 1656-1659.

Jiang G., Mysona B., Dun Y., Gnana-Prakasam J. P., Pabla N., Li W., Dong Z., Ganapathy V.

and Smith S. B. (2006) Expression, subcellular localization, and regulation of sigma

receptor in retinal muller cells. Invest. Ophthalmol. Vis. Sci. 47, 5576-5582.

Jin J. L., Fang M., Zhao Y. X. and Liu X. Y. (2015) Roles of sigma-1 receptors in Alzheimer's

disease. Int. J. Clin. Exp. Med. 8, 4808-4820.

Kikuchi-Utsumi K. and Nakaki T. (2008) Chronic treatment with a selective ligand for the

sigma-1 receptor chaperone, SA4503, up-regulates BDNF protein levels in the rat

hippocampus. Neurosci. Lett. 440, 19-22.

Klemas N. and Dowling R. (2015) Clinical Economics. Alzheimer's Disease. Med. Econ. 92, 14-

5, 17.

Klette K. L., DeCoster M. A., Moreton J. E. and Tortella F. C. (1995) Role of calcium in sigma-

mediated neuroprotection in rat primary cortical neurons. Brain Res. 704, 31-41.

Korte M., Carroll P., Wolf E., Brem G., Thoenen H. and Bonhoeffer T. (1995) Hippocampal

long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor. Proc.

Natl. Acad. Sci. U. S. A. 92, 8856-8860.

Kourrich S., Su T. P., Fujimoto M. and Bonci A. (2012) The sigma-1 receptor: roles in neuronal

plasticity and disease. Trends Neurosci. 35, 762-771.

129

Lee I. T., Chen S. and Schetz J. A. (2008) An unambiguous assay for the cloned human sigma1

receptor reveals high affinity interactions with dopamine D4 receptor selective compounds

and a distinct structure-affinity relationship for butyrophenones. Eur. J. Pharmacol. 578,

123-136.

Lewin G. R. and Barde Y. A. (1996) Physiology of the neurotrophins. Annu. Rev. Neurosci. 19,

289-317.

Lombardo S. and Maskos U. (2015) Role of the nicotinic acetylcholine receptor in Alzheimer's

disease pathology and treatment. Neuropharmacology. 96, 255-262.

Malik M., Rangel-Barajas C., Sumien N., Su C., Singh M., Chen Z., Huang R. Q., Meunier J.,

Maurice T., Mach R. H. and Luedtke R. R. (2015) The effects of sigma (sigma1) receptor-

selective ligands on muscarinic receptor antagonist-induced cognitive deficits in mice. Br. J.

Pharmacol. 172, 2519-2531.

Marrazzo A., Caraci F., Salinaro E. T., Su T. P., Copani A. and Ronsisvalle G. (2005)

Neuroprotective effects of sigma-1 receptor agonists against beta-amyloid-induced toxicity.

Neuroreport. 16, 1223-1226.

Matsumoto R. R., Bowen W. D., Tom M. A., Vo V. N., Truong D. D. and De Costa B. R. (1995)

Characterization of two novel sigma receptor ligands: antidystonic effects in rats suggest

sigma receptor antagonism. Eur. J. Pharmacol. 280, 301-310.

Matsuno K., Nakazawa M., Okamoto K., Kawashima Y. and Mita S. (1996) Binding properties

of SA4503, a novel and selective sigma 1 receptor agonist. Eur. J. Pharmacol. 306, 271-279.

Maurice T., Su T. P. and Privat A. (1998) Sigma1 (sigma 1) receptor agonists and neurosteroids

attenuate B25-35-amyloid peptide-induced amnesia in mice through a common mechanism.

Neuroscience. 83, 413-428.

130

Mendes C. C., Gomes D. A., Thompson M., Souto N. C., Goes T. S., Goes A. M., Rodrigues M.

A., Gomez M. V., Nathanson M. H. and Leite M. F. (2005) The type III inositol 1,4,5-

trisphosphate receptor preferentially transmits apoptotic Ca2+ signals into mitochondria. J.

Biol. Chem. 280, 40892-40900.

Monassier L., Manoury B., Bellocq C., Weissenburger J., Greney H., Zimmermann D., Ehrhardt

J. D., Jaillon P., Baro I. and Bousquet P. (2007) sigma(2)-receptor ligand-mediated

inhibition of inwardly rectifying K(+) channels in the heart. J. Pharmacol. Exp. Ther. 322,

341-350.

Nishimura T., Ishima T., Iyo M. and Hashimoto K. (2008) Potentiation of nerve growth factor-

induced neurite outgrowth by fluvoxamine: role of sigma-1 receptors, IP3 receptors and

cellular signaling pathways. PLoS One. 3, e2558.

Okuyama S. and Nakazato A. (1996) NE100: A novel sigma receptor antagonist. CNS Drug

Reviews. 2, 226-237.

Ostenfeld M. S., Fehrenbacher N., Hoyer-Hansen M., Thomsen C., Farkas T. and Jaattela M.

(2005) Effective tumor cell death by sigma-2 receptor ligand siramesine involves lysosomal

leakage and oxidative stress. Cancer Res. 65, 8975-8983.

Palacios G., Muro A., Vela J. M., Molina-Holgado E., Guitart X., Ovalle S. and Zamanillo D.

(2003) Immunohistochemical localization of the sigma1-receptor in oligodendrocytes in the

rat central nervous system. Brain Res. 961, 92-99.

Palacios G., Muro A., Verdu E., Pumarola M. and Vela J. M. (2004) Immunohistochemical

localization of the sigma1 receptor in Schwann cells of rat sciatic nerve. Brain Res. 1007,

65-70.

131

Peviani M., Salvaneschi E., Bontempi L., Petese A., Manzo A., Rossi D., Salmona M., Collina

S., Bigini P. and Curti D. (2014) Neuroprotective effects of the Sigma-1 receptor (S1R)

agonist PRE-084, in a mouse model of motor neuron disease not linked to SOD1 mutation.

Neurobiol. Dis. 62, 218-232.

Prince M., Wimo A., Guerchet M., Ali G. C., Wu Y. T. and Prina M. (2015) World Alzheimer

Report 2015: The global impact of dementia, an analysis of prevalence, incidence, cost and

trends. .

Robson M. J., Turner R. C., Naser Z. J., McCurdy C. R., O'Callaghan J. P., Huber J. D. and

Matsumoto R. R. (2014) SN79, a sigma receptor antagonist, attenuates methamphetamine-

induced astrogliosis through a blockade of OSMR/gp130 signaling and STAT3

phosphorylation. Exp. Neurol. 254, 180-189.

Rossi D., Pedrali A., Urbano M., Gaggeri R., Serra M., Fernandez L., Fernandez M., Caballero

J., Ronsisvalle S., Prezzavento O., Schepmann D., Wuensch B., Peviani M., Curti D.,

Azzolina O. and Collina S. (2011) Identification of a potent and selective sigma(1) receptor

agonist potentiating NGF-induced neurite outgrowth in PC12 cells. Bioorg. Med. Chem. 19,

6210-6224.

Ruscher K. and Wieloch T. (2015) The involvement of the sigma-1 receptor in

neurodegeneration and neurorestoration. J. Pharmacol. Sci. 127, 30-35.

Schetz J. A., Perez E., Liu R., Chen S., Lee I. and Simpkins J. W. (2007) A prototypical Sigma-1

receptor antagonist protects against brain ischemia. Brain Res. 1181, 1-9.

Schneider C. A., Rasband W. S. and Eliceiri K. W. (2012) NIH Image to ImageJ: 25 years of

image analysis. Nat. Methods. 9, 671-675.

132

Searovic P., Alonso M., Oses C., Pereira-Flores K., Velarde V. and Saez C. G. (2009) Effect of

tamoxifen and retinoic acid on bradykinin induced proliferation in MCF-7 cells. J. Cell.

Biochem. 106, 473-481.

Su T. P., Hayashi T., Maurice T., Buch S. and Ruoho A. E. (2010) The sigma-1 receptor

chaperone as an inter-organelle signaling modulator. Trends Pharmacol. Sci. 31, 557-566.

Takebayashi M., Hayashi T. and Su T. P. (2002) Nerve growth factor-induced neurite sprouting

in PC12 cells involves sigma-1 receptors: implications for antidepressants. J. Pharmacol.

Exp. Ther. 303, 1227-1237.

Villard V., Espallergues J., Keller E., Alkam T., Nitta A., Yamada K., Nabeshima T.,

Vamvakides A. and Maurice T. (2009) Antiamnesic and neuroprotective effects of the

aminotetrahydrofuran derivative ANAVEX1-41 against amyloid beta(25-35)-induced

toxicity in mice. Neuropsychopharmacology. 34, 1552-1566.

Vilner B. J., de Costa B. R. and Bowen W. D. (1995a) Cytotoxic effects of sigma ligands: sigma

receptor-mediated alterations in cellular morphology and viability. J. Neurosci. 15, 117-134.

Vilner B. J., John C. S. and Bowen W. D. (1995b) Sigma-1 and sigma-2 receptors are expressed

in a wide variety of human and rodent tumor cell lines. Cancer Res. 55, 408-413.

Whittemore E. R., Ilyin V. I. and Woodward R. M. (1997) Antagonism of N-methyl-D-aspartate

receptors by sigma site ligands: potency, subtype-selectivity and mechanisms of inhibition.

J. Pharmacol. Exp. Ther. 282, 326-338.

WHO (2016) Dementia fact sheet. 2016.

Wu Z. and Bowen W. D. (2008) Role of sigma-1 receptor C-terminal segment in inositol 1,4,5-

trisphosphate receptor activation: constitutive enhancement of calcium signaling in MCF-7

tumor cells. J. Biol. Chem. 283, 28198-28215.

133

Zeng C., Rothfuss J., Zhang J., Chu W., Vangveravong S., Tu Z., Pan F., Chang K. C., Hotchkiss

R. and Mach R. H. (2012) Sigma-2 ligands induce tumour cell death by multiple signalling

pathways. Br. J. Cancer. 106, 693-701.

Zeng C., Rothfuss J. M., Zhang J., Vangveravong S., Chu W., Li S., Tu Z., Xu J. and Mach R. H.

(2014) Functional assays to define agonists and antagonists of the sigma-2 receptor. Anal.

Biochem. 448, 68-74.

Zeng C., Vangveravong S., McDunn J. E., Hawkins W. G. and Mach R. H. (2013) Sigma-2

receptor ligand as a novel method for delivering a SMAC mimetic drug for treating ovarian

cancer. Br. J. Cancer. 109, 2368-2377.

Zheng P. (2009) Neuroactive steroid regulation of neurotransmitter release in the CNS: action,

mechanism and possible significance. Prog. Neurobiol. 89, 134-152.

Zou L. B., Yamada K., Sasa M., Nakata Y. and Nabeshima T. (2000) Effects of sigma(1)

receptor agonist SA4503 and neuroactive steroids on performance in a radial arm maze task

in rats. Neuropharmacology. 39, 1617-1627.

134

CHAPTER 4

MOLECULAR MECHANISMS OF SEROTONERGIC ACTION OF THE HIV-1 ANTIRETROVIRAL EFAVIRENZ

Dhwanil A. Dalwadia , Seongcheol Kima , Shahnawaz M. Amdania,c, Zhenglan Chena, Ren-Qi Huanga,b , John A. Schetza,b,1

aDepartment of Pharmacology & Neuroscience, Graduate School of Biomedical Sciences,

University of North Texas Health Science Center, 3500 Camp Bowie Blvd., Fort Worth, Texas,

76107

bInstitute for Healthy Aging, Center for Neuroscience Discovery,

University of North Texas Health Science Center, 3500 Camp Bowie Blvd., Fort Worth, Texas,

76107 cCurrent address: Lincoln Medical and Mental Health Center, 234 E. 149th St. Bronx, NY 10451

1To whom correspondence should be addressed. E-mail: [email protected]

1Abstract:

Efavirenz is highly effective at suppressing HIV-1, and the WHO guidelines list it as a component of the first-line antiretroviral (ARV) therapies for treatment-naïve patients. Though the pharmacological basis is unclear, efavirenz is commonly associated with a risk for

1 This project is taken in part from a dissertation submitted to the University of North Texas Health Science Center Graduate School of Biomedical Sciences in partial fulfilment of the requirements for the degree Doctor of Philosophy. This work has been published in the journal Pharmacological Research Dalwadi DA, Kim S, Shahnawas AM, Chen Z, Huang R, Schetz JA (2016) Molecular mechanisms of serotonergic action of the HIV-1 antiretroviral efavirenz. Pharmacological Research 110:10-24.

135 neuropsychiatric adverse events (NPAEs) when taken at the prescribed dose. In many patients these NPAEs appear to subside after several weeks of treatment, though long-term studies show that in some patients the NPAEs persist. In a recent study focusing on the abuse potential of efavirenz, its receptor psychopharmacology was reported to include interactions with a number of established molecular targets for known drugs of abuse, and it displayed a prevailing behavioral profile in rodents resembling an LSD-like activity. In this report, we discovered interactions with additional serotonergic targets that may be associated with efavirenz-induced

NPAEs. The most robust interactions were with 5-HT3A and 5-HT6 receptors, with more modest interactions noted for the 5-HT2B receptor and monoamine oxidase A. From a molecular mechanistic perspective, efavirenz acts as a 5-HT6 receptor inverse agonist of Gs-signaling, 5-

HT2A and 5-HT2C antagonist of Gq-signaling, and a blocker of the 5-HT3A receptor currents.

Efavirenz also completely or partially blocks agonist stimulation of the M1 and M3 muscarinic receptors, respectively. Schild analysis suggests that efavirenz competes for the same site on the

5-HT2A receptor as two known hallucinogenic partial agonists (±)-DOI and LSD. Prolonged exposure to efavirenz reduces 5-HT2A receptor density and responsiveness to 5-HT. Other ARVs such as zidovudine, nevirapine and emtricitabine did not share the same complex pharmacological profile as efavirenz, though some of them weakly interact with the 5-HT6 receptor or modestly block GABAA currents.

Keywords:

Efavirenz

5-HT2A

LSD

DOI

136

5-HT2C

5-HT6

Abbreviations:

NPAEs: Neuropsychiatric adverse events

ARV: Antiretroviral

HAART: Highly active antiretroviral therapy

ZDV: Zidovudine

FTC: Emtricitabine

NVP: Nevirapine

Chemical compounds:

Serotonin; 5-Hydroxytryptamine (PubChem CID: 5202)

Efavirenz (PubChem CID: 64139)

DOI (PubChem CID: 170617)

LSD Tartrate (PubChem CID: 159828)

Nevirapine (PubChem CID: 4463)

Emtricitabine (PubChem CID: 60877)

Zidovudine (PubChem CID: 35370)

TCB-2 (PubChem CID: 71433791)

α-Methyl-5HT (PubChem CID: 2107)

Atropine (PubChem CID: 174174)

Scopolamine (PubChem CID: 3000322)

137

1. Introduction:

Efavirenz [(4S)-6-chloro-4-(2-cyclopropylethynyl)-4-(trifluoromethyl)-2,4-dihydro-1H-

3,1-benzoxazin-2-one] is a potent non-nucleoside reverse transcriptase inhibitor (NNRTI) and one of the preferred components of highly active antiretroviral therapy (HAART) (WHO 2013;

Young et al. 1995). Though very effective at suppressing replication of the virus that causes

AIDS, a standard dose of efavirenz is known to carry a significant risk for CNS-mediated

NPAEs (Cohen et al. 2014; Gutierrez et al. 2005; Mills et al. 2013; Munoz-Moreno et al. 2009;

Shubber et al. 2013; Sustiva 1998). However, little is known from a mechanistic perspective as to why these NPAEs occur and which CNS off-targets might be involved. The findings of a previous report (Gatch et al. 2013) led us to examine the molecular mechanisms of efavirenz across a broader range of serotonergic targets.

Though efavirenz is considered to have good CNS penetration (Tozzi et al. 2009) consistent with its physicochemical properties, only low levels are detected in cerebral spinal fluid (CSF) with a mean concentration across studies of approximately 44 nM that corresponds to CSF levels 0.52% of plasma concentrations (Best et al. 2011; Tashima et al. 1999; Yilmaz et al. 2012). Efavirenz has a very high tendency to bind proteins (e.g, approximately 99.8% is bound to plasma proteins) (Avery et al. 2013; Tanaka et al. 2008), and healthy CSF contains very little protein compared to that in blood plasma (< 1%). Since the concentration of free efavirenz in the aqueous fraction is expected to be low and it binds to protein-rich brain tissue

(Dirson et al. 2006), one might expect that 44 nM in CSF, if it accounted for approximately 0.2%

(100%-99.8%) of the efavirenz dose, would correspond to an estimated 22 µM (44 nM ÷ 0.002) concentration of efavirenz in brain tissue. A study in rats suggests that efavirenz readily accumulates in the brain to levels that exceed 4.6 times the plasma levels within 1 hr of an i.p.

138 dose of 15 mg/kg (Dirson et al. 2006). Physiological effects have been reported in rodent studies at efavirenz doses in the same range: 10-30 mg/kg depending upon the behavioral measure examined (Gatch et al. 2013; Romao et al. 2011). High plasma levels of efavirenz (1 – 4 µg/mL) are needed to keep the HIV-1 virus suppressed to clinically meaningful levels (Marzolini et al.

2001), but efavirenz also has a narrow therapeutic window (Gutierrez et al. 2005): plasma levels less than 1 µg/mL result in virologic failure and while those greater than 2.74 µg/mL result in

NPAE (Gutierrez et al. 2005; Marzolini et al. 2001). Assuming a similar level of brain accumulation occurs in humans as in rats, then efavirenz plasma levels >2.74 µg/mL would correspond to brain concentrations >40 µM (2.74x10-3 g/L • 1/315.7 g/mole • 4.6). Based upon these types of assumptions and calculations, it would appear that efavirenz can rapidly accumulate in the brain to concentrations in the range of tens of micromolars. Rapid accumulation of relatively high concentrations of efavirenz in the brain and evidence of CNS behavioral effects in animals and humans suggest that this reverse transcriptase inhibitor has

CNS off-targets.

It has been almost two decades since efavirenz was approved by the FDA and to date, there is only one study that has sought to investigate the receptor neuropharmacology underpinning efavirenz’s CNS targets and this was in the context of its reported recreational use

(Gatch et al. 2013). In that study, a rationalized mechanistic approach was utilized in that receptor targets known to interact with drugs of abuse were selected to narrow the receptor profiling effort leading to a number of CNS receptors being identified as possible targets for efavirenz. At the receptor level, efavirenz was shown to interact with the 5-HT2A, 5-HT2C and

GABAA receptors, as well as DAT, SERT, and VMAT2 transporters (Gatch et al. 2013).

Mechanistically, efavirenz potentiated GABA-mediated chloride currents at the GABAA

139 receptor, and functioned as a DAT and SERT blocker (Gatch et al. 2013). However, in vivo, efavirenz failed to strongly substitute in tests of discrimination for drugs known to have these same functional properties on DAT, SERT, VMAT2 transporters, or the GABAA receptor (Gatch et al. 2013). In contrast, when efavirenz was used as the discriminative stimulus, LSD partially substituted for efavirenz. Moreover, in rats trained to discriminate LSD from saline, efavirenz partially substituted for LSD and this substitution was blocked by pre-treatment with the 5-HT2A receptor selective antagonist MDL100,907 (Gatch et al. 2013). In a rodent head-twitch assay, efavirenz produced head-twitch responses in wild type but not 5-HT2A-KO mice; though, the response was far weaker than for LSD and initiated much more rapidly (Gatch et al. 2013).

There is a strong positive correlation between compounds that are hallucinogens in humans and those that induce a head-twitch response in rodents (Colpaert et al. 1985; CORNE et al. 1963;

Singleton and Marsden 1981). Efavirenz also dose-dependently depressed novel open field locomotor activity, similar to LSD in the same strain of mice (Gatch et al. 2013). For these reasons, it was concluded that efavirenz’s predominate behavioral profile in rodents is most consistent with an LSD-like effect mediated by the 5-HT2A receptor (Gatch et al. 2013).

In this study we report for the first time that efavirenz interacts with the 5-HT3, 5-HT6 receptors and the enzyme MAO-A, and elucidate its molecular mechanisms of action at these targets as well as at the 5-HT2A and 5-HT2C receptors. Within a similar concentration range, efavirenz also blocks agonist responses at the M1 and M3 muscarinic receptors. Schild shift analysis suggests efavirenz acts at the same binding site on the 5-HT2A receptor as the partial agonists LSD and (±)-DOI, but it is unique in that it does not activate the Gq-signaling pathway.

Chronic exposure of the 5-HT2A receptor to efavirenz reduces receptor density and responsiveness to 5-HT. We also demonstrate here that efavirenz’s mechanism of action across a

140 range of CNS targets is distinct from other prominent HIV-1 medications like emtricitabine

(FTC), nevirapine (NVP) and zidovudine (ZDV).

2. Materials and Methods:

2.1. Chemicals:

Radiochemicals were from Perkin Elmer (Saint Louis, MO): 4-(2''-Methoxy)-phenyl-1-

[2''-(N-2-pyridinyl)-p-fluorobenzamido]ethyl-piperazine ([3H]MPPF, NET-1109, 80 Ci/mmol);

[3H]methylspiperone ([3H]MSP, NET-856, 84 Ci/mmol); [3H]Lysergic acid diethylamide

([3H]LSD, NET638, 70 Ci/mmol); [3H]Mesulergine (NET1148, 80 Ci/mmol); [3H]BRL-43694

(granisetron) (NET1030, 65 Ci/mmol); [3H]GR113808 (NET1152, 85 Ci/mmol). Efavirenz,

[(4S)-6-chloro-4-(2-cyclopropylethynyl)-4-(trifluoromethyl)-2,4-dihydro-1H-3,1-benzoxazin-2- one], and the other antiretroviral drugs (emtricitabine (FTC), zidovudine (ZDV) and nevirapine

(NVP)) were purchased from Sequoia Research Products Limited (Pangbourne, UK).

MDL100,907 was synthesized and kindly provided by Dr. Kenner C. Rice (NIDA/NIAAA).

Unless otherwise noted, all other drugs and reagents were purchased from Tocris Biosciences

(via R&D Systems, Inc. in Minneapolis, MN) or Sigma-Aldrich (St. Louis, MO). All ARV drugs were solubilized in DMSO at concentrations ranging from 10–100 mM and diluted at least

1:1000 v/v in the final assay solution.

2.2. Profiling Serotonin Receptors by Radioligand Binding

The interaction of efavirenz (10 µM) with serotonin receptors was measured by its ability to displace specifically bound radioligands from the metabotropic 5-HT1A, 5-HT2A, 5-HT2B, 5-

HT2C, 5-HT4, 5-HT5A, 5-HT6 and 5-HT7 receptors, and an ionotropic 5-HT3A receptor using the conditions outlined in Table 1. With the exception of the 5-HT4 receptor, which was sourced from Duncan Hartley guinea pig striatal tissue, all other serotonin receptors were cloned

141

receptors heterologously expressed in mammalian cell lines lacking the receptor subtypes of

interest. Radioligand, purified membranes expressing individual serotonin receptors and a fixed

concentration of 10 µM efavirenz in a total volume of 1 mL binding buffer were allowed to

equilibrate. Receptors were then isolated by rapid filtration through glass fiber filters pretreated

for 10 min with 0.5% polyethyleneimine (Sigma-Aldrich) and three rapid washes with 3 mL of

ice-cold wash buffer (50 mM Tris, pH 7.4 at 0-2°C). GF/C filters were used for membranes

derived from cell cultures and GF/B filters for membranes derived from brain tissue (Brandel,

Gaithersburg, MD). Dried filters were cut into individual scintillation vials, filled with 3.5 ml of

scintillation fluid, mixed, and the radioactivity bound to the filters was quantified via scintillation

spectroscopy. Membrane protein concentrations varied from 0.02-0.06 mg/mL. All data for each

experiment was measured in triplicate. Each experiment was then repeated two or three times

and the mean values of these experiments were reported with their associated SEM. A one-way

ANOVA followed by a Bonferroni post-hoc analysis set to high stringency (P < 0.001) was used

to determine significant displacement of the radioligand.

Table 1. Conditions for measuring efavirenz’s ability to displace radioligands specifically bound to the serotonin receptor subtypes. Cloned receptors expressed in cell lines served as the source of all subtypes, except for the 5-HT4 receptor. The letter(s) in front of the receptor subtype designate the species: h = human, m = mouse, r = rat, gp = guinea pig.

Drug for defining Receptor Cell type for non-specific Binding buffer and Subtype expression Radioligand binding conditions Buffer A; 90 minutes at h5-HT HEK293 [3H]MPPF 5 μM NAN-190 1A 25°C Buffer A; 90 minutes at h5-HT HEK293 [3H]MSP 5 μM mianserin 2A 25°C Buffer C; 90 minutes at h5-HT HEK293 [3H]LSD 30 μM mianserin 2B 25°C 3 h5-HT2C HEK293 [ H]mesulergine 5 μM mianserin Buffer A; 90 minutes at

142

25°C Buffer A; 90 minutes at m5-HT HEK293 [3H]BRL-43694 10 μM mianserin 3 25°C Buffer A; 30 minutes at gp5-HT Striatum [3H]GR113808 30 μM serotonin 4 25°C Buffer B; 60 minutes at h5-HT CHO-K1 [3H]LSD 100 μM serotonin 5A 37°C 30 μM Buffer C; 60 minutes at r5-HT HeLa [3H]LSD 6 chlorpromazine 37°C Buffer B; 120 minutes at h5-HT CHO [3H]LSD 10 μM serotonin 7 25°C Binding Buffer A = 50 mM Tris, pH = 7.4 Binding Buffer B = 10 mM MgSO4, 0.5 mM EDTA, 50 mM Tris, pH 7.4 Binding Buffer C = 5 mM MgCl2, 250 μM sodium ascorbate, 50 mM Tris, pH = 7.4 Approximate concentrations of the radioligands used in the displacement assays were 0.5 nM in all cases, except for [3H]GR113808 which was tested at 0.7 nM, [3H]GR113808 which was tested at 3 1.7 nM, and [ H]LSD which in the case of 5HT2B was tested at 0.25 nM and in the case of 5HT7 at 4 nM.

2.3. Measuring the affinity of efavirenz for selected serotonin receptor subtypes using

competition radioligand binding.

Selected receptors were examined further in competition-type concentration-response

curves to measure their affinity for efavirenz. The binding conditions and procedure were the

same as those described above and in Table 1 except that increasing concentrations, instead of a

fixed concentration, of efavirenz was used in order to generate concentration-response curves

and extract IC50 values. Equilibrium inhibition constants (Ki), representing binding affinities,

were calculated from the IC50 values using the Cheng-Prusoff equation: Ki = IC50/(1 +

[ligand]/KD). Concentration-response curves were fitted with a four parameter logistic equation

that included a variable slope using a 95% confidence interval for all curve-fits using Graphpad

Prism version 4.0.

143

2.4. Intracellular Calcium and IP-One Assay: measurement of Gq-coupled response

Individual cloned human 5-HT2A, 5-HT2B or 5-HT2c receptors were stably expressed in

HEK293 cells. Since an endogenous human M3 muscarinic receptor (Ancellin et al. 1999;

Atwood et al. 2011; Luo et al. 2008) is expressed in HEK293 cells, untransfected HEK293 cells served as the source of the M3 receptor. The cloned human M1 muscarinic receptor was stably expressed in CHO cells. HEK293 cells were grown in Dulbecco’s modified eagles medium

(DMEM) complete which is DMEM supplemented with 50 units/mL streptomycin, 50 µg/mL penicillin, 1 mM sodium pyruvate, and 10% fetal bovine serum. The media of transfected cells additionally contained 0.1 mg/mL G418 to maintain selection. In the case of the 5-HT2c receptor, dialyzed serum (Innovative Research Inc., Novi, MI) was employed as no signal could be obtained when using undialyzed FBS (ThermoFisher Scientific, Waltham, MA); however, undialyzed FBS was used in the case of the 5-HT2A receptor since virtually identical responses were observed regardless of the serum type. CHO cells stably expressing the cloned human M1 muscarinic receptor were grown in Ham’s F12 supplemented with 50 units/mL streptomycin, 50

µg/mL penicillin, 1 mM sodium pyruvate, and 10% fetal bovine serum, and 0.1 mg/mL G418 to maintain selection.

Cells expressing the receptor of interest were seeded into Poly-L-Lysine coated, black- walled 96 well plates at a density of 120,000 cells per well. The following day, media was removed and cells were loaded for three hours in the dark with the FLIPR Calcium-6 QF dye

144 calcium were detected as rapid changes in fluorescent signal measured at 485 nm excitation, 525 nm emission with a cutoff filter set at 515 nm using a Flexstation 3 (Molecular Devices,

Sunnyvale, CA). The baseline calcium signal was measured for 120 seconds followed by injection of the test drug and reading of any change in the calcium signal for an additional 780 seconds. At the termination of the experiment all signals were baseline subtracted and the change in intracellular calcium was quantified as the area under the curve, measured by integration

(Prism version 4.0). All data were normalized to maximal functional response defined as the asymptote of the sigmoidal curve.

IP-One levels were measured in HEK293 cells stably expressing the 5-HT2A receptor.

Cells were seeded into Poly-L-Lysine coated, black walled 96 well plates at a density of 20,000 cells per well. IP-One levels were measured using the IP-One HTRF® assay kit (cisbio Assays,

Bedford, MA) per manufacturer’s protocol.

2.5. Effect of prolonged exposure to efavirenz on receptor density and Gq-coupled

activation of the 5-HT2A receptor

HEK293 cells stably expressing the 5-HT2A receptor were split into four T-25 flasks to

20% confluency and were allowed 24 hrs to attach. Media was then removed by aspiration and replaced with either vehicle (1:1000 v/v DMSO), 30 µM 5-HT, or 30 µM efavirenz. Two flasks were designated for the vehicle treatment as one of them would be used later for a very short (15 min) exposure to efavirenz. For three days, the media was replaced with fresh media containing the same treatments every 12 hrs. At the end of the third day, one of the vehicle treated flasks was also treated with 30 µM efavirenz for only 15 min. This group served as a control to ensure all efavirenz could be removed with our washing procedure prior to measuring the change in

2+ intracellular Ca or receptor density (Bmax). For the washing procedure, media was removed

145 from all four flasks and cells were washed twice by incubating for 30 min in 10 mL DMEM complete. Following the second wash, cells were harvested for either the calcium assay as described above with no more than 8 hrs total between replating cells and performing the assay, or for radioligand binding to estimate receptor density (Bmax). For the radioligand binding assay, membranes were prepared as described above for each of the four treated groups. The 5-HT2A

3 receptor density was then estimated using 0.5 nM [ H]MSP followed by the calculation of Bmax at saturation using a rectangular hyperbola model Bmax = (Y •(KD+X))/X, where Y is [specific

3 binding] and X = [radioligand concentration]. The KD for [ H]MSP at the cloned human 5-HT2A receptor had been previously determined to be 246 pM (Cummings et al. 2010). The protein concentration for each sample was determined using the BCA assays as described in section 2.8.

2.6. Cyclic adenosine monophosphate (cAMP) Assay: measurement of Gs-coupled

responses

Changes in intracellular cAMP levels were measured in HeLa cells stably expressing the rat 5-HT6 receptor using a luminescent cAMP-Glo kit (Promega, Madison, WI) terminal assay as per the manufacturer’s protocol with the following modifications. A total of 10,000 cells were seeded per well in a 96 well plate and grown in DMEM complete (see above). The following day, media was removed and cells were pretreated for 30 min in DMEM supplemented with 500

µM of the phosphodiesterase inhibitors 3-Isobutyl-1-methylxanthine (IBMX) (Sigma-Aldrich,

St. Louis, MO) and 100 µM Ro 20-1724 (Santa Cruz Biotechnology, Inc., Dallas, Texas). Next, cells were treated either alone or with a combination of vehicle, efavirenz or 5-HT for 30 min and then lysed in 20 µL lysis buffer for an additional 30 min with vigorous shaking (600 rpm on an IKA MTS 2/4 digital microplate shaker). A total of 10 µL of this lysate was transferred to a half-area plate containing 20 µL cAMP-GloTM detection solution and the plate was mixed for 1

146 minute at 600 rpm and incubated at room temperature for 20 min. Finally, 40 µL of kinase glo® reagent was added to each well, mixed for 1 minute at 600 rpm and incubated at room temperature for 10 min. Luminescence measurements were performed using the Flexstation 3 set to luminescence mode with 100 millisecond integration time. All data were normalized to maximal stimulatory concentration of forskolin (10 µM).

2.7. Inhibition of monoamine oxidase A (MAO-A) activity

Recombinant human MAO-A was overexpressed in insect Hi5 cells and lysates of these cells containing approximately 4 µg/mL of enzyme were washed in 100 mM potassium phosphate, pH = 7.4, then pretreated for 15 min at 37°C with vehicle or 10 µM efavirenz, followed by treatment with 50 µM of the MAO-A substrate kynuramine for 60 min at 37°C. The reaction was terminated at by the addition of 6 N NaOH and the fluorescent 4-hydroxyquinoline cleavage product (Ex325/Em465) served as a measure of MAO-A activity.

2.8. Protein Assay

2.9. Measurement of 5-HT3A and GABAA receptor currents using electrophysiology

Currents were obtained in the whole cell configuration from HEK293 cells stably expressing the rat or human GABAA receptors (α1β2γ2 (short isoform of the γ2 subunit)

(Hawkinson et al. 1996), or the mouse 5HT3A receptors transiently transfected into HEK293 cells using a modified polyethylenimine method described previously (Hsu and Uludag 2012).

Cell were voltage-clamped at -60 mV and the recordings were made at room temperature in the whole cell configuration using borosilicate pipettes with a tip resistance of 1-2.5 M. The

147 pipette solution contained 140 mM CsCl, 10 mM EGTA, 10 mM HEPES, 4 mM Mg-ATP, pH =

7.2. Coverslips containing cultured cells were placed in a 1.5 mL chamber on the stage of an inverted light microscope and superfused continuously (5-8 mL/min) with an extracellular solution containing 125 mM NaCl, 5.5 mM KCl, 0.8 mM MgCl2, CaCl2 (1.2 mM in 5-HT3A studies and 3 mM in GABAA studies), 10 mM HEPES, 10 mM D-glucose pH = 7.3. GABA or 5-

HT-induced currents were low-pass filtered at 5 kHz, monitored on an oscilloscope and a chart recorder, and stored for subsequent analysis. A 60-80% series resistance compensation was applied at the amplifier. Any change in access resistance observed during the recording period resulted in the patch being aborted and that data was excluded from the analysis.

5-HT or GABA, with or without antiretroviral drugs, was dissolved in the extracellular solution and then applied to cells for 10-20 sec from independent reservoirs by gravity flow using a Y-shaped tube positioned within 100 µm of the cells. Using this system, the 10-90% rise time of the junction potential at the open tip is 12-51 milliseconds (Huang and Dillon 1999).

Receptors were activated with approximately an EC30 concentration of the appropriate n -HT3A receptors, 5 µM for rat α1β22 and 10 µM for human α1β22

GABAA receptors), because this level of occupancy elicits minimal desensitization. Once a positive (baseline) control GABA or 5-HT response was established, the effect of ARV drug was examined by co-application with GABA or 5-HT. 5-HT or GABA responses were again monitored after each ARV drug application to ensure adequate washout and to ensure no loss of signal and integrity of the patch. Using this strategy, multiple concentrations of ARV drug could generally be tested for each cell studied. ARV drug stocks were dissolved in DMSO and diluted in saline so that the final DMSO concentration (v/v) was ≤ 0.2%. The concentration-response curve for efavirenz’s modulatory effect on 5-HT3A receptors was fitted to the following equation:

148

n I/Imax= 1/(1+(IC50/[efavirenz]) ), where I is the peak 5-HT current normalized to control at a given efavirenz concentration and Imax represents the normalized 5-HT-induced current, IC50 is the half-maximal blocking concentration, and n is the Hill coefficient. All data are presented as mean values ± SEM of a minimum of six individual experiments. A student’s paired t-test was used to determine statistical significance (p< 0.05) for experiments comparing responses in the presence or absence of a single concentration ARV drug.

3. Results:

In a previous report (Gatch et al. 2013), CNS targets for efavirenz were identified that may contribute to its attractiveness as a drug of abuse. In this report, the molecular mechanisms of action for efavirenz were examined across a broader range of serotonergic targets in an effort to enhance our understanding of some of the observed NPAEs in patients taking it as prescribed.

This included a comparison of efavirenz’s mode of action with those of the other HIV-1 medications FTC, NVP and ZDV.

3.1. Efavirenz interacts with a number of receptors in the 5-HT family

Potential interactions of efavirenz with serotonin receptor subtypes from each of the seven subfamilies, 5-HT1 through 5-HT7, were probed initially by measuring the ability of a fixed concentration of efavirenz (10 µM) to displace radioligands specifically bound to individual receptor subtypes (Figure 1). The concentration selected was based upon our investigation of numerous receptor systems, including the 5-HT2A and 5HT2C receptors, and the estimated levels of efavirenz’s brain exposure (> 40 µM) in those at risk for experiencing NPAEs following a standard oral dose (see Introduction). Using this assay format, we showed in our previous report that efavirenz interacted with the 5-HT2A and 5-HT2C receptors, and in this report we show that efavirenz additionally had interactions with the 5-HT2B, and 5-HT6 receptors, which reached

149 statistical significance with a stringent cutoff (P < 0.001, ANOVA followed by a Bonferroni post-hoc) (Figure 1). Of these interactions, the most robust were with the 5-HT2A, 5-HT2C, and 5-

HT6 receptors as efavirenz displaced significantly more radioligand from these three 5-HT receptor subtypes than from the 5-HT2B receptor (P < 0.05, ANOVA followed by a Bonferroni post-hoc).

Those receptors experiencing ≥ 75% displacement of radioligand in the presence of 10

µM efavirenz (i.e., 5-HT2A, 5-HT2C and 5-HT6 receptors) were examined further in competition- type concentration-response curves to measure their affinity for efavirenz (Figure 2). The calculated affinities (Ki) for efavirenz were similar in magnitude ranging from 1.7 to 4.4 µM.

The highest affinity was for the 5-HT2C receptor and the lowest for the 5-HT2A, which had significantly lower affinity than for the other two receptors (P < 0.05 ANOVA with a Bonferroni post-hoc). While a variable slope model was not the statistically preferred model over a model with the pseudo Hill slope fixed to unity (e.g. -1.0), in order to evaluate the pseudo Hill slope parameter all the slopes were treated as a variable (instead of a fixed value) for the purpose of curve fitting. Using this approach, in all cases the pseudo Hill slopes were moderately greater than unity (-1.4 to -1.2) but the variance was such that none of them deviated significantly from unity when strict criteria were applied (P < 0.001, t-test), and further none of them differed significantly from one another (P < 0.05 ANOVA with a Bonferroni post-hoc). Slopes not different from unity are consistent with efavirenz binding to a single site on each of the 5-HT2A,

5-HT2C, and 5-HT6 receptors. As a reference, the affinities for serotonin at the 5-HT2A, 5-HT2C and 5-HT6 receptors were measured as well, yielding Ki ± SEM values equal to 360 ± 29 nM, 16

± 3.5 nM and 105 ± 8 nM, respectively, which are consistent with previous reports (Cussac et al.

2002; Hirst et al. 2003; Johnson et al. 1994).

150

3.2. Efavirenz functions as a 5-HT2A and 5-HT2C receptor antagonist, a 5-HT6 receptor inverse agonist, a 5-HT3 receptor pore blocker, and an MAO-A inhibitor

The individual functional properties of the stably expressed cloned Gq-coupled 5-HT2A and 5-HT2C, and the Gs-coupled 5-HT6 receptors were assessed starting with their ability to be stimulated by the endogenous full agonist serotonin (Figure 3). Serotonin activated the 5-HT2A receptor with a potency that was unaffected by the serum condition in the culture media: 13.3 ±

2.5 nM in absence of serum, 13.9 ± 4.1 nM in presence of dialyzed FBS, and 17.2 ± 7.5 nM in presence of FBS. In stark contrast, the potency of 5-HT at the 5-HT2C receptor spanned from the sub-nanomolar to the low nanomolar range: 0.039 ± 0.0039 nM in absence of serum, 1.5 ± 0.18 nM in presence of dialyzed FBS, and undetectable in the presence of FBS. Most notable was that no 5-HT response above baseline was observed when the 5-HT2C receptor was maintained in

FBS prior to measuring the Ca2+ response. Serotonin had potency in the lower nanomolar range,

12.7 nM, for the 5-HT6 receptor. The 5-HT2C receptor had a significantly higher potency for serotonin under serum free conditions compared with the 5-HT2A receptor (P < 0.05 Student’s t- test) or the 5-HT6 receptor (P < 0.05 Student’s t-test). Large differences in pseudo Hill slope values were measured as well ranging from 0.55 to 1.5, with the 5HT2A receptor having the shallowest slope and the 5HT2C receptor having the steepest slope. However, due to the variance in the data, these pseudo Hill slopes were not significantly different between receptors (P > 0.05

ANOVA with a Bonferroni post-hoc) and did not differ significantly from unity (P > 0.05,

Student’s t-test). More importantly, these values set a baseline for a full agonist response under our assays conditions and were useful for comparing and contrasting the ARV drug responses

151 mediated by these receptors. Also, the mean EC50 values reported here for the 5-HT2A, 5-HT2C

(for each of the serum conditions tested) and the 5-HT6 receptors are in agreement with previously published values (Chen et al. 2011; Porter et al. 1999; Purohit et al. 2003).

In a previous collaborative report, efavirenz was determined to have LSD-like properties mediated by the 5-HT2A receptor based primarily upon its behavioral properties in rodents (Gatch et al. 2013). In this report, we employed detection of intracellular calcium or IP1 accumulation as

2+ robust measures of 5-HT2A–mediated activation of the Gq-PLC-IP3-[Ca ]i signaling pathway.

2+ When a change in intracellular Ca was employed as the measure of Gq-signaling, efavirenz did not stimulate the 5-HT2A receptor at the maximum concentration (10 µM efavirenz) that could be tested as an agonist under the assay conditions (Figure 4A). To completely rule out the possibility that efavirenz might be a weak partial agonist, a higher concentration needed to be tested; due to efavirenz’s solubility restriction in the Ca2+ assay, the IP-One assay was employed.

Using the maximum concentration that could be tested, 100 µM efavirenz did not activate the

Gq-pathway when IP-One was used as the readout of Gq activity (Figure 4B). This prompted us to test whether efavirenz might have other actions given that the radioligand binding data indicated that it interacts with the 5-HT2A receptor. Accordingly, efavirenz was tested for its ability to modify functional responses of different classes of the 5-HT2A receptor agonists: 5-HT,

TCB-2 and α-methyl-5-hydroxytryptamine (α-Me-5HT). Like the selective 5-HT2A receptor antagonist MDL100,907, efavirenz was able to antagonize the effects of all three agonists

(Figure 4A). Given these results, efavirenz was also tested for its ability to modify the in vitro functional response of the well-known hallucinogens (±)-DOI and LSD by measuring changes in intracellular calcium. Both (±)-DOI and LSD behaved as partial agonists (Figure 5B & 5E), which is consistent with previously published reports (Berg et al. 1998; Egan et al. 1998; Marek

152 and Aghajanian 1996). In the presence of increasing concentrations of efavirenz, both (±)-DOI and LSD functional responses were progressively shifted to the right without a significant decrease in the maximal response (Figure 5A, P > 0.05 ANOVA followed by Bonferroni’s post- hoc for maximal responses) suggestive of either a negative heterotropic allosteric modulation with high cooperativity or a purely competitive interaction at a single site. Schild plot transformation of the dose-ratio shift data yielded slopes equal to 1.06 ± 0.054 and 1.20 ± 0.16 for (±)-DOI and LSD, respectively, that were not significantly different from unity (P > 0.05 t- test) suggesting that efavirenz competes with (±)-DOI and LSD for binding to a single site on the

5-HT2A receptor (Figure 5C, 5E). From the Schild plots, the pA2 value (also known as the antagonist potency (KB)) for efavirenz was calculated to be 1.3 ± 0.2 µM and 0.31 ± 0.18 µM using either (±)-DOI or LSD, respectively, as the agonist in competition with efavirenz.

However, these efavirenz antagonist potency values were not statistically different from one another or from the measured affinity (Ki = 4.37 ± 0.43 µM, Figure 2A) of efavirenz for the 5-

HT2A receptor (P > 0.01 ANOVA with a Bonferroni post-hoc). Collectively, the results suggest that efavirenz competitively interacts with a single site on the 5-HT2A receptor that is indistinguishable from the site where LSD and (±)-DOI bind.

While efavirenz produces NPAEs in the majority of patients, the initial clinical finding was that in many of those affected, the NPAEs appear to lessen or disappear after several weeks

(Sustiva 1998). Thus, we attempted to model a chronic exposure scenario in vitro and evaluate the possibility that efavirenz might induce a pharmacodynamic tachyphylaxis type response, meaning that its ability to antagonize a 5-HT2A receptor response would diminish upon repeated exposure. HEK293 cells stably expressing the cloned human 5-HT2A receptor were exposed to vehicle, 30 M 5-HT, or 30 µM efavirenz for 3 days. At the end of the 3 day treatment, cells

153 were thoroughly washed and suppression of 5-HT-induced Ca2+ response by efavirenz was assessed. In all the treatment groups, 10 µM efavirenz treatment resulted in a 48-49% suppression of the calcium response induced by a submaximal concentration of 5-HT (100 nM).

This suppression was statistically significant in the two vehicle and the serotonin pre-treatment groups (Figure 6A, P < 0.05 ANOVA followed by Bonferroni post-hoc), but was not significant in the 3 day 30 µM efavirenz pre-treatment group (Figure 6A, P > 0.05 ANOVA followed by

Bonferroni post-hoc) even though a 49% suppression is evident. Unexpectedly, the overwhelming effect of chronic treatment with efavirenz was a large (81%) and significant reduction in the 5-HT2A receptor’s responsiveness to 5-HT (Figure 6A, P < 0.05 ANOVA followed by Bonferroni post-hoc). This drastic reduction was not caused by the inability of efavirenz to dissociate from the receptor, because the responses in cells treated with 30 µM efavirenz for 15 min after 3 day treatment with DMSO were not significantly different from the responses in cells that underwent the 3 day treatment with DMSO only. To assess if the reduction in the 5-HT response in the 3 day efavirenz treatment group was due to the reduction in receptor density, Bmax was estimated following the same series of 3 day treatments. Three day exposure to efavirenz resulted in a statistically significant 52% reduction in Bmax (Figure 6B, P <

0.05 ANOVA followed by Bonferroni post-hoc), which is consistent with the observed reduction in the functional response to agonist following 3 days chronic treatment with efavirenz.

Like at the 5-HT2A receptor, efavirenz functioned as an antagonist at the 5-HT2C receptor.

Given that serum had a drastic effect on 5-HT’s potency at the 5-HT2C receptor (Figure 3B), we tested efavirenz’s ability to modify the in vitro functional response to 5-HT in the absence of serum. Using an EC50 concentration of 5-HT (0.03 nM) to stimulate the 5-HT2C receptor, up to 3

µM efavirenz had no significant effect on 5-HT responses; however, at concentrations ≥ 10 µM

154 efavirenz inhibited the 5-HT agonist effect, and at a concentration ≥ 60 µM efavirenz completely suppressed it (Figure 7A, P < 0.001 ANOVA followed by Dunnett’s multiple comparison). To verify the specificity of the results, we tested efavirenz’s ability to modulate the activity of an endogenous Gq-coupled M3 muscarinic receptors known to be expressed in HEK293 cells

(Ancellin et al. 1999; Atwood et al. 2011; Luo et al. 2008). While efavirenz alone had no effect on intracellular Ca2+ levels in untransfected HEK293 cells (Figure 8A), at concentrations ≥ 10

µM it was able to partially suppress the Gq response mediated by an EC50 concentration of acetylcholine (ACh) (Figure 8B, P < 0.05 ANOVA followed by Dunnett’s multiple comparison).

Since it was previously reported based upon binding studies that efavirenz does not interact with muscarinic receptors (Gatch et al. 2013), and our data indicated that it inhibits the agonist responses at the human M3 receptor, we investigated if efavirenz interacts with a different Gq- coupled muscarinic receptor. The M1 receptor was selected because it is the predominant muscarinic receptor in the brain, and because it plays an important role in cognition and learning and memory (Gould et al. 2015; Jiang et al. 2014; Lange et al. 2015; Puri et al. 2015). CHO cells stably expressing the human M1 receptor were used, to test for possible interactions of efavirenz with the M1 receptor. While efavirenz alone failed to stimulate the M1 receptor, at concentrations

≥ 10 µM it suppressed the Gq response mediated by an EC50 concentration of carbachol, and concentrations ≥ 30 µM resulted in complete inhibition (Figure 8C-D, P < 0.05 ANOVA followed by Dunnett’s multiple comparison). No Gq-mediated response was elicited by efavirenz, acetylcholine or carbachol in untransfected CHO cells (data not shown).

In HeLa cells stably expressing the Gs-coupled rat 5-HT6 receptor, efavirenz significantly reduced the basal level of cAMP activity (Figure 9A). At a lower concentration (3 µM), efavirenz first diminished 5-HT’s agonist effect, and at a higher concentration (30 µM), not only

155 completely overcame the agonist effect of 5-HT but also reduced activity to below basal levels

(Figure 9A). Clozapine is an atypical antipsychotic drug that has been reported to interact with the 5-HT6 receptor (Monsma et al. 1993) acting as an inverse agonist (Purohit et al. 2003). Like efavirenz, a high concentration of clozapine (30 µM) significantly reduced basal levels of cAMP, and not only completely overcame the agonist effects of 5-HT but also reduced cAMP levels to below basal (Figure 9A). However, in untransfected HeLa cells, which lack the 5-HT6 receptor, neither efavirenz (30 µM), clozapine (30 µM), nor 5-HT (1 µM) produced significant functional effects compared with the vehicle control (Figure 9B). This demonstrates that the functional responses we observed for efavirenz, clozapine, and serotonin were mediated via the 5-HT6 receptor. HeLa cells have been reported to express an endogenous Gs-coupled β-adrenergic receptor (Tallman et al. 1977). Thus, as a further control, we invoked a cAMP functional response in untransfected HeLa cells by stimulating an endogenous Gs-coupled β-adrenergic receptor with (-)-epinephrine, and this effect was blocked by the β-adrenergic receptor antagonist timolol (Figure 9B). In this case though, efavirenz failed to significantly modulate (-)- epinephrine’s stimulation of the β-adrenergic receptor response (Figure 9B). Thus, efavirenz exerts no cAMP-mediated functional effect in untransfected HeLa cells and it additionally failed to modulate functional responses produced by stimulation of the endogenous β-adrenergic receptors.

Although efavirenz did not significantly displace the binding of the 5-HT3 receptor

3 antagonist [ H]-granisetron from the 5-HT3A receptor (Figure 1), we reasoned that since this receptor subtype is a ligand-gated ion channel efavirenz might act as a pore blocker or a direct channel activator, and thus a follow-up functional assay was warranted. Consistent with the idea that it might act as a pore blocker, efavirenz (10 µM) alone had no effect (n = 5, data not shown),

156 but when co-applied with an EC30 concentration of 5-HT, which will open the channel pore, efavirenz inhibited the 5-HT3A receptor currents in a concentration-dependent manner (0.1-30

µM, Figure 10B). At higher concentrations, efavirenz had a greater blocking effect on the steady state current than peak current (Figure 10A), with 30 µM producing a complete inhibition of the steady-state currents. The efavirenz-induced inhibition of 5-HT activated current was completely reversible when treated with concentrations of efavirenz up to 10 µM and partially recovered at

30 µM after 2-3 min washout (data not shown).

We previously demonstrated (Gatch et al. 2013) that efavirenz (10 µM) blocks reuptake of serotonin via the serotonin transporter (SERT). Here we demonstrate that the same concentration of efavirenz also significantly, though modestly, inhibits (36%) monoamine oxidase A, an enzyme responsible for metabolizing 5-HT (Figure 11).

3.3. Other ARVs (FTC, NVP, ZDV) have a different receptor profile than efavirenz

In light of the findings that efavirenz has the most pronounced binding interactions with the 5-HT2A, 5-HT2C, and 5-HT6 receptors, we sampled other ARV drugs to see if they also had notable interactions with these receptor subtypes. Although no significant interactions for ZDV,

FTC or NVP were observed for any of these receptors at the original profiling concentration of

10 µM (data not shown), we tested these other ARVs to their limit of solubility for the particular assay (60 µM). At these highest concentrations, only NVP displayed a significant, albeit comparatively weak interaction, with the 5-HT6 receptor (Figure 12).

Given that efavirenz was shown previously to potentiate GABAA receptor currents

(Gatch et al. 2013), we also investigated the effects of these same three ARV drugs on the

GABAA receptor. Both ZDV and FTC inhibited the GABA-activated currents in α1β22GABAA receptors with reductions of 36-50% of control at 30 µM (Figure 13). The inhibitory effect had a

157 very rapid onset (msec), and was reversible upon removal of the drugs (Figure 13A). In contrast, co-application of NVP (30 µM) with GABA, caused only a small, though significant, 8% potentiation of GABA-activated currents that was also rapid and reversible (Figure 13B).

4. Discussion:

In a previous study focusing on its recreational use (Gatch et al. 2013), efavirenz was reported to interact with a number of known molecular targets for drugs of abuse (e.g., GABAA and 5-HT2A receptors, and DAT, SERT and VMAT2 transporters), and to have a prevailing behavioral effect in rodents most similar to an LSD-like activity mediated by the 5-HT2A receptor. Here our attention was primarily on discovering molecular mechanisms potentially relevant to the reported CNS toxicity of efavirenz in those taking a prescribed dose for the treatment of HIV-1. In this context, we additionally refined our focus on serotonergic systems as it was suspected that other components of this signaling family were likely involved. Indeed, efavirenz was found to interact with the entire 5-HT2 subfamily of receptors, the ionotropic 5-

HT3A receptor, the 5-HT6 receptor and MAO-A. The most prominent of the newly discovered interactions were followed up here with detailed mechanistic studies.

While efavirenz was reported to interact with the 5-HT2C receptor (Gatch et al. 2013), its activity at that receptor was never elucidated. We report here for the first time that efavirenz acts as an antagonist of the 5-HT2C receptor as it has no agonist activity alone but it concentration- dependently blocked the agonist response by serotonin. Given that ligands which interact with both the 5-HT2C and 5-HT2A receptors typically display similar functional properties, we re- evaluated the activity of efavirenz on the 5-HT2A receptors. Using cloned 5-HT2A receptors and either changes in intracellular calcium or IP1 accumulation as measures of activation of the Gq-

2+ PLC-IP3-[Ca ]i signaling pathway, efavirenz had no detectable agonist activity, but instead

158 blocked multiple different agonists (i.e., 5-HT, α-methyl-5-HT, TCB-2, (±)-DOI, and LSD) from activating the 5-HT2A receptor. Though it is unclear what the exact reason might be for the discrepancy in efavirenz’s activity reported here and a previous collaborative report employing different methodology and conditions (Gatch et al. 2013), it cannot be due to a greater sensitivity of activation as measured by radiolabeled inositol phosphates because measurements of IP3- mediated changes in intracellular calcium are much more sensitive than measures of changes in inositol phosphates (e.g., IP3) (Dickson et al. 2013). Since radioligand binding assays clearly demonstrate efavirenz interacts with the 5-HT2A receptor and in the present report efavirenz acts as an antagonist of Gq-mediated agonist responses, Schild analysis was performed to determine its antagonist potency (KB). No significant difference in efavirenz’s antagonist potency (KB) was observed using either (±)-DOI or LSD as the agonist, indicating an absence of allosteric probe dependence (Armour et al. 1999; Kenakin 2014). Regardless of whether the agonist employed was (±)-DOI or LSD, the slopes of the Schild plots for efavirenz were not different from unity.

The implication is that efavirenz competes for the same binding site on the 5-HT2A receptor as either of these known hallucinogens. Our finding that efavirenz behaves as an antagonist of the

5-HT2A receptor-mediated Gq signaling is consistent with efavirenz affecting human sleep patterns in a manner similar to the 5-HT2A receptor antagonist MK-0454 (Simen et al. 2015).

In large clinical studies (1008 patients), hallucinations have been reported as one of the NPAEs associated with efavirenz-containing treatment regimes, but the incidents were relatively low

(1.2%) compared to other NPAEs (Sustiva 1998). Further, the frequency and severity of all

NPAEs are highest during the first couple weeks of treatment and tend to disappear with continued treatment in some, but not all patients (Mills et al. 2013; Nelson et al. 2013). An even higher frequently (6-31%) of hallucinations associated with efavirenz treatment have been

159 reported in three other clinical studies, each involving between 114-197 patients (Gutierrez-

Valencia et al. 2009; Lochet et al. 2003; Mukonzo et al. 2013). An HIV-negative individual with a history of psychedelic drug experience reported, upon first time experimental ingestion of efavirenz (initially 300 mg stepped up to a full 600 mg dose p.o.), that it has a classical psychedelic quality with acute effects somewhat comparable to the hallucinogen mescaline

(Morris 2014). Thus, it seems clear that efavirenz, at the prescribed dose is capable of inducing hallucinations in humans in some instances. Given our finding that efavirenz competitively antagonizes the 5-HT2A receptor-mediated Gq signaling responses produced by the hallucinogenic 5-HT2A receptor partial agonists (±)-DOI and LSD, the question becomes whether efavirenz mediates hallucinogenic responses via the 5-HT2A receptor, as is the case for classical hallucinogenic drugs like LSD and DOI, or via some other receptor mechanism.

The 5-HT2A receptor is an established site for hallucinogenic drug action. The hallucinogenic indoleamine psilocybin increases cerebral metabolic rate in the frontal cortex and anterior cingulate cortex (Vollenweider et al. 1997), and pretreatment with the 5-HT2A antagonist ketanserin dose-dependently prevents hallucinations caused by psilocybin in humans

(Vollenweider et al. 1998). The frontal cortex-mediated head twitch response (Vickers et al.

2001; Willins and Meltzer 1997), a hallucinogenic behavioral proxy in rodents, is absent in 5-

HT2A knockout mice (Gonzalez-Maeso et al. 2007). In wild type mice the head-twitch response is blocked by the 5-HT2A antagonists, MDL100,907 and ketanserin, as is the increase in synaptic activity in the frontal cortex neurons induced by application of the hallucinogenic 5-HT2A agonists DOI and α-methyl-5-HT (Beique et al. 2007; Marek and Aghajanian 1996). To our knowledge, all known hallucinogens which signal via the 5-HT2A receptor happen to also act as

Gq signaling agonists; however, we show here that efavirenz functions as an antagonist of Gq

160 signaling. In Gq homozygous knockout mice, DOI is still capable of producing a significant, albeit blunted, head-twitch response, which suggests that the Gq signaling pathway is not the sole mediator of the 5-HT2A receptor dependent behavioral effects of hallucinogens (Garcia et al.

2007). In a previous report, efavirenz was shown to invoke a head-twitch response of remarkably reduced duration and intensity compared to other hallucinogens like LSD and DOI in wild type mice, but failed to induce this response in 5-HT2A knockout mice (Gatch et al. 2013). One possibility is that the comparatively diminished behavioral response for efavirenz, relative to other hallucinogens, might be related to efavirenz’s lack of Gq-mediated agonist signaling.

Efavirenz also has other behavioral properties similar to LSD, including in tests of discrimination where efavirenz partially substituted for LSD and vice versa (Gatch et al. 2013). Taken together, it seems clear that efavirenz interacts with the 5-HT2A receptor and has psychoactive properties, though it also has properties distinct from classical hallucinogens including that it fails to stimulate Gq signaling via the 5-HT2A receptor.

An alternative explanation is that the hallucinogenic properties of efavirenz are not exerted via its interactions with the 5-HT2A receptor, but instead might be attributed to its interactions with a different receptor system. A previous study using individual cloned receptors capable of mediating hallucinations revealed that efavirenz (10 µM) did not have significant interactions with dopamine receptors or NMDA receptors (Gatch et al. 2013). That same study also reported that efavirenz (10 µM) did not have high affinity interactions with mixed populations of muscarinic receptors present in brain tissue (Gatch et al. 2013); however, this mixed receptor population approach utilizing a radiolabeled antagonist is not well suited to reveal moderate interactions with a single subtype nor would it reveal allosteric interactions with agonists. Though efavirenz alone had no effect on intracellular calcium levels in untransfected

161

HEK293 cells, during the course of evaluating the possible modulation of endogenous Gq- coupled receptor-mediated responses in untransfected HEK293 cells, we discovered that efavirenz is able to partially antagonize agonist responses produced by endogenous M3 muscarinic receptors. The plateauing of the inhibitory response for efavirenz at the muscarinic

M3 receptor and the lack of any 5-HT response in untransfected cells, clearly demonstrates efavirenz has a different mechanistic profile at the endogenous M3 receptor than at the exogenously expressed 5-HT2A and 5-HT2C receptors. Given that efavirenz exerted a modest and partial attenuation of agonist actions on the M3 receptor, we investigated whether efavirenz might also affect the cloned human M1 receptor. In contrast to its effect on the M3 receptor, efavirenz was able to completely block agonist activation of the M1 receptor, like the muscarinic receptor antagonists scopolamine and atropine. In this context it is worth noting that scopolamine

(Golding and Stott 1997) has psychoactive properties that include hallucinations, vivid dreams, and memory impairment (Cotroneo 2013; Jalali et al. 2014; Lin et al. 2011; Seo et al. 2009;

Vallersnes et al. 2009). Similarly, atropine, another muscarinic receptor antagonist (Ma et al.

2013), is also known to cause hallucinations (Arthurs and Davies 1980; Baker and Silver 1984;

Fisher 1991). Hence, partial and complete blockade of the M3 and M1 receptors, respectively, by efavirenz might account for or contribute to its hallucinogenic properties. The M1 receptor is the most abundant muscarinic subtype in the brain and is known to play a role in cognition, learning and memory, and vigilance. M1 knockout mice exhibit cognitive deficits (Gould et al. 2015;

Jiang et al. 2014), hence blockade of the M1 receptor by efavirenz may contribute to the cognitive impairment observed in patients taking efavirenz (Ciccarelli et al. 2011; Sustiva 1998).

Although efavirenz reduces the 5-HT3A receptor currents, like a number of 5-HT3 receptor antagonists, it did not displace [3H]granisetron (also known as [3H]BRL-43694) binding and thus

162 is not a competitive antagonist for the 5-HT ligand binding site like granisetron is (Rojas et al.

2008). However, efavirenz does block 5-HT3 receptor mediated currents in the presence of 5-HT, but in contrast to ondansetron, a competitive antagonist of the 5-HT site that reduces peak currents more strongly than steady-state currents (Rammes et al. 2004), efavirenz reduces steady-state currents (as measured by end-application of current) more strongly than peak currents. This latter inhibition profile by efavirenz suggests it might act as an open-channel blocker of the 5-HT3A receptor similar to picrotoxin (Das and Dillon 2005), though the precise mechanism remains to be elucidated and we cannot rule out that efavirenz behaves as some sort of complex allosteric modulator.

In contrast to its actions on the 5-HT2A, 5-HT2C, and 5-HT3A receptors, at the 5-HT6 receptor efavirenz functions as an inverse agonist as evidenced by its ability to significantly reduce basal levels of cAMP in the absence of 5-HT. Also consistent with an inverse agonist, the

5-HT responses were first diminished then completely overcome by increasing concentrations of efavirenz even to the point where basal levels of activity were reduced. Clozapine, an accepted

5-HT6 receptor inverse agonist (Purohit et al. 2003), produced the same functional effect as efavirenz and served as our positive control for inverse agonism. These effects were observed using a rat receptor stably expressed in HeLa cells and did not require special measures to observe the inverse agonist activity as has been reported when the rat 5HT6 receptor is expressed in other cell lines (Romero et al. 2006). The functional responses we observed for efavirenz, clozapine, and serotonin were all mediated by the cloned 5-HT6 receptor, since these compounds produced no significant changes in cAMP levels in untransfected HeLa cells which lack any detectable level of 5-HT6 expression. Further, the agonist-induced increase in cAMP that is

163 mediated by endogenous β-adrenergic receptors was unaffected by efavirenz and suggests that efavirenz is not exerting a non-specific effect in these cells.

4.1 CNS targets of other ARVs

Given that efavirenz has multiple off-target activities and some of these appear to correlate with certain NPAEs observed in humans, we investigated other ARV drugs for possible interactions with some of these off-target receptors. ZDV and NVP have been occasionally associated with reports of CNS toxicity, while FTC has not. For example, ZDV use has been associated with malaise, fatigue and insomnia (Fischl et al. 1990; McLeod and Hammer 1992).

Further, a number of case reports have been published describing the onset of dose-related

NPAEs associated with ZDV for the treatment of HIV-1, including depressive and manic syndromes primarily accompanied by auditory hallucinations and paranoid delusions (Maxwell et al. 1988; Wright et al. 1989). Case reports of NPAEs for NVP include hallucinations, vivid dreaming or psychosis in conjunction with depression and concomitant delusions soon after initiating NVP in patients with no previous history of psychiatric illness or substance use

(Morlese et al. 2002; Wise et al. 2002). These NPAEs subsided after withdrawal of NVP. We report here that, in general, ZDV, NVP and FTC did not share the same complex profile as efavirenz with respect to interactions with 5-HT2A, 5-HT2C and 5-HT6 receptors. In fact, only in the case of the 5-HT6 receptor and NVP, and only at a much higher concentration, was an interaction even detectable (estimated affinity would correspond to ~65 µM or approximately 30- fold lower than for efavirenz). Since we had found previously that efavirenz acted as an allosteric potentiator of GABAA receptors (Gatch et al. 2013), the actions of ZDV, NVP and FTC on this receptor were examined as well. The most pronounced modulatory effect was observed for ZDV and FTC, and, in contrast to efavirenz, these ARV drugs inhibited, rather than potentiated the

164

GABA response. While the investigation of other ARV drugs was limited in scope, it seems clear that the complex multiple off-target profile for efavirenz is unique amongst the ARV drugs examined. Though no rodent behavior clearly related to the GABAA receptor potentiating effects of efavirenz has been elucidated, the finding that efavirenz potentiates and ZDV and FTC inhibit

GABA currents might still have relevance for HAART therapy given that the standard of HIV-1 care is combined drug treatments.

4.2 Speculation concerning efavirenz’s CNS targets possibly contributing to NPAEs

Given the molecular mechanistic insights gained from our studies, we became curious about which one(s) of the prominent NPAEs might be most likely to be mediated by 5-HT3A and

5-HT6 receptors. The following is educated speculation based upon inference drawn from the in vitro receptor pharmacology. Dizziness and headache are amongst the most frequently reported adverse events for 5-HT3 receptor antagonists (Goodin and Cunningham 2002; Haus et al. 2004;

Tralongo et al. 2004). Since efavirenz blocks 5-HT3 receptor currents, it seems reasonable that 5-

HT3 receptor antagonism may contribute in part to efavirenz-induced dizziness and headache

(Cohen et al. 2014; Mills et al. 2013; Smith et al. 2009; Sustiva 1998). HIV-1 patients are advised to take efavirenz before going to bed and sleep disturbances are commonly reported

NPAEs including both insomnia and somnolence (Clifford et al. 2005; Cohen et al. 2014; Nunez et al. 2001; Sustiva 1998; Waters et al. 2011). Several serotoninergic targets for efavirenz have been identified in this and a previous report (5-HT2A, 5-HT2C, 5-HT3, 5-HT6, 5-HT2B, MAO-A, and SERT) , and given the complex role that serotonin plays in the regulation of sleep and wakefulness and the known involvement of each of these serotonergic targets (Monti and Jantos

2011; Monti 2011; Morairty et al. 2008; Ponzoni et al. 1993; Real et al. 2007; Wisor et al.

2003), it seems reasonable to infer that one or more of these targets would be implicated in

165 efavirenz’s sleep-related NPAEs. Consistent with this idea, and as noted above, efavirenz alters certain aspects of human sleep patterns in a manner similar to that of a 5-HT2A receptor antagonist (Simen et al. 2015).

Other reported NPAEs for efavirenz include depression, anxiety and feelings of stress

(Cohen et al. 2014; Munoz-Moreno et al. 2009; Rihs et al. 2006; Sustiva 1998; Waters et al.

2011). Homozygous 5-HT2A knockout mice exhibit an anxiodepressive-like phenotype compared to wild type mice, possibly due to diminished 5-HT2A receptor mediated transmission (Petit et al.

2014). Though we demonstrated that efavirenz acts as an antagonist of 5-HT2A/2C–mediated Gq signaling, we also discovered that chronic application of efavirenz reduces 5-HT2A receptor density and the responsiveness of this receptor to serotonin. The inference here would be that efavirenz could also be reducing 5-HT2A mediated neurotransmission thus precipitating an anxiodepressive-like phenotype.

5. Conclusion:

In this report, we detail the molecular mechanisms of action of efavirenz across a family of serotonergic targets. Our radioligand binding studies revealed efavirenz’s interactions with the

5-HT2A, 5-HT2B, 5-HT2C and 5-HT6 receptors. Functional studies showed that efavirenz acts as an antagonist of Gq-coupling at the 5-HT2A and 5-HT2C receptors, an inverse agonist at the Gs- coupled 5-HT6 receptor, a 5-HT3A receptor pore blocker, and a modest inhibitor of MAO-A.

Within a similar concentration range, efavirenz also completely or partially blocks agonist responses at the M1 and M3 muscarinic receptors, respectively. Schild shift analysis suggests that efavirenz competes with LSD and (±)-DOI for binding to the same site on the 5-HT2A receptor.

5-HT2A receptor density and responsiveness to 5-HT are reduced upon prolonged treatment with efavirenz. Other ARVs like NVP, FTC, and ZDV did not share the same complex pharmacology

166 as efavirenz; only NVP interacted with the 5-HT6 receptor and potentiated GABAA receptors currents, like efavirenz, albeit with much smaller effect sizes. FTC and ZDV inhibited, instead of potentiating, GABAA receptor currents. Overall, the data suggests that efavirenz has complex receptor pharmacology with varied functional responses that is unique among ARV drugs tested in this report.

167

Figure 1. Efavirenz has significant interactions with the cloned metabotropic 5-HT2A, 5-

HT2B, 5-HT2C and 5-HT6 receptors. At a concentration of 10 μM, efavirenz displaced greater than 50% of specifically bound radioligand from the 5-HT6 receptor and 5-HT2 subfamily of receptors, while having a less pronounced or no detectable interaction with the other serotonin receptor subtypes: 5-HT1A, 5-HT3A, 5-HT4, 5-HT5, 5-HT7. A one-way ANOVA followed by a

Bonferroni post-hoc analysis set to high stringency (P < 0.001) indicated significant displacement for those groups having an asterisk. When considered as a group, efavirenz has similar robust interactions with 5-HT2A, 5-HT2C and 5-HT6 receptors, but significantly weaker interactions with the 5-HT2B receptor as determined by a one-way ANOVA followed by a

Bonferroni post-hoc analysis (P < 0.05) and indicated by the dagger. All data are presented as averaged values from multiple experiments ± SEM. Displacement data for the 5-HT1A, 5-HT2A and 5HT2C receptors were adapted from Gatch et al., 2013 (Gatch et al. 2013) and shown here for completeness. These series of experiments were performed by others but are included here for completeness.

168

Figure 1. Efavirenz has significant interactions with the cloned metabotropic 5-HT2A, 5-

HT2B, 5-HT2C and 5-HT6 receptors.

169

Figure 2. Efavirenz has low micromolar affinity for cloned metabotropic 5-HT2A, 5-HT2C and 5-HT6 receptors. The affinity of efavirenz for each receptor was determined at equilibrium by concentration-dependent competition binding of radiolabeled receptors using rapid filtration.

All data are presented as averaged values from multiple experiments ± SEM. A, The 5-HT2A receptor binds to efavirenz with an affinity (Ki) = 4.4 ± 0.43 µM and a pseudo Hill slope = -1.39

± -0.18. B, The 5-HT2C receptor binds to efavirenz with an affinity (Ki) = 1.7 ± 0.28 µM and a pseudo Hill slope = -1.22 ± -0.18. C, The 5-HT6 receptor binds to efavirenz with an affinity (Ki)

= 2.2 ± 0.27 µM and a pseudo Hill slope = -1.36 ± -0.14. In this series of experiments, the 5-

HT2A and 5-HT6 experiments were performed by me, and the 5-HT2C experiments were performed by other but are included here for completeness.

170

Figure 2. Efavirenz has low micromolar affinity for cloned metabotropic 5-HT2A, 5-HT2C

and 5-HT6 receptors.

171

Figure 3. The cloned metabotropic 5-HT2A, 5-HT2C and 5-HT6 receptors are potently activated by serotonin. Functional dose-responses to serotonin for the Gq-coupled 5-HT2 receptors were measured as an increase in intracellular calcium using the calcium sensitive fluorescent dye Calcium-6 QF (Ex485/Em525), while the functional dose-response for the Gs- coupled 5-HT6 receptor was measured as an increase in cAMP detected using the bioluminescent cAMP-Glo assay. All data are presented as averaged values from multiple experiments ± SEM.

A, Serotonin activates the 5-HT2A receptor with an average potency (EC50) range from 13-17 nM and a pseudo Hill slope range from 0.55-0.70 that was insensitive the serum condition used for culturing (P > 0.05 ANOVA with a Bonferroni post-hoc). B, Serotonin activates the 5-HT2C receptor with a potency (EC50) = 1.5 ± 0.18 nM and a pseudo Hill slope = 1.4 ± 0.19 when cells were cultured in dialyzed FBS and a potency (EC50) = 0.039 ± 0.0039 nM and a pseudo Hill slope = 1.5 ± 0.17 when cells were cultured in absence of serum. No response was observed when cells were culture in FBS. C, Serotonin activates the 5-HT6 receptor with a potency (EC50)

= 12.7 ± 6.2 nM and a pseudo Hill slope = 1.02 ± 0.35.

172

Figure 3. The cloned metabotropic 5-HT2A, 5-HT2C and 5-HT6 receptors are potently

activated by serotonin.

173

Figure 4. Efavirenz is a 5-HT2A receptor antagonist of Gq signaling. All data are presented as averaged values from multiple experiments ± SEM. A, Efavirenz behaves as an antagonist at the cloned 5-HT2A receptor when a change in intracellular calcium is used as a measure of Gq- signaling and the 5-HT2A receptor is stimulated with different classes of agonists: 5-HT, TCB-2 and α-Me-5HT. MDL100,907, a potent and selective 5-HT2A receptor antagonist was used as a positive control for antagonist activity. A one-way ANOVA followed by a Bonferroni post-hoc analysis set to high stringency (*P < 0.001) indicated significant inhibition of agonist responses by both 30 µM efavirenz and 30 µM MDL100,907. B, To rule out the possibility that weak partial agonist properties of efavirenz might only be observed at a very high concentration, a higher maximal concentration of efavirenz was tested using IP-One assay conditions than was possible under the Ca2+ assay conditions. Nevertheless, at the highest concentration that could be tested (100 µM) in the IP-One assay, efavirenz did not display any agonist activity. A one-way

ANOVA followed by Dunnett’s multiple comparison test set to high stringency (P < 0.001) indicated by an asterisk revealed only the response by 5-HT to be significantly different from vehicle.

174

Figure 4. Efavirenz is a 5-HT2A receptor antagonist of Gq signaling.

175

Figure 5. Schild shift analysis of efavirenz at the cloned 5-HT2A receptor suggests that it competes for the same binding site as (±)-DOI and LSD. A & D, Gq-signaling was measured as a change in intracellular calcium. Data are presented as averaged values from multiple experiments ± SEM. For clarity error bars are not shown and the Schild data are normalized to

100% response, even though both (±)-DOI and LSD are partial agonists of the 5-HT2A receptor,

P < 0.01, two-tailed t-test as shown in B and E. (±)-DOI and LSD concentration-response curves of the Gq-mediated functional responses in the presence of increasing fixed concentrations of efavirenz induce a progressive shift to the right with no reduction in efficacy. C, Schild plot transformation of the data shown in panel A suggests that efavirenz’s interaction with (±)-DOI is purely competitive given a slope equal to 1.06 ± 0.054. The pA2 or antagonist potency (KB) value for efavirenz competition of (±)-DOI is 1.3 ± 0.20 µM. Panel F suggests that efavirenz’s interaction with LSD is also purely competitive with a slope of 1.2 ± 0.16. The pA2 value for efavirenz competition of LSD is 0.31 ± 0.18 µM. In both panels C and F, the slopes of the Schild plot lines are not significantly different from unity (P > 0.05) and a two-site model is not significantly better than a one site model (P > 0.05).

176

A D

B E

C F

Figure 5. Schild shift analysis of efavirenz at the cloned 5-HT2A receptor suggests that it

competes for the same binding site as (±)-DOI and LSD.

177

Figure 6. Chronic exposure to efavirenz reduces the 5-HT2A receptor density and 5-HT mediated Gq-activation at the 5-HT2A receptor. Cells were exposed (pretreated) for 3 days to

DMSO vehicle [gray bars], serotonin (30 µM) [green bars] or efavirenz (30 µM) [blue bars], or to DMSO vehicle for 3 days and efavirenz for only 15 min on the third day [red bars]. The latter category was to ensure that our wash procedure was able to effectively remove all traces of efavirenz. After extensive washes, the cells from all groups were tested for their ability to mount functional responses to serotonin in the absence and presence of efavirenz (10 µM). All data were normalized to a maximal response generated by 100 µM 5-HT following the 3-day DMSO pretreatment and are presented as averaged values from multiple experiments ± SEM. A, Cells exposed to DMSO vehicle for 3 days elicited a robust calcium response when stimulated with a submaximal concentration of 5-HT (100 nM), and 10 µM efavirenz was able to antagonize this response by 48%. The same result was obtained when cells were exposed to efavirenz for only

15 min following 3 day exposure to DMSO, demonstrating that efavirenz could be easily removed by washing. Compared to the 3 day vehicle treated group, three day exposure to 30 µM

5-HT reduced the 100 µM maximal 5-HT response by 27%, and 10 µM efavirenz antagonized the 100 nM 5-HT submaximal response by 49%. In contrast to the other groups, 3 day exposure to 30 µM efavirenz drastically reduced the maximal 5-HT response by 81%, but the magnitude of efavirenz’s antagonist effect on the 100 nM 5-HT submaximal response was similar to all the other groups (i.e., a 49% reduction). The comparisons being made between specified groups within a category are indicated by interconnecting gray lines and show a significant inhibition by efavirenz. A bracket with asterisk indicates a significant difference between the 3 day DMSO category and the remaining categories for comparisons between matched groups. A two-way

ANOVA followed by Bonferroni multiple comparisons test was performed to identify significant

178 differences within and between the various pretreatment categories at P < 0.05. B, Three day exposure to 30 µM efavirenz resulted in a significant 52% decrease in receptor density (Bmax).

However, none of the other exposure conditions had a significant effect on Bmax. A one way

ANOVA followed by Bonferroni’s multiple comparisons test was performed to identify significant differences between various pretreatment categories. * P < 0.05 between the 3 day efavirenz pretreatment category and all the other pretreatment categories.

179

Figure 6. Chronic exposure to efavirenz reduces the 5-HT2A receptor density and 5-HT

mediated Gq-activation at the 5-HT2A receptor.

180

Figure 7. Efavirenz is an antagonist of the cloned 5-HT2C receptor. Gq-signaling was measured as a change in intracellular calcium. All data are presented as averaged values from multiple experiments ± SEM. A, Efavirenz dose-dependently antagonizes agonist responses produced by an EC50 concentration of 5-HT (EC50 data normalized to 100%). For these experiments cells were maintained in serum free media for at least 12 hours, pretreated with efavirenz, then stimulated with 0.03 nM serotonin, which is the EC50 for 5-HT for this receptor under serum free conditions (see Figure 3B). Values significantly different from 0.03 nM 5-HT,

(Dunnett’s multiple comparison test set to high stringency, P < 0.001) are denoted by asterisks.

B, Efavirenz did not have any agonist activity at the 5-HT2C receptor under similar conditions.

181

Figure 7. Efavirenz is an antagonist of the cloned 5-HT2C receptor.

182

Figure 8. Efavirenz partially inhibits the activation of the M3 muscarinic receptor, and completely inhibits the activation of the M1 muscarinic receptor. Gq-signaling was measured as a change in intracellular calcium. All data are presented as averaged values from multiple experiments ± SEM. A, In untransfected HEK293 cells, efavirenz does not stimulate Gq-coupling

2+ as measured by changes in intracellular Ca . However, these cells express an endogenous Gq- coupled M3 muscarinic receptor (Ancellin et al. 1999; Atwood et al. 2011; Luo et al. 2008) that was activated by acetylcholine, and this response was blocked by the muscarinic receptor antagonist scopolamine. B, Efavirenz partially suppresses ACh mediated M3 receptor activation at concentrations ≥ 10 µM but was unable to produce a complete block at the highest concentration that could be tested under these assay conditions. C, Efavirenz fails to stimulate

Gq-coupling in CHO cells stably expressing the human M1 receptor, but this receptor can be stimulated with the muscarinic receptor agonist carbachol and the muscarinic receptor antagonists scopolamine and atropine completely block carbachol’s effect. D, Efavirenz antagonizes carbachol-mediated M1 receptor activation in a concentration dependent manner. At

10 µM there is a significant inhibition of the response produced by an EC50 concentration of carbachol (normalized to 100%) and a complete inhibition at concentrations ≥ 30 µM. A one- way ANOVA followed by Dunnett’s multiple comparison test was performed to identify significant differences between the agonist response in the absence of efavirenz compared to the agonist response in the presence of different concentrations of efavirenz. * P < 0.05 for those responses in the presence of efavirenz that are significantly different from the agonist response in the absence of efavirenz.

183

A B

C D

Figure 8. Efavirenz partially inhibits the activation of the M3 muscarinic receptor, and

completely inhibits the activation of the M1 muscarinic receptor.

184

Figure 9. Efavirenz behaves as an inverse agonist of the cloned 5-HT6 receptor. Gs-signaling was measured as a change in intracellular cyclic AMP (cAMP) levels. All data are presented as averaged values from multiple experiments ± SEM. A, Efavirenz (30 µM) acts as an inverse agonist of the cloned 5-HT6 receptor stably expressed in HeLa cells as it lowers the basal levels of cAMP. Only at the higher concentrations (3-30 µM) can efavirenz overcome the stimulatory effect of 60 nM serotonin. The atypical antipsychotic clozapine (30 µM), a known inverse agonist, was used as a positive control for the 5-HT6 inverse agonist response. One-way ANOVA followed by a Dunnett’s post-hoc analysis comparing the 60 nM 5-HT group to groups containing increasing concentrations of efavirenz indicated that cAMP levels were significantly lower in experimental groups containing ≥ 3 µM efavirenz (P < 0.05). B, There is no stimulatory effect of 5-HT or inverse agonism by efavirenz or clozapine in untransfected HeLa cells. HeLa cells express endogenous β-adrenergic receptors whose stimulation by (-)-epinephrine can be blocked by the β-adrenergic receptor antagonist timolol but not by efavirenz.

185

Figure 9. Efavirenz behaves as an inverse agonist of the cloned 5-HT6 receptor.

186

Figure 10. Efavirenz inhibits cloned 5-HT3A receptor currents. All recording were performed on HEK293 cells transiently expressing mouse 5-HT3A receptors. A, Representative traces showing whole-cell currents activated by an EC30 concentration 0.6 µM 5-HT in the absence and presence of co-applied (20 sec) increasing concentrations (0.1-30 µM) of efavirenz. Note that efavirenz produces a greater inhibition on the steady-state current amplitude as the concentration increases. B, Summary data for effect of efavirenz on 5-HT-induced currents. The currents at the end of ligand application were normalized to the current amplitude at the same time point as in the control (5-HT alone). Efavirenz’s inhibitory concentration (IC50) is 4.9 ± 0.7 µM with a Hill coefficient of 1.2 ± 0.19 (n = 10). All data are presented as means ± SEM. These series of experiments were performed by others but included here for completeness.

187

Figure 10. Efavirenz inhibits cloned 5-HT3A receptor currents.

188

Figure 11. Efavirenz modestly inhibits monoamine oxidase-A. Efavirenz (10 μM) decreases the activity of MAO-A by 36 ± 1.7%. The mean value from multiple experiments ± SEM is plotted. Student’s t-test was used to determine statistical significance (P < 0.05). This series of experiments were performed by others but are included here for completeness.

189

Figure 11. Efavirenz modestly inhibits monoamine oxidase-A.

190

Figure 12. Nevirapine, but not emtricitabine or zidovudine, interacts with the cloned 5-HT6 receptor. All of the ARV drugs were tested at the highest concentration (60 µM) that would stay in solution under the assay conditions. All data are presented as averaged values from multiple experiments ± SEM. A two-way ANOVA followed by a Bonferroni post-hoc indicated that interaction of nevirapine with the 5HT6 receptor is significantly less than its vehicle control (P <

0.001). This series of experiments were performed by others but are included here for completeness.

191

Figure 12. Nevirapine, but not emtricitabine or zidovudine, interacts with the cloned 5-HT6 receptor.

192

Figure 13. The ARV drugs zidovudine (ZDV), emtricitabine (FTC), but not nevirapine

(NVP), inhibit cloned GABAA receptor currents. A, Representative traces showing whole cell currents activated by an EC30 concentration of GABA (5 µM) in HEK293 cells stably expressing

α1β2γ2GABAA receptors. Efavirenz was co-applied with GABA for 10 sec. Note that ARV drug- induced action was rapid and reversible. B, Summary data of the modulatory effect induced by

ZDV, FTC and NVP (30 µM). All currents were normalized to the GABA control response

(GABA alone). *P < 0.05, paired t-test, compared to the control, n=6-8. Data are presented as means ± SEM. This series of experiments were performed by others but are included here for completeness.

193

Figure 13. The ARV drugs zidovudine (ZDV), emtricitabine (FTC), but not nevirapine

(NVP), inhibit cloned GABAA receptor currents.

194

6. References:

Ancellin N., Preisser L., Le Maout S., Barbado M., Creminon C., Corman B. and Morel A.

(1999) Homologous and heterologous phosphorylation of the vasopressin V1a receptor.

Cell. Signal. 11, 743-751.

Armour S. L., Foord S., Kenakin T. and Chen W. J. (1999) Pharmacological characterization of

receptor-activity-modifying proteins (RAMPs) and the human calcitonin receptor. J.

Pharmacol. Toxicol. Methods. 42, 217-224.

Arthurs G. J. and Davies R. (1980) Atropine--a safe drug. Anaesthesia. 35, 1077-1079.

Atwood B. K., Lopez J., Wager-Miller J., Mackie K. and Straiker A. (2011) Expression of G

protein-coupled receptors and related proteins in HEK293, AtT20, BV2, and N18 cell lines

as revealed by microarray analysis. BMC Genomics. 12, 14-2164-12-14.

Avery L. B., Sacktor N., McArthur J. C. and Hendrix C. W. (2013) Protein-free efavirenz

concentrations in cerebrospinal fluid and blood plasma are equivalent: applying the law of

mass action to predict protein-free drug concentration. Antimicrob. Agents Chemother. 57,

1409-1414.

Baker J. H. and Silver J. R. (1984) Atropine toxicity in acute cervical spinal injury. Paraplegia.

22, 379-382.

Beique J. C., Imad M., Mladenovic L., Gingrich J. A. and Andrade R. (2007) Mechanism of the

5-hydroxytryptamine 2A receptor-mediated facilitation of synaptic activity in prefrontal

cortex. Proc. Natl. Acad. Sci. U. S. A. 104, 9870-9875.

Berg K. A., Maayani S., Goldfarb J., Scaramellini C., Leff P. and Clarke W. P. (1998) Effector

pathway-dependent relative efficacy at serotonin type 2A and 2C receptors: evidence for

agonist-directed trafficking of receptor stimulus. Mol. Pharmacol. 54, 94-104.

195

Best B. M., Koopmans P. P., Letendre S. L., Capparelli E. V., Rossi S. S., Clifford D. B., Collier

A. C., Gelman B. B., Mbeo G., McCutchan J. A., Simpson D. M., Haubrich R., Ellis R.,

Grant I. and CHARTER Group (2011) Efavirenz concentrations in CSF exceed IC50 for

wild-type HIV. J. Antimicrob. Chemother. 66, 354-357.

Chen G., Cho S. J., Huang X. P., Jensen N. H., Svennebring A., Sassano M. F., Roth B. L. and

Kozikowski A. P. (2011) Rational Drug Design Leading to the Identification of a Potent 5-

HT(2C) Agonist Lacking 5-HT(2B) Activity. ACS Med. Chem. Lett. 2, 929-932.

Ciccarelli N., Fabbiani M., Di Giambenedetto S., Fanti I., Baldonero E., Bracciale L.,

Tamburrini E., Cauda R., De Luca A. and Silveri M. C. (2011) Efavirenz associated with

cognitive disorders in otherwise asymptomatic HIV-infected patients. Neurology. 76, 1403-

1409.

Clifford D. B., Evans S., Yang Y., Acosta E. P., Goodkin K., Tashima K., Simpson D., Dorfman

D., Ribaudo H. and Gulick R. M. (2005) Impact of efavirenz on neuropsychological

performance and symptoms in HIV-infected individuals. Ann. Intern. Med. 143, 714-721.

Cohen C., Wohl D., Arribas J. R., Henry K., Van Lunzen J., Bloch M., Towner W., Wilkins E.,

Ebrahimi R., Porter D., White K., Walker I., Chuck S., De-Oertel S. and Fralich T. (2014)

Week 48 results from a randomized clinical trial of rilpivirine/emtricitabine/tenofovir

disoproxil fumarate vs. efavirenz/emtricitabine/tenofovir disoproxil fumarate in treatment-

naive HIV-1-infected adults. AIDS. 28, 989-997.

Colpaert F. C., Meert T. F., Niemegeers C. J. and Janssen P. A. (1985) Behavioral and 5-HT

antagonist effects of ritanserin: a pure and selective antagonist of LSD discrimination in rat.

Psychopharmacology (Berl). 86, 45-54.

196

CORNE S. J., PICKERING R. W. and WARNER B. T. (1963) A method for assessing the

effects of drugs on the central actions of 5-hydroxytryptamine. Br. J. Pharmacol.

Chemother. 20, 106-120.

Cotroneo C. (2013) Devil's Breath: Scopolamine, AKA Burundanga, Hailed As 'World's Scariest

Drug'. 2015.

Cummings D. F., Canseco D. C., Sheth P., Johnson J. E. and Schetz J. A. (2010) Synthesis and

structure-affinity relationships of novel small molecule natural product derivatives capable

of discriminating between serotonin 5-HT1A, 5-HT2A, 5-HT2C receptor subtypes. Bioorg.

Med. Chem. 18, 4783-4792.

Cussac D., Newman-Tancredi A., Quentric Y., Carpentier N., Poissonnet G., Parmentier J. G.,

Goldstein S. and Millan M. J. (2002) Characterization of phospholipase C activity at h5-

HT2C compared with h5-HT2B receptors: influence of novel ligands upon membrane-

bound levels of [3H]phosphatidylinositols. Naunyn Schmiedebergs Arch. Pharmacol. 365,

242-252.

Das P. and Dillon G. H. (2005) Molecular determinants of picrotoxin inhibition of 5-

hydroxytryptamine type 3 receptors. J. Pharmacol. Exp. Ther. 314, 320-328.

Dickson E. J., Falkenburger B. H. and Hille B. (2013) Quantitative properties and receptor

reserve of the IP(3) and calcium branch of G(q)-coupled receptor signaling. J. Gen. Physiol.

141, 521-535.

Dirson G., Fernandez C., Hindlet P., Roux F., German-Fattal M., Gimenez F. and Farinotti R.

(2006) Efavirenz does not interact with the ABCB1 transporter at the blood-brain barrier.

Pharm. Res. 23, 1525-1532.

197

Egan C. T., Herrick-Davis K., Miller K., Glennon R. A. and Teitler M. (1998) Agonist activity of

LSD and lisuride at cloned 5HT2A and 5HT2C receptors. Psychopharmacology (Berl). 136,

409-414.

Fischl M. A., Richman D. D., Hansen N., Collier A. C., Carey J. T., Para M. F., Hardy W. D.,

Dolin R., Powderly W. G. and Allan J. D. (1990) The safety and efficacy of zidovudine

(AZT) in the treatment of subjects with mildly symptomatic human immunodeficiency virus

type 1 (HIV) infection. A double-blind, placebo-controlled trial. The AIDS Clinical Trials

Group. Ann. Intern. Med. 112, 727-737.

Fisher C. M. (1991) Visual hallucinations on eye closure associated with atropine toxicity. A

neurological analysis and comparison with other visual hallucinations. Can. J. Neurol. Sci.

18, 18-27.

Garcia E. E., Smith R. L. and Sanders-Bush E. (2007) Role of G(q) protein in behavioral effects

of the hallucinogenic drug 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane.

Neuropharmacology. 52, 1671-1677.

Gatch M. B., Kozlenkov A., Huang R. Q., Yang W., Nguyen J. D., Gonzalez-Maeso J., Rice K.

C., France C. P., Dillon G. H., Forster M. J. and Schetz J. A. (2013) The HIV antiretroviral

drug efavirenz has LSD-like properties. Neuropsychopharmacology. 38, 2373-2384.

Golding J. F. and Stott J. R. (1997) Comparison of the effects of a selective muscarinic receptor

antagonist and hyoscine (scopolamine) on motion sickness, skin conductance and heart rate.

Br. J. Clin. Pharmacol. 43, 633-637.

Gonzalez-Maeso J., Weisstaub N. V., Zhou M., Chan P., Ivic L., Ang R., Lira A., Bradley-

Moore M., Ge Y., Zhou Q., Sealfon S. C. and Gingrich J. A. (2007) Hallucinogens recruit

198

specific cortical 5-HT(2A) receptor-mediated signaling pathways to affect behavior.

Neuron. 53, 439-452.

Goodin S. and Cunningham R. (2002) 5-HT(3)-receptor antagonists for the treatment of nausea

and vomiting: a reappraisal of their side-effect profile. Oncologist. 7, 424-436.

Gould R. W., Dencker D., Grannan M., Bubser M., Zhan X., Wess J., Xiang Z., Locuson C.,

Lindsley C. W., Conn P. J. and Jones C. K. (2015) Role for the M1 Muscarinic

Acetylcholine Receptor in Top-Down Cognitive Processing Using a Touchscreen Visual

Discrimination Task in Mice. ACS Chem. Neurosci. 6, 1683-1695.

Gutierrez F., Navarro A., Padilla S., Anton R., Masia M., Borras J. and Martin-Hidalgo A.

(2005) Prediction of neuropsychiatric adverse events associated with long-term efavirenz

therapy, using plasma drug level monitoring. Clin. Infect. Dis. 41, 1648-1653.

Gutierrez-Valencia A., Viciana P., Palacios R., Ruiz-Valderas R., Lozano F., Terron A., Rivero

A., Lopez-Cortes L. F. and Sociedad Andaluza de Enfermedades Infecciosas (2009)

Stepped-dose versus full-dose efavirenz for HIV infection and neuropsychiatric adverse

events: a randomized trial. Ann. Intern. Med. 151, 149-156.

Haus U., Spath M. and Farber L. (2004) Spectrum of use and tolerability of 5-HT3 receptor

antagonists. Scand. J. Rheumatol. Suppl. 119, 12-18.

Hawkinson J. E., Drewe J. A., Kimbrough C. L., Chen J. S., Hogenkamp D. J., Lan N. C., Gee

K. W., Shen K. Z., Whittemore E. R. and Woodward R. M. (1996) 3 alpha-Hydroxy-3 beta-

trifluoromethyl-5 alpha-pregnan-20-one (Co 2-1970): a partial agonist at the neuroactive

steroid site of the gamma-aminobutyric acidA receptor. Mol. Pharmacol. 49, 897-906.

Hirst W. D., Abrahamsen B., Blaney F. E., Calver A. R., Aloj L., Price G. W. and Medhurst A.

D. (2003) Differences in the central nervous system distribution and pharmacology of the

199

mouse 5-hydroxytryptamine-6 receptor compared with rat and human receptors investigated

by radioligand binding, site-directed mutagenesis, and molecular modeling. Mol.

Pharmacol. 64, 1295-1308.

Hsu C. Y. and Uludag H. (2012) A simple and rapid nonviral approach to efficiently transfect

primary tissue-derived cells using polyethylenimine. Nat. Protoc. 7, 935-945.

Huang R. Q. and Dillon G. H. (1999) Effect of extracellular pH on GABA-activated current in

rat recombinant receptors and thin hypothalamic slices. J. Neurophysiol. 82, 1233-1243.

Jalali F., Afshari R. and Babaei A. (2014) Smoking crushed hyoscine/scopolamine tablets as

drug abuse. Subst. Use Misuse. 49, 793-797.

Jiang S., Li Y., Zhang C., Zhao Y., Bu G., Xu H. and Zhang Y. W. (2014) M1 muscarinic

acetylcholine receptor in Alzheimer's disease. Neurosci. Bull. 30, 295-307.

Johnson M. P., Loncharich R. J., Baez M. and Nelson D. L. (1994) Species variations in

transmembrane region V of the 5-hydroxytryptamine type 2A receptor alter the structure-

activity relationship of certain ergolines and tryptamines. Mol. Pharmacol. 45, 277-286.

Kenakin T. (2014) What is pharmacological 'affinity'? Relevance to biased agonism and

antagonism. Trends Pharmacol. Sci. 35, 434-441.

Lange H. S., Cannon C. E., Drott J. T., Kuduk S. D. and Uslaner J. M. (2015) The M1

Muscarinic Positive Allosteric Modulator PQCA Improves Performance on Translatable

Tests of Memory and Attention in Rhesus Monkeys. J. Pharmacol. Exp. Ther. 355, 442-450.

Lin Y. G., Chen P. H., Chang F. Y., Wu L. T., Liao K. Y. and Wu T. C. (2011) Delirium due to

scopolamine patch in a 4-year-old boy. J. Formos. Med. Assoc. 110, 208-211.

200

Lochet P., Peyriere H., Lotthe A., Mauboussin J. M., Delmas B. and Reynes J. (2003) Long-term

assessment of neuropsychiatric adverse reactions associated with efavirenz. HIV. Med. 4,

62-66.

Luo J., Busillo J. M. and Benovic J. L. (2008) M3 muscarinic acetylcholine receptor-mediated

signaling is regulated by distinct mechanisms. Mol. Pharmacol. 74, 338-347.

Ma K., Yang Z. H., Yang L. M., Chen H. Z. and Lu Y. (2013) Activation of M1 mAChRs by

lesatropane rescues glutamate neurotoxicity in PC12 cells via PKC-mediated

phosphorylation of ERK1/2. Bosn J. Basic Med. Sci. 13, 146-152.

Marek G. J. and Aghajanian G. K. (1996) LSD and the phenethylamine hallucinogen DOI are

potent partial agonists at 5-HT2A receptors on interneurons in rat piriform cortex. J.

Pharmacol. Exp. Ther. 278, 1373-1382.

Marzolini C., Telenti A., Decosterd L. A., Greub G., Biollaz J. and Buclin T. (2001) Efavirenz

plasma levels can predict treatment failure and central nervous system side effects in HIV-1-

infected patients. AIDS. 15, 71-75.

Maxwell S., Scheftner W. A., Kessler H. A. and Busch K. (1988) Manic syndrome associated

with zidovudine treatment. JAMA. 259, 3406-3407.

McLeod G. X. and Hammer S. M. (1992) Zidovudine: five years later. Ann. Intern. Med. 117,

487-501.

Mills A. M., Antinori A., Clotet B., Fourie J., Herrera G., Hicks C., Madruga J. V., Vanveggel

S., Stevens M., Boven K. and ECHO and THRIVE study groups (2013) Neurological and

psychiatric tolerability of rilpivirine (TMC278) vs. efavirenz in treatment-naive, HIV-1-

infected patients at 48 weeks. HIV. Med. 14, 391-400.

201

Monsma F. J.,Jr, Shen Y., Ward R. P., Hamblin M. W. and Sibley D. R. (1993) Cloning and

expression of a novel serotonin receptor with high affinity for tricyclic psychotropic drugs.

Mol. Pharmacol. 43, 320-327.

Monti J. M. (2011) Serotonin control of sleep-wake behavior. Sleep Med. Rev. 15, 269-281.

Monti J. M. and Jantos H. (2011) Effects of the 5-HT(6) receptor antagonists SB-399885 and

RO-4368554 and of the 5-HT(2A) receptor antagonist EMD 281014 on sleep and

wakefulness in the rat during both phases of the light-dark cycle. Behav. Brain Res. 216,

381-388.

Morairty S. R., Hedley L., Flores J., Martin R. and Kilduff T. S. (2008) Selective 5HT2A and

5HT6 receptor antagonists promote sleep in rats. Sleep. 31, 34-44.

Morlese J. F., Qazi N. A., Gazzard B. G. and Nelson M. R. (2002) Nevirapine-induced

neuropsychiatric complications, a class effect of non-nucleoside reverse transcriptase

inhibitors? AIDS. 16, 1840-1841.

Morris H. (2014) Getting high on HIV medication (Full documentary). 2015.

Mukonzo J. K., Okwera A., Nakasujja N., Luzze H., Sebuwufu D., Ogwal-Okeng J., Waako P.,

Gustafsson L. L. and Aklillu E. (2013) Influence of efavirenz pharmacokinetics and

pharmacogenetics on neuropsychological disorders in Ugandan HIV-positive patients with

or without tuberculosis: a prospective cohort study. BMC Infect. Dis. 13, 261-2334-13-261.

Munoz-Moreno J. A., Fumaz C. R., Ferrer M. J., Gonzalez-Garcia M., Molto J., Negredo E. and

Clotet B. (2009) Neuropsychiatric symptoms associated with efavirenz: prevalence,

correlates, and management. A neurobehavioral review. AIDS. Rev. 11, 103-109.

Nelson M. R., Elion R. A., Cohen C. J., Mills A., Hodder S. L., Segal-Maurer S., Bloch M.,

Garner W., Guyer B., Williams S., Chuck S., Vanveggel S., Deckx H. and Stevens M.

202

(2013) Rilpivirine versus efavirenz in HIV-1-infected subjects receiving

emtricitabine/tenofovir DF: pooled 96-week data from ECHO and THRIVE Studies. HIV.

Clin. Trials. 14, 81-91.

Nunez M., Gonzalez de Requena D., Gallego L., Jimenez-Nacher I., Gonzalez-Lahoz J. and

Soriano V. (2001) Higher efavirenz plasma levels correlate with development of insomnia.

J. Acquir. Immune Defic. Syndr. 28, 399-400.

Petit A. C., Quesseveur G., Gressier F., Colle R., David D. J., Gardier A. M., Ferreri F., Lepine J.

P., Falissard B., Verstuyft C., Guiard B. P. and Corruble E. (2014) Converging translational

evidence for the involvement of the serotonin 2A receptor gene in major depressive

disorder. Prog. Neuropsychopharmacol. Biol. Psychiatry. 54, 76-82.

Ponzoni A., Monti J. M. and Jantos H. (1993) The effects of selective activation of the 5-HT3

receptor with m-chlorophenylbiguanide on sleep and wakefulness in the rat. Eur. J.

Pharmacol. 249, 259-264.

Porter R. H., Benwell K. R., Lamb H., Malcolm C. S., Allen N. H., Revell D. F., Adams D. R.

and Sheardown M. J. (1999) Functional characterization of agonists at recombinant human

5-HT2A, 5-HT2B and 5-HT2C receptors in CHO-K1 cells. Br. J. Pharmacol. 128, 13-20.

Puri V., Wang X., Vardigan J. D., Kuduk S. D. and Uslaner J. M. (2015) The selective positive

allosteric M1 muscarinic receptor modulator PQCA attenuates learning and memory deficits

in the Tg2576 Alzheimer's disease mouse model. Behav. Brain Res. 287, 96-99.

Purohit A., Herrick-Davis K. and Teitler M. (2003) Creation, expression, and characterization of

a constitutively active mutant of the human serotonin 5-HT6 receptor. Synapse. 47, 218-

224.

203

Rammes G., Eisensamer B., Ferrari U., Shapa M., Gimpl G., Gilling K., Parsons C., Riering K.,

Hapfelmeier G., Bondy B., Zieglgansberger W., Holsboer F. and Rupprecht R. (2004)

Antipsychotic drugs antagonize human serotonin type 3 receptor currents in a

noncompetitive manner. Mol. Psychiatry. 9, 846-58, 818.

Real C., Popa D., Seif I., Callebert J., Launay J. M., Adrien J. and Escourrou P. (2007) Sleep

apneas are increased in mice lacking monoamine oxidase A. Sleep. 30, 1295-1302.

Rihs T. A., Begley K., Smith D. E., Sarangapany J., Callaghan A., Kelly M., Post J. J. and Gold

J. (2006) Efavirenz and chronic neuropsychiatric symptoms: a cross-sectional case control

study. HIV. Med. 7, 544-548.

Rojas C., Stathis M., Thomas A. G., Massuda E. B., Alt J., Zhang J., Rubenstein E., Sebastiani

S., Cantoreggi S., Snyder S. H. and Slusher B. (2008) Palonosetron exhibits unique

molecular interactions with the 5-HT3 receptor. Anesth. Analg. 107, 469-478.

Romao P. R., Lemos J. C., Moreira J., de Chaves G., Moretti M., Castro A. A., Andrade V. M.,

Boeck C. R., Quevedo J. and Gavioli E. C. (2011) Anti-HIV drugs nevirapine and efavirenz

affect anxiety-related behavior and cognitive performance in mice. Neurotox Res. 19, 73-80.

Romero G., Sanchez E., Pujol M., Perez P., Codony X., Holenz J., Buschmann H. and Pauwels

P. J. (2006) Efficacy of selective 5-HT6 receptor ligands determined by monitoring 5-HT6

receptor-mediated cAMP signaling pathways. Br. J. Pharmacol. 148, 1133-1143.

Seo S. W., Suh M. K., Chin J. and Na D. L. (2009) Mental confusion associated with

scopolamine patch in elderly with mild cognitive impairment (MCI). Arch. Gerontol.

Geriatr. 49, 204-207.

Shubber Z., Calmy A., Andrieux-Meyer I., Vitoria M., Renaud-Thery F., Shaffer N., Hargreaves

S., Mills E. J. and Ford N. (2013) Adverse events associated with nevirapine and efavirenz-

204

based first-line antiretroviral therapy: a systematic review and meta-analysis. AIDS. 27,

1403-1412.

Simen A. A., Ma J., Svetnik V., Mayleben D., Maynard J., Roth A., Mixson L., Mogg R., Shera

D., George L., Mast T. C., Beals C., Stoch A., Struyk A., Shire N. and Fraser I. (2015)

Efavirenz modulation of sleep spindles and sleep spectral profile. J. Sleep Res. 24, 66-73.

Singleton C. and Marsden C. A. (1981) Circadian variation in the head twitch response produced

by 5-methoxy-N1,N1-dimethyltryptamine and p-chloroamphetamine in the mouse.

Psychopharmacology (Berl). 74, 173-176.

Smith S. R., Prosser W. A., Donahue D. J., Morgan M. E., Anderson C. M., Shanahan W. R. and

APD356-004 Study Group (2009) Lorcaserin (APD356), a selective 5-HT(2C) agonist,

reduces body weight in obese men and women. Obesity (Silver Spring). 17, 494-503.

Sustiva (1998) Sustiva product insert: Highlights of prescribing information. 2015, 21.

Tallman J. F., Smith C. C. and Henneberry R. C. (1977) Induction of functional beta-adrenergic

receptors in HeLa cells. Proc. Natl. Acad. Sci. U. S. A. 74, 873-877.

Tanaka R., Hanabusa H., Kinai E., Hasegawa N., Negishi M. and Kato S. (2008) Intracellular

efavirenz levels in peripheral blood mononuclear cells from human immunodeficiency

virus-infected individuals. Antimicrob. Agents Chemother. 52, 782-785.

Tashima K. T., Caliendo A. M., Ahmad M., Gormley J. M., Fiske W. D., Brennan J. M. and

Flanigan T. P. (1999) Cerebrospinal fluid human immunodeficiency virus type 1 (HIV-1)

suppression and efavirenz drug concentrations in HIV-1-infected patients receiving

combination therapy. J. Infect. Dis. 180, 862-864.

Tozzi V., Balestra P., Salvatori M. F., Vlassi C., Liuzzi G., Giancola M. L., Giulianelli M.,

Narciso P. and Antinori A. (2009) Changes in cognition during antiretroviral therapy:

205

comparison of 2 different ranking systems to measure antiretroviral drug efficacy on HIV-

associated neurocognitive disorders. J. Acquir. Immune Defic. Syndr. 52, 56-63.

Tralongo P., Dimari A., Conti G., Aiello R. and Mauceri G. (2004) Central nervous system side-

effects of 5-HT3-receptor antagonists in elderly cancer patients treated with chemotherapy.

Ann. Oncol. 15, 987-988.

Vallersnes O. M., Lund C., Duns A. K., Netland H. and Rasmussen I. A. (2009) Epidemic of

poisoning caused by scopolamine disguised as Rohypnol tablets. Clin. Toxicol. (Phila). 47,

889-893.

Vickers S. P., Easton N., Malcolm C. S., Allen N. H., Porter R. H., Bickerdike M. J. and Kennett

G. A. (2001) Modulation of 5-HT(2A) receptor-mediated head-twitch behaviour in the rat

by 5-HT(2C) receptor agonists. Pharmacol. Biochem. Behav. 69, 643-652.

Vollenweider F. X., Leenders K. L., Scharfetter C., Maguire P., Stadelmann O. and Angst J.

(1997) Positron emission tomography and fluorodeoxyglucose studies of metabolic

hyperfrontality and psychopathology in the psilocybin model of psychosis.

Neuropsychopharmacology. 16, 357-372.

Vollenweider F. X., Vollenweider-Scherpenhuyzen M. F., Babler A., Vogel H. and Hell D.

(1998) Psilocybin induces schizophrenia-like psychosis in humans via a serotonin-2 agonist

action. Neuroreport. 9, 3897-3902.

Waters L., Fisher M., Winston A., Higgs C., Hadley W., Garvey L., Mandalia S., Perry N.,

Nicola M. and Nelson M. (2011) A phase IV, double-blind, multicentre, randomized,

placebo-controlled, pilot study to assess the feasibility of switching individuals receiving

efavirenz with continuing central nervous system adverse events to etravirine. AIDS. 25, 65-

71.

206

WHO (2013) Consolidated guidelines on the use of antiretroviral drug for the treatment and

preventing HIV infection. Recommendations for a public health approach. 2015, 272.

Willins D. L. and Meltzer H. Y. (1997) Direct injection of 5-HT2A receptor agonists into the

medial prefrontal cortex produces a head-twitch response in rats. J. Pharmacol. Exp. Ther.

282, 699-706.

Wise M. E., Mistry K. and Reid S. (2002) Drug points: Neuropsychiatric complications of

nevirapine treatment. BMJ. 324, 879.

Wisor J. P., Wurts S. W., Hall F. S., Lesch K. P., Murphy D. L., Uhl G. R. and Edgar D. M.

(2003) Altered rapid eye movement sleep timing in serotonin transporter knockout mice.

Neuroreport. 14, 233-238.

Wright J. M., Sachdev P. S., Perkins R. J. and Rodriguez P. (1989) Zidovudine-related mania.

Med. J. Aust. 150, 339-341.

Yilmaz A., Watson V., Dickinson L. and Back D. (2012) Efavirenz pharmacokinetics in

cerebrospinal fluid and plasma over a 24-hour dosing interval. Antimicrob. Agents

Chemother. 56, 4583-4585.

Young S. D., Britcher S. F., Tran L. O., Payne L. S., Lumma W. C., Lyle T. A., Huff J. R.,

Anderson P. S., Olsen D. B. and Carroll S. S. (1995) L-743, 726 (DMP-266): a novel,

highly potent nonnucleoside inhibitor of the human immunodeficiency virus type 1 reverse

transcriptase. Antimicrob. Agents Chemother. 39, 2602-2605.

7. Acknowledgements: We thank the authors’ colleagues for providing the following as generous gifts: Dr. David

Julius (USCF, San Francisco, CA) for cDNA for the mouse 5-HT3A receptor (GenBank

accession no. S41757), Dr. David R. Sibley (NIH, Bethesda, MD) for the stable line

expressing cloned rat 5-HT6 receptor, and Dr. Juergen Wess (NIH, Bethesda, MD) for the

207

stable line expressing the cloned human M1 receptor. We also thank Dr. Glenn Dillon and

Cathy Bell-Horner for contributions to some of the preliminary aspects of this work. This

work was supported in part by the National Institutes of Health [Grant R01-MH063162

(JAS); Grant R41AG043243 (JAS)], ADDF grant 20140803 (JAS), UNTHSC Intramural

Grant RI-6015 (JAS) and institutional funds 67673 (JAS).

Author Contribution:

Dhwanil A. Dalwadi – designed, performed experiments, and helped with data interpretation

and writing of the manuscript

Seongcheol Kim – performed experiments for early phases of this work

Shahnawaz M. Amdani – performed experiments for early phases of this work

Zhenglan Chen – performed experiments

Ren-Qi Huang – designed experiments, interpreted data, and helped write the manuscript

John A. Schetz – conceived of the project, designed experiments, interpreted data, and wrote

the manuscript

Funding Sources:

National Institutes of Health [Grant R01-MH063162 (JAS); Grant R41AG043243 (JAS)]

ADDF grant 20140803 (JAS)

UNTHSC Intramural Grant RI-6015 (JAS) and institutional funds 67673 (JAS).

8. Conflict of Interest:

Authors declare no competing financial interest.

208

CHAPTER 5

CONCLUSION

In this dissertation the mechanism of actions of several novel small molecules at different receptor targets were evaluated and numerous new findings are reported here. These finding will be instrumental in the future development of novel compounds with the desired modes of action.

In the following sections the key findings of this dissertation are summarized and their implications are discussed.

Exploration of the structure-activity space of cloned α-like octopamine receptors from a marine barnacle and a terrestrial fly

The overall goal of this project was to develop a high-throughput model that can be used to evaluate the receptor pharmacology of OctR ligands to aid in the identification of compounds with antifouling / arthropod deterrent properties. The evidence in the literature suggests that

OctR agonists have arthropod deterrent / antifouling properties (Casida and Durkin 2013;

Dahlstrom and Elwing 2006; Dahlström et al. 2000; Dahms et al. 2004; Gross et al. 2015;

Hashemzadeh et al. 1985; Hollingworth and Murdock 1980; Lind et al. 2010; Matsumura and

Beeman 1976; Nathanson and Hunnicutt 1979; Roeder 1995; Swale et al. 2014), but all known

OctR agonists have α-AR off-target liability (Boyes and Moser 1988; Costa et al. 1988; Costa et al. 1991; Hsu and Kakuk 1984; Hsu et al. 1988). In an effort to identify OctR selective compounds, a high throughput in vitro functional assay was developed using the barnacle OctR

(BiOctR) as a proxy. The BiOctR was selected because preliminary data generated previously in

209

Dr. Schetz’s lab suggested that some compounds which induce hyperactivity in barnacle cyprids and also had mosquito deterrent properties (data not shown). Previous studies by other groups indicated that barnacle hyperactivity correlates with prevention of cyprid settlement

(Dahlstrom and Elwing 2006; Dahlström et al. 2000; Dahms et al. 2004; Lind et al. 2010). Since there is a strong correlation between BiOctR agonists and antifouling, the cloned BiOctR receptor was used to develop a stable cell line model that would facilitate high-throughput evaluation of OctR ligands.

The assay was thoroughly validated using reference compounds (clonidine, naphazoline, tizanidine, dexmedetomidine) and to provide further support for the hypothesis that OctR agonists function as deterrents, it was demonstrated that the reference compounds also induced hyperactivity in cyprids. In order to evaluate selectivity of the compounds, comparing affinities at receptors of interest would be the ideal and most convenient approach. The potencies cannot be directly compared to the affinity because potencies may be influenced by receptor reserve, which might lead to the selectivity appearing larger than it actually is. In this study [3H]- rauwolscine was used as the radioligand for performing competition binding experiments. It was determined that the BiOctR exhibits sodium sensitivity for rauwolscine and possibly other antagonists, and for agonists: sodium increased the receptor’s affinity for rauwolscine and decreased it for the agonists. This is the first report to explore the sodium sensitivity of the

BiOctR, and from an experimental perspective this is important information. As discussed in chapter 2, lowering concentration of salt from high to low (240 mM to 15 nM) had no effect on the affinity of the agonists, and when the agonist affinities determined by radioligand binding were compared to the Ka (agonist affinity calculated using the functional operational model of agonism), it was concluded that receptor reverse has a small effect on the potency and that

210 sodium is decreasing the affinity of the agonist. If an investigator performs a binding experiment without knowing that the receptor exhibits sodium sensitivity, then the affinities obtained from those experiments are likely to be incorrect. For example, Lind et al used [3H]-Rs79948, which is structurally similar to rauwolscine (see Table 1 in Chapter 2), as the radioligand to estimate Bmax and KD for the expression of BiOctR in their transfected CHO cells (Lind et al. 2010). They used salt free binding conditions and obtained a KD value of 7.3 nM. However, in this study the Ki for

Rs-79948 was measured to be 0.59 ± 0.08 nM in the presence of salt. Since rauwolscine interacts with the BiOctR at an allosteric site, it is possible that Rs-79948 and rauwolscine may not be competing for the same site, hence the difference in affinities between the two reports. However, the pseudo Hill slope for Rs-79948 was not statistically different from unity, suggesting that rauwolscine and Rs-79948 are competing for the same site. The pseudo Hill slope information should be interpreted with caution because if the two compounds are binding to different sites, but affinity states are not sufficiently different to be distinguishable using the methods utilized, then they would appear to be competitive. From a drug discovery perspective, obtaining rightward shifted, low affinities would likely result in the compound being excluded from the study because they would not meet the selection criteria (i.e. low affinity of agonists at the

BiOctR).

Though significant progress was made in understanding the ligand-BiOctR interactions, a suitable radioligand was not identified that would allow for obtaining accurate affinities for the

BiOctR agonists. When performing radioligand binding experiments, antagonist radioligands are the preferred choice because they are often sodium-insensitive (Schetz 2005). However, since

BiOctR exhibits sodium sensitivity for both agonists and antagonists, and binds to an allosteric site, the use of rauwolscine and possibly other antagonist radioligand is clearly not appropriate.

211

For the BiOctR, a radioligand that binds to the orthosteric site may be an appropriate choice for obtaining accurate affinities. Based on the Ka of dexmedetomidine (0.3 ± 0.1 nM), it seems like the ideal candidate for a radioligand. However, it is not known if it interacts with the receptor at the orthosteric site or an allosteric site. If it interacts with the allosteric site, it would reduce its attractiveness as a radioligand, but this issue can be resolved using Schild analysis once an antagonist that does not violate the Schild assumptions is identified.

By testing several different structural classes of ligands like imidazoles, imidazolines, substituted benzamides, catecholamines, phenothiazines, and an indolamine, as well as assessing stereoselectivity, valuable insight was gained into the structure-activity space of OctR ligands and their mechanism of action at the BiOctR. This information will be useful in designing new compounds with the desired functional and target selectivity. The assay developed in this study will facilitate high-throughput evaluation of compounds, and in the identification of potent OctR agonists with potential arthropod deterrent property.

Novel selective Sigma-1 receptor ligands facilitate BDNF release from a neuronal cell line

S1R agonists have neuroprotective effects in experimental models of neurodegeneration

(Antonini et al. 2011; Francardo et al. 2014; Lahmy et al. 2013; Mancuso et al. 2012; Marrazzo et al. 2005; Ono et al. 2014; Peviani et al. 2014; Takahashi et al. 1995; Takahashi et al. 1996;

Takahashi et al. 1997), and as a result the S1R is considered a promising target for neurodegenerative disorders like AD, Parkinson’s disease (PD) and amyotrophic lateral sclerosis

(ALS). In this chapter, SARs for novel S1R ligands were evaluated and identified S1R selective compounds that were functionally selective at activating the BDNF secretion pathway. The majority of the S1R ligands have been characterized as agonists or antagonists based on their in vivo functional profile, and from a drug discovery perspective, it is not cost-effective to test

212 every compound in vivo. Based on our survey of the literature, three different in vitro measures of S1R activation were identified, which seemed promising for functional characterization of

S1R ligands. These measures included S1R mediated secretion of BDNF (Fujimoto et al. 2012;

Malik et al. 2015), potentiation if IP3 mediated mobilization of ER to cytoplasmic calcium

(Hayashi et al. 2000; Hayashi and Su 2007a; Hong et al. 2004; Wu and Bowen 2008), and potentiation of NGF induced neurite sprouting (Ishima et al. 2008; Ishima and Hashimoto 2012;

Ishima et al. 2014; Nishimura et al. 2008; Rossi et al. 2011; Takebayashi et al. 2002). The rationale for evaluating three different functional measures was that at least one approach would be robust and reproducible. Based on evaluations of reference S1R agonists and antagonists, out of the three measures tested, only BDNF secretion appears to be a robust measure for evaluating

S1R ligands. Specifically, this method can be used to identify S1R agonists that activate the

BDNF secretion pathway. Several novel S1R ligands were characterized as S1R agonists based on their BDNF secretion profile. From a therapeutic perspective compounds that increase BDNF secretion have the potential to be neuroprotective since BDNF has been reported to protect against inflammatory insults, as well as having neurorestoration property like neurogenesis and reversing neuronal atrophy (Rothman and Mattson 2013).

Using the methodologies that were employed, S1R-mediated calcium mobilization and neurite sprouting were not reproducible measures of characterizing S1R ligands as previously reported (Hayashi et al. 2000; Hayashi and Su 2007b; Hong et al. 2004; Ishima et al. 2008;

Ishima and Hashimoto 2012; Ishima et al. 2014; Nishimura et al. 2008; Rossi et al. 2011;

Takebayashi et al. 2002; Wu and Bowen 2008). When S1R reference agonists and antagonists were assessed for potentiation or suppression of IP3-mediated calcium mobilization, respectively, no potentiation or suppression response was observed; the published results were not reproduced

213

2+ in this report. After thoroughly evaluating the literature on S1R-IP3-Ca interactions, it was concluded that the role of S1R in calcium mobilization is not as simple as stabilizing the IP3R and enhancing calcium mobilization (see Chapter 3 discussion for more detail). Hence, measuring changes in IP3 mediated calcium mobilization is not a generally useful approach for assigning functional roles to S1R ligands. Another functional measure that was evaluated was

S1R mediated potentiation of NGF-induced neurite sprouting. In this study a modified PC12 cell line model (PC-6-15, PC12 cell line overexpressing the trkA receptor) was utilized to reduce the duration of the assay by 3 to 4 days (from a 4 to 5 day assay to a 1 day assay). Though NGF could induce neurite sprouting in a concentration dependent manner, no potentiation effect was detected for any of the S1R reference agonists. Since the cellular model used here was different from the one that has been used to study this effect, it was concluded that the PC-6-15 is not an appropriate model for studying the effects of S1R ligands on neurite sprouting (see Chapter 2 discussion). If the PC12 cell line model was used, and was proven to be reproducible, it would not be a time effective assay from a drug discovery perspective because it would take 4-5 days just to complete the drug treatment, and then taking images and manually tracing neurites would take days to process. Since this project is in the early phase of the drug discovery and involves evaluating of large number of compounds, the PC12 neurite sprouting assay would not be cost or time effective. However, once the project has progressed past lead optimization, it might be feasible to test a small set of compounds for neurite sprouting activity using the regular PC-12 cells and the longer assay format.

Overall, the studies done here contributed to the discovery of four novel S1R selective lead compounds that are able to facilitate S1R-mediated BDNF secretion. One of the compounds, EPGN644 has moved on to animal testing for evaluation of its therapeutic value for

214 treating AD. These compounds are not just restricted to targeting AD. Other neurodegenerative diseases like Parkinson’s disease and ALS share a common inflammatory mechanism, and since

S1R ligands have been shown be protective against neuroinflammatory insults, the compounds studied here may also have therapeutic value for treating these diseases as well (Nguyen et al.

2015). In addition to neurodegenerative disorders, these compounds may also be useful for protection against stroke since the S1R agonist 4-PPBP was able to limit the damage induced by cerebral ischemic stroke in vivo (Goyagi et al. 2001; Goyagi et al. 2003; Takahashi et al. 1995;

Takahashi et al. 1996). Lastly, they may also be used as cough suppressants since other S1R ligands like dextromethorphan, carbetapentane, and butamirate are active ingredients of known antitussive medications (Brown et al. 2004; Klein and Musacchio 1989; Taylor et al. 2016).

Molecular mechanisms of serotonergic action of the HIV-1 antiretroviral efavirenz

Efavirenz is a highly efficacious HIV-1 antiretroviral mediation and is one of the preferred components of HAART (WHO 2013). However, since the approval of the drug in

1998, it was known that efavirenz is associated with NPAEs (Fumaz et al. 2005; Gutierrez et al.

2005; Lochet et al. 2003; Munoz-Moreno et al. 2009; Sustiva 1998). In this report new serotonergic and muscarinic receptor targets of efavirenz were identified, and its mechanism at those targets was evaluated. Efavirenz was shown to function as an antagonist of the M1, M3,5-

HT2A and 5-HT2C receptors, an inverse agonist of the 5-HT6 receptor, and a pore blocker of the

5-HT3 receptor. Schild analysis revealed that efavirenz interacts with the 5-HT2A receptor at the same site as known hallucinogens LSD and DOI. In vivo studies demonstrated that efavirenz induces a weak head-twitch response in rodents which is suggestive of hallucinogenic activity

(Gatch et al. 2013), and there are also reports of hallucinations in patients on efavirenz containing regimen (Gutierrez-Valencia et al. 2009; Lochet et al. 2003; Mukonzo et al. 2013;

215

Sustiva 1998). However, unlike LSD and DOI, which are agonists of Gq-signaling at the 5-HT2A receptor, efavirenz is the first example of a hallucinogenic drug that functions as an antagonist of

Gq-signaling at the 5-HT2A receptor. This is an important finding for the hallucinogen field because all known hallucinogen with 5-HT2A activity are agonists of Gq-signaling (Berg et al.

1998; Egan et al. 1998; Marek and Aghajanian 1996; McLean et al. 2006; Monte et al. 1997;

Sard et al. 2005; Urban et al. 2007). Since efavirenz is clearly capable of causing hallucinations in some humans taking the drug as an HIV medication, and in vivo, the head twitch response was absent in the 5-HT2A-KO mice, this suggests that 5-HT2A receptor is involved in inducing efavirenz mediated hallucinations (Gatch et al. 2013). Further, it was clearly demonstrated in this report that efavirenz is an antagonist of Gq-signaling, this suggests that Gq-signaling is not the sole mediator of hallucinogenic activity via the 5-HT2A receptor (See Chapter 4 discussion). This does not mean that the 5-HT2A receptor is the sole mediator of efavirenz induced hallucinations.

Muscarinic receptor antagonists like scopolamine and atropine are also able to induce hallucination (Arthurs and Davies 1980; Baker and Silver 1984; Cotroneo 2013; Fisher 1991;

Jalali et al. 2014; Lin et al. 2011; Seo et al. 2009; Vallersnes et al. 2009), and since efavirenz was shown to interact with the M1 and M3 receptors, the muscarinic receptor system may also be responsible for efavirenz’s hallucinogenic effect.

Identification of novel targets of efavirenz will be instrumental in understanding the mechanism of action of NPAEs. Once we have a better understanding of which receptor targets are responsible for which NPAEs, medications could be prescribed to counter the side effects, and if none exist, then new drugs could be developed targeting those receptor systems. Further, the knowledge of these off-targets could also be used to develop early off-target screening for novel antiretroviral. If the new compounds have the off-target receptor profile similar to

216 efavirenz, then there is a probability that the drug might induce NPAEs. Since efavirenz is such an efficacious drug, another possible future direction for the project could be to engineer out the off-target liability, which is now possible because several receptor off-targets of efavirenz have been identified in this study. Aside from the Gatch et al study, this is only report that has endeavored to evaluate CNS receptor targets of efavirenz. In this study, only two members of the muscarinic receptor family were evaluated, but other members of the muscarinic receptors may also be involved and should be explored further, which might help to explain the mechanism of additional NPAEs.

Summary statement

In this study three different systems were evaluated to achieve a detailed mechanistic understanding and SAR of small molecules by practically applying core pharmacology principles. This allowed for the identification of substructural features that have the desired functional and target selectivity, as well as elucidating the off-target liability of an approved drug. These new findings will be useful for designing novel small molecules with the desired functional effects, and minimal of-target liability. Further, during course of this study, new assays were also developed that will facilitate the discovery and development process.

217

References

Antonini V., Marrazzo A., Kleiner G., Coradazzi M., Ronsisvalle S., Prezzavento O., Ronsisvalle

G. and Leanza G. (2011) Anti-amnesic and neuroprotective actions of the sigma-1 receptor

agonist (-)-MR22 in rats with selective cholinergic lesion and amyloid infusion. J.

Alzheimers Dis. 24, 569-586.

Arthurs G. J. and Davies R. (1980) Atropine--a safe drug. Anaesthesia. 35, 1077-1079.

Baker J. H. and Silver J. R. (1984) Atropine toxicity in acute cervical spinal injury. Paraplegia.

22, 379-382.

Berg K. A., Maayani S., Goldfarb J., Scaramellini C., Leff P. and Clarke W. P. (1998) Effector

pathway-dependent relative efficacy at serotonin type 2A and 2C receptors: evidence for

agonist-directed trafficking of receptor stimulus. Mol. Pharmacol. 54, 94-104.

Boyes W. K. and Moser V. C. (1988) An alpha 2-adrenergic mode of action of chlordimeform on

rat visual function. Toxicol. Appl. Pharmacol. 92, 402-418.

Brown C., Fezoui M., Selig W. M., Schwartz C. E. and Ellis J. L. (2004) Antitussive activity of

sigma-1 receptor agonists in the guinea-pig. Br. J. Pharmacol. 141, 233-240.

Casida J. E. and Durkin K. A. (2013) Neuroactive insecticides: targets, selectivity, resistance,

and secondary effects. Annu. Rev. Entomol. 58, 99-117.

Costa L. G., Gastel J. and Murphy S. D. (1991) The formamidine pesticides chlordimeform and

amitraz decrease hepatic glutathione in mice through an interaction with alpha 2-

adrenoceptors. J. Toxicol. Environ. Health. 33, 349-358.

Costa L. G., Olibet G. and Murphy S. D. (1988) Alpha 2-adrenoceptors as a target for

formamidine pesticides: in vitro and in vivo studies in mice. Toxicol. Appl. Pharmacol. 93,

319-328.

218

Cotroneo C. (2013) Devil's Breath: Scopolamine, AKA Burundanga, Hailed As 'World's Scariest

Drug'. 2015.

Dahlstrom M. and Elwing H. (2006) Adrenoceptor and other pharmacoactive compounds as

putative antifoulants. Prog. Mol. Subcell. Biol. 42, 171-202.

Dahlström M., Mårtenssonb L. G., Jonssonc P. R., Arnebrantd T. and Elwinge H. (2000) Surface

active adrenoceptor compounds prevent the settlement of cyprid larvae of Balanus

improvisus. Volume 16, 191-203.

Dahms H. U., Jin T. and Qian P. Y. (2004) Adrenoceptor compounds prevent the settlement of

marine invertebrate larvae: Balanus amphitrite (Cirripedia), Bugula neritina (Bryozoa) and

Hydroides elegans (Polychaeta). Biofouling. 20, 313-321.

Egan C. T., Herrick-Davis K., Miller K., Glennon R. A. and Teitler M. (1998) Agonist activity of

LSD and lisuride at cloned 5HT2A and 5HT2C receptors. Psychopharmacology (Berl). 136,

409-414.

Fisher C. M. (1991) Visual hallucinations on eye closure associated with atropine toxicity. A

neurological analysis and comparison with other visual hallucinations. Can. J. Neurol. Sci.

18, 18-27.

Francardo V., Bez F., Wieloch T., Nissbrandt H., Ruscher K. and Cenci M. A. (2014)

Pharmacological stimulation of sigma-1 receptors has neurorestorative effects in

experimental parkinsonism. Brain. 137, 1998-2014.

Fujimoto M., Hayashi T., Urfer R., Mita S. and Su T. P. (2012) Sigma-1 receptor chaperones

regulate the secretion of brain-derived neurotrophic factor. Synapse. 66, 630-639.

Fumaz C. R., Munoz-Moreno J. A., Molto J., Negredo E., Ferrer M. J., Sirera G., Perez-Alvarez

N., Gomez G., Burger D. and Clotet B. (2005) Long-term neuropsychiatric disorders on

219

efavirenz-based approaches: quality of life, psychologic issues, and adherence. J. Acquir.

Immune Defic. Syndr. 38, 560-565.

Gatch M. B., Kozlenkov A., Huang R. Q., Yang W., Nguyen J. D., Gonzalez-Maeso J., Rice K.

C., France C. P., Dillon G. H., Forster M. J. and Schetz J. A. (2013) The HIV antiretroviral

drug efavirenz has LSD-like properties. Neuropsychopharmacology. 38, 2373-2384.

Goyagi T., Goto S., Bhardwaj A., Dawson V. L., Hurn P. D. and Kirsch J. R. (2001)

Neuroprotective effect of sigma(1)-receptor ligand 4-phenyl-1-(4-phenylbutyl) piperidine

(PPBP) is linked to reduced neuronal nitric oxide production. Stroke. 32, 1613-1620.

Goyagi T., Toung T. J., Kirsch J. R., Traystman R. J., Koehler R. C., Hurn P. D. and Bhardwaj

A. (2003) Neuroprotective kappa-opioid receptor agonist BRL 52537 attenuates ischemia-

evoked nitric oxide production in vivo in rats. Stroke. 34, 1533-1538.