Mouse Testicular Mkrn3 Expression Is Primarily Interstitial, Increases Peripubertally, and Is Responsive to LH/hCG

Abstract

Studies in humans and mice support a role for Makorin RING finger protein 3 (MKRN3) as an inhibitor of gonadotropin-releasing hormone (GnRH) secretion prepubertally, and its loss of function is the most common genetic cause of central precocious puberty in humans. Studies have shown that the gonads can synthesize neuropeptides and express MKRN3/Mkrn3 mRNA. Therefore, we aimed to investigate the spatiotemporal expression pattern of Mkrn3 in gonads during sexual development, and its potential regulation in the functional testicular compartments by gonadotropins. Mkrn3 mRNA was detected in testes and ovaries of wild-type mice at all ages evaluated, with a sexually dimorphic expression pattern between male and female gonads. Mkrn3 expression was highest peripubertally in the testes, whereas it was lower peripubertally than prepubertally in the ovaries. Mkrn3 is expressed primarily in the interstitial compartment of the testes but was also detected at low levels in the seminiferous tubules. In vitro studies demonstrated that Mkrn3 mRNA levels increased in human chorionic gonadotropin (hCG)–treated Leydig cell primary cultures. Acute administration of a GnRH agonist in adult mice increased Mkrn3 expression in testes, whereas inhibition of the hypothalamic–pituitary–gonadal axis by chronic administration of GnRH agonist had the opposite effect. Finally, we found that hCG increased Mkrn3 mRNA levels in a dose-dependent manner. Taken together, our developmental expression analyses, in vitro and in vivo studies show that Mkrn3 is expressed in the testes, predominantly in the interstitial compartment, and that Mkrn3 expression increases after puberty and is responsive to luteinizing hormone/hCG stimulation.

Gonadotropin-releasing hormone (GnRH), produced by specialized hypothalamic neurons in the preoptic area of the hypothalamus, has long been recognized as the direct regulator of the synthesis and secretion of the pituitary gonadotropins, follicle-stimulating hormone (FSH) and luteinizing hormone (LH). To date, several neuropeptides and other factors have been identified to play critical roles in the upstream regulation of GnRH secretion. Among them, kisspeptin (Kiss1, encoded by the Kiss1 gene) is a decisive stimulatory factor for the induction of pubertal development and, together with neurokinin B (Nkb, encoded by Tac2) and dynorphin A (Dyn, encoded by Pdyn), constitutes the most important circuit to drive pulsatile GnRH secretion, which then enters the pituitary portal circulation. Kiss1, Nkb, and Dyn are all secreted by a specialized neuronal population, known as KNDy neurons, located upstream of GnRH neurons in the arcuate hypothalamic nucleus in mice. Additional kisspeptin neurons, located in the anteroventral periventricular hypothalamic nucleus, mediate positive ovarian steroid feedback to induce the preovulatory LH peak in females (1). Furthermore, a hypothalamic neuropeptide with inhibitory effects on the secretion of gonadotropins induced by GnRH was identified in birds, named gonadotropin-inhibiting hormone, or RFamide-related peptide 3 (Rfrp3) in mammals. However, controversy remains regarding the role of Rfrp3 in mammals, including humans (2, 3).

In the last decade, the understanding of pubertal regulation has advanced with the discovery of the inhibitory role of MKRN3 (Makorin RING finger protein 3) on the prepubertal regulation of GnRH neurons in humans. Through exome sequencing analysis in individuals with familial central (GnRH dependent) precocious puberty, inactivating mutations in the MKRN3 gene were discovered in association with early pubertal development in humans (4). MKRN3 is located on the long arm of chromosome 15, in the Prader–Willi syndrome critical region and is maternally imprinted; therefore, only individuals carrying a mutated paternally inherited allele manifest the phenotype. The proteins of the MAKORIN family have 2 or 3 zinc finger domains: 2 C3H domains at the N-terminus, followed by a Cys-His configuration, a RING zinc finger (C3HC4), and a final C3H domain near the C terminus. The C3H zinc finger regions have been implicated in RNA binding, while RING zinc finger domains are E3 ubiquitin ligase domains. Proteins of the MAKORIN family are conserved among species, with high expression in the developing central nervous system. In the rodent hypothalamus, Mkrn3 expression is high until the second week of postnatal development, and then decreases sharply before puberty initiation, supporting the hypothesis that Mkrn3 acts as an inhibitor of GnRH in the hypothalamus prior to pubertal development (4). MKRN3 is ubiquitously expressed in adult human tissues, with the highest expression levels in adults in the testes (5). However, its function and mechanisms of action in the testes remain unclear.

The central role of hypothalamic neuropeptides (Mkrn3, Kiss1, Nkb, and Dyn) in the control of pubertal development and reproductive function is well-characterized. However, the roles of peripherally acting neuropeptides in the paracrine regulation of gametogenesis and/or in the production of gonadal steroids remain less explored. Similarly, the pathways mediating feedback of peripheral signals at the level of the hypothalamus are to date only partially elucidated. Studies in rodents, in nonhuman primates, and in humans have shown that the gonads also have the ability to synthesize neuropeptides, including GnRH and Rfrp3 (6), Kiss1 and its receptor (7-12), Nkb and its receptor (13, 14), and MKRN3/Mkrn3 mRNA (5, 15). However, the effects of these neuropeptides on gonadal function, and how their expression is controlled in the gonads by gonadotropins, have not yet been fully elucidated.

The gonadotropins FSH and LH are essential regulators of ovarian and testicular function. In males, LH stimulates the production of testosterone by Leydig cells, which in turn exerts sexual and anabolic actions, and in addition participates in the maintenance of spermatogenesis through its paracrine action on Sertoli cells. FSH, in turn, stimulates the proliferation of Sertoli cells in the prepubertal period, which play an important role in endocrine and paracrine control of spermatogenesis. Leydig and Sertoli cells are located in 2 different testicular compartments, the interstitium and seminiferous tubules. Despite the anatomical separation, these 2 cell groups interact with each other so that sperm production occurs through the processes of steroidogenesis and gametogenesis (16, 17).

Considering the crucial role of Mkrn3 in the pubertal regulation of the hypothalamic–pituitary–gonadal (HPG) axis and its high expression in human adult testes, we aimed to investigate the temporospatial expression pattern of Mkrn3 in gonads of mice across different stages of reproductive development. Since our results indicated a sex-specific pattern of Mkrn3 expression predominantly in the male gonad, we expanded our studies to further investigate its expression in Sertoli and Leydig cells and its response to selective gonadotropic stimulation.

Material and Methods

Reagents

Human chorionic gonadotropin (hCG) (Choriomon-M) was purchased from Meizler UCB Biopharma S.A. FSH (Gonal-F) was purchased from Merck Serono. GnRH agonist (Triptorelin) was purchased from Neo Decapeptyl, Aché. Chemiluminescence reagent (WesternBright ECL-spray) was purchased from Advansta Inc. Molecular weight ladder (Kaleidoscope 250 kD, 1610375) was purchased from Bio-Rad. Reagent Kit Advia Centaur Reproductive Endocrinology Assay Luteinizing Hormone was purchased from Siemens Healthineers. All other reagents, including antibodies, were purchased either from Thermo Fisher Scientific or Sigma-Aldrich. The oligonucleotides were synthesized by Integrated DNA Technologies IDT (Table 1). The Ishikawa cell line (code, 0364) was purchased from Cell Bank of Rio de Janeiro (BCRJ), Brazil.

Table 1.

Primer sequences used in quantitative real-time polymerase chain reaction analysis

Gene	Oligonucleotide Sequence (5′-3′)	Product length (bp)	Genbank access
Mkrn3	Forward: GGAGGGGATGAGCCAGAAAG	156	NM_011746.2
	Reverse: TGCCAGTATGCGCTTGATGA
Fshr	Forward: TACACAACTGTGCATTCAACGG	150	NM_013523.3
	Reverse: TGGGCAGGGAATAGACCTTTG
Lhr	Forward: CTCGCCCGACTATCTCTCAC	77	XM_011246310.2
	Reverse: ACGACCTCATTAAGTCCCCTG
Cyp17a1	Forward: GCCCAAGTCAAAGACACCTAAT	159	NM_007809.3
	Reverse: GTACCCAGGCGAAGAGAATAGA
Amh	Forward: TGGCTGAAGTGATATGGGAGC	191	NM_007445.2
	Reverse: TAGCACCAAATAGCGGGTGTC
Inhb-b	Forward: CTTCGTCTCTAATGAAGGCAACC	166	NM_008381.4
	Reverse: CTCCACCACATTCCACCTGTC
Wt1	Forward: GGGTCCTCGTGTTTGAAGGAA	128	NM_144783.2
	Reverse: GAGAGCCAGCCTACCATCC

Study Approval and Mouse Models

The mouse studies were approved by the Ethics Committee on Animal Use of the University of Brasilia (Approval Number: 118455/2014). Mice were housed in the animal facility of the School of Health Sciences, University of Brasilia, 5 animals per cage, at 25 °C, with a 12-hour light/dark cycle and free access to standard mouse chow diet and water. Wild-type C57BL/6 male and female mice on postnatal days (PNDs) 7, 14, 21 (prepubertal), 28, 35, 42 (peripubertal), 84 (sexual maturity), and 300 (senescence) were used to investigate the temporal pattern of Mkrn3/Mkrn3 mRNA/protein expression in testes, ovaries, and mediobasal hypothalamus (MBH). Mkrn3 knockout mice, generated by selective deletion of only the paternal allele (Mkrn3^+/−) or by biallelic deletion (Mkrn3^−/−) were used as controls to validate our Mkrn3 protein expression assay (18). These Mkrn3^+/− mice have the LacZ coding sequence inserted in place of the coding sequence of Mkrn3, through homologous recombination in embryonic stem cells, such that β-galactosidase protein replaces Mkrn3 protein expression. Adult male Mkrn3^+/− mice were used to determine spatial localization of Mkrn3 in the testis through X-gal staining (18). Other cohorts of peripubertal (PND 35) and adult wild-type C57BL/6 male mice were used for in vitro and in vivo experimental designs, respectively, to investigate whether Mkrn3 expression is modified in either of the 2 functional testicular compartments by selective gonadotropic stimulation.

Primary Sertoli and Leydig Cell Isolation and Characterization

Primary cultures of testicular Sertoli and Leydig cells were established according to previously described methods (19, 20) with some modifications. Briefly, testes of PND 35 mice were excised and decapsulated in Dulbecco's modified Eagle's Medium/Nutrient Mixture F-12 Ham culture medium supplemented with L-glutamine, HEPES 15 mM, gentamicin 0.02 g/L, and DNase (25 µg/mL). Testes were then cut into small fragments and digested by collagenase (0.05 mg/mL) plus 1% bovine serum albumin and 25 µg/mL DNase at 37 °C with shaking for 15 minutes. The solutions were settled for 2 minutes, and the supernatant was transferred to a new tube. Primary Leydig cells were pelleted from the supernatant by centrifugation at 2000 rpm for 5 minutes and then resuspended, plated and cultured in Dulbecco's modified Eagle's Medium/Nutrient Mixture F-12 Ham culture medium, with L-glutamine, 15 mM HEPES, and 0.02 g/L gentamicin supplemented with 10% fetal bovine serum (FBS), in a humidified atmosphere of 5% CO₂:95% air at 37 °C. The seminiferous tubules from the original tube were digested in hyaluronidase (0.05 mg/mL) plus 1% bovine serum albumin and 25 µg/mL DNase at 37 °C with shaking for 30 minutes. After several centrifugations at 1000 rpm for 2 minutes, the primary Sertoli cells were plated and culture in the same medium and conditions used for Leydig cells. In both primary Sertoli and Leydig cultures the media were changed after 24 hours to remove the remaining germ cells and nonadherent cells. After 7 days of culture the predominance of Sertoli or Leydig cells in each of the primary testicular cultures was determined by the expression of their specific genes: Fshr and Wt1 for Sertoli cells and Lhr for Leydig cells. The testicular primary culture containing primarily Sertoli cells with few Leydig cells was designated “primary Sertoli cells,” while the testicular primary culture containing primarily Leydig cells with few Sertoli cells was designated “primary Leydig cells.”

In Vitro Studies

Isolated primary Leydig and Sertoli cells were plated in a 6-well culture plate in medium supplemented with 10% FBS. After 7 days, when the cells reached 90% to 95% confluence, the primary lineages were stimulated with vehicle or hCG (0.03 IU/mL) or FSH (50 ng/mL) for 4 hours in medium without FBS. At the end of treatment, cells were harvested for total RNA isolation. All experiments were performed with cells isolated from 5 different animals.

In Vivo Studies

Adult wild-type C57BL/6 male mice were submitted to 3 treatment protocols with a GnRH agonist (Triptorelin) to induce different stimulation conditions of the HPG axis: (1) first, to induce acute HPG axis stimulation, animals were treated with a single intramuscular injection of vehicle (Treatment 1: V; n = 5) or GnRH agonist (Triptorelin) 5 µg/mouse (Treatment 1: GnRHa; n = 5); (2) second, to induce inhibition of the HPG axis, a separate cohort of male mice were treated for 15 days with daily intramuscular injections of vehicle (Treatment 2: V; n = 5) or GnRH agonist (Triptorelin) 5 µg/mouse/day (Treatment 2: 15-day GnRHa; n = 6); (3) finally, a third group of male mice received 5 µg GnRH agonist/mouse/day for 15 days, followed by the injection of a single dose of vehicle (intramuscular; Treatment 3: V; n = 5), hCG 10 IU (intramuscular; Treatment 3: hCG; n = 6), or FSH 4 IU (subcutaneous; Treatment 3: FSH; n = 7). The animals were euthanized 2 hours following the injections from each protocol described above, testes were collected, snap-frozen on liquid nitrogen and stored at −80 °C for mRNA expression analysis.

RNA Isolation, cDNA Synthesis, and Quantitative Real-Time Polymerase Chain Reaction Analysis

Total RNA was extracted from male and female gonads, MBH, and primary Sertoli and Leydig cells, using TRIzol reagent following the manufacturer's protocol. RNA was subsequently treated with DNase I to avoid genomic DNA contamination. RNA amount and purity were measured using a spectrophotometer (NanoVue Plus, GE Healthcare Life Sciences). Reverse transcription and quantitative real-time polymerase chain reaction (qPCR) were carried out with 5 ng of total RNA using Power SYBR Green RNA to CT 1-Step kit, along with (5 pmol/μL) primers, and were performed in an AB7500 PCR machine. Relative mRNA levels were measured using Gapdh as a reference gene and calculated by the comparative threshold cycle (Ct) method, according to the formula 2^−ΔΔCt (21). Primer sequences used are shown in Table 1.

Western Blot Analysis

Total proteins were extracted from testes of wild-type, Mkrn3^+/− or Mkrn3^−/− mice using RIPA buffer. After quantification, 10 µg of soluble proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a 10% polyacrylamide gel under reducing conditions at 4 °C and constant voltage (100 V). Proteins were then electrotransferred for 90 minutes at 4 °C and 160 mA onto polyvinylidene fluoride membranes. Nonspecific binding was blocked with 5% nonfat dry milk diluted in Tris-buffered saline with 0.05% of Tween-20 (TBS-T) for 2 hours at room temperature on a platform rocker. The blots were then incubated with a rabbit anti-MKRN3 antibody (1:1000, HPA029494, RRID:AB_10603873) overnight at 4 °C with gentle shaking. Subsequently, blots were rinsed with TBS-T and incubated with anti-rabbit secondary antibody linked to horseradish peroxidase for 1 hour and rinsed again in TBS-T. The antibody–antigen complexes were visualized and enhanced using a chemiluminescence reagent in a gel imaging system (Gel Doc XR + Imaging System, Bio-Rad). Finally, for normalization purpose, blots were rinsed in a stripping solution (Restore Western Blot Stripping Buffer, 21059) and re-probed with mouse anti-beta actin antibody (1:10,000, A3854, RRID:AB_262011). Molecular mass was approximated using a kaleidoscope 250 kD molecular weight ladder.

X-gal Staining

Testes (n = 3/group) from adult wild-type and Mkrn3^+/− mice were fixed in fresh 4% cold paraformaldehyde for 1 hour at 4 °C and then incubated 3 times, 30 minutes each, in rinse buffer (0.1 M sodium phosphate dibasic, 5 mM sodium phosphate monobasic, 3 mM MgCl₂-6H₂O, 1.5 mM sodium deoxycholate, 3% octylphenoxypolyethoxyethanol [IGEPAL, CA-630]) at 4 °C on a platform rocker. β-Galactosidase activity was detected using a LacZ staining solution (1 mg/mL X-gal [5-bromo-4-chloro-3-indolyl β-dgalactopyranoside], 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide in rinse buffer) at 37 °C overnight. Thereafter, the testes were embedded in paraffin (Harvard Medical School Rodent Pathology Core) and sectioned (7 µm). Sections were deparaffinized in xylene, hydrated in 100% and 70% ethanol, and counterstained with 1% Safranin.

Transactivation Assay and Reporter Gene

Human endometrial adenocarcinoma Ishikawa cells, previously shown to express LH/hCG receptor, were transfected with an expression vector containing the MKRN3 promoter linked to the luciferase reporter gene, or with the luciferase PGL2 empty vector as a negative control (22). Transfections were performed using lipofectamine (Lipofectamine 2000 Transfection Reagent) according to the manufacturer's instructions. Cells were exposed for 24 hours to vehicle (Medium) or different concentrations of hCG (0.015, 0.03, 0.06, 0.12, or 0.25 IU/mL) in medium without FBS. Luciferase activity was measured in a luminometer using a luciferase reporter assay kit according to the manufacturer's instructions. Results were reported as luciferase activity induced by the different compounds relative to vehicle. The experiment was repeated 3 times, each in triplicate.

LH Hormone Assay

Serum LH concentrations were measured using Reagent Kit Advia Centaur Reproductive Endocrinology Assay Luteinizing Hormone (RRID:AB_2895592), according to the manufacturer’s instructions, on the Advia Centaur XP (Siemens Healthineers).

Statistical Analysis

Data were analyzed using the GraphPad Prism 5.0 statistical package. Analyses were performed using unpaired Student's t-test or 1-way analysis of variance (ANOVA) followed by Tukey post hoc test for comparing the means of 2 or multiple groups, respectively. All data sets are presented as mean ± SEM. P < .05 was considered statistically significant.

Results

Mkrn3 Expression Pattern in Mouse Hypothalamus and Gonads

We initially validated our measurement of Mkrn3 mRNA levels by comparing prepubertal and adult expression patterns in the MBH of male and female wild-type mice. As previously reported (4), MBH Mkrn3 mRNA levels were markedly lower in the adult mice (PND 84) than prepubertal mice (PND 7) (Fig. 1A). In addition, at both ages, Mkrn3 mRNA levels were higher in the MBH of males than females (Fig. 1B).

Figure 1.

Mkrn3 mRNA levels in the hypothalamus vs gonads across the development. (A, B) Relative Mkrn3 mRNA levels in the mediobasal hypothalamus (MBH) of prepubertal (PND 7) and adult (PND 84) male and female mice. (A) Data are normalized to Mkrn3 mRNA levels at PND 7 (in both males and females). (B) Data are normalized to Mkrn3 mRNA levels in males (at PND 7 and PND 84). (C, D) Relative Mkrn3 mRNA levels in the MBH vs gonads of (C) prepubertal (PND 7) and (D) adult (PND 84) male and female mice. (C, D) Data are normalized to Mkrn3 mRNA levels in testis for males and ovary for females (at PND 7 and PND 84). Data are represented as mean ± SEM, n = 3-5/group. *P < .05, **P < .01, ***P < .001, ****P < .0001 by unpaired Student's t-test.

Next, we investigated Mkrn3 expression in the gonads compared with the MBH in both prepubertal (PND 7) and in adult, sexually mature (PND 84) mice. In prepubertal mice, Mkrn3 mRNA levels were higher in the MBH than in the gonads, both in males and females; however, this pattern was reversed in adult mice of both sexes (Fig. 1C and 1D).

Developmental Mkrn3 Gene Expression and Protein Levels in the Gonads of Mice

We next examined the expression pattern of Mkrn3 in the gonads of both male and female mice across different stages of postnatal development. Quantification by reverse transcription qPCR showed that Mkrn3 mRNA was detected in testes and ovaries of wild-type mice at all ages evaluated. However, the pattern of Mkrn3 gene expression across postnatal development differed between male and female gonads. In the testes, we observed a significant increase in Mkrn3 mRNA levels at PND 14 to PND 28 compared to PND 7, which then remained relatively stable at those higher levels in adulthood. In contrast, in the ovaries, Mkrn3 mRNA levels were lower on PND 21 and PND 28 than PND 14, and then remained unchanged between PND 28 and PND 300. Notably, Mkrn3 mRNA levels increased from PND 7 to PND 28 in the testes, whereas they decreased across these same postnatal ages in the ovaries (Fig. 2A and 2B).

Figure 2.

Comparative developmental Mkrn3 expression pattern in the mouse mediobasal hypothalamus (MBH), testis and ovary. (A, B) Relative Mkrn3 mRNA levels in mouse (A) testes and (B) ovaries at ages ranging from postnatal day (PND) 7 to PND 300. The phase of the estrous cycle was not monitored for sexually mature female mice at PND 84 and 300. (A) Data are normalized to Mkrn3 mRNA levels in the testis at PND 7. (B) Data are normalized to Mkrn3 mRNA levels in the ovary at PND 7. Data represent mean ± SEM, n = 5/group. Groups with different letters differ significantly from 1 another (P < .05) by 1-way ANOVA/Tukey. (C) Western blot showing Mkrn3 protein levels (black arrows, upper panel) in the mouse MBH and testes relative to β-actin protein levels (black arrows, lower panel). Nonspecific bands are indicated by red arrows. The Mkrn3 and β-actin protein sizes were determined based on size markers run on the same gel. Mkrn3^+/+ mice were compared to Mkrn3^+/− or Mkrn3^−/− mice at each age analyzed; n = 1. 56 kDa, expected Mkrn3 protein size; 42 kDa, expected β-actin protein size; Mkrn3^+/+, wild-type mice; Mkrn3^+/−, mice with deletion of the paternally inherited Mkrn3 allele; Mkrn3^−/−, biallelic Mkrn3 knockout mice.

Interestingly, Mkrn3 protein was detected only in the testes, but not in the ovaries. Mkrn3 protein levels followed a similar developmental pattern to that of Mkrn3 mRNA levels in the testes, with increased levels peripubertally (PND 28), then remaining high in adulthood (Fig. 2C). Of note, in contrast with the absence of detectable Mkrn3 protein by Western blot analysis in the MBH of mice with selective deletion of the paternally inherited Mkrn3 allele at PND 28 (Mkrn3^+/− knockout mice), testes from the same mice showed some remaining Mkrn3 protein at this age, suggesting incomplete imprinting/silencing of the maternal allele in the male gonad, in contrast to the complete imprinting pattern observed in the MBH. Indeed, total absence of Mkrn3 protein in the testis was observed only in biallelic knockout mice (Mkrn3^−/−), as seen at PND 100 (Fig. 2C).

Mkrn3 Localization in Testicular Compartments

Since we identified Mkrn3 protein only in the adult mouse testis, but not in the ovary, we next investigated the anatomical distribution of Mkrn3 in different testicular cell compartments. X-gal staining of testis sections from adult Mkrn3^+/− mice, in which β-galactosidase protein replaces Mkrn3 protein expression, showed the presence of LacZ (blue dots) primarily in the interstitial compartment, while it was also detected at lower levels in the seminiferous tubules, but not detected in sections from wild-type testes used as negative controls (Fig. 3).

Figure 3.

Mkrn3 expression pattern in adult mouse testis. (A, B) X-gal staining of testis sections from wild-type mice showing absence of LacZ (negative controls), and (C, D) from Mkrn3^+/− mice (n = 3/group). In the latter, β-galactosidase replaces Mkrn3 protein expression, which in turn results in detection of LacZ (blue dots) in the interstitial compartment (black arrows) and at lower levels in the seminiferous tubules (red arrows). Representative images are shown. Scale bars: 100 µm.

In Vitro Studies: Regulation of Mkrn3 Expression in Leydig and Sertoli Cells by Gonadotropins

The finding of progressively increasing Mkrn3 mRNA and protein levels in the testis, from early postnatal stages to puberty, together with the protein distribution pattern observed in 2 functionally distinct testicular compartments, led us to hypothesize that Mkrn3 expression might be regulated by gonadotropins. To test this, we established primary Sertoli and Leydig cell cultures from PND 35 mice. The predominance of Sertoli or Leydig cells in each of the primary testicular cell cultures was determined by the expression of their specific genes: Fshr and Wt1 for Sertoli cells and Lhr for Leydig cells (Fig. 4A).

Figure 4.

Differential regulation of Mkrn3 expression in testicular cell compartments (Leydig and Sertoli primary culture cells) by gonadotropins. (A) The predominance of Sertoli or Leydig cells in each of the primary testicular cultures was determined by the expression of their respective specific genes: Fshr and Wt1 for Sertoli cells and Lhr for Leydig cells. The testicular primary culture containing primarily Sertoli cells with few Leydig cells was designated “primary Sertoli cells,” while the testicular primary culture containing primarily Leydig cells with few Sertoli cells was designated “primary Leydig cells.” ND, not detected. (B, C) Cyp17a1 and Mkrn3 mRNA levels in primary Leydig cell cultures treated with vehicle, hCG or FSH. (D, E) Amh and Mkrn3 mRNA levels in primary Sertoli cell cultures treated with vehicle, hCG or FSH. (A) Data are normalized to Fshr and Wt1 mRNA levels in Sertoli cells, and to Lhr mRNA levels in Leydig cells. (B-E) Data are normalized to mRNA levels from the vehicle (V) Group. Data are represented as mean ± SEM. *P < .05, **P < .01 by 1-way ANOVA/Tukey, n = 5/group. Figure 4A was created from Servier medical arts—https://smart.servier.com.

After characterization of the primary cell cultures, we treated each with vehicle, hCG, or FSH for 4 hours. hCG treatment induced a significant increase in Cyp17a1 mRNA levels in Leydig cells (Fig. 4B). Cyp17a1 encodes a steroidogenic enzyme (17-alpha hydroxylase) and serves as a positive control for hCG-induced Leydig cell stimulation (23). Interestingly, Mkrn3 mRNA levels also increased in hCG-treated Leydig cell cultures compared with vehicle, while FSH had no significant effect (Fig. 4C). On the other hand, FSH treatment of Sertoli cell cultures significantly reduced Mkrn3 mRNA levels (Fig. 4E), although it did not change Amh expression, the positive control for FSH action (24) in those cells (Fig. 4D).

In Vivo Studies: Regulation of Mkrn3 Expression in Testicular Compartments by Gonadotropins

Based on our experiments using primary testicular cell cultures, which suggested a differential effect of hCG and FSH in the regulation of Mkrn3 gene expression in Leydig and Sertoli cells, respectively, we next investigated the effects of HPG axis suppression and subsequent gonadal stimulation by hCG or FSH in vivo. To assess whether acute stimulation by exogenous gonadotropins would affect Mkrn3 mRNA levels in the testis in vivo, we first injected adult male mice with either acute or prolonged GnRH agonist (Triptorelin) or vehicle (Fig. 5A). As expected, acute GnRH agonist treatment led to an increase on LH levels in the serum of the mice 2 hours later (Table 2). Consistent with the in vitro studies, we observed an increase in Cyp17a1 and Mkrn3 mRNA levels (Fig. 5B and 5C) but no change in Inhb mRNA levels, an FSH-dependent marker (25) (Fig. 5D) in the testes of mice treated acutely with GnRH agonist.

Figure 5.

Upregulation and downregulation of Mkrn3 mRNA levels in the testes by acute or chronic administration of GnRH agonist, respectively. (A) Schematic representation of the experimental design. (B) Cyp17a1, (C) Mkrn3, and (D) Inhb mRNA levels in the testes of adult mice 2 hours after acute administration of GnRH agonist. (E) Cyp17a1, (F) Mkrn3, and (G) Inhb mRNA levels in the testes of adult mice after chronic (daily for 15 days) administration of GnRH agonist. (B-G) Data are normalized to mRNA levels from the vehicle (V) group. Data are represented as mean ± SEM. *P < .05, **P < .01, ****P < .0001 by unpaired Student’s t-test, n = 5-6/group. Figure 5A was created from Servier medical arts—https://smart.servier.com.

Table 2.

Serum LH levels following acute or chronic GnRH agonist (Triptorelin) treatment in adult C57BL/6 male mice

Vehicle (saline)	Acute (2 hours) GnRH agonist (5 µg/mouse)	Chronic GnRH agonist (5 µg/mouse/day, 15 days)
<0.07 mUI/mL	0.12 mUI/mL	—
<0.07 mUI/mL	—	<0.07 mUI/mL
<0.07 mUI/mL	0.08 mUI/mL	<0.07 mUI/mL
<0.07 mUI/mL	0.08 mUI/mL	<0.07 mUI/mL
<0.07 mUI/mL	0.15 UI/mL	—

—, insufficient sample for LH assay.

Abbreviations: GnRH, gonadotropin-releasing hormone; LH, luteinizing hormone.

To further verify the finding of increased Mkrn3 mRNA levels induced by an acute GnRH agonist-mediated LH peak, we tested whether suppression of LH by prolonged GnRH agonist treatment would reverse this effect. To this end, another group of male mice were injected with GnRH agonist (Triptorelin) or vehicle daily for 15 days, in attempt to downregulate the HPG axis and consequently reduce serum LH and FSH levels in those mice. In the testes, Cyp17a1 and Mkrn3 mRNA levels were decreased, as expected (Fig. 5E and 5F). Inhb mRNA levels did not change after chronic GnRH agonist treatment (Fig. 5G).

Given that our in vitro and in vivo studies suggested that gonadotropins would exert an effect on the expression of Mkrn3 mRNA in the testis, we selectively tested hCG or FSH effects on testicular Mkrn3 and control gene expression markers after inducing downregulation of the HPG axis by chronic GnRH agonist treatment, as described above. We thus injected the HPG-suppressed animals with an acute dose of vehicle, hCG, or FSH (Fig. 6A). As expected, 2 hours later, Cyp17a1 and Mkrn3 mRNA levels increased following hCG injection. Inhb mRNA levels increased after FSH stimulation, but FSH did not affect Mkrn3 mRNA levels (Fig. 6B and 6D).

Figure 6.

Increased testicular Mkrn3 mRNA levels following acute hCG administration in HPG-suppressed mice. (A) Schematic representation of the experimental design. mRNA levels of (B) Cyp17a1, (C) Mkrn3, and (D) Inhb in the testes of adult mice after acute administration of vehicle, hCG or FSH. (B-D) Data are normalized to mRNA levels from the 15-day GnRHa group. Data are represented as mean ± SEM. *P < .05, ****P < .01 by 1-way ANOVA/Tukey, n = 5-7/group. Figure 6A was created from Servier medical arts—https://smart.servier.com.

In Vitro MKRN3 Transactivation Assay

Given that Mkrn3 gene expression is responsive to LH/hCG stimulation in vivo and in vitro, we evaluated the effects of increasing concentrations of hCG on the activity of the MKRN3 gene promoter. Ishikawa cells (known to express LH/hCG receptor) were transfected with an expression vector containing the MKRN3 promoter linked to the luciferase reporter gene or with the luciferase PGL2 empty vector (22), and then exposed for 24 hours to vehicle (medium) or increasing concentrations of hCG (0.015, 0.03, 0.06, 0.12, or 0.25 IU/mL). Consistent with our previous findings, we observed that hCG treatment increased Mkrn3 promoter activity in a dose-dependent manner (Fig. 7A) compared with a PGL2-luc empty vector control (Fig. 7B).

Figure 7.

Upregulation of MKRN3 promoter activity after hCG treatment in vitro. Luciferase activity after transfection of (A) MKRN3-Luc or (B) PGL2 empty vector control in human endometrium adenocarcinoma Ishikawa cells. Cells were exposed for 24 hours to vehicle (media) or increasing concentrations of hCG (0.015, 0.03, 0.06, 0.12, or 0.25 IU/mL). In A and B, data are normalized to MKRN3 promoter activity from the vehicle (V) group. Data are shown as mean ± SEM. ****P < .01 by 1-way ANOVA/Tukey. The experiment was repeated 3 times, with each performed in triplicate.

Discussion

The role of MKRN3 as a major hypothalamic regulator of the initial activation of the HPG axis during pubertal development in humans is well-established. Studies in humans and mice support this protein as a major inhibitor of GnRH secretion during prepubertal years, and its loss of function is the most common cause of central precocious puberty in humans (26). However, its mechanisms of action and regulation at multiple levels of the HPG axis are still debated. Prior studies have demonstrated that the gonads also have the ability to synthesize hypothalamic neuropeptides involved in the control of pubertal development and reproductive function, and that Mkrn3 expression does not decrease in adult testes (5-15, 27). We therefore sought to characterize the temporospatial pattern of Mkrn3 expression in the male and female gonads across the lifespan, using mice as a model. Given the clear peak of Mkrn3 expression in the testes peripubertally (PND 28), remaining high throughout adulthood, we investigated whether Mkrn3 expression differed in the 2 functional testicular compartments, the interstitium and seminiferous tubules, and whether it is differently regulated by gonadotropins. Through a comprehensive set of in vitro and in vivo experiments, we have demonstrated that in the testes, Mkrn3 is expressed primarily in the interstitial compartment and is responsive to LH/hCG stimulation.

Initially, we validated Mkrn3 expression data in the MBH by replicating those presented by Abreu and collaborators (4) in which Mkrn3 expression was markedly reduced from the prepubertal period to adulthood. In addition, we showed that, both prepubertally and in adulthood, Mkrn3 expression was higher in the MBH of males than in females. The reasons for this sex difference are not clear but may be related to sex-specific differences in hypothalamic neuronal circuits and regulation (1). In contrast, Roberts et al did not find evidence of sexual dimorphism of mouse hypothalamic Mkrn3 mRNA expression at PND 12 and PND 30 (28).

Next, we compared the expression of Mkrn3 in the hypothalamus to that in the gonads and determined the time point at which Mkrn3 expression appears in the gonads during reproductive maturation. In the testes, distinct from the expression pattern in the mouse hypothalamus, we found a progressive increase in Mkrn3/Mkrn3 mRNA/protein levels after PND 14, reaching a peak at PND 28, and then remaining relatively high throughout adulthood. Interestingly, we observed a mild decrease in Mkrn3 mRNA levels after PND 28, but not to the levels at PND 7. Lower levels of Mkrn3 at older ages may be due to the dilution effect from isolating Mkrn3 mRNA from whole testis as the number of germ cells increase. Although distinct from the expression pattern in the hypothalamus and from the decline in circulating MKRN3 levels during puberty in both healthy boys and girls (29, 30), our finding is consistent with previous descriptions of high MKRN3 expression in the adult human testis (5). Similarly, Mkrn2 protein, another member of the makorin family, has been shown to be ubiquitously expressed at low levels in multiple tissues, but preferentially in the testis (31). Interestingly, Anjum and collaborators (6) have described increased expression of kisspeptin, Gnrh and Rfrp3 in the testis after puberty. Moreover, 2 other studies have shown that the levels of kisspeptin increase in the testis during puberty and remain high during reproductive years (10, 11).

Mkrn3 protein was localized in both testicular compartments, the interstitium and seminiferous tubules, with the highest levels in the interstitial compartment. Mkrn2 protein was also identified by immunohistochemical analysis in the mouse testis but was localized primarily in the Sertoli cells and spermatids (31). Neuropeptides, such as kisspeptin, Gnrh, and Gnih have also been localized in the Leydig cells of mouse testes (6, 10, 11). Infan and collaborators (12) immunolocalized kisspeptin in Leydig cells and its receptor in Sertoli cells of nonhuman primate testis. Despite a lack of functional evidence, altogether these data raise the possibility that hypothalamic neuropeptides and other factors regulating the activation of the reproductive axis, including Mkrn3, are specifically expressed in different gonadal cellular compartments, to potentially influence gonadal function across development. This hypothesis is further supported by our findings of increased Mkrn3 mRNA levels in vitro in cultured Leydig cells after stimulation with hCG, which was corroborated in vivo, given that hCG treatment of adult mice induced a significant increase in testicular Mkrn3 mRNA levels. Of note, Mkrn3/Mkrn3 expression has been shown to increase as adult Leydig cells, which are more responsive to LH than fetal Leydig cells, begin to populate the interstitial compartment (32, 33).

In the ovaries, in turn, Mkrn3 mRNA levels were lower than in the testes, though higher at early postnatal ages and then decreasing progressively through development and into adulthood. However, Mkrn3 protein was undetectable in the ovaries at all ages analyzed, which lead us to believe that it likely does not play a significant role in ovary function.

It is noteworthy that Mkrn3^+/− testes, with deletion of the paternally inherited allele, have some preserved expression of Mkrn3 protein at PND 28, differently from the MBH of the same mice, which display complete absence of Mkrn3 protein. As proposed by Jong and collaborators (5), this finding suggests that Mkrn3 is not completely maternally imprinted in the testis as it is in the hypothalamus. As is the case for all imprinted genes, Mkrn3 protein expression may vary, depending on the imprinting pattern in a specific tissue and thereby on the proportion of “active” vs “silenced” alleles. Therefore, it is conceivable that a tissue-specific regulated imprinting pattern may be associated with Mkrn3 expression in the male gonad.

The observation of increased Mkrn3 expression during puberty in the testis and the differential expression of this protein in the steroidogenic and gametogenic compartments led us to hypothesize that Mkrn3 expression would be differentially responsive to gonadotropins. The studies conducted herein demonstrated that the Mkrn3 mRNA levels increased in hCG-treated Leydig cells primary cultures. Moreover, the acute administration of GnRH agonist in adult wild-type mice increased Mkrn3 expression in the testes. Recent studies have also shown that hypothalamic neuropeptides, in addition to being expressed in the gonads, have their expression in the gonads controlled by gonadotropins. Salehi and collaborators (11) showed that primary Leydig cell cultures expressed increasing amounts of Kiss1 mRNA and protein after LH stimulation. Moreover, in the same study, after acute injection of a GnRH agonist in adult mice, increased testicular StAR and Kiss1 expression was noted. Unexpectedly, FSH treatment of Sertoli cell cultures significantly reduced Mkrn3 mRNA levels, although Amh expression remained unchanged. The lack of Amh induction after FSH treatment as a positive control in this experiment might represent and important limitation of our study. Although evidence from older studies has indicated that anti-Müllerian hormone (AMH) gene transcriptional activation and AMH secretion in Sertoli cells are induced by FSH, more recent studies have underscored that FSH-induced AMH secretion occurs in a complex regulatory network and is partially dependent on sex steroids (24, 34-36). Moreover, AMH secretion is influenced by Sertoli cell maturation status and other factors, such as sex steroid exposure (37-39), and there is little evidence for FSH-induced AMH secretion specifically in mouse Sertoli cells primary cultures. Therefore, the lack of an FSH effect on Amh expression may be due to insufficient in vitro stimulation of the Sertoli cells, thus impacting the interpretation of the Mkrn3 mRNA expression in this experiment. Furthermore, the lack of Amh responsiveness is likely due to the use of older Sertoli cells (derived from PND 35 mice) that express low levels of AMH in vivo and are less responsive to FSH.

Indeed, other differentially expressed regulatory proteins have shown to be selectively responsive to either hCG or FSH stimulation. For instance, expression of Lin28, an RNA binding protein involved in the control of microRNA synthesis, was rescued in both Leydig and germ cells of hypogonadal Gpr54 knockout mice after chronic hCG or FSH treatment, respectively. These data further support the idea that multiple gonadotropin-dependent factors may be involved in the functional regulation of distinct testicular compartments and that a complementary communication among these cells is hypothesized (40, 41).

The inhibition of the HPG axis through long-term (15 days) treatment with a GnRH agonist in wild-type mice also corroborated that Mkrn3 mRNA levels decrease after removing the effects of gonadotropins on the testes, consistent with the observed hCG-dependent stimulation of Mkrn3 mRNA levels in Leydig cells in vitro. Moreover, rescue of Mkrn3 expression was observed in the group that received hCG injection after HPG downregulation, further demonstrating the induction of Mkrn3 mRNA levels by hCG in Leydig cells in vivo. Finally, we showed that acute hCG stimulation increased Mkrn3 transcriptional activity in a dose-dependent manner. Although most of the reported studies describe loss of function mutations in the coding region of MKRN3, defects in the regulatory regions of the gene were described in 3 recent studies (22, 42, 43) showing that genetic alterations in the promoter and 5′-UTR regulatory regions of the MKRN3 gene cause central precocious puberty. These findings support the idea that the modulation of the regulatory region of MKRN3 is important for it function.

In summary, our studies have shown that Mkrn3 is expressed in the testes and is responsive to LH/hCG stimulation in mice. Activation or inhibition of the HPG axis in vivo, by acute or chronic administration of GnRH agonist, respectively, affected testicular Mkrn3 mRNA levels. These findings were confirmed by our in vitro studies, in which Mkrn3 mRNA levels increased in primary Leydig cell cultures in response to hCG stimulation, and further confirmed by demonstrating Mkrn3 promoter activation in a luciferase reporter assay in the hCG-responsive Ishikawa cell line (Fig. 8). Taken together, our developmental expression analyses, in vitro and in vivo studies show that Mkrn3 is produced in the testes, predominantly in the interstitial compartment, and that testicular Mkrn3 gene expression increases peripubertally and is responsive to LH/hCG stimulation. Additional studies are necessary to further address the functional consequences of these findings.

Figure 8.

Testicular Mkrn3 expression is responsive to LH/hCG stimulation in mice (schematic representation). Stimulation or inhibition of the HPG axis, by acute or chronic administration of GnRH agonist in vivo, respectively, affect Leydig cell Mkrn3 mRNA levels in the testis. Mkrn3 mRNA levels and Mkrn3 promoter activation increase in hCG-treated Leydig cell primary cultures (created from Servier medical arts—https://smart.servier.com).

Abbreviations

AMH

anti-Müllerian hormone

ANOVA

analysis of variance

FBS

fetal bovine serum

FSH

follicle-stimulating hormone

GnRH

gonadotropin-releasing hormone

hCG

human chorionic gonadotropin

HPG

hypothalamic–pituitary–gonadal

Kiss1

kisspeptin

LH

luteinizing hormone

MBH

mediobasal hypothalamus

MKRN3

Makorin RING finger protein 3

PND

postnatal day

qPCR

quantitative real-time polymerase chain reaction

Rfrp3

RFamide-related peptide 3

Funding

This work was supported by NIH grant R01 HD082314 (to U.B.K.) and by NIH grant R00 HD091381 (to A.P.A.).

Disclosures

The authors have nothing to disclose.

Data Availability

Original data generated and analyzed during this study are included in this published article or in the data repositories listed in References.