The Role of Kisspeptin in Reproduction

Thesis submitted for the degree of Doctor of Philosophy Imperial College London

Alexander Comninos 2014

Section of Investigative Medicine Division of Diabetes, Endocrinology & Metabolism Imperial College London

Declaration of Copyright ______

The copyright of this thesis rests with the author and is made available under a Creative

Commons Attribution Non-Commercial No Derivatives Licence. Researchers are free to copy, distribute or transmit the thesis on the condition that they attribute it, that they do not use it for commercial purposes and that they do not alter, transform or build upon it. For any reuse or redistribution, researchers must make clear to others the licence terms of this work.

Abstract ______

Kisspeptin is a recently identified hypothalamic neuropeptide hormone which is a critical regulator of mammalian reproductive physiology. In addition, kisspeptin administration stimulates gonadotrophin release in all species tested. Kisspeptin may therefore serve as a potential therapy for reproductive pathologies.

The role of kisspeptin has been studied extensively within the hypothalamus but little is known about its extra-hypothalamic effects. Using manganese-enhanced MRI to detect neuronal activity, I demonstrate marked decreases in neuronal activity in the amygdala of adult mice after kisspeptin administration. This data reveals a novel effect of kisspeptin with pharmacological implications.

The human kisspeptin peptides kisspeptin-10, -13, -14, and -54 are named according to their constituent amino acids. Kisspeptin-10 is the smallest and most easily synthesised peptide from a pharmacological point of view. By investigating the effects of the kisspeptin-10 peptide in healthy men and women, I uncover a sexual dimorphism in gonadotrophin response to kisspeptin-10 as well as a greater potency of kisspeptin-54 over kisspeptin-10 in healthy women.

Having established the effects of acute kisspeptin administration in healthy women, I investigated the effects of chronic administration of kisspeptin on reproductive function.

Using biochemical and radiological parameters of menstrual cyclicity, I demonstrate that 7 days of kisspeptin administration advances the menstrual cycle in healthy women.

In summary, in this thesis I have identified a novel effect of kisspeptin in rodents involving the amygdala. My clinical studies have identified for the first time the acute and chronic effects of kisspeptin administration in humans. Combined, these data have important implications for the development of kisspeptin as a potential therapy for reproductive pathologies.

Declaration of Contributors ______

Declaration of Originality: The work described in this thesis is my own. Any collaborations and assistance are detailed below. Contributors are within the Section of Investigative

Medicine (Imperial College London) unless stated otherwise.

Radioimmunoassays were established and maintained by Professor Mohammad Ghatei.

Chapter 2: I designed the study with the assistance of Professor Waljit Dhillo, Professor

Jimmy Bell (Metabolic and Molecular Imaging Group, Imperial College London) and

Professor Paul Matthews (Division of Brain Sciences, Imperial College London). I performed the animal studies with the assistance of Dr. Jelena Anastasovska and Dr.

Meliz Sahuri-Arisoylu (both from the Metabolic and Molecular Imaging Group, Imperial

College London).

Chapter 3: The study was designed in collaboration with Professor Waljit Dhillo, Dr.

Channa Jayasena and Dr. Monica Nijher. Recruitment and studies were carried out by myself with assistance from Dr. Channa Jayasena, Dr. Monica Nijher, Dr. Ali Abbara, Dr.

Adam Januszewki, Ms. Meriel Vaal and Ms. Zohreh Farzad. Assays were performed by myself together with Dr. Channa Jayasena and Dr. Monica Nijher.

Chapter 4: The study was designed in collaboration with Professor Waljit Dhillo and Dr.

Channa Jayasena. Study visits were carried out by myself with assistance from Dr.

Channa Jayasena, Dr. Monica Nijher, Dr. Ali Abbara, Dr. Akila de Silva, Dr. Risheka

Ratnasabapathy and Dr. Chioma Izzi-engbeaya. The ultrasound scans were performed by

Dr. Adrian Lim and Ms. Daksha Patel. Professor Johannes Veldhuis performed the pulsatility analyses. Assays were performed by myself together with Dr. Channa Jayasena and Dr. Monica Nijher.

Acknowledgements ______

Firstly, I would like to thank Professor Waljit Dhillo for the opportunity to conduct research on kisspeptin in his research group and for his constant support and enthusiasm throughout my research studies.

I am extremely grateful to the senior researchers Dr. Channa Jayasena, Professor Sir

Stephen Bloom, Professor Jimmy Bell, Professor Paul Matthews and Professor

Mohammad Ghatei for their advice and help with various technical aspects of the work.

For the animal studies, I am indebted to Dr. Jelena Anastasovska and Dr. Meliz Sahuri-

Arisoylu for teaching me many new techniques and being so generous with their time.

For the clinical studies, I am indebted to my colleagues Dr. Ali Abbara, Dr. Monica Nijher,

Dr. Akila de Silva, Dr. Risheka Ratnasabapathy, Dr. Chioma Izzi-engbeaya Ms. Meriel

Vaal and Ms. Zohreh Farzad for all their help and moral support.

Special thanks are due to the staff at the Imperial College Clinical Research Facility for providing invaluable support during the screening of study participants for the clinical studies.

I am grateful to the Wellcome Trust and GlaxoSmithKline for awarding me a Translational

Medicine Training Fellowship which allowed me to pursue my PhD research studies.

Finally I would like to thank the study participants, without whom none of the clinical studies would have been possible.

This thesis is dedicated to my parents and Georgia for all their love and support.

Table of Contents ______

ABSTRACT ...... 3 DECLARATION OF CONTRIBUTORS ...... 4 ACKNOWLEDGEMENTS ...... 5 TABLE OF CONTENTS ...... 6 INDEX OF FIGURES ...... 10 INDEX OF TABLES ...... 12 LIST OF ABBREVIATIONS ...... 13 CHAPTER 1: GENERAL INTRODUCTION ...... 15

1.1 THE REPRODUCTIVE AXIS: AN OVERVIEW ...... 16 1.2 THE REPRODUCTIVE AXIS: THE GNRH PULSE GENERATOR ...... 20 1.3 KISSPEPTIN: THE PEPTIDE PRODUCT OF THE KISS1 GENE ...... 23 1.4 KISS1 EXPRESSION: CNS DISTRIBUTION ...... 24 1.4.1 Mouse ...... 24 1.4.2 Rat ...... 24 1.4.3 Primates (including humans) ...... 25 1.5 KISS1 EXPRESSION: PERIPHERAL DISTRIBUTION ...... 25 1.6 THE KISSPEPTIN RECEPTOR (KISS1R) ...... 27 1.7 KISS1R EXPRESSION: CNS DISTRIBUTION ...... 28 1.7.1 Mouse ...... 28 1.7.2 Rat ...... 28 1.7.3 Primates (including humans) ...... 29 1.8 KISS1R EXPRESSION: PERIPHERAL DISTRIBUTION ...... 29 1.9 THE IMPORTANCE OF THE KISS1/KISS1R SYSTEM IN THE REPRODUCTIVE AXIS ...... 30 1.9.1 KISS1R ...... 30 1.9.2 KISS1 ...... 31 1.10 KISSPEPTIN ADMINISTRATION CAN INDUCE PUBERTY IN ANIMALS ...... 32 1.11 KISSPEPTIN STIMULATES THE RELEASE OF GNRH WHICH IN TURN STIMULATES GONADOTROPHIN RELEASE ...... 33 1.12 DIRECT EFFECTS OF KISSPEPTIN ON PITUITARY GONADOTROPHIN SECRETION ...... 36 1.13 KISSPEPTIN STIMULATES THE PRE-OVULATORY LH SURGE AND OVULATION IN ANIMALS ...... 36 1.14 THE PRE-OVULATORY LH SURGE AND OVULATION IN HUMANS ...... 39 1.15 KISSPEPTIN: FIRST ADMINISTRATION TO HUMANS...... 40 1.16 KISSPEPTIN EFFECTS ARE MORE PRONOUNCED DURING THE PRE-OVULATORY PHASE ...... 41 1.17 KISSPEPTIN-10 VS. KISSPEPTIN-54 AND THE GNRH PULSE GENERATOR ...... 41 1.18 KISSPEPTIN ADMINISTRATION IN DISORDERS OF REPRODUCTION ...... 43 1.18.1 Hypothalamic Amenorrhoea ...... 43 1.18.2 Hypogonadotrophic Hypogonadism secondary to TAC3 or TAC3R mutations ...... 44 1.18.3 Hypogonadism secondary to Type 2 Diabetes Mellitus (T2DM) ...... 44 1.18.4 Hyperprolactinaemia ...... 45 1.19 CHRONIC KISSPEPTIN ADMINISTRATION AND TACHYPHYLAXIS ...... 45 1.20 THERAPEUTIC APPLICATIONS OF KISSPEPTIN TO STIMULATE THE REPRODUCTIVE AXIS ...... 46 1.20.1 In Vitro Fertilisation ...... 46 1.20.2 Other applications ...... 47 1.21 THERAPEUTIC APPLICATIONS OF KISSPEPTIN TO DOWN-REGULATE THE REPRODUCTIVE AXIS . 48

1.21.1 Kisspeptin receptor antagonists ...... 48 1.21.2 Kisspeptin analogues ...... 49 1.22 NON-REPRODUCTIVE EFFECTS OF KISSPEPTIN ...... 50 1.22.1 Cancer metastasis ...... 50 1.22.2 Other neuroendocrine effects ...... 51 1.22.3 Other CNS roles ...... 51 1.22.4 Metabolism ...... 52 1.22.5 Cardiovascular system ...... 54 1.22.6 Fluid homeostasis ...... 54 1.22.7 Summary ...... 55 1.23 AIMS AND HYPOTHESES OF THE THESIS ...... 56

CHAPTER 2: THE EFFECTS OF KISSPEPTIN ADMINISTRATION ON NEURONAL ACTIVATION AS DETERMINED BY MANGANESE-ENHANCED MAGNETIC RESONANCE IMAGING ...... 58

2.1 INTRODUCTION ...... 59 2.1.1 Regions of Interest (ROIs) ...... 60 2.1.2 Principles of Magnetic Resonance Imaging (MRI) ...... 61 2.1.3 Manganese-enhanced Magnetic Resonance Imaging (MEMRI) ...... 63 2.1.4 Comparison with other techniques for assessing neuronal activity ...... 65 2.2 AIMS AND HYPOTHESIS ...... 68 2.3 METHODS ...... 69 2.3.1 Animal preparation ...... 69 2.3.2 Kisspeptin-54 peptide synthesis ...... 69 2.3.3 Study 1: Effects of peripheral kisspeptin administration on plasma kisspeptin and serum LH levels in adult male mice ...... 70 2.3.3.1 Study 1A: Effects of peripheral kisspeptin administration on plasma kisspeptin levels ...... 70 2.3.3.2 Study 1B: Effects of peripheral kisspeptin administration on serum LH levels ...... 71 2.3.4 Study 2: Effects of peripheral kisspeptin-54 administration on CNS neuronal activity in adult male mice using MEMRI ...... 72 2.3.4.1 Protocol ...... 72 2.3.4.2 Regions of Interest (ROIs) and Image Analysis ...... 73 2.3.4.3 Statistical Analysis ...... 74 2.4 RESULTS ...... 76 2.4.1 Study 1: Peripheral kisspeptin administration increases circulating kisspeptin and LH levels ...... 76 2.4.1.1 Study 1A: Peripheral kisspeptin administration increases circulating kisspeptin levels ..... 76 2.4.1.2 Study 1B: Peripheral kisspeptin administration increases circulating LH levels ...... 76 2.4.2 Study 2: Peripheral kisspeptin administration modulates CNS neuronal activity as determined by MEMRI ...... 78 2.4.2.1 Amygdala ...... 78 2.4.2.2 ARC and AVPV...... 78 2.4.2.3 POA ...... 79 2.4.2.4 Anterior Pituitary ...... 79 2.5 DISCUSSION ...... 82 Limitations ...... 86 Summary ...... 87 2.6 FUTURE WORK ...... 88

CHAPTER 3: THE EFFECTS OF KISSPEPTIN-10 IN HEALTHY MEN AND WOMEN ...... 90

3.1 INTRODUCTION ...... 91 3.2 AIM AND HYPOTHESIS ...... 93 3.3 METHODS ...... 94 3.3.1 Subjects ...... 94 3.3.2 Study days ...... 94 3.3.3 LH, FSH, Oestradiol and Testosterone measurement ...... 95 3.3.4 Kisspeptin radioimmunoassay ...... 97 3.3.5 Kisspeptin-10 and -54 peptide synthesis ...... 97 3.3.6 Study 1: Effects of intravenous bolus injection of saline or kisspeptin-10 in healthy male volunteers ...... 97 3.3.7 Study 2: Effects of IV bolus injection of saline, kisspeptin-10 or kisspeptin-54 in healthy female volunteers ...... 98 3.3.8 Study 3: Determining the half-life of kisspeptin-10 in healthy male and female volunteers ...... 98 3.3.9 Data analysis ...... 100 3.4 RESULTS ...... 101 3.4.1 Study 1: Effects of IV bolus injection of saline or kisspeptin-10 in healthy male volunteers ...... 101 3.4.2 Study 2: Effects of IV bolus injection of saline, kisspeptin-10 or kisspeptin-54 in healthy female volunteers ...... 108 3.4.2.1: Follicular phase of the menstrual cycle ...... 108 3.4.2.2: Pre-ovulatory phase of the menstrual cycle ...... 109 3.4.3 Study 3: Determining the half-life of kisspeptin-10 in healthy male and female volunteers (follicular and pre-ovulatory phase)...... 114 3.5 DISCUSSION ...... 117 Limitations ...... 128 Summary ...... 128 3.6 FUTURE WORK ...... 129

CHAPTER 4: THE EFFECTS OF CHRONIC KISSPEPTIN-54 ADMINISTRATION IN HEALTHY WOMEN ...... 131

4.1 INTRODUCTION ...... 132 4.2 AIMS AND HYPOTHESIS ...... 134 4.3 METHODS ...... 135 4.3.1 Subjects ...... 135 4.3.2 Protocol ...... 135 4.3.3 Kisspeptin-54 peptide synthesis and testing ...... 137 4.3.4 Kisspeptin injections ...... 137 4.3.5 Gonadotropin-releasing hormone (GnRH) test protocol ...... 138 4.3.6 Baseline period ...... 138 4.3.7 Treatment period ...... 138 4.3.8 Post-treatment period ...... 138 4.3.9 Four hour blood sampling post-injection of saline or kisspeptin ...... 139 4.3.10 Eight hour blood sampling for assessments of LH pulsatility ...... 139 4.3.11 Assessment of pituitary sensitivity before and after injections of saline or kisspeptin140 4.3.12 Basal measurement of reproductive hormones ...... 140 4.3.13 Collection and processing of blood samples ...... 140 4.3.14 Ultrasound scans ...... 140

4.3.15 Other measurements ...... 141 4.3.16 Analytical methods ...... 141 4.3.17 Kisspeptin radioimmunoassay ...... 141 4.3.18 Data analysis ...... 142 4.4 RESULTS ...... 143 4.4.1 Subjects ...... 143 4.4.2 Acute effects of saline or kisspeptin-54 injection on plasma kisspeptin at the commencement, mid-point and end of twice-daily administration in healthy women ...... 144 4.4.2.1 Acute changes in plasma kisspeptin IR following injection of saline or kisspeptin-54 ...... 144 4.4.2.2 Plasma kisspeptin IR pre-injection of saline or kisspeptin-54 ...... 144 4.4.3 Acute effects of saline or kisspeptin-54 injection on serum reproductive hormones at the commencement, mid-point and end of twice-daily administration in healthy women ...... 146 4.4.3.1 Commencement of treatment period (menstrual day 7) ...... 146 4.4.3.2 Mid-point of treatment period (menstrual day 11) ...... 146 4.4.3.3 End of treatment period (menstrual day 14) ...... 146 4.4.3.4 Serum LH increase on different study days ...... 147 4.4.4 Effects of twice-daily saline or kisspeptin injections on length of the menstrual cycle and biochemical markers of reproductive activity in healthy women ...... 150 4.4.4.1 Length of menstrual cycle ...... 150 4.4.4.2 Timing of peak serum LH ...... 150 4.4.4.3 Timing of luteal phase of menstrual cycle ...... 150 4.4.5 Effects of twice-daily saline or kisspeptin injections on radiological markers of reproductive activity in healthy women ...... 153 4.4.6 Effects of twice-daily saline or kisspeptin on LH pulsatility in healthy women ...... 156 4.4.7 Assessment of pituitary sensitivity before and after injections of saline or kisspeptin158 4.5 DISCUSSION ...... 160 Limitations ...... 164 Summary ...... 166 4.6 FUTURE WORK AND THERAPEUTIC IMPLIATIONS ...... 166

CHAPTER 5: GENERAL DISCUSSION ...... 168

5.1 GENERAL DISCUSSION ...... 169 5.2 FUTURE WORK SUMMARY ...... 175

APPENDIX 1: PRINCIPLES OF RADIOIMMUNOASSAY ...... 177 APPENDIX 2: PRINCIPLES OF XMAP IMMUNOASSAY ...... 179 REFERENCES ...... 180

PUBLICATIONS & COMMUNICATIONS ...... 202

Index of Figures ______

Chapter 1

Figure 1-1: The amino acid sequence of the GnRH decapeptide ...... 18 Figure 1-2: Overview of the reproductive axis before the identification of kisspeptin...... 18 Figure 1-3: The female menstrual cycle ...... 19 Figure 1-4: Human kisspeptin isoforms...... 26 Figure 1-5: Neuroanatomy and feedback system of the kisspeptin-reproductive axis in rodents and humans...... 35

Chapter 2

Figure 2-1: Representative cross-sectional images through 3-dimensional ROIs in adult mouse brain from which signal intensity (SI) profiles were generated ...... 75 Figure 2-2: Effect of peripheral kisspeptin or vehicle administration on circulating kisspeptin and LH levels...... 77 Figure 2-3: Changes in CNS neuronal activity following kisspeptin or saline administration...... 80

Chapter 3

Figure 3-1: Study Protocol for Studies 1 and 2...... 99 Figure 3-2: Study Protocol for Study 3...... 99 Figure 3-3: Plasma kisspeptin IR after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers...... 104 Figure 3-4: Serum LH after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers ...... 105 Figure 3-5: Serum FSH after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers...... 106 Figure 3-6: Serum Testosterone after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers ...... 107 Figure 3-7: Plasma kisspeptin IR after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women...... 110

Figure 3-8: Serum LH after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women...... 111 Figure 3-9: Serum FSH after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women ...... 112 Figure 3-10: Serum Oestradiol after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women...... 113 Figure 3-11: Plasma kisspeptin IR before, during and after a 90 minute infusion of kisspeptin-10 (360 pmol/kg/min) to healthy males and females ...... 115 Figure 3-12: Natural Logarithm kisspeptin IR against time immediately after cessation of a 90 minute infusion of 360 pmol/kg/min kisspeptin-10...... 116 Figure 3-13: Sex steroid feedback loops have different effects on the two identified kisspeptin neuronal populations in the ARC and AVPV of the hypothalamus in rodents...... 123

Chapter 4

Figure 4-1: Study protocol diagram...... 136 Figure 4-2: Acute changes in plasma kisspeptin levels in healthy women receiving twice-daily injections of saline or kisspeptin-54...... 145 Figure 4-3: Acute changes in serum reproductive hormones in healthy women receiving twice- daily injections of saline or kisspeptin-54...... 148 Figure 4-4: Time classification of acute serum LH responses to kisspeptin-54 in healthy women receiving twice-daily injections of kisspeptin-54...... 149 Figure 4-5: Levels of serum reproductive hormones throughout the menstrual cycle in healthy women receiving twice-daily injections of saline or kisspeptin-54...... 152 Figure 4-6: Changes in radiological markers during the menstrual cycle of healthy women receiving twice-daily injections of saline or kisspeptin-54...... 155 Figure 4-7: Changes in LH pulsatility during the menstrual cycle of healthy women receiving twice-daily injections of saline or kisspeptin-54...... 157 Figure 4-8: Pituitary responsiveness in healthy women receiving twice-daily injections of saline or kisspeptin-54...... 159

Index of Tables

______

Chapter 1

Table 1-1: LH pulse frequency during various phases of the menstrual cycle ...... 21 Table 1-2: Neurotransmitters modulating GnRH secretion from GnRH neurones ...... 22

Chapter 3

Table 3-1: Healthy recruitment inclusion criteria...... 96 Table 3-2: Baseline characteristics of healthy male and female volunteers recruited to the study...... 103

Chapter 4

Table 4-1: Baseline characteristics of the 5 healthy female subjects at first study visit...... 143 Table 4-2: Table showing first menstrual day that corpus luteum was visualised by ultrasonography in each participant and highest recorded progesterone level during that menstrual cycle (saline or kisspeptin-54 treatment)...... 154

List of Abbreviations ______

ANOVA Analysis of variance

ARC Arcuate nucleus

AUC Area under the curve

AVPV Anteroventral periventricular nucleus

BBB Blood brain barrier

BP Blood pressure

CNS Central nervous system

CSF Cerebrospinal fluid

CVO Circumventricular organ

DNA Deoxyribonucleic acid

E2 Oestradiol

FOV Field of view

FSH Follicle stimulating hormone

GABA Gamma-aminobutyric acid

GEE General estimated equation

GH Growth hormone

GLP-1 Glucagon-like peptide-1

GnRH Gonadotrophin-releasing hormone

GPR54 G-protein coupled receptor 54

HA Hypothalamic amenorrhoea

E2 Oestradiol icv Intracerebroventricular iv Intravenous

IR Immunoreactivity

KP10 Kisspeptin-10

KP54 Kisspeptin-54

LH Luteinising hormone

MEMRI Manganese-enhanced magnetic resonance imaging

ME Median eminence

MRI Magnetic resonance imaging

MR Magnetic resonance

NKB Neurokinin B

NMDA N-methyl-D-aspartate

NMV Net magnetisation vector

OXM Oxyntomodulin

PCOS Polycystic ovarian syndrome

PeN Periventricular nucleus

PET Positron emission tomography

PE Percentage enhancement

POA Preoptic area

PRL Prolactin

PSA Prostate-specific antigen

PVN Paraventricular nucleus

RF Radiofrequency

RIA Radioimmunoassay

RP3V Rostral periventricular area

RPM Revolutions per minute sc Subcutaneous

SI Signal intensity

SEM Standard error of mean

T2DM Type 2 diabetes mellitus

TEeff Effective echo time

TR Repetition time

TSH Thyroid stimulating hormone

VMH Ventromedial hypothalamus

Chapter 1 ______

General Introduction

1.1 The reproductive axis: An overview

The release of sex hormones in animals and humans is tightly regulated by the reproductive axis. This axis relies on interplay between the hypothalamus, where the signals originate, the pituitary gland and the gonads.

Specialised neurones within the medial preoptic area (in rodents and sheep) or the mediobasal hypothalamus (in humans and primates) release gonadotrophin-releasing hormone (GnRH), a ten amino acid peptide (Figure 1-1), in a pulsatile manner into the hypophyseal-portal circulation (Carmel et al. 1976, Barry 1979, Rance et al. 1994).

Through this hypophyseal-portal circulation, the GnRH arrives at the anterior pituitary gonadotroph cells which are consequently stimulated to release the gonadotrophin hormones, luteinising hormone (LH) and follicle stimulating hormone (FSH), into the systemic circulation (Wildt et al. 1981, Clarke and Cummins 1982). These gonadotrophins then act on the gonads to stimulate gametogenesis and the release of the gonadal sex steroids (androgens in males and oestrogens in females). FSH predominantly acts to stimulate gametogenesis, while LH predominantly promotes the synthesis of the sex steroids (Reichlin 1998).

Tight regulation of this axis is achieved by a negative feedback system whereby the gonadotrophins and sex steroids inhibit further gonadotrophin and GnRH release (Figure

1-2).

In males, LH activates receptors on leydig cells triggering the synthesis and secretion of androgens, primarily testosterone. Subsequently testosterone stimulates spermatogenesis and the development of male secondary characteristics.

In females, LH activates receptors on theca cells of the ovary triggering the release of testosterone. Testosterone is subsequently converted into oestrogens by adjacent granulosa cells. Meanwhile, FSH stimulates ovarian follicle production. Ovulation occurs due to a mid-cycle surge of LH also termed the pre-ovulatory phase. During this phase the

16 normal negative feedback exerted by LH on the reproductive axis (Figure 1-2) switches to a positive feedback due to the high levels of oestrogen and progesterone (from the developing follicles). This results in further LH release and eventually ovulation. After ovulation the follicle degenerates into a corpus luteum which secretes progesterone and further oestradiol (Figure 1-3).

Failure at any point in this reproductive axis can lead to a range of pathologies including delayed puberty and infertility.

Figure 1-1: The amino acid sequence of the GnRH decapeptide

pGlu – His – Trp – Ser – Tyr – Gly – Leu – Arg – Pro – Gly – NH2

Figure 1-2: Overview of the reproductive axis before the identification of kisspeptin. The secretion of the reproductive hormones is tightly regulated by a classical negative feedback loop.

negative feedbacknegative

feedback

negative

Testosterone Oestradiol

Figure 1-3: The female menstrual cycle (Aitken et al. 2008). The first day of menstrual bleeding signifies the start of the follicular phase. Here the breakdown and shedding of the endometrial lining is a consequence of a decrease in oestrogen and progesterone. In this follicular phase, FSH increases stimulating the development of multiple ovarian follicles. Subsequently FSH levels decrease resulting in only one or two follicles developing further. Oestrogen is released from these developing follicles which initiates endometrial thickening. The pre-ovulatory phase then begins at around day 13; here LH/FSH levels increase dramatically resulting in increases in oestrogen and progesterone. The high LH levels stimulate ovulation. Subsequently the luteal phase begins where LH/FSH levels fall and the ruptured follicle forms a corpus luteum, which produces progesterone. The corpus luteum degenerates during this phase (if there is no fertilisation) resulting in decreased levels of progesterone and oestrogen hence triggering a new menstrual cycle.

1.2 The reproductive axis: The GnRH pulse generator

GnRH is a key player in the reproductive axis and indeed displays structural conservation across all mammals except the guinea pig. GnRH is synthesised and secreted by specialised GnRH neurones in the mediobasal hypothalamus in humans. These neurones have projections to the median eminence where GnRH is then released into the hypophyseal-portal circulation (Gore 2002). This GnRH release is pulsatile and it is this pulsatile release that is key to the normal functioning of the reproductive axis. This is demonstrated by the chronic (non-pulsatile) administration of GnRH to rhesus monkeys with hypothalamic lesions failing to stimulate LH and FSH release due to the down- regulation of the GnRH receptor (Belchetz et al. 1978). Thereafter when hourly intermittent GnRH pulses were administered there was restoration of LH and FSH release

(Belchetz et al. 1978).

The frequency and amplitude of GnRH pulses varies throughout human development and the menstrual cycle:

Human Development

The pulsatile secretion of GnRH has been noted in early gestation and continues until 6 months of age in males and 2 years in females. Thereafter GnRH pulsatility returns around puberty, initially with nocturnal pulses and then with additional daytime pulses

(Gore 2002).

Menstrual Cycle

There is marked variation in GnRH pulsatility during the different phases of the menstrual cycle and this is required for normal functioning of the reproductive axis (Table 1).

Measurement of GnRH is difficult as it is only released into the local hypophyseal-portal

20 circulation. Therefore downstream serum LH pulsatility is used as a surrogate marker of

GnRH pulsatility. Hence pulsatile LH release, stimulated by GnRH, increases in frequency during the follicular phase and peaks around ovulation.

The pulsatile nature of GnRH results from a number of complex pathways. Although

GnRH neurones release GnRH inherently in a pulsatile manner (Funabashi et al. 2000), there are a number of excitatory and inhibitory neurotransmitters that modulate this secretion in vivo as shown below (Table 2).

Table 1-1: LH pulse frequency during various phases of the menstrual cycle. Adapted from (Gore 2002).

LH pulse frequency Menstrual Phase (minutes) Early Follicular 110 Mid Follicular 70 Ovulatory 65 Early Luteal 100 Mid Luteal 200 Late Luteal 300

Table 1-2: Neurotransmitters modulating GnRH secretion from GnRH neurones. Adapted from (Evans 1999) (Gore 2002).

Excitatory Inhibitory Excitatory and Inhibitory

Angiotensin II Dopamine Galanin Acetylcholine Glutamate Cholecystokinin Gamma-aminobutyric acid Histamine Dopamine Opioids Leptin Oxytocin Corticotropin releasing factor Neuropeptide Y Vasopressin Noradrenaline Serotonin Substance P Vasoactive intestinal peptide

Stress, nutrition and sex steroid feedback are also important modulators of GnRH secretion, an effect partly mediated by the above neurotransmitters. There do however exist a number of functional limitations to the GnRH neuronal network and the reproductive axis as displayed in Figure 1-2. For instance, GnRH neurones lack oestrogen receptor (ERα) and so this suggests the existence of an intermediary signal for mediating gonadal steroid feedback (Herbison and Theodosis 1992).

Recently the peptide kisspeptin has been identified to have a powerful role in the regulation of GnRH secretion and therefore the reproductive axis.

1.3 Kisspeptin: The peptide product of the KISS1 gene

The KISS1 gene was first identified in 1996 as a metastasis-suppressor gene in melanoma cell lines. As it was first discovered in Hershey (Pennsylvania, USA), the gene

(KISS1) and gene product (kisspeptin) were named after the famous chocolate ‘Kisses’ manufactured in the city.

The original work employed subtractive hybridisation to demonstrate that KISS1-mRNA was selectively overexpressed in metastasis-suppressed human melanoma cell lines.

Furthermore, transfection of full length human KISS1 cDNA into melanoma cell lines suppressed metastasis in an expression-dependent manner (Lee et al. 1996). Soon after, the KISS1 gene was localised to chromosome 1q32 with a total of four exons identified, although the first two exons are not translated (West et al. 1998). It was several years later that the peptide products of the KISS1 gene were identified (Kotani et al. 2001, Muir et al. 2001, Ohtaki et al. 2001).

The KISS1 gene encodes a precursor peptide containing 145 amino acids. Plasma proteolytic cleavage of the 145 amino acid gene product results in kisspeptin peptides of varying amino acid length. They are named according to their amino acid length, as kisspeptin-10, -13, -14, and -54 in humans. These kisspeptins all share a common carboxy-terminal decapeptide sequence, where they harbour an Arg-Phe-NH2 motif required for biological activity (Kirby et al. 2010) (Figure 1-4). In rodents, the long kisspeptin isoform is 52 amino acids in length, with Arg-Tyr-NH2 at the carboxy-terminal instead of Arg-Phe-NH2 as in humans (Tena-Sempere 2006).

Several studies have examined the distribution of KISS1 expression in rodents and humans. The standard nomenclature is KISS1 in humans and Kiss1 in non-human species. If both KISS1 and Kiss1 are referred to in a single sentence then just KISS1 is used for simplicity.

1.4 KISS1 expression: CNS distribution

1.4.1 Mouse

Kisspeptin has been identified within the mouse CNS, with high expression of both Kiss1 mRNA and the kisspeptin protein particularly within the arcuate nucleus (ARC), anteroventral periventricular nucleus (AVPV), and the preoptic periventricular nucleus

(PeN) of the hypothalamus which are all areas implicated in the regulation of GnRH and gonadotrophin secretion (Gottsch et al. 2004, Clarkson and Herbison 2006). Interestingly levels of Kiss1 expression in the AVPV and PeN are ten-fold higher in female mice when compared to male mice (Clarkson and Herbison 2006).

In addition to the locations detailed above, a more recent study using a polyclonal rabbit antibody directed towards the final carboxy-terminal decapeptide sequence of the kisspeptin peptide, identified kisspeptin within the dorsomedial nucleus, lateral septum

(ventral aspect), and posterior hypothalamus (Clarkson et al. 2009). This study also identified Kiss1 expression outside the hypothalamus in the mouse brain including the bed nucleus of the stria terminalis, subfornical organ, medial amygdala, paraventricular thalamus, periaqueductal grey and locus coerulus. The presence of Kiss1 expression in these areas suggests a role for kisspeptin signalling in functions beyond reproduction that remains to be explored.

1.4.2 Rat

Kiss1 mRNA has been identified in the same neuroanatomical locations as in the mouse above, including the key reproductive centres of the ARC, AVPV and PeN (Irwig et al.

2004, Smith et al. 2006, Adachi et al. 2007, Kauffman et al. 2007). In addition the kisspeptin peptide has been identified in the ARC, paraventricular and ventromedial nuclei of the rat hypothalamus (Brailoiu et al. 2005).

1.4.3 Primates (including humans)

In the primate CNS (including humans), hypothalamic KISS1 mRNA is predominantly located within the infundibular nucleus and pre-optic area (POA), which are equivalent to the rodent ARC. No equivalent anatomical area to the rodent AVPV kisspeptin neurones has yet been identified in humans (Rometo et al. 2007). However in female monkeys and women, kisspeptin neurones have been identified in the rostral periventricular zone

(RP3V) which are positively stimulated by sex steroids similar to the AVPV Kiss1 neurones in rodents (Hrabovszky et al. 2010, Smith et al. 2010) (Figure 1-5). This population of KISS1 neurones is not present in men or agonadal male monkeys suggesting a physiological role for these neurones in females in generating the positive- feedback driven mid-cycle gonadotrophin surge (Ramaswamy et al. 2008, Hrabovszky et al. 2010).

1.5 KISS1 expression: Peripheral distribution

KISS1 expression has been identified peripherally in the testis, ovary, pancreas, liver, small intestine, carotid body (main peripheral arterial chemoreceptor), dorsal root ganglion and placenta in primates (including humans) and rodents (Muir et al. 2001, Ohtaki et al.

2001, Richard et al. 2008, Gaytan et al. 2009, Mi et al. 2009, Porzionato et al. 2011, Tariq et al. 2013). Out of these sites, expression is highest in the placenta with human maternal plasma kisspeptin levels in the third trimester of pregnancy up to 7000-fold greater than in non-pregnant women (Muir et al. 2001, Horikoshi et al. 2003). The role of elevated plasma kisspeptin levels in pregnancy is yet to be explained but it is possible that these high kisspeptin levels may help to down-regulate the reproductive axis and regulate trophoblastic invasion of the uterus during pregnancy.

Figure 1-4: Human kisspeptin isoforms. All the kisspeptin peptides share the highlighted carboxy-terminal decapeptide sequence which is necessary for biological activity (highlighted in yellow below).

Kisspeptin-54

GTSLSPPPESSGSRQQPGLSAPHSRQIPAPQGAVLVQREKDLPNYNWNSFGLRF-NH2

Kisspeptin-14

DLPNYNWNSFGLRF-NH2

Kisspeptin-13

LPNYNWNSFGLRF-NH2

Kisspeptin-10

YNWNSFGLRF-NH2

1.6 The kisspeptin receptor (KISS1R)

The kisspeptin receptor (formerly known as GPR54, AXOR12 or hOT7T175) is the product of the KISS1R gene, and was identified 5 years after the KISS1 gene by three independent groups (Kotani et al. 2001, Muir et al. 2001, Ohtaki et al. 2001). KISS1R maps to chromosome 19p13.3 and contains five exons encoding a 398 amino acid protein with seven hydrophobic transmembrane domains (Muir et al. 2001). It is a member of the rhodopsin family of seven transmembrane G-protein-coupled receptors (GPCR) with structural similarities (40% identity) to the galanin receptor although it does not bind galanin or galanin-like peptide (Lee et al. 1999, Muir et al. 2001).

The kisspeptin receptor is a class A GPCR coupled to Gq/11. Binding of kisspeptin to the kisspeptin receptor results in activation of phospholipase C, resulting in phosphatidylinositol 4,5-bisphosphate hydrolysis followed by accumulation of inositol-

(1,4,5)-triphosphate and diacylyglycerol. This results in Ca2+ mobilisation which mediates kisspeptin’s function (Muir et al. 2001, Liu et al. 2008, Constantin et al. 2009). Thereafter the kisspeptin receptor undergoes internalisation, recycling and subsequent recruitment from an intracellular pool (Min et al. 2014).

The standard nomenclature is KISS1R in humans and Kiss1r in non-human species. If both KISS1R and Kiss1r are referred to in a single sentence then just KISS1 or KISS1R is used for simplicity.

1.7 KISS1R expression: CNS distribution

1.7.1 Mouse

Similarly to Kiss1, Kiss1r is expressed widely throughout the CNS. Importantly in reproduction, Kiss1r is expressed in 70-90% of GnRH neurons (Figure 1-5), the majority of whom reside in the POA (Irwig et al. 2004, Han et al. 2005, Messager et al. 2005).

Utilising X-gal histochemistry, whereby β-galactosidase was knocked into the Kiss1r gene, expression has been detected in additional areas including the dentate gyrus of the hippocampus (highest expression), septum, anteroventral nucleus of the thalamus, posterior hypothalamus, periaqueductal grey, supramammilary and pontine nuclei, and dorsal cochlear nucleus. Of note, no Kiss1r expression was detected in the ARC or AVPV

(both sites with strong Kiss1 expression) (Herbison et al. 2010).

1.7.2 Rat

In the rat, Kiss1r mRNA has been identified through northern blot and in situ hybridisation in similar regions including the pons, midbrain, thalamus, hypothalamus, hippocampus, amygdala, cortex, frontal cortex, and striatum (Lee et al. 1999). Other studies have localised Kiss1r to additional areas including the diagonal band of Broca, medial septum, medial preoptic area, lateral preoptic area, median preoptic nucleus and anterior and lateral hypothalamus (Irwig et al. 2004). Interestingly, Kiss1r mRNA has also been detected in the pituitary gonadotrophs, the site of gonadotrophin secretion (Richard et al.

2008).

1.7.3 Primates (including humans)

In humans, high levels of KISS1R mRNA have been identified in the pituitary, cerebral cortex, amygdala, putamen, globus pallidus, cingulate gyrus, medulla, thalamus, pons, caudate, substantia nigra and cerebellum (Clements et al. 2001, Kotani et al. 2001, Muir et al. 2001, Ohtaki et al. 2001). A variety of methodologies have been used including reverse transcriptase polymerase chain reaction, northern blotting and immunohistochemistry.

1.8 KISS1R expression: Peripheral distribution

Kiss1r has been detected peripherally in the liver and intestine in rats (Lee et al. 1999) while studies in humans have identified a number of other sites. The highest peripheral expression has been detected in the placenta and pancreas, while other sites include the spleen, kidney, testis, ovary, spinal cord and lymph nodes (Kotani et al. 2001, Muir et al.

2001, Ohtaki et al. 2001).

The widespread detection of KISS1R (in a similar way to KISS1) suggests an importance for KISS1/KISS1R signalling beyond simply reproduction which will be covered later.

1.9 The importance of the KISS1/KISS1R system in the reproductive axis

1.9.1 KISS1R

The interest in the KISS1/KISS1R system amongst the reproductive endocrine community originated in 2003 with two seminal publications describing deletions and inactivating mutations in the KISS1R gene in a French and Saudi Arabian family respectively (de

Roux et al. 2003, Seminara et al. 2003). Deletions/mutations of the KISS1R gene result in hypogonadotrophic hypogonadism and hence a failure to go through puberty in humans and rodents (de Roux et al. 2003, Funes et al. 2003, Seminara et al. 2003, Lanfranco et al. 2005, Semple et al. 2005, Cerrato et al. 2006, Tenenbaum-Rakover et al. 2007).

Humans with these mutations have normal pituitary function except for reduced gonadotrophin release (Seminara et al. 2003). Meanwhile, treatment of individuals with

KISS1R deletions/mutations with GnRH or gonadotrophins, hence by-passing the kisspeptin system, has resulted in temporary restoration of the reproductive axis with successful gametogenesis, placental function and lactation (Lanfranco et al. 2005, Pallais et al. 2006, Tenenbaum-Rakover et al. 2007).

In addition Kiss1r null mice display hypogonadotrophic hypogonadism although they have anatomically normal GnRH neurones and are phenotypically normal in all other respects

(de Roux et al. 2003, Funes et al. 2003, Seminara et al. 2003). As in humans with KISS1R mutations, administration of GnRH to these Kiss1r null mice results in normal gonadotrophin release from the pituitary (Funes et al. 2003, Seminara et al. 2003).

Furthermore kisspeptin administration does not stimulate gonadotrophin release in these mice suggesting that kisspeptin acts upstream of GnRH (Messager et al. 2005).

Conversely activating mutations of the KISS1R gene stimulate the reproductive axis. In humans a KISS1R activating mutation has been reported resulting in central precocious

(early) puberty as a result of extended kisspeptin receptor activation (Teles et al. 2008).

1.9.2 KISS1

More recently mutations have been identified in the KISS1 gene in humans, resulting in similar phenotypes to those observed in humans with KISS1R mutations. In 2012, an inactivating mutation in the KISS1 gene in a large Turkish consanguineous family resulting in failure of pubertal progression was described by Topaloglu and colleagues (Topaloglu et al. 2012). Conversely two different activating mutations in the KISS1 gene have been described in children with central precocious puberty (Silveira et al. 2010).

Kiss1 null mice are healthy and have no phenotypic abnormality other than a failure of pubertal progression. This is accompanied by low circulating gonadotrophins and sex steroids as well as small gonads. Peripheral administration of kisspeptin to these Kiss1 null mice robustly stimulates gonadotrophin secretion as expected due to presence of functional Kiss1r (d'Anglemont de Tassigny et al. 2007). The very similar phenotype of

Kiss1r and Kiss1 null mice provides evidence that kisspeptins are the physiological ligand for the Kiss1r in vivo.

These studies of the KISS1 and KISS1R gene together demonstrate the fundamental importance of the KISS1/KISS1R system in the normal functioning of the reproductive axis.

1.10 Kisspeptin administration can induce puberty in animals

The KISS1/KISS1R system can also be manipulated to induce pubertal progression. In immature pre-pubertal female rats, precocious puberty can be initiated by the administration of kisspeptin (Matsui et al. 2004, Navarro et al. 2004). While in juvenile rhesus monkeys, administration of kisspeptin induced pulsatile LH secretion characteristic of the pubertal state which was subsequently abolished by the GnRH antagonist acycline providing further evidence that kisspeptin acts upstream of GnRH (Plant et al. 2006). In keeping with this, another study in rodents (using dual label immunofluorescence techniques) demonstrated that Kiss1 neurones have axonal projections to GnRH neurones, with a marked increase in the number of Kiss1 neurones juxtaposing GnRH cell bodies occurring at the onset of puberty (Clarkson and Herbison 2006). Interestingly, in humans, it has recently been observed that circulating kisspeptin is elevated in both boys and girls compared with adults with a peak concentration occurring between 9 and 12 years of age (Jayasena et al. 2014). Together these data suggest that increased kisspeptin signalling may indeed be part of the trigger for pubertal maturation.

1.11 Kisspeptin stimulates the release of GnRH which in turn stimulates gonadotrophin release

Administration either centrally (intracerebroventricular) or peripherally of the short form of kisspeptin, kisspeptin-10, in a number of animal models (including rodents, sheep, pigs, and primates) stimulates gonadotrophin release with the effect on LH being more marked than that on FSH (Gottsch et al. 2004, Irwig et al. 2004, Matsui et al. 2004, Navarro et al.

2004, Thompson et al. 2004, Messager et al. 2005, Plant et al. 2006, Seminara et al.

2006, Arreguin-Arevalo et al. 2007, Caraty et al. 2007, Ramaswamy et al. 2007, Lents et al. 2008). The mechanism for this is now established to be via the stimulation of GnRH release through activation of the kisspeptin receptor on GnRH neurones (Figure 1-5). This is evidenced by several studies:

1. The GnRH neurones of primates, rodents and sheep are found in close apposition with kisspeptin neurones in the hypothalamus (Clarkson and Herbison 2006).

2. GnRH neurons express the kisspeptin receptor in rodents (Irwig et al. 2004, Messager et al. 2005).

3. The treatment of rodent GnRH neurones with kisspeptin-10 increases their electrophysiological activation, and triggers GnRH release (Irwig et al. 2004, Han et al.

2005). However, this effect of kisspeptin on GnRH release is not observed in Kiss1r null mice (d'Anglemont de Tassigny et al. 2008).

4. Intracerebroventricular administration of kisspeptin-10 to sheep stimulates an increase in cerebrospinal fluid GnRH (Messager et al. 2005).

5. Intracerebroventricular administration of kisspeptin-52 to rats induces c-Fos (a marker of neuronal activation) expression in the majority of GnRH neurones (Irwig et al. 2004).

6. GnRH antagonists abolish the stimulatory effects of kisspeptin-10 on gonadotrophin release (Gottsch et al. 2004, Irwig et al. 2004, Matsui et al. 2004, Shahab et al. 2005).

7. Administration of GnRH to humans and mice with kisspeptin receptor mutations results in gonadotrophin release (Seminara et al. 2003).

Peripheral kisspeptin-10 is therefore believed to stimulate the release of GnRH from the hypothalamus, which then stimulates gonadotrophin release from the pituitary gland

(Figure 1-5). The longer kisspeptin peptide, kisspeptin-54, has also been shown to stimulate gonadotrophin release in rodents and humans as detailed below in Section 1.15

(Dhillo et al. 2005, Patterson et al. 2006, Tovar et al. 2006, Dhillo et al. 2007).

Figure 1-5: Neuroanatomy and feedback system of the kisspeptin-reproductive axis in rodents and humans. In the rodent, kisspeptin (Kiss1) is released by the kisspeptin neurones in the AVPV and ARC which activates kisspeptin receptors on GnRH neurone cell bodies (in the preoptic area, POA) and dendritic terminals (in the median eminence, ME). There is subsequent release of GnRH into the hypophyseal portal circulation which stimulates anterior pituitary gonadotrophs to release the gonadotrophins LH and FSH. Sex steroids exert negative feedback on the ARC kisspeptin neurones and positive feedback on the AVPV kisspeptin neurones. In humans, the Infundibular nucleus contains the majority of kisspeptin neurones. A second population in the rostral periventricular area (RP3V) has been identified in women which may be positively regulated by sex steroids(Hrabovszky et al. 2010).

1.12 Direct effects of kisspeptin on pituitary gonadotrophin secretion

There is evidence that kisspeptin may also directly stimulate pituitary gonadotrophs to release the gonadotrophins independent of GnRH. Firstly, both kiss1 and kiss1r are expressed by the pituitary gonadotrophs (Kotani et al. 2001, Richard et al. 2008).

Secondly, kisspeptin administration (at pharmacological levels) to pituitary explants results in gonadotrophin secretion (Gutierrez-Pascual et al. 2007). And finally, low levels of kisspeptin have been detected in the hypophyseal-portal circulation in sheep (Smith et al.

2008). However, based on the pharmacological levels of kisspeptin required to stimulate gonadotrophin secretion from pituitary explants and the low kisspeptin levels in the hypophyseal-portal circulation, it is generally accepted that indirect stimulation through increased GnRH secretion is the principal physiological pathway for gonadotrophin secretion.

1.13 Kisspeptin stimulates the pre-ovulatory LH surge and ovulation in animals

Ovulation in animals (including humans) occurs as a result of a pre-ovulatory LH surge which stimulates final maturation and release of the oocyte from the ovarian follicle. The

LH surge occurs due to positive feedback of oestrogen (and progesterone) on GnRH secretion.

Wild type mice which are ovariectomised and supplemented with oestrogen (and progesterone) are able to generate LH surges, however Kiss1 null and Kiss1r null mice under the same experimental condition are unable to generate LH surges (Clarkson et al.

2008). Interestingly and in direct contradiction to this, a study a year earlier reported that

Kiss1r null mice could indeed mount an oestrogen-induced LH surge with accompanied

36 activation of GnRH neurones (as determined by c-Fos) (Dungan et al. 2007). This disparity in results may be partly explained by the different methods of generation of the kiss1r mouse. The latter study utilised retroviral insertion, which may in some cases generate a hypomorphic allele with residual gene function (Couldrey et al. 1999) whereas the former study used an established mouse strain with complete kiss1r knock-out

(Seminara et al. 2003). This suggests that the former study may be the preferred model and highlights the essential role for the kiss/kiss1r system in co-ordinating sex steroid signals and GnRH neuronal activation.

Rodent studies have provided strong evidence that the AVPV kisspeptin neurones (which are positively regulated by oestrogen) are integral in the pre-ovulatory GnRH/LH surge

(Smith et al. 2006, Herbison 2008). Studies in female rats demonstrated increased Kiss1 expression in AVPV hypothalamic kisspeptin neurons during the evening of proestrus

(high oestrogen state) and during an oestradiol/progesterone-induced LH surge in ovariectomised animals (Smith et al. 2006, Herbison 2008). In addition, c-Fos expression

(a marker of neuronal activation) was also increased in these specific AVPV kisspeptin neurons during the LH surge while this surge as well as oestrous cyclicity was abolished by immunoneutralisation of kisspeptin neurones using a specific kisspeptin monoclonal antibody directed to the POA close to the GnRH neurones (Kinoshita et al. 2005). In keeping with this, oestrogen administration to female rodents results in increased Kiss1 mRNA in the AVPV (Smith et al. 2005, Smith et al. 2006, Adachi et al. 2007).

Interestingly the ARC kisspeptin neurones exhibit the opposite effects, with reduced Kiss1 and c-Fos expression under the same experimental conditions (Smith et al. 2006).

However, in female rodents (compared to males) there are substantially more AVPV than

ARC kisspeptin neurones helping to explain the predominant positive feedback of oestradiol during the LH surge when oestradiol levels are high (Clarkson and Herbison

2006).

Given the importance of oestrogen in modulating the reproductive axis (Figure 1-5), it is interesting to note the absence of oestrogen receptor α (ERα) on GnRH neurons

(Petersen et al. 2003). This may be explained by the presence of ERα on the ARC and

AVPV kisspeptin neurones which provide afferent signals to the GnRH neurones. Indeed, blockade of the ERα receptor blocks the LH surge as well as decreasing the effects of exogenously administered kisspeptin (Roa et al. 2008, Roa et al. 2008). These studies therefore suggest that ER signalling in kisspeptin neurons contributes to the LH surge in the rodent pre-ovulatory phase.

The above studies demonstrate the importance of kiss1/kiss1r in generating the pre- ovulatory surge. In addition subcutaneous kisspeptin administration can itself trigger ovulation in gonadotrophin-primed pre-pubertal rats, an effect which is abolished by pre- treatment with a GnRH antagonist (Matsui et al. 2004). In keeping with this, intravenous infusion of kisspeptin can induce timed LH surges and ovulation in cyclical ewes as well as anoestrous (acyclic) ewes (Caraty et al. 2007). Most recently kisspeptin has been shown to stimulate successful ovulation in a dose-dependent manner in the musk shrew

(Inoue et al. 2011).

1.14 The pre-ovulatory LH surge and ovulation in humans

Although it may be expected that the human LH surge may follow a similar mechanism to the above, it is important to acknowledge inter-species variation. The ARC population of kisspeptin neurones in rodents are equivalent to the infundibular and preoptic kisspeptin neurones in humans, however an anatomical equivalent to the AVPV rodent kisspeptin neurones have not been conclusively identified in humans. However, recent studies have identified a population of kisspeptin neurones located in the rostral periventricular zone that are positively regulated by oestradiol and are only present in women. This RP3V kisspeptin population may therefore serve as a functional equivalent to the rodent AVPV kisspeptin neurones and therefore play a role in the human pre-ovulatory LH surge and ovulation (Hrabovszky et al. 2010, Smith et al. 2010).

Another point to consider is that the LH surge in non-human mammals is known to be a result of GnRH stimulation (Pau et al. 1993). However in humans some studies suggest that the pituitary gonadotrophs respond to rising circulating oestrogen levels to generate the LH surge themselves rather than via an increase in hypothalamic GnRH as seen in rodents (Knobil et al. 1980, Adams et al. 1994). Simply put, in humans the surge may result from a pituitary mechanism rather than a hypothalamic one. Kisspeptin may, however, still modulate this pituitary-based system as pituitary gonadotrophs also express the kisspeptin receptor. Hence it is feasible that kisspeptin signalling at the pituitary level may augment gonadotroph responses to oestrogen to generate the LH surge. Further work is needed to clarify the mechanism for the human pre-ovulatory LH surge.

1.15 Kisspeptin: First administration to humans

The kisspeptin receptor can be activated by all the kisspeptins as they share the common carboxy-terminal decapeptide necessary for biological activity. In humans kisspeptin-10 and kisspeptin-54 have been studied to date.

In 2005, following on from animal studies demonstrating the significant effects of kisspeptin-10 on the reproductive axis, the first human study examining the effects of kisspeptin-54 was published. This double-blind, placebo-controlled, crossover trial in healthy male volunteers showed that a 90 minute intravenous (IV) infusion of 4 pmol/kg/min kisspeptin-54 increased mean serum LH at 90 minutes by more than 2-fold when compared to saline controls. In addition serum FSH also increased by 18%. The increase in these gonadotrophins contributed to a 13% rise in testosterone at 180 minutes

(Dhillo et al. 2005). These changes in sex hormones did not result in any behavioural changes, while reverse-phase fast protein liquid chromatography (FPLC) confirmed that the elevated plasma kisspeptin levels were due to the administered peptide itself rather than any other endogenous sources or cleavage of the administered peptide (Dhillo et al.

2005).

Two years later, in 2007, a similar result was seen in healthy female volunteers when subcutaneous kisspeptin-54 was administered (0.2-6.4 nmol/kg) during the follicular phase of the menstrual cycle (Dhillo et al. 2007). Plasma kisspeptin levels were elevated in a dose-dependent manner following injection with peak levels achieved within 30 minutes. Akin to the male study above, subsequent increases in LH and FSH were also dose-dependent. Consistent with the data from rodent (Thompson et al. 2004, Navarro et al. 2005) and the above human male study, the release of LH was almost 7-fold higher than that of FSH, demonstrating the more pronounced effect of kisspeptin on LH than

FSH. Similar to the male study, no adverse effects on heart rate or blood pressure were noted during or after the studies (Dhillo et al. 2007). This was of particular importance

40 given the vasoconstrictor properties of kisspeptin that had been demonstrated in vitro

(Mead et al. 2007).

1.16 Kisspeptin effects are more pronounced during the pre- ovulatory phase

The brains of males and females have marked physiological and anatomical differences.

These include sex differences in neurone size, number, morphology and patterns of gene expression. GnRH neurones where many afferent signals converge are not themselves sexually dimorphic which suggests that there are other factors responsible for the sex differences seen in physiology and behaviours such as the timing of sexual maturation and the ability of females to generate an LH surge in the pre-ovulatory phase.

Human studies have shown for the first time that females have a more pronounced gonadotrophin response to kisspeptin-54 injection during the pre-ovulatory phase (15-16 days before the next predicted period) followed by the luteal phase and then the follicular phase of the menstrual cycle (Dhillo et al. 2007). This is in keeping with animal data which suggests that kisspeptin signalling could be crucial for generating the mid-cycle LH surge in females.

1.17 Kisspeptin-10 vs. Kisspeptin-54 and the GnRH pulse generator

The human studies described above have focused on the effects of the 54-amino acid peptide kisspeptin-54. The 145-amino acid precursor of kissepeptin-54 encoded by KISS1 is also processed to 14, 13 and 10 amino acid sequences (Kotani et al. 2001). While there have been many studies investigating the effects of kisspeptin-10 in non-human animals

41 as listed above, there have only been 2 studies in human males and importantly none in human females. Animal studies to date suggest that kisspeptin-10 and kisspeptin-54 act similarly to stimulate reproductive hormone release however kisspeptin-10 is characterised by a shorter duration of action (on testosterone levels) and faster onset of action after IP administration in rodents (Mikkelsen et al. 2009). This suggests that there are differences in the pharmacokinetic handling of these two forms of kisspeptin. In humans, the half-life of kisspeptin-54 is 28 minutes. Until the work in this thesis, the precise half-life of kisspeptin-10 in humans was unknown (see Chapter 3).

Recent human male studies have demonstrated that IV bolus kisspeptin-10 potently stimulates LH secretion (with a smaller stimulatory effect on FSH) while continuous infusion also increases testosterone, LH pulse frequency and amplitude (George et al.

2011). In addition, administration of kisspeptin-10 to men resets the GnRH pulsatile clock by inducing an immediate LH pulse regardless of the timing of the previous pulse and these pulses are on average of greater amplitude than endogenous pulses (Chan et al.

2011). The relationship between kisspeptin and LH pulses has also been examined in females, but with the longer isoform (kisspeptin-54) and with a subcutaneous mode of administration. Together with my colleagues, I showed that a single injection of kisspeptin-

54 increases the LH pulse frequency in healthy women (mean increase in number of LH pulses / 4 hours, following injection: -0.17 ± 0.54, saline; +2.33 ± 0.56 kisspeptin-54

(0.3nmol/kg)) (Jayasena et al. 2013).

In summary, both exogenous kisspeptin-10 and kisspeptin-54 stimulate gonadotrophin release in healthy men, with a greater effect seen on LH compared to FSH. In addition kisspeptin-10 can increase LH pulse amplitude and frequency as well as resetting the

GnRH pulse generator. In women, subcutaneous kisspeptin-54 administration also stimulates gonadotrophin release (most potently in the pre-ovulatory phase) and increases LH pulse frequency but the effects of kisspeptin-10 were unknown in women until work from this thesis (see Chapter 3).

1.18 Kisspeptin administration in disorders of reproduction

1.18.1 Hypothalamic Amenorrhoea

Women with functional hypothalamic amenorrhoea are hypogonadal or eugonadal, have low oestradiol levels and lack menstrual cycles without any associated organic pathology.

This results from a failure of pulsatile GnRH secretion from the hypothalamus and may be triggered by energy deficits associated with weight loss, exercise or psychological stress.

These women are infertile and studies indicate that hypothalamic amenorrhoea is a major cause of infertility, accounting for 30% of cases of amenorrhoea (Reindollar et al. 1986).

In 2009, it was reported that kisspeptin-54 acutely stimulates gonadotrophin secretion in women with functional hypothalamic amenorrhoea (HA) due to low body weight (Jayasena et al. 2009). In this study kisspeptin-54 (6.4 nmol/kg) was administered subcutaneously twice daily for 2 weeks. The effect on gonadotrophin secretion was most marked following the first injection, with much diminished effect (tachyphylaxis) after 2 weeks of injections

(Jayasena et al. 2009). However, when the injection frequency was changed from twice daily to twice weekly, the gonadotrophin response was this time sustained (Jayasena et al. 2010). This study demonstrated for the first time a possible therapeutic role for kisspeptin as well as revealing tachyphylaxis after twice daily kisspeptin injection that will be discussed in further detail in Chapter 4. More recently intravenous infusion of kisspeptin-54 for 8 hours has been demonstrated to temporarily increase LH pulse frequency and amplitude in women with hypothalamic amenorrhoea (Jayasena et al.

2014). In this study, higher doses and longer durations of infusions were associated with tachyphylaxis therefore highlighting once more the importance of dose and duration of kisspeptin administration in future clinical applications.

1.18.2 Hypogonadotrophic Hypogonadism secondary to TAC3 or TAC3R mutations

TAC3 and TACR3 encode the tachykinin neurokinin B (NKB) and its primary receptor

(NK3R) respectively. Humans with mutations in these genes exhibit a phenotype almost identical to that observed with KISS1 or KISS1R mutations (Topaloglu et al. 2009).

Furthermore kisspeptin neurones in the ARC of rodents (Navarro et al. 2011) and infundibular nucleus of humans (Hrabovszky et al. 2012) express both genes. In animals,

NKB has been observed to stimulate, inhibit or have no effect on reproductive hormone secretion depending on the species, route of administration and sex steroid milieu

(Navarro 2013). In humans, together with my colleagues I demonstrated that intravenous

NKB administration to healthy men and women at doses up to 5.12 nmol/kg/h did not significantly alter reproductive hormone secretion (Jayasena et al. 2014). However, continuous infusion of kisspeptin-10 (1.5 mcg/kg/h) for 12 hours restored pulsatile LH secretion in men with TAC3 and TAC3R mutations (Young et al. 2013). This suggests that

NKB signalling has a permissive role in the human reproductive axis and that NKB action is proximal to kisspeptin in this axis.

1.18.3 Hypogonadism secondary to Type 2 Diabetes Mellitus (T2DM)

Low serum testosterone is commonly observed in up to half of men with T2DM (Dandona and Dhindsa 2011) although the precise mechanism for this deficit remains unclear. In a small study of 5 men (mean age 34 years) with T2DM and concomitant biochemical hypogonadism, intravenous kisspeptin-10 infusion was able to stimulate increases in LH secretion and LH pulse frequency as well as serum testosterone. This suggests a potential therapeutic role for kisspeptin and kisspeptin agonists in restoring endogenous testosterone secretion in men with T2DM and hypogonadism (George et al. 2013).

1.18.4 Hyperprolactinaemia

Hyperprolactinaemia accounts is the commonest cause of hypogonadotrophic anovulation and is therefore one of the leading causes of female infertility (Molitch 2010). Kisspeptin neurones have recently been identified as the main conduit for this mechanism as they express the prolactin receptor (Kokay et al. 2011) whereas very few GnRH neurones do

(Grattan et al. 2007, Kokay et al. 2011). In mice, hyperprolactinaemia results in anovulation via reduced GnRH and gonadotrophin secretion, and diminished kisspeptin expression. In these mice administration of kisspeptin restores gonadotrophin secretion and ovarian cyclicity suggesting that kisspeptin neurones have a fundamental role in hyperprolactinaemic anovulation (Sonigo et al. 2012).

1.19 Chronic kisspeptin administration and tachyphylaxis

Tachyphylaxis is a well-described phenomenon in mammalian endocrinology. GnRH analogues achieve this phenomenon by continuously stimulating the GnRH receptor resulting in down-regulation of the reproductive axis. This ability of GnRH analogues is exploited by their use for medical castration as a treatment in prostate cancer (Labrie

2014).

In a similar way chronic kisspeptin administration can cause tachyphylaxis resulting in a down-regulation of the reproductive axis. For example high-dose continuous intravenous infusion of kisspeptin-10 to juvenile and adult male rhesus monkeys results in an initial acute stimulation of LH lasting 3 hours, followed by a rapid drop to baseline levels

(Seminara et al. 2006, Ramaswamy et al. 2007). During the final 3 hours of these studies they administered single boluses of kisspeptin, N-methyl-DL-aspartic acid (NMDA) and

GnRH and showed that the latter two were still able to elicit LH pulses but kisspeptin was not. This suggests desensitisation to the effects of kisspeptin at the level of the kisspeptin

45 receptor. A similar increase in LH followed by tachyphylaxis has also been seen in rodents after 2 days of a subcutaneous infusion of kisspeptin (Thompson et al. 2006). However when administered as intermittent injections, kisspeptin can induce chronic stimulation of the reproductive axis resulting in precocious puberty, as seen in juvenile rats and juvenile male monkeys, when administered twice daily injections for 5 days and hourly injections for 2 days respectively (Navarro et al. 2004, Plant et al. 2006).

Hence there exists a spectrum from stimulation to suppression of the reproductive axis that appears to be dependent on animal model as well as mode of administration, isoform, dose and duration of kisspeptin treatment. Tachyphylaxis in humans will be considered further in Chapter 4.

1.20 Therapeutic applications of kisspeptin to stimulate the reproductive axis

The shorter amino acid sequence of kisspeptin-10 makes it simpler and cheaper to synthesize than kisspeptin-54. In addition, kissepeptin-10 has potentially greater options for pharmaceutical development since agonists (with greater bioactivity than kisseptin-10) and antagonists have already been developed based on its decapeptide sequence (Curtis et al. 2010, Millar et al. 2010). Until this thesis the effects of kisspeptin-10 in healthy women were unknown. This is addressed in chapter 3.

1.20.1 In Vitro Fertilisation

In the UK one in seven couples suffer with infertility (HEFA Fertility Facts and Figures

2008). This has a profound psychological and social impact on the couple. The mainstay of treatment is the use of in vitro fertilisation (IVF) which involves using gonadotrophin

46 injections to stimulate growth of ovarian follicles followed by hCG injection to induce egg maturation. However hCG has a prolonged direct action on the LH receptors in the ovary and can lead to uncontrolled stimulation of the ovaries resulting in ovarian hyperstimulation syndrome (OHSS) with potentially life-threatening consequences.

Kisspeptin, however, may stimulate the ovaries in a more physiological and controlled manner given that its role is upstream of GnRH. Kisspeptin injections can hence be used to stimulate reproductive hormone release in reproductive pathologies as demonstrated in the study of women with KISS1R mutations or hypothalamic amenorrhoea (Seminara et al. 2003, Jayasena et al. 2010), and so potentially serve as a treatment for infertility. Very recently, together with colleagues, I have demonstrated that a single injection of kisspeptin-54 can induce egg maturation in women with subfertility undergoing in vitro fertilisation (Jayasena et al. 2014). Furthermore subsequent fertilisation of eggs matured following kisspeptin-54 administration and transfer of resulting embryos can lead to successful human pregnancy with biochemical and clinical pregnancy rates of 40% and

23% respectively (Jayasena et al. 2014).

1.20.2 Other applications

Translating the previously mentioned studies (Section 1.18) that have explored kisspeptin administration in reproductive disorders suggests a potential role for kisspeptin administration in the treatment of hypothalamic amenorrhoea, hypogonadism (due to

T2DM) and hyperprolactinaemia. Additional studies have demonstrated that kisspeptin can initiate puberty in monkeys and so kisspeptin may be a useful tool in treating children with delayed puberty (Plant et al. 2006). Evidence that kisspeptin administration can restore LH pulsatility in individuals with hypogonadotrophic hypogonadism secondary to

TAC3 or TAC3R mutations provides a further potential application (Young et al. 2013).

1.21 Therapeutic applications of kisspeptin to down-regulate the reproductive axis

As demonstrated by the different results (stimulation or tachyphylaxis) of administering kisspeptin at different time-intervals, the frequency and mode of administration can be manipulated to provide other therapeutic uses. For instance, down-regulating the reproductive axis with more frequent kisspeptin administration (tachyphylaxis) may have therapeutic benefit by providing medical castration in sex-hormone dependent cancers such as prostate cancer or as a novel form of contraception.

In addition as GnRH pulse frequency primarily determines LH rather than FSH secretion, then slowing GnRH pulses might normalise the hyper-secretion of LH commonly observed in Polycystic Ovarian Syndrome (PCOS) (Tsutsumi and Webster 2009).

1.21.1 Kisspeptin receptor antagonists

A potent kisspeptin receptor antagonist was developed in 2009 by amino acid substitution within the kisspeptin-10 decapeptide sequence (peptide-234) (Roseweir et al. 2009).

Peptide-234 provided a unique opportunity to interrogate the kisspeptin-GnRH system further. For example, peptide-234 inhibited the firing of GnRH neurones in the rodent brain indicating that kisspeptin signalling occurs proximal to GnRH neurones. In addition, peptide-234 reduced pulsatile GnRH secretion in female pubertal monkeys, confirming an important role for kisspeptin in puberty onset. Finally peptide-234 blocked the physiological ovulatory LH surge and post-castration rise in LH in sheep, rats and mice, indicating that kisspeptin neurones mediate the positive and negative feedback effect of sex steroids on gonadotrophin secretion respectively (Millar et al. 2010, Silveira et al.

2010).

In addition to aiding our understanding of kisspeptin physiology, kisspeptin receptor antagonists may also serve in future as therapeutic agents for treating hormone- dependent reproductive disorders such as precocious puberty, endometriosis and prostate cancer. The ability of kisspeptin antagonists to limit follicular development and so inhibit ovulation also suggests a potential role in female contraception. Indeed kisspeptin antagonists may be specifically advantageous in scenarios where exogenous oestrogen is contraindicated.

1.21.2 Kisspeptin analogues

Kisspeptin analogues with high potency agonist activity have also been developed. Both

KISS1-305 and TAK-448 initially stimulate LH and subsequent testosterone release, however continued (chronic) administration resulted in abrupt reductions of both hormones to castrate levels. In addition 3 weeks of KISS1-305 administration resulted in a reduction of hypothalamic GnRH content to 10-20% of control suggesting that chronic administration of kisspeptin analogues disrupts endogenous kisspeptin-GnRH signalling

(Matsui et al. 2012).

The ability of kisspeptin analogues to down-regulate the reproductive axis has been recently exploited in human phase 1 studies in men with prostate cancer. Intermittent

(monthly) subcutaneous depot TAK-448 injections resulted in sustained testosterone suppression and >50% decrease in PSA (prostate specific antigen)(Maclean et al. 2014).

1.22 Non-reproductive effects of kisspeptin

Given the widespread expression of both kisspeptin and the kisspeptin receptor throughout the human body (see above), it is interesting to briefly consider the non- reproductive roles/effects of kisspeptin signalling.

1.22.1 Cancer metastasis

The pioneering identification of the KISS1 gene related to its ability to suppress metastasis in human melanoma cell lines (Lee et al. 1996). Since then, there have been numerous studies exploring the role of KISS1/KISS1R in numerous different types of cancer. These studies have demonstrated that kisspeptin has anti-metastasis activity in a wide range of cancers including oesophageal, papillary thyroid, and ovarian cancers

(Ringel et al. 2002, Masui et al. 2004, Jiang et al. 2005, Castellano et al. 2006, Lin et al.

2011). In addition, loss of KISS1 gene expression has been identified as a poor prognostic marker for tumour progression and metastasis in several cancers including pancreatic, gastric and bladder cancers (Sanchez-Carbayo et al. 2003, Dhar et al. 2004, Nagai et al.

2009). Furthermore together with my colleagues, I recently demonstrated that plasma kisspeptin may be a potential biomarker of tumour metastasis in patients with ovarian cancer; patients with lower plasma kisspeptin levels having a higher rate of late stage ovarian cancer (Jayasena et al. 2012). It is likely that KISS1 exerts its anti-metastatic effects through a variety of mechanisms including suppression of NFkB binding to the matrix metalloproteinase-9 (MMP9) gene promoter therefore reducing the ability of the cancer cells to invade through the extra-cellular matrix (Ji et al. 2013).

1.22.2 Other neuroendocrine effects

The kisspeptin receptor, KISS1R, is expressed in the pituitary gland and so it is reasonable to hypothesise that kisspeptin signalling may play a role in the secretion of other non-gonadotrophin pituitary hormones such as growth hormone (GH) and prolactin

(PRL) (Muir et al. 2001). Kisspeptin stimulates GH secretion but not TSH secretion when administered to pituitary cells of female baboons (Smith et al. 2005) and kisspeptin stimulates PRL secretion when administered to goldfish pituitary cells in vitro (Yang et al.

2010). In vivo studies have observed increases in GH in heifers (Kadokawa et al. 2008), and increases in PRL in rats (Szawka et al. 2010) when administered kisspeptin intravenously and intracerebroventricularly respectively. In contrast, several studies have failed to identify any non-gonadotrophin secretory effects. For example intravenous bolus administration of kisspeptin failed to alter circulating levels of GH, PRL and thyroid- stimulating hormone (TSH) in adult male rhesus monkeys (Ramaswamy et al. 2009).

Similarly no effect by kisspeptin on GH secretion was seen in goats (Hashizume et al.

2010), cows (Ezzat Ahmed et al. 2009), and prepubertal gilts (Lents et al. 2008).

Consistent with these latter negative studies, I demonstrated together with my colleagues no effect of acute or chronic subcutaneous kisspeptin administration on GH, PRL and

TSH in healthy women (Jayasena et al. 2014). Hence in humans, at least, it is unlikely that kisspeptin administration has significant effects on non-gonadotropin pituitary secretion and this has beneficial implications for the safety of kisspeptin as a potential therapeutic for patients with reproductive disorders.

1.22.3 Other CNS roles

As mentioned earlier, KISS1 and KISS1R expression has been identified in many areas of the brain including extra-hypothalamic areas. For example, both genes are highly expressed in the dentate gyrus of the hippocampus in rodents, where kisspeptin may

51 increase synaptic transmission in dentate granule cells through various signalling cascades suggesting a role in cognition and the pathogenesis of epilepsy (Arai 2009). In keeping with this, kisspeptin stimulation of hippocampal slices has been observed to enhance synaptic transmission in the hippocampus (Arai et al. 2005). A possible role in autonomic and sensory neuronal signalling has also been suggested based on the expression pattern of Kiss1r in the medulla and spinal cord of rodents (Dun et al. 2003).

Another study in mice, based on the identification of Kiss1r in the dorsal root ganglia and dorsal horns of the spinal cord, has demonstrated that kisspeptin increases pain sensitivity using hot plate and formalin (inflammatory) pain tests (Spampinato et al. 2011).

More recently, a unique role for the kisspeptin system in the zebrafish habenula (a highly conserved brain region that mediates behavioural responses to stressful conditions) in inhibiting fear responses has been demonstrated (Ogawa et al. 2014). Finally, the finding of Kiss1 and Kiss1r expression in the rodent amygdala has heightened interest in the possible roles of the kisspeptin system in reproductive and non-reproductive behaviours which will be covered in further detail in Chapter 2 (Kim et al. 2011). Putting these studies together suggests that the kisspeptin system has several non-reproductive roles in the

CNS although further exploration is required to clarify them precisely.

1.22.4 Metabolism

The identification of KISS1 and KISS1R in alpha (glucagon-secreting) and beta (insulin- secreting) pancreatic islets in humans and rodents suggests a role for the kisspeptin system in glucose metabolism (Hauge-Evans et al. 2006). However, studies in human and rodent islets and perfused rat pancreata have demonstrated both stimulatory and inhibitory effects of kisspeptin on glucose-induced insulin secretion (Hauge-Evans et al.

2006, Silvestre et al. 2008, Bowe et al. 2009, Vikman and Ahren 2009, Bowe et al. 2012,

Song et al. 2014). In vivo, peripheral but not central kisspeptin administration to rats

52 increased circulating insulin levels indicating a peripheral mechanism of action (Bowe et al. 2009). A further study in adult monkeys demonstrated no effect of kisspeptin administration on basal insulin secretion but a potentiation of glucose-induced insulin secretion (Wahab et al. 2011). The different effects of kisspeptin administration on insulin secretion may relate to the differing protocols with the recent suggestion that high doses of kisspeptin stimulate, whereas low doses of kisspeptin inhibit islet insulin secretion

(Song et al. 2014).

Further metabolic data has recently emerged from a study that used Kiss1r null mice.

Female (but not male) Kiss1r mice exhibited dramatically higher body weight and adiposity along with impaired glucose tolerance therefore supporting an insulin stimulatory role for kisspeptin (Tolson et al. 2014). However the reason for the observed sexually dimorphic metabolic phenotype given that no sex differences in pancreatic kiss1r expression have been reported, remains unclear.

Another interesting finding relates to the expression of Kiss1 in white adipose tissue, which is increased by sex steroids and fasting but reduced by a high fat diet in rodents suggesting a role for kisspeptin signalling in adipose metabolism (Brown et al. 2008).

Along these lines peripheral kisspeptin administration increases circulating adiponectin and decreases plasma resistin and leptin levels in fasted monkeys (Wahab et al. 2010).

In conclusion, although there are clearly metabolic roles for the kisspeptin system, the available studies so far are conflicting and fragmentary highlighting a need for further study in this area.

1.22.5 Cardiovascular system

KISS1 as well as KISS1R has been identified in smooth muscle cells of the human aorta, coronary arteries with ex vivo vasoconstrictor actions following kisspeptin binding (Mead et al. 2007). In keeping with this kisspeptin administration to mice decreased peripheral blood flow as well as inducing oedema formation after intradermal injection (Sawyer et al.

2011). KISS1 and KISS1R expression has also been demonstrated in the endothelial and smooth muscles cells of intramyocardial blood vessels and myocytes in humans.

Furthermore, kisspeptin administration to human paced atrial tissue revealed a positive inotropic effect (Maguire et al. 2011). It is important to note that although these studies suggest notable cardiovascular effects of kisspeptin, in vivo studies of kisspeptin administration to healthy men and women, have not revealed any effect on heart rate or blood pressure (Dhillo et al. 2005, Dhillo et al. 2007).

1.22.6 Fluid homeostasis

There exists data suggesting that kisspeptin may increase vasopressin levels in rats

(leading to inhibition of diuresis) (Scott and Brown 2011), and aldosterone secretion in vitro from human adrenal cells (Nakamura et al. 2007, Takahashi et al. 2010). In addition a further study observed alterations in Kiss1/Kiss1r expression in the kidney in rodent models of chronic renal impairment (Shoji et al. 2010). Again, no fluid changes have been reported in human kisspeptin administration studies although there is no study as yet examining this directly.

1.22.7 Summary

In summary, there are a number of confirmed and unconfirmed roles for the kisspeptin system based on the above animal and human studies. Their significance is worthy of future study given the prospect that kisspeptin may in future be used as a therapeutic agent for reproductive disorders (Sonigo et al. 2012, Ratnasabapathy and Dhillo 2013,

Jayasena et al. 2014). Importantly no adverse effects have been reported as yet following kisspeptin administration in humans (Dhillo et al. 2005, Dhillo et al. 2007, Chan et al.

2011, George et al. 2011, Jayasena et al. 2013, Jayasena et al. 2013, Jayasena et al.

2014).

1.23 Aims and Hypotheses of the thesis

The kisspeptin system is critical for normal functioning of the reproductive axis. Since its identification almost two decades ago, almost 1000 papers have been published describing various aspects. This brings further questions and the possibility that kisspeptin can be exploited therapeutically in disorders of reproduction. To this end, in this thesis I have endeavoured to address several outstanding issues that are of fundamental importance to both kisspeptin physiology and the development of kisspeptin therapeutics:

A: To explore the extra-hypothalamic effects of kisspeptin administration using neuroimaging techniques

Most of the previous literature has focussed on the hypothalamic effects of kisspeptin, with a paucity of data exploring the extra-hypothalamic effects. KISS1 and KISS1R are expressed widely in the rodent and human brain suggesting roles beyond the hypothalamus (Muir et al. 2001, Clarkson et al. 2009, Herbison et al. 2010). An area of particular interest is the amygdala as it has an established role in various reproductive behaviours and hormonal effects (Murray 2007). To develop kisspeptin therapeutics, a better understanding of the extra-hypothalamic effects of kisspeptin administration is required.

Recent advances in functional neuroimaging now permit the assessment of neuronal activity following peptide administration. I propose to optimise these techniques in an attempt to address the aim of exploring the extra-hypothalamic effects of kisspeptin administration particularly within the amygdala.

HYPOTHESIS: Peripheral kisspeptin administration alters extra-hypothalamic neuronal activity.

B: To investigate the effects of kisspeptin-10 on the reproductive axis in healthy men and women

Kisspeptin-54 administration to men and women has been studied in detail previously

(Dhillo et al. 2005, Dhillo et al. 2007). The shorter amino acid sequence of kisspeptin-10 makes it simpler and hence cheaper to synthesise and so may preferable from a therapeutic point of view. Continuous kisspeptin-10 administration to healthy men results in increased LH secretion and pulsatility, and subsequent increased testosterone secretion (Chan et al. 2011, George et al. 2011). However, the effects of kisspeptin-10 in healthy women are unknown and I therefore aim to investigate this in detail.

HYPOTHESIS: Kisspeptin-10 stimulates reproductive hormone secretion in healthy men and women in a similar manner to kisspeptin-54.

C: To investigate the effects of chronic kisspeptin-54 administration in healthy women

The literature so far suggests that chronic kisspeptin administration to animals and women with hypothalamic amenorrhoea results in tachyphylaxis and a subsequent down- regulation of the reproductive axis (Seminara et al. 2006, Jayasena et al. 2009, Jayasena et al. 2010). However, the effects of chronic kisspeptin administration have not been investigated in healthy women and may have clinical implications for the use of kisspeptin to stimulate or down-regulate the reproductive axis therapeutically.

HYPOTHESIS: Chronic kisspeptin-54 administration to healthy women will result in tachyphylaxis (similar to that observed in women with hypothalamic amenorrhoea) and a resultant down-regulation of the reproductive axis.

Chapter 2 ______

The effects of kisspeptin administration on neuronal activation as determined by manganese-enhanced magnetic resonance imaging

2.1 Introduction

KISS1 and KISS1R expression is not limited to the hypothalamus (Lee et al. 1999,

Clarkson et al. 2009), yet there is a paucity of data exploring the brain more widely. Given the established role of the KISS1/KISS1R system in reproduction, an area of particular interest is the amygdala, whose functions contribute to a broad range of social and reproductive behaviours (Murray 2007), as well as gonadotropin secretion and oestrous cyclicity (Bar-Sela and Critchlow 1966, Lawton and Sawyer 1970, Bagga et al. 1984).

More recently a pivotal role for the amygdala in the stress-induced suppression of gonadotrophin secretion has been demonstrated (Lin et al. 2011). Furthermore, neuroanatomical studies have demonstrated neuronal projections between the amygdala and hypothalamic regions that regulate reproductive hormone release such as the ARC,

AVPV and POA (Canteras et al. 1995, Hahn et al. 2003, Keshavarzi et al. 2014). Hence it is evident that the amygdala plays a key role in many behaviours and hormonal mechanisms related to reproduction.

The finding of KISS1R expression within the human and rodent amygdala (Lee et al.

1999, Muir et al. 2001) as well as the more recent observation of sex steroid-dependent

KISS1 expression within the rodent amygdala (Kim et al. 2011), has heightened interest in the extra-hypothalamic roles of kisspeptin. Examining the effects of kisspeptin on neuronal activation in the hypothalamus and beyond, particularly within the amygdala, is therefore of great interest.

To this end, I applied recent advances in neuroimaging to optimise manganese-enhanced magnetic resonance imaging (MEMRI) to detect signal intensity as a surrogate for neuronal activity after peripheral administration of kisspeptin in rodents. Regions of interest were selected for detailed investigation as detailed below.

2.1.1 Regions of Interest (ROIs)

5 regions of interest were selected for study:

Amygdala

The left and right amygdalae were selected for reasons detailed above.

Arcuate (ARC) and Anteroventral Periventricular (AVPV) Nuclei

The ARC and the AVPV nuclei in the hypothalamus contain the most important and studied populations of Kiss1-expressing neurones that play a critical role in reproduction

(Clarkson et al. 2009). No Kiss1r expression is observed in these sites in mice (Herbison et al. 2010). The left and right ARC were analysed separately to examine for any lateralising difference in response to kisspeptin.

Preoptic Area (POA)

The POA is the predominant location of GnRH neurones, up to 90% of which express the kisspeptin receptor (Irwig et al. 2004, Han et al. 2005). The POA ROI used for analysis also included the nearby medial septal area which contains a smaller additional population of GnRH neurones (Herbison et al. 2010).

Anterior Pituitary

The anterior pituitary contains gonadotrophs that release LH in response to GnRH. In addition, there exists a smaller effect due to direct kisspeptin action on Kiss1r-expressing gonadotrophs (Gutierrez-Pascual et al. 2007).

2.1.2 Principles of Magnetic Resonance Imaging (MRI)

MRI is based on the physical phenomenon of spin, which describes the angular momentum inherent in protons and neutrons. Nuclei with an odd total number of protons and neutrons will possess this spin. The hydrogen nucleus which consists of a single free proton therefore possesses spin and serves as the focus for MRI. In addition, in biological tissues, hydrogen is associated with water which forms the basis for differentiating between different tissue densities.

Associated with the spin of the proton is a magnetic dipole moment, running parallel to the spin axis. When a static external magnetic field (B0) is applied across a tissue, the free protons align with the direction of the field, resulting in a net magnetic moment (also termed net magnetisation vector, NMV) parallel with B0.

Generation of MRI images requires intermittent application of a radiofrequency (RF) pulse perpendicular to B0. The protons ‘tip’ to align with the RF pulse, thus gaining energy. After the RF pulse stops, the protons then re-align with B0 and return (‘relax’) to the low energy state. This ‘relaxation’ results in changes to the magnetic field that can be detected though induction of a current in a receiving coil and converted to images. To acquire spatial information, the strength of the magnetic field is modulated by gradient coils across the volume to be imaged.

The T1 is the longitudinal relaxation time, therefore indicating the time required for re- alignment of spin with B0 in the longitudinal plane after the RF pulse is switched off. This

T1 time is long for protons in water molecules whereas it is typically short in the case of protons in fat.

At the same time as this longitudinal relaxation, there is de-phasing of individual precessing spins in the transverse plane. The time constant relating to this de-phasing is termed T2. The magnetic resonance signal from a tissue is therefore dependent primarily

61 dependent on the proton density and the T1 and T2 relaxation times. These properties therefore define the visible contrast on the resulting MR images. By modifying the time between successive RF pulses and so the degree of T1 and T2 relaxation, the MR images can be weighted in favour of T1 or T2 which have distinctly different contrast properties.

The repetition time (TR) represents the time elapsed between successive RF pulses.

Shortening the TR permits full recovery of tissues with a short T1 but only a partial recovery of tissues with a long T1. Full recovery results in a greater net signal following subsequent RF pulses. The time interval between the RF pulse and the first measurement is termed the echo delay time (TE). These can be manipulated so that a short TR increases T1 effects and a short TE minimises T2 effects. By selecting the appropriate TR and TE, it is therefore possible to weight an image to accentuate signal from particular tissue types that have different proton densities. In T1-weighted images, fat is bright and solid/high-water content structures are dark. In T2-weighted images, the situation is reversed with fat appearing dark and solid/high water content structures appearing bright.

The relaxation properties of the various tissues undergoing MRI can also be modified by the administration of exogenous paramagnetic agents.

2.1.3 Manganese-enhanced Magnetic Resonance Imaging

(MEMRI)

The paramagnetic properties of the divalent manganese ion (Mn2+) make it an excellent

MRI contrast agent. This results from the ability of the unpaired electrons in the outer shell of Mn2+ ions to alter the relaxation properties of protons in a magnetic field leading to an increased T1-weighted signal (Mendonca-Dias et al. 1983). Hence the signal is proportional to the concentration of Mn2+ in the tissue. In addition, Mn2+ ions exhibit the biological capacity to mimic calcium ions (Ca2+) and enter L-type voltage-gated calcium channels (Hunter et al. 1981, Narita et al. 1990) . These channels are located on various excitable cells such as neurones. Therefore when an action potential arrives at a neuronal presynaptic membrane, it results in the opening of voltage-gated calcium channels and influx of Ca2+ together with any available Mn2+ into the neurone. Therefore Mn2+ accumulates in neurones in an activity-dependent manner. Mn2+ ions also cause marked

2+ shortening of the T1, resulting in image contrast at the sites of Mn accumulation.

Increased MRI signal intensity (SI) is due to intracellular rather than extracellular Mn2+ accumulation as tissue enhancement lasts longer than blood enhancement and the volume of distribution of Mn2+ is equivalent to the intracellular volume (Mori et al. 2004).

The accumulation of Mn2+ in a region of the CNS is dependent on both the kinetics of Mn2+ diffusion to that region as well as neuronal activity. The kinetics of Mn2+ entry into the CNS are complex in the presence of an intact blood-brain-barrier (BBB). The passage of Mn2+ ions into the CNS occurs predominantly via the blood-cerebrospinal fluid barrier at the choroid plexus (Rabin et al. 1993). Other important sites are the circumventricular organs

(CVO) which are regions of the CNS that are characterised by incomplete coverage by the

BBB and a comprehensive vesicular transport mechanism for the movement of molecules between blood and cerebrospinal fluid (CSF) (Wagner et al. 1974). The CVO include the median eminence (ME), which is located at the base of the hypothalamus, and the area

63 postrema located adjacent to the 4th ventricle in the brainstem. Therefore both these areas are important crossing points for circulating Mn2+ in the blood to access the CNS; as a consequence the hypothalamus and brainstem are both exposed to circulating Mn2+ ions to a greater extent than other areas of the CNS. This property is exploited by numerous studies of hypothalamic nuclei activity using MEMRI.

In some studies, particularly those examining response to physical stimuli, hyperosmotic agents have been used to disrupt the BBB and facilitate faster entry of Mn2+ into the CNS

(Aoki et al. 2002). However when studying the effects of peripheral administration of peptides, it is preferable to keep the BBB intact as is physiological.

These properties of Mn2+ make MEMRI well-suited to imaging studies of neuronal connections and neuro-architecture in animal species (Lin and Koretsky 1997, Pautler

2004, Watanabe et al. 2004, Jung et al. 2014, Majid et al. 2014) and in the assessment of neuronal activity in response to pharmacological (Lu et al. 2007, Anastasovska et al.

2012, Arora et al. 2012) and physical stimuli (Aoki et al. 2002, Chuang et al. 2009).

MEMRI has also been used to study appetite. Kuo and colleagues demonstrated an increase in signal intensity (SI, as a marker of neuronal activity) in the ARC, PVN and

VMH of fasted compared to non-fasted rodents (Kuo et al. 2006). The effects of gut hormone administration on neuronal activity have also been studied. Peripherally administered ghrelin increased SI in the ARC and VMH in fed mice, while PYY3-36 administration reduced SI in the same regions in fasted mice (Kuo et al. 2007).

Administration of the anorexigenic glucagon-like peptide-1 (GLP-1) reduced SI in the PVN and increased SI in the VMH whereas oxyntomodulin (OXM) reduced SI in the ARC and supraoptic nuclei of the hypothalamus (Parkinson et al. 2009). Neuronal activity in response to gut hormone administration has also been studied outside the hypothalamus with both GLP-1 and OXM significantly increasing SI in the area postrema of fasted mice, reflecting an increase in neuronal activity within the brainstem (Parkinson et al. 2009).

Other MRI-based techniques for assessing neuronal activity such as blood oxygenation level-dependent (BOLD) fMRI assess haemodynamic changes as an indirect marker of neuronal activity. MEMRI therefore has a significant advantage over BOLD fMRI as it measures neuronal activity directly rather than relying on the complex and poorly understood relationship between neuronal activity and haemodynamic changes (Silva et al. 2007). There do however exist a number of non-MRI based techniques for assessing neuronal activity as below.

2.1.4 Comparison with other techniques for assessing neuronal activity

c-Fos techniques

Studies examining the expression of immediate-early gene products, most notably c-Fos, have been the mainstay in attempts to determine neuronal activity following pharmacological stimuli. However the association between c-Fos expression and neuronal activity is indirect. This is evidenced by increased c-Fos expression in hypothalamic magnocellular neurones following transynaptic but not antidromic activation indicating that c-Fos is coupled to signal transduction pathways rather than neuronal depolarisation per se (Luckman et al. 1994). Therefore in neurones in which activation is not coupled to signal transduction pathways that promote c-Fos, it might be expected that the measurement of c-Fos expression would not accurately reflect neuronal activity (Bullitt

1990).

Another consideration is that c-Fos expression is influenced by other stimuli such as growth factors (Sigmund et al. 1990, Yamamori 1991), neuronal development (Sano et al.

1991), and cellular stress (Roca et al. 2004). In addition, repeated c-Fos measurements

65 cannot be made in the same animal so temporal information regarding neuronal activity cannot be obtained unless large number of animals are used and culled at sequential time-points.

Hence MEMRI offers several advantages over c-Fos techniques as it provides temporal data based directly on in vivo neuronal voltage-gated channel opening (and hence neuronal activity).

Electrophysiological techniques

Assessing the electrical response of a neurone to a stimulus is arguably the most direct measurement of neuronal activity. This technique can be applied in vivo and can be adapted to measure the electrical activity in a single neurone or multiple neurones within a given area (Wyneken et al. 2004). These techniques have also been applied to the study of kisspeptin neurones within the ARC and AVPV, and GnRH neurones in acute brain slices (Piet et al. 2014). For example, NMDA, leptin and neurokinin B are potent activators of kisspeptin neurone electrical activity while GABA inhibits kisspeptin neurones (Piet et al. 2014). In addition, the firing pattern of kisspeptin neurones located in the AVPV fluctuates with the oestrous cycle and is modulated by both glutamate and GABA (Alreja

2013). To date, there are no electrophysiological studies of extra-hypothalamic neuronal activity in response to kisspeptin.

Electrophysiological techniques, however, have some important limitations. In vivo experiments require implantation of electrodes which is invasive and can disrupt the structures being interrogated. Furthermore, the brain regions need to be pre-determined with little scope for expanding the region of interest subsequently (Wyneken et al. 2004).

By comparison MEMRI is a relatively non-invasive technique and allows for the assessment of neuronal activity within larger CNS areas without physical disruption of the area in question.

Positron emission tomography (PET)

Neuroimaging modalities such as positron emission tomography (PET) and single photon emission computed tomography (SPECT) have previously been used to investigate the various cognitive and homeostatic mechanisms that influence feeding (Anthony et al.

2006, Wang et al. 2006). However, these techniques are expensive and suffer from low signal/noise detection efficiency. Furthermore the spatial resolution of PET and SPECT is several millimetres compared to that of MEMRI which is less than 1mm (depending on equipment), and so can only investigate neuronal activity in gross areas of a rodent brain.

In addition, PET and SPECT require exposure to radiation unlike MEMRI.

2.2 Aims and Hypothesis

The widespread expression of KISS1 and KISS1R in the rodent and human brain suggests roles and effects throughout the brain and not just within the established hypothalamic reproductive areas of the ARC, AVPV and POA (Muir et al. 2001, Clarkson et al. 2009, Herbison et al. 2010). As mentioned in Section 2.1, one area of particular interest is the amygdala based on its established role in reproductive behaviours, its connections with hypothalamic nuclei involved in reproduction and its expression of KISS1 and KISS1R genes (Canteras et al. 1995, Murray 2007, Keshavarzi et al. 2014).

Aims

1. To assess time-course of effects of peripheral kisspeptin administration on

circulating kisspeptin and luteinising hormone (LH) levels.

2. To optimise the MEMRI protocol (based on aim 1) to assess neuronal activity in

response to peripheral kisspeptin administration.

3. To assess neuronal activity in the amygdala as well as established areas of the

reproductive axis (ARC and AVPV which are the predominant sites of kisspeptin

neurones, POA which houses the majority of GnRH neurones and the Anterior

Pituitary which secretes the gonadotrophins).

Hypothesis

Peripheral kisspeptin administration alters extra-hypothalamic neuronal activity.

2.3 Methods

2.3.1 Animal preparation

All animal studies were performed in accordance with the United Kingdom Animals

(Scientific Procedures) Act of 1986 and were subject to local ethical review (Project

Licence No. 70/7236). C57/BL6 male mice (8-10 weeks of age; Harlan, Bicester, UK) were maintained at 21–23°C under a 12 hour light/dark cycle (lights on at 0700) and were allowed ad libitum access to standard chow (RM3; Special Diets Services, Essex, UK) and drinking water. Adult male mice were used so as to minimise variation in gonadotrophin response to kisspeptin which is seen pre-pubertally and in females at different oestrous phases (Roa et al. 2006). Body weights of the mice in the different experimental groups, measured on the morning of the scan, were not significantly different

(p=0.32).

2.3.2 Kisspeptin-54 peptide synthesis

Kisspeptin-54 was synthesised by the Advanced Biotechnology Centre, Imperial College

London and purified by reverse-phase high performance liquid chromatography.

Electrospray mass spectroscopy and amino acid analysis confirmed identity of the peptide

(Dhillo et al. 2005). Although kisspeptin-10, -13, -14, and -54 display similar potency in vitro, kisspeptin-54 was selected due to its higher in vivo potency than the other kisspeptin fragments (Tovar et al. 2006, Jayasena et al. 2011). Kisspeptin was dissolved in saline containing gelofusin (5% vol/vol) (B.Braun Medical Ltd, Sheffield, UK) to minimise peptide adsorption to the injection system (Kraegen et al. 1975, Dhillo et al. 2005) and was administered by intraperitoneal injection over 10 seconds.

2.3.3 Study 1: Effects of peripheral kisspeptin administration on plasma kisspeptin and serum LH levels in adult male mice

I investigated the temporal effects of kisspeptin administration on circulating kisspeptin

(Study 1A) and LH levels (Study 1B) during a 2 hour time-course to determine the optimal period for MEMRI scanning to assess the effects of kisspeptin administration. Since the blood volume obtained from repeated tail clip sampling in mice is too small for multiple hormone analysis, two complementary studies were carried out (one for measurement of plasma kisspeptin and the other for measurement of serum LH following kisspeptin injection). This study also served to confirm bioactivity of the kisspeptin-54 compound.

2.3.3.1 Study 1A: Effects of peripheral kisspeptin administration on plasma kisspeptin levels

Mice were exposed to equivalent conditions to MEMRI scanning but outside the scanner.

Mice (n=8/group) were anaesthetised with 2% isoflurane in oxygen at flow-rate of 2 L/min and maintained by 1% isoflurane–oxygen. Blood (20µl) was obtained from tail tip pre- (0 minutes) and post-bolus intraperitoneal injection (at time 20, 40, 60, 120 minutes) of either kisspeptin-54 (0.04 nmol/g), or equivalent vehicle volume (gelofusin). Kisspeptin was dissolved in saline containing gelofusin and was administered by intraperitoneal injection.

This dose of kisspeptin was chosen as it has been previously shown to robustly increase

LH secretion following intraperitoneal injection in mice (Curtis et al. 2010, Jayasena et al.

2011).

Blood was collected in lithium-heparin cuvettes and stored on ice until centrifugation at

1000g for 4 minutes. Plasma was separated and stored at -20°C until assay.

Measurement of plasma kisspeptin-immunoreactivity (kisspeptin IR) was performed using a manual radioimmunoassay. Antibody GQ2 was raised in sheep immunised with

70 synthetic human kisspeptin-54 (Bachem UK Ltd.) conjugated to bovine serum albumin

(BSA) by glutaraldehyde and used at a final dilution of 1:3,500,000. The antibody cross- reacted 100% with human kisspeptin-54, kisspeptin-14, and kisspeptin-10 and less than

0.01% with other related RF amide proteins, including prolactin-releasing peptide, RF amide-related peptide 1 (RFRP1, human and rat), RFRP2 (human), RFRP3 (human),

QRFP43 (human), neuropeptide FF (human), and neuropeptide AF (human)(Dhillo et al.

2005). The iodogen method was used to prepare the 125I-kisspeptin-54 label, which was subsequently purified by High Performance Liquid Chromatography (Salacinski et al.

1981). The assay was carried out in duplicate using dilutions of neat plasma in 0.7ml of

0.06M phosphate buffer (pH 7.2) containing 0.3% BSA. Incubation was for 3 days at 4oC.

Subsequently free and antibody-bound label were separated by charcoal adsorption. The limit of detectability was 2pmol/l of plasma kisspeptin with 95% confidence interval, and the intra- and inter-assay coefficients of variation were 8.3 and 10.2%, respectively. (see

Appendix 1: Principles of Radioimmunoassay).

Circulating kisspeptin levels were compared using two-way ANOVA. p<0.05 was considered statistically significant.

2.3.3.2 Study 1B: Effects of peripheral kisspeptin administration on serum LH levels

A separate but identical experiment to Study 1A was performed to determine the time- profile of LH levels following kisspeptin administration (n=8/group). Blood was collected into plain cuvettes and stored on ice until centrifugation at 1000g for 10 minutes. Serum was separated and stored at -20oC until assay. Serum LH was assayed by magnetic-bead luminex immunoassay (Milliplex Mouse LH Magnetic-Bead Assay, Merck Millipore, UK).

Because of the small samples of blood obtained from the mice during the study, only single measurements were performed using 10µl of serum per individual time-point for each mouse. The LH was analysed using xMap technology (Millipore) with the mouse

71 pituitary magnetic bead panel. A standard curve was generated using six-fold serial dilutions of the LH hormone standard provided by the company. Standards and samples were incubated together with the antibody-coated beads on a microplate shaker overnight at 40C and washed manually with use of a hand-held magnet to keep the beads in the wells. Detection antibody was then added to the wells and incubated on a microplate shaker for 30 minutes at room temperature. The plate was then washed three times and sheath fluid added to each well. Beads were re-suspended on the microplate shaker for 5 minutes. Plates were then read on a Luminex 200IS system. Data were analysed by the supplied software. Inter- and intra-assay variation of the kit was <20% and <15% respectively. Circulating LH levels were compared using two-way ANOVA. p<0.05 was considered statistically significant.

2.3.4 Study 2: Effects of peripheral kisspeptin-54 administration on CNS neuronal activity in adult male mice using MEMRI

Having established a time-course for the effects of kisspeptin administration on circulating kisspeptin and LH, I proceeded to examine central neuronal activity during this time- course in pre-determined 3-dimensional regions of interest (ROIs).

2.3.4.1 Protocol

Mice were acclimatised in a holding room adjacent to the MRI-laboratory for 24 hours pre- scanning. Mice were anaesthetised with 2% isoflurane in oxygen at a flow-rate of 2 L/min and maintained by 1% isoflurane-oxygen via a facemask for the duration of scanning (as in Study 1). The rectal temperature of each mouse was monitored and maintained at 36 ±

0.50C by a heating system (SA Instruments, Stony Brook, New York, USA) throughout the scanning period. MEMRI was performed by a 9.4T MRI-scanner (Agilent Technologies)

72 with quadrature birdcage coil (Magnetic Resonance Labs) (Kuo et al. 2007). Transverse slices covering whole brain, were acquired repeatedly in an array, 66 times (127min), using a 2-dimensional Fast Spin Echo multi-slice sequence with parameters; Repetition

Time (TR) 1.8s, effective Echo Time (TE) (TEeff) 5.6ms (6 echoes, spacing 5.6ms, k- space center=1), Field of View (FOV) 25×25mm, matrix 192 × 192, 46 contiguous axial,

0.4mm thick slices and 2 averages. After the third acquisition, 100mM MnCl2 was infused by intravenous cannula (tail vein), at a rate of 0.2ml/h and total dose of 0.5μmol/g

(Chaudhri et al. 2006, Kuo et al. 2006, Kuo et al. 2007). Based on the results of Study 1, an effect of kisspeptin on circulating kisspeptin and LH levels is readily observed at 20 minutes post-kisspeptin ip administration. Manganese uptake reaches steady state within

20-60 minutes depending on brain region (Kuo et al. 2007), therefore kisspeptin-54

(0.04nmol/g, n=8) or equivalent volume vehicle (n=7) ip injection was administered simultaneously with initiation of MnCl2 infusion to ensure kisspeptin-induced changes in manganese uptake were detectable within the scanning period.

2.3.4.2 Regions of Interest (ROIs) and Image Analysis

Image-processing software (FSL, FMRIB, UK) was used to define ROIs on a standardized mouse brain template (a composite of many control C57BL/6 mouse brains). 3- dimensional ROIs were neuroanatomically defined and corresponded to; left amygdala, right amygdala, left ARC, right ARC, AVPV, POA, and anterior pituitary with reference to a standard mouse brain atlas (Paxinos and Franklin 2004)(see Figure 2-1). To ensure adequate manganese entry into the circulation only scans where SI in the fourth ventricle increased >20% over baseline within 10 acquisitions were included. 4D brain images were extracted and spatially normalized (SPM5, FIL Methods Group; AFNI, Medical College of

Wisconsin and FSL, UK), before SI time-course measurements were performed in the above ROIs.

2.3.4.3 Statistical Analysis

Data were normally distributed (assessed by Kolmogorov-Smirnov test). Percentage change SI from baseline (pre-manganese) to post-manganese infusion was calculated for each ROI (Percentage enhancement, PE). PE was analyzed between kisspeptin and vehicle-treated groups by comparing entire time-course SI enhancement curve by general estimated equation (GEE) analysis as previously published (Chaudhri et al. 2006, Kuo et al. 2007, Anastasovska et al. 2012). GEE performs general linear regression analysis using all values in each animal and therefore accounts for within-animal correlation between adjacent values (Zeger and Liang 1986). GEE analysis permitted assessment of significance of effect between kisspeptin and vehicle-treated groups. Area under curve analysis (AUC) was used to assess quantitative size of effect. p<0.05 was considered statistically significant.

Amygdala ARC

R AMYG L AMYG R ARC L ARC

AVPV POA

Anterior Pituitary

Figure 2-1: Representative cross-sectional images through 3-dimensional ROIs in adult mouse brain from which signal intensity (SI) profiles were generated corresponding to the amygdala, ARC, AVPV, POA (including medial septum), and anterior pituitary (AP).

2.4 Results

2.4.1 Study 1: Peripheral kisspeptin administration increases circulating kisspeptin and LH levels

2.4.1.1 Study 1A: Peripheral kisspeptin administration increases circulating kisspeptin levels

Vehicle administration had no effect on plasma kisspeptin IR. Intraperitoneal administration of 0.04 nmol/g kisspeptin-54 under general anaesthesia resulted in dramatic increases in plasma kisspeptin IR compared to vehicle between 20-120 minutes post-administration (p<0.0001, Figure 2-2A).

2.4.1.2 Study 1B: Peripheral kisspeptin administration increases circulating LH levels

Vehicle administration alone had no effect on serum LH. However, kisspeptin administration increased LH compared to vehicle administration between 20-120 minutes post-administration (p<0.0001) (Figure 2-2B). The peak LH increase occurred at 120 minutes post-kisspeptin (vehicle +0.3±0.2ng/ml, kisspeptin +3.7±0.5ng/ml, p<0.0001 vs. vehicle). The timings of the observed increases in LH (20-120 minutes) were similar to those at which increased plasma kisspeptin IR was observed after kisspeptin administration (Figure 2-2A).

10000

1000

Kisspeptin *** 100 Vehicle

(pmol/L) 10

Plasma Kisspeptin-IR Plasma 1 0 20 40 60 80 100 120 Time from Injection (min)

4.5 4.0 3.5 3.0

*** 2.5 2.0

1.5 1.0 LH (ng/mL) Serum 0.5 0.0 0 20 40 60 80 100 120 Time from Injection (min)

Figure 2-2: Effect of peripheral kisspeptin or vehicle administration on circulating kisspeptin and LH levels. Time-course of increases in (A) plasma kisspeptin IR and (B) serum luteinising hormone (LH) after intraperitoneal injection of kisspeptin-54 (0.04 nmol/g) at time 0 minutes in adult mice. n=8/group. *** p<0.0001. Arrow represents bolus injection of kisspeptin/vehicle. Data presented as mean ± SEM.

2.4.2 Study 2: Peripheral kisspeptin administration modulates CNS neuronal activity as determined by MEMRI

Having demonstrated increased LH levels after kisspeptin administration, I proceeded to assess the effects of kisspeptin on central neuronal activity using MEMRI. I assessed the relative signal intensity (SI) as a marker of neuronal activation in selected, neuroanatomically defined 3-dimensional regions of interest over a 120 minute time- course (to encompass the hormonal changes observed in Study 1).

2.4.2.1 Amygdala

A markedly reduced SI was observed in the amygdala after kisspeptin administration compared to vehicle alone (GEE analysis: vehicle left amygdala vs. kisspeptin left amygdala p=0.0257; vehicle right amygdala vs. kisspeptin right amygdala p=0.0085,

Figure 2-3A and B). The control and kisspeptin lines diverged from approximately 20 minutes post-injection onwards.

The mean SI decreased by 20% in the left amygdala and 22% in the right amygdala

(mean SI AUC in PE.min, vehicle left amygdala 885.1±82.2, kisspeptin left amygdala

704.1±40.3; vehicle right amygdala 983.4±76.6, kisspeptin right amygdala 762.3±51.9).

This reduction of SI in the amygdala (Figure 2-3A and B) occurred at a similar time to increases in circulating kisspeptin (Figure 2-2A) and LH (Figure 2-2B).

2.4.2.2 ARC and AVPV

No significant differences in SI between kisspeptin and vehicle administration were observed in either the ARC (left and right, Figure 2-3C and D) or AVPV (Figure 2-3E) for the duration of the study.

2.4.2.3 POA

A reduced SI was observed in the POA after kisspeptin administration compared to vehicle (p=0.0234, Figure 2-3F).

2.4.2.4 Anterior Pituitary

Although there was a trend towards an initial increase in SI in the first 20 minutes after kisspeptin injection, no significant differences were observed over the full time-course of the experiment (Figure 2-3G).

A: Left Amygdala B: Right Amygdala 30 30 **

* 20 20

10 10 Vehicle Kisspeptin

Signal Intensity (PE) Intensity Signal 0 Signal Intensity (PE) Intensity Signal 0 MnCl 2 MnCl 2 0 20 40 60 80 100 120 0 20 40 60 80 100 120 Time (min) Time (min) C: Left ARC D: Right ARC 50 50

40 40 30 30 20 20

10 10

Signal Intensity (PE) Intensity Signal

Signal Intensity (PE) Intensity Signal 0 MnCl 2 0 MnCl 2 0 20 40 60 80 100 120 0 20 40 60 80 100 120 Time (min) Time (min) E: AVPV F: POA 40 40 * 30

20 20 10

Signal Intensity (PE) Intensity Signal Signal Intensity (PE) Intensity Signal 0 0 MnCl 2 MnCl 2

0 20 40 60 80 100 120 0 20 40 60 80 100 120 Time (min) Time (min) G: Anterior Pituitary 20

Signal Intensity (PE) Intensity Signal MnCl 2

0 20 40 60 80 100 120 Time (min) Figure 2-3: Changes in CNS neuronal activity following kisspeptin or saline administration. Time-course of T1-weighted MRI signal change in the left amygdala (A), right amygdala (B), left ARC (C), right ARC (D), AVPV (E), POA (F) and anterior pituitary

(G) after intravenous MnCl2 infusion into adult mice also receiving an intraperitoneal injection of kisspeptin (n=8), or vehicle (n=7). Arrows indicate start of MnCl2 infusion and bolus injection of kisspeptin or vehicle. Grey bar represents duration of MnCl2 infusion. SI was measured as PE over baseline. **p<0.01, *p<0.05. Data presented as mean ± SEM.

2.5 Discussion

Although our understanding of the KISS1/KISS1R system has improved dramatically in the past decade, the vast majority of work has focussed on the hypothalamus. The effects of kisspeptin in other areas of the brain have been less well-characterised. Furthermore, as work continues to develop the KISS1 gene product, kisspeptin, into a potential therapeutic agent (Jayasena et al. 2010, Sonigo et al. 2012, George et al. 2013) it is of paramount importance that we better understand the effects of in vivo exogenous kisspeptin administration on other neuronal populations outside the hypothalamus. These effects may provide physiological insights as well as having important therapeutic implications. To this end, MEMRI offers a unique opportunity to examine neuronal activation in living tissue with several advantages over more traditional c-Fos techniques

(where results may be comparable (Hsu et al. 2007, Lehallier et al. 2012)) as detailed earlier as well as over functional MRI techniques that rely purely on haemodynamic variations (Lin and Koretsky 1997). In addition, extensive work has focussed on robustly confirming the validity of MEMRI including non-specific peptide effects on neuronal Mn2+ accumulation as well as discounting any potential effect of Mn2+ administration itself on neuronal activation as determined by c-Fos (Kuo et al. 2007). Hence in this study I was able to apply MEMRI to study the effects of exogenous kisspeptin administration on Mn2+ uptake as a surrogate for neuronal activation in pre-selected ROIs.

Peripheral kisspeptin administration decreases neuronal activity in the amygdala.

Current evidence suggests that the amygdala exerts a predominantly inhibitory effect on the reproductive axis and on reproductive behaviour. For example, electrical stimulation of the amygdala delays puberty (Bar-Sela and Critchlow 1966). Conversely, lesioning of the amygdala results in hypersexuality (Kluver and Bucy 1997), increased circulating LH

(Lawton and Sawyer 1970), and prevents the stress-induced suppression of LH pulses

(Lin et al. 2011). Here, I report for the first time that, in adult male mice, peripherally administered kisspeptin inhibits neuronal activation in the amygdala and is temporally associated with an increase in circulating LH levels. These data therefore suggest that kisspeptin administration may release the tonic “inhibitory brake” exerted by the amygdala on the reproductive axis to stimulate gonadotropin secretion.

Recent work by (Kim et al. 2011) has brought to prominence the importance of the amygdala and its involvement with the Kiss1/Kiss1r system. In adult mice Kiss1 expression has been observed in the amygdala (Clarkson et al. 2009). This amygdala

Kiss1 expression is greater in males compared to females (where expression also varies with oestrous cycle) (Kim et al., 2011). By contrast, Kiss1r expression has not yet been demonstrated in the amygdala of adult mice (Herbison et al. 2010) but has been observed in the amygdalae of rats and humans (Lee et al. 1999, Muir et al. 2001). It is remains unclear if this species difference in Kiss1r amygdala expression represents true inter- species variation or a methodological difference (LacZ-Xgal assay in a transgenic Kiss1r knock-in mouse model in mice (Herbison et al. 2010) versus northern blot/in-situ hybridisation in rats (Lee et al. 1999)).

Although Kiss1/Kiss1r system may be present to some degree in the amygdala of mice, peripherally administered kisspeptins are not thought to cross the blood-brain-barrier

(Herde et al. 2011). However, it has previously been shown that peripheral kisspeptin administration can alter the neuronal activity of central neurones (Scott and Brown 2011).

It is also interesting that we observed an overall inhibitory effect of peripherally administered kisspeptin on neuronal activity in the amygdala in association with an increase in LH release. Kisspeptin can activate γ-aminobutyric acidergic (GABA) neurones to then inhibit downstream neuronal activity (Fu and van den Pol 2010,

Pielecka-Fortuna and Moenter 2010). In addition, the amygdala has a well-established network of GABA inter-neurones as well as GABA and non-GABA projections to

83 reproductive centres of the hypothalamus (Canteras et al. 1995, Hahn et al. 2003,

Keshavarzi et al. 2014). Finally, it is important to consider that MEMRI cannot distinguish directly between activation of one population of neurons and deactivation of another within the same ROI, and hence represents an overall net affect. Therefore, in the current study the net neuronal deactivation observed in the amygdala following kisspeptin administration may represent the predominant inhibition (via GABA interneurones) of

GABA/non-GABA neurones and projections to reproductive centres of the hypothalamus.

This in turn would relieve GABA inhibitory tone on the GnRH neurones, and therefore stimulate reproductive hormone release (although the precise effects of GABA signalling on GnRH neurones remain controversial as reviewed in (Herbison and Moenter 2011)).

Indeed, a similar mechanism has been reported whereby kisspeptin indirectly inhibits neuropeptide Y neurons via GABA neurones (Fu and van den Pol 2010). Further non- imaging studies are required to clarify this.

Peripheral kisspeptin administration modulates neuronal activity in other brain areas.

MEMRI can be used to monitor changes in hypothalamic neuronal activity in mice. For example, MEMRI was used to demonstrate changes in neuronal activity in specific hypothalamic nuclei following gut hormone administration (Kuo et al. 2007). The POA contains the vast majority of GnRH neurones and between 60-90% of these express

KISS1R in rodents (Irwig et al. 2004). Since kisspeptin stimulates gonadotropin release through activation of GnRH neurones, we anticipated that exogenous kisspeptin administration might result in increased neuronal activation in the POA. In fact, neuronal activity in the hypothalamic POA decreased after kisspeptin administration. Previous studies have shown that, although central administration of kisspeptin increases c-Fos expression in GnRH neurones (Irwig et al. 2004), no change in GnRH neuronal c-Fos

84 expression has been observed after peripheral kisspeptin administration in adult mice

(Sanchez-Carbayo et al. 2003). This is in keeping with both our current study and previous studies suggesting that peripherally administered kisspeptin may predominantly act on GnRH neurone dendritic terminals outside the blood-brain-barrier which therefore do not undergo cell body activation (Herde et al. 2011, Xu et al. 2012). In addition, GnRH neurones represent only a fraction of the neurones within the POA that can be modulated by kisspeptin. For example, GABA neuronal activity within the POA is decreased by kisspeptin in rodents. And this inhibition of GABA neuronal activity in the POA contributes to the increased gonadotropin release (Neal-Perry et al. 2009, Zhang et al. 2009). This is in keeping with the finding from the current study of overall decreased neuronal activity within the POA after peripheral kisspeptin administration.

I also examined manganese uptake in other hypothalamic regions with well-established reproductive functions. Neuronal activity in the ARC and AVPV was unaffected by peripheral kisspeptin administration compared to vehicle. This is consistent with the absence of expression of KISS1R in these areas in mice as well as the inability of peripheral kisspeptin to cross the blood-brain-barrier to affect these areas directly

(Herbison et al. 2010). However, modulation of KISS1 neuronal activity in the ARC and

AVPV is predominantly by testosterone feedback in male rodents and so a change in neuronal activity may occur in response to changing testosterone levels (negative on

ARC, positive on AVPV) (Smith et al. 2005). Although circulating testosterone levels are increased within 60 minutes after peripheral kisspeptin administration (Mikkelsen et al.

2009) several days of testosterone exposure are required to achieve feedback effects on the ARC and AVPV in male mice (Smith et al. 2005) consistent with the lack of effect observed in the current study.

Finally, I evaluated the influence of kisspeptin on manganese uptake within the anterior pituitary, the site of LH release from the gonadotrophs in response to GnRH. Although there was an initial non-significant increase over the first 20 minutes, this did not continue

85 through the full time-course. This could be due to rapid habituation of the gonadotroph cell population by the pharmacological concentrations of kisspeptin administered which are known to cause sustained GnRH release which stimulates the gonadotrophs (Han et al.

2005). However, given that gonadotrophs represent a fraction of the cell population of the anterior pituitary, it also is possible that any effect on gonadotroph activity was masked by other, non-gonadotroph activity.

Limitations

One of the main limitations in this study (common to all neuroimaging techniques) is the inability for distinction between different neuronal populations within an anatomically defined region of interest. Thus neuroimaging measures the net neuronal activity within that region. As mentioned in the discussion above there may hence be increased activity in one population and decreased activity in another co-localised population; however the data will only reflect the net change in activity. This may explain for instance why although

GABA neurones may be activated by kisspeptin, the net decrease in activity may reflect the downstream effects of these inhibitory GABA neurones on another larger neuronal population.

Another interesting consideration is that an increase in signal intensity may represent an increase in activity of either stimulatory or inhibitory neurones and vice versa. Hence other non-imaging experimental techniques are required to determine the precise identities of the neuronal sub-types.

The use of anaesthesia typically suppresses neuronal activity and hence would make any changes in response to an external stimulus harder to detect. However, a degree of anaesthesia is required to keep the animal stationary and minimise any effects of stress on the reproductive axis (Chand and Lovejoy 2011). The depth of anaesthesia correlates

86 with suppression of MEMRI signal intensity (Silva and Bock 2008). Hence the dose used in this study was based on previous experience and dose-titration experiments to ascertain the lowest dose at which reliable immobility is achieved (Chaudhri et al. 2006,

Kuo et al. 2006, Kuo et al. 2007, Parkinson et al. 2009). This should not however limit the results of the current study significantly as both vehicle and kisspeptin treated mice received the same anaesthesia protocol.

Summary

These findings open up future directions in which roles of kisspeptin outside the hypothalamus may be studied further using neuroimaging techniques. Given the observation of changes in amygdala neuronal activity in response to kisspeptin and a direct association with LH release, future work may focus on the KISS1/KISS1R system and sexual behaviours attributed to the amygdala. Along these lines, Kauffman and colleagues demonstrated that Kiss1r expression is essential for olfactory-mediated partner preference behaviour but further reproductive behavioural studies are now warranted

(Kauffman et al. 2007). Other behaviours such as feeding, grooming, locomotion and sleeping have been shown to be unaffected by central kisspeptin administration

(Thompson et al. 2004)

This is the first study to use neuroimaging techniques to study the effects of kisspeptin administration. In summary, I demonstrate using MEMRI that peripheral kisspeptin administration decreases neuronal activity within the amygdala and increases LH secretion in adult male mice. This study also demonstrated decreases in neuronal activity in the POA with no effect observe in the ARC, AVPV and anterior pituitary. These findings highlight potentially interesting extra-hypothalamic effects of kisspeptin on neuronal activity in the amygdala. MEMRI technique has previously been validated in the study of gut hormone effects (Kuo et al. 2007) and here I demonstrate the use in the study of

87 kisspeptin signalling. This raises the possibility that MEMRI can be used to identify neuronal effects of kisspeptin administration in other brain areas. This is of important physiological and pharmacological importance as kisspeptin emerges as a possible therapeutic agent for reproductive pathologies.

2.6 Future Work

To confirm the findings from this study, similar experiments could be performed using alternative techniques for the measurement of neuronal activity such as c-Fos measurement in the amygdala after kisspeptin administration. In addition, it would be interesting to repeat a similar protocol as in this study but using a kisspeptin antagonist

(such as peptide-234) instead of kisspeptin; one could hypothesise that the opposite effects on neuronal activity may be observed.

Further work is needed to determine the physiological importance of these novel actions of kisspeptin. For example, one possible future study may examine the effects of direct kisspeptin and kisspeptin antagonist administration into the amygdala (via indwelling catheter) with measurement of gonadotrophin responses. Another avenue to explore is the precise identities of the neuronal subtypes involved- to what extent is GABA involved?

Immunohistochemical studies are required to address this and clarify the precise pathways involved.

In addition, given the extensive role for the amygdala in social and reproductive behaviours, it may be fruitful to perform extensive behavioural studies following central and peripheral kisspeptin. In addition closer examination of behaviours of Kiss1 and

Kiss1r null mice may reveal subtle roles for these genes in various behaviours.

Another important future study would be to repeat this protocol in female mice to see if there is any sexual dimorphism. Given that Kiss1 expression within the amygdala is lower

88 in female compared with male rodents and peaks during proestrous, it would be important to perform the studies in the same oestrous phase (Kauffman et al. 2007).

From the findings of this study and the widespread identification of Kiss1 and Kiss1r in the

CNS, future study of other extra-hypothalamic CNS areas using MEMRI may be interesting. These may include the dentate gyrus, supramammillary nucleus, dorsal cochlear nucleus which highly express Kiss1r (Herbison et al. 2010).

Chapter 3 ______

The effects of kisspeptin-10 in healthy men and women

3.1 Introduction

The hormone kisspeptin plays a key role in human reproductive physiology and is a potential therapy for reproductive disorders as detailed in the introduction. In the previous chapter, I investigated the effects of kisspeptin-54 on neuronal activity in rodents.

However, kisspeptin therapeutics may be based on the various human kisspeptin isoforms

(Kisspeptin-54, -14, -13, -10). Kisspeptin-54, is the longest form and has been studied in healthy men (Dhillo et al. 2005), healthy women (Dhillo et al. 2007) and women with hypothalamic amenorrhoea (Jayasena et al. 2009, Jayasena et al. 2010, Jayasena et al.

2014) and subfertility (Jayasena et al. 2014). However, the shorter peptide sequence of kisspeptin-10 makes it easier and therefore cheaper to synthesise. In addition several kisspeptin analogues have been developed based on its decapeptide sequence (Millar and Newton 2013).

Kisspeptin-10 administration stimulates robust gonadotrophin release in both male and female rodents (Irwig et al. 2004, Thompson et al. 2004, Roa et al. 2006). In humans, kisspeptin-10 had been administered to healthy men but not women at the time of performing these studies. Due to the cyclical changes of the female menstrual cycle, the first studies of both kisspeptin-54 and kisspeptin-10 were performed in men as they lack cyclical changes in reproductive axis function (Dhillo et al. 2005, Chan et al. 2011).

In men an intravenous bolus kisspeptin-10 at a dose as low as 0.008nmol/kg stimulates

LH release with a maximal response observed between 0.23 and 0.76nmol/kg (George et al. 2011). In addition, kisspeptin-10 has been observed to enhance LH pulsatility. A single intravenous bolus dose of kisspeptin-10 more than doubles LH pulse amplitude

(compared to endogenous pulses) (Chan et al. 2011). Furthermore a single intravenous kisspeptin-10 bolus induces an immediate LH pulse therefore resetting the GnRH pulse generator (with a delay to the next pulse approximating the normal interpulse interval)

(Chan et al. 2011).

Examining a different duration of administration, continuous infusions of kisspeptin-10 for

22.5 hours have also been shown to increase LH several-fold with accompanying increases in LH pulse frequency and amplitude (George et al. 2011). Furthermore continuous intravenous infusion of kisspeptin resulted in greater increases in circulating testosterone compared to bolus intravenous administration (George et al. 2011).

Similar to the tachyphylaxis first seen with kisspeptin-54 administration in women with hypothalamic amenorrhoea (Jayasena et al. 2009, Jayasena et al. 2010), higher intravenous bolus doses of kisspeptin-10 (3µg/kg) also result in a reduced LH secretion compared to lower doses (1µg/kg)(George et al. 2011). Hence tailoring of doses, route of administration and isoform of kisspeptin are all of paramount importance for the development of kisspeptin therapeutics.

In this chapter, I examine for the first time the effects of kisspeptin-10 in healthy women and compare this directly with those observed in healthy men. Based on the study demonstrating different potencies of kisspeptin-54 on LH secretion in healthy women depending on their menstrual phase (pre-ovulatory>luteal>follicular) (Dhillo et al. 2007), I will address the same question with kispeptin-10. In addition I will determine the plasma half-life for kisspeptin-10 which is of key importance therapeutically and will likely be shorter than that of kisspeptin-54 which is 27.6 ± 11 minutes (Dhillo et al. 2005) due to its shorter sequence.

3.2 Aim and Hypothesis

Aims

1. To investigate the effects of kisspeptin-10 administration on reproductive hormone secretion (LH, FSH and E2) in healthy men.

2. To investigate the effects of kisspeptin-10 administration on reproductive hormone secretion (LH, FSH and E2) in healthy women in different phases of the menstrual cycle.

3. To determine the plasma half-life of kisspeptin-10 in healthy men and women.

Hypothesis

1. Kisspeptin-10 stimulates reproductive hormone secretion in healthy men and women.

2. Kisspeptin-10 stimulates reproductive hormone secretion most potently during the pre- ovulatory phase.

3. Kisspeptin-10 has a shorter plasma half-life than kisspeptin-54.

3.3 Methods

3.3.1 Subjects

The study was conducted with Local Ethics Committee approval (reference 08/H0707/95) in accordance with The Declaration of Helsinki. Recruitment was via advertisements in the local press. Full written informed consent was obtained from all subjects. Fifteen healthy female volunteers and eleven male volunteers were recruited after medical screening, using the criteria below (Table 3-1).

3.3.2 Study days

Subjects were admitted to the Clinical Research Unit and asked to lay supine for the duration of each study. Urine was tested to exclude pregnancy in women (Clearview

Easy-HCG; Inverness Medical Innovations Inc., Waltham, MA). All blood samples were analysed for measurement of serum LH, FSH, Oestradiol (in women) or Testosterone (in men), and plasma kisspeptin immunoreactivity (IR). Blood samples for serum analysis were collected in plain serum Vacutainer tubes (Beckton Dickson, Franklin Lakes, NJ,

USA). Samples were allowed to clot prior to centrifugation at room temperature for 10 minutes at 3000rpm (Hettich EBA 20 machine, Hettich International, Tuttlingen, Germany) and separation of serum. Blood samples for plasma kisspeptin analysis were collected in lithium heparin tubes (Beckton Dickson, Franklin Lakes, NJ, USA) containing 5000 kallikrein inhibitor units of aprotinin (0.2ml Trasylol; Bayer, Newbury, UK). Samples were immediately centrifuged at room temperature for 4 minutes at 4000rpm, and then separated. Serum and plasma samples were stored at -20oC until analysis. Heart rate, blood pressure, and the presence of adverse symptoms were recorded at regular intervals

(10-15 minutes).

3.3.3 LH, FSH, Oestradiol and Testosterone measurement

LH, FSH, Oestradiol and total Testosterone were measured using automated chemiluminescent immunoassays (Abbott Laboratories, Abbott Park, IL). Reference ranges for males were as follows: LH, 4–14 iU/l; FSH, 1.5–8 iU/l; testosterone, 10–28 nmol/l. Reference ranges for females were as follows: LH in iU/l, 2–10 follicular, 20–60 midcycle, 4–14 luteal; FSH in iU/l, 10-50 mid-cycle, 1.5–8 follicular and luteal; oestradiol in pmol/l, <300 early follicular, 400-1500 midcycle, 200-1000 luteal. The respective intra- and interassay coefficients of variation for each assay were: 4.1 and 2.7% (LH); 4.1 and 3.0%

(FSH); 3.3% and 3.0% (oestradiol); 4.2 and 2.8% (total testosterone). Analytical sensitivities were: 0.5 iU/l (LH); 0.05 iU/l (FSH); 37 pmol/l (oestradiol); 2 nmol/l (total testosterone).

Table 3-1: Healthy recruitment inclusion criteria. Recruitment involved a clinical history (including menstrual history for females), examination (including blood pressure, heart rate), electrocardiogram, and blood tests (full blood count, renal profile, liver and bone profile, thyroid profile, random glucose, prolactin, gonadotrophins and sex hormone profile as appropriate; males- testosterone, sex hormone binding globulin; females- oestradiol, progesterone, androstenedione, dehydroepiandrosterone, extracted testosterone, sex hormone binding globulin).

Male and female subjects

age 18 to 40 years

no clinical or biochemical evidence of hypogonadism

normal thyroid function

normal serum prolactin

no therapeutic or recreational drug use

no systemic disease co-morbidity

Female subjects

regular menstrual cycles

no oral contraceptive pill therapy within the last year

no clinical or biochemical evidence of polycystic ovarian syndrome

3.3.4 Kisspeptin radioimmunoassay

Plasma kisspeptin immunoreactivity (IR) was measured using a manual radioimmunoassay as described previously in Section 2.3.3.1.

3.3.5 Kisspeptin-10 and -54 peptide synthesis

Human sequence kisspeptin-54 peptide was synthesised by Advanced Biotechnology

Centre, Imperial College London. Human sequence kisspeptin-10 was synthesised by

Bachem Holding AG (Bubendorf, Switzerland). Both peptides were purified by reverse- phase High Performance Liquid Chromatography (HPLC). Electrospray mass spectroscopy and amino acid analysis confirmed identity of the peptide. Toxicology testing in animals was conducted prior to administration to human volunteers. The limulus amoebocyte lysate test (LAL) detected no endotoxin (Associates of Cape Cod, Liverpool,

UK), and bacterial culture was sterile (Department of Microbiology, Hammersmith

Hospital, London, UK), in samples of kisspeptin-10 and kisspeptin-54 peptide. Vials of freeze-dried kisspeptin-10 and -54 were stored at minus 20oC and reconstituted in 0.9% saline.

3.3.6 Study 1: Effects of intravenous bolus injection of saline or kisspeptin-10 in healthy male volunteers

Intravenous bolus injection of 0.9% saline or kisspeptin-10 (at doses 0.3, 1.0, 3.0, or 10 nmol/kg) was administered at time 0 minutes. For all studies the intravenous bolus was given via a cannulated antecubital vein over 10 seconds and subsequently flushed with

10ml of 0.9% saline. Blood samples were taken at -30, 0, 10, 20, 30, 40, 50, 60, 75, 90,

120, 150, and 180 minutes (n=5 per group) (Figure 3-1).

3.3.7 Study 2: Effects of IV bolus injection of saline, kisspeptin-10 or kisspeptin-54 in healthy female volunteers

A: Follicular phase of the menstrual cycle

Women between day 2-10 of their menstrual cycle were administered an intravenous bolus injection of 0.9% saline, kisspeptin-10 (at doses 1.0, 3.0, or 10nmol/kg), or kisspeptin-54 (1.0nmol/kg) at time 0 minutes, and blood samples were taken at -30, 0, 10,

20, 30, 40, 50, 60, 75, 90, 120, 150, and 180 minutes (n=4-7 per group) (Figure 3-1).

B: Pre-ovulatory phase of the menstrual cycle

Women 15-16 days before their next predicted period received intravenous bolus injection of 0.9% saline or 10nmol/kg kisspeptin-10 as described for the follicular phase study above (n=5) (Figure 3-1).

3.3.8 Study 3: Determining the half-life of kisspeptin-10 in healthy male and female volunteers

To determine the plasma half-life of kisspeptin-10 in men, and in women during the follicular and pre-ovulatory phases of the menstrual cycle, frequent blood sampling was performed during and after intravenous infusion of 360pmol/kg/min of kisspeptin-10 to ensure steady initial plasma kisspeptin levels. Detailed blood sampling was performed at

1 minute (from 91-100 minutes) and 2 minute (from 102-120 minutes) intervals immediately after stopping the kisspeptin-10 infusion (at time 90 minutes) (Figure 3-2).

Blood samples were assayed for plasma kisspeptin IR. The linear regression line gradient of natural log plasma kisspeptin IR was used to calculate the half-time of disappearance

(t1/2) for infused kisspeptin-10 in healthy males and females.

Figure 3-1: Study Protocol for Studies 1 and 2. Healthy male and female (in follicular or pre-ovulatory phase) volunteers where cannulated at time -30 minutes. An intravenous bolus of 0.9% saline, kisspeptin-10 or kisspeptin-54 was administered at time 0 minutes.

Figure 3-2: Study Protocol for Study 3. Healthy male and female (in follicular or pre- ovulatory phase) volunteers where cannulated at time -30 minutes. An intravenous infusion (360 pmol/kg/min) of kisspeptin was administered from 0 to 90 minutes.

3.3.9 Data analysis

Data are presented as mean ± SEM. All data of serum reproductive hormones during treatment are presented as increases in serum levels after injection when compared with pre-injection levels. Area Under Curves (AUC) were calculated to provide a cumulative measure of kisspeptin/reproductive hormone change throughout the timecourse. Time profiles of hormone levels were compared using two-way ANOVA with Bonferroni's multiple comparison test. Pairs of means were compared with unpaired t tests, and multiple means of AUC reproductive hormone release were compared using one-way

ANOVA with Bonferonni's multiple comparison test. Slopes of linear regression lines were compared using an F test. Half-lives were calculated using (natural log 2)/gradient of linear regression line and compared using one-way ANOVA with Bonferonni’s multiple comparison test. For all statistical tests, p<0.05 was considered statistically significant.

100

3.4 Results

Baseline characteristics of the subjects recruited after medical screening are summarised below (Table 3-2). There were no significant difference in age or body mass index (BMI) between the male and female volunteers. The female reproductive hormonal milieu (LH,

FSH, and Oestradiol) was significantly higher in the pre-ovulatory phase of the menstrual cycle as expected.

3.4.1 Study 1: Effects of IV bolus injection of saline or kisspeptin-10 in healthy male volunteers

Plasma kisspeptin IR

Plasma kisspeptin IR was elevated after intravenous bolus injection of kisspeptin-10 at all doses in healthy male volunteers (Figure 3-3A). The peak occurred for all doses at 10 minutes after injection and was statistically significant (vs. saline) for 3 and 10 nmol/kg of kisspeptin-10 (Figure 3-3A). For the latter peak kisspeptin IR was 3350 ± 725 pmol/l.

Subsequently plasma kisspeptin IR returned to undetectable levels 50 minutes after injection (Figure 3-3A).

Similarly mean kisspeptin IR Area Under Curve (AUC) was increased significantly for 3 and 10 nmol/kg doses of kisspeptin-10 (Figure 3-3B). For the latter mean AUC kisspeptin

IR was 700 ± 160 h.pmol/l, (p<0.001 vs. saline).

Serum LH

Serum LH was significantly elevated after administration of each tested dose of intravenous bolus kisspeptin-10 compared with saline from 20-120 minutes (Figure 3-4A).

Peak stimulation of serum LH was observed 30-50 minutes after injection (therefore 20-40

101 minutes after peak plasma kisspeptin IR). Thereafter serum LH levels returned to baseline

180 minutes after injection (Figure 3-4A). Maximal stimulation of LH was observed after intravenous bolus of 10 nmol/kg (the highest dose) kisspeptin-10 (mean AUC LH increase was 5.3 ± 1.2 h.iU/l, p<0.001 vs. saline), although there was significant increase for all doses (Figure 3-4B).

Serum FSH

Serum FSH was significantly increased compared with saline injection after intravenous bolus injection of 0.3 and 3.0 nmol/kg kisspeptin-10 (Figures 3-5A, B) up to 90 minutes after injection. Peak FSH stimulation occurred slightly later than for LH, between 40-150 minutes after injection depending on dose (Figure 3-5A). Maximal stimulation of FSH was observed after intravenous bolus of the smallest dose, 0.3 nmol/kg kisspeptin-10 (mean

AUC FSH increase was 1.2 ± 0.5 h.iU/l, p<0.05 vs. saline), although significant stimulation was also seen at 3 nmol/kg (Figure 3-5B).

Serum testosterone

Serum testosterone AUC was significantly increased compared with saline injection after intravenous bolus injection of 1.0 nmol/kg kisspeptin-10 (Figure 3-6B). Serum levels of testosterone at this dose steadily increased to peak levels 150-180 minutes after injection

(later than for LH and FSH) (Figure 3-6A).

102

Table 3-2: Baseline characteristics of healthy male and female volunteers recruited to the study. Female endocrine profiles are presented during the follicular and pre- ovulatory phases of the menstrual cycle. Data are shown as mean ± SEM.

Healthy male Healthy female Baseline volunteers volunteers Characteristic (n=11) (n=15)

Age (yr) 28.8 ± 2.1 31.8 ± 1.4 p=0.23

BMI (kg/m2) 24.5 ± 0.5 22.6 ± 0.8 p=0.09 Length of menstrual 28 ± 0.3 cycle (d)

LH (iU/l) 2.9 ± 0.2

Follicular 3.9 ± 0.4

Pre-ovulatory 28.0 ± 4.5a

FSH (iU/l) 2.6 ± 0.2 Follicular 3.8 ± 0.4 Pre-ovulatory 9.8 ± 0.9a

Testosterone (nmol/l) 21.6 ± 1.5

Oestradiol (pmol/l) Follicular 228 ± 53 Pre-ovulatory 726 ± 71b

a p<0.0001 vs. follicular phase of the menstrual cycle b p<0.001 vs. follicular phase of the menstrual cycle

103

 A  4000 Saline 0.3nmol/kg 3000  1nmol/kg 3nmol/kg 2000 10nmol/kg

1000

Kisspeptin IR (pmol/l) IR Kisspeptin 0 0 30 60 90 120 150 180

Time after Injection (min)

B *** 1000

800

600 * 400

200

AUC Kisspeptin IR (h.pmol/l) IR Kisspeptin AUC 0 Saline 0.3 1 3 10

Kisspeptin-10 Dose (nmol/kg)

Figure 3-3: A- Plasma kisspeptin IR after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers. For 10 nmol/kg vs. saline: δδδ, p<0.001. For 3 nmol/kg vs. saline: γγγ, p<0.001. B- Mean kisspeptin IR AUC post-intravenous injection of saline, 0.3, 1, 3, 10 nmol/kg kisspeptin-10 over the 180 min timecourse. *, p<0.05; **, p<0.01; ***, p<0.001.

104

    A  5      Saline 4 0.3nmol/kg 3 1nmol/kg 2 3nmol/kg 10nmol/kg 1 0

LH Increase (iU/l) Increase LH -1 -2 0 30 60 90 120 150 180

Time after Injection (min)

*** B

-2

AUC LH Increase (h.iU/l) LH AUC -4

Saline 0.3 1 3 10

Kisspeptin-10 Dose (nmol/kg)

Figure 3-4: A- Serum LH after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers. For 0.3 nmol/kg vs. saline: α, p<0.05; ααα, p<0.001. For 1 nmol/kg vs. saline: β, p<0.05; ββ, p<0.01; βββ, p<0.001. For 3 nmol/kg vs. saline: γγ, p<0.01; γγγ, p<0.001. For 10 nmol/kg vs. saline: δ, p< 0.05; δδ, p< 0.01, δδδ, p< 0.001. B- Mean serum LH increase AUC for saline, 0.3, 1, 3, 10 nmol/kg kisspeptin-10 over the 180 min timecourse post-intravenous bolus. *, p<0.05; **, p<0.01; ***, p<0.001.

105

     A   1.2 Saline 0.8 0.3nmol/kg 1nmol/kg 0.4 3nmol/kg 10nmol/kg 0.0

FSH Increase (iU/l) Increase FSH -0.4

-0.8 0 30 60 90 120 150 180

Time after Injection (min)

B **

2 *

AUC FSH Increase (h.iU/l) FSH AUC -1 Saline 0.3 1 3 10

Kisspeptin-10 Dose (nmol/kg)

Figure 3-5: A- Serum FSH after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers. For 0.3 nmol/kg vs. saline: α, p<0.05; αα, p<0.01; ααα, p<0.001. For 3 nmol/kg vs. saline: γ, p<0.05; γγ, p<0.01; γγγ, p<0.001. B- Mean serum FSH increase AUC for saline, 0.3, 1, 3, 10 nmol/kg kisspeptin-10 over the 180 min timecourse post-intravenous bolus. *, p<0.05; **, p<0.01; ***, p<0.001.

106

A       10 Saline 0.3nmol/kg 5 1nmol/kg 3nmol/kg 0 10nmol/kg

-5

Testosterone Increase (nmol/l) Increase Testosterone -10 0 30 60 90 120 150 180

Time after Injection (min)

B *

-10

-20 Saline 0.3 1 3 10

AUC Testosterone Increase (h.nmol/l) Increase Testosterone AUC Kisspeptin-10 Dose (nmol/kg)

Figure 3-6: A- Serum Testosterone after intravenous bolus injection of saline or 0.3, 1, 3, 10 nmol/kg kisspeptin-10 to healthy male volunteers. For 0.3 nmol/kg vs. saline: α, p<0.05. For 1 nmol/kg vs. saline: β, p<0.05; ββ, p<0.01. For 10 nmol/kg vs. saline: δ, p<0.05. B- Mean serum FSH increase AUC for saline, 0.3, 1, 3, 10 nmol/kg kisspeptin-10 over the 180 minute timecourse post-intravenous bolus. *, p<0.05; **, p<0.01; ***, p<0.001.

107

3.4.2 Study 2: Effects of IV bolus injection of saline, kisspeptin-10 or kisspeptin-54 in healthy female volunteers

3.4.2.1: Follicular phase of the menstrual cycle

The plasma kisspeptin IR was elevated after intravenous bolus injection of kisspeptin-10 at all doses in healthy female volunteers during the follicular phase of the menstrual cycle

(Figure 3-7). The highest plasma kisspeptin IR was observed after intravenous bolus injection of 10 nmol/kg kisspeptin-10 (mean AUC kisspeptin in healthy females during the follicular phase was 527 ± 108 h.pmol/l, p<0.001 vs. saline). Although this was lower when compared with kisspeptin IR after the same dose of kisspeptin-10 to men (in Study 1, 700

± 160 h.pmol/l for men), this difference was not statistically significant (p=0.42 vs. men). At the dose of 10 nmol/kg kisspeptin-10, mean peak kisspeptin IR (2638 ± 302 pmol/l) was observed 10 minutes after injection (as with males). Plasma kisspeptin IR subsequently returned to undetectable levels 50 minutes after injection (as with males) (Fig 3-7A).

Kisspeptin-54 was also administered as an IV bolus in this study as a positive control as previous work has demonstrated that it increases reproductive hormones in females. This larger form of kisspeptin had a smoother more prolonged profile with peak kisspeptin IR occurring 30-40 minutes after injection in contrast to kisspeptin-10 which peaked at 10 minutes after injection (Figure 3-7A).

Unlike in the male study, no significant changes in serum LH, FSH or oestradiol were observed after intravenous bolus injection of kisspeptin-10 at all doses tested in the follicular phase (Figures 3-8, 3-9, 3-10). However mean LH and FSH AUC were significantly elevated after intravenous bolus injection of 1 nmol/kg of the longer form of kisspeptin, kisspeptin-54 to healthy women in the follicular phase (Figures 3-8, 3-9).

108

3.4.2.2: Pre-ovulatory phase of the menstrual cycle

Kisspeptin IR was significantly elevated in women during the pre-ovulatory phase after kisspeptin-10 injection compared with saline. This elevation was not significantly different when compared with kisspeptin IR after the same dose of kisspeptin-10 in follicular phase women or men (mean AUC kisspeptin IR in pre-ovulatory phase was 320 ± 56 h.pmol/l, p=0.13 vs. follicular phase and p=0.06 vs. men) (Figure 3-7).

Serum LH and FSH were both significantly elevated after intravenous bolus injection of 10 nmol/kg kisspeptin-10 in female volunteers during the pre-ovulatory phase of the menstrual cycle (mean AUC increase was 30.4 ± 11.1 h.iU/l (LH), p<0.05 vs. saline, and

6.9 ± 0.9 h.iU/l (FSH), p<0.01 vs. saline) (Figures 3-8, 3-9). This in contrast to the lack of response seen in the follicular phase with kisspeptin-10.

Peak LH and FSH stimulation occurred 40 minutes after injection (similar time frame to that observed in males) of kisspeptin-10. Significant elevation in LH and FSH was maintained until 75 minutes after which the fall in levels was steeper compared to males

(Figures 3-8A, 3-9A).

Serum oestradiol, however, was not altered significantly by kisspeptin-10 in the pre- ovulatory phase (mean AUC oestradiol increase was 111 ± 96 h/pmol/l, p=0.13 vs. saline)

(Figure 3-10). This in contrast to the gonadal sex steroid testosterone which was elevated towards the end of the timecourse after 1 nmol/kg kisspeptin-10 in males (Figure 3-6A).

109

A  3500     Saline follicular  3000 1nmol/kg KP10 follicular 2500 3nmol/kg KP10 follicular 2000 10nmol/kg KP10 follicular 10nmol/kg KP10 preov 1500 Saline preov 1000 KP54 follicular 500

Kisspeptin IR (pmol/L) IR Kisspeptin 0 0 30 60 90 120 150 180 Time after Injection (min)

B 1000 ** *** *** 800 ** 600

400

200

AUC Kisspeptin IR (h.pmol/l) IR Kisspeptin AUC 0 Saline 1 3 10 KP54 Saline 10 FOLLICULAR PREOV

Figure 3-7: A- Plasma kisspeptin IR after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women. For 3 nmol/kg vs. saline: φφφ, p<0.001. For 10 nmol/kg follicular vs. saline: λλλ, p<0.001. For 10 nmol/kg pre-ovulatory vs. saline: µµµ, p<0.001. For kisspeptin-54 (1nmol/kg) vs. saline: θ, p<0.05; θθθ, p<0.001. B- Mean kisspeptin IR AUC for saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 over the 180 minute timecourse post-intravenous bolus in follicular and pre-ovulatory phase. *, p<0.05; **, p<0.01; ***, p<0.001.

110

A 35   Saline follicular 30 1nmol/kg KP10 follicular 25 3nmol/kg KP10 follicular 20 10nmol/kg KP10 follicular 15 10nmol/kg KP10 preov 10 Saline preov 5 KP54 follicular

LH Increase (iU/l) Increase LH 0 -5 0 30 60 90 120 150 180

Time after Injection (min)

B * * 50 40 30 20 10 0 -10

AUC LH Increase (h.iU/l) LH AUC -20 Saline 1 3 10 KP54 Saline 10

FOLLICULAR PREOV

Figure 3-8: A- Serum LH after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women. For 10 nmol/kg pre-ovulatory vs. saline: µµ, p<0.01; µµµ, p<0.001. B- Mean serum LH AUC for saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 over the 180 minute timecourse post-intravenous bolus in follicular and pre-ovulatory phase. *, p<0.05; **, p<0.01; ***, p<0.001.

111

A     6 Saline follicular 1nmol/kg KP10 follicular 4 3nmol/kg KP10 follicular 10nmol/kg KP10 follicular 2 10nmol/kg KP10 preov Saline preov 0 KP54 follicular

FSH Increase (iU/l) Increase FSH

-2 0 30 60 90 120 150 180

Time after Injection (min)

B * ** 12 10 8 6 4 2 0 -2

AUC FSH Increase (h.iU/l) FSH AUC -4 Saline 1 3 10 KP54 Saline 10

FOLLICULAR PREOV

Figure 3-9: A- Serum FSH after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women. For 10 nmol/kg pre-ovulatory vs. saline: µ, p<0.05; µµ, p<0.01; µµµ, p<0.001. B- Mean serum FSH AUC for saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 over the 180 minute time-course post-intravenous bolus in follicular and pre-ovulatory phase. *, p<0.05; **, p<0.01; ***, p<0.001.

112

A 150 Saline follicular 100 1nmol/kg KP10 follicular 50 3nmol/kg KP10 follicular 0 10nmol/kg KP10 follicular 10nmol/kg KP10 preov -50 Saline preov -100 KP54 follicular -150

Oestradiol Increase (pmol/l) Increase Oestradiol -200 0 30 60 90 120 150 180

Time after Injection (min)

300

200

100

-100

-200

-300 AUC Oestradiol Increase (h.iU/l) Increase Oestradiol AUC Saline 1 3 10 KP54 Saline 10

FOLLICULAR PREOV

Figure 3-10: A- Serum Oestradiol after intravenous bolus injection of saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 in follicular or pre-ovulatory phase to healthy women. B- Mean serum Oestradiol AUC for saline, 1, 3, 10 nmol/kg kisspeptin-10 and 1 nmol/kg kisspeptin-54 over the 180 min time-course post- intravenous bolus in follicular and pre-ovulatory phase.

113

3.4.3 Study 3: Determining the half-life of kisspeptin-10 in healthy male and female volunteers (follicular and pre-ovulatory phase)

To determine the plasma half-life of kisspeptin-10, frequent blood sampling was performed in men and, women during the follicular and pre-ovulatory phases of the menstrual cycle, after cessation of an intravenous infusion of 360 pmol/kg/min of kisspeptin-10 in order to achieve a steady state (Figure 3-11). During the infusion males had higher plasma kisspeptin IR than females in both the follicular and pre-ovulatory phases (Figure 3-11).

When plotted on a natural log scale, linear regression slopes were not significantly different among the three groups (F = 0.13; Degrees of Freedom = 2; Degrees of

Freedom denominator = 137; p=0.88) (Figure 3-12). Plasma half-lives of kisspeptin-10 were calculated using these regression gradients and were also statistically similar between all three groups: male, 4.62 ± 0.34 minutes; female follicular phase 4.72 ± 0.43 minutes; female pre-ovulatory phase 4.40 ± 0.34 minutes (Figure 3-12).

114

Infusion 2500   Male 2000 Female: Follicular      1500 Female: Preovulatory 

1000

500

Kisspeptin IR (pmol/L) IR Kisspeptin 0 -30 0 30 60 90 120 150 180 Time (min)

Figure 3-11: Plasma kisspeptin IR before, during and after a 90 minute infusion of kisspeptin-10 (360 pmol/kg/min) to healthy males and females. Blood was sampled every 1 min between 90-100 minutes and then every 2 minutes between 100-120 minutes. For men vs. women during the follicular phase: ψ, p<0.05; ψψ, p<0.01; ψψψ, p<0.001. For men vs. women during the pre-ovulatory phase: ω, p<0.05; ωωω, p<0.001.

115

7.5

Male

7.0 Female: Follicular

Female: Preovulatory 6.5

6.0

5.5

Natural Log (kisspeptin IR pmol/l) IR (kisspeptin Log Natural

5.0 90.0 90.5 91.0 91.5 92.0 92.5 93.0 93.5 94.0 94.5 95.0 95.5 96.0 96.5 97.0 97.5 98.0 98.5 99.0 99.5 100.0 Time (min)

Figure 3-12: Natural Logarithm kisspeptin IR against time immediately after cessation of a 90 minute infusion of 360 pmol/kg/min kisspeptin-10. Vertical detached lines indicate calculated half-lives for females in the follicular and pre-ovulatory phase and males.

116

3.5 Discussion

This study reveals a previously unknown sexual dimorphism in responsiveness to the short form of kisspeptin, kisspeptin-10, by intravenous bolus injection in healthy men and women. Numerous studies in various male and female animal species (rodents, ewes, cows) have demonstrated that kisspeptin-10 robustly stimulates gonadotrophin release, suggesting kisspeptin signalling as a potential therapeutic target for treating reproductive disorders (Irwig et al. 2004, Navarro et al. 2004, Shahab et al. 2005, Plant et al. 2006,

Arreguin-Arevalo et al. 2007, Caraty et al. 2007, Ramaswamy et al. 2007, Kadokawa et al.

2008, Lents et al. 2008). Work has also been done on male (but not female) primate models showing gonadotrophin stimulation (Shahab et al. 2005, Plant et al. 2006). More recently two studies have shown that kisspeptin-10 also stimulates gonadotrophin release in healthy men (Chan et al. 2011, George et al. 2011). However, the effects of administration of kisspeptin-10 in female primates or women had not previously been studied at the time of commencing these studies.

Kisspeptin-10 stimulates reproductive hormone release in healthy men.

My study has shown that intravenous bolus injection of kisspeptin-10 significantly stimulated LH release in healthy men at all doses tested. These results are consistent with the previous literature on kisspeptin-10 using similar doses (Chan et al. 2011, George et al. 2011) and the longer form kisspeptin-54 (Dhillo et al. 2005) in healthy men. This effect is likely to be due to direct stimulation by kisspeptin of the kisspeptin receptor on GnRH neurones as detailed in the introduction.

There was also stimulation of FSH although this was to a lesser degree and later than that observed for LH. This will be discussed together with the female study later.

117

In the current study, all doses of kisspeptin-10 tested (0.3–10 nmol/kg) were associated with similar degrees of gonadotropin secretion in healthy male subjects. A recent study suggests that kisspeptin-10 stimulates serum LH secretion at doses as low as 0.01 μg/kg

(equivalent to 0.008 nmol/kg), with a maximal response seen at 1 μg/kg (0.8 nmol/kg) in men (George et al. 2011). Taking these data into consideration, all doses of kisspeptin-10 selected during my study may have stimulated near-maximal levels of gonadotropin secretion in healthy male volunteers. Interestingly, the study by George et al. observed that the increase in serum LH at 3 μg/kg (2.4 nmol/kg) was lower than the increase in serum LH at 1 μg/kg (0.8 nmol/kg), which they speculated was attributable to tachyphylaxis, a phenomenon that has been observed after chronic kisspeptin-54 injections and discussed earlier (Jayasena et al. 2009, Jayasena et al. 2010). Therefore I cannot exclude that the similarity in LH responses at all doses between 0.3-10 nmol/kg kisspeptin-10 in men might be explained in part by tachyphylaxis to kisspeptin-10 at the higher tested doses.

An increase in testosterone release was also observed in the current study, occurring towards the end of the study time course as this is a downstream resultant effect of the preceding LH rise.

Kisspeptin-10 does not stimulate reproductive hormone release in healthy women in the follicular phase unlike the longer form kisspeptin-54.

It has previously been observed that kisspeptin-10 does indeed stimulate LH release in female adult mice although to a lesser extent than kisspeptin-54 (Jayasena et al. 2011).

Furthermore, studies in female cows and ewes have demonstrated reproductive hormone stimulation by kisspeptin-10 (Arreguin-Arevalo et al. 2007, Kadokawa et al. 2008). Hence it was surprising that the doses of kisspeptin-10 that stimulated LH release in healthy males failed to stimulate reproductive hormone release in healthy females during the

118 follicular phase of the menstrual cycle. Indeed, it is noteworthy that only a marginal elevation of plasma kisspeptin IR (approximately 10 h.pmol/l) was necessary to stimulate significant LH secretion in men after injection of kisspeptin-10 (0.3 nmol/kg iv bolus); by contrast, a 50-fold greater elevation in plasma kisspeptin IR failed to stimulate LH release in women given the high dose 10 nmol/kg kisspeptin-10 during the follicular phase. It may be expected therefore, that levels of kisspeptin-10 may have been modified by factors known to differ between the sexes, such as body fat content or clearance of the peptide from the plasma (Soldin et al. 2011). However against this, the doses used in the study produced similar cumulative plasma kisspeptin IR in both males and females (p=0.42 for

10 nmol/kg kisspeptin-10 in men vs. women). Furthermore, my study demonstrated no differences in plasma half-life and linear regression gradient for kisspeptin-10 between males and females.

Interestingly and consistent with the previous literature, kisspeptin-54 was able to significantly stimulate LH and FSH increases in women despite similar cumulative exposure to plasma kisspeptin (AUC Kisspeptin IR between 10 nmol/kg kisspeptin-10 vs.

1 nmol/kg kisspeptin-54, p=0.21). It is important to note here that the radioimmunoassay used to detect plasma kisspeptin IR detects all forms of kisspeptin (-54, -14, -13, -10).

Previous studies have described the increased potency of kisspeptin-54 over kisspeptin-

10 in stimulating LH release in female rodents (Matsui et al. 2004, Jayasena et al. 2011).

It is interesting to speculate as to the possible explanations for the different female gonadotrophin response to kisspeptin-10 and kisspeptin-54:

1. An increased affinity or efficacy of kisspeptin-54 vs. kisspeptin-10 for the kisspeptin receptor. This explanation seems unlikely given the fact that a ten-fold increase in molar dose for kisspeptin-10 over kisspeptin -54 still failed to stimulate reproductive hormone release in this study. In addition, kisspeptin-10 has been reported

119

Ohtaki et al. Nature, 2001, Asami et al. Bioorg Med Chem Lett, 2012 to have a similar binding affinity and efficacy compared to kisspeptin-54 for the kisspeptin receptor in vitro (Kotani et al. 2001). However, it is possible that the binding affinity may be different in the in vivo setting.

2. A different time-profile of exposure of kisspeptin-54 vs. kisspeptin-10 to the kisspeptin receptor. There was clearly a smoother time-profile of plasma kisspeptin IR with significant increases (compared to saline) noted up to 90 minutes after injection after administration of kisspeptin-54. This was in contrast to kisspeptin-10 administration that resulted in significantly increased kisspeptin IR only up to 20 minutes post-injection. This difference between kisspeptin-10 and kisspeptin-54 time-profiles may be a consequence of the rapid breakdown of kisspeptin-10 in the circulation. Indeed the in vivo plasma half- life of iv kisspeptin-10 in this study was calculated as approximately 4 minutes, which is 7- fold shorter than the calculated in vivo plasma half-life of kisspeptin-54 (28 minutes,

(Dhillo et al. 2005)). Hence it is possible that the kisspeptin receptor requires a more prolonged exposure to raised plasma kisspeptin levels for activation. However, data which is not in keeping with this explanation is that administering a subcutaneous dose or a 90 minute infusion of kisspeptin-10 which resulted in a similar time-profile of plasma kisspeptin IR as that seen with kisspeptin-54 here, also failed to cause reproductive hormone stimulation (Jayasena et al. 2011).

3. The potency of kisspeptin-10 may be more dependent on the background hormonal milieu of testosterone/oestradiol than kisspeptin-54. This is further evidenced by the robust responses of gonadotrophin release to kisspeptin-10 during the higher oestradiol state of the pre-ovulatory phase seen in the current study. This may relate to the sexual dimorphism seen in different populations of kisspeptin neurones which is detailed in the next section.

120

Kisspeptin-10 stimulates reproductive hormone release in pre-ovulatory phase healthy women.

It has previously been demonstrated that women in the pre-ovulatory phase of the menstrual cycle are significantly more sensitive to the effects of kisspeptin-54 on gonadotrophin release when compared with women in the follicular phase of the menstrual cycle (Dhillo et al. 2007). In keeping with this observation, intravenous bolus administration of kisspeptin-10 significantly stimulated LH and FSH release in women during the pre-ovulatory phase of the menstrual cycle in this study.

Given that the pharmacokinetic profiles of plasma kisspeptin IR in both the follicular and pre-ovulatory phases were similar, these results suggest that women have heightened sensitivity to kisspeptin-10 during the pre-ovulatory phase of the menstrual cycle. Hence it is plausible that the different background hormonal milieu with higher oestradiol levels in the pre-ovulatory phase may somehow influence the responsiveness to kisspeptin-10.

To explore the different gonadotrophin release responses to kisspeptin in the pre- ovulatory and follicular phase in healthy women, it is interesting to consider previous rodent work that has identified two distinct kisspeptin neuronal populations that have contrasting responses to oestradiol. In female rodents, c-Fos expression within kisspeptin neurones and levels of Kiss1 expression are increased within the AVPV nucleus of the hypothalamus in the proestrus phase as a result of positive feedback on this nucleus by oestradiol (Smith et al. 2006). This contrasts against the negative feedback by oestradiol on a different distinct population of kisspeptin neurones in the ARC as evidenced by reduced levels of ARC Kiss1 expression in ovariectomised female rats administered oestradiol (Smith et al. 2005). This suggests a complex interplay between AVPV and ARC kisspeptin neurones with GnRH neurones in female rodents (Figure 3-13).

Of note, the AVPV (but not the ARC) in rodents is sexually dimorphic with a greater number of kisspeptin neurones in females compared to males (Clarkson and Herbison

121

2006, Kauffman et al. 2007). This dimorphism is a result of neonatal exposure to testosterone, which is aromatised into oestrogen and causes defeminisation of the AVPV region (Homma et al. 2009). However this does not explain the sexual dimorphism in gonadotrophin responses to kisspeptin-10 administration.

122

Figure 3-13: Sex steroid feedback loops have different effects on the two identified kisspeptin neuronal populations in the ARC and AVPV of the hypothalamus in rodents.

123

It is interesting to consider how these points may relate to human kisspeptin signalling. To date, KISS1 expressing neurones have only been identified in humans in the infundibular nucleus and to a lesser extent the medial preoptic area (Rometo et al. 2007). These neurones express the Oestrogen Receptor α (ERα). It is important to note that GnRH neurones express ERβ but this is not involved in feedback mechanisms and hence it is thought that gonadal steroid feedback on the hypothalamus in the reproductive axis is via the kisspeptin neurones. In the low oestrogen state of menopause there is a significant increase in KISS1 expression in the infundibular nucleus together with hypertrophy and increased numbers of these KISS1-expressing neurones when compared to pre- menopausal specimens (Rance 2009). Hence this suggests that higher oestrogen states may down-regulate KISS1 expression in humans at least when comparing pre- and postmenopausal states.

Recently a putative human equivalent of the AVPV which contains a prominent kisspeptin- immunoreactive cell group in the female but not the male hypothalamus has been identified in the RP3V (Hrabovszky et al. 2010). This may help partly explain the sexual dimorphism seen in response to kisspeptin-10 in the current study. However further work is needed to confirm whether its kisspeptin neurones are positively stimulated by increased oestrogen levels similar to the non-primate AVPV kisspeptin neurones which could help explain the pre-ovulatory gonadotrophin surge in human females.

In my study I have administered kisspeptin exogenously. Hence it is important to also consider that there may exist sexual dimorphism in the levels of expression of the kisspeptin receptor in humans, which may help to explain the different results in men and women in response to kisspeptin-10 in my study. However, there are currently no studies which have investigated kisspeptin receptor expression in humans.

Another possible explanation as to why there was a greater gonadotrophin response in the pre-ovulatory rather than follicular phase of the menstrual cycle may be provided by

124 studies in animals. One study investigating Kiss1r expression by GnRH neurones in the median eminence of young female adult macaques, showed no variation in different phases of the menstrual cycle as well as no influence of gonadal status or short/long-term oestradiol treatment (Eghlidi et al. 2010). However, more recent data suggests that the key may lie in changes found in the circadian expression of the Kiss1r gene that earlier studies had not examined. In female rodents, high levels of oestradiol elicited circadian expression profiles of Kiss1r in vitro. These profiles demonstrated that GnRH-expressing and GnRH-secreting cells (GT1-7) exposed to oestradiol exhibited both increased absolute Kiss1r expression as well as increased Kiss1r expression rhythmicity when compared to those exposed to vehicle only (Tonsfeldt et al. 2011). In this study the authors also demonstrated that oestradiol treatment potentiated kisspeptin-induced GnRH secretion. Hence it is plausible that the higher oestradiol levels in rodents just before ovulation may increase the GnRH neuronal sensitivity to various afferent signals including kisspeptin. Indeed, this may partly explain my findings of increased gonadotrophin response to kisspeptin-10 in the pre-ovulatory compared to the follicular phase in healthy women.

Along these lines, even large doses of kisspeptin-10 failed to elicit a gonadotrophin response in the early-follicular phase in women providing evidence that GnRH neurones are relatively resistant to kisspeptin in this phase. This suggests another possibility: that endogenous kisspeptin secretion is already at maximal levels providing near-maximal stimulation of the GnRH neurones in the early follicular phase. Hence the GnRH neurones have only a limited ability to mount any further response to additional exogenous kisspeptin. This may indicate that there is maximal ‘kisspeptin tone’ during the early follicular phase of the menstrual cycle. In keeping with this, a study in rhesus monkeys has observed that kisspeptin secretion is negatively regulated by circulating oestradiol levels (Guerriero et al. 2012). Hence there may indeed be higher ‘kisspeptin tone’ during the early follicular phase of the menstrual cycle when oestradiol levels are at their lowest.

125

Another factor to consider is KISS1/KISS1R expression outside the hypothalamus, downstream in the pituitary gonadotrophs. In rat pituitary gonadotrophs, oestradiol increases Kiss1 expression (similarly to in the AVPV), while Kiss1r expression is positively regulated by GnRH and negatively controlled by chronic exposure to oestradiol (Richard et al. 2008). Hence taking this latter point into account, it is plausible that the sustained lower oestradiol in the follicular phase which precedes the pre-ovulatory phase may indeed lead to increased Kiss1r expression at a pituitary level which is maximal leading into the pre-ovulatory phase. Hence exogenous kisspeptin may have more receptor to bind to in the pre-ovulatory phase compared to the follicular phase resulting in greater gonadotrophin release.

Putting this all together to explain the menstrual phase-dependent responses to kisspeptin-10 in the current study, it is likely that there is a complex interaction dependent on background oestradiol levels between kisspeptin neurones at various sites, exogenous kisspeptin-10 and the kisspeptin receptors on GnRH neurones and gonadotrophs. This requires further study of the responses to exogenous kisspeptin-10 under different oestradiol conditions to clarify this further.

Kisspeptin-10 stimulates LH more potently than FSH.

Similar to the male study, kisspeptin-10 in the pre-ovulatory phase of healthy women stimulated LH secretion more potently than FSH. This is consistent with previous studies of kisspeptin-10 and kisspeptin-54 in healthy men and women (Dhillo et al. 2005, Dhillo et al. 2007, Chan et al. 2011, George et al. 2011). In addition, in men the peak FSH increase occurred slightly later than the peak LH increase following intravenous bolus of kisspeptin-

10 (40-150 minutes for FSH vs. 30-50 minutes for LH). This difference has previously been observed for both males and females (Dhillo et al. 2005, Dhillo et al. 2007). This

126 study therefore suggests that both kisspeptin-10 and kisspeptin-54 stimulate LH secretion more potently and more rapidly than FSH.

Administration of GnRH demonstrates a similar pattern to kisspeptin-10 and kisspeptin-54 of differing LH/FSH potency and rapidity of effect. Combining this with the additional similarity that LH and FSH responses to exogenous GnRH are also heightened in the pre- ovulatory phase, provides further evidence to suggest that GnRH is the intermediary messenger for kisspeptin to stimulate the anterior pituitary production of LH and FSH.

Kisspeptin-10 and Testosterone/Oestradiol.

It is interesting to consider why in my study there was no consistent stimulation of testosterone secretion after intravenous bolus kisspeptin-10 injection in healthy men when compared with the robust increases in serum LH and FSH observed at all tested doses.

Significant increases in serum testosterone were observed only at 1.0 nmol/kg iv bolus kisspeptin-10, but these rises were marginal (approximately 10 h.nmol/l above baseline).

Similarly, intravenous bolus injection of kisspeptin-10 stimulated gonadotrophin release in women during the pre-ovulatory phase of the menstrual cycle but did not increase serum oestradiol during 3 hours after injection. Previous data suggest that at least 4 hours are required for serum levels of sex steroids to peak after a subcutaneous bolus injection of kisspeptin-54 (Dhillo et al. 2005, Dhillo et al. 2007, Jayasena et al. 2009). A longer period of blood sampling after injection may have revealed more pronounced alterations in sex steroid secretion in subjects after injection of kisspeptin-10. Alternatively it is possible that intravenous bolus injection of kisspeptin-10 and kisspeptin-54 have a duration of action inadequate to stimulate significant gonadal sex steroid release.

127

Limitations

Interestingly my study demonstrated non-significant reductions in serum LH (p=0.19) and

FSH (p=0.91) after just saline injection in men compared with women. Although both men and women underwent identical study protocols, it is possible that minor differences in factors inhibiting reproductive hormone release such as stress may have been greater in the male group. However such a difference does not explain why men are more responsive to kisspeptin-10 than women in the follicular phase of the menstrual cycle.

As mentioned above, the effects of kisspeptin on sex steroid (testosterone/oestradiol) take several hours to manifest, unlike the fast response in LH secretion (Dhillo et al. 2005,

Dhillo et al. 2007). Hence a longer protocol of blood sampling may be required for the detection of sex steroid changes after kisspeptin administration.

Summary

In summary, this is the first clinical study to directly compare the effects of kisspeptin-10 administration on reproductive hormone release in healthy men and women. Kisspeptin-

10 robustly stimulates gonadotrophin release in men and women in the pre-ovulatory phase but fails to stimulate gonadotrophin release in the follicular phase. This sexual dimorphism as well as the different responses between the follicular and pre-ovulatory phase require further study to elucidate the mechanisms involved. However these findings have important clinical implications for the potential therapeutic use of kisspeptin-10 to treat disorders of reproduction suggesting that kisspeptin-54 may offer a more potent alternative in women.

128

3.6 Future Work

This study revealed both sexual dimorphism as well as different responses to kisspeptin-

10 in the follicular and pre-ovulatory phase. In keeping with the findings of my study, another study published recently also observed greater gonadotrophin response to kisspeptin-10 during the pre-ovulatory phase, with only half of the ten women responding to kisspeptin-10 during the early follicular phase (Chan et al. 2012). Furthermore, evidence of sexual dimorphism was also observed in that unlike in men where kisspeptin-

10 resets the endogenous GnRH pulse generator, this effect was not observed in women.

These data suggest further work is required to clarify the varied responses to kisspeptin according to sex and background sex steroid levels. A recent study suggests that kisspeptin-10 stimulates LH secretion in post-menopausal women (low endogenous sex steroids/high gonadotrophin secretion) but does not in women taking the combined

(oestrogen and progesterone) sex steroid contraceptive pill (high sex steroid/low gonadotrophin secretion) (George et al. 2012). By contrast, in my study and that by Chan

(Chan et al. 2012), LH response to kisspeptin-10 was increased during the higher sex- steroid conditions of the pre-ovulatory phase. This variance may reflect the effects of longer-term sex steroid supplementation/withdrawal on LH response to kisspeptin-10, rather than the shorter term hormonal fluctuations observed during the menstrual cycle.

Furthermore, gonadotrophin secretion is suppressed in the high sex steroid environment of sex steroid contraceptive pill supplementation compared to the pre-ovulatory phase where gonadotrophin secretion is increased. In addition lower doses of kisspeptin-10 were used in the study by George et al. (George et al. 2012) compared to my study. Further work is required to clarify this for example by assessing LH response to lower doses

(<1nmol/kg) of kisspeptin-10 during various menstrual phases and after only a few days of oestrogen supplementation (rather than longer term supplementation). In addition, further human studies are required to identify kisspeptin neuronal subpopulations that may have different responses to oestradiol feedback similar to the ARC and AVPV in rodents.

129

Another issue to address is the route of administration of kisspeptins. Subcutaneous administration is cheaper and more convenient for the patient than intravenous injection.

A study comparing the responses to equimolar kisspeptin-10 and kisspeptin-54 via subcutaneous versus intravenous route would be of clinical interest.

Finally, the potential use of repeated injections of kisspeptin-54 has demonstrated tachyphylaxis in women with hypothalamic amenorrhoea (Jayasena et al. 2009, Jayasena et al. 2010). However, this repeated (chronic) kisspeptin administration has not been investigated in healthy women. I address this issue in the next chapter.

130

Chapter 4 ______

The Effects of Chronic Kisspeptin-54 administration in Healthy Women

131

4.1 Introduction

Central or peripheral administration of kisspeptin induces gonadotrophin and sex steroid release in all mammalian species investigated, including rats (Thompson et al. 2004,

Patterson et al. 2006, Tovar et al. 2006), mice (Gottsch et al. 2004, Messager et al. 2005), hamsters (Greives et al. 2011), monkeys (Seminara et al. 2006), cows (Kadokawa et al.

2008), pigs (Lents et al. 2008), goats (Hashizume et al. 2010), sheep (Caraty et al. 2007,

Redmond et al. 2011), and goldfish (Chang et al. 2012).

Administration of kisspeptin-54 or kisspeptin-10 acutely stimulates gonadotrophin secretion in healthy male (Dhillo et al. 2005, Chan et al. 2011, George et al. 2011) and female volunteers (Dhillo et al. 2007, Jayasena et al. 2011, Chan et al. 2012, Jayasena et al. 2013), and women with functional HA (hypothalamic amenorrhea) (Jayasena et al.

2009, Jayasena et al. 2010, Jayasena et al. 2014). However, twice-daily subcutaneous administration of kisspeptin-54 to women with HA causes profound tachyphylaxis within

24 hours of commencing treatment (Jayasena et al. 2009). In fact after 2 weeks of treatment kisspeptin no longer elicited a detectable LH response (Jayasena et al. 2009).

However, the response to exogenously administered GnRH was intact, suggesting that continuous exposure to kisspeptin resulted in desensitisation at the level of the kisspeptin receptor. As plasma kisspeptin IR remains elevated for over 4 hours following subcutaneous kisspeptin-54 (6.4 nmol/kg), then it is likely that this dosing regimen resulted in continuous exposure to kisspeptin. A subsequent study attempted to avoid inducing desensitisation by employing twice weekly subcutaneous injections of kisspeptin

(6.4nmol/kg) rather than twice daily (Jayasena et al. 2010). Responses to kisspeptin were still observed after 8 weeks on this protocol, although they were dampened in comparison to day 1.

Furthermore continuous intravenous infusion of kisspeptin-10 to monkeys is associated with tachyphylaxis within 3 hours of commencing administration (Seminara et al. 2006). A

132 similar desensitisation has been observed in rodents during continuous intracerebral infusion (Roa et al. 2008). It has therefore been assumed that chronic treatment with kisspeptin-54 in healthy women may have limited therapeutic potential as a stimulator of human reproductive activity, due to tachyphylaxis observed following chronic administration in rodents, primates and women with HA. However, this hypothesis has not been tested previously in normal women with regular menstrual cycles. To test this hypothesis I used the exact same dose of kisspeptin-54 that previously caused tachyphylaxis within 24 hours in women with hypothalamic amenorrhoea (Jayasena et al.

2010). As kisspeptin secretion is itself pulsatile (Keen et al. 2008), it is likely that pulsatile delivery of kisspeptin in a similar way to GnRH is required to stimulate the reproductive axis (Belchetz et al. 1978). I therefore designed a protocol similar to the previous study in women with hypothalamic amenorrhoea with twice daily subcutaneous kisspeptin administration.

I performed a single-blinded placebo-controlled one-way crossover study to determine the effects of twice-daily administration of kisspeptin for 7 days on the menstrual cycle in healthy human female subjects with regular menstrual cycles.

133

4.2 Aims and Hypothesis

Aims

1. To determine the effects of repeated (chronic) kisspeptin injection on menstrual cyclicity in healthy women.

2. To assess the feasibility for women to self-administer subcutaneous injections of kisspeptin at home.

Hypothesis

1. Chronic kisspeptin administration to healthy women will cause tachyphylaxis in a similar way to that seen in women with hypothalamic amenorrhoea.

134

4.3 Methods

4.3.1 Subjects

Ethical approval was granted by the Hammersmith and Queen Charlotte’s and Chelsea

Hospitals Research Ethics Committee (registration number: 05/Q0406/142). Written informed consent was obtained from all subjects. This study was performed in accordance with the Declaration of Helsinki. Five healthy female subjects with regular menstrual cycles were recruited through advertisements placed in local newspapers and a subsequent medical screening appointment (age in years: 31.6 ± 2.6, range 24-37; weight in kg: 60.4 ± 2.5, range 50.4-63.8) (Table 4-1)

4.3.2 Protocol

A single-blinded, placebo-controlled, one-way crossover design study of ten menstrual cycles was performed (Figure 4-1).

During month 1 of the study protocol, five healthy female subjects self-administered twice- daily subcutaneous saline injections between days 7 and 14 of their menstrual cycle.

During month 2 of the study protocol, the same five healthy female subjects self- administered twice-daily subcutaneous kisspeptin-54 injections (6.4 nmol/kg; equivalent to

37 mcg/kg) between days 7 and 14 of their menstrual cycle. Day 1 of each month was defined as the first day of menstrual bleeding.

135

DAY OF CYCLE: 1 6 7 11 14 15 18 21 24 28 Month 1 TWICE-DAILY SALINE USS USS USS USS USS USS

8h blood 8h blood 8h blood sampling sampling sampling

GnRH test 4h blood sampling

USS USS USS USS USS USS Month 2 TWICE-DAILY KP54 DAY OF CYCLE: 1 6 7 11 14 15 18 21 24 28

Figure 4-1: Study protocol diagram.

Five healthy women underwent a one-way crossover protocol. Twice-daily injections of saline and kisspeptin-54 were administered between days 7 and 14 during month 1 and 2 of the protocol, respectively. During each month of the protocol, subjects underwent a baseline GnRH test (day 1) and 8 hour blood sampling (day 6) prior to commencing injections. During the injection period (days 7-14), subjects underwent a 4 hour study (day 7) and two further 8 hour studies (days 11 and 14) immediately following injection of saline or kisspeptin-54. A GnRH test was performed 24 hours after cessation of saline or kisspeptin treatment (day 15). ‘USS’ denotes ultrasound examination and measurement of reproductive hormones. Day 1 was defined as the first day of menstrual bleeding.

136

4.3.3 Kisspeptin-54 peptide synthesis and testing

Although kisspeptin-10, -13, -14, and -54 display similar potency in vitro, we used kisspeptin-54 due to its higher in vivo potency than the other kisspeptin fragments

(Gottsch et al. 2004, Mikkelsen et al. 2009, Jayasena et al. 2011). Kisspeptin-54 was synthesised and tested as described in Section 3.3.5.

4.3.4 Kisspeptin injections

All subjects were trained in preparation and self-administration of subcutaneous injections at the start of the study protocol. At the beginning of each week when injections were to be performed, a box containing unlabelled vials of freeze-dried saline (month 1) or vials of freeze-dried kisspeptin-54 (month 2), alcohol wipes, saline ampoules for reconstitution of freeze-dried vial contents, 0.5ml insulin syringes with needles, and needle disposal bins was given to each subject. For injection, vial contents were reconstituted in 0.5ml of 0.9% saline by the subject. A 0.5ml insulin syringe was then used to inject saline alone (month

1) or 6.4 nmol/kg of kisspeptin-54 (month 2) into the lower abdominal region subcutaneously. Injection sites were rotated around the lower anterior abdominal region below the umbilicus. This kisspeptin-54 dose was the same used in previous work in healthy women and women with HA (Dhillo et al. 2007, Jayasena et al. 2009). Subjects were instructed to prepare and perform injections in the morning after breakfast (unless attending for a study in which case it was done by the subject at time=0 of the study) and in the evening before bed. The volume of the saline or kisspeptin injections was identical.

Subjects were instructed to refrigerate vials stored at home. Prior to commencing study visits, injection sites were inspected, and numbers of returned vials, insulin syringes, and saline or kisspeptin vials were counted in order to monitor compliance. Plasma kisspeptin immunoreactivity (IR) was also assessed throughout the study protocol to confirm compliance with injections.

137

4.3.5 Gonadotropin-releasing hormone (GnRH) test protocol

Subjects were cannulated and given a 100 mcg iv bolus injection of GnRH (HRF®,

Intrapharm Ltd, Kent, UK) at 0 minutes. Blood was sampled for measurement of serum

LH, FSH and oestradiol at -30, 0, 15, 30, 45, 60, 90, 120 minutes. This protocol was based on a standard in-hospital endocrine GnRH test.

4.3.6 Baseline period

During menstrual days 1 to 6 of each month of the study protocol, pituitary responsiveness was assessed by a GnRH test, and baseline reproductive hormones and ultrasound markers were measured (Figure 4-1). The collection, processing and analysis of blood samples is detailed below. The baseline period also allowed the acclimatisation of subjects to study conditions to minimise any effects of stress. No injections were administered during this time-period.

4.3.7 Treatment period

During menstrual days 7 to 14 of each month of the study protocol, subjects self- administered twice-daily, single-blinded subcutaneous injections of saline (month 1 of study protocol) or kisspeptin-54 (month 2 of study protocol) as above (Figure 4-1)

4.3.8 Post-treatment period

During menstrual days 15 to 28 of each month of the study protocol, subjects underwent a post-treatment observation period, in order to assess pituitary responsiveness (GnRH

138 test), and measure circulating reproductive hormones and ultrasound markers (Figure 4-

1). No injections were administered during this period.

4.3.9 Four hour blood sampling post-injection of saline or kisspeptin

All subjects underwent blood sampling during the 4 hour period immediately following the first injection of the saline (day 7, month 1) or kisspeptin-54 (day 7, month 2) treatment period in order to confirm that kisspeptin-54 acutely stimulated LH release as previously shown in healthy women (Dhillo et al. 2007). Saline or kisspeptin-54 6.4 nmol/kg was subcutaneously administered at 0 minutes by the subject, and blood was sampled for serum LH, FSH, oestradiol, and plasma kisspeptin IR at -30, 0, 10, 20, 40, 60, 90, 120,

150, 180, 210 and 240 minutes.

4.3.10 Eight hour blood sampling for assessments of LH pulsatility

Subjects underwent three assessments of LH pulsatility during month 1 and month 2 of the study protocol. Baseline assessment of LH pulsatility was performed on menstrual day

6 (1 day prior to commencing injections). LH pulsatility was also assessed following injection of saline or kisspeptin-54 on menstrual days 11 and 14. The saline or kisspeptin-

54 was reconstituted using one of the vials given to each subject at the beginning of the treatment period for home storage. Blood was sampled every 10 minutes. All studies commenced between 8.00am and 9.00am.

139

4.3.11 Assessment of pituitary sensitivity before and after injections of saline or kisspeptin

Subjects each underwent two GnRH tests during month 1 and month 2 of the study protocol. A baseline GnRH test was performed on menstrual day 1 and was repeated in each subject 24 hours after final saline or kisspeptin injection (menstrual day 15).

4.3.12 Basal measurement of reproductive hormones

Basal measurements of serum LH, FSH, oestradiol, progesterone, and plasma kisspeptin

IR were taken from subjects during days 1, 6, 7, 11, 14, 15, 18, 21 and 28 during month 1 and month 2 of the study protocol. On days 7, 11 and 14 the basal blood sample was taken before the morning injection of saline or kisspeptin-54.

4.3.13 Collection and processing of blood samples

Blood samples for serum analysis were collected and processed as in Section 3.3.2.

4.3.14 Ultrasound scans

Trans-abdominal ultrasound scans were performed on days 1, 7, 15, 18, 21 and 28 during month 1 and month 2 of the study protocol. Trans-vaginal scans were not routinely used to minimise discomfort to volunteers during repeated examinations. However trans-vaginal scans were performed if there was poor image quality on trans-abdominal scanning. The ultrasonographer was blinded to treatment for all subjects. During each scan the following parameters were measured: endometrial thickness in millimetres (mm); mean ovarian volume in cubic centimetres (cm3); mean follicle number; maximum diameter of largest follicle in each ovary (mm). Ovulation was defined as a rise in serum progesterone

140

>10nmol/l together with suggestive radiological features (visualisation of a dominant follicle with subsequent appearance of a pre-ovulatory follicle and/or corpus luteum)(Welt et al. 2004).

4.3.15 Other measurements

During each study visit, urine was tested in order to exclude pregnancy (Clearview easy-

HCG, Inverness Medical Innovations Inc., Waltham, MA). Blood pressure and heart rate were recorded regularly during all 4 hour studies and 8 hour studies. Two senior physicians were in attendance during 4 hour and 8 hour studies.

4.3.16 Analytical methods

Serum LH, FSH, and oestradiol were analysed as described in Section 3.3.3.

Progesterone was also analysed using an automated chemiluminescent immunoassay

(Abbot Laboratories, Abbott Park, IL). Inter- and intra-assay coefficient of variation were

1.8% and 2.9% respectively with an analytical sensitivity of 0.1nmol/l.

4.3.17 Kisspeptin radioimmunoassay

Plasma kisspeptin immunoreactivity (IR) was measured using a manual radioimmunoassay as described in Section 2.3.3.1.

141

4.3.18 Data analysis

Investigators performing the clinical studies were blinded to results until all subjects had completed the study protocol. An established, blinded deconvolution method with 93% sensitivity and specificity was employed to identify LH pulses, and calculate the secretory mass of LH pulses (Liu et al. 2009). Mean secretory mass is defined as the integral over time of the total amount of hormone released during a secretory burst normalised per litre of distribution volume (Liu et al. 2009). Cumulative levels of basal, pulsatile and total

(basal + pulsatile) LH secretion were also estimated during each study. These pulsatility analyses were designed and performed by Professor Johannes Veldhuis at the Mayo

Clinic, Rochester, Minnesota, USA. Data are presented as mean ± standard error of mean (SEM). Kisspeptin IR data was log-transformed to normalise data prior to data analysis. All other analyses included data series, the majority of which had normal distributions assessed using the Komogorov-Smirnov test with Dallal-Wilkinson-Lillie analysis. Hormone profiles during 4 hour blood sampling studies were analysed using repeated measures 2-way ANOVA with Bonferonni post hoc correction. Pairs of means were analysed using unpaired/paired two-tailed t-tests as appropriate. Multiple means were compared using one-way ANOVA with Bonferonni’s Multiple Comparison Test. In all cases, p<0.05 was considered statistically significant.

142

4.4 Results

4.4.1 Subjects

Baseline characteristics of the subjects recruited after medical screening are summarised below (Table 4-1).

All female subjects had regular periods for >24 months, normal reproductive hormone levels, had not been on contraceptive medication for >12 months, and a BMI 18.5-24.9 kg/m2. Normal medical history, examination and blood tests were required at screening for study entry. Mean normal menstrual cycle length reported at screening was 28.6 ± 1.1 days.

Participant Age Body Mass Normal cycle length (years) Index (kg/m2) (±1 day) 1 24 22.9 31 2 34 24.2 31 3 27 22.0 25 4 36 18.6 28 5 37 22.0 28 Table 4-1: Baseline characteristics of the 5 healthy female subjects at first study visit.

143

4.4.2 Acute effects of saline or kisspeptin-54 injection on plasma kisspeptin at the commencement, mid-point and end of twice-daily administration in healthy women

4.4.2.1 Acute changes in plasma kisspeptin IR following injection of saline or kisspeptin-

Plasma kisspeptin IR was unchanged at approximately 10pmol/l following injection of saline on the first (Figure 4-2A), fourth (Figure 4-2B) and final (seventh; Figure 4-2C) days of twice-daily administration. Kisspeptin-54 injection acutely elevated plasma kisspeptin IR on the first day of administration, with peak mean kisspeptin-IR levels of 2421±392pmol/l at 45 minutes post injection (p<0.001 vs. saline) (Figure 4-2A). Similar elevations of kisspeptin IR were observed following injection of kisspeptin on the fourth and last injection days (Figures 4-2B-C).

4.4.2.2 Plasma kisspeptin IR pre-injection of saline or kisspeptin-54

In subjects receiving saline, plasma kisspeptin IR measured prior to the morning saline injection remained approximately 10pmol/l throughout the study protocol (Figure 4-2D). In subjects receiving kisspeptin injections, plasma kisspeptin IR was not elevated on menstrual days 1, 6 or 7 since these blood samples were taken prior to the first kisspeptin-54 injection. Plasma kisspeptin IR was elevated on menstrual days 11 and 14

(during the twice daily kisspeptin treatment period but just prior to the morning kisspeptin injection), which suggested compliance with kisspeptin injection the previous evening. As expected, plasma kisspeptin IR was not elevated on the morning of day 15, since it was approximately 24 hours following final kisspeptin-54 injection and the half-life of kisspeptin-54 is approximately 28 minutes after intravenous administration (Dhillo et al.

2005). Furthermore plasma kisspeptin IR remained <10pmol/l on days 18-28 of the study protocol (Figure 4-2D).

144

A: Day 7 B: Day 11 C: Day 14 (First day of injections) (Mid-point of injections) (Last day of injections)

10000 10000 *** ** * 10000 *** ** Saline ** *** ** KP54 1000 1000 1000

100 100 100

10 10 10

1 1 1 0 60 120 180 240 0 120 240 360 480 0 120 240 360 480

Kisspeptin-IR (pmol/l)

Kisspeptin-IR (pmol/l) Time (Minutes) Time (Minutes) Time (Minutes) Injection Injection Injection D 70 Saline KP54 50

1 6 7 11 14 15 18 21 28 Kisspeptin-IR (pmol/l) Day of menstrual cycle

Figure 4-2: Acute changes in plasma kisspeptin levels in healthy women receiving twice-daily injections of saline or kisspeptin-54.

A-C: Time-profiles of plasma kisspeptin immunoreactivity (IR) following sc bolus injection of saline or kisspeptin-54 at time 0 minutes on menstrual days 7 (A), 11 (B) and 14 (C) of the study protocol. D: Bar graph comparing mean pre-injection kisspeptin IR following saline or kisspeptin-54 throughout the study protocol. All subjects commenced twice-daily treatment with saline or kisspeptin-54 on the morning of menstrual day 7 following their basal blood sample, and finished treatment on the morning of day 14. Data are presented as mean ± SEM. *, p<0.05; **, p<0.01.

145

4.4.3 Acute effects of saline or kisspeptin-54 injection on serum reproductive hormones at the commencement, mid-point and end of twice-daily administration in healthy women

4.4.3.1 Commencement of treatment period (menstrual day 7)

Saline injection did not change serum LH levels when compared with baseline (Figure 4-

3A, baseline LH 5.20±0.64IU/L). Kisspeptin-54 injection acutely increased serum LH levels in subjects when compared with saline (Figure 4-3A, p<0.05 at time-points 180 minutes to 210 minutes, baseline LH 5.68 ± 0.98IU/L). The mean maximal increase in LH from baseline after kisspeptin injection was observed at 180 minutes, and was 8.6 ±

3.4IU/L above baseline.

4.4.3.2 Mid-point of treatment period (menstrual day 11)

Saline injection did not change serum LH compared with baseline (Figure 4-3B, baseline

LH 5.74±1.45IU/L). Kisspeptin-54 injection acutely increased serum LH levels in subjects when compared with saline (Figure 4-3B, p<0.01 at 210 minutes, and p<0.05 at 220 to

240 minutes, baseline LH 9.41 ± 4.89IU/L). The mean maximal increase in LH from baseline after kisspeptin injection was observed at 210 minutes, and was 8.3 ± 2.4IU/L above baseline.

4.4.3.3 End of treatment period (menstrual day 14)

Saline injection did not change serum LH compared with baseline (Figure 4-3C, baseline

LH 17.79 ± 9.69IU/L). The mean maximal increase in LH from baseline after kisspeptin injection was observed at 220 minutes, and was 12.7 ± 8.1IU/L above baseline (baseline

LH 6.01 ± 1.60IU/L). Kisspeptin-54 injection showed a trend towards stimulating serum LH

146 but this did not reach statistical significance compared with saline. During days 7, 11 and

14 of the treatment period, total LH secretion was increased significantly following kisspeptin-54 when compared with saline (p<0.01) (Figure 4-3D).

4.4.3.4 Serum LH increase on different study days

I also compared the magnitude of serum LH increase 1 hour following kisspeptin-54 injection, with the peak increase in serum LH following kisspeptin-54 injection (Figure 4-4).

On menstrual day 7, the peak LH response was 3-fold higher when compared with the LH response 1 hour following kisspeptin-54 injection (increase in serum LH: 9.2 ± 4.5IU/L, peak; 3.1 ± 1.8IU/L, 1 hour, p=0.063 vs.1 hour). On menstrual days 11 and 14, there was virtually no change in mean serum LH 1 hour following kisspeptin-54 injection (<1IU/L), whereas peak increases in serum LH of 19.7 ± 9.5IU/L (p=0.059 vs. 1 hour) and 23.0 ±

12.9IU/L (p=0.067 vs. 1 hour) were later observed, respectively (Figure 4-4).

147

A: Day 7 B: Day 11 C: Day 14 (First day of injections) (Mid-point of injections) (Last day of injections) 30 30 30 Saline * KP54 20 20 ** * 20

10 10 10

0 0 0

0 30 60 90 120 150 180 210 240 0 30 60 90 120 150 180 210 240 0 30 60 90 120 150 180 210 240

Serum LH increase (IU/L)

Serum LH increase (IU/L) Serum LH increase (IU/L) Time (Minutes) Time (Minutes) Time (Minutes) Injection Injection Injection D ** 2000

1500 Saline 1000 KP54

500

-500 d7 d11 d14 d7 d11 d14

AUC LH increase LH (h.IU/L) AUC Day of menstrual cycle

Figure 4-3: Acute changes in serum reproductive hormones in healthy women receiving twice-daily injections of saline or kisspeptin-54.

A-C: Acute increases in serum LH following sc bolus injection of saline or kisspeptin-54 at time 0 minutes on menstrual days 7 (A), 11 (B) and 14 (C) of the study protocol. All subjects underwent treatment during menstrual days 7 and 14 with twice-daily sc bolus of saline or kisspeptin-54 injections. D: Summary bar graph comparing mean area under curve (AUC) LH increase following saline or kisspeptin-54, on days 7, 11 and 14 of the study protocol. Data are presented as mean ± SEM. *, p<0.05; **, p<0.01.

148

Figure 4-4: Time classification of acute serum LH responses to kisspeptin-54 in healthy women receiving twice-daily injections of kisspeptin-54.

Mean levels of serum LH increase 1 hour following kisspeptin-54 injection, were compared with peak increases in serum LH following kisspeptin-54 injection, on the 1st injection day (menstrual day 7), 4th injection day (menstrual day 11), and the 7th injection day (menstrual day 14). Data are presented as mean ± SEM. Paired, two-tailed t-tests did not reveal any significant differences between peak and 1 hour LH responses.

149

4.4.4 Effects of twice-daily saline or kisspeptin injections on length of the menstrual cycle and biochemical markers of reproductive activity in healthy women

4.4.4.1 Length of menstrual cycle

Subjects had the same menstrual cycle length prior to commencing the study when compared with menstrual cycle length during saline administration (mean menstrual cycle length in days: 28.6 ± 1.1, prior to study commencement; 28.6 ± 1.4, saline month; p=1.00). However all subjects had a shorter menstrual cycle (by approximately 2 days) during kisspeptin-54 treatment when compared with saline (mean length of menstrual cycle in days: saline 28.6 ± 1.4 vs. kisspeptin 26.8 ± 3.1, p<0.01) (Figure 4-5A).

4.4.4.2 Timing of peak serum LH

During kisspeptin-54 treatment, observed peak levels of serum LH and oestradiol, but not

FSH, were earlier during the menstrual cycle when compared with the saline group

(Figures 4-5B-D). Furthermore the menstrual day of highest recorded serum LH was approximately 2 days earlier during kisspeptin-54 treatment when compared with saline treatment (mean menstrual day of highest recorded serum LH: saline 15.2 ± 1.3 vs. kisspeptin 13.0 ± 1.9, p<0.05) (Figure 4-5E).

4.4.4.3 Timing of luteal phase of menstrual cycle

I examined the onset of the luteal phase which is characterised by release of a mature oocyte from the ovary and secretion of progesterone by the residual corpus luteum.

During kisspeptin-54 treatment, levels of serum progesterone became elevated

(>10nmol/L) earlier during the menstrual cycle when compared with the saline group

(Figure 4-5F). The menstrual day of onset of luteal phase (defined as starting when serum

150 progesterone was elevated >10nmol/L) was approximately 2 days earlier during kisspeptin-54 treatment when compared with saline treatment (mean menstrual day of serum progesterone increase >10nmol/l: saline 18.0 ± 2.1 vs. kisspeptin 15.8 ± 0.9, p<0.05) (Figure 4-5G).

151

Figure 4-5: Levels of serum reproductive hormones throughout the menstrual cycle in healthy women receiving twice-daily injections of saline or kisspeptin-54.

A: Comparison of cycle length in individual healthy women undergoing twice-daily subcutaneous bolus injections of saline or kisspeptin-54 between menstrual days 7 and 14 of the study protocol. B-D: Levels of serum LH (B), FSH (C) and oestradiol (D) during the menstrual cycle in healthy women undergoing twice-daily subcutaneous bolus injections of saline or kisspeptin-54 between menstrual days 7 and 14. E: Comparison of peak serum LH in individual healthy women undergoing twice-daily subcutaneous bolus injections of saline or kisspeptin-54 between menstrual days 7 and 14 of the study protocol. F-G: Levels of serum progesterone (F), and comparison of menstrual day of onset of luteal phase (G) in individual healthy women undergoing twice-daily subcutaneous bolus injections of saline or kisspeptin-54 between menstrual days 7 and 14 of the study protocol. The luteal phase of menstrual cycle was defined as beginning when serum progesterone was elevated >10nmol/l. *, p<0.05; **, p<0.01. Data are presented as mean ± SEM.

152

4.4.5 Effects of twice-daily saline or kisspeptin injections on radiological markers of reproductive activity in healthy women

A corpus luteum, indicating recent ovulation, was observed in all women following kisspeptin-54 treatment, and in 3/5 women following saline treatment (Table 4-2). The mean day of visualisation was day 23 for the saline-month and day 19 for the kisspeptin- month (p=0.29). Elevated serum progesterone (>10nmol/L) was observed in all subjects receiving saline or kisspeptin-54 treatment (Table 4-2). Immediately following 7 days of kisspeptin-54 treatment (on menstrual day 15), the mean diameter of the largest follicle seen during ultrasonography was significantly higher when compared with women during saline treatment (mean diameter of largest follicle in mm: saline 10.0 ± 2.2 vs. kisspeptin

15.5 ± 1.2, p<0.05) (Figure 4-6A). No significant differences in number of follicles or endometrial thickness were observed during kisspeptin-54 treatment when compared with saline (Figure 4-6B, C).

153

SALINE KISSPEPTIN-54 Participant Day CL Highest Day CL Highest observed Progesterone observed Progesterone (nmol/L) (nmol/L) 1 28 14 28 28 2 21 37 15 28 3 Not seen 47 21 21 4 Not seen 44 18 40 5 21 41 15 32 Mean ± SEM 23.3 ± 2.3 36.6 ± 13.2 19.4 ± 2.4 29.8 ± 6.9

Table 4-2: Table showing first menstrual day that corpus luteum (CL) was visualised by ultrasonography in each participant and highest recorded progesterone level during that menstrual cycle (saline or kisspeptin-54 treatment).

154

Figure 4-6: Changes in radiological markers during the menstrual cycle of healthy women receiving twice-daily injections of saline or kisspeptin-54.

Mean values for diameter of largest ovarian follicle (A), number of ovarian follicles (B) and endometrial thickness (C) in healthy women undergoing twice-daily sc bolus injections of saline or kisspeptin-54 (6.4 nmol/kg) between menstrual days 7 and 14. Data are presented as mean ± SEM.

155

4.4.6 Effects of twice-daily saline or kisspeptin on LH pulsatility in healthy women

LH pulsatility was determined immediately prior to commencing saline or kisspeptin-54 treatment (menstrual day 6 of study protocol), and during saline or kisspeptin-54 treatment

(menstrual days 11 and 14 of study protocol). These calculations were performed by

Professor Johannes Veldhuis at the Mayo Clinic (Rochester, Minnesota, USA). Mean secretory mass was slightly lower before kisspeptin-54 treatment when compared with saline treatment (mean secretory mass in IU/L: saline 5.9 ± 0.5 vs. kisspeptin 4.1 ± 0.4, p<0.01) (Figure 4-7A). Despite this, mean secretory mass was increased significantly during menstrual day 11 (4th day of treatment) during kisspeptin-54 treatment when compared with the saline treatment (mean secretory mass in IU/L: saline 3.4 ± 0.4 vs. kisspeptin 14.5 ± 3.6, p<0.01). On the last day of treatment (menstrual day 14) no significant differences in secretory mass were observed between saline or kisspeptin-54 treatment. No significant differences in pulse frequency were observed between saline and kisspeptin-54 treatment (Figure 4-7B). Pulsatile LH secretion was increased significantly during menstrual day 11 (4th day of treatment) in the kisspeptin-54 group when compared with the saline group (mean secretory mass in IU/L: saline 17.9 ± 3.6 vs. kisspeptin 78.2 ± 22.8, p<0.05) (Figure 4-7C). However basal and total LH secretion were not significantly different between groups (Figure 4-7D-E).

156

Figure 4-7: Changes in LH pulsatility during the menstrual cycle of healthy women receiving twice-daily injections of saline or kisspeptin-54.

All women underwent frequent blood sampling for 8h on menstrual days 6, 11 and 14 of the study protocol. Twice-daily sc bolus injections of saline or kisspeptin-54 (6.4 nmol/kg) were self-administered between menstrual days 7 and 14. A-B: Mean values for secretory mass (A), and estimated pulses per 24h (B) are presented. C-E: Levels of pulsatile (C), basal (D) and total (E) LH secretion during each 8h sampling study are presented. Data are presented as mean ± SEM.*, p<0.05; **, p<0.01.

157

4.4.7 Assessment of pituitary sensitivity before and after injections of saline or kisspeptin

I examined sensitivity of all healthy female subjects to intravenous GnRH, both 6 days before (menstrual day 1 of study protocol) and 24 hours after saline or kisspeptin treatment (menstrual day 15 of study protocol). On menstrual day 1, no significant difference in pituitary sensitivity was observed following GnRH administration between subjects prior to commencing saline of kisspeptin treatment (mean maximal LH increase during first 2 hours post-GnRH injection in IU/L: 13.1 ± 1.1; pre-saline, 13.4 ± 1.1 pre- kisspeptin-54; p=ns) (Figure 4-8A). Furthermore 24 hours following cessation of twice- daily saline or kisspeptin-54 injections, no significant difference in pituitary sensitivity was observed following GnRH administration between saline or kisspeptin-54 treatment (mean maximal LH increase during first 2 hours post-GnRH injection in IU/L: 39.9 ± 8.9; post- saline, 41.0 ± 16.4 post-kisspeptin-54; p=ns) (Figure 4-8B).

The expected physiological increase in pituitary sensitivity to GnRH during menstrual day

15 versus menstrual day 1 was observed in healthy female subjects, whether receiving saline or kisspeptin treatment (Figure 4-8C).

158

Figure 4-8: Pituitary responsiveness in healthy women receiving twice-daily injections of saline or kisspeptin-54.

A-B: Serum LH following 100 mcg intravenous bolus injection of GnRH at time 0 minutes in women on (A) menstrual day 1 prior to commencing saline or kisspeptin treatment and (B) menstrual day 15 of the study protocol 24 hours after stopping saline or kisspeptin treatment). C: Summary bar graph comparing mean AUC LH increase following GnRH administration in women before/after saline or kisspeptin-54 treatment. Data are presented as mean ± SEM. *, p<0.05.

159

4.5 Discussion

Previous studies have demonstrated that exogenous kisspeptin acutely stimulates gonadotrophin secretion in women (Dhillo et al. 2007, Chan et al. 2012, Jayasena et al.

2013). However the pharmacological effects of chronic kisspeptin exposure to healthy women with regular menstrual cycles have not been previously studied. Chronic kisspeptin administration causes profound tachyphylaxis in male monkeys and in women with functional HA (Seminara et al. 2006, Jayasena et al. 2009, Jayasena et al. 2010). In this study, contrary to my hypothesis, I present novel data suggesting that menstrual cyclicity persists in healthy women following twice-daily kisspeptin-54 treatment during the follicular phase of the menstrual cycle and furthermore the menstrual cycle is advanced.

Chronic kisspeptin administration advances the menstrual cycle in healthy women.

In the current study, kisspeptin treatment was associated with an advanced timing of serum progesterone increase, and onset of menstrual bleeding by approximately 2 days when compared with saline. Animal studies demonstrate that kisspeptin stimulates gonadotrophin secretion in a GnRH-dependent manner, since its action is abolished by

GnRH antagonist (Tovar et al. 2006). In keeping with my study exogenous GnRH has previously been shown to be sufficient to trigger ovulation (Knobil et al. 1980). Further studies with daily follicle morphology assessment, and daily serum LH measurement

(rather than every 2-6 days as in the current study) would be required to confirm these findings. However this study data raises the possibility that kisspeptin-54 administration may advance the onset of ovulation in healthy women by stimulating endogenous hypothalamic GnRH secretion.

In addition to stimulating GnRH, kisspeptin itself is implicated in generation of the LH surge needed for ovulation. In rodents, a subpopulation of hypothalamic anteroventral periventricular nucleus (AVPV) kisspeptin neurons are implicated in generating the LH

160 surge needed for ovulation, and are positively rather than negatively regulated by oestradiol (Smith et al. 2006, Clarkson et al. 2008, Herbison 2008). Central administration of a monoclonal antibody to kisspeptin is sufficient to block ovulation in rats (Kinoshita et al. 2005). Furthermore, administration of kisspeptin-10 has been shown to stimulate ovulation in the musk shrew (Inoue et al. 2011), rat (Matsui et al. 2004) and sheep (Caraty et al. 2007). Humans have no anatomical equivalent of the AVPV although there is data to suggest an RP3V kisspeptin neurone population that may be positively mediated by oestradiol and is only present in women (Hrabovszky et al. 2010). It is possible that exogenous kisspeptin-54 may have advanced the onset of the luteal phase in these female subjects by increasing kisspeptin signalling in this subpopulation of hypothalamic kisspeptin neurons which stimulate the LH surge.

Several lines of evidence suggest that in humans unlike in lower species, a change in pituitary responsiveness to GnRH rather than an actual GnRH surge is responsible for the

LH surge (Adams et al. 1994, Hall et al. 1994). It is therefore also possible that kisspeptin-

54 advances ovulation in women by increasing the prevailing levels of oestradiol, which are needed to trigger the LH surge at a pituitary level. Put simply, there is earlier induction of the oestrogen positive feedback event.

Chronic kisspeptin administration does not cause marked tachyphylaxis in healthy women.

Comparison of my data with other clinical studies of kisspeptin administration is complicated by the investigation of its two peptide forms (kisspeptin-54 and -10), and sexual dimorphism of the effects of kisspeptin-10. Intravenous bolus administration consistently stimulates LH in men (George et al. 2011, Jayasena et al. 2011, George et al.

2013), but women are much less sensitive to the effects of kisspeptin-10 during the follicular phase of menstrual cycle when compared with the pre-ovulatory phase as

161 demonstrated in Chapter 3 and confirmed by a subsequent study (Chan et al. 2012).

Although kisspeptin-54 stimulates LH in both sexes, its effects have never been compared directly between men and women. Consistent with our findings in healthy women, George et al. recently observed that 22.5 hour intravenous infusion of the shorter form of kisspeptin, kisspeptin-10, stimulates LH secretion without apparent tachyphylaxis in healthy men (George et al. 2011). However in contrast to the current findings, twice daily kisspeptin-54 treatment (at the same dose as in my study, 6.4 nmol/kg) has been shown to rapidly cause tachyphylaxis in women with hypothalamic amenorrhoea. Women with hypothalamic amenorrhoea are acutely 4-fold more sensitive to the effects of kisspeptin-

54 when compared with healthy women in the follicular phase of their menstrual cycle. A higher dose of kisspeptin-54 may therefore be necessary to achieve tachyphylaxis in the healthy female subjects in my study (Dhillo et al. 2005, Jayasena et al. 2009).

Nevertheless, my data suggest that menstrual cyclicity is able to persist in healthy women using the currently tested regimen of follicular phase kisspeptin-54 treatment. New data has emerged suggesting that chronic subcutaneous administration of kisspeptin analogues causes tachyphylaxis in healthy men (and rodents) (Matsui et al. 2012,

Maclean et al. 2014). It would therefore be interesting to determine if tachyphlyaxis to kisspeptin-54 is also sexually dimorphic, or dependent on the dose and precise form of kisspeptin receptor agonist administered.

It is interesting to note that while intravenous bolus injection of the shorter kisspeptin fragment, kisspeptin-10, acutely stimulates LH secretion within an hour of administration, in this study subcutaneous injection of kisspeptin-54 acutely stimulated peak LH secretion

3-4 hours following administration despite a rapid elevation of kisspeptin IR within minutes of administration. On the first injection day, the increase in serum LH 1 hour following kisspeptin-54 injection was 3-fold lower when compared with the peak increase in serum

LH following kisspeptin-54 injection (Figure 4-4). Furthermore on the 4th and final injection

162 days, there was negligible stimulation of LH secretion 1 hour following kisspeptin-54 injection; however serum LH later increased to peak levels 3-4 hours after kisspeptin injection. By comparison, Chan et al. observed that kisspeptin-10 stimulated peak LH secretion within an hour of commencing administration in healthy men (Chan et al. 2011).

It is possible that these data suggest differences between the biological actions of kisspeptin-10 and -54. For instance, one could speculate that kisspeptin-10 stimulates a readily releasable pool of GnRH stored within nerve terminals of the median eminence

(d'Anglemont de Tassigny et al. 2008, d'Anglemont de Tassigny et al. 2010), whereas kisspeptin-54 might act more proximally in the GnRH neurone to stimulate de novo GnRH synthesis. Alternatively, it is possible kisspeptin-54 might require breakdown into a smaller kisspeptin fragment before becoming biologically active. However against this, liquid chromatography has only identified kisspeptin-54 in the human circulation acutely following kisspeptin-54 administration (Dhillo et al. 2005).

It is worthwhile appraising the evidence that exogenous kisspeptin stimulates basal and pulsatile GnRH secretion. Although kisspeptin stimulates rat pituitary gonadotrophin secretion in vitro (Gutierrez-Pascual et al. 2007), animal studies using GnRH antagonists support the view that exogenous kisspeptin stimulates basal LH secretion through a

GnRH-dependent mechanism (Shahab et al. 2005), which is possibly mediated through

GnRH nerve terminals at the median eminence (d'Anglemont de Tassigny et al. 2008, d'Anglemont de Tassigny et al. 2010). However it is less clear whether exogenous kisspeptin stimulates endogenous GnRH pulsatility. Sustained exposure to exogenous kisspeptin-10 stimulates pulsatile LH secretion for several hours in healthy men (George et al. 2011), and patients with inactivating mutations of the NKB signalling pathway

(Young et al. 2013). However since these studies merely measure LH (rather than GnRH) pulsatility, it is not possible to exclude that these effects reflect increasing circulating oestrogen and pituitary responsiveness to unchanged, endogenous GnRH pulsatility.

Chan and colleagues have demonstrated that intravenous bolus kisspeptin-10 increased

163 the time to the next endogenous LH pulse in healthy men (Chan et al. 2011); this represents the only current data suggesting that exogenous kisspeptin can directly modulate endogenous GnRH pulsatility in humans (by increasing the latency period to next LH/ GnRH pulse).

Mean peak levels of LH during the menstrual cycle were lower during kisspeptin treatment when compared with saline treatment. Furthermore it is important to recognize that kisspeptin-54 did not acutely stimulate significant LH secretion during the final day of twice-daily kisspeptin-54 treatment. I therefore cannot exclude that kisspeptin-54 treatment caused partial tachyphylaxis in healthy female volunteers in this protocol.

Alternatively, the short baseline sampling period involving just two blood samples may have reduced our power to detect stimulation of LH following kisspeptin-54 injection on day 14. LH levels were highly variable during menstrual day 14; this may have been caused by some but not all subject experiencing an LH surge, and by the potential effect of kisspeptin-54 to advance the onset of LH surge by a different degree in different subjects.

It is noteworthy that all subjects had peak progesterone levels ≥21mmol/l following kisspeptin-54 treatment, which is highly suggestive of successful ovulation. Furthermore a corpus luteum was observed in all five subjects during the month of kisspeptin-54 treatment (Table 4-2). This data therefore suggest that all five subjects had ovulatory cycles during kisspeptin treatment. Further studies are required to determine if menstrual cycles during kisspeptin therapy have subtle physiological differences when compared with natural menstrual cycles, and to determine if ovulation is observed at a different frequency during kisspeptin treatment when compared with placebo treatment.

164

Limitations

It is important to recognise that the current study is based on observations within a small number of subjects, so may have insufficient statistical power to reveal more subtle effects of kisspeptin-54 administration on menstrual cyclicity or LH pulsatility. Furthermore, gonadotrophin response to kisspeptin treatment is known to alter with the phase of menstrual cycle in healthy women (Dhillo et al. 2007, Chan et al. 2012). A larger study is needed to confirm my findings and determine what factors influence the variability in response of subjects to kisspeptin treatment. In addition, it is possible that chronic kisspeptin-54 treatment might have effects which last beyond the study period and so a longer protocol may detect any longer term effects.

I designed a one-way crossover protocol in which subjects self-administered saline during the first month followed by kisspeptin-54 during the second month. It therefore remains possible that an order effect might have contributed to these results. However it is noteworthy that LH pulse secretory mass was marginally lower at the start of the second month when compared with LH pulse secretory mass at the start of month one (when greater stress levels would be expected). Furthermore, subjects had the same menstrual cycle length prior to commencing the study when compared with menstrual cycle length during saline administration (month 1).

Another limitation relates to the inability to statistically confirm LH stimulation on the last day of injections (day 14; Figure 4-3C). This relates to the high level of inter-individual variation as reflected by the wide error bars on Figure 4-3C. This is likely explained by the fact that having completed a full week of kisspeptin injections, the menstrual cycles may have been advanced to different extents and as it is known that LH response to kisspeptin varies according to phase of the menstrual cycle (Dhillo et al. 2007). In addition there are also wide error bars on the saline data as some subjects were at different points of their

LH surge representing a normal variation in cycle timings between different women.

165

Summary

In summary, my study has important pharmacological implications; menstrual cyclicity persists in healthy women during a treatment regime of kisspeptin-54 previously demonstrated to cause tachyphylaxis in women with HA. Furthermore, kisspeptin-54 treatment is associated with an advanced timing of the luteal phase of menstrual cycle when compared with placebo. Further studies are required to determine if kisspeptin-54 therapy could have potential to treat patients with anovulatory reproductive disorders.

4.6 Future Work and Therapeutic Impliations

Further studies with larger cohorts, longer durations and with daily blood/ultrasound testing would be of interest in confirming my findings and examining for any subtle effects on the menstrual cycle that were not detected here.

Chronic kisspeptin administration has been implicated as a potential novel therapy for inhibiting reproductive hormone secretion in contraception or the treatment of hormone sensitive cancer; my data suggest that kisspeptin-54 may have limited therapeutic potential in this regard, at least in women at the dose of kisspeptin tested here. However the observation that kisspeptin may advance the menstrual cycle, raises the possibility that women with certain forms of infertility could be treated with kisspeptin. A recent study suggests that kisspeptin is sufficient to restore ovulation in a mouse model of anovulatory hyperprolactinaemia (Sonigo et al. 2012). Further studies are required to determine if kisspeptin-54 treatment could be used to stimulate ovulation in women with anovulatory reproductive disorders other than HA.

Traditional ovulatory drugs such as clomiphene citrate have low pregnancy rates (Kousta et al. 1997). Although efficacious, in vitro fertilisation (IVF) confers a risk of ovarian hyperstimulation (OHSS) (Hart and Norman 2006). Kisspeptin can stimulate endogenous,

166 rather than pharmacological, levels of reproductive hormone secretion. Kisspeptin might therefore offer a fertility treatment with lower risk of OHSS when compared with IVF.

Indeed, together with colleagues, I have recently demonstrated that a single subcutaneous injection of kisspeptin-54 is able to trigger egg maturation in IVF and result in successful embryo transfer and live births (Jayasena et al. 2014). Further work is needed however to establish if kisspeptin-54 has a lower risk of OHSS compared to conventional triggers such as βHCG. In this regard, it would be useful to study populations with a high risk of OHSS such as women with PCOS.

Given the sexual dimorphism in response to kisspeptin-10 observed in Chapter 3, another interesting future study may directly compare acute kisspeptin-54 and chronic kisspeptin-

54 administration (in a similar protocol to this chapter) between the sexes. This would have important clinical implications for the selection of appropriate doses of kisspeptin-54 depending on clinical indication (i.e. to stimulate or down-regulate the reproductive axis).

167

Chapter 5 ______

General Discussion

168

5.1 General Discussion

The kisspeptin system was first identified in cancer studies in 1996 (Lee et al. 1996), however it was not until 2003 that a critical role for this kisspeptin system was identified in reproduction. Since then in humans, inactivating mutations of the KISS1R (de Roux et al.

2003, Seminara et al. 2003) and KISS1 (Topaloglu et al. 2012) have been identified resulting in hypogonadotrophic hypogonadism and so a failure to go through puberty with resultant infertility. Conversely activating mutations of the KISS1R (Teles et al. 2008) and

KISS1 (Silveira et al. 2010) gene have also been described which lead to central precocious (early) puberty. These findings together confirm the pivotal role for the kisspeptin system in mammalian reproduction.

This pivotal role has been exploited by administering kisspeptin, the peptide product of the

KISS1 gene, to interrogate the physiological mechanisms of reproduction. Kisspeptin administration increases LH secretion but this effect is blocked by the pre-administration of a GnRH antagonist (Irwig et al. 2004). Together with other studies (detailed in the introduction), this provides evidence that kisspeptin stimulates gonadotrophin release via stimulation of GnRH neurones in the hypothalamus.

As well as providing physiological insights into the inner workings of the reproductive axis, kisspeptin administration is now emerging as a potential clinical treatment for reproductive disorders in men and women. To this end there remain several unanswered questions that this thesis has attempted to address.

I applied recent advances in neuroimaging to visualise manganese uptake (as a surrogate for neuronal activity) in pre-selected brain regions following intraperitoneal kisspeptin administration in adult male mice. This technique has previously been validated in the

169 interrogation of gut hormones and hypothalamic neuronal activation (Chaudhri et al. 2006,

Kuo et al. 2007, Parkinson et al. 2009).

The major finding of decreased neuronal activity within the amygdala was fascinating especially considering the established inhibitory effects that the amygdala normally exerts on the reproductive axis (Kluver and Bucy 1997). Although functional neuroimaging does not permit the precise identification of the pathways and neuronal sub-types involved, this data suggests that further study in this region of the brain may be fruitful (see further work).

In addition, I demonstrated that peripheral kisspeptin adminsitration does not affect ARC or AVPV neuronal activity but decreases POA (the site of the majority of GnRH neurones) neuronal activity. These findings are consitent with the literature and also suggest that the inhibition seen in the POA and the amygdala may represent the ‘inhibition of inhibitory neurones’, with GABAergic neurones the prime candidates for the inhibitory neurones.

The recent studies examining GABA signalling and kisspeptin pay testament to this possibility (Zhang et al. 2009, Herbison and Moenter 2011, Kokay et al. 2011, Alreja 2013,

Navarro 2013, Di Giorgio et al. 2014, Keshavarzi et al. 2014).

Almost one thousand papers have been published on the kisspeptin system, very few to date have examined the roles in non-hypothalamic neuronal populations. Determining these roles as well as the pharmacological effects of kispeptin administration in these areas is of paramount importance for our understanding of reproducitve physiology and the future development of kisspeptin as a clinical treatment for reproductive pathologies in humans.

Given the different isoforms of human kisspeptin (-10, -13, -14, -54) and the animal data highlighting sexual dimorphism in the neuroanatomical layout of the kisspeptin system; I proceeded to explore this in humans.

170

Studies in humans can be challenging to complete and bring with them a raft of safety, ethical and regulatory issues. However, this is offset by several factors. Only human studies can truly identify the physiological effects of kisspeptin in humans. Furthermore, only human studies can identify safety issues regarding human kisspeptin administration.

In addition humans can undergo various dosing regimes through various routes as well as having frequent blood sampling which can provide data on reproductive hormone pulsatility. I exploited these advantages to address my research aims and hypotheses.

The data from my study in Chapter 3 provides evidence for a sexual dimorphism in gonadotrophin response to peripheral administration of kisspeptin-10. This was later confirmed by another research group (Chan et al. 2012). While healthy men respond to kisspeptin-10 with robust LH secretion, healthy women only have detectable LH increases during the pre-ovulatory phase of their menstrual cycle. Although there are several possible explanations (detailed in Section 3.5), the most likely explanation is that the higher background oestradiol levels of the pre-ovulatory phase increase GnRH neuronal response to exogenous kisspeptin (Tonsfeldt et al. 2011). In addition to being interesting from a therapeutic point of view, this finding provides evidence that kisspeptin signalling is altered during the pre-ovulatory phase in women and this may play a role in generating the physiological LH surge.

Another interesting finding relates to the higher potency of intravenous kisspeptin-54 compared to kisspeptin-10 in healthy women and this can be largely explained by differences in the pharmacokinetics between the two isoforms. When kisspeptin-10 and kisspeptin-54 are applied directly to cultured GnRH cells in vitro, they have similar activity and potency (Kotani et al. 2001). However, in vivo experiments in rodents have demonstrated that kisspeptin-54 has a longer onset and duration of action compared to kisspeptin-10 when given at the same molar concentration and route (Tovar et al. 2006,

Mikkelsen et al. 2009). These differences between kisspeptin-10 and kisspeptin-54 are likely explained by differences in kisspeptin pharmacokinetics in vivo.

171

My study demonstrated that kisspeptin-10 has an in vivo half-life of 4 minutes in healthy men and women (during all menstrual phases). Interestingly kisspeptin-10 added to human plasma degrades with a half-life of 55 seconds in vitro at 370C (Chan et al. 2011).

The difference in calculated half-life for kisspeptin-10 between these two studies may relate to methodological differences. The radioimmunoassay used in my study as well as detecting the different human isoforms, may also detect the breakdown products such as the previously observed 9-amino acid breakdown product (Chan et al. 2011) and so result in a longer apparent half-life. Another explanation is the difference between the in vivo and in vitro setting. In my in vivo setting, there may exist factors that prolong the half-life of kisspeptin-10 such as binding proteins or distribution outside the blood. Either way, kisspeptin-10 has a half-life between under 5 minutes.

The half-life of kisspeptin-54 in humans has previously been calculated as 28 minutes

(Dhillo et al. 2005) and is therefore significantly longer than kisspeptin-10. The slower decay of kisspeptin-54 provides the most likely explanation for the longer duration of action observed with kisspeptin-54 compared to kisspeptin-10. Furthermore in my study

(Chapter 4), the finding of continued high plasma kisspeptin IR 8 hours after subcutaneous kisspeptin-54 administration suggests that kisspeptin-54 can behave as a sustained-release formulation when administered subcutaneously.

However this does not fully explain why kisspeptin-54 elicits a more robust LH response compared to kisspeptin-10 after intravenous bolus administration in women, as continuous intravenous infusion of kisspeptin-10 to mimic the longer plasma kisspeptin-IR observed after intravenous bolus kisspeptin-54 still does not elicit an LH response (Jayasena et al.

2011). Further studies on the pharmacokinetics and pharmacodynamics of these two kisspeptin isoforms are required to explore these differences in further detail and provide information of therapeutic relevance.

172

Following on from this study, I proceeded to investigate subcutaneous kisspeptin-54 administration in further detail. This final study in my thesis, provided very encouraging results for the use of chronic subcutaneous kisspeptin injections in women. Not only were women able to self-administer injections reliably in an unsupervised non-clinical environment but chronic kisspeptin administration also brought forward their menstrual cycle demonstrating an activation of the reproductive axis. This was against my hypothesis, as I had hypothesised that chronic kisspeptin administration would result in tachyphylaxis simlar to that observed in women with hypothalamic amenorrhoea in a similar protocol (Jayasena et al. 2009). Furthermore, given that chronic kisspeptin exposure can also result in down-regulation of the reproductive axis in primates

(Seminara et al. 2006), it was interesting to observe that twice daily injections of kisspeptin-54 for one week brought forward rather than delayed the menstrual cycle.

These data firstly suggest that patients may be able to self-administer kisspeptin-54 as part of a treatment to upregulate their reproductive axis. Target pathologies may for example include PCOS with the aim of regulating menstrual cyclicity, or restoration of ovarian cyclicity in patients with hyperprolactinaemia intolerant of standard dopamine agonist treatments.

Given the spectrum of differing response to the same dose of chronic kisspeptin administration between healthy women here and in women with hypothalamic amenorrhoea in the previous study, there is likely to be differences in response to kisspeptin dependent on dose. Higher doses may indeed result in tachyphylaxis or a down-regulation of the reproductive axis. Putting this together with the sexual dimorphism and differences between kisspeptin-10 and kisspeptin-54 observed in chapter 2, suggest that future clinical treatments using kisspeptin may exploit the different effects observed depending on kisspeptin isoform, route of administration, duration of administration, gender of patient, reproductive pathology of patient and background sex steroid milieu.

173

In summary, I have investigated the effects of kisspetin in rodents and humans with different aims. I provide evidence that peripheral kisspeptin adminsitration modulates neuronal activity outside the hypothalamus, a finding that suggests further exploration of the extra-hypothalamic effects of kissepeptin. I also provide evidence of a sexual dimorphism in response to kisspeptin administration in humans as well as investigating the effects of kisspeptin-10 compared to kisspeptin-54 and the acute and chronic effects of kisspeptin-54 in healthy women. This body of work provides some answers to several questions relating to kisspeptin physiology and the development of kisspeptin therapeutics, however there remain several questons for future study.

174

5.2 Future Work

I have finished each chapter with a separate section on future work, however it is perhaps appropriate to end the thesis with an outline of future directions for the study of kisspeptin based on my work.

Neuroimaging is a powerful and evolving tool in neuroscience. In this thesis I employed

MEMRI to determine neronal activity in pre-selected regions of interest following kisspeptin administration to male rodents. The finding of decreased neuronal activity in the amygdala after kisspeptin administration may suggest a role for kisspeptin signalling in social and reproductive behaviours attributable to the amygdala which could form the basis of future study. In addition confirming my findings using other techniques such as c-

Fos expression may be of value in the validation of MEMRI use in kisspeptin neuroscience.

While it would be of interest to repeat these MEMRI experiments in female rodents in different oestrous phases given that amygdala kisspeptin signalling is influenced by sex steroids (Kim et al. 2011), it would also be of great interest to explore other brain areas such as the hippocampus which exhibits marked Kiss1r expression and has functions related to memory. It would also be worthwhile assessing if similar neuronal activity is observed with central kisspeptin administration given that peripherally administered kisspeptin is not thought to cross the blood-brain-barrier itself (Herde et al. 2011).

My clinical findings regarding different kisspeptin isoforms, gender effects and acute versus chronic administration suggests many future directions for study and clinical applications which I have detailed at the end of each chapter. Given that kisspeptin sits at the apex of the reproductive axis, it may have several advantages compared to treatments lower down the axis (GnRH injection, LH/FSH injections, Oestradiol/Testosterone treatments). These other treatments typically result in supraphysiological levels of downstream hormones. For example the ovarian hyperstimulation syndrome (OHSS)

175 observed during IVF treatment is thought to result from the use of hCG to trigger egg maturation. hCG acts directly on LH receptors in the ovary and has a much longer duration of action than endoenous LH surge which causes egg maturation in a natural menstrual cycle. By contrast, kisspeptin as it acts at the top of the axis may be able to stimulate (depending on dose) a more physiological magnitude of downstream hormones by stimulating the release of an endogenous pool of GnRH. This needs further study, and

I am participating in studies administering kisspeptin to women with subfertility due to

PCOS at high risk of OHSS to attempt to answer this research question.

It is clear that even 11 years after the identification of kisspeptin as a crucial reproductive neuropeptide there remains much work to be done, fuelled by the exciting results observed to date.

176

Appendix 1: Principles of Radioimmunoassay ______

Radioimmunoassay (RIA) is a technique used to measure the concentration of an unknown quantity of peptide (in this case kisspeptin) in plasma.

Unlabelled antigen on the peptide in the plasma sample and radioloabelled (using I-125) antigen compete for an antibody binding site:

Ag* + Ab + Ag ↔ Ag*Ab + AgAb

Ag*= radioloabelled antigen

Ag= unlabelled antigen (from plasma sample)

Ab= antibody

The specificity of the antibody is dependent on its ability to recognise structural features of the antigen on the peptide and bind to it. The antibody for kisspeptin in my assay was produced by immunising rabbits with kisspeptin.

In the RIA, both the radioloabelled peptide and antibody are at specific, constant concentrations, with the concentration of antibody being limiting. The known amount of radiolabelled antigen will compete with unlabelled antigen (in the plasma sample) for antibody binding, with the amount of radiolabelled antigen bound to the antibody being inversely proportional to the amount of unlabelled antigen in the plasma sample being assayed.

A ‘standard curve’ is produced at the same time as assay of samples. This ‘standard curve’ uses known concentrations of pure peptide to calculate the percentage of labelled

177 peptide bound for each concentration of unlabelled pure peptide. Separation of the unbound radiolabelled antigen from the radiolabelled antigen-antibody complex and counting the proportion of radiolabel present in the two fractions enables direct measurement of the amount of unlabelled antigen that has bound to the antibody. By directly referencing the ‘standard curve’, the unknown concentrations of peptide in the plasma sample can be deduced by interpolation.

To distinguish antibody-bound antigen from free antigen, and to determine the distribution of radioactive antigen between the two fractions, the method of charcoal separation is used as detailed below.

Dextran is added to a charcoal suspension to block the larger holes in porous charcoal.

The suspension is then added to the RIA tubes, where it traps the free (unbound) radiolabelled antigen. The tubes are then centrifuged and the supernatant (containing antibody-antigen complex) and charcoal pellet (free radiolabelled antigen) are separated by aspiration with a pipette. The bound and free labels are counted in a gamma counter.

All samples are added in duplicate. ‘Non-specific binding’ tubes (without antibody) are added at the beginning of the assay rack. Tubes containing half and twice the standard volume of added labelled antigen are also included so as to confirm label quality. Tubes with no plasma sample (‘zero’ tubes) are placed at regular intervals throughout the assay, and ‘standard curves’ (containing known quantities of unlabelled antigen) are placed at the beginning and end of each assay, in order to calibrate the assay. The analytical sensitivity of the assay is determined by the smallest change in hormone concentration that can be reliably detected. This is governed by the steepness of the standard curve, and the error of a known value on the curve. The analytical sensitivity is calculated as two standard deviations from the mean concentration of antigen measured in the zero standards (which contain no unlabelled peptide).

178

Appendix 2: Principles of xMAP Immunoassay ______xMAP technology is used to perform a variety of bioassays including immunoassays on the surface of fluorescent-coded magnetic beads.

Proprietary techniques are used to internally colour-code microspheres with two fluorescent dyes. Through precise concentrations of these dyes, 100 distinctly coloured bead sets can be created, each of which is coated with a specific capture antibody, in this case to mouse LH.

After an analyte from a test sample is captured by the bead, a biotinylated detection antibody is introduced. The reaction mixture is then incubated with Streptavidin-PE conjugate, the reporter molecule, to complete the reaction on the surface of each microsphere. The microspheres are allowed to pass rapidly through a laser which excites the internal dyes marking the microsphere set. A second laser excites PE, the fluorescent dye on the reporter molecule. Finally, high-speed digital-signal processors identify each individual microsphere and quantify the result of its bioassay based on fluorescent reporter signals.

See Section 2.3.3.2 for experimental protocol.

179

References ______

Adachi, S., S. Yamada, Y. Takatsu, H. Matsui, M. Kinoshita, et al. (2007). Involvement of anteroventral periventricular metastin/kisspeptin neurons in estrogen positive feedback action on luteinizing hormone release in female rats. J Reprod Dev 53(2): 367-378.

Adams, J. M., A. E. Taylor, D. A. Schoenfeld, W. F. Crowley, Jr. and J. E. Hall (1994). The midcycle gonadotropin surge in normal women occurs in the face of an unchanging gonadotropin-releasing hormone pulse frequency. J Clin Endocrinol Metab 79(3): 858- 864.

Aitken, R. J., M. A. Baker, G. F. Doncel, M. M. Matzuk, C. K. Mauck, et al. (2008). As the world grows: contraception in the 21st century. J Clin Invest 118(4): 1330-1343.

Alreja, M. (2013). Electrophysiology of kisspeptin neurons. Adv Exp Med Biol 784: 349- 362.

Anastasovska, J., T. Arora, G. J. Sanchez Canon, J. R. Parkinson, K. Touhy, et al. (2012). Fermentable carbohydrate alters hypothalamic neuronal activity and protects against the obesogenic environment. Obesity (Silver Spring) 20(5): 1016-1023.

Anthony, K., L. J. Reed, J. T. Dunn, E. Bingham, D. Hopkins, et al. (2006). Attenuation of insulin-evoked responses in brain networks controlling appetite and reward in insulin resistance: the cerebral basis for impaired control of food intake in metabolic syndrome? Diabetes 55(11): 2986-2992.

Aoki, I., C. Tanaka, T. Takegami, T. Ebisu, M. Umeda, et al. (2002). Dynamic activity- induced manganese-dependent contrast magnetic resonance imaging (DAIM MRI). Magn Reson Med 48(6): 927-933.

Arai, A. C. (2009). The role of kisspeptin and GPR54 in the hippocampus. Peptides 30(1): 16-25.

Arai, A. C., Y. F. Xia, E. Suzuki, M. Kessler, O. Civelli, et al. (2005). Cancer metastasis- suppressing peptide metastin upregulates excitatory synaptic transmission in hippocampal dentate granule cells. J Neurophysiol 94(5): 3648-3652.

Arora, T., R. L. Loo, J. Anastasovska, G. R. Gibson, K. M. Tuohy, et al. (2012). Differential effects of two fermentable carbohydrates on central appetite regulation and body composition. PLoS One 7(8): e43263.

Arreguin-Arevalo, J. A., C. A. Lents, T. A. Farmerie, T. M. Nett and C. M. Clay (2007). KiSS-1 peptide induces release of LH by a direct effect on the hypothalamus of ovariectomized ewes. Anim Reprod Sci 101(3-4): 265-275.

180

Bagga, N., G. S. Chhina, V. M. Kumar and B. Singh (1984). Cholinergic activation of medial preoptic area by amygdala for ovulation in rat. Physiol Behav 32(1): 45-48.

Bar-Sela, M. and V. Critchlow (1966). Delayed puberty following electrical stimulation of amygdala in female rats. Am J Physiol 211(5): 1103-1107.

Barry, J. (1979). Immunohistochemistry of luteinizing hormone-releasing hormone- producing neurons of the vertebrates. Int Rev Cytol 60: 179-221.

Belchetz, P. E., T. M. Plant, Y. Nakai, E. J. Keogh and E. Knobil (1978). Hypophysial responses to continuous and intermittent delivery of hypopthalamic gonadotropin- releasing hormone. Science 202(4368): 631-633.

Bowe, J. E., V. L. Foot, S. A. Amiel, G. C. Huang, M. W. Lamb, et al. (2012). GPR54 peptide agonists stimulate insulin secretion from murine, porcine and human islets. Islets 4(1): 20-23.

Bowe, J. E., A. J. King, J. S. Kinsey-Jones, V. L. Foot, X. F. Li, et al. (2009). Kisspeptin stimulation of insulin secretion: mechanisms of action in mouse islets and rats. Diabetologia 52(5): 855-862.

Brailoiu, G. C., S. L. Dun, M. Ohsawa, D. Yin, J. Yang, et al. (2005). KiSS-1 expression and metastin-like immunoreactivity in the rat brain. J Comp Neurol 481(3): 314-329.

Brown, R. E., S. A. Imran, E. Ur and M. Wilkinson (2008). KiSS-1 mRNA in adipose tissue is regulated by sex hormones and food intake. Mol Cell Endocrinol 281(1-2): 64-72.

Bullitt, E. (1990). Expression of c-fos-like protein as a marker for neuronal activity following noxious stimulation in the rat. J Comp Neurol 296(4): 517-530.

Canteras, N. S., R. B. Simerly and L. W. Swanson (1995). Organization of projections from the medial nucleus of the amygdala: a PHAL study in the rat. J Comp Neurol 360(2): 213-245.

Caraty, A., J. T. Smith, D. Lomet, S. Ben Said, A. Morrissey, et al. (2007). Kisspeptin synchronizes preovulatory surges in cyclical ewes and causes ovulation in seasonally acyclic ewes. Endocrinology 148(11): 5258-5267.

Carmel, P. W., S. Araki and M. Ferin (1976). 'Pituitary stalk portal blood collection in rhesus monkeys: evidence for pulsatile release of gonadotropin-releasing hormone (GnRH)'. Endocrinology 99(1): 243-248.

181

Castellano, J. M., V. M. Navarro, R. Fernandez-Fernandez, J. Roa, E. Vigo, et al. (2006). Expression of hypothalamic KiSS-1 system and rescue of defective gonadotropic responses by kisspeptin in streptozotocin-induced diabetic male rats. Diabetes 55(9): 2602-2610.

Cerrato, F., J. Shagoury, M. Kralickova, A. Dwyer, J. Falardeau, et al. (2006). Coding sequence analysis of GNRHR and GPR54 in patients with congenital and adult-onset forms of hypogonadotropic hypogonadism. Eur J Endocrinol 155 Suppl 1: S3-S10.

Chan, Y. M., J. P. Butler, N. E. Pinnell, F. P. Pralong, W. F. Crowley, Jr., et al. (2011). Kisspeptin resets the hypothalamic GnRH clock in men. J Clin Endocrinol Metab 96(6): E908-915.

Chan, Y. M., J. P. Butler, V. F. Sidhoum, N. E. Pinnell and S. B. Seminara (2012). Kisspeptin administration to women: a window into endogenous kisspeptin secretion and GnRH responsiveness across the menstrual cycle. J Clin Endocrinol Metab 97(8): E1458- 1467.

Chand, D. and D. A. Lovejoy (2011). Stress and reproduction: controversies and challenges. Gen Comp Endocrinol 171(3): 253-257.

Chang, J. P., A. Mar, M. Wlasichuk and A. O. Wong (2012). Kisspeptin-1 directly stimulates LH and GH secretion from goldfish pituitary cells in a Ca(2+)-dependent manner. Gen Comp Endocrinol 179(1): 38-46.

Chaudhri, O. B., J. R. Parkinson, Y. T. Kuo, M. R. Druce, A. H. Herlihy, et al. (2006). Differential hypothalamic neuronal activation following peripheral injection of GLP-1 and oxyntomodulin in mice detected by manganese-enhanced magnetic resonance imaging. Biochem Biophys Res Commun 350(2): 298-306.

Chuang, K. H., J. H. Lee, A. C. Silva, L. Belluscio and A. P. Koretsky (2009). Manganese enhanced MRI reveals functional circuitry in response to odorant stimuli. Neuroimage 44(2): 363-372.

Clarke, I. J. and J. T. Cummins (1982). The temporal relationship between gonadotropin releasing hormone (GnRH) and luteinizing hormone (LH) secretion in ovariectomized ewes. Endocrinology 111(5): 1737-1739.

Clarkson, J., X. d'Anglemont de Tassigny, W. H. Colledge, A. Caraty and A. E. Herbison (2009). Distribution of kisspeptin neurones in the adult female mouse brain. J Neuroendocrinol 21(8): 673-682.

182

Clarkson, J., X. d'Anglemont de Tassigny, A. S. Moreno, W. H. Colledge and A. E. Herbison (2008). Kisspeptin-GPR54 signaling is essential for preovulatory gonadotropin- releasing hormone neuron activation and the luteinizing hormone surge. J Neurosci 28(35): 8691-8697.

Clarkson, J. and A. E. Herbison (2006). Postnatal development of kisspeptin neurons in mouse hypothalamus; sexual dimorphism and projections to gonadotropin-releasing hormone neurons. Endocrinology 147(12): 5817-5825.

Clements, M. K., T. P. McDonald, R. Wang, G. Xie, B. F. O'Dowd, et al. (2001). FMRFamide-related neuropeptides are agonists of the orphan G-protein-coupled receptor GPR54. Biochem Biophys Res Commun 284(5): 1189-1193.

Constantin, S., C. S. Caligioni, S. Stojilkovic and S. Wray (2009). Kisspeptin-10 facilitates a plasma membrane-driven calcium oscillator in gonadotropin-releasing hormone-1 neurons. Endocrinology 150(3): 1400-1412.

Couldrey, C., M. B. Carlton, P. M. Nolan, W. H. Colledge and M. J. Evans (1999). A retroviral gene trap insertion into the histone 3.3A gene causes partial neonatal lethality, stunted growth, neuromuscular deficits and male sub-fertility in transgenic mice. Hum Mol Genet 8(13): 2489-2495.

Curtis, A. E., J. H. Cooke, J. E. Baxter, J. R. Parkinson, A. Bataveljic, et al. (2010). A kisspeptin-10 analog with greater in vivo bioactivity than kisspeptin-10. Am J Physiol Endocrinol Metab 298(2): E296-303.

d'Anglemont de Tassigny, X., K. J. Ackroyd, E. E. Chatzidaki and W. H. Colledge (2010). Kisspeptin signaling is required for peripheral but not central stimulation of gonadotropin- releasing hormone neurons by NMDA. J Neurosci 30(25): 8581-8590.

d'Anglemont de Tassigny, X., L. A. Fagg, M. B. Carlton and W. H. Colledge (2008). Kisspeptin can stimulate gonadotropin-releasing hormone (GnRH) release by a direct action at GnRH nerve terminals. Endocrinology 149(8): 3926-3932.

d'Anglemont de Tassigny, X., L. A. Fagg, J. P. Dixon, K. Day, H. G. Leitch, et al. (2007). Hypogonadotropic hypogonadism in mice lacking a functional Kiss1 gene. Proc Natl Acad Sci U S A 104(25): 10714-10719.

Dandona, P. and S. Dhindsa (2011). Update: Hypogonadotropic hypogonadism in type 2 diabetes and obesity. J Clin Endocrinol Metab 96(9): 2643-2651.

de Roux, N., E. Genin, J. C. Carel, F. Matsuda, J. L. Chaussain, et al. (2003). Hypogonadotropic hypogonadism due to loss of function of the KiSS1-derived peptide receptor GPR54. Proc Natl Acad Sci U S A 100(19): 10972-10976.

183

Dhar, D. K., H. Naora, H. Kubota, R. Maruyama, H. Yoshimura, et al. (2004). Downregulation of KiSS-1 expression is responsible for tumor invasion and worse prognosis in gastric carcinoma. Int J Cancer 111(6): 868-872.

Dhillo, W. S., O. B. Chaudhri, M. Patterson, E. L. Thompson, K. G. Murphy, et al. (2005). Kisspeptin-54 stimulates the hypothalamic-pituitary gonadal axis in human males. J Clin Endocrinol Metab 90(12): 6609-6615.

Dhillo, W. S., O. B. Chaudhri, E. L. Thompson, K. G. Murphy, M. Patterson, et al. (2007). Kisspeptin-54 stimulates gonadotropin release most potently during the preovulatory phase of the menstrual cycle in women. J Clin Endocrinol Metab 92(10): 3958-3966.

Dhillo, W. S., C. N. Jayasena, J. E. Jackson, J. A. Lynn, S. R. Bloom, et al. (2005). Localization of gastrinomas by selective intra-arterial calcium injection in patients on proton pump inhibitor or H2 receptor antagonist therapy. Eur J Gastroenterol Hepatol 17(4): 429-433.

Di Giorgio, N. P., S. J. Semaan, J. Kim, P. V. Lopez, B. Bettler, et al. (2014). Impaired GABAB receptor signaling dramatically up-regulates Kiss1 expression selectively in nonhypothalamic brain regions of adult but not prepubertal mice. Endocrinology 155(3): 1033-1044.

Dun, S. L., G. C. Brailoiu, A. Parsons, J. Yang, Q. Zeng, et al. (2003). Metastin-like immunoreactivity in the rat medulla oblongata and spinal cord. Neurosci Lett 335(3): 197- 201.

Dungan, H. M., M. L. Gottsch, H. Zeng, A. Gragerov, J. E. Bergmann, et al. (2007). The role of kisspeptin-GPR54 signaling in the tonic regulation and surge release of gonadotropin-releasing hormone/luteinizing hormone. J Neurosci 27(44): 12088-12095.

Eghlidi, D. H., G. E. Haley, N. C. Noriega, S. G. Kohama and H. F. Urbanski (2010). Influence of age and 17beta-estradiol on kisspeptin, neurokinin B, and prodynorphin gene expression in the arcuate-median eminence of female rhesus macaques. Endocrinology 151(8): 3783-3794.

Evans, J. J. (1999). Modulation of gonadotropin levels by peptides acting at the anterior pituitary gland. Endocr Rev 20(1): 46-67.

Ezzat Ahmed, A., H. Saito, T. Sawada, T. Yaegashi, T. Yamashita, et al. (2009). Characteristics of the stimulatory effect of kisspeptin-10 on the secretion of luteinizing hormone, follicle-stimulating hormone and growth hormone in prepubertal male and female cattle. J Reprod Dev 55(6): 650-654.

Fu, L. Y. and A. N. van den Pol (2010). Kisspeptin directly excites anorexigenic proopiomelanocortin neurons but inhibits orexigenic neuropeptide Y cells by an indirect synaptic mechanism. J Neurosci 30(30): 10205-10219.

184

Funabashi, T., S. Daikoku, K. Shinohara and F. Kimura (2000). Pulsatile gonadotropin- releasing hormone (GnRH) secretion is an inherent function of GnRH neurons, as revealed by the culture of medial olfactory placode obtained from embryonic rats. Neuroendocrinology 71(2): 138-144.

Funes, S., J. A. Hedrick, G. Vassileva, L. Markowitz, S. Abbondanzo, et al. (2003). The KiSS-1 receptor GPR54 is essential for the development of the murine reproductive system. Biochem Biophys Res Commun 312(4): 1357-1363.

Gaytan, F., M. Gaytan, J. M. Castellano, M. Romero, J. Roa, et al. (2009). KiSS-1 in the mammalian ovary: distribution of kisspeptin in human and marmoset and alterations in KiSS-1 mRNA levels in a rat model of ovulatory dysfunction. Am J Physiol Endocrinol Metab 296(3): E520-531.

George, J. T., R. A. Anderson and R. P. Millar (2012). Kisspeptin-10 stimulation of gonadotrophin secretion in women is modulated by sex steroid feedback. Hum Reprod 27(12): 3552-3559.

George, J. T., J. D. Veldhuis, A. K. Roseweir, C. L. Newton, E. Faccenda, et al. (2011). Kisspeptin-10 is a potent stimulator of LH and increases pulse frequency in men. J Clin Endocrinol Metab 96(8): E1228-1236.

George, J. T., J. D. Veldhuis, M. Tena-Sempere, R. P. Millar and R. A. Anderson (2013). Exploring the pathophysiology of hypogonadism in men with type 2 diabetes: kisspeptin- 10 stimulates serum testosterone and LH secretion in men with type 2 diabetes and mild biochemical hypogonadism. Clin Endocrinol (Oxf) 79(1): 100-104.

Gore, A. (2002). GnRH The Master Molecule of Reproduction.

Gottsch, M. L., M. J. Cunningham, J. T. Smith, S. M. Popa, B. V. Acohido, et al. (2004). A role for kisspeptins in the regulation of gonadotropin secretion in the mouse. Endocrinology 145(9): 4073-4077.

Grattan, D. R., C. L. Jasoni, X. Liu, G. M. Anderson and A. E. Herbison (2007). Prolactin regulation of gonadotropin-releasing hormone neurons to suppress luteinizing hormone secretion in mice. Endocrinology 148(9): 4344-4351.

Greives, T. J., K. L. Long, C. M. Burns and G. E. Demas (2011). Response to exogenous kisspeptin varies according to sex and reproductive condition in Siberian hamsters (Phodopus sungorus). Gen Comp Endocrinol 170(1): 172-179.

Guerriero, K. A., K. L. Keen and E. Terasawa (2012). Developmental increase in kisspeptin-54 release in vivo is independent of the pubertal increase in estradiol in female rhesus monkeys (Macaca mulatta). Endocrinology 153(4): 1887-1897.

185

Gutierrez-Pascual, E., A. J. Martinez-Fuentes, L. Pinilla, M. Tena-Sempere, M. M. Malagon, et al. (2007). Direct pituitary effects of kisspeptin: activation of gonadotrophs and somatotrophs and stimulation of luteinising hormone and growth hormone secretion. J Neuroendocrinol 19(7): 521-530.

Hahn, J. D., T. Kalamatianos and C. W. Coen (2003). Studies on the neuroanatomical basis for stress-induced oestrogen-potentiated suppression of reproductive function: evidence against direct corticotropin-releasing hormone projections to the vicinity of luteinizing hormone-releasing hormone cell bodies in female rats. J Neuroendocrinol 15(8): 732-742.

Hall, J. E., A. E. Taylor, K. A. Martin, J. Rivier, D. A. Schoenfeld, et al. (1994). Decreased release of gonadotropin-releasing hormone during the preovulatory midcycle luteinizing hormone surge in normal women. Proc Natl Acad Sci U S A 91(15): 6894-6898.

Han, S. K., M. L. Gottsch, K. J. Lee, S. M. Popa, J. T. Smith, et al. (2005). Activation of gonadotropin-releasing hormone neurons by kisspeptin as a neuroendocrine switch for the onset of puberty. J Neurosci 25(49): 11349-11356.

Hart, R. and R. Norman (2006). Polycystic ovarian syndrome--prognosis and outcomes. Best Pract Res Clin Obstet Gynaecol 20(5): 751-778.

Hashizume, T., H. Saito, T. Sawada, T. Yaegashi, A. A. Ezzat, et al. (2010). Characteristics of stimulation of gonadotropin secretion by kisspeptin-10 in female goats. Anim Reprod Sci 118(1): 37-41.

Hauge-Evans, A. C., C. C. Richardson, H. M. Milne, M. R. Christie, S. J. Persaud, et al. (2006). A role for kisspeptin in islet function. Diabetologia 49(9): 2131-2135.

Herbison, A. E. (2008). Estrogen positive feedback to gonadotropin-releasing hormone (GnRH) neurons in the rodent: the case for the rostral periventricular area of the third ventricle (RP3V). Brain Res Rev 57(2): 277-287.

Herbison, A. E., X. de Tassigny, J. Doran and W. H. Colledge (2010). Distribution and postnatal development of Gpr54 gene expression in mouse brain and gonadotropin- releasing hormone neurons. Endocrinology 151(1): 312-321.

Herbison, A. E. and S. M. Moenter (2011). Depolarising and hyperpolarising actions of GABA(A) receptor activation on gonadotrophin-releasing hormone neurones: towards an emerging consensus. J Neuroendocrinol 23(7): 557-569.

Herbison, A. E. and D. T. Theodosis (1992). Localization of oestrogen receptors in preoptic neurons containing neurotensin but not tyrosine hydroxylase, cholecystokinin or luteinizing hormone-releasing hormone in the male and female rat. Neuroscience 50(2): 283-298.

186

Herde, M. K., K. Geist, R. E. Campbell and A. E. Herbison (2011). Gonadotropin-releasing hormone neurons extend complex highly branched dendritic trees outside the blood-brain barrier. Endocrinology 152(10): 3832-3841.

Homma, T., M. Sakakibara, S. Yamada, M. Kinoshita, K. Iwata, et al. (2009). Significance of neonatal testicular sex steroids to defeminize anteroventral periventricular kisspeptin neurons and the GnRH/LH surge system in male rats. Biol Reprod 81(6): 1216-1225.

Horikoshi, Y., H. Matsumoto, Y. Takatsu, T. Ohtaki, C. Kitada, et al. (2003). Dramatic elevation of plasma metastin concentrations in human pregnancy: metastin as a novel placenta-derived hormone in humans. J Clin Endocrinol Metab 88(2): 914-919.

Hrabovszky, E., P. Ciofi, B. Vida, M. C. Horvath, E. Keller, et al. (2010). The kisspeptin system of the human hypothalamus: sexual dimorphism and relationship with gonadotropin-releasing hormone and neurokinin B neurons. Eur J Neurosci 31(11): 1984- 1998.

Hrabovszky, E., M. T. Sipos, C. S. Molnar, P. Ciofi, B. A. Borsay, et al. (2012). Low degree of overlap between kisspeptin, neurokinin B, and dynorphin immunoreactivities in the infundibular nucleus of young male human subjects challenges the KNDy neuron concept. Endocrinology 153(10): 4978-4989.

Hsu, Y. H., W. T. Lee and C. Chang (2007). Multiparametric MRI evaluation of kainic acid- induced neuronal activation in rat hippocampus. Brain 130(Pt 12): 3124-3134.

Hunter, D. R., R. A. Haworth and H. A. Berkoff (1981). Cellular manganese uptake by the isolated perfused rat heart: a probe for the sarcolemma calcium channel. J Mol Cell Cardiol 13(9): 823-832.

Inoue, N., K. Sasagawa, K. Ikai, Y. Sasaki, J. Tomikawa, et al. (2011). Kisspeptin neurons mediate reflex ovulation in the musk shrew (Suncus murinus). Proc Natl Acad Sci U S A 108(42): 17527-17532.

Irwig, M. S., G. S. Fraley, J. T. Smith, B. V. Acohido, S. M. Popa, et al. (2004). Kisspeptin activation of gonadotropin releasing hormone neurons and regulation of KiSS-1 mRNA in the male rat. Neuroendocrinology 80(4): 264-272.

Jayasena, C. N., A. Abbara, A. N. Comninos, G. M. Nijher, G. Christopoulos, et al. (2014). Kisspeptin-54 triggers egg maturation in women undergoing in vitro fertilization. J Clin Invest.

Jayasena, C. N., A. Abbara, J. D. Veldhuis, A. N. Comninos, R. Ratnasabapathy, et al. (2014). Increasing LH pulsatility in women with hypothalamic amenorrhoea using intravenous infusion of Kisspeptin-54. J Clin Endocrinol Metab 99(6): E953-961.

187

Jayasena, C. N., A. N. Comninos, A. De Silva, A. Abbara, J. D. Veldhuis, et al. (2014). Effects of neurokinin B administration on reproductive hormone secretion in healthy men and women. J Clin Endocrinol Metab 99(1): E19-27.

Jayasena, C. N., A. N. Comninos, A. Januszewski, H. Gabra, A. Taylor, et al. (2012). Plasma kisspeptin: a potential biomarker of tumor metastasis in patients with ovarian carcinoma. Clin Chem 58(6): 1061-1063.

Jayasena, C. N., A. N. Comninos, S. Narayanaswamy, S. Bhalla, A. Abbara, et al. (2014). Acute and chronic effects of kisspeptin-54 administration on GH, prolactin and TSH secretion in healthy women. Clin Endocrinol (Oxf).

Jayasena, C. N., A. N. Comninos, G. M. Nijher, A. Abbara, A. De Silva, et al. (2013). Twice-daily subcutaneous injection of kisspeptin-54 does not abolish menstrual cyclicity in healthy female volunteers. J Clin Endocrinol Metab 98(11): 4464-4474.

Jayasena, C. N., A. N. Comninos, J. D. Veldhuis, S. Misra, A. Abbara, et al. (2013). A single injection of kisspeptin-54 temporarily increases luteinizing hormone pulsatility in healthy women. Clin Endocrinol (Oxf) 79(4): 558-563.

Jayasena, C. N., G. M. Nijher, A. Abbara, K. G. Murphy, A. Lim, et al. (2010). Twice- weekly administration of kisspeptin-54 for 8 weeks stimulates release of reproductive hormones in women with hypothalamic amenorrhea. Clin Pharmacol Ther 88(6): 840-847.

Jayasena, C. N., G. M. Nijher, O. B. Chaudhri, K. G. Murphy, A. Ranger, et al. (2009). Subcutaneous injection of kisspeptin-54 acutely stimulates gonadotropin secretion in women with hypothalamic amenorrhea, but chronic administration causes tachyphylaxis. J Clin Endocrinol Metab 94(11): 4315-4323.

Jayasena, C. N., G. M. Nijher, A. N. Comninos, A. Abbara, A. Januszewki, et al. (2011). The effects of kisspeptin-10 on reproductive hormone release show sexual dimorphism in humans. J Clin Endocrinol Metab 96(12): E1963-1972.

Jayasena, C. N., G. M. Nijher, S. Narayanaswamy, A. De Silva, A. Abbara, et al. (2014). Age-dependent elevations in plasma kisspeptin are observed in boys and girls when compared with adults. Ann Clin Biochem 51(Pt 1): 89-96.

Ji, K., L. Ye, M. D. Mason and W. G. Jiang (2013). The Kiss-1/Kiss-1R complex as a negative regulator of cell motility and cancer metastasis (Review). Int J Mol Med 32(4): 747-754.

Jiang, Y., M. Berk, L. S. Singh, H. Tan, L. Yin, et al. (2005). KiSS1 suppresses metastasis in human ovarian cancer via inhibition of protein kinase C alpha. Clin Exp Metastasis 22(5): 369-376.

188

Jung, D. J., M. Han, S. U. Jin, S. H. Lee, I. Park, et al. (2014). Functional mapping of the auditory tract in rodent tinnitus model using manganese-enhanced magnetic resonance imaging. Neuroimage.

Kadokawa, H., M. Matsui, K. Hayashi, N. Matsunaga, C. Kawashima, et al. (2008). Peripheral administration of kisspeptin-10 increases plasma concentrations of GH as well as LH in prepubertal Holstein heifers. J Endocrinol 196(2): 331-334.

Kauffman, A. S., M. L. Gottsch, J. Roa, A. C. Byquist, A. Crown, et al. (2007). Sexual differentiation of Kiss1 gene expression in the brain of the rat. Endocrinology 148(4): 1774-1783.

Kauffman, A. S., J. H. Park, A. A. McPhie-Lalmansingh, M. L. Gottsch, C. Bodo, et al. (2007). The kisspeptin receptor GPR54 is required for sexual differentiation of the brain and behavior. J Neurosci 27(33): 8826-8835.

Keen, K. L., F. H. Wegner, S. R. Bloom, M. A. Ghatei and E. Terasawa (2008). An increase in kisspeptin-54 release occurs with the pubertal increase in luteinizing hormone- releasing hormone-1 release in the stalk-median eminence of female rhesus monkeys in vivo. Endocrinology 149(8): 4151-4157.

Keshavarzi, S., R. K. Sullivan, D. J. Ianno and P. Sah (2014). Functional properties and projections of neurons in the medial amygdala. J Neurosci 34(26): 8699-8715.

Kim, J., S. J. Semaan, D. K. Clifton, R. A. Steiner, S. Dhamija, et al. (2011). Regulation of Kiss1 expression by sex steroids in the amygdala of the rat and mouse. Endocrinology 152(5): 2020-2030.

Kinoshita, M., H. Tsukamura, S. Adachi, H. Matsui, Y. Uenoyama, et al. (2005). Involvement of central metastin in the regulation of preovulatory luteinizing hormone surge and estrous cyclicity in female rats. Endocrinology 146(10): 4431-4436.

Kirby, H. R., J. J. Maguire, W. H. Colledge and A. P. Davenport (2010). International Union of Basic and Clinical Pharmacology. LXXVII. Kisspeptin receptor nomenclature, distribution, and function. Pharmacol Rev 62(4): 565-578.

Kluver, H. and P. C. Bucy (1997). Preliminary analysis of functions of the temporal lobes in monkeys. 1939. J Neuropsychiatry Clin Neurosci 9(4): 606-620.

Knobil, E., T. M. Plant, L. Wildt, P. E. Belchetz and G. Marshall (1980). Control of the rhesus monkey menstrual cycle: permissive role of hypothalamic gonadotropin-releasing hormone. Science 207(4437): 1371-1373.

Kokay, I. C., S. L. Petersen and D. R. Grattan (2011). Identification of prolactin-sensitive GABA and kisspeptin neurons in regions of the rat hypothalamus involved in the control of fertility. Endocrinology 152(2): 526-535.

189

Kotani, M., M. Detheux, A. Vandenbogaerde, D. Communi, J. M. Vanderwinden, et al. (2001). The metastasis suppressor gene KiSS-1 encodes kisspeptins, the natural ligands of the orphan G protein-coupled receptor GPR54. J Biol Chem 276(37): 34631-34636.

Kousta, E., D. M. White and S. Franks (1997). Modern use of clomiphene citrate in induction of ovulation. Hum Reprod Update 3(4): 359-365.

Kraegen, E. W., L. Lazarus, H. Meler, L. Campbell and Y. O. Chia (1975). Carrier solutions for low-level intravenous insulin infusion. Br Med J 3(5981): 464-466.

Kuo, Y. T., A. H. Herlihy, P. W. So and J. D. Bell (2006). Manganese-enhanced magnetic resonance imaging (MEMRI) without compromise of the blood-brain barrier detects hypothalamic neuronal activity in vivo. NMR Biomed 19(8): 1028-1034.

Kuo, Y. T., J. R. Parkinson, O. B. Chaudhri, A. H. Herlihy, P. W. So, et al. (2007). The temporal sequence of gut peptide CNS interactions tracked in vivo by magnetic resonance imaging. J Neurosci 27(45): 12341-12348.

Labrie, F. (2014). GnRH agonists and the rapidly increasing use of combined androgen blockade in prostate cancer. Endocr Relat Cancer 21(4): R301-R317.

Lanfranco, F., J. Gromoll, S. von Eckardstein, E. M. Herding, E. Nieschlag, et al. (2005). Role of sequence variations of the GnRH receptor and G protein-coupled receptor 54 gene in male idiopathic hypogonadotropic hypogonadism. Eur J Endocrinol 153(6): 845- 852.

Lawton, I. E. and C. H. Sawyer (1970). Role of amygdala in regulating LH secretion in the adult female rat. Am J Physiol 218(3): 622-626.

Lee, D. K., T. Nguyen, G. P. O'Neill, R. Cheng, Y. Liu, et al. (1999). Discovery of a receptor related to the galanin receptors. FEBS Lett 446(1): 103-107.

Lee, J. H., M. E. Miele, D. J. Hicks, K. K. Phillips, J. M. Trent, et al. (1996). KiSS-1, a novel human malignant melanoma metastasis-suppressor gene. J Natl Cancer Inst 88(23): 1731-1737.

Lehallier, B., O. Rampin, A. Saint-Albin, N. Jerome, C. Ouali, et al. (2012). Brain processing of biologically relevant odors in the awake rat, as revealed by manganese- enhanced MRI. PLoS One 7(10): e48491.

Lents, C. A., N. L. Heidorn, C. R. Barb and J. J. Ford (2008). Central and peripheral administration of kisspeptin activates gonadotropin but not somatotropin secretion in prepubertal gilts. Reproduction 135(6): 879-887.

190

Lin, Y., X. Li, M. Lupi, J. S. Kinsey-Jones, B. Shao, et al. (2011). The role of the medial and central amygdala in stress-induced suppression of pulsatile LH secretion in female rats. Endocrinology 152(2): 545-555.

Lin, Y. J. and A. P. Koretsky (1997). Manganese ion enhances T1-weighted MRI during brain activation: an approach to direct imaging of brain function. Magn Reson Med 38(3): 378-388.

Liu, P. Y., D. M. Keenan, P. Kok, V. Padmanabhan, K. T. O'Byrne, et al. (2009). Sensitivity and specificity of pulse detection using a new deconvolution method. Am J Physiol Endocrinol Metab 297(2): E538-544.

Liu, X., K. Lee and A. E. Herbison (2008). Kisspeptin excites gonadotropin-releasing hormone neurons through a phospholipase C/calcium-dependent pathway regulating multiple ion channels. Endocrinology 149(9): 4605-4614.

Lu, H., Z. X. Xi, L. Gitajn, W. Rea, Y. Yang, et al. (2007). Cocaine-induced brain activation detected by dynamic manganese-enhanced magnetic resonance imaging (MEMRI). Proc Natl Acad Sci U S A 104(7): 2489-2494.

Luckman, S. M., R. E. Dyball and G. Leng (1994). Induction of c-fos expression in hypothalamic magnocellular neurons requires synaptic activation and not simply increased spike activity. J Neurosci 14(8): 4825-4830.

Maclean, D. B., H. Matsui, A. Suri, R. Neuwirth and M. Colombel (2014). Sustained Exposure to the Investigational Kisspeptin Analog, TAK-448, Downregulates Testosterone into the Castration Range in Healthy Males and in Patients with Prostate Cancer: Results From Two Phase 1 Studies. J Clin Endocrinol Metab: jc20134236.

Maguire, J. J., H. R. Kirby, E. J. Mead, R. E. Kuc, X. d'Anglemont de Tassigny, et al. (2011). Inotropic action of the puberty hormone kisspeptin in rat, mouse and human: cardiovascular distribution and characteristics of the kisspeptin receptor. PLoS One 6(11): e27601.

Majid, T., Y. O. Ali, D. V. Venkitaramani, M. K. Jang, H. C. Lu, et al. (2014). In vivo axonal transport deficits in a mouse model of fronto-temporal dementia. Neuroimage Clin 4: 711- 717.

Masui, T., R. Doi, T. Mori, E. Toyoda, M. Koizumi, et al. (2004). Metastin and its variant forms suppress migration of pancreatic cancer cells. Biochem Biophys Res Commun 315(1): 85-92.

Matsui, H., Y. Takatsu, S. Kumano, H. Matsumoto and T. Ohtaki (2004). Peripheral administration of metastin induces marked gonadotropin release and ovulation in the rat. Biochem Biophys Res Commun 320(2): 383-388.

191

Matsui, H., A. Tanaka, K. Yokoyama, Y. Takatsu, K. Ishikawa, et al. (2012). Chronic administration of the metastin/kisspeptin analog KISS1-305 or the investigational agent TAK-448 suppresses hypothalamic pituitary gonadal function and depletes plasma testosterone in adult male rats. Endocrinology 153(11): 5297-5308.

Mead, E. J., J. J. Maguire, R. E. Kuc and A. P. Davenport (2007). Kisspeptins are novel potent vasoconstrictors in humans, with a discrete localization of their receptor, G protein- coupled receptor 54, to atherosclerosis-prone vessels. Endocrinology 148(1): 140-147.

Mendonca-Dias, M. H., E. Gaggelli and P. C. Lauterbur (1983). Paramagnetic contrast agents in nuclear magnetic resonance medical imaging. Semin Nucl Med 13(4): 364-376.

Messager, S., E. E. Chatzidaki, D. Ma, A. G. Hendrick, D. Zahn, et al. (2005). Kisspeptin directly stimulates gonadotropin-releasing hormone release via G protein-coupled receptor 54. Proc Natl Acad Sci U S A 102(5): 1761-1766.

Mi, W. L., Q. L. Mao-Ying, Q. Liu, X. W. Wang, X. Li, et al. (2009). The distribution of kisspeptin and its receptor GPR54 in rat dorsal root ganglion and up-regulation of its expression after CFA injection. Brain Res Bull 78(4-5): 254-260.

Mikkelsen, J. D., A. H. Bentsen, L. Ansel, V. Simonneaux and A. Juul (2009). Comparison of the effects of peripherally administered kisspeptins. Regul Pept 152(1-3): 95-100.

Millar, R. P. and C. L. Newton (2013). Current and future applications of GnRH, kisspeptin and neurokinin B analogues. Nat Rev Endocrinol 9(8): 451-466.

Millar, R. P., A. K. Roseweir, J. A. Tello, R. A. Anderson, J. T. George, et al. (2010). Kisspeptin antagonists: unraveling the role of kisspeptin in reproductive physiology. Brain Res 1364: 81-89.

Min, L., K. Soltis, A. C. Reis, S. Xu, W. Kuohung, et al. (2014). Dynamic kisspeptin receptor trafficking modulates kisspeptin-mediated calcium signaling. Mol Endocrinol 28(1): 16-27.

Molitch, M. E. (2010). Pituitary gland: can prolactinomas be cured medically? Nat Rev Endocrinol 6(4): 186-188.

Mori, T., R. Doi, M. Koizumi, E. Toyoda, D. Ito, et al. (2004). CXCR4 antagonist inhibits stromal cell-derived factor 1-induced migration and invasion of human pancreatic cancer. Mol Cancer Ther 3(1): 29-37.

Muir, A. I., L. Chamberlain, N. A. Elshourbagy, D. Michalovich, D. J. Moore, et al. (2001). AXOR12, a novel human G protein-coupled receptor, activated by the peptide KiSS-1. J Biol Chem 276(31): 28969-28975.

192

Murray, E. A. (2007). The amygdala, reward and emotion. Trends Cogn Sci 11(11): 489- 497.

Nagai, K., R. Doi, F. Katagiri, T. Ito, A. Kida, et al. (2009). Prognostic value of metastin expression in human pancreatic cancer. J Exp Clin Cancer Res 28: 9.

Nakamura, Y., S. Aoki, Y. Xing, H. Sasano and W. E. Rainey (2007). Metastin stimulates aldosterone synthesis in human adrenal cells. Reprod Sci 14(8): 836-845.

Narita, K., F. Kawasaki and H. Kita (1990). Mn and Mg influxes through Ca channels of motor nerve terminals are prevented by verapamil in frogs. Brain Res 510(2): 289-295.

Navarro, V. M. (2013). Interactions between kisspeptins and neurokinin B. Adv Exp Med Biol 784: 325-347.

Navarro, V. M., J. M. Castellano, R. Fernandez-Fernandez, M. L. Barreiro, J. Roa, et al. (2004). Developmental and hormonally regulated messenger ribonucleic acid expression of KiSS-1 and its putative receptor, GPR54, in rat hypothalamus and potent luteinizing hormone-releasing activity of KiSS-1 peptide. Endocrinology 145(10): 4565-4574.

Navarro, V. M., J. M. Castellano, R. Fernandez-Fernandez, S. Tovar, J. Roa, et al. (2005). Effects of KiSS-1 peptide, the natural ligand of GPR54, on follicle-stimulating hormone secretion in the rat. Endocrinology 146(4): 1689-1697.

Navarro, V. M., R. Fernandez-Fernandez, J. M. Castellano, J. Roa, A. Mayen, et al. (2004). Advanced vaginal opening and precocious activation of the reproductive axis by KiSS-1 peptide, the endogenous ligand of GPR54. J Physiol 561(Pt 2): 379-386.

Navarro, V. M., M. L. Gottsch, M. Wu, D. Garcia-Galiano, S. J. Hobbs, et al. (2011). Regulation of NKB pathways and their roles in the control of Kiss1 neurons in the arcuate nucleus of the male mouse. Endocrinology 152(11): 4265-4275.

Neal-Perry, G., D. Lebesgue, M. Lederman, J. Shu, G. D. Zeevalk, et al. (2009). The excitatory peptide kisspeptin restores the luteinizing hormone surge and modulates amino acid neurotransmission in the medial preoptic area of middle-aged rats. Endocrinology 150(8): 3699-3708.

Ogawa, S., F. M. Nathan and I. S. Parhar (2014). Habenular kisspeptin modulates fear in the zebrafish. Proc Natl Acad Sci U S A 111(10): 3841-3846.

Ohtaki, T., Y. Shintani, S. Honda, H. Matsumoto, A. Hori, et al. (2001). Metastasis suppressor gene KiSS-1 encodes peptide ligand of a G-protein-coupled receptor. Nature 411(6837): 613-617.

193

Pallais, J. C., Y. Bo-Abbas, N. Pitteloud, W. F. Crowley, Jr. and S. B. Seminara (2006). Neuroendocrine, gonadal, placental, and obstetric phenotypes in patients with IHH and mutations in the G-protein coupled receptor, GPR54. Mol Cell Endocrinol 254-255: 70-77.

Parkinson, J. R., O. B. Chaudhri, Y. T. Kuo, B. C. Field, A. H. Herlihy, et al. (2009). Differential patterns of neuronal activation in the brainstem and hypothalamus following peripheral injection of GLP-1, oxyntomodulin and lithium chloride in mice detected by manganese-enhanced magnetic resonance imaging (MEMRI). Neuroimage 44(3): 1022- 1031.

Patterson, M., K. G. Murphy, E. L. Thompson, S. Patel, M. A. Ghatei, et al. (2006). Administration of kisspeptin-54 into discrete regions of the hypothalamus potently increases plasma luteinising hormone and testosterone in male adult rats. J Neuroendocrinol 18(5): 349-354.

Pau, K. Y., M. Berria, D. L. Hess and H. G. Spies (1993). Preovulatory gonadotropin- releasing hormone surge in ovarian-intact rhesus macaques. Endocrinology 133(4): 1650- 1656.

Pautler, R. G. (2004). In vivo, trans-synaptic tract-tracing utilizing manganese-enhanced magnetic resonance imaging (MEMRI). NMR Biomed 17(8): 595-601.

Paxinos, G. and K. B. J. Franklin (2004). The mouse brain in stereotaxic coordinates. Amsterdam ; Boston, Elsevier Academic Press.

Petersen, S. L., E. N. Ottem and C. D. Carpenter (2003). Direct and indirect regulation of gonadotropin-releasing hormone neurons by estradiol. Biol Reprod 69(6): 1771-1778.

Pielecka-Fortuna, J. and S. M. Moenter (2010). Kisspeptin increases gamma-aminobutyric acidergic and glutamatergic transmission directly to gonadotropin-releasing hormone neurons in an estradiol-dependent manner. Endocrinology 151(1): 291-300.

Piet, R., S. de Croft, X. Liu and A. E. Herbison (2014). Electrical properties of kisspeptin neurons and their regulation of GnRH neurons. Front Neuroendocrinol.

Plant, T. M., S. Ramaswamy and M. J. Dipietro (2006). Repetitive activation of hypothalamic G protein-coupled receptor 54 with intravenous pulses of kisspeptin in the juvenile monkey (Macaca mulatta) elicits a sustained train of gonadotropin-releasing hormone discharges. Endocrinology 147(2): 1007-1013.

Porzionato, A., G. Fenu, M. Rucinski, V. Macchi, A. Montella, et al. (2011). KISS1 and KISS1R expression in the human and rat carotid body and superior cervical ganglion. Eur J Histochem 55(2): e14.

Rabin, O., L. Hegedus, J. M. Bourre and Q. R. Smith (1993). Rapid brain uptake of manganese(II) across the blood-brain barrier. J Neurochem 61(2): 509-517.

194

Ramaswamy, S., R. B. Gibbs and T. M. Plant (2009). Studies of the localisation of kisspeptin within the pituitary of the rhesus monkey (Macaca mulatta) and the effect of kisspeptin on the release of non-gonadotropic pituitary hormones. J Neuroendocrinol 21(10): 795-804.

Ramaswamy, S., K. A. Guerriero, R. B. Gibbs and T. M. Plant (2008). Structural interactions between kisspeptin and GnRH neurons in the mediobasal hypothalamus of the male rhesus monkey (Macaca mulatta) as revealed by double immunofluorescence and confocal microscopy. Endocrinology 149(9): 4387-4395.

Ramaswamy, S., S. B. Seminara, C. R. Pohl, M. J. DiPietro, W. F. Crowley, Jr., et al. (2007). Effect of continuous intravenous administration of human metastin 45-54 on the neuroendocrine activity of the hypothalamic-pituitary-testicular axis in the adult male rhesus monkey (Macaca mulatta). Endocrinology 148(7): 3364-3370.

Rance, N. E. (2009). Menopause and the human hypothalamus: evidence for the role of kisspeptin/neurokinin B neurons in the regulation of estrogen negative feedback. Peptides 30(1): 111-122.

Rance, N. E., W. S. Young, 3rd and N. T. McMullen (1994). Topography of neurons expressing luteinizing hormone-releasing hormone gene transcripts in the human hypothalamus and basal forebrain. J Comp Neurol 339(4): 573-586.

Ratnasabapathy, R. and W. S. Dhillo (2013). The effects of kisspeptin in human reproductive function - therapeutic implications. Curr Drug Targets 14(3): 365-371.

Redmond, J. S., G. G. Macedo, I. C. Velez, A. Caraty, G. L. Williams, et al. (2011). Kisspeptin activates the hypothalamic-adenohypophyseal-gonadal axis in prepubertal ewe lambs. Reproduction 141(4): 541-548.

Reichlin, S. (1998). 'Neuroendocrinology'. Williams Textbook of Endocrinology 10th edition: 165-248.

Reindollar, R. H., M. Novak, S. P. Tho and P. G. McDonough (1986). Adult-onset amenorrhea: a study of 262 patients. Am J Obstet Gynecol 155(3): 531-543.

Richard, N., G. Galmiche, S. Corvaisier, A. Caraty and M. L. Kottler (2008). KiSS-1 and GPR54 genes are co-expressed in rat gonadotrophs and differentially regulated in vivo by oestradiol and gonadotrophin-releasing hormone. J Neuroendocrinol 20(3): 381-393.

Ringel, M. D., E. Hardy, V. J. Bernet, H. B. Burch, F. Schuppert, et al. (2002). Metastin receptor is overexpressed in papillary thyroid cancer and activates MAP kinase in thyroid cancer cells. J Clin Endocrinol Metab 87(5): 2399.

195

Roa, J., E. Vigo, J. M. Castellano, F. Gaytan, D. Garcia-Galiano, et al. (2008). Follicle- stimulating hormone responses to kisspeptin in the female rat at the preovulatory period: modulation by estrogen and progesterone receptors. Endocrinology 149(11): 5783-5790.

Roa, J., E. Vigo, J. M. Castellano, F. Gaytan, V. M. Navarro, et al. (2008). Opposite roles of estrogen receptor (ER)-alpha and ERbeta in the modulation of luteinizing hormone responses to kisspeptin in the female rat: implications for the generation of the preovulatory surge. Endocrinology 149(4): 1627-1637.

Roa, J., E. Vigo, J. M. Castellano, V. M. Navarro, R. Fernandez-Fernandez, et al. (2006). Hypothalamic expression of KiSS-1 system and gonadotropin-releasing effects of kisspeptin in different reproductive states of the female Rat. Endocrinology 147(6): 2864- 2878.

Roa, J., E. Vigo, D. Garcia-Galiano, J. M. Castellano, V. M. Navarro, et al. (2008). Desensitization of gonadotropin responses to kisspeptin in the female rat: analyses of LH and FSH secretion at different developmental and metabolic states. Am J Physiol Endocrinol Metab 294(6): E1088-1096.

Roca, A., K. J. Shin, X. Liu, M. I. Simon and J. Chen (2004). Comparative analysis of transcriptional profiles between two apoptotic pathways of light-induced retinal degeneration. Neuroscience 129(3): 779-790.

Rometo, A. M., S. J. Krajewski, M. L. Voytko and N. E. Rance (2007). Hypertrophy and increased kisspeptin gene expression in the hypothalamic infundibular nucleus of postmenopausal women and ovariectomized monkeys. J Clin Endocrinol Metab 92(7): 2744-2750.

Roseweir, A. K., A. S. Kauffman, J. T. Smith, K. A. Guerriero, K. Morgan, et al. (2009). Discovery of potent kisspeptin antagonists delineate physiological mechanisms of gonadotropin regulation. J Neurosci 29(12): 3920-3929.

Salacinski, P. R., C. McLean, J. E. Sykes, V. V. Clement-Jones and P. J. Lowry (1981). Iodination of proteins, glycoproteins, and peptides using a solid-phase oxidizing agent, 1,3,4,6-tetrachloro-3 alpha,6 alpha-diphenyl glycoluril (Iodogen). Anal Biochem 117(1): 136-146.

Sanchez-Carbayo, M., P. Capodieci and C. Cordon-Cardo (2003). Tumor suppressor role of KiSS-1 in bladder cancer: loss of KiSS-1 expression is associated with bladder cancer progression and clinical outcome. Am J Pathol 162(2): 609-617.

Sanchez-Carbayo, M., N. D. Socci, J. J. Lozano, W. Li, E. Charytonowicz, et al. (2003). Gene discovery in bladder cancer progression using cDNA microarrays. Am J Pathol 163(2): 505-516.

196

Sano, K., T. Motoike, H. Nakamura, Y. Matsui, Y. Teranishi, et al. (1991). Differential expression of the N-myc, c-fos, and smg p25A genes in human neuroblastoma cells during neuronal and Schwannian differentiation. Brain Res Mol Brain Res 9(1-2): 149-152.

Sawyer, I., S. J. Smillie, J. V. Bodkin, E. Fernandes, K. T. O'Byrne, et al. (2011). The vasoactive potential of kisspeptin-10 in the peripheral vasculature. PLoS One 6(2): e14671.

Scott, V. and C. H. Brown (2011). Kisspeptin activation of supraoptic nucleus neurons in vivo. Endocrinology 152(10): 3862-3870.

Seminara, S. B., M. J. Dipietro, S. Ramaswamy, W. F. Crowley, Jr. and T. M. Plant (2006). Continuous human metastin 45-54 infusion desensitizes G protein-coupled receptor 54-induced gonadotropin-releasing hormone release monitored indirectly in the juvenile male Rhesus monkey (Macaca mulatta): a finding with therapeutic implications. Endocrinology 147(5): 2122-2126.

Seminara, S. B., S. Messager, E. E. Chatzidaki, R. R. Thresher, J. S. Acierno, Jr., et al. (2003). The GPR54 gene as a regulator of puberty. N Engl J Med 349(17): 1614-1627.

Semple, R. K., J. C. Achermann, J. Ellery, I. S. Farooqi, F. E. Karet, et al. (2005). Two novel missense mutations in g protein-coupled receptor 54 in a patient with hypogonadotropic hypogonadism. J Clin Endocrinol Metab 90(3): 1849-1855.

Shahab, M., C. Mastronardi, S. B. Seminara, W. F. Crowley, S. R. Ojeda, et al. (2005). Increased hypothalamic GPR54 signaling: a potential mechanism for initiation of puberty in primates. Proc Natl Acad Sci U S A 102(6): 2129-2134.

Shoji, I., T. Hirose, N. Mori, K. Hiraishi, I. Kato, et al. (2010). Expression of kisspeptins and kisspeptin receptor in the kidney of chronic renal failure rats. Peptides 31(10): 1920- 1925.

Sigmund, O., Z. Naor, D. J. Anderson and R. Stein (1990). Effect of nerve growth factor and fibroblast growth factor on SCG10 and c-fos expression and neurite outgrowth in protein kinase C-depleted PC12 cells. J Biol Chem 265(4): 2257-2261.

Silva, A. C. and N. A. Bock (2008). Manganese-enhanced MRI: an exceptional tool in translational neuroimaging. Schizophr Bull 34(4): 595-604.

Silva, A. C., A. P. Koretsky and J. H. Duyn (2007). Functional MRI impulse response for BOLD and CBV contrast in rat somatosensory cortex. Magn Reson Med 57(6): 1110- 1118.

Silveira, L. G., S. D. Noel, A. P. Silveira-Neto, A. P. Abreu, V. N. Brito, et al. (2010). Mutations of the KISS1 gene in disorders of puberty. J Clin Endocrinol Metab 95(5): 2276- 2280.

197

Silvestre, R. A., E. M. Egido, R. Hernandez and J. Marco (2008). Kisspeptin-13 inhibits insulin secretion without affecting glucagon or somatostatin release: study in the perfused rat pancreas. J Endocrinol 196(2): 283-290.

Smith, J. T., M. J. Cunningham, E. F. Rissman, D. K. Clifton and R. A. Steiner (2005). Regulation of Kiss1 gene expression in the brain of the female mouse. Endocrinology 146(9): 3686-3692.

Smith, J. T., H. M. Dungan, E. A. Stoll, M. L. Gottsch, R. E. Braun, et al. (2005). Differential regulation of KiSS-1 mRNA expression by sex steroids in the brain of the male mouse. Endocrinology 146(7): 2976-2984.

Smith, J. T., S. M. Popa, D. K. Clifton, G. E. Hoffman and R. A. Steiner (2006). Kiss1 neurons in the forebrain as central processors for generating the preovulatory luteinizing hormone surge. J Neurosci 26(25): 6687-6694.

Smith, J. T., A. Rao, A. Pereira, A. Caraty, R. P. Millar, et al. (2008). Kisspeptin is present in ovine hypophysial portal blood but does not increase during the preovulatory luteinizing hormone surge: evidence that gonadotropes are not direct targets of kisspeptin in vivo. Endocrinology 149(4): 1951-1959.

Smith, J. T., M. Shahab, A. Pereira, K. Y. Pau and I. J. Clarke (2010). Hypothalamic expression of KISS1 and gonadotropin inhibitory hormone genes during the menstrual cycle of a non-human primate. Biol Reprod 83(4): 568-577.

Soldin, O. P., S. H. Chung and D. R. Mattison (2011). Sex differences in drug disposition. J Biomed Biotechnol 2011: 187103.

Song, W. J., P. Mondal, A. Wolfe, L. C. Alonso, R. Stamateris, et al. (2014). Glucagon regulates hepatic kisspeptin to impair insulin secretion. Cell Metab 19(4): 667-681.

Sonigo, C., J. Bouilly, N. Carre, V. Tolle, A. Caraty, et al. (2012). Hyperprolactinemia- induced ovarian acyclicity is reversed by kisspeptin administration. J Clin Invest 122(10): 3791-3795.

Spampinato, S., A. Trabucco, A. Biasiotta, F. Biagioni, G. Cruccu, et al. (2011). Hyperalgesic activity of kisspeptin in mice. Mol Pain 7: 90.

Szawka, R. E., A. B. Ribeiro, C. M. Leite, C. V. Helena, C. R. Franci, et al. (2010). Kisspeptin regulates prolactin release through hypothalamic dopaminergic neurons. Endocrinology 151(7): 3247-3257.

Takahashi, K., I. Shoji, A. Shibasaki, I. Kato, K. Hiraishi, et al. (2010). Presence of kisspeptin-like immunoreactivity in human adrenal glands and adrenal tumors. J Mol Neurosci 41(1): 138-144.

198

Tariq, A. R., M. Shahab, I. J. Clarke, A. Pereira, J. T. Smith, et al. (2013). Kiss1 and Kiss1 receptor expression in the rhesus monkey testis: a possible local regulator of testicular function. Central European Journal of Biology 8(10): 968-974.

Teles, M. G., S. D. Bianco, V. N. Brito, E. B. Trarbach, W. Kuohung, et al. (2008). A GPR54-activating mutation in a patient with central precocious puberty. N Engl J Med 358(7): 709-715.

Tena-Sempere, M. (2006). GPR54 and kisspeptin in reproduction. Hum Reprod Update 12(5): 631-639.

Tenenbaum-Rakover, Y., M. Commenges-Ducos, A. Iovane, C. Aumas, O. Admoni, et al. (2007). Neuroendocrine phenotype analysis in five patients with isolated hypogonadotropic hypogonadism due to a L102P inactivating mutation of GPR54. J Clin Endocrinol Metab 92(3): 1137-1144.

Thompson, E. L., K. G. Murphy, M. Patterson, G. A. Bewick, G. W. Stamp, et al. (2006). Chronic subcutaneous administration of kisspeptin-54 causes testicular degeneration in adult male rats. Am J Physiol Endocrinol Metab 291(5): E1074-1082.

Thompson, E. L., M. Patterson, K. G. Murphy, K. L. Smith, W. S. Dhillo, et al. (2004). Central and peripheral administration of kisspeptin-10 stimulates the hypothalamic- pituitary-gonadal axis. J Neuroendocrinol 16(10): 850-858.

Tolson, K. P., C. Garcia, S. Yen, S. Simonds, A. Stefanidis, et al. (2014). Impaired kisspeptin signaling decreases metabolism and promotes glucose intolerance and obesity. J Clin Invest 124(7): 3075-3079.

Tonsfeldt, K. J., C. P. Goodall, K. L. Latham and P. E. Chappell (2011). Oestrogen induces rhythmic expression of the Kisspeptin-1 receptor GPR54 in hypothalamic gonadotrophin-releasing hormone-secreting GT1-7 cells. J Neuroendocrinol 23(9): 823- 830.

Topaloglu, A. K., F. Reimann, M. Guclu, A. S. Yalin, L. D. Kotan, et al. (2009). TAC3 and TACR3 mutations in familial hypogonadotropic hypogonadism reveal a key role for Neurokinin B in the central control of reproduction. Nat Genet 41(3): 354-358.

Topaloglu, A. K., J. A. Tello, L. D. Kotan, M. N. Ozbek, M. B. Yilmaz, et al. (2012). Inactivating KISS1 mutation and hypogonadotropic hypogonadism. N Engl J Med 366(7): 629-635.

Tovar, S., M. J. Vazquez, V. M. Navarro, R. Fernandez-Fernandez, J. M. Castellano, et al. (2006). Effects of single or repeated intravenous administration of kisspeptin upon dynamic LH secretion in conscious male rats. Endocrinology 147(6): 2696-2704.

199

Tsutsumi, R. and N. J. Webster (2009). GnRH pulsatility, the pituitary response and reproductive dysfunction. Endocr J 56(6): 729-737.

Vikman, J. and B. Ahren (2009). Inhibitory effect of kisspeptins on insulin secretion from isolated mouse islets. Diabetes Obes Metab 11 Suppl 4: 197-201.

Wagner, H. J., C. Pilgrim and J. Brandl (1974). Penetration and removal of horseradish peroxidase injected into the cerebrospinal fluid: role of cerebral perivascular spaces, endothelium and microglia. Acta Neuropathol 27(4): 299-315.

Wahab, F., R. Bano, S. Jabeen, S. Irfan and M. Shahab (2010). Effect of peripheral kisspeptin administration on adiponectin, leptin, and resistin secretion under fed and fasting conditions in the adult male rhesus monkey (Macaca mulatta). Horm Metab Res 42(8): 570-574.

Wahab, F., T. Riaz and M. Shahab (2011). Study on the effect of peripheral kisspeptin administration on basal and glucose-induced insulin secretion under fed and fasting conditions in the adult male rhesus monkey (Macaca mulatta). Horm Metab Res 43(1): 37- 42.

Wang, G. J., J. Yang, N. D. Volkow, F. Telang, Y. Ma, et al. (2006). Gastric stimulation in obese subjects activates the hippocampus and other regions involved in brain reward circuitry. Proc Natl Acad Sci U S A 103(42): 15641-15645.

Watanabe, T., J. Frahm and T. Michaelis (2004). Functional mapping of neural pathways in rodent brain in vivo using manganese-enhanced three-dimensional magnetic resonance imaging. NMR Biomed 17(8): 554-568.

Welt, C. K., J. L. Chan, J. Bullen, R. Murphy, P. Smith, et al. (2004). Recombinant human leptin in women with hypothalamic amenorrhea. N Engl J Med 351(10): 987-997.

West, A., P. J. Vojta, D. R. Welch and B. E. Weissman (1998). Chromosome localization and genomic structure of the KiSS-1 metastasis suppressor gene (KISS1). Genomics 54(1): 145-148.

Wildt, L., A. Hausler, G. Marshall, J. S. Hutchison, T. M. Plant, et al. (1981). 'Frequency and amplitude of gonadotropin-releasing hormone stimulation and gonadotropin secretion in the rhesus monkey'. Endocrinology 109(2): 376-385.

Wyneken, U., J. J. Marengo and F. Orrego (2004). Electrophysiology and plasticity in isolated postsynaptic densities. Brain Res Brain Res Rev 47(1-3): 54-70.

Xu, Z., S. Kaga, A. Mochiduki, J. Tsubomizu, S. Adachi, et al. (2012). Immunocytochemical localization of kisspeptin neurons in the rat forebrain with special reference to sexual dimorphism and interaction with GnRH neurons. Endocr J 59(2): 161- 171.

200

Yamamori, T. (1991). CDF/LIF selectively increases c-fos and jun-B transcripts in sympathetic neurons. Neuroreport 2(4): 173-176.

Yang, B., Q. Jiang, T. Chan, W. K. Ko and A. O. Wong (2010). Goldfish kisspeptin: molecular cloning, tissue distribution of transcript expression, and stimulatory effects on prolactin, growth hormone and luteinizing hormone secretion and gene expression via direct actions at the pituitary level. Gen Comp Endocrinol 165(1): 60-71.

Young, J., J. T. George, J. A. Tello, B. Francou, J. Bouligand, et al. (2013). Kisspeptin restores pulsatile LH secretion in patients with neurokinin B signaling deficiencies: physiological, pathophysiological and therapeutic implications. Neuroendocrinology 97(2): 193-202.

Zeger, S. L. and K. Y. Liang (1986). Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42(1): 121-130.

Zhang, C., M. A. Bosch, O. K. Ronnekleiv and M. J. Kelly (2009). Gamma-aminobutyric acid B receptor mediated inhibition of gonadotropin-releasing hormone neurons is suppressed by kisspeptin-G protein-coupled receptor 54 signaling. Endocrinology 150(5): 2388-2394.

201

Publications & Communications ______

202

Publications and Communications ______

Directly arising from thesis:

Publications (attached at back of thesis)

The Effects of Kisspeptin-10 on Reproductive Hormone Release Show Sexual Dimorphism in Humans. CN Jayasena, GM Nijher, AN Comninos, A Abbara, A Januszewki, ML Vaal, L Sriskandarajah, KG Murphy, Z Farzad, MA Ghatei, SR Bloom, WS Dhillo. J Clin Endocrinol Metab. 2011;96:1963-72.

Twice-daily subcutaneous injection of kisspeptin-54 does not abolish menstrual cyclicity in healthy female volunteers. AN Comninos*, CN Jayasena*, GMK Nijher, A Abbara, A De Silva, JD Veldhuis, R Ratnasabapathy, C Izzi-Engbeaya, A Lim, DA Patel, MA Ghatei, SR Bloom & WS Dhillo. J Clin Endocrinol Metab. 2013;98(11):4464-74.

Oral Communications

Contrasting effects of kisspeptin-10 between men and women reveal sexual dimorphism in the hypothalamic regulation of human reproduction. CN Jayasena, GM Nijher, AN Comninos, A Abbara, A Januszewki, ML Vaal, L Sriskandarajah, KG Murphy, Z Farzad, MA Ghatei, SR Bloom, WS Dhillo Society for Endocrinology Annual Conference, March 2012, Harrogate

Men and women respond differently to the novel hormone kisspeptin-10. AN Comninos, CN Jayasena, WS Dhillo Wellcome Trust Clinical Fellows’ 2012 Translational Research Meeting, London

Kisspeptin advances ovulation in healthy women. AN Comninos, CN Jayasena CN, G Nijher, A Abbara , A De Silva, JD Veldhuis JD, R Ratnasabapathy, C Izzi-Engbeaya C, A Lim, DA Patel, M Ghatei, SR Bloom & WS Dhillo. Society for Endocrinology Annual Conference, March 2013, Harrogate Awarded Prize for Highly Commended Oral Communication

Kisspeptin advances the menstrual cycle in healthy women. AN Comninos, CN Jayasena CN, G Nijher, A Abbara , A De Silva, JD Veldhuis JD, R Ratnasabapathy, C Izzi-Engbeaya C, A Lim, DA Patel, M Ghatei, SR Bloom & WS Dhillo. American Endocrine Society Annual Conference, June 2013, San Francisco

203

Poster Communications

Chronic kisspeptin administration advances the menstrual cycle in healthy women. AN Comninos, CN Jayasena CN, GMK Nijher, A Abbara , A De Silva, JD Veldhuis JD, R Ratnasabapathy, C Izzi-Engbeaya C, A Lim, DA Patel, MA Ghatei, SR Bloom & WS Dhillo. European Society of Human Reproduction and Embryology Annual Meeting, June 2013, London

The effects of kisspeptin on the menstrual cycle in healthy women. AN Comninos, CN Jayasena CN, GMK Nijher, A Abbara , A De Silva, JD Veldhuis JD, R Ratnasabapathy, C Izzi-Engbeaya C, A Lim, DA Patel, MA Ghatei, SR Bloom, WS Dhillo. Shortlisted for Academy of Medical Sciences Spring Meeting Poster Competition, February 2014, London

Kisspeptin advances ovulation in healthy women. AN Comninos, CN Jayasena CN, GMK Nijher, A Abbara , A De Silva, JD Veldhuis JD, R Ratnasabapathy, C Izzi-Engbeaya C, A Lim, DA Patel, MA Ghatei, SR Bloom, WS Dhillo. Awarded Travel Prize to present at The European Cooperation in Science and Technology GnRH Network Training School, March 2014, Berlin

Kisspeptin inhibits neuronal activity in the amygdala of male mice. AN Comninos, J Anastasovska, M Sahuri-Arisoylu, CN Jayasena, MA Ghatei, SR Bloom, PM Matthews, JD Bell, WS Dhillo. American Endocrine Society Annual Conference, June 2014, Chicago

Other related publications during my PhD studies:

First author

Plasma Kisspeptin: A Potential Biomarker of Tumor Metastasis in Patients with Ovarian Carcinoma. AN Comninos*, CN Jayasena*, A Januszewski, H Gabra, A Taylor, RA Harvey, MA Ghatei, SR Bloom, WS Dhillo. Clin Chem. 2012;58(6):1061-3.

Effects of neurokinin B administration on reproductive hormone secretion in healthy men and women. AN Comninos*, CN Jayasena*, A De Silva, A Abbara, JD Veldhuis, GMK Nijher, Z Ganiyu-Dada, M Vaal, G Stamp, MA Ghatei, SR Bloom, WS Dhillo. J Clin Endocrinol Metab. 2014;99(1):E19-27.

The relationship between gut and adipose hormones, and reproduction. AN Comninos, CN Jayasena, WS Dhillo. Human Reproduction Update. 2014;20(2):153-74.

204

Acute and chronic effects of kisspeptin-54 administration on GH, prolactin and TSH secretion in healthy women. AN Comninos*, CN Jayasena*, S Narayanaswamy, S Bhalla, A Abbara, Z Ganiyu- Dada1, M Busbridge, MA Ghatei, SR Bloom, WS Dhillo. Clin Endocrinol 2014 May 24 (Epub ahead of print)

The identification of elevated urinary kisspeptin-immunoreactivity during pregnancy. AN Comninos*, CN Jayasena*, S Narayanaswamy, A Abbara, GM Nijher, M Cheema, Z Malik, MA Ghatei, SR Bloom, WS Dhillo. Ann Clin Biochem. 2014 Aug 28 (Epub ahead of print)

Non-first author publications

The effects of long-term GH and IGF-1 exposure on the development of cardiovascular, cerebrovascular and metabolic co-morbidities in treated patients with acromegaly. CN Jayasena, AN Comninos, H Clarke, WS Dhillo. Clin Endocrinol 2011;72(2):220-5.

A single injection of kisspeptin-54 temporarily increases LH pulsatility in healthy women. CN Jayasena, AN Comninos, J Veldhuis, S Misra, A Abbara, C Izzi-Engbeaya, M Donaldson, MA Ghatei, SR Bloom, WS Dhillo. Clin Endocrinol 2013;79(4):558-63.

Increasing LH pulsatility in women with hypothalamic amenorrhoea using intravenous infusion of kisspeptin-54. CN Jayasena*, A Abbara*, J Veldhuis, AN Comninos, R Ratnasabapathy, A De Silva, GMK Nijher, Z Ganiyu-Dada, A Mehta, C Todd, MA Ghatei, SR Bloom, WS Dhillo. J Clin Endocrinol Metab 2014 Feb 11 (Epub ahead of print)

Kisspeptin-54 triggers egg maturation in women undergoing in vitro fertilization. CN Jayasena*, A Abbara*, AN Comninos, GMK Nijher, G Christopoulos, S Narayanaswamy, C Izzi-Engbeaya, M Sridharan, A Mason, J Warwick, D Ashby, MA Ghatei, SR Bloom, A Carby, G Trew, WS Dhillo. J Clin Invest 2014 Jul 18 (Epub ahead of print)

Reduced levels of plasma kisspeptin during the antenatal booking visit are associated with increased risk of miscarriage. CN Jayasena*, A Abbara*, C Izzi-Engbeaya, AN Comninos, RA Harvey, J Gonzalez Maffe, Z Sarang, Z Ganiyu-Dada, AI Padilha, M Dhanjal, C Williamson, L Regan, MA Ghatei, SR Bloom, WS Dhillo. J Clin Endocrinol Metab 2014 (Accepted for publication)

*=joint first authors

205

Publications directly arising from thesis J C E M O N L I N E

H o t T o p i c s i n T r a n s l a t i o n a l E n d o c r i n o l o g y — E n d o c r i n e R e s e a r c h

The Effects of Kisspeptin-10 on Reproductive Hormone Release Show Sexual Dimorphism in Humans

Channa N. Jayasena, Gurjinder M. K. Nijher, Alexander N. Comninos, Ali Abbara, Adam Januszewki, Meriel L. Vaal, Labosshy Sriskandarajah, Kevin G. Murphy, Zohreh Farzad, Mohammad A. Ghatei, Stephen R. Bloom, and Waljit S. Dhillo

Section of Investigative Medicine, Imperial College London, Hammersmith Hospital, London W12 ONN, United Kingdom

Background: Kisspeptin peptides are critical in human reproductive physiology and are potential therapies for infertility. Kisspeptin-10 stimulates gonadotropin release in both male and female rodents. However, few studies have investigated the effects of kisspeptin-10 on gonadotropin release in humans, and none have investigated the effect in women. If kisspeptin is to be useful for treating reproductive disease, its effects in both men and women must be established.

Aim: To compare the effects of kisspeptin-10 administration on reproductive hormone release in healthy men and women.

Methods: Intravenous bolus kisspeptin-10 was administered to men and women (n = 4 –5 per group). Subcutaneous bolus and iv infusion of kisspeptin-10 was also administered to female women (n = 4 –5 per group). Circulating reproductive hormones were measured.

Results: In healthy men, serum LH and FSH were elevated after iv bolus kisspeptin-10, at doses as low as 0.3 and 1.0 nmol/kg, respectively. In healthy women during the follicular phase of the menstrual cycle, no alterations in serum gonadotropins were observed after iv bolus, sc bolus, or iv infusion of kisspeptin-10 at maximal doses of 10 nmol/kg, 32 nmol/kg, and 720pmol/kg/min, respectively. In women during the pre- ovulatory phase, serum LH and FSH were elevated after iv bolus kisspeptin-10 (10 nmol/kg).

Conclusion: Kisspeptin-10 stimulates gonadotropin release in men as well as women during the preovulatory phase of menstrual cycle but fails to stimulate gonadotropin release in women during the follicular phase. The sexual dimorphism of the responsiveness of healthy men and women to kisspeptin-10 administration has important clinical implications for the potential of kisspeptin-10 to treat disorders of reproduction. (J Clin Endocrinol Metab 96: E1963–E1972, 2011)

he kisspeptins are critical regulators of mammalian re- Central or peripheral administration of the shorter kiss- T productive physiology (1, 2). In humans, inactivating peptin peptide, kisspeptin-10, has been demonstrated to mutations of the kisspeptin receptor (KISS1R) cause puber- stimulate gonadotropin release in several mammalian spe- tal failure (3, 4), and activating mutations can lead to pre- cies (8 –18). This effect is abolished by preadministration cocious puberty (5). The human kisspeptin peptides, kiss- of an antagonist to the hypothalamic hormone GnRH (10, peptin-10, -13, -14, and -54, are named according to their 15). Kisspeptin-10 is therefore thought to stimulate the number of constituent amino acids (6). All kisspeptin pep- release of GnRH from the hypothalamus, which then stim- tides share the C-terminal decapeptide sequence, kisspeptin- ulates gonadotropin release from the pituitary gland. The 10, which is required for biological activity in vitro (7). longer kisspeptin peptide, kisspeptin-54, has also been

ISSN Print 0021-972X ISSN Online 1945-7197 Abbreviations: AUC, Area under the curve; IR, immunoreactivity. Printed in U.S.A. Copyright © 2011 by The Endocrine Society doi: 10.1210/jc.2011-1408 Received May 3, 2011. Accepted September 13, 2011. First Published Online October 5, 2011

J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972 jcem.endojournals.org E1963

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. E1964 Jayasena et al. Kisspeptin-10 in Men and Women J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972

shown to stimulate gonadotropin release in rodents (19 – Study 2: effects of iv bolus injection of kisspeptin- 21). In addition, administration of kisspeptin-54 stimu- 10 or kisspeptin-54 in healthy female volunteers lates gonadotropin secretion in humans (22–25). Kisspep- Follicular phase of the menstrual cycle tin therefore has the potential to become a novel therapy Women between d 2–10 of their menstrual cycle were ad- for treatment of reproductive disorders in humans. ministered an iv bolus injection of 0.9% saline, kisspeptin-54 The shorter amino acid sequence of kisspeptin-10 (1.0 nmol/kg), or kisspeptin-10 (at doses of 1.0, 3.0, or 10 nmol/ makes it simpler and cheaper to synthesize than kisspep- kg) at time 0 min, and blood samples were taken at -30, 0, 10, 20, 30, 40, 50, 60, 75, 90, 120, 150, and 180 min (n = 4 –5 per tin-54. Future kisspeptin therapies may therefore be based group). upon kisspeptin-10 rather than kisspeptin-54. It is therefore therapeutically important to determine whether kiss- Preovulatory phase of the menstrual cycle peptin-10 can stimulate reproductive hormone release Women 15–16 d before their next predicted period received in healthy men and women. Kisspeptin-10 stimulates iv bolus injection of 10 nmol/kg kisspeptin-10 as described for the follicular phase (n = 5). gonadotropin release in male rhesus monkeys (13–15). Furthermore, two recent reports have suggested that Study 3: effects of sc bolus injection of saline or kisspeptin-10 administration to healthy men stimulates kisspeptin-10 in healthy female volunteers in the gonadotropin release (26, 27). Although kisspeptin-10 follicular phase of the menstrual cycle is known to stimulate gonadotropin release in animals Kisspeptin-10 (at doses 2, 4, 8, 16, or 32 nmol/kg) or 0.9% (8, 9, 12, 17), there are no published data examining the saline was administered sc at time 0 min, and blood samples were effects of administering kisspeptin-10 to female pri- taken at -30, 0, 15, 30, 45, 60, 75, 90, 120, 150, 180, 210, and 240 min (n = 4 –5 per group). mates or humans. This study aimed to determine the effects of kisspep- Study 4: effects of iv infusion of saline or tin-10 administration on reproductive hormone release in kisspeptin-10 in healthy female volunteers in the healthy men and, for the first time, in healthy women. follicular phase of the menstrual cycle Kisspeptin-10 was dissolved in saline containing gelofusine

(5% vol/vol) (B. Braun Medical, Sheffield, UK) and was infused Subjects and Methods iv over 90 min. During the first 30 min of infusion, the volunteers were administered 20, 50, 90, 180, 360, or 720 pmol/kg/min.

The infusion rate for each volunteer was then halved for the Subjects remaining 60 min of each infusion. Blood samples were taken at The study was conducted with Ethics Committee approval -30, 0, 15, 30, 45, 60, 75, 90, 120, 150, 180, 210, and 240 min. (reference 08/H0707/95) in accordance with The Declaration of Helsinki. Written informed consent was obtained from all sub- Study 5: determining the half-life of kisspeptin-10 jects. Thirty-five healthy female subjects and 11 healthy male in healthy female and male volunteers subjects were recruited, using criteria summarized in Supple- To determine the plasma half-life of kisspeptin-10 in men, mental Table 1 (published on The Endocrine Society’s Journals and in women during the follicular and preovulatory phases of Online web site at http://jcem.endojournals.org). the menstrual cycle, frequent blood sampling was performed during iv infusion of 360 pmol/kg/min kisspeptin-10. The pro- Study days tocol was identical to that used in study 4; except that detailed Subjects were admitted to our Clinical Investigation Unit and blood sampling was performed at 1-min (from 91–100 min) and asked to lay supine for the duration of each study. Urine was 2-min (from 102–120 min) intervals immediately after stopping tested to exclude pregnancy in women (Clearview Easy-HCG; kisspeptin-10 infusion. Blood samples were assayed for plasma Inverness Medical Innovations Inc., Waltham, MA). All blood kisspeptin IR. The decay curve of kisspeptin IR was used to samples were analyzed for measurement of serum LH, FSH, es- calculate the half-time of disappearance (t1/2) for infused kiss- tradiol (in women) or testosterone (in men), and plasma kisspeptin-10 as described previously (22). peptin immunoreactivity (IR). Heart rate, blood pressure, and the presence of adverse symptoms were recorded at regular Data analysis intervals. Data are presented as mean :± SEM. Time profiles of hormone levels were compared using two-way ANOVA with Bonferroni’s Study 1: effects of iv bolus injection of saline or multiple-comparison test. Pairs of means were compared with unpaired t tests (or Mann-Whitney U test if nonparametric), and kisspeptin-10 in healthy male volunteers multiple means of area under curve (AUC) reproductive hor- Intravenous bolus injection of 0.9% saline or kisspeptin-10 mone release were compared using one-way ANOVA with Bon- (at doses of 0.3, 1.0, 3.0, or 10 nmol/kg) was administered at time feronni’s multiple-comparison test (or Kruskall-Wallis with 0 min. Blood samples were taken at -30, 0, 10, 20, 30, 40, 50, Dunn’s multiple-comparison tests if nonparametric). Slopes of 60, 75, 90, 120, 150, and 180 min (n = 4 –5 per group). linear regression curves were compared using an F test. In all

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972 jcem.endojournals.org E1965

cases, P < 0.05 was considered statistically significant. All data nificantly increased compared with saline injection after iv of serum reproductive hormones during treatment are presented bolus injection of 1.0 or 3.0 nmol/kg kisspeptin-10 (Fig. 1, as increases in serum levels after injection when compared with C and G). Serum testosterone was significantly in- preinjection levels. creased compared with saline injection after iv bolus injection of 0.3 or 1.0 nmol/kg kisspeptin-10 (Fig. 1, D

Results and H). Serum levels of testosterone at these doses steadily increased to peak levels 150 –180 min after Baseline characteristics of the subjects recruited to the injection (Fig. 1D). study are summarized in Table 1. Study 2: effects of iv bolus injection of saline or Study 1: effects of iv bolus injection of saline or kisspeptin-10 or kisspeptin-54 in healthy female kisspeptin-10 in healthy male volunteers volunteers Plasma kisspeptin IR was elevated after iv bolus injec- Follicular phase of the menstrual cycle tion of kisspeptin-10 at all doses in healthy male volunteers (Fig. 1, A and E). The highest plasma kisspeptin IR Plasma kisspeptin IR was elevated after iv bolus injec- was observed after 10 nmol/kg kisspeptin-10 (mean AUC tion of kisspeptin-10 at all doses in healthy female volun- kisspeptin IR was 700 :± 160 h·pmol/liter, P < 0.001 vs. teers during the follicular phase of menstrual cycle (Fig. 2, saline). At this dose, mean peak kisspeptin IR (3350 :± 725 A and E). The highest plasma kisspeptin IR was observed pmol/liter) was observed 10 min after injection, and after iv bolus injection of 10 nmol/kg kisspeptin-10 (mean plasma kisspeptin IR returned to undetectable levels 50 AUC kisspeptin IR in women during follicular phase was min after injection. Serum LH was elevated significantly 527 :± 108 h·pmol/liter, P < 0.001 vs. saline); this was after administration of each tested dose of iv bolus kiss- lower when compared with kisspeptin IR after injection of peptin-10 injection compared with saline (Fig. 1, B and F). the same dose of kisspeptin-10 to men, but this difference Peak stimulation of serum LH was observed 30 – 40 min was not significant (P = 0.42 vs. men). At this dose, mean after injection, and levels of serum LH gradually returned peak kisspeptin IR (2638 :± 302 pmol/liter) was observed to baseline 180 min after injection (Fig. 1B). Maximal 10 min after injection, and plasma kisspeptin IR returned stimulation of LH was observed after iv bolus 10 nmol/kg to undetectable levels 50 min after injection. Gonadotro- kisspeptin-10 (mean AUC LH increase was 6.1 :± 1.3 h·IU/ pin release was elevated significantly after iv bolus injec- liter, P < 0.001 vs. saline) (Fig. 1F). Serum FSH was sig- tion of 1 nmol/kg kisspeptin-54 [mean AUC increase (in h·IU/liter): 27.0 :± 11.8 (LH), P < 0.05 vs. saline, and 7.9 :± 4.1 (FSH), P < 0.05 vs. saline). Unexpectedly, no TABLE 1. Baseline characteristics of healthy male and significant changes in serum levels of reproductive hor- female volunteers administered kisspeptin-10 mones were observed after iv bolus injection of kisspep- Healthy male Healthy female tin-10 at doses up to 10 nmol/kg (Fig. 2, B–D and F–H). Baseline volunteers (n volunteers characteristic = 11) (n = 35) Preovulatory phase of the menstrual cycle Age (yr) 28.8 :± 2.1 30.4 :± 0.19 Kisspeptin IR was elevated significantly in women dur- BMI (kg/m2) 24.5 :± 0.5 22.9 :± 0.1 Length of menstrual 27 :± 0.1 ing the preovulatory phase after kisspeptin-10 injection cycle (d) compared with saline, and this elevation was nonsignifi- LH (IU/liter) 2.9 :± 0.2 cantly different when compared with kisspeptin IR after Follicular 3.8 :± 0.6 a the same dose of kisspeptin-10 in follicular-phase women Preovulatory 6.8 :± 1.6 FSH (IU/liter) 2.6 :± 0.2 or men (mean AUC kisspeptin IR in preovulatory phase Follicular 3.7 :± 0.4 was 320 :± 56 h·pmol/liter, P = 0.13 vs. follicular phase, b Preovulatory 6.8 :± 1.6 and P = 0.06 vs. men) (Fig. 2E). Serum LH and FSH were Testosterone 21.6 :± 1.5 (nmol/liter) elevated significantly after iv bolus injection of 10 nmol/kg Estradiol (pmol/liter) kisspeptin-10 in women during the preovulatory phase of Follicular 256 :± 43 a the menstrual cycle [mean AUC increase was 30.3 :± 7.7 Preovulatory 562 :± 195 h·IU/liter (LH), P < 0.05 vs. saline, and 6.9 :± 0.9 h·IU/liter Female endocrine profiles are presented during the follicular and (FSH), P < 0.01 vs. saline] (Fig. 2, B, C, F, and G); how- preovulatory phases of the menstrual cycle. Data are shown as mean :± SEM. ever, serum estradiol was not altered significantly (mean a P < 0.05 vs. follicular phase of the menstrual cycle. AUC estradiol increase was 111 :± 96 h·pmol/liter, P = b P < 0.01 vs. follicular phase of the menstrual cycle. 0.14 vs. saline) (Fig. 2, D and H).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. E1966 Jayasena et al. Kisspeptin-10 in Men and Women J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972

Study 4: effects of iv infusion of kisspeptin-10 in healthy female volunteers in the follicular phase of the menstrual cycle Intravenous infusion of peptides delivers the peptide directly into the circulation, avoiding possible degra- dation in sc tissue, and results in a sustained increase in circulating levels of the administered peptide. Plasma kisspeptin IR increased during iv infusion of kisspeptin-10 at all doses (Fig. 4A). The highest plasma kisspeptin IR was observed during iv infusion of 720 pmol/kg/min kisspeptin-10 (mean AUC kisspeptin IR was 2518 :± 100 h·pmol/liter). All iv infusion doses of kisspeptin-10 were associated with a higher mean plasma kisspeptin IR than the highest studied sc dose of kisspeptin-10 (32 nmol/kg, mean AUC kisspeptin IR was 201 :± 16 h·pmol/liter). No significant changes in serum reproductive hormone levels were observed after iv infusion of any dose of kisspeptin-10 in healthy female volunteers in the follicular phase of the menstrual cycle (Fig. 4, B–H).

Study 5: pharmacokinetic profile of kisspeptin IR during iv infusion FIG. 1. Plasma kisspeptin IR and serum reproductive hormone levels after iv bolus injection of of kisspeptin-10 in healthy male kisspeptin-10 to healthy male volunteers. A–D, Time profiles for plasma kisspeptin IR (A) and and female volunteers changes in serum LH (B), FSH (C), and testosterone (D) during 4 h after iv bolus injection of To determine the plasma half-life of saline or kisspeptin-10 to healthy male volunteers (n = 4 –5 per group). For 10 nmol/kg vs. saline: >\, P < 0.05; >\>\, P < 0.01; >\>\>\, P < 0.001. For 3 nmol/kg vs. saline: >, P < 0.05; >>>, kisspeptin-10, frequent blood sampling P < 0.001. For 1 nmol/kg vs. saline: a, P < 0.05; aaa, P < 0.001. E–H, AUC for plasma was performed in women during the kisspeptin IR (E) and changes in serum LH (F), FSH (G), and testosterone (H) during 4 h after iv follicular and preovulatory phases of bolus injection of saline or kisspeptin-10 to healthy male volunteers (n = 4 –5 per group). *, P < 0.05; **, P < 0.01; ***, P < 0.001. T, Testosterone. Data are shown as mean :± SEM. menstrual cycle, and in men, after cessation of an iv infusion of 360 pmol/ kg/min kisspeptin-10 (Fig. 5). When Study 3: effects of sc bolus injection of saline or plotted on a natural log scale, linear regression slopes were kisspeptin-10 in healthy female volunteers in the not significantly different among the three groups (F = follicular phase of the menstrual cycle 0.099; degrees of Freedom numerator = 2; degrees of Free- Plasma kisspeptin IR was significantly elevated during dom denominator = 111; P = 0.91), and plasma half-lives the 4 h after sc injection of kisspeptin-10 at doses of 4 of kisspeptin-10 were calculated as 3.8 :± 0.3 min (men), nmol/kg and higher, when compared with saline (Fig. 3A). 4.1 :± 0.4 min (follicular-phase women), and 4.1 :± 0.4 The highest plasma kisspeptin IR after sc injection was min (preovulatory-phase women) (Fig. 5E). observed after injection of 32 nmol/kg kisspeptin-10 (mean AUC kisspeptin IR was 201 :± 16 h·pmol/liter, P <

0.001 vs. saline). Discussion No significant changes in serum reproductive hormone levels were observed after sc bolus injection of kisspep- These studies reveal a previously unknown sexual dimor- tin-10 at any dose (Fig. 3, B–H). phism in responsiveness to kisspeptin-10 administration

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972 jcem.endojournals.org E1967

tested. These results are consistent with our previous observation that administration of the longer form of kisspeptin, kisspeptin-54, stimulates reproductive hormone release in men (22). Further- more, two recent reports have shown that iv injection of kisspeptin-10 potently stimulates LH release in healthy male volunteers at doses similar to those used in this study (26, 27) and that prolonged infusion of kisspeptin-10 significantly stimulates LH pulsatility in healthy male volunteers (27). Studies in rodents suggest that kisspeptin-10 stimulates gonadotropin release in both males and females (8, 9). Furthermore, we observed that the kisspeptin-10 peptide used in this study stimulated LH release in female adult mice (Supplemental Fig. 1), although with a much lower potency when compared with kisspeptin-54. We were therefore surprised to observe that doses of kisspeptin-10 identical to those used in men failed to stimulate reproductive hormone release in healthy female volunteers during the follicular phase of the menstrual cycle when administered as an iv bolus injection. It is possible that the lack of effect of iv administration of kiss-

peptin-10 to stimulate reproductive hor- FIG. 2. Plasma kisspeptin IR and serum reproductive hormone levels after iv bolus injection of kisspeptin-10 to healthy female volunteers. A–D, Time profiles for plasma kisspeptin IR (A) mone release in women in the follicular and changes in serum LH (B), FSH (C), and estradiol (D) during 4 h after iv bolus injection of phase of their menstrual cycle was due saline, kisspeptin-10 (KP10), or kisspeptin-54 (KP54) to healthy female volunteers. For 10 to the short period of time that circu- nmol/kg KP10 vs. saline: >\, P < 0.05; >\>\, P < 0.01; >\>\>\, P < 0.001. For 3 nmol/kg KP10 vs. saline: >>>, P < 0.001. For 1 nmol/kg KP54 vs. saline: K, P < 0.05; KK, P < 0.01; KKK, P < lating kisspeptin-10 levels were ele- 0.001. For 10 nmol/kg KP10 (Preov) vs. saline: *, P < 0.05; **, P < 0.01; ***, P < 0.001. E– vated after iv bolus administration. H, AUC for plasma kisspeptin IR (E) and changes in serum LH (F), FSH (G), and estradiol (H) Subcutaneous bolus administration of during 4 h after iv bolus injection of saline or kisspeptin-10 to healthy female volunteers. kisspeptin-10 is likely to result in a lon- *, P < 0.05; **, P < 0.01; ***, P < 0.001. Data are shown as mean :± SEM. E2, Estradiol; Preov, preovulatory phase of the menstrual cycle. ger period of elevated circulating levels of kisspeptin-10 than iv bolus administration (22, 23). Furthermore, we in healthy men and women. Numerous studies in animals have previously determined that sc bolus injection of kiss- have demonstrated that kisspeptin-10 robustly stimulates peptin-54 at doses as low as 0.4 nmol/kg potently stimu- gonadotropin release, thus implicating kisspeptin signaling lates serum LH in healthy women during the follicular as a potential therapeutic target for treating female patients with infertility (8, 9, 12, 17). Kisspepin-10 stimulates gonad- phase of the menstrual cycle (23). However, serum go- otropin release in male rhesus monkeys (13–15), and recent nadotropin levels were not elevated by sc bolus injection reports have suggested that kisspepeptin-10 administration of kisspeptin-10 in healthy women in the follicular phase stimulates gonadotropin release in healthy men (26, 27). of the menstrual cycle, despite elevations in plasma kiss- However, the effects of administration of kisspeptin-10 in peptin IR for up to 90 min after injection. female primates or women have not previously been studied. The lack of effect of sc bolus administration of kiss- We observed that iv bolus injection of kisspeptin-10 peptin-10 to stimulate reproductive hormone release in robustly stimulated LH release in healthy men at all doses women in the follicular phase of their menstrual cycle may

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. E1968 Jayasena et al. Kisspeptin-10 in Men and Women J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972

the follicular phase of the menstrual cycle, despite markedly raised plasma levels of kisspeptin IR up to concentrations of 2000 pmol/liter. Our data therefore suggest that women in the follicular phase of the menstrual cycle are markedly less responsive to kisspeptin-10 administration than men. The underlying mechanism for this observation is unclear. However, sexual dimorphism of hypothalamic kisspeptin signaling pathways has been demonstrated in rodents (28 –30). It is important to consider whether the contrasting effects of kisspeptin-10 in men and women may have been attributable to factors other than differential sensitivity to the peptide. We observed that levels of plasma kisspeptin IR appeared slightly lower in women when compared with men when identical weight-adjusted doses of kisspeptin-10 were administered. It is possible that levels of kisspeptin-10 may have been modified by factors known to differ between the sexes, such as body fat content or clearance of peptide from the circulation (19). However, it is noteworthy that only a marginal elevation of plasma kisspeptin IR (approximately 10 h·pmol/liter) was necessary to stimulate significant LH secretion in men after injection of kisspeptin-10 (0.3 nmol/kg iv bolus); by contrast, a 200- fold greater elevation in plasma kiss-

peptin IR failed to stimulate LH release FIG. 3. Plasma kisspeptin IR and serum reproductive hormone levels after sc bolus injection of kisspeptin-10 to healthy female volunteers in the follicular phase of the menstrual cycle. in women during the follicular phase of A–D, Time profiles for plasma kisspeptin IR (A) and changes in serum LH (B), FSH (C), and the menstrual cycle. Furthermore, we estradiol (D) during 4 h after sc bolus injection of kisspeptin-10 to healthy female volunteers observed no significant differences in in the follicular phase of the menstrual cycle (n = 4 –5 per group). For 32 vs. 2 nmol/kg: *, P < 0.05; ***, P < 0.001. For 16 vs. 2 nmol/kg: ƒƒƒ, P < 0.001. For 8 vs. 2 nmol/kg: >\>\>\, the plasma half-lives of kisspeptin-10 P < 0.001. For 4 vs. 2 nmol/kg: >, P < 0.05. E–H, AUC for plasma kisspeptin IR (E) and between men and women. It is therefore changes in serum LH (F), FSH (G), and testosterone (H) during 4 h after sc bolus injection of unlikely that the striking contrast be- kisspeptin-10 to healthy female volunteers in the follicular phase of the menstrual cycle (n = tween male and female responsiveness 4 –5 per group). *, P < 0.05; **, P < 0.01; ***, P < 0.001. E2, Estradiol. Data are shown as mean :± SEM. to kisspeptin-10 administration reflects differences in metabolism of kisspeptin-10. We observed nonsignificant re- have been due to breakdown of kisspeptin-10 in the sc ductions in serum LH (P = 0.19) and FSH (P = 0.91) after tissue. Intravenous infusion of peptides delivers the pep- saline injection in men when compared with saline injec- tide directly into the circulation, avoiding possible degra- dation in sc tissue, and results in a sustained increase in tion in women. Although both men and women under- circulating levels of the administered peptide. However, went identical study protocols, it is possible that minor serum gonadotropin levels were not elevated during iv differences in factors inhibiting reproductive function infusion of kisspeptin-10 to healthy female volunteers in such as stress may be greater in the male group. However,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972 jcem.endojournals.org E1969

kisspeptin-54 when compared with kisspeptin-10 may explain why exogenous kisspeptin-54 injection (but not exogenous kisspeptin-10) stimulates reproductive hormone release in women during the follicular phase of the menstrual cycle. The differences between kisspeptin-10 and kisspeptin-54 may be a consequence of the rapid breakdown of kisspeptin-10 in the circulation. The in vivo plasma half-life of iv kisspeptin-10 in this study was calculated to approximately 4 min, which is 7-fold shorter than the calculated in vivo plasma half-life of kisspeptin-54 (22). It is interesting to consider whether men and women have differential responses to kisspeptin-54 (as they appear to be to kisspeptin-10) based on previous studies. Intravenous infusion of kisspeptin-54 at doses as low as 1.2 nmol (0.25 pmol/kg/min) stimulate LH secretion in healthy men (22). By comparison, sc bolus injection of kisspeptin-54 at doses of 28 nmol (0.4 nmol/ kg) or more stimulate LH secretion in healthy women during the follicular phase of the menstrual cycle (23); fur-

thermore, this study suggests that iv bo- FIG. 4. Plasma kisspeptin IR and serum reproductive hormone levels during iv infusion of kisspeptin-10 to healthy female volunteers in the follicular phase of the menstrual cycle. A–D, lus injection of approximately 60 nmol Time profiles for plasma kisspeptin IR (A) and changes in serum LH (B), FSH (C), and estrogen (1 nmol/kg) kisspeptin-54 stimulates sig- (D) during 4 h after commencement of a 90-min iv infusion of kisspeptin-10 to healthy female nificant LH secretion in these women. volunteers in the follicular phase of the menstrual cycle (n = 4 –5 per group). For 720 vs. 20 pmol/kg · min: ***, P < 0.001. For 360 vs. 20 pmol/kg · min: ƒƒƒ, P < 0.001. For 180 vs. 20 These data demonstrate that, unlike pmol/kg · min: >\, P < 0.05; >\>\>\, P < 0.001. For 50 vs. 20 pmol/kg · min: a, P < 0.05. E–H, kisspeptin-10, kisspeptin-54 can stim- AUC for plasma kisspeptin IR (E) and changes in serum LH (F), FSH (G), and testosterone (H) ulate reproductive hormone release in during 4 h after commencement of a 90-min iv infusion of kisspeptin-10 to healthy female volunteers in the follicular phase of the menstrual cycle (n = 4 –5 per group). **, P < 0.01; women in the follicular phase of their ***, P < 0.001. E2, Estradiol. Data are shown as mean :± SEM. menstrual cycle. In the current study, gonadotropin responses during kisspeptin-10 infu- such a difference would not explain why men are more sion appeared more variable than after iv or sc bolus in- responsive to kisspeptin-10 than women in the follicular jections. These alterations in gonadotropin secretion did phase of the menstrual cycle. not appear to be related to either the dose or to the onset In the current study, we observed that exogenous kiss- of kisspeptin infusion; however, it is possible that subtle peptin-54 stimulated significant gonadotropin secretion in women during the follicular phase of the menstrual effects of kisspeptin-10 infusion on gonadotropin release cycle, despite a failure of exogenous kisspeptin-10 to stim- were not detected in the current study protocol. ulate gonadotropin secretion at a 10-fold higher molar We have previously demonstrated that women in the dose. Furthermore, kisspeptin-54 injection stimulated LH preovulatory phase of the menstrual cycle are significantly in female mice more potently when compared with kiss- more sensitive to the effects of kisspeptin-54 on gonado- peptin-10, and it has been previously observed that kiss- tropin release when compared with women in the follic- peptin-52 (31) and kisspeptin-54 (21) stimulate LH more ular phase of the menstrual cycle (23). In keeping with potently when compared with kisspeptin-10 in male rats. these observations, iv bolus injection of kisspeptin-10 sig- Collectively, these data suggest that the greater potency of nificantly stimulated LH and FSH release in women dur-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. E1970 Jayasena et al. Kisspeptin-10 in Men and Women J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972

secretion (30 min after injection); this phenomenon was also observed after sc bolus injection of kisspeptin-54 in healthy women (23). Our results therefore suggest that both kisspeptin-10 and -54 stimulate LH secretion more potently and more rapidly than FSH. It is interesting to consider why we did not observe any consistent stimulation of testosterone secretion after iv bolus kisspeptin-10 injection in healthy men when compared with the robust increases in serum LH and FSH observed at all tested doses. Significant increases in serum testosterone were observed only at 0.3 and 1.0 nmol/kg iv bolus kisspeptin-10, but these rises were marginal (no more than 10 h·nmol/liter above baseline). Furthermore, iv bolus injection of kisspeptin-10 stimulated go- FIG. 5. Detailed time profiles of plasma kisspeptin IR after iv infusion of kisspeptin-10 to nadotropin release in women during healthy male and female volunteers. A–D, Blood sampling was performed for measurement the follicular phase of the menstrual cy- of plasma kisspeptin IR during 4 h after commencement of a 90-min iv infusion of kisspeptin- cle but did not increase serum estradiol 10 (360 pmol/kg · min) to healthy male volunteers (A and D) and female volunteers during the follicular (B and D) and preovulatory (C and D) phases of the menstrual cycle (n = 4 –5 per during 3 h after injection. Our previous group). For men vs. women during follicular phase: <1<1<1, P < 0.001; <1, P < 0.05. For men vs. data suggest that at least 4 h are re- women during preovulatory phase: ***, P < 0.001; *, P < 0.05. E, Blood sampling was quired for serum levels of sex steroids performed at 1-min intervals immediately after stopping infusion of kisspeptin-10. Linear curve fits of natural log plasma kisspeptin IR (dotted lines) were similar for all three groups to peak after a sc bolus injection of (P = 0.91). Data are shown as mean :± SEM. kisspeptin-54 (23–25). A longer period of blood sampling after injection may have revealed more pronounced ing the preovulatory phase of the menstrual cycle in the alterations in sex steroid secretion in subjects after injec- current study. Because we observed that kisspeptin-10 has tion of kisspeptin-10. We observed that serum gonado- a similar pharmacokinetic profile in both follicular and tropins almost returned to baseline levels within 3 h after preovulatory phases of the menstrual cycle, our results iv bolus injection of kisspeptin-10; it is possible that iv suggest that women have heightened sensitivity to kiss- bolus injection of kisspeptin-10 has a duration of action peptin-10 during the preovulatory phase of the menstrual inadequate to stimulate significant gonadal sex steroid cycle. In female rodents, c-fos expression within kisspeptin release. neurons and levels of kiss1 expression are increased within the anteroventral periventricular nucleus of the hypothal- In the current study, all tested iv bolus doses of kiss- amus immediately before ovulation (35). Furthermore, peptin-10 (0.3–10 nmol/kg) were associated with similar levels of kiss1r expression are increased in rat hypotha- degrees of gonadotropin secretion in healthy male sub- lamic fragments at diestrus when compared with proestrus jects. A recent study by George et al. (27) suggests that (36). It is therefore possible that women in the preovula- kisspeptin-10 stimulates serum LH secretion at doses as tory phase of the menstrual cycle have heightened respon- low as 0.01 fLg/kg (equivalent to 0.008 nmol/kg), with a siveness to kisspeptin-10 administration when compared maximal response seen at 1 fLg/kg (0.8 nmol/kg) in men. with women during the follicular phase of the menstrual Taking these data into consideration, all doses of kisspep- cycle. tin-10 selected during our study may have stimulated near- In the current study, we observed that kisspeptin-10 maximal levels of gonadotropin secretion in healthy male stimulated LH secretion more potently than FSH. This is volunteers. Interestingly, George et al. (27) observed that consistent with our previous studies of kisspeptin-54 ad- the increase in serum LH at 3 fLg/kg (2.4 nmol/kg) was ministration in healthy men and women. After iv bolus lower than the increase in serum LH at 1 fLg/kg, which they kisspeptin-10 injection, peak FSH secretion was observed speculated was attributable to tachyphylaxis, a phenom- 45–150 min after injection, which was later than peak LH enon that we have previously observed after chronic kiss-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972 jcem.endojournals.org E1971

peptin-54 injections (24, 25, 27). We therefore cannot P, Steplewski K, Shabon U, Miller JE, Middleton SE, Darker JG, exclude that the similarity between LH responses at doses Larminie CG, Wilson S, Bergsma DJ, Emson P, Faull R, Philpott KL, Harrison DC 2001 AXOR12, a novel human G protein-coupled between 0.3 and 10 nmol/kg kisspeptin-10 in men might receptor, activated by the peptide KiSS-1. J Biol Chem 276:28969 – be explained in part by tachyphylaxis to kisspeptin-10 at 28975 the higher tested doses. Additional studies are required to 3. de Roux N, Genin E, Carel JC, Matsuda F, Chaussain JL, Milgrom E 2003 Hypogonadotropic hypogonadism due to loss of function of investigate these observations. the KiSS1-derived peptide receptor GPR54. Proc Natl Acad Sci USA In summary, this is the first clinical study to report the 100:10972–10976 effects of kisspeptin-10 administration on gonadotropin 4. Seminara SB, Messager S, Chatzidaki EE, Thresher RR, Acierno Jr release in women and to compare the effects of kisspep- JS, Shagoury JK, Bo-Abbas Y, Kuohung W, Schwinof KM, Hendrick AG, Zahn D, Dixon J, Kaiser UB, Slaugenhaupt SA, Gusella JF, tin-10 between men and women. Kisspeptin-10 robustly O’Rahilly S, Carlton MB, Crowley Jr WF, Aparicio SA, Colledge stimulates gonadotropin release in men but fails to stim- WH 2003 The GPR54 gene as a regulator of puberty. N Engl J Med ulate gonadotropin release in healthy female volunteers in 349:1614 –1627 5. Teles MG, Bianco SD, Brito VN, Trarbach EB, Kuohung W, Xu S, the follicular phase of the menstrual cycle when adminis- Seminara SB, Mendonca BB, Kaiser UB, Latronico AC 2008 A tered by iv bolus injection, sc bolus injection, or iv infu- GPR54-activating mutation in a patient with central precocious pu- sion. These experiments reveal sexual dimorphism in the berty. N Engl J Med 358:709 –715 6. Bilban M, Ghaffari-Tabrizi N, Hintermann E, Bauer S, Molzer S, responsiveness of healthy human volunteers to kisspep- Zoratti C, Malli R, Sharabi A, Hiden U, Graier W, Kno¨ fler M, tin-10 administration. These findings have important clin- Andreae F, Wagner O, Quaranta V, Desoye G 2004 Kisspeptin-10, ical implications for the potential therapeutic use of kiss- a KiSS-1/metastin-derived decapeptide, is a physiological invasion peptin-10 to treat disorders of reproduction. inhibitor of primary human trophoblasts. J Cell Sci 117:1319 –1328 7. Kotani M, Detheux M, Vandenbogaerde A, Communi D, Vander-

winden JM, Le Poul E, Bre´ zillon S, Tyldesley R, Suarez-Huerta N, Vandeput F, Blanpain C, Schiffmann SN, Vassart G, Parmentier M 2001 The metastasis suppressor gene KiSS-1 encodes kisspeptins, Acknowledgments the natural ligands of the orphan G protein-coupled receptor GPR54. J Biol Chem 276:34631–34636 We are grateful to the Wellcome Trust and Sir John McMichael 8. Irwig MS, Fraley GS, Smith JT, Acohido BV, Popa SM, Cunningham Centre for providing infrastructure for this study. MJ, Gottsch ML, Clifton DK, Steiner RA 2004 Kisspeptin activation of gonadotropin releasing hormone neurons and regulation of Address all correspondence and requests for reprints to: Prof. KiSS-1 mRNA in the male rat. Neuroendocrinology 80:264 –272 Stephen R. Bloom, Section of Investigative Medicine, Imperial 9. Navarro VM, Castellano JM, Ferna´ ndez-Ferna´ ndez R, Barreiro College London, 6th Floor, Commonwealth Building, Hammer- ML, Roa J, Sanchez-Criado JE, Aguilar E, Dieguez C, Pinilla L, smith Hospital, Du Cane Road, London W12 ONN, United Tena-Sempere M 2004 Developmental and hormonally regulated Kingdom. E-mail: [email protected]. messenger ribonucleic acid expression of KiSS-1 and its putative receptor, GPR54, in rat hypothalamus and potent luteinizing hor- The Section of Investigative Medicine is funded by grants mone-releasing activity of KiSS-1 peptide. Endocrinology 145: from the Medical Research Council (MRC), Biotechnology and 4565– 4574 Biological Sciences Research Council, National Institute of 10. Gottsch ML, Cunningham MJ, Smith JT, Popa SM, Acohido BV, Health Research (NIHR), an Integrative Mammalian Biology Crowley WF, Seminara S, Clifton DK, Steiner RA 2004 A role for Capacity Building Award, and an FP7-HEALTH-2009-241592 kisspeptins in the regulation of gonadotropin secretion in the mouse. EurOCHIP Grant and is supported by the NIHR Imperial Bio- Endocrinology 145:4073– 4077 medical Research Centre Funding Scheme. This work was also 11. Messager S, Chatzidaki EE, Ma D, Hendrick AG, Zahn D, Dixon J, funded by an MRC project grant. C.N.J. is supported by an Thresher RR, Malinge I, Lomet D, Carlton MB, Colledge WH, NIHR Clinical Lectureship, Society for Endocrinology Early Ca- Caraty A, Aparicio SA 2005 Kisspeptin directly stimulates gonadotropin-releasing hormone release via G protein-coupled receptor reer Grant, and Academy of Medical Sciences/Wellcome Starter 54. Proc Natl Acad Sci USA 102:1761–1766 Grant for Clinical Lecturers. G.M.K.N. is supported by a Well- 12. Caraty A, Smith JT, Lomet D, Ben Saïd S, Morrissey A, Cognie J, come Clinical Training Fellowship. A.N.C. is supported by an Doughton B, Baril G, Briant C, Clarke IJ 2007 Kisspeptin synchro- Imperial College Healthcare NHS Trust Charity Award. W.S.D. nizes preovulatory surges in cyclical ewes and causes ovulation in is supported by an NIHR Career Development Fellowship. seasonally acyclic ewes. Endocrinology 148:5258 –5267, Disclosure Summary: The authors have nothing to disclose. 13. Plant TM, Ramaswamy S, Dipietro MJ 2006 Repetitive activation of hypothalamic G protein-coupled receptor 54 with intravenous pulses of kisspeptin in the juvenile monkey (Macaca mulatta) elicits a sustained train of gonadotropin-releasing hormone discharges. References Endocrinology 147:1007–1013 14. Ramaswamy S, Seminara SB, Pohl CR, DiPietro MJ, Crowley Jr WF, 1. Ohtaki T, Shintani Y, Honda S, Matsumoto H, Hori A, Kanehashi Plant TM 2007 Effect of continuous iv administration of human K, Terao Y, Kumano S, Takatsu Y, Masuda Y, Ishibashi Y, Wa- metastin 45–54 on the neuroendocrine activity of the hypothalamic- tanabe T, Asada M, Yamada T, Suenaga M, Kitada C, Usuki S, pituitary-testicular axis in the adult male rhesus monkey (Macaca Kurokawa T, Onda H, Nishimura O, Fujino M 2001 Metastasis mulatta). Endocrinology 148:3364 –3370 suppressor gene KiSS-1 encodes peptide ligand of a G-protein-cou- 15. Shahab M, Mastronardi C, Seminara SB, Crowley WF, Ojeda SR, pled receptor. Nature 411:613– 617 Plant TM 2005 Increased hypothalamic GPR54 signaling: a poten- 2. Muir AI, Chamberlain L, Elshourbagy NA, Michalovich D, Moore tial mechanism for initiation of puberty in primates. Proc Natl Acad DJ, Calamari A, Szekeres PG, Sarau HM, Chambers JK, Murdock Sci USA 102:2129 –2134

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:20 For personal use only. No other uses without permission. . All rights reserved. E1972 Jayasena et al. Kisspeptin-10 in Men and Women J Clin Endocrinol Metab, December 2011, 96(12):E1963–E1972

16. Greives TJ, Mason AO, Scotti MA, Levine J, Ketterson ED, Kriegs- Subcutaneous injection of kisspeptin-54 acutely stimulates gonad- feld LJ, Demas GE 2007 Environmental control of kisspeptin: im- otropin secretion in women with hypothalamic amenorrhea, but plications for seasonal reproduction. Endocrinology 148:1158 – chronic administration causes tachyphylaxis. J Clin Endocrinol 1166 Metab 94:4315– 4323 17. Lents CA, Heidorn NL, Barb CR, Ford JJ 2008 Central and periph- 25. Jayasena CN, Nijher GM, Abbara A, Murphy KG, Lim A, Patel D, eral administration of kisspeptin activates gonadotropin but not Mehta A, Todd C, Donaldson M, Trew GH, Ghatei MA, Bloom SR, somatotropin secretion in prepubertal gilts. Reproduction 135: Dhillo WS 2010 Twice-weekly administration of kisspeptin-54 for 879 – 887 8 weeks stimulates release of reproductive hormones in women with 18. Kadokawa H, Matsui M, Hayashi K, Matsunaga N, Kawashima C, hypothalamic amenorrhea. Clin Pharmacol Ther 88:840 – 847 Shimizu T, Kida K, Miyamoto A 2008 Peripheral administration of 26. Chan YM, Butler JP, Pinnell NE, Pralong FP, Crowley Jr WF, Ren kisspeptin-10 increases plasma concentrations of GH as well as LH C, Chan KK, Seminara SB 2011 Kisspeptin resets the hypothalamic in prepubertal Holstein heifers. J Endocrinol 196:331–334 GnRH clock in men. J Clin Endocrinol Metab 96:E908 –E915 19. Tovar S, Va´ zquez MJ, Navarro VM, Ferna´ ndez-Ferna´ ndez R, Cas- 27. George JT, Veldhuis JD, Roseweir AK, Newton CL, Faccenda E, tellano JM, Vigo E, Roa J, Casanueva FF, Aguilar E, Pinilla L, Di- Millar RP, Anderson RA 2010 Kisspeptin 10 rapidly and potently eguez C, Tena-Sempere M 2006 Effects of single or repeated intra- stimulates LH secretion in men. J Clin Endocrinol Metab 96:E1228 – venous administration of kisspeptin upon dynamic LH secretion in E1236 conscious male rats. Endocrinology 147:2696 –2704 28. Clarkson J, Herbison AE 2006 Postnatal development of kisspeptin 20. Patterson M, Murphy KG, Thompson EL, Patel S, Ghatei MA, neurons in mouse hypothalamus; sexual dimorphism and projec- Bloom SR 2006 Administration of kisspeptin-54 into discrete re- tions to gonadotropin-releasing hormone neurons. Endocrinology gions of the hypothalamus potently increases plasma luteinising hor- 147:5817–5825 mone and testosterone in male adult rats. J Neuroendocrinol 18: 29. Adachi S, Yamada S, Takatsu Y, Matsui H, Kinoshita M, Takase K, 349 –354 Sugiura H, Ohtaki T, Matsumoto H, Uenoyama Y, Tsukamura H, 21. Thompson EL, Murphy KG, Patterson M, Bewick GA, Stamp GW, Inoue K, Maeda K 2007 Involvement of anteroventral periventricu- Curtis AE, Cooke JH, Jethwa PH, Todd JF, Ghatei MA, Bloom SR lar metastin/kisspeptin neurons in estrogen positive feedback action 2006 Chronic subcutaneous administration of kisspeptin-54 causes on luteinizing hormone release in female rats. J Reprod Dev 53: testicular degeneration in adult male rats. Am J Physiol Endocrinol 367–378 Metab 291:E1074 –E1082 30. Kauffman AS, Gottsch ML, Roa J, Byquist AC, Crown A, Clifton 22. Dhillo WS, Chaudhri OB, Patterson M, Thompson EL, Murphy DK, Hoffman GE, Steiner RA, Tena-Sempere M 2007 Sexual dif- KG, Badman MK, McGowan BM, Amber V, Patel S, Ghatei MA, ferentiation of Kiss1 gene expression in the brain of the rat. Endo- Bloom SR 2005 Kisspeptin-54 stimulates the hypothalamic-pitu- crinology 148:1774 –1783 itary gonadal axis in human males. J Clin Endocrinol Metab 90: 31. Soldin OP, Chung SH, Mattison DR 2011 Sex differences in drug 6609 – 6615 disposition. J Biomed Biotechnol 2011:187103 23. Dhillo WS, Chaudhri OB, Thompson EL, Murphy KG, Patterson 32. Smith JT, Popa SM, Clifton DK, Hoffman GE, Steiner RA 2006 M, Ramachandran R, Nijher GK, Amber V, Kokkinos A, Donald- Kiss1 neurons in the forebrain as central processors for generating son M, Ghatei MA, Bloom SR 2007 Kisspeptin-54 stimulates go- the preovulatory luteinizing hormone surge. J Neurosci 26:6687– nadotropin release most potently during the preovulatory phase of 6694 the menstrual cycle in women. J Clin Endocrinol Metab 92:3958 – 33. Roa J, Vigo E, Castellano JM, Navarro VM, Ferna´ ndez-Ferna´ ndez 3966 R, Casanueva FF, Dieguez C, Aguilar E, Pinilla L, Tena-Sempere M 24. Jayasena CN, Nijher GM, Chaudhri OB, Murphy KG, Ranger A, 2006 Hypothalamic expression of KiSS-1 system and gonadotropin- Lim A, Patel D, Mehta A, Todd C, Ramachandran R, Salem V, releasing effects of kisspeptin in different reproductive states of the Stamp GW, Donaldson M, Ghatei MA, Bloom SR, Dhillo WS 2009 female Rat. Endocrinology 147:2864 –2878

E n d o c r i n e R e s e a r c h

Twice-Daily Subcutaneous Injection of Kisspeptin-54 Does Not Abolish Menstrual Cyclicity in Healthy Female Volunteers

C. N. Jayasena,* A. N. Comninos,* G. M. K. Nijher, A. Abbara, A. De Silva, J. D. Veldhuis, R. Ratnasabapathy, C. Izzi-Engbeaya, A. Lim, D. A. Patel, M. A. Ghatei, S. R. Bloom, and W. S. Dhillo

Department of Investigative Medicine (C.N.J., A.N.C., G.M.K.N., A.A., A.D.S., R.R., C.I.-E., M.A.G., S.R.B., W.S.D.), Imperial College London, Hammersmith Hospital, London W12 0NN, United Kingdom; Endocrine Research Unit (J.D.V.), Center for Translational Science Activities, Mayo Clinic, Rochester, Minnesota 55905; and Imaging Department (A.L., D.A.P.), Imperial College Healthcare National Health Service Trust, Charing Cross Hospital, London W6 8RF, United Kingdom

Background: Kisspeptin is a critical hypothalamic regulator of reproductive function. Chronic kisspeptin administration causes profound tachyphylaxis in male monkeys and in women with functional hypothalamic amenorrhea. The pharmacological effects of chronic kisspeptin exposure in healthy women with normal menstrual cycles have not been studied previously.

Aim: Our aim was to determine the effects of follicular-phase kisspeptin-54 treatment on menstrual cyclicity in healthy women.

Methods: We performed a prospective, single-blinded, 1-way crossover study. Healthy women received twice-daily sc injections of kisspeptin (6.4 nmol/kg) or 0.9% saline during menstrual days 7–14 (n = 5 per treatment arm). Serial assessments of basal reproductive hormones, ultrasound parameters, LH pulsatility, and acute sensitivity to GnRH and kisspeptin-54 injection were performed.

Results: Menstrual cyclicity persisted in all women after follicular-phase kisspeptin-54 treatment. Chronic exposure to kisspeptin-54 did not abolish acute stimulation of LH after injection of kisspeptin-54 or GnRH. In addition, kisspeptin-54 treatment was associated with a shorter mean length of the menstrual cycle (mean length of menstrual cycle was 28.6 ± 1.4 days with saline vs 26.8 ± 3.1 days with kisspeptin, P < .01), earlier onset of highest recorded serum LH (mean menstrual day of highest LH was 15.2 ± 1.3 with saline vs 13.0 ± 1.9 with kisspeptin, P < .05), and earlier onset of the luteal phase (mean menstrual day of progesterone increase was 18.0 ± 2.1 with saline vs 15.8 ± 0.9 with kisspeptin, P < .05).

Conclusion: Our data suggest that 1 week of exogenous kisspeptin-54 does not abolish menstrual cyclicity in healthy women. Further work is needed to determine whether kisspeptin could be used to treat certain anovulatory disorders. (J Clin Endocrinol Metab 98: 4464 – 4474, 2013)

he kisspeptins are a group of arginine-phenylalanine (RF) Mice and humans lacking kisspeptin or the kisspeptin receptor T amide peptides encoded by the KISS1 gene and are en- fail to undergo puberty and are infertile (8 –10). Central or pe- dogenous ligands for the kisspeptin receptor (KISS1R) (1– 4). ripheral administration of kisspeptin induces gonadotropin and KISS1 and KISS1R are expressed predominantly in the hypo- sex steroid release in all mammalian species investigated, in- thalamus, pituitary, and placenta (3, 5–7). Kisspeptin signaling cluding rats (11–13), mice (14, 15), monkeys (16), and sheep exerts powerful effects on the mammalian reproductive system. (17, 18).

ISSN Print 0021-972X ISSN Online 1945-7197 * C.N.J. and A.N.C. are joint first authors. Printed in U.S.A. Abbreviations: HA, hypothalamic amenorrhea; IR, immunoreactivity. Copyright © 2013 by The Endocrine Society Received January 7, 2013. Accepted August 29, 2013. First Published Online September 12, 2013

4464 jcem.endojournals.org J Clin Endocrinol Metab, November 2013, 98(11):4464 – 4474 doi: 10.1210/jc.2013-1069

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 August 2014. at 07:17 For personal use only. No other uses without permission. . All rights reserved. doi: 10.1210/jc.2013-1069 jcem.endojournals.org 4465

consent was obtained from all subjects. This study was performed in accordance with the Declaration of Helsinki. Five healthy female subjects with regular menstrual cycles were recruited through advertisements placed in local newspapers (age 31.6 ± 2.6, range 24 –37 years; weight 60.4 ± 2.5, range 50.4 – 63.8 kg) as described previously (22, 23).

Protocol A single-blinded, placebo-controlled, 1-way crossover design study of 10 menstrual cycles was performed (Figure 1). During month 1 of the study protocol,

five healthy female subjects self-admin- Figure 1. Study protocol diagram. Five healthy women underwent a 1-way crossover protocol. istered twice-daily sc saline injections be- Twice-daily injections of saline and kisspeptin-54 (KP54) were administered between days 7 and tween days 7 and 14 of their menstrual 14 during months 1 and 2 of the protocol, respectively. During each month of the protocol, subjects underwent a baseline GnRH test (day 1) and 8-hour blood sampling (day 6) before cycle (see Subject Characteristics in Sup- commencing injections. During the injection period (days 7–14), subjects underwent a 4-hour plemental Table 1, published on The study (day 7) and two additional 8-hour studies (days 11 and 14) immediately after injection of Endocrine Society’s Journals Online saline or kisspeptin-54. A GnRH test was performed 24 hours after cessation of saline or website at http://jcem.endojournals.org). kisspeptin treatment (day 15). USS denotes ultrasound examination and measurement of During month 2 of the study protocol, reproductive hormones. Day 1 was defined as the first day of menstrual bleeding. the same five healthy female subjects self- administered twice-daily sc kisspep- Administration of kisspeptin-54 or kisspeptin-10 acutely tin-54 injections (6.4 nmol/kg; equivalent to 37 [Lg/kg) (19) be- stimulates gonadotropin secretion in healthy male (19 –22) tween days 7 and 14 of their menstrual cycle; (see Supplemental and female (23, 24) volunteers and women with functional Methods for kisspeptin-54 peptide synthesis and testing). Day 1 hypothalamic amenorrhea (HA) (25, 26). However, twice- of each month was defined as the first day of menstrual bleeding. daily sc administration of kisspeptin-54 to women with HA Kisspeptin injections causes profound tachyphylaxis within 24 hours of com- All subjects were trained in self-administration of sc injections mencing treatment (25). Furthermore, continuous iv infu- by an investigator at the start of the study protocol. At the be- sion of kisspeptin-10 to monkeys is associated with tachy- ginning of each week when injections were to be performed, a phylaxis within 3 hours of commencing administration (16). box containing unlabeled vials of freeze-dried saline (month 1) or It has therefore been assumed that chronic treatment with vials of freeze-dried kisspeptin-54 (month 2), alcohol wipes, sa- kisspeptin-54 in healthy women may have limited therapeu- line ampoules for reconstitution of freeze-dried vial contents, 0.5-mL insulin syringes with needles, and needle disposal bins tic potential as a stimulator of human reproductive activity was given to each subject. For injection, vial contents were re- because tachyphylaxis has been observed after chronic ad- constituted in 0.5 mL of 0.9% saline. A 0.5-mL insulin syringe ministration in primates and women with HA. However, this was then used to inject saline alone (month 1) or 6.4 nmol/kg hypothesis had not been tested previously. kisspeptin-54 (month 2) into the lower anterior abdominal re- In this study, we aimed to determine whether chronic gion sc. This kisspeptin-54 dose was the same as used in our exogenous kisspeptin was sufficient to alter menstrual cy- previous work in healthy women and women with HA (25, 26). Volunteers were instructed to prepare and perform injections in clicity, using the exact dose previously demonstrated to the morning after breakfast (unless attending for a study in which cause tachyphylaxis within 24 hours in women with HA case it was done by the volunteer at time 0 of the study) and in (6.4 nmol/kg) (25, 26). We performed a single-blinded the evening before bed. The volume of the saline or kisspeptin placebo-controlled 1-way crossover study to determine injections was identical. Subjects were instructed to refrigerate the effects of twice-daily administration of kisspeptin for vials stored at home. Before commencing study visits, injection 7 days on the menstrual cycle in healthy human female sites were inspected, and numbers of returned vials, insulin syringes, and saline or kisspeptin vials were counted to monitor subjects with regular menstrual cycles. compliance. Plasma kisspeptin immunoreactivity (IR) was also

assessed throughout the study protocol to confirm compliance.

Subjects and Methods Baseline period During menstrual days 1 to 6 of each month of the study Subjects protocol, pituitary responsiveness was assessed by a GnRH test Ethical approval was granted by the Hammersmith and (see Supplemental Methods for protocol), and baseline repro- Queen Charlotte’s and Chelsea Hospitals Research Ethics Com- ductive hormones and ultrasound markers were measured. The mittee (registration number 05/Q0406/142). Written informed collection, processing, and analysis of blood samples are detailed

4466 Javasena et al Kisspeptin-54 Advances Menstrual Cycle J Clin Endocrinol Metab, November 2013, 98(11):4464 – 4474

in Supplemental Methods. The baseline period also allowed the protocol. Transvaginal scans were not used to minimize discom- acclimatization of subjects to study conditions. No injections fort to volunteers during repeated examinations. The ultra- were administered during this period. sonographer was blinded to treatment for all subjects. During each scan, the following parameters were measured: endometrial Treatment period thickness (in millimeters), mean ovarian volume (in cubic cen- During menstrual days 7 to 14 of each month of the study timeters), mean follicle number, and maximum diameter of larg- protocol, subjects self-administered twice-daily, single-blinded est follicle in each ovary (in millimeters). Ovulation was defined sc injections of saline (month 1 of study protocol) or kisspep- as a rise in serum progesterone >10 nmol/L together with sug- tin-54 (month 2 of study protocol) as above. gestive radiological features (visualization of a dominant follicle with subsequent appearance of a preovulatory follicle and/or corpus luteum). Posttreatment period During menstrual days 15 to 28 of each month of the study Data analysis protocol, subjects underwent a posttreatment observation period to assess pituitary responsiveness (GnRH test) and measure Investigators performing the clinical studies were blinded to circulating reproductive hormones and ultrasound markers. No results until all subjects had completed the study protocol. J.D.V. injections were administered during this period. used an established, blinded deconvolution method with 93% sensitivity and specificity (27) to identify LH pulses and calculate the secretory mass of LH pulses (integral of LH secretion over Four-hour blood sampling after injection of saline or time during a secretory burst normalized per liter of distribution kisspeptin volume). Cumulative levels of basal, pulsatile, and total (basal All subjects underwent blood sampling during the 4-hour plus pulsatile) LH secretion were also estimated during each period immediately after the first injection of the saline (day 7, study. Data are presented as mean ± SEM. Kisspeptin IR data month 1) or kisspeptin-54 (day 7, month 2) treatment period to were log-transformed to normalize data before data analysis. All confirm that kisspeptin-54 acutely stimulated LH release as pre- other analyses included data series, most which had normal dis- viously shown in healthy women (23). Saline or kisspeptin-54 tributions assessed using the Kolmogorov-Smirnov test with (6.4 nmol/kg) was sc administered at 0 minutes by the subject, Dallal-Wilkinson-Lillie analysis. Hormone profiles during and blood was sampled for serum LH, FSH, estradiol, and 4-hour blood sampling studies were analyzed using repeated- plasma kisspeptin IR at -30, 0, 10, 20, 40, 60, 90, 120, 150, 180, measures 2-way ANOVA with Bonferroni post hoc correction. 210, and 240 minutes. Pairs of means were analyzed using the unpaired two-tailed t test. Multiple means were compared using 1-way ANOVA with Bon- Eight-hour blood sampling for assessment of LH feronni’s multiple-comparison test. In all cases, P < .05 was pulsatility considered statistically significant.

Subjects underwent 3 assessments of LH pulsatility during month 1 and month 2 of the study protocol. Baseline assessment of LH pulsatility was performed on menstrual day 6 (1 day before Results commencing injections). LH pulsatility was also assessed after injection of saline or kisspeptin-54 on menstrual days 11 and 14. Acute effects of saline or kisspeptin-54 injection on The saline or kisspeptin-54 was reconstituted using one of the plasma kisspeptin at the commencement, vials given to each subject at the beginning of the treatment pe- midpoint, and end of twice-daily administration in riod for home storage. Blood was sampled every 10 minutes. healthy women Studies commenced between 8:00 and 9:00 AM.

Acute changes in plasma kisspeptin IR after injection Assessment of pituitary sensitivity before and after of saline or kisspeptin-54 injections of saline or kisspeptin Plasma kisspeptin was unchanged at approximately 10 Subjects each underwent 2 GnRH tests during month 1 and pmol/L after injection of saline on the first (Figure 2A), month 2 of the study protocol (see Supplemental Methods for fourth (Figure 2B), and final (seventh; Figure 2C) days of protocol). A baseline GnRH test was performed on menstrual day 1 and was repeated in each subject 24 hours after final saline twice-daily administration. Kisspeptin-54 injection acutely el- or kisspeptin injection (menstrual day 15). evated plasma kisspeptin IR on the first day of administration, with peak mean kisspeptin IR levels of 2421 ± 392 Basal measurement of reproductive hormones pmol/L at 45 minutes after injection (P < .001 vs saline) Basal measurements of serum LH, FSH, estradiol, progester- (Figure 2A). Similar elevations of kisspeptin IR were ob- one, and plasma kisspeptin IR were taken from subjects during served after injection of kisspeptin on the fourth and last days 1, 6, 7, 11, 14, 15, 18, 21, and 28 during month 1 and month injection days (Figure 2, B and C). 2 of the study protocol. On days 7, 11, and 14, the basal blood sample was taken before the morning injection. Plasma kisspeptin IR preinjection of saline or Ultrasound scans kisspeptin-54 Transabdominal ultrasound scans were performed on days 1, In subjects receiving saline, plasma kisspeptin IR mea- 7, 15, 18, 21, and 28 during month 1 and month 2 of the study sured before the morning saline injection remained ap-

doi: 10.1210/jc.2013-1069 jcem.endojournals.org 4467

Figure 2. Acute changes in plasma kisspeptin levels in healthy women receiving twice-daily injections of saline or kisspeptin-54. A–C, Time profiles of plasma kisspeptin IR after sc bolus injection of saline or kisspeptin-54 (KP54) at time 0 on menstrual days 7 (A), 11 (B), and 14 (C) of the study protocol. D, Bar graph comparing mean preinjection kisspeptin IR after saline or kisspeptin-54 throughout the study protocol. All subjects commenced twice-daily treatment with saline or kisspeptin-54 on the morning of menstrual day 7 after their basal blood sample and finished treatment on the morning of day 14. Data are presented as mean ± SEM. *, P < .05; **, P < .01. proximately 10 pmol/L throughout the study protocol 0.98 IU/L). The mean maximal increase in LH from base- (Figure 2D). In subjects receiving kisspeptin injections, line after kisspeptin injection was observed at 180 minutes plasma kisspeptin IR was not elevated on menstrual days and was 8.6 ± 3.4 IU/L above baseline. 1, 6, or 7 because these blood samples were taken before the first kisspeptin-54 injection. Plasma kisspeptin IR was Midpoint of treatment period (menstrual day 11) elevated on menstrual days 11 and 14 (during the twice- Saline injection did not change serum LH or FSH com- daily kisspeptin treatment period but just before the morn- pared with baseline (Figure 3B, baseline LH 5.74 ± 1.45 ing kisspeptin injection), which suggested compliance IU/L). Kisspeptin-54 injection acutely increased serum LH with kisspeptin injection the previous evening. As ex- levels in subjects when compared with saline (Figure 3B, pected, plasma kisspeptin IR was not elevated on the P < .01 at 210 minutes, and P < .05 at 220 to 240 minutes, morning of day 15, because it was approximately 24 hours baseline LH 9.41 ± 4.89 IU/L). The mean maximal in- after the final kisspeptin-54 injection. Furthermore, crease in LH from baseline after kisspeptin injection was plasma kisspeptin IR remained <10 pmol/L on days 18 to 28 of the study protocol (Figure 1D). observed at 210 minutes and was 8.3 ± 2.4 IU/L above baseline. Acute effects of saline or kisspeptin-54 injection on serum reproductive hormones at the End of treatment period (menstrual day 14) commencement, midpoint, and end of twice-daily Saline injection did not change serum LH or FSH com- administration in healthy women pared with baseline (Figure 3C, baseline LH 17.79 ± 9.69 IU/L). The mean maximal increase in LH from baseline Commencement of treatment period (menstrual after kisspeptin injection was observed at 220 minutes and day 7) Saline injection did not change serum LH levels when was 12.7 ± 8.1 IU/L above baseline (baseline LH 6.01 ± compared with baseline (Figure 3A, baseline LH 5.20 ± 1.60 IU/L). Kisspeptin-54 injection showed a trend to- 0.64 IU/L). Kisspeptin-54 injection acutely increased se- ward stimulating serum LH, but this did not reach statis- rum LH levels in subjects when compared with saline (Fig- tical significance compared with saline. During days 7, 11, ure 3A, P < .05 at 180 –210 minutes, baseline LH 5.68 ± and 14 of the treatment period, total LH secretion was

4468 Javasena et al Kisspeptin-54 Advances Menstrual Cycle J Clin Endocrinol Metab, November 2013, 98(11):4464 – 4474

: Day 7 : Day 11 : Day 14

) ) A ) B C

L L

/ / (First day of injections) / (Mid-point of injections) (Last day of injections)

U U

I I

(

( (

30 30 30

s * s KP54 20 20 ** * 20

r r

i 10 i 10 i 10

H H

L LH 0 0 0

m m

um 0 30 60 90 120 150 180 210 240 0 30 60 90 120 150 180 210 240 0 30 60 90 120 150 180 210 240 ru

Se Time (M inutes) Se Time (M inutes) Time (M inutes) Injection Injection Injection D )

L / ** 2000 .IU

(

1500 Saline 1000 KP54

ease

c 500

i 0

L -500

UC d7 d11 d14 d7 d11 d14

A Day of menstrual cycle

Figure 3. Acute changes in serum reproductive hormones in healthy women receiving twice-daily injections of saline or kisspeptin-54. A–C, Acute increases in serum LH after sc bolus injection of saline or kisspeptin-54 (KP54) at time 0 on menstrual days 7 (A), 11 (B), and 14 (C) of the study protocol. All subjects underwent treatment during menstrual days 7 and 14 with twice-daily sc bolus of saline or kisspeptin-54 injections. D, Summary bar graph comparing mean area under curve (AUC) LH increase after saline or kisspeptin-54 on days 7, 11, and 14 of the study protocol. Data are presented as mean ± SEM. *, P < .05; **, P < .01. increased significantly after kisspeptin-54 when compared all subjects had a shorter menstrual cycle (by approxi- with saline (P < .01 using 2-way ANOVA) (Figure 3D). mately 2 days) during kisspeptin-54 treatment when com- We also compared the magnitude of serum LH increase pared with saline (mean length of menstrual cycle was 1 hour after kisspeptin-54 injection with the peak increase 28.6 ± 1.4 days with saline vs 26.8 ± 3.1 days with kiss- in serum LH after kisspeptin-54 injection (Supplemental peptin, P < .01) (Figure 4A). Figure 1). On menstrual day 7, the peak LH response was 3-fold higher when compared with the LH response 1 hour Timing of peak serum LH after kisspeptin-54 injection (increase in serum LH was During kisspeptin-54 treatment, observed peak levels 9.2 ± 4.5 IU/L at 1 hour; peak 3.1 ± 1.8 IU/L, P = .063 of serum LH and estradiol, but not FSH, were earlier dur- vs 1 hour). On menstrual days 11 and 14, there was vir- ing the menstrual cycle when compared with the saline tually no change in mean serum LH 1 hour after kisspep- group (Figure 4, B–D). Furthermore the menstrual day of tin-54 injection (<1 IU/L), whereas peak increases in se- highest recorded serum LH was approximately 2 days ear- rum LH of 19.7 ± 9.5 IU/L (P = .059 vs 1 hour) and 23.0 ± lier during kisspeptin-54 treatment when compared with 12.9 IU/L (P = .067 vs 1 hour) were later observed, saline treatment (mean menstrual day of highest recorded respectively. serum LH was 15.2 ± 1.3 with saline vs 13.0 ± 1.9 with kisspeptin, P < .05) (Figure 4E). Effects of twice-daily saline or kisspeptin injections on length of the menstrual cycle and Timing of luteal phase of menstrual cycle biochemical markers of reproductive activity in We examined the onset of the luteal phase, which is healthy women characterized by release of a mature oocyte from the ovary

Length of menstrual cycle and secretion of progesterone by the residual corpus lu- Subjects had the same menstrual cycle length before teum. During kisspeptin-54 treatment, levels of serum commencing the study when compared with menstrual progesterone became elevated (>10 nmol/L) earlier dur- cycle length during saline administration (mean menstrual ing the menstrual cycle when compared with the saline cycle length was 28.6 ± 1.1 days before study commence- group (Figure 4F). The menstrual day of onset of the luteal ment and 28.6 ± 1.4 days with saline; P = 1.00). However phase (defined as beginning when serum progesterone was

doi: 10.1210/jc.2013-1069 jcem.endojournals.org 4469

Effects of twice-daily saline or kisspeptin on LH pulsatility in healthy women LH pulsatility was determined immediately before commencing saline or kisspeptin-54 treatment (menstrual day 6 of study protocol) and during saline or kisspeptin-54 treatment (menstrual days 11 and 14 of study protocol). Mean secretory mass was slightly lower before kisspeptin-54 treatment when compared with saline treatment (mean secretory mass was 5.9 ± 0.5 IU/L with saline vs 4.1 ± 0.4 IU/L with kisspeptin, P < .01) (Figure 6A). Despite this, mean secretory mass was increased significantly during menstrual day 11 (fourth day of treatment) during Figure 4. Levels of serum reproductive hormones throughout the menstrual cycle in healthy kisspeptin-54 treatment when com- women receiving twice-daily injections of saline or kisspeptin-54. A, Comparison of cycle length in individual healthy women undergoing twice-daily sc bolus injections of saline or kisspeptin-54 pared with the saline treatment (KP54) between menstrual days 7 and 14 of the study protocol. B–D, Levels of serum LH (B), FSH (mean secretory mass was 3.4 ± 0.4 (C), and estradiol (D) during the menstrual cycle in healthy women undergoing twice-daily sc IU/L with saline vs 14.5 ± 3.6 IU/L bolus injections of saline or kisspeptin-54 between menstrual days 7 and 14. E, Comparison of peak serum LH in individual healthy women undergoing twice-daily sc bolus injections of saline with kisspeptin, P < .01). On the last or kisspeptin-54 between menstrual days 7 and 14 of the study protocol. F and G, Levels of day of treatment (menstrual day 14), serum progesterone (F) and comparison of menstrual day of onset of the luteal phase (G) in no significant differences in secre- individual healthy women undergoing twice-daily sc bolus injections of saline or kisspeptin-54 tory mass were observed between sa- between menstrual days 7 and 14 of the study protocol. The luteal phase of the menstrual cycle was defined as beginning when serum progesterone was elevated >10 nmol/L. Data are line or kisspeptin-54 treatment. No presented as mean ± SEM. *, P < .05; **, P < .01. significant differences in pulse frequency were observed between sa- elevated >10 nmol/L) was approximately 2 days earlier line and kisspeptin-54 treatment (Figure 6B). Pulsatile LH during kisspeptin-54 treatment when compared with sa- secretion was increased significantly during menstrual day line treatment (mean menstrual day of serum progesterone 11 (fourth day of treatment) in the kisspeptin-54 group increase >10 nmol/L was 18.0 ± 2.1 with saline vs 15.8 ± when compared with the saline group (mean secretory 0.9 with kisspeptin, P < .05) (Figure 4G). mass was 17.9 ± 3.6 IU/L with saline vs 78.2 ± 22.8 IU/L

with kisspeptin, P < .05) (Figure 6C). However, basal and Effects of twice-daily saline or kisspeptin injections on radiological markers of reproductive total LH secretion were not significantly different between activity in healthy women groups (Figure 6, D and E). A corpus luteum, indicating recent ovulation, was observed in all women after kisspeptin-54 treatment and in Assessment of pituitary sensitivity before and 3 of 5 women after saline treatment. Elevated serum pro- after injections of saline or kisspeptin gesterone (>10 nmol/L) was observed in all subjects re- We examined the sensitivity of all healthy female sub- ceiving saline or kisspeptin-54 treatment. Immediately af- jects to iv GnRH, both 6 days before (menstrual day 1 of ter 7 days of kisspeptin-54 treatment (on menstrual day study protocol) and 24 hours after saline or kisspeptin 15), the mean diameter of the largest follicle seen during treatment (menstrual day 15 of study protocol). On men- ultrasonography was significantly higher when compared strual day 1, no significant difference in pituitary sensi- with women during saline treatment (mean diameter of tivity was observed after GnRH administration between largest follicle was 10.0 ± 2.2 mm with saline vs 15.5 ± subjects before commencing saline of kisspeptin treatment 1.2 mm with kisspeptin, P < .05) (Figure 5A). No signif- (mean maximal LH increase during first 2 hours after icant differences in number of follicles or endometrial GnRH injection was 13.1 ± 1.1 IU/L before saline and thickness were observed during kisspeptin-54 treatment 13.4 ± 1.1 IU/L before kisspeptin-54; P value was not when compared with saline (Figure 5, B and C). significant) (Supplemental Figure 2A). Furthermore, 24

4470 Javasena et al Kisspeptin-54 Advances Menstrual Cycle J Clin Endocrinol Metab, November 2013, 98(11):4464 – 4474

In the current study, kisspeptin treatment was associated with an advanced timing of the serum progesterone increase and onset of menstrual bleeding by approximately 2 days when compared with saline. Animal studies demonstrate that kisspeptin stimulates gonadotropin secretion in a GnRH-dependent manner because its action is abolished by GnRH antagonist (14). Ex- ogenous GnRH is sufficient to trigger ovulation (29). More studies with daily follicle morphology assessment and daily serum LH measurement would be required to confirm our findings. However, our

study data raise the possibility that Figure 5. Changes in radiological markers during the menstrual cycle of healthy women receiving twice-daily injections of saline or kisspeptin-54. A–C Mean values for diameter of kisspeptin-54 administration may largest ovarian follicle (A), number of ovarian follicles (B), and endometrial thickness (C) in advance the onset of ovulation in healthy women undergoing twice-daily sc bolus injections of saline or kisspeptin-54 (KP54) (6.4 healthy women by stimulating en- nmol/kg) between menstrual days 7 and 14. Data are presented as mean ± SEM. dogenous hypothalamic GnRH secretion. In addition to stimulating hours after cessation of twice-daily saline or kisspep- GnRH, kisspeptin itself is implicated in generation of the tin-54 injections, no significant difference in pituitary LH surge needed for ovulation. In rodents, a subpopula- sensitivity was observed after GnRH administration be- tion of hypothalamic anteroventral periventricular nu- tween saline and kisspeptin-54 treatment (mean maxi- cleus kisspeptin neurons are implicated in generating the mal LH increase during first 2 hours after GnRH injec- LH surge needed for ovulation (30) and are positively tion was 39.9 ± 8.9 IU/L after saline and 41.0 ± 16.4 rather than negatively regulated by estradiol (31). Central IU/L after kisspeptin-54; P value was not significant) administration of a monoclonal antibody to kisspeptin is (Supplemental Figure 2B). sufficient to block ovulation in rats (32). Furthermore, The expected physiological increase in pituitary sensi- administration of kisspeptin-10 has been shown to stim- tivity to GnRH during menstrual day 15 vs menstrual day ulate ovulation in the musk shrew (33), rat (34), and sheep 1 (28) was observed in healthy female subjects, whether (17). Humans have no anatomical equivalent of the an- receiving saline or kisspeptin treatment (Supplemental teroventral periventricular nucleus. Nevertheless, it is pos- Figure 2C). sible that exogenous kisspeptin-54 may have advanced the

onset of the luteal phase in our female subjects by increasing kisspeptin signaling in a subpopulation of hypotha- Discussion lamic kisspeptin neurons that stimulate the LH surge. Sev-

Genetic studies demonstrate that kisspeptin peptides are eral lines of evidence suggest that in humans, unlike in necessary for pubertal maturation in humans (8 –10). We lower species, a change in pituitary responsiveness to and other investigators have recently demonstrated that GnRH rather than an actual GnRH surge is responsible exogenous kisspeptin acutely stimulates gonadotropin se- for the LH surge (35, 36). It is therefore possible that cretion in women (23–26). However, the pharmacological kisspeptin-54 advances ovulation in women by increasing effects of chronic kisspeptin exposure in healthy women the prevailing levels of estradiol, which are needed to trig- with regular menstrual cycles have not been studied pre- ger the LH surge at a pituitary level. viously. Chronic kisspeptin administration causes pro- Chronic kisspeptin administration has also been impli- found tachyphylaxis in male monkeys and in women in cated as a potential novel therapy for inhibiting reproduc- functional HA (16, 25, 26). We present novel data sug- tive hormone secretion in contraception or the treatment gesting that menstrual cyclicity persists in healthy women of hormone-sensitive cancer; our data suggest that kiss- after twice-daily kisspeptin-54 treatment during the fol- peptin-54 may have limited therapeutic potential in this licular phase of the menstrual cycle. regard, at least in women at the dose of kisspeptin tested.

doi: 10.1210/jc.2013-1069 jcem.endojournals.org 4471

Figure 6. Changes in LH pulsatility during the menstrual cycle of healthy women receiving twice-daily injections of saline or kisspeptin-54. All women underwent frequent blood sampling for 8 hours on menstrual days 6, 11, and 14 of the study protocol. Twice-daily sc bolus injections of saline or kisspeptin-54 (KP54) (6.4 nmol/kg) were self-administered between menstrual days 7 and 14. A and B, Mean values for secretory mass (A) and estimated pulses per 24 hours (B). C–E, Levels of pulsatile (C), basal (D), and total (E) LH secretion during each 8-h sampling study. Data are presented as mean ± SEM. *, P < .05; **, P < .01.

However, our observation that kisspeptin may advance effects have never been compared directly between men the menstrual cycle raises the possibility that women with and women (19, 23). Consistent with our findings in certain forms of infertility could be treated with kisspep- healthy women, George et al (21) recently observed that a tin. A recent study suggests that kisspeptin is sufficient to 22.5-hour iv infusion of the shorter form of kisspeptin, restore ovulation in a mouse model of anovulatory hyper- kisspeptin-10, stimulates LH secretion without apparent prolactinemia (37). More studies are required to deter- tachyphylaxis in healthy men. However, in contrast to the mine whether kisspeptin-54 treatment could be used to current findings, twice-daily kisspeptin-54 treatment has stimulate ovulation in women with anovulatory repro- been shown to rapidly cause tachyphylaxis in women with ductive disorders other than HA. HA. Women with HA are acutely 4-fold more sensitive to Traditional ovulatory drugs such as clomiphene citrate the effects of kisspeptin-54 when compared with healthy have low pregnancy rates (38, 39). Although efficacious, women in the follicular phase of their menstrual cycle. A in vitro fertilization confers a risk of ovarian hyperstimu- higher dose of kisspeptin-54 may therefore be necessary to lation (40). Kisspeptin acts by stimulating endogenous, achieve tachyphylaxis in healthy female volunteers (23, rather than pharmacological, levels of reproductive hor- 25). Nevertheless, our data suggest that menstrual cyclic- mone secretion. Kisspeptin might therefore offer a fertility ity persists in healthy women using the currently tested treatment with lower risk of ovarian hyperstimulation regimen of follicular-phase kisspeptin-54 treatment. Pre- syndrome when compared with in vitro fertilization. liminary data have emerged suggesting that chronic ad- Comparison of our data with other clinical studies of ministration of a kisspeptin analog causes tachyphylaxis kisspeptin administration is complicated by the investiga- in healthy men (41). It would be interesting to determine tion of its 2 peptide forms (kisspeptin-54 and -10) and whether tachyphlyaxis to kisspeptin-54 is also sexually sexual dimorphism of the effects of kisspeptin-10; iv bolus dimorphic or dependent on the dose and precise form of administration consistently stimulates LH men (20 –22), kisspeptin receptor agonist administered. but women are much less sensitive to the effects of kiss- It is interesting to note that although iv bolus injection peptin-10 during the follicular phase of the menstrual cy- of the shorter kisspeptin fragment, kisspeptin-10, acutely cle when compared with the preovulatory phase (22, 24). stimulates LH secretion within an hour of administration, Although kisspeptin-54 stimulates LH in both sexes, its we observed that sc injection of kisspeptin-54 acutely

4472 Javasena et al Kisspeptin-54 Advances Menstrual Cycle J Clin Endocrinol Metab, November 2013, 98(11):4464 – 4474

stimulated peak LH secretion 3 to 4 hours after adminis- effects of kisspeptin-54 administration on menstrual cy- tration despite a rapid elevation of kisspeptin IR within clicity or LH pulsatility. Furthermore, the gonadotropin minutes of administration. On the first injection day, the response to kisspeptin treatment is known to alter with the increase in serum LH 1 hour after kisspeptin-54 injection phase of the menstrual cycle in healthy women (23, 24). A was 3-fold lower when compared with the peak increase larger study is needed to confirm our findings and deter- in serum LH after kisspeptin-54 injection. Furthermore, mine what factors influence the variability in the response on the fourth and final injection days, there was negligible of subjects to kisspeptin treatment. It is possible that stimulation of LH secretion 1 hour after kisspeptin-54 chronic kisspeptin-54 treatment might have effects that injection; however, serum LH later increased to peak lev- last beyond the treatment period. For this reason, we de- els 3 to 4 hours after kisspeptin injection. By comparison, signed a 1-way crossover protocol in which subjects self- Chan et al (20) observed that kisspeptin-10 stimulated administered saline during the first month followed by peak LH secretion within an hour of commencing admin- kisspeptin-54 during the second month. It therefore re- istration in healthy men. It is possible that these data sug- mains possible that an order effect might have contributed gest differences between the biological actions of kisspep- to our results. However, it is noteworthy that LH pulse tin-10 and -54. For instance, one might speculate that secretory mass was marginally lower at the start of the kisspeptin-10 stimulates a readily releasable pool of second month when compared with LH pulse secretory GnRH stored within nerve terminals of the median emi- mass at the start of the first month (when greater stress nence (42, 43), whereas kisspeptin-54 might act more levels would be expected). Furthermore, subjects had the proximally in the GnRH neuron to stimulate de novo same menstrual cycle length before commencing the study GnRH synthesis. Alternatively, it is possible kisspeptin-54 when compared with menstrual cycle length during saline might require breakdown into a smaller kisspeptin frag- administration. We also recognize that although estradiol ment before becoming biologically active. However, liq- measurements were compared with placebo-controlled uid chromatography has only identified kisspeptin-54 in values, the automated platform assay used to analyze se- the human circulation acutely after kisspeptin-54 admin- rum estradiol can cross-react with other steroids. istration (19). Mean peak levels of LH during the menstrual cycle were It is interesting to appraise the evidence that exogenous lower during kisspeptin treatment when compared with kisspeptin stimulates basal and pulsatile GnRH secretion. saline treatment. Furthermore, it is important to recognize Although kisspeptin stimulates rat pituitary gonadotropin that kisspeptin-54 did not acutely stimulate significant LH secretion in vitro (44), animal studies using GnRH antag- secretion during the final day of twice-daily kisspeptin-54 onists support the view that exogenous kisspeptin stimu- treatment. We therefore cannot exclude that kisspep- lates basal LH secretion through a GnRH-dependent tin-54 treatment caused partial tachyphylaxis in healthy mechanism (45, 46), which is possibly mediated through female volunteers. Alternatively, the short baseline sam- GnRH nerve terminals at the median eminence (42, 43). pling period involving just 2 blood samples may have re- However, it is less clear whether exogenous kisspeptin duced our power to detect stimulation of LH after kiss- stimulates endogenous GnRH pulsatility. Sustained expo- peptin-54 injection on day 14. LH levels were highly sure to exogenous kisspeptin-10 or -54 stimulates pulsatile variable during menstrual day 14, which may have been LH secretion for several hours in healthy men (21) and caused by some but not all subjects experiencing an LH patients with inactivating mutations of the neurokinin B surge and by the potential effect of kisspeptin-54 to ad- signaling pathway (47). However, because these studies vance the onset of the LH surge in subjects. merely measure LH pulsatility, it is not possible to exclude It is noteworthy that all subjects had peak progesterone that effects reflect increasing circulating estrogen and pi- levels >21 mmol/L after kisspeptin-54 treatment, which is tuitary responsiveness to unchanged, endogenous GnRH highly suggestive of ovulation. Furthermore, a corpus lu- pulsatility. Chan and colleagues (20) recently demon- teum was observed in all 5 subjects during the month of strated that iv bolus kisspeptin-10 increased the time to the kisspeptin-54 treatment (Supplemental Table 2). Our data next endogenous LH pulse in healthy men; this may rep- therefore suggest that all 5 subjects had ovulatory cycles resent the only current data suggesting that exogenous during kisspeptin treatment. More studies are required to kisspeptin can directly modulate endogenous GnRH pul- determine whether menstrual cycles during kisspeptin satility in humans (by increasing the latency period to the therapy have subtle physiological differences when com- next LH/GnRH pulse). pared with natural menstrual cycles and to determine It is important to recognize that the current study is whether anovulation is observed at a different frequency based on observations within a small number of subjects during kisspeptin treatment when compared with placebo so may have insufficient statistical power to reveal subtle treatment.

doi: 10.1210/jc.2013-1069 jcem.endojournals.org 4473

In summary, our data have important pharmacological 6. Rometo AM, Krajewski SJ, Voytko ML, Rance NE. Hypertrophy implications; menstrual cyclicity persists in healthy and increased kisspeptin gene expression in the hypothalamic infundibular nucleus of postmenopausal women and ovariectomized women during a treatment regime of kisspeptin-54 previ- monkeys. J Clin Endocrinol Metab. 2007;92:2744 –2750. ously demonstrated to cause tachyphylaxis in women with 7. Richard N, Galmiche G, Corvaisier S, Caraty A, Kottler ML. KiSS-1 HA. Furthermore, kisspeptin-54 treatment is associated and GPR54 genes are co-expressed in rat gonadotrophs and differentially regulated in vivo by oestradiol and gonadotrophin-releasing with an advanced timing of the luteal phase of the men- hormone. J Neuroendocrinol. 2008;20:381–393. strual cycle when compared with placebo. More studies 8. de Roux N, Genin E, Carel JC, Matsuda F, Chaussain JL, Milgrom are required to determine whether kisspeptin-54 therapy E. Hypogonadotropic hypogonadism due to loss of function of the could have potential to treat patients with anovulatory KiSS1-derived peptide receptor GPR54. Proc Natl Acad Sci U S A. 2003;100:10972–10976. reproductive disorders. 9. Seminara SB, Messager S, Chatzidaki EE, et al. The GPR54 gene as a regulator of puberty. N Engl J Med. 2003;349:1614 –1627. 10. Topaloglu AK, Tello JA, Kotan LD, et al. Inactivating KISS1 mutation and hypogonadotropic hypogonadism. N Engl J Med. 2012; Acknowledgments 366:629 – 635. 11. Thompson EL, Patterson M, Murphy KG, et al. Central and pe- We are grateful to the National Institute for Health Research/ ripheral administration of kisspeptin-10 stimulates the hypothalam- Wellcome Trust Clinical Research Facility at Imperial College ic-pituitary-gonadal axis. J Neuroendocrinol. 2004;16:850 – 858. Healthcare NHS Trust for providing infrastructure for this 12. Patterson M, Murphy KG, Thompson EL, Patel S, Ghatei MA, study. Bloom SR. Administration of kisspeptin-54 into discrete regions of the hypothalamus potently increases plasma luteinising hormone

and testosterone in male adult rats. J Neuroendocrinol. 2006;18: Address all correspondence and requests for reprints to: Prof 349 –354. Waljit S. Dhillo, Department of Investigative Medicine, Imperial 13. Tovar S, Vázquez MJ, Navarro VM, et al. Effects of single or re- College London, Sixth Floor, Commonwealth Building, Ham- peated intravenous administration of kisspeptin upon dynamic LH mersmith Hospital, Du Cane Road, London W12 ONN, United secretion in conscious male rats. Endocrinology. 2006;147:2696 – Kingdom. E-mail: [email protected]. 2704. This work was funded by a Medical Research Council project 14. Gottsch ML, Cunningham MJ, Smith JT, et al. A role for kisspeptins grant. The department is funded by an Integrative Mammalian in the regulation of gonadotropin secretion in the mouse. Endocri- Biology Capacity Building Award and the National Institute nology. 2004;145:4073– 4077. for Health Research (NIHR) Biomedical Research Centre 15. Messager S, Chatzidaki EE, Ma D, et al. Kisspeptin directly stimu- Funding Scheme. C.N.J. is supported by an NIHR Clinical lates gonadotropin-releasing hormone release via G protein-coupled receptor 54. Proc Natl Acad Sci U S A. 2005;102:1761–1766. Lectureship, an Academy of Medical Sciences/Wellcome 16. Seminara SB, Dipietro MJ, Ramaswamy S, Crowley WF Jr, Plant Starter Grant for Clinical Lecturers, and a Society for Endo- TM. Continuous human metastin 45–54 infusion desensitizes G crinology Early Career Grant. A.N.C. and A.D.S. are sup- protein-coupled receptor 54-induced gonadotropin-releasing hor- ported by Wellcome/GlaxoSmithKline Clinical Research Fel- mone release monitored indirectly in the juvenile male Rhesus mon- lowships. G.M.K.N. and A.A. are supported by Wellcome key (Macaca mulatta): a finding with therapeutic implications. En- Clinical Research Training Fellowships. R.R. and C.I. are sup- docrinology. 2006;147:2122–2126. ported by NIHR Academic Clinical Fellowships. W.S.D. is 17. Caraty A, Smith JT, Lomet D, et al. Kisspeptin synchronizes pre- supported by an NIHR Career Development Fellowship. ovulatory surges in cyclical ewes and causes ovulation in seasonally Disclosure Summary: The authors have no conflicts of inter- acyclic ewes. Endocrinology. 2007;148:5258 –5267. est to disclose. 18. Redmond JS, Macedo GG, Velez IC, Caraty A, Williams GL, Am- stalden M. Kisspeptin activates the hypothalamic-adenohypophyseal-gonadal axis in prepubertal ewe lambs. Reproduction. 2011; 141:541–548. References 19. Dhillo WS, Chaudhri OB, Patterson M, et al. Kisspeptin-54 stimulates the hypothalamic-pituitary gonadal axis in human males. J Clin 1. Clements MK, McDonald TP, Wang R, et al. FMRFamide-related Endocrinol Metab. 2005;90:6609 – 6615. neuropeptides are agonists of the orphan G-protein-coupled recep- 20. Chan YM, Butler JP, Pinnell NE, et al. Kisspeptin resets the hypo- tor GPR54. Biochem Biophys Res Commun. 2001;285:1189 –1193. thalamic GnRH clock in men. J Clin Endocrinol Metab. 2011;96: 2. Kotani M, Detheux M, Vandenbogaerde A, et al. The metastasis E908 –E915. suppressor gene KiSS-1 encodes kisspeptins, the natural ligands of 21. George JT, Veldhuis JD, Roseweir AK, et al. Kisspeptin-10 is a the orphan G protein-coupled receptor GPR54. J Biol Chem. 2001; potent stimulator of LH and increases pulse frequency in men. J Clin 276:34631–34636. Endocrinol Metab. 2011;96:E1228 –E1236. 3. Ohtaki T, Shintani Y, Honda S, et al. Metastasis suppressor gene 22. Jayasena CN, Nijher GM, Comninos AN, et al. The effects of kiss- KiSS-1 encodes peptide ligand of a G-protein-coupled receptor. Na- peptin-10 on reproductive hormone release show sexual dimor- ture. 2001;411:613– 617. phism in humans. J Clin Endocrinol Metab. 2011;96:E1963– 4. Muir AI, Chamberlain L, Elshourbagy NA, et al. AXOR12, a novel E1972. human G protein-coupled receptor, activated by the peptide KiSS-1. 23. Dhillo WS, Chaudhri OB, Thompson EL, et al. Kisspeptin-54 stim- J Biol Chem. 2001;276:28969 –28975. ulates gonadotropin release most potently during the preovulatory 5. Clarkson J, Herbison AE. Postnatal development of kisspeptin neu- phase of the menstrual cycle in women. J Clin Endocrinol Metab. rons in mouse hypothalamus; sexual dimorphism and projections to 2007;92:3958 –3966. gonadotropin-releasing hormone neurons. Endocrinology. 2006; 24. Chan YM, Butler JP, Sidhoum VF, Pinnell NE, Seminara SB. Kiss- 147:5817–5825. peptin administration to women: a window into endogenous kiss-

4474 Javasena et al Kisspeptin-54 Advances Menstrual Cycle J Clin Endocrinol Metab, November 2013, 98(11):4464 – 4474

peptin secretion and GnRH responsiveness across the menstrual 36. Adams JM, Taylor AE, Schoenfeld DA, Crowley WF Jr, Hall JE. The cycle. J Clin Endocrinol Metab. 2012;97:E1458 –E1467. midcycle gonadotropin surge in normal women occurs in the face of 25. Jayasena CN, Nijher GM, Chaudhri OB, et al. Subcutaneous injec- an unchanging gonadotropin-releasing hormone pulse frequency. tion of kisspeptin-54 acutely stimulates gonadotropin secretion in J Clin Endocrinol Metab. 1994;79:858 – 864. women with hypothalamic amenorrhea, but chronic administration 37. Sonigo C, Bouilly J, Carré N, et al. Hyperprolactinemia-induced causes tachyphylaxis. J Clin Endocrinol Metab. 2009;94:4315– ovarian acyclity is reversed by kisspeptin administration. J Clin In- 4323. vest. 2012;122:3791–3795. 26. Jayasena CN, Nijher GM, Abbara A, et al. Twice-weekly adminis- 38. Kousta E, White DM, Franks S. Modern use of clomiphene citrate tration of kisspeptin-54 for 8 weeks stimulates release of reproduc- in induction of ovulation. Hum Reprod Update. 1997;3:359 –365. tive hormones in women with hypothalamic amenorrhea. Clin Phar- 39. Hart R, Norman R. Polycystic ovarian syndrome–prognosis and macol Ther. 2010;88:840 – 847. outcomes. Best Pract Res Clin Obstet Gynaecol. 2006;20:751–778. 27. Liu PY, Keenan DM, Kok P, Padmanabhan V, O’Byrne KT, Veld- 40. Min JK, Breheny SA, MacLachlan V, Healy DL. What is the most huis JD. Sensitivity and specificity of pulse detection using a new relevant standard of success in assisted reproduction? The singleton, deconvolution method. Am J Physiol Endocrinol Metab. 2009;297: term gestation, live birth rate per cycle initiated: the BESST endpoint E538 –E544. for assisted reproduction. Hum Reprod. 2004;19:3–7. 28. Knobil E, Plant TM, Wildt L, Belchetz PE, Marshall G. Control of 41. Matsui H 2012 Testosterone depletion by continuous administra- the rhesus monkey menstrual cycle: permissive role of hypothalamic tion of kisspeptin analogues. Paper presented at: Second World Con- gonadotrophin-releasing hormone. Science. 1980;207:1371–1373. ference of Kisspeptin Signaling in the Brain; November 9, 2012; 29. Grimes EM, Taymor ML, Thompson IE. Induction of timed ovu- Tokyo, Japan. lation with synthetic luteinizing hormone-releasing hormone in 42. d’Anglemont de Tassigny X, Fagg LA, Carlton MB, Colledge WH. women undergoing insemination therapy. I. Effect of a single par- Kisspeptin can stimulate gonadotropin-releasing hormone (GnRH) enteral administration at midcycle. Fertil Steril. 1975;26:277–282. release by a direct action at GnRH nerve terminals. Endocrinology. 30. Clarkson J, d’Anglemont de Tassigny X, et al. Kisspeptin-GPR54 2008;149:3926 –3932. signaling is essential for preovulatory gonadotropin-releasing hor- 43. d’Anglemont de Tassigny X, Ackroyd KJ, Chatzidaki EE, Colledge mone neuron activation and the luteinizing hormone surge. J Neu- WH. Kisspeptin signaling is required for peripheral but not central rosci. 2008;28:8691– 8697. stimulation of gonadotropin-releasing hormone neurons by 31. Smith JT, Popa SM, Clifton DK, Hoffman GE, Steiner RA. Kiss1 NMDA. J Neurosci. 2010;30:8581– 8590. neurons in the forebrain as central processors for generating the 44. Matsui H, Tanaka A, Yokoyama K, et al. Chronic administration of preovulatory luteinizing hormone surge. J Neurosci. 2006;26: the metastin/kisspeptin analog KISS1–305 or the investigational 6687– 6694. agent TAK-448 suppresses hypothalamic pituitary gonadal function 32. Adachi S, Yamada S, Takatsu Y, et al. Involvement of anteroventral and depletes plasma testosterone in adult male rats. Endocrinology. periventricular metastin/kisspeptin neurons in estrogen positive 2012;153:5297–5308. feedback action on luteinizing hormone release in female rats. J 45. Gutiérrez-Pascual E, Martínez-Fuentes AJ, Pinilla L, Tena-Sempere Reprod Dev. 2007;53:367–378. M, Malagón MM, Castaño JP. Direct pituitary effects of kisspeptin: 33. Inoue N, Sasagawa K, Ikai K, et al. Kisspeptin neurons mediate activation of gonadotrophs and somatotrophs and stimulation of reflex ovulation in the musk shrew (Suncus murinus). Proc Natl luteinising hormone and growth hormone secretion. J Neuroendo- Acad Sci U S A. 2011;108:17527–17532. crinol. 2007;19:521–530. 34. Matsui H, Takatsu Y, Kumano S, Matsumoto H, Ohtaki T. Periph- 46. Shahab M, Mastronardi C, Seminara SB, Crowley WF, Ojeda SR, eral administration of metastin induces marked gonadotropin re- Plant TM. Increased hypothalamic GPR54 signaling: a potential lease and ovulation in the rat. Biochem Biophys Res Commun. mechanism for initiation of puberty in primates. Proc Natl Acad Sci 2004;320:383–388. U S A. 2005;102:2129 –2134. 35. Hall JE, Taylor AE, Martin KA, Rivier J, Schoenfeld DA, Crowley 47. Young J, George JT, Tello JA, et al. Kisspeptin restores pulsatile LH WF Jr. Decreased release of gonadotropin-releasing hormone during secretion in patients with neurokinin B signaling deficiencies: phys- the preovulatory midcycle luteinizing hormone surge in normal iological, pathophysiological and therapeutic implications. Neu- women. Proc Natl Acad Sci U S A. 1994;91:6894 – 6898. roendocrinology. 2013;97:193–202.