Effect of relaxin‐3 on Kiss‐1, gonadotropin‐releasing hormone, and gonadotropin subunit gene expression

Abstract Purpose Relaxin‐3 is a hypothalamic neuropeptide that belongs to the insulin superfamily. We examined whether relaxin‐3 could affect hypothalamic Kiss‐1, gonadotropin‐releasing hormone (GnRH), and pituitary gonadotropin subunit gene expression. Methods Mouse hypothalamic cell models, mHypoA‐50 (originated from the hypothalamic anteroventral periventricular region), mHypoA‐55 (originated from arcuate nucleus), and GT1‐7, and the mouse pituitary gonadotroph LβT2 were used. Expression of Kiss‐1, GnRH, and luteinizing hormone (LH)/follicle‐stimulating hormone (FSH) β‐subunits was determined after stimulation with relaxin‐3. Results RXFP3, a principle relaxin‐3 receptor, was expressed in these cell models. In mHypoA‐50 cells, relaxin‐3 did not exert a significant effect on Kiss‐1 expression. In contrast, the Kiss‐1 gene in mHypoA‐55 was significantly increased by 1 nmol/L relaxin‐3. These cells also express GnRH mRNA, and its expression was significantly stimulated by relaxin‐3. In GT1‐7 cells, relaxin‐3 significantly upregulated Kiss‐1 expression; however, GnRH mRNA expression in GT1‐7 cells was not altered. In primary cultures of fetal rat neuronal cells, 100 nmol/L relaxin‐3 significantly increased GnRH expression. In pituitary gonadotroph LβT2, both LHβ‐ and FSHβ‐subunit were significantly increased by 1 nmol/L relaxin‐3. Conclusions Our findings suggest that relaxin‐3 exerts its effect by modulating the expression of Kiss‐1, GnRH, and gonadotropin subunits, all of which are part of the hypothalamic‐pituitary‐gonadal axis.


| INTRODUC TI ON
It is well established that female reproduction is under the dynamic control of hypothalamic gonadotropin-releasing hormone (GnRH) and pituitary gonadotropin secretion via regulatory feedback loops mediated by sex steroid hormones within the hypothalamic-pituitarygonadal (HPG) axis. Hypothalamic kisspeptin, which is encoded by the Kiss-1 gene, is currently believed to be positioned at the highest level in the HPG axis and controls GnRH. 1 GnRH released into portal circulation, in turn, reaches the anterior pituitary and stimulates the synthesis and release of the gonadotropins luteinizing hormone (LH) and follicle-stimulating hormone (FSH). Luteinizing hormone and FSH individually or cooperatively act on gonads to regulate gametogenesis and steroidogenesis. In mammals, Kiss-1 neurons, which regulate GnRH, are located in two different areas of the hypothalamus, the anteroventral periventricular (AVPV) and arcuate nucleus (ARC). 2 At present, it is speculated that Kiss-1 neurons in the AVPV region are responsible for the estradiol-induced (E2-induced) GnRH-LH surge (positive feedback), whereas those in the ARC region are involved in an E2-induced negative feedback mechanism. This concept is based on the observations that Kiss-1 expression in the AVPV region is upregulated by E2, whereas that in the ARC is inhibited by E2. 3 Relaxin-3 is a hypothalamic neuropeptide that belongs to the insulin superfamily. 4 Prior to its discovery, it was known that there are two genes encoding relaxin-1 and -2. 5,6 Unlike relaxin-1 and -2, relaxin-3 is predominantly expressed in the brain, with particularly strong expression in the pontine nucleus incertus (NI). 4 Classically, relaxin-1 and -2 have been known to be produced by the ovary and target the mammalian reproductive system to ripen the uterine cervix, elongate the pubic symphysis, and inhibit uterine contraction. They also have roles in enhancing sperm motility and regulating blood pressure. 7 As for the characteristics of relaxin-3, it seems to have orexigenic effects; however, the details of the physiological role of relaxin-3 are largely unknown. Previous literature reported that chronic intracerebroventricular injection of relaxin-3 induced hyperphagia and increased body weight and fat mass in male rats. 8 Repeated relaxin-3 administration has also been reported to increase cumulative food intake with an increase in leptin serum level in male rats. 9 Chronic stress and repeated food restriction have been reported to increase body weight in female rats, which was associated with a significant increase in relaxin-3 mRNA levels in the brain. 10 Relaxin-3 is expressed in the NI of the brain stem, which has projections to many hypothalamic areas, and it has also been found via immunostaining in the paraventricular nucleus (PVN), ARC, and preoptic area (POA), all of which are areas known to be important in the regulation of the HPG axis. 11 In 2008, McGowan et al demonstrated that intracerebroventricular administration of relaxin-3 to adult male rats significantly increased plasma levels of LH. Because this effect was blocked by pretreatment with a peripheral GnRH agonist, they concluded that relaxin-3 stimulates the reproductive axis by stimulating GnRH neurons. Using hypothalamic explants and a mouse GnRH-producing hypothalamic cell model, namely GT1-7 cells, they demonstrated that relaxin-3 stimulates the release of GnRH. 12 Relaxin-3 binds mainly to the receptor RXFP3, 13 but it also binds and activates the receptor RXFP1. 14 These receptors are expressed predominantly in the central nervous system, particularly in the PVN, POA, and ARC regions of the hypothalamus. [15][16][17] These areas are known to play an important role in the regulation of appetite and/or the HPG axis. Although relaxin-3 binds to both RXFP3 and RXFP1, RXFP1 does not participate in appetite control in rats. 8 RXFP3 exerts its receptor effects by inhibiting cAMP, normally associated with Gi-protein coupling, 15 whereas RXFP1 is thought to mediate Gs-protein-dependent modulation of cAMP production. 18 In the present study, we focused on the functional role of relaxin-3/ RXFP3 in the components of the HPG axis by investigating the direct effect of relaxin-3 on Kiss-1, GnRH, and pituitary gonadotropin subunit expression using hypothalamic neuronal cell models and a pituitary gonadotroph cell model.

| Materials and cell models
The following chemicals and reagents were obtained from the indi-

| Cell culture and stimulation
Cells were plated in 35-mm tissue culture dishes and incubated with high-glucose DMEM containing 10% heat-inactivated FBS and 1% penicillin-streptomycin at 37°C under a humidified atmosphere of 5% CO 2 in air. After 24 hours, cells were used for each experiment.
While stimulated with the test reagents, cells were incubated with or without (control) the test reagents in high-glucose DMEM containing 1% heat-inactivated FBS and 1% penicillin-streptomycin for the indicated concentrations and time periods.

| Western blot analysis
Cell extracts were lysed on ice with RIPA buffer (phosphate-buffered saline, 1% NP-40, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate [SDS]) containing 0.1 mg/mL phenylmethyl sulfonyl fluoride, 30 mg/mL aprotinin, and 1 mmol/L sodium orthovanadate, scraped for 20 seconds, and centrifuged at 14 000 g for 10 minutes at 4°C. Protein concentration in the cell lysates was measured using the Bradford method. Denatured protein (30 µg per well) was resolved in 10% SDS polyacrylamide gel electrophoresis (SDS-PAGE) gels according to standard protocols. Protein was then transferred onto polyvinylidene difluoride membranes (Hybond-P PVDF; Amersham Biosciences), which were blocked for 2 hours at room temperature in Blotto (5% milk in Tris-buffered saline). Membranes were incubated with anti-RXFP3 antibody (1:100 dilution; Santa Cruz Biotechnology, Inc) in Blotto overnight at 4°C and washed three times for 10 minutes per wash with Tris-buffered saline/1% Tween. Subsequent incubation with horseradish peroxidase-conjugated (HRP-conjugated) antibodies was performed for 1 hour at room temperature in Blotto, and additional washes were performed as needed. Following enhanced chemiluminescence detection (Amersham Biosciences), membranes were exposed to X-ray film (Fujifilm). Tissues from rat cerebral cortex were used as positive controls.

| RNA preparation, reverse transcription, PCR, and quantitative real-time PCR
Total RNA from stimulated cells was extracted using TRIzol-LS Thermal cycling conditions were as follows: 10 minutes denaturation at 95°C, followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. Reactions were followed by melting curve analysis (55-95°C). To determine PCR efficiency, a 10-fold serial dilution of cDNA was performed as previously described. 19 PCR conditions were optimized to generate >95% PCR efficiency, and only those reactions with between 95% and 105% efficiency were included in subsequent analyses. Relative differences in cDNA concentration between baseline and experimental conditions were then calculated using the comparative threshold cycle (Ct) method. 20 Briefly, for each sample, a ΔCt was calculated to normalize to the internal control using the following equation: ΔCt = ΔCt(gene) -Ct (GAPDH). To obtain differences between experimental and control conditions, ΔΔCt was calculated as ΔCt(sample) -ΔCt(control). Relative mRNA levels were then calculated using the following equation: fold difference = 2 ΔΔCt .

| Statistical analysis
All experiments were repeated independently at least three times. F I G U R E 1 Expression of the relaxin-3 receptor RXFP3 in the hypothalamic and pituitary gonadotroph cell models. A, Total RNA was prepared, and RT-PCR was carried out for 40 cycles using Rcfp3-specific primers. PCR products were resolved in a 1.5% agarose gel and visualized with ethidium bromide staining. cDNAs from COS7 cells were used as the negative control. B, Cell lysates (30 μg protein) from the model cells were analyzed by SDS-PAGE followed by immunoblotting and incubation with an antibody against RXFP3. The bands were visualized using an HRP-conjugated secondary antibody

| Expression of relaxin-3 receptor RXFP3 in hypothalamic and pituitary gonadotroph cell models
As neuronal cell models from the hypothalamus, we used mHy-poA-50 and mHypoA-55 cells. These cells originate from the AVPV and ARC regions of the hypothalamus, and respond differently to E2. 21 We also used GT1-7 hypothalamic neurons and the pituitary gonadotroph cell model LβT2. RT-PCR using specific primers for the relaxin-3 receptor RXFP3 demonstrated that all of these cells express the RXFP3 gene ( Figure 1A). Western blotting analysis using anti-RXFP3 antibody revealed that RXFP3 protein was also expressed in these cell models ( Figure 1B). Extracts or cDNAs from rat brain tissue were used as positive controls.

| Effect of relaxin-3 on GnRH gene expression in primary cultures of fetal rat neuronal cells
To further examine the effect of relaxin-3 on GnRH mRNA expression, we used primary cultures of neuronal cells from fetal rat brain. These cells contain GnRH-expressing neurons. Similar to our observations in mHypoA-50 and mHypoA-55 cells, relaxin-3 treatment increased GnRH mRNA expression in fetal rat neuronal cells 1.84 ± 0.35-fold at 1 nmol/L and 2.41 ± 0.35-fold at 100 nmol/L ( Figure 5).

| Effect of relaxin-3 on the expression of gonadotropin LHβ-and FSHβ-subunits
Lastly, we examined the effect of relaxin-3 on the expression of pituitary gonadotropin LHβ-and FSHβ-subunits using a gonadotroph cell model. These subunits determine the specificity of LH and FSH by binding the α-subunit, which is common to both LH and FSH.
Our results show that relaxin-3 also has an effect on gonadotropin subunit gene expression. In the mouse gonadotroph cell model LβT2, 1 nmol/L relaxin-3 increased LHβ-subunit mRNA expression 1.73 ± 0.34-fold, which was statistically significant ( Figure 6A).

| D ISCUSS I ON
Energy balance and reproduction are intrinsically linked, and a number of food-intake-related hormones such as leptin and ghrelin have been reported to be involved in controlling the reproductive axis. 24 Neuropeptide Y and galanin-like peptide can also modulate the HPG axis. 25,26 McGowan et al 8,9 first reported that central administration is upregulated by E2. 21 Interestingly, although both mHypoA-50 and mHypoA-55 were first developed as Kiss-1-producing neurons, these cells also express GnRH. 27 Similarly, another hypothalamic cell model, GT1-7, was found to not only express GnRH, but also to express and secrete Kiss-1/kisspeptin. 23,28 Using these cell models, we found that relaxin-3 could be a potent stimulator of are also present in the pituitary gland. 31 We also demonstrated that all of the Kiss-1-, GnRH-, and gonadotropin-subunit-expressing cell models used in this study express the relaxin-3 receptor RXFP3 and responded to relaxin-3.
There are several limitations of this study. We did not examine It is plausible that relaxin-3 is an orexigenic peptide because it induced hyperphagia whereas its antagonist induced hypophagia.
Although one report showed that chronic stress and repeated food restriction increased body weight, which was associated with a significant increase in relaxin-3 mRNA levels in female rat brain, 10 no changes in the expression of relaxin-3 mRNA or its receptors in satiated and starved animals have been reported to date. If relaxin-3 plays a physiological role in controlling appetite, one might expect increased relaxin-3 expression during food restriction and decreased in the satiated situation. However, the pulsatility of GnRH/ gonadotropin secretion is known to be suppressed by starvation. 32 Because the physiology of this peptide is still largely obscure, it would be interesting to study this peptide in starved animals. A recently published clinical study assessing serum levels of relaxin-3 in patients with delayed puberty revealed that relaxin-3 levels were significantly positively correlated with LH, FSH, and sex steroids. 33 Because relaxin-3 could stimulate the expression of Kiss-1, GnRH, and pituitary gonadotropins by itself, it may be involved in coordinating feeding or energy balance with reproduction.

ACK N OWLED G M ENTS
This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to HK and AO).

CO N FLI C T O F I NTE R E S T
All authors declare that they have no conflict of interest.

H U M A N A N D A N I M A L R I G HT S
This study's protocol was approved by the committee of the