Cardiotonic steroids trigger non-classical testosterone signaling in Sertoli cells via the α4 isoform of the sodium pump

Biochimica et Biophysica Acta 1813 (2011) 2118–2124

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b b a m c r

Cardiotonic steroids trigger non-classical testosterone signaling in Sertoli cells via the α4 isoform of the sodium pump Lutz Konrad a, Raimund Dietze b, Ulrike Kirch b, Herbert Kirch b, Alexander Eva b, Georgios Scheiner-Bobis b,⁎ a b

Zentrum f. Frauenheilkunde und Geburtshilfe, Fachbereich Medizin, Germany Institut für Veterinär-Physiologie und -Biochemie, Fachbereich Veterinärmedizin, Justus-Liebig-Universität Giessen, Germany

a r t i c l e

i n f o

Article history: Received 12 April 2011 Received in revised form 21 June 2011 Accepted 19 July 2011 Available online 23 July 2011 Keywords: Cardiotonic steroid Ouabain Sertoli cell CREB Erk1/2 Alpha4 isoform

a b s t r a c t The α4 isoform of the Na +,K+-ATPase (sodium pump) is known to be expressed in spermatozoa and to be critical for their motility. In the investigation presented here, we find that the rat-derived Sertoli cell line 93RS2 also expresses considerable amounts of the α4 isoform in addition to the α1 isoform. Since Sertoli cells are not motile, one can assume that the function of the α4 isoform in these cells must differ from that in spermatozoa. Thus, we assessed a potential involvement of this isoform in signaling pathways that are activated by the cardiotonic steroid (CTS) ouabain, a highly specific sodium pump ligand. Treatment of 93RS2 cells with ouabain leads to activation of the c-Src/c-Raf/Erk1/2 signaling cascade. Furthermore, we show for the first time that the activation of this cascade by ouabain results in phosphorylation and activation of the transcription factor CREB. This signaling cascade is induced at low nanomolar concentrations of ouabain, consistent with the involvement of the α4 isoform. This is further supported by experiments involving siRNA: silencing of α4 expression entirely blocks ouabain-induced activation of Erk1/2 whereas silencing of α1 has no effect. The findings of this study unveil new aspects in CTS/sodium pump interactions by demonstrating for the first time ouabain-induced signaling through the α4 isoform. The c-Src/c-Raf/Erk1/2/CREB cascade activated by ouabain is identical to the so-called non-classical signaling cascade that is normally triggered in Sertoli cells by testosterone. Taking into consideration that CTS are produced endogenously, our results may help to gain new insights into the physiological mechanisms associated with male fertility and reproduction. © 2011 Elsevier B.V. All rights reserved.

1. Introduction The Na +,K +-ATPase (sodium pump) is a membrane-bound enzyme that hydrolyzes ATP to transport 3 Na + ion out of the cell and 2 K + ions into the cell. The oligomeric enzyme consists of α and β subunits. Four different α subunit isoforms (α1–α4) and three β subunit isoforms (β1–β3) have been identified thus far. Aside from these two subunits that are absolutely required for ion transport, a third regulatory lipoprotein similar to phospholamban has been found in some tissues to be associated with the α and β subunits and is referred to as the γ subunit. Whereas the α1 isoform of the sodium pump seems to be expressed in all types of animal cells, some of the other α isoforms are found together with the α1 isoform in certain cell types.

⁎ Corresponding author at: Institut für Veterinär-Physiologie und -Biochemie, Fachbereich Veterinärmedizin, Justus-Liebig-Universität Giessen, Frankfurter Str. 100, D-35392 Giessen, Germany. Tel.: + 49 641 9938172; fax: + 49 641 9938179. E-mail address: [email protected] (G. Scheiner-Bobis). 0167-4889/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamcr.2011.07.012

Cardiotonic steroids (CTS) are specific inhibitors of the sodium pump that have been used for more than 200 years for the treatment of congestive heart failure. At low concentrations that do not cause a global inhibition of the sodium pump, CTS induce through their binding to the α subunit various intracellular signaling events that may be mediated by specific α subunit isoforms. Thus, in the neuroblastoma cell line SK-N-AS, Erk1/2 activation through the CTS ouabain is mediated via the α3 isoform [1], and in astrocytes it appears to be the α2 isoform that is involved in the generation of signaling cascades [2]. While α1, α2, and α3 isoforms are widely present in various tissues, the α4 isoform seems to be present only in the male reproductive organs. After its original identification in crude testis homogenates it was found to be coexpressed with the α1 isoform in sperm cells [3,4]. Its biological function is possibly associated with sperm motility [5–7]; however, as in a recent publication the α4 isoform of Na+,K +-ATPase was shown to be involved in signaling events during capacitation of bovine sperm [8], its physiological function in germ cells might go beyond mere involvement in cell motility. In the investigation presented here we demonstrate that the α4 subunit isoform of the sodium pump is coexpressed with α1 in rat Sertoli cell lines and mediates the activation of signaling cascades by

L. Konrad et al. / Biochimica et Biophysica Acta 1813 (2011) 2118–2124

CTS that overlap with the non-classical signaling cascades induced in Sertoli cells by testosterone. 2. Materials and methods 2.1. Cell culture The Sertoli cell line 93RS2 [9] was used for most experiments. For the detection of α4 specific mRNA the Sertoli cell line SCIT-C8 [10] was used in parallel. Cells were cultured in DMEM/Ham's F-12 (1/1 by vol.) with L-glutamine supplemented with 10% (by vol.) fetal bovine serum (FBS) and 1% Pen/Strep. Culture was maintained in a humidified incubator at 37 °C in 5% CO2. The medium was replaced every 2 days. Cells were harvested after removing the medium, followed by a washing step with Dulbecco's PBS without Ca2+ and Mg 2+ and incubation with accutase (0.25%) for 3–5 min at 37 °C. All cell culture reagents were purchased from PAA Laboratories (Pasching, Austria). 2.2. Preparation of cell lysates Approximately 1 × 10 5 cells of the Sertoli cell line 93RS2 were plated in 10-cm culture dishes and grown as described above. After 24 h, incubation was continued with 5% FBS and otherwise unchanged conditions. Cells were then serum-starved (i.e. 1% serum) for another 24 h. Ouabain was added to the cells at various concentrations and the cells were further incubated at 37 °C for various times as described in the figure legends. The medium was then aspirated and after a washing step with PBS, cells were lysed in 750 μl of a commercially available cell lysis buffer (Cell Signaling Technology, Frankfurt, Germany) according to the protocol of the provider. Immediately before use, 1 μM PMSF, 1× protease inhibitor cocktail (Roche, Mannheim, Germany) and 2 μg/ml pepstatin were added to the lysis buffer. All lysis steps were carried out on ice. After 5 min of incubation cells were detached from the dish with a cell scraper and the mixture was transferred into reaction vials and sonicated twenty times for 1 s each with intervals of 1 s. After centrifugation of the lysates at 4 °C for 20 min at 13,000 ×g, the protein concentration in the supernatant was determined using the bicinchoninic acid (BCA) protein assay reagent kit (Pierce, Rockford, IL, USA) with inclusion of the lysis buffer in the standard curve. Aliquots taken for further analysis were stored at −20 °C. 2.3. Western blot detection of isolated proteins A total of 10 μg of protein was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on gels containing 10% acrylamide and 0.3% N,N′-methylene-bis-acrylamide. Biotinylated proteins (Cell Signaling Technology, Frankfurt a.M., Germany) were run in parallel as molecular weight markers. After SDS-PAGE, proteins were electro-blotted onto nitrocellulose membranes (Schleicher & Shull, Dassel, Germany) at 500 mA for 30 min. Detection of proteins was performed by following the protocols of the providers of the primary antibodies (Cell Signaling Technology, Santa Cruz Inc., or Dianova) and the enhanced chemiluminescence (ECL) kit (GE HealthCare, München, Germany) that also includes the biotinylated secondary antibody. Chemiluminescence was visualized and quantified by the Molecular Imager ChemiDoc XRS System (BioRad, München, Germany). 2.4. Detection of α subunit isoform-specific mRNA in Sertoli cells Total mRNA was isolated from 93RS2 or SCIT-C8 cells that were grown to 50% confluency by using the RNeasy Mini Kit (Qiagen, Hilden, Germany). RNase-free DNase I (Qiagen, Germany) was used to remove any contaminating DNA. Concentration and purity of total

2119

mRNA were determined by measuring the absorbance at 260 and 280 nm. Reverse transcription and PCR amplification were carried out by the OneStep RT-PCR kit (Qiagen). A total of 60 ng of mRNA, 1 μl (20 pmol/ml) of each primer, 10 μl of five-fold concentrated buffer (20 mmol/L Tris HCl, 100 mmol/L KCl, 12.5 mmol/L MgCl2, 10 mmol/L dNTPs), 2 μl of a mixture of Ominiscript and Sensiscripts reverse transcriptases and HotStar Taq DNA polymerase in a final volume of 25 μl were incubated in a MasterCycler Gradient (Eppendorf, Hamburg, Germany) at 50 °C for 30 min to accomplish reverse transcription. For PCR amplification the mixture was heated at 95 °C for 15 min, followed by 40 cycles of denaturation at 94 °C for 1 min, annealing at 58 °C for 1 min, and cDNA extension at 72 °C for 1 min. After amplification, a final extension at 72 °C was performed for 10 min. For specific amplification of α1, α2, α3, and α4 isoforms and GAPDH, forward and reverse primers were used as published [4]. The RT-PCR amplificates were analyzed by electrophoresis on 1.7% agarose gels followed by ethidium bromide staining. To ascertain whether α4-specific mRNA is present not only in Sertoli cell lines but also in normal Sertoli cells, mRNA was isolated from primary Sertoli cell cultures of 19-day-old Wistar rats as published [10], or from SCO (Sertoli Cells Only) humans. These samples were amplified by RTR-PCR as described above. For amplification of the human α4, PCR primers used were as previously published [7]. 2.5. Silencing α1 or α4 mRNA by siRNA siRNA used for silencing either of the α1 or α4 isoform-specific mRNAs is commercially available (Stealth™ RNAi; Invitrogene, Karlsruhe, Germany). Transfection into the 93RS2 cells was carried out according to the protocol of the provider. Briefly, a total of 5 × 10 3 32RS2 cells in culture medium without antibiotics were added to each well of 6-well culture dishes and cultured to 30% confluency. Immediately before use, Stealth™ RNAi and Lipofectamine 2000 were each diluted with Opti-MEM I Reduced Serum Media and incubated for 5 min at room temperature. Then the two solutions were combined, mixed gently, and after a further 20 min at room temperature added to the cells. The final concentration of the Stealth™ RNAi in the media was 100 nM. For control, cells were either treated with Lipofectamine 2000 alone or transfected with Stealth™ RNAi Negative Control Medium GC (negative control for either α1 or α4 Na +,K +-ATPase Stealth™ RNAi). After incubation for 72 h under the conditions described above, transfection efficiencies of about 82% were estimated by using the Block-iT™ Transfection Kit (Invitrogen, Germany) according the protocol of the provider. Total RNA was then isolated from 93RS2 cells and the extracted RNA was subjected to RT-PCR to amplify α1- and α4-specific mRNA/DNA fragments as described above. The Stealth™ RNAi oligonucleotides used were 5′-UGU UUC UUG GCU UAU GGC AUC CGA A-3′ plus 5′-UUC GGA UGC CAU AAG CCA AGA AAC A-3′ against the α1 isoform mRNA, and 5′-GCC UCA UCA GGA GUC UCU UCC CAU A-3′ plus 5′-UAU GGG AAG AGA CUC CUG AUG AGG C-3′ against the α4 isoform mRNA. In a different set of experiments, after the 72 h-treatment with Stealth™ RNAi, cells were incubated with 100 nM ouabain for 30 min and were used to isolate lysates for western blot experiments. 2.6. MTT assay of cells transfected with siRNA against α1 or α4 subunit isoforms 93RS2 Sertoli cells were treated with Stealth™ RNAi (2 × 10 3 cells per well in 24-well culture dishes) as described above. After 72 h of incubation, transfection medium was replaced by the complete growth medium. Cells were then cultured as described for 1–7 days. To determine cell viability at the end of each measuring period, the medium was aspirated and replaced by 300 μl of fresh culture

2120

L. Konrad et al. / Biochimica et Biophysica Acta 1813 (2011) 2118–2124

medium containing 0.5 mg/ml MTT (3-[4,5-dimethylthiazol-2-yl]2,5-diphenyltetra-zolium bromide). After 4 h of incubation the medium containing MTT was removed by inverting the plate. The resulting formazan crystals were then solubilized by adding 200 μl DMSO to each well. After 10 min of shaking, absorbance in each well was determined in a microplate reader at 540 nm. 2.7. Statistical analysis Optical density of protein bands of the various Western blots were corrected for any gel loading differences by taking into consideration the optical density of actin or unphosphorylated Erk1/2 bands, which were detected in Western blots that were run in parallel. Data were analyzed by one-way ANOVA with repeated measures and by applying Dunnett's comparison of all data to the control. Significance was accepted at p b 0.05. 3. Results

α1 93RS2

bp

3.3. c-Src activation by ouabain Ouabain-induced Erk1/2 activation has been shown to be mediated by activation of c-Src [16,17] via phosphorylation at Tyr418 [18]. A similar mechanism is apparently involved in the ouabain-stimulated induction of signaling cascades in the 93RS2 Sertoli cell line. Ouabain causes

α4 SCIT-C8

2000 1500 1200 1031 900 800 700 3000 2000 1500 1200 1031 900 800 700

α4 Sertoli rat

α4 sperm human

α4 SCO human

Fig. 1. Detection of α subunit isoforms in Sertoli cell lines 93RS2 and SCIT8 by RT-PCR and agarose gel electrophoresis. Primers used for RT-PCR have been published previously [4]. The expected amplificates for the sodium pump α1 isoform (upper left panel; 1447 bp) and α4 isoform (upper right panel; 935 bp) are both found in 93RS2 and SCIT8 Sertoli cell lines. The α4-specific amplificate is also found in rat Sertoli cells (lower left panel; 935 bp) and in Sertoli cells of SCO human donors and in sperm of healthy human donors (lower middle and right panels; 748 bp). Neither the α2 nor the α3 isoforms were detectable in these cells.

phosphorylation of c-Src at Tyr418 (Fig. 3) that can be visualized by western blotting using a monoclonal antibody that specifically recognizes the phosphorylated form of this amino acid. At a concentration as low as 1 nM, ouabain induces a significant activation of Src kinase.

A

kDa 60

B

total Erk1/2

phospho-Erk1/2

50 40

3.2. Erk1/2 activation in 93RS2 Sertoli cells by ouabain

30 Ouabain (nM) 0

C

Erk1/2 activation (% control)

Low concentrations of the CTS ouabain trigger signaling cascades in various cells [14,15] that in many cases lead to the activation of Erk1/2. Thus, in order to investigate whether ouabain induces similar signaling events in the Sertoli cell line 93RS2, cells were incubated with various concentrations of ouabain for 30 min and cell lysates were isolated and investigated in western blots for Erk1/2 activation. At the low concentration of 1 nM, ouabain was found to trigger a clear activation of Erk1/2 (Fig. 2). Considering the fact that the 93RS2 cells are derived from the rat, which expresses an α1 isoform with very low affinity for ouabain (Ki = 1.5 ± 1.3 × 10 −4 M) [4], it is very likely that the observed activation of Erk1/2 is mediated through the interaction of ouabain with the α4 isoform, which displays a much higher affinity for the glycoside (Ki = 1.6 ± 1 × 10 −9 M) [4].

α4 93RS2

3000

3.1. Expression of α subunit isoforms in Sertoli cells The α4 subunit isoform of the sodium pump has been hitherto detected in male gonads only, where it is coexpressed with the α1 isoform [3,11,12]. The other two isoforms, α2 and α3, have not been detected in testis homogenates [11,12]. On the cellular level, coexpression of α1 and α4 isoforms has been shown for germ cells [4,8,13]. The function of the α4 has been linked to the motility of the spermatozoa [5–7] and with sperm capacitation [8]. We demonstrate here, however, that spermatozoa are not the sole cell type expressing the α4 isoform. Using RT-PCR we find that the Sertoli cell lines 93RS2 and SCIT-C8 express -like spermatozoa-α4 isoform-specific mRNA in addition to α1 isoform-specific mRNA (Fig. 1, upper panels). The α4 isoform seems to be the predominant form, although the results shown in the figure are not the product of quantitative PCR. The two other isoforms, α2 and α3, were not detectable (not shown). To verify that the α4 isoform is not only expressed in the Sertoli cell lines but also in normal Sertoli cells, rat Sertoli cells isolated as described [10], Sertoli cells from SCO humans, and sperm cells from healthy human donors were included in the investigation. The lower panels in Fig. 1 show that Sertoli cells from all three sources express the α4 subunit isoform-specific mRNA. Since Sertoli cells are not motile, the presence of α4 must be associated with physiological roles other than involvement in motility.

α1 SCIT-C8

1

10 100

0

1

10 100

** **

200

* 100

0 Ouabain (nM) 0

1

10

100

Fig. 2. Erk1/2 activation by ouabain. 93RS2 cells were treated for 30 min with various concentrations of ouabain. Cell lysates were then prepared and 10 μg of protein was separated on SDS polyacrylamide gels, blotted onto nitrocellulose, and subsequently probed with a monoclonal antibody against either the total (A) or the phosphorylated (activated) Erk1/2 (B). Visualization of Erk1/2 or phospho-Erk1/2 was achieved by using horseradish peroxidase-conjugated antibodies against the primary antibodies. (C) Mean data from n = 4 western blots like the one shown in B (± SEM); * = p ≤ 0.05; ** = p ≤ 0.01. The Western blots of total Erk1/2 served as loading controls. Values shown in (C) were corrected accordingly.

L. Konrad et al. / Biochimica et Biophysica Acta 1813 (2011) 2118–2124

A

2121

3.5. Ouabain-induced activation of CREB c-Src (60 kDa)

60 kDa 50kDa Ouabain (nM)

c-Src activation (% control)

B

0

1

10

**

300

100

**

*

200 100

0 Ouabain (nM)

0

1

10

100

Fig. 3. Activation of c-Src by ouabain. A) 93RS2 cells were incubated for 30 min with various concentrations of ouabain. Thereafter, 10 μg of cell lysates were separated by SDS gel electrophoresis and blotted onto nitrocellulose. To identify activated c-Src, blots were probed with an antibody that recognizes the phosphorylated Tyr418 of this kinase. B) Activation of c-Src is significant at all ouabain concentrations tested (n = 4; mean ± SEM; * = p ≤ 0.05; ** = p ≤ 0.01). Western blots of actin served as loading controls (not shown). Values were corrected for loading controls.

3.4. Effect of ouabain on c-Raf Activation of c-Raf occurs downstream of c-Src activation and precedes phosphorylation of Erk1/2. Using an antibody against the inhibitory phosphorylation site at Ser259 of c-Raf [19], we observed an ouabain-induced and concentration-dependent dephosphorylation of this amino acid (Fig. 4). Since the dephosphorylation of this amino acid represents a critical step in the Ras-dependent activation of c-Raf [20], one can assume that ouabain triggers activation of c-Raf that is mediated through Ras.

MW

A 100 80

c-Raf (74 kDa)

60 50

Ouabain (nM)

0

1

10

100

phospho-cRaf (Ser259) (%)

B 120 100

*

80 60

3.6. Silencing α1 or α4 isoforms by means of siRNA The neuroblastoma cell line SK-N-AS expresses α1 and α3 isoforms of the sodium pump α subunit. While both isoforms are essential for the survival of the SK-N-AS cells, ouabain-induced activation of Erk1/2 is mediated exclusively through α3 [1]. In order to investigate the physiological significance of α1 and α4 for the Sertoli cell line 93RS2, siRNA against the α1 or the α4 isoform was introduced in the cells and the successful silencing of either isoform and the effects of silencing on cell growth and ouabain-induced signaling were analyzed. The PCR amplification demonstrated that either of the α subunit isoforms can be successfully silenced with the appropriate siRNA, since RT-PCR did not produce any detectable amplificates for the corresponding cDNA fragments (Fig. 6). Both α subunit isoforms, however, are important for cell growth and proliferation: silencing either α1 or α4 isoforms results in a considerable decline of cell numbers as determined by the MTT assay (Fig. 7), whereby the influence of α1 appears to be stronger. Silencing α1 or α4 isoforms, however, produces diverse effects on ouabain-induced signaling. When 100 nM ouabain is added to the cells 72 h after siRNA transfection, Erk1/2 activation is only possible when the α4 isoform is still expressed (Fig. 8C and D). Silencing of this isoform results in a signal (phosphorylated Erk1/2) that is even weaker than that of the control that is obtained in the absence of ouabain and in the presence of both α1 and α4 isoforms (Fig. 8C and D). The reduction of the signal is not due to an overall reduction of Erk1/2 expression (Fig. 8B) or a change in the amount of constitutive proteins like actin (Fig. 8A). Silencing of the α1 subunit does not influence ouabain-induced Erk1/2 phosphorylation (Fig. 8C and D). Thus, as in the neuroblastoma cell line SK-N-AS [1], the α1 isoform does not mediate any ouabain-induced signaling in the Sertoli cell line 93RS2. 4. Discussion

40

*

20

0 Ouabain (nM)

Activation of the c-Src/Ras/Raf/Erk1/2 cascade could result in the activation of the transcription factor CREB. In Sertoli cells this reaction cascade is triggered by testosterone via a so-called non-classical pathway of testosterone action [21]. In this reaction cascade, binding of testosterone to the androgen receptor induces the activation of CREB via the c-Src/Ras/c-Raf/Erk1/2-cascade [22]. CREB-inducible transcription in Sertoli cells is essential for the survival of spermatocytes and the production of mature spermatozoa [23]. For these reasons it was of particular interest to investigate a possible CREB activation by ouabain. A time-dependent activation of CREB (phosphorylation at Ser133) was induced by 10 nM ouabain (Fig. 5). Phosphorylation of CREB after 30 min is significantly higher than the basal phosphorylation (at 0 min) and reaches an almost 2.5-fold increase after 180 min (Fig. 5D). Phosphorylation of CREB is not seen in the absence of ouabain (Fig. 5B and D), and ouabain does not have any significant effect on the expression of the reference protein actin (Fig. 5A), indicating that the observed phosphorylation of CREB is not due to increased expression of this protein.

0

1

10

100

Fig. 4. Activation of c-Raf by ouabain. A) Incubation of 93RS2 cells with various concentrations of ouabain leads to the dephosphorylation of the c-Raf inhibitory site at Ser259, as detected by a specific antibody against the phosphorylated form of this amino acid in a western blot. B) Statistical analysis of the results of independent experiments (n = 4; mean ± SEM; * = p ≤ 0.05). Western blots of actin served as loading controls (not shown). Values were corrected for loading controls.

The α4 subunit isoform of the sodium pump is known to be expressed in testis [3,5,13] and mainly in the middle piece of the flagellum [7] or the head of spermatozoa [8]. Its function has been associated with the motility of the sperm cells [5,7] or with sperm capacitation [8]. Our finding that there is considerable expression of the α4 subunit isoform in rat Sertoli cell lines and in rat Sertoli cells or SCO humans (Fig. 1), however, raises the question about the involvement of this isoform in additional physiological processes, since Sertoli cells are not motile and do not undergo capacitation.

2122

L. Konrad et al. / Biochimica et Biophysica Acta 1813 (2011) 2118–2124

A

50 kDa Actin (10 nM ouabain)

40 kDa

B C

50 kDa 40 kDa

p-CREB (no ouabain)

50 kDa 40 kDa 0

60

30

120

180

D

p-CREB (10 nM ouabain) min

CREB activation (%)

** 200

*

*

100

0

-

Ouabain

+ 0

-

+ 30

-

+ 60

-

+

-

120

+ 180

min

Fig. 5. Ouabain-induced activation of CREB. A) Western blot of actin serving as gel loading control. B) Western blot of phospho-CREB in the absence of ouabain. This control demonstrates that phospho-CREB formation does not occur spontaneously over the time period addressed in the experiment. C) In the so-called non-classical pathway of testosterone action, activation of the c-Src/Ras/Raf/Erk1/2 cascade leads to the activation of the transcription factor CREB [21]. The same occurs when 93RS2 cells are incubated with 10 nM ouabain for various times. C) Maximum formation of phospho-CREB (gray bars; phospho-Ser 133) after 2 h is 2.5-fold over control (n = 4; mean ± SEM; * = p ≤ 0.05; ** = p ≤ 0.01). There are no significant changes in phospho-CREB formation in the absence of ouabain (white bars).

The results of our investigation support this notion. We specifically show here that the expression of the α4 isoform (and not of the α1 isoform) of the sodium pump in Sertoli cells is directly linked to ouabain-induced signaling that results in the activation of the Src/cRaf/Erk1/2 cascade. We further show for the first time that this signaling cascade leads to the phosphorylation and activation of the transcription factor CREB. These results might help in gaining new insights into the physiological mechanisms associated with male fertility and reproduction. Sertoli cells are a significant part of the male reproductive system. They surround the diploid spermatogonia that at the onset of puberty

undergo divisions to either produce stem cells for spermatogonia or to differentiate into haploid spermatozoa. During gonadal development Sertoli cells produce anti-Mullerian (AMH) hormone, form the bloodtestis barrier, and also eliminate defective germ cells. Sertoli cells express receptors for the follicle-stimulating hormone (FSH), and under FSH influence they produce androgen binding protein and inhibin, which prevents further FSH release. Leydig cells that are found in the interstitial tissue express luteinising hormone (LH)

+ α1 siRNA

Control

+ α4 siRNA

A

B

9 8 7

OD 540 nm

1500 kb

1000 kb

6 5 4 3 2

500 kb siRNA:

α4 +

+

α1

Fig. 6. Silencing α1 or α4 mRNA expression by means of siRNA. 93RS2 cells were treated with siRNA against the α1 or α4 isoforms of the sodium pump α subunit according to the manufacturer's protocol. Control cells were treated with negative control siRNA, provided in the kit of the manufacturer. Total RNA was then isolated and subjected to RT-PCR to amplify α1- or α4-specific mRNA/cDNA fragments. A) RT-PCR for the amplification of α4-specific DNA fragment. Treatment with α4-specific siRNA abolishes α4-specific mRNA expression. B) RT-PCR for the amplification of α1-specific DNA fragment. After transfection with α1-specific siRNA, expression of α1-specific mRNA is not detectable.

1 0 24

48

72

96

120

144

168

hours of incubation Fig. 7. Viability and proliferation of 93RS2 cells. Sertoli cells were treated with α1specific siRNA (open circles) or α4-specific siRNA (open squares). Viability and proliferation of Sertoli cells was followed by the MTT test over a period of 7 days. Control cells were treated with negative control siRNA (solid squares). The proliferation of cells treated with either α1 or α4 siRNA after 48 h is clearly lower than that of control cells (n = 8–12; mean ± SEM).

L. Konrad et al. / Biochimica et Biophysica Acta 1813 (2011) 2118–2124

A

Actin

B Erk1/2 (total)

C α α4 siRNA + ouabain

control + ouabain

α1 siRNA + α ouabain

D

Erk1/2 activation (% control)

control

p-Erk1/2

250 200

*

*

150 100 50

**

0

Fig. 8. Ouabain induced activation of Erk1/2, after silencing α1 or α4 isoforms. A) Detection of total actin in a Western blot. Silencing of either α1 or α2 isoforms has no effect on actin expression. B) The same applies for the expression of total Erk1/2. C) Treatment of 93RS2 cells with 100 nM ouabain (control+ ouabain) for 30 min leads to an approximate 2-fold activation of Erk1/2 over control. Transfection of cells with siRNA against the α1 isoform does not affect the activation of Erk1/2 (α1 siRNA+ ouabain). When the cells are treated with siRNA against the α4 isoform, no ouabain induced activation of Erk1/2 occurs (α4 siRNA+ ouabain). Under these conditions, the total phospho-Erk1/2 signal is reduced bellow to that obtained in the absence of ouabain (control), thus implying that the Erk1⁄2 basal activation might be partially contributed by the α4 subunit isoform of the sodium pump. D) Mean data from n = 3 western blots like the one shown in A (±SEM); * = significantly higher than control; **= significantly lower than control. The Western blots of actin served as loading controls. Values were corrected for loading controls.

receptors and produce the androgens testosterone and 5α-dihydrotestosterone almost exclusively. The process of spermatogenesis, including maturation of spermatogonia to spermatozoa, seems to be tightly associated with the activation Erk1/2 and other mitogen-activated protein kinases (MAPK) [24,25]. Differentiation and proliferation of Sertoli cells depend on FSH, which activates an Erk1/2 signaling cascade [26]. Androgens, in contrast, trigger a so-called non-classical testosterone signaling pathway in Sertoli cells that also leads to Erk1/2 activation and gene expression [21]. This latter process is an absolute requirement for the production of haploid spermatozoa [27,28]. At that stage, however, spermatozoa are not yet completely mature. After their development in the epididymis and after ejaculation, they progressively gain motility. Upon their passage through the cervix they undergo the process of capacitation, reach a new form of motility known as hyperactivation, bind to the oocyte, undergo the acrosomal reaction, and fertilize the oocyte. The molecular mechanisms that mediate forward motility, hyperactivation, or the acrosomal reaction of spermatozoa are not completely understood. All of these activities, however, seem to be mediated via signaling cascades involving MAPKs [29–31]. Ouabain and other CTS trigger signaling cascades in various cells that often result in the activation of Erk1/2. The same applies for the 93RS2 Sertoli cell line where there is clear activation (phosphorylation) of Erk1/2 upon exposure to low concentrations of ouabain (Fig. 2). Taking into consideration that rat cells express an α1 isoform with very low affinity for ouabain [4] whereas the α4 isoform has a high affinity for ouabain [4], one can assume that the Erk1/2 activation observed is mediated through the interaction of ouabain with the α4

2123

subunit. This assumption was confirmed by using siRNA to silence either the α1 or the α4 subunit isoform. While ouabain-induced activation of Erk1/2 was not affected by silencing the α1 isoform, it was greatly affected when α4 expression is reduced (Fig. 8). Based on this finding one can speculate that α1 and α4 isoforms might be targeted to different microdomains of the plasma membrane and that α4 interactions with proteins in its immediate environment lead to the generation of the ouabain-induced signaling cascades we observe. Although the sorting mechanism for the α1 or α4 subunit proteins is not known, the fact that in sperm cells the α4 is localized only at the base of the flagellum whereas the α1 is spread out over the entire structure [7] indicates the existence of a functioning sorting mechanism ensuring the targeting of the two proteins to defined areas of the plasma membrane. In general, the results obtained from the experiments involving siRNA in the Sertoli cell line 93RS2 resemble earlier results obtained with the neuroblastoma cell line SK-N-AS [1]. There too, ouabaininduced Erk1/2 activation was mediated not by the α1 isoform but through a second subunit isoform, the co-expressed α3 subunit. And, as in SK-N-AS cells where both α1 and α3 are needed for cell growth, both α1 and α4 are required for the proliferation of the Sertoli cell line 93RS2 (Fig. 7). Apparently, under normal conditions the two α subunit isoforms occupy distinct and essential roles for the physiology of the cells, so that the function of one cannot be compensated by the other when one of the two isoforms is silenced by siRNA. Ouabain-induced Erk1/2 activation is often mediated by c-Src kinase [32,33]. Ouabain-induced activation of Erk1/2 in the 93RS2 Sertoli cell line via the α4 isoform seems to follow the same pathway. A low concentration of ouabain leads to activation of c-Src (Fig. 3) and also the next kinase in this cascade, c-Raf (Fig. 4). This cascade coincides to this point with the non-classical signaling pathway triggered by testosterone in Sertoli cells [21,34]. The non-classical testosterone pathway leads to phospho-CREB formation and gene activation in Sertoli cells [21,34] and is also induced in the experiments described here by low concentrations of ouabain (Fig. 5). The activation of the transcription factor by ouabain CREB occurs by phosphorylation at Ser133, the same amino acid that is also phosphorylated by the action of testosterone [22]. To our knowledge this is the first report on ouabain-induced activation of CREB. Phosphorylation of CREB in the Sertoli cell line 93RS2 is clearly detectable after 30 min of incubation with ouabain and increases for up to three hours, the end point of the measurement. A change in total CREB expression was not observed during this time (Fig. 5). Taken together, these data indicate that ouabain triggers a signaling cascade in Sertoli cells that is to a great extent identical to the non-classical signaling pathway triggered by testosterone, but the physiological significance of this must still be evaluated. Ouabain is apparently produced endogenously: the highest immunoreactivity for this compound has been found thus far in adrenals, hypophysis, and hypothalamus [35–37]. The genuine location for ouabain biosynthesis, however, seems to be the adrenals, since adrenalectomy leads to the reduction of circulating ouabain levels [36,38,39]. Therefore, it might be that endogenously produced ouabain participates via its interaction with the α4 isoform in signaling events that influence the physiology of sperm or Sertoli cells. Although at this stage the concept of endogenous CTS acting in a hormone-like fashion on various cells of the male reproduction system is rather tentative, the presence of digoxin-like immunoreactivity in human seminal fluid [40] supports this notion and encourages further investigation. Acknowledgements The work was supported by a grant from the DFG, Sche 307/6-1. The authors wish to Thank Dr. Martin Bergmann for mRNA from SCO and healthy human sperm and Dr. Klaus Failing for his help regarding the statistical analysis of data.

2124

L. Konrad et al. / Biochimica et Biophysica Acta 1813 (2011) 2118–2124

References [1] L. Karpova, A. Eva, U. Kirch, A. Boldyrev, G. Scheiner-Bobis, Sodium pump α1 and α3 subunit isoforms mediate distinct responses to ouabain and are both essential for survival of human neuroblastoma, Febs. J. 277 (2010) 1853–1860. [2] V.A. Golovina, H. Song, P.F. James, J.B. Lingrel, M.P. Blaustein, Na+ pump α2subunit expression modulates Ca2+ signaling, Am. J. Physiol. Cell Physiol. 284 (2003) C475–C486. [3] A.L. Woo, P.F. James, J.B. Lingrel, Characterization of the fourth alpha isoform of the Na,K-ATPase, J. Membr. Biol. 169 (1999) 39–44. [4] K. Wagoner, G. Sanchez, A.N. Nguyen, G.C. Enders, G. Blanco, Different expression and activity of the α1 and α4 isoforms of the Na,K-ATPase during rat male germ cell ontogeny, Reproduction 130 (2005) 627–641. [5] A.L. Woo, P.F. James, J.B. Lingrel, Sperm motility is dependent on a unique isoform of the Na,K-ATPase, J. Biol. Chem. 275 (2000) 20693–20699. [6] A.L. Woo, P.F. James, J.B. Lingrel, Roles of the Na, K-ATPase α4 isoform and the Na+/H + exchanger in sperm motility, Mol. Reprod. Dev. 62 (2002) 348–356. [7] G. Sanchez, A.N. Nguyen, B. Timmerberg, J.S. Tash, G. Blanco, The Na, K-ATPase α4 isoform from humans has distinct enzymatic properties and is important for sperm motility, Mol. Hum. Reprod. 12 (2006) 565–576. [8] L.D. Newton, S. Krishnakumar, A.G. Menon, J.P. Kastelic, F.A. van der Hoorn, J.C. Thundathil, Na+/K+ATPase regulates sperm capacitation through a mechanism involving kinases and redistribution of its testis-specific isoform, Mol. Reprod. Dev. 77 (2010) 136–148. [9] C. Jiang, S.J. Hall, K. Boekelheide, Development and characterization of a prepubertal rat Sertoli cell line, 93RS2, J. Androl. 18 (1997) 393–399. [10] L. Konrad, M. Munir Keilani, A. Cordes, E. Volck-Badouin, L. Laible, M. Albrecht, H. Renneberg, G. Aumuller, Rat Sertoli cells express epithelial but also mesenchymal genes after immortalization with SV40, Biochim. Biophys. Acta 1722 (2005) 6–14. [11] O.I. Shamraj, J.B. Lingrel, A putative fourth Na+, K + -ATPase α-subunit is expressed in testis, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 12952–12956. [12] G. Blanco, R.J. Melton, G. Sanchez, R.W. Mercer, Functional characterization of a testes-specific alpha-subunit isoform of the sodium/potassium adenosinetriphosphatase, Biochemistry 38 (1999) 13661–13669. [13] G. Blanco, G. Sanchez, R.J. Melton, W.G. Tourtellotte, R.W. Mercer, The α4 isoform of the Na,K-ATPase is expressed in the germ cells of the testes, J. Histochem. Cytochem. 48 (2000) 1023–1032. [14] W. Schoner, G. Scheiner-Bobis, Endogenous cardiac glycosides: hormones using the sodium pump as signal transducer, Semin. Nephrol. 25 (2005) 343–351. [15] Z. Xie, J. Xie, The Na/K-ATPase-mediated signal transduction as a target for new drug development, Front. Biosci. 10 (2005) 3100–3109. [16] P. Kometiani, L. Liu, A. Askari, Digitalis-induced signaling by Na+/K+-ATPase in human breast cancer cells, Mol. Pharmacol. 67 (2005) 929–936. [17] J. Tian, T. Cai, Z. Yuan, H. Wang, L. Liu, M. Haas, E. Maksimova, X.-Y. Huang, Z.-J. Xie, Binding of Src to Na+/K+-ATPase forms a functional signaling complex, Mol. Biol. Cell 17 (2006) 317–326. [18] M. Haas, H. Wang, J. Tian, Z. Xie, Src-mediated inter-receptor cross-talk between the Na+/K+-ATPase and the epidermal growth factor receptor relays the signal from ouabain to mitogen-activated protein kinases, J. Biol. Chem. 277 (2002) 18694–18702. [19] T. Pisitkun, V. Jacob, S.M. Schleicher, C.L. Chou, M.J. Yu, M.A. Knepper, Akt and ERK1/2 pathways are components of the vasopressin signaling network in rat native IMCD, Am. J. Physiol. Renal Physiol. 295 (2008) F1030–F1043. [20] M. Jaumot, J.F. Hancock, Protein phosphatases 1 and 2A promote Raf-1 activation by regulating 14-3-3 interactions, Oncogene 20 (2001) 3949–3958. [21] W.H. Walker, Molecular mechanisms of testosterone action in spermatogenesis, Steroids 74 (2009) 602–607.

[22] C. Fix, C. Jordan, P. Cano, W.H. Walker, Testosterone activates mitogen-activated protein kinase and the cAMP response element binding protein transcription factor in Sertoli cells, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 10919–10924. [23] M. Scobey, S. Bertera, J. Somers, S. Watkins, A. Zeleznik, W. Walker, Delivery of a cyclic adenosine 3′,5′-monophosphate response element-binding protein (creb) mutant to seminiferous tubules results in impaired spermatogenesis, Endocrinology 142 (2001) 948–954. [24] M.W. Li, D.D. Mruk, C.Y. Cheng, Mitogen-activated protein kinases in male reproductive function, Trends Mol. Med. 15 (2009) 159–168. [25] T. Almog, Z. Naor, Mitogen activated protein kinases (MAPKs) as regulators of spermatogenesis and spermatozoa functions, Mol. Cell. Endocrinol. 282 (2008) 39–44. [26] P. Crepieux, S. Marion, N. Martinat, V. Fafeur, Y.L. Vern, D. Kerboeuf, F. Guillou, E. Reiter, The ERK-dependent signaling is stage-specifically modulated by FSH, during primary Sertoli cell maturation, Oncogene 20 (2001) 4696–4709. [27] C. Sette, M. Barchi, A. Bianchini, M. Conti, P. Rossi, R. Geremia, Activation of the mitogen-activated protein kinase ERK1 during meiotic progression of mouse pachytene spermatocytes, J. Biol. Chem. 274 (1999) 33571–33579. [28] S. Di Agostino, F. Botti, A. Di Carlo, C. Sette, R. Geremia, Meiotic progression of isolated mouse spermatocytes under simulated microgravity, Reproduction 128 (2004) 25–32. [29] T. Almog, S. Lazar, N. Reiss, N. Etkovitz, E. Milch, N. Rahamim, M. Dobkin-Bekman, R. Rotem, M. Kalina, J. Ramon, A. Raziel, H. Breitbart, R. Seger, Z. Naor, Identification of extracellular signal-regulated kinase 1/2 and p38 MAPK as regulators of human sperm motility and acrosome reaction and as predictors of poor spermatozoan quality, J. Biol. Chem. 283 (2008) 14479–14489. [30] T. Almog, Z. Naor, The role of Mitogen activated protein kinase (MAPK) in sperm functions, Mol. Cell. Endocrinol. 314 (2010) 239–243. [31] M. Luconi, T. Barni, G.B. Vannelli, C. Krausz, F. Marra, P.A. Benedetti, V. Evangelista, S. Francavilla, G. Properzi, G. Forti, E. Baldi, Extracellular signal-regulated kinases modulate capacitation of human spermatozoa, Biol. Reprod. 58 (1998) 1476–1489. [32] J. Tian, T. Cai, Z. Yuan, H. Wang, L. Liu, M. Haas, E. Maksimova, X.Y. Huang, Z.J. Xie, Binding of Src to Na+/K+-ATPase forms a functional signaling complex, Mol. Biol. Cell 17 (2006) 317–326. [33] M. Liang, T. Cai, J. Tian, W. Qu, Z.J. Xie, Functional characterization of Srcinteracting Na/K-ATPase using RNA interference assay, J. Biol. Chem. 281 (2006) 19709–19719. [34] W.H. Walker, Non-classical actions of testosterone and spermatogenesis, Philos. Trans. R Soc. Lond. B Biol. Sci. 365 (2010) 1557–1569. [35] J.M. Hamlyn, M.P. Blaustein, S. Bova, D.W. DuCharme, D.W. Harris, F. Mandel, W.R. Mathews, J.H. Ludens, Identification and characterization of a ouabain-like compound from human plasma, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 6259–6263. [36] J.M. Hamlyn, Z.R. Lu, P. Manunta, J.H. Ludens, K. Kimura, J.R. Shah, J. Laredo, J.P. Hamilton, M.J. Hamilton, B.P. Hamilton, Observations on the nature, biosynthesis, secretion and significance of endogenous ouabain, Clin. Exp. Hypertens. 20 (1998) 523–533. [37] S. Li, C. Eim, U. Kirch, R.E. Lang, W. Schoner, Bovine adrenals and hypothalamus are a major source of proscillaridin A- and ouabain-immunoreactivities, Life Sci. 62 (1998) 1023–1033. [38] B.R. Boulanger, M.P. Lilly, J.M. Hamlyn, J. Laredo, D. Shurtleff, D.S. Gann, Ouabain is secreted by the adrenal gland of the awake dogs, Am. J. Physiol. 264 (1993) E413–E419. [39] F. Masugi, T. Ogihara, T. Hasegawa, K. Sagakuchi, Y. Kumahara, Normalization of high plasma level of ouabain-like immunoreactivity in primary aldosteronism after removal of adenoma, J. Hum. Hypertens. 2 (1988) 17–20. [40] A. Vadazs, P. Jakobi, J. Stoler, A. Makler, N. Krivoy, Endogenous digoxin-like immunoreactivity measured in seminal fluid from a normal male population, Gynecol. Obstet. Invest. 33 (1992) 236–238.