Kisspeptin neurones in the posterodorsal medial amygdala modulate sexual partner preference and anxiety in male mice

The posterodorsal medial amygdala (MePD) is a neural site in the limbic brain involved in regulating emotional and sexual behaviours. There is, however, limited information available on the specific neuronal cell type in the MePD functionally mediating these behaviours in rodents. The recent discovery of a significant kisspeptin neurone population in the MePD has raised interest in the possible role of kisspeptin and its cognate receptor in sexual behaviour. The present study therefore tested the hypothesis that the MePD kisspeptin neurone population is involved in regulating attraction towards opposite sex conspecifics, sexual behaviour, social interaction and the anxiety response by selectively stimulating these neurones using the novel pharmacosynthetic DREADDs (designer receptors exclusively activated by designer drugs) technique. Adult male Kiss‐Cre mice received bilateral stereotaxic injections of a stimulatory DREADD viral construct (AAV‐hSyn‐DIO‐hM3D(Gq)‐mCherry) targeted to the MePD, with subsequent activation by i.p. injection of clozapine‐N‐oxide (CNO). Socio‐sexual behaviours were assessed in a counter‐balanced fashion after i.p. injection of either saline or CNO (5 mg kg‐1). Selective activation of MePD kisspeptin neurones by CNO significantly increased the time spent by male mice in investigating an oestrous female, as well as the duration of social interaction. Additionally, after CNO injection, the mice appeared less anxious, as indicated by a longer exploratory time in the open arms of the elevated plus maze. However, levels of copulatory behaviour were comparable between CNO and saline‐treated controls. These data indicate that DREADD‐induced activation of MePD kisspeptin neurones enhances both sexual partner preference in males and social interaction and also decreases anxiety, suggesting a key role played by MePD kisspeptin in sexual motivation and social behaviour.

odour was hampered by lesioning the MePD in mice, 12 rats 13 and hamsters. 14,15 The resultant deficit in mate preference as a result of MePD lesion was accompanied by a prolonged latency to mount or ejaculate, 15 which suggests a critical role for the MePD in sexual motivation and copulatory behaviour. Additionally, the MePD is functionally related to fear and anxiety. The induction of c-Fos has been reported in the MePD of rats subjected to anxiety using the elevated plus maze. 16 Also, social defeat robustly activates the MePD, 17 whereas its lesioning decreased playful fighting behaviour in rats. 18 Collectively, these data indicate that the MePD serves an overlapping function integrating emotional and sexual behaviour.
Although lesion and immediate early gene expression studies have identified an important role played by the MePD in sociosexual behaviour, they are limited in clarifying the definitive function of the specific neuronal cell type involved and, as such, there is a lack of adequate information on how sexual cues processed within the MePD bring about the appropriate behaviour. The recent discovery of a significant kisspeptin neurone population in the MePD 19 has raised interest in the probable role of kisspeptin and its cognate receptor in sexual behaviour. Kisspeptin receptor (Kiss1r) knockout male mice lack the characteristic preference for oestrous female odour, 20 suggesting an essential role of kisspeptin signalling in mediating mate preference. We have recently shown the specificity of the MePD as a neural site for the role of kisspeptin in sexual behaviour, with micro-infusion of kisspeptin into the MePD causing ex-copula erections in male rats. 21 On the other hand, little is known about the role of kisspeptin in the fear and anxiety response, although it was recently reported that habenula kisspeptin modulates fear in zebrafish. 22 In rats, the role of kisspeptin on anxiety is controversial, 23,24 whereas it attenuated negative mood in humans. 25 In the present study, we employed the pharmacosynthetic DREADDs (designer receptors exclusively activated by designer drugs) technique to selectively stimulate kisspeptin neurones in the MePD of Kiss-Cre male mice aiming to investigate whether endogenous MePD kisspeptin signalling potentiates a preference for oestrous female mice, in addition to regulating sexual behaviour, the anxiety response and social interaction.

| Animals
Breeding pairs of Kiss-Cre heterozygous transgenic mice 26

| Stereotaxic injection of DREADD virus
Surgical procedures for stereotaxic injection of stimulatory Credependent DREADD viral construct (AAV-hSyn-DIO-hM 3 D(Gq)-mCherry, Serotype:5; University of North Carolina at Chapel Hill Vector Core, NC, USA) to express the hM 3 Dq-DREADD in MePD Kiss1 neurones were performed under aseptic conditions with general anaesthesia induced by ketamine (Vetalar, 100 mg kg -1 , i.p.; Pfizer, Sandwich, UK) and xylazine (Rompun, 10 mg kg -1 , i.p.; Bayer, Leverkusen, Germany). Kiss-Cre male mice (age 7-8 weeks, n = 12) were secured in a David Kopf stereotaxic frame and small holes were drilled into the skull at a location above the MePD after a midline incision of the scalp. The stereotaxic injection coordinates used to target the MePD were obtained from the mouse brain atlas of Paxinos and

| Behavioural tests
Behavioural tests include sexual partner preference, sexual behaviour, anxiety and social interaction tests. A cross-over design was employed to test all mice with vehicle (saline) and clozapine-N-oxide (CNO; Tocris Bioscience, Bristol, UK). CNO was administered i.p. in saline as vehicle at a dose of 5 mg kg -1 . 29 On the day of testing, half of the mice received vehicle or CNO injections, whereas the other half received injections in the reverse order on subsequent testing, leading to an animal being tested twice for a given behavioural test in a counter-balanced order, with each test separated by 3-5 days. The behavioural tests were designed and ordered to first access olfactoryrelated processes such as sexual partner preference and social interaction, followed by sexual behaviour and a standard anxiety test. All behavioural tests were conducted in a dimly lit room between 12.00 and 14.00 hours, and tests commenced 30 minutes post vehicle or CNO injection to enable activation of hM 3 Dq receptor by CNO. 30 Behavioural events were manually scored by 2 experimenters who were blind to the treatment.

| Sexual partner preference
Sexual partner preference testing was carried out as described by Dresroziers et al. 31 and Angoa-Perez et al. 32 using a 3 chamber compartment as described below. A rectangular plexiglass cage (60 × 13 × 30 cm; Techniplast, Buguggiate, Italy) was divided into 3 equal compartments by an opaque partition with an opening (5 × 5 cm) at floor level. Mice were habituated to the 3-compartment box for 10 minutes prior to the commencement of the test. Once testing began, an oestrous female mouse confirmed by vaginal cytology and gonadally-intact male mouse housed in a wire mesh cup were randomly introduced into each of the lateral compartments of the box.
The mesh cup does not permit physical contact and only allows visual, olfactory and vocal communications. The number of entries by the test mouse into each lateral compartment containing either the male or female conspecific, as well as the time spent actively sniffing or poking its nose near the holes of the mesh cup, was recorded over a 10-minute test period. Preference score was determined by subtracting the time spent with the oestrous female from that spent with the male. A positive score indicates preference for female, whereas a negative score indicates preference for male. 33

| Sexual behaviour
Sexual behaviour testing was conducted in a Techniplast cage (32 × 16 × 14 cm) with clean wood chip bedding. Males were given sexual experience by co-habitation with receptive females for about 2 weeks before the test. On the day of the test, mice were habituated to the test arena for 10 minutes before introducing the oestrous female. The latency for mounting, intromission and ejaculation, as well as the number of mounts and intromissions, was recorded. The test was terminated once the mouse ejaculated or after 30 minutes of testing.
If no sexual behaviour was displayed within the 30-minute test period, the latency was scored as 1800 seconds. A mount was scored as the male climbing and grabbing the female from behind with both paws.
Intromission was designated as vaginal penetration during mounting accompanied by pelvic thrusting, whereas ejaculation was scored as intromission with a longer lasting thrust resulting in the male immobilising and falling off the female followed by a period of disinterest in the female. 34

| Anxiety-like behaviour
The elevated plus maze (EPM) was used to assess anxiety. The EPM and closed arms, as well as the number of entries into each arm, was recorded during the 5-minute period. An entry into the arm was defined as all 4 paws in the arm, whereas an exit was defined as at least 2 paws out of the arm. Anxiety index (AI) was determined from total activity on the EPM using the formula described by Cohen et al 35 The AI score ranges from 0 to 1; a higher index indicates increased anxiety-like behaviour.

| Social interaction
Test mice were singly housed for 1 hour in a holding room adjacent to the test room. At the commencement of the test, the test mouse and a same sex and strain juvenile conspecific (23-28 days old) 36 were placed simultaneously in the test arena, comprising a Techniplast cage (32 × 16 × 14 cm) with clean wood chip bedding. The total time spent sniffing, following, grooming and mounting the conspecific was recorded over a period of 5 minutes. The use of a juvenile mouse as a social stimulus should trigger less aggressive bouts in adult mice 37 and limit confounding aggressive responses typical of adult male-male interaction 38 in the social repertoire.

| Validation of injection site
Mice were anaesthetised with a lethal dose of ketamine and transcardially perfused with heparinised saline for 5 minutes, followed by Gottingen, Germany). The neuroanatomical landmarks bordering the MePD were determined using a reference guide from the mouse brain atlas. 27 The number of mCherry positive neurones was counted in the MePD of each animal and the total number was used to calculate the group mean (mean ± SEM). Images were taken using Axioskop 2 Plus microscope (Zeiss) equipped with axiovision, version 4.7 (Zeiss).

| Statistical analysis
Comparisons between vehicle and CNO-induced DREADDs activation on behavioural events were made by subjecting data to Mann-Whitney U test (Systat Software Inc., San Joses, CA, USA). Student's t test was used to compare the mean numbers of positive mCherry fluorescent neurones in the MePD from mice observed bilaterally with the 1 from mice observed unilaterally. Data are reported as the mean ± SEM. P < .05 was considered statistically significant.

| Selective targeting of MePD Kiss1 neurones
The Cre-dependent DREADDs confine to the Cre locus of targeted neural site 30

| Effect of CNO on sexual partner preference
When presented with a choice between an oestrous female and a male, test mice preferred the oestrous female rather than the male.
The mate preference score was significantly higher following CNO

| DISCUSSION
The present study utilised a chemogenetic approach via expression Male mice are characteristically attracted to a sexually receptive female and this approach behaviour is induced by female pheromones. 43 The results from the present study demonstrate a site-specific regulatory role for kisspeptin in the motivation to approach an oestrous female. This finding is in close agreement with studies involving Kiss1r knockout mice, where male mice displayed equal preference for both male and female conspecifics. 20 Previous studies have also shown that to men than those collected at other times of the menstrual cycle, 47,48 and that this exposure to body odour is associated with mating motivation. 49 Furthermore, mate odour preference is associated with reward-seeking activities 50 and is used as a measure of pleasurable behaviour in male mice. 51 Female odour activates reward centres in the brain, causing dopamine release from the nucleus accumbens of male mice. 50  In men, kisspeptin-induced enhancement of limbic brain activation in response to sexual images is independent of testosterone secretion. 25 We suggest that the enhanced sexual partner preference observed in the present study may derive from kisspeptin synaptic control rather than from any androgenic influence.
Sexual partner preference positively correlates with copulatory behaviour in rodents. A deficit in sexual partner preference is usually accompanied by prolonged latency to mount or ejaculate. 15,53 The outcome of enhanced preference score in the present study was, however, not facilitatory to the expression of male sexual behaviour. In male rats, Fos immunoreactivity indicated that a cluster of neurones in the MePD is involved in modulating copulatory behaviour. 54 Because the levels of male sexual behaviour (mount, intromission and ejaculation) were equivalent in the presence and absence of DREADDs activation of MePD kisspeptin neurones, it is plausible to attribute a limited role for kisspeptin in the MePD on male coital behaviour or to assume that sexual partner preference and sexual behaviour are differentially regulated. We have recently reported a dose-dependent effect of kisspeptin in the MePD on ex-copula erections in rats, which is analogous to human psychogenic erections caused by erotica. 21 Given that the previously utilised pharmacological approach may involve a mechanistic pathway distinct from the DREADDs technique, it is possibile that the endogenous kisspeptin signalling induced by hM 3 Dq receptor activation over a short term may not be sufficient to elicit any change in sexual behaviour.
Furthermore, the MePD is also considered as a neural hub that controls anxiety and social behaviour. 16 There is a parallel interaction between anxiety and sexual behaviour. Anxious male mice exhibit reduced sexual motivation 58 and treatment with anxiolytic agents exerts a corrective effect on sexual interest in depressed mice. 59 In men, anxiety-related disorders occur in tandem with sexual dysfunction. 60 It is therefore not surprising that kisspeptin may coordinate sexual preference and anxiety behaviour in an integrated fashion that is positive towards copulation. Moreover, the enhancement of limbic brain activity by kisspeptin in men viewing sexual images correlates with the attenuation of negative mood and reduced sexual aversion, 25 which indicates that, by modulating limbic brain activity, kisspeptin is critical for normal reproductive behaviour.
The sexual dimorphic nature of kisspeptin in the MePD could not be explored in the present study because we only investigated the sociosexual response of male mice without a female. This limited scope narrows the far-reaching significance of the present study, although our planned future work aims to ascertain the behavioural outcome in the female. Speculatively, a similar trend may be obtainable in the female, given the response of hypothalamic kisspeptin neurones of female mice to olfactory cues, 45  MePD of males in the present study was greater and supports the sex variation in MePD kisspeptin population. 19 However, differences in tissue preparation, the number of sections counted and section thickness cannot be ignored as variable factors. 26 Future studies are required to address this technical ambiguity.
In conclusion, the data obtained in the present study suggest that activation of kisspeptin neurones in the MePD synchronises socio-sexual behaviour by enhancing preference for a sexual partner and the anxiolytic response to promote maximal reproductive success in the male.