ORIGINAL RESEARCH ARTICLE

Journal of
2079
Cellular
Physiology
Nandrolone and Stanozolol
Induce Leydig Cell Tumor
Proliferation Through an
Estrogen-Dependent Mechanism
Involving IGF-I System
ADELE CHIMENTO,1y ROSA SIRIANNI,1y FABIANA ZOLEA,1 ARIANNA DE LUCA,1
MARILENA LANZINO,1 STEFANIA CATALANO,1 SEBASTIANO ANDÒ,2 AND
VINCENZO PEZZI1*
1
Department of Pharmaco-Biology, University of Calabria, Cosenza, Italy
2
Department of Cell Biology, University of Calabria, Cosenza, Italy

Several substances such as anabolic androgenic steroids (AAS), peptide hormones like insulin-like growth factor-I (IGF-I), aromatase
inhibitors and estrogen antagonists are offered via the Internet, and are assumed without considering the potential deleterious effects that
can be caused by their administration. In this study we aimed to determine if nandrolone and stanozolol, two commonly used AAS, could
have an effect on Leydig cell tumor proliferation and if their effects could be potentiated by the concomitant use of IGF-I. Using a rat Leydig
tumor cell line, R2C cells, as experimental model we found that nandrolone and stanozolol caused a dose-dependent induction of
aromatase expression and estradiol (E2) production. When used in combination with IGF-I they were more effective than single molecules
in inducing aromatase expression. AAS exhibited estrogenic activity and induced rapid estrogen receptor (ER)-dependent pathways
involving IGF1R, AKT, and ERK1/2 phosphorylation. Inhibitors for these kinases decreased AAS-dependent aromatase expression. Up-
regulated aromatase levels and related E2 production increased cell proliferation as a consequence of increased cyclin E expression. The
observation that ER antagonist ICI182,780 was also able to significantly reduce ASS- and AAS þ IGF-induced cell proliferation, confirmed a
role for estrogens in AAS-dependent proliferative effects. Taken together these data clearly indicate that the use of high doses of AAS, as it
occurs in doping practice, enhances Leydig cell proliferation, increasing the risk of tumor development. This risk is higher when AAS are
used in association with IGF-I. To our knowledge this is the first report directly associating AAS and testicular cancer.
J. Cell. Physiol. 227: 2079–2088, 2012. ß 2011 Wiley Periodicals, Inc.

Substance abuse has become increasingly widespread among reports highlight a link between AAS abuse and various types of
athletes at subcompetitive and recreational level, raising cancer, mainly to the liver such as hepatocellular adenomas and
concern for human health. In addition to the illicit use of adenocarcinomas (Socas et al., 2005), however other types of
substances to increase performance of athletes and to enhance cancer such as Wilms’ tumors have been reported (Bronson
the muscular mass and strength, more recently the use of some and Matherne, 1997; Modlinski and Fields, 2006).
agents has been extended to non-athletes with the aim to Androgens exert their biological effect through an
combat ageing, obesity and improve appearance or libido intracellular receptor, the androgen receptor (AR), that is
(Calfee and Fadale, 2006). Since its discovery in 1935, numerous present in the reproductive tract as well as in many non-
derivatives of testosterone have been synthesized, with the reproductive tissues, including bone, skeletal muscle, brain,
goals of prolonging its biological activity in vivo, producing orally liver, kidney, and adipocytes. Binding of androgens to AR
active androgens, and developing products, commonly referred determines receptor dimerization, nuclear translocation and
to as anabolic androgenic steroids (AAS), that are more binding to specific responsive elements (ARE) present in the
anabolic and less androgenic than the parent molecule. AAS promoter region of target genes (Li and Al-Azzawi, 2009).
doping is undeniably rampant worldwide. The doses of
testosterone or other androgens used by athletes are
substantially larger than those prescribed for the treatment of The authors have declared that no conflict of interest exists.
androgen deficiency. In one survey (Evans, 1997a), 50% of y
androgen users reported using at least 500 mg of testosterone Adele Chimento and Rosa Sirianni contributed equally to this
study.
weekly or an equivalent dose of another androgen; in another
survey (Pope and Katz, 1994), almost one-fourth of androgen Contract grant sponsor: Ministero della Salute
users assumed 1,000 mg of testosterone weekly or an Contract grant sponsor: Associazione Italiana per la Ricerca sul
equivalent dose of other androgens. It is becoming increasingly Cancro
clear that the abuse of AAS is associated with serious adverse *Correspondence to: Vincenzo Pezzi, Department of Pharmaco-
effects to the liver (Wakabayashi et al., 1984) and the Biology, University of Calabria, 87036 Arcavacata di Rende,
cardiovascular (Fineschi et al., 2001, 2007) central nervous Cosenza, Italy. E-mail: v.pezzi@unical.it
(Shahidi, 2001), musculoskeletal (Al-Ismail et al., 2002), Received 23 May 2011; Accepted 7 July 2011
endocrine (Shahidi, 2001) and reproductive (Dohle et al., 2003; Published online in Wiley Online Library
Eklöf et al., 2003) systems. As a consequence of their effects on (wileyonlinelibrary.com), 18 July 2011.
the endocrine and reproductive systems AAS cause suppressed DOI: 10.1002/jcp.22936
spermatogenesis, gynecomastia, and virilization. Clinical

ß 2 0 1 1 W I L E Y P E R I O D I C A L S , I N C .
2080 CHIMENTO ET AL.

Androgens mechanism of action in skeletal muscle cells is well present study we evaluated on Leydig R2C cells the effects of
documented and includes up-regulation of markers of myogenic commonly used ASS, differentially metabolized by aromatase,
differentiation, such as MyoD and myosin heavy chain II (Singh such as nandrolone (aromatizable) and stanozolol (non-
et al., 2003, 2006; Bhasin et al., 2003). However, androgens can aromatizable), used alone or in association with IGF-I, on
be converted to estrogens through the action of the aromatase aromatase expression and Leydig cell tumor proliferation.
enzyme. In the human, aromatase is expressed in a number of
cells including brain, skin fibroblasts, bone, adipose tissue, in Materials and Methods
steroidogenic tissues such as placenta and gonads (Simpson Cell cultures
et al., 1994), in particular in man aromatase is present in most of
the testicular cells. Estrogens are required for a normal Cells were bought from American Type Culture Collection (LGC
spermatogenesis, which seems extremely sensitive to estrogen Standards, Teddington, Middlesex UK), grown for 2 weeks (four
concentration. Transgenic mice lacking aromatase expression passages) before freezing aliquots. Each aliquot was used for no
(ArKO mice) show an age-dependent disruption of more than ten passages. R2C cells (a rat Leydig tumor cell line)
spermatogenesis, a significant reduction in testis weight and were cultured in Ham/F-10 (Sigma, St. Louis, MO) medium
compromised fertility (Honda et al., 1998; Robertson et al., supplemented with 15% horse serum (HS), 2.5% fetal bovine serum
2002). Similarly, men with inactivating mutations of the (FBS), and antibiotics (Invitrogen, S.r.l., San Giuliano Milanese,
aromatase gene, leading to the lack of the estrogen synthesis, Italy). Cell monolayers were subcultured onto 30 mm dishes for
are infertile (Faustini-Fustini et al., 1999). On the other hand, protein or RNA extraction (1
106 cells/plate), 12-well culture
about half of the male transgenic mice over-expressing plates for steroid measurement (2
105 cells/well) and 24-well
aromatase and presenting enhancement of circulating 17b- culture plates proliferation assay (1
105 cells/well), and used for
estradiol (E2) levels are infertile and/or have enlarged testis and experiments 48 h later.
show Leydig cell hyperplasia and Leydig cell tumors (Fowler Human embryonic kidney (HEK)-293 cells were cultured in
et al., 2000). In a previous study we have shown that Leydig cell Dulbecco’s modified Eagle’s/Ham F12 (DME/F12) medium (GIBCO
tumor is characterized by aromatase over-expression and BRL, Grand Island, NY) supplemented with 5% FBS (Invitrogen,
consequent increased estrogen production, that contributes to S.r.l.) and antibiotics. Cell monolayers were subcultured onto 24-
inducing tumor cell proliferation (Sirianni et al., 2007). well culture plates for transfection experiments.
Testosterone, nandrolone, stanozolol, methandienone, and Cell cultures were treated for the indicated times with
methenolol are the most frequently abused androgens (Pope PD98059, LY294002, GF109203X (Calbiochem, VWR
and Katz, 1994; Evans, 1997a,b). These androgens can be International S.r.l., Milano, Italy), AG1024, ICI182,780 nandrolone,
differentially metabolized by aromatase, specifically nandrolone stanozolol, IGF-I, 17b-estradiol, testosterone, and
can be converted to estrogens, while stanozolol is a non- dihydrotestosterone (Sigma).
aromatizable androgen.
Radioimmunoassay
In addition to the use of androgens, athletes also abuse other
drugs to purportedly enhance muscle building, muscle shaping Prior to experiments, R2C cells were maintained in Ham/F-10
or athletic performance (Evans, 1997a). These accessory drugs without serum or antibiotics (serum-free medium). Cells were
include stimulants, such as amphetamine, clenbuterol, then treated as necessary and estradiol content of medium
ephedrine, and thyroxine, anabolic agents such as growth recovered from each well was determined against standards
hormone (GH), insulin and insulin-like growth factor-I (IGF-I) prepared in serum-free medium using a radioimmunoassay kit
and drugs perceived to reduce adverse effects such as human (Diagnostic System Laboratories, Webster, TX). Results were
chorionic gonadotropin (hCG), aromatase inhibitors or normalized to the cellular protein content per well. For IGF-I
estrogen antagonists (Evans, 1997a). In particular, IGF-I, which determination cells were cultured for 48 h before treatment for an
is the main effector for the action of GH, is a peptide additional 72 h in Ham/F-10 containing 1% dextrane charcoal
physiologically produced by the liver. The potential benefits of coated (DCC) FBS. IGF-I content in medium recovered from each
IGF-I administration include increased muscle protein synthesis well of R2C cells was determined following extraction and assay
and the sparing of glycogenolysis with glycogen synthesis and protocol provided with the rat IGF-I radioimmunoassay kit (DSL
increased fatty acid availability. IGF-I is known to have a role in 2900; Diagnostic System Laboratories).
testicular growth and development and in the control of Leydig
cell number (Saez, 1994). IGF-I is produced locally in the testis, Aromatase activity assay
in Sertoli, Leydig and peritubular cells derived from the
immature testis and cultured in vitro (Casella et al., 1987; Zhou The aromatase activity in subconfluent R2C cell culture medium
and Bondy, 1993). The crucial role of IGF-I in the development was measured by tritiated water-release assay using 0.5 mM
and function of Leydig cells was highlighted by studies on IGF-I [1b-3H(N)]androst-4-ene-3,17-dione (DuPont NEN, Boston, MA)
gene knockout mice (Liu et al., 1993; Baker et al., 1996). The as a substrate as previously shown (Lephart and Simpson, 1991;
failure of adult Leydig cells to mature and the reduced capacity Sirianni et al., 2009).
for testosterone production in IGF-I knock out (KO) mice are Western blot analysis
caused by deregulated expression of testosterone biosynthetic
and metabolizing enzymes (Wang et al., 2003), expression levels Methods for protein extraction and blots preparation have been
of all mRNA species associated with testosterone biosynthesis previously published (Sirianni et al., 2007). Blots were incubated
are lower in the absence of IGF-I. Furthermore, IGF-I plays a overnight at 48C with (a) anti-human P450 aromatase antibody
central role in inducing aromatase expression in Leydig tumor (1:50, Serotec, Oxford, UK), (b) anti-cyclin E (M-20) antibody
cells, consequently IGF-I increases estrogen production that (1:1,000, Santa Cruz Biotechnology, Santa Cruz, CA), (c) anti-
contributes to the induction of tumor cell proliferation (Sirianni pIGF1R antibody (1:500; Cell Signaling Technology, Beverly, MA),
et al., 2007). (d) anti-pERK antibody (1:500; Cell Signaling Technology), (e) anti-
Taking into account that Leydig cell tumors are common in pAKT antibody (1:500; Santa Cruz Biotechnology), (f) anti-IGF1R
young men, the same age group commonly abusing AAS, we antibody (1:500; Santa Cruz Biotechnology), (g) anti-ERK antibody
wanted to investigate the effects of AAS and IGF-I on Leydig cell (1:1,000; Cell Signaling Technology), (h) anti-AKT antibody
tumors. Our hypothesis was that AAS can induce Leydig cell (1:1,000; Santa Cruz Biotechnology), (i) anti-GAPDH antibody
tumor proliferation and that this effect could be potentiated by (1:1,000; Santa Cruz Biotechnology). Membranes were incubated
the concomitant use of IGF-I. To verify this hypothesis in the with horseradish peroxidase (HRP)-conjugated secondary

JOURNAL OF CELLULAR PHYSIOLOGY

 AAS INDUCE LEYDIG CELL TUMOR 2081

antibodies (Amersham Pharmacia Biotech, Piscataway, NJ) and Results

immunoreactive bands were visualized with the ECL Western Nandrolone and stanozolol control Leydig cell
blotting detection system (Amersham Biosciences, Cologno proliferation through the induction of aromatase
Monzese, Italy). To assure equal loading of proteins, membranes expression and estradiol production
were stripped and incubated overnight with GAPDH antibody.
We first evaluated if nandrolone and stanozolol were able to
RNA extraction, reverse transcription and PCR induce R2C cell proliferation. Cell proliferation was measured
in R2C cells treated with increasing doses of nandrolone
The TRizol RNA isolation system (Invitrogen, S.r.l.) was used to (Fig. 1A) and stanozolol (Fig. 1B), revealing a significant
extract RNA from R2C cells. Each RNA sample was treated with induction with all tested doses of both androgens, with 1 mM
DNase I (Ambion, Austin, TX), and purity and integrity of the RNA being the most effective. Since proliferation of R2C cells
were confirmed spectroscopically and by gel electrophoresis depends on estrogen production (Sirianni et al., 2007), we
before use. One microgram of total RNA was reverse transcribed evaluated the effects of nandrolone and stanozolol on
in a final volume of 30 ml using the ImProm-II reverse transcriptase aromatase expression. For this purpose R2C cells were treated
system kit (Promega Italia S.r.l., Milano, Italy). Samples were for 24 h with increasing doses of the two androgens and
aliquoted and stored at 208C. PCR amplification was performed aromatase protein expression was evaluated by Western blot
using 1.5 U of Taq DNA polymerase (Promega Italia S.r.l.) in PCR analysis (Fig. 1C,D). Both androgens were able to significantly
buffer containing 200 mM dNTP, 1.5 mM MgCl2, and 25 pmoles of increase aromatase levels, with maximum induction seen at
each primer in a total volume of 50 ml. Primers sequence for the rat 1 mM (Fig. 1C,D). We also evaluated endogenous estrogen
aromatase gene, PCR conditions and number of cycles were production in response to nandrolone and stanozolol, revealing
previously published (Catalano et al., 2010). L19 ribosomal protein the ability of both androgens to increase estradiol production as
mRNA was used as housekeeping gene, PCR conditions, number of a consequence of effects on aromatase expression (Fig. 1E).
cycles and primers sequence were previously published (Chimento To confirm that proliferative effects depended on the ability of
et al., 2010). To avoid products due to DNA contamination, the two androgens to induce estradiol production, cells were
primers were designed to amplify a region across different exons. treated with nandrolone and stanozolol in the presence of
PCR products were analyzed by electrophoresis on a 2% agarose estrogen receptor (ER) antagonist ICI182,780 (ICI). As seen in
gel stained with ethidium bromide (Sigma–Aldrich Italia S.r.l., Figure 1F ICI reduced the effects of both androgens.
Milano, Italy).
Nandrolone or stanozolol in combination with IGF-I
Assessment of cell proliferation further induce Leydig cell proliferation and aromatase
expression
3-[4,5-Dimethylthiaoly]-2,5-diphenyltetrazolium bromide (MTT)
assay was conducted to detect cell proliferation (Sylvester, 2011). Since we have previously shown a role for IGF-I in Leydig cell
A total of 1
105 cells were seeded onto 24-well plates in proliferation (Sirianni et al., 2007), we also evaluated effects of
complete medium and let grow for 2 days. Prior to experiments, IGF-I treatment combined with nandrolone or stanozolol on
cells were maintained overnight in Ham/F-10 serum-free medium proliferation of R2C cells (Fig. 2). The dose of 1 mM was used in
and the day after treated. There were triplicates for each combination with IGF-I (100 ng/ml) and cell proliferation was
concentration. Forty-eight hours after treatment fresh MTT re- measured (Fig. 2A). Results obtained showed that both
suspended in PBS was added to each well (final concentration nandrolone and stanozolol have an additive effect with IGF-I in
0.33 mg/ml). After 1 h incubation, the culture media were discarded inducing proliferation (Fig. 2A). Also in this case, the estrogen
and replaced with 100 ml of DMSO. The optical density was antagonist ICI was able to significantly reduce both AAS- and
measured at 570 nm in a spectrophotometer. AAS þ IGF-I-dependent proliferative effects (Fig. 2A). An
additive effect in inducing aromatase expression was found
Transfection assay when IGF was combined with nandrolone or stanozolol
(Fig. 2B,C). Data obtained on aromatase expression were also
Before transfection, complete medium was removed, and 0.5 ml of mirrored by effects on aromatase activity (Fig. 2D).
DMEM/F12 without phenol red, serum or antibiotics was added to IGF-I receptor (IGF1R) activation by direct ligand binding or
the plates. Transfection was performed using Fugene6 reagent by indirect transactivation determines activation of three major
(Roche Diagnostics, Mannheim, Germany), following the transductional pathways: Ras/Raf/mitogen-activated protein
manufacturer’s instruction. Plasmids were used at the kinase (MAPK), phosphatidylinositol 3-kinase (PI3K)/AKT, and
concentration of 0.5 mg/well for the XETL (Bunone et al., 1996) phospholipase C (PLC)/protein kinase C (PKC). We tested the
promoter-luciferase reporter plasmid, of 0.1 mg/well for ERa effects on cell proliferation of specific inhibitors such as
expression vector (Tora et al., 1989), of 10 ng/well for the b- PD98059 (PD) for MAPK; LY294002 (LY) for PI3K/AKT,
galactosidase control vector (Promega Italia S.r.l.). Four hours after GF109203X (GFX) for PLC/PKC and AG1024 (AG) for IGF1R,
transfection, the medium was removed and replaced with DMEM/ used in combination with nandrolone or stanozolol (Fig. 3A).
F12 without phenol red, serum or antibiotics and supplemented The use of AG, LY, and GFX was able to reverse nandrolone
with the indicated concentrations of treatments for 24 h. Cells and stanozolol-induced cell proliferation (Fig. 3A). PD was the
were lysed using the passive lysis buffer (Promega Italia S.r.l.), and only inhibitor that did not interfere with cell proliferation. To
enzymatic activities were assayed using the Luciferase (Promega demonstrate the link between cell proliferation and estrogen
Italia S.r.l.) and b-galactosidase (Ambion) assay systems following production we evaluated the effects of the inhibitors on
the manufacturer’s instructions. Firefly luciferase values of each aromatase expression (Fig. 3B–D). AG, LY, and GFX were able
sample were normalized by b-galactosidase activity and data were to reduce nandrolone- and stanozolol-induced aromatase
reported as relative light units (RLU) values. mRNA (Fig. 3B) and protein (Fig. 3C,D) levels.
Statistical analysis The ability of inhibitors for IGF-I-dependent pathways to
control nandrolone and stanozolol effects led us to suppose
All experiments were conducted at least three times and results that the two AAS could regulate IGF-I production in R2C cells.
were from representative experiments. Data were expressed as Basal IGF-I content in R2C cell culture medium was 80 mg/ml/
mean values  standard deviation (SD) and statistical significance mg protein and was increased by 2- and 3.4-fold by nandrolone
was analyzed by Student–Newman–Keuls multiple comparison and stanozolol, respectively (Fig. 4A). When IGF-I was
methods, using SigmaStat version 3.0 (SPSS, Chicago, IL). neutralized with a specific antibody cell proliferation was

JOURNAL OF CELLULAR PHYSIOLOGY

2082 CHIMENTO ET AL.

Fig. 1. Effects of nandrolone and stanozolol on aromatase expression and estradiol production in R2C cells. A,B: R2C cells were left untreated
(basal) or treated for 24 h with nandrolone (nandro) (A) or stanozolol (stano) (B) at the indicated concentrations. Cell proliferation was assessed
using the MTT method as indicated in the Materials and Methods Section. Final results represent mean W SD of three independent experiments
each performed in triplicate (MP < 0.001 compared with basal). C,D: R2C cells were left untreated (basal) or treated for 24 h with nandrolone
(nandro) (C) or stanozolol (stano) (D) at the indicated concentrations. Western blot analysis for aromatase was performed on 50 mg of total
proteins. Blots are representative of three independent experiments with similar results. GAPDH was used as a loading control. Graphs represent
means of aromatase optical densities normalized to GAPDH content of the same sample (MP < 0.001 compared with basal). E: Cells were treated
with nandrolone (nandro) (1 mM) or stanozolol (stano) (1 mM) for 48 h. E2 content in cell culture medium, measured by RIA, was normalized to the
well protein content. Values represent means W SD of three different experiments, each performed in triplicate (MP < 0.001 compared with basal).
F: R2C cells were left untreated (basal) or treated for 48 h with nandrolone (nandro) (1 mM), stanozolol (stano) (1 mM), and ICI182,780 (ICI) (10 mM)
alone or in combination. Proliferation was assessed using the MTT method as indicated in the Materials and Methods Section. Final results
represent mean W SD of three independent experiments (MP < 0.001 compared with basal; MMP < 0.001 compared with nandro; -P < 0.005 compared
with stano).

markedly decreased (Fig. 4B). Nandrolone and stanozolol could of regulation of gene expression (Foradori et al., 2008). For this
not overcome the inhibitory effect produced by the antibody reason we decided to investigate the ability of nandrolone and
(Fig. 4B). These data confirm that R2C proliferation is tightly stanozolol to activate rapid signaling involving IGF1R
dependent on IGF-I and indicate that a mechanism through phosphorylation and activation of kinases involved in cell
which nandrolone and stanozolol control R2C proliferation is proliferation. R2C cells were treated for 10 min with increasing
through the induction of IGF-I production. doses of the two androgens and IGF1R, ERK and AKT
phosphorylation was evaluated by Western blot analysis. Both
Nandrolone and stanozolol activate rapid signaling androgens were capable of increasing kinase activity at all used
through ERa transactivation doses (Fig. 5A,B).
The effects produced by the two AAS on the activation of
Recently it has been demonstrated that androgens can activate IGF1R-dependent pathways were also reproduced by E2,
alternative pathways, in addition to classic genomic mechanisms dihydrotestosterone (DHT), and testosterone (T) (Fig. 5C). In

JOURNAL OF CELLULAR PHYSIOLOGY

 AAS INDUCE LEYDIG CELL TUMOR 2083

Fig. 2. Additive effects of nandrolone or stanozolol and IGF-I on cell proliferation and aromatase expression and activity. A: R2C cells were left
untreated (basal) or treated for 48 h with IGF-I (100 ng/ml), nandrolone (nandro) (1 mM), stanozolol (stano) (1 mM), and ICI182,780 (ICI) (10 mM)
used alone or in combination. Cell proliferation was assessed using the MTT method as indicated in the Materials and Methods Section. Final results
represent mean W SD of three independent experiments each performed in triplicate (MP < 0.001 compared with basal; -P < 0.001 compared with
IGF-I; MMP < 0.005 compared with IGF-I R nandrolone; #P < 0.001 compared with IGF-I R stano). B,C: R2C cells were left untreated (basal) or treated
for 24 h with IGF-I (100 ng/ml) used alone or in combination with nandro (1 mM) (B) or stano (1 mM) (C). Western blot analysis for aromatase was
performed on 50 mg of total proteins. Blots are representative of three independent experiments with similar results. GAPDH was used as a loading
control. Aromatase optical densities were normalized to GAPDH content of the same sample. Graphs represent means of values from three blots,
where basal values were assumed as 100 (MP < 0.001 compared with basal). D: R2C cells were left untreated (basal) or treated for 48 h with nandro
(1 mM) or stano (1 mM) and IGF-I (100 ng/ml) alone or in combination. Aromatase activity was assessed using the modified tritiated water method.
Result obtained were measured in pmoles of released [3H]H2O, normalized to the well protein content (pmol/mg protein/h) and expressed as fold
over basal. Values represent the mean W SD of three independent experiments each performed in triplicate (MP < 0.001 compared with basal;
MM
P < 0.001 compared with IGF-I).

JOURNAL OF CELLULAR PHYSIOLOGY

2084 CHIMENTO ET AL.

Fig. 3. Effects of IGF-I pathway inhibitors on nandrolone and stanozolol induced cell proliferation and aromatase expression. A–D: R2C cells were
left untreated (basal) or treated in serum-free medium for 48 h (A) or 24 h (B–D) with nandrolone (nandro) (1 mM) or stanozolol (stano) (1 mM)
alone or in combination with AG1024 (10 mM) (AG), LY294002 (10 mM) (LY), PD98059 (10 mM) (PD) and GF109203X (10 mM) (GFX). A: Cell
proliferation was assessed using the MTT method as indicated in the Materials and Methods Section. Final results represent mean W SD of three
independent experiments each performed in triplicate (MP < 0.001 compared with basal; MMP < 0.001 compared with nandrolone; RP < 0.001
compared with stanozolol). B: Total RNA was extracted from cells untreated (basal) and treated as indicated. RT-PCR was used to analyze mRNA
levels of CYP19. L19 was used as housekeeping gene. Negative control (neg) was obtained using water instead of cDNA. Image is representative of
three independent experiments with similar results. C,D: Western blot analysis for aromatase was performed on 50 mg of total proteins. Blots are
representative of three independent experiments with similar results. GAPDH was used as a loading control. Aromatase optical densities were
normalized to GAPDH content of the same sample. Graphs represent means of values from three blots, where basal values were assumed as 100
(MP < 0.001 compared with basal; MMP < 0.001 compared with nandrolone; RP < 0.001 compared with stanozolol).

addition the presence of specific inhibitors for IGF1R, ERK1/2, Our results demonstrate the ability of both androgens to
and AKT were able to prevent kinases activation (Fig. 5C). transactivate ERa (Fig. 5E).
There are numerous reports suggesting that androgens
might act through ERa by themselves (Maggiolini et al., 1999) Nandrolone and stanozolol influence cell proliferation by
and our data shown in Figures 1F and 2A indicate that this could increasing cyclin E expression levels
also be the case for the two AAS in R2C cells. To define if the To explain the observed effects on cell proliferation we
rapid AAS-dependent signaling was mediated by ER, R2C cells evaluated expression of cyclin E, that we have previously shown
were treated with the two androgens in the presence of the ER to be up-regulated by IGF-I in R2C cells (Sirianni et al., 2007).
antagonist ICI182,780. As shown in Figure 5D ICI Cyclin E protein expression was increased by all used doses of
administration reduced the effect of both androgens on IGF1R both nandrolone (Fig. 6A) and stanozolol (Fig. 6B). In addition,
phosphorylation. combined treatments with IGF-I and either steroid, caused a
To evaluate the specificity of nandrolone and stanozolol to stronger induction of cyclin E expression (Fig. 6C,D).
transactivate ERa we performed co-transfection experiments
using a luciferase reporter plasmid (XETL), a construct Discussion
containing an estrogen-responsive element (Bunone et al.,
1996), and the human ERa expression vector (HEGO) (Tora Leydig cells are the main site of testosterone biosynthesis in the
et al., 1989). To avoid interference by endogenous ERs male mammals, as well as a site of conversion of testosterone to
transfections were performed in HEK293 cells that do not estradiol through aromatase activity (Carreau et al., 2006).
express ERs. Eighteen hours after transfection cells were Alterations in local estrogen synthesis may have significant
exposed to increasing concentrations of the two androgens. consequences in malignancy of these cells. In a previous study,

JOURNAL OF CELLULAR PHYSIOLOGY

 AAS INDUCE LEYDIG CELL TUMOR 2085

the effects of high doses (0.1–10 mM) of nandrolone and

stanozolol. When we evaluated the effects of these two
androgens on aromatase expression we found that they are
both able to increase the enzyme protein levels and
consequently estrogen production in R2C cells, with stanozolol
being more effective in inducing the enzyme but nandrolone
more effective in increasing estrogen levels. A plausible
explanation for this behavior is that nandrolone is immediately
metabolized to estradiol by R2C cells, where aromatase is
constitutively active, and the amount left to activate the
mechanism responsible for increasing aromatase transcription
is less than the concentration effectively used to treat cells. On
the other side, stanozolol is not metabolized by aromatase, and
the concentration used to treat cells is effectively bio-available
to determine the effects on aromatase transcription. Thus, both
AAS contribute to estrogen production determining an effect
on tumor cell proliferation that we could define ‘‘indirect,’’ not
related to their androgenic properties, but estrogen-
dependent. This effect is confirmed by the observation that ICI
reduces AAS-dependent cell proliferation. However, since we
found that treatments with inhibitors for IGF1R, PKC, and PI3K
were able to reduce AAS effects on aromatase expression and
cell proliferation, it could also be suggested that these
molecules directly activate non-genomic hormonal signals. We
supposed that the two AAS could activate ER-dependent
pathways. The estrogenic actions of androgens are established
(Rochefort and Garcia, 1983), and the possibility for AAS to
activate ERs is consistent, also in our experimental system, with
the ability of ICI to reduce AAS-dependent cell proliferation. In
addition, AAS acting on a membrane bound ER activate IGF1R
and downstream kinases, as seen in several cell systems (Song
et al., 2006). These events could partially explain the inhibition
produced by AG1024, the IGF1R inhibitor, and by downstream
IGF-I signaling inhibitors (GFX for PKC and LY for PI3K). The
use of high doses of AAS could cause in our cell system ER/AR
desensitization, explaining the reduced effects seen with 10 mM
of both nandrolone and stanozolol. Our results give interesting
Fig. 4. Nandrolone and stanozolol effects on cell proliferation indications and open several hypothesis on possible non-
depend on IGF-I production. A: R2C cells were left untreated (basal) genomic molecular mechanisms induced by AAS involving IGF-I
or treated for 48 h with nandrolone (nandro) (1 mM) or stanozolol dependent signaling pathways. For example, GPR30, a novel ER
(stano) (1 mM) in 1% DCC. IGF-I levels in culture medium were (Prossnitz and Maggiolini, 2009), could also be activated by AAS,
determined by RIA and IGF-I content was normalized to the cell
culture well protein content. Data represent the mean W SEM of explaining why ICI was not able to completely block
values from three separate cell culture wells expressed as fold over nandrolone- and stanozolol-dependent IGF1R activation.
basal. (MP < 0.01 compared with basal conditions). B: Cell proliferation Additionally, the two AAS can also work through a membrane
was assessed using the MTT method as indicated in the Materials and AR to activate IGF1R signaling.
Methods Section. IGF-I antibody (IGF-I Ab) was added to the medium
at 5 mg/ml 24 h before being treated for an additional 24 h with the It was shown for human primary prostatic stromal cell
indicated concentrations of nandrolone and stanozolol. Columns, cultures, that administration of DHT and T, but not of E2,
mean percent of untreated (basal) cells (100%) from three modulated IGF-I protein expression (Le et al., 2006). In our cell
independent experiments each done in triplicate; bars, SD. (MP < 0.01 model the two androgens induce IGF-I production, and when
compared with basal).
this growth factor was immunoneutralized, cell proliferation
was deeply reduced and nandrolone and stanozolol lost their
ability to increase cell proliferation, confirming our previous
report indicating IGF-I as indispensable for R2C cell
we observed that R2C tumor Leydig cells release a conspicuous proliferation (Sirianni et al., 2007). In addition, the observation
amount of E2, significantly higher than normal Leydig cell that stanozolol was more effective than nandrolone in inducing
cultures, as a consequence of aromatase over-expression in IGF-I production, supports the hypothesis that nandrolone
tumor cells (Sirianni et al., 2007). It can be suggested that in the metabolism to estradiol reduces its bio-available amount.
presence of increased androgens availability, their metabolism IGF-I signaling is highly involved in cancer development and
through aromatase, expressed in Leydig cells, increases local progression (Larsson et al., 2005) exerting powerful effects on
estrogen levels, contributing to the initiation or progression of each of the key stages of cancer formation: cellular
Leydig cell tumor. We tested the effects of two commonly used proliferation, apoptosis, angiogenesis, metastasis and resistance
AAS nandrolone (aromatizable) and stanozolol (non- to chemotherapeutic agents. We previously showed that IGF-I,
aromatizable) and evaluated the effects on aromatase endogenously produced by Leydig tumor cells, activating PI3K/
expression and Leydig cell proliferation. Since muscle mass and AKT and PLC/PKC (but not MAPK) pathways determines an
strength are correlated with the administered dose and the increase in aromatase expression and estrogen production
circulating concentrations of AAS (Bhasin et al., 2001, 2005), inducing cell proliferation (Sirianni et al., 2007). In the current
the doses commonly used in practice are extremely high and study we confirmed these data and tested also the effect of
range from 500 to 1,000 mg/weekly (Pope and Katz, 1994; combined treatments of nandrolone or stanozolol with IGF-I.
Evans, 1997a). These observations support our decision to test Effects of the two AAS were potentiated by the presence of

JOURNAL OF CELLULAR PHYSIOLOGY

2086 CHIMENTO ET AL.

Fig. 5. Effects of nandrolone or stanozolol on ER-activated IGF1R signaling. R2C cells were left untreated (basal) or treated for 10 min with
nandrolone (nandro) (A) or stanozolol (stano) (B) at the indicated concentrations. C: R2C cells were left untreated (basal) or treated for 10 min with
E2 (100 nM), dihydrotestosterone (DHT) (100 nM), testosterone (T) (100 nM), nandrolone (1 mM), stanozolol (1 mM) combined with AG1024
(10 mM) (AG), PD98059 (10 mM) (PD), and LY294002 (10 mM) (LY). D: R2C cells were left untreated (basal) or treated for 10 min with nandrolone
(1 mM), stanozolol (1 mM) and ICI182,780 (ICI) (10 mM) used alone or in combination. Western blot analysis was performed on 50 mg of total
proteins. Blots are representative of three independent experiments with similar results. GAPDH (A,B,D) or alternatively IGF1R, ERK1/2, and
AKT (C) were used as loading control. E: HEK293 cells were transiently transfected using XETL reporter plasmid and ERa expression vector and
treated with the indicated doses of nandrolone or stanozolol. Data were normalized to the coexpressed b-galactosidase expression vector and
expressed as RLU. Results represent the mean W SD of data from three independent experiments, each performed in triplicate. (MP < 0.001
compared with basal).

IGF-I. The evaluation of aromatase activity revealed that the mass (Bhasin et al., 2001) in correlation with the administered
combined treatments of nandrolone or stanozolol with IGF-I dose and the circulating concentrations (Bhasin et al., 2005), it
increased the enzyme activity to levels above those seen with should be seriously taken into account the potential dangerous
the single treatments. All the effects seen on aromatase were effects produced by the use of AAS on the activation of
mirrored by changes in R2C proliferative behavior; nandrolone pathways involved in the progression of a type of cancer, such as
and stanozolol increase cell proliferation and their effects are Leydig cell tumor, with high incidence in young people, the same
additive with IGF-I. One mechanism through which estrogens age people abusing AAS. In this study we have shown that
and IGF-I induce cell proliferation is by increasing protein levels aromatizable and non-aromatizable androgens can promote
of G1 regulatory cyclins A, B1, D1, D3, and E in target cells (Prall testicular tumor development through the induction of
et al., 1997; Ma et al., 2009). In our study, we showed that the aromatase expression, estrogen and IGF-I production. In
expression of one of the most important regulators of Leydig addition, the two tested androgens binding ERa transactivated
cell cycle, cyclin E, can be increased by nandrolone and IGF1R activating PI3K and PKC pathways determining an
stanozolol and significantly increased to a greater extent by the induction of R2C cell proliferation. The examined AAS effects
combined treatment with IGF-I. These results further confirm are potentiated by the concomitant use of IGF-I. It would be
that AAS could contribute to activate expression of estrogen- interesting to determine the effect of the two AAS in human
and IGF-I-dependent genes involved in cell cycle progression. cultures of Leydig cells, that, however are currently not
In conclusion, despite the growing body of data over the past available, as well as in vivo in an animal model. Before beginning
decade that has established that androgens increase muscle any illegal and self-determined use of doping agents deep

JOURNAL OF CELLULAR PHYSIOLOGY

 AAS INDUCE LEYDIG CELL TUMOR 2087

Fig. 6. Additive effects of nandrolone or stanozolol and IGF-I on cyclin E expression in R2C cells. R2C cells were left untreated (basal) or treated for
24 h with nandrolone (nandro) (A) or stanozolol (stano) (B) at the indicated concentrations, nandrolone (nandro) (1 mM) plus IGF-I (100 ng/ml) (C)
or stanozolol (stano) (1 mM) plus IGF-I (100 ng/ml) (D). Western blot analysis for cyclin E was performed on 50 mg of total proteins. Blots are
representative of three independent experiments with similar results. GAPDH was used as a loading control. Cyclin E optical densities were
normalized to GAPDH content of the same sample. Graphs represent means of values from three blots, where basal values were assumed as 1.
(MP < 0.001 compared with basal; RP < 0.001 compared with IGF-I).

consideration should be given to the deleterious effects Baker J, Hardy MP, Zhou J, Bondy C, Lupu F, Bellve AR, Efstratiadis A. 1996. Effects of an Igf1
associated with their use, among which we can now include gene null mutation on mouse reproduction. Mol Endocrinol 10:903–918.
Bhasin S, Woodhouse L, Casaburi R, Singh AB, Bhasin D, Berman N, Chen X, Yarasheski KE,
Leydig cell tumor. Magliano L, Dzekov C, Dzekov J, Bross R, Phillips J, Sinha-Hikim I, Shen R, Storer TW. 2001.
Testosterone dose-response relationships in healthy young men. Am J Physiol-Endocrinol
Metab 281:E1172–E1181.
Acknowledgments Bhasin S, Taylor W, Singh R, Artaza J, Sinha-Hikim I, Jasuja R, Choi H, Gonzalez-Cadavid N.
2003. The mechanisms of androgen effects on body composition: Mesenchymal
This work was supported by Ministero della Salute— pluripotent cell as the target of androgen action. J Gerontol Ser A Biol Sci Med Sci
58:M1103–M1110.
Dipartimento dell’Innovazione, Direzione Generale della Bhasin S, Woodhouse L, Casaburi R, Singh AB, Mac RP, Lee M, Yarasheski KE, Sinha-Hikim I,
Ricerca Scientifica e Tecnologica—Commissione per la Dzekov C, Dzekov J, Magliano L, Storer TW. 2005. Older men are as responsive as young
men to the anabolic effects of graded doses of testosterone on the skeletal muscle. J Clin
Vigilanza ed il Controllo sul Doping e per la Tutela Della Salute Endocrinol Metab 90:678–688.
Nelle Attività Sportive. Convenzione n. 2007-11 and by Bronson FH, Matherne CM. 1997. Exposure to anabolic-androgenic steroids shortens life
Associazione Italiana per la Ricerca sul Cancro. span of male mice. Med Sci Sports Exerc 29:615–619.
Bunone G, Briand P, Miksicek R, Picard D. 1996. Activation of the unliganded estrogen
receptor by EGF involves the MAP kinase pathway and direct phosphorylation. EMBO J
15:2174–2183.
Literature Cited Calfee R, Fadale P. 2006. Popular ergogenic drugs and supplements in young athletes.
Pediatrics 117:e577–589.
Al-Ismail K, Torreggiani W, Munk P, Nicolaou S. 2002. Gluteal mass in a bodybuilder: Carreau S, Delalande C, Silandre D, Bourguiba S, Lambard S. 2006. Aromatase and estrogen
Radiological depiction of a complication of anabolic steroid use. Eur Radiol 12:1366–1369. receptors in male reproduction. Mol Cell Endocrinol 246:65–68.

JOURNAL OF CELLULAR PHYSIOLOGY

2088 CHIMENTO ET AL.

Casella SJ, Smith EP, van Wyk JJ, Joseph DR, Hynes MA, Hoyt EC, Lund PK. 1987. Isolation of Prall OWJ, Sarcevic B, Musgrove EA, Watts CKW, Sutherland RL. 1997. Estrogen-induced
rat testis cDNAs encoding an insulin-like growth factor I precursor. DNA 6:325–330. activation of Cdk4 and Cdk2 during G1-S phase progression is accompanied by increased
Catalano S, Malivindi R, Giordano C, Gu G, Panza S, Bonofiglio D, Lanzino M, Sisci D, Panno cyclin D1 expression and decreased cyclin-dependent kinase inhibitor association with
ML, Andò S. 2010. Farnesoid X receptor, through the binding with steroidogenic factor 1- cyclin E-Cdk2. J Biol Chem 272:10882–10894.
responsive element, inhibits aromatase expression in tumor Leydig cells. J Biol Chem Prossnitz ER, Maggiolini M. 2009. Mechanisms of estrogen signaling and gene expression via
285:5581–5593. GPR30. Mol Cell Endocrinol 308:32–38.
Chimento A, Sirianni R, Delalande C, Silandre D, Bois C, Andò S, Maggiolini M, Carreau S, Robertson KM, O’Donnell L, Simpson ER, Jones MEE. 2002. The Phenotype of the aromatase
Pezzi V. 2010. 17b-estradiol activates rapid signaling pathways involved in rat pachytene knockout mouse reveals dietary phytoestrogens impact significantly on testis function.
spermatocytes apoptosis through GPR30 and ERa. Mol Cell Endocrinol 320:136–144. Endocrinology 143:2913–2921.
Dohle GR, Smit M, Weber RF. 2003. Androgens and male fertility. World J Urol 21:341–345. Rochefort H, Garcia M. 1983. The estrogenic and antiestrogenic activities of androgens in
Eklöf AC, Thurelius AM, Garle M, Rane A, Sjöqvist F. 2003. The anti-doping hot-line, a means female target tissues. Pharmacol Ther 23:193–216.
to capture the abuse of doping agents in the Swedish society and a new service function in Saez JM. 1994. Leydig cells: Endocrine, paracrine, and autocrine regulation. Endocr Rev
clinical pharmacology. Eur J Clin Pharmacol 59:571–577. 15:574–626.
Evans NA. 1997a. Gym and tonic: A profile of 100 male steroid users. Br J Sport Med 31:54–58. Shahidi NT. 2001. A review of the chemistry, biological action, and clinical applications of
Evans NA. 1997b. Local complications of self administered anabolic steroid injections. Br J anabolic-androgenic steroids. Clin Ther 23:1355–1390.
Sport Med 31:349–350. Simpson ER, Mahendroo MS, Means GD, Kilgore MW, Hinshelwood MM, Lorence S,
Faustini-Fustini M, Rochira V, Carani C. 1999. Oestrogen deficiency in men: Where are we Amarneh B, Ito Y, Fisher CR, Michael MD. 1994. Aromatase cytochrome P450, the enzyme
today? Eur J Endocr 140:111–129. responsible for estrogen biosynthesis. Endocr Rev 15:342–355.
Fineschi V, Baroldi G, Monciotti F, Reattelli LP, Turillazzi E. 2001. Anabolic steroid abuse and Singh R, Artaza JN, Taylor WE, Gonzalez-Cadavid NF, Bhasin S. 2003. Androgens stimulate
cardiac sudden death. Arch Pathol Lab Med 125:253–255. myogenic differentiation and inhibit adipogenesis in C3H 10T1/2 pluripotent cells through
Fineschi V, Riezzo I, Centini F, Silingardi E, Licata M, Beduschi G, Karch S. 2007. Sudden an androgen receptor-mediated pathway. Endocrinology 144:5081–5088.
cardiac death during anabolic steroid abuse: Morphologic and toxicologic findings in two Singh R, Artaza JN, Taylor WE, Braga M, Yuan X, Gonzalez-Cadavid NF, Bhasin S. 2006.
fatal cases of bodybuilders. Int J Legal Med 121:48–53. Testosterone inhibits adipogenic differentiation in 3T3-L1 cells: Nuclear translocation
Foradori CD, Weiser MJ, Handa RJ. 2008. Non-genomic actions of androgens. Front of androgen receptor complex with beta-catenin and T-cell factor 4 may bypass canonical
Neuroendocrinol 29:169–181. wnt signaling to down-regulate adipogenic transcription factors. Endocrinology 147:141–
Fowler KA, Gill K, Kirma N, Dillehay DL, Tekmal RR. 2000. Overexpression of aromatase 154.
leads to development of testicular Leydig cell tumors: An in vivo model for hormone- Sirianni R, Chimento A, Malivindi R, Mazzitelli I, Ando S, Pezzi V. 2007. Insulin-like growth
mediated testicular cancer. Am J Pathol 156:347–353. factor-I, regulating aromatase expression through steroidogenic factor 1, supports
Honda SI, Harada N, Ito S, Takagi Y, Maeda S. 1998. Disruption of sexual behavior in male estrogen-dependent tumor Leydig cell proliferation. Cancer Res 67:8368–8377.
aromatase-deficient mice lacking exons 1 and 2 of the cyp19 gene. Biochem Biophys Res Sirianni R, Chimento A, De Luca A, Zolea F, Carpino A, Rago V, Maggiolini M, Andò S, Pezzi V.
Commun 252:445–449. 2009. Inhibition of cyclooxygenase-2 down-regulates aromatase activity and decreases
Larsson O, Girnita A, Girnita L. 2005. Role of insulin-like growth factor 1 receptor signalling in proliferation of Leydig tumor cells. J Biol Chem 284:28905–28916.
cancer. Br J Cancer 92:2097–2101. Socas L, Zumbado M, Pérez-Luzardo O, Ramos A, Pérez C, Hernández JR, Boada LD. 2005.
Le H, Arnold JT, McFAnn KK, Blackman MR. 2006. DHT and testosterone, but not DHEA or Hepatocellular adenomas associated with anabolic androgenic steroid abuse in
E2, differentially modulate IGF-I, IGFBP-2, and IGFBP-3 in human prostatic stromal cells. bodybuilders: A report of two cases and a review of the literature. Br J Sports Med 39:e27.
Am J Physiol Endocrinol Metab 290:E952–E960. Song RXD, Fan P, Yue W, Chen Y, Santen RJ. 2006. Role of receptor complexes in the
Lephart ED, Simpson ER. 1991. Assay of aromatase activity. Methods Enzymol 206:477–483. extranuclear actions of estrogen receptor alpha in breast cancer. Endocr Relat Cancer
Li J, Al-Azzawi F. 2009. Mechanism of androgen receptor action. Maturitas 63:142–148. 13:S3–S13.
Liu JP, Baker J, Perkins AS, Robertson EJ, Efstratiadis A. 1993. Mice carrying null mutations of Sylvester PW. 2011. Optimization of the tetrazolium dye (MTT) colorimetric assay for
the genes encoding insulin-like growth factor I (Igf-1) and type 1 IGF receptor (Igf1r). Cell cellular growth and viability. Methods Mol Biol 716:157–168.
75:59–72. Tora L, Mullick A, Metger D, Ponglikitmongkol M, Park I, Chambon P. 1989. The cloned
Ma QL, Yang TL, Yin JY, Peng ZY, Yu M, Liu ZQ, Chen FP. 2009. Role of insulin-like growth human oestrogen receptor contains a mutation which alters its hormone binding
factor-1 (IGF-1) in regulating cell cycle progression. Biochem Biophys Res Commun properties. EMBO J 8:1981–1986.
389:150–155. Wakabayashi T, Onda H, Tada T, Iijima M, Itoh Y. 1984. High incidence of peliosis hepatis in
Maggiolini M, Donzè O, Jeannin E, Andò S, Picard D. 1999. Adrenal androgens stimulate the autopsy cases of aplastic anemia with special reference to anabolic steroid therapy. Acta
proliferation of breast cancer cells as direct activators of estrogen receptor a. Cancer Res Pathol Jpn 34:1079–1086.
59(19):4864–4869. Wang GM, O’Shaughnessy PJ, Chubb C, Robaire B, Hardy MP. 2003. Effects of insulin-like
Modlinski R, Fields K. 2006. The effect of anabolic steroids on the gastrointestinal system, growth factor I on steroidogenic enzyme expression levels in mouse Leydig cells.
kidneys, and adrenal glands. Curr Sports Med Rep 5:104–109. Endocrinology 144:5058–5064.
Pope H, Katz D. 1994. Psychiatric and medical effects of anabolic-androgenic steroid use: A Zhou J, Bondy C. 1993. Anatomy of the insulin-like growth factor system in the human testis.
controlled study of 160 athletes. Arch Gen Psychiatry 51:375–382. Fertil Steril 60:897–904.

JOURNAL OF CELLULAR PHYSIOLOGY