Silymarin protects from varicocele-induced damages in testis and improves sperm quality: evidence for E2f1 involvement

Abstract

This study was designed to evaluate the protective effect of silymarin (SMN) on varicocele-induced damage in testis and its effects on sperm parameters and on antioxidant status. Wistar rats were divided into three groups: control-sham, varicocele-induced, and SMN-treated varicocelized (50mg/kg, orally) rats. The sperm count, DNA integrity, and histone-protamine transition was evaluated after 42 days. The antioxidant status was analyzed by determining testicular malondialdehyde (MDA) and total thiol molecules (TTM). The endocrine status of the testicular tissue was estimated by counting the normal Leydig cell distribution/mm² and by determination of serum testosterone. The expression of E2f1 mRNA was analyzed using RT-PCR. Carbohydrate depletion and lipid foci replacement in germinal cells were examined by histochemical analyses. Silymarin rehabilitated the varicocele-induced Leydig cell degeneration and testosterone reduction. In addition, SMN recovered the varicocele-induced reduction of TTM and lowered significantly (P < 0.05) the varicocele-elevated content of MDA. The SMN treatment resulted in a significant (P < 0.05) down-regulation of the VCL-up-regulated E2f1 mRNA. Silymarin-treated animals were protected from varicocele-induced testicular atrophy and these animals showed a significant (P < 0.05) increase in the percentage of seminiferous tubules with positive tubular differentiation, repopulation, and spermiogenesis indices. Furthermore, SMN improved the varicocele-induced carbohydrate reduction in germinal cells. Our data suggest that in addition to oxidative stress, alteration in the testicular endocrine function plays a crucial role in the pathogenesis of varicocele. Moreover, the protective effects of SMN on varicocele-induced damage may reflect its antioxidant property, which may be mediated via the E2f1 transcription factor.

Keywords:

• oxidative stress
• silymarin
• spermatogenesis
• varicocele

Introduction

Varicocele (VCL) is characterized with abnormal tortuosity of the veins of the pampiniform plexus, which drains the testicular tissue [French et al. 2008]. This impairment is one of the physical causes of infertility in men. Although the exact pathogenesis of VCL is not fully understood, elevated testicular temperature, severe hypoxia by venous stasis and small vessel occlusion, damaged Leydig cells, remarkable reduction in testosterone level, and androgen receptor defects have been shown as possible causes [Hendin et al. 1999; Naughton et al. 2001].

Testicular inflammation and infiltration of polymorphonuclear leukocytes have also been reported in varicocelized testes [Evers and Collins 2008]. Venous stasis-induced hypoxia, iNOS up-regulation, malondialdehyde (MDA), and nitric oxide (NO) generation in the varicocelized testes have been previously reported [Romeo et al. 2003; Turkyilmaz et al. 2004].

The plasma membrane of mammalian spermatozoa consists of a high level of phospholipids, and saturated and unsaturated fatty acids [Saalu et al. 2009] which makes them susceptible to increased reactive oxygen species (ROS). At the same time damaged germinal epithelium, abnormal spermatozoa, and apoptotic spermatogenic cells significantly elevate ROS production in the testicular tissue. There are various studies indicating that varicocele-enhanced ROS products are cytotoxic and cause testicular damage [French et al. 2008; Saalu et al. 2009; Razi et al. 2010].

The spermatogenic lineage is characterized by discrete stages of mitosis and meiosis in which carbohydrates play an essential role in supplying energy. Lack of energy can suppress mitosis and ultimately reduce sperm production. We recently observed that in the rodent model of VCL, the cells involved in spermatogenesis exhibited decreased cytoplasmic carbohydrate [Razi et al. 2011].

The transcription factor E2F1 is a potent inducer of apoptosis. Endogenous E2f1 promotes apoptosis in vivo and in vitro E2f1 enhances various types of apoptosis. Overexpression of E2f1 sensitizes cells to apoptosis. E2f1-dependent apoptosis can be activated by chemotoxic agents. Previous reports have indicated that E2f1 can promote both cell cycle progression and apoptosis and dysregulated E2f1 can cooperate with P53 to induce apoptosis [Hallstrom and Nevins 2003]. E2f1 enhances the P53-dependent apoptosis through the induction of the P53 co-factors and the up-regulation of procaspases [Zhang et al. 2010]. Varicocele-related apoptosis has been recognized as one of the pathways to male infertility, and the expression profile of E2f1 in the rodent model should help clarify its role in varicocele.

Previous reports indicate that varicocelectomy has been identified as an appropriate method in order to treat infertility in varicoceles [Mostafa et al. 2001; Hurtado de Catalfo et al. 2007]. There are, however, reports indicating that a significant number of human patients remain infertile after varicocelectomy [Sofikitis and Miyagawa 1992; Shiraishi and Naito 2007]. This leads us to seek pharmacological agents that can reduce VCL-induced injuries. Antioxidants have been used successfully to reduce oxidative stress related injuries in many organs including the testis [Suziki and Sofikitis 1999; Ozokutan et al. 2000].

Silymarin (SMN) is a flavonoids complex extracted from seeds of the milk thistle (Silybum marianum). Silibinin (SBN) is the major active substituent of silymarin. Silymarin currently is used successfully for liver disorders and also as an anticancer and anti-inflammatory agent [Anjeli et al. 2010; Mata-Santos et al. 2010]. It acts as a potent antioxidant by reacting with ROS, and also potentiates the effects of the physiological antioxidants such as glutathione and superoxide dismutase [El-Shitany et al. 2008; Nencini et al. 2007]. Previous reports have also shown the protective effect of SMN on DOX-induced cardio- and hepatotoxicity in rat [Malekinejad et al. 2012a; Rašković et al. 2011]. The present study was designed to evaluate the protective effect of SMN on VCL-induced histochemical alterations in the testis, DNA damage in sperm cells, antioxidant status, and VCL-induced changes in E2f1 expression in the testes.

Results

General characteristics of SMN treated animals

After 42 days of VCL induction, body weight gain did not change statistically (P > 0.05) in the valicocelized and SMN-treated groups compared to the control-sham group. The testicular to body weight ratio in the VCL group showed a remarkable decline compared to the control-sham group, while SMN administration resulted in a significant (P < 0.05) improvement in testicular weight (Supplemental Fig. 1).

Figure 1.  Effect of silymarin (SMN) on the varicocel-induced changes in serum level of testosterone. A) SMN administration significantly (P < 0.05) increased the serum concentration of testosterone; B) positive correlation between the number of Sudan Black-B positive Leydig cells and serum level of testosterone (r² = 0.973, P < 0.05); and C) mean average of Sudan Black-B positive Leydig cell number per one mm² of the interstitial connective tissue. ^a: significant difference (P < 0.05) between the varicocele-induced and control-sham group, ^b: remarkable difference (P < 0.05) between the varicocele-induced and SMN-treated groups (n = 8)

Biochemical analyses revealed that in the VCL-induced animals, serum testosterone declined remarkably, while the SMN-treated group showed significantly (P < 0.05) higher levels of testosterone in comparison to the non-treated VCL-induced group (Fig. 1A). Statistical analyses revealed that there was a positive correlation between the lipid-positive Leydig cell number per mm² with serum levels of testosterone in different groups (Fig. 1B, C).

The sperm parameters are summarized in Table 1. The VCL-induced animals showed a significant (P < 0.05) decrease in sperm count in comparison to the SMN-treated group. Aniline-blue staining for chromatin condensation revealed that animals in the SMN-treated group exhibited a (P < 0.05) higher percentage of sperm with nuclear maturity (74.03 ± 2.19) versus the VCL-induced non-treated group (66.25 ± 3.94). No changes in sperm count and sperm chromatin condensation were observed in the control-sham animals. Fluorescent staining for DNA damage indicated that VCL-induced sperm damage shown in Figure 2, was significantly attenuated by SMN administration (P < 0.05). The eosin-nigrosin staining for sperm viability showed that the percentage of dead sperm was significantly (P < 0.05) reduced in the SMN-treated group and that SMN-treatment also attenuated the effects of VCL on sperm mobility.

Figure 2.  Representative profile of sperm shape in various staining techniques. A) light green fluorescent architecture from normal sperm with double strand DNA; B) light yellow fluorescent from sperm with partly denatured single-stranded DNA; C) reddish fluorescent from sperm with single strand DNA; D) sperm with immature chromatin condensation; E) sperm with condensed chromatin; F) live sperm with colorless cytoplasm; and G) dead sperm with stained cytoplasm. A-C) acridine-orange staining, D-E) aniline-blue staining, and F-G) eosin-nigrosin staining

Table 1. Sperm parameters.

Parameters	Control-sham	Varicocelized	SMN-treated
Sperm count (*10^6)	67.80 ± 2.28	44.50 ± 2.25^a	50.33 ± 1.63^b
Chromatin condensation (%) 87.21 ± 2.63		66.25 ± 3.94^a	74.03 ± 2.19^b
DS-DNA (%)	85.50 ± 3.87	64.25 ± 3.30^a	78.75 ± 4.11^b
PSS-DNA (%)	9.75 ± 1.70	7.50 ± 1.29^a	11.54 ± 1.91^b
SS-DNA (%)	6.00 ± 1.41	28.75 ± 4.11^a	11.50 ± 1.95^b
Sperm Motility (%)	92.8 ±3.1	21.0 ± 2.5^a	49.1 ± 3.9 ^b
Sperm Viability (%)	91.8 ± 2.1	58.5 ± 3.1^a	69.5 ± 1.1^b

Average percentage of sperm count, viability, motility, chromatin condensation, and DNA double strands integrity in different groups. ^a,b: significant differences (P < 0.05) between data in the same row. All data are presented in Mean ± SD. SMN: silymarin; DS-DNA: double strand DNA; PSS-DNA: partly denatured single stranded-DNA; SS-DNA: single stranded- DNA

The level of testicular MDA (4.8 ± 0.13 nmol/mg) significantly increased (Supplemental Fig. 2A) in the VCL-induced animals in comparison to the control-sham (0.85 ± 0.13 nmol/mg) and SMN-administrated groups (2.45 ± 0.15 nmol/mg). The total thiol content was reduced significantly (P < 0.05) when compared to the control-sham animals, while the SMN treatment maintained TTM in the testis (Supplemental Fig. 2B).

SMN regulated VCL-induced E2f1 expression in the testis

The mRNA level of E2f1 was determined using a semi-quantitative real time PCR technique and the results were normalized against the mRNA level of GAPDH. E2f1 was expressed in the testis of adult and healthy rats. After 42 days of VCL-induction the level of E2f1 mRNA increased significantly (P < 0.05). In comparison the level of E2f1 mRNA in the animals that received SMN (50 mg/kg for 6 weeks) showed a significant (P < 0.05) reduction (Supplemental Fig. 3).

Figure 3.  Microphotogram from rat testes. A) control-sham group: the seminiferous tubules with positive tubular differentation and spermiogenesis indexes (SP) with no edema in connective tissue are observed; B) varicocele-induced non-treated group: seminiferous tubules are presented with arrested spermatigenesis. A severe edema in connective tissue (SE) and decreased number of Leydig cells are shown; C) silymarin (SMN)-treated group: the seminiferous tubules are exhibiting normal spermatogenesis with positive tubular differentiation and spermiogenesis (SP) indices. An improved edema (IE) in connective tissue are seen; D) higher magnification from seminiferous tubules with spermatogenesis arrest (SA) on varicocelized testes: the giant cell (head arrow); and E) the tubule from varicocele-positve testes exhibites negative tubular differentiation index (TDI). The spermiogenesis from previous stages are presented with arrow. Iron-Weigert staining, A-C: 400× magnification, D-E: 600× magnification

SMN reduced VCL-induced histopathological damages

As summarized in Table 2, histopathological analyses demonstrated that in the VCL-induced animals, 25.50 ± 1.56% of the seminiferous tubules displayed dissociated germinal epithelium. In comparison, this was reduced to 15.83 ± 1.47 % in the SMN-treated group. The VCL-induced animals showed significantly (P < 0.05) higher tubular depletion in comparison to the SMN-treated and control-sham animals (Fig. 3). The tubular negative TDI remarkably (P < 0.05) increased in the VCL-induced group (23.00 ± 1.54%) versus the SMN-treated (13.34 ± 2.33%) and control-sham (3.00 ± 3.70%) animals. SMN administration significantly (P < 0.05) decreased the percentage of tubules with negative spermiogenesis index (SPI), while animals in the VCL group showed a remarkably higher (P < 0.01) percentage of tubules with negative SPI. Staining directed to nuclei of active and inactive spermatogonia (RI: repopulation index) showed that, the percentage of tubules with negative RI significantly (P < 0.05) increased in the VCL-induced group (25.12 ± 1.41%) in comparison to the SMN-treated (18.50 ± 1.51%) and control-sham (5.20 ± 1.09) animals.

Table 2. Summary of histological analysis.

Parameters	Control-sham	Varicocelized	SMN-treated
T.D (%)	0.02 ± 0.44	8.00 ± 2.60^a	3.16 ± 0.75^b
G.E.D (%)	0.05 ± 0.02	25.50 ± 1.56^a	15.83 ± 1.47^b
Negative TDI (%)	3.00 ± 0.70	23.00 ± 1.54^a	13.34 ± 2.33^b
Negative RI (%)	5.20 ± 1.09	25.00 ± 1.41^a	18.50 ± 1.51^b
Negative SPI (%)	5.00 ± 1.51	41.16 ± 1.16^a	27.66 ± 2.60^b
Leydig cells	27.20 ± 1.78	17.66 ± 1.50^a	21.00 ± 1.78^b
(no/ mm^2)
Sertoli cells	31.80 ± 1.78	17.33 ± 1.36^a	21.36 ± 1.35^b
(no/mm^2)

Average of histomorphometric analyses for tubular depletion (T.D), germinal epithelium dissociation (GED), tubular differentiation index (TDI), repopulation index (RI), spermiogenesis index (SPI), and Leydig cells distribution in different groups. SMN: silymarin; ^a,b: indicate significant differences (P < 0.05) between data in the same row. All data are presented as Mean ± SD.

The testes of VCL-induced animals showed severe edema in interstitial connective tissue, while antioxidant therapy (SMN-administration) attenuated the edema of connective tissue. The number of Leydig cells per mm² of the interstitial tissue decreased in the VCL group, while the SMN administration resulted in a significantly higher Leydig cell survival. Accordingly the SMN-administrated animals showed more than 21 cells/mm² versus approximately 17 cells in the VCL-induced and 27 cells in the control-sham animals. Comparing Sertoli cell number/ seminiferous tubule showed that, in the VCL-induced group the number of Sertoli cells significantly (P < 0.05) decreased, while the animals treated with SMN were observed with a remarkably higher number of Sertoli cells per tubule. No histopathological alterations were found in the control-sham animals for Sertoli cell number.

Effect of SMN on intracytoplasmic carbohydrate depletion and lipid foci

Histochemical carbohydrate analyses showed that the cells of the first three layers in the seminiferous tubules faintly reacted with periodic acid shift (PAS) staining in the VCL-induced animals, while the same layers in the SMN-treated group showed an intensive reaction against carbohydrate staining (Fig. 4). Unlike the VCL-positive animals (8.83 ± 1.94), the rats in the SMN-treated group showed a significantly (P < 0.05) higher number of PAS positive Sertoli cells (19.52 ± 3.08) per tubule. No changes were observed in the number of Sertoli cells in the control-sham animals (24.60 ± 2.19). Assessment of PAS positive Sertoli cells (cells with normal supplementation of intracytoplasmic carbohydrate) with a PAS positive spermatogenesis cell series clarified that there was a positive correlation between decreased PAS positive Sertoli cells and a weak reaction for carbohydrate staining in the first three layers of the VCL-induced rats (Supplemental Fig. 4A,B). In the varicocele-induced animals spermatogonia and spermatocytes exhibited dense lipid staining. In comparison the SMN-administrated group showed a faint reaction for lipid staining (Fig. 5). As summarized in Supplemental Table 1 the animals in VCL-induced group manifested a significantly (P < 0.05) lower number of Leydig cells with intracytoplasmic lipid accumulation versus the animals in the SMN-treated group.

Figure 4.  Microphotogram from rat testes. A) control-sham group: a dense periodic acid shift (PAS) reaction in spermatogenesis cells lineage cytoplasm is presented; B) varicocele-induced group: in * marked tubules the first three layers are presented with faint PAS reaction sites indicating a severe tubular depletion (TD); and C) silymarin (SMN)-administrated group: a protected intracytoplasmic carbohydrate content in spermatogenesis cells series is shown. PAS staining, A-C: 400× magnification

Figure 5.  Effect of silymarin (SMN) on varicocele-induced alteration in energy source of the spermatocytogenesis and spermatogenesis cell series. A) control-sham group: the spermatocytogenesis and spermatogenesis cell series are presented with faint reaction against lipid staining (black arrow), while the upper layers (post meiosis cells series) are exhibited with intensive intacytoplasmic lipids accumulation (yellow arrow); B) varicocele-induced group: depleted tubules are presented with intensive lipid positive cells. Note giant cells with dark lipid stained cytoplasm (red arrows); C) SMN-administrated group: note faint lipidophilic sites in first three layers (red arrow) of tubules and dense Sudan black-B stained upper layers; D) higher magnification from testes of the control-sham group: Leydig cells are presented with dark stained cytoplasm; E) higher magnification from varicocelized testis: note faint Sudan black-B stained Leydig cells; and F) improved Leydig cells biosteroidogeneis in the SMN-treated group, which are exhibiting dense lipidophilic cytoplasm. Sudan black-B staining, A-C: 400 × , D-F: 600× magnification

Discussion

This study showed that SMN decreased the VCL-induced damage in testicular tissue and improved sperm parameters. The SMN-treated group showed significantly increased testosterone with elevated TTM and reduced MDA content. Animals in the SMN-treated group exhibited improved sperm parameters such as sperm count, percentage of nuclear maturity, and DNA damage. Moreover, the up-regulation of E2f1 mRNA in the VCL-induced animals appeared regulated by SMN administration. Additionally, the SMN-treated animals showed improved TDI, RI, SPI, and increased carbohydrate levels in germinal epithelium as essential basic factors in a healthy testis tissue.

Reactive oxygen species generation positively correlates with damaged germinal cells and abnormal spermiogenesis, which presents with an increased number of morphologically abnormal sperm [Griveau and Lannuo 1997; Padron et al. 1997]. The antioxidant status analyses in this study showed that after SMN administration the TTM level significantly increased. This suggests that SMN administration remarkably decreased ROS stress by reducing damaged germinal cells and spermatozoa. The lower percentage of morphologically abnormal and dead sperm in the SMN-treated animals and an improvement in the testicular tissue structure would be supported by recovered antioxidant status. Elevated lipid peroxidation has been known as a pathophysiologic derangement in VCL patients [Hendin et al. 1999; Shiraishi and Naito 2007]. It has been identified that the elevated oxidative stress in VCL patients leads to severe sperm cell membrane peroxidation and ultimately mortality [Saalu et al. 2009; Razi et al. 2011]. The imbalance in the oxidative system results in an intensive decrease in axonemal proteins phosphorylation, which in turn leads to sperm immobility [Kessopoulou et al. 1992]. Sperm quality analyses in the current study showed that the percentage of dead sperm significantly decreased in the SMN-administrated group, while the motile sperm percentage was ameliorated. The level of MDA was reduced in the SMN-treated animals, suggesting that SMN administration resulted in sperm viability and motility improvement by promoting antioxidant defence. Previous studies demonstrated that incomplete and/or defective chromatin condensation and lipid peroxidation of sperm membrane leads to intensive sperm DNA fragmentation in VCL patients [Saleh et al. 2003]. Moreover, it has been suggested that severe sperm DNA damage positively correlates with VCL-induced oxidative stress [Smith et al. 2003; Moskovtsev et al. 2009]. Special staining for double stranded DNA showed that the percentage of sperm with partly denatured single-stranded-DNA and SSDNA significantly decreased in the SMN-administrated group. Simultaneously the percentage of sperm with condensed chromatin remarkably increased in the SMN-treated animals. Therefore, these findings suggest that SMN protects the sperm from DNA damage both by stimulating the chromatin condensation and also by attenuation of the DNA denaturation.

We in this study and other researchers previously showed that the transcription factor E2f1 is expressed in the rat testis [El-Darwish et al. 2006]. Additionally, we found that in the varicocelized animals the mRNA level of E2f1 was up-regulated significantly (P < 0.05). One should note the possible pathways, which may alter the E2f1 expression and that hypoxia seems to be the most likely. In this study we did not show the direct link between hypoxia and its effect on E2f1 expression in varicocelized animals and this is considered a limitation. However, increasing evidence indicates that hypoxia is one of the key influential factors in the VCL-induction [Gat et al. 2005]. Similarly, other studies have demonstrated that hypoxia-induced cyclin-dependent protein kinase activation promotes cardiomyocyte apoptosis via the induction of E2f1 [Hauck et al. 2002; Angelis et al. 2011]. Assuming that due to exceeding the venous blood pressure, in arterioles, that could result in the hypoxic condition [Gat et al. 2005]; we may conclude that the E2f1 up-regulation likely is mediated through the hypoxic condition in the varicocelized testis. Moreover, as shown above SMN was able to regulate the VCL-induced E2f1 up-regulation, which may be related to its potent antioxidant effects. Additionally, we recently reported that SMN has the capability to reduce remarkably the hypoxia induced damages in the rainbow trout brain [Malekinejad et al. 2012b].

In the current study our findings showed that following the SMN administration VCL-induced spermatogenic arrest diminished and spermiogenesis increased significantly, as the morphometric studies in the testes of SMN-treated group showed an increased percentage of tubules with positive TDI, RI, and SPI versus the VCL-induced non-treated rats. It has been shown that testes subject to long-term VCL-exposure loose Leydig cells, leading to severe reduction in testosterone [Razi et al. 2011]. As testosterone controls the Sertoli cells physiologic function, the VCL-induced reduction in testosterone level enhances the Sertoli cells degeneration. Therefore, it might be suggested that the SMN administration resulted in Leydig cell survival and consequently a remarkable increase in testosterone and Sertoli cell physiological activities. Our recent study on the long-term effects of induced VCL showed the profound changes in the germinal cell intracytoplasmic carbohydrate level, indicating a significant decrease in spermatogonia cell carbohydrate supplementation [Razi et al. 2011]. Similar findings were observed in the current study after 42 days of VCL induction. The animals in the VCL-induced group exhibited a faint reaction for PAS staining in spermatogonia and spermatocytes, while the SMN-treated group showed densely stained cells in the first three layers of germinal epithelium, suggesting that SMN protected from glucose depletion. The other novel finding of this study was that, in the VCL-induced group the germinal cells changed the energy source from glucose to lipids, which were identified by increased lipid accumulation in cells of the spermatogenic lineage. Therefore, the severe lack of energy promotes the degenerative processes in VCL and antioxidant therapy (SMN administration) can protect from destructive events by preventing glucose depletion and consequently lipid accumulation. The data presented showed that SMN was able to moderate varicocele-induced damage on testicular tissue and on sperm parameters. The SMN protected the testicular tissue against the VCL-induced derangement by elevating tubular TDI, RI, and SPI along with increasing the sperm chromatin condensation, DNA integrity, and motility. The mechanism for the SMN-induced protective activities may be attributed to its antioxidant potential, which was represented by the reduction of VCL-increased MDA and by enhancing the VCL-reduced TTM concentrations in the testis (Fig. 6). This study also showed the regulatory effect of SMN on VCL-induced up-regulation of E2f1 transcription factor as a novel therapeutic target in VCL patients.

Figure 6.  Mechanism by which silymarin (SMN) protects testicular tissue in the varicocelized rats. Varicocele decreases testosterone level by Leydig cell degeneration that result in a remarkable loss of Sertoli cell physiologic function. Furthermore an increase in germinal cell apoptosis and abnormality in spermiogenesis occurred. In addition, varicocele (VCL)-induced decrease in intracytoplasmic carbohydrate lowers the cellular essential energy for mitosis. These impairments increase reactive oxygen species (ROS) stress, which in turn propel normal cells through apoptosis. SMN protects testicular tissue by both reducing cellular damages and by a remarkable decrease in ROS level.

Materials and Methods

Chemicals

Silymarin standard (S 0292, containing 80% silibin), guanidine hydrochloride, and 5.5'-Dithiobis-2-nitrobenzoic acid (DTNB), were purchased from Sigma Chemical Co. (St. Luis, MO, USA). Thiobarbituric acid, phosphoric acid (85%), trichloracetic acid (TCA), dimethyl sulfoxide (DMSO) and ethanol were obtained from Merck (Germany). N-butanol was obtained from Carl Roth, GmbH Co. (Germany). Commercially available standard kit was used for the determination of alkaline phosphatase (ALP, 744, Man Inc. Tehran, Iran). All other chemicals were commercial products of analytical grade.

Animals and experimental design

Twenty four male Wistar rats (200-220 g) were obtained from the animal resource of the Faculty of Veterinary Medicine, Urmia University. The rats were in good health. The animals were acclimatized for a week and had free access to food and water. The experimental protocols were approved by the ethical committee of Urmia University in accordance with principles of laboratory animal care. The animals were assigned to control-sham, varicocelized (VCL) non-treated, and SMN-treated (50 mg/kg, orally) groups (n = 8). The rats in the control-sham group received only normal saline during the experimental period. The varicocelized animals were divided into two subgroups and the first group (VCL) also received normal saline and the rats in the last group (VCL⁺SMN⁺) after the VCL induction received SMN (50 mg/kg, orally) for 42 d.

Varicocele induction

The left varicocele was induced as previously reported [Pasqualotto et al. 2003]. Briefly, the animals were anaesthetized with intra-peritoneal injection of 7mg/kg ketamine hydrochloride 5% (Trittau, Germany) and 5mg/kg xylazine 2% (Woerden, The Netherlands). To reduce the diameter of the renal vein to approximately 1 mm, the left renal vein ligation was performed at a direct medial to the junction of the adrenal and spermatic veins. Then the anastomotic branch between the left testicular vein and the left common iliac vein was ligated.

Autopsy, organ weight

Following 42 d the rats were weighed and euthanized by using a special CO₂ device (Uromdaco, Iran). The left side testicular tissues were excised, dissected free from surrounding tissues under the high magnification (40×) provided by a stereo zoom microscope (model TL2, Olympus Co., Tokyo, Japan) and their weight was determined. Dissected testes samples were washed with chilled saline normal and half of the specimens were fixed in bouin's fixative and kept for further histological analyses. The remaining samples were immediately frozen and stored at -70°C for further biochemical analyses.

Assessment of serum levels of testosterone

Blood samples were collected and serum samples were prepared with centrifugation (3000 g for 5 min), and the serum level of testosterone was assessed by using a competitive chemiluminescent immunoassay kit (PishtazTeb, Iran).

Evaluating the epididymal sperm parameters

The epididymis was separated carefully from the testicles under a 20-times magnification provided by a stereo zoom microscope (model TL2, Olympus Co., Tokyo, Japan). The epididymis was divided into 3 segments; head, body, and tail. The epididymal tail was trimmed and the content was added to 5 mL pre-warmed Hams F10 medium. After 20 min the epididymal tissue was separated from the released spermatozoa. The sperm count was performed according to standard hemocytometric method as described previously by Pant and Srivastava [2003].

Assessment of sperm chromatin condensation

The aniline-blue staining was performed in order to evaluate sperm chromatin condensation [Terquem and Daduone 1983]. Briefly, after sperm preparation, 5 µl of the prepared spermatozoa were spread onto glass slides, and allowed to dry. The smears were fixed in 3% buffered glutaraldehyde in 0.2 M phosphate buffer (pH 7.2) for 30 min. Slides were then stained with 5% aqueous aniline blue mixed with 4% acetic acid (pH 3.5) for 5 min and then 100 sperm cells per slide were evaluated and the percentage of unstained sperm heads was calculated. Two classes of staining intensities were distinguished: unstained and completely stained. The percentage of unstained and stained sperm was compared between the study groups.

Sperm motility, viability, and abnormality

The sperm motility was examined based on the WHO [1999] standard method for manual examination of sperm motility. Briefly, the sperm samples were diluted 1:8 in Ham's F10 before examination. A 20µl of sperm sample was placed on sperm examination area and examined under 10× magnification loop. Only the motile sperm with forward progression counted within 10 boxes and recorded. Finally, motility was evaluated based on the following equation:

[Motility (%) = motile sperm/motile+ non-motile sperm] × 100.

The eosin-nigrosin staining was performed for sperm viability assay [WHO 1999]. In brief, 50µl of epididymal sperm was mixed with 20 µl of eosin in sterile test tube. After 5 s 50 µl of nigrosin was added and mixed thoroughly. The mixture of stained sperm was smeared on the slide and examined under bright field microscope (1,000× magnification, Olympus, Germany). The colorless sperm were considered as live and stained sperm were marked as dead spermatozoa. A hundred spermatozoa from each animal were counted and the abnormal (abnormalities on head and tail) and normal sperm were identified. The sperm viability, motility, and morphology were reported in percentage.

Acridine orange (AO) staining for sperm DNA strands

The AO staining was performed in order to estimate the sperm DNA fragmentation [Tejada et al. 1984]. Briefly, air dried slides were stained for 10 min with freshly prepared AO (0.19 mg/ml), washed in distilled water, then the cover-slip was applied on the slides. The slides were evaluated on the same day using epi-fluorescent microscope (Model GS7, Nikon co., Japan). In all preparations, at least 100 spermatozoa were evaluated at 40× magnification. Spermatozoa with green fluorescence were considered to have native double strand DNA (DS-DNA) and the spermatozoa with yellow fluorescence were marked as having partly denatured single-stranded-DNA (PSS-DNA), and with red fluorescence as completely denatured single-stranded- DNA (SS-DNA). Percentage of green, yellow, and red spermatozoa were assessed and compared between the groups.

Testicular MDA analyses

To determine the lipid peroxidation rate, MDA content of the collected testis samples was measured using the thiobarbituric acid (TBA) reaction as described previously [Niehaus and Samuelsson 1968]. In short, 0.3-0.4 g of the testis samples were homogenized in ice-cold KCl (150 mM), and then the mixture was centrifuged at 3,000 × g for 10 min. Thereafter 0.5 ml of the supernatant was mixed with 3 ml phosphoric acid (1% v/v) and then following vortex mixing, 2 ml of 6.7 g L⁻¹ TBA was added to the samples. The samples were heated at 100°C for 45 min, then chilled on ice. Finally, 3 ml N-butanol was added and the samples were further centrifuged at 3,000 × g for 10 min again. The absorbance of supernatant was measured spectrophotometerically at 532 nm and the MDA concentration calculated according to the simultaneously prepared calibration curves using MDA standards. The amount of MDA was expressed as nmol per mg protein of the samples. The protein content of the samples was measured according to the Lowry method [Lowry et al. 1951].

Measurement of total thiol molecules (TTM)

Total sulfhydryl level in the testes samples was measured as described previously [Hu and Dillared 1994]. Briefly, 0.3-0.4 g of the testes samples was homogenized in ice-cold KCl (150 mM), and the mixture was centrifuged at 3,000 × g for 10 min. To 0.2 ml of the supernatant of the tissue homogenate, 0.6 ml Tris-EDTA buffer (Tris base 0.25 M ethylendiamintetraacetic acid 20 mM, pH 8.2) and thereafter 40 µl 5.5'-Dithiobis-2-nitrobenzoic acid (10 mM in pure methanol) were added. The final volume of this mixture was made up to 4.0 ml by addition of pure methanol. After 15 min incubation at room temperature, the samples were centrifuged at 3,000 × g for 10 min and ultimately the absorbance of the supernatant was read at 412 nm. The TTM capacity was expressed as nmol per mg of protein in samples. The protein content of the samples was measured according to Lowry et al. [1951].

RNA isolation and RT-PCR

To evaluate the effect of varicocele induction and the effect of SMN on mRNA level of E2f1 in VCL-induced animals, total RNA was isolated from the testis samples using the standard TRIZOL method [Chomczynski and Sacchi 2006]. To avoid genomic DNA contamination extra care was taken when the colorless aqueous phase collected after chloroform extraction. The RNA amount was determined spectrophotometrically (260 nm and A260/280 = 1.8-2.0), and the samples were stored at -70°C. For RT-PCR, cDNA was synthesized in a 20 µl reaction mixture containing 1µg RNA, oligo(dT) primer (1µl), 5× reaction buffer (4µl), RNAse inhibitor (1µl), 10mM dNTP mix (2µl) and M-MuLV reverse transcriptase (1µl) according to the manufacturer's protocol (Fermentas, GmbH, Germany). The cycling protocol for 20 µl reaction mix was 5 min at 65°C, followed by 60 min at 42°C, and 5 min at 70°C to terminate the reaction.

Second strand cDNA synthesis

The RT-PCR reaction was carried out in a total volume of 25µl containing PCR master mix (12.5 µl), FWD and REV specific primers (each 0.75 µl), and cDNA as a template (1µl) and nuclease free water (10 µl). PCR conditions were run as follows: general denaturation at 95°C for 3 min, 1 cycle, followed by 40 cycles of 95°C for 20s; annealing temperature (63°C for GAPDH and 58°C for E2F1) for 30s; elongation: 72°C for 1 min and 72°C for 5 min. The products of RT-PCR were separated on 1.5 % agarose gel containing ethidium bromide and visualized using Gel Doc 2000 system (Bio-Rad). The specific primers for Ratus E2f1 and GAPDH were designed [Wong et al. 2011; Shibata et al. 1999] and manufactured by CinnaGen (CinnaGen Co. Tehran, Iran). The E2f1 forward and reversed primers were TTGACCCCTCTGGATTTCTG and CCCTTTGGTCTGCTCAATGT respectively yielding a 198 bp product and the GAPDH forward and reversed primers were GTTACCAGGGCTGCCTTCT and GGGTTTCCCGTTGATGACC, respectively yielding a 167 bp product.

Histological analyses

The testicular samples were paraffin embedded and cut (5-6µm) by rotary microtome (Microtome GmbH, Germany). The sections (5-6 µm) were stained with Iron-Weigert (Pajohesh Asia., Iran) for detection of germinal cell nucleuses in the testis and were analyzed under light microscope by multiple magnifications (400× and 1,000×).The Leydig cell number per one mm² of the interstitial connective tissue was counted and compared between groups.

Tubular differentiation (TDI), repopulation (RI), spermiogenesis (SPI), and Sertoli cell indices

The percentage of seminiferous tubules with more than 3-4 germinal layers and percentage of tubules with normal spermiogenesis were considered as positive TDI and SPI. The percentage of tubules with positive RI, as the ratio of active spermatogonia (spermatogonia type B with light nucleus in Iron-Weigert staining technique) to inactive spermatogonia (spermatogonia type A with dark nucleus in Iron-Weigert staining technique), was estimated. The Sertoli cell number per one seminiferous tubule was counted. All mentioned parameters were analyzed in 20 prepared sections of left testes.

Estimating intra-cytoplasmic carbohydrate level in germinal epithelium

To evaluate the carbohydrate level, the PAS technique was conducted according to the manufacturer's instructions (Pajohesh Asia kits, Iran). Briefly, the paraffin sectioned slides were deparaffinized and hydrated. The hydrated slides were oxidized in 5% periodic acid solution for 5 min. After rinsing with distilled water the slides were placed in Schiff reagent for 15 min and washed with lukewarm water. After 5 min the slides were counterstained with Meyer's hematoxylin. The percentage of tubules with PAS positive stained spermatogenesis cell lineage was evaluated.

Intra-cytoplasmic lipid foci evaluation in germinal epithelium

To examine the lipid level of the tissue samples, the Sudan Black B (Pajohesh Asia kits, Iran) staining was performed. In this procedure, the fresh tissue specimens were sectioned with cryostat microtome (Bright 00361, England) and stained through a special technique separately. In short, the frozen section prepared specimens were picked-up on the clean glass slides, and the slides were fixed in 6% formalin freshly. Then the slides were washed well in tap water, rinsed with distilled water, and the excess water was drained. The slides were merged in propylene glycol for 5 min and passed to Sudan Black B solution. Thereafter, the 85% propylene glycol was used for sinking the slides for 7 min. After rinsing the slides with distilled water, the nuclear fast-red staining solution was used for contrast. The percentage of tubules with Sudan Black B stained spermatogenesis cell lineage was estimated.

Statistical analyses

For the measured parameters, mean and standard deviations were calculated. Results were analyzed using Graph Pad Prism software (version 2.01. Graph Pad software Inc. San Diego, California, USA). The comparisons between groups were made by analysis of variance (ANOVA) followed by Bonferroni post-hoc test. A P value < 0.05 was considered significant.

Supplemental Table 1. Average percentage of seminiferous tubules with PAS and Sudan Black B stained spermatogenesis cell series in different groups.

Parameters (%)	Control-sham	Varicocelized	SMN-treated
PAS positive	98.00 ± 9.01	52.83 ± 6.72^a	73.16 ± 5.04^b
Sudan Black B positive	7.12 ± 1.23	40.16 ± 3.72^a	28.83 ± 6.48^b

^{a b} indicate significant differences (P < 0.05) between data in the same row. All data are presented as Mean ± SD.

Abbreviations

AO:	=	acridine –orange
DS-DNA:	=	double strand DNA
MDA:	=	malondialdehyde
NO:	=	nitric oxide
PAS:	=	periodic acid shift
PSS-DNA:	=	partly single strand-DNA
ROS:	=	reactive oxygen species
RI:	=	repopulation index
SBN:	=	silibinin
SMN:	=	silymarin
SPI:	=	spermiogenesis index
SS-DNA:	=	single strand-DNA
TBA:	=	tthiobarbituric acid
TDI:	=	tubular differentiation index
TTM:	=	total thiol molecules
VCL:	=	varicocele.

Acknowledgments

We wish to thank Mr. Ali Karimi and Mr. Hamed Tabatabaie, and the staffs of Histology and Pharmacology laboratories for their kind technical support.

Declaration of interest: The authors report no declarations of interest.

Author contributions: Designed and performed the experiments and wrote the manuscript: HM; Performed the experiments: MSM; Performed the experiments and wrote the manuscript: MR; Analyzed the data: VS-I.