17β-estradiol prevents experimentally-induced oxidative damage to membrane lipids and nuclear DNA in porcine ovary

Abstract

Estrogens, with their principle representative 17β-estradiol, contribute to the redox state of cells showing both pro- and antioxidative properties. In the ovary, being the main source of estrogens, maintaining balance between the production and detoxification of ROS is crucial. Whereas ovary estrogen concentration is difficult to estimate, its circulating concentration in women may reach the nanomolar level. The aim of the study was to evaluate the effects of 17β-estradiol on oxidative damage to membrane lipids (lipid peroxidation, LPO) and to nuclear DNA in the porcine ovary under basal conditions and in the presence of Fenton reaction (Fe²⁺+H₂O₂→Fe³⁺+^•OH + OH⁻) substrates. Ovary homogenates and DNA were incubated in the presence of 17β-estradiol (1 mM–1 pM), without/with FeSO₄ (30 μM) + H₂O₂ (0.5 mM). Malondialdehyde + 4-hydroxyalkenals (MDA + 4-HDA) concentration (LPO index) was measured spectrophotometrically. The concentration of 8-oxo-7,8-dihydro-2’-deoxyguanosine (8-oxodG) (DNA damage index) was measured by HPLC. We observed that 17β-estradiol did not alter the basal level of oxidative damage, but reduced Fe²⁺+H₂O₂-induced oxidative damage to membrane lipids when ≥10 nM and to DNA at concentrations ≥1 nM. In the ovary at near physiological concentration, 17β-estradiol prevents experimentally induced oxidative damage. This suggests that under physiological conditions this hormone may contribute to protecting the ovary against oxidative damage.

• 17β-estradiol
• lipid peroxidation
• nuclear DNA
• ovary
• oxidative damage

Introduction

Estrogens, of which 17β-estradiol is a principle representative, are hormones belonging to a group of steroid derivatives of cholesterol. Estrogens are synthesized from androgens, primarily in the ovaries, and, to a lesser extent, in the placenta, testes, and adrenal cortex. Their functions are very broad. They regulate growth, development, metabolism, sexual functions, and reproduction, etc. In addition, estrogens have a certain impact on the redox state of cells showing both pro- and antioxidative properties.

Epidemiological as well as experimental evidence indicates that estrogens, as prooxidants, may play a role in cancer development in estrogen-responsive tissues, such as the breast and the endometrium. Two potential mechanisms of the estrogen-mediated carcinogenesis are considered. The first originates from hormonal effects via receptor-mediated pathways which can stimulate the proliferation of cancer cells [Bolton and Thatcher 2008]. The second potential mechanism is independent of the interaction between hormone and its receptor and is associated with direct action of estrogen metabolites. These catechol metabolites are generated by the 2-, 4- and 16α-hydroxylation pathways that produce – via redox cycles – semiquinone and quinone intermediates, which in turn generate reactive oxygen species (ROS) [Spencer et al. 2012]. Both quinone metabolites may contribute to cancer initiation via oxidative damage to macromolecules. Oxidative damage to macromolecules in response to in vivo treatment with 17β-estradiol has been documented, [Karbownik et al. 2001a; Karbownik et al. 2001b].

Estrogens also possess antioxidative properties. Current evidence indicates that estrogens can reduce oxidative stress (i.e., disrupted redox homeostasis) at two levels, by preventing ROS generation and by scavenging free radicals. Estrogens can upregulate the endogenous antioxidative defense by indirect action on the expression and activity of such enzymes as SOD, GPx1, and GPx4 [Zhang et al. 2007]. Considering the role of estrogens as endogenous antioxidants it is worth recalling that females live longer than males. This has been attributed to higher levels of estrogen in females and supported in part by the observation in females that at low estrogen they are not likely to exhibit a significant antioxidant capacity [Vina et al. 2011].

Oxidative stress can lead to damage of lipids, proteins, and DNA (mitochondrial DNA included). In the ovary these processes can have a negative impact on female fertility and gamete health [Borowiecka et al. 2012; Devine et al. 2012; Karuputhula et al. 2013]. Therefore, maintaining balance between the production and detoxification of ROS is crucial for proper functioning of the ovary, being the main source of estrogens.

The most basic reaction of oxidative stress is the Fenton reaction (Fe²⁺ + H₂O₂ → Fe³⁺ + ^•OH + OH⁻) that produces the most harmful free radical, hydroxyl radical (^•OH). In our previous study [Karbownik-Lewinska et al. 2010] we have shown that external hydrogen peroxide (H₂O₂) is not indispensible for the experimental induction of lipid peroxidation in porcine ovary homogenates. This may suggest that H₂O₂, although present in the ovary in probably low concentrations (but being not compartmentalized), contributes quite strongly to the redox state of this organ. Iron, in turn, is an essential element for many metabolic processes and thus normally present in cells. Under normal conditions, iron is not toxic because it is bound to proteins (transferrin, ferritin) and, as such, it is not reactive. However, iron overload in certain disorders, such as polycystic ovary syndrome (PCOS) accompanied by obesity or endometriosis, can enhance oxidative stress in the ovary [Higashiura et al. 2012; Luque-Ramírez et al. 2007; Otsuki et al. 2012].

Bivalent iron (ferrous ion; Fe²⁺) and/or H₂O₂ − the substrates of the Fenton reaction − are frequently used, also by the authors of the present study [Karbownik-Lewinska et al. 2010, 2012; Karbownik-Lewinska et al. 2015; Mehta et al. 2011; Stepniak et al. 2013], to experimentally induce oxidative damage to macromolecules, membrane lipids, and DNA. The aim of this study was to evaluate the effects of 17β-estradiol on Fenton reaction-induced oxidative damage to membrane lipids (lipid peroxidation, LPO) and to nuclear DNA in the porcine ovary.

Results and Discussion

As shown in Figures 1A and 2B, 17β-estradiol did not affect the basal level of oxidative damage to either membrane lipids (Figure 1A) or DNA (Figure 2A) in the ovary. However, such results do not mean that 17β-estradiol is ineffective as an antioxidant and they do not rule out its potential to prevent oxidative changes due to the action of any pro-oxidative agent. This action may provide an additional advantage since oxidative processes occur with certain intensity in biological structures and they are indispensable for physiological processes, e.g., cell signaling. Thus, it is probably not required and, in consequence, not recommended to reduce auto-oxidation under physiological conditions. That 17β-estradiol does not change the basal level of oxidative damage to macromolecules makes this hormone an excellent antioxidant, the properties of which can be compared for example with the well-known antioxidant – melatonin [Karbownik et al. 2001a; Stasiak et al. 2010; Maganhin et al. 2013].

Figure 1. Concentrations of malondialdehyde + 4-hydroxyalkenals (MDA + 4-HDA) in porcine ovary. Homogenates were incubated in the presence of 17β-estradiol [1 mM, 100 μM, 10 μM, 1 μM, 100 nM, 10 nM, 1 nM, 100 pM, 10 pM, 1 pM] alone (A) or in the presence of 17β-estradiol together with Fenton reaction substrates, i.e., FeSO₄ [30 μM] plus H₂O₂ [0.5 mM] (B). Data are expressed as the amount of MDA + 4-HDA (nmol) per mg of protein. Bars represent the mean ± SE of three independent experiments run in duplicates. No significant differences were found. *p<0.05 vs. Control (in the absence of both Fe²⁺+H₂O₂ and 17β-estradiol); **p<0.05 vs. Fe²⁺+H₂O₂ (in the absence of 17β-estradiol).

Figure 2. Oxidative damage to nuclear DNA in porcine ovary. DNA was incubated in the presence of 17β-estradiol [1 mM, 1 μM, 1 nM, 100 pM, 10 pM, 1 pM] alone (A) or in the presence of 17β-estradiol together with Fenton reaction substrates, i.e., FeSO₄ [30 μM] plus H₂O₂ [0.5 mM] (B). Data are expressed as the ratio 8-oxodG/dG × 10⁵. Data are from three independent experiments. Values are expressed as mean ± SE (error bars). No significant differences were found. *p<0.05 vs. Control (in the absence of both Fe²⁺+H₂O2 and 17β-estradiol); **p<0.05 vs. Fe²⁺+H₂O₂ (in the absence of 17β-estradiol).

The protective effects of 17β-estradiol on the oxidative damage of membrane lipids and DNA caused by Fenton reaction substrates extends over a similar range of concentrations. With respect to membrane lipids, 17β-estradiol reduced the level of Fenton reaction-induced oxidative damage in a concentration-dependent manner, with statistical significance found for the following hormone concentrations: 1 mM, 100 μM, 10 μM, 1 μM, 100 nM, and 10 nM (Figures 1B). Similarly, 17β-estradiol reduced the Fenton reaction-induced oxidative DNA damage in a concentration-dependent manner, with statistical significance found for three highest concentrations of this hormone, i.e., 1 mM, 1 μM, and 1 nM (Figure 2B).

Fenton reaction substrates, especially iron, play a very important role in oxidative stress-related pathological processes occurring in the ovary. For example, it is documented that endometriosis – which is considered as a benign condition – is associated with an approximately three-fold increased risk of endometrioid and clear cell ovarian cancer [Kobayashi et al. 2011]. This endometriosis-associated ovarian cancer seems to be a distinct clinical entity, in which patients are younger, have lesions of a lower grade, and a better survival. It is postulated that this malignant transformation is the result of persistent iron-induced oxidative stress initiated by endometriosis-dependent retrograde menstruation and ovarian hemorrhage [Higashiura et al. 2012]. A significantly higher level of oxidative damage (presumably caused by the Fenton reaction product) was detected in endometriotic lesions than in normal endometrium, when mitochondrial DNA rearrangements and the level of oxidized guanine nucleosides and lipoperoxides were measured [Kao et al. 2005]. These results are in line with our earlier work in which Fe²⁺, when used alone, significantly increased the level of lipid peroxidation in porcine ovary exactly to the same extent as when the metal was used together with H₂O₂. Thus, exogenous H₂O₂ was not required to enhance lipid peroxidation in porcine ovary homogenates [Karbownik-Lewinska et al. 2010]. Another well known risk factor for ovary cancer is PCOS accompanied by obesity, which is also associated with increased body iron stores [Luque-Ramírez et al. 2007]. All of the above may stress a particular role of 17β-estradiol as a direct antioxidant, which favorably affects oxidative damage caused by the Fenton reaction substrate in the ovary.

Some evidence for protective effects of 17β-estradiol may be also drawn from studies in postmenopausal women, in whom the amount of 17β-estradiol produced by the ovaries drops significantly. This is considered as one of the mechanisms of the enhanced oxidative stress observed during this period of life [Sánchez-Rodríguez et al. 2012]. Additionally, the highest incidence of ovarian cancer has been observed in perimenopausal and postmenopausal women [Holschneider and Berek 2000] and may also, to a certain extent, be related to the increased level of oxidative damage. Antioxidative effects of 17β-estradiol administered as a hormone replacement therapy [Gómez and Mora 2013] has been suggested although this is not without controversy [Victorino et al. 2013]. Additionally, 17β-estradiol decreased the level of lipid peroxidation in ovariectomized rats [Oztekin et al. 2007].

Several in vitro studies have highlighted the antioxidative properties of 17β-estradiol. Examples include a reduction of experimentally-induced peroxide production in response to 17β-estradiol in isolated mitochondria [Borrás et al. 2010] as well as higher levels of superoxide dismutase and glutathione peroxidase (at both protein and gene levels) in mitochondria from females compared with those of males [Borrás et al. 2003]. The antioxidative effects of this hormone were also shown in aortic endothelial cells, in which oxidative damage induced by repeated H₂O₂-treatment was reduced by pretreatment with 17β-estradiol [Song et al. 2009]. In human skin fibroblasts, 17β-estradiol was found to counteract the detrimental effects of oxidative stress on the dermal compartment during skin aging [Bottai et al. 2013]. These authors suggested that the dramatic lowering of estrogens during menopause could render skin more susceptible to oxidative damage

In the present study the protective effects of 17β-estradiol against oxidative damage to both membrane lipids and to DNA damage induced by Fenton reaction substrates were observed. These protective effects were only observed when relatively high, although physiologically achievable, concentrations were used. The lowest concentration of 17β-estradiol showing antioxidative properties in our study was 10 nM in the case of LPO (Figure 1B) and 1 nM in the case of DNA (Figure 2B). For comparison, circulating estrogen peaks around 1 nM in preovulatory premenopausal women [Adlercreutz et al. 1986]. However, in postmenopausal women levels of estradiol are much lower, which can be partially responsible for the increased incidence of malignancies in this period of life. It should be noted that the contribution of 17β-estradiol to the antioxidative defense may seem negligible in light of the millimolar concentrations of glutathione, being the main intracellular antioxidant or the high micromolar concentrations of such well known antioxidants as Vitamin C and Vitamin E circulating in a healthy woman. Nonetheless, the ability of 17β-estradiol to contribute to antioxidative defense may be of particular importance in the place of its production, i.e., in the ovary.

The potential primary mechanisms of the protective effects of 17β-estradiol, observed in the present study, should be considered. The antioxidative effects of 17β-estradiol observed in the present study are receptor-independent and its non-genomic direct effect of this hormone. As the incubation time in the presence of 17β-estradiol and Fenton reaction substrates was rather short, i.e., 30 minutes for LPO and 60 minutes for DNA, the activation of antioxidative enzymes by 17β-estradiol should be rather negligible. Instead, it is assumed that the observed protective effects were due to scavenging hydroxyl radicals being a product of Fenton reaction. This ability of 17β-estradiol is most likely related to its unique chemical structure among steroids. Their chemical structure contains four cycloalkane rings, one of which (A-ring) is an aromatic ring, and a methyl group on the 13 carbon. Such a chemical structure is similar to that of simple phenolic antioxidants, such as vitamin E or butylated hydroxytoluene [Prokai-Tatrai et al. 2009a]. It has been shown that only the free phenolic hydroxyl group of the A-ring is the prerequisite for this activity [Moosman and Behl 1999]. Additionally, an important new antioxidant cycle for estrogens to shield against free radicals was discovered showing that estrogens can actually be regenerated after free radical scavenging [Prokai et al. 2003; Prokai-Tatrai et al. 2009b]. Proposed mechanisms of antioxidative activity of 17β-estradiol also include the synergistic interaction between this molecule and glutathione [Hum and Macrae 2000]. In addition, 17β-estradiol may also weakly function as a metal ion chelator, however these properties are rather associated with the presence of unsaturated B-ring of other estrogens.

The present work extends our previous studies that have also used Fenton reaction substrates to induce oxidative damage to macromolecules (e.g., [Karbownik-Lewinska et al. 2010; Karbownik-Lewinska et al. 2012; Karbownik-Lewinska et al. 2015; Stepniak et al. 2013; Kokoszko-Bilska et al. 2014]). For example, in one of these previous studies [Karbownik-Lewińska et al. 2012] we used Fenton reaction substrates to induce oxidative damage to mitochondrial DNA in the thyroid. The present work extends the previous in such a way that it confirms that Fenton reaction substrates may be used – in a similar range of concentrations – to induce oxidative damage to other macromolecules and in other tissues. It is of note that the thyroid examined in the former study, is characterized by substantial oxidative stress as it is required for thyroid hormone synthesis. Interestingly, under the same experimental conditions, the same concentration of Fenton reaction substrates may induce oxidative damage in both the thyroid and ovary. It is worth stressing, however, that as we clearly documented before [Karbownik-Lewinska et al. 2010; Stepniak et al. 2013] the distinct differences between ovary and the thyroid tissue concerning the susceptibility to Fenton reaction substrates used together or separately. Kokoszko-Bilska et al. [2014], evaluated the protective effects of caffeic acid phenethyl ester. We now show the effectiveness of an endogenous antioxidant that is even effective at physiological concentrations.

In conclusion, 17β-estradiol at concentrations close to physiological prevents experimentally induced oxidative damage to macromolecules in the ovary. This suggests that in the ovary and under physiological conditions this hormone may contribute to minimizing the negative effects of oxidative mechanisms.

Materials and Methods

The procedures, used in the study, were approved by the Ethics Committee of the Medical University of Lodz, Poland.

Chemicals

Ferrous sulfate (FeSO₄), hydrogen peroxide (H₂O₂), 17β-estradiol, alkaline phosphatase, and nuclease P₁ were purchased from Sigma (St. Louis, MO, USA). MilliQ-purified H₂O was used for preparing all solutions. All the used chemicals were of analytical grade and came from commercial sources.

Animals

Porcine ovaries were collected from twenty-one animals at a slaughter-house, frozen on solid CO₂, and stored at −l80°C until assay.

Method

Assay of lipid peroxidation. Ovary tissue was homogenized in ice cold 50 mM Tris-HCl buffer (pH = 7.4) (10%, w/v). Ovary homogenates were incubated for 30 min at 37 °C in the presence of 17β-estradiol (1 mM; 100 uM; 10 uM; 1 uM; 100 nM; 10 nM; 1 nM; 100 pM; 10 pM; 1 pM) alone (to check its effect on the basal LPO) or with the addition of Fenton reaction substrates [FeSO₄ (30 μM) + H₂O₂ (0.5 mM)].

The concentrations of FeSO₄ (30 μM) and H₂O₂ (0.5 mM) were chosen on the basis of the results of our previous study [Karbownik-Lewinska et al. 2010]. The reactions were stopped by cooling the samples on ice. Each experiment was repeated three times (with the use of seven ovaries for each homogenate pool) and was run in duplicate.

Measurement of lipid peroxidation products. Concentrations of malondialdehyde + 4-hydroxyalkenals (MDA + 4-HDA), as an index of LPO, were measured in tissue homogenates by the use of the ALDetect™ Lipid Peroxidation Assay Kit for LPO obtained from Enzo Life Science (Farmingdale, NY, USA), as described before [Milczarek et al. 2013]. The level of LPO was expressed as the amount of MDA + 4-HDA (nmol) per mg of protein. Protein was measured, using Bradford’s method [Bradford 1976], with bovine albumin as the standard.

Nuclear DNA isolation and incubation. Nuclear DNA was isolated and purified using a phenol extraction method [Shigenaga et al. 1994] with some modifications described by us before [Stepniak et al. 2013]. Nuclear DNA was incubated for 60 min at 37°C in the presence of 17β-estradiol (1 mM; 1 μM; 1 nM; 100 pM; 10pM; 1 pM) without or with addition of Fenton reaction substrates, i.e., FeSO₄ (30 μM) + H₂O₂ (0.5 mM), as described before [Stepniak et al. 2013]. Three independent experiments were performed, and in each experiment DNA was isolated from seven different ovaries.

Evaluation of the 8-oxo-7,8-dihydro-2′deoxyguanosine/2′-deoxyguanosine (8-oxodG/dG) ratio. The evaluation of the 8-oxo-7,8-dihydro-2′deoxyguanosine/2′-deoxyguanosine (8-oxodG/dG) ratio was as described before [Stepniak et al. 2013]. The results are expressed as the ratio of 8-oxodG to dG ×10⁵.

Statistical analyses

Results are expressed as means ± SE. Data were statistically analyzed, using a one-way analysis of variance (ANOVA), followed by the Student-Neuman-Keuls’ test. The level of p < 0.05 was accepted as statistically significant.

Declaration of interest

The research was supported by a grant from the Medical University of Lodz (Project No. 503/1-168-01/503-01). The authors report no conflicts of interest.

Author contribution

Designed the study and prepared the final version of the manuscript: JS, MK-L; Carried out the experiments and performed the statistical evaluation: JS; Supervised the conduction of the study: MK-L. Both authors read and approved the final manuscript.