Oestrogens ameliorate mitochondrial dysfunction in Leber's hereditary optic neuropathy

BRAINA JOURNAL OF NEUROLOGY

Oestrogens ameliorate mitochondrial dysfunctionin Leber’s hereditary optic neuropathyCarla Giordano,1,* Monica Montopoli,2,* Elena Perli,1 Maurizia Orlandi,1 Marianna Fantin,3

Fred N. Ross-Cisneros,4 Laura Caparrotta,2 Andrea Martinuzzi,3 Eugenio Ragazzi,2 Anna Ghelli,5

Alfredo A. Sadun,4 Giulia d’Amati1 and Valerio Carelli6

1 Dipartimento di Medicina Sperimentale e Patologia, Sapienza, Universita di Roma, 00161 Rome, Italy

2 Dipartimento di Farmacologia ed Anestesiologia, Universita di Padova, 35131 Padova, Italy

3 IRCCS ‘E. Medea’, 31015 Conegliano, Treviso, Italy

4 Departments of Ophthalmology and Neurosurgery, Keck School of Medicine at University of Southern California, Los Angeles, 90033 CA, USA

5 Dipartimento di Biologia Evoluzionistica Sperimentale, Universita di Bologna, 40123 Bologna, Italy

6 Dipartimento di Scienze Neurologiche, Universita’ di Bologna, 40123 Bologna, Italy

*These authors contributed equally to this work.

Correspondence to: Dr Valerio Carelli,

Dipartimento di Scienze Neurologiche,

Universita di Bologna,

Via Ugo Foscolo 7,

40123 Bologna, Italy

E-mail: [email protected]

Leber’s hereditary optic neuropathy, the most frequent mitochondrial disease due to mitochondrial DNA point mutations in

complex I, is characterized by the selective degeneration of retinal ganglion cells, leading to optic atrophy and loss of central

vision prevalently in young males. The current study investigated the reasons for the higher prevalence of Leber’s hereditary

optic neuropathy in males, exploring the potential compensatory effects of oestrogens on mutant cell metabolism. Control and

Leber’s hereditary optic neuropathy osteosarcoma-derived cybrids (11778/ND4, 3460/ND1 and 14484/ND6) were grown in

glucose or glucose-free, galactose-supplemented medium. After having shown the nuclear and mitochondrial localization of

oestrogen receptors in cybrids, experiments were carried out by adding 100 nM of 17b-oestradiol. In a set of experiments, cells

were pre-incubated with the oestrogen receptor antagonist ICI 182780. Leber’s hereditary optic neuropathy cybrids in galactose

medium presented overproduction of reactive oxygen species, which led to decrease in mitochondrial membrane potential,

increased apoptotic rate, loss of cell viability and hyper-fragmented mitochondrial morphology compared with control cybrids.

Treatment with 17b-oestradiol significantly rescued these pathological features and led to the activation of the antioxidant

enzyme superoxide dismutase 2. In addition, 17b-oestradiol induced a general activation of mitochondrial biogenesis and a

small although significant improvement in energetic competence. All these effects were oestrogen receptor mediated. Finally, we

showed that the oestrogen receptor b localizes to the mitochondrial network of human retinal ganglion cells. Our results

strongly support a metabolic basis for the unexplained male prevalence in Leber’s hereditary optic neuropathy and hold promises

for a therapeutic use for oestrogen-like molecules.

Keywords: LHON; oestrogen; mitochondrial disorders; oestrogen receptors; oxidative stress

Abbreviations: COIV = cytochrome oxidase c subunit IV; DMEM = Dulbecco’s modified eagle medium; LHON = Leber’s hereditaryoptic neuropathy; PCR = polymerase chain reaction; SOD2 = superoxide dismutase 2

doi:10.1093/brain/awq276 Brain 2010: Page 1 of 15 | 1

Received March 11, 2010. Revised July 27, 2010. Accepted August 9, 2010.

� The Author (2010). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved.

For Permissions, please email: [email protected]

Brain Advance Access published October 13, 2010

at Universit? di B

ologna - Sistem

a Bibliotecario d'A

teneo on October 14, 2010

brain.oxfordjournals.orgD

ownloaded from

IntroductionMore than 20 years ago, the first point mutation in the mitochon-

drial genome was associated with a maternally inherited disease,

Leber’s hereditary optic neuropathy (LHON; Wallace et al., 1988).

LHON is characterized by the selective degeneration of retinal

ganglion cells, in particular those contributing to the papillomacu-

lar bundle, leading to optic atrophy and loss of central vision

(Carelli et al., 2004). LHON is now recognized as the most fre-

quent mitochondrial disorder (Man et al., 2003) and over the last

two decades intense work has been carried out to elucidate its

clinical and molecular basis (Carelli et al., 2004; Man et al., 2009).

However, crucial features of this disease still remain elusive. In

particular, it is difficult to explain the incomplete penetrance

when all individuals in a maternal lineage carry a homoplasmic

mutation (100% of mitochondrial genomes are mutant in each

cell). Furthermore, male prevalence also remains unexplained.

To date, three mitochondrial DNA point mutations at positions

11778/ND4, 3460/ND1 and 14484/ND6 are found in 490% of

patients with LHON (Carelli et al., 2004; Man et al., 2009). There

is also well-established evidence that the non-synonymous popu-

lation polymorphisms found in mitochondrial DNA haplogroups

J1c and J2b increase the penetrance of LHON mutations

11778/ND4 and 14484/ND6, respectively (Carelli et al., 2006;

Hudson et al., 2007b). It is also assumed that further genetic

(Carelli et al., 2003; Phasukkijwatana et al., 2010) and environ-

mental (Sadun et al., 2003; Kirkman et al., 2009) factors play a

role in modulating the variability of penetrance and possibly

gender prevalence.

The possibility that nuclear genes on chromosome X play a role

has repeatedly been pursued with controversial results (Vilkki

et al., 1991; Chalmers et al., 1996). The hypothesis of an

X-linked modifying gene is particularly attractive (two-loci hypoth-

esis) as it would explain both features of variable penetrance and

male prevalence (Bu et al., 1991). Recently, linkage analysis iden-

tified two loci on chromosome X (Hudson et al., 2005; Shankar

et al., 2008), suggesting multiple modifying genes that remain

unidentified.

However, some data suggest a different scenario. Assuming that

an X-linked gene plays a major modifying role in the expression of

LHON in males, in the case of affected females homozygosity or

skewed inactivation favouring the modifying allele should be

assumed (Bu et al., 1991). Comprehensive examination of af-

fected females by independent studies failed to document any

excess of skewed inactivation of the X-chromosome (Hudson

et al., 2007a). Thus, based on the different hormonal metabolism

between genders, it may be that in females, oestrogens play a

protective role in modifying the severity of the mitochondrial

defect (Carelli et al., 2004, 2007), which is characterized by the

combination of defective ATP synthesis driven by complex I sub-

strates (Baracca et al., 2005), increased oxidative stress (Beretta

et al., 2004; Floreani et al., 2005) and enhanced sensitivity to

apoptotic cell death (Ghelli et al., 2003; Zanna et al., 2005).

Oestrogens may modify this scenario, given the increasing

body of evidence of their direct action on the mitochondrial

respiratory chain, oxidative stress and mitochondrial biogenesis

(Simpkins et al., 2008; Chen et al., 2009), and the suggestion

that they are responsible for female longevity (Vina et al.,

2006). Oestrogen receptors are present on mitochondria

(Chen et al., 2004a; Yang et al., 2004) and oestrogen-responsive

elements are localized within the D-loop, a major regulatory

region for human mitochondrial DNA transcription and replication

(Chen et al., 2004b). These observations suggest that oestrogens

may influence mitochondrial functions both by a classic long-term

genomic mechanism and also by rapid non-genomic signalling

involving membrane-associated oestrogen receptors, including

direct effects on mitochondria (Chen et al., 2009).

The current study uses the cybrid cell model to investigate

whether oestrogens modify the mitochondrial dysfunction in

LHON and provides new insights on the gender bias of this

still-elusive disease and possible new therapeutic approaches.

Materials and methods

Cell lines and reagentsControl and LHON cybrids (11778/ND4, 3460/ND1 and 14484/ND6)

were grown in Dulbecco’s modified eagle medium (DMEM) supple-

mented with 10% dialysed foetal bovine serum, 2mM L-glutamine,

100U/ml penicillin, 100 mg/ml streptomycin and 0.1mg/ml bromo-

deoxyuridine (referred to as DMEM-glucose). Experiments were

performed both in DMEM-glucose and in glucose-free DMEM,

supplemented with 10% dialysed foetal bovine serum, 5mM galactose

and 110mg/ml sodium pyruvate (referred to as DMEM-galactose).

17b-Oestradiol was purchased from Sigma-Aldrich (St Louis, MO,

USA) and the oestrogen receptor antagonist ICI 182780 from Tocris

Bioscience (Bristol, UK). 17b-Oestradiol and ICI 182780 were dissolved

in ethanol at a final concentration of 100 mM and diluted to appro-

priate concentrations in culture medium as required. Untreated

cells were maintained at the same final ethanol concentration. The

oestrogen receptor antagonist ICI 182780 was added 30min before

17b-oestradiol.

Immunostaining and western blotanalysisFor immunocytochemistry, cells grown on coverslips were fixed with

4% formaldehyde freshly prepared from paraformaldehyde in

phosphate-buffered saline (pH 7.4) with 0.1% Triton X-100. Primary

antibodies were visualized using secondary fluorescein isothiocyanate-

and Cy3-conjugated antibodies (Jackson Laboratories, Bar Harbor,

Maine, USA). Immunofluorescence was performed on formalin-fixed,

paraffin-embedded human retinal sections obtained from one patient

with LHON (male, aged 52 years) and two control individuals (1 male

and 1 female, aged 59 and 70 years, respectively; obtained from the

Lions Eye Bank of Oregon, Portland, OR, USA). For western blot ana-

lysis, cells were rinsed twice with ice-cold phosphate-buffered saline,

lysed in ice-cold radioimmunoprecipitation assay buffer (50mM Tris–

HCl pH 8, 150mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 1%

sodium dodecyl sulphate, 1mM phenylmethanesulphonylfluoride,

10mg/ml aprotinin, 10mg/ml leupeptin and 10mg/ml pepstatin)

and centrifuged at 10 000g for 10min at 4�C. For some experiments

mitochondria were isolated from 4�106 cybrid cells by standard dif-

ferential centrifugation. Protein concentration was measured by

bicinchoninic acid (Beyotime Biotechnology, Haimen, China). Equal

2 | Brain 2010: Page 2 of 15 C. Giordano et al.

at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

amounts of protein (50mg) were separated by 12% sodium dodecyl

sulphate polyacrylamide gel electrophoresis and transferred onto a

nitrocellulose membrane (Millipore, Bedford, MO). Primary antibodies

were visualized using horseradish peroxidase-conjugated secondary

antibodies (Dako, Glostrup, Denmark). Signals were detected by

enhanced chemiluminescence (Amersham Biosciences, UK).

The following primary antibodies were used: rabbit polyclonal

antibody raised against amino acids 2–185 of human oestrogen

receptor a (Santa Cruz Biotechnology, Santa Cruz, CA, USA);

rabbit polyclonal antibody raised against a synthetic peptide

(CSSTEDSKNKESSQNLQSQ) corresponding to amino acids 485–503

of rat oestrogen receptor b (ERb 485-503; Calbiochem, MerckKGaA,

Darmstadt, Germany); rabbit polyclonal antibody raised against amino

acids 1–150 of human oestrogen receptor b (ERb H-150, Santa Cruz

Biotechnology); rabbit polyclonal antibody anti-superoxide dismutase 2

(Assay Designs; Ann Arbor, MI, USA) mouse monoclonal antibody

anti-cytochrome c oxidase subunit IV (COIV) (Mitosciences, Eugene,

OR, USA) mouse monoclonal antibody anti-porin and mouse mono-

clonal antibody anti-ND6 (Invitrogen, Paisley, UK), mouse monoclonal

antibody anti-b-actin (Sigma, Saint Luise, MO, USA) and mouse

monoclonal antibody anti-mitochondria extract clone MTC (UCS

Diagnostic, Morlupo, Italy).

Intracellular level of reactive oxygenspecies and mitochondrial/transmembrane potential (�j)Direct detection of intracellular steady-state levels of reactive oxygen

species was carried out on living cells by cytofluorimetry using

20,70-dichlorofluorescin-diacetate (H2-DCF-DA; Molecular Probes,

Invitrogen Corp., Carlsbad, CA, USA; Wu et al., 2007).

Mitochondrial membrane potential was measured by cytofluorimetry

using the cationic lipophilic green fluorochrome rhodamine-123

(Rh123; Molecular Probes, Invitrogen Corp., Carlsbad, CA, USA;

Ferlini et al., 2007). Sample fluorescence was analysed by Epics XL

Coulter Systems (Fullerton, CA) equipped with a 488 nm Argon laser.

Dead cells were excluded by electronically gating data on the basis of

forward-versus-side scatter profiles; a minimum of 104 cells/sample

were analysed further. Logarithmic detectors were used for the FL1

fluorescence channel necessary for 20,70-dichlorofluorescin detection.

Mean fluorescence intensity values were obtained by the analysis

EXPO 32 software (Coulter Systems, Fullerton, CA, USA).

Superoxide dismutase 2 activitySuperoxide dismutase (SOD) activity was evaluated as reported

previously (Oberley et al., 1984), with minor modifications. Briefly,

1ml of medium consisting of 50mm KH2PO4 (pH 7.8), 0.1mM

ethylenediaminetetraacetic acid and a superoxide generating system

(0.15mM xanthine plus 0.02U xanthine oxidase) was used together

with 50 mM nitroblue tetrazolium to monitor superoxide formation by

following the changes in colorimetric absorbance at 550 nm for 5min

at 25�C. To assess the specific mitochondrial superoxide dismutase

2 (SOD2) activity, cell fractions were pre-incubated for 60min at

0�C in the presence of 2mM KCN, which induces inhibition of the

cytoplasmic Cu/ZnSOD. The catalytic activities of the samples were

evaluated as their ability to inhibit the rate of nitroblue tetrazolium

reduction. SOD2 activity was expressed as unit per milligram of

protein. One unit of SOD2 is defined as the amount of enzyme

needed to exhibit 50% dismutation of the superoxide radical.

Cell viability and ATP assaysCell proliferation was measured by the trypan blue dye exclusion

assay. Multiple series of 60mm dishes were seeded with a constant

number of cells (106) for 24 and 48 h. Cells were detached by 0.25%

trypsin and 0.2% ethylenediaminetetraacetic acid, washed, suspended

in phosphate-buffered saline in the presence of trypan blue solution

(Sigma, St Louis, MO, USA) at 1:1 ratio and counted using a haema-

tocytometer. Viability was expressed as percentage of untreated cell

number in DMEM-glucose. ATP cellular content was measured by

using the luciferin/luciferase assay (Zanna et al., 2005).

Annexin V/propidium iodide stainingfor apoptotic cellsCells were seeded at 1.5�105 cells/well, incubated overnight, treated

according to experimental protocol and incubated for 24 h. Cells were

harvested by quick trypsinization to minimize potentially high annexin

V background levels in adherent cells, were washed and then stained

with Alexa 488/annexin V/propidium iodide (Molecular Probes,

Invitrogen UK). Cells were analysed on an Epics XL-flow cytometer

using the Analysis software (both hardware and software were from

Beckman Coulter, Miami, FL, USA), with the laser excitation wave-

length set at �=488nm. The green signal from Alexa 488/annexin V

was measured at �=525nm and the red signal from propidium iodide

was measured at �=620 nm. Cells staining negative for both annexin

V and propidium iodide are viable, cells that are annexin V-positive/

propidium iodide-negative are in early apoptosis, whereas cells that are

annexin V-positive/propidium iodide-positive are necrotic or in late

apoptosis (Stadelmann, et al., 2000).

Mitochondrial morphologyCells were stained with MitoTracker Orange (Molecular Probes,

Invitrogen, Carlsbad, CA, USA) for 30min at 37�C, then fixed with

4% paraformaldehyde in phosphate-buffered saline for 15min, coun-

terstained with 4’,6-diamidino-2-phenylindole and mounted on glass

slides by using Mowiol 40-88 (Sigma, St Louis, MO, USA). Images

were acquired with a Nikon C1 confocal microscope and analysed

using Nikon EZ-C1 software (version 2.10; NIKON Corporation,

Tokyo, Japan).

Quantification of mitochondrial DNAby quantitative real-time polymerasechain reactionTotal DNA was isolated by Wizard Genomic DNA Purification Kit

(Promega, Madison, WI, USA). Evaluation of mitochondrial DNA

copy number by quantitative real-time polymerase chain reaction

(PCR) was performed as previously described (Mussini et al., 2005).

Briefly, a mitochondrial DNA fragment (nt 4625–4714) and a nuclear

DNA fragment (FasL gene) were co-amplified by using TaqMan�

probe system and Platinum� Quantitative PCR SuperMix-UDG

(Invitrogen, Life Technologies, Carlsbad, CA, USA). With each assay,

a standard curve for mitochondrial and nuclear DNA was generated

using serial known dilutions of a vector (a kind gift from Dr Andrea

Cossarizza) in which the regions used as template for the two ampli-

fications were cloned tail to tail to have a ratio of 1:1 of the reference

molecules. The absolute mitochondrial DNA copy number per cell was

obtained by the ratio of mitochondrial to nuclear DNA values

LHON and oestrogens Brain 2010: Page 3 of 15 | 3

at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

multiplied by two (as two copies of the nuclear gene are present in a

cell). PCR was carried out in an iCycler Thermal cycler (BioRad,

Hercules, CA, USA) and at least three measurements were obtained

for each sample.

Gene expression by quantitativereal-time polymerase chain reactionTotal RNA was isolated by using SV Total RNA Isolation System

(Promega, Madison, WI, USA) and measured with a NanoDrop

ND-1000 spectrophotometer (NanoDrop Technologies, Inc.

Wilmington, DE, USA). One microgram of total RNA was reverse-

transcribed to complementary DNA using random hexamer primers

(AMV Reverse Transcriptase, Promega, Fitchburg, WI, USA). Relative

expression of each gene was determined by quantitative real-time PCR

using the Platinum SYBR Green qPCR Super Mix-UDG (Invitrogen, Life

Technologies, Paisley, UK). Primers for PGC1-�, PGC1-�, NRF1, NRF2

and Tfam were as previously reported (Sebastiani et al., 2008). Primers

for COI, COIV and ND6 are reported in Supplementary Table 1. They

were carefully designed with Beacon Designer Software (Biorad) and

checked by using Blast software (http://www.ncbi.nlm.nih.gov/

BLAST/) to avoid cross-homology. The specificity of the amplicons

was confirmed by direct sequencing using an ABI Prism 310 Genetic

analyser following standard procedures. Quantitative real-time PCR

was performed in triplicate using 1 ml of complementary DNA template

in a 50-ml reaction. Linearity and efficiency of PCR amplifications were

assessed using standard curves generated by serial dilution of comple-

mentary DNA; melt-curve analysis was used to confirm the specificity

of amplification and absence of primer dimers. In all samples, the

relative expression of each target gene with respect to one control

(reference sample) was evaluated with the comparative threshold

cycle (�Ct) method. All values were normalized to the 18S ribosomal

RNA housekeeping gene.

Oxygen consumptionOxygen consumption was measured in intact cells (5�106) using a

Clark type oxygen electrode, in 1.85ml DMEM lacking glucose

supplemented with 5% dialysed foetal bovine serum at 37�C, as

previously described (Carelli et al., 2002).

Statistical analysisAll data are expressed as mean� SEM. Standard ANOVA procedures

followed by multiple pair-wise comparison adjusted with Bonferroni

corrections were performed for continuous variables, chi-square test

was used for frequency data. Multi-way ANOVA procedures were

performed to analyse SOD2 activity and total cellular ATP content.

Paired t-tests were used to analyse the rate of oxygen consumption.

Significance was considered at P50.05.

Results

Characterization of cybrid cellsWe investigated the effects of 17b-oestradiol on the cellular

phenotype of cybrids obtained from four unrelated patients carry-

ing one of the three classic LHON pathogenic mutations

and compared these with two control cybrids (Supplementary

Table 2). Osteosarcoma-derived (143B.TK-) cybrid cell lines were

previously established and characterized (Ghelli et al., 2003;

Baracca et al., 2005; Floreani et al., 2005; Zanna et al., 2005),

and the complete mitochondrial DNA sequence had been deter-

mined for all the clones used in this study (Pello et al., 2008; Ghelli

et al., 2009). The LHON pathogenic mutations were homoplasmic

in all cell lines. Furthermore, we established by immunofluores-

cence that both oestrogen receptor a and oestrogen receptor bwere localized to 143B.TK-nucleus, whereas immunofluorescence

and western blot analyses showed localization of the oestrogen

receptor b to cybrid-derived mitochondria (Fig. 1A and B). This

characterization for oestrogen receptors was carried out prior to

the cells’ exposure to 17b-oestradiol to verify their constitutive

expression.

17b-Oestradiol decreases levels ofreactive oxygen speciesPrevious results have shown that production of mitochondrial re-

active oxygen species was significantly increased in cybrids bearing

LHON mutations (Beretta et al. 2004; Floreani et al., 2005). We

first confirmed that in glucose medium, 11778/ND4 cybrids pre-

sent an �3-fold increase in levels of reactive oxygen species as

compared with controls (Fig. 2A). After 1 h of incubation in

glucose-free medium containing galactose, a condition forcing

cells to rely predominantly on oxidative phosphorylation for ATP

synthesis, production of reactive oxygen species further increased

by �2-fold in both cell lines. We then added 17b-oestradiol to the

galactose medium at increasing concentrations ranging from 10 to

200 nM, and evaluated levels of reactive oxygen species after 1 h

of incubation. As shown in Fig. 2A, addition of 17b-oestradiolcaused a decrease in levels of reactive oxygen species in both

LHON and control cybrids, in a concentration-dependent

manner. The lowest levels of reactive oxygen species were

reached with 100 nM 17b-oestradiol, and this concentration was

chosen for the further experiments. This experiment was replicated

in all the LHON cybrid cell lines. As shown in Fig. 2A, supplemen-

tation of medium with 17b-oestradiol induced significant decrease

in levels of reactive oxygen species in all cell lines. This phenom-

enon was evident both in cells incubated in glucose and in galact-

ose medium. Finally, we looked as to whether the observed effects

were mediated by oestrogen receptors. We pre-incubated cells

with the oestrogen receptor antagonist ICI 182780, and found

that the antioxidant effect of 17b-oestradiol was fully abolished

(Fig. 2A). As a control experiment, cells were also exposed to ICI

182780 alone without significant effect (data not shown). Similar

results were obtained after 3 h of incubation in galactose medium

(Supplementary Fig. 1). In all of these experiments, the LHON

cybrids carrying the 11778/ND4 mutation displayed the highest

levels of reactive oxygen species compared with the other cell

lines.

17b-Oestradiol induces superoxidedismutase 2 activityManganese SOD2 is the main mitochondrial antioxidant enzyme.

Oestrogens can modulate SOD2 activity both by a rapid,


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

non-genomic action (Pedram et al., 2006, Gottipati and

Cammarata, 2008) and by long-term regulation of gene transcrip-

tion (Borras et al., 2005). We previously showed that LHON cybrids

have a significant reduction of SOD2 activity as compared with

controls, despite an increased amount of protein (Floreani et al.,

2005). To shed light on the mechanism involved in the

17b-oestradiol-dependent decrease in levels of reactive oxygen

species described above, we evaluated SOD2 activity in a

time-course experiment (1, 6, 12 and 24 h) using 11778/ND4

and control cybrids grown in glucose and galactose medium sup-

plemented with 100 nM 17b-oestradiol. We first confirmed that the

SOD2 activity of 11778/ND4 cybrids maintained in glucose

medium was significantly lower than in controls (Fig. 2B).

Supplementation with 17b-oestradiol led to a significant increase

of SOD2 activity both in control and 11778/ND4 cybrids. This in-

crease was more marked in LHON cybrids and remained stable for

24 h. Gene expression of SOD2 was enhanced by 17b-oestradiol inboth control and 11778/ND4 cybrids after 6 h of incubation, and a

significant SOD2 protein increase was observed after 24 h (Fig. 2C).

Incubation in galactose medium induced a significant increase of

SOD2 activity in control cells (Fig. 1B), which was not observed in

LHON cybrids, despite the induction of gene expression and

increased protein (Fig. 2C). Supplementation of galactose

medium with 17b-oestradiol led to a further rapid (1 h)

Figure 1 Oestrogen receptor localization to osteosarcoma-derived (143B. TK-) cybrid cell lines. (A) Fluorescence microscopy localization

of oestrogen receptors to the nucleus and mitochondria of cybrid cells. Oestrogen receptor b is stained red by oestrogen receptor bBIO1974 antibody (a) and oestrogen receptor b H150 antibody (e); oestrogen receptor a is stained red by H-184 antibody (i);

mitochondria are stained green using anti-mitochondria extract antibody (b, f and l); merged image of oestrogen receptors and

mitochondria (c, g andm); negative controls loaded with 4’,6-diamidino-2-phenylindole (blue) (d, h and n). Overlap is seen in bright green

(original magnification �40). (B) Western blot analysis of mitochondrial preparation from 143B.TK-derived wild-type cybrids and MCF7

oestrogen–dependent breast cancer cells used as positive control. The oestrogen receptor b (ERb) protein (identified with oestrogen

receptor b 485–503 antibody) is constitutively over-expressed in the mitochondrial fraction of MCF7 cells, whereas oestrogen receptor a(ERa) is present at lower levels, confirming previous data (Pedram et al., 2006). In wild-type cybrids, oestrogen receptor b is less abundant

than in MCF7 cells, whereas oestrogen receptor a is not expressed.


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

Figure 2 17b-Oestradiol decreases levels of reactive oxygen species and induces SOD2 activity. (A) Left: Control and 11778/ND4 cybrids

were incubated for 1 h in glucose (glu) or galactose (gal) medium� 10–200 nM 17b-oestradiol (E2), then the intracellular steady-state

levels of reactive oxygen species was evaluated. Right: In a further experiment control and LHON cybrids were incubated for 1 h in glucose

(glu) or galactose (gal) medium� 100nM 17b-oestradiol (E2). The oestrogen receptor antagonist ICI 182780 (I) was added 30min before

17b-oestradiol. Untreated cells were maintained at the same final ethanol concentration. Data are expressed as mean fluorescence

intensity (MFI) (� SEM) and are from three experiments. �P50.05; ��P50.01; ��P50.001 LHON versus control; †P50.05 for

glu + E2 versus glu; *P50.05, ***P50.001 for gal versus glu; +P50.05; ++P50.01 for gal + E2 versus gal and versus gal + E2+ I (see

Supplementary Fig. 1). (B) Time course of SOD2 activity of control and 11778/ND4 (HPE9) cybrids incubated in glucose or galactose

medium� 100nM 17b-oestradiol (E2). Incubation in galactose medium induced a significant increase of SOD2 activity in control cybrids

(P50.001) that was not observed in LHON. SOD2 activity is expressed as unit per milligram of protein. Each data point is the mean of

quadruplicate replicates. ��P50.001 for LHON versus control; +++P50.001 for E2 versus ethanol. (C) Left: The relative expression of

mitochondrial SOD2 gene was evaluated by real-time PCR in control (HGA and HP27) and 11778/ND4 (HPE9 and HFF3) cybrids

incubated for 6 h in glucose (glu) or galactose (gal) medium � 100 nM 17b-oestradiol (E2). The oestrogen receptor antagonist ICI 182780

was added 30min before 17b-oestradiol. Data represent mean arbitrary units (� SEM) normalized to control values in glucose medium

and are from three experiments. †P50.05 and ††P50.01 for glu + E2 versus glu; ***P50.001 for gal versus glu; +++P50.001 for gal+E2versus gal. Right: Western blot analysis of SOD2 protein performed on extract from control (HGA) and 11778/ND4 (HPE9) cybrids

incubated for 24 h in glucose or galactose medium� 100nM 17b-oestradiol. A representative blot from three is shown.


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

upregulation of SOD2 activity, both in control and 11778/ND4

cybrids, which remained stable for 24 h, and was not linked with

increased protein levels despite increased gene expression. The

latter increase was fully abolished by the oestrogen antagonist

ICI 182780.

17b-Oestradiol ameliorates cell viabilityand reduces apoptosis in galactosemediumPrevious work by our group showed that the growth of LHON

cybrids in galactose medium suffered a marked decrease in cell

number compared with controls, secondary to apoptotic cell

death (Ghelli et al., 2003; Zanna et al., 2005). As shown in

Fig. 3A, in both LHON and control cybrids, supplementation of

medium with 100 nM 17b-oestradiol significantly reduced the

galactose-dependent loss of viability, as evaluated by the

trypan blue dye assay. As an additional marker of cell viability,

we measured the mitochondrial membrane potential (mt�c).After 24-h incubation in galactose medium, LHON cybrids

displayed a significant decrease (35%) of mt�c compared with

control cybrids. Supplementation with 100 nM 17b-oestradiolprevented the mt�c reduction and this effect was lost by

pre-incubation with the oestrogen receptor antagonist ICI

182780 (Fig. 3B).

The next set of experiments was designed to examine whether

protection of cell viability by 17b-oestradiol was related to reduc-

tion of the apoptotic cell death rate by labelling the cells

with annexin V. Annexin V is an early marker of apoptosis

and binds phosphatidylserine exposed on the cytoplasmic surface

of the cell membrane of apoptotic cells (Stadelmann and

Lassmann, 2000). After incubation in galactose medium for 24 h,

a significant number of LHON cybrids shifted from the viable

to the apoptotic state, thus confirming our previous results

(Ghelli et al., 2003; Zanna et al., 2005). Incubation with

17b-oestradiol prevented the shift towards apoptosis, corroborat-

ing the increase in cell viability reported above (Supplementary

Fig. 2). Figure 3C summarizes the percentages of apoptotic cells

in control and LHON cybrids, and demonstrates the statistically

significant decrease in cell death following treatment with

17b-oestradiol.

17b-Oestradiol reduces mitochondrialnetwork fragmentationThe importance of the mitochondrial network organization in

both mitochondrial respiratory chain defects and apoptosis is

now well recognized (Bernard et al., 2007). We therefore investi-

gated the morphology of the mitochondrial network under

the conditions used in the previous experiments. Control and

11778/ND4 LHON cybrids were stained with Mitotraker

Orange, counterstained with 40,6-diamidino-2-phenylindole and

scored into three categories based on mitochondrial morphology

as previously reported (Zanna et al., 2008). Briefly, Class I cells

showed a typical filamentous network, Class II cells showed

filamentous mitochondria containing balloon-like structures and

Class III cells showed complete fragmentation resulting in only

isolated mitochondrial balloons (Fig. 4A). In the present study,

we added Class IV to take into account the presence of shrunken

cells with dense, fluorescent cytoplasm and fragmented nucleus,

representing end-stage, dying apoptotic cells. In galactose

medium, the percentage of the four classes was significantly dif-

ferent between control and LHON cybrids. In fact, control cybrids

exhibited a combination of Class I (�47%) and Class II (�31%)

cells, with about the same limited amount of Class III and IV cells

(�10%). In contrast, incubation of the 11778/ND4 cybrids in gal-

actose medium for 24 h led to a dramatic increase of Class III

(�20%) and IV (�47%) cells, with parallel reduction of Class I

and II cells. This phenomenon was significantly compensated

by incubation of the 11778/ND4 cybrids with 17b-oestradiol(Fig. 4B). The results of these experiments are compatible and

complementary to those obtained from cell growth rates and

the annexin assay.

17b-Oestradiol induces mitochondrialbiogenesisRecent studies showed oestrogen modulation of mitochondrial

biogenesis by transcriptional regulation of nuclear and mitochon-

drial genes (Mattingly et al., 2008). Thus, we investigated whether

induction of mitochondrial biogenesis and respiration occurs in

LHON cybrids grown in glucose and galactose medium when

treated with 100 nM 17b-oestradiol. First, we evaluated, as a

marker of mitochondrial biogenesis, the mitochondrial DNA con-

tent of cells. Control and LHON cybrids showed a similar increase

in mitochondrial DNA amount after 3–6 h of incubation in galact-

ose medium. A further increase in mitochondrial DNA content was

observed when glucose or galactose medium was supplemented

with 17b-oestradiol (Fig. 5A). This event was evident after

15–30min of incubation with 17b-oestradiol, reached a plateau

(up to 2.5-fold increase) after �3 h and did not change over

the following 72h (Supplementary Fig. 3). The increase in mito-

chondrial DNA content was inhibited by pre-incubation with ICI

182780 (Fig. 5A). Among LHON cybrids, the one carrying

the 3460/ND1 mutation displayed the highest increase of mito-

chondrial DNA content after 17b-oestradiol treatment, so we

chose this cell line for the further experiments. To gain insight

into the mechanism involved in the 17b-oestradiol-mediated

mitochondrial DNA copy number increase, we evaluated the

gene expression level of the master regulators of mitochondrial

biogenesis: PPAR-� coactivator 1-a (PGC1-�) and its homologue

PGC-1�; nuclear respiratory factor 1 and 2 (NRF1 and NRF2)

and mitochondrial transcription factor A (Tfam). We observed a

coordinated gene induction (�up to 12-fold) after 6 h of treat-

ment both in control and LHON cybrids (Fig. 5B). The induction

was completely inhibited by the 17b-oestradiol antagonist ICI

182780. The increase in mitochondrial DNA content and the

induction of the key regulators of mitochondrial biogenesis

were paralleled by the upregulation of two mitochondrial-encoded

messenger RNAs: cytochrome c oxidase (COX) subunits I

(MTCOI), and NADH dehydrogenase subunit 5 (MTND5;

Fig. 5B). A postponed upregulation (after 12–24 h of


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

17b-oestradiol treatment) was observed for the nuclear-encoded

COX subunit IV (COIV; Fig. 5C). The induction of nuclear and

mitochondrial genes was inhibited by ICI 182780. Despite

gene induction, control and LHON cybrids grown in galactose

medium revealed a decrease of COIV and ND6 protein, with

the LHON cybrids showing the lowest protein amount.

Supplementation of medium with 17b-oestradiol partially

prevented the galactose-dependent COIV and ND6 protein

Figure 3 17b-Oestradiol ameliorates cell viability in galactose medium by reducing apoptosis. (A) Growth curves of control and LHON

cybrids maintained in glucose (glu) or galactose (gal) medium� 100nM 17b-oestradiol (E2). ***P50.001 for gal versus glu; +P50.05,++P50.01, +++P50.001 for gal + E2 versus gal. Data are expressed as % of untreated cell number in glucose medium, and are

mean� SEM from four different experiments in duplicate. Growth curves for control and 11778/ND4 are from HP27 and HFF3 clones.

Similar results were obtained with HPE9 and HGA clones (data not shown). (B) Mitochondrial membrane potential (mt�j) of control andLHON cybrids incubated for 24 h in glucose (glu) or galactose (gal) medium� 100nM 17b-oestradiol (E2). In a subset of experiments, cells

were pre-incubated with ICI 182780 (I). Data are expressed as mean fluorescence intensity (MFI;� SEM) (% of values of untreated cells in

glucose medium) and are from three independent experiments. Control and 11778/ND4 values are the mean from two clones. *P50.05,

***P50.001 for gal versus glu; +P50.05, +++P50.001 for gal + E2 versus gal and versus gal + E2+ I. (C) Percentages of apoptotic cells in

control and LHON cybrids incubated in glucose (glu) or galactose (gal) medium� 100 nM 17b-oestradiol (E2), as evaluated by labelling the

cells with annexin V. Data are the mean� SEM of the percent number of apoptotic cells from three repeated experiments. Control and

11778/ND4 values are the mean values from two clones. ��P50.001 LHON versus control; ***P50.001 for gal versus glu; ++P50.01

for gal + E2 versus gal (see Supplementary Fig. 2).


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

reduction and led to a remarkably increased amount of protein

when cells were maintained in glucose medium (Fig. 5C).

17b-Oestradiol enhances energeticcompetenceTo assess whether oestrogens improved the energetic competence

of cybrid cell lines, we measured the rate of oxygen consumption

and the total cellular ATP content in the presence or absence

of 17b-oestradiol. Supplementation of glucose medium with

17b-oestradiol led to a small although significant increase in

the rate of oxygen consumption both in LHON and control

cybrids after 24 h of incubation (Fig. 6A). Similar results were

obtained after 48 h of incubation (not shown). Similarly, supple-

mentation of glucose medium with 100 nM 17b-oestradiol led to

a �20% increase of cellular ATP content in both 11778/ND4

and control cybrids (Fig. 6B). After 24 h of incubation with

galactose medium, the ATP content significantly decreased

in both control and LHON cybrids as previously described

(Zanna et al., 2005), the LHON cybrids performing slightly

worse. Supplementation of medium with 100 nM 17b-oestradiolpartially prevented the galactose-induced ATP-content decrease.

This phenomenon was particularly evident in LHON cybrids

(540% increase in total ATP content in galactose medium

plus oestrogens compared with galactose medium plus vehicle;

Fig. 6B).

Oestrogen receptor b localizes to retinalganglion cellsSeveral studies have localized both oestrogen receptor a and

oestrogen receptor b to mitochondria of many cell types, including

rat primary neurons, human lens epithelial cells and human

foetal cortical neurons (for a review see Simpkins et al., 2008

Figure 4 17b-Oestradiol reduces mitochondrial network fragmentation. (A) Representative images of the four classes of cells (see

‘Results’ section) as observed in 11778/ND4 cybrids incubated for 24 h in galactose medium, loaded with MitoTracker Orange and

counterstained with 4’,6-diamidino-2-phenylindole. The inset shows the corresponding nuclear morphology. (B) Bar graphs showing

quantification of the four categories by blind test. Cybrids analysed were from 30 photos obtained from two controls (HGA, HP27) and

two 11778/ND4 (HFF3, HPE9) cybrid cell lines grown in galactose medium (gal)� 100nM 17b-oestradiol (E2). ��P50.001, ��P50.05 for

LHON versus WT; +++P50.001 for gal + E2 versus gal.


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

Figure 5 17b-Oestradiol induces mitochondrial biogenesis. (A) Amount of mitochondrial DNA in control and LHON cybrids maintained

for 3 h in glucose (glu) or galactose (gal) medium� 100nM 17b-oestradiol (E2). In a subset of experiments, cells were pre-incubated with

ICI 182780 (I). Bar graph represents the mean plus SEM from three experiments. †††P50.001 for glu + E2 versus glu and versus glu + E2+ I;

**P50.01 for gal versus glu; +++P50.001 for gal + E2 versus gal (see Supplementary Fig. 3). (B) Control and LHON cybrids were

(continued)


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

and Chen et al., 2009). To evaluate whether oestrogen receptors

localize to human retinal ganglion cells, the main target of LHON,

we performed immunoperoxidase staining on formalin-fixed,

paraffin-embedded retinal sections obtained post-mortem from

one patient with LHON and two control individuals of both gen-

ders. A punctuate staining typical of mitochondrial pattern was

observed in the somata of retinal ganglion cells and the

unmyelinated portion of the axons in the retinal nerve fibre

layer, as well as in the inner and outer plexiform layers (Fig. 7A

and B). Residual retinal ganglion cells in the patient with LHON

retained a similar expression pattern. A punctuate cytoplasm stain-

ing was also observed in neoplastic cells from a paraffin breast

cancer section, similar to what has been reported in BRC7 cells

(Pedram et al., 2006; Fig. 7C). In addition, double immunofluor-

escence with anti-mitochondrial and anti-oestrogen receptor bantibodies demonstrated co-localization of mitochondria and oes-

trogen receptor b in the somata of retinal ganglion cells (Fig. 7D).

Immunostaining with oestrogen receptor a antibodies gave nega-

tive results (not shown).

DiscussionThe current study, part of a long-standing investigation to char-

acterize the cellular behaviour of mitochondrial DNA mutations in

LHON using the cybrid cell model, had the dual aim of investigat-

ing the reasons for the higher prevalence of LHON in males and

exploring the potential compensatory effects of oestrogens on

mutant cell metabolism. Firstly, we provided evidence that both

oestrogen receptor a and oestrogen receptor b are present in the

nuclei of 143B.TK-cells, whereas only the oestrogen receptor blocalized to the mitochondria of osteosarcoma-derived cybrids.

These results extend previous studies on osteosarcoma-derived

cell lines (Solakidi et al., 2005) and confirm that oestrogen recep-

tor b is enriched in the mitochondria of different cell types (Yang

et al., 2004). We then showed that oestrogens present a multi-

layer effect on complex I-defective LHON cybrids, leading to

reduced production of reactive oxygen species, partially rescuing

cell viability in galactose medium by restoring membrane potential

and limiting apoptotic cell death. In addition, we also documented

a coordinated activation of mitochondrial biogenesis and a small

but significant improvement in the energetic competence of

cybrids induced by oestrogens. Finally, we showed that oestrogen

receptor b is localized to the mitochondrial network of human

retinal ganglion cells and the unmyelinated portion of their

axons in the retinal nerve fibre layer, and their expression is re-

tained in the surviving retinal ganglion cells of patients with

LHON. Thus, our results on the cybrid cell model apply to this

target tissue. Our study provides an explanatory framework for

male prevalence in LHON and a potential new avenue for thera-

peutic interventions.

Previous work by our group with cybrids has shown that mito-

chondrial DNA point mutations in LHON affecting different

subunits of complex I essentially lead to increased oxidative

Figure 5 Continuedincubated for 6 h in glucose (glu) or galactose (gal) medium� 100 nM 17b-oestradiol (E2). In a subset of experiments, cells were

pre-incubated with ICI 182780 (I). The relative expression of the following genes was evaluated by real-time PCR analysis: PGC1-� and its

homologue PGC-1�, NRF1, NRF2, Tfam, MTCOI, MTND5. (C) In a subsequent experiment, cells were incubated for 24 h in glucose (glu)

or galactose (gal) medium� 100nM 17b-oestradiol (E2), and the relative gene and protein expression of the nuclear encoded respiratory

chain subunits COIV evaluated by real-time PCR (top) and western blot analysis of mitochondrial fraction (bottom) along with the protein

expression of the mitochondrial encoded complex I subunit ND6. Data represent mean arbitrary units (� SEM) normalized to control

values in glucose medium and are from three experiments. A representative blot out of three is shown. †††P50.001, ††P50.01, †P50.05

for glu + E2 versus glu and versus glu + E2+ I; ***P50.001, **P50.01, *P50.05 for gal versus glu; +++P50.001 for gal + E2 versus gal.

Figure 6 17b-Oestradiol ameliorates energetic competence

of cybrid cells. (A) Control, 11778/ND4, 3460/ND1 and

14484/ND6 cybrids were incubated for 24 h in glucose

(glu)� 100nM 17b-oestradiol (E2) and the rate of oxygen

consumption measured. Data are mean� SEM from three to

four separate experiments. For each clone, the rate of oxygen

consumption in glucose medium supplemented with

17b-oestradiol has been normalized to the rate of oxygen

consumption in glucose medium plus ethanol. *P50.03;

**P50.003 for glu + E2 versus glu. (B) Control and 11778/ND4

LHON cybrids were incubated for 24 h in glucose (glu)� 100nM

17b-oestradiol and the total ATP cellular content measured by

luciferin/luciferase assay. Data are mean� SEM from four

separate experiments. For each clone, values are normalized for

the total ATP content in glucose medium. +++P50.001 for E2versus ethanol; *** for galactose (gal) versus glucose.


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

stress (Beretta et al., 2004; Floreani et al., 2005), defective com-

plex I-driven bioenergetics (Baracca et al., 2005) and an enhanced

predisposition to apoptotic cell death (Ghelli et al., 2003; Zanna

et al., 2005). Most of these pathologic cell phenotypes become

full blown when cybrids are grown in glucose-free, galactose-

supplemented medium, a condition forcing cells to rely on

oxidative phosphorylation for ATP production (Robinson et al.,

1992). In the present study, we further showed that under this

stressful metabolic condition, cybrids undergo a rapid increase in

production of reactive oxygen species. This was associated with

early increase in mitochondrial DNA copy number and, over time,

with a general upregulation of mitochondrial biogenesis, as

Figure 7 Oestrogen receptor b localizes to retinal ganglion cells. (A) Immunoperoxidase stain with anti-oestrogen receptor b antibodies

(ERb-H150) on horizontal retinal sections from a normal individual (left: male, 59-years-old) and a patient with LHON with the 11778/

ND4 mutation (Middle: male, 52-years-old). In the normal individual, a positive stain is observed in the somata of retinal ganglion cells

(between arrows) and the unmyelinated portion of the axons in the retinal nerve fibre layer, as well as in the inner and outer plexiform

layers. Residual retinal ganglion cells in the patient with LHON (arrows) show a similar expression pattern. Similar results were obtained

with the antibody ERb 485-503 both on male and female control individuals (not shown). (B) Higher magnification showing a typical

mitochondrial punctuate pattern of oestrogen receptor b in the somata of retinal ganglion cells. (C) Immunoperoxidase stain with

anti-oestrogen receptor b antibodies (ERb-H150) on a formalin fixed, paraffin embedded section of breast cancer used as a positive

control. Note the faint nuclear and the strong cytoplasm staining, similar to that reported in BRC7 breast cancer cells (Pedram et al., 2006).

(D) Double immunofluorescence stain with anti-mitochondrial (green) and anti-oestrogen receptor b (red) antibodies. The negative

control (right) shows a non-specific background fluorescence induced by fixation media, yet on top of this background is the evident

co-localization of mitochondria and oestrogen receptor b in the somata of retinal ganglion cells. Overlap is seen in yellow. NFL= nerve

fiber layer; IPL = inner plexiform layer; OPL=outer plexiform layer.


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

revealed by the increased expression of both mitochondrial and

nuclear genes coding for components of the mitochondrial respira-

tory chain. These changes are tightly linked to and driven by the

induction of the upstream master regulators of mitochondrial bio-

genesis, PGC1-� and PGC1-� and their targets, confirming that

these transcription co-activators can be powerfully induced by

reactive oxygen species (St-Pierre et al., 2006). The forced use

of oxidative phosphorylation in galactose medium ultimately led

to a massive increase of reactive oxygen species in LHON cybrids,

significantly higher than controls, and reduced cell viability due to

an increased rate of apoptotic cell death. Using this metabolic

paradigm, we investigated the effects of oestrogen treatment.

Exposure of LHON and control cybrids to 100 nM of

17b-oestradiol led to a rapid decrease in levels of reactive

oxygen species, both in glucose and galactose medium, which

was particularly remarkable for LHON cells. This decline in reactive

oxygen species was paralleled by an increase in SOD2 activity,

which was rapid and not linked with higher protein levels in the

first hours of 17b-oestradiol incubation. After 24 h, SOD2 protein

increase was evident only in cells maintained in glucose medium.

These effects also applied to control cells and were oestrogen

receptor dependent. Our results are compatible with previous

studies showing that oestrogens may activate SOD2 both by a

direct effect and by increased transcription (Borras et al., 2004;

Pedram et al., 2006; Gottipati and Cammarata, 2008). Similarly,

LHON cybrid viability, loss of membrane potential and rate of

apoptotic cell death in galactose medium were all significantly

rescued by 17b-oestradiol treatment. These results were mirrored

by changes in mitochondrial network dynamics of LHON cybrids

in galactose. In fact, the hyper-fragmented mitochondrial morph-

ology, typically associated with pre-apoptotic or apoptotic cell

morphology, was clearly reduced by 17b-oestradiol treatment.

The administration of 17b-oestradiol also resulted in the

powerful activation of mitochondrial biogenesis, orchestrated by

the upstream transcription machinery including transcriptional

co-activators PGC1� and PGC1�, and transcription factors NRF1,

NRF2 and Tfam. The timing of this activation indicates early

events, possibly induced by direct action of oestrogen on mito-

chondria, and long-term events due to genomic effects of oestro-

gen (Mattingly et al., 2008). In fact, the increase in mitochondrial

DNA copy number occurred as early as 15min after treatment.

This is most probably driven by a direct effect of oestrogen on

mitochondrial DNA replication as suggested by the described

oestrogen responsive elements in the D-loop region (Chen et al.,

2004b). However, this increase in mitochondrial biogenesis ultim-

ately resulting in increased respiratory chain complexes was not

mirrored by an equivalent improvement in cell bioenergetics as

measured by oxygen consumption and total ATP content in this

model system.

Overall, our results using cybrid cells demonstrate that oestro-

gens can improve the mitochondrial dysfunction in LHON, most

prominently by counteracting the excess of reactive oxygen spe-

cies production, which leads to rapid loss of viability and apoptotic

cell death in the time-course experiments after switching to gal-

actose medium. In fact, under this condition of over-imposed

metabolic stress, administration of oestrogens improved all param-

eters measured. In contrast, the beneficial effect of oestrogens did

not seem to be equally efficient in ameliorating the bioenergetic

competence of LHON cybrids. All results obtained after oestrogen

treatment were mediated by oestrogen receptors as counteracted

by the use of the oestrogen receptor antagonist ICI 182780.

Our observations in the cell model of cybrids most probably

apply to the in vivo target tissue, i.e. retinal ganglion cells and

their axons. We demonstrate here for the first time that oestrogen

receptors are expressed in human retina, particularly abundant in

retinal ganglion cells. The oestrogen receptor b was shown by

immunofluorescence to decorate the mitochondrial network

co-localizing with mitochondrial directed antibodies. The expres-

sion of oestrogen receptor b was also retained in the surviving

retinal ganglion cells from patients with LHON. There is clinical

evidence that pathologic expression of subclinical LHON differs in

male and female unaffected carriers of the mitochondrial DNA

mutation. In particular, optical coherence tomography measure-

ment of nerve-fibre-layer swelling in the temporal–inferior quad-

rants (papillomacular bundle) of unaffected mutation carriers was

significantly thicker in males compared with females (Savini et al.,

2005). Similarly, subclinical colour vision defects observed in

asymptomatic carriers were prevalent in males (Ventura et al.,

2007). A further supporting element is the slightly older age of

onset in females compared with males (Carelli et al., 2004). This is

possibly due to a subgroup of females developing LHON after

menopause. Furthermore, cybrids exposure to testosterone did

not show relevant changes in cell growth and respiration

(Andrea Martinuzzi, data not shown), pointing to a specific role

of oestrogen on mitochondrial metabolism.

In conclusion, our study provides new insights for LHON,

strongly supporting a metabolic basis for the observed and, to

date, unexplained male prevalence. The different exposure to oes-

trogens between males and females is sufficient to modify the

severity of mitochondrial dysfunction induced by mitochondrial

DNA mutations affecting complex I in LHON, as clearly demon-

strated by our results. Furthermore, these observations hold prom-

ise for a therapeutic use of molecules with oestrogen-like activities,

such as phyto-oestrogens (Kuiper et al., 1997) that preferentially

bind to oestrogen receptor b, thus limiting their oestrogenic activ-

ity to cells expressing these receptors, such as retinal ganglion

cells.

AcknowledgementsThe authors wish to thank Claudia Travaglini, Mariangela

Sebastiani, Enrico Secchi, Daniela Catanzaro and Silvia Vidali for

their excellent technical assistance.

FundingThis work was supported by Telethon Grant GGP06233 (V.C.),

Associazione Serena Talarico per i giovani nel mondo and

Fondazione Giuseppe Tomasello O.N.L.U.S. (C.G.), Research to

Prevent Blindness (F.N.R.-C. and A.A.S.), Struggling Within

Leber’s Foundation (F.N.R.-C. and A.A.S.), Eierman Foundation


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

(F.N.R.-C. and A.A.S.) and National Institutes of Health grant

EY03040 (F.N.R.-C. and A.A.S.).

Supplementary materialSupplementary material is available at Brain online.

ReferencesBaracca A, Solaini G, Sgarbi G, Lenaz G, Baruzzi A, Schapira AH, et al.

Severe impairment of complex I-driven adenosine triphosphate synthe-

sis in Leber hereditary optic neuropathy cybrids. Arch Neurol 2005; 62:

730–6.

Bernard G, Bellance N, James D, Parrone P, Fernandez H, Letellier T,

et al. Mitochondrial bioenergetics and structural network organization.

Journal of Cell Science 2007; 120: 838–48.

Beretta S, Mattavelli L, Sala G, Tremolizzo L, Schapira AH, Martinuzzi A,

et al. Leber hereditary optic neuropathy mtDNA mutations disrupt

glutamate transport in cybrid cell lines. Brain 2004; 127: 2183–92.Borras C, Gambini J, Gomez-Cabrera MC, Sastre J, Pallardo FV,

Mann GE, et al. 17beta-oestradiol up-regulates longevity-related anti-

oxidant enzyme expression via the ERK1 and ERK2[MAPK]/NFkappaB

cascade. Aging Cell 2005; 4: 113–18.

Bu XD, Rotter JI. X chromosome-linked and mitochondrial gene control

of Leber hereditary optic neuropathy: evidence from segregation

analysis for dependence on X chromosome inactivation. Proc Natl

Acad Sci USA 1991; 88: 8198–202.

Carelli V, Achilli A, Valentino ML, Rengo C, Semino O, Pala M, et al.

Haplogroup effects and recombination of mitochondrial DNA: novel

clues from the analysis of Leber hereditary optic neuropathy pedigrees.

Am J Hum Genet 2006; 78: 564–74.Carelli V, Giordano C, d’Amati G. Pathogenic expression of homoplasmic

mtDNA mutations needs a complex nuclear-mitochondrial interaction.

Trends Genet 2003; 19: 257–62.

Carelli V, La Morgia C, Iommarini L, Carroccia R, Mattiazzi M,

Sangiorgi S, et al. Mitochondrial optic neuropathies: how two gen-

omes may kill the same cell type? Biosci Rep 2007; 27: 173–84.

Carelli V, Ross-Cisneros FN, Sadun AA. Mitochondrial dysfunction as a

cause of optic neuropathies. Prog Retin Eye Res 2004; 23: 53–89.

Carelli V, Vergani L, Bernazzi B, Zampieron C, Bucchi L, Valentino M,

et al. Respiratory function in cybrid cell lines carrying European mtDNA

haplogroups: implications for Leber’s hereditary optic neuropathy.

Biochim Biophys Acta 2002; 1588: 7–14.

Chalmers RM, Davis MB, Sweeney MG, Wood NW, Harding AE.

Evidence against an X-linked visual loss susceptibility locus in Leber

hereditary optic neuropathy. Am J Hum Genet 1996; 59: 103–8.Chen JQ, Cammarata PR, Baines CP, Yager JD. Regulation of mitochon-

drial respiratory chain biogenesis by estrogens/estrogen receptors and

physiological pathological and pharmacological implications. Biochim

Biophys Acta 2009; 1793: 1540–70.

Chen JQ, Delannoy M, Cooke C, Yager JD. Mitochondrial localization of

ERa and ERb in human MCF7 cells. Am J Physiol Endocrinol Metab

2004a; 286: E1011–22.

Chen JQ, Eshete M, Alworth WL, Yager JD. Binding of MCF-7 cell mito-

chondrial proteins and recombinant human estrogen receptors alpha

and beta to human mitochondrial DNA estrogen response elements.

J Cell Biochem 2004b; 93: 358–73.Ferlini C, Scambia G. Assay for apoptosis using the mitochondrial probes,

Rhodamine123 and 10-N-nonyl acridine orange. Nat Protoc 2007; 2:

3111–14.

Floreani M, Napoli E, Martinuzzi A, Pantano G, De Riva V, Trevisan R,

et al. Antioxidant defences in cybrids harboring mtDNA mutations

associated with Leber’s hereditary optic neuropathy. FEBS J 2005;

272: 1124–35.

Ghelli A, Porcelli AM, Zanna C, Vidoni S, Mattioli S, Barbieri A, et al. The

background of mitochondrial DNA haplogroup J increases the sensitiv-

ity of Leber’s hereditary optic neuropathy cells to 2,5-hexanedione

toxicity. PLoS One 2009; 4: e7922.

Ghelli A, Zanna C, Porcelli AM, Schapira AH, Martinuzzi A, Carelli V,

et al. Leber’s hereditary optic neuropathy (LHON) pathogenic muta-

tions induce mitochondrial-dependent apoptotic death in transmito-

chondrial cells incubated with galactose medium. J Biol Chem 2003;

278: 4145–50.

Gottipati S, Cammarata PR. Mitochondrial superoxide dismutase activa-

tion with 17 beta-estradiol-treated human lens epithelial cells. Mol Vis

2008; 14: 898–905.

Hudson G, Carelli V, Horvath R, Zeviani M, Smeets HJ, Chinnery PF.

X-Inactivation patterns in females harboring mtDNA mutations

that cause Leber hereditary optic neuropathy. Mol Vis 2007a; 13:

2339–43.

Hudson G, Carelli V, Spruijt L, Gerards M, Mowbray C, Achilli A, et al.

Clinical expression of Leber hereditary optic neuropathy is affected by

the mitochondrial DNA-haplogroup background. Am J Hum Genet

2007b; 81: 228–33.Hudson G, Keers S, Yu-Wai Man P, Griffiths P, Huoponen K,

Savontaus ML, et al. Identification of an X-chromosomal locus and

haplotype modulating the phenotype of a mitochondrial DNA disorder.

Am J Hum Genet 2005; 77: 1086–91.Kirkman MA, Yu-Wai-Man P, Korsten A, Leonhardt M, Dimitriadis K, De

Coo IF, et al. Gene-environment interactions in Leber hereditary optic

neuropathy. Brain 2009; 132: 2317–26.Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S,

et al. Comparison of the ligand binding specificity and transcript tissue

distribution of estrogen receptors alpha and beta. Endocrinology 1997;

138: 863–70.Man PY, Griffiths PG, Brown DT, Howell N, Turnbull DM, Chinnery PF.

The epidemiology of Leber hereditary optic neuropathy in the North

East of England. Am J Hum Genet 2003; 72: 333–9.Man PY, Griffiths PG, Hudson G, Chinnery PF. Inherited mitochondrial

optic neuropathies. J Med Genet 2009; 46: 145–58.Mattingly KA, Ivanova MM, Riggs KA, Wickramasinghe NS, Barch MJ,

Klinge CM. Estradiol stimulates transcription of nuclear respiratory

factor-1 and increases mitochondrial biogenesis. Mol Endocrinol

2008; 22: 609–22.Mussini C, Pinti M, Bugarini R, Borghi V, Nasi M, Nemes E, et al. Effect

of CD4-monitored treatment interruption on mitochondrial DNA

content in HIV-infected patients: a prospective study. AIDS 2005;

19: 1627–33.Oberley LW, Spitz DR. Assay of superoxide dismutase activity in tumor

tissue. Methods Enzymol 1984; 105: 457–64.Phasukkijwatana N, Kunhapan B, Stankovich J, Chuenkongkaew WL,

Thomson R, Thornton T, et al. Genome-wide linkage scan and asso-

ciation study of PARL to the expression of LHON families in Thailand.

Hum Genet 2010; 128: 39–49.Pedram A, Razandi M, Wallace DC, Levin ER. Functional estrogen

receptors in the mitochondria of breast cancer cells. Mol Biol Cell

2006; 17: 2125–37.Pello R, Martın MA, Carelli V, Nijtmans LG, Achilli A, Pala M, et al.

Mitochondrial DNA background modulates the assembly kinetics of

OXPHOS complexes in a cellular model of mitochondrial disease.

Hum Mol Genet 2008; 17: 4001–11.

Robinson BH, Petrova-Benedict R, Buncic JR, Wallace DC. Nonviability of

cells with oxidative defects in galactose medium: a screening test for

affected patient fibroblasts. Biochem Med Metab Biol 1992; 48:

122–6.

Sadun AA, Carelli V, Salomao SR, Berezovsky A, Quiros PA, Sadun F,

et al. Extensive investigation of a large Brazilian pedigree of 11778/

haplogroup J Leber hereditary optic neuropathy. Am J Ophthalmol

2003; 136: 231–8.

Savini G, Barboni P, Valentino ML, Montagna P, Cortelli P,

De Negri AM, et al. Retinal nerve fiber layer evaluation by optical


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

coherence tomography in unaffected carriers with Leber’s hereditaryoptic neuropathy mutations. Ophthalmology 2005; 112: 127–31.

Sebastiani M, Giordano C, Nediani C, Travaglini C, Borchi E, Zani M,

et al. Induction of mitochondrial biogenesis is a maladaptive mechan-

ism in mitochondrial cardiomyopathies. J Am Coll Cardiol 2007; 50:1362–9.

Shankar SP, Fingert JH, Carelli V, Valentino ML, King TM, Daiger SP,

et al. Evidence for a novel X-linked modifier locus for Leber hereditary

optic neuropathy. Ophthalmic Genet 2008; 29: 17–24.Simpkins JW, Yang SH, Sarkar SN, Pearce V. Estrogen actions on

mitochondria-physiological and pathological implications. Cell

Endocrinol 2008; 290: 51–9.Solakidi S, Psarra AM, Sekeris CE. Differential subcellular distribution of

estrogen receptor isoforms: localization of ERalpha in the nucleoli and

ERbeta in the mitochondria of human osteosarcoma SaOS-2 and

hepatocarcinoma HepG2 cell lines. Biochim Biophys Acta 2005;1745: 382–92.

Stadelmann C, Lassmann H. Detection of apoptosis in tissue sections.

Cell Tissue Res 2000; 301: 19–31.

St-Pierre J, Drori S, Uldry M, Silvaggi JM, Rhee J, Jager S, et al.Suppression of reactive oxygen species and neurodegeneration by

the PGC-1 transcriptional coactivators. Cell 2006; 127: 379–408.

Ventura DF, Gualtieri M, Oliveira AG, Costa MF, Quiros P, Sadun F,

et al. Male prevalence of acquired color vision defects in asymptomaticcarriers of Leber’s hereditary optic neuropathy. Invest Ophthalmol Vis

Sci 2007; 48: 2362–70.

Vilkki J, Ott J, Savontaus ML, Aula P, Nikoskelainen EK. Optic atrophy inLeber hereditary optic neuroretinopathy is probably determined by an

X-chromosomal gene closely linked to DXS7. Am J Hum Genet 1991;

48: 486–91.

Vina J, Sastre J, Pallardo FV, Gambini J, Borras C. Role of mitochondrialoxidative stress to explain the different longevity between gen-

ders: protective effect of estrogens. Free Radic Res 2006; 40:

1359–65.

Wallace DC, Singh G, Lott MT, Hodge JA, Schurr TG, Lezza AM, et al.Mitochondrial DNA mutation associated with Leber’s hereditary optic

neuropathy. Science 1988; 242: 1427–30.

Wu CW, Ping YH, Yen JC, Chang CY, Wang SF, Yeh CL, et al. Enhancedoxidative stress and aberrant mitochondrial biogenesis in human

neuroblastoma SH-SY5Y cells during methamphetamine induced

apoptosis. Toxicol Appl Pharmacol 2007; 220: 243–51.

Yang SH, Liu R, Perez EJ, Wen Y, Stevens SM Jr, Valencia T, et al.Mitochondrial localization of estrogen receptor beta. Proc Natl Acad

Sci USA 2004; 101: 4130–5.

Zanna C, Ghelli A, Porcelli AM, Martinuzzi A, Carelli V, Rugolo M.

Caspase-independent death of Leber’s hereditary optic neuropathycybrids is driven by energetic failure and mediated by AIF and

Endonuclease G. Apoptosis 2005; 10: 997–1007.

Zanna C, Ghelli A, Porcelli AM, Karbowski M, Youle RJ, Schimpf S, et al.

OPA1 mutations associated with dominant optic atrophy impairoxidative phosphorylation and mitochondrial fusion. Brain 2008; 131:

352–67.


at Universit? di B

ologna - Sistem

a Bibliotecario d'A



ownloaded from

Oestrogens ameliorate mitochondrial dysfunction in Leber's hereditary optic neuropathy

Documents