Conditional Oprk1-dependent Kiss1 deletion in kisspeptin neurons caused estrogen-dependent LH pulse disruption and LH surge attenuation in female rats

The gonadotropin-releasing hormone (GnRH) pulse and surge are considered to be generated by arcuate kisspeptin/neurokinin B/dynorphin A (KNDy) neurons and anteroventral periventricular nucleus (AVPV) kisspeptin neurons, respectively, in female rodents. The majority of KNDy and AVPV kisspeptin neurons express κ-opioid receptors (KORs, encoded by Oprk1) in female rodents. Thus, this study aimed to investigate the effect of a conditional Oprk1-dependent Kiss1 deletion in kisspeptin neurons on the luteinizing hormone (LH) pulse/surge and fertility using Kiss1-floxed/Oprk1-Cre rats, in which Kiss1 was deleted in cells expressing or once expressed the Oprk1/Cre. The Kiss1-floxed/Oprk1-Cre female rats, with Kiss1 deleted in a majority of KNDy neurons, showed normal puberty while having a one-day longer estrous cycle and fewer pups than Kiss1-floxed controls. Notably, ovariectomized (OVX) Kiss1-floxed/Oprk1-Cre rats showed profound disruption of LH pulses in the presence of a diestrous level of estrogen but showed apparent LH pulses without estrogen treatment. Furthermore, Kiss1-floxed/Oprk1-Cre rats, with Kiss1 deleted in approximately half of AVPV kisspeptin neurons, showed a lower peak of the estrogen-induced LH surge than controls. These results suggest that arcuate and AVPV kisspeptin neurons expressing or having expressed Oprk1 have a role in maintaining normal GnRH pulse and surge generation, the normal length of the estrous cycle, and the normal offspring number in female rats.

www.nature.com/scientificreports/(AVPV) serve as a target of estrogen positive feedback action on GnRH/LH surge generation in rodents because estrogen increases AVPV Kiss1 expression or c-Fos (a marker of neuronal activation) in AVPV kisspeptin neurons in female rats [10][11][12] and mice [13][14][15] .The notion that AVPV kisspeptin neurons serve as the GnRH surge generator has also been verified by sex differences in the numbers of AVPV kisspeptin neurons (males < females) and the absence of an estrogen-induced LH surge in male rodent models 4,11,[16][17][18][19] .Importantly, Wiegand et al. 20 demonstrated that AVPV lesion disrupted estrogen + progesterone-induced LH surge in female rats.Furthermore, our and other previous studies showed that kisspeptin antagonism by chronic intracerebroventricular infusion of a GPR54 antagonist or by chronic infusion of an anti-kisspeptin antibody into the preoptic area blocked spontaneous or estrogen-induced LH surge in female rats 10,21,22 .
It has been suggested that KNDy neurons synchronize with each other in a paracrine and autocrine manner through stimulatory NKB-neurokinin 3 receptor (NK3R, the primary receptor for NKB) signaling and inhibitory Dyn-κ-opioid receptor (KOR, the primary receptor for Dyn) signaling to form GnRH/gonadotropin pulses because the majority of KNDy neurons express NK3Rs and KORs in rats and sheep [23][24][25] .Indeed, central administration of NKB or a KOR antagonist increased the frequency of multiple unit activity (MUA) volleys accompanied by LH pulses, and central Dyn administration decreased the frequency of MUA volleys in female goats 26 , in which MUA volleys were taken from the cluster of KNDy neurons.In addition, Ruka et al. 27 and de Croft et al. 28 showed that an NK3R agonist increased and Dyn or a KOR agonist decreased the firing frequency of GFP-labeled KNDy neurons in male mice.Moreover, the LH pulse frequency was increased by peripheral administration of the KOR antagonist and was suppressed by peripheral administration of the NK3R antagonist in female goats 29 and ewes 30 .Importantly, KOR expression was found in some KNDy neurons in rats (62%) 25 , sheep (98%) 24 , and mice (6-65%) 27,[31][32][33] , but it has not been clarified whether KOR-expressing KNDy neurons play an indispensable role in fertility and GnRH pulse generation.Therefore, it is worth clarifying the effects of a conditional Oprk1-dependent Kiss1 deletion in KNDy cells on fertility and LH pulses in newly generated gene-modified female rats.
The sex steroidal milieu should be taken into account when studying the role of Dyn-KOR signaling in GnRH pulse generation by KNDy neurons because Dyn-KOR signaling mediates the negative feedback action of sex steroids and stress-induced suppression of GnRH/gonadotropin pulses in several mammals, including rodents, ruminants, primates, and humans 34 .Indeed, Han et al. 35 showed that central KOR antagonism promoted synchronization between KNDy neurons in testis-intact Kiss1-Cre male mice, but not in castrated males.Furthermore, our recent study showed that central administration of a KOR antagonist restored glucoprivic suppression of LH pulses in ovariectomized rats treated with a diestrous level of estradiol-17β (OVX + low E2 rats) 25 .Interestingly, low E2 treatment significantly increased the number of Pdyn (encoding Dyn)-expressing cells in the paraventricular nucleus (PVN) but not in the ARC of OVX rats 25 , suggesting that PVN Dyn neurons mainly mediate the inhibitory effect of sex steroids and/or malnutrition on KNDy neural activity and GnRH/ LH pulses.Therefore, the contribution of Dyn-KOR signaling in controlling pulsatile GnRH/LH release could be changed depending on the steroidal milieu.
It is also known that KORs are expressed in some (50%) AVPV kisspeptin neurons in mice 32 , while the role of Dyn-KOR signaling in AVPV kisspeptin neurons has not yet been clarified.If KOR-expressing AVPV kisspeptin neurons are needed for GnRH/LH surge generation, a conditional Oprk1-dependent Kiss1 deletion in AVPV kisspeptin cells may affect the LH surge and ovulation.
Thus, the present study aimed to investigate the effect of a conditional Oprk1-dependent Kiss1 deletion in kisspeptin cells on LH pulses, the LH surge, and fertility using newly generated Kiss1-floxed/Oprk1-Cre rats (Fig. 1a) to clarify whether KOR-negative kisspeptin neurons could maintain fertility and whether KOR-positive kisspeptin neurons are a prerequisite for the pulsatile and surge modes of GnRH/LH release.Puberty onset, ovarian weight, vaginal cyclicity, and fertility were examined in ovary-intact Kiss1-floxed/Oprk1-Cre rats and Kiss1floxed control rats.Furthermore, ARC Kiss1 mRNA expression; pulsatile LH release; and pituitary Gnrhr, Lhb, and Fshb mRNA (encoding GnRH receptor, and LH and FSH β-subunit, respectively) expression were examined in Kiss1-floxed/Oprk1-Cre rats and Kiss1-floxed control rats under different steroidal milieus (OVX vs. physiological estrogenic condition, i.e., OVX + low E2 condition).In addition, we examined effects of the conditional Kiss1 deletion on the number of AVPV Kiss1-expressing cells and on the LH surge induced by a proestrous level of E2 (high E2) by comparing Kiss1-floxed/Oprk1-Cre rats and Kiss1-floxed control rats.Moreover, for comparison with Kiss1-floxed/Oprk1-Cre rats, we generated rats with conditionally deleted Kiss1 in all kisspeptin neurons (Kiss1-floxed/Kiss1-Cre rats) by crossing Kiss1-Cre rats 36 with Kiss1-floxed rats (Fig. 1b), and reproductive indicators, such as puberty onset, ovarian weight, ARC Kiss1 mRNA expression, pulsatile LH release, and pituitary Gnrhr, Lhb, and Fshb mRNA expression, were also examined in the Kiss1-floxed/Kiss1-Cre rats.

Colocalization of Oprk1 and Oprk1-Cre-activated tdTomato mRNA expression
We first generated Oprk1-Cre rats (see Supplementary Fig. S1 online) and confirmed the specificity of Cre expression in Oprk1-expressing cells by crossing Oprk1-Cre rats and Cre-dependent tdTomato reporter rats 37 (see Supplementary Fig. S2a online).Double in situ hybridization (ISH) revealed the presence of a number of Oprk1-expressing cells and tdTomato-expressing cells in the hypothalamus, including the ARC (Fig. S2b), PVN (Fig. S2c), and supraoptic nucleus (SON, Fig. S2d), in OVX + low E2 Oprk1-Cre/tdTomato reporter female rats in adulthood.Quantitative analysis revealed that the majority of Oprk1-expressing cells coexpressed tdTomato in the ARC (92.96 ± 1.35%), PVN (98.06 ± 0.35%), and SON (97.95 ± 0.40%) in Oprk1-Cre/tdTomato reporter rats, indicating that most Oprk1-expressing cells in these regions were labeled with the tdTomato reporter (Fig. S2b-d).On the other hand, many tdTomato-expressing cells without Oprk1 expression were observed in these regions because theoretically cells that have expressed Oprk1/Cre at least once should exhibit persistent tdTomato Oprk1-driven Cre-activated tdTomato expression in KNDy neurons in female rats at prepubertal and adult stages Double ISH for tdTomato and Tac3 (encoding NKB, a marker of KNDy neurons) revealed Oprk1-Cre-activated tdTomato expression in a small population of ARC Tac3-expressing cells (19.48 ± 1.93%) in ovary-intact Oprk1-Cre/tdTomato reporter rats in the prepubertal period (3 weeks of age), whereas tdTomato expression was found in most Tac3-expressing cells (92.14 ± 1.12%) in OVX + low E2 Oprk1-Cre/tdTomato reporter rats in adulthood (9-10 weeks of age, Fig. S2e).The number of ARC Tac3-and tdTomato-coexpressing cells in Oprk1-Cre/tdTomato reporter rats in prepubertal period was significantly lower than that in adulthood (t (4) = 11.34,p < 0.01, Student's t-test), although the total number of Tac3-expressing cells was comparable between the prepubertal and adult groups (t (4) = 1.13, p = 0.32, Student's t-test).On the other hand, many tdTomato-expressing cells without Tac3 expression were observed in these regions, consistent with our previous study showing that Oprk1 expression was found in both KNDy and non-KNDy neurons in the ARC of female rats 25 .

Kiss1 deletion in cells expressing or having expressed Oprk1 reduced the number of ARC and AVPV Kiss1-expressing cells
Double ISH for Kiss1 and Oprk1 revealed that a few Kiss1-expressing cells without Oprk1 signals were located in the ARC of OVX Kiss1-floxed/Oprk1-Cre rats, regardless of E2 treatment, whereas many Kiss1-expressing cells with/without Oprk1 signals were found in the ARC of OVX Cre(−)/Kiss1-floxed control rats, regardless of E2 treatment (Fig. 3a).Two-way ANOVA revealed a significant main effect of the genotype, but not E2 treatment, without the interaction effect between the genotype and E2 treatment on the number of ARC Kiss1-expressing cells.Specifically, the number of ARC Kiss1-expressing cells was significantly lower in Kiss1-floxed/Oprk1-Cre rats (2.87 ± 0.65% of the control on average with/without low E2 treatment; 8.00 ± 0.71 and 27.20 ± 4.73 cells in Kiss1-floxed/Oprk1-Cre rats with or without low E2 treatment, respectively) than in Cre(−)/Kiss1-floxed control rats (F (1,14) = 246.03,p < 0.01, Fig. 3b).Quantitative analysis revealed the absence of Oprk1 expression in the remaining ARC Kiss1-expressing cells in OVX Kiss1-floxed/Oprk1-Cre rats with/without low E2 treatment, whereas Oprk1 coexpression was found in Cre(−)/Kiss1-floxed control rats (52.58 ± 4.40% of the control on average with/without low E2 treatment).
Figure 4c shows the relative Gnrhr, Lhb, and Fshb mRNA expression levels in the anterior pituitary gland in Kiss1-floxed/Oprk1-Cre rats and Cre(−)/Kiss1-floxed control rats.Two-way ANOVA revealed significant main effects of E2 treatment and the genotype, without the interaction effects between E2 treatment and the genotype on pituitary Gnrhr, Lhb, and Fshb mRNA expression (Fig. 4c).Specifically, Gnrhr, Lhb, and Fshb mRNA expression levels were significantly lower in OVX + low E2 rats than in OVX rats (E2 treatment as the main effect; Gnrhr, F (1,14)

Discussion
The present study demonstrated that Kiss1-floxed/Oprk1-Cre female rats, in which the Kiss1 gene was deleted in KNDy and AVPV kisspeptin neurons expressing or having expressed Oprk1, showed profound disruption of LH pulses in the presence of a diestrous level of E2 and a reduction in the AUC of the E2-induced LH surge.These results suggest that the Kiss1-deleted kisspeptin neurons in both the ARC and AVPV have a physiological role in maintaining normal reproductive function.Indeed, ovary-intact Kiss1-floxed/Oprk1-Cre female rats showed a 5-day estrous cycle (vs.a 4-day estrous cycle in Cre(−)/Kiss1-floxed control rats), a lower ovarian weight, and a smaller number of pups compared to Cre(−)/Kiss1-floxed controls.These findings suggest that Kiss1 expression in ARC and AVPV kisspeptin neurons, which are expressing or have expressed Oprk1, is likely needed to fully generate the GnRH/LH pulse and surge under physiological estrogen conditions.It is speculated that the estrogen-dependent suppression of LH pulses and LH surge attenuation may have caused a lower ovarian weight and a smaller number of pups in ovary-intact Kiss1-floxed/Oprk1-Cre female rats than in Cre(−)/Kiss1floxed control rats.This is the first report to show that kisspeptin neurons receiving direct Dyn-KOR signaling are needed to fully generate the GnRH/LH pulse and estrogen-induced GnRH/LH surge in OVX female rats under E2 replacement regimens that mimic diestrous and proestrous levels of estrogen, respectively.Notably, the current ovary-intact Kiss1-floxed/Oprk1-Cre female rats may have maintained effective levels of LH pulses and surges to maintain ovarian function, estrous cycles and fertility even though a longer estrous cycle, a lower ovarian weight, and a smaller number of pups than the wild-type rats.
Importantly, apparent LH pulses were observed in the current OVX Kiss1-floxed/Oprk1-Cre rats without estrogen replacement, suggesting that the remaining small portion of KOR-negative KNDy neurons are sufficient to maintain GnRH/gonadotropin pulse generation, at least in the estrogen-free condition.This notion is consistent with our previous study showing that the rescue of 3-50% of ARC KNDy neurons resulted in the restoration of GnRH/LH pulses in OVX global Kiss1 KO rats 6 .Indeed, pituitary Gnrhr, Lhb, and Fshb mRNA expression levels, which largely correlate with GnRH/gonadotropin pulses, in OVX Kiss1-floxed/Oprk1-Cre rats seemed to be comparable to OVX Kiss1-floxed control rats treated with low E2.Therefore, we speculate that the remaining KOR-negative KNDy neurons may synchronize with "Kiss1-deficient NDy" neurons in Kiss1-floxed/Oprk1-Cre female rats via mutual contact among KNDy and "Kiss1-deficient NDy" neurons through gap junctions and/or glial connections.Indeed, our previous study showed that gap junction inhibitors blocked synchronized activities among mouse KNDy neurons in vitro 38 , and Moore et al. 9 and Han et al. 35 showed using in vivo Ca 2+ imaging at a single-cell resolution that all recorded KNDy neurons exhibited synchronous activation prior to pulsatile LH release in female and male mice, respectively.Furthermore, Liu et al. 39 showed that KNDy neurons can generate synchronized episodic activity even in Kiss1 KO mice.A recent study by Kauffman and colleagues showed that Kiss KORKO mice, in which KOR signaling in Kiss1 neurons was disrupted, exhibited apparent LH pulses in males, normal attainment of puberty in both sexes, and fertility in females 32 .In addition, previous studies showed that both sexes of global Oprk1 KO mice [40][41][42] and global Pdyn KO mice 43,44 were fertile.In this context, other opioid receptor signaling pathways, such as enkephalin-δ-opioid receptor (DOR) and β-endorphin-μ-opioid receptor (MOR), may be able to compensate for the lack of KOR signaling in KNDy neurons to maintain GnRH pulse generation.In support of this notion, central DOR and MOR antagonism rescued frequent LH pulses in female rats under glucoprivation 45,46 .
The low E2 treatment disrupted the LH pulse in the current OVX Kiss1-floxed/Oprk1-Cre rats.This could be simply due to estrogen-dependent suppression of ARC Kiss1 expression in remaining KOR-negative KNDy neurons, as the E2 treatment tended to reduce the number of ARC Kiss1-expressing cells in OVX Kiss1-floxed/Oprk1-Cre rats.Indeed, the diestrous level of E2 caused a reduction of Gnrhr, Lhb, and Fshb mRNA expression levels in the pituitary of OVX Kiss1-floxed/Oprk1-Cre rats.In addition, it is tempting to speculate that the estrogendependent LH pulse disruption in Kiss1-floxed/Oprk1-Cre rats may be due to estrogen-dependent activation of some inhibitory signaling pathway(s).In support of this, our previous studies showed that the suppression of LH pulses by fasting or glucoprivation was dependent on the negative feedback level of E2 in OVX rats 47,48 and that central DOR or MOR antagonism restored LH pulses in OVX + low E2 rats under glucoprivation 45,46 .Further studies are required to identify the inhibitory signaling pathway(s) activated by the diestrous level of E2 to suppress GnRH/LH pulse generation in Kiss1-floxed/Oprk1-Cre rats.
The present study showed that Oprk1 expression in KNDy neurons increased after the peripubertal period, as Oprk1-driven Cre-activated tdTomato expression was found in approximately 20% of ARC Tac3 (a marker of KNDy neurons)-expressing cells at 3 weeks of age (prepubertal period) and 92% of ARC Tac3-expressing cells in adulthood.These findings suggest that Kiss1 may have been knocked out in only 20% of KNDy neurons in Kiss1-floxed/Oprk1-Cre female rats at the onset of puberty, which may explain why the timing of puberty onset in the current ovary-intact Kiss1-floxed/Oprk1-Cre female rats was comparable to that in Cre(−)/Kiss1-floxed control rats.It is speculated that after puberty onset, the Kiss1 gene may be gradually knocked out as Oprk1 expression gradually increases in KNDy neurons.Indeed, the findings in the current Oprk1-Cre-activated tdTomato reporter female rats revealed that approximately 90% of KNDy neurons, at least once (even transiently), expressed Oprk1 mRNA by adulthood.Thus, the Kiss1 gene was knocked out in a majority of KNDy neurons www.nature.com/scientificreports/ in adult Kiss1-floxed/Oprk1-Cre female rats, consequently resulting in a longer estrous cycle, smaller ovarian weight, and smaller litter size than those in Cre(−)/Kiss1-floxed control rats.
The current study suggests that AVPV kisspeptin neurons, which are expressing or have expressed Oprk1, are needed to generate a full amount of LH surge because the AUC of the high E2-induced LH surge in Kiss1floxed/Oprk1-Cre rats was approximately half of that in Cre(−)/Kiss1-floxed control rats.The reduction in the LH surge could be simply due to the reduction in the number of AVPV Kiss1-expressing cells (approximately half of the control) in the current Kiss1-floxed/Oprk1-Cre rats, in which the ratio of Kiss1 KO in AVPV kisspeptin neurons is largely consistent with a previous study showing Oprk1 mRNA expression in 50% of AVPV kisspeptin neurons in female mice 32 .On the other hand, it is likely that KOR-negative AVPV kisspeptin neurons still function as the GnRH/LH surge generator because the Kiss1-floxed/Oprk1-Cre rats showed a surge-like www.nature.com/scientificreports/increase in LH levels in the afternoon, and the ratios of the peak levels of LH surge to the baseline LH levels were comparable to those in Kiss1-floxed control rats.In support of this notion, previous studies showed that ~ 60% of AVPV kisspeptin neurons were activated at the proestrous and E2-induced LH surge in female rodents 10,12 .
Our findings should be interpreted in the context of the following limitation.Kiss1 gene was deleted in > 95% of KNDy neurons and > 50% of AVPV kisspeptin neurons in Kiss1-floxed/Oprk1-Cre female rats, whereas ISH showed Oprk1 mRNA expression only in approximately 50% of KNDy neurons and 5% of AVPV kisspeptin neurons in Kiss1-floxed controls.In addition, Oprk1-Cre/tdTomato reporter rats showed many Cre-activated tdTomato-expressing cells, that did not coexpress Oprk1 mRNA, in the ARC, PVN, and SON.These inconsistencies may be due to the detection threshold of ISH for Oprk1.Namely, Oprk1 mRNA expression might be below the threshold for the current Oprk1 ISH in some Oprk1-expressing cells.Improvement of the methodology to detect Oprk1 mRNA may solve the inconsistency between the Oprk1 expression ratio in ARC and AVPV kisspeptin neurons and the Kiss1 KO rate in Kiss1-floxed/Oprk1-Cre female rats and between the number of Cre-activated tdTomato-expressing cells and Oprk1-expressing cells in Oprk1-Cre/tdTomato reporter rats.
In conclusion, the present study demonstrated that kisspeptin neurons, which are expressing and have expressed Oprk1, in both the ARC and AVPV had a role in maintaining GnRH pulse and surge generation in female rats because the current Kiss1-floxed/Oprk1-Cre female rats showed profound disruption of LH pulses in the presence of a diestrous level of E2 and a reduction in the peak of the E2-induced LH surge.This suggests that KNDy and AVPV kisspeptin neurons, which are expressing and have expressed Oprk1, seem to be needed for normal GnRH/LH pulse generation and for normal peak levels of the estrogen-induced GnRH/LH surge, respectively, to support a normal estrous cycle length and offspring number in ovary-intact female rats.Another important finding of the study is that Kiss1-floxed/Oprk1-Cre rats, in which Kiss1 was expressed only in KOR-negative kisspeptin neurons, were still fertile and maintained the GnRH/gonadotropin pulse-generating mechanism, especially under the steroid-free condition.

Methods Animals
Oprk1-Cre/tdTomato reporter rats were generated by crossing newly generated Oprk1-Cre rats (see below for details) and tdTomato reporter rats [LE-Tg(Gt(ROSA)26Sor-CAG-tdTomato)24Jfhy rats 37 , supplied by the National BioResource Project-Rat, Kyoto University, Kyoto, Japan (NBRP Rat No. 0734) with support in part by the National BioResource Project of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Tokyo, Japan].Kiss1-floxed/Oprk1-Cre rats were generated by crossing Oprk1-Cre rats and Kiss1-floxed rats (Rat Genome Database identification (RGD ID): 125097497) 6 .The resulting Kiss1-floxed/Oprk1-Cre (specifically, Kiss1 fl/fl /Oprk1 Cre+ ) female rats and Cre(−)/Kiss1-floxed (specifically, Kiss1 fl/fl /Oprk1 Cre-) female control rats were used.Kiss1-floxed/Kiss1-Cre rats were generated by crossing Kiss1-Cre rats 36 and Kiss1-floxed rats.The resulting Kiss1-floxed/Kiss1-Cre (specifically, Kiss1 fl/Cre+ ) female rats and Cre(−)/Kiss1-floxed (specifically, Kiss1 fl/Cre-) female control rats were used.Wild-type Iar:Wistar-Imamichi male rats (RGD ID: 125097496, Institute for Animal Reproduction, Kasumigaura, Japan) were used for the mating test.All rats were maintained in a room with a 14:10-h light/dark cycle (lights on at 05:00 h) at 22 ± 3 °C and had free access to food (CE-2; CLEA Japan, Tokyo, Japan) and water.Female rats were housed by groups of 2-4 animals per cage (if not otherwise specified).Female rats with two consecutive 4-or 5-day estrous cycles, as determined by vaginal smears, were mated overnight with male rats on the day of the proestrus.The resultant pregnant females, as determined by the presence of vaginal plugs, were housed individually.The day when the vaginal plug was found in the morning was designated gestational Day 0, and the day when the newborn litter was found at noon was designated postnatal Day 0. Genotypes were analyzed by PCR with DNA obtained from newborn pups.The primers are listed in Supplementary Table S1 online.The litter size was recorded and adjusted to eight by postnatal Day 5 to minimize the growth variation within and between litters.The pups were weaned on postnatal Day 20 or 21.Animals were subjected to brain, pituitary, ovary, or blood collection at 8-13 weeks of age (adulthood) or 21 days of age (prepubertal period).If not otherwise specified, surgeries were conducted under aseptic conditions and anesthesia with an intraperitoneal injection of a ketamine (27 mg/kg)-xylazine (5.3 mg/kg) mixture, followed by inhalation of isoflurane (1-2% in air).After the surgery, female rats were housed individually.Care of the animals and all experimental procedures performed in the present study were reviewed and approved

Gene targeting and generation of Oprk1-Cre rats
To generate Oprk1-Cre rats, a targeting vector was designed to replace the stop codon of the Oprk1 gene with the T2A-Cre sequence (see Supplementary Fig. S1 online).The 5'-and 3'-homology arms and T2A-Cre sequences were inserted into pAAV-MCS3.Production of adeno-associated virus (AAV) type 6 carrying the targeting vector was performed as described previously 49 .A guide RNA targeted around the stop codon of the rat Oprk1 gene (5′-TAA GCC AGT ATG ACT AGT CA-3′) was synthesized by and purchased from Integrated DNA Technologies (Coralville, IA).The CRISPR/Cas9 components were introduced by electroporation, and the AAV targeting vector was transfected at the pronuclear stage of rat zygotes collected from Crlj:WI rats (RGD ID: 2312504, Charles River Laboratories Japan, Yokohama, Japan), as described previously 50 .Two-cell stage embryos were transferred to pseudopregnant female rats under isoflurane anesthesia (2-3% in air).Genotypes of the obtained animals were confirmed by PCR and sequencing.The primers for PCR are listed in Supplementary Table S1 online.

Brain sampling from prepubertal and adult Oprk1-Cre/tdTomato reporter female rats
Adult female Oprk1-Cre/tdTomato reporter rats (n = 3, collected from 3 litters) were OVX and immediately received subcutaneous Silastic tubing (1.57mm inner diameter; 3.18 mm outer diameter; 25 mm in length; Dow Corning, Midland, MI) filled with E2 (Sigma-Aldrich, St. Louis, MO) dissolved in peanut oil (Sigma-Aldrich) at 20 μg/ml for 1 week to produce a negative feedback level of plasma E2 (low E2) 47 .Ovary-intact prepubertal Oprk1-Cre/tdTomato reporter rats (n = 3, collected from 2 litters) and OVX + low E2 adult Oprk1-Cre/tdTomato reporter rats were deeply anesthetized with sodium pentobarbital (40 mg/kg, Tokyo Chemical Industry, Tokyo, Japan) and then intracardially perfused with phosphate-buffered saline, followed by 4% paraformaldehyde (Sigma-Aldrich).The brains were immediately removed and postfixed in the same fixative overnight at 4 °C, followed by immersion in 30% sucrose in 0.05 M phosphate buffer until the brains sank at 4 °C.Frozen frontal sections containing the ARC (50-μm thickness) were prepared using a cryostat (CM1800, Leica Biosystems, Wetzlar, Germany).Every fourth ARC section was used for double ISH to visualize tdTomato/Oprk1 or Tac3/tdTomato.Tac3 was used as a marker of KNDy neurons because our previous study showed that Tac3 expression was evident in KNDy neurons before and after puberty onset in female rats 51 .www.nature.com/scientificreports/Cre(−)/Kiss1-floxed rats (n = 3), were subjected to frequent blood sampling to determine LH pulses.A silicon cannula for blood sampling (inner diameter 0.5 mm and outer diameter 1.0 mm; Shin-Etsu Polymer, Tokyo, Japan) was inserted into the right atrium through the jugular vein 1 day before the onset of blood sampling.Blood samples (100 μL) were collected from freely moving conscious OVX rats every 6 min for 3 h (13:00 to 16:00 h), and plasma samples were collected after the centrifugation at 4 °C.An equivalent volume of rat red blood cells, which were taken from donor rats and diluted with heparinized saline, was replaced through the atrial catheter after each blood collection to keep the hematocrit constant.One day after blood sampling, the brains and anterior pituitary were collected.The brains were processed as described above.Frozen frontal sections containing the ARC (50-μm thickness) were prepared using a cryostat.Every fourth ARC section was used for ISH to visualize Kiss1 and Oprk1 in Kiss1-floxed/Oprk1-Cre rats or Kiss1 in Kiss1-floxed/Kiss1-Cre rats.The anterior pituitary was stored at − 80 °C for reverse transcription-quantitative PCR (RT-qPCR) analyses for Gnrhr, Lhb, and Fshb mRNA expression.

Analyses of body
Blood, pituitary, and brain sampling from E2-treated OVX Kiss1-floxed/Oprk1-Cre rats Kiss1-floxed/Oprk1-Cre rats (n = 5) and their counterpart Cre(−)/Kiss1-floxed control rats (n = 6) were OVX (as described above) and immediately received Silastic tubing filled with E2 dissolved in peanut oil at 20 μg/mL for 1 week to serve as OVX + low E2 rats.OVX + low E2 Kiss1-floxed/Oprk1-Cre rats and Cre(−)/Kiss1-floxed control rats were subjected to frequent blood sampling to determine LH pulses as described above.One day after the frequent blood sampling, the estrogen tubing was replaced with another tube containing E2 dissolved in peanut oil at 1000 μg/mL to produce a positive feedback level of plasma E2 (high E2) 6,52 .Blood samples (100 μL) were collected from Kiss1-floxed/Oprk1-Cre rats (n = 4) and their counterpart Cre(−)/Kiss1-floxed control rats (n = 6) every 1 h from 10:00 to 21:00 h 2 days after the high E2 replacement to determine the afternoon LH surge.One day after the second blood sampling to detect the LH surge, the estrogen tubing was replaced with another tube containing E2 dissolved in peanut oil at 20 μg/mL (low E2).Estrogen tubing replacement was performed under isoflurane anesthesia (2-3% in air).One week later, the brains and anterior pituitary of Kiss1floxed/Oprk1-Cre rats (n = 4) and their counterpart Cre(−)/Kiss1-floxed control rats (n = 5) were collected and processed as described above.Every fourth ARC section and every second AVPV section were used for double ISH to visualize Kiss1 and Oprk1.

RT-qPCR analyses of ARC Kiss1 mRNA expression and pituitary Gnrhr, Lhb, and Fshb mRNA expression
Total RNA was extracted from the ARC using ISOGEN (Nippon Gene, Tokyo, Japan).Total RNA was extracted from the fixed hemipituitary by using the RNeasy FFPE kit (QIAGEN) as described previously 6 .cDNA from each sample was synthesized with oligo (deoxythymidine) primers at 37 °C using the high-capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA).Gene expression levels were determined using QuantStudio 3 (Applied Biosystems) with the Thunderbird SYBR Green qPCR Mix (TOYOBO, Osaka, Japan) as described previously 38,52 .The forward and reverse primers for Kiss1, Gnrhr, Lhb, Fshb, and Actb (encoding β-actin) are listed in Supplementary Table S1 online.The specificity of the amplification products was confirmed by dissociation curve analysis.The relative gene expression levels of Kiss1, Gnrhr, Lhb, and Fshb were normalized to Actb, and the fold changes between the groups were calculated using the 2 -ΔΔCT method.

Figure 2 .
Figure 2. Kiss1-floxed/Oprk1-Cre female rats showed puberty onset, estrous cycles, and fertility, whereas Kiss1floxed/Kiss1-Cre female rats lacked puberty onset.Growth curves and puberty onset in Kiss1-floxed/Oprk1-Cre female rats and their counterpart Cre(−)/Kiss1-floxed female rats (a) or Kiss1-floxed/Kiss1-Cre female rats and their counterpart Cre(−)/Kiss1-floxed female rats (b).Body weights are the means ± SEM.Numbers in parentheses indicate the number of animals used.The timing of vaginal opening as an external sign of puberty onset is expressed as a percentage of the total number of animals in each group.Representative photographs of ovaries and ovarian weights of Kiss1-floxed/Oprk1-Cre rats and their counterpart Cre(−)/Kiss1-floxed rats (c) or Kiss1-floxed/Kiss1-Cre rats and their counterpart Cre(−)/Kiss1-floxed rats (d).Scale bars, 5 mm.Open circles indicate the individual data.(e) Individual data of estrous cycles from the Kiss1-floxed/Oprk1-Cre group and Cre(−)/Kiss1-floxed control group.E, estrus; P, proestrus; and D, diestrus.(f) Cycle length (on average of two consecutive cycles) of Kiss1-floxed/Oprk1-Cre and Cre(−)/Kiss1-floxed groups.A dagger indicates a statistically significant difference (p < 0.05) between the groups based on the Wilcoxon rank sum test.Open circles indicate the individual data.(g) Litter sizes of Kiss1-floxed/Oprk1-Cre and Cre(−)/Kiss1-floxed female rats, which were mated with Iar:Wistar-Imamichi stud male rats.Values are the means ± SEM.Open (pups were delivered on gestational Day 22) and closed (pups were delivered on gestational Day 23) circles indicate the individual data.Numbers in (or on) each column indicate the number of animals used.Asterisks indicate statistically significant differences (p < 0.05) between the groups based on Student's t-test.
13:20495 | https://doi.org/10.1038/s41598-023-47222-5www.nature.com/scientificreports/by the Animal Experiment Committees of Nagoya University and the National Institutes of Natural Sciences.This study was conducted in accordance with the Nagoya University Regulations on Animal Care and Use in Research and ARRIVE guidelines.

K is s 1 Figure 5 .
Figure 5. Treatment with a proestrous level of E2 (high E2) induced a surge-like increase in LH in OVX Kiss1floxed/Oprk1-Cre rats, but the LH surge was attenuated by the conditional Kiss1 deletion in cells expressing or having expressed Oprk1.(a) Mean plasma LH profiles of OVX + high E2 Kiss1-floxed/Oprk1-Cre rats and Cre(−)/Kiss1-floxed control rats.The area under the curve of the afternoon LH surge (b), the peak levels of the LH surge (c), the baseline LH levels (d), and the ratios of the peak levels of the LH surge to the baseline LH levels (e) in OVX + high E2 Kiss1-floxed/Oprk1-Cre rats and Cre(−)/Kiss1-floxed control rats.Values are the means ± SEM.Numbers in each column indicate the number of animals used.An asterisk indicates a statistically significant difference (p < 0.05) between the groups based on Student's t-test.