|Home | About | Journals | Submit | Contact Us | Français|
Estradiol (E2) is a very potent cytoprotectant against a wide variety of cellular insults in numerous different cell models, including a Friedreich’s ataxia (FRDA) model. Previously, we demonstrated that estrogen-like compounds are able to prevent cell death in an FRDA model independent of any known estrogen receptor (ER) by reducing reactive oxygen species (ROS) and the detrimental downstream effects of ROS buildup including oxidative damage to proteins and lipids and impaired mitochondrial function.
We have previously demonstrated by western blot that our cell model lacks ERα and expresses only very low levels of ERβ. Using L-buthionine (S,R)-sulfoximine (BSO) to induce oxidative stress in human FRDA fibroblasts, we determine the potency and efficacy of the soy-derived ERβ agonist S-equol and its ERα-preferring enantiomer, R-equol in vitro on cell viability and ROS accumulation. Here we demonstrate that these equol biphenolic compounds, while significantly less potent and efficacious than E2, provide statistically similar attenuation of ROS and cytoprotection against a BSO-induced oxidative insult.
These preliminary data demonstrate that estrogen and soy-derived equols could have a beneficial effect in delaying the onset and decreasing the severity of symptoms in FRDA patients by an antioxidant mechanism. In addition, these data confirm that the protection seen previously with E2 was indeed unrelated to ER binding.
First recognized in 1863 , Friedreich’s Ataxia (FRDA) is the most common hereditary form of ataxia characterized by an autosomal recessive GAA trinucleotide repeat in the FXN gene, resulting in the absence of frataxin protein [2,3]. The exact function of frataxin is unclear, however it is necessary for iron metabolism within cells, Fe-S cluster assembly in proteins, and maintenance of cellular redox state. Without sufficient levels of frataxin, reactive oxygen species (ROS) begin to accumulate and cells are unable to maintain function of Fe-S cluster proteins essential for mitochondrial respiration leading to mitochondrial dysfunction, insufficient energy production and ultimately cell death, beginning in organs with greater energy requirements and thus more dependent on aerobic ATP production, such as the heart, brain and spinal cord. Symptoms usually begin in the second decade of life and include ataxia, neural hearing and ocular abnormalities, scoliosis, diabetes and cardiomyopathy, which is the most common cause of premature death in FRDA patients [for review see Ref .
First detected in humans in 1982 , equol is a biphenolic isoflavone metabolized from the soy product daidzein by intestinal flora [6-8] in 14-59% of the human population . Equol is known to act as an antioxidant [10,11], decreases circulating estrogens and androgens , inhibits DHT binding to its receptor  and decreases risks of prostate [9,11,14] and breast cancer . Separation of racemic equol mixtures shows that S-equol binds with very high affinity to ERβ (Kd ~ 0.73 nM), while its enantiomer, R-equol has a far lower affinity for ERβ, instead showing a preference for ERα (Kd ~ 15.4 nM), while E2 has a Kd ~ 0.05-0.1 nM [16,17]. These enantiomers allow for the discrimination between effects due to antioxidant effects and those due to ERβ activation.
We have previously shown that phenolic estrogens are able to prevent BSO-induced FRDA skin fibroblast death, as well as block the formation of ROS , prevent lipid peroxidation, protein damage, depletion of ATP and support the mitochondria and oxidative phosphorylation . In the present study, we provide further evidence that E2 acts by an ERα- and ERβ-independent mechanism. In addition, we demonstrated a lack of ERα and a very low level of ERβ in FRDA fibroblasts by western blot . Here, we show pharmacologically that ERβ is not contributing to this process, as R- and S-equol have statistically equivalent efficacies and potencies, represented here as EC50 values. These data indicate that it is the phenolic ring present in the compound structure of equol and E2 and not intrinsic receptor binding ability that is responsible for cytoprotective effects in this FRDA cell model. Although these compounds are substantially less efficacious and potent than compounds previously used , this pharmacologic model lends support to the non-receptor mediated, non-genomic antioxidant mechanism of E2.
To determine the effect of R- and S-equol (Figure (Figure1)1) on cell viability, we first assessed their protective potential compared to 17β-estradiol (E2) at 100nM, a concentration previously shown to be very protective in this cell model . At 100nM, both R- and S-equol provided statistically significant protection compared to the BSO-alone treated group, however the two groups did not differ significantly from each other (Figure (Figure2a).2a). E2 also provided significantly more protection than either of these two compounds (Figure (Figure2a).2a). A dose–response assessment showed that R- and S-equol have almost identical cytoprotective profiles at all concentrations (Figure (Figure2b),2b), and EC50 evaluation demonstrated that the two have statistically equivalent EC50 values (Table (Table1),1), indicating that the cytoprotective effect is not due to stimulation of ERβ.
To determine the effects of R- and S-equol on ROS attenuation, these two compounds were again compared to E2 (Figure (Figure3a).3a). BSO induced a 2-fold increase of ROS, which was prevented by 100nM concentrations of E2, R-equol and S-equol. None of these groups differed from each other. In addition, a dose response curve for R- and S-equol shows that there is no significant difference in the ROS attenuation profiles of these two compounds at any concentration (Figure (Figure3b),3b), and the EC50 values do not differ significantly (Table (Table11).
FRDA is the most common of the inherited ataxias world wide, affecting an estimated 1:50,000 to 1:20,000 people [2,4]. With the effective loss of functional frataxin throughout all organ systems, and the resulting ROS proliferation and mitochondrial respiration impairment, cells in organs most dependent on ATP production begin to degenerate [4,20]. This results in the loss of cells in the posterior columns and spinocerebellar tracts of the spinal cord, resulting in tremor and ataxia, as well as lateral and kyphoscoliosis, weakness, speech problems, pes cavitus, an increased incidence of diabetes mellitus and glucose intolerance and cardiac disorders, such as hypertrophic cardiomyopathy with interstitial fibrosis . Disease onset and severity is variable depending on the number of GAA trinucleotide repeats present in the first intron of the FXN gene, although this alone is not able to account for the full course of the disease process . There is little difference between males and females in terms of disease onset, progress and severity as this is inherited in an autosomal recessive manner and symptoms begin in the first 2 decades of life, before hormone level changes in puberty .
Estrogen and non-feminizing estrogens have been shown to be potently cytoprotective in many different cell and animal models of disease states [23,24], including a FRDA cell model . Previous observations have demonstrated that antioxidants, especially mitochondrially targeted antioxidants [25,26], including estrogen receptor agonists and non-feminizing estrogens  are protective against FRDA. These effects have been shown to be ER independent and are instead based in the antioxidant properties of phenolic estrogens [27,28].
Estrogens exert both genomic and non-genomic effects on redox status of cells for reviews see [29-34]. Unfortunately, no studies on the genomic effects of estrogens has been published using FRDA cells, but we have reported that these cells respond to estrogens even in the presence of a pan-estrogen receptor inhibitors, ICI 182780 , have no detectable ERα and low levels of ERβ , and exhibit these effects at concentrations in excess of the ED50 for 17β-estradiol . Nonetheless, genomic effects of estrogens on antioxidant enzymes have been reported, which could contribute to estrogen’s antioxidant effects. For example, tamoxifen is reported to up-regulated the quinine reductase, NQO1 , and estrogens up-regulate expression of peroxidase-1 and MnSOD . In contrast, Pajovic and Saicic  have reported that MnSOD, glutathione peroxidase, glutathione-S-transferase and glutathione reductase are decreased by estradiol, whereas catalase is increased. The extent to which the non-genomic effects of estrogens influence these paradoxical decreases in antioxidant enzyme expression is not known.
Estrogens are highly lipid soluble (the logarithm of the octanol/water partition coefficient, log P, is 3.35) and largely reside in the membrane component of cells  where they are ideally suited to affect oxidation of unsaturated bonds in phospholipids. Indeed, estrogens appear to intercalate into the membrane with their phenolic A ring situated near the site of lipid peroxidation . We reasoned that estrogens may interrupt lipid peroxidation chain reactions via oxidation in a manner that could be redox-cycled back to the parent estrogen, using a plentiful and regenerable source of cellular reducing potential, such as glutathione or NADPH. We discovered that estrogens were converted via hydroxyl radical exposure to a quinol product that was, in turn, enzymatically reduced back to the parent estrogen in the presence of NAD(P)H as a co-factor [27,28]. This estrogen redox cycle is operative in the central nervous system  where it serves, together with the “classical” antioxidant mechanism for phenolic compounds, as a defense mechanism against ROS.
Equol is a naturally derived biphenolic (Figure (Figure1)1) product of soy digestion in a substantial percentage of the American population . It is created by intestinal flora as a racemic mixture of the R- and S-forms, with the S-form being very selective for ERβ, the only ER present in FRDA fibroblasts  while the R-form is only a very weak agonist at this receptor [16,17]. Our results indicate that, while not as potent or efficacious as E2 (Figure (Figure2a2a and and33a) , the R- and S-forms of equol are equally effective in attenuating ROS (Figure (Figure3b,3b, Table Table1)1) and preventing cell death (Figure (Figure2b,2b, Table Table1).1). These data indicate that equol, specifically the non-feminizing R-equol, could potentially be used to prevent or delay cell death and pathologic symptoms in FRDA and supports our previous hypothesis that estrogen-like compounds are acting in a manner unrelated to any known ER [18,19].
Because the biphenolic compounds R- and S-equol have statistically equal cytoprotective profiles despite extremely different ERβ binding profiles, these data confirm that ERβ is not involved in the protective effects of E2 seen previously in this FRDA fibroblast model [18,19]. Furthermore, this study demonstrates that estrogen and soy-derived equols are effective at reducing ROS and improving cell viability in FRDA fibroblasts and shows that naturally derived soy estrogens could have a beneficial effect in delaying the onset and decreasing the severity of symptoms in FRDA patients by an antioxidant mechanism. These data add more weight to the neuroprotective hypothesis of estrogen and provide evidence that E2 and other phenol ring-containing estrogens should be considered as candidate drugs for the treatment and prevention of the symptoms of FRDA.
Fibroblasts from a 30 year old Friedreich’s Ataxia (FRDA) patient (Coriell Institute, Camden NJ, USA) were maintained in Dulbecco’s Modified Eagle Medium (DMEM; ThermoScientific, Waltham, MA, USA) with 10% charcoal-stripped fetal bovine serum (CSFBS; ThermoScientific, Waltham, MA, USA), 1% GlutaMAX (ThermoScientific, Waltham, MA, USA) and 1% penicillin-streptomycin (Invitrogen, Carlsbad, CA, USA) at 37°C in 90% humidity and 5% CO2. At the time of treatment, the FRDA fibroblast media was changed to phenol red- and sodium pyruvate-free DMEM (ThermoScientific, Waltham, MA, USA) and 1% penicillin-streptomycin. Experiments were conducted using cell passages 15–19.
17β-Estradiol (E2) was acquired from Steraloids, Inc. (Newport, RI, USA). L-buthionine (S,R)-sulfoximine (BSO) was obtained from Sigma-Aldrich (St Louis, MO, USA). R- and S- Equol were obtained from the laboratory of Dr Robert J Handa at The University of Arizona.
FRDA fibroblasts were removed from culture with 0.25% Trypsin-EDTA (Invitrogen, Carlsbad, CA, USA) and plated on 96-well plates at a density of 3000 cells per well in DMEM with 10% CSFBS, 1% GlutaMAX and 1% penicillin-streptomycin. After 24 hours the media was removed and replaced with phenol red- and sodium pyruvate-free DMEM with 1% penicillin-streptomycin. The cells were then treated for 12 to 48 hours with either dimethyl sulfoxide vehicle control (DMSO; Sigma-Aldrich, St Louis, MO, USA) or 1mM BSO in the presence of E2, R-equol or S-equol ((3S)-3-(4-Hydroxyphenyl)-7-chromanol). This duration of exposure was chosen based on our observation of BSO-induced enhacement of ROS at 12 hours and cell death at 48 hours .
Cells were plated on a 96-well plate at a density of 5,000 cells per well, then treated with vehicle or 1mM BSO. After 48 hours of BSO treatment, the media was removed, and 1 μg/mL Calcein AM (CalBiochem, San Diego, CA, USA) in phosphate buffer pH 7.2 (PBS; Fisher Scientific, Pittsburg, PA, USA) was added to each well and the plate was incubated for 10 minutes at 37°C. Cell viability was determined with a Tecan Infinite M200 (Tecan Systems, Inc., San Jose, CA) plate reader with an excitation of 490nm and emission of 520nm at 48 hours.
After 12 hours of treatment the media was removed from each well of the 96-well plate, and 100μL of a 1μM 2’,7’-Dichlorodihydrofluorescein diacetate (DCFDA; AnaSpec Inc., Fremont, CA, USA) in PBS was added to each well. The plates were returned to a 37°C incubator for 20 minutes, then each well was washed three times with PBS and the resulting reaction was read on a Tecan Infinite M200 plate reader with an absorbance of 495 nm and an emission of 529 nm.
All data are displayed as mean ± 1 standard deviation. These data were analyzed using the ANOVA against an alpha level of 0.05. All bar graphs were made using GraphPad Prism 5 and EC50 calculations were made with GraphPad Prism 5. For all groups, n=8 wells and experiments were repeated three times to ensure consistency.
BSO: L-buthionine (S,R)-sulfoximine; FRDA: Friedreich’s Ataxia; E2: 17β-Estradiol; ROS: Reactive oxygen species; ER: Estrogen receptor; ERα: Estrogen receptor α; ERβ: Estrogen receptor β.
The authors declare that they have no competing interests.
TER carried out the experiments, performed statistical analysis and wrote the initial draft of the manuscript. JWS revised and approved the final manuscript. All authors were involved with the conception and design of the studies. Both authors read and approved the final manuscript.
The pre-publication history for this paper can be accessed here:
The authors would like to thank Dr Robert J Handa for providing us with the R- and S-equol compounds. We would also like to thank Yogesh Mishra for help with ChemDraw software. Supported in part by NIH Grants P01 AG100485, P01 AG22550, and P01 AG027956 (to JWS) and NIA Grant T31 AG020494 (to TER).