The interplay between kisspeptin and endocannabinoid systems modulates male hypothalamic and gonadic control of reproduction in vivo

Introduction Male reproduction is under the control of the hypothalamus–pituitary–gonadal (HPG) axis. The endocannabinoid system (ECS) and the kisspeptin system (KS) are two major signaling systems in the central and peripheral control of reproduction, but their possible interaction has been poorly investigated in mammals. This manuscript analyzes their possible reciprocal modulation in the control of the HPG axis. Materials and methods Adolescent male rats were treated with kisspeptin-10 (Kp10) and endocannabinoid anandamide (AEA), the latter alone or in combination with the type 1 cannabinoid receptor (CB1) antagonist rimonabant (SR141716A). The hypothalamic KS system and GnRH expression, circulating sex steroids and kisspeptin (Kiss1) levels, and intratesticular KS and ECS were evaluated by immunohistochemical and molecular methods. Non-coding RNAs (i.e., miR145-5p, miR-132-3p, let7a-5p, let7b-5p) were also considered. Results Circulating hormonal values were not significantly affected by Kp10 or AEA; in the hypothalamus, Kp10 significantly increased GnRH mRNA and aromatase Cyp19, Kiss1, and Kiss1 receptor (Kiss1R) proteins. By contrast, AEA treatment affected the hypothalamic KS at the protein levels, with opposite effects on the ligand and receptor, and SR141716A was capable of attenuating the AEA effects. Among the considered non-coding RNA, only the expression of miR145-5p was positively affected by AEA but not by Kp10 treatment. Localization of Kiss1+/Kiss1R+ neurons in the arcuate nucleus revealed an increase of Kiss1R-expressing neurons in Kp10- and AEA-treated animals associated with enlargement of the lateral ventricles in Kp10-treated animals. In the brain and testis, the selected non-coding RNA was differently modulated by Kp10 or AEA. Lastly, in the testis, AEA treatment affected the KS at the protein levels, whereas Kp10 affected the intragonadal levels of CB1 and FAAH, the main modulator of the AEA tone. Changes in pubertal transition-related miRNAs and the intratesticular distribution of Kiss1, Kiss1R, CB1, and CB2 following KP and AEA treatment corroborate the KS-ECS crosstalk also showing that the CB1 receptor is involved in this interplay. Conclusion For the first time in mammals, we report the modulation of the KS in both the hypothalamus and testis by AEA and revealed the KP-dependent modulation of CB1 and FAAH in the testis. KP involvement in the progression of spermatogenesis is also suggested.

Introduction: Male reproduction is under the control of the hypothalamuspituitary-gonadal (HPG) axis.The endocannabinoid system (ECS) and the kisspeptin system (KS) are two major signaling systems in the central and peripheral control of reproduction, but their possible interaction has been poorly investigated in mammals.This manuscript analyzes their possible reciprocal modulation in the control of the HPG axis.
Results: Circulating hormonal values were not significantly affected by Kp10 or AEA; in the hypothalamus, Kp10 significantly increased GnRH mRNA and aromatase Cyp19, Kiss1, and Kiss1 receptor (Kiss1R) proteins.By contrast, AEA treatment affected the hypothalamic KS at the protein levels, with opposite effects on the ligand and receptor, and SR141716A was capable of attenuating the AEA effects.Among the considered non-coding RNA, only the expression of miR145-5p was positively affected by AEA but not by Kp10 treatment.Localization of Kiss1+/Kiss1R+ neurons in the arcuate nucleus revealed an increase of Kiss1R-expressing neurons in Kp10-and AEA-treated animals

Introduction
The hypothalamus-pituitary-gonadal (HPG) axis controls reproduction in both sexes and finds the main actors in the hypothalamic gonadotropin-releasing hormone (GnRH), pituitary gonadotropins, and gonadal sex steroids (1).Endocrine feedback mechanisms and also autocrine and paracrine communications along the HPG axis regulate reproductive functions in response to endogenous and exogenous environmental factors to guarantee the production of high-quality gametes (1).In males, the production of spermatozoa (SPZ) requires the coordination between mitosis, meiosis, and differentiation of germ cells inside the testis.All these processes are coordinated by an endocrine route involving the hypothalamic GnRH, the pituitary follicle-stimulating hormone (FSH), and the luteinizing hormone (LH).Gonadotropins target the somatic cells within the gonad (i.e., Sertoli cells and Leydig cells) which in turn produce sex steroids (i.e., mainly testosterone but also estradiol at a lesser amount) and sustain spermatogenesis progression to form high-quality SPZ.
The endocannabinoid system (ECS) and the kisspeptin system (KS) are two major signaling systems in the central and peripheral control of reproduction (2)(3)(4)(5)(6)(7)(8).The ECS comprises ligands [i.e., e n d o c a n n a b i n o i d s l i k e a n a n d a m i d e ( A E A ) a n d 2arachidonoylglycerol (2-AG) and other endocannabinoid-like mediators such as N-acylethanolamines (NAEs)], canonical and non-canonical cannabinoid receptors [i.e., the G-coupled membrane receptors CB1 and CB2, the transient receptor potential cation channel subfamily V member 1 (TRPV1), and the nuclear peroxisome proliferator-activated receptors (PPARs) among others] (9), biosynthetic and hydrolyzing enzymes, and transporters (9).The KS currently includes kisspeptins, the cleavage products of the Kiss1 precursor, and the membrane kisspeptin receptor (Kiss1R) (5).In general, KS is considered the gatekeeper of the HPG axis from puberty onset onward (5) since central precocious puberty or hypogonadotropic hypogonadism occurs in humans and rodents as a consequence of gain of function and inactivating mutations of KISS1/KISS1R, respectively (10)(11)(12).Also, the ECS modulates the release of the hypothalamic GnRH, but its activity is opposite to KS.In fact, in vivo and ex vivo, the endocannabinoid AEA acts as retrograde signals to inhibit the release of GnRH (13)(14)(15).The characterization of the reproductive phenotype of CB1 knockout (CB1-/-) animals revealed the downregulation of the reproductive axis (16,17).At the periphery, the ECS directly affects the postnatal development of Leydig cells (18, 19) steroidogenic ability (20), spermatogenesis progression (21,22), and sperm maturation and quality (23)(24)(25)(26).CBs are differently expressed in testicular germ cells and somatic cells in mammals (21,22), and the modulation of endocannabinoid tone, primarily via the fatty acid amide hydrolase (FAAH), represents a key feature in endocannabinoid signaling.In general, CB1 has been linked to the control of the activity of Leydig cells (18-20), the recruitment of spermatogonial stem cells, chromatin remodeling, and the maturation of post-meiotic stages (22, 24, 26).Conversely, CB2 activity has been linked to the control of Sertoli cell physiology (27) and to spermatogenesis progression toward mitotic and meiotic stages (21).Endocannabinoid tone within the epididymis controls the acquisition of sperm motility in the cauda epididymis (23,28), and data on CB1 pharmacological inhibition by its antagonist rimonabant (SR141716A) in human sperm and CB1 −/− mice linked the inactivity of CB1 to reduced sperm survival and quality with defects in nuclear sperm size (25,29) and disulfide bond formation (26).Involvement of the ECS in acrosome reaction and capacitation has also been reported in vertebrates including humans (2,30).
Despite the well-known role of the KS in the central control of reproduction, at the periphery, the expression of the KS has been reported in cell lines, testis, SPZ, and epididymis collected from several species (31)(32)(33)(34)(35)(36)(37)(38)(39)(40)(41)(42)(43)(44), but the real functions of the KS in the testis and SPZ are far to be fully elucidated (8,45,46).In general, Leydig cells and germinal epithelium are the main sources and targets of Kiss1 in several species (8), and Kiss1 secretion from testis has been suggested (47).However, due to different experimental designs, species, and methodologies, data reporting Kiss1/Kiss1R intratesticular localization, effects on steroidogenesis, spermatogenesis progression, and SPZ function are still debated (8,45).Similarly, the involvement of the KS in Ca 2+ mobilization and SPZ physiology has been investigated in humans and mice (36,37) and is related to human sperm motility and hyperactivation (36) and fertilizing ability (37).Very recently, the involvement of Kiss1R in the cytoskeletal-nucleoskeletal pathway occurring in mouse Leydig cells to modulate gene expression was reported (40); in parallel, the possible role of the KS in epididymal SPZ maturation and storage was demonstrated in dogs (38) and rats (39), with significant Kiss1 levels measured within the epididymis (39).Based on the overall considerations, it appears that both ECS and KS centrally and locally regulate male reproduction; however, their possible interplay is still unraveled.
Hence, in this manuscript, we used an adolescent rat model treated in vivo with kisspeptin-10 (Kp10), or the main endocannabinoid AEA, alone or in combination with CB1 antagonist SR141716A, to investigate the possible reciprocal modulation of ECS and KS in rat testis.For the first time in mammals, we reported the modulation of the KS in both the hypothalamus and testis by AEA and revealed the KP-dependent modulation of CB1 and FAAH in the testis.The effects on the intratesticular distribution of Kiss1, Kiss1R, CB1, and CB2 following KP and AEA treatment are also reported.

Animals and tissue collection
Early peripubertal Wistar male rats (body weight 250-300 g) [38 post-natal days (pnd), n = 10/experimental group (Harlan Laboratories, Italy)] were housed two per cage with food and water

Gonadosomatic index and animal growth during the treatment
At sacrifice, the gonadosomatic index (GSI) was calculated using the formula: ½testes weight (g)=body weight (g) at sacrif ice Â 100 The growth of the animals was evaluated as: 100 Â ½tibial lenght (mm)=body weight (g) at sacrif ice Plasma collection and measure of circulating Kiss1, testosterone, and estradiol Blood was centrifuged at 800×g for 15 min at 4°C for plasma collection.Plasma was immediately frozen at −80°C until used for the measure of Kiss1 and sex steroid by enzyme-linked immunosorbent assay (ELISA).

Total protein extraction and Western blot
Rat testes (n = 6) and hypothalamus (n = 4) from in-vivotreated animals were processed for the extraction of total proteins by the RIPA Lysis Buffer System (Santa Cruz Biotechnology sc-24948) supplied with protease inhibitor cocktail (10 μl/ml of RIPA), 200 mM of phenylmethylsulfonyl fluoride (PMSF, 10 μl/ml of RIPA), and 100 mM of sodium orthovanadate (10 μl/ml RIPA) following the manufacturer's instructions.Cleared protein extracts were collected by centrifugation at 11,000×g for 30 min at 4°C and quantified by Lowry assay (52).
Total proteins (40 μg) were resolved on precast 4%-20% Mini-Protean TxG gels (Bio-Rad Laboratories, CA, USA) and electrotransferred into PVDF filters by the TransBlot Turbo Transfer System (Bio-Rad Laboratories).Membranes were blocked in UltraCruz Blocking solution (sc-516214, Santa Cruz Biotechnology) for 1 h at room temperature (RT) and incubated overnight with primary antiserum (details in Table 1).After washing three times for 10 min in T-TBS (0.25% Tween-20 in Tris-buffered saline, TBS, pH 7.6) at RT, membranes were incubated 1 h at RT with the appropriate horseradish peroxidaseconjugated (HRP) secondary antiserum (details in Table 1) in UltraCruz Blocking solution and then washed again in T-TBS.Western blotting luminol reagent was used to detect the signals (Santa Cruz Biotechnology sc-2048).To quantify the protein signal, each membrane was stripped at RT for 30 min in stripping buffer (Santa Cruz Biotechnology sc-281698) and then reprobed with the mouse monoclonal anti-a-tubulin (details in Table 1).Autoradiographic films were scanned; signals were plotted and quantified with ImageJ.Western blot data are reported as specific protein optic density (OD)/housekeeping OD ± SD (n = 3 at least).

RT-qPCR
Total RNA was extracted by TRIzol reagent (Thermo Fisher Scientific, Milan, Italy) following the manufacturer's instructions and treated with DNAseI to eliminate any contamination of genomic DNA (Invitrogen Life Technologies: Carlsbad, California, USA).After quantification and quality evaluation on agarose gel, 2.5 μg of total RNA was reverse-transcribed in a final volume of 20 μl using SuperScript-III RnaseH -Reverse Transcriptase (Thermo Fisher Scientific), 0.5 μg of oligo dT (18), 0.5 mM of dNTP mix, 5 mM of DTT, 1X first-strand buffer (Thermo Fisher Scientific), and 40 U of RNase Out (Thermo Fisher Scientific) following the manufacturer's instructions.As a negative control, total RNA not treated with reverse transcriptase was used.
The 2 −DDCt method was used for the relative quantification of gene expression (53).Assays also included a melting curve analysis for which all samples displayed single peaks for each primer pair.Data were reported as mean fold change ± SEM over the value one arbitrarily assigned to the control group.

miRNA expression analysis
For the analysis of miRNA expression, miRNAs were extracted from the testis by the miRNeasy tissue/cells/advanced mini kit and cleared from DNA by the gDNA Eliminator Spin Columns (Qiagen, Milan, Italy).Following the manufacturer's instructions, reverse transcription was carried out using the miRCURY LNA-RT Kit; qPCR was carried out in miRCURY LNA miRNA PCR Assay for the custom-validated primers for hsa-miR145-5p, has-miR-132-3p, has-let7a-5p, and has-let7b-5p and the housekeeping U6 (reverse transcription kit, master mix, and validated primers all from Qiagen).

Nissl staining of the hypothalamic nuclei and dimension of the lateral ventricle
Hypothalamus coronal sections were processed for Nissl staining as already described (54).Briefly, after deparaffinization and rinsing in distilled H 2 O, sections were incubated for 1 min in thionine.Slides were then dehydrated through serial ethanol solutions (70% to 100%), cleared in xylene, and mounted with Eukitt® Quick-hardening mounting medium.Images were acquired by using a bright-field microscope (Olympus SC180) at ×4 and ×10 magnification.Sections obtained from the left cerebral hemisphere were imaged for bright-field microscopy using the Olympus SC180 microscope.The areas of the lateral ventricle were manually traced and measured using ImageJ, and the measurements (n = 4 animals for each group) were recorded in square millimeters (54).

IFL
For paraffin-embedded sections of the testis and hypothalamus, the IF protocol included deparaffination in Clear-Rite 3 (Thermo Scientific, Waltham, MA, USA) rehydration through descending degrees of alcohol up to water and non-enzymatic antigen retrieval in sodium citrate buffer 10 mM, pH 6.0, for 30 min at 95°C.Slides were washed three times with ice-cold PBS for 10 min and permeabilized with 0.1% Triton X-100 in PBS.For Kiss1 and Kiss1R analysis, a treatment with HCl 2 M for 30 min at 37°C was performed.Then, the slides were washed in PBS and blocked with a solution containing 10% of appropriate normal serum in PBS for 1 h at RT.The testis sections were incubated with anti-Kiss1, anti-Kiss1R, anti-CB1, and/or anti-CB2 or stained with 4′,6diamidine-2′-phenylindole dihydrochloride solution (DAPI, 1.5 μg/ml in distilled H 2 O), while the hypothalamus sections were processed using anti-neuronal nuclei (NeuN) and anti-Kiss1.All sections were incubated with primary antisera overnight at 4°C in a humidified chamber.After three washing steps, the slides were incubated with the appropriate fluorescent dye-conjugated secondary antibodies for 1 h at RT. Details of the primary and secondary antisera are given in Table 1.
Negative controls were performed using all reagents except the primary antibody.Samples were then coverslipped with mounting gel (80% glycerol in PBS).
The samples were analyzed using a confocal laser scanning microscope (Leica SP5, Leica Microsystems, Wetzlar, Germany).Images were acquired in sequential scan mode by using the same acquisition parameters (laser intensities; gain photomultipliers; pinhole aperture; magnification ×20, ×40, or ×63).Figures were assembled using Adobe Photoshop 7 and Adobe Illustrator 10.

IHC
Ten-micron-thick brain sections, mounted on poly-L-lysinecoated glass slides, were analyzed by IHC using the anti-Kiss1R antibody (details and working conditions in Table 1).The IHC protocol included deparaffination, non-enzymatic antigen retrieval, and permeabilization as reported in the previous section.Then, the slides were immersed in 3.6% hydrogen peroxidase solution for 10 min to inactivate endogenous peroxidases and blocked with a solution containing 10% of normal serum in PBS for 1 h at RT.After rinsing, the samples were incubated overnight at 4°C with anti-Kiss1R antibody.Then, the sections were washed twice in PBS and incubated with the secondary antibody for 1 h at RT, as in Table 1.Slides were then incubated with streptavidin-horseradish peroxidase (Vector Laboratories) for 1 h at RT.The immunoreaction signal was detected using a substrate chromogen solution containing 3,3-diaminobenzidine tetrahydrochloride (DAB substrate kit, Vector Laboratories).Finally, the sections were coverslipped with the Eukitt® Quick-hardening mounting medium (Sigma-Aldrich).For histological analysis, H&E staining was performed on the testis sections following standard procedures (54).Slides were analyzed using a brightfield microscope (Olympus SC180) at ×10 magnification.The number of the Kiss1R-positive cells within the arcuate nucleus (ARC) was counted in each section for n = 4 per experimental group using ImageJ software, and the mean number of cells/animal was used for further statistical analysis.

Statistics
Student's t-test and ANOVA followed by Duncan's test were carried out to assess the significance of the differences in gene/ protein expression, IFL, and IHC signal.Student's t-test was used to assess the statistically significant differences between the treatment groups and the control group; ANOVA followed by Duncan's test was used for multigroup comparison.Data were considered significant at p<0.05.

Kp10 and AEA treatments do not affect the morphometric parameters of the treated animals
Weight gain is a recognized tool of puberty and SR has a possible anorectic effect (55).Hence, we measured the body weight and the length of the tibia to evaluate the correct growth of the treated animals vs. the control group during the 3-week-long treatments.At sacrifice, no differences in body weight, tibial length, testes weight, GSI, and growth were observed between the control and treated animals (Table 2).
Kp10 and AEA differentially modulated Kiss1R, Kiss1, and Cyp19 proteins, GnRH mRNA, and puberty-related miRs in the hypothalamus Kisspeptin is the main upstream modulator of GnRH, from puberty onset onward (5); both kisspeptin and estradiol (5,6,31,40), the sex steroid produced by the irreversible aromatization of testosterone by Cyp19 aromatase, were found as positive modulators of the KS.Conversely, the HPG axis is downregulated in CB1-/-mice (16,17).To test the possible interplay between KS and ECS on the above issues, we evaluated the effects of Kp10 and AEA on Kiss1R, Kiss1, and Cyp19 proteins and GnRH mRNA in the mediobasal hypothalamus of the treated animals by Western blot and qPCR, respectively.The KP administration caused a slight increase of Kiss1R, Kiss1, and Cyp19 proteins and GnRH mRNA (Figures 1A-E), but the change was statistically significant only for Kiss1 and Cyp19 proteins and GnRH mRNA.AEA significantly increased Kiss1R protein (p< 0.01), decreased Kiss1 and Cyp19 proteins (p< 0.01), and increased GnRH mRNA, but the trend was not statistically significant (p = 0.065).Pretreatment with SR reverted the AEAdependent effect on both Kiss1R and Kiss1 proteins but lowered Kiss1R protein below the control group, had no effect on Cyp19, and increased GnRH mRNA (p< 0.05).
Since treatments were carried out in pubertal animals and stopped at sex maturation, the effects of Kp10 and AEA treatments on the expression of miRs notably involved in the hypothalamic control of reproduction at puberty (56) were analyzed.The expression rate of miR-132-3p, let-7a-5p, and let-7b-5p did not change vs. the control group following KP or AEA treatment (Figures 1F-I); the expression of miR-145-5p was significantly higher in the AEA-treated group only and SR reverted this AEA-dependent effect (Figure 1G).
Kp10 and AEA affect the number of Kiss1and Kiss1R-expressing neurons in the hypothalamus, but only Kp10 increases the size of the lateral ventricle Based on the Western blot results, we investigated whether the treatments specifically affected Kiss1 and Kiss1R proteins in the hypothalamic ARC, where GnRH is preferentially produced.Hence, we evaluated Kiss1 and Kiss1R in the ARC of treated and untreated rats (Figures 2A-C).Both KP and AEA treatments increased the number of Kiss1R-expressing neurons in the ARC compared with controls (p< 0.01 and 0.05, respectively) (Figure 2B), and the AEA-mediated effect was attenuated by SR administration.KP administration and the combined AEA/SR treatment increased the number of Kiss1-expressing neurons in the ARC (p< 0.01 and p< 0.05, respectively) (Figure 2C).Interestingly, Kiss1 was present not only within the ARC but also along the ependyma of the lateral ventricle as shown in the representative image of the control in Figure 2C.The KP treatment exacerbated this condition suggesting a role for this protein also in cerebrospinal fluid homeostasis (Figure 2C).To further investigate this issue, we analyzed the morphology and size of the lateral ventricle in KP-treated animals.A significant expansion of the ventricle area was observed compared with controls (p< 0.01) (Figure 3).

The administration of Kp10 and AEA does not affect circulating hormonal levels
Changes in the circulating gonadotropins and sex steroids in turn are the main consequences of the overstimulation or inhibition of the HPG axis (1), and the possible effects of KP and AEA administration on the circulating levels of Kiss1 have never been investigated before in rodents.Thus, we measured Kiss1 and sex steroids in the blood to ascertain to what extent treatments affected the physiology of the HPG axis.
The circulating levels of Kiss1, testosterone, and estradiol were measured by ELISA following the in-vivo treatment with KP, AEA, and SR + AEA (Table 3) revealing the lack of any statistically significant change in the measured hormonal levels.

Kp10 decreases Cyp19 and GnRH mRNA in the testis
To further confirm the functionality of the HPG axis and the lack of significant effects on steroid biosynthesis, we investigated by Western blot the effects of the treatments on 3bHsd and Cyp19, the two main enzymes involved in testosterone biosynthesis and conversion into estradiol (57).In addition, the expression rate of intratesticular GnRH mRNA, involved in the control of the physiology of both Leydig cells and spermatogenesis (57), was also considered.Results revealed the absence of any statistically significant difference between the control group and the KP-treated group (Figures 4A-C).The trend of the corresponding mRNA was toward a decrease in the KP group for both 3bHsd and Cyp19 (Figure 4D) but was statistically significant only for Cyp19 mRNA (p< 0.05 vs. the control group).The intratesticular expression of GnRH mRNA was significantly lower in the KP-treated group (p< 0.05).

Effects on spermatogenesis progression: Kp10 increases Lin28A and Sirt1 proteins but has opposite effects than AEA on puberty-related miRs
Due to the possible acceleration of spermatogenesis and tissue damage related to prolonged KP treatments (58, 59), molecular markers of spermatogenesis and testis morphology were investigated.The protein levels of molecular markers of the S phase of the cell cycle (i.e., PCNA) (60), spermatogenesis progression at puberty (i.e., Lin28A) (61), and cell survival (i.e., Sirt1) (62) were analyzed by Western blot (Figure 5A).The protein level of PCNA did not differ between the control and the KP-treated group (Figure 5B), while both Lin28A and Sirt1 protein levels increased following KP treatment (p< 0.05) (Figures 5C, D).
Testis morphology in the control and treated animals was comparable (Figure 5F).

Kp10 increases and AEA reduces the expression of Kiss1R in the testis
The possible self-modulatory loop of KP on the intratesticular KS was analyzed by Western blot (Figure 6A).The KP treatment significantly increased the protein level of the corresponding receptor (Figure 6B) (p< 0.05), without any significant effect on the ligand (Figure 6C).
Following the in-vivo treatment with AEA ± SR141716A, Kiss1R and Kiss1 proteins were analyzed by Western blot (Figure 6D).The AEA treatment significantly reduced the protein Modulation of the intratesticular markers of steroidogenesis and GnRH.Representative Western blots for 3bHsd and Cyp19 level of Kiss1R (Figure 6E) but did not have any significant effect on Kiss1 (Figure 6F) although an increasing trend could be observed.Pretreatment with SR141716A fully reversed the AEA-dependent effects on Kiss1R.

Kp10 and AEA treatments differentially change the intratesticular localization of Kiss1 and Kiss1R
In control animals, the intratesticular localization of Kiss1 and Kiss1R was quite overlapping within the germinal epithelium (Figure 7).It marked mostly elongating spermatids close to the lumen and, to a lesser extent, premeiotic stages (i.e., preleptotene spermatocytes) close to the basal compartment within the germinal epithelium.In the interstitial compartment, a few Kiss1 and numerous Kiss1R-positive elements were detectable.Following the KP treatment, more intense staining was observed for both Kiss1 and Kiss1R in the interstitial compartment and the germinal epithelium.In detail, within the germinal epithelium, the immunoreactivity of Kiss1 and Kiss1R still characterized elongating spermatids but was also detectable in spermatocytes and a few spermatogonia.Following the AEA treatment, Kiss1 and Kiss1R localization in the spermatids was limited to a few scattered spots but was still evident in spermatocytes (Figure 7).A stronger staining for Kiss1 and Kiss1R was detectable in the SR + AEAtreated animals in both preleptotene and pachytene spermatocytes.

Kp10 increases CB1 and FAAH proteins
The possible effects of in-vivo administration of KP on the testicular ECS were evaluated by Western blot focusing on the main cannabinoid receptors, CB1 and CB2, on FAAH and Nacylphosphatidylethanolamine-phospholipase D (NAPE-PLD), the main enzymes responsible for AEA degradation and biosynthesis, respectively.KP treatment significantly increased the protein level of CB1 (Figure 8A) (p< 0.05) but did not have any effect on CB2 (Figure 8B).Kp10 treatment significantly increased the protein level of FAAH (Figure 8C) (p< 0.05), without any statistically significant effect on NAPE-PLD (Figure 8D).
Since CB1 and FAAH were responsive to the KP treatment, a possible self-modulatory loop of AEA on its canonical receptor and its hydrolyzing enzyme was analyzed by Western blot following the AEA treatment (Figure 8E); a lack of any statistically significant change in CB1 and FAAH proteins was observed (Figures 8F, G).

Kp10 and AEA treatments differentially change the intratesticular localization of CB1 and CB2
In control animals, CB1 was localized in both somatic (i.e., interstitial Leydig cells and Sertoli cells) and germ cells (i.e., spermatogonia in the basal compartment and round/ elongating spermatids in the luminal compartment).CB2 had a more scattered and diffuse localization within the interstitium and the germinal epithelium, particularly in Sertoli cells and germ cells at mitotic and meiotic stages.CB1 and CB2 merge revealed a quite overlapping distribution (Figure 9).Following the KP treatment, CB1 was detectable in the interstitium and along the contour of the seminiferous tubule; a scattered and punctuate signal for CB2 was detectable in the germinal epithelium (Figure 9).In AEA-treated animals, CB1 and CB2 spotted signals were well evident in the interstitial and the germinal compartment, with overlapping signals (Figure 9).Pretreatment with rimonabant reduced CB1 but not CB2 staining in both interstitial and germinal compartments (Figure 9) corroborating the CB1-dependent action of AEA at the testicular level.

Discussion
KS and ECS are two main signaling pathways in the central and peripheral control of reproduction (2-8), but their possible interrelation in the control of spermatogenesis and spermatogenesis progression has never been investigated in mammals.Currently, the only available data in the field are related to the control of spermatogenesis in the anuran amphibian Pelophilax esculentus (63), an experimental model used as an archetype for the study of master systems in the physiology of reproduction (64).Nevertheless, in this non-mammalian animal model, the reciprocal communication between KS and ECS occurs centrally to regulate the expression of GnRH and locally, via the estradiol synthesis and FAAH, to modulate the intratesticular tone of endocannabinoids, endocannabinoid signaling, and sex steroids responsible for spermatogenesis progression and maturation of post-meiotic stages (32, 33, 63).In mammals, few reports on the specific hypothalamic communication between KS and ECS are currently available.A gonadal cycledependent role of endocannabinoid signaling in the neuronal circuits involving Kiss1 neurons has been recently reported in female rats demonstrating that i) CB1 is expressed at the mRNA level in the GABAergic and glutamatergic subpopulations of Kiss1 neurons, ii) CB1 is localized in approximately 66% of the Kiss1 afferents to GnRH neurons, and iii) CB1 expression is modulated by estradiol (65).Despite data from females, studies on males are limited to the CB1dependent decrease of Kiss1 transcript in the medial preoptic area and ARC following acute stress, a condition that inhibits the HPG axis (66), and to the participation of hypothalamic CB1 in the disruption of reproductive axis by the inhibition of Kiss1 mRNA and the increase of gonadotrophin-inhibitory hormone expression in a rat model of systemic infections (67).
Puberty is the endpoint of a maturational continuum that leads to reproductive competence and finds the main gatekeeper in the KS.In the present study, we demonstrated that in the hypothalamus of adolescent rats, Kp10 increased the expression of GnRH at the mRNA level as previously reported in vertebrates and cell lines (5,6,32,63).Conversely, AEA, via CB1, modulated the KS, increasing Kiss1R and decreasing the corresponding ligand at the protein level, thus demonstrating the involvement of the ECS in the regulation of KS signaling.In mice, endocannabinoids tonically inhibit GABA (A)-R drive onto GnRH neurons (15,68), and CB1 −/− mice express lower GnRH mRNA level than wild type (17).Furthermore, the administration of AEA-in vivo by intracerebral injection or ex vivo in mediobasal hypothalamus explants-had negative effects on GnRH release following inflammation or alcohol (69,70).These effects were reversed by sex steroids (13).In the present study, despite an increasing trend, the in-vivo AEA treatment was not significant for GnRH mRNA, contrary to previous data in nonmammalian species (14, 63).However, pretreatment with rimonabant, a drug acting as a CB1 antagonist, did not simply revert the AEA effect to the control level; instead, it seemed to exert an effect on itself by significantly reducing Kiss1R and CyP19 protein expression levels and inducing GnRH and Kiss1 in the ARC.This suggests that SR might also behave as a weak inverse agonist by binding receptors other than the CB1.Accordingly, it has been shown that in CB1-/-mice and human post-mortem brains, rimonabant could produce inverse agonist effects by binding further G-coupled receptors (mainly Gi3, Go, and Gz subtypes) (71).In addition, SR has been shown to function as a low-potency mixed agonist/antagonist of the CB1 and the transient receptor potential vanilloid receptor 1 (TRPV1) (4,72,73).This finding suggests the involvement of receptors other than CB1 in the complex neuronal network directly and indirectly controlling the activity of the HPG axis.In fact, the blockade of vanilloid receptors TRPV1 under basal or acute inflammatory conditions decreases GnRH, Kiss1, and Rfrp3 (GNIH) at the mRNA level in male rats (74).
In the hypothalamus, Kiss1 is notably produced in a sex-specific manner by neuronal populations mainly located within the hypothalamic ARC (both females and males) and the anteroventral-peri-ventricular nucleus (AVPV) (females only) to modulate GnRH pulse and mediate sex steroid feedbacks (5,6).The localization of Kiss1 and Kiss1R carried out in the present study confirmed that the ARC is the main brain region responsible for kisspeptin signaling in males, and the specific increase in the number of Kiss1+/Kiss1R+ neurons following Kp10 administration is consistent with the previously reported selfmodulatory loop of the KS (31,40,75,76).However, Western blot analysis of the basal hypothalamus did not show the same increase of Kiss1R found in the IHC of ARC.In addition, according to Western blot data, AEA significantly increased the number of Kiss1R+ cells in ARC, but the expression of Kiss1 protein decreased while remaining quite constant in the Western The apparent discrepancies in these results could be ascribable to the different tests used as Western blot was performed on total hypothalamic homogenates including all other nuclei, whereas immunostaining referred specifically to ARC.The fact remains that the ECS and KP systems strictly communicate with each other at the hypothalamic level, and the effects of AEA treatment are mediated by the CB1 receptor since its pharmacological inhibition by rimonabant attenuated the effects.Our results also showed that the co-treatment with SR141716 reduced Kiss1R (see Western blot data) and increased Kiss1 expression (IF analysis of ARC) to a higher extent than would be expected (i.e., results are significant compared with the control group).These findings suggest that the observed effects may be partially CB1 independent and could be mediated by other receptors such as TRPV1 (74, 77).
Notably, the present study revealed the Kiss1 localization along the ependyma of the lateral ventricle, and KP treatment exacerbated this condition.To the best of our knowledge, this is the first report showing the presence of Kiss1 in this brain region even though we were unable to discriminate if this accumulation represented the preferential route by which the administered Kp10 reached the brain during the treatment or was the result of an increase in endogenous Kiss1 as a result of positive feedback.This finding suggests additional roles for the peptide, including the homeostasis of the cerebrospinal fluid.Accordingly, a significant expansion of the ventricle area was observed in the brain of KP-treated animals in parallel to the increased expression of Cyp19, the enzyme that irreversibly converts testosterone into estradiol.Because estrogens are involved in post-natal brain development (4, 50), the enlargement of cerebral ventricles can be a sign of estrogenization in male rats (54); thus, the higher protein levels of Cyp19 might indicate an increase of estrogen-and estrogen-related signaling in the brain of KP-treated adolescent rats.
The exogenous modulation of the KS usually finds an endpoint in the altered levels of pituitary gonadotropins and sex steroids in plasma.Usually, kisspeptin (i.e., Kp10 or Kp54) administration increases gonadotropins and sex steroid production in vertebrates, including male rhesus monkeys, and in men and as a consequence of HPG axis stimulation (5,8,45,(78)(79)(80).However, long-term KP treatment causes a desensitization of the HPG axis and reduces or has no effect on the production of sex steroids, and in some cases, it causes spermatogenesis damage impairing the physiology of Leydig cells, i.e., the intratesticular sources of testosterone [for a review, see (8,45)].In particular, Aytürk et al. reported opposite effects on testosterone biosynthesis following chronic or acute intraperitoneal administration of Kp10; the chronic administration was not effective, while the acute administration increased testosterone biosynthesis (59).Conversely, low circulating levels of both testosterone and 17bestradiol characterize CB1-/-mice (16,17).In the present study, Kp10 and AEA were administered chronically twice a week for 3 weeks, and both treatments did not have any significant effect in terms of the circulating levels of sex steroid and the intratesticular levels of 3bHsd and aromatase Cyp19.Conversely, KP treatment reduced Cyp19 and GnRH mRNA, suggesting the existence of different modulatory mechanisms in the control of transcription and translation.Nevertheless, in both treatments, the circulating levels of Kiss1 did not differ from the control group, probably due to its short half-life, thus excluding the possible desensitization of the HPG axis in treated animals.In addition, the testicular morphology did not reveal any anomaly following the KP treatment.Taken together, the measured hormonal values and the normal testis morphology exclude the previously reported acceleration of spermatogenesis due to the KP-dependent stimulation of the HPG axis [see (8) for a review].Nevertheless, molecular markers of testis physiology were affected by KP administration.In fact, the testis expression of PCNA, a wellknown marker of DNA replication occurring during the S phase of the cell cycle (60), was unchanged in KP-treated animals, whereas KP produced an increase in protein levels for both Lin28A and Sirt1.In this respect, Lin28A notably marks the spermatogonial progenitor population and regulates the cyclic expansion of the spermatogonial progenitor population as demonstrated by the reduced proliferation of spermatogonial stem cells in Lin28A knockout mice (81).Instead, Sirt1 is highly expressed from late spermatogonia onward with a role in acrosome biogenesis (82), and its expression level drops in response to environmental factors causing the impairment of the HPG axis and testis damage (62).Hence, these findings confirm the hypothesis that intratesticular KP signaling may represent a molecular switch for spermatogenesis progression in mammals, independently from testosterone biosynthesis, as demonstrated in co-cultures of spermatogonial cells with somatic cells (83), and differently from what happens in non-mammalian vertebrates (32, 33).
In this study, we also characterized the intratesticular KS in rats, currently a knowledge gap in the field, confirming previous reports in mammals that addressed Leydig cells, the main intratesticular source of Kiss1 and the main target for KS signaling [see (8,45) for a review].The present data revealed that the localization of Kiss1/Kiss1R in the germinal epithelium is quite scanty in the control testis and limited to a few elements in the basal and luminal compartments.
In contrast to the hypothalamus, in the testis, KP and AEA treatments got opposite effects on Kiss1R only, without any effect on Kiss1.In fact, Western blot revealed that Kp10 increased Kiss1R protein, whereas AEA, via CB1, decreased it.However, in the testis of Kp10-treated animals, both Kiss1 and Kiss1R immunoreactivity rates were significantly increased, not only in Leydig cells but also in pre-/early-meiotic stages.Conversely, the AEA treatment confined the localization of Kiss1 and Kiss1R just in pre-meiotic/ meiotic spermatocytes and drastically reduced Kiss1/Kiss1R immunoreactivity in post-meiotic stages.
Furthermore, KP increased the intratesticular protein levels of CB1 and FAAH, the main AEA hydrolyzing enzyme; no effect was observed for CB2 or NAPE-PLD, the main enzyme in AEA biosynthesis.Hence, an intragonadic reciprocal crosstalk between the KS and ECS signaling occurs, thus suggesting that KP and AEA get direct effects in the gonad independently from the endocrine route responsible for the maintenance of spermatogenesis.The intratesticular localization of Kiss1, Kiss1R, CB1, and CB2 in the control and treated animals was confirmed by Western blot data, revealing that the KS and ECS systems are deeply involved in the intratesticular paracrine and autocrine communications between somatic cells and germ cells.Furthermore, KS had the ability to switch on a self-modulatory loop in both interstitial and germinal compartments to drive the progression of spermatogenesis and, in parallel, to direct AEA responsivity mainly to the interstitial compartment; conversely AEA switched off KS in postmeiotic stages and interstitium, directing Kiss1R signaling toward pre-/ early-meiotic stages.
Lastly, in this study, we included the expression analysis of miRNAs related to the post-natal maturation of the brain and testis (i.e., miR132, miR145, let-7a, and let7-b) and the epigenetic modulation of pubertal transition (56,61).In physiological conditions, miR132, miR145, and let7-b age-dependently increased from late infantile to puberty and reached the highest expression levels in the hypothalamus of adult rats (56).In contrast, in the testis, the expression levels of let7a, let7-b, miR132, and miR145 decreased from the neonatal-infantile stages to the transition from puberty to adulthood (61).Interestingly, KP and AEA via CB1 differently regulated puberty-related miRNAs in the brain and testis.In fact, this study revealed that the hypothalamic expression of miR145 only resulted in a significant increase following AEA treatment; conversely, in the testis, all the analyzed miRNAs with the exception of let7b were significantly higher than controls in KPtreated animals and were significantly lower in the AEA-treated group.This point gets particular interest for further studies in the field.In fact, miR-132 modulates ovarian and adrenal steroidogenesis (84) and promotes estradiol synthesis in granulosa cells (85), and its expression decreases in Sertoli cells following heat stress, a condition that heavily damages spermatogenesis (86).Furthermore, the increased expression of miR145, notably related to the progression of meiosis (61), parallels the increased levels of Lin28A protein, a marker of spermatogonial proliferation (81), in the testis of KP-treated animals.Consistently, no effect was observed on let7b, a testicular marker of meiotic and early post-meiotic stages (61), in all experimental conditions and in both the brain and testis.
In conclusion, in the present study, we demonstrated that in both the hypothalamus and testis of adolescent rats, AEA modulates KS at the protein level, and that in the testis, KP affects CB1 and FAAH, the main regulator of AEA tone.A KP involvement in the progression of spermatogenesis is also suggested.A schematic representation of the main outcomes and functional interplay of KS and ECS stimulations along the HPG axis is reported in Figure 10.
Further investigations are certainly needed in the near future to clarify some points.The first limitation of this study concerns the real outcomes on gamete quality following the in-vivo modulation of KS/ ECS.The additional limitations are to clearly identify which effects are hypothalamus-mediated and which are direct intragonadic effects of the tested drugs, together with the definition of the signaling pathway in KS/ECS crosstalk.In this respect, ex-vivo treatments of mediobasal hypothalamus and testis with specific agonists and the corresponding antagonists or the treatments with kisspeptin agonist and CB1 antagonists and vice versa have to be taken into consideration in the next studies.Nevertheless, the main strength of this work is the determination of a crosstalk between the two systems under examination, not only at the central level, but also for the first time in mammals, at the gonadal level.Taking into consideration the possible clinical of KS and ECS in the treatment of infertility and in-vitro fertilization, the present data provide new molecular mechanisms and targets in the intricate communication network controlling male reproduction at the central and peripheral levels.

FIGURE 3 2
FIGURE 3 Morphological evaluation of the lateral ventricle.Representative images (n = 4 for each animal group) of Nissl staining showing the lateral ventricle of the control and treated animals.C, control; KP, animals injected with Kp10; AEA, animals injected with AEA; SR + AEA, animals first received SR141716A and 30 min later received AEA treatment.HP, hippocampus; LV, lateral ventricle; PL, choroid plexus.Magnification ×4.**p< 0.01.
(A) and normalization of 3bHsd (B) and Cyp19 (C) signals carried out against a-tubulin.Quantitative expression of 3bHsd, Cyp19, and GnRH mRNA carried out by qPCR (D).Data are representative of n = 6 different animals and are expressed as protein OD/tubulin OD ± standard deviation or normalized fold expression (n.f.e.) ± SEM against the housekeeping genes HPRT/b-actin and the control group used as a reference sample.C, control group (placebo-injected animals); KP, animals injected with Kp10; *p< 0.05.

6
FIGURE 6 Modulation of the KS by KP (A-C) and AEA (D-F) in rat testis.Representative Western blots for Kiss1R and Kiss1 (A, D) and normalization of Kiss1R (B, E) and Kiss1 signals (C, F) carried out against a-tubulin.N = 6 different animals in KP treatments and n = 4 in AEA treatments; data are expressed as protein OD/tubulin OD ± standard deviation.C, control group (placebo-injected animals); KP, animals injected with Kp10; AEA, animals injected with AEA; SR + AEA, animals first received SR141716A and 30 min later received AEA treatment.*p< 0.05.

8
FIGURE 8 Modulation of the ECS by KP (A-D) and AEA (E-G) in rat testis.Representative Western blots and normalization of the signals carried out against atubulin.n = 6 different animals in KP treatments and n = 4 in AEA treatments.Data are expressed as protein OD/tubulin OD ± standard deviation.C, control group (placebo-injected animals); KP, animals injected with Kp10; AEA, animals injected with AEA; SR + AEA, animals first received SR141716A and 30 min later received AEA treatment.*p< 0.05.

TABLE 1
Primary and secondary antisera used for Western blot, IFL, and IHC.

TABLE 2
Morphometric parameters of the treated animals at sacrifice.