*For correspondence:rross4@partners.org (RAR);vnavarro@bwh.harvard.edu (VMN)Competing interests: Theauthors declare that nocompeting interests exist.Funding: See page 15Received: 16 February 2018Accepted: 14 June 2018Published: 15 June 2018Reviewing editor: Joel KElmquist, University of TexasSouthwestern Medical Center,United StatesCopyright Ross et al. Thisarticle is distributed under theterms of the Creative CommonsAttribution License, whichpermits unrestricted use andredistribution provided that theoriginal author and source arecredited.PACAP neurons in the ventralpremammillary nucleus regulatereproductive function in the femalemouseRachel A Ross1,2,3,4*, Silvia Leon3,5, Joseph C Madara3,6, Danielle Schafer7,Chrysanthi Fergani3,5, Caroline A Maguire5, Anne MJ Verstegen3,6,Emily Brengle5,6, Dong Kong8,9, Allan E Herbison7, Ursula B Kaiser3,5,Bradford B Lowell3,6, Victor M Navarro3,5*1Department of Psychiatry, Beth Israel Deaconess Medical Center, Boston, MA,United States; 2Department of Psychiatry, Massachusetts General Hospital,Massachusetts, United States; 3Harvard Medical School, Massachusetts, UnitedStates; 4McLean Hospital, Boston, United States; 5Department of Medicine, Divisionof Endocrinology, Diabetes and Hypertension, Brigham and Women’s Hospital,Boston, United States; 6Department of Medicine, Division of Endocrinology,Diabetes and Metabolism, Beth Israel Deaconess Medical Center, Massachusetts,United States; 7Centre for Neuroendocrinology, Otago School of Medical Sciences,University of Otago, Dunedin, New Zealand; 8Department of Neuroscience, TuftsUniversity School of Medicine, Massachusetts, United States; 9Sackler School ofGraduate Biomedical Sciences, Tufts University, Boston, United StatesAbstract Pituitary adenylate cyclase activating polypeptide (PACAP, Adcyap1) is aneuromodulator implicated in anxiety, metabolism and reproductive behavior. PACAP globalknockout mice have decreased fertility and PACAP modulates LH release. However, its source androle at the hypothalamic level remain unknown. We demonstrate that PACAP-expressing neuronsof the ventral premamillary nucleus of the hypothalamus (PMVPACAP) project to, and make directcontact with, kisspeptin neurons in the arcuate and AVPV/PeN nuclei and a subset of these neuronsrespond to PACAP exposure. Targeted deletion of PACAP from the PMV through stereotaxicvirally mediated cre- injection or genetic cross to LepR-i-cre mice with Adcyap1fl/fl mice led todelayed puberty onset and impaired reproductive function in female, but not male, mice. Wepropose a new role for PACAP-expressing neurons in the PMV in the relay of nutritional stateinformation to regulate GnRH release by modulating the activity of kisspeptin neurons, therebyregulating reproduction in female mice.DOI: https://doi.org/10.7554/eLife.35960.001IntroductionThe hypothalamic-pituitary-gonadal (HPG) axis is tightly regulated throughout development toensure the proper timing of puberty onset and attainment of fertility. Metabolic cues play a criticalrole in this regulatory process by modulating the release of kisspeptin and/or gonadotropin-releas-ing hormone (GnRH) at the hypothalamic level (Navarro and Kaiser, 2013). Situations of energy def-icit and surfeit are known to cause fertility impairments (e.g. hypothalamic amenorrhea in anorexianervosa) through the suppression of pulsatile LH secretion, but the mechanism by which metabolicRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 1 of 18RESEARCH ARTICLEfactors regulate gonadotrophic axis function is largely unknown (Laughlin et al., 1998). Kisspeptins,secreted from kisspeptin (Kiss1) neurons in the arcuate (ARC), anteroventral periventricular and peri-ventricular (AVPV/PeN), are the main secretagogues of GnRH. Kiss1 neurons convey most of the reg-ulatory cues of the gonadotropic axis, which determines the proper pulsatile or surge-like pattern ofrelease of GnRH (Herbison, 2016). Despite this primary role of kisspeptin and GnRH neurons in reg-ulating the HPG axis, the signal transduction of metabolic information to these neurons is yet to befully elucidated.In this context, pituitary adenylate cyclase-acting polypeptide (PACAP, official gene symbolAdcyap1) has emerged as a neuroendocrine factor involved in the regulation of food intake andgonadotropin release (Hawke et al., 2009; Resch et al., 2013; Tanida et al., 2013; Halvor-son, 2014). PACAP has been implicated in central processes such as development of anxiety behav-ior (Mustafa, 2013) and circadian rhythms (Mertens et al., 2007) and, importantly, both the ligandand the receptor are present in a number of metabolic and reproductive nuclei in the hypothalamus,including the ventromedial hypothalamus (VMH) (Vaudry et al., 2000). The presence of PACAP inthese nuclei has been linked to the anorectic action of leptin at the hypothalamic level by modulat-ing the activity of POMC neurons directly (Hawke et al., 2009; Resch et al., 2013; Tanida et al.,2013; Martin et al., 2014; Mounien et al., 2006), but this has not been linked to its role in fertility.It is known that whole body deletion of PACAP signaling decreases fertility (Jamen et al., 2000) andcentral PACAP administration delays puberty onset in rodents (Szabo´ et al., 2002); however, itsaction on gonadotropin release remains controversial as both stimulatory and inhibitory actions havebeen reported in rodents after central PACAP infusion (Ko¨ves et al., 2014; Ko¨ves et al., 1998),indicating a degree of complexity that may be species- and/or sex-specific. Interestingly, a numberof studies have invoked PACAP in the metabolic role of leptin (Hawke et al., 2009; Tanida et al.,2013); however, whether this action is direct (PACAP released from leptin receptor (LepR) express-ing neurons) or indirect remains to be determined. In the context of reproduction, while LepR isexpressed in a subset of kisspeptin neurons (Smith et al., 2006), genetic studies have demonstratedthat the main site of leptin’s action to regulate reproduction is not on kisspeptin neurons directly,but rather through cells in the ventral premammillary nucleus (PMV) (Cavalcante et al., 2014;Donato et al., 2011; Donato et al., 2009; Cravo et al., 2013). We observed that PACAP is highlyexpressed in this nucleus and therefore we sought to investigate if PACAP in the PMV is importantin the regulation of reproductive function and whether it serves as a mediator of leptin to exert itsreproductive role.We generated conditional PACAP knockout (Adcyap1fl/fl) mice that allow for the deletion of thePACAP gene in the presence of cre-recombinase by genetic cross and viral injection, and we investi-gated whether PACAP is essential for full reproductive capabilities of female mice.We also used PACAP-i-cre mice (Krashes et al., 2014) that allowed us to investigate whetherPMVPACAP neurons synaptically connect to kisspeptin neurons at different neuroanatomical levels(i.e. ARC and AVPV/PeN) to determine how these PMVPACAP neurons fit into the central regulationof the reproductive axis.ResultsPACAP release from leptin-responsive neurons is essential for normaltiming of puberty onset and fertilityAn intact PMV is required to transduce metabolic information through leptin signaling to allow forHPG activity (Donato et al., 2011). We found that the PMV is the region of the brain with the high-est level of co-localization of pStat3 (an indirect marker of LepR activity) and PACAP (Figure 1a);approximately 87% of the PACAP neurons also express pStat3 in that region, though only 70% ofthe LepR neurons co-express PACAP. There are two other regions, both in the hypothalamus, thatalso show co-localization of pStat3 and PACAP, although to a lesser extent: the central nucleus ofthe ventromedial hypothalamus and the supramammillary nucleus (Figure 1—figure supplement 1).To investigate if PACAP is an important relay for leptin we produced mice with PACAP deleted con-ditionally from leptin receptor expressing neurons using the leptin receptor cre knock-in mouse(Leshan et al., 2006) crossed to the Adcyap1fl/fl mouse that we made. The LepR-i-cre mouseexpresses cre recombinase in cells that produce the long-form of the leptin receptor, which areRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 2 of 18Research article NeuroscienceFigure 1. PACAP release from leptin-responsive neurons is essential for normal timing of puberty onset and fertility. (A) Representativemicrophotograph of a coronal section depicting PACAP (green) and pSTAT (red) immunoreactivity in the PMV of adult female PACAP-i-cre;L10-GFPmice treated with 5 mg/kg leptin two hours before perfusion. Yellow cells show co-localization. (B–D) Puberty onset is delayed in LepR- Adcyap1fl/flmice. (B) Accumulated percentage of mice with vaginal opening (VO), an indirect assessment of puberty onset by daily visualization for two weeks postwean (Adcyap1fl/fl n = 9; LepRcre- Adcyap1fl/fl n = 9). (C) Mean postnatal day of VO (Adcyap1fl/fl n = 9; LepRcre- Adcyap1fl/fl n = 9). **p<0.01, Studentt-test. (D) Mean postnatal day of first estrous as determined by histology samples after VO taken over the course of one month. (Adcyap1fl/fl n = 4;LepRcre-Adcyap1fl/fl n = 7). *p<0.05, Student t-test. (E–G) Estrous cyclicity is dysregulated in LepR-Adcyap1fl/fl mice. (E) Representative examples ofdaily (30 days) estrous cycle phases of control Adcyap1fl/fl (n = 9) and LepRcre- Adcyap1fl/fl mice (n = 9). Mouse estrous cycle is 4 days on average,including estrous phase (E), proestrous phase (P) and met/diestrous phase (M/D) which are combined due to similarity by histology. (F) Mean number ofestrous cycles in 30 days. *p<0.05 Student t-test (Adcyap1fl/fl n = 9; LepRcre-Adcyap1fl/fl n = 9). (G) Mean cycle length in days. *p<0.05 Student t-test(Adcyap1fl/fl n = 9; LepRcre-Adcyap1fl/fl n = 9). (H) Number of pups per litter for three sets of breeding pairs over five months of ongoing breeding issignificantly lower in LepR- Adcyap1fl/fl mice. *p<0.05 Student t-test (Adcyap1fl/fl n = 18; LepRcre- Adcyap1fl/fl n = 18). (I) Preovulatory LH surge isblunted in LepR-Adcyap1fl/fl mice. Circulating levels of LH in the morning (AM 10:00 hr) and afternoon (PM 19:00 hr) under an LH-surge inducingprotocol. Adcyap1fl/fl n = 5; LepRcre- Adcyap1fl/fl n = 5). Different letters indicate statistically different values (2 Way ANOVA followed by Fisher’s posthoc test, p<0.05). (J) Circulating LH levels before and one week after OVX. The lack of steroid hormone feedback in control animals leads to a markedincrease in LH, which is blunted in LepR-Adcyap1fl/fl mice. Different letters indicate statistically different values (2 Way ANOVA followed by Bonferronipost hoc test, p<0.05), (intact: Adcyap1fl/fl n = 7; LepRcre-Adcyap1fl/fl n = 4. OVX: Adcyap1fl/fl n = 3; LepRcre-Adcyap1fl/fl n = 5). (K) Circulating LH levels25 min after the injection of vehicle (saline) or 1 nmol kisspeptin 10 (kp10) are not significantly different between the knockout and control animals.Different letters indicate statistically different values (2 Way ANOVA followed by Bonferroni post hoc test, p<0.05), (saline: Adcyap1fl/fl n = 7; LepRcre-Adcyap1fl/fl n = 3. Kp10: Adcyap1 n = 4; LepRcre-Adcyap1fl/fl n = 9). (L) Circulating LH levels after overnight fast and 30 min after central leptinadministration are blunted in LepR-Adcyap1fl/fl mice. Different letters indicate statistically different values (2 Way ANOVA followed by Fisher’s post hoctest, p<0.05), (fasting: Adcyap1fl/fl n = 3; LepRcre-Adcyap1fl/fl n = 7. Fasting + leptin: Adcyap1fl/fl n = 3; LepRcre-Adcyap1fl/fl n = 9).DOI: https://doi.org/10.7554/eLife.35960.002The following figure supplements are available for figure 1:Figure supplement 1. Representatives images of hypothalamic nuclei coronal sections depicting areas of co-labeling (yellow) of PACAP-ir (green) andpSTAT-ir (red) in adult female PACAP-i-cre;L10-GFP mice treated with 5 mg/kg IP leptin two hours before perfusion.DOI: https://doi.org/10.7554/eLife.35960.003Figure supplement 2. Validation of PACAP conditional allele.Figure 1 continued on next pageRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 3 of 18Research article Neurosciencefound primarily in the brain (Leshan et al., 2006), thus producing a conditional knockout of PACAPfrom neurons that express the leptin receptor. We validated this conditional knockout by qPCR andby RNA in situ hybridization (Figure 1—figure supplements 2 and 3a).Because this genetic recombination occurs before puberty onset, we were able to address thequestion of the role that PACAP in LepR neurons may play in puberty, which relies on a normal func-tioning HPG axis. Compared to non-cre expressing littermate controls, LepR-i-cre; Adcyap1fl/flfemales had delayed onset of puberty, measured both by vaginal opening and day of first estrus(Figure 1b–d). There was no significant change in body weight in any of the animals on regular chowor high fat diet (Figure 1—figure supplement 3b). As expected, deletion of PACAP from LepR neu-rons leads to dysregulated estrous cycling (Figure 1e–g), though less markedly than the lesion ofthe PMV region (Donato et al., 2011); these LepR-i-cre;Adcyap1fl/fl females were able to get preg-nant, but had fewer pups per litter (Figure 1h). Consistent with the ability to become pregnant, allstages of follicle development were present, including corpora lutea (Figure 1—figure supplement3c).This conditional genetic KO of PACAP from LepR neurons displayed marked HPG axis dysregula-tion. The LepR-i-cre;Adcyap1fl/fl animals showed a blunted LH surge (Figure 1i), which correlateswith the dysregulated estrous cycle (Figure 1e–g). They also had a blunted response to ovariectomyand fewer pups per litter (Figure 1h,j). Since neurons located in the PMV have been described tocontact GnRH neurons in addition to kisspeptin neurons (Donato et al., 2011), it is possible thatGnRH neurons require either PACAP or other PMV signals to elicit a proper response to kisspeptin,which could lead to the reproductive impairment seen in these mice. Therefore, in order to deter-mine if the function of GnRH neurons (and gonadotropes) remains intact in this model, we adminis-tered kisspeptin intracerebroventricularly (ICV) to these animals and compared LH release with theircorresponding controls. Kisspeptin administration led to a similar induction of LH in both groups(Figure 1k). This provides further evidence that GnRH neurons are able to respond normally to kiss-peptin and, therefore, the reproductive impairment observed in these mice must rely on, or relaythrough another neuron that stimulates kisspeptin release. Finally, leptin administration to these ani-mals in the morning after a fast, induced a lower than expected rise in LH release (Figure 1l); bycomparison to the littermate controls, the LH induction in response to leptin treatment was bluntedby 30%. This indicates that PACAP is responsible for transducing some, but not all, of the metabolicinformation relayed by leptin to the HPG axis.PACAP deletion from the PMV in adulthood leads to impairedgonadotrophic functionBecause the PMV is the site of greatest co-localization of LepR activity and PACAP, and prior studiesdetailed the importance of the region to the reproductive action of leptin (Donato and Elias, 2011),we investigated the role of PACAP in the PMV directly. To determine if PACAP secreted from thePMV is necessary for HPG axis function, we directly deleted PACAP from the PMV of adult femaleAdcyap1fl/fl mice with bilateral stereotaxic injections of an adenovirus carrying cre-recombinase(Figure 2a,b). We again found marked dysregulation of the normal estrous cycle when validatedPMVPACAP knockout (KO) mice were compared to littermate controls injected with a control virus(same serotype without cre, Figure 2c). Mice lacking PACAP in the PMV had fewer estrous cycles ina 25 day period and resultant longer cycle lengths (Figure 2d,e). To note, PMVPACAP KO animalsmaintained equivalent body weight to control animals throughout the study (Figure 2—figure sup-plement 1a).In order to determine if this deletion had effects on fertility, we mated the female mice to 8–12week old male C57Bl/6 wild-type mice for 1 week at a time. Males were removed after this period.Nearly all control mice became pregnant and successfully produced litters within 21 days of mating,indicating successful copulation on the first day of mating to the male. In contrast, the PMVPACAP KOFigure 1 continuedDOI: https://doi.org/10.7554/eLife.35960.004Figure supplement 3. Additional characterization of the metabolic and reproductive phenotype of LepRCre-Adcyap1fl/fl male and female mice.DOI: https://doi.org/10.7554/eLife.35960.005Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 4 of 18Research article Neurosciencemice had reduced fecundity, with decreased capacity to produce a litter after mating, and significantdelay in the time to become pregnant (Figure 2f). Furthermore, those pups that were produced bythe PMVPACAP KO females did not survive more than one day post partum (data not shown), whichwas thought to be due to neglect/infanticide by the dam based on observation of the animals inFigure 2. Ablation of PACAP expression from the PMV leads to impairment in the gonadotropic axis. (A) Schematic representation of the sites ofinjection of AAV8.CMV.HI.eGFP-Cre showing the extent of viral spread around the PMV for Adcyap1fl/fl study animals. Animals were included inanalyses only if they had bilateral injection across >70% of the rostro-caudal diameter of the PMV that did not spread into surrounding areas of PACAPexpression (i.e. VMH, dorsal premammillary nucleus). Control virus (AAV8.CMV.PI.eGFP.WPRE.bGH) was injected bilaterally in littermate controls of thesame genotype (data not shown). (B) Representative ISH depicting coronal section with Adcyap1 mRNA in the PMV of AAV8.CMV.PI.eGFP.WPRE.bGHcontrol injected (left panel) and AAV8.CMV.HI.eGFP-Cre injected Adcyap1fl/fl female mice. Only animals with bilateral deletion of >70% Adcyap1 mRNAwere included in analyses. (C – E) Assessment of estrous cyclicity shows marked dysregulation in the AAV-cre treated Adcyap1fl/fl mice. (C)Representative examples of daily (25 days) estrous cycle phases of control (n = 8) and AAV-cre injected mice (n = 6). (D) Mean number of estrous cyclesin 25 days. *p<0.05 Student t-test (AAV-ctrl:Adcyap1fl/fl n = 8; AAV-cre:Adcyap1fl/fl n = 10). (E) Mean cycle length in days. *p<0.05 Student t-test (controln = 6; AAV-cre n = 8). (F) AAV-cre mediated deletion of PACAP from PMV prolongs parturition latency represented as accumulated percentage ofanimals giving birth per day after mating with C57Bl/6 male for 1 week (AAV-ctrl:Adcyap1fl/fl n = 8; AAV-cre:Adcyap1fl/fl n = 6). (G) Analysis of themagnitude of the preovulatory LH surge. Circulating levels of LH in the morning (AM 10:00 hr) and afternoon (PM 19:00 hr) under an LH-surge inducingprotocol. (AAV-ctrl:Adcyap1fl/fl n = 3; AAV-cre:Adcyap1fl/fl n = 3). Different letters indicate statistically different values. Only AAV-Ctrl injected miceshowed a significant increase from the AM to the PM. (2 Way ANOVA followed by Bonferroni post hoc test, p<0.05). (H) Circulating LH levels beforeand one week after OVX are not significantly different between the AAV-cre and AAV-ctrl treated Adcyap1fl/fl mice. Different letters indicate statisticallydifferent values (2 Way ANOVA followed by Bonferroni post hoc test, p<0.05), (AAV-ctrl:Adcyap1fl/fl n = 8; AAV-cre:Adcyap1fl/fl n = 10). (I) Circulating LHlevels 25 min after the injection of vehicle (saline) or 1 nmol kisspeptin 10 (kp10) are intact. Different letters indicate statistically different values (2 WayANOVA followed by Bonferroni post hoc test, p<0.05), (AAV-ctrl:Adcyap1 fl/fl n = 8; AAV-cre:Adcyap1fl/fl n = 10). (J) Circulating LH levels after overnightfast and 30 min after central leptin administration indicate PACAP is involved in, but is not necessary for leptin signal transduction through the PMV.Different letters indicate statistically different values (2 Way ANOVA followed by Fisher’s post hoc test, p<0.05), (AAV-ctrl:Adcyap1fl/fl n = 12; AAV-cre:Adcyap1fl/fl n = 8).DOI: https://doi.org/10.7554/eLife.35960.006The following figure supplement is available for figure 2:Figure supplement 1. PACAP deletion from PMV neurons in adult females does not affect body weight but decreases the number of corpora lutea.DOI: https://doi.org/10.7554/eLife.35960.007Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 5 of 18Research article Neurosciencetheir home cages. Though we could not assess the litter size, we observed a decrease in the numberof corpora lutea in these mice (Figure 2—figure supplement 1b), which suggests an impairment inovulation, consistent with a smaller number of follicles maturing per cycle, as a consequence ofreduced gonadotropins. In control-injected animals the increase in LH between baseline (10:00) andsurge (19:00) is significant, whereas for the KO animals only a trend to increased LH was observed at19:00, suggesting an alteration in the ability of these mice to mount a normal ovulatory LH surge(Figure 2g). On the other hand, the compensatory rise in LH after ovariectomy was not differentbetween the two groups (Figure 2h), indicating that the circuitry downstream of gonadal feedback,particularly the ARCkisspeptin population, responds normally to the absence of sex steroids.Similar to what we observed in response to kisspeptin administration in the LepR-i-cre; Adcyap1fl/fl ablated mice, PMVPACAP KO mice had a normal response to kisspeptin administration (Figure 2i),indicating intact signaling between kisspeptin and GnRH neurons, and supporting a role for PACAPto modulate kisspeptin release. To consider if PMVPACAP is necessary to relay the permissive signalof leptin to gonadotropin release, we administered leptin centrally after an overnight fast to assesswhether, as in LepR-i-cre; Adcyap1fl/fl animals, the magnitude of the LH increase is reduced com-pared to controls. Unexpectedly, the LH response was not reduced, instead there was a slight butsignificant increase in LH induction by leptin treatment (Figure 3j). Overall these data show thatPACAP from the PMV is necessary for normal reproductive function but not for the relay of leptinsignal from this nucleus.PMVPACAP neurons are glutamatergic, monosynaptically connect toARC and AVPV/PeN kisspeptin neurons, and PACAP directly stimulatesactivity in a regionally distinct subset of these neurons.PMVPACAP neurons send projections to both regions of kisspeptin neuron populations, the AVPV/PeN and the ARC (Figure 3a,b, Figure 3—figure supplement 1a,b). These PMVPACAP neurons areglutamatergic (Figure 3c), though there are also PMV glutamatergic neurons that do not expressPACAP. We used channelrhodopsin-assisted circuit mapping in order to determine if these closeprojections from glutamatergic, PACAPergic PMV neurons make direct monosynaptic contact witheach subpopulation of kisspeptin neurons. In the ARC, where kisspeptin neurons participate in theregulation of GnRH pulsatility (Clarkson et al., 2017), and the expression of the PACAP receptorhas been demonstrated (Campbell et al., 2017), blue light stimulation of PMVPACAP neurons at reg-ular intervals (1 s) in brain slices led to immediate (<6 ms latency) excitatory postsynaptic currents inkisspeptin neurons. In female mice (n = 5), there was a monosynaptic glutamatergic connectionbetween PMVPACAP neurons and ARCkisspeptin neurons (14/19 kisspeptin neurons tested throughoutthe ARC), and this was less frequently observed in non-kisspeptin ARC neurons (only 5/17 neuronsreceived direct monosynaptic input from PMVPACAP neurons) (Figure 3d). Similarly, in the AVPV/PeN, where the kisspeptin neurons are thought to drive the large GnRH release that leads to thepreovulatory LH surge (Herbison, 2016), the majority of AVPV/PeNkisspeptin neurons tested receiveddirect input from PMVPACAP neurons (11/18 neurons connected) but only a small fraction of non-kiss-peptin neurons assayed showed direct connectivity in response to activation of PMVPACAP neurons(2/13 neurons connected).Because channelrhodopsin-assisted circuit mapping does not resolve the effects of slow neuro-peptidergic neurotransmission, we isolated the effect of PACAP applied directly to kisspeptin neu-rons in both the AVPV/PeN and the ARC. We used brain slice calcium imaging to examine theresponses of kisspeptin neurons located throughout both regions to PACAP (10 nM). Brain slicescontaining the AVPV and PeN or the rostral, middle and caudal aspects of the ARC were preparedfrom Kiss1-Cre;GCaMP6 mice (Figure 3e). PACAP was found to induce different effects on the basaland spontaneous calcium fluctuations observed in 34 AVPV/PeNkisspeptin neurons (four slices fromeach of 4 diestrous mice). Nine cells (25%) exhibited increases in the numbers of calcium fluctuationswith, typically, no change in basal fluorescence, ten cells (29%) showed decreases in basal levels aswell as the numbers of calcium fluctuations, and 15 neurons (46%) showed no response to PACAPapplication (Figure 3f). PACAP was found to induce region-dependent effects on fluorescence levelsin ARCkisspeptin neurons. A total of 27 rostral, 64 middle, and 68 caudal neurons (each from 4 slicesfrom four diestrous mice) were tested. While no cells in the rostral and middle ARC responded toPACAP, 17/68 (25%) of caudal kisspeptin neurons exhibited increases in fluorescence levels duringPACAP application, with a decrease to basal fluorescence upon washout (Figure 3h). The remainderRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 6 of 18Research article Neuroscienceshowed no response to PACAP application. Together, with the channelrhodopsin circuit mappingexperiments, these findings indicate that PMVPACAP neurons project directly to kisspeptin neurons,but PACAP has direct stimulatory effect on a subpopulation of neurons located in the caudal aspectof the ARC, and interspersed through the AVPV/PeN. Furthermore, PACAP has a net inhibitoryeffect on another subpopulation of AVPV/PeNkisspeptin neurons. Our experiments cannot distinguishif this PACAP activity originates from a particular population of PMVPACAP neurons (e.g. LepR), yetFigure 3. PMVPACAP neurons monosynaptically contact a subset of ARC and AVPV Kisspeptin neurons. (A) Schematic representation of the site ofinjection of AAV8.2-eF1a-DIO-syp-mCherry-WPRE (projection tracing, n = 3) or AAV8-hSyn-DIO-ChR2(H134R)-mCherry (circuit mapping, n = 5). (B)Representative microphotographs depicting immunoreactivity of projections from PMVPACAP neurons (red) and Kiss1 neurons (green) in the ARC andAVPV/PeN nuclei after injection of AAV8.2-eF1a-DIO-syp-mCherry-WPRE into the PMV in PACAP-i-cre female mice. (C) Representative double label ISHdepicting co-localization of Adcyap1 (white) and Vglut2 (pink) mRNA in the PMV of female C57Bl/6 mice in a coronal section. Right image is anenlarged view of the boxed section in the left image. (D) Channelrhodopsin assisted circuit mapping (CRACM) analysis of projections from PACAP-i-cremice injected with AAV-DIO-syp-ChR2-mcherry photo-stimulated in the ARC or AVPV of Kiss1-GFP mice. The majority of tested kisspeptin neurons inboth regions receive direct excitatory input from PMVPACAP neurons (14/19 in ARC, 11/18 in AVPV/PeN), and a small number of non-kisspeptin neuronsin each region (5/17 and 2/13 respectively) do as well. Arrows depict blue light pulse (473 nm, 5 ms epoch) administered 1 s apart during the first 4 s ofan 8 s sweep, repeated for a total of 30 sweeps per recorded cell. Black line shows the average of all sweeps, and gray lines are individual sweeps. (E)Representative photomicrograph of coronal section showing GCaMP6f fluorescence in the AVPV/PeN specific to kisspeptin neurons of a diestrousfemale Kiss1-GCaMP6f mouse. (F) Traces showing the effect of 10 nM PACAP (grey bar) on GCaMP6f fluorescence levels (delta F/F) in three kisspeptincells recorded simultaneously from the AVPV/PeN indicating PACAP causes different responses in different populations of neurons in this region. (G)Representative photomicrograph of coronal section showing GCaMP6f fluorescence in the caudal arcuate nucleus specific to kisspeptin neurons of adiestrous female Kiss1-GCaMP6f mouse. (H) Traces showing the effect of 10 nM PACAP (grey bar) on GCaMP6f fluorescence levels (delta F/F) in fivekisspeptin cells recorded simultaneously from the caudal ARC indicating PACAP causes increased activity in a subpopulation of kisspeptin neurons.DOI: https://doi.org/10.7554/eLife.35960.008The following figure supplement is available for figure 3:Figure supplement 1. Representative images of PMV injection site for tracing and electrophyisiology experiments.DOI: https://doi.org/10.7554/eLife.35960.009Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 7 of 18Research article Neurosciencethis is further evidence of the important role of the complex interactivity of kisspeptin neurons ingenerating both pulsatile and surge LH secretion (Clarkson et al., 2017).DiscussionIn this study, we show that a large percentage of PMV neurons express PACAP (and glutamate),which modulates kisspeptin neurons, thus affecting fertility, estrous cycle, and puberty onset (Fig-ure 4). The most striking finding is that deletion of PACAP from the PMV of adult female mice leadsto >50% reduction in fecundity, a novel role for the neuromodulator specific to the PMV. Previouswork has shown that PACAP is critical for fertility in whole body animal knockouts or in whole braininjection studies (Jamen et al., 2000; Ko¨ves et al., 2003; Shintani et al., 2003), but the mechanismand localization of action was unknown. This is the first study to show that a small population ofhypothalamic glutamatergic LepR-expressing and PACAPergic neurons must express PACAP fornormal timing of puberty onset and fertility in female mice.PACAP effects have been shown to be sexually dimorphic, which is thought to be due to theestrogen response element of the PACAP receptor (Ramikie and Ressler, 2016). Interestingly, inthis study, we observed that deletion of PACAP from the leptin-responsive neurons in female miceleads to delayed vaginal opening and delayed first estrus, both indirect markers of puberty onset,despite both groups having similar body weight. These data point to a crucial role of PACAP in thetiming of sexual maturation in female mice, and aver the importance of future studies in males.Figure 4. Schematic representation of the hierarchical inputs from PMVPACAP neurons to AVPV/PeN and ARC Kiss1 neurons. Briefly, PMVPACAP neuronsregulate reproductive function by directly modulating subpopulations of AVPV/PeNkisspeptin and caudal ARCkisspeptin neurons, which leads to theregulation of the preovulatory GnRH/LH surge and pulsatile GnRH release, respectively, through the action of kisspeptin at the level of GnRH neurons.Through the direct action of leptin, PMVPACAP neurons mediate in part the metabolic regulation of reproductive function.DOI: https://doi.org/10.7554/eLife.35960.010Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 8 of 18Research article NeuroscienceBecause PACAP and LepR are expressed in the gonads and other peripheral tissues, the potentialdeletion of PACAP from those areas could be possible in our LepR-i-cre; Adcyap1fl/fl mouse model.Use of the LepR-i-Cre mouse should mostly restrict deletion of PACAP to the brain, however, thelong form of the leptin receptor may be expressed at low levels in the periphery (Leshan et al.,2006) and use of genetic cre-dependent manipulations can also lead to off-target deletion in neu-rons that express the LepR promoter during development but not in adulthood. Interestingly, in thegonads, LepR and PACAP are expressed in different cell types (Duggal et al., 2002; Reglodi et al.,2012; Ryan et al., 2003). There is a chance as well for off-target deletion of PACAP from cells (cen-tral or peripheral) that do not express LepR in adult animals. Still, both the LepR-i-cre mediated dele-tion of PACAP and the central AAV-delivery based manipulations recapitulate the effects ofpreviously published ablation of PMV neurons (Donato et al., 2011). Taken together, our studydemonstrates that the role of PACAP on the HPG axis is centrally mediated, specifically at the PMV.As described above, PACAP seems to mediate the action of leptin (Hawke et al., 2009;Tanida et al., 2013; Ko¨ves et al., 2003). Circulating levels of leptin are essential for gonadotropinrelease (Ahima et al., 1996) and, as such, acute decreases in circulating leptin during fasting orundernourishment lead to hypogonadotropic hypogonadism. In this study, removal of PACAP fromleptin-responsive neurons blunted the ability of exogenous leptin to increase LH levels after fasting,but direct deletion of PACAP from the PMV did not, suggesting that PACAP from the PMV is onlypartially necessary for the normal permissive role of leptin on the gonadotropic axis. We surmisethat other factors (neuromodulators or neurons), or other leptin-responsive, PACAP-expressingregions of the brain, such as the VMH or supramammillary nucleus are likely involved in the transmis-sion of the reproductive role of leptin (Padilla et al., 2017). Notably, PACAP neurons—and the PMVin general—are predominantly glutamatergic, yet our recent studies indicated that leptin’s metabolicand reproductive actions are mediated by GABAergic transmission (Martin et al., 2014;Vong et al., 2011). While the absence of marked metabolic differences in these mice is in line withearlier studies (Vong et al., 2011), the hypogonadism observed in these mice are the first evidenceof a divergent pathway of leptin to regulate metabolism and reproduction. Our data indicate thatPACAP likely acts together with other neuromodulators produced in the same or separate LepR-expressing neurons through which leptin exerts its reproductive effect. Given that only a subset ofkisspeptin neurons are directly activated by PACAP, and that a subpopulation of kisspeptin neuronsis GABAergic, it is possible that these two populations of neurons (i.e. PMVPACAP and kisspeptin neu-rons) work in concert in response to different types of input from the PMV (i.e. PACAP-ergic or glu-tamatergic). Kiss1 and GnRH neuron firing patterns are very tightly regulated (Herbison, 2016) andfurther studies focusing on the neuromodulatory effects of PACAP (and other co-expressed peptidesfrom the PMVPACAP neurons) will be required to further disentangle the role of PMV neurons in thecontrol of GnRH release.We observed that the majority of AVPV/PeNkisspeptin and ARCkisspeptin neurons tested receivedirect contact from PMVPACAP neurons, supporting a role for this neuropeptide in the control of theLH surge in females. Indeed, deletion of PACAP from the PMV either through viral delivery of crerecombinase to Adcyap1fl/fl mice or through crossing with LepR-i-cre mice led to a blunted LH surgeand a smaller litter size, indicating an impairment in the ability of the mouse to ovulate and the pres-ence of fewer mature follicles, which is also reinforced by fewer corpora lutea in the AAV-creinjected model. Moreover, the subfertility induced by both models of PACAP deletions led to irregu-lar estrous cycles, indicating a role in LH pulsatility. Estrous cyclicity is determined by the ability ofthe mouse to mount an ovulatory process in the afternoon of proestrus and is the result of increasinglevels of estradiol as a consequence of low frequency GnRH/LH pulses during diestrus (thought tobe driven by ARCkisspeptin) (Clarkson et al., 2017), which drives more follicles to mature. The matur-ing follicles in turn synthesize more estradiol, which stimulates AVPV/PeNkisspeptin neurons(Smith et al., 2005; Pinilla et al., 2012), inducing an LH surge. These reproductive impairments sug-gest that PACAP from the PMV plays a role in both pulsatile and surge-like release of GnRH(Figure 4).Within ARCkisspeptin neurons, the direct response to PACAP is specific to regional clusters of thecaudal arcuate. These subpopulations of ARCkisspeptin cells have not been separately defined fromthe rest of the ARC population and this is the first time that PACAP activity has been shown todirectly affect the activity of the ARCkisspeptin population, which express PACAP receptor(Campbell et al., 2017). ARCkisspeptin neurons are primarily involved in setting the tone for GnRHRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 9 of 18Research article Neurosciencepulsatility (Clarkson et al., 2017), a requisite for normal HPG axis function, therefore indicating alikely role of PMVPACAP in the fine tuning of kisspeptin and GnRH pulses. Similarly, the AVPV/PeNkiss-peptin neurons are a heterogenous group based on their response to PACAP. This indicates thatthere are likely different receptors and second messenger systems activated by PACAP in subpopu-lations of AVPV/PeNkisspeptin neurons that may play different roles in the timing and magnitude ofthe preovulatory LH surge. As above, the ability of GnRH neurons to respond normally to kisspeptinin the absence of PMVPACAP signaling, strongly suggests an action on (as evidenced by Figure 3) orupstream of Kiss1 neurons. Furthermore, the compensatory LH rise in response to ovariectomy inthese animals is also intact suggesting that PMVPACAP neurons are not required for sex steroid nega-tive feedback, but instead may be involved in a separate pathway, such as the nutritional regulationof reproductive function. Together, these data indicate the neuromodulatory role of PACAP in theHPG axis is likely more subtle than the steroid hormone feedback, but remains important in itsfunction.In summary, we document the first evidence of a role for PACAP from the PMV in ovulatorycycling and subsequent fertility in females, which is required for normal reproduction activity (Fig-ure 4). PACAP may also contribute to the reproductive role of leptin through the activation of gluta-matergic, PACAPergic, leptin responsive neurons, but within the PMV, leptin signal transductiondoes not require PACAP expression. We also show for the first time that ARCkisspeptin and AVPV/PeNkisspeptin neurons are both heterogeneous populations, which can be defined by their responseto PACAP, and this circuit is likely important for fine tuning of the HPG axis. These findings shedlight on the complex mechanisms that underlie the neuroendocrine regulation of reproduction andmay offer the platform to develop new treatments for reproductive disorders associated with meta-bolic impairments, such as anorexia nervosa, polycystic ovarian syndrome, or obesity.Materials and methodsAnimals and surgeriesSubjectsAll animal care and experimental procedures were approved by the National Institute of Health,Beth Israel Deaconess Medical Center and Brigham and Women’s Hospital Institutional Animal Careand Use Committee, protocol #05165. The Brigham and Women’s Hospital is a registered researchfacility with the U.S. Department of Agriculture (#14–19), is accredited by the American Associationfor the Accreditation of Laboratory Animal Care, and meets the National Institutes of Health stand-ards as set forth in the Guide for the Care and Use of Laboratory Animals (DHHS Publication No.(NIH) 85–23 Revised 1985). Mice were housed at 22–24˚C with a 12 hr light (06:00)/dark (18:00) cyclewith standard mouse chow (Teklad F6 Rodent Diet 8664) and water provided ad libitum. For behav-ioral studies, females between age 8 and 16 weeks were used. For electrophysiologic studies, femalemice between age 5 and 12 weeks were used. All cre-driver and cre-reporter mice were used in theheterozygous state. All transgenic mice were maintained on a mixed background and have beendescribed previously, except for the pacap-lox mice. Wild-type, Adcyap1fl/fl mice and PACAP-i-cre(Krashes et al., 2014) were maintained as separate litters and group housed according to sex.LepR-i-cre mice (Leshan et al., 2006) (Martin Myers, University of Michigan) were crossed toAdcyap1fl/fl and non-cre expressing homozygous flox allele offspring of the cross were used as litter-mate controls. Crosses were made between PACAP-i-cre and Kiss1hr-GFP (humanized Renilla GreenFluorescent Protein) (Cravo et al., 2013) (gift of Carol Elias, University of Michigan) for electrophysi-ology studies, and PACAP-i-cre or vglut2-cre and Rosa26/CMV/Actin-loxSTOPlox-L10GFP mice(L10-GFP, David Olson) for immunohistological detection.To make the Adcyap1fl/fl mice, the lox-modified PACAP targeting construct was made by recom-bineering technology. To engineer the targeting vector, 5’ homology arm, 3’ homology arm andCKO region were amplified from mouse Sv129 BAC genomic DNA and confirmed by end sequenc-ing (Cyagen biosciences, Santa Clara, CA). The two loxP sites flank the second exon and whenrecombined, create a frameshift mutation and truncated protein. The plasmid was electroporatedinto W4 ES cells and cells expanded from targeted ES clones were injected into C57Bl/6 blastocysts.Germline transmitting chimeric animals were obtained and then mated with flpE mice to delete thefrt-site flanked neomycin selection cassette. The resulting heterozygous offspring were crossed toRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 10 of 18Research article Neurosciencegenerate homozygous Adcyap1fl/fl study subjects. All mice are thus on a mixed C57Bl/6J and 129Svbackground. Offspring were genotyped by PCR using 2 primers (F: CCGATTGATTGACTACAGGCTCC and R: GTGTTAAACACCAGTTAGCCACGC) which detect the presence or absence of the 5’loxP site and a 3rd primer was used in conjunction with the forward primer (CKO-R GGGCTTTGATCTGGGAACTGAAG) to detect the recombination event. By generating mice homozygous for a germ-line cre-deleted allele (by cross to E2A-cre, Jax labs), we have established that the cre-deleted alleledoes not express intact Adcyap1 mRNA (by PCR and by ISH, Figure 2—figure supplement 1).Subject historyFor viral mediated PACAP deletion studies, Adcyap1fl/fl mice were randomly assigned to control orAAV-cre treated groups. For transgenic mice, groups with the presence or absence of Cre weredetermined by genotype of mouse and then randomly distributed. Animals were naı¨ve to experi-mental testing before beginning the study. Multiple LH induction studies were conducted in thesame animal subjects for each cohort, but each type of study was performed only once per cohort.Viral InjectionsStereotaxic injections were performed as previously described in female mice between ages 6–8weeks. Mice were anaesthetized with xylazine (5 mg/kg) and ketamine (75 mg/kg) diluted in saline(350 mg per kg) and placed into a stereotaxic apparatus (KOPF Model 963 or Stoelting). For postop-erative care, mice were injected intraperitoneally with meloxicam (5 mg/kg). After exposing the skullvia small incision, a small hole was drilled for injection. A pulled-glass pipette with a 20–40 nm tipdiameter was inserted into the brain, and virus was injected by an air pressure system. A microma-nipulator (Grass Technologies, Model S48 Stimulator) was used to control injection speed at 25 nlminone and the pipette was withdrawn 5 min after injection. For electrophysiology experimentsAAV8-hSyn-DIO-ChR2(H134R)-mCherry (University of North Carolina Vector Core; titer 1.3 1012genome copies per ml) was injected unilaterally into the ventral premamillary region (PMV 25 nL,AP: 2.47, DV: 5.6, ML:+0.55) of PACAP-i-cre and Kiss1hr-GFP mice. For PACAP in vivo deletionstudies, AAV8.CMV.HI.eGFP-Cre.WPRE.SV40 or AAV8.CMV.PI.eGFP.WPRE.bGH for control (Univer-sity of Pennsylvania Vector Core; titer 1.01 1013 and 1.39 1013 respectively) was injected bilater-ally into the PMV of Adcyap1fl/fl mice (same volume and coordinates). Bilateral hits were definedpost-hoc by visual comparison with Allen Brain Atlas delineation of PMV region, using immunohis-tochemistry to check the spread of injection, and RNA ISH to confirm that Adcyap1 was in factdeleted. Animals with only unilateral PMV Adcyap1 deletion, or with deletion that extended to theVMH, were discarded from analysis.For tracing studies from PACAP-ergic neurons, AAV8.2-eF1a-DIO-syp-mCherry-WPRE (Massachu-setts Institute of Technology Virovek core; titer 2.23 1013 genome copies per ml) was injected uni-laterally to the PMV (same volume and coordinates). Mice were given a minimum of 2 weeks forrecovery and 1 week for acclimation before being used in any experiments.Intracerebroventricular (ICV) injection2–3 days before the experiment, the mice were briefly anesthetized with isoflurane and a small holewas bored in the skull 1 mm lateral and 0.5 mm posterior to bregma with a Hamilton syringeattached to a 27-gauge needle fitted with polyethylene tubing, leaving 3.5 mm of the needle tipexposed. Once the initial hole was made, all subsequent injections were made at the same site. ForICV injections, mice were anesthetized with isoflurane for a total of 2–3 min, during which time 5 mlof solution were slowly and continuously injected into the lateral ventricle. The needle remainedinserted for approximately 60 s after the injection to minimize backflow up the needle track. Micetypically recovered from the anesthesia within 3 min after the injection. For hormonal analyses, bloodsamples (4 ml) were obtained from the tail before and 30 min after injection. Recombinant mouseLeptin (2 ug/5 ul) was purchased from Preprotech. Mouse kisspeptin-10 (Kp-10) (1 nmol/5 ul) wasobtained from Phoenix Pharmaceuticals. The drugs were dissolved in saline (0.9% NaCl). The doseand time of collection were selected based on our previous studies (Navarro et al., 2015).Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 11 of 18Research article NeuroscienceEstrous cycle monitoringFemale mice are spontaneous ovulators and typically have a 4 day estrous cycle that consists of fourstages: proestrus, estrus, metestrus/diestrus I and diestrus II. Mice were habituated to handling andvaginal smears before behavioral experiments. Mice were swabbed daily at 9:00am to check for con-sistent cycling (Caligioni, 2009) and underwent blood sampling at 10:00am, except for LH surge,which occurred just after the onset of the dark cycle. Using a disposable pipette, a small volume ofphysiological sterile water was placed near the opening of the vagina and 10 ul was flushed in threerepetitions with minimal insertion to avoid pseudopregnancy. The smear was displaced onto a glassslide and after air-drying, was examined under brightfield microscope at 20X. The identification ofestrous stages was based on characteristic cell type appearance observed and the density of eachcell type in the vaginal secretion (Caligioni, 2009). Estrous stage was identified by changes in cellappearance across the cycle that reflects circulating gonadal steroids. A cycle was considered as thetime between two consecutive estrous phases with at least one day of diestrous phase in betweenthem.FecundityFemale study mice were paired with 8 week old C57Bl/6 male mice for 1 week, at which time themales were removed from the cages and females were left individually housed with cage enrich-ment, and monitored for signs of pregnancy. Post-birth, females were paired again for 1 week with anew C57Bl/6 male to ensure any potential infertility was due to the female only.Tail-tip bleedingBlood sampling was performed after a single excision of the tip of the tail. Tail was cleaned withsaline and then massaged to take a 4 ul blood sample from the tail tip with a pipette. Whole bloodwas immediately diluted in 116 ul of 0.05% PBST, vortexed, and frozen on dry ice. Samples werestored at 80˚C for a subsequent LH ELISA (Steyn et al., 2013).OvariectomiesThe time-course of changes in the circulating levels of LH in response to ovarectomy (OVX) wasexplored in female mice to assay for feedback mechanisms of gonadal steroids on HPG axis. Groupsof adult female mice were subjected to bilateral OVX via abdominal route, as previously described(Garcı´a-Galiano et al., 2012). Blood samples were obtained before and 7-d after OVX to use for LHELISA (Garcı´a-Galiano et al., 2012).Induction of LH surge/E2 replacementImmediately after OVX, capsules containing 0.625 ug of 17-b estradiol dissolved in sesame seed oilwere implanted subcutaneously via a small midscapular incision at the base of the neck. Silastic tub-ing (15 mm long, 0.078 in inner diameter, 0.125 in outer diameter; Dow Corning) was used for cap-sule preparation. Whole blood samples were collected for LH analysis 2 days after surgery in themorning (10:00am EST) and in the evening just after lights off (19:00) (Dror et al., 2013).ElectrophysiologyFemale animals in diestrous stage were deeply anesthetized and decapitated. Brains were quicklyremoved into ice-cold cutting solution consisting of (in mM) 72 sucrose, 83 NaCl, 2.5 KCl, 1NaH2PO4, 26 NaHCO3, 22 glucose, 5 MgCl2, 1 CaCl2, oxygenated with 95% O2/5% CO2, mea-sured osmolarity 310–320 mOsm/l. 300-m-thick coronal sections were cut with a Leica VT1000Svibratome and incubated in oxygenated cutting solution at 34˚C for 45 min. Slices were transferredto oxygenated aCSF (126 mM NaCl, 21.4 mM NaHCO3, 2.5 mM KCl, 1.2 mM NaH2PO4, 1.2 mMMgCl2, 2.4 mM CaCl2, 10 mM glucose) and stored in the same solution at room temperature (20–24˚C) for at least 60 min before recording. A single slice was placed in the recording chamber whereit was continuously superfused at a rate of 3–4 ml per min with oxygenated aCSF. Neurons werevisualized with an upright microscope (SliceScope, Scientifica) equipped with infrared differentialinterference contrast and fluorescence optics. Borosilicate glass microelectrodes (5–7 M) were filledwith internal solution.Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 12 of 18Research article NeuroscienceFor CRACM experiments, recordings were obtained using a Cs+-based low-Cl internal solutionconsisting of (in mM) 135 cesium methanesulfonate, 10 HEPES, 1 EGTA, 4 MgCl2, 4 Na2-ATP, 0.4Na2-GTP, 10 disodium phosphocreatine (pH 7.3 adjusted with CsOH; 295 mOsm kg1; ECl = 70mV). Initially, light- evoked EPSCs or IPSCSs were recorded in whole-cell voltage-clamp mode, withmembrane potential clamped at Vh = 70 mV or 0 mV, respectively. Once the excitatory nature ofthe recorded inputs was established, picrotoxin was included to isolate glutamatergic currents. Thelight-evoked EPSC protocol consisted of four blue light pulses (473 nm wavelength, 5 ms) adminis-tered 1 s apart during the first 4 s of an 8 s sweep, repeated for a total of 30 sweeps. Evoked EPSCswith short latency (<6 ms) upon light stimulation were considered to be light-driven. As discussed byothers, such currents are most likely monosynaptic (Petreanu et al 2007). Number of animals usedfor CRACM experiments: PMVPACAPfi ARCkisspeptin, n = 5; PMVPACAPfi AVPV/PeNkisspeptin, n = 5.The PMVPACAP neurons were labeled with AAV8-DIO-ChR2-mCherry injected unilaterally into thePMV of PACAP-i-cre;kiss1hr-GFP mice. To visualize kisspeptin neurons in the AVPV/PeN, estrogenwas injected subcutaneously 2 days prior to experiments where AVPV/PeN neurons were targetedfor recording (this was not required for visualizing the ARC kiss1 population). For consistency, werecorded from both regions in the setting of estrogen treatment, and there was no difference in theconnections between neurons when comparing connections in animals that received estrogen andthose that did not.Calcium imagingMice were generated by crossing Kiss1-Cre+/- (Kiss1tm2(Cre-GFP)Coll), (Yeo et al., 2016) generated inthe Herbison lab (University of Otago, NZ), and homozygous floxed GCaMP6f (Ai95(RCL-GCaMP6f)-D, also known as 129S-Gt(ROSA)26Sortm95.1(CAG-GCaMP6f)Hze) (Madisen et al., 2015) lines to generatemixed background 129S6Sv/Ev C57Bl/6 Kiss1-Cre;GCaMP6f-lox-STOP-lox (Kiss1-GCaMP) mice.Mice (n = 4 per region of interest) were housed under a 12:12 hr lighting schedule (lights on at 6:00A.M.) with ad libitum access to food and water and females investigated at the diestrous phase ofthe cycle. These experiments were approved by the University of Otago Animal Welfare and EthicsCommittee.Calcium imaging of kisspeptin neurons was undertaken in acute brain slices using previously pub-lished methodology (Piet et al., 2015). In brief, 300 mm-thick brain slices containing the AVPV andPeN rostral, middle and caudal regions of the ARN were prepared and constantly perfused (1 mL/min) with 30˚C, 95%O2/5%CO2 equilibrated, artificial cerebrospinal fluid (aCSF) comprised of (mM)NaCl 120, KCl 3, NaHCO3 26, NaH2PO4 1, CaCl2 2.5, MgCl2 1.2 and glucose 10, and containing 0.5mM TTX, 100 mM picrotoxin, 10 mM CNQX, and 40 mM AP5 to block all synaptic transmission. Sliceswere placed under an upright Olympus BX51W1 microscope and multiple individual cells in a planeof focus visualized through a 40x immersion objective using a xenon arc light source (300 W; filteredby a GFP filter cube (excitation 470–490 nm, Chroma)) and a DG-4 shutter (Sutter Instruments). Epi-fluorescence (495 nm long pass and emission 500–520 nm) was collected at 2 Hz using a HamamatsuORCA-ER digital CCD camera. The effects of PACAP on kisspeptin neuron fluorescence wereassessed by performing basal image acquisition (4 min) and adding 10 nM PACAP to the aCSF for atwo-min period before switching back to aCSF only for a further 4 min wash period. Regions of inter-est over individual, non-overlapping, and in-focus fluorescent somata were selected and analyzedusing ImageJ software and custom R scripts (Clarkson et al., 2017). Individual cells were consideredto have responded if they exhibited an increase in fluorescence that was greater than their meanbasal level +2 standard deviations (4 min prior to drug). For AVPV/PeNkisspeptin neurons, a 25%change in the frequency of individual calcium events between the basal and test +wash periods wasconsidered a response. For visualizing data, values are presented as relative fluorescence changesusing DFF¼FtFF100 where F is the baseline fluorescence intensity calculated as the mean fluores-cence intensity over the 4 min period preceding drug applications and Ft is the fluorescencemeasured.Fluorescence in situ hybridizationBrains were sectioned coronally at 12 mm using a cryostat, thaw- mounted onto electrostaticallyclean slides, and stored at 80˚C until postfixed. Prior to hybridization, sections were postfixed in 4%paraformaldehyde, rinsed in 0.1 M PBS (pH 7.4), equilibrated in 0.1 M triethanolamine (pH 8.0), andRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 13 of 18Research article Neuroscienceacetylated in triethanolamine containing 0.25% acetic anhydride. Standard in vitro transcriptionmethods were used to generate both sense and antisense riboprobes recognizing PACAP transcript,which were subsequently diluted in hybridization cocktail (Amresco) with tRNA. Sections werehybridized overnight at 60˚C with either digoxigenin (DIG)-labeled riboprobes. After hybridization,slides were treated with RNase A and stringently washed in 0.5x SSC at 56˚C for 30 min. Slides werethen incubated with an antibody against DIG conjugated to horseradish peroxidase (HRP; Roche)overnight at 4˚C. Riboprobe signal was further enhanced using the TSA-Plus fluorophore systemwith Alexa Fluor conjugated streptavidin (LifeTechnologies). Image capture was performed usingfluorescent microscopy (Olympus VS120 slide scanner microscope) and analyzed with image J soft-ware. The primer sequences for the riboprobe generation are 5’-ccaatgaccatgtgtagcg and 5’- atca-gaccagaagacgaggc. For dual fluorescence ISH, we used probes for Adcyap1 and Vglut2 obtainedfrom ACDBio and used the RNAscope method per their protocol (ACDBio).Brain tissue preparation for immunohistochemistryAnimals were terminally anesthetized with 7% chloral hydrate diluted in saline (350 mg/kg) and trans-cardially perfused with phosphate-buffered saline (PBS) followed by 10% neutral buffered formalin(PFA). Brains were removed, stored in the same fixative for 2 hr, transferred into 20% sucrose at 4˚Covernight, and cut into 30 mm sections on a freezing microtome (Leica) coronally into four equalseries. A single series of sections per animal was used in the histological studies.ImmunohistochemistryBrain sections were washed in PBS with Tween-20, pH 7.4 (PBST) and blocked in 3% normal donkeyserum/PBST for 1 hr at room temperature. Then, brain sections were incubated overnight at roomtemperature in blocking solution containing primary antiserum (rat anti-mCherry, Life Technologies,1:1,000; chicken anti- GFP, Life Technologies, 1:1000 (Millipore, #AB9754). The next morning sec-tions were extensively washed in PBS and then incubated in Alexa- fluorophore secondary antibody(Molecular Probes, 1:1,000) for 1 hr at room temperature. After several washes in PBS, sections weremounted on gelatin-coated slides and fluorescence images were captured with Olympus VS120 slidescanner microscope.For pStat3 immunohistochemistry studies, Mice were injected with 5 mg/kg recombinant leptin 2hr before perfusion (as above). Brain sections were washed in 0.1 M phosphate-buffered saline, pH7.4, followed by incubation in 5% NaOH and 0.3% H2O2 for 2 min, then with 0.3% glycine (10 min),and finally with 0.03% SDS (10 min), all made up in PBS. Sections were blocked in 3% normal donkeyserum/0.25% Triton X-100 in PBS for 1 hr at room temperature and then incubated overnight atroom temperature in blocking solution containing 1/250 rabbit anti-pStat3 (Cell Signaling, #9145)and 1/1,000 chicken anti-GFP (Life Technologies, #A10262). The next morning sections were exten-sively washed in PBS and then incubated in 1/250 donkey anti-rabbit 594 (Molecular Probes,R37119) and 1/1,000 donkey anti-chicken 488 (Jackson ImmunoResearch, 703-545-155) for 2 hr atroom temperature. After several washes in PBS, sections were mounted onto gelatin-coated slidesand fluorescent images were captured with Olympus VS120 slide scanner microscope.Brain punchesBrains were rapidly extracted, cooled in ice-cold DMEM/F12 medium for 5 min and then placed,ventral surface up, into a chilled stainless steel brain matrix (catalog no. SA-2165, Roboz SurgicalInstrument Co., Gaithersburg, MD). Using anatomical markers, brains were blocked to obtain a sin-gle coronal section containing the entire PMV,~2 mm thick. The PMV was microdissected by knifecuts at its visually approximated dorsolateral borders under 5x magnification.RNA Expression (qRT-PCR)Total RNA from PMV was extracted from frozen tissue collected using TRIzol reagent (Invitrogen) fol-lowed by chloroform/isopropanol extraction. RNA was quantified using a NanoDrop 2000 spectro-photometer (Thermo Scientific) and 1 mg of RNA was reverse transcribed using Superscript III cDNAsynthesis kit (Invitrogen). Quantitative real-time PCR assays were performed on an ABI Prism 7000sequence detection system, and analyzed using ABI Prism 7000 SDS software (Applied Biosystems).The cycling conditions were as follows: 2 min incubation at 95˚C (hot start), 45 amplification cyclesRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 14 of 18Research article Neuroscience(95˚C for 30 s, 60˚C for 30 s, and 45 s at 75˚C, with fluorescence detection at the end of each cycle),followed by melting curve of the amplified products obtained by ramped increase of the tempera-ture from 55˚C to 95˚C to confirm the presence of single amplification product per reaction.Adcyap1 expression was detected using primers PACAP F- GAA ACC CGC TGC AAG ACTT/PACAPR – CGA CAT CTC TCC TGT CCGC and was normalized with housekeeping gene Rpl19.Statistical AnalysisAll data are expressed as the mean ± SEM for each group. A two tailed unpaired t-Student test or aone- or two-way ANOVA test followed by Bonferroni or Fisher’s post-hoc test was used to assessvariation among experimental groups. Significance level was set at p<0.05. All analyses were per-formed with GraphPad Prism Software, Inc (San Diego, CA).AcknowledgementsThis work was supported by NIH R01 HD019938 and R01 HD082314 to UBK; R00 HD071970 andR01 HD090151 to VMN; NHLBI 5T32HL007374-36, NCATS UL1 TR001102, T32 HL007374-36 andHarvard Medical School Dupont Warren Fellowship to RAR; R01 DK075632, R01 DK096010, R01DK089044, R01 DK111401, P30 DK046200, P30 DK057521 to BBL.Additional informationFundingFunder Grant reference number AuthorNational Institutes of Health UL1 TR001102 Rachel A RossNational Institutes of Health 5T32HL007374-36 Rachel A RossNational Institutes of Health R01 HD082314 Ursula B KaiserNational Institutes of Health R01 HD019938 Ursula B KaiserNational Institutes of Health R01 DK075632 Bradford B LowellNational Institutes of Health R01 DK089044 Bradford B LowellNational Institutes of Health R01 DK111401 Bradford B LowellNational Institutes of Health P30 DK046200 Bradford B LowellNational Institutes of Health P30 DK057521 Bradford B LowellNational Institutes of Health R01 DK096010 Bradford B LowellNational Institutes of Health R01 HD090151-A1 Victor M NavarroNational Institutes of Health R00 HD071970 Victor M NavarroThe funders had no role in study design, data collection and interpretation, or thedecision to submit the work for publication.Author contributionsRachel A Ross, Conceptualization, Resources, Data curation, Formal analysis, Validation, Investiga-tion, Visualization, Methodology, Writing—original draft, Project administration, Writing—reviewand editing; Silvia Leon, Resources, Data curation, Formal analysis, Investigation, Methodology;Joseph C Madara, Danielle Schafer, Anne MJ Verstegen, Resources, Methodology; Chrysanthi Fer-gani, Resources, Formal analysis, Investigation, Methodology; Caroline A Maguire, Investigation,Methodology; Emily Brengle, Data curation, Methodology; Dong Kong, Resources; Allan E Herbison,Bradford B Lowell, Conceptualization, Resources, Methodology; Ursula B Kaiser, Conceptualization,Supervision, Visualization; Victor M Navarro, Conceptualization, Resources, Data curation, Software,Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodol-ogy, Writing—original draft, Project administration, Writing—review and editingRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 15 of 18Research article NeuroscienceAuthor ORCIDsAllan E Herbison http://orcid.org/0000-0002-9615-3022Victor M Navarro http://orcid.org/0000-0002-5799-219XEthicsAnimal experimentation: All animal care and experimental procedures were approved by theNational Institute of Health, Beth Israel Deaconess Medical Center and Brigham and Women’s Hos-pital Institutional Animal Care and Use Committee. protocol #05165.Decision letter and Author responseDecision letter https://doi.org/10.7554/eLife.35960.013Author response https://doi.org/10.7554/eLife.35960.014Additional filesSupplementary files. Transparent reporting formDOI: https://doi.org/10.7554/eLife.35960.011Data availabilityAll data generated or analysed during this study are included in the manuscript and supporting files.ReferencesAhima RS, Prabakaran D, Mantzoros C, Qu D, Lowell B, Maratos-Flier E, Flier JS. 1996. Role of leptin in theneuroendocrine response to fasting. Nature 382:250–252. DOI: https://doi.org/10.1038/382250a0, PMID:8717038Caligioni CS. 2009. Assessing reproductive status/stages in mice. Current Protocols in Neuroscience Appendix 4:Appendix 4I. DOI: https://doi.org/10.1002/0471142301.nsa04is48, PMID: 19575469Campbell JN, Macosko EZ, Fenselau H, Pers TH, Lyubetskaya A, Tenen D, Goldman M, Verstegen AM, ReschJM, McCarroll SA, Rosen ED, Lowell BB, Tsai LT. 2017. A molecular census of arcuate hypothalamus andmedian eminence cell types. Nature Neuroscience 20:484–496. DOI: https://doi.org/10.1038/nn.4495, PMID: 28166221Cavalcante JC, Bittencourt JC, Elias CF. 2014. Distribution of the neuronal inputs to the ventral premammillarynucleus of male and female rats. Brain Research 1582:77–90. DOI: https://doi.org/10.1016/j.brainres.2014.07.034, PMID: 25084037Clarkson J, Han SY, Piet R, McLennan T, Kane GM, Ng J, Porteous RW, Kim JS, Colledge WH, Iremonger KJ,Herbison AE. 2017. Definition of the hypothalamic GnRH pulse generator in mice. PNAS 114:E10216–E10223.DOI: https://doi.org/10.1073/pnas.1713897114Cravo RM, Frazao R, Perello M, Osborne-Lawrence S, Williams KW, Zigman JM, Vianna C, Elias CF. 2013. Leptinsignaling in Kiss1 neurons arises after pubertal development. PLoS One 8:e58698. DOI: https://doi.org/10.1371/journal.pone.0058698, PMID: 23505551Donato J, Cravo RM, Fraza˜o R, Gautron L, Scott MM, Lachey J, Castro IA, Margatho LO, Lee S, Lee C,Richardson JA, Friedman J, Chua S, Coppari R, Zigman JM, Elmquist JK, Elias CF. 2011. Leptin’s effect onpuberty in mice is relayed by the ventral premammillary nucleus and does not require signaling in Kiss1neurons. Journal of Clinical Investigation 121:355–368. DOI: https://doi.org/10.1172/JCI45106, PMID: 21183787Donato J, Elias CF. 2011. The ventral premammillary nucleus links metabolic cues and reproduction. Frontiers inEndocrinology 2:57. DOI: https://doi.org/10.3389/fendo.2011.00057, PMID: 22649378Donato J, Silva RJ, Sita LV, Lee S, Lee C, Lacchini S, Bittencourt JC, Franci CR, Canteras NS, Elias CF. 2009. Theventral premammillary nucleus links fasting-induced changes in leptin levels and coordinated luteinizinghormone secretion. Journal of Neuroscience 29:5240–5250. DOI: https://doi.org/10.1523/JNEUROSCI.0405-09.2009, PMID: 19386920Dror T, Franks J, Kauffman AS. 2013. Analysis of multiple positive feedback paradigms demonstrates a completeabsence of LH surges and GnRH activation in mice lacking kisspeptin signaling. Biology of Reproduction 88:146. DOI: https://doi.org/10.1095/biolreprod.113.108555, PMID: 23595904Duggal PS, Weitsman SR, Magoffin DA, Norman RJ. 2002. Expression of the long (OB-RB) and short (OB-RA)forms of the leptin receptor throughout the oestrous cycle in the mature rat ovary. Reproduction 123:899–905.DOI: https://doi.org/10.1530/rep.0.1230899, PMID: 12052244Garcı´a-Galiano D, van Ingen Schenau D, Leon S, Krajnc-Franken MA, Manfredi-Lozano M, Romero-Ruiz A,Navarro VM, Gaytan F, van Noort PI, Pinilla L, Blomenro¨hr M, Tena-Sempere M. 2012. Kisspeptin signaling isRoss et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 16 of 18Research article Neuroscienceindispensable for neurokinin B, but not glutamate, stimulation of gonadotropin secretion in mice.Endocrinology 153:316–328. DOI: https://doi.org/10.1210/en.2011-1260, PMID: 22067321Halvorson LM. 2014. PACAP modulates GnRH signaling in gonadotropes. Molecular and Cellular Endocrinology385:45–55. DOI: https://doi.org/10.1016/j.mce.2013.09.029, PMID: 24095645Hawke Z, Ivanov TR, Bechtold DA, Dhillon H, Lowell BB, Luckman SM. 2009. PACAP neurons in the hypothalamicventromedial nucleus are targets of central leptin signaling. Journal of Neuroscience 29:14828–14835.DOI: https://doi.org/10.1523/JNEUROSCI.1526-09.2009, PMID: 19940178Herbison AE. 2016. Control of puberty onset and fertility by gonadotropin-releasing hormone neurons. NatureReviews Endocrinology 12:452–466. DOI: https://doi.org/10.1038/nrendo.2016.70, PMID: 27199290Jamen F, Rodriguez-Henche N, Pralong F, Jegou B, Gaillard R, Bockaert J, Brabet P. 2000. PAC1 null femalesdisplay decreased fertility. Annals of the New York Academy of Sciences 921:400–404. DOI: https://doi.org/10.1111/j.1749-6632.2000.tb07004.x, PMID: 11193864Ko¨ves K, Ka´ntor O, Lakatos A, Szabo´ E, Kirilly E, Heinzlmann A, Szabo´ F. 2014. Advent and recent advances inresearch on the role of pituitary adenylate cyclase-activating polypeptide (PACAP) in the regulation ofgonadotropic hormone secretion of female rats. Journal of Molecular Neuroscience 54:494–511. DOI: https://doi.org/10.1007/s12031-014-0294-7, PMID: 24696167Ko¨ves K, Ka´ntor O, Molna´r J, Heinzlmann A, Szabo´ E, Szabo´ F, Nemeske´ri A, Horva´th J, Arimura A. 2003. Therole of PACAP in gonadotropic hormone secretion at hypothalamic and pituitary levels. Journal of MolecularNeuroscience 20:141–152. DOI: https://doi.org/10.1385/JMN:20:2:141, PMID: 12794308Ko¨ves K, Molna´r J, Ka´ntor O, Lakatos A, Fo´gel K, Kausz M, Vandermeers-Piret MC, Somogyva´ri-Vigh A, ArimuraA. 1998. Role of PACAP in the regulation of gonadotroph hormone secretion during ontogenesis: a singleneonatal injection of PACAP delays puberty and its intracerebroventricular administration before the criticalperiod of proestrous stage blocks ovulation in adulthood. Annals of the New York Academy of Sciences 865:590–594. DOI: https://doi.org/10.1111/j.1749-6632.1998.tb11242.x, PMID: 9928075Krashes MJ, Shah BP, Madara JC, Olson DP, Strochlic DE, Garfield AS, Vong L, Pei H, Watabe-Uchida M, UchidaN, Liberles SD, Lowell BB. 2014. An excitatory paraventricular nucleus to AgRP neuron circuit that driveshunger. Nature 507:238–242. DOI: https://doi.org/10.1038/nature12956, PMID: 24487620Laughlin GA, Dominguez CE, Yen SS. 1998. Nutritional and endocrine-metabolic aberrations in women withfunctional hypothalamic amenorrhea. Journal of Clinical Endocrinology & Metabolism 83:25–32. DOI: https://doi.org/10.1210/jc.83.1.25, PMID: 9435412Leshan RL, Bjo¨rnholm M, Mu¨nzberg H, Myers MG. 2006. Leptin receptor signaling and action in the centralnervous system. Obesity 14, :208S–212. DOI: https://doi.org/10.1038/oby.2006.310, PMID: 17021368Madisen L, Garner AR, Shimaoka D, Chuong AS, Klapoetke NC, Li L, van der Bourg A, Niino Y, Egolf L, MonettiC, Gu H, Mills M, Cheng A, Tasic B, Nguyen TN, Sunkin SM, Benucci A, Nagy A, Miyawaki A, Helmchen F, et al.2015. Transgenic mice for intersectional targeting of neural sensors and effectors with high specificity andperformance. Neuron 85:942–958. DOI: https://doi.org/10.1016/j.neuron.2015.02.022, PMID: 25741722Martin C, Navarro VM, Simavli S, Vong L, Carroll RS, Lowell BB, Kaiser UB. 2014. Leptin-responsive GABAergicneurons regulate fertility through pathways that result in reduced kisspeptinergic tone. Journal of Neuroscience34:6047–6056. DOI: https://doi.org/10.1523/JNEUROSCI.3003-13.2014, PMID: 24760864Mertens I, Husson SJ, Janssen T, Lindemans M, Schoofs L. 2007. PACAP and PDF signaling in the regulation ofmammalian and insect circadian rhythms. Peptides 28:1775–1783. DOI: https://doi.org/10.1016/j.peptides.2007.05.005, PMID: 17586087Mounien L, Bizet P, Boutelet I, Gourcerol G, Fournier A, Vaudry H, Je´gou S. 2006. Pituitary adenylate cyclase-activating polypeptide directly modulates the activity of proopiomelanocortin neurons in the rat arcuatenucleus. Neuroscience 143:155–163. DOI: https://doi.org/10.1016/j.neuroscience.2006.07.022, PMID: 16962718Mustafa T. 2013. Pituitary adenylate cyclase-activating polypeptide (PACAP): a master regulator in central andperipheral stress responses. Advances in pharmacology 68:445–457. DOI: https://doi.org/10.1016/B978-0-12-411512-5.00021-X, PMID: 24054157Navarro VM, Bosch MA, Leo´n S, Simavli S, True C, Pinilla L, Carroll RS, Seminara SB, Tena-Sempere M,Rønnekleiv OK, Kaiser UB. 2015. The integrated hypothalamic tachykinin-kisspeptin system as a centralcoordinator for reproduction. Endocrinology 156:627–637. DOI: https://doi.org/10.1210/en.2014-1651,PMID: 25422875Navarro VM, Kaiser UB. 2013. Metabolic influences on neuroendocrine regulation of reproduction. CurrentOpinion in Endocrinology & Diabetes and Obesity 20:335–341. DOI: https://doi.org/10.1097/MED.0b013e32836318ce, PMID: 23807606Padilla SL, Qiu J, Nestor CC, Zhang C, Smith AW, Whiddon BB, Rønnekleiv OK, Kelly MJ, Palmiter RD. 2017.AgRP to Kiss1 neuron signaling links nutritional state and fertility. PNAS 114:2413–2418. DOI: https://doi.org/10.1073/pnas.1621065114, PMID: 28196880Piet R, Fraissenon A, Boehm U, Herbison AE. 2015. Estrogen permits vasopressin signaling in preoptic kisspeptinneurons in the female mouse. Journal of Neuroscience 35:6881–6892. DOI: https://doi.org/10.1523/JNEUROSCI.4587-14.2015, PMID: 25926463Pinilla L, Aguilar E, Dieguez C, Millar RP, Tena-Sempere M. 2012. Kisspeptins and reproduction: physiologicalroles and regulatory mechanisms. Physiological Reviews 92:1235–1316. DOI: https://doi.org/10.1152/physrev.00037.2010, PMID: 22811428Ramikie TS, Ressler KJ. 2016. Stress-related disorders, pituitary adenylate cyclase-activating peptide (PACAP)ergic system, and sex differences. Dialogues in Clinical Neuroscience 18:403–413. PMID: 28179812Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 17 of 18Research article NeuroscienceReglodi D, Tamas A, Koppan M, Szogyi D, Welke L. 2012. Role of PACAP in female fertility and reproduction atgonadal level - recent advances. Frontiers in Endocrinology 3:155. DOI: https://doi.org/10.3389/fendo.2012.00155, PMID: 23248616Resch JM, Maunze B, Gerhardt AK, Magnuson SK, Phillips KA, Choi S. 2013. Intrahypothalamic pituitaryadenylate cyclase-activating polypeptide regulates energy balance via site-specific actions on feeding andmetabolism. American Journal of Physiology-Endocrinology and Metabolism 305:E1452–E1463. DOI: https://doi.org/10.1152/ajpendo.00293.2013, PMID: 24148346Ryan NK, Van der Hoek KH, Robertson SA, Norman RJ. 2003. Leptin and leptin receptor expression in the ratovary. Endocrinology 144:5006–5013. DOI: https://doi.org/10.1210/en.2003-0584, PMID: 12959975Shintani N, Tomimoto S, Hashimoto H, Kawaguchi C, Baba A. 2003. Functional roles of the neuropeptide PACAPin brain and pancreas. Life Sciences 74:337–343. DOI: https://doi.org/10.1016/j.lfs.2003.09.021,PMID: 14607262Smith JT, Acohido BV, Clifton DK, Steiner RA. 2006. KiSS-1 neurones are direct targets for leptin in the ob/obmouse. Journal of Neuroendocrinology 18:298–303. DOI: https://doi.org/10.1111/j.1365-2826.2006.01417.x,PMID: 16503925Smith JT, Cunningham MJ, Rissman EF, Clifton DK, Steiner RA. 2005. Regulation of Kiss1 gene expression in thebrain of the female mouse. Endocrinology 146:3686–3692. DOI: https://doi.org/10.1210/en.2005-0488,PMID: 15919741Steyn FJ, Wan Y, Clarkson J, Veldhuis JD, Herbison AE, Chen C. 2013. Development of a methodology for andassessment of pulsatile luteinizing hormone secretion in juvenile and adult male mice. Endocrinology 154:4939–4945. DOI: https://doi.org/10.1210/en.2013-1502, PMID: 24092638Szabo´ F, Horva´th J, Heinzlmann A, Arimura A, Ko¨ves K. 2002. Neonatal PACAP administration in rats delayspuberty through the influence of the LHRH neuronal system. Regulatory Peptides 109:49–55. DOI: https://doi.org/10.1016/S0167-0115(02)00185-4, PMID: 12409214Tanida M, Hayata A, Shintani N, Yamamoto N, Kurata Y, Shibamoto T, Morgan DA, Rahmouni K, Hashimoto H.2013. Central PACAP mediates the sympathetic effects of leptin in a tissue-specific manner. Neuroscience 238:297–304. DOI: https://doi.org/10.1016/j.neuroscience.2013.02.016, PMID: 23454538Vaudry D, Gonzalez BJ, Basille M, Yon L, Fournier A, Vaudry H. 2000. Pituitary adenylate cyclase-activatingpolypeptide and its receptors: from structure to functions. Pharmacological Reviews 52:269–324. PMID: 10835102Vong L, Ye C, Yang Z, Choi B, Chua S, Lowell BB. 2011. Leptin action on GABAergic neurons prevents obesityand reduces inhibitory tone to POMC neurons. Neuron 71:142–154. DOI: https://doi.org/10.1016/j.neuron.2011.05.028, PMID: 21745644Yeo SH, Kyle V, Morris PG, Jackman S, Sinnett-Smith LC, Schacker M, Chen C, Colledge WH. 2016. Visualisationof Kiss1 neurone distribution using a Kiss1-CRE transgenic mouse. Journal of Neuroendocrinology 28.DOI: https://doi.org/10.1111/jne.12435, PMID: 27663274Ross et al. eLife 2018;7:e35960. DOI: https://doi.org/10.7554/eLife.35960 18 of 18Research article Neuroscience

Reviewed by:

Abstract

Extracted data

Reviewed by:

Abstract

Extracted data

Related items

Related items