neurotoxic properties of the anabolic androgenic steroids ... · neurotoxic properties of the...

10
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/49806988 Neurotoxic properties of the anabolic androgenic steroids nandrolone and methandrostenolone in primary neuronal cultures Article in Journal of Neuroscience Research · April 2011 DOI: 10.1002/jnr.22578 · Source: PubMed CITATIONS 19 READS 179 8 authors, including: Some of the authors of this publication are also working on these related projects: Computational modeling to avoid drug resistance in melanoma patients View project Filippo Caraci University of Catania 114 PUBLICATIONS 2,583 CITATIONS SEE PROFILE Antonino Corsaro University of Catania 171 PUBLICATIONS 1,655 CITATIONS SEE PROFILE Ferdinando Nicoletti University of Catania 758 PUBLICATIONS 23,570 CITATIONS SEE PROFILE All content following this page was uploaded by Filippo Caraci on 15 January 2017. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

Upload: others

Post on 25-Jun-2020

18 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/49806988

Neurotoxicpropertiesoftheanabolicandrogenicsteroidsnandroloneandmethandrostenoloneinprimaryneuronalcultures

ArticleinJournalofNeuroscienceResearch·April2011

DOI:10.1002/jnr.22578·Source:PubMed

CITATIONS

19

READS

179

8authors,including:

Someoftheauthorsofthispublicationarealsoworkingontheserelatedprojects:

ComputationalmodelingtoavoiddrugresistanceinmelanomapatientsViewproject

FilippoCaraci

UniversityofCatania

114PUBLICATIONS2,583CITATIONS

SEEPROFILE

AntoninoCorsaro

UniversityofCatania

171PUBLICATIONS1,655CITATIONS

SEEPROFILE

FerdinandoNicoletti

UniversityofCatania

758PUBLICATIONS23,570CITATIONS

SEEPROFILE

AllcontentfollowingthispagewasuploadedbyFilippoCaracion15January2017.

Theuserhasrequestedenhancementofthedownloadedfile.Allin-textreferencesunderlinedinblueareaddedtotheoriginaldocumentandarelinkedtopublicationsonResearchGate,lettingyouaccessandreadthemimmediately.

Page 2: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

Neurotoxic Properties of the AnabolicAndrogenic Steroids Nandrolone andMethandrostenolone in Primary NeuronalCultures

Filippo Caraci,1 V. Pistara,2 A. Corsaro,2 Flora Tomasello,3

Maria Laura Giuffrida,1 Maria Angela Sortino,4 Ferdinando Nicoletti,5,6

and Agata Copani1,7*1Department of Pharmaceutical Sciences, University of Catania, Catania, Italy2Department of Chemical Sciences, University of Catania, Catania, Italy3PhD Program in Neuropharmacology, University of Catania, Catania, Italy4Department of Experimental and Clinical Pharmacology, University of Catania, Catania, Italy5Department of Human Physiology and Pharmacology, University of Rome ‘‘Sapienza,’’ Rome, Italy6I.N.M. Neuromed, Pozzilli, Italy7I.B.B., CNR-Catania, Italy

Anabolic-androgenic steroid (AAS) abuse is associatedwith multiple neurobehavioral disturbances. The sites ofaction and the neurobiological sequels of AAS abuseare unclear at present. We investigated whether twodifferent AASs, nandrolone and methandrostenolone,could affect neuronal survival in culture. The endoge-nous androgenic steroid testosterone was used forcomparison. Both testosterone and nandrolone wereneurotoxic at micromolar concentrations, and theireffects were prevented by blockade of androgen recep-tors (ARs) with flutamide. Neuronal toxicity developedonly over a 48-hr exposure to the steroids. The cell-impermeable analogues testosterone-BSA and nandro-lone-BSA, which preferentially target membrane-associ-ated ARs, were also neurotoxic in a time-dependentand flutamide-sensitive manner. Testosterone-BSA andnandrolone-BSA were more potent than their parentcompounds, suggesting that membrane-associatedARs were the relevant sites for the neurotoxic actionsof the steroids. Unlike testosterone and nandrolone,toxicity by methandrostenolone and methandrosteno-lone-BSA was insensitive to flutamide, but it was pre-vented by the glucocorticoid receptor (GR) antagonistRU-486. Methandrostenolone-BSA was more potentthan the parent compound, suggesting that its toxicityrelied on the preferential activation of putative mem-brane-associated GRs. Consistently with the evidencethat membrane-associated GRs can mediate rapideffects, a brief challenge with methandrostenolone-BSAwas able to promote neuronal toxicity. Activation of pu-tative membrane steroid receptors by nontoxic (nano-molar) concentrations of either nandrolone-BSA ormethandrostenolone-BSA became sufficient to increaseneuronal susceptibility to the apoptotic stimulus pro-vided by b-amyloid (the main culprit of AD). We specu-

late that AAS abuse might facilitate the onset or pro-gression of neurodegenerative diseases not usuallylinked to drug abuse. VVC 2011 Wiley-Liss, Inc.

Key words: testosterone; nandrolone; methandro-stenolone; beta-amyloid; neuronal death

Anabolic-androgenic steroids (AASs) are syntheticderivatives of testosterone that are abused in the worldof sport to build muscle and boost athletic performance(Yesalis and Bahrke, 1995). Abuse of AASs causes seriousside effects that involve the cardiovascular system, theliver, and the reproductive system (van Amsterdamet al., 2010). A major concern regards their neurobeha-vioral actions, which are associated with stroke, mooddisturbances, and psychotic symptoms (Uzych, 1992;Hall et al., 2005; Santamarina et al., 2008).

At present, neurobiological mechanisms and sites ofaction of AASs are unclear. In vitro, low concentrationsof AASs amplify excitotoxic neuronal death (Orlandoet al., 2007). In male normal volunteers, high doses ofAASs induce cognitive impairment (Su et al., 1993; Dalyet al., 2003). The main endogenous androgenic steroid,testosterone, has both neuroprotective effects (Hammond

Contract grant sponsor: Italian Ministry of Health; Contract grant num-

ber: 2006-08 (to A.C.).

*Correspondence to: Agata Copani, MD, PhD, Department of Pharma-

ceutical Sciences, University of Catania, Viale Andrea Doria 6, 95125,

Catania, Italy. E-mail: [email protected]

Received 17 September 2010; Revised 22 October 2010; Accepted 8

November 2010

Published online 2 February 2011 in Wiley Online Library

(wileyonlinelibrary.com). DOI: 10.1002/jnr.22578

Journal of Neuroscience Research 89:592–600 (2011)

' 2011 Wiley-Liss, Inc.

Page 3: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

et al., 2001; Pike, 2001; Nguyen et al., 2005; Pike et al.,2008) and neurotoxic effects (Estrada et al., 2006; Cun-ningham et al., 2009), depending on the experimentalparadigm. Endogenous testosterone appears to exhacer-bate some types of neuronal injury, including ischemia-reperfusion injury (Yang et al., 2002) and methamphet-amine-induced neurodegeneration of nigrostriatal dopa-minergic neurons (Dluzen and McDermott, 2006). Inculture paradigms, a few studies indicate that testosteronecan be directly neurotoxic both at supraphysiological(Estrada et al., 2006) and at physiological (Cunninghamet al., 2009) concentrations. Despite evidence that testos-terone may be neurotoxic, there are conditions in whichtestosterone supports neuronal viability (Pike et al.,2009). In particular, it has been demonstrated thatandrogens selectively protect neurons against apoptosis-inducing insults (Nguyen et al., 2010), including b-amy-loid (Ab) protein, the main contributor to Alzhemer’sdisease (AD) pathology.

Androgens mediate their classical genomic effectsthrough binding to the androgen receptor (AR), a mem-ber of the nuclear receptor superfamily that functions asa ligand-activated transcription factor (Heinlein andChang, 2002). A CAG repeat expansion within the firstexon of the AR gene is responsible for testosterone-de-pendent nuclear accumulation of ARs, with ensuingmotor neuron degeneration in spinal and bulbar muscu-lar atrophy (Katsuno et al., 2010). Androgens also medi-ate rapid nongenomic effects via the activation of signal-ing pathways (i.e., the MAPK pathway and the AKTpathway) triggered either by classical ARs (Heinlein andChang, 2002; Nguyen et al., 2005; Foradori et al., 2008)or by membrane-associated ARs (Gatson et al., 2006).How the different AR activities are related to the exist-ing discrepancies regarding whether androgens are pro-tective or damage promoting is unclear.

Given that AAS abuse poses a significant publichealth problem, we investigated the potential neurotoxiceffects of suprapharmacological concentrations of AASs,taking into consideration the existence of both classicaland membrane-associated ARs and focusing on twolargely abused steroids with different pharmacologicalprofiles, nandrolone and methandrostenolone. Nandro-lone binds to ARs to a greater degree than testosterone,whereas methadrostenolone is a weak agonist of theARs (for review see Fragkaki et al., 2009). Hence, theirneurotoxic properties were compared with those of tes-tosterone, and the attempt to differentiate between intra-cellular and membrane functions of ARs was carried outby using cell-impermeable steroid analogues that prefer-entially bind to membrane-associated receptors.

MATERIALS AND METHODS

All animal experimental procedures were carried out inaccordance with the directives of the Italian and EU regula-tions for care and use of experimental animals and wereapproved by the Institutional Animal Care and Use Commit-tee of the University of Catania.

Drugs

Testosterone, testosterone-BSA conjugate, nandrolone,methandrostenolone, RU-486, formestane, and flutamidewere purchased from Sigma-Aldrich (Milan, Italy). Nandro-lone- and methandrostenolone-3-(O-carboxymethyloxime)-BSA conjugates were synthetized in two steps involving ste-roid-(O-carboxymethyl)-oxime derivatives, obtained in turnfrom the two steroids with hemihydrate carboxymethylaminein pyridine at room temperature. Derivatives were covalentlylinked to BSA by a modified mixed anhydride method(Erlanger et al., 1957), using ethyl chloroformate and triethyl-amine in THF. Derivatives were then purified and character-ized by IR, 1H-NMR and mass spectra.

Testosterone, nandrolone, methandrostenolone, fluta-mide, and formestane were dissolved in DMSO at the initialconcentration of 10 mM. The final concentration of DMSOapplied to the cultures was 0.1%. AAS-BSA conjugates weredissolved in 50 mM Tris-HCl (pH 8.5), at 0.2–1 mg/ml.

Pure Cultures of Cortical Neurons

Cultures of pure cortical neurons were obtained fromrats at embryonic day 15 (Morini, s.p.a., Reggio Emilia, Italy).Briefly, cortices were dissected in Ca21/Mg21-free buffer andmechanically dissociated. Cortical cells were plated at a densityof 400 3 103/well on 24-well plates precoated with 0.1 mg/ml poly-D-lysine in DMEM/Ham’s F12 (1:1) medium sup-plemented with the following components: 10 mg/ml bovineserum albumin, 10 lg/ml insulin, 100 lg/ml transferrin, 100lM putrescine, 30 nM selenium, 2 mM glutamine, 6 mg/mlglucose, 100 U/ml penicillin, and 100 lg/ml streptomycin.Cytosine-1b-D-arabinofuranoside (10 lM) was added to thecultures 18 hr after plating to avoid the proliferation of non-neuronal elements and was kept for 3 days before mediumreplacement. This method yields >99% pure neuronal cultures(Copani et al., 1999).

Mixed Cultures of Cortical Cells

Cells dissected from the cortices of rat embryos and dis-sociated as described above were grown into MEM supple-mented with horse serum (10%), fetal calf serum (FCS; 10%),glutamine (2 mM), and glucose (6 mg/ml). Cultures werekept at 378C in a humified 5% CO2 atmosphere. After 3–5days in vitro, non-neuronal cell division was halted by 3 daysof exposure to cytosine-1b-D-arabinofuranoside (10 lM), andcultures were shifted to a maintenance medium identical toplating medium but lacking fetal serum. Subsequent partialmedium replacement was carried out twice per week. Onlymature cultures (12–14 days in vitro) were used for theexperiments. Mature cultures cointained about 40% neurons.

Cultures of Cortical Astrocytes

Cortical glial cells were prepared from 1–3-day-oldSprague-Dowley rats (Morini, s.p.a.). After removal ofmeninges and isolation of cortices, cells were dispersed by me-chanical and enzymatic dissociation using a 0.25% solution oftrypsin. Cells were plated onto 75-mm2 flasks and maintainedin DMEM culture medium, supplemented with 10% FCS,penicillin/streptomycin (100 U/ml to 100 lg/ml), and gluta-

Neurotoxic Effects of AASs 593

Journal of Neuroscience Research

Page 4: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

mine (2 mM). All medium constituents were from Invitrogen(Carlsbad, CA), and all plastic materials were from CorningLife Sciences (Acton, MA). Confluent cultures at 8–10 days invitro were shaked overnight at 378C to remove microglia andoligodendrocytes. Astrocytes were collected by trypsin diges-tion, seeded onto 24 multiwell plates and used 6–8 days afterreplating.

Handling of Ab

Different lots of Ab(25–35) were tested, and the samebatch was used throughout the entire study to rely on a con-sistent profile of toxicity. Peptides were solubilized in sterile,double-distilled water at an initial concentration of 2.5 mMand stored frozen at –208C. Ab(25–35) was used at a finalconcentration of 25 lM in the presence of the glutamate re-ceptor antagonists MK-801 (1 lM) and DNQX (30 lM) toavoid the potentiation of endogenous glutamate toxicity.

Assessment of Neuronal Injury

In mixed cortical cultures, neuronal injury was estimatedby examination of the cultures by phase-contrast microscopy24 hr after the incubation with Ab. Neuronal damage wasquantitatively assessed in all experiments by estimation of deadneurons by trypan blue staining. Stained neurons werecounted from three random microscopic fields per well. Inpure neuronal cultures, neuronal injury was assessed by the 3-[4,5-dimethylthioazol-2-yl]-2,5-diphenyl tetrazolium bromide(MTT) assay. Cultures were incubated with MTT (0.9 mg/mlfinal concentration) for 2 hr at 378C. A solubilization solutioncontaining 20% sodium dodecyl sulfate was then added for anadditional 1 hr, and formazan production was evaluated in aplate reader (absorbance 5 560 nm).

Western Blot Analysis

Cells were harvested in lysis buffer containing a cocktailof protease inhibitors (P2714; Sigma-Aldrich). After sonica-tion, an aliquot of the sample was processed for protein con-centrations according to the method of Bradford. Sampleswere concentrated and boiled for 5 min. Proteins were sepa-rated electrophoretically on polyacrylamide gel (30 mA/hr)using 60–80 lg of cell proteins per lane. Proteins were trans-ferred to nitrocellulose membranes (Hybond ECL; AmershamBiosciences Europe GmbH, Milan, Italy) at room temperatureusing a transblot semidry transfer cell. After blocking, themembranes were incubated with rabbit antiandrogen receptorantibody (1:500; Ab3509; Abcam, Cambridge, United King-dom) overnight at 48C. Membranes were then thoroughlywashed and incubated with HRP-conjugated secondary anti-bodies. Specific bands were visualized using the SuperSignalchemiluminescent detection system (Pierce Biotechnology,Rockford, IL).

Immunofluorescence Labelling and ConfocalMicroscopy

Mixed cortical cultures were plated onto 35-mm disheswith glass slides (Amniodish Euroclone, Milan, Italy) pre-coated with 0.1 mg/ml poly-D-lysine. Cells were fixed for 20min in paraformaldehyde and permeabilized with 0.1% Triton

X-100 for 5 min. Non-specific binding was blocked by incu-bation with PBS containing 3% BSA at room temperature for1 hr. After blocking, rabbit anti-androgen receptor antibody(1:100 dilution) was added overnight at 48C. After washing, aTexas red-conjugated donkey anti-rabbit antibody was usedfor 1 hr (1:400 dilution; Santa Cruz Biotechnology). Sampleswere analyzed through a 360 objective on a confocal laserscanning microscope Olympus FV1000.

RESULTS

To begin investigating the hypothesis that supra-pharmacologic doses of AASs may be neurotoxic, wehave compared the effects of two synthetic AASs (nan-drolone and methandrostenolone) and their BSA conju-gates (nandrolone-BSA and methandrostenolone-BSA)with those of testosterone, using as a model rat corticalcultures, both pure neuronal and mixed neuron–gliatypes. Consistently with in vivo evidence (DonCarloset al., 2006; Sarkey et al., 2008), ARs were expressed bycultured astrocytes and neurons (Fig. 1A), and AR im-munoreactivity was observed by confocal microscopy inneuronal (Fig. 1B) and glial (Fig. 1C) processes. Initially,pure neuronal cultures were exposed to different con-centrations (0.01–10 lM) of either testosterone or nan-drolone, and toxicity was assessed by MTT assay 48 hrlater. Testosterone and nandrolone exerted significanttoxicity only at concencentrations of 10 lM (Fig. 2A).Instead, the cell-impermeable derivatives, testosterone-BSA and nandrolone-BSA, were toxic at 1 lM (Fig.2B). Unlike testosterone and nandrolone, methandroste-nolone had a toxic effect at 1 lM, whereas its derivative,methandrostenolone-BSA, was toxic at concentrations aslow as 100 nM (Fig. 2C). Toxicity by testosterone, nan-drolone, or their BSA conjugates was abrogated by theAR antagonist flutamide (Fig. 3). Instead, toxicity bymethandrostenolone or methandrostenolone-BSA wasinsensitive to flutamide (not shown), but was preventedentirely by the glucocorticoid receptor antagonist RU-486 (Fig. 3). Concentrations of nandrolone, methandros-tenolone, or AAS conjugates, which were neurotoxicfollowing a 48-hr exposure, were also applied to the cul-tures in a 20-min pulse and then washed out 24 hr priorto the assessment of neuronal death. Under this experi-mental condition, only methandrostenolone-BSA wastoxic, and its toxicity was again prevented by RU486(Fig. 4). Finally, testosterone, nandrolone, methandroste-nolone, or their BSA-conjugates were tested for theirability to increase neuronal vulnerability to the apoptoticinsult provided by Ab protein (Loo et al., 1993).Because Ab is able to potentiate glutamate toxicity (Kohet al., 1990; Copani et al., 1991), and AASs mightamplify glutamate toxicity (Orlando et al., 2007), experi-ments were carried out in the presence of a cocktail ofionotropic glutamate receptor antagonists (1 lM MK-801 1 30 lM DNQX) to exclude the contribution ofendogenous excitotoxicity to the overall process of neu-ronal death. Ab(25–35) was used at concentrations(25 lM) able to induce a 45% of neuronal death over a

594 Caraci et al.

Journal of Neuroscience Research

Page 5: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

24-hr period. Prior to the treatment with Ab(25–35),pure neuronal cultures were exposed for 24 hr to thehighest concentrations of the drugs devoid of intrinsictoxicity. With this particular experimental protocol,methandrostenolone-BSA amplified Ab-induced toxicityat concentrations of 10 nM (Fig. 5).

Fig. 2. Neuronal death induced in pure neuronal cultures by increas-ing concentrations of testosterone and nandrolone (A), their BSA-conjugated analogues (B), and either methandrostenolone or methan-drostenolone-BSA (C). All drugs were applied for 48 hr. Neuronaldeath is represented as the percentage reduction of neuronal survivalmeasured by MTT assay. Values are means 6 SEM of six to ninedeterminations. *P < 0.05 vs. controls (one-way ANOVA 1 Fisher’sPLSD).

Fig. 1. AR expression in cultured neural cells. A: Western blot anal-ysis of AR expression in pure cultures of rat cortical neurons or incultured astrocytes. Samples have been loaded in duplicate. A singleband of about 110 kD, corresponding to the molecular weight ofARs, was observed in both cases. Confocal images with 360 magni-fication of AR immunoreactivity in neurons (B) or astrocytes (C)from mixed cortical cultures. [Color figure can be viewed in theonline issue, which is available at wileyonlinelibrary.com.]

Neurotoxic Effects of AASs 595

Journal of Neuroscience Research

Page 6: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

The set of experiments described above wasrepeated in mixed rat cortical cultures, which include aglial component. In mixed cultures, neuronal death wasassessed by Trypan blue staining. In the presence of glialcells, testosterone and nandrolone exerted significanttoxicity already at 1 lM (Fig. 6A), and nandrolone-BSAbecame toxic at a concentration as low as 100 nM (Fig.6B). Instead, the presence of glia cells did not appear toaffect the toxicity profile of either methandrostenoloneor methandrostenolone-BSA (Fig. 6C). Similarly to whatwas observed in cultures of pure neurons, flutamideabrogated toxicity by testosterone, nandrolone, and theirBSA conjugates (Fig. 7), whereas RU-486 preventedentirely the toxicity of methandrostenolone or methan-drostenolone-BSA (Fig. 7). When tested in the brief-insult paradigm (i.e., 24 hr after a 20-min pulse), meth-androstenolone-BSA was toxic once more, and itstoxicity was prevented by RU486 (Fig. 8).

Finally, all drugs were tested for their ability toaffect neuronal susceptibility to Ab-induced toxicity inmixed cortical cultures. For hippocampal cultures, testos-terone has been reported to be neuroprotective againstAb(25–35) at 10 nM concentrations (Pike, 2001). Atthese concentrations, testosterone protected mixed corti-cal cultures against Ab toxicity (Fig. 9), and protection

was abolished by the combined addition of flutamideand the aromatase inhibitor formestane (Fig. 9). How-ever, at higher concentrations (100 nM), which werenot toxic per se, testosterone did not influence Ab tox-icity (Fig. 10). At the highest concentrations devoid ofintrinsic toxicity, nandrolone-BSA, methandrostenolone,and methandrostenolone-BSA all amplified Ab-inducedtoxicity in mixed cortical cultures (Fig. 10).

DISCUSSION

We have demonstrated that suprapharmacologicdoses of two different AASs, namely, nandrolone andmethandrostenolone, are toxic for cultured cortical neu-

Fig. 3. Neuronal death induced in pure neuronal cultures by a 48-hrexposure to testosterone, nandrolone, methandrostenolone, or theirBSA-conjugated analogues alone or combined with either 10 lMflutamide or RU486. Neuronal death is represented as the percentagereduction of neuronal survival measured by MTT assay. Values aremeans 6 SEM of six determinations. *P < 0.05 vs. controls and #P< 0.05 vs. the values obtained in the absence of flutamide or RU486(one-way ANOVA 1 Fisher’s PLSD).

Fig. 4. Neuronal death by nandrolone and methandrostenolone (A)and their BSA-conjugated analogues (B) applied for 20 min to pureneuronal cultures. In B, RU486 (10 lM) was coadded with methan-drostenolone-BSA during the 20-min pulse. Neuronal death is repre-sented as the percentage reduction of neuronal survival measured byMTT assay. Values are means 6 SEM of four determinations. *P <0.05 vs. control and #P < 0.05 vs. methandrostenolone-BSA alone(one-way ANOVA 1 Fisher’s PLSD).

596 Caraci et al.

Journal of Neuroscience Research

Page 7: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

rons. However, the neurotoxic properties of nandroloneand methandrostenolone diverge.

Similarly to the case with testosterone, micromolarconcentrations of nandrolone were detrimental to corticalneurons. The actions of testosterone and nandrolone wereprevented by pharmacological blockade of ARs with flu-tamide, suggesting that toxicity was dependent on ARs.The evidence that the neurotoxic effects of testosteroneand nandrolone required a 48-hr exposure suggests thathigh concentrations of androgens may affect neuronal via-bility by acting through AR-mediated genomic mecha-nisms. Both testosterone and nandrolone were morepotent in cortical cultures containing glia cells, suggestingthat AR-expressing glia cells could be implicated in theregulation of neuronal survival. The role of AR activationin glia cells in currently unclear. Based on the evidencethat androgens can promote the induction of nuclear fac-tor-jB-dependent proinflammatory genes, which lead tobrain inflammation (Gonzales et al., 2009), it cannot beexcluded that glial response factors synergize with andro-gens in reducing neuronal viability.

Confocal studies using an antibody against the clas-sical nuclear AR revealed the localization of ARs toextranuclear sites, including neuronal axons and gliaprocesses. AR immunoreactivity also profiled plasmamembranes, indicating the presence of putative mem-brane receptors both in neurons and in astrocytes. Evi-dence exists that membrane-associated ARs may beeither related or unrelated to the classical nuclear recep-tors (for review see DonCarlos et al., 2006). We found

that testosterone-BSA and nandrolone-BSA, which pref-erentially target membrane ARs, exerted neurotoxiceffects both in pure neuronal and in mixed cultures.These effects were prevented by flutamide, thus support-

Fig. 5. Modulation of Ab(25–35)-induced toxicity in pure neuronalcultures by a 24-hr preexposure to testosterone, nandrolone, methan-drostenolone, or their BSA-conjugated analogues. Ab(25–35) wasapplied for 24 hr, and the amount of neuronal death induced by 25lM Ab was set as 100%. Values are means 6 SEM of six determina-tions. *P < 0.05 vs. Ab alone (one-way ANOVA 1 Fisher’s PLSD).

Fig. 6. Neuronal death induced in mixed cortical cultures by increas-ing concentrations of testosterone and nandrolone (A), their BSA-conjugated analogues (B), and either methandrostenolone or methan-drostenolone-BSA (C). All drugs were applied for 48 hr. Neuronaldeath was assessed by trypan blue staining. After trypan blue staining,dead neurons were counted in three random microscopic fields/well.Values are means 6 SEM of six to nine determinations. *P < 0.05vs. controls (one-way ANOVA 1 Fisher’s PLSD).

Neurotoxic Effects of AASs 597

Journal of Neuroscience Research

Page 8: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

ing the notion that membrane and intracellular ARsmight share a similar pharmacological profile.

It has been proposed that activation of ARs mayelicit opposite effects on cell survival (i.e., detrimental orbeneficial) depending on whether membrane ARs or in-tracellular ARs are activated (Gatson et al., 2006; Gatsonand Singh, 2007). In pure neuronal cultures, both testos-terone-BSA and nandrolone-BSA were toxic, with agreater potency than their parent compounds. A similareffect was also observed following the exposure of mixedcultures to nandrolone-BSA, indicating that AAS-relatedtoxicity depends on the preferential activation of puta-tive membrane ARs over intracellular ARs. Interestingly,activation of putative membrane ARs by low nanomolarconcentrations (10 nM) of nandrolone-BSA did not itselflead to neuronal death, but was sufficient to increaseneuronal susceptibility to the apoptotic stimulus providedby Ab(25–35).

Consistently with the hypothesis that androgensmay activate two competing pathways for the regulationof neuronal survival, physiological concentrations (10nM) of testosterone were instead protective againstAb(25–35)-induced apoptosis in mixed cultures. Theprotective effect of 10 nM testosterone was fully blockedby combining flutamide with the aromatase inhibitorformestane. We speculate that, under physiological con-ditions (i.e., low concentrations of the hormone in a

neuron–glia interplay), testosterone might undergo aro-matization to the neuroprotective molecule 17b-estradiol(Garcia-Segura et al., 2003). After that, the residual con-centrations of testosterone might promote neuronal sur-vival if intracellular ARs are abundant with respect tomembrane-associated ARs.

A difference between methandrostostenolone andeither nandrolone or testosterone could be observedwhen the drug was first tested for its intrinsic toxicity.The action of methandrostenolone was not blocked byflutamide, suggesting that toxicity was independent ofAR activation. Consistently with the notions that meth-androstenolone is only a weak agonist of the ARs andthat most of its anabolic activity likely comes from non-AR-mediated effects (Fragkaki et al., 2009), the neuro-toxicity of the drug was fully prevented by the glucocor-ticoid receptor (GR) antagonist RU486. RU486 alsohas a remarkable antiprogesterone activity (Baulieu,1989); however, unlike other AASs (McRobb et al.,

Fig. 7. Neuronal death induced in mixed cortical cultures by a 48-hrexposure to testosterone, nandrolone, methandrostenolone, or theirBSA-conjugated analogues alone or combined with either 10 lMflutamide or RU486. After trypan blue staining, dead neurons werecounted in three random microscopic fields/well. Values are means6 SEM of six determinations. *P < 0.05 vs. controls and #P < 0.05vs. the values obtained in the absence of flutamide or RU486 (one-way ANOVA 1 Fisher’s PLSD).

Fig. 8. Neuronal death by nandrolone and methandrostenolone (A)and their BSA-conjugated analogues (B) applied for 20 min to mixedcortical cultures. In B, RU486 (10 lM) was coadded with methan-drostenolone-BSA during the 20-min pulse. Neuronal death wasassessed by trypan blue staining. Values are means 6 SEM of fourdeterminations. *P < 0.05 vs. control and #P < 0.05 vs. methan-drostenolone-BSA alone (one-way ANOVA 1 Fisher’s PLSD).

598 Caraci et al.

Journal of Neuroscience Research

Page 9: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

2008; Fragkaki et al., 2009), methandrostenolone lackssignificant progestational properties (Wynn and Landon,1961; Fragkaki et al., 2009), suggesting that progesteronereceptors (PRs) were not involved in the neurotoxicaction of the drug. Notably, nandrolone is among theAASs endowed with progesterone-like actions (McRobbet al., 2008; Fragkaki et al., 2009); the evidence that flu-tamide prevented entirely the toxicity of nandroloneprovides hints that PRs might not be relevant to theneurotoxic properties of AASs.

Activation of GRs has been shown to exacerbate avariety of neuronal insults, including excitotoxicity andAb toxicity (Goodman et al., 1996). In mixed cultures,methandrostenolone was able to exacerbate Ab(25–35)-induced toxicity at concentrations that were not toxicper se (100 nM), and exhibited intrinsic toxicity at 1lM concentrations independently of the presence of gliacells. Methandrostenolone-BSA, which likely bindsmembrane-associated GRs, was always more potent thanthe parent compound, suggesting that its toxicity reliedon the preferential activation of putative membrane GRsover intracellular GRs. Recently, the activation of puta-tive membrane-associated GRs has been shown to medi-ate rapid, nongenomic, effects able to potentiateNMDA-evoked toxicity in hippocampal neurons (Xiaoet al., 2010). Consistently with this evidence, a briefchallenge with methandrostenolone-BSA, but not withnandrolone-BSA, was sufficient to promote neuronaltoxicity both in pure and in mixed neuronal cultures.The identity of membrane-bound GRs has not been

elucidated yet, and both an unknown receptor type andclassical intracellular GRs that associate with the mem-brane have been proposed (for review see Tasker et al.,2006). The ‘‘fast’’ toxicity induced by methandrosteno-lone-BSA was prevented by RU486, suggesting that thedrug was acting at classical intracellular receptors associ-ated with the plasma membrane.

Overall, we have provided evidence that two AASswith a different pharmacological profile, namely, nan-drolone and methandrostenolone, can affect neuronalsurvival at suprapharmacologic doses, raising a seriousconcern for steroid abusers, who have micromolar con-centrations of AASs in their brain (Lukas, 1996; Wu,1997; Daly et al., 2001). The relevant sites for the neu-rotoxic action of nandrolone and methandrostenoloneappear to be membrane-associated ARs and membrane-associated-GRs, respectively. Noteworthy, concentra-tions of the drugs that were not directly neurotoxicwere, however, able to increase neuronal susceptibilityto the apoptotic stimulus provided by Ab(25–35).Hence, in vivo, exposure to AASs may result in a com-promised brain, more susceptible, later in life, to theonset or progression of diseases not usually linked todrug abuse, especially neurodegenerative diseases (e.g.,Alzheimer’s disease).

REFERENCES

Baulieu EE. 1989. Contragestion and other clinical applications of RU

486, an antiprogesterone at the receptor. Science 245:1351–1357.

Fig. 9. Low concentrations of testosterone protected against Ab(25–35)-induced toxicity in mixed cortical cultures. Testosterone wasapplied 24 hr before the addition of 25 lM Ab(25–35) alone or to-gether with flutamide (10 lM), formestane (1 lM), or a combinationof flutamide 1 formestane. Ab(25–35) was applied for 24 hr, and theamount of neuronal death induced by 25 lM Ab was set as 100%.Values are means 6 SEM of four determinations. *P < 0.05 vs. Abalone and #P < 0.05 vs. Ab 1 testosterone (one-way ANOVA 1Fisher’s PLSD).

Fig. 10. Modulation of Ab(25–35)-induced toxicity in mixed corticalcultures by a 24-hr preexposure to testosterone, nandrolone, methan-drostenolone or their BSA-conjugated analogues. Ab(25–35) wasapplied for 24 hr, and the amount of neuronal death induced by 25 lMAb was set as 100%. Values are means6 SEM of six to nine determina-tions. *P < 0.05 vs. Ab alone (one-way ANOVA1 Fisher’s PLSD).

Neurotoxic Effects of AASs 599

Journal of Neuroscience Research

Page 10: Neurotoxic properties of the anabolic androgenic steroids ... · Neurotoxic Properties of the Anabolic Androgenic Steroids Nandrolone and Methandrostenolone in Primary Neuronal Cultures

Copani A, Koh JY, Cotman CW. 1991. Beta-amyloid increases neuronal

susceptibility to injury by glucose deprivation. Neuroreport 2:763–765.

Copani A, Condorelli F, Caruso A, Vancheri C, Sala A, Giuffrida Stella

AM, Canonico PL, Nicoletti F, Sortino MA. 1999. Mitotic signaling

by beta-amyloid causes neuronal death. FASEB J 13:2225–2234.

Cunningham RL, Giuffrida A, Roberts JL. 2009. Androgens induce do-

paminergic neurotoxicity via caspase-3-dependent activation of protein

kinase Cdelta. Endocrinology 150:5539–5548.

Daly RC, Su TP, Schmidt PJ, Pickar D, Murphy DL, Rubinow DR.

2001. Cerebrospinal fluid and behavioral changes after methyltestoster-

one administration: preliminary findings. Arch Gen Psychiatry 58:172–

177.

Daly RC, Su TP, Schmidt PJ, Pagliaro M, Pickar D, Rubinow DR.

2003. Neuroendocrine and behavioral effects of high-dose anabolic ste-

roid administration in male normal volunteers. Psychoneuroendocrinol-

ogy 28:317–331.

Dluzen DE, McDermott JL. 2006. Estrogen, testosterone, and metham-

phetamine toxicity. Ann N Y Acad Sci 1074:282–294.

DonCarlos LL, Sarkey S, Lorenz B, Azcoitia I, Garcia-Ovejero D, Hup-

penbauer C, Garcia-Segura LM. 2006. Novel cellular phenotypes and

subcellular sites for androgen action in the forebrain. Neuroscience

138:801–807.

Erlanger BF, Borek F, Beiser SM, Lieberman S. 1957. Steroid-protein

conjugates. I. Preparation and characterization of conjugates of bovine

serum albumin with testosterone and with cortisone. J Biol Chem

228:713–727.

Estrada M, Varshney A, Ehrlich BE. 2006. Elevated testosterone induces

apoptosis in neuronal cells. J Biol Chem 281:24501–25492.

Foradori CD, Weiser MJ, Handa RJ. 2008. Non-genomic actions of

androgens. Front Neuroendocrinol 29:169–181.

Fragkaki AG, Angelis YS, Koupparis M, Tsantili-Kakoulidou A, Kokotos

G, Georgakopoulos C. 2009. Structural characteristics of anabolic

androgenic steroids contributing to binding to the androgen receptor

and to their anabolic and androgenic activities. Applied modifications in

the steroidal structure. Steroids 74:172–197.

Garcia-Segura LM, Veiga S, Sierra A, Melcangi RC, Azcoitia I. 2003.

Aromatase: a neuroprotective enzyme. Prog Neurobiol 71:31–41.

Gatson JW, Singh M. 2007. Activation of a membrane-associated andro-

gen receptor promotes cell death in primary cortical astrocytes. Endo-

crinology 148:2458–2464.

Gatson JW, Kaur P, Singh M. 2006. Dihydrotestosterone differentially

modulates the mitogen-activated protein kinase and the phosphoinosi-

tide 3-kinase/Akt pathways through the nuclear and novel membrane

androgen receptor in C6 cells. Endocrinology 147:2028–2034.

Gonzales RJ, Duckles SP, Krause DN. 2009. Dihydrotestosterone stimu-

lates cerebrovascular inflammation through NFkappaB, modulating con-

tractile function. J Cereb Blood Flow Metab 29:244–253.

Goodman Y, Bruce AJ, Cheng B, Mattson MP. 1996. Estrogens attenu-

ate and corticosterone exacerbates excitotoxicity, oxidative injury, and

amyloid beta-peptide toxicity in hippocampal neurons. J Neurochem

66:1836–1844.

Hall RC, Hall RC, Chapman MJ. 2005. Psychiatric complications of an-

abolic steroid abuse. Psychosomatics 46:285–290.

Hammond J, Le Q, Goodyer C, Gelfand M, Trifiro M, LeBlanc A.

2001. Testosterone-mediated neuroprotection through the androgen re-

ceptor in human primary neurons. J Neurochem 77:1319–1326.

Heinlein CA, Chang C. 2002. The roles of androgen receptors and

androgen-binding proteins in nongenomic androgen actions. Mol

Endocrinol 16:2181–2187.

Katsuno M, Banno H, Suzuki K, Adachi H, Tanaka F, Sobue G. 2010.

Clinical features and molecular mechanisms of spinal and bulbar muscu-

lar atrophy (SBMA). Adv Exp Med Biol 685:64–74.

Koh JY, Yang LL, Cotman CW. 1990. Beta-amyloid protein increases

the vulnerability of cultured cortical neurons to excitotoxic damage.

Brain Res 533:315–320.

Loo DT, Copani A, Pike CJ, Whittemore ER, Walencewicz AJ, Cot-

man CW. 1993. Apoptosis is induced by beta-amyloid in cultured cen-

tral nervous system neurons. Proc Natl Acad Sci U S A 90:7951–7955.

Lukas SE. 1996. CNS effects and abuse liability of anabolic-androgenic

steroids. Annu Rev Pharmacol Toxicol 36:333–357.

McRobb L, Handelsman DJ, Kazlauskas R, Wilkinson S, McLeod MD,

Heather AK. 2008. Structure–activity relationships of synthetic proges-

tins in a yeast-based in vitro androgen bioassay. J Steroid Biochem Mol

Biol 110:39–47.

Nguyen TV, Yao M, Pike CJ. 2005. Androgens activate mitogen-acti-

vated protein kinase signaling: role in neuroprotection. J Neurochem

94:1639–1651.

Nguyen TV, Jayaraman A, Quaglino A, Pike CJ. 2010. Androgens selec-

tively protect against apoptosis in hippocampal neurones. J Neuroen-

docrinol 22:1013–1022.

Orlando R, Caruso A, Molinaro G, Motolese M, Matrisciano F, Togna

G, Melchiorri D, Nicoletti F, Bruno V. 2007. Nanomolar concentra-

tions of anabolic-androgenic steroids amplify excitotoxic neuronal death

in mixed mouse cortical cultures. Brain Res 1165:21–29.

Pike CJ. 2001. Testosterone attenuates beta-amyloid toxicity in cultured

hippocampal neurons. Brain Res 919:160–165.

Pike CJ, Nguyen TV, Ramsden M, Yao M, Murphy MP, Rosario ER.

2008. Androgen cell signaling pathways involved in neuroprotective

actions. Horm Behav 53:693–705.

Pike CJ, Carroll JC, Rosario ER, Barron AM. 2009. Protective actions

of sex steroid hormones in Alzheimer’s disease. Front Neuroendocrinol

30:239–258.

Santamarina RD, Besocke AG, Romano LM, Ioli PL, Gonorazky SE.

2008. Ischemic stroke related to anabolic abuse. Clin Neuropharmacol

31:80–85.

Sarkey S, Azcoitia I, Garcia-Segura LM, Garcia-Ovejero D, DonCarlos

LL. 2008. Classical androgen receptors in non-classical sites in the brain.

Horm Behav 53:753–764.

Su TP, Pagliaro M, Schmidt PJ, Pickar D, Wolkowitz O, Rubinow DR.

1993. Neuropsychiatric effects of anabolic steroids in male normal vol-

unteers. JAMA 269:2760–2764.

Tasker JG, Di S, Malcher-Lopes R. 2006. Minireview: rapid glucocorti-

coid signaling via membrane-associated receptors. Endocrinology

147:5549–5556.

Uzych L. 1992. Anabolic-androgenic steroids and psychiatric-related

effects: a review. Can J Psychiatry 37:23–28.

van Amsterdam J, Opperhuizen A, Hartgens F. 2010. Adverse health

effects of anabolic-androgenic steroids. Regul Toxicol Pharmacol

57:117–123.

Wu FC. 1997. Endocrine aspects of anabolic steroids. Clin Chem

43:1289–1292.

Wynn V, Landon J. 1961. A study of the androgenic and some related

effects of methandienone. Br Med J 1:998–1003.

Xiao L, Feng C, Chen Y. 2010. Glucocorticoid rapidly enhances

NMDA-evoked neurotoxicity by attenuating the NR2A-containing

NMDA receptor-mediated ERK1/2 activation. Mol Endocrinol

24:497–510.

Yang SH, Perez E, Cutright J, Liu R, He Z, Day AL, Simpkins JW.

2002. Testosterone increases neurotoxicity of glutamate in vitro and is-

chemia-reperfusion injury in an animal model. J Appl Physiol 92:195–

201.

Yesalis CE, Bahrke MS. 1995. Anabolic-androgenic steroids. Current

issues. Sports Med 19:326–340.

600 Caraci et al.

Journal of Neuroscience Research

View publication statsView publication stats