The Effect of Cordycepin on Steroidogenesis and Apoptosis in MA-10 Mouse Leydig Tumor Cells

Cordycepin is a natural pure compound extracted from Cordyceps sinensis (CS). We have demonstrated that CS stimulates steroidogenesis in primary mouse Leydig cell and activates apoptosis in MA-10 mouse Leydig tumor cells. It is highly possible that cordycepin is the main component in CS modulating Leydig cell functions. Thus, our aim was to investigate the steroidogenic and apoptotic effects with potential mechanism of cordycepin on MA-10 mouse Leydig tumor cells. Results showed that cordycepin significantly stimulated progesterone production in dose- and time-dependent manners. Adenosine receptor (AR) subtype agonists were further used to treat MA-10 cells, showing that A1, A 2A , A 2B , and A3, AR agonists could stimulate progesterone production. However, StAR promoter activity and protein expression remained of no difference among all cordycepin treatments, suggesting that cordycepin might activate AR, but not stimulated StAR protein to regulate MA-10 cell steroidogenesis. Meanwhile, cordycepin could also induce apoptotic cell death in MA-10 cells. Moreover, four AR subtype agonists induced cell death in a dose-dependent manner, and four AR subtype antagonists could all rescue cell death under cordycepin treatment in MA-10 cells. In conclusion, cordycepin could activate adenosine subtype receptors and simultaneously induce steroidogenesis and apoptosis in MA-10 mouse Leydig tumor cells.


Introduction
Cordyceps sinensis (CS) is an ingredient of traditional Chinese medicine and is prescribed for replenish the kidney and soothe the lung and for the treatment of fatigue [1]. Cordycepin (3 -deoxyadenosine, an adenosine analogue) is a pure component extracted from the mycelia of CS, and it is well known to possess anticancer ability which induce apoptosis in HeLa cells, oral cancer cells, breast cancer cells, leukemia, and lymphoma cell lines [2][3][4][5]. Previous studies demonstrated that CS alone could stimulate steroid production in both normal and tumor mouse Leydig cells [6,7] and activate apoptosis in MA-10 mouse Leydig tumor cells [8]. It is highly possible that cordycepin is the main component in CS modulating Leydig cell functions. Thus, the aim of the present study was to investigate the steroidogenic and apoptotic effects with potential mechanism of cordycepin on MA-10 mouse Leydig tumor cells.
Steroidogenesis, steroid hormone biosynthesis, occurs mainly in the adrenal glands, brain, placenta, testes, and ovaries [9]. In the male reproduction system, steroidogenesis in Leydig cells is regulated by luteinizing hormone (LH)/human chorionic gonadotropin (hCG). LH and hCG activate its cognate receptors and coupling to the adenylate cyclase (AC) through the heterotrimeric guanine nucleotidebinding regulatory protein (G-protein) [10,11]. The activated GTP-bound α subunit of G-protein would be able to activate adenylyl cyclase, which results in the hydrolysis of ATP to cyclic AMP. Once cAMP is synthesized, the following activation of protein kinase A (PKA) pathway would phosphorylate steroidogenic acute regulatory protein (StAR) [12]. The StAR protein, a 30 kDa phosphoprotein, is the rate-limiting step which delivers cholesterol from the outer to the inner mitochondrial membrane [13]. After translocation into mitochondrial, P450 side chain cleave enzyme (P450scc) converts cholesterol to pregnenolone [14]. When pregnenolone is formed, it may be metabolized to progesterone by mitochondrial 3β-hydroxysteriod dehydrogenase (3β-HSD), or it may exit the mitochondria and undergo further metabolism with the final steroid hormone product dependent upon the nature of the tissue [15,16].
It has been well demonstrated that adenosine acts through four G-protein-coupled membrane receptors, the A 1 , A 2A , A 2B , and A 3 adenosine receptors [17]. The adenosine 2 Evidence-Based Complementary and Alternative Medicine A 1 and A 3 receptors (A 1 -AR and A 3 -AR) inhibit adenylate cyclase via G i and activate phospholipase C (PLC) [18,19], and the adenosine A 2A and A 2B receptors (A 2A -AR and A 2B -AR) stimulate adenylyl cyclase via G s [20,21], respectively. It is also shown that adenosine appears to induce cell death through apoptosis, whereas ATP appears to cause both necrosis and apoptosis [22].
Although some reports have showed that cordycepin possesses anticancer ability, there is still no research about cordycepin on gonadal steroidogenesis. In fact, we have demonstrated that cordycepin could stimulate testosterone production in normal mouse primary Leydig cells without any phenomenon of cell death [23]. Since the isolation of primary Leydig cell is complicated with very low yield from mouse testis, MA-10 mouse Leydig tumor cell line was then used to further examine the regulatory mechanisms regarding the effects of cordycepin. Interestingly, we did find that cordycepin could induce MA-10 cell steroidogenesis and apoptosis. Thus, our objective was to investigate the effect of cordycepin on Leydig cell steroidogenesis and apoptosis with the preliminarily possible mechanisms.

Cell
Culture. The MA-10 cell line was a gift from Dr. Mario Ascoli (The University of Iowa, Iowa City, USA) and is maintained at 37 • C in a humidified environment containing 95% air and 5% CO 2 for all the following experiments. MA-10 cells (5 × 10 3 cells/100 μL medium) or (6 × 10 5 cells/2 mL medium) were plated into 96-well plates or 6 cm dish grown for 24 hr in Waymouth medium containing 10% fetal bovine serum, respectively. The medium was removed and the cells were washed twice with 1X PBS, and then treated with various concentrations of cordycepin in serum free Waymouth medium for indicated time periods. The cells were then isolated for total protein. The expression of target protein was determined by Western blot analysis. Cytotoxicity assay and cell morphological analysis of the MA-10 cells were determined by MTT assay. The media were withdrawn and progesterone levels were determined by radioimmunoassay. Finally, the adenosine receptor subtypes mRNA expression were performed by RT-PCR.

Morphology Study.
MA-10 cells were seeded in 96-well plate (Techno Plastic Products AG, Trasadingen, Switzerland) containing 5 × 10 3 cells with 100 μL serum medium in each well. After 70-80% confluence, cells were treated without or with 100 μM and 1 mM cordycepin for 24 hr, respectively. Cell morphology was then observed and recorded under light microscopy (Olympus, CK40). Apoptosis was characterized by the loss of cellular contact with the matrix and the appearance of plasma membrane blebbing.

MTT Cytotoxicity Assay.
Cordycepin-induced MA-10 cells cytotoxicity was determined by measuring mitochondrial succinate dehydrogenase activity using a modification of the MTT assay. After MA-10 cells reach 70%-80% confluence, cells were treated with serum free medium containing 100 μM and 1 mM cordycepin for indicated time points (24 and 48 hr). MTT was added to wells (0.5 mg/mL final concentration) after time points of culture at 37 • C in a 5% CO 2 humidified atmosphere in the presence of the desired reagents. Mitochondrial dehydrogenases of viable cells will convert MTT into a color-dense formazan. Four hours later, the DMSO was added in wells to dissolve the formazan. The DMSO solutions were added and the absorbance was measured at 590 nm by an ELISA reader (VersaMax, MDS Inc., Toronto, Canada).

DNA Fragmentation Assay.
Cells (1 × 10 6 ) were lysed in a 0.6 mL cell lysis solution containing 20 mM Tris-HCl, 10 mM EDTA, pH 8.0 and 0.3% Triton X-100. DNA was extracted with 0.6 mL phenol/chloroform (1 : 1), and the mixture was centrifuged at 12,500 rpm for 10 min. DNA in the aqueous phase was extracted with phenol/chloroform Evidence-Based Complementary and Alternative Medicine 3 (1 : 1) again. The aqueous (DNA containing) phase was mixed with isopropanol at −20 • C overnight. After centrifugation, DNA pellets were washed with 70% ethanol and air-dried. DNA pellets were dissolved in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0), and RNase A (3 mg/mL) was added to remove RNA at 37 • C for 30 min. DNA electrophoresis was carried out in 2% agarose gel. The gel was stained with ethidium bromide. DNA fragments were visualized by exposing the gel to UV light.

Reverse Transcription-Polymerase Chain Reaction (RT-PCR).
Total RNA was isolated from MA-10 cells using Trizol reagent as recommended by the manufacturer (Invitrogen, Carlsbad, Calif, USA). Reverse transcription was performed in a mixture containing 5 μM random primer, 200 μM dNTP, 2 U/μL MMLV reverse transcriptase together with 5 μL tRNA as the template. The corresponding buffer was performed at 42 • C for 90 min followed by 95 • C for 10 min. PCR was performed in a mixture containing 2 μL 10X PCR buffer, 0.4 μL 10 mM dNTP, 0.4 μL 20 mM specific forward and reverse primers (primer sequence and corresponding sequence of specific genes were listed in Table 1), 14.7 μL ddH 2 O, 0.1 μL 0.5 U Taq with 2 μL RT product as template for each reaction. Thermocontrolling program was set up as the following: denature at 95 • C for 30 sec, annealing at 55 • C for 30 sec, elongation at 72 • C for 30 sec with another 5 min of elongation at 72 • C. The whole mixture was subjected to 30 cycles for amplification of L19, 32 cycles for amplification of A 1 -AR, 34 cycles for amplification of A 2A -AR and A 2B -AR and 38 cycles for amplification of A 3 -AR. The PCR product was then separated on a 1.5% agarose gel at 120 V for 30 min in 1X TBE buffer (0.09 M Tris, 0.09 M boric acid, 0.001 M EDTA, pH 8.0). The gel was then stained with ethidium bromide for 10 min and destained with mili-Q water. The gel image was captured by using Labwork imager system (Digital CCD Camera, Hamamtsu Photonics system, Bridgewater, USA). (RIA). Media from cultures with different treatments were harvested. Twenty μL of sample was loaded into a glass tube and 100 μL each of progesterone antiserum and 3 H progesterone were loaded. Equilibrium reaction occurred at 37 • C for 30 min and was stopped by putting the tubes in ice. Charcoal was added into the tubes at 4 • C for 15 min and centrifuged for 10 min in order to spin down the charcoal-3 H progesterone complex. The supernatant was poured into 2 mL of scintillation fluid and samples would be counted in a β-counter (LS 5000TA, Beckman Inc., Fullerton, Calif, USA).

Effects of Cordycepin on Steroidogenesis in MA-10 Mouse Leydig Tumor Cells.
To test the hypothesis that cordycepin influences the production of steroid hormone in MA-10 mouse Leydig tumor cells; we initially determined the effect of cordycepin on the production of progesterone. MA-10 cells were incubated with different dosages (1, 10, 100 μM, and 1 mM) of cordycepin for 24 hr. Results showed that the progesterone production induced by 100 μM cordycepin was more than 3 folds significantly compared to the control (217.5 ± 101.7 versus 675.9 ± 185.4 pg/μg protein; P < .05) (Figure 1(a)). As shown in Figure 1(b), cordycepin at 100 μM significantly stimulated progesterone to a maximum at 24 hr (P < .05). According to the results, 100 μM cordycepin for 24 hr treatment was used to investigate the possible cellular mechanism.   about 3 folds (P < .05) (Figure 2(e)). However, both 100 μM and 1 mM cordycepin would downregulate A 2B -AR mRNA expression by 60% and 80%, respectively (P < .05) (Figure 2(d)).

Effects of Cordycepin on StAR Protein and Promoter
Expressions in MA-10 Mouse Leydig Tumor Cells. It has been previously shown that steroidogenesis induced by Cordycep sinensis in MA-10 cells requires de novo protein synthesis [27]. To further understand the mechanism of cordycepin-induced steroidogenesis in MA-10 cells, StAR protein expression was investigated by Western blotting analysis. MA-10 cells were incubated with different dosages (1∼100 μM) of cordycepin for 3 hr or 100 μM cordycepin for 1, 3, 6, and 12 hr, respectively. Results showed that the expression of StAR protein was not activated by cordycepin with different dosages (Figure 5(a)) and different time treatments ( Figure 5(b)) in MA-10 mouse Leydig tumor cells (P > .05).

Discussion
In this study, we demonstrate that cordycepin, a pure substance isolated from the Cordyceps sinensis, could stimulate steroidogenesis in MA-10 mouse Leydig tumor cells. Besides, cordycepin could induce the antitumor effect possibly through adenosine receptors in MA-10 mouse Leydig tumor cells.
First, we found that cordycepin increased the expression of A 1 -, A 2A -, and A 3 -AR mRNA but decreased the expression of A 2B -AR mRNA at 24 hr treatment. Under microscopic observation, we found that cordycepin-treated MA-10 cells exhibited cellular shrinkage and membrane blebbing, and finally cells detached from the dish. By MTT and DNA ladder assays, adenosine also significantly reduced MA-10 cell viability in a dose-dependent manner and induced DNA fragmentation. The effective concentration (EC50) of adenosine which could cause 50% inhibition of MA-10 cells growth was 5 mM after 48 hr treatment. However, we observed that the death effect of cordycepin was somewhat greater than adenosine in the same concentration (100 μM and 1 mM). These phenomena indicate that the apoptotic effect by adenosine in MA-10 cells was comparable.
Nakamura and coworkers have demonstrated that cordycepin inhibited lung carcinoma cells and melanoma cells growth by stimulating A 3 -AR [25,26]. Many evidences have also illustrated that A 1 -and A 2A -AR expression mediated cell death by inducing apoptosis in breast carcinoma cells, astrocytoma cells, mouse thymocytes [29][30][31]. In this study, we showed that cordycepin could significantly stimulate the expression of A 3 -AR mRNA in cordycepin-treated MA-10 cells. Moreover, this effect also occurred in A 1 -AR and A 2A -AR mRNA expressions. In addition, we also demonstrated that cordycepin induced MA-10 mouse Leydig tumor cell apoptosis through caspase-9 and caspase-3 pathways [28]. These data suggested that AR might participate in cordycepin-induced apoptosis pathway in MA-10 mouse Leydig cells.
We continued to investigate the AR agonists to cotreat with cordycepin in MA-10 cells. We found that A 3 -AR agonists could significantly rescue cordycepin-induced apoptosis in MA-10 cells. Many reports indicate that adenosine displays contradictory effects such as the induction of cell apoptosis or stimulation of cell proliferation [32,33]. It has been reported that the A 3 -AR agonist (CL-IB-MECA) reduced apoptosis in human astroglioma D384 cells [34]. In fact, some evidences have shown that adenosine via A 1 -ARand A 3 -AR-mediated cytoprotection involves phospholipase C, PKC, and p38 MAPK pathways, and reduced ROS production in cardiomyocytes [35][36][37]. In this study, we demonstrated that activation of AR would induce apoptosis in MA-10 cells. However, activation of different subtypes of   AR could trigger apoptosis/survival pathway in MA-10 cells, which must be further investigated.
Recent reports demonstrated that adenosine receptor antagonists can resume cell viability on toxicant druginduced apoptosis in thyroid cancer cells [38,39]. We used the selective AR antagonists to cotreat with cordycepin in MA-10 cells. We also demonstrated that A 1 -, A 2A -, A 2B -, and A 3 -AR antagonist could significantly rescue 100 μM cordycepin-induced apoptosis in MA-10 cells. These results indicate that the apoptotic effects of cordycepin on MA-10 cells could be mediated by AR. However, 1 mM cordycepin cotreated with adenosine antagonists could not reverse cell viability, but even promote cell death. Recent developments of potent and selective antagonists of AR subtypes have been valuable for further defining the physiological effects of the various AR subtypes. Moreover, there are substantial species differences in the affinity of these compounds, and these selective compounds may fairly potent as antagonists of another AR subtypes [40].
It has been shown that adenosine-stimulated steroidogenesis might be involved in the A 2A -and A 2B -AR and phosphorylation of MAPK ERK 1/2 signal pathway in rat adrenal cells [41]. Moreover, it is proposed that ligand binding results in a change in receptor state from an inactive to an active state will ultimately elicit its biological response based on the receptor's conformation [42]. In this model, agonists are thought to have selective binding affinity for the preexisting resting and active states or can induce a conformational change to a different receptor state and effects binding affinity of a ligand [43]. In the present study, cordycepin cotreated with A 2B -AR agonist (NECA) would significantly decrease NECA-stimulated progesterone production. It is probable that cordycepin may bind with other receptors and induce conformation change to effect binding affinity of A 2B -AR agonist. This result is consistent with the observation that cordycepin (100 μM) decreased the expression of A 2B -AR mRNA at 24 hr treatment. Although cordycepin (100 μM) did not affect the expression of A 1and A 3 -AR mRNA, cordycepin cotreated with A 1 -and A 3 -AR agonist (CPA and IB-MECA) also significantly decreased their progesterone production. On the other hand, cordycepin did not affect the A 2A -AR agonist (CGS-21680)stimulated progesterone production. It is reasonable that cordycepin may compete with CGS-21680 in the same biding site of A 2A -AR to stimulate progesterone production. Here, we demonstrated that activation of A 1 -, A 2A -, A 2B -, and A 3 -AR could induce steroidogenesis in MA-10. However, we used the AR antagonists to cotreat with cordycepin, and results illustrated huge variations. AR antagonists only slightly increased progesterone production than control group in MA-10 cells, which may decrease the accuracy. More experiments should be conducted to decrease the inconsistence.
It is well established that StAR protein is essential for steroidogenesis, and that ERK 1/2 phosphorylation driven by mitochondrial PKA will induce StAR protein expression and then steroidogenesis [46]. The previous data showed that cordycepin upregulates the expression of StAR mRNA and StAR protein to induce steroidogenesis through the PKA signaling pathway in normal mouse Leydig cells [55]. It has also been documented that the downstream effectors of PKA signaling pathway include several transcription factors [56]. However, our results indicated that cordycepin could not induce StAR protein and StAR promoter expressions in MA-10 cells. Our data are somewhat inconsistent to several studies that steroidogenesis could be trigger by cAMP and StAR-independent pathways [57][58][59]. Therefore, it is possible that the cordycepin could possibly acitvate MAPK-ERK 1/2 and PKC pathways without increasing of StAR promoter and StAR protein expression to induce steroidogenesis in MA-10 cells, which will be worth to further investigate.
In conclusion, the present studies demonstrate that cordycepin is one of the active constituent of Cordyceps sinensis, which can stimulate progesterone production in MA-10 cells. Meanwhile, cordycepin could also activate AR and simultaneously induce steroidogenesis and apoptosis in MA-10 mouse Leydig tumor cells.