Unacylated ghrelin stimulates steroidogenesis in lean rats and reverses reproductive dysfunction in high fat diet-fed rats

ABSTRACT

This study investigated the effect of sub-chronic administration of unacylated ghrelin (UAG) on steroidogenesis, sperm parameter, and reproductive function in lean and high fat diet (HFD)-induced obese male rats. Rats were divided into 4 groups (n = 12 each) as 1) Control-fed standard diets (STD): (10 kcal%), 2) STD + UAG (200 ng/kg, i.p.), 3) HFD obese: fed HFD (45 kcal%), and 4) HFD + UAG. Diet regimen was continued for 16 weeks after which normal saline as a vehicle or UAG was administered to desired groups for the next 4 weeks. In vitro, testicular slices were incubated with increasing concentrations of UAG (10⁻⁸–10⁻⁶ M) in the presence or absence of GSH-R1a antagonist, [D-Lys-3]-GHRP-6 (10⁻⁶ M). UAG significantly increased the circulatory levels of FSH, LH and testosterone, increased testicular testosterone levels and sperm count and motility in lean and obese rats and reduced sperm morphological abnormalities and increased pregnancy rate and number of pups at birth in HFD-obese rats. Associated with the reduction in the final body and fat masses weights and independent of food intake, UAG post-therapy to both lean and HFD-fed rats significantly lowered fasting blood glucose and insulin levels, lowered HOMA-IR value, enhanced OGTT and ITT, lowered circulatory leptin levels, downregulated aromatase expression in adipose and testicular tissue and inhibited HFD-induced testicular oxidative stress and activation of cleaved caspase-3. Dysregulation of testicular levels of StAR, SF-1, CYP11A1 in the testis of both groups as well as in the in vitro preparation, in a dose-dependent manner, independent of GSH-R1a and not associated with activation of STAT3, a mediator of leptin signaling was apparent. In conclusion, administration of UAG can enhance reproductive function in lean rats and reverses HFD-induced reproductive dysfunction in obese rats.

Abbreviations: AG: acylated ghrelin; BMI: body mass index; CHOL: cholesterol;FSH: follicular stimulating hormone; GHS: growth hormone secretagogues; GSH: reduced glutathione; HFD: high fat diet; HOMA-IR: homeostasis model assessment of insulin resistance: IR: insulin resistance; OGTT: oral glucose tolerance test; ITT: insulin tolerance test; LH: luteinizing hormone; MDA: malondialdehyde; STAT3: signal transducer and activator of transcription; SOD: superoxide dismutase; STD: standard diet; SF-1: steroidogenic factor-1; StAR: steroidogenic acute regulatory protein; CYP11A1: cholesterol side-chain cleavage enzyme (or P450scc); TGs: triglycerides; UAG: unacylated ghrelin.

KEYWORDS:

• Ghrelin
• HFD
• obese
• sperm
• steroidogenesis
• rats

Introduction

Obesity has been listed as a major contributing factor to male infertility in both human and animals (Kasturi et al. 2008), even after controlling other risk factors including age of both partners, body mass index (BMI) of the female, smoking, alcohol abuse and exposure to toxins (Sallmén et al. 2006; Ramlau-Hansen et al. 2007; Nguyen et al. 2007; Ohwaki et al., 2009). Indeed, obese males had low pregnancy rates and number of births, oligospermia, azoospermia and sperm motility associated with increased morphological abnormalities, erectile dysfunction, and increased testicular and germ cells apoptosis (Jensen et al. 2004; Fejes et al. 2006; Fernandez et al. 2011; Fariello et al. 2012; Sermondade et al. 2012; Yan et al. 2015; Cui et al. 2017).

To date, although multiple mechanisms by which obesity-induced male infertility have been identified, it was shown that obesity-induced hypogonadotropic hypogonadism is the major mechanisms leading to male infertility (Hammoud et al. 2012). In obesity, over-activation of cytochrome P450 aromatase enzyme, which converts androgens to estrogen, as well as the insulin resistance (IR) and leptin resistance are major pathways leading to hypogonadotropic hypogonadism (Phillips and Tanphaichitr 2010). Indeed, it was shown that estrogen can directly act on the hypothalamus to inhibit gonadotropin release (Phillips and Tanphaichitr 2010). However, while the coexistence of hypothalamic leptin resistance leads to inhibition of luteinizing hormone (LH) release, the spontaneous increases in testicular leptin in obesity inhibits steroidogenesis and testosterone synthesis (Tena-Sempere et al. 2001; Phillips and Tanphaichitr 2010). Similarly, hyperinsulinemia can directly inhibit testosterone synthesis, induce sperm DNA damage and inhibit the hepatic synthesis of SHBG which eventually increases circulatory estrogen levels (Pasquali et al. 1995; Dhindsa et al. 2004; Tsai et al. 2004; Agbaje et al. 2007).

Ghrelin is a 28-amino-acid peptide that synthesized and secreted mainly by X/A-like cells of the oxyntic mucosa of the stomach (70%) and by the X/A-like cells of the small intestine (30%) (Date et al. 2000). Ghrelin is the endogenous ligand for growth hormone secretagogues receptor (GHS-R) (Kojima et al. 1999), a G-protein-coupled receptor that is found mainly in the pituitary and hypothalamus (Howard et al. 1996) and also in multiple other tissues (Gnanapavan et al. 2002), implicating its important roles it plays in the body. In addition, mRNA of ghrelin and its receptors have been detected in the male’s gonads indicating its regulatory role in reproductive function (Gaytan et al. 2004; Ishikawa et al. 2007).

In the blood, ghrelin circulates in 2 forms; an acylated (AG, 5–20%) and an unacylated ghrelin (UAG, 80–90%) (Van der Lely et al. 2004; Pacifico et al. 2009). While AG acts peripherally throughout its own receptors, reduced glutathione (GSH)-Ra1, UAG has no definite receptors and may compete for AG on GSH-Ra1 or could act via other unidentified receptors (Van der Lely et al. 2004). However, AG and UAG have an antagonistic effect on numerous metabolic activities (Gauna et al. 2004; Barazzoni et al. 2007).

Indeed and under pharmacological concentrations, while AG enhanced appetite, increases food intake, stimulates hepatic glucose output, inhibits insulin release and impairs insulin sensitivity, UAG produced opposite effects independently or counteract AG metabolic effects (Broglio et al. 2004; Gauna et al. 2004, 2005, 2007; Sun et al. 2006; Barazzoni et al. 2007). In addition, AG is a lipogenic factor that can stimulate lipogenesis and inhibit fat oxidation in the liver and adipose tissue of rodents (Choi et al. 2003; Thompson et al. 2004; Giovambattista et al. 2008). On the contrary and independent of food intake or increase in body weight, opposite adipogenic effects were seen in after treatment or overexpression of UAG (Zhang et al. 2008; Delhanty et al. 2010).

Interestingly, BMI is positively correlated with circulatory AG, and a high ratio of circulatory AG/UAG was shown in obese children or adults (St-Pierre et al. 2007; Pacifico et al. 2009; Rodríguez et al. 2009). Also, GHS-Ra1 protein levels were significantly increased in ob/ob mice (Zhu et al. 2013). Concomitant with its adipogenic effect, in vitro and in vivo evidences in males have shown that AG delays puberty, impairs reproductive function and reduces testosterone secretion in several ways including inhibition of Leydig cells proliferation, suppression of steroidogenic enzymes, suppression of LH and follicular stimulating hormone (FSH) release and induction of germ cell apoptosis (reviewed in Alves et al. 2016). On the contrary, inhibition of ghrelin signaling by GHS-R-1a [D-Lys-3]-GHRP-6 (D-GHRP), a specific GHS-R1a antagonist, restored androgen synthesis, reduced germ cell apoptosis, and improved sperm production in ob/ob mice (Zhu et al. 2013).

In spite of these findings, the effect of UAG on lean or obese males’ reproductive function is poorly described. Given its antagonistic effect to AG and its ability to inhibit adipose tissue lipogenesis, hepatic lipid accumulation, and improve insulin sensitivity, in this study, it was hypothesized that administration of UAG will improve testosterone levels enhance reproductive function in lean rats and reverses high fat diet (HFD)-induced reproductive alterations in obese rats and mechanisms of actions were clarified.

Result and discussion

This study confirms that UAG can stimulate steroidogenesis in both lean and obese rats and reverses HFD-induced alterations in sperm parameters and reproductive function. Using an in vivo animal model of lean and obese rats and independent of food intake, UAG significantly lowers adipose tissue mass, increased testis weight, enhanced steroidogenesis and increased both circulatory and testicular testosterone levels, all of which lead to improving sperm count and sperm parameters. This was parallel with stimulating the HPG axis and mediated by significant downregulation of adipose and testicular tissue-derived aromatase decreases in circulatory estradiol and leptin levels and improvement of insulin sensitivity in both populations. To further test if UAG stimulate steroidogenesis directly and independent of GSH-R1a, in vitro incubation of testicular slices with increasing concentrations of UAG in the presence or absence of GSH-R1a antagonist, [D-Lys-3]-GHRP-6 confirmed that UAG leads to a dose-dependent upregulation of the steroidogenesis key enzymes and factors including steroidogenic factor-1 (SF-1), steroidogenic acute regulatory protein (StAR) and cholesterol side-chain cleavage enzyme (CYP11A1), and down-regulation of aromatase expression, independent of leptin signaling, and GSH-R1a.

In general, HFD-induced obesity in rodents is a valid model to study obesity and its allied symptoms and its effect on male infertility (Yan et al. 2015; Cui et al. 2017; Miao et al. 2018). In this study, as expected and over a period of 16 weeks of HFD feeding, rats fed our HFD formula (Table 1) developed significant increases in their final body and fat deposits weights (epididymal, visceral and retroperitoneal fats) and had a higher value of adiposity index as compared to standard diets (STD)-fed rats (Table 2). They also showed significant increases in plasma levels of glucose and insulin as well as calculated homeostasis model assessment of insulin resistance (HOMA-IR) and had higher serum levels of triglycerides (TGs) and cholesterol (CHOL) (Figure 1A-E). In addition, they had significantly higher levels of plasma glucose and insulin levels after OGGT and higher glucose levels after ITT (Figure 2A-C) as compared to control rats fed STD, indicating a state of peripheral IR. This is basically attributed to increased caloric intake as HFD-fed rats consumed more food/day (Table 2). However, administration of UAG to obese rats didn’t alter food intake in both lean and obese rats (Table 2) which strongly argues that the observed changes all measured parameters are not accentuated to decreased appetite nor to increased caloric input but rather due to a direct effect of UAG on fat metabolism or peripheral tissue studied here.

Table 1. Ingredient and nutrient composition of the diets.

	Low-Fat Diet (LFD)		High-Fat Diet (HFD)
Ingredients (g/kg)	Weight	kcal	Weight	kcal
Casein	200.0	800	200.0	800
L-Cystine	3.0	12	3.0	12
Corn Starch	315.0	1260	72.8	291
Maltodextrin 10	35.0	140	100	400
Sucrose	350.0	1400	172.8	691
Soybean Oil	25.0	225	25.0	225
Lard	20.0	180	177.5	1598
Cellulose	50.0	00	50.0	0.0
Vitamin-Mineral Premix	10	40	10	40
Potassium Citrate, 1 H2O	16.5	0.0	16.5	0.0
DiCalcium Phosphate	13	0.0	13	0.0
Calcium Carbonate	5.5	0.0	5.5	0.0
Mineral Mix	10	0.0	10	0.0
Choline chloride	2.0	0.0	2.0	0.0
Total	1055.	4057	858.15	4057
Cholesterol (mg)/4057 kcal	14.4		127.8

*Cholesterol in lard = 0.72 mg/gram.

Table 2. Body, testes, epididymis, and fat weights, as well as calculated adiposity index in all groups of rats.

Parameter	STD	UAG	HFD	HFD + UAG
Daily food intake (g)	26.5 ± 4.2	25.2 ± 3.0	32.3 ± 4.1^ab	31.4 ± 2.8
Final body weight (g)	354.4 ± 11.2	5.4^a ± 328.3	441 ± 14.2^ab	397 ± 2.2^abc
Fat deposits (g)
Epididymal	.3 ± 0.65	4.2 ± 0.3^a	8.7 ± 0.7^ab	6.1 ± 0.6^abc
Visceral	4.1 ± 0.3	3.4 ± 0.4^a	7.8 ± 0.4^ab	4.8 ± 0.3^abc
Retroperitoneal	7.2 ± 0.6	6.1 ± 0.3^a	10.5 ± 0.7^ab	8.1 ± 0.4^abc
Adiposity index (%)	4.8 ± 0.4	4.1 ± 0.3^a	6.1 ± 0.5^ab	4.8 ± 0.4^bc
Testes (g)	3.31 ± 0.4	3.41 ± 0.5	2.9 ± 0.3^ab	3.23 ± 0.2^c
Testes (g/100 g)	0.96 ± 0.1	1.04 ± 0.1	0.66 ± 0.07^ab	0.81 ± 0.09^abc
Epididymes (mg)	1.3 ± 0.1	1.6 ± 0.2^a	1.01 ± 0.07^ab	1.21 ± 0.08^abc
Epididymes (mg/100 g)	0.36 ± 0.04	0.49 ± 0.07 ^a	0.23 ± 0.03^ab	0.30 ± 0.03^abc

Data are expressed as mean ± SD (n = 12/group). Values were considered significantly different at P < 0.05. ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. STD: standard diet. HFD: high-fat diet. UAG: unacylated ghrelin.

Figure 1. Metabolic profile of all experimental groups. Rats were fasted for 12 h at the end of the experimental procedure after being anesthetized with sodium pentobarbital (60–70 mg/kg, intraperitoneally) and blood samples were directly collected by cardiac puncture. (A) Fasting plasma glucose levels (mg/dl) as measured calorimetrically. (B) Fasting plasma insulin levels (ng/ml) as measured by ELISA. (C) Calculated homeostasis model assessment of IR (HOMA-IR) as determined using the following equation (HOMA-IR = FPG (mg/dl) × fasting insulin (U/ml)]/405]). (D) Fasting serum levels of triglycerides (TGs) (mg/dl). (E) Fasting serum levels of cholesterol (CHOL) (mg/dl). The serum TGs and CHOL were measured calorimetrically using a biochemical analyzer (ADVIA 1200 S analyzer). Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’st test. Data are expressed as mean ± SD of n = 12/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. STD: standard diet. HFD: high-fat diet. UAG: unacylated ghrelin.

Figure 2. Oral glucose tolerance test (OGTT) and insulin tolerance test (ITT) outputs of all experimental groups. For the OGTT, an oral glucose solution (2 g/kg) was administered to 12 h-fasted rats and then both glucose (mg/dl) and insulin (µU/ml) levels were examined before glucose loading and immediately at 15, 30, 60, 90 and 120 min. ITT was performed 1 day after the OGTT where the 12-h fasted rats were injected intraperitoneally with insulin (0.75 U/rat) followed by the direct measurements of glucose (mg/dl) at similar time intervals like those of OGTT. (A) Plasma glucose levels (mg/dl) at different time intervals before and after glucose loading. (B) Area under the curve (AUC) for glucose levels presented in graph A and expressed as percentages of the AUC of the control rats fed the standard diet (STD). (C) Plasma insulin levels (µU/ml) at different time intervals before and after glucose loading. (D) AUC for insulin levels presented in graph C and expressed as percentages of the AUC of the control rats fed the STD. (E) Plasma glucose levels (mg/dl) at different time intervals before and after intraperitoneal insulin administration. (F) AUC for glucose levels presented in graph E and expressed as percentages of the AUC of the control rats fed the STD). Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’st test. Data are expressed as mean ± SD of n = 12/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. HFD: high-fat diet. UAG: unacylated ghrelin.

In addition, the current study supports the results of others performed in human and animal models that previously assumed that obesity is associated and linked with reduced levels of both AG and UAG and concomitant increases in AG/UAG ratio (Barazzoni et al. 2007; St-Pierre et al. 2007; Pacifico et al. 2009; Rodríguez et al. 2009). In the plasma of STD-fed rats, circulatory UAG concentration was dominant over that of AG where the ratio of AG/UAG was 0.431 (1:2.3) (Figure 3A-C). In support to above-mentioned studies, HFD feeding significantly reduced the plasma circulatory levels of both AG and UAG, thus increasing the ratio of AG/UAG to 0.718 (1:1.4) (Figure 3A-C). However, although UAG has no effect of circulatory AG levels (Figure 3A), it significantly increased the circulatory level of UAG in both STD and HFD-fed rats by 3.47 and 2.7 folds beyond its baseline value measured in STD-fed rats (Figure 3B), thus significantly decreased the ratio of AG/UAG in both lean and obese rats to 0.125 (1:8) and 0.164 (1:6.2), respectively (Figure 3C). These findings suggest that improvements in all metabolic and reproductive parameters studied are associated with the high UAG/AG ratio after UAG therapy.

Figure 3. Plasma levels of acylated ghrelin (AG) and unacylated ghrelin (UAG) and serum leptin levels in all experimental groups. Rats were fasted for 12 h, anesthetized with sodium pentobarbital (60–70 mg/kg, intraperitoneally) and then blood sample were collected by cardiac puncture. (A) Plasma levels of AG (pg/ml) as measured by ELISA. (B) Plasma levels of UAG (pg/ml) as measured by ELISA. (C) Average ratio of AG/UAG as calculated by dividing individual level of AG over corresponding level of UAG for each rat. (D) Average ratio of UAG/AG as calculated by dividing individual level of UAG over corresponding level of AG for each rat. (E) Serum levels of leptin (ng/ml) as measured by ELISA. Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’st test. Data are expressed as mean ± SD of n = 12/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. HFD: high-fat diet. UAG: unacylated ghrelin.

In the current study, the initial established positive correlation between infertility and obesity has been confirmed by the lower pregnancy rate and index, number and weights of pubs at birth or at age of 7 days, number of fetus per female (Table 3) and sperm count, impaired motility, increased morphological abnormalities (Table 4) and altered seminiferous tubules, and germ cells structures in obese rats (Figure 4C). This is in accordance with many other studies (Jensen et al. 2004; Fejes et al. 2006; Fernandez et al. 2011; Fariello et al. 2012; Yan et al. 2015; Cui et al. 2017). Interestingly, this was associated with a pathological oxidative stress response and increased apoptosis in the testis of obese rats as evident by the significant increases in malondialdehyde (MDA) levels, decreases in GSH levels and superoxide dismutase (SOD) activity and upregulated protein levels of cleaved caspase-3 (Figure 5A-D), confirming the role of obesity as an independent factor of male infertility which acts by increasing testicular oxidative stress-induced damage of testicular tissue, germ cells apoptosis and decreased sperm count (Agarwal et al. 2006; Rato et al., 2014; Wagner et al. 2017).

Table 3. Mating outcome in all groups of rats.

Parameter	STD	UAG	HFD	HFD + UAG
No. of females	12	12	12	12
No. of Pregnant females	12	12	5	10
Pregnancy index (%)	100%	100%	41.6%	83%
No. of pubs at birth	109	115	32	89
No. of pups per female	9.2 ± 0.4	9.6 ± 0.69	6.95 ± 0.59	7.32 ± 0.83
The weight of pubs at birth	7.25 ± 0.75	7.34 ± 0.69	6.95 ± 0.59	7.32 ± 0.83
Survival ratio at day 7 (%)	104/109 (95.4%)	109/115 (94.7%)	31/32 (96.8%)	84/89 (94.3%)
The weight of pubs on day 7	10.4 ± 1.2	10.7 ± 1.1	10.3 ± 1.1	10.6 ± 1.4

Data are expressed as mean ± SD (n = 12/group). Values were considered significantly different at P < 0.05. STD: standard diet. HFD: high-fat diet. UAG: unacylated ghrelin.

Table 4. Characterization of epididymal sperm morphology in groups of rats (%).

Parameter	STD	UAG	HFD	HFD + UAG
Sperm count 10^6/0.1g epididymis	143.4 ± 12.4	177 ± 18.4^a	74.6 ± 8.9^ab	113.4 ± 10.2^abc
Sperm motility (%)	73.2 ± 4.9	84.2 ± 6.3^a	34.7 ± 4.3^ab	64.8 ± 5.8 ^abc
Absence of tail (%)	1.74 ± 0.42	1.89 ± 0.45	10.80 ± 1.3^ab	4.87 ± 0.62 ^abc
Absence of head (%)	1.98 ± 0.32	1.88 ± 0.38	4.93 ± 0.33^ab	1.77 ± 0.47^ab
Tail bending (%)	2.92 ± 0.61	2.87 ± 0.38	2.86 ± 0.31	2.99 ± 0.44
Tail coiling (%)	1.09 ± 0.22	1.33 ± 0.18	22.11 ± 4.44^ab	7.89 ± 1.12^abc
Midpiece bending (%)	1.48 ± 0.31	1.56 ± 0.4	1.74 ± 0.44	1.47 ± 0.5
Total abnormality (%)	9.2 ± 0.43	9.5 ± 0.35	42.4 ± 0.71 ^ab	19.0 ± 0.78^abc

Data are expressed as mean ± SD (n = 12/group). Values were considered significantly different at P < 0.05. ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. STD: standard diet. HFD: high-fat diet. UAG: unacylated ghrelin.

Figure 4. Histopathological changes by hematoxylin and eosin (H and E) in the testis of all experimental groups (200X). (A) Taken from a control rat. (B) Taken from a control rats treated with UAG. Both A&B showed normal testicular architecture characterized by well-preserved seminiferous tubule structures with a preserved definite membrane with a small lumen filled with mature sperm. Spermatogonia, primary spermatocytes, early spermatids, late spermatids, Sertoli and Leydig cells were also present and well preserved. (C) Taken from a HFD fed obese rat and showing degeneration and shrinked seminiferous tubules with wide lumens some of which are empty of mature sperm whereas the others have reduced sperm. Loss of spermatogonial cells, presence of giant and welled primary spermatocytes, loss of Sertoli and Leydig cells and intestinal edema were also present. (D) Taken from a HFD+ UAG treated rat showing improvements in the structure of seminiferous tubules and size, regeneration of Sertoli and Leydig cells, and absence of edema. All stages of spermatogonial cells with central sperms in the lumen of tubules were seen.

Figure 5. Oxidative stress and apoptosis in the testes of all experimental groups. Oxidative stress determinations were made in homogenized testicular tissue (50 mg/rat). (A) Levels of malondialdehyde (MDA) (µM) as determined calorimetrically. (B) Reduced glutathione (GSH) (µM) levels as determined calorimetrically. (C) Activity of SOD (U/mg tissue) as determined calorimetrically. (D) Photomicrographs of protein expression of cleaved caspase-3 and the reference protein, β-actin, in the testis of all experimental groups as detected by western blotting. (E) Calculated average relative expression of protein levels of cleaved caspase-3 presented in graph D as normalized to their individual corresponding levels of β-actin. For western blotting, equal protein samples (60µg) from each group were separated on nitrocellulose membranes. Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’s t test. Data are expressed as mean ± SD of n = 6/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. HFD: high-fat diet. UAG: unacylated ghrelin.

In the testis of obese males, major sources of pathological levels of ROS include adipocytes-induced inflammatory cytokines including TNF-α and IL-6, IR, and dyslipidemia (Dandona et al. 2005; Davi et al., 2005; Agarwal et al. 2006; Lampiao and Du Plessis 2008). As previously mentioned, HFD-fed rats of the current study developed IR and had higher levels of circulatory TGs and CHOL which may explain the origin of the oxidative stress response in their testis. Interestingly and given the observation that UAG administration to STD-fed lean rats have no impact on all measured markers of oxidative stress and on the levels of cleaved caspase-3 (Figure 5A-D), it is expected that the amelioration of oxidative stress parameters in obese rats is not due to an antioxidant potential of UAG but rather secondary to the anti-adipogenic role of UAG mediated by improvements of insulin sensitivity and serum lipids.

The second observation of this study is the reduced levels of LH, FSH and testosterone levels in the serum of HFD-induced obese rats (Figure 6A-E), which can be described as of hypogonadotropic hypogonadism, a common symptom seen in obese males (Roth et al. 2008; Phillips and Tanphaichitr 2010; Hammoud et al. 2012). Steroidogenesis and testosterone synthesis requires an adequate amount of circulatory FSH and LH levels which is tightly regulated by the HPG axis (Jana et al. 2006). In addition to their roles in spermatogenesis, testosterone and FSH can inhibit germ cells apoptosis (Blanco-Rodriguez and Martinez-Garcia 1998; Yan et al. 2015). Such decreases in LH and FSH levels observed in obese rats could explain why these rats had less sperm count, low circulatory testosterone levels and reduced expression of steroidogenic enzymes and factors including StAR, CYP11A1 (P450^scc) and SF-1 (Figure 7A). On the contrary, UAG administration (higher UAG/AG ratio) significantly increased the levels of these sex hormones in STD and HFD-fed rats and upregulated the expression of all these mediators (Figure 6A-E, Figure 7A), leading to higher sperm count and less apoptosis. Hence, it was of interest to investigate the mechanisms by which obesity and UAG modulate the activity of the HPG axis and steroidogenesis.

Figure 6. Hormonal profile analysis the serum and of intratesticular testosterone levels in all experimental groups. (A) Serum levels of testosterone (ng/ml) as determined by ELISA (A). (B) Serum levels of luteinizing hormone (ng/ml) as determined by ELISA. (C) Serum levels of follicular stimulating hormone (FSH) (ng/ml) as determined by ELISA. (D) Serum levels of estradiol (pg/ml) as determined by ELISA. (E) Testosterone levels (ng/mg) in the right testis of all experimental groups. For intratesticular testosterone determination, testicular slices (40-50mg) were incubated in appropriate buffered medium 199 (M199) containing bovine serum albumin (BSA) and soybean trypsin inhibitor at 34°C for 2 h. Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’s t test. Data are expressed as mean ± SD of n = 6/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. HFD: high-fat diet. UAG: unacylated ghrelin.

Figure 7. Effect of Unacylated ghrelin (UAG) on the in vivo and in vitro expression of major steroidogenesis genes. (A) Photomicrographs of protein expression of StAR, CF-1, CYP11A1 and the reference protein, β-actin, in the testis of all experimental groups as detected by western blotting. (B) Calculated average relative expression of protein levels of StAR, CF-1, CYP11A1 presented in graph A as normalized to their individual corresponding levels of β-actin. (C) Photomicrographs of protein expression of StAR, CF-1, CYP11A1 and the reference protein, β-actin, in the isolated testicular tissue after treatment with increasing concentrations of UAG (10⁻⁸–10⁻⁶ M) in the presence or absence of GSH-R1a antagonist, [D-Lys-3]-GHRP-6 (10⁻⁶ M). Lane 1: control cells incubated with the medium only; lane 2: cells incubated with [D-Lys-3]-GHRP-6 (10⁻⁶ M); lane 3: cell incubated with UAG (10⁻⁸ M); lane 4: cell incubated with UAG (10⁻⁷ M); lane 5: cell incubated with UAG (10⁻⁶ M), lane 6: cells incubated with both [D-Lys-3]-GHRP-6 (10⁻⁶ M) and UAG (10⁻⁶ M). (D) Calculated average relative expression of protein levels of StAR, CF-1, CYP11A1 presented in graph C as normalized to their individual corresponding levels of β-actin. For western blotting, equal protein samples (60µg) from each group were separated on nitrocellulose membranes. Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’s t test. Data are expressed as mean ± SD of n = 6/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. HFD: high-fat diet.

Circulatory insulin, estrogen, and leptin are major modulators of the HPG axis in males. Indeed, increased fat mass in obesity induces direct activation of adipose tissue-derived aromatase cytochrome P450 enzyme which can convert androgen to estrogen, which in turn has a negative feedback signal on the HPG axis (Schneider et al. 1979; Cohen 1991; Akingbemi 2005; Phillips and Tanphaichitr 2010). In the same line, HFD-induced obese rats of this study had higher levels of circulatory estradiol (Figure 8E) and upregulated levels aromatase in both their adipose tissue and testis (Figure 8A-B). Interestingly, even UAG treatment to lean or obese rats significantly decreased circulatory levels of estradiol and downregulated aromatase expression in both tissues (Figure 8A-B), which could be due to decreases in fat masses. It significantly downregulated aromatase expression in a dose-dependent manner, in vitro, when isolated testis tissue was incubated with increasing concentrations of UAG (10⁻⁸–10⁻⁶ M) (Figure 8C). This effect was independent of GSH-R1a as evident by the similar protein expression levels of aromatase that were seen in the isolate testicular slices pretreated with [D-Lys-3]-GHRP-6 a specific GHS-R1a antagonist and then incubated with UAG as compared to the testicular preparation of the same tissue incubated with UAG at a dose of 10⁻⁶ M (Figure 8C).

Figure 8. Effect of Unacylated ghrelin (UAG) on the in vivo and in vitro expression of aromatase. (A) Photomicrographs of protein expression of aromatase and the reference protein, GAPDH, in the white adipose tissue (WAT) of all experimental groups. (B) Calculated average relative expression of WAT protein levels of aromatase presented in graph A as normalized to their individual corresponding levels of GAPDH. (C) Photomicrographs of protein expression of aromatase and the reference protein, β-actin, in the testis of all experimental groups. (D) Calculated average relative expression of testicular protein levels of aromatase presented in graph C as normalized to their individual corresponding levels of β-actin. (E) Photomicrographs of protein expression of aromatase and the reference protein, β-actin, in the isolated testicular tissue after treatment with increasing concentrations of UAG (10⁻⁸–10⁻⁶ M) in the presence or absence of GSH-R1a antagonist, [D-Lys-3]-GHRP-6 (10⁻⁶ M). Lane 1: control cells incubated with the medium only; lane 2: cells incubated with [D-Lys-3]-GHRP-6 (10⁻⁶ M); lane 3: cell incubated with UAG (10⁻⁸ M); lane 4: cell incubated with UAG (10⁻⁷ M); lane 5: cell incubated with UAG (10⁻⁶ M), lane 6: cells incubated with both [D-Lys-3]-GHRP-6 (10⁻⁶ M) and UAG (10⁻⁶ M). F: Calculated average relative expression of protein levels of aromatse presented in graph E as normalized to their individual corresponding levels of β-actin. For western blotting, equal protein samples (60µg) from each group were separated on nitrocellulose membranes. Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’s t test. Data are expressed as mean ± SD of n = 6/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. HFD: high-fat diet.

Similarly, IR impairs reproductive function in males directly via inhibition of testosterone and induction of germ cell apoptosis or indirectly, by the inhibition of HPG axis (Dhindsa et al. 2004; Tsai et al. 2004; Pitteloud et al., 2005; Agbaje et al. 2007). Indeed, hyperinsulinemia in obesity reduces the hepatic synthesis of SHBG which ultimately increases estradiol levels (Jensen et al. 2004). Hence, we could speculate that IR developed in the HFD-induced obese rats of this study to participate at least in the observed increases in circulatory estradiol levels in their serum, which was ameliorated in both STD and HFD-induced obsess rats treated with UAG and had a higher ratio of UAG/AG (Figure 2A-C).

Indeed, UAG treatment significantly reduced fasting glucose and insulin levels, improved HOMA-IR and improved oral glucose tolerance test (OGTT) and ITT output in both STD and HFD-induced obese rats (Figures 1 and 2). In support to our findings, UAG inhibited hepatic glucose output and insulin release and improved insulin sensitivity in culture as well as lean rats in vivo (Broglio et al. 2004; Gauna et al. 2004, 2005, 2007; Sun et al. 2006; Barazzoni et al. 2007), an effect that has been attributed to the inhibition of lipid synthesis or accumulation in both the liver and adipose tissue (Barazzoni et al. 2007; Cederberg et al. 2012). Indeed, mice overexpressing UAG have lower fat body mass and improved insulin sensitivity compared to control (Zhang et al. 2008). In this regard, it has been found that UAG treatment enhanced peripheral insulin sensitivity, inhibited lipid synthesis and stimulated lipolysis in adipose tissue and significantly increases fat oxidation in the liver of GHS-Ra1-deleted mice (Zhang et al. 2008; Delhanty et al. 2010). This could explain the reduction in fat mass and low circulatory levels of TGs and CHOL in the livers of lean or HFD-fed rats treated with UAG (Table 2 and Figure 1E-D).

On the one hand, circulatory leptin levels are directly proportional to the adipose mass and BMI (Margetic et al. 2002). HFD-induced obese rats in this study had higher levels of circulatory leptin levels (Figure 3E). Similar to increases in estradiol levels and IR, hyperleptinemia due to increase fat mass in obesity inhibits HPG axis (Phillips and Tanphaichitr 2010). Under physiological concentrations, leptin stimulates the hypothalamus increasing the synthesis and releases FSH and LH Phillips and Tanphaichitr 2010). However, in obesity, hypothalamic leptin insufficiency (resistance) can induce hypogonadotropic hypogonadism which could also participate in the observed decreases in the circulatory levels of sex hormones in obese rats in this study. An inverse relationship between serum levels of leptin and testosterone has been observed in overweight and obese males (Vettor et al. 1997; Isidori and Fabbri 1999; Zorn et al. 2007). Hence, the reduction in fat masses induced by UAG (higher UAG/AG ratio) in both control and obese treated rats shown in this study may explain why those rats have reduced levels of circulatory leptin as compared to their controls (Figure 3E), and subsequently, have higher levels of testosterone.

On the other hand, insulin receptors, as well as leptin receptors, have been described in Leydig cells and they can modulate testosterone synthesis (Lin et al. 1986; Ishikawa et al. 2007; Herrid et al. 2008). Interestingly, testis blood barrier (TBB) is unsaturable and any significant increase in serum leptin level induces a dose-dependent increase in testicular leptin levels (Banks et al. 1999). Similar to the decreases in circulatory LH levels, IR and testicular hyperleptinemia can inhibit testicular steroidogenesis by downregulation of including StAR, P450^scc, and SF-1 (Lin et al. 1986; Bebakar et al. 1990; Tena-Sempere et al. 2001), which could also participate in the observed decreases in the expression of these proteins in the testis of HFD-obese rats.

The unanswered point remained in this study is which mechanisms mediates the upregulation of StAR, P450^scc, and SF-1 in lean and obese rats treated with UAG. Additionally, it was of our interest to investigate if UAG can directly stimulate testicular tissue steroidogenesis. UAG candidate signaling of the Ob-Rb pathway in the testis includes downstream activation SOCS3 and signal transducer and activator of transcription (STAT3) (Gema Frühbeck, 2006). Hence, we examined STAT3 activation in the testis of all groups of rats and in vitro after incubating the testicular tissue with increasing concentrations of UAG (10⁻⁶–10⁻⁸ M). In addition, to investigating if UAG may act through GSH-R1a, we have added one group where the testicular tissue slices were pre-incubated with GSH-R1a antagonist, [D-Lys-3]-GHRP-6 and then incubated with the highest effective concentration of UAG (10⁻⁶ M).

Protein expression of SF-I, StAR, and CYP11A1 have significantly increased in a dose-dependent manner in the isolated testicular slices incubated with increasing concentrations of UAG (10⁻⁸–10⁻⁶ M) with the maximum expression of all these proteins to be seen with the highest concentration of UAG of 10⁻⁶ M (Figure 7B). In addition, significant increases in protein levels of SF-1, StAR and CYP11A1 were also observed when testicular tissue was treated with [D-Lys-3]-GHRP-6 and then incubated with UAG at a concentration of 10–⁶ M, which were not significantly different to their protein expression seen in testicular tissues treated only with UAG (Figure 7B), indicating that UAG is able to directly stimulate steroidogenesis, independent of GSH-R1a.

Interestingly, levels of STAT3 were significantly increased in the testis of obese rats and significantly decreased in the testis of HFD-induced obese rats treated with UAG (Figure 9A). These results confirm that hyperleptinemia can suppress steroidogenesis in obesity, in vivo. However, UAG administration to lean rats didn’t or after incubation with the testicular slices, in vitro, didn’t affect STAT3 levels, indicating the absence of any effect on testicular Ob-Rb signaling pathway. This confirms that the improvements steroidogenesis in obese rats treated with UAG is secondary to decrease circulatory leptin levels after UAG treatments. Indeed, using our in vitro testis model, stable non-statistically significant protein levels of STAT3 and p-STAT3 were also seen with all concentrations of UAG used with or without [D-Lys-3]-GHRP-6 treatment (Figure 9B).

Figure 9. Effect of Unacylated ghrelin (UAG) on the in vivo and in vitro expression of STAT3 and phospho-STAT3 (p-STAT = 3). (A) Photomicrographs of protein expression of STAT3 and p-STAT3 and the reference protein, β-actin in the testis of all groups of rats. (B) Calculated average relative expression of STAT3 and p-STAT3 presented in graph A as normalized to their individual corresponding levels of β-actin. (C) Photomicrographs of protein expression of STAT3 and p-STAT3 in incubated isolated testicular tissue after treatment with increasing concentrations of UAG (10⁻⁸–10⁻⁶ M) in the presence or absence of GSH-R1a antagonist, [D-Lys-3]-GHRP-6 (10⁻⁶ M) and normalized to β-actin. Lane 1: control cells incubated with the medium only; lane 2: cells incubated with [D-Lys-3]-GHRP-6 (10⁻⁶ M); lane 3: cell incubated with UAG (10⁻⁸ M); lane 4: cell incubated with UAG (10⁻⁷ M); lane 5: cell incubated with UAG (10⁻⁶ M), lane 6: cells incubated with both [D-Lys-3]-GHRP-6 (10⁻⁶ M) and UAG (10⁻⁶ M). (D) Calculated average relative expression of STAT3 and p-STAT3 presented in graph C as normalized to their individual corresponding levels of β-actin. For western blotting, equal protein samples (60µg) from each group were separated on nitrocellulose membranes. Statistical analysis between the various groups was done using a one-way ANOVA test, followed by Tukey’s t test. Data are expressed as mean ± SD of n = 6/group and values were considered significantly different at P < 0.05 were ^a:vs. STD, ^b:vs. STD+ UAG. ^c:vs. HFD. HFD: high-fat diet.

In conclusion, the data presented in this study indicate that UAG can stimulate steroidogenesis in both obese and lean rats by stimulating HPG axis, a mechanism mediated by lowering leptin and estradiol levels and improving IR, direct inhibition of aromatase expression in adipose and testicular tissue and by direct upregulation of StAR, CYP11A1 (P450^scc) in the testicular tissue. However, further studies are needed to determine if UAG can act directly at the levels of the hypothalamus and/or pituitary level or is mediated by modulation of IR.

Materials and methods

Animals

Healthy adult male Wistar rats (weighing 80–90 g) were obtained from the animal facility of the Medical College at King Khalid University, Abha, Kingdom of Saudi Arabia (KSA). During the adaptation period of 1 week and during the experimental procedure, all rats were housed in groups of four in a well-controlled separate room of temperature of 23 ± 1°C and humidity of 55 ± 10%, under a 12-h light/dark cycle. Rats were always allowed free access to food and water ad labium. All procedures performed here were approved by KKU animal ethical committee which follows the guidelines published by the US National Institutes of Health for animal use (NIH Publication No. 85–23, revised 1996).

Diets

Both standard (STD) and high-fat diets (HFD) were prepared in our labs as previously described by Li et al. (2013) with some modifications (Table 1). Each 100 g of the STD provides about 3.85 kcal/g of fat and contained 67.3 g carbohydrates (70% kcal), 19.2 g proteins (20% kcal) and 4.3 g fat (10 % kcal), whereas each 100 g of the HFD provides about 4.73 kcal/g of fat and contained 40 g carbohydrates (35% kcal) 24 g proteins (20% kcal) and 24 g fat (45 kcal%). Both diets were always prepared every week and were always maintained at 4°C in a cold chamber for a maximum period of 7 days.

Animals and experimental design

Rats (n = 48) were divided into four groups (12 rats each) as 1) STD: fed STD for 16 weeks and the administered normal saline as vehicle for the next 4 weeks; 2) STD + UAG: fed STD for 16 weeks and then administered UAG (200 ng/kg) for the next 4 weeks (Cat No. 2951 Tocris Bioscience, USA); 3) HFD model group: fed HFD for 16 weeks and then administered normal saline for the next 4 week; ands 4) HFD then UAG: fed HFD for 16 weeks and then administered UAG (200 ng/kg, i.p) for the next 4 weeks. All treatments were administered i.p. in a final volume of 200 µl. Rats were weighed every week, and food consumption was monitored daily. However, the dose of UAG was selected based on our previous study (Dallak 2018) where we have shown that UAG at this dose is able to inhibit fasting glucose levels and hepatic lipogenesis and IR in lean rats (Dallak 2018).

Mating and pregnancy rate

Directly after the end of all treatments and for a period of 1 week, each rat per group was cohabitated with two proestrus females and a successful mating was confirmed by the presence of sperm in female’s vaginas after being flushed with normal saline. Over a period of 3 weeks, the number of pregnant females and the number of pups and their weights at birth and after 1 week of birth were recorded.

OGTT, ITT, and HOMA-IT

Twelve hours post the last treatment; rats of all groups’ rats were fasted for 12 h and then administered either oral glucose solution (2 g/kg) to measure OGTT. A similar procedure was followed 1 days later were the rats were injected intraperitoneally with of insulin (0.75 U/rat) to measure the insulin tolerance test (ITT). In each test, blood samples (250 µl each) were collected from the tail tip after 0, 15, 30, 60, 90 and 120 min. For each time interval, blood glucose and insulin levels were measured after OGTT and only glucose levels were measured for ITT. Blood samples that were collected at 0.0 min before glucose administration was used to measure for fasting plasma glucose (Cat. No. ab65333, Abcam, Cambridge, UK) and insulin levels (Cat No. ERINS, Thermo Fisher Scientific, Waltham, USA). The areas under the curve (AUC) for both glucose and insulin levels obtained by each test were calculated using GraphPad (version 6) prism software and relatively presented as compared to AUC of the control rats. Insulin sensitivity was assessed by calculating the value of homeostasis model assessment of IR (HOMA-IR) according to this formula:

$\mathbf{H}\mathbf{O}\mathbf{M}\mathbf{A}\mathbf{-}\mathbf{I}\mathbf{R}=[\mathrm{F}\mathrm{P}\mathrm{G}(\mathrm{m}\mathrm{g}/\mathrm{d}\mathrm{l})\mathrm{f}\mathrm{a}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{s}\mathrm{u}\mathrm{l}\mathrm{i}\mathrm{n}(\mathrm{U}/\mathrm{m}\mathrm{l})]/405.$

Blood and tissue collection and adiposity index

Twenty-four hours after ITT, each 12 h-fasted rat was anesthetized with sodium pentobarbital (60–70 mg/kg, i.p.) and a 2 ml blood sample was collected directly from the heart into either an EDTA or plain tubes to prepare plasma and sera, respectively, which were stored at −80°C for further analysis. Then, animals were killed by euthanasia with an overdose of sodium pentobarbital (160 mg/kg, i.p.) and exsanguinations. Epididymal, visceral and retroperitoneal pad fats were carefully collected and weighted. Adiposity index was calculated by the sum of all these fats divided by the final body weight and then multiplied by 100% (Eleawa et al. 2014). Then after, both testes and epididymis were quickly removed, cleaned from any extra tissue, weighed and washed with ice-cold phosphate buffered saline (PBS, pH 7.4) containing heparin (0.16 mg/m) to remove any red blood cells (erythrocytes) and clots. Absolute and relative (to final body weights) weights for both testes and epididymides were calculated. The right epididymis of each rat was used freshly to determine sperm parameters as discussed below. On the other hands, the left testis was cut into small pieces, embedded in liquid nitrogen, stored at −80°C and used later for biochemical analysis, qPCR, western blotting studies. The right testis was cut into two equal halves, one used for histological evaluation and the other half used to determine intratesticular testosterone levels.

Measurements in the sera

The serum levels of total TGs and CHOL were measured using a biochemical analyzer (ADVIA 1200 S analyzer). ELISA rat’s specific kits were used to measure testosterone (Cat. No. E0930Ra, Shanghai Crystal Day Biotech Co., Ltd. China), FSH (Cat. No. ab108641, Cambridge, UK), LH (Cat. No. CSB-E12654r, Cusabio Biotech Co., Ltd., China) and estradiol (Cat. No. ab108641, Cambridge, UK). Fresh plasma samples were used within 3 days after collection to measure levels of AG and UAG using special rat’s ELISA kits (A05117 and A05118, respectively, SPI Bio, France). Samples were collected into special EDTA tubes (Cat. No. D31009). All samples were stabilized by the addition p-hydroxy Mercuri benzoic acid (PHMP) (final concentration of 1mM) and 1N HCl (100 µl/ml plasma, pH between 4–5) to prevent AG degradation and stabilize both forms (Dezaki et al., 2004). Leptin levels in the serum samples were determined using rat’s ELISA kit (Cat. No. 90,040, Crystal Chem, IL, USA).

Intratesticular testosterone levels

Levels of testosterone in each rat’s testis were measured according to the method described by Laskey et al. (1994) and Fernandez et al. (2011). In brief, the half of the right testis was decapsulated and then the parenchyma cuts into slices each of about 40–50mg. Then they were incubated with containing 1.0 ml of buffered Medium 199 (M199) (2.1 g/l Hepes and 0.71 g/L sodium bicarbonate (NaHCO3)) containing 0.1% bovine serum albumin (BSA) and 25 mg/L soybean trypsin inhibitor. The parenchyma was incubated in the media at 34°C for 2 h and centrifuged 10000g for 5 min to collect supernatants which were stored at −80°C for testosterone determinations using the same kits used for the assay of this hormone in the serum.

Semen parameters

Sperm count, viability, motility and morphology were done according to Eleawa et al. (2014). In brief, the right cauda epididymis was freshly minced and kept in normal saline (1:20, wt/v, 37°C) for 5 min. Motility, morphology, and count were done under the light microscope manually. The total number of sperms was counted using a glass hemocytometer. Immotile and motile sperms were counted in a total of 400 sperm field expressed as percentages (%). For morphological assessments, sperm suspension was mixed with Eosin at a ratio of 1:1 and the number of headless, tailless, bent tail, coiled tail, curved mid-piece sperm were counted and then presented a percentage (%). Counts were done in duplicates for each sperm sample and then averaged. Final data were presented as average readings of the 12 rats/group.

Oxidative stress analysis in the testis

A colorimetric kit was used to determine testicular levels of MDA (Cat. No. NWK-MDA01, NWLSS, USA) and reduced glutathione (GSH) (Cat. No. 703,002, Cayman Chemical, Ann Arbor, MI, USA), as well as the activity of SOD (Cat. No. 706,002, Cayman Chemical, Ann Arbor, MI, USA). All procedures were performed in accordance with the manufacturer’s instructions and were run in duplicates. Final data were presented as average readings of the 12 rats/group.

In vitro testicular tissue incubation

This part was done after the initial observation that in vivo administration of UAG enhanced mRNA and protein levels of some steroidogenic-related genes and enhance circulatory testosterone levels in both lean and HFD-fed rats. It aimed to investigate if the effect of UAG on steroidogenesis occurs directly (not mediated by leptin or leptin receptors) and if it is mediated by GHS-R1a or not. Testicular tissue incubation was done in details as previously described by Tena-Sempere et al. (2001). Testes from adult healthy males (140 ± 10g) of the same genetic background and parents were isolated, decapsulated and sliced into four equal pieces with an average weight of 349 ± 38 mg). Each two parenchyma slices were incubated with 2 ml of DMEM–F12 medium (ratio of 1:1) in a glass tube. The medium was supplemented with gentamicin (0 · 1 g/l). The whole mixture was incubated for 1 h at 32°C under an atmosphere of 5% CO2–95% with continuous shaking. Then, the media were changed and replaced individually with a fresh media containing increasing concentrations of UAG (10⁻⁸–10⁻⁶ M) in the presence or absence of GSH-R1a antagonist, [D-Lys-3]-GHRP-6 (10⁻⁶ M). In addition, an additional group was challenged with [D-Lys-3]-GHRP-6 (10⁻⁶ M) alone. The control group received no treatment. In this study, [D-Lys-3]-GHRP-6 was added 20 min before the addition of UAG. All incubations were done for 12 h. Doses of [D-Lys-3]-GHRP-6 was selected based on the study of Fukushima et al. (2005) where such dose was able to inhibit the effect of AG on primary osteoblast-like cell proliferation in culture. After a 12-h incubation period, testicular tissues were frozen in liquid nitrogen and stored at −80°C for q-PCR and western blot studies for the detection of steroidogenic enzymes and leptin signaling. Also, supernatants were preserved at −80°C

Western blotting

Total protein was extracted from frozen testicular and visceral fat samples (in vivo or in vitro) using a Millipore extraction kit (Cat. No. 2140, Merck Millipore, USA) to which protease inhibitor cocktail (Cat. No. P8340, Sigma-Aldrich, St. Louis, MO, USA) was added according to manufacturer’s instructions. Protein levels in all samples were determined using a Pierce BCA protein assay kit (Cat. No.23225, Thermo Fisher Scientific). The procedure was done in accordance with our established method (Eid et al., 2018). In brief, each protein sample of 60 µg was separated on 8–15% SDS-PAGE gel and then transferred manually to nitrocellulose membranes. After successful washing with TBS-T buffer and blocking with 5% skimmed milk (in TBS-T buffer), membranes were incubated overnight at 4°C with primary antibodies against StAR (Cat. No. sc-166,821 4399, 30 kDa, 1:1000), CYP11A1 (P450^scc) (Cat. No. sc-18,043, 60 kDa, 1:1000), STAT3 (Cat. No. F-2:sc-8019, 91/86 kDa, 1:500), p-STAT3 (Tyr 705, Cat. No. B7:sc-8059, 91/86 kDa, 1:250), and β-actin (Cat. No. 4970, 45 kDa, 1:200), all purchased from Santa Cruze Biotechnology. Antibody against SF-1 (Cat. No. ab65815, 52 kDa, 1:500) purchased from Abcam, UK and antibodies against cleaved caspase-3 (Cat. No 9662, 17, 19,35 kDa, 1:500), BCL-2 (Cat. No, 2876, 28kDa, 1:1000), aromatase (Cat. NO. 8799, 61 kDa, 1:1000) and GAPDH (Cat. No. 2188, 37 kDa), all purchased from Cell Signaling Technology. Membranes were then successfully washed and incubated with the corresponding secondary HRP-conjugated secondary antibody. Developments of bands were detected by chemiluminescence (Pierce ECL reagents, Thermofisher, USA, Piscataway, NJ) and all band intensities were quantified using C-DiGit blot scanner (LI-COR, USA) and its image studio DiGits software. Protein expressions were presented as relative expressions to the expression of β-actin. Data were performed in duplicate for six rats/group.

Histological evaluation

Formalin preserved testes sections were embedded in paraffin, cut into sections of 5 μm, processed and finally stained with hematoxylin and eosin (H and E). All sections were examined under a light microscope by different a histologist who was unaware of the groups.

Statistical analysis

GraphPad statistical software (Version 6) was used for all analysis. All data were collected and analyzed using a one-way ANOVA test, followed by Tukey’s t test to determine the significant difference between the various groups. For OGTT and IPGTT, we have used two-way ANOVA with repeated measures to detect the significance between groups and within the same group. All data are expressed as mean ± SD and P < 0.05 and was considered to be statistically significant.

Acknowledgments

The author would like to thank Mr. Mahmoud Alkhateeb from the department of physiology at KKU for his contribution in the measurements of some biochemical and molecular parameters of this study and would like to thank the technical staff of the animal house at the and the medical college of King Khalid University for their help in the current study.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Mohammad Dallak

Dr. Mohammad Dallak is the only author for this article. He designed the experimental study, performed both the in vivo and in vitro animal experiments, analyzed semen analysis, supervised the technical staff, measured some biochemical and molecular parameters, analyzed the data, graphed the figures, wrote the initial draft, and finalized the manuscript.