Jaclyn N. Highland Jaclyn.Highland@Gmail.Com Education Phd in Toxicology, Graduate Program in Life Sciences, December 2020

Home , Hydroxynorketamine

Characterization of Ketamine’s (2,6)-Hydroxynorketamine Metabolites: Pharmacokinetic and Behavioral Considerations for Antidepressant Applications

Item Type dissertation

Authors Highland, Jaclyn

Publication Date 2020

Abstract Despite numerous available treatments for depression, most are not only slow to take effect but also ineffective in many patients, highlighting the need for novel and effective antidepressant treatments. While (R,S)-ketamine (ketamine) has gained att...

Keywords antidepressants; hydroxynorketamine; Antidepressive Agents; Behavior; Depression; Drug Development; Pharmacokinetics

Download date 28/09/2021 20:31:32

Link to Item http://hdl.handle.net/10713/14503 Jaclyn N. Highland [email protected]

Education PhD in Toxicology, Graduate Program in Life Sciences, December 2020 University of Maryland Baltimore, Baltimore, MD

BS in Chemistry, magna cum laude, May 2015 Slippery Rock University of Pennsylvania, Slippery Rock, PA

Research Experience Graduate Dissertation Research, University of Maryland Baltimore May 2016 – August 2020 PI: Todd Gould, M.D., Associate Professor, Department of Psychiatry − Assessed the pharmacokinetic profiles of 12 hydroxynorketamine (HNK) metabolites of ketamine and evaluated the antidepressant-relevant behavioral effectiveness of the (2,6)-HNKs. − Evaluated the role of gonadal hormones in mediating the sex-dependent differences in the metabolism of ketamine and (2R,6R)-HNK. − Evaluated the oral bioavailability of (2R,6R)-hydroxynorketamine and six novel prodrug candidates, and assessed the oral antidepressant-like behavioral efficacy of (2R,6R)-hydroxynorketamine in mice (Highland et al, Journal of Psychopharmacology 2018). − Investigated the role of group II metabotropic glutamate (mGlu2/3) receptor activity in modulating stress resilience versus susceptibility and evaluating an mGlu2/3 antagonist for its ability to protect against the onset and recurrence of maladaptive stress-induced behaviors in mice (Highland et al, Neuropharmacology 2019).

Laboratory Rotations, University of Maryland Baltimore Jan. 2016 – April 2016 PI: Jessica Mong Ph.D., Associate Professor, Department of Pharmacology − Evaluated the distribution of estrogen receptors in brain regions active during sleep in rats.

Aug. 2015 – Dec. 2015 PI: Marilyn Huestis Ph.D., Chief, Chemistry and Drug Metabolism Section, National Institute on Drug Abuse Intramural Research Program − Contributed to the validation of an updated LC/MS/MS method for the quantification of cannabinoids in exhaled breath samples.

Undergraduate Research Projects, Slippery Rock University Jan. 2014 – May 2015 PI: Paul Birckbichler Ph.D., Associate Professor, Chemistry Department − Investigated the induction of apoptosis by synthetic compounds derived from vitamin A in leukemia cell lines. − Maintained suspension cell cultures using aseptic techniques and assayed cell survival and apoptosis using microscopy and flow cytometry.

Aug. 2014 – May 2015 PI: Donald Zapien, Ph.D., Professor, Chemistry Department − Developed an undergraduate-level laboratory experiment in forensic chemistry. − Adapted a protocol for the detection of acetaminophen in samples of simulated urine samples using ELISA screening and GC/MS confirmation.

Publications Highland, J.N., Zanos, P., Georgiou, P., and Gould, T.D. “Group II metabotropic glutamate receptor blockade promotes stress resilience in mice.” Neuropharmacology 44(10):1788-1796, 2019.

Highland, J.N., Morris, P.J., Zanos, P., Lovett, J., Gosh, S., Wang, A.Q., Zarate, C.A. Jr., Thomas, C.J., Moaddel, R., and Gould, T.D. “Mouse, rat, and dog bioavailability and mouse oral antidepressant efficacy of (2R,6R)-hydroxynorketamine.” Journal of Psychopharmacology 33(1):12-24, 2019.

Zanos, P., Highland, J.N., Liu, X., Troppoli, T.A., Georgiou, P., Lovett, J., Morris, P.J., Stewart, B.S., Thomas, C.J., Thompson, S.M., Moaddel, R., and Gould, T.D. “(R)- ketamine exerts antidepressant actions partly via conversion to (2R,6R)- hydroxynorketamine, while causing adverse effects at sub-anesthetic doses.” British Journal of Pharmacology 176(14):2573-2592, 2019.

Zanos, P., Highland, J.N., Georgiou, P., Stewart, B.W., Jenne, C.E., Kang, X.J., Zarate, C.A., and Gould, T.D. “(2R,6R)-hydroxynorketamine exerts mGlu2 receptor- dependent antidepressant actions.” PNAS 116(13):6441-6450, 2019.

Zanos, P., Moaddel, R., Morris P.J., Riggs, L.M., Highland, J.N., Georgiou, P., Pereira, E.F.R., Albuquerque, E.X., Thomas, C.J., Zarate, C.A. Jr., and Gould, T.D. “Ketamine and ketamine metabolite pharmacology: Insights into therapeutic mechanisms.” Pharmacological Reviews 70(3):621-660, 2018.

Zanos, P., Nelson, M.E., Highland, J.N., Krimmel, S.R., Georgiou, P., Gould, T.D., and Thompson, S.M. “A negative allosteric modulator for α5 subunit-containing GABA receptors exerts a rapid and persistent antidepressant-like action without the side effects of the NMDA receptor antagonist ketamine in mice.” eNeuro 4(1): ENEURO.0285-16.2017, 2017.

Oral presentations “Characterization of ketamine’s (2,6)-hydroxynorketamine metabolites: pharmacokinetic and behavioral considerations for antidepressant applications.” Data blitz pipeline presentation, ASENT Annual Meeting, Bethesda, MD, March 4, 2020.

“Characterization of ketamine’s hydroxynorketamine metabolites and their antidepressant-relevant synaptic properties.” Training Program in Integrative Membrane Biology Retreat, Baltimore, MD, May 3, 2019.

Poster presentations Highland, J.N., Morris, P.J., Zanos, P., Riggs, L.M., Lovett, J., Gosh, S., Konrath, K., Wang, A.Q., Zarate, C.A., Thomas, C.J., Moaddel, R., and Gould, T.D. “Characterization of ketamine’s (2,6)-hydroxynorketamine metabolites: pharmacokinetic and behavioral considerations for antidepressant applications.” ASENT Annual Meeting, Bethesda, MD, March 4, 2020.

Highland, J.N., Morris, P.J., Zanos, P., Lovett, J., Gosh, S., Wang, A., Zarate, C.J., Moaddel, R., and Gould, T.D. “Oral bioavailability and antidepressant efficacy of (2R,6R)-hydroxynorketamine.” Annual Department of Psychiatry Research Day, University of Maryland School of Medicine, May 30, 2019.

Highland J.N., Morris, P.J., Riggs, L.M., Zanos, P., Konrath, K., Moaddel, R., Thomas, C.J., Wang, A.Q., and Gould T.D. “Hydroxynorketamine metabolites in the treatment of depression: pharmacokinetics and behavioral activity in mice.” Annual Department of Psychiatry Research Day, University of Maryland School of Medicine, June 7, 2018.

Highland, J.N., Zanos, P., Georgiou, P., and Gould, T.D. “Group II metabotropic glutamate receptor blockade promotes stress resilience.” Society for Biological Psychiatry Annual Meeting, New York, NY, 2018.

Highland, J.N., Zanos, P., Georgiou, P., and Gould, T.D. “Group II metabotropic glutamate receptor blockade promotes stress resilience.” Society for Neuroscience Annual Meeting, Washington, DC, 2017.

Highland, J.N., Zanos, P., Georgiou, P., and Gould, T.D. “Group II metabotropic glutamate receptor blockade promotes stress resilience.” Program in Neuroscience Annual Retreat, University of Maryland School of Medicine, June 5, 2017.

Honors and awards Outstanding Trainee Poster Award, Annual Psychiatry Research Day, June 2018

T32 GM008181 Training Program in Integrative Membrane Biology NIH/NIGMS, Sep. 2017-Aug. 2019 Abstract

Characterization of Ketamine’s (2,6)-Hydroxynorketamine Metabolites: Pharmacokinetic and Behavioral Considerations for Antidepressant Applications.

Jaclyn N. Highland, Doctor of Philosophy, 2020

Dissertation Directed by:

Todd D. Gould, MD Professor, Psychiatry, Pharmacology, and Anatomy & Neurobiology University of Maryland School of Medicine

Despite numerous available treatments for depression, most are not only slow to take effect but also ineffective in many patients, highlighting the need for novel and effective antidepressant treatments. While (R,S)-ketamine (ketamine) has gained attention for its rapid-acting antidepressant effects in previously treatment-resistant patients, its widespread antidepressant use is limited by adverse effects and abuse potential. The ketamine metabolite (2R,6R;2S,6S)-hydroxynorketamine (HNK), and most potently the

(2R,6R)-HNK stereoisomer, shares the antidepressant-like behavioral effectiveness, but lacks the adverse effect burden and abuse potential, of ketamine in rodents, suggesting that

(2R,6R)-HNK, and possibly other HNKs (12 have been identified, but only (2R,6R)- and

(2S,6S)-HNK were previously studied), may be favorable antidepressant drug candidates.

However, several important considerations for the development of HNKs as antidepressants were not previously evaluated. Such considerations were addressed in the present study. First, the pharmacokinetic profiles of the 12 HNKs were characterized, revealing robust differences in plasma and brain concentrations among the various HNKs, despite similar brain-to-plasma ratios and rapid elimination profiles. Second, it was demonstrated that (2R,6R)-HNK has favorable oral bioavailability (45-52%) and orally administered (2R,6R)-HNK exerts antidepressant-relevant behavioral effects, but not overt adverse effects, in mice. Third, the sex-dependent differences in the metabolism of ketamine and (2R,6R)-HNK in mice were characterized. It was demonstrated that, following ketamine administration, female mice have lower levels of ketamine and higher levels of HNK than male mice. In addition, following direct dosing, female mice have higher levels of (2R,6R)-HNK than males. Male gonadal hormones at least partly mediate these differences. Further, the relative effectiveness of the four (2,6)-HNKs in the mouse forced swim test were compared, revealing that both (2R,6S)- and (2S,6R)-HNK exert effects at lower doses compared to either (2R,6R)- or (2S,6S)-HNK, and establishing a relative rank-order of effectiveness of (2R,6R)- > (2S,6R)- > (2R,6R)- > (2S,6S)-HNK.

Finally, it was identified that the novel compound, (5R)-methyl-(2R,6R)-HNK recapitulates the antidepressant dose-response relationship of (2R,6S)-HNK, with which it shares a similar three-dimensional structure, suggesting a critical role of three-dimensional structure in mediating antidepressant-relevant effectiveness. Altogether, these studies represent several important insights which may support the development of (2,6)-HNKs as antidepressant treatments.

Characterization of Ketamine’s (2,6)-Hydroxynorketamine Metabolites: Pharmacokinetic and Behavioral Considerations for Antidepressant Applications.

by Jaclyn N. Highland

Dissertation submitted to the faculty of the Graduate School of the University of Maryland, Baltimore in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2020

Dedication

Dedicated to my grandmother, who always told me to “go and get your dream.”

iii

Acknowledgements

I am grateful to all those who have contributed to my experience and helped me along this journey. First, I would like to thank Dr. Todd Gould for his mentorship. Thank you for all that you have taught me, especially how to think like a scientist. I will always remember your encouragement to “do the best experiment possible,” and continue to apply that philosophy to do the best that is possible given the tools at hand, even outside the lab.

I also want to thank all of the members of the Gould lab, past and present, who have taught me so much over the years. A special thank you to Drs. Panos Zanos and Polymnia

Georgiou, who have been there from my first day in the lab and helped me in more ways than I can count; thank you for your advice, insights, and friendship.

Additionally, I want to thank the many collaborators who contributed to the completion of my research projects, including Drs. Patrick Morris and Craig Thomas who synthesized and characterized numerous compounds for these studies, and Drs. Ruin

Moaddel and Amy Wang who helped with the many LC/MS-MS and pharmacokinetic analyses for my projects. I would also like to thank my committee members, Drs. Edna

Pereira, Scott Thompson, Greg Elmer, and Craig Thomas, for all of their constructive insights and guidance throughout this process.

Finally, I would like to thank my family and friends for their continued support. To my parents and my family, thank you for always believing in me, supporting my goals, and being there for me no matter what. I would not be here today without you. To my fiancé,

Dylan, thank for your unwavering love and support. You give me all that I need, whether its encouragement, laughter, or your signature “it will all work out,” and I cannot imagine making it through this journey without you. Last but not least, thank you to my friends, too many to name, for all your support over the years. I am lucky to have you all. iv

Table of Contents

Dedication ...... iii

Acknowledgements ...... iv

List of Tables ...... vii

List of Figures ...... viii

List of Abbreviations ...... ix

Chapter 1. Introduction and background ...... 1

1.1. Depression and treatment strategies ...... 1

1.2. Hydroxynorketamines ...... 4 Formation and metabolism...... 4 Pharmacodynamics and putative sites of actions ...... 11 Behavioral actions ...... 19

1.3. Objectives ...... 26

Chapter 2. Pharmacokinetic characterization of the twelve hydroxynorketamines formed in vivo from ketamine ...... 28

2.1. Introduction ...... 28 2.2. Methods ...... 29 2.3. Results ...... 33 2.4. Discussion ...... 43

Chapter 3. Oral bioavailability and antidepressant-relevant behavioral effects of (2R,6R)-hydroxynorketamine ...... 46

3.1. Introduction ...... 46 3.2. Methods ...... 48 3.3. Results ...... 55 3.4. Discussion ...... 63

Chapter 4. Sex-dependent metabolism and behavioral effectiveness of ketamine and (2R,6R)-hydroxynorketamine ...... 68

4.1. Introduction ...... 68 4.2. Methods ...... 69 4.3. Results ...... 73 4.4 Discussion ...... 80

Chapter 5. Evaluation of the antidepressant-relevant behavioral effectiveness of the (2,6)-hydroxynorketamines ...... 89 v

5.1. Introduction ...... 89 5.2. Methods ...... 91 5.3. Results ...... 93 5.4. Discussion ...... 97

Chapter 6. Discussion and conclusions ...... 103

References ...... 107

List of Tables

Table 1.1. Antidepressant-relevant behavioral actions of (2R,6R)-HNK...... 21

Table 1.2. Antidepressant-relevant behavioral actions of (2S,6S)-HNK...... 22

Table 2.1. Hydroxynorketamine pharmacokinetics in male and female mice...... 35

Table 2.2. Sex-dependent differences in the levels of and exposures to hydroxynorketamines...... 37

Table 3.1. Distribution of (2R,6R)-HNK in the plasma and brain of mice...... 57

Table 3.2. Bioavailability and brain penetrance of (2R,6R)-HNK in mice...... 57

Table 3.3. AUC, Cmax, and relative bioavailability of (2R,6R)-HNK and prodrug candidates in the plasma of mice...... 60

Table 3.4. Adverse effects screening outcomes following oral (2R,6R)-HNK...... 61

Table 5.1 Statistical analyses of effects of the (2,6)-HNKs in the forced swim test...... 95

vii

List of Figures

Figure 1.1. Metabolic formation of HNKs from ketamine...... 6

Figure 2.1. Time course of plasma levels of hydroxynorketamines in mice...... 34

Figure 2.2. Brain levels of and exposure to hydroxynorketamines...... 42

Figure 3.1. Plasma and brain concentrations of (2R,6R)-HNK in mice...... 56

Figure 3.2. Candidate prodrug modifications...... 59

Figure 3.3. Candidate prodrugs do not improve the oral bioavailability of (2R,6R)-HNK in mice...... 59

Figure 3.4. (2R,6R)-HNK does not alter open-field locomotor activity in mice...... 60

Figure 3.5. Oral (2R,6R)-HNK reduces immobility time in the mouse forced-swim test. 62

Figure 3.6. Oral (2R,6R)-HNK reverses learned helplessness in mice...... 63

Figure 4.1. Plasma concentrations of ketamine and metabolites following ketamine administration to male and female mice...... 74

Figure 4.2. Plasma concentrations of (2R,6R)-hydroxynorketamine following direct administration to male and female mice...... 75

Figure 4.3. Gonadectomy alters plasma levels of ketamine and hydroxynorketamine following ketamine treatment in male, but not female, mice...... 76

Figure 4.4. Plasma concentrations of (2R,6R)-hydroxynorketamine following direct administration to gonadectomized or sham male mice...... 77

Figure 4.5. Testosterone did not alter the behavioral effectiveness of ketamine in the mouse forced swim test...... 78

Figure 4.6. Gonadectomy alters weight gain in mice...... 80

Figure 5.1. (2,6)-hydroxynorketamines reduce immobility time in the mouse forced swim test...... 94

Figure 5.2. (5R)-methyl-(2R,6R)-hydroxynorketamine reduces forced swim test immobility time...... 96

viii

List of Abbreviations

2a acetate ester of (2R,6R)-hydroxynorketamine

2b propionate ester of (2R,6R)-hydroxynorketamine

2c pentanoate ester of (2R,6R)-hydroxynorketamine

2d octanoate ester of (2R,6R)-hydroxynorketamine

2e benzoate ester of (2R,6R)-hydroxynorketamine

AMPAR α-amino-3-hydoxy-5-methyl-isoxazolepropionic acid receptor

ANOVA analysis of variance AUC area under the curve

BA bioavailability

BDNF brain derived neurotrophic factor cAMP cyclic adenosine monophosphate

Cmax maximum concentration

CORT corticosterone

CSDS chronic social defeat stress CYP cytochrome P450 enzyme

DHNK dehydronorketamine

ED50 half-maximal effective dose eEF2 eukaryotic elongation factor 2

GDX gonadectomized

ERK extracellular signal related kinase fEPSP field excitatory postsynaptic potential GABA gamma aminobutyric acid GluA1/2/3/4 glutamate ionotropic receptor AMPA type subunit 1/2/3/4 GluN1/2 glutamate ionotropic receptor NMDA type subunit ½

HCl hydrochloride

HK hydroxyketamine

HILIC hydrophobic interaction liquid chromatography HNK hydroxynorketamine hr hours

IC50 half-maximal inhibitory concentration i.c.v. intracerebroventricular i.p. intraperitoneal i.v. intravenous ketamine/KET (R,S)-ketamine

Ki inhibition constant

LC/MS liquid chromatography-mass spectrometry

LLOQ lower limit of quantitation

LTP long-term potentiation

Me methyl mEPSC miniature excitatory postsynaptic current mGlu2/3 metabotropic glutamate receptor subtype 2/3 min minutes

MS/MS tandem mass spectrometry mTOR mammalian target of rapamycin mTORC1 mammalian target of rapamycin complex 1

NA not applicable

ND not determined NMDAR N-methyl-D-aspartate receptor

NMR nuclear magnetic resonance spectrometry norKET norketamine

NS no statistical difference

x p.o. oral

SAL saline

SEM standard error of the mean

SNRI selective norepinephrine reuptake inhibitor

SSRI selective serotonin reuptake inhibitor t1/2 half-life

Tmax time of maximum observed concentration

TrkB tropomyosin receptor kinase B

UGT uridine 5`-diphospho- glucuronosyltransferase

UPLC ultra-performance liquid chromatography

VEH vehicle vlPAG ventrolateral periaqueductal gray

VTA ventral tegmental area

Chapter 1. Introduction and background

1.1. Depression and treatment strategies

Major depression is a leading cause of disability worldwide (Reddy, 2010), with a lifetime risk of approximately 16% (Kessler et al., 2003). A diverse range of symptoms, including depressed mood, anhedonia (decreased capacity to feel pleasure), sleep disturbances, fatigue, cognitive impairments, and suicidal thoughts or behaviors (Kennedy,

2008; Uher et al., 2014), characterize this chronic, debilitating, and even life-threatening disorder. While the etiology of depression is complex and not fully understood, evidence suggests that: (i) weakened synaptic function in various mood-relevant brain regions, including the prefrontal cortex and the hippocampus, contributes to the development of depression-relevant behaviors, and (ii) antidepressant treatments act to restore these synaptic deficits (reviewed in Thompson et al., 2015).

For decades, pharmacotherapies for major depression have targeted the monoaminergic neurotransmitter systems. These classical antidepressants include the monoamine oxidase inhibitors, tricyclic antidepressants, and selective serotonin and/or norepinephrine reuptake inhibitors (SSRIs and SNRIS, respectively), which all act to increase synaptic monoamine levels (reviewed in Racagni & Popoli, 2008; Taylor et al.,

2005). These elevated monoamines are thought to initiate synaptic changes which, over time, restore synaptic transmission in mood-relevant brain regions to induce antidepressant effects (reviewed in Duman et al., 1997; Racagni & Popoli, 2008; Thompson et al., 2015).

However, these classical antidepressants have a long latency to exert their full therapeutic effects, requiring weeks to months of chronic treatment to achieve remission, and up to

30% of depressed patients fail to reach remission despite treatment with multiple classical

1 antidepressants (Gaynes et al., 2009; Insel & Wang, 2009; Rush et al., 2006). Finally, among those patients who do reach remission, the vast majority (approximately 85%) will experience a recurrence of symptoms (Hardeveld et al., 2010). Taken together, these factors highlight the need for novel, effective, and rapid-acting antidepressant treatments.

A profound development in antidepressant research has been the finding that the

N-methyl-D-aspartate receptor (NMDAR) antagonist (R,S)-ketamine (ketamine) rapidly reverses core symptoms of depression in patients classified as treatment-resistant (i.e. patients that previously failed to respond to multiple types of antidepressant treatments;

DiazGranados et al., 2010; Price et al., 2014; Zarate et al., 2012; Zarate et al., 2006). In particular, ketamine exerts antidepressant effects within hours of a single sub-anesthetic infusion, with effects typically lasting one-week post-treatment (e.g. Berman et al., 2000;

DiazGranados et al., 2010; Price et al., 2014; Zarate et al., 2006).

While its full mechanisms are still being elucidated, ketamine is thought to exert its antidepressant actions by rapidly modulating glutamatergic neurotransmission and restoring synaptic transmission in those regions affected by depression (reviewed in

Thompson et al., 2015; Zanos & Gould, 2018). In particular, several hypotheses for ketamine’s mechanism of action have emerged: a) direct inhibition of NMDAR-mediated spontaneous synaptic transmission resulting in disinhibition of eukaryotic elongation factor

2 (eEF2) and subsequent increases in brain-derived neurotrophic factor (BDNF)-mediated protein synthesis (Autry et al., 2011; Gideons et al., 2014; Nosyreva et al., 2013), b) preferential inhibition of NMDARs on gamma aminobutyric acid (GABA)-ergic interneurons resulting in disinhibition of glutamatergic neurotransmission (Homayoun &

Moghaddam, 2007; Moghaddam et al., 1997; Seamans, 2008; Widman & McMahon,

2018), and c) inhibition of extrasynaptic NMDARs resulting in disinhibition of mammalian target of rapamycin complex 1 (mTORC1)-mediated protein synthesis (Miller et al., 2014).

Altogether, these mechanisms—which are not mutually exclusive and may collectively contribute to ketamine’s biological actions—converge upon the upstream enhancement of glutamatergic synaptic transmission to rapidly reverse symptoms of depression (reviewed in Zanos & Gould, 2018).

Nonetheless, while the rapid-acting antidepressant effectiveness of ketamine exceeds that of classical antidepressant drugs, the widespread use of ketamine as an antidepressant is limited by its dissociative side effects and abuse potential (Krystal et al.,

1994; Zanos et al., 2018), which have been attributed to its actions as a potent NDMAR inhibitor (reviewed in Zanos et al., 2018). Thus, identifying compounds which share the rapid-acting antidepressant effectiveness of ketamine, but lack its adverse effect burden, has emerged as a goal in the field of antidepressant research, leading to the study of a range of compounds that target glutamatergic signaling. These potential, novel, rapid-acting antidepressants include NDMAR antagonists selective for the GluN2B subunit (Ates-

Alagoz & Adejare, 2013; Ibrahim et al., 2012; Preskorn et al., 2008; Vasilescu et al., 2017), negative allosteric modulators of α5 subunit-containing GABA receptors (Fischell et al.,

2015; Zanos et al., 2017), group II metabotropic glutamate (mGlu2/3) receptor antagonists

(reviewed in Chaki & Fukumoto, 2018; Swanson et al., 2005; Witkin et al., 2007), and, the focus of this work, hydroxynorketamines (HNKs) (Chen et al., 2020; Chou et al., 2018;

Fukumoto et al., 2019; Highland et al., 2018; Lumsden et al., 2019; Nelson & Trainor,

2007; Pham et al., 2018; Rahman et al., 2020; Yokoyama et al., 2020; Zanos, Highland,

Stewart, et al., 2019; Zanos et al., 2016).

1.2. Hydroxynorketamines

Formation and metabolism

Ketamine metabolism to form hydroxynorketamines

HNKs are metabolites of ketamine formed in vivo following its administration to humans and other animals (Leung & Baillie, 1986; Moaddel et al., 2016; Moaddel et al.,

2010; Zanos et al., 2016; Zarate et al., 2012; Zhao et al., 2012). Twelve individual HNK stereoisomers (Fig 1.1), which differ from each other by the position of cyclohexyl ring hydroxylation and their unique stereochemistry at two stereocenters, have been identified in the plasma of humans (Moaddel et al., 2010; Zarate et al., 2012; Zhao et al., 2012) and in the plasma and brains of rodents (Leung & Baillie, 1986; Moaddel et al., 2016; Zanos et al., 2016) following systemic administration of racemic (R,S)-ketamine (ketamine).

Microsomal studies have identified that ketamine initially undergoes stereoselective N-demethylation primarily catalyzed by the cytochrome P450 (CYP) liver enzymes CYP2B6 and CYP3A4 to (R,S)-norketamine, which is then further metabolized to dehydronorketamine and the HNKs (Desta et al., 2012; Dinis-Oliveira, 2017; Kharasch

& Labroo, 1992; Woolf & Adams, 1987) (Fig 1.1). The cyclohexyl ring of (R,S)- norketamine can be hydroxylated at the 4, 5, or 6 position, resulting in the (2,4)-, (2,5)-, and (2,6)-HNKs, respectively (Desta et al., 2012; Woolf & Adams, 1987). Via this pathway, the production of the (2R,4R;2S,4S)-HNK, (2R,5R;2S,5S)-HNK, and

(2R,6R;2S,6S)-HNK stereoisomers is predominantly catalyzed by CYP2A6 and CYP2B6, while the formation of (2R,4S;2S,4R)-HNK is primarily carried out by CYP3A4 and

CYP3A5, and that of (2R,5S;2S,5R)-HNK, by CYP2B6 (Desta et al., 2012). Through a

4 minor metabolic pathway, ketamine additionally undergoes direct hydroxylation to form the 6-hydroxyketamines (HKs; primarily catalyzed by CYP2A6, CYP3A4, and CYP3A5,

(Desta et al., 2012)), which are then N-demethylated to form the (2,6)-HNKs (Desta et al.,

2012; Woolf & Adams, 1987) (Fig 1.1). The (2R,6R;2S,6S)-HK metabolite is readily demethylated by CYP2B6 to (2R,6R;2S,6S)-HNK, whereas the conversion of

(2R,6S;2S,6R)-HK to (2R,6S;2S,6R)-HNK is reported to be much slower (Desta et al.,

2012).

Following ketamine administration, HNKs have been detected in humans (Fassauer et al., 2017; Grunebaum et al., 2019; Hasan et al., 2017; Moaddel et al., 2010; Zarate et al.,

2012; Zhao et al., 2012), in addition to several other animal species, including mice (Pham et al., 2018; Yamaguchi et al., 2018; Zanos, Highland, Liu, et al., 2019; Zanos et al., 2016;

Zanos et al., 2018), rats (Leung & Baillie, 1986; Moaddel et al., 2015; Moaddel et al.,

2016), dogs (Sandbaumhuter et al., 2016; Sandbaumhuter, Theurillat, & Thormann, 2017;

Theurillat et al., 2016), horses and ponies (Lankveld et al., 2006; Sandbaumhuter,

Theurillat, & Thormann, 2017; Schmitz et al., 2009). Despite differences in study designs, rapid metabolism of ketamine to HNKs appears to be conserved across species, as HNKs were detected at the earliest sampling time points, within 2.5-20 minutes of intraperitoneal

(i.p.) or intravenous (i.v.) ketamine dosing in rodents, dogs, and horses, (Leung & Baillie,

1986; Moaddel et al., 2015; Pham et al., 2018; Yamaguchi et al., 2018; Zanos, Highland,

Liu, et al., 2019; Zanos et al., 2016; Zarate et al., 2012; Zhao et al., 2012) and immediately upon completion of i.v. ketamine infusion in humans (Zarate et al., 2012; Zhao et al., 2012).

Also conserved across those species studied, the (2,6)-HNKs are more abundant than either the (2,5)- or (2,4)-HNKs in the plasma of humans (Moaddel et al., 2010; Zarate et al.,

2012), in the plasma of dogs and horses (Sandbaumhuter, Theurillat, & Thormann, 2017), and in the plasma and brains of mice and rats (Moaddel et al., 2016; Zanos et al., 2016) after ketamine administration.

Figure 1.1. Metabolic formation of HNKs from ketamine. (R,S)-ketamine (KET) undergoes rapid and stereoselective metabolism, catalyzed by the cytochrome P450 (CYP) liver enzymes, to form (R,S)- norketamine (norKET), (R,S)-dehydronorketamine (DHNK), the (2,6)-hydroxyketamines (HKs), and twelve individual hydroxynorketamines (HNKs).

Hydroxynorketamine pharmacokinetics

Following i.p. or i.v. ketamine administration, circulating HNK concentrations reach their maximum within 5-10 min in mice (Yamaguchi et al., 2018; Zanos, Highland,

Liu, et al., 2019; Zanos et al., 2016), 10-36 min in rats (Leung & Baillie, 1986; Moaddel et al., 2016), 49-68 min in dogs (Sandbaumhuter et al., 2016), and 40-230 min (3.83 h) in

6 humans (Zhao et al., 2012). Following direct i.p. administration to rodents, peak plasma levels of (2R,6R)- and (2S,6S)-HNK have been observed at the earliest time studied, within

2.5-10 min of dosing (Lumsden et al., 2019; Yamaguchi et al., 2018; Zanos et al., 2016).

After oral administration, (2S,6S)-HNK has also been detected in the plasma of rats at the earliest time points studied, 10 min post-administration, with peak concentrations occurring at approximately 25 min after dosing (Moaddel et al., 2015). The oral bioavailability of (2S,6S)-HNK was determined to be 46.3% in rats (Moaddel et al., 2015).

(2R,6R)- and (2S,6S)-HNK readily penetrate the brain, where they are detected as early as 2-10 min after peripheral ketamine dosing and 2.5-15 min after their direct administration in mice and rats (at the earliest time points tested; Leung & Baillie, 1986;

Lilius et al., 2018; Lumsden et al., 2019; Moaddel et al., 2015; Yamaguchi et al., 2018;

Zanos, Highland, Liu, et al., 2019; Zanos et al., 2016). Of note, local metabolism of ketamine to HNKs does not appear to occur in the brain (Moaddel et al., 2015; Zanos et al., 2018), and therefore brain levels reflect penetration from the periphery. Peak brain concentrations of HNKs are achieved within 10-15 min of ketamine dosing and within 5-

10 minutes of direct administration in rodents (Leung & Baillie, 1986; Lumsden et al.,

2019; Moaddel et al., 2015; Yamaguchi et al., 2018; Zanos, Highland, Liu, et al., 2019;

Zanos et al., 2016). The ratio of brain to plasma exposures for both (2R,6R)- and (2S,6S)-

HNK are approximately 1:1 (reported between 0.9-1.3) in rodents (Leung & Baillie, 1986;

Moaddel et al., 2015; Yamaguchi et al., 2018; Zanos et al., 2016).

HNKs undergo glucuronide conjugation, at least partly catalyzed by the UDP- glucuronosyltransferase isoform UGT2B4 (Moaddel et al., 2010), and are eliminated both in their unconjugated and conjugated forms in urine and bile (Chang & Glazko, 1974;

Dinis-Oliveira, 2017; Lankveld et al., 2006; Moaddel et al., 2010; Sandbaumhuter &

Thormann, 2018; Turfus et al., 2009). Plasma elimination half-lives were determined to be

13.2 ± 3.7 h (mean ± SD) for (2R,6R;2S,6S)-HNK following i.p. injection of ketamine to rats (Moaddel et al., 2016), 0.56 h for (2R,6R)-HNK following i.p. injection of (R)- ketamine to mice (Zanos, Highland, Liu, et al., 2019), approximately one h (median 56.95 min) for HNKs following i.v. injection of ketamine in horses (Lankveld et al., 2006), and

14.25 ± 3.95 h (mean ± SD) for (2R,6R;2S,6S)-HNK following i.v. infusion of ketamine in humans (Farmer et al., 2020). However, HNK half-lives were reported to show considerable inter-individual variability between human subjects (Zhao et al., 2012).

HNKs remain detectable in the plasma and brain of mice for 2-4 h after i.p. ketamine dosing (Zanos, Highland, Liu, et al., 2019; Zanos et al., 2016). One study reported that in rats, (2R,6R;2S,6S)-HNK could be detected (≥ 5.0 ng/ml) in the plasma up to 48 h after a single i.p. dose of ketamine, while levels of all other HNKs were below levels of quantitation within 4 h of ketamine dosing (Moaddel et al., 2016). Following a single i.v. ketamine dose, HNKs were detected in the plasma of dogs for at least 7.5 h

(Sandbaumhuter et al., 2016; Theurillat et al., 2016) and in the plasma of horses up to 2 h

(Lankveld et al., 2006). In humans, circulating (2R,5R;2S,5S)-, (2R,5S;2S,5R)-, and

(2R,6S;2S,6R)-HNKs could be detected up to 230 min (3.83 h) after a single ketamine infusion (0.5 mg/kg, 40 min infusion; Zarate et al., 2012), while the (2R,6R;2S,6S)-HNK stereoisomers were measurable in plasma (Grunebaum et al., 2019; Zarate et al., 2012;

Zhao et al., 2012) and urine (Fassauer et al., 2017) for 1-3 days post-infusion. Following their direct administration, (2R,6R)- and (2S,6S)-HNK remain detectable in the plasma and

8 brain up to 2-4 h after a single dose in mice (Zanos et al., 2016) and up to 8 h in rats

(Moaddel et al., 2015).

Factors altering hydroxynorketamine metabolism and exposure

Several factors have been reported to impact the metabolism of HNKs, including biological sex (Zanos et al., 2016), circadian cycle (Martinez-Lozano Sinues et al., 2017), and prior treatment with other compounds (Lilius et al., 2018; Sandbaumhuter et al., 2016;

Sandbaumhuter, Theurillat, Bettschart-Wolfensberger, et al., 2017; Yamaguchi et al.,

2018). Importantly, these factors are important variables that may influence the pharmacodynamic effects of HNKs, including their behavioral actions, and the possible

HNK-dependent actions of ketamine.

Peak and total concentrations of (2R,6R;2S,6S)-HNK following ketamine administration are higher in female mice relative to males (Zanos et al., 2016). The circadian cycle has also been reported to alter ketamine metabolism to HNKs in mice, with approximately two-fold higher concentrations of (2R,6R)-HNK detected in exhaled breath of mice receiving a ketamine injection in the evening compared to those injected with an equivalent dose in the morning (Martinez-Lozano Sinues et al., 2017). However, this finding has not been confirmed with direct assessment of HNK concentrations in plasma or other tissues.

Prior treatment with or co-administration of various compounds have also been reported to alter ketamine metabolism and HNK exposure (Lilius et al., 2018;

Sandbaumhuter et al., 2016; Sandbaumhuter, Theurillat, Bettschart-Wolfensberger, et al.,

2017; Yamaguchi et al., 2018). Namely, in dogs sedated with the α2-agonist medetomidine

9 prior to ketamine administration, both metabolism of ketamine to (2R,6R;2S,6S)-HNK and elimination of (2R,6R;2S,6S)-HNK were faster (Sandbaumhuter et al., 2016). However, an in vitro study found that medetomidine attenuated the formation of (2R,6R;2S,6S)-HNK from ketamine in liver microsomes, while three other α2-agonists, detomidine, xylazine, and romifidine, had the opposite effect, resulting in higher levels of (2R,6R;2S,6S)-HNK

(Sandbaumhuter, Theurillat, Bettschart-Wolfensberger, et al., 2017). Although these interactions require further clarification, the effects of medetomidine are likely CYP- mediated, as CYP3A plays an important role in the metabolism of both medetomidine and ketamine (Desta et al., 2012; Duhamel et al., 2010; Sandbaumhuter et al., 2015).

A recent study also demonstrated that prior treatment with morphine resulted in higher serum and brain levels of (2R,6R;2S,6S)-HNK in ketamine-treated rats (Lilius et al., 2018). In addition, as expected, prior treatment with the CYP inhibitors ticlopidine and

1-aminobenzotriazole led to a robust attenuation of (2R,6R)-HNK formation following (R)- ketamine administration in mice (Yamaguchi et al., 2018).

A chemical modification to the structure of ketamine has also been utilized to alter its metabolism to HNKs; namely, di-deuterium substitution at the C6 position of the hydroxyl ring of ketamine or (R)-ketamine (designated as d2-ketamine and (R)-d2- ketamine, respectively) results in a selective and robust attenuation of the metabolism of ketamine to the (2,6)-HNKs or (2R,6R)-HNK, respectively (Zanos, Highland, Liu, et al.,

2019; Zanos et al., 2016; Zhang et al., 2018).

Pharmacodynamics and putative sites of actions

Although the mechanisms underlying the actions of HNKs are still being elucidated, several potential sites of action have been identified. Numerous studies have demonstrated HNK-induced effects on the glutamatergic system, including enhanced glutamate release (Pham et al., 2018; Riggs et al., 2019), increased α-amino-3-hydroxy-5- methyl-4-isoxazolepropionic acid receptor (AMPAR) expression, rapid facilitation of

AMPAR-dependent synaptic transmission (Ho et al., 2018; Riggs et al., 2019; Shaffer et al., 2019; Zanos et al., 2016), possible effects on metabotropic glutamate (mGlu) receptor signaling (Wray et al., 2019; Zanos, Highland, Stewart, et al., 2019), and actions on pathways downstream of glutamatergic signaling, including the mTORC1 and BDNF pathways (Fred et al., 2019; Fukumoto et al., 2019; Paul et al., 2014; Zanos et al., 2016).

In addition, effects on other neurotransmitters, including serotonin and norepinephrine have also been reported (Ago et al., 2019), and changes in neuronal morphology and structural plasticity have been observed (Cavalleri et al., 2018; Collo et al., 2018). Outside of these effects on synaptic function, effects on other cellular processes, including inflammatory responses (Ho et al., 2019; Rahman et al., 2020; Xiong et al., 2019) and mitochondrial function (Faccio et al., 2018; Rahman et al., 2020) have also been reported.

Synaptic actions

NDMAR inhibition

Although ketamine is a potent NMDAR antagonist, it has been demonstrated that its HNK metabolites have substantially lower affinity to bind to or inhibit NMDARs

(Lumsden et al., 2019; Moaddel et al., 2013; Morris et al., 2017b). In particular, based on

MK-801 binding displacement, while still lower than that of ketamine, (2S,6S)-HNK exhibits a greater NMDAR binding affinity (Ki: 7.34-21.19 µM) compared to any other

HNK stereoisomers (Kis of (2R,6R)-, (2R,6S)-, (2S,6R)-, (2R,5R)-, (2S,5S)-, (2R,5S)-,

(2S,5R)-, (2R,4R)-, (2S,4S)-, (2R,4S)-, and (2S,4R)-HNK are >100 µM) (Moaddel et al.,

2013; Morris et al., 2017b).

Compared to (2R,6R)-HNK, (2S,6S)-HNK has greater potency to suppress

NMDAR-mediated currents in cultured hippocampal neurons and oocytes (11-22-fold greater potency), NMDAR-mediated field excitatory potentials (fEPSPs) in hippocampal slices (4.5-fold greater potency), and NMDA-induced lethality in mice (an in vivo measure of functional NMDAR inhibition, 12-fold greater potency; Lumsden et al., 2019).

However, compared to ketamine, (2S,6S)-HNK has lower potency to inhibit NMDARs, demonstrating 5-13-fold lower potency to inhibit NMDAR-mediated currents, 10-fold lower potency to inhibit NMDAR-mediated fEPSPs, and nearly 3-fold lower potency to attenuate NMDA-induced lethality in mice (Lumsden et al., 2019). (2R,6R)-HNK has even lower potency compared to ketamine, demonstrating 57-150-fold, 47-fold, and 35-fold lower potency to inhibit NMDAR-mediated currents (Abbott & Popescu, 2020; Lumsden et al., 2019), NMDA fEPSPs (Lumsden et al., 2019), and NMDA-induced lethality

(Lumsden et al., 2019), respectively, compared to ketamine. Altogether, the literature demonstrates that (2S,6S)-HNK exerts moderate NMDAR inhibition, albeit with lower potency compared to ketamine, while (2R,6R)-HNK has substantially lower potency to inhibit NMDARs. While the other HNKs have not been extensively studied, their low

NMDAR binding affinities (similar to (2R,6R)-HNK) suggest low potential for NMDAR inhibition.

Effects on glutamatergic neurotransmission

Evidence suggests that (2R,6R)-HNK can enhance glutamatergic neurotransmission (Chou et al., 2018; Riggs et al., 2019; Ye et al., 2019; Zanos et al., 2016), at least partly via increased glutamate release (Ho et al., 2018; Pham et al., 2018; Riggs et al., 2019; Shaffer et al., 2019; Wray et al., 2019; Zanos et al., 2016). For instance, administration of (2R,6R)-HNK (10 mg/kg, i.p. injection or 1 nmol/side intracortical perfusion) results in increased glutamate release in the prefrontal cortex of mice (Pham et al., 2018). Also consistent with (2R,6R)-HNK-induced increases in glutamate release, it has been demonstrated that (2R,6R)-HNK (10 µM) increases the frequency, but not amplitude, of miniature excitatory postsynaptic currents (mEPSCs) in CA1 pyramidal neurons in hippocampal slices (Riggs et al., 2019), and leads to a potentiation of glutamatergic synaptic transmission in hippocampal slices (0.3-30 µM; Riggs et al., 2019;

Zanos et al., 2016) which is concurrent with reductions in paired pulse facilitation and occluded by high release probability (Riggs et al., 2019). It has also been demonstrated that

(2R,6R)-HNK increases the frequency and amplitude of mEPSCs (10 µM; also consistent with a post-synaptic component to its effects in addition to a pre-synaptic increase in release probability) and reverses chronic stress-induced decreases in AMPAR-mediated currents in the ventrolateral periaqueductal gray (10 mg/kg, i.p.; vlPAG; Chou et al., 2018;

Ye et al., 2019). Of note, neither (2R,6R)- nor (2S,6S)-HNK appear to directly bind to or activate AMPARs (10 µM; Shaffer et al., 2019).

In other brain regions, including the nucleus accumbens and ventral tegmental area

(VTA), (2R,6R)-HNK has also been demonstrated to modulate glutamatergic neurotransmission, although an attenuation of glutamatergic activity has been reported in

13 these areas (Yao et al., 2017), in contrast to the effects described above. Namely, (2R,6R)-

HNK decreased mEPSC frequency and amplitude and also decreased AMPAR- relative to

NMDAR-mediated transmission in the VTA, and attenuated the magnitude and prevalence of LTP in the nucleus accumbens (10 mg/kg, i.p.; Yao et al., 2018). Nevertheless, the literature supports that (2R,6R)-HNK modulates glutamatergic neurotransmission in various brain regions, actions which may lead to network-wide excitation, as evidenced by an increase in high-frequency gamma (30-80 Hz) electrocorticographic oscillations (30-80

Hz) following (2R,6R)-HNK administration in mice (10 mg/kg, i.p.), a marker of cortical activation (Zanos, Highland, Stewart, et al., 2019; Zanos et al., 2016).

There is some evidence to suggest that cyclic adenosine monophosphate (cAMP)- dependent mechanisms (Wray et al., 2019), and in particular, type II metabotropic glutamate (mGlu2) receptor signaling (Zanos, Highland, Stewart, et al., 2019), may be implicated in the (2R,6R)-HNK-induced synaptic potentiation of excitatory synapses. mGlu2 receptors are presynaptic autoreceptors that are negatively coupled to adenylyl cyclase and suppress cAMP-dependent glutamate release under normal conditions

(reviewed in Schoepp, 2001). It has been demonstrated that (2R,6R)-HNK has a convergent mechanism of action with mGlu2 receptors (Zanos, Highland, Stewart, et al., 2019). In particular, both the (2R,6R)-HNK-induced increases in cortical gamma oscillations and antidepressant-relevant behavioral actions are bi-directionally modulated by mGlu2 receptor activity: the effects of (2R,6R)-HNK are blocked by mGlu2/3 activation, whereas inhibiting mGlu2/3 activity enhances the effects of subthreshold (2R,6R)-HNK (Zanos,

Highland, Stewart, et al., 2019). Further, it was demonstrated that (2R,6R)-HNK prevented mGlu2/3 receptor agonist-induced hyperthermia in mice, a physiological assay for

14 functional mGlu2/3 receptor inhibition (Zanos, Highland, Stewart, et al., 2019). Finally,

(2R,6R)-HNK has been shown to induce redistribution of the G protein, GαЅ, to non-lipid raft regions, resulting in a robust increase in cAMP accumulation in C6 cells (Wray et al.,

2019), suggesting that (2R,6R)-HNK may acts via an mGlu2 receptor- and cAMP- dependent pathway to modulate glutamate release and enhance network activity.

Targets downstream of glutamatergic signaling

(2R,6R)-HNK administration has been shown to alter several targets downstream of glutamate receptor expression and/or activity, including BDNF (Fukumoto et al., 2019;

Lumsden et al., 2019; Zanos et al., 2016) and mTOR (Fukumoto et al., 2019; Lumsden et al., 2019; Paul et al., 2014). In particular, (2R,6R)-HNK has been reported to increase

BDNF protein expression in the hippocampus of mice (10 mg/kg, i.p.; observed 24 h post- treatment; Zanos et al., 2016) and increase BDNF release in primary neuronal cultures (10-

50 nM; 1 h exposure; Fukumoto et al., 2019). Additionally, it has been reported that

(2R,6R)-HNK may enhance tropomyosin receptor kinase B (TrkB) surface expression by disrupting adaptor protein complex 2-mediated TrkB endocytosis (10-100 µM, 15 min exposure; Fred et al., 2019). Notably, there is evidence that BDNF/TrkB signaling contributes to the antidepressant-relevant behavioral outcomes of (2R,6R)-HNK in mice

(Fukumoto et al., 2019).

Subsequent to BDNF/TrkB signaling, mTOR pathway activity and downstream signaling has been demonstrated to be increased by HNKs (Fukumoto et al., 2019;

Lumsden et al., 2019; Paul et al., 2014; Singh et al., 2016). Namely, both (2R,6R)-HNK increased mTOR phosphorylation in the mouse hippocampus (10 mg/kg, i.p, 30 min post-

15 treatment; Lumsden et al., 2019) and prefrontal cortex (30 mg/kg, i.p.; 30 min post- treatment; Fukumoto et al., 2019). Similarly, (2S,6S)-HNK increased mTOR phosphorylation in the rat prefrontal cortex (20 mg/kg, i.v., 20 min post-treatment) and in

PC-12 cells (0.01-1 nM, 1 h exposure; Paul et al., 2014). Downstream targets of mTOR signaling, including extracellular signal related kinase (ERK; 1 h exposure of rat primary neurons to 1-50 nM (2R,6R)-HNK or 0.5-1 nM exposure of PC-12 cells of (2S,6S)-HNK) and protein kinase B (1 h exposure of PC-12 cells to 0.1-1 nM (2S,6S)-HNK) also demonstrate enhanced activation following (2R,6R)-HNK and (2S,6S)-HNK exposure

(Fukumoto et al., 2019; Paul et al., 2014). The (2R,6R)-HNK-induced increase in ERK phosphorylation was determined to be dependent on TrkB and AMPAR activity

(Fukumoto et al., 2019).

In addition to AMPAR-depending synaptic transmission and effects on intracellular signaling cascades, there is evidence that HNKs may exert their sustained actions via increases the expression AMPARs (Ho et al., 2018; Shaffer et al., 2019; Zanos et al., 2016).

In particular, increases in GluA1 and GluA2 protein expression has been observed in the hippocampus of mice following (2R,6R)-HNK dosing (10 mg/kg, i.p.; 24 h post-treatment) in vivo (Zanos et al., 2016), while increases in GluA1 surface protein expression (0.1-10

µM, 90-180 min exposure; rat primary neurons) and in GluA1, GluA2, and GluA4 mRNA expression (200-400 nM, 24 h exposure; U251-MG glioblastoma cells or human iPSC- derived astrocytes) were observed following (2R,6R)-HNK exposure in cell cultures (Ho et al., 2018; Shaffer et al., 2019). Similarly, (2S,6S)-HNK (200-400 nM, 24 h exposure;

U251 MG cells or human iPSC-derived astrocytes) increased GluA1, GluA2, and GluA4 mRNA expression in cell cultures (Ho et al., 2018). Notably, AMPAR activity has been

16 demonstrated to be required for (2R,6R)-HNK to exert its antidepressant-relevant behavioral effects in mice (Zanos et al., 2016).

Other neurotransmitters

In addition to the glutamatergic actions of HNKs, the effects of HNKs on several other neurotransmitters have been studied (Ago et al., 2019; Can et al., 2016; Pham et al.,

2018). In particular, (2R,6R)-HNK, but not (2S,6S)-HNK, increased extracellular serotonin

(10-20 mg/kg, i.p.; Ago et al., 2019; Pham et al., 2018) and norepinephrine (20 mg/kg, i.p.;

Ago et al., 2019) levels in the medial prefrontal cortex of mice, while neither compound altered dopamine levels (20 mg/kg, i.p.; Ago et al., 2019). However, neither (2R,6R)- nor

(2S,6S)-HNK had functional activity on monoamine transporters or dopamine receptors

(up to 10 µM; Can et al., 2016).

Effects on cellular morphology and structural plasticity

There is some evidence that (2R,6R)-HNK can induce morphological changes and structural plasticity (Collo et al., 2018), although some results have been mixed (Fukumoto et al., 2019; Michaelsson et al., 2019). Of note, these changes may be involved in the antidepressant effects of HNKs. In particular, it has been demonstrated that the formation of dendritic spines in the prefrontal cortex is necessary to maintain ketamine’s antidepressant-relevant behavioral effects following corticosterone treatment in mice

(Moda-Sava et al., 2019). It has been demonstrated that (2R,6R)-HNK enhances dendritic outgrowth in primary mouse and human induced pluripotent stem cell-derived dopaminergic neurons (0.5 µM, observed 3 days after 1-6 h exposure; Cavalleri et al.,

2018), a process which was shown to be to be AMPAR- and mTOR-dependent (Collo et al., 2018). However, additional studies have reported that racemic (2R,6R;2S,6S)-HNK (1-

100 µM) did not alter cellular proliferation in neurosphere cultures (Michaelsson et al.,

2019), and (2R,6R)-HNK (30 mg/kg, i.p.) did not alter spine density measured 24 h after treatment in vivo in mice (Fukumoto et al., 2019). Altogether, the effects of (2R,6R)-HNK on morphological changes require further validation, and potential effects of additional

HNKs should also be assessed.

Non-synaptic effects

While mechanistic studies of HNKs have largely focused on actions relevant to the modulation of synaptic transmission, there is also evidence that (2R,6R)- and (2S,6S)-HNK may alter several non-synaptic cellular processes including inflammation (Ho et al., 2019;

Xiong et al., 2019), as well as mitochondrial function and energy metabolism (Faccio et al., 2018; Rahman et al., 2020). In particular, there is some evidence that (2R,6R)- and

(2S,6S)-HNK may increase anti-inflammatory markers in HMC3 human microglial cells

(400 nM, 24 h exposure; Ho et al., 2019), or decrease mediators of inflammation in mice

(10 mg/kg, i.p.; Rahman et al., 2020), although one study reported that (2R,6R)-HNK (10 mg/kg, i.p.) did not alter systemic markers of inflammation in socially defeated mice

(Xiong et al., 2019). In addition to these potential anti-inflammatory actions, there is some evidence that (2R,6R)- and (2S,6S)-HNK alter mitochondrial function and energy metabolism, including increasing indicators of glycolysis and altering fatty acid metabolism in PC12 cells (36 h exposure to 5 nM of (2R,6R)-HNK or 0.5 nM of (2S,6S)-

HNK; Faccio et al., 2018). Indicators of glycolysis were also increased following (2R,6R)-

HNK treatment (10 mg/kg, i.p.) in vivo in mice (Rahman et al., 2020). However, these effects and their therapeutic implications require further clarification.

Behavioral actions

Preclinical behavioral studies

Antidepressant-relevant actions

To date, HNKs have primarily received attention for their therapeutic potential as novel antidepressant compounds, given a growing number of preclinical studies demonstrating that (2R,6R)- and (2S,6S)-HNK induce behavioral actions in preclinical test that predict antidepressant effectiveness (Chen et al., 2020; Chou et al., 2018; Fukumoto et al., 2019; Highland et al., 2018; Lumsden et al., 2019; Nelson & Trainor, 2007; Pham et al., 2018; Rahman et al., 2020; Yokoyama et al., 2020; Zanos, Highland, Stewart, et al.,

2019; Zanos et al., 2016).

One of the first indications of the antidepressant-relevant behavioral effects of

HNKs was the finding that metabolism of ketamine to form the (2,6)-HNKs is critical for its sustained antidepressant-relevant behavioral effects in mouse models (Zanos et al.,

2016). In particular, it was demonstrated that deuterium substitution at the C6 position of ketamine, a chemical modification which selectively and robustly hinders metabolism to the (2,6)-HNKs, prevents the sustained behavioral effects of ketamine in the learned helplessness and forced swim tests assessed 24 h post-treatment (Zanos et al., 2016). The analogous deuterium substitution to (R)-ketamine (specifically preventing metabolism to

(2R,6R)-HNK) attenuated its behavioral effects in the same tests (Zanos, Highland, Liu, et al., 2019). Further, it has been demonstrated that female mice are more sensitive to

19 ketamine’s antidepressant-like behavioral effects, responding at lower doses than are required to exert effects in males (Saland et al., 2017; Zanos et al., 2016), concurrent with higher peak brain levels of (2R,6R;2S,6S)-HNK observed in female mice (Zanos et al.,

2016). However, the roles of (2R,6R)-HNK in mediating ketamine’s antidepressant- relevant behavioral effects have been questioned. In particular, one study reported that deuterium substitution did not attenuate the antidepressant-like actions of (R)-ketamine in mice (Zhang et al., 2018), and a second study reported that prior treatment with CYP inhibitors, while attenuating metabolism to (2R,6R)-HNK, did not abolish (R)-ketamine’s antidepressant-like behavioral effects (Yamaguchi et al., 2018). It should be noted, however, that (R)-ketamine itself exerts HNK-independent behavioral effects (Zanos,

Highland, Liu, et al., 2019); thus, the contribution of (R)-ketamine vs. HNK in modulating the observed behavioral outcomes requires further clarification.

Nonetheless, independent of a potential role in mediating ketamine’s antidepressant effects, a growing number of studies have demonstrated that direct administration of

(2R,6R)-HNK and, to a lesser extent, (2S,6S)-HNK, induces behavioral effects in a variety preclinical rodent studies used to predict antidepressant effectiveness (summarized in

Tables 1.1 and 1.2; Chou et al., 2018; Fukumoto et al., 2019; Highland et al., 2018;

Lumsden et al., 2019; Nelson & Trainor, 2007; Pham et al., 2018; Rahman et al., 2020;

Yokoyama et al., 2020; Zanos, Highland, Stewart, et al., 2019; Zanos et al., 2016). These rapid-acting antidepressant-relevant behavioral effects have been observed within hours of

Table 1.1. Antidepressant-relevant behavioral actions of (2R,6R)-HNK.

Note: Sub-effective doses and studies that did not identify an effect of (2R,6R)-HNK were not summarized in this table; please refer to the text for these details. Abbreviations: CORT, corticosterone; CSDS, chronic social defeat stress; HNK, hydroxynorketamine; i.c.v., intracerebroventricular; i.p., intraperitoneal; hr, hours; vlPAG, ventrolateral periaqueductal gray.

21 treatment and are also sustained for days to weeks following HNK administration (Chou et al., 2018; Fukumoto et al., 2019; Highland et al., 2018; Lumsden et al., 2019; Nelson &

Trainor, 2007; Pham et al., 2018; Rahman et al., 2020; Yokoyama et al., 2020; Zanos,

Highland, Stewart, et al., 2019; Zanos et al., 2016).

Table 1.2. Antidepressant-relevant behavioral actions of (2S,6S)-HNK.

Note: Sub-effective doses and studies that did not identify an effect of (2S,6S)-HNK were not summarized in this table; please refer to the text for these details. Abbreviations: CORT, corticosterone; HNK, hydroxynorketamine; intraperitoneal; hr, hours.

In particular, (2R,6R)-HNK has been demonstrated to reduce immobility time in the forced swim test (Chen et al., 2020; Chou et al., 2018; Fukumoto et al., 2019; Lumsden et al., 2019; Pham et al., 2018; Zanos, Highland, Liu, et al., 2019; Zanos, Highland,

Stewart, et al., 2019; Zanos et al., 2016), reverse escape deficits in the learned helplessness test (Elmer et al., 2020; Zanos, Highland, Liu, et al., 2019; Zanos et al., 2016), reduce latency to eat in the novelty-suppressed feeding test (Fukumoto et al., 2019; Lumsden et al., 2019; Zanos et al., 2016), and exert anti-anhedonic effects following chronic stress

(Fukumoto et al., 2019; Zanos, Highland, Stewart, et al., 2019; Zanos et al., 2016). (2S,6S)-

HNK, while not studied as extensively as (2R,6R)-HNK, has also been demonstrated to exert antidepressant-like behavioral effects, including those observed in the forced swim

(Yokoyama et al., 2020; Zanos et al., 2016) and learned helplessness tests (Zanos et al.,

2016), as well as anti-anhedonic effects following chronic stress (Yokoyama et al., 2020).

Experimental details for each outcome, including effective doses and timing of testing following administration of (2R,6R)- and (2S,6S)-HNK, are listed in Table 1.1. and Table

1.2, respectively. In these behavioral tests, (2R,6R)-HNK has demonstrated superior behavioral effectiveness, with lower minimal effective doses when compared to (2S,6S)-

HNK under the same experimental conditions (5 vs. 25 mg/kg, i.p. in the forced swim test and 5 vs. 75 mg/kg, i.p. in the learned helplessness test; Zanos et al., 2016). Also consistent with its greater antidepressant effectiveness relative to (2S,6S)-HNK, it has also been reported that (2R,6R)-, but not (2S,6S)-HNK, exerted behavioral effects in the forced swim and learned helplessness tests when tested at equivalent doses (Chou, 2020; Chou et al.,

2018).

Of note, some discrepancies in the antidepressant-like behavioral effects of

(2R,6R)- and (2S,6S)-HNK have been reported. Namely, several studies, primarily from one group of researchers, have failed to detect effects of (2R,6R)-HNK in behavioral tests of antidepressant effectiveness (Herzog et al., 2020; Shirayama & Hashimoto, 2018; Xiong et al., 2019; Yamaguchi et al., 2018; Yang et al., 2017; Yokoyama et al., 2020). However, the lack of a complete dose response (i.e. testing only a single dose) in most of those studies may explain these inconsistences, in contrast to the number of studies reporting effects in similar behavioral tests (reviewed in Hillhouse et al., 2019). In addition, in contrast to the enhanced effectiveness of (2R,6R)-HNK noted above, one study reported that (2S,6S)-, but not (2R,6R)-HNK, reduced immobility time in the forced swim test and exerted anti- anhedonic effects following chronic corticosterone treatment when tested at equivalent

23 doses (Yokoyama et al., 2020). It is unclear whether this discrepancy may be explained by differences in experimental methods.

Consistent with a site of action localized in the central nervous system, it has been demonstrated that direct administration of (2R,6R)-HNK into the brain, including infusion into the cerebral ventricles, (Zanos, Highland, Liu, et al., 2019), prefrontal cortex

(Fukumoto et al., 2019; Pham et al., 2018), or vlPAG (Chou et al., 2018), induced antidepressant-relevant behavioral effects.

Other behavioral effects

There is some preclinical evidence to suggest that (2R,6R)- and (2S,6S)-HNK may exert analgesic actions (Kroin et al., 2019; Lilius et al., 2018) or enhanced aggressive behaviors and social dominance (Chou, 2020; Ye et al., 2019) in rodents. However, additional studies are needed to determine the potential therapeutic implications of these effects.

Characterization of adverse effects

Preclinical studies have studied adverse behavioral effects following treatment of rodents with (2R,6R)- and (2S,6S)-HNK, demonstrating an innocuous adverse effect profile for (2R,6R)-HNK to date (Lilius et al., 2018; Zanos et al., 2016). In particular, (2R,6R)-

HNK did not alter locomotor activity (up to 125 mg/kg, i.p.; Zanos et al., 2016), sensorimotor gating (up to 375 mg/kg, i.p.; Zanos et al., 2016), or memory performance

(10 mg/kg, i.p; Herzog et al., 2020), and did not induce motor incoordination (up to 125 mg/kg, i.p.; Zanos et al., 2016) in mice. Further, (2R,6R)-HNK was not self-administered by mice, suggesting low abuse liability (Zanos et al., 2016). Overall, it has been 24 demonstrated that (2R,6R)-HNK lacks overt adverse effects, including the NMDAR- mediated dissociative properties and abuse liability characteristic of ketamine. In contrast, some adverse effects have been observed following dosing of (2S,6S)-HNK (Zanos et al.,

2016). Namely, (2S,6S)-HNK has been reported to induce hyperlocomotion (at doses >25 mg/kg, i.p.) and motor incoordination (125 mg/kg, i.p.) in mice (Zanos et al., 2016), effects likely attributed to its moderate potency to inhibit NMDARs. The potential adverse effect burdens of the other HNKs have not yet been characterized.

Associations between plasma HNK levels and clinical outcomes following ketamine treatment for depression

Although no clinical studies have directly evaluated the antidepressant effects of

HNKs in humans to date, several retrospective studies have assessed associations between antidepressant response and plasma levels of HNKs following ketamine infusion. Although

Zarate et al. (2012) detected a trend for higher levels of (2R,6R;2S,6S)-HNK in plasma of patients with bipolar depression who responded to ketamine treatment, this effect did not reach statistical significance, and there was no difference in plasma (2R,6R;2S,6S)-HNK levels detected between responders and non-responders in patients with unipolar depression. The same study reported that higher plasma (2R,5R;2S,5S)-HNK levels were associated with non-response in the bipolar group, and that plasma levels of both

(2R,5R;2S5S)- and (2R,5S;2S,5R)-HNK were inversely correlated with psychotic and dissociative symptoms (Zarate et al., 2012). A second study reported that, in ketamine- treated patients, plasma levels of both (2R,6R)- and (2S,6S)-HNK were inversely correlated with improvement in depression and suicidal ideation, but neither was associated with

25 acute dissociative or anxiogenic side effects (Grunebaum et al., 2019). Similarly, a third study reported that plasma (2R,6R;2S,6S)-HNK levels were inversely correlated with antidepressant response (Farmer et al., 2020). These negative associations between plasma

(2R,6R;2S,6S)-HNK levels and improvements in depression symptoms following ketamine infusion in clinical studies seemingly contradict the preclinical findings reporting antidepressant effects of HNKs. It is important to note, however, that the HNK-independent antidepressant effects reported earlier (Zanos, Highland, Liu, et al., 2019; Zhang et al.,

2018) can confound interpretation of the relationship between plasma levels of HNKs and the magnitude of antidepressant responses in ketamine-treated subjects. The possibility that direct administration of HNKs, including but not limited to (2R,6R)-HNK, exerts antidepressant actions in human awaits testing in clinical trials.

1.3. Objectives

While a growing number of studies have demonstrated that (2R,6R)-HNK, and, to a lesser extent, (2S,6S)-HNK exert antidepressant-relevant behavioral effects in preclinical rodent models (Chen et al., 2020; Chou et al., 2018; Fukumoto et al., 2019; Highland et al.,

2018; Lumsden et al., 2019; Nelson & Trainor, 2007; Pham et al., 2018; Rahman et al.,

2020; Yokoyama et al., 2020; Zanos, Highland, Stewart, et al., 2019; Zanos et al., 2016), a number of important considerations for the development of HNKs as antidepressant treatments had not previously been evaluated. This work aimed to address some of these factors. Namely, the central goals of this dissertation were:

1) To characterize the pharmacokinetic profile of all 12 HNKs

2) To evaluate the oral bioavailability and antidepressant-like behavioral efficacy of oral (2R,6R)-HNK

3) To characterize the sex-dependent differences in the metabolism and behavioral effectiveness of ketamine and (2R,6R)-HNK

4) To evaluate the relative effectiveness of the (2,6)-HNKs to induce antidepressant-relevant behavioral actions

Chapter 2. Pharmacokinetic characterization of the twelve hydroxynorketamines formed in vivo from ketamine

2.1. Introduction

Ketamine undergoes rapid and stereoselective metabolism forming a number of metabolites that include twelve distinct hydroxynorketamines (HNKs). (2R,6R)-HNK and, to a lesser extent, (2S,6S)-HNK have gained attention for their rapid-onset antidepressant- relevant effects in preclinical studies (Chen et al., 2020; Chou et al., 2018; Elmer et al.,

2020; Fukumoto et al., 2019; Lumsden et al., 2019; Pham et al., 2018; Yokoyama et al.,

2020; Zanos, Highland, Liu, et al., 2019; Zanos, Highland, Stewart, et al., 2019; Zanos et al., 2016).

In particular, a growing number of studies demonstrate that (2R,6R)-HNK exerts behavioral effects thought to predict antidepressant efficacy in a number of rodent tests

(Chen et al., 2020; Chou et al., 2018; Elmer et al., 2020; Fukumoto et al., 2019; Lumsden et al., 2019; Pham et al., 2018; Zanos, Highland, Liu, et al., 2019; Zanos, Highland,

Stewart, et al., 2019; Zanos et al., 2016). While (2R,6R)-HNK has been more thoroughly studied to date, several studies have demonstrated that (2S,6S)-HNK also exerts antidepressant-like behavioral effects (Chen et al., 2020; Yokoyama et al., 2020; Zanos et al., 2016), albeit with lower apparent effectiveness when compared to (2R,6R)-HNK under similar experimental conditions (Zanos et al., 2016).

Following their direct administration to rodents, (2R,6R)- and (2S,6S)-HNK readily penetrate the brain where they can be detected within minutes of systemic dosing (Leung

& Baillie, 1986; Lilius et al., 2018; Lumsden et al., 2019; Moaddel et al., 2015; Yamaguchi et al., 2018; Zanos, Highland, Liu, et al., 2019; Zanos et al., 2016) and exhibit similar rapid elimination profiles (Moaddel et al., 2015; Zanos et al., 2016). However, little was

28 previously known about the pharmacokinetic profile of other HNK stereoisomers. It is possible that other HNKs may more effectively exert antidepressant-relevant actions or have more favorable pharmacological profiles than either (2R,6R)- or (2S,6S)-HNK. It is essential to understand whether HNK stereoisomers other than (2R,6R)- or (2S,6S)-HNK also cross the blood brain barrier prior to evaluating their behavioral effects. Here, the pharmacokinetic profiles of the 12 HNKs formed in vivo were evaluated in the plasma and brains of male and female mice.

2.2. Methods

Animals

Male and female CD-1 mice (n=3-4/group; Charles Rivers Laboratories; Raleigh,

NC, USA), 8-10 weeks old at the time of testing, were habituated to the University of

Maryland (Baltimore, MD, USA) animal facility for at least one week prior to testing. Mice were group housed with 4-5 per cage with a constant 12-hour light cycle (lights on/off at

07:00/19:00). Food and water were available ad libitum. All experiments were performed during the light phase. All studies were approved by the University of Maryland School of

Medicine Institutional Animal Care and Use Committee and conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Drugs

(2R,6R)-hydroxynorketamine (HNK) hydrochloride, (2S,6S)-HNK hydrochloride,

(2R,6S)-HNK, (2S,6R)-HNK, (2R,5R)-HNK, (2S,5S)-HNK, (2R,5S)-HNK, (2S,5R)-HNK,

(2R,4R)-HNK, (2S,4S)-HNK, (2R,4S)-HNK, (2S,4R)-HNK hydrochloride were

29 synthesized and characterized (molecular structures were confirmed via 1H and 13C NMR, the absolute and relative stereochemistry was assigned by x-ray crystallography, and the purity was determined via LC/MS) at the National Center for Advancing Translation

Sciences (National Institutes of Health, Rockville, MD, USA) as previously described

(Morris et al., 2017b). All compounds were dissolved in 20% (w/v) cyclodextrin (Sigma-

Aldrich, USA) in saline. All compounds were administered at a dose of 5 mg/kg (free base dose for compounds provided as a hydrochloride salt—(2R,6R)-HNK, (2S,6S)-HNK, and

(2S,4R)-HNK—is equivalent to 4.31 mg/kg) in a volume of 7.5 ml/kg.

Dosing and sample collection

Compounds were administered intraperitoneally (i.p.) in separate cohorts of mice

(n=4 per sex, compound, and collection time point). At 10 min (0.167 h), 30 min (0.5 h), and 1, 2, and 4 h post-treatment, mice were deeply anesthetized (4% isoflurane for approximately 2 min). Trunk blood was collected into 1.5-ml polypropylene tubes containing 30 µl of disodium EDTA (0.5 M, pH 8.0) and kept on ice until plasma collection

(<30 minutes). Brains were then removed, flash-frozen in isopentane, and stored at -80°C until analysis. Blood was centrifuged at 8000 x g for 6 min at 4°C to obtain plasma. Plasma was collected into clean tubes and stored at -80°C until analysis.

UPLC-MS/MS analysis of plasma and brain samples

The ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-

MS/MS) analysis of HNKs was performed by an external collaborator, Dr. Amy Wang at the National Center for Advancing Translational Sciences (National Institutes of Health,

Rockville, Maryland 20850, USA). UPLC-MS/MS methods were developed and optimized to determine HNKs concentrations in mouse plasma and brain samples. Mass spectrometric analysis was performed on a Waters Xevo TQ-S triple quadrupole instrument using electrospray ionization in positive mode with the selected reaction monitoring. The separation of HNKs from endogenous components was performed on a Waters Acquity

UPLC with 0.5 mL/min flow rate and gradient elution. The mobile phase A and B were

0.1% formic acid in water and 0.1% formic acid in acetonitrile, respectively. Plasma and brain concentrations of (2R,6R)-, (2S,6S)-, (2R,4S)-, and (2S,4R)-HNK were determined by hydrophilic interaction liquid chromatography-tandem mass spectrometry (HILIC

UPLC-MS/MS) using a BEH amide column (1.7 µ, 2.1 x 50 mm). Plasma and brain concentrations of (2R,6S)-, (2S,6R)-, (2R,5R)- (2S,5S)-, (2R,5S)-, (2S,5R)-, (2R,4R)-, and

(2S,4S)-HNK were determined by reversed phase UPLC-MS/MS based upon a previously described protocol with several modifications (Moaddel et al., 2010; Zanos et al., 2016) using a BEH C18 column (1.7 µM, 2.1 x 50 mm). The gradient for HILIC UPLC method was: 0–0.2 min 98% B; 0.2-1.4 min 98% to 35% B; 1.4–1.8 min 35% B to 10%B, 1.9-2.4 min 98%B. The gradient for BEH C18 UPLC method was: 0–0.2 min 1% B; 0.2-1.4 min

1% to 35% B; 1.4–1.8 min 35% to 98%B, 1.8-2.2 min 98%B, 2.3 min 1%B. The analytical run time of the UPLC methods was 2.5 min. The calibration standards and quality control samples were prepared in blank mouse plasma and brain homogenate. Calibration standards ranged from 5.0-5000 ng/ml. A linear regression with 1/x2 weighting was used to construct the calibration curve. Quality control samples were prepared at 10.0 ng/ml,

100 ng/ml, and 2500 ng/ml in the corresponding matrix. Brains were weighed and homogenized in 3 volumes of water. d4-HNK was used as internal standard. 200 µL of 50

31 ng/ml d4-HNK solution in acetonitrile was used to precipitate proteins. The supernatant was transferred for UPLC-MS/MS analysis. The data were acquired using MassLynx and analyzed using TargetLynx version 4.1. The lower limit of quantification (LLOQ) was 5.0 mg/ml.

Pharmacokinetic analysis

Pharmacokinetic analyses were performed by collaborator Dr. Amy Wang. The pharmacokinetic parameters of (2R,6R)- (2S,6S)-, (2R,6S)-, (2S,6R)-, (2R,5R)- (2S,5S)-,

(2R,5S)-, (2S,5R)-, (2R,4R)- (2S,4S)-, (2R,4S)-, (2S,4R)-HNK were calculated using non- compartmental analysis (Model 200) in the pharmacokinetic software Phoenix WinNonlin

(version 7.0, Certara, St. Louis, MO). The area under the plasma and tissue concentration versus time curve (AUC) was calculated using the linear trapezoidal method. The slope of the elimination phase was estimated by log linear regression using at least 3 data points and the terminal rate constant (λ) was derived from the slope. AUC0→∞ was calculated as the sum of the AUC0→t (where t is the time of the last measurable concentration).

Statistical analysis

Statistical analyses were performed using GraphPad Prism software (v8; GraphPad

Software, Inc.). All statistical tests were two-tailed, and significance was assigned at p<0.05. Data are presented as the mean ± standard error of the mean (SEM). Sex-dependent differences in concentrations over time were analyzed using a two-way ANOVA with sex and time as factors, followed by Holm–Šídák post-hoc comparison when significance was reached. The experimental design (each mouse provided plasma and brain for a single

32 experimental time point) resulted in a single experimentally derived AUC per compound and sex. Thus, to enable the comparison of AUCs between sexes, a previously described bootstrapping method (Shen & Machado, 2017) was utilized to estimate the mean and SEM for each AUC. Briefly, to calculate the mean and SEM of the AUC, N=1000 of 1024 possible combinations were selected to form N time series, and then 1000 AUCs and the corresponding mean and SEM determined for each sex. Sex-dependent differences in

AUCs were determined using a standard t-test.

2.3. Results

Plasma pharmacokinetics of HNKs

All HNKs were detected in the plasma between 10 min and 4 h after their i.p. injection in male and female mice. Peak plasma levels (Cmax) for all HNKs were observed at the earliest time point, i.e. 10 min post-treatment, in both sexes (Figure 2.1 and Table

2.1). Robust differences were observed in the peak plasma concentrations and total plasma concentrations (AUC; Table 2.1). Approximately 6-fold differences were observed between the dose-normalized peak plasma concentrations of HNKs in both sexes, with the highest levels observed for (2S,6S)-HNK in both sexes, and the lowest observed for (2S,6R)-HNK in males and (2S,5S)-HNK in females (Table 2.1).

Approximately 11- and 12-fold differences were observed in the dose-normalized HNK total plasma concentrations among male and female mice, respectively (Table 2.1).

Consistent with the trends noted for peak levels, the highest total plasma concentrations were observed following (2S,6S)-HNK injection in male and female mice, while the lowest total plasma concentrations were observed for (2S,6R)-HNK in males and (2S,5S)-

HNK in females (Table 2.1). Plasma levels decreased rapidly after dosing and plasma elimination half-lives ranged from 0.4-0.8 h in females and from 0.5-0.9 h in males

(Table 2.1).

Figure 2.1. Time course of plasma levels of hydroxynorketamines in mice. Plasma concentrations following intraperitoneal administration of (A) (2R,6R)-hydroxynorketamine (HNK; 4.3 mg/kg free base dose), (B) (2S,6S)-HNK (4.3 mg/kg free base dose), (C) (2R,6S)-HNK (5 mg/kg), (D) (2S,6R)-HNK (5 mg/kg), (E) (2R,5R)-HNK (5 mg/kg), (F) (2S,5S)-HNK (5 mg/kg), (G) (2R,5S)-HNK (5 mg/kg), (H) (2S,5R)- HNK (5mg/kg), (I) (2R,4R)-HNK (5 mg/kg), (J) (2S,4S)-HNK (5 mg/kg), (K) (2R,4S)-HNK (5 mg/kg), (L) (2S,4R)-HNK (4.3 mg/kg free base dose) to male (M; solid lines) and female (F; dashed lines) mice. Inset area-under-the-curve (AUC) of the plots of concentration vs. time. Data points and error bars represent mean and SEM, respectively, of results obtained from 3-4 mice/group, as described in the footnote of Table 2.2. *p<0.05, **p<0.01, ***p<0.001.

Table 2.1. Hydroxynorketamine pharmacokinetics in male and female mice.

Abbreviations: AUC/Dose, area under the concentration vs. time curve normalized to free base dose; B:P ratio; brain to plasma AUC ratio; Cmax/Dose, maximum observed concentration normalized to free base dose; F, female; hr, hours; HNK, hydroxynorketamine; Tmax, time of observed maximal concentrations; M, male; t1/2, terminal half-life.

For most HNKs, peak and total plasma levels were sex dependent (Figure 2.1 and summarized in Table 2.2). Female mice had greater peak and total plasma concentrations of (2R,6R)-HNK (Figure 2.1A), (2S,6S)-HNK (Figure 2.1B), (2R,5S)-HNK (Figure 2.1G), and (2S,4S)-HNK (Figure 2.1J), compared to males that received the same compound at the same dose. Following administration of (2S,6R)-HNK, total plasma concentrations were greater in females compared to males, although differences in plasma concentrations at fixed time points did not reach statistical significance (Figure 2.1D). Conversely, following administration of (2S,4R)-HNK, plasma concentrations were greater in females than males at 10 min post-injection, but total exposures were not different between the

35 sexes (Figure 2.1L). While plasma concentrations were higher during the first 10 minutes following (2R,4R)-HNK treatment in females than males, total plasma concentrations were greater in males (Figure 2.1I). Compared to females, male mice had greater peak and total plasma levels of (2R,6S)-HNK (Figure 2.1C), (2R,5R)-HNK (Figure 2.1E), (2S,5R)-HNK

(Figure H), and (2R,4S)-HNK (Figure 2.1K). No sex-dependent differences in peak or total plasma concentrations were detected following treatment with (2S,5S)-HNK (Figure 2.1F).

Table 2.2. Sex-dependent differences in the levels of and exposures to hydroxynorketamines.

Table 2.2. Continued

Table 2.2. Note: When a significant interaction (sex x time) detected in the two-way ANOVA, the Holm- Šídák post-hoc comparison test was performed to compare the effect of sex at each time point. All experimental groups had n=4 (number per compound, sex, and time), with the following exceptions, where n=3: (2S,6S)-HNK male plasma at 2 and 4 h and brain at 4 h; (2S,6R)-HNK female plasma at 2 h; and (2S,4R)- HNK male brain at 1 h. For comparison of AUCs, n=1000 (see Methods section for bootstrapping approach). Abbreviations: AUC, area under the concentration vs. time curve; F, female; HNK, hydroxynorketamine; M, male, NS, not statistically different.

Total and peak brain concentrations of HNKs

All 12 HNKs readily penetrated the brains of mice, where they were detected within the first 10 min following i.p. dosing (Figure 2.2). Peak brain levels were observed at the earliest sampling time point (10 min post-injection) in male and female mice for all HNKs, except for (2R,5S)-HNK in females, (2S,5R)-HNK in males, and (2R,4R)-HNK in males, where peak levels were observed at 30 min post-injection (Figure 2.2). Similar to the plasma, robust differences in the peak and total brain concentrations were observed (see

Table 2.1). Namely, approximately 7- and 8-fold differences were observed in the dose- normalized peak brain concentrations in male and female mice, respectively (Table 2.1).

The highest brain levels (normalized to dose) were observed following (2S,6S)-HNK dosing in females and (2R,4S)-HNK in males, while the lowest were observed following

(2R,4S)-HNK in males and (2S,5S)-HNK in females (Table 2.1). Approximately 9- and 12- fold differences were observed in the dose-normalized total brain concentrations of HNKs among male and female mice, respectively (Table 2.1). Males and females exhibited the highest total brain concentrations of (2R,4R)-HNK and (2S,6S)-HNK, respectively, and the lowest total brain concentrations of (2S,6R)-HNK and (2S,5S)-HNK, respectively. The ratios of brain to plasma total concentrations (AUC) of HNKs ranged between 0.6-1.3 for males and 0.7-1.4 for females (Table 2.1 and Figure 2.3). Brain levels decreased rapidly, with elimination half-lives ranging between 0.3 and 0.8 h in both sexes (Table 2.1). Most

HNKs could be detected in the brain up to 4 h post-treatment, with the following

41 exceptions: (2S,5S)- and (2S,4R)-HNK were below LLOQs in both sexes, while (2R,6S)- and (2R,4S)-HNK were below LLOQs in females, but not males, at 4 h post-treatment.

Figure 2.2. Brain levels of and exposure to hydroxynorketamines. Plasma concentrations following intraperitoneal administration of (A) (2R,6R)-hydroxynorketamine (HNK; 4.3 mg/kg free base dose), (B) (2S,6S)-HNK (4.3 mg/kg free base dose), (C) (2R,6S)-HNK (5 mg/kg), (D) (2S,6R)-HNK (5 mg/kg), (E) (2R,5R)-HNK (5 mg/kg), (F) (2S,5S)-HNK (5 mg/kg), (G) (2R,5S)-HNK (5 mg/kg), (H) (2S,5R)-HNK (5mg/kg), (I) (2R,4R)-HNK (5 mg/kg), (J) (2S,4S)-HNK (5 mg/kg), (K) (2R,4S)-HNK (5 mg/kg), (L) (2S,4R)- HNK (4.3 mg/kg free base dose) to male (M; solid lines) and female (F; dashed lines) mice. Inset area under the concentration vs. time curve (AUC). Data are the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001.

Peak and total concentrations of HNKs in the brain were also sex dependent (Figure

2.2 and summarized in Table 2.2). Female mice had greater peak and total brain concentrations of (2R,6R)-HNK (Figure 2.2A), (2S,6S)-HNK (Figure 2.2B), (2S,6R)-HNK

(Figure 2.2D), (2R,5S)-HNK (Figure 2.2G), and (2S,4S)-HNK (Figure 2.2J), compared to

42 males. Although (2S,4R)-HNK resulted in higher brain concentrations in female mice during the first 10 min post-administration, no differences were observed in its total total brain concentrations in between male and female mice (Figure 2.2L).

Consistent with its plasma profile (Figure 2.1I), (2R,4R)-HNK dosing resulted in higher brain levels in female than male mice during the first 10 min post-treatment, but total brain concentrations were higher in male mice (Figure 2.2I). By contrast, male mice had greater peak and total brain concentrations of (2R,4S)-HNK than females (Figure

2.2K). Following systemic administration, , total brain concentrations of (2R,6S)-HNK

(Figure 2.2C) and (2S,5R)-HNK (Figure 2.2H) were greater in male mice relative to females, although no statistically significant sex-dependent differences in brain concentrations over time were observed. No sex-dependent differences were observed in peak or total brain concentrations of (2R,5R)- and (2S,5S)-HNK (Figure 2.2E and F, respectively).

2.4. Discussion

Following a single i.p. injection, all 12 HNKs were detected in the plasma of male and female mice between 10 min and 4 h post-treatment (Figure 2.1). Peak plasma levels were observed at the earliest time point tested, 10 min post-treatment, for all HNKs. All

HNKs readily penetrated the brain, where they were detected at the earliest timepoint test,

10 min post-injection (Figure 2.2). The peak brain levels occurred within 10 min of dosing for all HNKs except for (2R,4R)-HNK and (2S,5R)-HNK in males and (2R,5S)-HNK in females, for which peak brain levels were observed at 30 min post-treatment (Table 2.1 and Figure 2.2). Although brain to plasma ratios demonstrated approximately two-fold

43 differences between the individual HNK stereoisomers, all were between 0.6-1.4 (Table

2.1), suggesting good brain penetrance, with no accumulation, of all HNKs. Robust differences were observed in the peak and total concentrations of the HNKs the HNKs in both the plasma (Cmax, ~6-fold differences; AUC, ~11-12-fold differences) and the brain

(Cmax, 7-8-fold differences; AUC, 9-12-fold differences; Table 2.1). Elimination from the plasma and brain was rapid and all half-lives were determined to be shorter than one hour, ranging from 0.4-0.9 h in the plasma and 0.3-0.8 h in the brain (Table 2.1). Most HNKs were quantifiable in the brain up to 4 h post-treatment (during the time frame that they were present in the plasma), with few exceptions. Namely, brain levels were below the

LLOQ (5 ng/ml) by 4 h post-treatment for (2S,5S)- and (2S,4R)-HNK in both sexes and for

(2R,6S)- and (2R,4S)-HNK in female mice.

Of note, sex-dependent differences were observed in the pharmacokinetic profiles of most HNKs. In both the plasma and the brain, total concentrations of (2R,6R)-, (2S,6S)-

, (2S,6R)-, (2R,5S)-, and (2S,4S)-HNK were greater in female mice, whereas total concentrations of (2R,6S)-, (2S,5R)-, (2R,4R)-, and (2R,4S)-HNK were greater in males.

Total plasma concentrations of (2R,5R)-HNK were also greater in male mice compared to females, but there were no sex-dependent differences in total brain concentrations of this

HNK.

These data provide the first in-depth characterization of the pharmacokinetics of all

12 HNKs that can be formed in vivo from ketamine following their direct administration to male and female mice. Overall, the data demonstrate that although the 12 HNKs exhibit robust differences in peak and total concentrations in the plasma and brains of mice, both rapid elimination and capacity to penetrate the brain are well conserved. Notably, the

44 ability of the 12 HNKs to rapidly penetrate the brain following systemic dosing suggest that these compounds would be capable of exerting pharmacodynamic effects in the brain, including antidepressant-relevant effects similar to those which have previously been observed for (2R,6R)- and (2S,6S)-HNK (Chen et al., 2020; Chou et al., 2018; Elmer et al.,

2020; Fukumoto et al., 2019; Lumsden et al., 2019; Pham et al., 2018; Yokoyama et al.,

2020; Zanos, Highland, Liu, et al., 2019; Zanos, Highland, Stewart, et al., 2019; Zanos et al., 2016). Additionally, most HNKs demonstrated sex-dependent difference in their pharmacokinetic profiles, an important consideration for future behavioral studies.

Chapter 3. Oral bioavailability and antidepressant-relevant behavioral effects of (2R,6R)-hydroxynorketamine 1

3.1. Introduction

Ketamine has gained attention for its rapid-acting antidepressant actions in human clinical trials (Berman et al., 2000; DiazGranados et al., 2010; Price et al., 2014; Zarate et al., 2006). Specifically, in treatment-resistant depressed patients, a single infusion of ketamine reverses depressive symptoms within hours, with effects lasting up to one week.

While its antidepressant efficacy is superior, in terms of rate of onset and effectiveness in previously treatment-resistance patients, to that of classical antidepressants—which typically require weeks to take effect and are ineffective in a large number of patients

(Gaynes et al., 2009; Insel & Wang, 2009)—the widespread use of ketamine as an antidepressant is limited by its dissociative side effects and abuse potential (Krystal et al.,

1994). Moreover, due to its poor oral bioavailability (approximately 17-24% in humans)

(Chong et al., 2009; Clements et al., 1982; Grant et al., 1981; Yanagihara et al., 2003), ketamine is most frequently administered via intravenous (i.v.) infusion, making outpatient treatment challenging.

Ketamine is rapidly and stereoselectively metabolized into a number of metabolites, including (R,S)-norketamine, (R,S)-dehydronorketamine, hydroxyketamines, and hydroxynorketamines (HNKs) (Adams et al., 1981; Chang & Glazko, 1974; Moaddel et al., 2010; Woolf & Adams, 1987). A number of HNK metabolites have been described,

1 Originally published in Highland et al. “Mouse, rat, and dog bioavailability and mouse oral antidepressant efficacy of (2R,6R)-hydroxynorketamine“ Journal of Psychopharmacology Highland, J. N., Morris, P. J., Zanos, P., Lovett, J., Ghosh, S., Wang, A. Q., Zarate, C. A., Jr., Thomas, C. J., Moaddel, R., & Gould, T. D. (2018). Mouse, rat, and dog bioavailability and mouse oral antidepressant efficacy of ( 2R,6R)- hydroxynorketamine. J Psychopharmacol, 269881118812095. https://doi.org/10.1177/0269881118812095 and reproduced with journal permission. 46 including the (2,4)-, (2,5)-, and (2,6)-HNKs, (Desta et al., 2012; Moaddel et al., 2010;

Woolf & Adams, 1987), with (2R,6R;2S,6S)-HNK being the most prevalent in the plasma of humans (Zarate et al., 2012) and in the plasma and brain of rodents (Can et al., 2016;

Moaddel et al., 2016; Zanos et al., 2016) following ketamine administration. Although early studies classified (2R,6R;2S,6S)-HNK as an inactive metabolite due to its lack of anesthetic effects (Leung & Baillie, 1986), more recent studies have established the biological activity of this metabolite (Cavalleri et al., 2018; Kroin et al., 2019; Moaddel et al., 2013; Paul et al., 2014; Pham et al., 2018; Singh et al., 2016; Wray et al., 2019; Yao et al., 2018; Zanos et al., 2016). Notably, it has been demonstrated that, in mice, the ketamine metabolite (2R,6R)-HNK exerts rapid and sustained antidepressant actions (Pham et al.,

2018; Zanos et al., 2016), and lacks the side effects and abuse potential associated with ketamine (Zanos et al., 2016).

We evaluated the exposure to (2R,6R)-HNK in the plasma and brains of mice following oral (p.o.) administration and compared it to that of intraperitoneal (i.p.) and i.v. administration, the routes of administration most commonly utilized when studying the antidepressant actions of ketamine in rodents and humans, respectively. To determine if a prodrug strategy could improve its oral bioavailability, we evaluated the exposure to

(2R,6R)-HNK in plasma following administration of five prodrug candidates. We confirmed that (2R,6R)-HNK did not exert overt adverse effects when administered orally and evaluated the oral efficacy of (2R,6R)-HNK in two assays predictive of antidepressant efficacy in mice, the forced-swim and learned helplessness tests.

3.2. Methods

Animals

Male and female CD-1 mice (Charles River, Raleigh, NC, USA), 9-12 weeks old, were used for pharmacokinetic studies of (2R,6R)-HNK and prodrug candidates, and for all behavioral testing. Mice were acclimated to the University of Maryland Baltimore animal facility (Baltimore, MD, USA) for at least one week prior to testing. Mice were group housed in cages of 4-5 per cage with a constant 12-h light cycle (lights on/off at

07:00/19:00). Food and water were available ad libitum. All mouse pharmacokinetic and behavioral tests were performed during the light phase at the University of Maryland

Baltimore, were approved by the University of Maryland Baltimore Animal Care and Use

Committee and were conducted in full accordance with the National Institutes of Health

Guide for the Care and Use of Laboratory Animals. All mouse tests of antidepressant efficacy were performed by a male experimenter.

Drugs

(2R,6R)-HNK HCl (1) was synthesized and characterized as described (Morris et al., 2017; Zanos et al., 2016). Prodrug esters (acetate (2a), propionate (2b), pentanoate (2c), octanoate (2d) and benzoate (2e) esters) of (2R,6R)-HNK, substituted at the hydroxyl group, were synthesized and characterized (see supplementary information for Highland et al., 2018) at the National Center for Advancing Translational Sciences (National Institutes of Health, Rockville, Maryland 20850, USA). Ketamine hydrochloride (HCl) was purchased from Sigma-Aldrich (USA). All drugs were dissolved in 0.9% saline.

In an effort to improve the oral bioavailability of (2R,6R)-HNK, five prodrug candidates were synthesized by introducing a series of ester modifications onto the free hydroxyl group of (2R,6R)-HNK (see Figure 3.2). It was hypothesized that increasing the lipophilicity of (2R,6R)-HNK via these modifications may increase its oral absorption (see

Gao, 2017). Once absorbed, the prodrug is converted to the parent drug by circulating esterases (reviewed in Hamada, 2017; Zawilska et al., 2013).

Dosing and sample collection

(2R,6R)-HNK was administered i.p., p.o. or i.v. (via the tail vein). In one experimental cohort, 50 mg/kg of (2R,6R)-HNK HCl (43.1 mg/kg free base) was administered either p.o. or i.p. to male mice. In separate experiments, the same dose was administered either p.o. or i.v. to male or female mice to determine absolute bioavailability.

Prodrug compounds were administered at the dose of 50 mg/kg (HCl salt; free base concentration 44-45 mg/kg), i.p. and p.o., in separate cohorts of male mice. All compounds were administered in a volume of 7.5 ml/kg for i.p. and p.o. administration, and in a volume of 4.0 ml/kg for i.v. injection. Prior to p.o. administration, mice were briefly anesthetized

(<2 min) under 3% isoflurane. For i.v. administration, mice were briefly restrained (<2 min) using a ventilated restraint tube and the tail was gently warmed to encourage vasodilation.

Following (2R,6R)-HNK administration, samples were collected at 0.167, 0.5, 1, 2, and 4 h after drug administration. For all prodrugs, samples were collected at 0.167, 0.5, and 2 h after administration. Mice were deeply anesthetized under 3-4% isoflurane (placed into induction chamber for approximately 2 min) before being decapitated. Trunk blood

49 and whole brains were immediately collected. Blood was collected into 1.5-ml Eppendorf tubes containing 30 μl disodium EDTA (0.5 M, pH 8.0) and kept on ice until plasma collection (<30 min). Blood was then centrifuged at 8000 x g for 6 min at 4°C to obtain plasma. Plasma was collected into clean Eppendorf tubes and stored at -80°C until analysis.

Brains were immediately frozen on dry ice and stored at -80°C until analysis.

LC/MS/MS analysis of plasma and brain samples

The concentrations of (2R,6R)-HNK in plasma and brain were determined by achiral liquid chromatography-tandem mass spectrometry (LC-MS/MS) using a previously described protocol with slight modifications (Moaddel et al., 2015; Moaddel et al., 2010;

Paul et al., 2014). LC-MS/MS analyses were performed by an external collaborator, Dr.

Ruin Moaddel at the National Institute on Aging (National Institutes of Health Intramural

Research Program, Baltimore, MD, USA). The determination of (2R,6R)-HNK was accomplished using an Eclipse XDB-C18 guard column (4.6 mm x 12.5 mm) and Varian

Pursuit XRs 5 C18 analytical column (250 x 4.0 mm ID, 5 um). The mobile phase consisted of ammonium acetate (5 mM, pH 7.6) as Component A and acetonitrile as Component B.

A linear gradient was run as follows: 0 min 20% B; 5 min 20% B; 15 min 90% B; 20 min

20% B at a constant flow rate of 0.4 ml/min. The total run time was 30 min per sample.

For plasma samples, the calibration standard for (2R,6R)-HNK ranged from 39.1 to

20,000 ng/ml for (2R,6R)-HNK and 39.1 ng/ml to 5,000 ng/ml for the prodrugs. The quantification of (2R,6R)-HNK was accomplished by calculating area ratios using d4- ketamine (10 μl of 50 μg/ml solution) as the internal standard. Brains were weighed and suspended in 990 μl of water:methanol (3:2, v/v), d4-ketamine (10 μl of 10 μg/ml) was

50 added and the resulting mixture was homogenized on ice with a polytron homogenizer and centrifuged at 21,000 x g for 30 min. The supernatant was collected and processed using

1-ml Oasis HLB solid phase extraction cartridges (Waters Corp., Waltham, MA). The cartridges were preconditioned with 1 ml of methanol, followed by 1 ml of water and then

1 ml ammonium acetate (10 mM, pH 9.5). The supernatants were added to the cartridges, followed by 1 ml of water and the compounds were eluted with 1 ml of methanol. The eluent was transferred to an autosampler vial for analysis. Quality control standards were prepared at 78.1 ng/ml, 625 ng/ml and 2500 ng/ml spiked in the corresponding matrix.

MS/MS analysis was performed using a triple quadrupole mass spectrometer model

API 4000 system from Applied Biosystems/MDS Sciex equipped with Turbo Ion Spray®

(TIS) (Applied Biosystems, Foster City, CA, USA). The data was acquired and analyzed using Analyst version 1.4.2 (Applied Biosystems). Positive electrospray ionization data were acquired using multiple reaction monitoring (MRM; see supplementary information for Highland et al., 2018). The TIS instrumental source settings for temperature, curtain gas, ion source gas 1 (nebulizer), ion source gas 2 (turbo ion spray) and ion spray voltage were 600°C, 25 psi, 60 psi, 60 psi and 5500 V, respectively.

Bioanalysis of (2R,6R)-HNK

The bioanalysis of (2R,6R)-HNK and candidate prodrugs was carried out for three time points (0.167, 0.5 and 2 h). The area under the plasma concentration versus time curve

(AUC) was calculated using Phoenix WinNonlin (version 7.0, Certara, St. Louis, MO).

AUClast was calculated as AUC0→t (where t is the time of the last measurable concentration). Bioanalysis was performed by Dr. Amy Wang at the National Center for

Advancing Translational Sciences (National Institutes of Health, Rockville, Maryland

20850, USA). Relative oral bioavailability was calculated as the ratio of the plasma AUClast of p.o. to that of i.p.

The pharmacokinetic parameters of (2R,6R)-HNK (5 time points) were calculated using non-compartmental analysis (Model 200) in the pharmacokinetic software Phoenix

WinNonlin (version 7.0, Certara, St. Louis, MO). Pharmacokinetic analyses were performed by Dr. Amy Wang. The area under the plasma and tissue concentration versus time curve (AUC) was calculated using the linear trapezoidal method. The slope of the apparent terminal phase was estimated by log linear regression using at least 3 data points and the terminal rate constant (λ) was derived from the slope. AUC0→∞ was calculated as the sum of the AUC0→t (where t is the time of the last measurable concentration).

Relative bioavailability was calculated as the ratio of the plasma AUC0→∞ of p.o. to that of i.p. administration. Absolute bioavailability was calculated as the ratio of the plasma

AUC0→∞ of p.o. to that of i.v. administration. The apparent terminal half-life (t½) was calculated as 0.693/λ.

Behavioral studies

For all behavioral studies, (2R,6R)-HNK was administered i.p. or p.o. (anesthesia was not utilized for behavioral studies). Ketamine was administered i.p.

Open-field test

Mice were placed into an open field arena (50 cm x 50 cm x 38 cm; length x width x height) for a 60-min habituation period. Mice then received saline or (2R,6R)-HNK (15,

50, 150, and 450 mg/kg, p.o.) and open-field locomotor activity was assessed for an additional 60 min. Distance travelled and time spent in the central 50% of the arena were analyzed using TopScan software (v2.0; CleverSys, Inc.).

Adverse effects screening

Mice received a single dose of (2R,6R)-HNK (450 mg/kg, p.o.) and were immediately placed into a clear Plexiglass arena (40 cm x 40 cm x 40 cm) and observed at pre-defined intervals over 60 min. Specifically, each mouse was observed for 30 s every 2- min for the first 20 min following injection, and then 30s every 10 min from 30-60 min post-injection. The occurrence of the following behaviors, which are associated with discomfort, sickness, and/or stereotypy were scored during each 30-sec observation period

(adapted from Kelley, 2001): tail flicking (score 0 (absent) or 1 (present)), eye ptosis (i.e. orbital tightening; score 0 (absent) or 1 (present)), piloerection (hair erected; score 0

(absent) or 1 (present)), sniffing (defined as sniffing of the ground of greater intensity than baseline activity; score number bouts), backwards movement (score number bouts), rearing

(score number bouts) and grooming (score number bouts). The total score for each behavior was calculated per animal.

Forced-swim test

Each mouse received an oral gavage (p.o.) immediately followed by a systemic

(i.p.) injection. The following treatment groups were included: p.o. saline and i.p. saline; p.o. saline and i.p. ketamine (15 mg/kg); p.o. (2R,6R)-HNK (15, 50, and 150 mg/kg) and i.p. saline. The forced-swim test was performed as previously described (Zanos et al.,

2016). Briefly, mice were placed into a clear Plexiglass cylinder (20 cm height x 15 cm diameter) filled to a depth of 15 cm with water (23 ± 1°C) and subjected to a 6-min swim session. Mice were tested in an initial session one hour after treatment and were re-tested

24 hours after treatment using the same procedure. Swim sessions were recorded using a digital video camera. The time spent immobile (defined as passive floating with no movements other than those necessary to keep the head above water) was scored during the final 4 min of the session. Sessions were scored by a trained experimenter blind to the treatment groups.

Learned helplessness test

The learned helplessness test was performed as previously described (Zanos et al.,

2016). In brief, mice were placed into one side of a two-chambered shuttle box (Coulbourn instruments) and administered inescapable shocks (day 1; 120 shocks, 0.45 mA). Twenty- four hours later (day 2), mice were screened for “helplessness,” wherein mice were placed into one side of the shuttle box and administered a series of escapable shock trials (i.e. the shuttle door opened with the shock to allow escape to the other side; 30 trials, 0.45 mA).

Helpless mice were defined as those with at least 20 total escape failures and at least 5 escape failures in the last 10 trials. Helpless mice were assigned to treatment groups such that there were no baseline differences in the number of escape failures during screening

(p.o. saline and i.p. saline, 25.55±1.02; p.o. saline and i.p. (2R,6R)-HNK, 25.36±0.97; p.o.

(2R,6R)-HNK (50 mg/kg) and i.p. saline, 25.91±1.18; p.o. (2R,6R)-HNK (150 mg/kg) and i.p. 25.90±1.03; data reported as mean ± SEM). On day 3, all helpless mice received an oral gavage (p.o.) immediately followed by an i.p. injection. The following groups were

54 included: p.o. saline and i.p. saline; p.o. saline and i.p. (2R,6R)-HNK (50 mg/kg); and p.o.

(2R,6R)-HNK (50 and 150 mg/kg) and i.p. saline. Twenty-four hours after treatment (day

4), mice underwent escapable shock testing (45 trials, 0.45 mA) to assess reversal of helplessness. During the testing phase, the door opened simultaneously with shock onset for the first 5 trials and with a 2 sec delay for the remaining 40 trials. The number of escape failures during these last 35 trials and the final 10 trials were used to assess helplessness.

Escape failures were recorded using GraphicState software (v3.1; Coulbourn instruments).

Statistical analysis

All behavioral tests were performed with the experimenter blind to treatment groups and mice were randomly assigned to treatment groups. Statistical analyses were performed using GraphPad Prism software (v6; GraphPad Software, Inc.). All statistical tests were two-tailed, and significance was assigned at p<0.05. When three or more groups were included (open-field test, learned helplessness test, and forced-swim test) one-way

ANOVA was performed followed by Holm–Šídák post-hoc comparison (with all groups compared to saline control group) when significance was reached. When two groups were included (i.e. adverse effects screening outcomes) an unpaired Student’s t-test was used.

Significant results are indicated with asterisks in the figures (*p<0.05, **p<0.01,

***p<0.001).

3.3. Results

Bioavailability and pharmacokinetic profile of (2R,6R)-HNK in mice

(2R,6R)-HNK (50 mg/kg HCl salt) was administered via three different routes (i.p., p.o., and i.v.) to evaluate its bioavailability and pharmacokinetic profile in the plasma and 55 brain. Samples were collected at five time points, ranging from 10 min (0.167 h) to 4 h following administration and concentrations of (2R,6R)-HNK were quantitated via LC-

MS/MS. For all routes of administration, peak concentrations in the plasma and brain were achieved within 10 min of administration (Tmax = 0.167 h; see Figure 3.1). The half-life

(t1/2) was between 0.2 and 0.8 h in the plasma and between 0.5 and 0.7 h in the brain.

Maximum concentrations (Cmax), area under the curve (AUC), bioavailability (BA), and brain penetrance (brain to plasma ratio) are summarized in Tables 3.1 and 3.2.

Figure 3.1. Plasma and brain concentrations of (2R,6R)-HNK in mice.

(A-C) Plasma concentrations of (2R,6R)-HNK following (A) p.o. or i.p. administration of a single 50 mg/kg dose to male mice, (B) p.o. or i.v. administration of 50 mg/kg to male mice, and (C) p.o. or i.v. administration of 50 mg/kg to female mice. (D-F) Brain concentrations corresponding to dosing parameters described in A- C, respectively. Inset: AUC. Data are mean ± SEM. n=3-4/group. Abbreviations: p.o., oral; i.p., intraperitoneal; i.v., intravenous; AUC, area under the curve.

Table 3.1. Distribution of (2R,6R)-HNK in the plasma and brain of mice.

Note: Parameters calculated from five sampling time points following a single dose of 50 mg/kg (43.1 mg/kg a free base). Concentrations measured 10 min after i.v. dosing; values not representative of actual peak. Abbreviations: AUC, area under the curve; Cmax, maximum concentration; ND, not determined; t1/2, half-life.

Table 3.2. Bioavailability and brain penetrance of (2R,6R)-HNK in mice.

Note: Parameters calculated from five sampling time points following a single dose of 50 mg/kg (43.1 mg/kg free base). Abbreviations: BA, bioavailability; ND, not determined.

Following p.o. or i.p. administration of 50 mg/kg of (2R,6R)-HNK to male mice,

(Figure 3.1A and 3.1D and see Tables 3.1-3.2), the relative oral BA was 62%. Following p.o. administration, the maximal plasma concentrations (5.90 μg/ml) were 44.1% of that of i.p. administration (13.4 μg/ml). In the brain, the maximal concentrations (6.49 μg/g) were 51% of that of i.p. administration (12.7 μg/g). The ratios of brain to plasma exposure

(brain penetrance) were similar for p.o. (0.90) and i.p. (0.96) administration.

In separate experiments, 50 mg/kg of (2R,6R)-HNK HCl was administered either p.o. or i.v. to male (Figure 3.1B and 3.1E and see Tables 3.1-3.2) and female (Figure 3.1C and 3.1F and see Tables 3.1-3.2) mice to assess its absolute oral bioavailability. The absolute oral bioavailability was 46% in males and 52% in females. At the earliest sampling time point (10 min after dosing) concentrations following p.o. administration were 35% and 42% of those achieved following i.v. administration, in the plasma (5.64 vs 16.0 μg/ml)

57 and brain (8.60 vs 20.4 μg/g), respectively. Brain penetrance was similar for p.o. (1.16) and i.v. (0.86) administration. Following i.v. administration, the clearance of (2R,6R)-HNK was 74 ml/min/kg in males and 51 ml/min/kg in females. The apparent steady-state volume of distribution was estimated to be 1.7 and 0.80 L/kg in male and female mice, respectively.

Relative oral bioavailability of (2R,6R)-HNK prodrug candidates in mice

The relative oral bioavailability of five (2R,6R)-HNK prodrug candidates (see

Figure 3.2) was assessed following the administration of a single 50 mg/kg dose (p.o. and i.p.). Plasma was collected at three time points (0.167, 1, and 2 h) and the concentrations of the prodrug and (2R,6R)-HNK were quantitated (Figure 3.3). To directly compare

(2R,6R)-HNK to the prodrugs, the data obtained for (2R,6R)-HNK was reanalyzed with the same three time points included for the prodrug candidates (Figure 3.3A). The choice of three time points allowed for a preliminary assessment of whether any prodrug had a robust improvement in the oral bioavailability.

Based on the three time points, 2d had the lowest systemic exposure following p.o. administration (AUC 0.782 μg/ml•hr) and 2a had the highest (AUC 2.26 μg/ml•hr), whereas (2R,6R)-HNK itself had an AUC of 2.80 μg/ml•hr. We note that the areas calculated from three time points in our screening procedure should only be interpreted as relative values and cannot be directly compared with values calculated using the greater number of time points. The relative oral bioavailability of the prodrug candidates varied from 10% for 2d to 44% for 2e. When compared at identical time points, the relative bioavailability of 2e was similar to, but did not exceed, that of (2R,6R)-HNK itself (47%).

Figure 3.2. Candidate prodrug modifications. Chemical structures of the parent compound (2R,6R)-HNK (1) and ester prodrug candidates (2a, 2b, 2c, 2d, and 3e).

Figure 3.3. Candidate prodrugs do not improve the oral bioavailability of (2R,6R)-HNK in mice. Plasma concentrations of (2R,6R)-HNK following p.o. (open circles with dashed lines) and systemic (i.p.; filled circles with solid lines) administration of a single dose (50 mg/kg) of (A) (2R,6R)-HNK and the (B) acetate (2a), (C) propionate (2b), (D) pentanoate (2c), (E) octanoate (2d), and (F) benzoate (2e) prodrug esters of (2R,6R)-HNK to male mice. Inset: AUC. Date are the mean ± SEM. n=4/group. Abbreviations: p.o., oral; i.p. intraperitoneal; AUC, area under the curve.

Table 3.3. AUC, Cmax, and relative bioavailability of (2R,6R)-HNK and prodrug candidates in the plasma of mice.

Note: Parameters were calculated from three sampling time points following a dose of 50 mg/kg. Abbreviations: AUC, area under the curve; BA, bioavailability; Cmax, maximum concentration.

Lack of adverse effects of (2R,6R)-HNK in mice

To assess drug effects on locomotion and exploratory behavior, open-field locomotor activity was evaluated following p.o. administration of (2R,6R)-HNK. After a

60-min habituation period in the open-field arena, mice received (2R,6R)-HNK (15, 50,

150, and 450 mg/kg p.o.) or saline (p.o.). The total distance traveled and time spent in the central 50% of the arena were measured over an additional 60 min after administration.

(2R,6R)-HNK did not alter the total distance travelled at any dose tested (p>0.05; Figure

3.4A). Similarly, no significant effect was observed in time spent in the center of the open- field arena (p>0.05; Fig 3.4B).

Figure 3.4. (2R,6R)-HNK does not alter open-field locomotor activity in mice. Oral administration of (2R,6R)-HNK did not alter (A) the total distance travelled (F(4,34)=0.17, p=0.95) or (B) the time spent in the center (central 50%) of the arena (F(4,34)=0.69, p=0.61) at any dose tested (15-450 mg/kg), relative to saline- treated controls in the open-field test. Data are the mean ± SEM. n=7-8/group. Abbreviations: SAL, saline. 60

To evaluate additional possible adverse drug effects at a high oral dose, mice were systematically observed for the occurrence of behaviors associated with pain or discomfort, sickness, and stereotypy for one hour following administration of 450 mg/kg of (2R,6R)-

HNK or saline (see Table 3.4). The only behaviors observed consistently were grooming and rearing. Rearing was observed with comparable frequency in both treatment groups, while grooming scores were significantly lower among mice treated with (2R,6R)-HNK

(p<0.05). Piloerection was observed in two animals but occurred equally among saline- (1 animal) and (2R,6R)-HNK-treated (1 animal) mice, and thus was not attributed to drug effects. Taken together, these observations demonstrate that (2R,6R)-HNK, up to a dose of

450 mg/kg (p.o.), has minimal, if any, adverse effects in mice.

Table 3.4. Adverse effects screening outcomes following oral (2R,6R)-HNK.

Note: Outcomes are following a single oral dose of 450 mg/kg. Data are the mean ± SEM. n=4/group. *p<0.05.

Oral antidepressant efficacy of (2R,6R)-HNK in mice

The antidepressant efficacy of orally administered (2R,6R)-HNK was evaluated in two established mouse models: the forced-swim test and the learned helplessness test.

Injected (i.p.) ketamine and/or (2R,6R)-HNK served as positive controls in both experiments. All mice received an oral gavage immediately followed by an i.p. injection to control for the route of administration.

In the forced-swim test, systemic injections of ketamine (15 mg/kg, i.p.) and

(2R,6R)-HNK (15 mg/kg, i.p.) reduced immobility time in the forced-swim test both 1 hour after administration (p<0.05; Figure 3.5A) and in a 24-hour re-test (p<0.01; Figure 3.5B), consistent with the rapid and sustained antidepressant actions of these compounds in this test, as has been previously described (Autry et al., 2011; Li et al., 2010; Pham et al., 2018;

Won et al., 2011; Zanos et al., 2016). Orally administered (2R,6R)-HNK also exerted rapid

(1 hour; Figure 3.5A) and sustained (24 hours; Figure 3.5B) reductions in immobility time at doses ranging from 15-150 mg/kg (1-h test, p<0.05; 24-h test, p<0.01), establishing the oral efficacy of this compound.

Figure 3.5. Oral (2R,6R)-HNK reduces immobility time in the mouse forced-swim test. Oral (2R,6R)- HNK exerted (A) acute (1 h after treatment, F(5,54)=4.59, p=0.0015) and (B) sustained (24 h after treatment, F(5,54)=3.84, p=0.0048) reductions in immobility time in the forced-swim test, relative to saline-treated controls, at all doses tested (15-150 mg/kg). Data are the mean ± SEM. n=10/group. *p<0.05, **p<0.01, ***p<0.0001. Abbreviations: p.o., oral; SAL, saline; i.p., intraperitoneal.

Following inescapable shock induction of learned helplessness, mice received

(2R,6R)-HNK and were tested for reversal of learned helplessness 24 h later. Consistent with a previous report (Zanos et al., 2016), a systemic injection of (2R,6R)-HNK (50 mg/kg, i.p.) reduced the total number of escape failures (p<0.01; Figure 3.6A and 3.6B) and those in the last 10 trials (p<0.001; Figure 3.6C) relative to saline-treated controls.

Orally administered (2R,6R)-HNK (50 and 150 mg/kg, p.o.) was also effective at reversing learned helplessness, as demonstrated by a significant reduction in the number of escape failures (total failures, p<0.05; last 10 trials, p<0.001 Figure 3.6).

Figure 3.6. Oral (2R,6R)- HNK reverses learned helplessness in mice. Oral (2R,6R)-HNK was effective in reversing learned helplessness, assessed via a reduction in (A) the number of escape failures over time (effect of treatment, F(3,39)=4.20, p=0.011; effect of time, F(8,312)=3.89, p=0.0002; interaction, F(24,312)=2.51, p=0.0002) (B) the total number of escape failures (F(3,39)=4.20, p=0.011), and (C) the escape failures in the last 10 testing trials (F(3,39)=10.12, p<0.001). Data are the mean ± SEM. n=10-11/group. *p<0.05, **p<0.01, ***p<0.001. Abbreviations: p.o., oral; i.p., intraperitoneal; SAL, saline.

3.4. Discussion

It has previously been demonstrated that, following i.p. administration, the ketamine metabolite (2R,6R)-HNK exerts rapid and sustained antidepressant effects in mice (Pham et al., 2018; Zanos et al., 2016), without the side effects or abuse potential associated with the parent compound (Zanos et al., 2016). Here, we show for the first time that (2R,6R)-HNK has absolute oral bioavailability of between 46-52% in mice. Additional studies (data not shown here) revealed that (2R,6R)-HNK has oral bioavailability of 42%

63 in rats and 58% in dogs2. The oral bioavailability of (2R,6R)-HNK is similar to that of the

(2S,6S)-HNK enantiomer previously reported in rats (approximately 46%; Moaddel et al.,

2015). Compared to i.p. administration, (2R,6R)-HNK has a relative oral bioavailability of

62% in male mice. None of the prodrug modifications provided an improvement over the bioavailability for the parent compound. Further, we demonstrate that (2R,6R)-HNK has oral antidepressant efficacy in the mouse forced-swim and learned helplessness tests.

Circulating plasma and brain concentrations of (2R,6R)-HNK were determined in mice following p.o., i.p., or i.v. administration using five sampling time points to compare maximal concentrations and exposure between routes of administration and to determine the oral bioavailability. At a dose of 50 mg/kg in male mice, the highest measured concentrations following oral administration (plasma, 5.64-5.90 μg/ml; brain, 6.49-8.60

μg/g) were approximately 2-3-fold lower than those measured after i.p. (plasma 13.4

μg/ml; brain, 12.7 μg/g) and i.v. (plasma, 16.2 μg/ml; brain, 20.4 μg/g) administration. At the same dose in female mice, the highest observed plasma concentrations following p.o.

(14.7 μg/ml) and i.v. (24.7 μg/ml) administration were higher (1.5-2.6x) than those measured in males. Comparing within the female cohort, however, similar to the trends observed in males, the maximal levels achieved following p.o. administration were nearly two-fold (1.7x) lower than those following i.v. administration. The absolute oral bioavailability was 46% in males and 52% in females. Because i.p. injection is most commonly used in rodent studies of ketamine and (2R,6R)-HNK, we also calculated the relative oral bioavailability, a comparison of systemic exposure following p.o. to that of

2 Rat and dog bioavailability data is published in Highland et al. Ibid.. These studies were designed by external collaborators and completed by contract research organizations; as such, data are not included in this dissertation. 64 i.p. administration. Relative oral bioavailability of (2R,6R)-HNK was 62%. (2R,6R)-HNK readily penetrated the brain, with brain to plasma ratios ranging from 0.67 to 1.2 for both sexes and all routes of administration. Overall, the data demonstrate that (2R,6R)-HNK has favorable oral bioavailability (see Cyriac & James, 2014) in three mammalian species and readily penetrates the brain of rodents.

In an effort to improve its oral bioavailability, a series of prodrugs were synthesized by introducing a variety of ester modifications onto the free hydroxyl group of (2R,6R)-

HNK (see Figure 3.2). A preliminary screen (three sampling time points) for each of the prodrugs was carried out in mice to determine if a meaningful improvement in relative oral bioavailability was observed, compared to (2R,6R)-HNK. Of note, the maximum circulating concentrations of (2R,6R)-HNK were increased following i.p. administration of

2a compared to (2R,6R)-HNK itself (17.5 μg/ml and 13.4 μg/ml, respectively). Similarly, systemic exposure was increased following i.p. administration of 2a (AUC 7.96 μg/ml•hr),

2d (AUC 7.59 μg/ml•hr), and 2c (AUC 7.18 μg/ml•hr), compared to (2R,6R)-HNK itself

(AUC 5.91 μg/ml•hr). However, for p.o. administration, maximal concentrations and systemic exposures were decreased with increasing acyl chain length for the prodrug candidates relative to (2R,6R)-HNK, with 2a having the highest concentrations and systemic exposure (4.18 μg/ml and 2.26 μg/ml•hr, respectively) and 2d having the lowest

(1.23 μg/ml and 0.782 μg/ml•hr, respectively) (Table 3.3). Replacing the acyl chain with a benzoyl group (2e) resulted in a slight increase in systemic exposure (2.40 μg/ml•hr) following p.o. administration, compared to the other prodrug candidates. As a result, 2e had the highest relative oral bioavailability (44%) of all the prodrugs tested (see Table 3.3), which was similar to that of the parent drug (2R,6R)-HNK (47% using identical time

65 points). Notably, while ester cleavage in vivo was exceedingly rapid to reveal the parent drug, none of the ester prodrug candidates tested improved the oral bioavailability of

(2R,6R)-HNK. Upon retrospective analysis, (2R,6R)-HNK itself exhibited relatively high oral bioavailability (absolute bioavailability 42-52%) and good oral antidepressant efficacy in mice, which may enable the use of the parent drug itself as an oral treatment.

Our data demonstrate that (2R,6R)-HNK has antidepressant-like effectiveness in the mouse forced swim and learned helplessness tests following oral administration.

Although maximal plasma and brain concentrations and exposure following oral administration of (2R,6R)-HNK were lower than those achieved following i.p. injection, equivalent doses administered via each route produced similar antidepressant effects in the forced swim test (15 mg/kg) and the learned helplessness test (50 mg/kg). While the minimal effective doses (not determined in the present study) may vary between routes of administration, these data suggest that the oral bioavailability of (2R,6R)-HNK is sufficiently high such that large doses are not required to exert antidepressant effects via the oral route. Notably, oral (2R,6R)-HNK did not alter locomotor activity or induce behavioral changes indicative of discomfort, sickness, and/or stereotypy up to a dose of

450 mg/kg, at least nine-fold higher than those doses required for antidepressant effects

(15-50 mg/kg). These results are consistent with the previous finding that (2R,6R)-HNK lacks the hyperlocomotion and dissociative side effects associated with ketamine (Zanos et al., 2016) and, taken together, suggest a favorable safety profile, although future studies are needed to thoroughly investigate potential side effects following oral administration of

(2R,6R)-HNK.

Although the current study establishes the oral bioavailability and demonstrates oral antidepressant effectiveness of (2R,6R)-HNK in mouse behavioral tests, several limitations should be noted. First, a complete dose response relationship of behaviors for oral administration, compared to i.p. injection, has not been established. It is possible that the range of antidepressant doses, including the minimal effective dose needed to exert antidepressant effects, varies between these routes of administration. Thus, in order to further study the antidepressant and/or potential side effects at the most appropriate doses, future studies should establish the full dose response of oral (2R,6R)-HNK. Additionally, while none of the prodrug candidates tested improved the oral bioavailability of (2R,6R)-

HNK, these data do not preclude the possibiltiy that alternative prodrug strategies may enhance oral uptake of this compound. For example, amino acid modifications may enable active transport across the intestinal barrier (reviewed in Vig et al., 2013).

While there has been some progress towards the use of intranasal (Lapidus et al.,

2014), sublingual (Lara et al., 2013), and subcutaneous (George et al., 2017; Loo et al.,

2016) ketamine, most clinical studies utilize i.v. administration. These limitations for its administration, taken together with its side effects and abuse potential, limit the widespread use of ketamine in depression. Here, we show that the ketamine metabolite (2R,6R)-HNK has promising oral bioavailability, and that it exerts antidepressant-like effects via oral administration, without causing overt adverse behavioral effects.

Chapter 4. Sex-dependent metabolism and behavioral effectiveness of ketamine and (2R,6R)-hydroxynorketamine

4.1. Introduction

In preclinical studies, female rodents have been reported to be more sensitive to the antidepressant-relevant behavioral effects of ketamine, with lower doses required to exert behavioral effects in females than males (Carrier & Kabbaj, 2013; Franceschelli et al.,

2015; Saland et al., 2016; Sarkar & Kabbaj, 2016; Zanos et al., 2016). Higher levels of the

(2R,6R;2S,6S)-HNK metabolite were observed in the brains of female than male mice following ketamine dosing (Zanos et al., 2016), and higher levels of (2R,6R)-HNK have also been observed in the plasma and brains of female mice following its direct administration (Highland et al., 2018; and see Chapters 1-2). Taken together, these findings suggest that sex-dependent differences in the metabolism of ketamine to (2R,6R)-HNK contribute to the differences in ketamine’s antidepressant-like behavioral effects observed between male and female rodents.

While some studies have suggested a role of estradiol and/or progesterone in mediating the sex-dependent differences ketamine’s antidepressant-like effectiveness in rodents (Chen et al., 2020; Saland et al., 2016), the mechanisms underlying the sex- dependent differences in metabolism of ketamine and (2R,6R)-HNK and their role in modulating the observed behavioral differences have not previously been studied in detail.

Moreover, the potential role of testosterone in modulating differences in metabolism or behavioral effectiveness of ketamine and its metabolites have not been evaluated.

In the present study, we compared the metabolism of ketamine and its major metabolites, including (2R,6R;2S,6S)-HNK, following ketamine dosing and of (2R,6R)-

HNK following its direct administration, in intact male and female mice. Further, we

68 evaluated the role of circulating gonadal hormones in mediating sex-dependent pharmacokinetic differences. We observed that in intact male mice, peak and total plasma concentrations of ketamine were higher and those of HNK were lower, compared to females. Gonadectomy prevented the observed sex-dependent metabolism differences in male mice but did not alter ketamine or HNK levels in females, suggesting a role of testosterone in modulating metabolism. Finally, we evaluated whether gonadectomy and testosterone replacement altered behavioral sensitivity to ketamine in the forced swim test.

4.2. Methods

Animals

Male and female CD-1 mice (Charles Rivers Laboratories; Raleigh, NC, USA), 8-

10 weeks old at the time of testing, were habituated to the University of Maryland

(Baltimore, MD, USA) animal facility for at least one week prior to testing. Mice were group housed in cages of 4-5 per cage with a constant 12-hour light cycle (lights on/off at

07:00/19:00). Food and water were available ad libitum. All experiments were performed during the light phase. All studies were approved by the University of Maryland School of

Medicine Animal Care and Use Committee and conducted in accordance with the National

Institutes of Health Guide for the Care and Use of Laboratory Animals.

Drugs

(R,S)-ketamine hydrochloride was purchased from Sigma Aldrich (USA). (2R,6R)- hydroxynorketamine (HNK) hydrochloride was synthesized and characterized at the

National Center for Advancing Translation Sciences (National Institutes of Health,

Rockville, MD, USA) as previously described (Morris et al., 2017a). All compounds were dissolved in saline and administered intraperitoneally (i.p.) n a volume of 7.5 ml/kg.

Surgical procedures

Gonadectomy

Male and female mice were gonadectomized according to previously described methods (Idris, 2012). A surgical plane of anesthesia was induced (3.5-4.5%) and maintained (1.5-3%) with inhaled isoflurane delivered via a precision vaporizer. In male mice, a small incision was made in the midline of the scrotum, the testes were extracted, and the blood supply was clamped and cauterized. The incision was closed with non- absorbable monofilament sutures. In female mice, a small dorsal, midline incision was made in the skin and the fascia gently separated. Small incisions were then made in the abdominal muscle, the ovaries were extracted, and blood supply was clamped and cauterized. The muscle incisions were closed with absorbable monofilament sutures and the skin incisions with non-absorbable monofilament sutures. Sham surgeries were performed in male and female mice as described above, except that testes and ovaries, respectively, were left intact. Mice received carprofen (5 mg/kg, subcutaneous) pre- operatively as well as 24 and 48 h after surgery. All experiments were performed 10 days after surgery.

Subcutaneous implants

Testosterone was administered via subcutaneous, testosterone-releasing implants based upon published methods (Bowen et al., 2011; Daan et al., 1975). Lengths of silastic

70 tubing (12 mm length x 2.15 mm outer diameter) were filled with testosterone (12 mg) and sealed with silicone. Control implants were prepared to the same specifications but left empty. Implants were incubated at 37°C in saline overnight (approximately 12 h) prior to implantation. Implants were inserted at the time of gonadectomy or sham surgery. A small, dorsal incision (approximately 5 mm) was made, the implant was carefully inserted subcutaneously, and the incision closed with non-absorbable monofilament sutures.

Bioanalysis of plasma and brain levels

In four separate experiments, ketamine (10 mg/kg, i.p.) was administered to: 1) intact male and female mice, 2) gonadectomized or sham control male mice, 3) gonadectomized or sham control female mice, and 4) gonadectomized male mice with testosterone-releasing implants, gonadectomized male mice with control implants, and sham control male mice and control implants. In two additional experiments, (2R,6R)-HNK

(10 mg/kg, i.p.) was administered to 1) intact male or female mice, and 2) gonadectomized or sham control male mice. At 10 min, 30 min, 1 h, and 2 h after ketamine treatment, and at 5 min, 10 min, 30 min, 1 h and 2 h after (2R,6R)-HNK treatment, mice were deeply anesthetized (4% isoflurane for approximately 2 min). Trunk blood was immediately collected into 1.5-ml polypropylene tubes containing 30 µl of disodium EDTA (0.5 M, pH

8.0) and kept on ice until plasma collection (<30 minutes). Brains were harvested, immediately frozen in isopentane, and stored at -80°C until analysis. Blood was centrifuged at 8000 x g for 6 minutes at 4°C to obtain plasma. Plasma was collected into clean tubes and stored at -80°C until analysis. The concentrations of ketamine, norketamine, and HNK were determined by achiral liquid chromatography-tandem mass spectrometry (LC-

MS/MS) using a previously described protocol (Moaddel et al., 2010; Zanos et al., 2016).

LC-MS/MS analyses were performed by Dr. Ruin Moaddel (National Institute on Aging,

National Institutes of Health Intramural Research Program, Baltimore, MD).

Behavioral testing in the forced swim test

Ten days after the completion of surgeries (as described above), male mice

(gonadectomized with control implant, gonadectomized with testosterone-releasing implant, or sham surgery with control implant) received an i.p. injection of saline or ketamine (3 mg/kg, i.p.; administered by a male experimenter). Twenty-four hours later, mice were tested in the forced swim test, according to previously described methods

(Highland et al., 2018; Zanos et al., 2016). Briefly, mice were placed into a clear Plexiglass cylinder (20 cm height x 15 cm diameter) filled to a depth of 15 cm with water (23 ± 1°C) and subjected to a 6-min swim session. Swim sessions were recorded using a digital video camera. The time spent immobile (defined as passive floating with no movements other than those necessary to keep the head above water) was scored during the final 4 min of the session. Sessions were scored by a trained experimenter blind to the treatment groups.

Statistical analysis

All experiments and data collection were performed in a blinded and randomized manner. Statistical analyses were performed using GraphPad Prism software (v8;

GraphPad Software, Inc.). All statistical tests were two-tailed, and significance was assigned at p<0.05. Differences in concentrations of ketamine and metabolites over time were analyzed using a two-way ANOVA with sex or surgical condition and time as factors,

72 followed by Holm–Šídák post-hoc comparison when significance was reached. Differences in forced swim test immobility time were analyzed via two-way ANOVA with surgical condition and treatment as factors. Differences in weight gain were assessed via an unpaired t-test when two groups were compared and via one-way ANOVA when three groups were included. Statistically significant results are indicated with asterisks in the figures (*p<0.05, **p<0.01, ***p<0.001). Data are presented as mean ± SEM.

4.3. Results

Plasma levels of HNK are higher and those of ketamine are lower in female mice, compared to males.

Following administration of ketamine (10 mg/kg, i.p) to intact male and female mice, plasma concentrations of ketamine (effect of sex, F[1,24]=50.85, p<0.0001; effect of time, F[3,24]=187.8, p<0.0001; interaction, F[3,24]=21.11, p<0.0001; post-hoc comparison, p<0.0001 at 10 min and p=0.0493 at 30 min post-treatment; Figure 4.1A) and norketamine

(effect of sex, F[1,24]=6.823, p=0.0153; effect of time, F[3,24]=56.23, p<0.0001; interaction,

F[3,24]=0.4628, p=0.7109; Figure 4.1B) were lower, while concentrations of (2R,6R;2S,6S)-

HNK were higher (effect of sex, F[1,24]=23.06, p<0.0001; effect of time, F[3,24]=40.16, p<0.0001; interaction, F[3,24]=8.394, p=0.0005; post-hoc comparison, p<0.0001 at 10 min and p=0.0189 at 30 min post-treatment; Figure 4.1C), in females compared to males.

Robust, approximately two-fold, sex-dependent differences in the peak plasma concentrations (observed 10 min post-treatment) were observed for ketamine and HNK, whereas the differences in norketamine levels were more subtle (Figure 4.1).

The total plasma concentrations (area under the concentration vs. time curve; AUC) of ketamine and norketamine were approximately two-fold and 1.5-fold greater in males than in females. By contrast, the total plasma concentrations of HNK were approximately

1.5-fold greater in females than males. Plasma levels of dehydronorketamine (DHNK) were higher in females at 10 min post-treatment (effect of sex, F[1,24]=2.159, p=0.1548; effect of time, F[3,24]=31.32, p<0.0001; interaction, p=0.0484; post-hoc comparison, p=0.0324 at 10 minutes post-treatment; Figure 4.1D).

Following the direct administration of (2R,6R)-HNK (10 mg/kg, i.p.) to intact male and female mice, females had higher plasma levels of (2R,6R)-HNK at 5 and 10 min post- treatment (effect of sex, F[1,32]=31.45, p<0.0001; effect of time, F[4,32]=0.0001, interaction,

F[4,32]=8.178, p=0.0001; post-hoc comparison, p<0.0001 at 5 and 10 minutes post- treatment; Figure 4.2). Peak plasma levels (observed 5 min post-treatment) were approximately 1.3-fold higher in females, while concentrations 10 min post-treatment were approximately two-fold higher in females, compared to males (Figure 4.2). Total plasma concentrations of (2R,6R)-HNK were approximately 1.8-fold higher in female mice, relative to males.

Figure 4.1. Plasma concentrations of ketamine and metabolites following ketamine administration to male and female mice. Plasma concentrations of (A) ketamine (KET), (B) norketamine (norKET), (C) hydroxynorketamine (HNK), and (D) dehydronorketamine (DHNK) following a single intraperitoneal injection of ketamine (10 mg/kg) to intact male (M) and female (F) mice. Data are the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001. Inset area under the concentration vs. time curve (AUC).

Figure 4.2. Plasma concentrations of (2R,6R)- hydroxynorketamine following direct administration to male and female mice. Plasma concentrations of (2R,6R)- hydroxynorketamine (HNK) following a single intraperitoneal injection (10 mg/kg) to intact male (M) and female (F) mice. Data are the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001. Inset: values of the area-under-the-concentration vs. time curve (AUC) for male and female mice.

Gonadectomy alters metabolism of ketamine and (2R,6R)-HNK in males but not females

To determine whether gonadal hormones mediated the observed sex-dependent differences in the metabolism of ketamine, male and female mice underwent gonadectomy

10 days prior to ketamine dosing. In female mice, gonadectomy did not significantly alter plasma levels of ketamine (effect of gonadectomy, F[1,24]=2.403, p=0.1341; effect of time,

F[3,24]=262.9, p<0.0001; interaction, F[3,24]=1.306, p=0.2953), norketamine (effect of gonadectomy, F[1,24]=0.9115, p=0.3492; effect of time, F[3,24]=50.11, p<0.0001; interaction,

F[3.24]=0.7321, p=0.5430), HNK (effect of gonadectomy, F[1,24]=2.449, p=0.1307; effect of time, F[3,24]=50.35, p<0.0001; interaction, F[3,24]=1.454, p=0.2520), or DHNK (effect of gonadectomy, F[1,24]=1.179, p=0.284; effect of time, F[3,24]=36.09, p<0.0001; interaction,

F[3,24]=1.203, p=0.3299; Figure 4.3A-D), compared to sham-operated controls.

In male mice, gonadectomy resulted in lower plasma levels of ketamine (Figure

4.3E; effect of gonadectomy, F[1,22]=7.167, p=0.0138; effect of time, F[3,22]=72.30, p<0.0001; interaction, F[3,22]=2.183, p=0.1187) and higher levels of HNK (effect of gonadectomy, F[1,22]=7.056, p=0.0144; effect of time, F[3,22]=16.43, p<0.0001; interaction,

F[3,22]=5.004, p=0.0085; post-hoc comparison, p=0.0026 at 10 minutes) in the first 10 min

75 post-treatment (Figure 4.3G), compared to sham-operated controls. Gonadectomy did not significantly alter levels of norketamine (Figure 4.3F; effect of gonadectomy

F[1,22]=0.4235, p=0.5219; effect of time, F[3,22]=18.92, p<0.0001; interaction,

F[3,22]=0.3883, p=0.7626) or DHNK (Figure 4.3H; effect of gonadectomy F[1,22]=2.051, p=0.1661; effect of time, F[3,22]=2.776, p=0.0653; interaction, F[3,22]=0.5624, p=0.6455).

Relative to sham-operated controls, gonadectomy resulted in approximately 1.4-fold and

1.5-fold lower peak and total plasma ketamine levels, respectively. Peak HNK plasma levels were approximately two-fold higher in gonadectomized males, while total HNK plasma concentrations were only slightly higher in gonadectomized males (approximately

1.1-fold increase), relative to sham controls.

Figure 4.3. Gonadectomy alters plasma levels of ketamine and hydroxynorketamine following ketamine treatment in male, but not female, mice. (A-D) Plasma concentrations of (A) ketamine (KET), (B) norketamine (norKET), (C) hydroxynorketamine (HNK), and (D) dehydronorketamine (DHNK) following a single intraperitoneal injection of ketamine (10 mg/kg) to female mice 10 days after sham surgery or gonadectomy (GDX). (E-H) Plasma concentrations of (E) KET, (F) norKET, (G) HNK, and (H) DHNK following a single intraperitoneal injection of ketamine (10 mg/kg) to male mice 10 days after sham surgery or gonadectomy (GDX). Data are the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001. Inset area under the concentration vs. time curve (AUC).

To determine if gonadectomy also altered (2R,6R)-HNK levels following its direct administration, male mice underwent gonadectomy or sham surgery 10 days prior to

(2R,6R)-HNK dosing. Similar to the effects observed following ketamine dosing, gonadectomy resulted in higher plasma levels of (2R,6R)-HNK following its direct administration (Figure 4.4; effect of gonadectomy, F[1,30]=12.09, p=0.0016; effect of time,

F[4,30]=145.0, p<0.0001; interaction, F[4,30]=3.740, p=0.0139; post-hoc comparison, p=0.0011 at 5 min and p=0.0052 at 10 min post-treatment). Plasma levels were approximately 1.3-fold higher at their peak, 5 min post-treatment, and approximately 1.5- fold higher 10 min post-treatment, in gonadectomized male mice, relative to sham controls.

Following its direct administration, total plasma exposure to (2R,6R)-HNK was approximately 1.4-fold greater in gonadectomized males compared to sham controls.

Figure 4.4. Plasma concentrations of (2R,6R)- hydroxynorketamine following direct administration to gonadectomized or sham male mice. Plasma concentrations of (2R,6R)-hydroxynorketamine (HNK) following a single intraperitoneal injection (10 mg/kg) to male mice 10 days after gonadectomy (GDX) or sham surgery. Data points and error bars represent mean and SEM, respectively. *p<0.05, **p<0.01, ***p<0.001. Inset: values of area under the concentration vs. time curve (AUC) obtained from sham- operated and gonadectomized animals.

Testosterone did not increase behavioral effectiveness of low-dose ketamine in male mice

Based upon the elevated plasma levels of HNK observed in gonadectomized male mice following ketamine administration, it was hypothesized that gonadectomy would enhance the behavioral sensitivity to ketamine in male mice in the forced swim test (i.e.

77 reduce the minimal effective dose required to decrease immobility time) and that testosterone would reverse this effect. To test this hypothesis, male mice (gonadectomized with control implant, gonadectomized with testosterone-releasing implant, or sham surgery with control implant) received a single dose of ketamine (3 mg/kg, i.p.) previously demonstrated to be sub-effective in intact males and effective in females (Zanos et al.,

2016), and 24 h later, were tested in the forced swim test. It was expected that gonadectomized males (with control implants) would exhibit reduced immobility time at this dose. However, none of the experimental groups demonstrated any significant changes in immobility time following this dose of ketamine (Figure 4.5; effect of treatment,

F[1,65]=0.2099, p=0.6484; effect of surgery, F[2,65]=0.6236, p=0.5392; interaction,

F[2,65]=0.5889, p=0.5578).

Figure 4.5. Testosterone did not alter the behavioral effectiveness of ketamine in the mouse forced swim test. Twenty-four hours after a dose of ketamine that is typically sub- effective in intact male mice (3 mg/kg, i.p.), no changes in immobility time were observed for sham controls, gonadectomized male mice with control implants (GDX), or gonadectomized male mice with testosterone-releasing implants (GDX+T) in the forced swim test. Data are the mean ± SEM.

Gonadectomy alters weight gain in both sexes

There was an overall effect of gonadectomy on body weight observed in both male

(Figure 4.6A; effect of gonadectomy, F[1,136]=4.79, p=0.0304; effect of time, F[1,136]=1.94, p=0.167; interaction, F[1,136]=1.65, p=0.201) and female (Figure 4.6C; effect of gonadectomy, F[1,60]=4.47, p=0.0386; effect of time, F[1,60]=74.9, p<0.0001; interaction,

F[1,60]=4.47, p=0.0503) mice. However, in male mice that received subcutaneous implants,

78 the effect of surgical treatment group did not reach statistical significance (effect of surgery, F[2,136]=1.31, p=0.272; effect of time, F[1,136]=10.9, p=0.0012; interaction,

F[2,136]=0.489, p=0.614).

To evaluate the effects of gonadectomy on weight gain during the post-operative period, changes in body weight were calculated as the difference in body weight on post- operative day 9 (24 h prior to drug dosing) and the pre-operative body weight (Figure 4.6B,

4.6D, and 4.6F). In male mice, gonadectomy decreased weight gain relative to sham controls (Figure 4.6B; t[68]=4.06, p=0.0001). This effect was not altered by the placement of control subcutaneous implant but was reversed with testosterone-releasing implants

(Figure 4.6F; on-way ANOVA, effect of surgery, F[2,68]=6.99, p=0.0017; post-hoc comparison, sham vs. gonadectomized with control implant, p=0.0013; sham vs. gonadectomized with testosterone implant, p=0.176; gonadectomized with control implant vs. testosterone implant, p=0.0430). In females, gonadectomy increased post-operative weight gain relative to sham controls (Figure 4.6D; t[30]=2.93, p=0.0065).

Figure 4.6. Gonadectomy alters weight gain in mice. Gonadectomy (GDX) altered body weight in (A-B) male and (C-D) female mice. (E) In male mice with subcutaneous testosterone (T)-releasing or control (Ctrl) implants, the effect of surgical condition (gonadectomy and subcutaneous implant) on body weight did not reach statistical significance. Weight gain during the post-operative period (measured as the difference between pre-operative (pre-) body weight and weight at post-operative day 9 (post-)) was attenuated in (B,F) gonadectomized male mice, an effect that was reversed by (F) testosterone replacement. (D) In females, gonadectomy increased post-operative weight gain relative to sham controls. Graph and error bars represent mean and SEM, respectively. *p<0.05, **p<0.01, ***p<0.0001.

4.4 Discussion

The present study demonstrated that female mice have higher circulating levels of

HNK, both following ketamine treatment (Figure 4.1) and the direct administration of

(2R,6R)-HNK (Figure 4.2), consistent with previous studies (Highland et al., 2018; Zanos et al., 2016; and see Chapters 2-3). After ketamine treatment, female mice also had lower plasma levels of ketamine and norketamine, compared to males (Figure 4.1). Of note, one earlier study reported that female rats had higher levels of ketamine and norketamine, compared to males (Saland & Kabbaj, 2018), in contrast with the present data. The apparent discrepancy could be, at least in part, accounted for by species-specific activity of different

CYP isoforms, including CYP3A (Martignoni et al., 2006), which is a major player in the metabolism of ketamine.

The present data are consistent with sex-dependent differences in both the phase I metabolism of ketamine to (2R,6R)-HNK, and in the phase II metabolism and/or elimination of (2R,6R)-HNK. In particular, the data suggest that, in female mice, ketamine is metabolized more rapidly and/or extensively to (2R,6R)-HNK, and (2R,6R)-HNK is eliminated more slowly than in males. As a result, total plasma concentrations of (2R,6R)-

HNK are higher in females than in males, which may explain the enhanced behavioral sensitivity to ketamine that has been observed in female relative to male rodents (Carrier

& Kabbaj, 2013; Franceschelli et al., 2015; Saland et al., 2017; Saland et al., 2016; Sarkar

& Kabbaj, 2016; Zanos et al., 2016).

Additionally, we demonstrated that male gonadal hormones modulate the metabolism of ketamine and (2R,6R)-HNK. We found that gonadectomy did not alter the levels of ketamine or its metabolites in female mice but resulted in lower levels of ketamine and higher levels of HNK after ketamine treatment in males (Figure 4.3). This is in contrast with an earlier report that the estrous cycle influenced ketamine’s metabolism in rats, suggesting a role of ovarian hormones in modulating the metabolism of ketamine, at least in rats (Saland & Kabbaj, 2018). It is possible that this discrepancy is explained by differences in metabolism that exist between mice and rats, but this possibility requires further testing.

Following direct administration of (2R,6R)-HNK, gonadectomized male mice had higher plasma levels of (2R,6R)-HNK compared to sham controls (Figure 4.4). Thus, comparing gonadectomized vs. sham-operated males, the differences in ketamine and

(2R,6R)-HNK levels were similar to those observed when comparing females vs. intact males. However, circulating male gonadal hormones may only partly explain the differences in ketamine metabolism. Following ketamine treatment, both gonadectomized male and female mice exhibited approximately two-fold greater peak HNK plasma levels relative to sham or intact males, respectively. Yet, the differences in peak ketamine plasma levels and in total ketamine and HNK plasma levels were more robust when comparing intact females vs. males (2-fold reduction in peak ketamine levels, 2-fold reduction in ketamine exposure, and 1.5-fold increase in HNK exposure) than for gonadectomized vs. sham males (1.4-fold reduction in peak ketamine levels, 1.5-fold reduction in ketamine

81 exposure, and only 1.1-fold reduction in HNK exposure). Similarly, when comparing total plasma concentrations following direct administration of (2R,6R)-HNK, more robust differences were observed for intact females vs. males (1.8-fold) than for gonadectomized vs. sham males (1.4-fold), although both females and gonadectomized males had similar

(1.3-fold) increases in peak (2R,6R)-HNK concentrations compared to intact or sham males, respectively.

Of note, all experiments were conducted 10 days after gonadectomy. It is possible that effects of circulating gonadal hormones persist to some degree up to 10 days post- gonadectomy and, thus, it remains possible that gonadectomy may result in more robust metabolism changes after a more prolonged period prior to treatment. It is also possible that organizational effects of gonadal hormones, that are initiated during earlier stages in development and not altered by gonadectomy in adulthood, may partially contribute to metabolism differences. Furthermore, it is important to note that gonadectomy resulted in changes in body weight compared to sham controls, particularly an attenuation of weight gain in male mice and an increase in weight gain in females (Figure 4.6), consistent with previous reports (e.g. Harada et al., 2016; Inoue et al., 2010; Masui & Tamura, 1926; Ornoy et al., 1994). However, in contrast, gonadectomy has also been reported to increase body weight in male mice (Harada et al., 2016; Ornoy et al., 1994), while at least one study reported no effect of gonadectomy on body weight (Masui & Tamura, 1926). Differences in experimental strain or age may explain these discrepancies. Nevertheless, it is possible that the decreased weight gain observed in gonadectomized males, potentially due to reduced body fat (Krotkiewski et al., 1980), may itself result in pharmacokinetic changes.

While the precise effects of body fat composition on drug metabolism have not fully

82 elucidated, some of the cytochrome P450 (CYP450) enzymes involved in ketamine’s metabolism, including CYP3A4, CYP2A6, and CYP2B6 (Desta et al., 2012) have been reported to be altered in models of obesity (reviewed in Tomankova et al., 2017). In particular, both CYP3A4 expression and activity was decreased in a mouse model of obesity (Tomankova et al., 2017). Conversely, CYP2B6 activity and expression, as well as overall CYP2A activity were increased in mouse models of obesity (Tomankova et al.,

2017). Altogether, while there is some evidence to suggest that changes in body fat composition may alter CYP-mediated drug metabolism, the precise effects on CYP expression and/or activity, and their potential role in mediating the changes observed here, require further clarification. However, it remains possible that the observed metabolism differences are an indirect effect of gonadectomy, which is mediated by body weight/composition changes.

In addition to potential indirect effects mediated via body fat changes, additional and/or alternative mechanisms may underlie the effects of gonadal hormones, namely testosterone, on the metabolism of ketamine to (2R,6R;2S,6S)-HNK. First, because both testosterone (Kandel et al., 2017; Krauser et al., 2004; Maenpaa et al., 1993; Sohl et al.,

2009) and ketamine (Desta et al., 2012) undergo CYP3A4-mediated metabolism, it is possible that circulating testosterone attenuates metabolism of ketamine to form

(2R,6R;2S,6S)-HNK via competition for the available enzyme. Similarly, competition with testosterone may also underlie the changes in phase II metabolism or elimination of

(2R,6R)-HNK. A second possibility, not mutually exclusive with the former, is that testosterone alters the activity and/or expression of enzymes involved in the metabolism of ketamine and (2R,6R)-HNK. Sex-dependent differences have been observed for several of

83 the CYP450 enzymes involved in ketamine’s metabolism to HNKs, including increased

CYP2A6, CYP2B6, and CYP3A4 expression and activity in females relative to males, in both humans and rodents (Gochfeld, 2017; Waxman & Holloway, 2009; Yang et al.,

2012). Evidence suggest that sex-dependent differences in CYP450 activity are modulated by the gonadal hormones, including testosterone, via effects on growth hormone release from the pituitary gland (reviewed in Waxman & Holloway, 2009). Thus, a testosterone- dependent attenuation of the activity and/or expression of CYP2A6, CYP2B6, or CYP3A4 may underlie decreased metabolism of ketamine to form (2R,6R;2S,6S)-HNK in intact males, relative to females or gonadectomized controls, but future studies are required to understand the precise enzymes and mechanisms involved.

The apparent differences in the elimination of (2R,6R)-HNK may be mediated via similar mechanisms. The HNKs undergo glucuronide conjugation, which is partly catalyzed by the glucuronosyltransferase isoform UGT2B4 (Moaddel et al., 2010). While testosterone has not been reported as a substrate for this specific isoform (Barre et al.,

2007), it does undergo glucuronide conjugation catalyzed by other glucuronosyltransferase isoforms, predominantly UGT2B17 (Sten et al., 2009; Turgeon et al., 2001). If UGT2B17 also contributes to the glucuronidation of (2R,6R)-HNK, then competition with testosterone may also underlie the lower (2R,6R)-HNK levels following its direct administration to intact males, compared with females or gonadectomized males.

Additionally, it has been reported that males have higher expression of UGT2B17, in addition to UGT2B28 and UGT2A3 (Yang et al., 2012), which may additionally contribute to these differences. However, the contribution of UGT2B17, or other isoforms, to the glucuronidation of (2R,6R)-HNK remains unknown.

Based upon the relative increase in (2R,6R)-HNK observed in gonadectomized male mice compared to sham controls, it was hypothesized that gonadectomy would enhance the behavioral effectiveness of ketamine in male mice. Further, we hypothesized that testosterone replacement would reverse this effect, consistent with testosterone modulating the observed differences in metabolism. However, at the single dose of ketamine tested here (3 mg/kg, i.p.), which was selected based upon an earlier study demonstrating it to be sub-effective in intact male mice but effective in female mice (Zanos et al., 2016), none of the experimental groups exhibited significant ketamine-induced reductions in immobility time in the forced swim test (Figure 4.5). While it is possible that gonadectomy does not enhance behavioral sensitivity to ketamine despite its effects on metabolism and resultant HNK levels, a second possibility is that gonadectomy does increase the behavioral effectiveness to some degree, but that our experimental design using only a single dose (which may still be a sub-effective dose in gonadectomized male mice) did not capture this effect. Thus, additional experiments are required to elucidate the effects of gonadal hormones, particularly testosterone, on the behavioral actions of ketamine and (2R,6R)-HNK in CD1 mice.

Notably, several previous studies have demonstrated a role of the ovarian hormones in modulating the antidepressant-relevant behavioral effects of ketamine (Carrier &

Kabbaj, 2013; Chen et al., 2020; Dossat et al., 2018; Saland et al., 2016) or (2R,6R)-HNK

(Chen et al., 2020). In particular, it has been demonstrated that estradiol and progesterone modulate the antidepressant-relevant behavioral responses to ketamine in female mice and rats (Carrier & Kabbaj, 2013; Chen et al., 2020; Saland et al., 2016) and to (2R,6R)-HNK in female mice (Chen et al., 2020). It has also been reported that the estrous cycle influences

85 the antidepressant-relevant behavioral effects, assessed in the forced swim test, following systemic treatment of female mice with ketamine (Dossat et al., 2018). Finally, one study demonstrated that progesterone enhanced the anti-anhedonic effects of ketamine in male rats (Saland et al., 2016).

While the effects of testosterone have not been as thoroughly studied, one study demonstrated that testosterone blocked the effects of low-dose ketamine in females, but consistent with the present results, testosterone did not increase sensitivity to a low dose of ketamine in male rats (Saland et al., 2016). However, similar to the limitations of our experimental design, Saland et al. (2016) tested only a single dose of ketamine in both sexes. Altogether, the role of testosterone in modulating antidepressant-like effects in male and female mice requires further clarification, and future studies should evaluate a full range of doses, and possibly various time points relative to testosterone depletion/replacement to fully understand the effects of testosterone on ketamine’s, and possibly (2R,6R)-HNK’s, behavioral actions.

While effect of biological sex on ketamine’s metabolism and antidepressant efficacy has not been studied extensively in humans, several studies have reported sex- dependent differences in the levels of ketamine and its metabolites following an infusion of ketamine (Zarate et al., 2012) or in ketamine’s antidepressant outcomes or adverse effects, though the results have been mixed (Coyle & Laws, 2015; Freeman et al., 2019;

Niciu et al., 2014; Rybakowski et al., 2017). Zarate et al. (2012) observed higher plasma levels of (2R,6R;2S,6S)-HNK, (2R,5R;2S,5S)-HNK, and DHNK in female patients, compared to males, while male subjects had higher levels of hydroxyketamine. It has also been reported that the antidepressant response to ketamine is more prevalent among males

86 compared to females (Rybakowski et al., 2017), and a meta-analysis indicated that male sex was predictive of sustained antidepressant response 1-week after ketamine treatment

(Coyle & Laws, 2015). However, other studies have reported no significant differences in the antidepressant response to ketamine between the sexes (Freeman et al., 2019; Niciu et al., 2014). Although a study of recreational ketamine users reported that females were more likely to experience discontinuation symptoms, including anxiety and dysphoria (Chen et al., 2014), following an antidepressant infusion of ketamine, one clinical trial found no significant difference in the frequency of adverse events between males and females

(Freeman et al., 2019), while a second reported that dissociative side effects were more prevalent in males (Derntl et al., 2019). Altogether, potential sex-dependent differences in the metabolism of ketamine and (2R,6R)-HNK, as well as in the antidepressant response and adverse effect burden of these compounds, requires further study.

Overall, we demonstrated that female mice have higher levels of HNK and lower levels of ketamine following ketamine treatment and higher levels of (2R,6R)-HNK following its direct administration, and that male gonadal hormones, at least partly, mediate these sex-dependent differences. These sex-dependent differences in ketamine and HNK levels are an important variable that should be considered in future studies. In particular, while their relevance to the antidepressant-relevant behavioral effects of ketamine and

(2R,6R)-HNK requires further clarification, it is also possible that adverse effects may exhibit sex-dependent differences. Namely, the adverse effects associated with ketamine, which have been attributed to its ability to inhibit NMDARs (reviewed in Zanos et al.,

2018), may be less prevalent or severe in females, due to their relatively lower levels of and exposure to ketamine compared to males. On the contrary, females may be more

87 sensitive to antidepressant effects mediated by (2R,6R)-HNK, due to their greater exposure compared to males. These possibilities, while not an exhaustive list of potential clinical implications, exemplify the importance the sex-dependent differences in metabolism of ketamine and HNK as an experimental variable when designing and interpreting future studies.

Chapter 5. Evaluation of the antidepressant-relevant behavioral effectiveness of the (2,6)-hydroxynorketamines

5.1. Introduction

Several studies have provided evidence that (2R,6R)-hydroxynorketamine (HNK) and (2S,6S)-HNK induce antidepressant-relevant effects in a variety of preclinical rodent behavioral tests thought to predict antidepressant efficacy (Chen et al., 2020; Chou et al.,

2018; Fukumoto et al., 2019; Highland et al., 2018; Lumsden et al., 2019; Nelson &

Trainor, 2007; Pham et al., 2018; Rahman et al., 2020; Yokoyama et al., 2020; Zanos,

Highland, Stewart, et al., 2019; Zanos et al., 2016). However, the potential behavioral effects of other HNKs have not been studied to date.

A total of twelve HNK metabolites are formed from the metabolism of ketamine in vivo (Leung & Baillie, 1986; Moaddel et al., 2016; Moaddel et al., 2010; Zanos et al., 2016;

Zarate et al., 2012; Zhao et al., 2012), with the (2,6)-HNKs being the most prevalent in the majority species studied (Moaddel et al., 2016; Moaddel et al., 2010; Sandbaumhuter,

Theurillat, & Thormann, 2017; Zanos et al., 2016; Zarate et al., 2012). There is evidence to suggest that the (2,6)-HNKs play a critical role in mediating the sustained antidepressant-relevant actions of the parent compound ketamine (Zanos, Highland, Liu, et al., 2019; Zanos et al., 2016). Namely, it was demonstrated that deuterium substitution at the C6 position of racemic ketamine, a chemical modification which specifically attenuates its metabolism to form the (2,6)-HNKs, prevents the sustained (observed 24 h post-treatment) antidepressant-relevant behavioral effects induced by ketamine in the mouse forced swim and learned helplessness tests (Zanos et al., 2016). Likewise, the analogous modification to (R)-ketamine, which specifically attenuates its metabolism to

(2R,6R)-HNK and (2R,6S)-HNK, suppressed its antidepressant-relevant behavioral

89 effectiveness compared to that of unmodified (R)-ketamine (Zanos, Highland, Liu, et al.,

2019), further supporting a role of (2R,6R)-HNK (the predominant HNK) and possibly

(2R,6S)-HNK in at least partly mediating the antidepressant-relevant effects of ketamine.

However, the contribution of the (2,6)-HNKs to mediating ketamine’s actions has been called into question by some contrasting reports (Yamaguchi et al., 2018; Zhang et al.,

2018).

In particular, it was reported that attenuating (R)-ketamine’s metabolism to

(2R,6R)- and (2R,6S)-HNK, either via deuterium substitution (Zhang et al., 2018) or prior administration with a CYP inhibitor cocktail (Yamaguchi et al., 2018), did not attenuate ketamine’s antidepressant-like behavioral effects. Nonetheless, independent of their role in mediating ketamine’s actions, a growing number of studies have demonstrated that direct administration of (2R,6R)-HNK induces antidepressant-relevant behavioral effects in rodents (Chen et al., 2020; Chou et al., 2018; Elmer et al., 2020; Fukumoto et al., 2019;

Lumsden et al., 2019; Pham et al., 2018; Zanos, Highland, Liu, et al., 2019; Zanos,

Highland, Stewart, et al., 2019; Zanos et al., 2016). Several studies have also demonstrated antidepressant-relevant behavioral effects following (2S,6S)-HNK administration to rodents (Yokoyama et al., 2020; Zanos et al., 2016), albeit with lower apparent effectiveness when compared to (2R,6R)-HNK under similar experimental conditions

(Zanos et al., 2016).

Consistent with the greater relative antidepressant effectiveness of (2R,6R)-HNK compared to (2S,6S)-HNK, it has also been reported that (2R,6R)-, but not (2S,6S)-HNK, exerts behavioral effects when tested at equivalent doses in rodent tests thought to predict antidepressant efficacy (Chou, 2020; Chou et al., 2018). On the contrary, at least one study

90 has reported that (2S,6S)-, but not (2R,6R)-HNK, reduced forced swim test immobility time and reversed measures of anhedonia following chronic stress in mice (Yokoyama et al., 2020). It is unclear whether this discrepancy may be explained by differences in experimental methods. Nonetheless, it is possible that additional HNK metabolites exert antidepressant-relevant behavioral actions with even greater effectiveness compared to

(2R,6R)-HNK or (2S,6S)-HNK. Moreover, identifying the structural characteristics that confer enhanced behavioral effectiveness may inform the development of novel drug candidates.

Here, we compared the antidepressant-like behavioral effects of (2R,6R)-, (2S,6S)-

, (2R,6S)-, and (2S,6R)-HNK in the mouse forced swim test and established their relative rank-order of effectiveness by comparing their minimally effective doses under the same experimental conditions. Additionally, we utilized the novel drug candidate (5R)-methyl-

(2R,6R)-HNK, which shares a three-dimensional structure similar to (2R,6S)-HNK, to evaluate whether three-dimensional structure confers the behavioral effectiveness of the

(2,6)-HNKs.

5.2. Methods

Animals

Male CD-1 mice (Charles Rivers Laboratories; Raleigh, NC, USA), 8-11 weeks old at the time of testing, were habituated to the University of Maryland (Baltimore, MD, USA) animal facility for at least one week prior to testing. Mice were group housed in cages of

4-5 per cage with a constant 12-hour light cycle (lights on/off at 07:00/19:00). Food and water were available ad libitum. All experiments were performed during the light phase.

All studies were approved by the University of Maryland School of Medicine Animal Care and Use Committee and conducted in accordance with the National Institutes of Health

Guide for the Care and Use of Laboratory Animals.

Drugs

(2R,6R)-HNK hydrochloride, (2S,6S)-HNK hydrochloride, (2R,6S)-HNK, and

(2S,6R)-HNK were synthesized and characterized (molecular structures were confirmed via 1H and 13C NMR, the absolute and relative stereochemistry was assigned by x-ray crystallography, and the purity was determined via LC/MS) at the National Center for

Advancing Translation Sciences (National Institutes of Health, Rockville, MD, USA). The synthesis and characterization of (2R,6R)-HNK hydrochloride, (2S,6S)-HNK hydrochloride, (2R,6S)-HNK, and (2S,6R)-HNK were performed as previously described

(Morris et al., 2017b). (5R)-methyl-(2R,6R)-HNK was obtained from Organix Inc.

(Woburn, MA, USA). All compounds were dissolved in 20% (w/v) cyclodextrin (Sigma

Aldrich, USA) in saline and administered intraperitoneally (i.p.) in a volume of 7.5 ml/kg.

Forced swim test

The (2,6)-HNKs were tested in separate cohorts of mice. (2R,6R)-HNK, (2S,6S)-

HNK, (2R,6S)-HNK, (2S,6R)-HNK were administered at the doses of 1, 3, 10, 30, and 100 mg/kg. (5R)-methyl-(2R,6R)-HNK was administered at the doses of 0.3, 1, 3, and 10 mg/kg. Vehicle and (2R,6R)-HNK (30 mg/kg) were included as controls in each cohort.

Injections were performed by a male experimenter. Twenty-four h after treatment, mice were tested in the forced swim test, according to previously described methods (Highland

92 et al., 2018; Zanos et al., 2016). Briefly, mice were placed into a clear Plexiglass cylinder

(20 cm height x 15 cm diameter) filled to a depth of 15 cm with water (23 ± 1°C) and subjected to a 6-min swim session. Swim sessions were recorded using a digital video camera. The time spent immobile (defined as passive floating with no movements other than those necessary to keep the head above water) was scored during the final 4 min of the session. Sessions were scored by a trained experimenter blind to the treatment groups.

Statistical analysis

All experiments and analyses were performed in a blinded and randomized manner.

All injections were performed by a male experimenter. Statistical analyses were performed using GraphPad Prism software (v8; GraphPad Software, Inc.). All statistical tests were two-tailed, and significance was assigned at p<0.05. Differences in immobility time were analyzed via one-way ANOVA followed by Holm–Šídák post-hoc comparison (with all groups compared to saline control group) when significance was reached. Significant results are indicated with asterisks in the figures (*p<0.05, **p<0.01, ***p<0.001). Data are presented as the mean ± standard error of the mean (SEM). Individual group sizes and statistical analyses for each experiment are summarized in Table 5.1.

5.3. Results

(2,6)-HNKs reduce forced swim test immobility time with differing effectiveness

The behavioral effects of (2R,6R)-HNK (Figure 5.1A), (2S,6S)-HNK (Figure 5.1B),

(2R,6S)-HNK (Figure 5.1C), and (2S,6R)-HNK (Figure 5.1D) were evaluated in the mouse forced swim test, 24 h after treatment (see Table 5.1 for statistical analyses). Relative to

93 vehicle, (2R,6R)-HNK reduced immobility time at 10, 30, and 100 mg/kg (Figure 5.1A), whereas (2S,6S)-HNK reduced immobility at 30 and 100 mg/kg, consistent with previous reports of their relative behavioral effectiveness (Zanos et al., 2016). (2R,6S)-HNK (Figure

5.1C) and (2S,6R)-HNK (Figure 5.1D) reduced immobility time at the doses of 1, 3, and

10 mg/kg and at 3 and 10 mg/kg, respectively. This demonstrates, for the first time, that

(2R,6S)- and (2S,6R)-HNK are capable of exerting antidepressant-relevant behavioral effects, both with enhanced effectiveness compared to (2R,6R)-HNK. Of note, both

(2R,6S)- and (2S,6R)-HNK exhibited U-shaped dose response curves, reducing immobility time at doses ≤10 mg/kg, but not at doses ≥30 mg/kg (Figure 5.1C-D), in contrast with

(2R,6R)- and (2S,6S)-HNK, which did not demonstrate U-shaped dose responses within the tested range (up to 100 mg/kg).

Figure 5.1. (2,6)- hydroxynorketamines reduce immobility time in the mouse forced swim test. (A) (2R,6R)- hydroxynorketamine (HNK), (B) (2S,6S)- HNK, (C) (2R,6S)-HNK, and (D) (2S,6R)-HNK reduced immobility time in the mouse forced swim test with minimal effective doses of 10 mg/kg, 30 mg/kg, 1 mg/kg, and 3 mg/kg, respectively, 24 h after intraperitoneal injection. Graphs and error bars represent mean and SEM, respectively, of results obtained from 10-18 mice/group. *p<0.05, **p<0.01, ***p<0.001.

Table 5.1 Statistical analyses of effects of the (2,6)-HNKs in the forced swim test.

Note: samples sizes are listed in respective order of experimental groups. Abbreviations: HNK, hydroxynorketamine; Me, methyl; NA, not applicable; VEH, vehicle.

(5R)-methyl-(2R,6R)-HNK has comparable antidepressant-like behavioral effectiveness to (2R,6S)-HNK

Based upon the finding that (2R,6S)-HNK reduced forced swim test immobility with greater effectiveness compared to (2R,6R)-HNK (Figure 5.1), it was hypothesized that differences in the three-dimensional structures of these compounds (Figure 5.2 A-B) underlie their differential behavioral effectiveness. While in (2R,6R)-HNK the amine and hydroxyl groups are preferentially oriented equatorial and the aryl group axial to the

95 cyclohexyl ring (Figure 5.2A), in (2R,6S)-HNK the aryl group is preferentially oriented equatorial to the cyclohexyl ring (Figure 5.2B; see Morris et al., 2017b). Thus, the novel compound (5R)-methyl-(2R,6R)-HNK was synthesized, where the methyl group substitution causes the compound to adopt a three-dimensional structure akin to (2R,6S)-

HNK (Figure 5.2C). Because it was predicted that (5R)-methyl-(2R,6R)-HNK would exert antidepressant-relevant behavioral effects at low doses, it was tested within the range of

0.3-10 mg/kg.

Similar to (2R,6S)-HNK, (5R)-methyl-(2R,6R)-HNK reduced immobility time in the forced swim test with a minimum effective dose of 1 mg/kg, and was effective within the range of 1-10 mg/kg (with a trend to reduce immobility, p=0.06, observed at the dose of 3 mg/kg; Figure 5.2D). The present data demonstrate that the novel compound (5R)- methyl-(2R,6R)-HNK exerts antidepressant-relevant behavioral effects in the forced swim test, (Figure 5.2D) with similar effectiveness compared to (2R,6S)-HNK (Figure 5.1C), consistent with the hypothesis that the three-dimensional structure of (2R,6S)-HNK confers its enhanced effectiveness compared to (2R,6R)-HNK.

Figure 5.2. (5R)-methyl-(2R,6R)-hydroxynorketamine reduces forced swim test immobility time. (A-C) Three-dimensional structures of (A) (2R,6R)-hydroxynorketamine (HNK; adapted from Morris et al., 2017b), and), (B) (2R,6S)-HNK (adapted from Morris et al., 2017b), and (C) (5R)-methyl (Me)-(2R,6R)- HNK. (D) (5R)-Me-(2R,6R)-HNK reduces immobility time in the mouse forced swim test 24 hours after treatment, with minimal effective dose of 1 mg/kg (i.p.). Data are the mean ± SEM. n=13-15/group. *p<0.05, **p<0.01.

5.4. Discussion

For the first time, antidepressant-relevant behavioral effects of the four (2,6)-HNKs were compared in the mouse forced swim test. Results presented here demonstrate that

(2R,6R)-HNK has greater behavioral effectiveness (minimum effective dose 10 mg/kg, i.p.;

Figure 5.1A) compared to (2S,6S)-HNK (minimum effective dose 30 mg/kg, i.p.; Figure

5.1B). This is consistent with an earlier study (Zanos et al., 2016), which reported that

(2R,6R)-HNK reduced forced swim test immobility time at doses as low as 5 mg/kg, i.p. and reduced escape failures in the learned helplessness test at doses as low as 3 mg/kg, i.p., whereas (2S,6S)-HNK required doses of 25 and 75 mg/kg, i.p., respectively, to exert effects in the same behavioral tests. The relatively greater behavioral effectiveness of (2R,6R)-

HNK compared to that of (2S,6S)-HNK is also consistent with earlier studies demonstrating that, when tested at equivalent doses (10 mg/kg, i.p.), (2R,6R)-, but not (2S,6S)-HNK, reduced forced swim test immobility time in mice (Chou et al., 2018) and reversed learned helplessness in rats (Chou, 2020). However, these findings are in contrast to one study which reported that (2S,6S)-, but not (2R,6R)-HNK, reduced forced swim test immobility time and measures of anhedonia following chronic stress in mice when each was tested at the same dose (20 mg/kg, i.p.; Yokoyama et al., 2020). It is unclear whether differences in experimental design, species, or strain may underlie these apparent discrepancies.

Nevertheless, by testing across a 100-fold range of doses and comparing compounds under the same experimental conditions, the present data support that (2R,6R)-HNK exhibits greater behavioral effectiveness, relative to (2S,6S)-HNK, in the forced swim test.

A novel finding is that both (2R,6S)-HNK (effective at 1 mg/kg, i.p.; Figure 5.1C) and (2S,6R)-HNK (minimum effective dose 3 mg/kg, i.p.; Figure 5.1D) exerted behavioral

97 effects in the forced swim test at lower doses compared to either (2R,6R)- or (2S,6S)-HNK.

Altogether, the rank-order of effectiveness in the forced swim test was determined to be

(2R,6S)-, (2S,6R)-, (2R,6R)-, and (2S,6S)-HNK, from most to least effective. Although these findings await further validation, these data are the first to highlight the potential superior potency of (2R,6S)-HNK.

The novel compound (5R)-methyl-(2R,6R)-HNK adopts a three-dimensional structure similar to (2R,6S)-HNK (Figure 5.2). In particular, the methyl group substitution at the C5 position causes (5R)-methyl-(2R,6R)-HNK to localize the aryl group equatorial to the cyclohexyl ring while the amine and hydroxyl groups are localized axial, similar to

(2R,6S)-HNK (Figure 5.2B) and in contrast with (2R,6R)-HNK (Figure 5.2A; see Morris et al., 2017b). Notably, (5R)-methyl-(2R,6R)-HNK reduced immobility time (minimum effective dose 1 mg/kg, i.p.; Figure 5.2D) with similar effectiveness as (2R,6S)-HNK, consistent with the hypothesis that the three-dimensional structure of (2R,6S)-HNK confers enhanced behavioral effectiveness. It is unclear whether (5R)-methyl-(2R,6R)-HNK will share the U-shaped dose response curve observed for (2R,6S)-HNK when tested at doses

≥30 mg/kg. Additionally, the range of doses tested did not allow the determination of a true minimally effective dose of (2R,6S)-HNK (which reduced immobility at the lowest dose tested, 1 mg/kg). Thus, future studies should aim to more fully characterize the dose- response of both of these compounds and to determine whether (5R)-methyl-(2R,6R)-HNK recapitulates the complete dose response characteristics of (2R,6S)-HNK.

There are several important limitations to the present study. First, only one behavioral outcome, immobility time in the forced swim test, was evaluated and only at a single time point. It is possible that the relative effectiveness of the (2,6)-HNKs may differ

98 in additional behavioral tests thought to be predictive of antidepressant efficacy or at different times relative to testing (e.g. it is possible that some HNKs may have longer- lasting actions or exert more robust effects when evaluated earlier or later relative to dosing). It is important for future studies to evaluate the effects of the (2,6)-HNKs as well as the other HNKs that have not yet been tested, in a variety of behavioral tests, including those that predict antidepressant effectiveness and in those that aim to characterize adverse behavioral effects. Second, only male mice were included in these experiments, in order to evaluate the relative effectiveness of the test compounds independent of the previously observed sex-dependent differences in their brain concentrations (described in Chapter 2).

Of note, total brain concentrations of (2R,6R)-, (2S,6S)-, and (2S,6R)-HNK were greater in female mice compared to males, while total brain concentrations of (2R,6S)-

HNK were greater in male than female mice (see Table 2.2). Thus, it is important for future studies to evaluate the behavioral effects of the (2,6)-HNKs in females, in order to evaluate how the sex-dependent pharmacokinetic differences may impact antidepressant-like behavioral effects, and also to determine if their rank-order of effectiveness is conserved between both sexes. Finally, the effects of the HNKs on biochemical and/or synaptic outcomes thought to underlie antidepressant-relevant behavioral effects are yet to be studied (discussed below).

In addition to the sex-dependent differences in their pharmacokinetics profiles, it is important to note that within-sex differences between the four (2,6)-HNKs have been observed, including differences in their peak and total brain concentrations (described in

Chapter 2). In particular, following systemic treatment with equivalent doses, peak and total brain levels of (2S,6S)-HNK are approximately 2-3-fold greater than those of (2R,6R)-

HNK in male mice (see Table 2.1). However, despite its greater brain concentrations,

(2S,6S)-HNK was less effective in reducing immobility time in the forced swim test, requiring three-fold higher doses to exert effects, compared to (2R,6R)-HNK (minimum effective doses of 30 vs. 10 mg/kg, i.p., respectively; Figure 5.1A-B). Thus, in light of the observed pharmacokinetic differences, it is possible that (2R,6R)-HNK is substantially more potent than (2S,6S)-HNK with respect to its engagement of targets in the brain. By the same token, it is possible that (2R,6S)- and (2S,6R)-HNK are even more potent compared to (2R,6R)-HNK, since both compounds exhibit greater behavioral effectiveness

(effective at 1 mg/kg, i.p. and 3 mg/kg, i.p., respectively), despite reaching lower brain concentrations. Namely, peak and total brain levels were approximately two-fold lower following (2R,6S)-HNK dosing and approximately three-fold lower following (2S,6R)-

HNK dosing, compared to equivalent dosing with (2R,6R)-HNK (see Table 2.1).

While these findings require further testing, it is possible that differences in three- dimensional structure of HNKs confer distinct potencies to engage targets underlying their behavioral effects. The precise mechanisms underlying the antidepressant-relevant effects of the HNKs are still being elucidated. However, several putative targets have been identified. Specifically, (2R,6R)-HNK has been shown to potentiate glutamatergic transmission via a facilitation of glutamate release (Chou et al., 2018; Pham et al., 2018;

Riggs et al., 2019; Ye et al., 2019; Zanos et al., 2016). HNK-induced facilitation glutamatergic transmission has also been associated with subsequent activation of intracellular signaling pathways, including the BDNF/TrkB (Fukumoto et al., 2019;

Lumsden et al., 2019; Zanos et al., 2016) and mTOR (Fukumoto et al., 2019; Lumsden et al., 2019; Paul et al., 2014) pathways, which can lead to increased expression of synaptic

100 proteins including AMPARs (Ho et al., 2018; Shaffer et al., 2019; Zanos et al., 2016) that ultimately promote a long-lasting increase in synaptic strength in mood-regulating brain regions (see Chapter 1). Future studies should evaluate the effects of (2R,6S)-, (2S,6R)-, and (5R)-methyl-(2R,6R)-HNK on these various biochemical and synaptic outcomes, and, importantly, compare their relative potencies to induce such effects, in order to understand:

1) how the potency to engage putative targets is related to the observed rank-order of behavioral effectiveness, and 2) how the three-dimensional structures of the (2,6)-HNKs modulates their effectiveness to engage these putative targets. Moreover, additional studies are needed to characterize the behavioral, biochemical, and synaptic actions of the (2,4)- and (2,5)-HNKs in order to fully understand the structure-function relationship underlying the antidepressant-relevant behavioral effects of the HNKs, which may inform the development of HNKs, or structurally similar novel compounds, as novel antidepressant drugs.

Of note, the present data are consistent with an NMDAR inhibition-independent mechanism contributing to the actions of the (2,6)-HNKs. In particular, (2S,6S)-HNK has been demonstrated to have moderate inhibitory actions on NMDARs (Ki = 7.34-21.19 µM) compared to ketamine (Ki = 0.25-1.06 µM; Lumsden et al., 2019; Moaddel et al., 2013;

Morris et al., 2017b). By contrast, (2R,6R)-HNK has markedly lower potency to bind or inhibit NMDARs (Ki > 100 µM) (Abbott & Popescu, 2020; Lumsden et al., 2019; Moaddel et al., 2013; Morris et al., 2017b), and does not meaningfully inhibit NMDARs at concentrations similar to those produced following the behaviorally active dose of 10 mg/kg, i.p. (estimated to be <20 µmol/kg in brain tissue and <10 µM in the extracellular hippocampal space; Lumsden et al., 2019).

101

The relatively low binding affinities of (2R,6S)- and (2S,6R)-HNK are similar to that of (2R,6R)-HNK (all were reported to be >100 µM; Morris et al., 2017b). Both compounds reduced immobility time in the forced swim test at doses ≤10 mg/kg, i.p., despite producing lower brain concentrations that similar doses of (2R,6R)-HNK, suggesting that the brain concentrations of (2R,6S)- and (2S,6R)-HNK necessary to induce antidepressant-like actions are even lower than those required for (2R,6R)-HNK, and, therefore, also below the threshold for NMDAR inhibition. Altogether, these findings suggest that (2R,6S)- and (2S,6R)-HNK exert antidepressant-relevant behavioral actions at brain concentrations that are insufficient to inhibit NMDARs, consistent with an NMDAR- independent mechanism of action. However, the mechanistic actions of (2R,6S)- and

(2S,6R)-HNK, in addition to the other HNKs, require further clarification.

Altogether, the data presented here demonstrate, for the first time, that the relative rank-order of effectiveness of the four (2,6)-HNKs to reduce forced swim test immobility time, from most to least effective, is (2R,6R)-, (2S,6R)-, (2R,6R)-, and (2S,6S)-HNK. This is the first indication that (2R,6S)-HNK has greater effectiveness compared to (2R,6R)-

HNK, exerting antidepressant-relevant behavioral effects at 10-fold lower doses, despite its lower relative brain concentrations. Consistent with the three-dimensional structure of

(2R,6S)-HNK conferring its enhanced effectiveness, the structurally similar, novel compound (5R)-methyl-(2R,6R)-HNK recapitulated the effectiveness of (2R,6S)-HNK.

While these compounds require further study, the data suggest that (2R,6S)-HNK, (5R)- methyl-(2R,6R)-HNK, or additional HNKs yet to be tested, may represent novel, potent drug candidates for the treatment of depression.

102

Chapter 6. Discussion and conclusions

The studies presented here aimed to address several important considerations for the development of HNKs as novel antidepressant treatments. First, the pharmacokinetic profiles of the 12 HNKs formed in vivo from ketamine were characterized. Robust differences in peak concentrations (6-fold differences in the plasma, 7-8-fold differences in the brain) and total levels (11-12-fold differences in the plasma, 9-12-fold differences in the brain) in the plasma and the brain were observed between the various HNKs, in addition to sex-dependent differences in peak and total concentrations that were identified for most

HNKs. However, all 12 HNKs readily penetrated the brain (with brain to plasma exposure ratios between 0.6-1.4) and shared similar rapid elimination profiles (half-lives between

0.4-0.8 h in the plasma and 0.3-0.8 h in the brain). These data provide the first in-depth assessment of the pharmacokinetics of the 12 HNKs, and, importantly, highlight the ability of all 12 HNKs to readily penetrate the brain suggesting that these compounds may be capable of exerting pharmacodynamic effects in the brain, including antidepressant- relevant actions.

Second, it was demonstrated that (2R,6R)-HNK has favorable oral bioavailability of approximately 50% (between 45-52%) in mice, which was not improved by an ester prodrug strategy. Notably, orally administered (2R,6R)-HNK exerted antidepressant- relevant behavioral actions (15-50 mg/kg) but did not exert overt adverse effects (up to 450 mg/kg, at least 9-fold fold higher than doses required for antidepressant-relevant effects) in mice. Taken together, these data may support the use of (2R,6R)-HNK as an oral antidepressant drug, although its antidepressant efficacy and oral bioavailability in humans awaits testing in clinical trials.

103

Additionally, sex-dependent differences in the pharmacokinetics of ketamine and

(2R,6R)-HNK were characterized. It was demonstrated that, compared to males, female mice have lower levels of ketamine and higher levels of HNK following ketamine administration, as well as higher levels of (2R,6R)-HNK following its direct administration, and that male gonadal hormones at least partly mediate these differences. While the precise mechanisms by which male gonadal hormones modulate metabolism and the impact of these metabolic differences on the behavioral effectiveness of ketamine and/or (2R,6R)-

HNK require further clarification, the observed sex-dependent pharmacokinetic differences may contribute to differences in antidepressant efficacy and/or adverse effect burden between the sexes, and highlight the importance of sex as a biological variable when designing and interpreting studies of ketamine and (2R,6R)-HNK.

Further, the relative behavioral effectiveness of the four (2,6)-HNKs was evaluated in the mouse forced swim test in male mice. The data demonstrated for the first time that both (2R,6S)-HNK (effective at 1 mg/kg, i.p.) and (2S,6R)-HNK (minimum effective dose

3 mg/kg, i.p.) exhibit greater effectiveness compared to either (2R,6R)-HNK (minimum effective dose 10 mg/kg, i.p.) or (2S,6S)-HNK (minimum effective dose 30 mg/kg, i.p.).

They also indicated a relative rank-order of effectiveness of (2R,6S)- > (2S,6R)- > (2R,6R)-

> (2S,6S)-HNK to reduce immobility time in the forced swim test. Interesting, enhanced behavioral effectiveness of (2R,6S)-HNK compared to (2S,6R)-HNK was observed despite

(2R,6S)-HNK dosing resulting in lower peak and total brain concentrations compared to equivalent dosing of (2R,6R)-HNK. This may suggest that, if these compounds were tested at equal concentrations and compared independent of pharmacokinetic differences (for example testing in in vitro models), (2R,6S)-HNK may demonstrate an even more robust

104 enhancement in potency to engage targets compared to (2R,6R)-HNK. However, this possibility requires further testing in future studies.

Finally, the novel compound (5R)-methyl-(2R,6R)-HNK, which shares a similar three-dimensional structure with (2R,6S)-HNK, recapitulated the behavioral effectiveness of (2R,6S)-HNK in the forced swim test, suggesting a critical role of three-dimensional structure in mediating antidepressant-relevant effectiveness in this assay. While these compounds require further study, including additional evaluations of their antidepressant- relevant effects in other behavioral tests, characterization of potential adverse effects, and assessment of their mechanisms of action, the data suggest that (2R,6S)-HNK, (5R)- methyl-(2R,6R)-HNK, or additional HNKs yet to be tested, may represent novel, potent, antidepressant drug candidates. Notably, it is possible that (2R,6S)-HNK and/or (5R)- methyl-(2R,6R)-HNK may have greater potency to engage targets involved in mediating their antidepressant-relevant behavioral effects. Understanding the how the relative potencies to engage putative antidepressant targets is related to behavioral effectiveness may provide valuable insights into the mechanisms of action underlying the effects of

HNKs.

Future studies are needed to validate and expand upon these findings. These should include studies focused on: (i) identifying the mechanisms underlying the antidepressant- relevant behavioral actions observed here, (ii) evaluating the full range of behavioral actions including potential adverse effects of the (2,6)-HNKs, (iii) understanding the mechanisms underlying sex-dependent differences in the pharmacokinetic profile of ketamine and HNKs and how these differences impact behavioral outcomes, and (iv)

105 characterizing the behavioral actions of the (2,4)-, and (2,5)-HNKs that were not tested here.

The present studies provide important insights into the pharmacokinetic profiles and behavioral effects of the HNKs, particularly, the (2,6)-HNKs, and lay the foundation for additional studies to elucidate the potential clinical relevance of HNKs as novel antidepressant treatments.

106

References

Abbott, J. A., & Popescu, G. K. (2020). Hydroxynorketamine Blocks NMDA Receptor Currents by Binding to Closed Receptors. Mol Pharmacol. https://doi.org/10.1124/mol.120.119784

Adams, J. D., Jr., Baillie, T. A., Trevor, A. J., & Castagnoli, N., Jr. (1981). Studies on the biotransformation of ketamine. 1-Identification of metabolites produced in vitro from rat liver microsomal preparations. Biomed Mass Spectrom, 8(11), 527-538. http://www.ncbi.nlm.nih.gov/pubmed/7317567

Ago, Y., Tanabe, W., Higuchi, M., Tsukada, S., Tanaka, T., Yamaguchi, T., Igarashi, H., Yokoyama, R., Seiriki, K., Kasai, A., Nakazawa, T., Nakagawa, S., Hashimoto, K., & Hashimoto, H. (2019). (R)-Ketamine Induces a Greater Increase in Prefrontal 5-HT Release Than (S)-Ketamine and Ketamine Metabolites via an AMPA Receptor-Independent Mechanism. Int J Neuropsychopharmacol, 22(10), 665-674. https://doi.org/10.1093/ijnp/pyz041

Ates-Alagoz, Z., & Adejare, A. (2013). NMDA Receptor Antagonists for Treatment of Depression. Pharmaceuticals (Basel), 6(4), 480-499. https://doi.org/10.3390/ph6040480

Autry, A. E., Adachi, M., Nosyreva, E., Na, E. S., Los, M. F., Cheng, P. F., Kavalali, E. T., & Monteggia, L. M. (2011). NMDA receptor blockade at rest triggers rapid behavioural antidepressant responses. Nature, 475(7354), 91-95. https://doi.org/10.1038/nature10130

Barre, L., Fournel-Gigleux, S., Finel, M., Netter, P., Magdalou, J., & Ouzzine, M. (2007). Substrate specificity of the human UDP-glucuronosyltransferase UGT2B4 and UGT2B7. Identification of a critical aromatic amino acid residue at position 33. FEBS J, 274(5), 1256-1264. https://doi.org/10.1111/j.1742-4658.2007.05670.x

Berman, R. M., Cappiello, A., Anand, A., Oren, D. A., Heninger, G. R., Charney, D. S., & Krystal, J. H. (2000). Antidepressant effects of ketamine in depressed patients. Biol Psychiatry, 47(4), 351-354. https://doi.org/S0006-3223(99)00230-9 [pii]

Bowen, R. S., Ferguson, D. P., & Lightfoot, J. T. (2011). Effects of Aromatase Inhibition on the Physical Activity Levels of Male Mice. J Steroids Horm Sci, 1(1), 1-7. https://doi.org/10.4172/2157-7536.S1-001

Can, A., Zanos, P., Moaddel, R., Kang, H. J., Dossou, K. S., Wainer, I. W., Cheer, J. F., Frost, D. O., Huang, X. P., & Gould, T. D. (2016). Effects of Ketamine and Ketamine Metabolites on Evoked Striatal Dopamine Release, Dopamine Receptors, and Monoamine Transporters. J Pharmacol Exp Ther, 359(1), 159- 170. https://doi.org/10.1124/jpet.116.235838

107

Carrier, N., & Kabbaj, M. (2013). Sex differences in the antidepressant-like effects of ketamine. Neuropharmacology, 70, 27-34. https://doi.org/10.1016/j.neuropharm.2012.12.009

Cavalleri, L., Merlo Pich, E., Millan, M. J., Chiamulera, C., Kunath, T., Spano, P. F., & Collo, G. (2018). Ketamine enhances structural plasticity in mouse mesencephalic and human iPSC-derived dopaminergic neurons via AMPAR-driven BDNF and mTOR signaling. Mol Psychiatry, 23(4), 812-823. https://doi.org/10.1038/mp.2017.241 Chaki, S., & Fukumoto, K. (2018). mGlu receptors as potential targets for novel antidepressants. Curr Opin Pharmacol, 38, 24-30. https://doi.org/10.1016/j.coph.2018.02.001

Chang, T., & Glazko, A. J. (1974). Biotransformation and disposition of ketamine. Int Anesthesiol Clin, 12(2), 157-177. http://www.ncbi.nlm.nih.gov/pubmed/4603048

Chen, B. K., Luna, V. M., LaGamma, C. T., Xu, X., Deng, S. X., Suckow, R. F., Cooper, T. B., Shah, A., Brachman, R. A., Mendez-David, I., David, D. J., Gardier, A. M., Landry, D. W., & Denny, C. A. (2020). Sex-specific neurobiological actions of prophylactic (R,S)-ketamine, (2R,6R)-hydroxynorketamine, and (2S,6S)- hydroxynorketamine. Neuropsychopharmacology, 45(9), 1545-1556. https://doi.org/10.1038/s41386-020-0714-z

Chen, W. Y., Huang, M. C., & Lin, S. K. (2014). Gender differences in subjective discontinuation symptoms associated with ketamine use. Subst Abuse Treat Prev Policy, 9, 39. https://doi.org/10.1186/1747-597X-9-39

Chong, C., Schug, S. A., Page-Sharp, M., Jenkins, B., & Ilett, K. F. (2009). Development of a sublingual/oral formulation of ketamine for use in neuropathic pain: Preliminary findings from a three-way randomized, crossover study. Clin Drug Investig, 29(5), 317-324. https://doi.org/10.2165/00044011-200929050-00004 Chou, D. (2020). Brain-derived neurotrophic factor in the ventrolateral periaqueductal gray contributes to (2R,6R)-hydroxynorketamine-mediated actions. Neuropharmacology, 170, 108068. https://doi.org/https://doi.org/10.1016/j.neuropharm.2020.108068

Chou, D., Peng, H. Y., Lin, T. B., Lai, C. Y., Hsieh, M. C., Wen, Y. C., Lee, A. S., Wang, H. H., Yang, P. S., Chen, G. D., & Ho, Y. C. (2018). (2R,6R)- hydroxynorketamine rescues chronic stress-induced depression-like behavior through its actions in the midbrain periaqueductal gray. Neuropharmacology, 139, 1-12. https://doi.org/S0028-3908(18)30335-6 [pii] Clements, J. A., Nimmo, W. S., & Grant, I. S. (1982). Bioavailability, pharmacokinetics, and analgesic activity of ketamine in humans. J Pharm Sci, 71(5), 539-542. https://doi.org/S0022-3549(15)44170-X [pii]

108

Collo, G., Cavalleri, L., Chiamulera, C., & Merlo Pich, E. (2018). (2R,6R)- Hydroxynorketamine promotes dendrite outgrowth in human inducible pluripotent stem cell-derived neurons through AMPA receptor with timing and exposure compatible with ketamine infusion pharmacokinetics in humans. Neuroreport, 29(16), 1425-1430. https://doi.org/10.1097/WNR.0000000000001131

Coyle, C. M., & Laws, K. R. (2015). The use of ketamine as an antidepressant: a systematic review and meta-analysis. Hum Psychopharmacol, 30(3), 152-163. https://doi.org/10.1002/hup.2475

Cyriac, J. M., & James, E. (2014). Switch over from intravenous to oral therapy: A concise overview. J Pharmacol Pharmacother, 5(2), 83-87. https://doi.org/10.4103/0976-500X.130042 Daan, S., Damassa, D., Pittendrigh, C. S., & Smith, E. R. (1975). An effect of castration and testosterone replacement on a circadian pacemaker in mice (Mus musculus). Proc Natl Acad Sci U S A, 72(9), 3744-3747. https://doi.org/10.1073/pnas.72.9.3744

Derntl, B., Hornung, J., Sen, Z. D., Colic, L., Li, M., & Walter, M. (2019). Interaction of Sex and Age on the Dissociative Effects of Ketamine Action in Young Healthy Participants. Front Neurosci, 13, 616. https://doi.org/10.3389/fnins.2019.00616

Desta, Z., Moaddel, R., Ogburn, E. T., Xu, C., Ramamoorthy, A., Venkata, S. L., Sanghvi, M., Goldberg, M. E., Torjman, M. C., & Wainer, I. W. (2012). Stereoselective and regiospecific hydroxylation of ketamine and norketamine. Xenobiotica, 42(11), 1076-1087. https://doi.org/10.3109/00498254.2012.685777

DiazGranados, N., Ibrahim, L. A., Brutsche, N. E., Ameli, R., Henter, I. D., Luckenbaugh, D. A., Machado-Vieira, R., & Zarate, C. A., Jr. (2010). Rapid resolution of suicidal ideation after a single infusion of an N-methyl-D-aspartate antagonist in patients with treatment-resistant major depressive disorder. J Clin Psychiatry, 71(12), 1605-1611. https://doi.org/10.4088/JCP.09m05327blu

Dinis-Oliveira, R. J. (2017). Metabolism and metabolomics of ketamine: a toxicological approach. Forensic Sci Res, 2(1), 2-10. https://doi.org/10.1080/20961790.2017.1285219

Dossat, A. M., Wright, K. N., Strong, C. E., & Kabbaj, M. (2018). Behavioral and biochemical sensitivity to low doses of ketamine: Influence of estrous cycle in C57BL/6 mice. Neuropharmacology, 130, 30-41. https://doi.org/10.1016/j.neuropharm.2017.11.022

Duhamel, M. C., Troncy, E., & Beaudry, F. (2010). Metabolic stability and determination of cytochrome P450 isoenzymes' contribution to the metabolism of medetomidine in dog liver microsomes. Biomed Chromatogr, 24(8), 868-877. https://doi.org/10.1002/bmc.1379

109

Duman, R. S., Heninger, G. R., & Nestler, E. J. (1997). A molecular and cellular theory of depression. Arch Gen Psychiatry, 54(7), 597-606. https://doi.org/10.1001/archpsyc.1997.01830190015002

Elmer, G. I., Tapocik, J. D., Mayo, C. L., Zanos, P., & Gould, T. D. (2020). Ketamine metabolite (2R,6R)-hydroxynorketamine reverses behavioral despair produced by adolescent trauma. Pharmacol Biochem Behav, 172973. https://doi.org/10.1016/j.pbb.2020.172973

Faccio, A. T., Ruperez, F. J., Singh, N. S., Angulo, S., Tavares, M. F. M., Bernier, M., Barbas, C., & Wainer, I. W. (2018). Stereochemical and structural effects of (2R,6R)-hydroxynorketamine on the mitochondrial metabolome in PC-12 cells. Biochim Biophys Acta Gen Subj, 1862(6), 1505-1515. https://doi.org/10.1016/j.bbagen.2018.03.008

Farmer, C. A., Gilbert, J. R., Moaddel, R., George, J., Adeojo, L., Lovett, J., Nugent, A. C., Kadriu, B., Yuan, P., Gould, T. D., Park, L. T., & Zarate, C. A., Jr. (2020). Ketamine metabolites, clinical response, and gamma power in a randomized, placebo-controlled, crossover trial for treatment-resistant major depression. Neuropsychopharmacology. https://doi.org/10.1038/s41386-020-0663-6

Fassauer, G. M., Hofstetter, R., Hasan, M., Oswald, S., Modess, C., Siegmund, W., & Link, A. (2017). Ketamine metabolites with antidepressant effects: Fast, economical, and eco-friendly enantioselective separation based on supercritical- fluid chromatography (SFC) and single quadrupole MS detection. J Pharm Biomed Anal, 146, 410-419. https://doi.org/10.1016/j.jpba.2017.09.007

Fischell, J., Van Dyke, A. M., Kvarta, M. D., LeGates, T. A., & Thompson, S. M. (2015). Rapid Antidepressant Action and Restoration of Excitatory Synaptic Strength After Chronic Stress by Negative Modulators of Alpha5-Containing GABAA Receptors. Neuropsychopharmacology, 40(11), 2499-2509. https://doi.org/10.1038/npp.2015.112

Franceschelli, A., Sens, J., Herchick, S., Thelen, C., & Pitychoutis, P. M. (2015). Sex differences in the rapid and the sustained antidepressant-like effects of ketamine in stress-naive and "depressed" mice exposed to chronic mild stress. Neuroscience, 290, 49-60. https://doi.org/10.1016/j.neuroscience.2015.01.008

Fred, S. M., Laukkanen, L., Brunello, C. A., Vesa, L., Goos, H., Cardon, I., Moliner, R., Maritzen, T., Varjosalo, M., Casarotto, P. C., & Castren, E. (2019). Pharmacologically diverse antidepressants facilitate TRKB receptor activation by disrupting its interaction with the endocytic adaptor complex AP-2. J Biol Chem, 294(48), 18150-18161. https://doi.org/10.1074/jbc.RA119.008837

Freeman, M. P., Papakostas, G. I., Hoeppner, B., Mazzone, E., Judge, H., Cusin, C., Mathew, S., Sanacora, G., Iosifescu, D., DeBattista, C., Trivedi, M. H., & Fava, M. (2019). Sex differences in response to ketamine as a rapidly acting

110

intervention for treatment resistant depression. J Psychiatr Res, 110, 166-171. https://doi.org/10.1016/j.jpsychires.2019.01.010

Fukumoto, K., Fogaca, M. V., Liu, R. J., Duman, C., Kato, T., Li, X. Y., & Duman, R. S. (2019). Activity-dependent brain-derived neurotrophic factor signaling is required for the antidepressant actions of (2R,6R)-hydroxynorketamine. Proc Natl Acad Sci U S A, 116(1), 297-302. https://doi.org/10.1073/pnas.1814709116

Gaynes, B. N., Warden, D., Trivedi, M. H., Wisniewski, S. R., Fava, M., & Rush, A. J. (2009). What did STAR*D teach us? Results from a large-scale, practical, clinical trial for patients with depression. Psychiatr Serv, 60(11), 1439-1445. https://doi.org/10.1176/ps.2009.60.11.1439 George, D., Galvez, V., Martin, D., Kumar, D., Leyden, J., Hadzi-Pavlovic, D., Harper, S., Brodaty, H., Glue, P., Taylor, R., Mitchell, P. B., & Loo, C. K. (2017). Pilot Randomized Controlled Trial of Titrated Subcutaneous Ketamine in Older Patients with Treatment-Resistant Depression. Am J Geriatr Psychiatry, 25(11), 1199-1209. https://doi.org/10.1016/j.jagp.2017.06.007

Gideons, E. S., Kavalali, E. T., & Monteggia, L. M. (2014). Mechanisms underlying differential effectiveness of memantine and ketamine in rapid antidepressant responses. Proc Natl Acad Sci U S A, 111(23), 8649-8654. https://doi.org/10.1073/pnas.1323920111

Gochfeld, M. (2017). Sex Differences in Human and Animal Toxicology. Toxicol Pathol, 45(1), 172-189. https://doi.org/10.1177/0192623316677327

Grant, I. S., Nimmo, W. S., & Clements, J. A. (1981). Pharmacokinetics and analgesic effects of i.m. and oral ketamine. Br J Anaesth, 53(8), 805-810. https://doi.org/S0007-0912(17)46897-8 [pii]

Grunebaum, M. F., Galfalvy, H. C., Choo, T. H., Parris, M. S., Burke, A. K., Suckow, R. F., Cooper, T. B., & Mann, J. J. (2019). Ketamine metabolite pilot study in a suicidal depression trial. J Psychiatr Res, 117, 129-134. https://doi.org/10.1016/j.jpsychires.2019.08.005

Hamada, Y. (2017). Recent progress in prodrug design strategies based on generally applicable modifications. Bioorg Med Chem Lett, 27(8), 1627-1632. https://doi.org/S0960-894X(17)30218-4 [pii] Harada, N., Hanaoka, R., Horiuchi, H., Kitakaze, T., Mitani, T., Inui, H., & Yamaji, R. (2016). Castration influences intestinal microflora and induces abdominal obesity in high-fat diet-fed mice. Sci Rep, 6, 23001. https://doi.org/10.1038/srep23001

Hardeveld, F., Spijker, J., De Graaf, R., Nolen, W. A., & Beekman, A. T. (2010). Prevalence and predictors of recurrence of major depressive disorder in the adult population. Acta Psychiatr Scand, 122(3), 184-191. https://doi.org/10.1111/j.1600-0447.2009.01519.x

111

Hasan, M., Hofstetter, R., Fassauer, G. M., Link, A., Siegmund, W., & Oswald, S. (2017). Quantitative chiral and achiral determination of ketamine and its metabolites by LC-MS/MS in human serum, urine and fecal samples. J Pharm Biomed Anal, 139, 87-97. https://doi.org/10.1016/j.jpba.2017.02.035

Herzog, D. P., Mellema, R. M., Remmers, F., Lutz, B., Muller, M. B., & Treccani, G. (2020). Sexually Dimorphic Behavioral Profile in a Transgenic Model Enabling Targeted Recombination in Active Neurons in Response to Ketamine and (2R,6R)-Hydroxynorketamine Administration. Int J Mol Sci, 21(6). https://doi.org/10.3390/ijms21062142

Highland, J. N., Morris, P. J., Zanos, P., Lovett, J., Ghosh, S., Wang, A. Q., Zarate, C. A., Jr., Thomas, C. J., Moaddel, R., & Gould, T. D. (2018). Mouse, rat, and dog bioavailability and mouse oral antidepressant efficacy of ( 2R,6R)- hydroxynorketamine. J Psychopharmacol, 269881118812095. https://doi.org/10.1177/0269881118812095

Hillhouse, T. M., Rice, R., & Porter, J. H. (2019). What role does the (2R,6R)- hydronorketamine metabolite play in the antidepressant-like and abuse-related effects of (R)-ketamine? Br J Pharmacol, 176(19), 3886-3888. https://doi.org/10.1111/bph.14785

Ho, M. F., Correia, C., Ingle, J. N., Kaddurah-Daouk, R., Wang, L., Kaufmann, S. H., & Weinshilboum, R. M. (2018). Ketamine and ketamine metabolites as novel estrogen receptor ligands: Induction of cytochrome P450 and AMPA glutamate receptor gene expression. Biochem Pharmacol, 152, 279-292. https://doi.org/10.1016/j.bcp.2018.03.032

Ho, M. F., Zhang, C., Zhang, L., Li, H., & Weinshilboum, R. M. (2019). Ketamine and Active Ketamine Metabolites Regulate STAT3 and the Type I Interferon Pathway in Human Microglia: Molecular Mechanisms Linked to the Antidepressant Effects of Ketamine. Front Pharmacol, 10, 1302. https://doi.org/10.3389/fphar.2019.01302

Homayoun, H., & Moghaddam, B. (2007). NMDA receptor hypofunction produces opposite effects on prefrontal cortex interneurons and pyramidal neurons. J Neurosci, 27(43), 11496-11500. https://doi.org/10.1523/JNEUROSCI.2213- 07.2007

Ibrahim, L., Diaz Granados, N., Jolkovsky, L., Brutsche, N., Luckenbaugh, D. A., Herring, W. J., Potter, W. Z., & Zarate, C. A., Jr. (2012). A Randomized, placebo- controlled, crossover pilot trial of the oral selective NR2B antagonist MK-0657 in patients with treatment-resistant major depressive disorder. J Clin Psychopharmacol, 32(4), 551-557. https://doi.org/10.1097/JCP.0b013e31825d70d6

Idris, A. I. (2012). Ovariectomy/orchidectomy in rodents. Methods Mol Biol, 816, 545- 551. https://doi.org/10.1007/978-1-61779-415-5_34 112

Inoue, T., Zakikhani, M., David, S., Algire, C., Blouin, M. J., & Pollak, M. (2010). Effects of castration on insulin levels and glucose tolerance in the mouse differ from those in man. Prostate, 70(15), 1628-1635. https://doi.org/10.1002/pros.21198

Insel, T. R., & Wang, P. S. (2009). The STAR*D trial: revealing the need for better treatments. Psychiatr Serv, 60(11), 1466-1467. https://doi.org/10.1176/appi.ps.60.11.1466 Kandel, S. E., Han, L. W., Mao, Q., & Lampe, J. N. (2017). Digging Deeper into CYP3A Testosterone Metabolism: Kinetic, Regioselectivity, and Stereoselectivity Differences between CYP3A4/5 and CYP3A7. Drug Metab Dispos, 45(12), 1266- 1275. https://doi.org/10.1124/dmd.117.078055

Kelley, A. E. (2001). Measurement of rodent stereotyped behavior. Curr Protoc Neurosci, Chapter 8, Unit 8 8. https://doi.org/10.1002/0471142301.ns0808s04

Kennedy, S. H. (2008). Core symptoms of major depressive disorder: relevance to diagnosis and treatment. Dialogues Clin Neurosci, 10(3), 271-277. https://www.ncbi.nlm.nih.gov/pubmed/18979940

Kessler, R. C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K. R., Rush, A. J., Walters, E. E., & Wang, P. S. (2003). The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA, 289(23), 3095-3105. https://doi.org/10.1001/jama.289.23.3095 Kharasch, E. D., & Labroo, R. (1992). Metabolism of ketamine stereoisomers by human liver microsomes. Anesthesiology, 77(6), 1201-1207. https://doi.org/10.1097/00000542-199212000-00022

Krauser, J. A., Voehler, M., Tseng, L. H., Schefer, A. B., Godejohann, M., & Guengerich, F. P. (2004). Testosterone 1 beta-hydroxylation by human cytochrome P450 3A4. Eur J Biochem, 271(19), 3962-3969. https://doi.org/10.1111/j.1432-1033.2004.04339.x

Kroin, J. S., Das, V., Moric, M., & Buvanendran, A. (2019). Efficacy of the ketamine metabolite (2R,6R)-hydroxynorketamine in mice models of pain. Reg Anesth Pain Med, 44(1), 111-117. https://doi.org/10.1136/rapm-2018-000013

Krotkiewski, M., Kral, J. G., & Karlsson, J. (1980). Effects of castration and testosterone substitution on body composition and muscle metabolism in rats. Acta Physiol Scand, 109(3), 233-237. https://doi.org/10.1111/j.1748-1716.1980.tb06592.x

Krystal, J. H., Karper, L. P., Seibyl, J. P., Freeman, G. K., Delaney, R., Bremner, J. D., Heninger, G. R., Bowers, M. B., Jr., & Charney, D. S. (1994). Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans. Psychotomimetic, perceptual, cognitive, and neuroendocrine responses. Arch Gen Psychiatry, 51(3), 199-214. http://www.ncbi.nlm.nih.gov/pubmed/8122957

113

Lankveld, D. P., Driessen, B., Soma, L. R., Moate, P. J., Rudy, J., Uboh, C. E., van Dijk, P., & Hellebrekers, L. J. (2006). Pharmacodynamic effects and pharmacokinetic profile of a long-term continuous rate infusion of racemic ketamine in healthy conscious horses. J Vet Pharmacol Ther, 29(6), 477-488. https://doi.org/10.1111/j.1365-2885.2006.00794.x

Lapidus, K. A., Levitch, C. F., Perez, A. M., Brallier, J. W., Parides, M. K., Soleimani, L., Feder, A., Iosifescu, D. V., Charney, D. S., & Murrough, J. W. (2014). A randomized controlled trial of intranasal ketamine in major depressive disorder. Biol Psychiatry, 76(12), 970-976. https://doi.org/10.1016/j.biopsych.2014.03.026 Lara, D. R., Bisol, L. W., & Munari, L. R. (2013). Antidepressant, mood stabilizing and procognitive effects of very low dose sublingual ketamine in refractory unipolar and bipolar depression. Int J Neuropsychopharmacol, 16(9), 2111-2117. https://doi.org/10.1017/S1461145713000485 Leung, L. Y., & Baillie, T. A. (1986). Comparative pharmacology in the rat of ketamine and its two principal metabolites, norketamine and (Z)-6-hydroxynorketamine. J Med Chem, 29(11), 2396-2399. http://www.ncbi.nlm.nih.gov/pubmed/3783598

Li, N., Lee, B., Liu, R. J., Banasr, M., Dwyer, J. M., Iwata, M., Li, X. Y., Aghajanian, G., & Duman, R. S. (2010). mTOR-dependent synapse formation underlies the rapid antidepressant effects of NMDA antagonists. Science, 329(5994), 959-964. https://doi.org/10.1126/science.1190287 Lilius, T. O., Viisanen, H., Jokinen, V., Niemi, M., Kalso, E. A., & Rauhala, P. V. (2018). Interactions of (2S,6S;2R,6R)-Hydroxynorketamine, a Secondary Metabolite of (R,S)-Ketamine, with Morphine. Basic Clin Pharmacol Toxicol, 122(5), 481-488. https://doi.org/10.1111/bcpt.12941

Loo, C. K., Galvez, V., O'Keefe, E., Mitchell, P. B., Hadzi-Pavlovic, D., Leyden, J., Harper, S., Somogyi, A. A., Lai, R., Weickert, C. S., & Glue, P. (2016). Placebo- controlled pilot trial testing dose titration and intravenous, intramuscular and subcutaneous routes for ketamine in depression. Acta Psychiatr Scand, 134(1), 48-56. https://doi.org/10.1111/acps.12572

Lumsden, E. W., Troppoli, T. A., Myers, S. J., Zanos, P., Aracava, Y., Kehr, J., Lovett, J., Kim, S., Wang, F. H., Schmidt, S., Jenne, C. E., Yuan, P., Morris, P. J., Thomas, C. J., Zarate, C. A., Jr., Moaddel, R., Traynelis, S. F., Pereira, E. F. R., Thompson, S. M., Albuquerque, E. X., & Gould, T. D. (2019). Antidepressant- relevant concentrations of the ketamine metabolite (2R,6R)-hydroxynorketamine do not block NMDA receptor function. Proc Natl Acad Sci U S A, 116(11), 5160- 5169. https://doi.org/10.1073/pnas.1816071116

Maenpaa, J., Pelkonen, O., Cresteil, T., & Rane, A. (1993). The role of cytochrome P450 3A (CYP3A) isoform(s) in oxidative metabolism of testosterone and benzphetamine in human adult and fetal liver. J Steroid Biochem Mol Biol, 44(1), 61-67. https://doi.org/10.1016/0960-0760(93)90152-m

114

Martignoni, M., Groothuis, G. M., & de Kanter, R. (2006). Species differences between mouse, rat, dog, monkey and human CYP-mediated drug metabolism, inhibition and induction. Expert Opin Drug Metab Toxicol, 2(6), 875-894. https://doi.org/10.1517/17425255.2.6.875

Martinez-Lozano Sinues, P., Kohler, M., Brown, S. A., Zenobi, R., & Dallmann, R. (2017). Gauging circadian variation in ketamine metabolism by real-time breath analysis. Chem Commun (Camb), 53(14), 2264-2267. https://doi.org/10.1039/c6cc09061c

Masui, K., & Tamura, Y. (1926). The Effect of Gonadectomy on the Weight of the Kidney, Thymus, and Spleen of Mice. Journal of Experimental Biology, 3(3), 207-223. https://jeb.biologists.org/content/jexbio/3/3/207.full.pdf

Michaelsson, H., Andersson, M., Svensson, J., Karlsson, L., Ehn, J., Culley, G., Engstrom, A., Bergstrom, N., Savvidi, P., Kuhn, H. G., Hanse, E., & Seth, H. (2019). The novel antidepressant ketamine enhances dentate gyrus proliferation with no effects on synaptic plasticity or hippocampal function in depressive-like rats. Acta Physiol (Oxf), 225(4), e13211. https://doi.org/10.1111/apha.13211

Miller, O. H., Yang, L., Wang, C. C., Hargroder, E. A., Zhang, Y., Delpire, E., & Hall, B. J. (2014). GluN2B-containing NMDA receptors regulate depression-like behavior and are critical for the rapid antidepressant actions of ketamine. Elife, 3, e03581. https://doi.org/10.7554/eLife.03581

Moaddel, R., Abdrakhmanova, G., Kozak, J., Jozwiak, K., Toll, L., Jimenez, L., Rosenberg, A., Tran, T., Xiao, Y., Zarate, C. A., & Wainer, I. W. (2013). Sub- anesthetic concentrations of (R,S)-ketamine metabolites inhibit acetylcholine- evoked currents in alpha7 nicotinic acetylcholine receptors. Eur J Pharmacol, 698(1-3), 228-234. https://doi.org/10.1016/j.ejphar.2012.11.023 Moaddel, R., Sanghvi, M., Dossou, K. S., Ramamoorthy, A., Green, C., Bupp, J., Swezey, R., O'Loughlin, K., & Wainer, I. W. (2015). The distribution and clearance of (2S,6S)-hydroxynorketamine, an active ketamine metabolite, in Wistar rats. Pharmacol Res Perspect, 3(4), e00157. https://doi.org/10.1002/prp2.157

Moaddel, R., Sanghvi, M., Ramamoorthy, A., Jozwiak, K., Singh, N., Green, C., O'Loughlin, K., Torjman, M., & Wainer, I. W. (2016). Subchronic administration of (R,S)-ketamine induces ketamine ring hydroxylation in Wistar rats. J Pharm Biomed Anal, 127, 3-8. https://doi.org/10.1016/j.jpba.2016.03.030 Moaddel, R., Venkata, S. L., Tanga, M. J., Bupp, J. E., Green, C. E., Iyer, L., Furimsky, A., Goldberg, M. E., Torjman, M. C., & Wainer, I. W. (2010). A parallel chiral- achiral liquid chromatographic method for the determination of the stereoisomers of ketamine and ketamine metabolites in the plasma and urine of patients with complex regional pain syndrome. Talanta, 82(5), 1892-1904. https://doi.org/10.1016/j.talanta.2010.08.005

115

Moda-Sava, R. N., Murdock, M. H., Parekh, P. K., Fetcho, R. N., Huang, B. S., Huynh, T. N., Witztum, J., Shaver, D. C., Rosenthal, D. L., Alway, E. J., Lopez, K., Meng, Y., Nellissen, L., Grosenick, L., Milner, T. A., Deisseroth, K., Bito, H., Kasai, H., & Liston, C. (2019). Sustained rescue of prefrontal circuit dysfunction by antidepressant-induced spine formation. Science, 364(6436). https://doi.org/10.1126/science.aat8078

Moghaddam, B., Adams, B., Verma, A., & Daly, D. (1997). Activation of glutamatergic neurotransmission by ketamine: a novel step in the pathway from NMDA receptor blockade to dopaminergic and cognitive disruptions associated with the prefrontal cortex. J Neurosci, 17(8), 2921-2927. https://www.ncbi.nlm.nih.gov/pubmed/9092613

Morris, P. J., Moaddel, R., Zanos, P., Moore, C. E., Gould, T., Zarate, C. A., Jr., & Thomas, C. J. (2017a). Synthesis and N-Methyl-d-aspartate (NMDA) Receptor Activity of Ketamine Metabolites. Org Lett, 19(17), 4572-4575. https://doi.org/10.1021/acs.orglett.7b02177

Morris, P. J., Moaddel, R., Zanos, P., Moore, C. E., Gould, T. D., Zarate, C. A., Jr., & Thomas, C. J. (2017b). Synthesis and N-Methyl-d-aspartate (NMDA) Receptor Activity of Ketamine Metabolites. Org Lett, 19(17), 4572-4575. https://doi.org/10.1021/acs.orglett.7b02177

Nelson, R. J., & Trainor, B. C. (2007). Neural mechanisms of aggression. Nat Rev Neurosci, 8(7), 536-546. https://doi.org/10.1038/nrn2174

Niciu, M. J., Luckenbaugh, D. A., Ionescu, D. F., Guevara, S., Machado-Vieira, R., Richards, E. M., Brutsche, N. E., Nolan, N. M., & Zarate, C. A., Jr. (2014). Clinical predictors of ketamine response in treatment-resistant major depression. J Clin Psychiatry, 75(5), e417-423. https://doi.org/10.4088/JCP.13m08698

Nosyreva, E., Szabla, K., Autry, A. E., Ryazanov, A. G., Monteggia, L. M., & Kavalali, E. T. (2013). Acute suppression of spontaneous neurotransmission drives synaptic potentiation. J Neurosci, 33(16), 6990-7002. https://doi.org/10.1523/JNEUROSCI.4998-12.2013

Ornoy, A., Giron, S., Aner, R., Goldstein, M., Boyan, B. D., & Schwartz, Z. (1994). Gender dependent effects of testosterone and 17 beta-estradiol on bone growth and modelling in young mice. Bone Miner, 24(1), 43-58. https://doi.org/10.1016/s0169-6009(08)80130-4

Paul, R. K., Singh, N. S., Khadeer, M., Moaddel, R., Sanghvi, M., Green, C. E., O'Loughlin, K., Torjman, M. C., Bernier, M., & Wainer, I. W. (2014). (R,S)- Ketamine metabolites (R,S)-norketamine and (2S,6S)-hydroxynorketamine increase the mammalian target of rapamycin function. Anesthesiology, 121(1), 149-159. https://doi.org/10.1097/ALN.0000000000000285

116

Pham, T. H., Defaix, C., Xu, X., Deng, S. X., Fabresse, N., Alvarez, J. C., Landry, D. W., Brachman, R. A., Denny, C. A., & Gardier, A. M. (2018). Common Neurotransmission Recruited in (R,S)-Ketamine and (2R,6R)- Hydroxynorketamine-Induced Sustained Antidepressant-like Effects. Biol Psychiatry, 84(1), e3-e6. https://doi.org/10.1016/j.biopsych.2017.10.020

Preskorn, S. H., Baker, B., Kolluri, S., Menniti, F. S., Krams, M., & Landen, J. W. (2008). An innovative design to establish proof of concept of the antidepressant effects of the NR2B subunit selective N-methyl-D-aspartate antagonist, CP- 101,606, in patients with treatment-refractory major depressive disorder. J Clin Psychopharmacol, 28(6), 631-637. https://doi.org/10.1097/JCP.0b013e31818a6cea

Price, R. B., Iosifescu, D. V., Murrough, J. W., Chang, L. C., Al Jurdi, R. K., Iqbal, S. Z., Soleimani, L., Charney, D. S., Foulkes, A. L., & Mathew, S. J. (2014). Effects of ketamine on explicit and implicit suicidal cognition: a randomized controlled trial in treatment-resistant depression. Depress Anxiety, 31(4), 335-343. https://doi.org/10.1002/da.22253

Racagni, G., & Popoli, M. (2008). Cellular and molecular mechanisms in the long-term action of antidepressants. Dialogues Clin Neurosci, 10(4), 385-400. https://www.ncbi.nlm.nih.gov/pubmed/19170396

Rahman, S. U., Hao, Q., He, K., Li, Y., Yang, X., Ye, T., Ali, T., Zhou, Q., & Li, S. (2020). Proteomic Study Reveals the Involvement of Energy Metabolism in the Fast Antidepressant Effect of (2R, 6R)-Hydroxy Norketamine. Proteomics Clin Appl, e1900094. https://doi.org/10.1002/prca.201900094

Reddy, M. S. (2010). Depression: the disorder and the burden. Indian J Psychol Med, 32(1), 1-2. https://doi.org/10.4103/0253-7176.70510 Riggs, L. M., Aracava, Y., Zanos, P., Fischell, J., Albuquerque, E. X., Pereira, E. F. R., Thompson, S. M., & Gould, T. D. (2019). (2R,6R)-hydroxynorketamine rapidly potentiates hippocampal glutamatergic transmission through a synapse-specific presynaptic mechanism. Neuropsychopharmacology. https://doi.org/10.1038/s41386-019-0443-3

Rush, A. J., Trivedi, M. H., Wisniewski, S. R., Nierenberg, A. A., Stewart, J. W., Warden, D., Niederehe, G., Thase, M. E., Lavori, P. W., Lebowitz, B. D., McGrath, P. J., Rosenbaum, J. F., Sackeim, H. A., Kupfer, D. J., Luther, J., & Fava, M. (2006). Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry, 163(11), 1905-1917. https://doi.org/163/11/1905 [pii] Rybakowski, J. K., Permoda-Osip, A., & Bartkowska-Sniatkowska, A. (2017). Ketamine augmentation rapidly improves depression scores in inpatients with treatment- resistant bipolar depression. Int J Psychiatry Clin Pract, 21(2), 99-103. https://doi.org/10.1080/13651501.2017.1297834

117

Saland, S. K., Duclot, F., & Kabbaj, M. (2017). Integrative analysis of sex differences in the rapid antidepressant effects of ketamine in preclinical models for individualized clinical outcomes. Curr Opin Behav Sci, 14, 19-26. https://doi.org/10.1016/j.cobeha.2016.11.002

Saland, S. K., & Kabbaj, M. (2018). Sex Differences in the Pharmacokinetics of Low- dose Ketamine in Plasma and Brain of Male and Female Rats. J Pharmacol Exp Ther, 367(3), 393-404. https://doi.org/10.1124/jpet.118.251652

Saland, S. K., Schoepfer, K. J., & Kabbaj, M. (2016). Hedonic sensitivity to low-dose ketamine is modulated by gonadal hormones in a sex-dependent manner. Sci Rep, 6, 21322. https://doi.org/10.1038/srep21322

Sandbaumhuter, F. A., Theurillat, R., Bektas, R. N., Kutter, A. P. N., Bettschart- Wolfensberger, R., & Thormann, W. (2016). Pharmacokinetics of ketamine and three metabolites in Beagle dogs under sevoflurane vs. medetomidine comedication assessed by enantioselective capillary electrophoresis. J Chromatogr A, 1467, 436-444. https://doi.org/10.1016/j.chroma.2016.07.060

Sandbaumhuter, F. A., Theurillat, R., Bettschart-Wolfensberger, R., & Thormann, W. (2017). Effect of the alpha2 -receptor agonists medetomidine, detomidine, xylazine, and romifidine on the ketamine metabolism in equines assessed with enantioselective capillary electrophoresis. Electrophoresis, 38(15), 1895-1904. https://doi.org/10.1002/elps.201700017

Sandbaumhuter, F. A., Theurillat, R., & Thormann, W. (2015). Effects of medetomidine and its active enantiomer dexmedetomidine on N-demethylation of ketamine in canines determined in vitro using enantioselective capillary electrophoresis. Electrophoresis, 36(21-22), 2703-2712. https://doi.org/10.1002/elps.201500147

Sandbaumhuter, F. A., Theurillat, R., & Thormann, W. (2017). Separation of hydroxynorketamine stereoisomers using capillary electrophoresis with sulfated beta-cyclodextrin and highly sulfated gamma-cyclodextrin. Electrophoresis, 38(15), 1878-1885. https://doi.org/10.1002/elps.201700016

Sandbaumhuter, F. A., & Thormann, W. (2018). Enantioselective capillary electrophoresis provides insight into the phase II metabolism of ketamine and its metabolites in vivo and in vitro. Electrophoresis, 39(12), 1478-1481. https://doi.org/10.1002/elps.201800012

Sarkar, A., & Kabbaj, M. (2016). Sex Differences in Effects of Ketamine on Behavior, Spine Density, and Synaptic Proteins in Socially Isolated Rats. Biol Psychiatry, 80(6), 448-456. https://doi.org/10.1016/j.biopsych.2015.12.025

Schmitz, A., Theurillat, R., Lassahn, P. G., Mevissen, M., & Thormann, W. (2009). CE provides evidence of the stereoselective hydroxylation of norketamine in equines. Electrophoresis, 30(16), 2912-2921. https://doi.org/10.1002/elps.200900221

118

Schoepp, D. D. (2001). Unveiling the functions of presynaptic metabotropic glutamate receptors in the central nervous system. J Pharmacol Exp Ther, 299(1), 12-20. http://www.ncbi.nlm.nih.gov/pubmed/11561058

Seamans, J. (2008). Losing inhibition with ketamine. Nat Chem Biol, 4(2), 91-93. https://doi.org/10.1038/nchembio0208-91

Shaffer, C. L., Dutra, J. K., Tseng, W. C., Weber, M. L., Bogart, L. J., Hales, K., Pang, J., Volfson, D., Am Ende, C. W., Green, M. E., & Buhl, D. L. (2019). Pharmacological evaluation of clinically relevant concentrations of (2R,6R)- hydroxynorketamine. Neuropharmacology, 153, 73-81. https://doi.org/10.1016/j.neuropharm.2019.04.019

Shen, M., & Machado, S. G. (2017). Bioequivalence evaluation of sparse sampling pharmacokinetics data using bootstrap resampling method. J Biopharm Stat, 27(2), 257-264. https://doi.org/10.1080/10543406.2016.1265543

Shirayama, Y., & Hashimoto, K. (2018). Lack of Antidepressant Effects of (2R,6R)- Hydroxynorketamine in a Rat Learned Helplessness Model: Comparison with (R)-Ketamine. Int J Neuropsychopharmacol, 21(1), 84-88. https://doi.org/10.1093/ijnp/pyx108

Singh, N. S., Rutkowska, E., Plazinska, A., Khadeer, M., Moaddel, R., Jozwiak, K., Bernier, M., & Wainer, I. W. (2016). Ketamine Metabolites Enantioselectively Decrease Intracellular D-Serine Concentrations in PC-12 Cells. PLoS One, 11(4), e0149499. https://doi.org/10.1371/journal.pone.0149499 Sohl, C. D., Cheng, Q., & Guengerich, F. P. (2009). Chromatographic assays of drug oxidation by human cytochrome P450 3A4. Nat Protoc, 4(9), 1252-1257. https://doi.org/10.1038/nprot.2009.122

Sten, T., Finel, M., Ask, B., Rane, A., & Ekstrom, L. (2009). Non-steroidal anti- inflammatory drugs interact with testosterone glucuronidation. Steroids, 74(12), 971-977. https://doi.org/10.1016/j.steroids.2009.07.004

Swanson, C. J., Bures, M., Johnson, M. P., Linden, A. M., Monn, J. A., & Schoepp, D. D. (2005). Metabotropic glutamate receptors as novel targets for anxiety and stress disorders. Nat Rev Drug Discov, 4(2), 131-144. https://doi.org/10.1038/nrd1630

Taylor, C., Fricker, A. D., Devi, L. A., & Gomes, I. (2005). Mechanisms of action of antidepressants: from neurotransmitter systems to signaling pathways. Cell Signal, 17(5), 549-557. https://doi.org/10.1016/j.cellsig.2004.12.007

Theurillat, R., Sandbaumhuter, F. A., Bettschart-Wolfensberger, R., & Thormann, W. (2016). Microassay for ketamine and metabolites in plasma and serum based on enantioselective capillary electrophoresis with highly sulfated gamma- cyclodextrin and electrokinetic analyte injection. Electrophoresis, 37(9), 1129- 1138. https://doi.org/10.1002/elps.201500468

119

Thompson, S. M., Kallarackal, A. J., Kvarta, M. D., Van Dyke, A. M., LeGates, T. A., & Cai, X. (2015). An excitatory synapse hypothesis of depression. Trends Neurosci, 38(5), 279-294. https://doi.org/10.1016/j.tins.2015.03.003 Tomankova, V., Anzenbacher, P., & Anzenbacherova, E. (2017). Effects of obesity on liver cytochromes P450 in various animal models. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub, 161(2), 144-151. https://doi.org/10.5507/bp.2017.026

Turfus, S. C., Parkin, M. C., Cowan, D. A., Halket, J. M., Smith, N. W., Braithwaite, R. A., Elliot, S. P., Steventon, G. B., & Kicman, A. T. (2009). Use of human microsomes and deuterated substrates: an alternative approach for the identification of novel metabolites of ketamine by mass spectrometry. Drug Metab Dispos, 37(8), 1769-1778. https://doi.org/10.1124/dmd.108.026328

Turgeon, D., Carrier, J. S., Levesque, E., Hum, D. W., & Belanger, A. (2001). Relative enzymatic activity, protein stability, and tissue distribution of human steroid- metabolizing UGT2B subfamily members. Endocrinology, 142(2), 778-787. https://doi.org/10.1210/endo.142.2.7958

Uher, R., Payne, J. L., Pavlova, B., & Perlis, R. H. (2014). Major depressive disorder in DSM-5: implications for clinical practice and research of changes from DSM-IV. Depress Anxiety, 31(6), 459-471. https://doi.org/10.1002/da.22217

Vasilescu, A. N., Schweinfurth, N., Borgwardt, S., Gass, P., Lang, U. E., Inta, D., & Eckart, S. (2017). Modulation of the activity of N-methyl-d-aspartate receptors as a novel treatment option for depression: current clinical evidence and therapeutic potential of rapastinel (GLYX-13). Neuropsychiatr Dis Treat, 13, 973-980. https://doi.org/10.2147/NDT.S119004

Vig, B. S., Huttunen, K. M., Laine, K., & Rautio, J. (2013). Amino acids as promoieties in prodrug design and development. Adv Drug Deliv Rev, 65(10), 1370-1385. https://doi.org/10.1016/j.addr.2012.10.001 Waxman, D. J., & Holloway, M. G. (2009). Sex differences in the expression of hepatic drug metabolizing enzymes. Mol Pharmacol, 76(2), 215-228. https://doi.org/10.1124/mol.109.056705

Widman, A. J., & McMahon, L. L. (2018). Disinhibition of CA1 pyramidal cells by low- dose ketamine and other antagonists with rapid antidepressant efficacy. Proc Natl Acad Sci U S A, 115(13), E3007-E3016. https://doi.org/10.1073/pnas.1718883115

Witkin, J. M., Marek, G. J., Johnson, B. G., & Schoepp, D. D. (2007). Metabotropic glutamate receptors in the control of mood disorders. CNS Neurol Disord Drug Targets, 6(2), 87-100. https://doi.org/10.2174/187152707780363302

Won, J., Marin de Evsikova, C., Smith, R. S., Hicks, W. L., Edwards, M. M., Longo- Guess, C., Li, T., Naggert, J. K., & Nishina, P. M. (2011). NPHP4 is necessary for

120

normal photoreceptor ribbon synapse maintenance and outer segment formation, and for sperm development. Hum Mol Genet, 20(3), 482-496. https://doi.org/10.1093/hmg/ddq494 Woolf, T. F., & Adams, J. D. (1987). Biotransformation of ketamine, (Z)-6- hydroxyketamine, and (E)-6-hydroxyketamine by rat, rabbit, and human liver microsomal preparations. Xenobiotica, 17(7), 839-847. https://doi.org/10.3109/00498258709043993

Wray, N. H., Schappi, J. M., Singh, H., Senese, N. B., & Rasenick, M. M. (2019). NMDAR-independent, cAMP-dependent antidepressant actions of ketamine. Mol Psychiatry, 24(12), 1833-1843. https://doi.org/10.1038/s41380-018-0083-8

Xiong, Z., Fujita, Y., Zhang, K., Pu, Y., Chang, L., Ma, M., Chen, J., & Hashimoto, K. (2019). Beneficial effects of (R)-ketamine, but not its metabolite (2R,6R)- hydroxynorketamine, in the depression-like phenotype, inflammatory bone markers, and bone mineral density in a chronic social defeat stress model. Behav Brain Res, 368, 111904. https://doi.org/10.1016/j.bbr.2019.111904

Yamaguchi, J. I., Toki, H., Qu, Y., Yang, C., Koike, H., Hashimoto, K., Mizuno- Yasuhira, A., & Chaki, S. (2018). (2R,6R)-Hydroxynorketamine is not essential for the antidepressant actions of (R)-ketamine in mice. Neuropsychopharmacology, 43(9), 1900-1907. https://doi.org/10.1038/s41386- 018-0084-y

Yanagihara, Y., Ohtani, M., Kariya, S., Uchino, K., Hiraishi, T., Ashizawa, N., Aoyama, T., Yamamura, Y., Yamada, Y., & Iga, T. (2003). Plasma concentration profiles of ketamine and norketamine after administration of various ketamine preparations to healthy Japanese volunteers. Biopharm Drug Dispos, 24(1), 37- 43. https://doi.org/10.1002/bdd.336

Yang, C., Qu, Y., Abe, M., Nozawa, D., Chaki, S., & Hashimoto, K. (2017). (R)- Ketamine Shows Greater Potency and Longer Lasting Antidepressant Effects Than Its Metabolite (2R,6R)-Hydroxynorketamine. Biol Psychiatry, 82(5), e43- e44. https://doi.org/10.1016/j.biopsych.2016.12.020

Yang, L., Li, Y., Hong, H., Chang, C. W., Guo, L. W., Lyn-Cook, B., Shi, L., & Ning, B. (2012). Sex Differences in the Expression of Drug-Metabolizing and Transporter Genes in Human Liver. J Drug Metab Toxicol, 3(3), 1000119. https://doi.org/10.4172/2157-7609.1000119

Yao, N., Skiteva, O., Zhang, X., Svenningsson, P., & Chergui, K. (2017). Ketamine and its metabolite (2R,6R)-hydroxynorketamine induce lasting alterations in glutamatergic synaptic plasticity in the mesolimbic circuit. Mol Psychiatry. https://doi.org/10.1038/mp.2017.239 Yao, N., Skiteva, O., Zhang, X., Svenningsson, P., & Chergui, K. (2018). Ketamine and its metabolite (2R,6R)-hydroxynorketamine induce lasting alterations in

121

glutamatergic synaptic plasticity in the mesolimbic circuit. Mol Psychiatry, 23(10), 2066-2077. https://doi.org/10.1038/mp.2017.239

Ye, L., Ko, C. Y., Huang, Y., Zheng, C., Zheng, Y., & Chou, D. (2019). Ketamine metabolite (2R,6R)-hydroxynorketamine enhances aggression via periaqueductal gray glutamatergic transmission. Neuropharmacology, 157, 107667. https://doi.org/10.1016/j.neuropharm.2019.107667

Yokoyama, R., Higuchi, M., Tanabe, W., Tsukada, S., Naito, M., Yamaguchi, T., Chen, L., Kasai, A., Seiriki, K., Nakazawa, T., Nakagawa, S., Hashimoto, K., Hashimoto, H., & Ago, Y. (2020). (S)-norketamine and (2S,6S)- hydroxynorketamine exert potent antidepressant-like effects in a chronic corticosterone-induced mouse model of depression. Pharmacol Biochem Behav, 191, 172876. https://doi.org/10.1016/j.pbb.2020.172876

Zanos, P., & Gould, T. D. (2018). Mechanisms of ketamine action as an antidepressant. Mol Psychiatry, 23(4), 801-811. https://doi.org/10.1038/mp.2017.255 Zanos, P., Highland, J. N., Liu, X., Troppoli, T. A., Georgiou, P., Lovett, J., Morris, P. J., Stewart, B. W., Thomas, C. J., Thompson, S. M., Moaddel, R., & Gould, T. D. (2019). (R)-Ketamine exerts antidepressant actions partly via conversion to (2R,6R)-hydroxynorketamine, while causing adverse effects at sub-anaesthetic doses. Br J Pharmacol, 176(14), 2573-2592. https://doi.org/10.1111/bph.14683

Zanos, P., Highland, J. N., Stewart, B. W., Georgiou, P., Jenne, C. E., Lovett, J., Morris, P. J., Thomas, C. J., Moaddel, R., Zarate, C. A., Jr., & Gould, T. D. (2019). (2R,6R)-hydroxynorketamine exerts mGlu2 receptor-dependent antidepressant actions. Proc Natl Acad Sci U S A, 116(13), 6441-6450. https://doi.org/10.1073/pnas.1819540116

Zanos, P., Moaddel, R., Morris, P. J., Georgiou, P., Fischell, J., Elmer, G. I., Alkondon, M., Yuan, P., Pribut, H. J., Singh, N. S., Dossou, K. S., Fang, Y., Huang, X. P., Mayo, C. L., Wainer, I. W., Albuquerque, E. X., Thompson, S. M., Thomas, C. J., Zarate, C. A., Jr., & Gould, T. D. (2016). NMDAR inhibition-independent antidepressant actions of ketamine metabolites. Nature, 533(7604), 481-486. https://doi.org/10.1038/nature17998

Zanos, P., Moaddel, R., Morris, P. J., Riggs, L. M., Highland, J. N., Georgiou, P., Pereira, E. F. R., Albuquerque, E. X., Thomas, C. J., Zarate, C. A., Jr., & Gould, T. D. (2018). Ketamine and Ketamine Metabolite Pharmacology: Insights into Therapeutic Mechanisms. Pharmacol Rev, 70(3), 621-660. https://doi.org/10.1124/pr.117.015198

Zanos, P., Nelson, M. E., Highland, J. N., Krimmel, S. R., Georgiou, P., Gould, T. D., & Thompson, S. M. (2017). A Negative Allosteric Modulator for alpha5 Subunit- Containing GABA Receptors Exerts a Rapid and Persistent Antidepressant-Like Action without the Side Effects of the NMDA Receptor Antagonist Ketamine in Mice. eNeuro, 4(1). https://doi.org/10.1523/ENEURO.0285-16.2017

122

Zarate, C. A., Jr., Brutsche, N., Laje, G., Luckenbaugh, D. A., Venkata, S. L., Ramamoorthy, A., Moaddel, R., & Wainer, I. W. (2012). Relationship of ketamine's plasma metabolites with response, diagnosis, and side effects in major depression. Biol Psychiatry, 72(4), 331-338. https://doi.org/10.1016/j.biopsych.2012.03.004 Zarate, C. A., Jr., Singh, J. B., Carlson, P. J., Brutsche, N. E., Ameli, R., Luckenbaugh, D. A., Charney, D. S., & Manji, H. K. (2006). A randomized trial of an N-methyl- D-aspartate antagonist in treatment-resistant major depression. Arch Gen Psychiatry, 63(8), 856-864. https://doi.org/63/8/856 [pii] Zawilska, J. B., Wojcieszak, J., & Olejniczak, A. B. (2013). Prodrugs: a challenge for the drug development. Pharmacol Rep, 65(1), 1-14. http://www.ncbi.nlm.nih.gov/pubmed/23563019

Zhang, K., Toki, H., Fujita, Y., Ma, M., Chang, L., Qu, Y., Harada, S., Nemoto, T., Mizuno-Yasuhira, A., Yamaguchi, J. I., Chaki, S., & Hashimoto, K. (2018). Lack of deuterium isotope effects in the antidepressant effects of (R)-ketamine in a chronic social defeat stress model. Psychopharmacology (Berl), 235(11), 3177- 3185. https://doi.org/10.1007/s00213-018-5017-2

Zhao, X., Venkata, S. L., Moaddel, R., Luckenbaugh, D. A., Brutsche, N. E., Ibrahim, L., Zarate, C. A., Jr., Mager, D. E., & Wainer, I. W. (2012). Simultaneous population pharmacokinetic modelling of ketamine and three major metabolites in patients with treatment-resistant bipolar depression. Br J Clin Pharmacol, 74(2), 304-314. https://doi.org/10.1111/j.1365-2125.2012.04198.x

123