Effect of curcumin on methotrexate-induced ovarian damage and follicle reserve in rats: the role of PARP-1 and P53

Abstract

Background

Methotrexate (MTX) is an agent used in the treatment of many neoplastic and non-neoplastic diseases and is known to cause oxidative damage in normal tissues. Curcumin (Cur) is a natural polyphenol compound with powerful antioxidant and antiapoptotic effects. In this study we investigate the effects of Cur on MTX-induced ovarian damage.

Materials and methods

Thirty-two young adult female Wistar albino rats were divided into four groups: (1) Control (n = 8): only vehicle group, (2) Cur (n = 8): Cur-only group (200 mg/kg/day), (3) MTX (n = 8): MTX-only group (0.35 mg/kg/day), (4) MTX+Cur (n = 8): The group was given MTX (0.35 mg/kg/day) and Cur (200 mg/kg/day) for 28 days. Then, SOD, CAT, MDA, AMH levels were measured using ELISA kits. Follicle count was performed on H&E stained slides. In addition, the expressions of P53 and PARP-1 were analysed by immunohistochemistry.

Results

MDA levels were seen to be higher in the MTX group than in the MTX+Cur group (p < 0.05). Cur treatment lowered MDA levels and increased SOD and CAT levels (p < 0.05 for all). In the MTX+Cur group, atretic follicle count decreased (p < 0,05), however, primordial follicle count increased (p < 0,01). Secondary follicle count and AMH levels were higher in MTX-treated groups (p < 0,05 and p < 0,01, respectively). Expressions of p53 and Poly [ADP-ribose] polymerase 1 (PARP-1) increased significantly in the MTX group compared to the other groups (p < 0,05).

Conclusion

Cur pretreatment prior to MTX administration may be an effective option in preserving the ovarian follicle pool by regulating P53 and PARP-1 expressions with its antioxidant effect.

KEY MESSAGES

• MTX has a negative effect on ovarian follicle reserve in rats

• Curcumin, with its antioxidant activity, prevents MTX-induced follicle loss through the p53 and PARP-1 mediated pathway.

• In the MTX ovarian follicle reserve model in rats, AMH may be insufficient to reflect follicle reserve.

Introduction

Mammalian ovaries are extremely dynamic organs in which most of the follicles are effectively removed during their reproductive life [1]. Follicular growth and atresia in mammalian ovaries are regulated by gonadotropins, growth factors, cytokines, and intracellular proteins [2]. In particular, it has been shown that granulosa cells play an important role in deciding the fate of follicles by producing molecules necessary for follicular growth and development and by killing themselves through an apoptotic process resulting in follicular atresia. Although follicular apoptosis is a natural process that determines the fate of ovarian follicles, women undergoing cancer chemotherapy experience negative effects on their ovarian follicle reserves, resulting in a significant decrease in pregnancy rates [3, 4].

Methotrexate (MTX) is a folic acid antagonist that blocks thymidylate and purine synthesis by suppressing dihydrofolate reductase enzyme activity. It shows that suppression of dihydrofolate reductase activity with increasing and repeated doses of MTX inhibits DNA and RNA synthesis and secondary protein production and is therefore lethal for cells in both G and S phases. It is known that the use of MTX has negative effects on ovarian function and follicle reserve. MTX can cause an increase in the level of intracellular reactive oxygen species (ROS) and histopathological changes in the ovaries due to mitochondrial dysfunction and depletion of antioxidants [5, 6].

Poly [ADP-ribose] polymerase 1 (PARP-1) is a nuclear DNA repair enzyme that is involved in the DNA damage response, creates the first response by detecting DNA strand breaks, and plays a role in maintaining genomic integrity [7, 8]. PARP-1 activation represents an important mechanism of tissue damage in various pathological conditions related to oxidative stress [9, 10]. P53 is a key transcription factor involved in cellular pathways such as suppression of tumorigenesis, repair of DNA damage, regulation of the cell cycle, apoptosis, angiogenesis, and aging [11, 12]. The main function of p53 tumor suppressor protein is to restrict abnormal or stressed cells before the damage in DNA is converted into heritable mutation. P53 is regulated by a post-translational regulatory mechanism as a result of induction of DNA damage. It is also a transcription factor that provides DNA repair by inducing G1 phase blockade [13].

Oxidative stress is the result of an imbalance between ROS production and removal via the antioxidant defense system [14]. At the cellular and tissue level, endogenous antioxidant enzymes are present to mitigate or prevent free radical toxicity (e.g. superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase, glutathione S-transferase, catalase (CAT), thioredoxin reductase, peroxyredoxins (Prx) and NAD(P)H: quinone oxidoreductase (NQO1)) [15,16]. Exceeding the capacity of the antioxidant defense system and the excessive presence of superoxide radicals in the environment lead to the formation of reactive oxygen species (ROS). Lipid peroxidation is the breakdown reaction of polyunsaturated fatty acids in mammalian cell membranes by free oxygen radicals to various products such as peroxides, alcohols, aldehydes, hydroxy fatty acids, ethane, pentane, malondialdehyde (MDA) [17]. With low ROS levels, p53 exhibits antioxidant activity to scavenge free radicals that maintain cell survival, but can induce cell death in response to high oxidative stress [18]. Recent studies have shown that there is excessive production of ROS and decreased antioxidant enzymes in some tissues are associated with the effects of MTX [19–21]. MTX increases the levels of malondialdehyde (MDA), the main product of lipid peroxidation. Lipid peroxidation causes decreased membrane permeability, decreased neurotransmitter-receptor interaction, and induction of cell apoptosis [22, 23].

Curcumin (Cur) is a natural polyphenol compound produced from the rhizomes of the Curcuma longa plant. Recent studies have shown that Cur has anticancer, antiviral, antiarthritic, and antiamyloid properties as well as strong antioxidant and anti-inflammatory effects [24–26]. With its potent antioxidant properties, Curcumin is capable of binding to reactive oxygen species, thus inhibiting peroxidation and degradation of biomolecules such as nucleic acids, peptides, and lipids. This effect can prevent the onset of apoptosis, senescence, and malignant transformation in healthy cells [27]. It has been reported that the molecular mechanism of these therapeutic effects of Cur is mediated by transcription factors (mainly nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB)), endothelial cell growth factors, inflammatory cytokines and protein kinases [28, 29]. Experimental animal studies have shown that Cur prevents oxidative stress damage by inhibiting superoxide radicals, hydrogen peroxide and nitric oxide radicals, and consequently reduces proapoptotic proteins and P53-related gene product expressions [24, 30]. Notably, it is well known that while Cur induces apoptosis in cancer cells through p53 and PARP-mediated mechanisms, it has a protective effect against oxidative damage in normal cells through the same proteins [31, 32]. These data suggests that the use of Cur in patients receiving chemotherapy may protect normal cells while enhancing the effects of the chemotherapeutic agent.

In the light of the aforementioned data, the starting point of our study is that the negative effects of MTX-induced oxidative stress on ovarian folliculogenesis can be prevented by Cur, a molecule with undeniable antioxidant activity. The study aims to investigate the protective impact of Cur on the ovarian follicle reserve of rats subjected to MTX.

Materials and methods

Animals and ethics

In our study, female Wistar albino rats (n = 32, weighing 250–300 g each, 8 weeks old) produced by Bülent Ecevit University Animal Care and Research Unit (Zonguldak, Turkey) were used. During the course of the experiment, the requisite environmental conditions for the care of the animals were provided (room temperature of 20 ± 1 °C, humidity of 60 ± 10%, and a 12/12 h light/dark cycle). The subjects were permitted free access to food and water. Throughout the experiment, the Guide for the Care and Use of Laboratory Animals published by the US Public Health Service was followed. All sections of this report adhere to the ARRIVE Guidelines for reporting animal research [33]. A completed ARRIVE guidelines checklist is included in Checklist S1.

Determination of estrus cycle

To determine the estrous cycle of the rats, vaginal smears were taken daily for 10 consecutive days [34]. The morphology of epithelial cells obtained from the smears was examined using a light microscope and photographed, too. (Zeiss Axio Lab. A1, Germany) (Figure 1A–D).

Figure 1.

H&E staining results of vaginal smears collected from subjects to determine the stage of the oestrous cycle. (A) Proestrus; (B) Estrus; (C) Metestrus; (D) Diestrus. Scale bar: 50 μm.

Experimental design

The rats were randomly divided into four groups after determining the estrus stage for each animal using daily vaginal swabs including 1 control group and 3 experimental groups, with 8 rats in each group, namely;

I. Control (n = 8): Cur solvent, olive oil (1 ml) was given by intragastric gavage and MTX solvent Phosphate-Buffered Saline (PBS) was given intraperitoneally (i.p.) daily for 28 days.

II. Cur (n = 8): Cur (Sigma Aldrich, St. Louis, Missouri, USA) was dissolved in olive oil and administered by intragastric gavage at 200 mg/kg/day for 28 days [28].

III. MTX (n = 8): MTX was dissolved in PBS and administered i.p. at a dose of 0.35 mg/kg/day for 28 days [28].

IV. MTX + Cur (n = 8): MTX was dissolved in PBS and injected i.p. at a dose of 0.35 mg/kg/day for 28 days. Cur was also dissolved in olive oil and administered by intragastric tube at 200 mg/kg/day for 28 days.

A total of 8 cages containing the rats included in the experiment (2 cages for each group and 4 rats in each cage) were kept in an independent room with the appropriate physical features listed above throughout the experiment. All cages were numbered. All subjects were given a protocol number and these numbers were written on the subjects’ tails. Moreover, all tissue and serum samples obtained from the subjects were immediately labelled, and protocol numbers were written on the labels and stored in separate bags in groups.

The sample size was calculated by utilizing previous similar studies and using the G*Power program. Appropriate randomization methods should be used when allocating to groups to ensure that each experimental unit has an equal probability of receiving a particular treatment. For this purpose, the online GraphPad application was used.

Once the investigation was complete, the rats were weighed and subsequently euthanised under general anaesthesia. General anaesthesia was achieved via intraperitoneal injection of xylazine (10 mg/kg, Rompun, Bayer) and ketamine (90 mg/kg, Ketalar, Eczacıbaşı). The ovarian tissue samples were excised, weighed, and subsequently placed in a 10% formalin solution. The flowchart was described in Figure 2.

Figure 2.

Flow chart of the study.

The Wistar Albino rats were randomly divided into four groups after the estrus stage of each animal was determined using daily vaginal swabs. The groups were as follows: control, curcumin (Cur), methotrexate (MTX) and MTX+Cur (n = 8 for each group). At the end of the study, in the estrus phase of the sexual cycle, rats were euthanized under general anesthesia with intraperitoneal injection of xylazine (10 mg/kg, Rompun, Bayer) and ketamine (90 mg/kg, Ketalar, Eczacıbaşı). The levels of superoxide dismutase (SOD), catalase (CAT), malondialdehyde (MDA) and anti-müllerian hormone (AMH) were quantified in right ovarian tissues excised from the subjects using the ELISA method. The left ovarian tissues were subjected to histopathological analysis using the haematoxylin and eosin (H&E) and periodic acid-Schiff (PAS) staining methods. The protein levels of poly ADP-ribose polymerase-1 (PARP-1) and p53 in ovarian tissue were determined using the immunohistochemical method.

Histopathological evaluations

Paraffin blocks were obtained from the left ovarian tissues following overnight formalin fixation for subsequent light microscopic examination. The blocks were sectioned at a thickness of 5 microns. The sections were deparaffinised and stained with H&E (Harris’ haematoxylin; Merck, Darmstadt, Germany and eosin Y; Sigma-Aldrich, St. Louis, Missouri, USA) and PAS (periodic acid and basic fuchsin; Merck, Darmstadt, Germany) stains in the objective of revealing the general histological characteristics of the ovarian tissue. All staining was performed as described by Suvarna et al. [35]. Slides were evaluated and photographed using a light microscope (Carl Zeiss Axio Lab A1, Jena, Germany).

Follicle counting

Serial sections of 5 µm thickness were obtained from the prepared paraffin blocks using a Shandon Finesse 325 cylindrical microtome, continuing until the entire block was sectioned. Every fifth section throughout the entire ovary was stained with H&E, and only the follicles displaying a clearly visible oocyte nucleolus were included in the count, which was performed using the blinded method [36]. All sections were prepared by the same author (NK) and were blinded to the reader (MK). The criteria of Devine et al. were used to classify ovarian follicles [37]. Furthermore, the criteria defined by Wang et al. were used to define atretic follicles [38]. The number of follicles per ovary was estimated by multiplying the total number of follicles in each ovary by a correction factor of five and dividing the result by the total number of sections examined for each ovary, as previously described [39]. Follicle counting was performed by two histologists. Both histologists were unaware of the groups during follicle counting.

Evaluation of AMH, SOD, CAT and MDA levels

Serum samples were obtained by centrifuging the blood samples taken from the subjects during the sacrifice. Furthermore, the right ovarian tissues of the rats were also subjected to biochemical analysis. The AMH level in serum and the SOD, CAT and MDA levels in tissue were quantified using ELISA kits, in accordance with the protocol provided by the manufacturer. (AMH; Elabscience, Houston, TX, USA, SOD, CAT and MDA; Rel Assay Diagnostics, Gaziantep, Turkey). Both intra-coefficients of variation (CV) and inter-CV were determined as <10% by manufacturer. Tissue and serum samples were stored at −80 °C until analysis.

Immunohistochemical evaluations

The expression of PARP-1 and p53 were demonstrated by immunohistochemical method. Five micrometre thick sections obtained from paraffin blocks were mounted on positively charged slides for subsequent analysis. Following deparaffinisation and immersion, the sections in citrate buffer were maintained at a temperature just below the boiling point for 15 min in a microwave oven, thereby facilitating the unmasking of the antigenic binding sites. Subsequently, the sections were immersed in distilled water and permitted to cool at room temperature for a period of 20 min. Following three washes with PBS, the sections were treated for a period of ten minutes with 3% hydrogen peroxide, with the objective of eliminating any endogenous peroxidase activity. The application of Ultra V block (Thermo Fisher, Massachusetts, USA) for a period of seven minutes was undertaken in order to block non-specific binding sites. Following this, the sections were treated with anti-P53 (1:250 dilution, rabbit polyclonal IgG, Abcam, Cambridge, UK) and anti-PARP-1 (1:250 dilution, rabbit polyclonal IgG, Abcam, Cambridge, UK) primary antibodies for a duration of 24 h at a temperature of +4 °C. The secondary antibody (Thermo Fisher, Massachusetts, USA) and streptavidin peroxidase (Thermo Fisher, Massachusetts, USA) were then applied for 30 and 10 min, respectively, at room temperature. The final stage of the procedure involved the application of a diaminobenzidine (DAB) chromogen solution (Vector Laboratories, Newark, USA) until the desired staining intensity was achieved under microscopic examination. Subsequently, the sections were counterstained with haematoxylin [36]. Throughout the incubation periods, the tissue samples were maintained in a humid environment to inhibit drying and prevent the formation of background stains. Sections closed using entellan and coverslip were examined under a light microscope. The photographic documentation of findings was conducted using the Zeiss Axio Lab. A1 brand photo-microscope. All immunohistochemical analyses were conducted in accordance with the protocols recommended by the manufacturer.

A histological scoring system (H score) was employed to define the immunohistochemical results, utilising the following criteria: 0, no staining; 1+, weak but identifiable staining; 2+, medium or pronounced staining; 3+, intense staining. The H score value for each section was calculated by multiplying the percentage of stained cells within each density category by the corresponding density. The scoring was conducted manually on 20 randomly selected areas in each section at x40 objective magnification under a light microscope. The mean scores were utilized for the statistical analysis. H-score = ∑i i xPi, i; density score, Pi; cell percentage [40].

In addition, a ‘secondary antibody only control’ staining was performed to verify that the secondary antibody did not bind non-specifically to cellular components and that the staining results were specific. To this end, rat ovarian tissue was treated with antibody diluent only at the primary antibody stage. Further steps of immunohistochemical staining were performed as described above.

Statistical analysis

The statistical evaluation was conducted using the Jamovi Desktop 2.3.21 software. The data were subjected to a Kolmogorov-Smirnov test to ascertain their normality. Descriptive statistics for data that were normally distributed were expressed as mean ± standard deviation, while other data were expressed as median (minimum-maximum). Parametric data were subjected to one-way ANOVA and post hoc Tukey tests. The Kruskal-Wallis test was employed for the analysis of non-parametric data. Subgroup comparisons were conducted using the Dunn test in accordance with the Kruskal-Wallis analysis of variance. A value of p < 0.05 was considered statistically significant.

Results

On the last day of the experiment, relative ovarian weights were calculated using each subject’s body weight (gr) and right and left ovarian weights (mg) (ovarian weight/last measured body weightX100). There was no significant difference between the groups in the comparison of body weights of animals (p > 0,05) and ovarian weights per 100 g animal weight (p > 0,05) (Figure 3).

Figure 3.

Relative right and left ovarian weights.

Values are given as mean ± s.d. (n = 8/group). There was no statistically significant difference between the groups in terms of both right and left ovarian weights (p > 0.05 both).

Histopathological findings

Follicle cells surrounding the primordial follicles in the control and Cur groups appeared to be normal with elliptical and euchromatic nuclei. Moreover, there were many developing follicles with normal-looking oocyte and granulosa cells in these groups (Figure 4A–F).

Figure 4.

H&E staining results of all groups.

Control (A, B, C) and curcumin (D, E, F) groups seem to preserve their normal histological architecture. In the MTX (G, H, I) group, numerous normal-looking secondary follicles (black arrowhead), a normal-looking preantral follicle (black thin arrow), a primordial follicle with condensed and spindle-shaped nucleated follicle cells indicating early-stage atresia (red arrowhead), just below the germinal epithelium numerous atretic primordial follicles (double-headed arrow). A normal-looking primary and secondary follicle in the MTX+Cur (J, K, L) group (hollow black arrow and solid black arrow, respectively). The MTX+Cur group shows normal-looking primordial follicles (hollow star) and a normal-looking preantral follicle (filled red arrow) at high magnification. Scale bars; A, D, G, J; 200 µm, B, E, H, K; 50 µm, C, F, I, L; 20 µm.

However, atretic follicles were observed from all follicular developmental stages, especially primordial follicles in the MTX group. Many primordial follicles containing follicle cells with pycnotic nuclei and spindle-like appearance were suggestive of early stage atresia. It was also observed that some primordial follicles exhibited signs of advanced atresia such as vacuolization in the oocyte cytoplasm and loss of oocyte nuclei. Larger atretic follicles could be easily distinguished by the loosening of the connections between the granulosa cells and their separation from each other, the observation of apoptotic bodies in the granulosa cell layer, the arrangement of apoptotic bodies along the antrum border in some places, and the spread to the entire antrum in some atretic follicles. Vacuolization in the oocyte cytoplasm in some atratic follicles was also remarkable. However, this group had a large number of normal-appearing antral follicles. The germinal epithelium was intact, similar to the control and Cur groups, and no edema or hemorrhage was observed in the stroma. A large number of adipocytes was also noticed in the cortical stroma in the MTX group (Figure 4G–I).

Primordial follicles in the MTX+Cur group contained normal-appearing oocytes, surrounded by elliptical and euchromatic nucleated follicle cells, similar to control and Cur groups. Follicles at other developmental stages mostly had normal follicle morphology. As in the MTX group, there were many secondary follicles with intact granulosa layer and normal-looking oocytes and no signs of atresia in this group (Figure 4J–L).

Follicle counts of each group are given in Figure 5. Primordial and primary follicle numbers showed a significant decrease in the MTX group compared to the other groups (p > 0.01 for all comparisons). Secondary follicle numbers showed a significant increase in the MTX-treated groups compared to the control and Cur groups (p > 0.05 for all comparisons). No significant difference was found between groups in terms of tertiary follicle numbers (p > 0,05 for all comparisons). The number of atretic follicles in the MTX group was found to be higher than the control, Cur and MTX+Cur groups (p < 0,01, p < 0,05 and p < 0,01, respectively).

Figure 5.

Follicle count comparison.

Values are given as mean ± s.d. (n = 8/group). *Different from control, MTX and MTX+Cur groups (p < 0,01), **Different from MTX+Cur group (p < 0,05), ***Different from Control and Cur groups (p < 0,01 and p < 0,05, respectively), ****Different from Control and Cur groups (p < 0,05 both), *****Different from control, cur and MTX+Cur groups (p < 0,01, p < 0,05 and p < 0,01, respectively).

In order to evaluate the integrity of the zona pellucida, PAS staining method was applied to the ovarian sections of all groups. It was observed that the zona pellucida preserved its integrity in the control and Cur groups (Figure 6A,B). It was noted that in many atretic follicles in the ovaries of the MTX group (Figure 6C), the ZP thickness lost its homogeneity, became very thin in some areas, and floated freely in the antrum following a wavy course. In the MTX+Cur group, it was observed that ZP exhibited homogeneous thickness, similar to the control and Cur groups, and followed a smooth course between the basement membrane and the oocyte (Figure 6D).

Figure 6.

PAS staining results of all groups.

The zona pellucida (hollow black arrow) appeared normal in the control (A) and curcumin (B) groups. In the MTX group, it was observed that ZP lost its relationship with the granulosa cells and oocyte, became wavy, was released in the antrum, and had heterogeneous thickness (C). Similar to the control and curcumin groups, in the MTX+Cur group, ZP seemed to maintain a tight relationship with the granulosa cells and oocyte and had a homogeneous thickness. (D). Scale bars; 50 µm.

Oxidative stress parameters and antioxidant status

Tissue MDA levels increased significantly in the MTX group compared to the Cur and MTX+Cur groups (p < 0,05 both). Although MDA levels were lower in the Cur-treated groups compared to the control group, the difference was not statistically significant (p > 0.05 both). SOD levels of MTX group decreased significantly compared to Cur and MTX+Cur groups (p < 0,01 and p < 0,05, respectively). In the comparison of CAT levels between groups, the MTX+Cur group showed a significant increase compared to the control, Cur and MTX groups (p < 0,05 each) (Table 1).

Table 1.

Effect of Cur on SOD, CAT and MDA levels.

	Control (n = 8)	Cur (n = 8)	MTX (n = 8)	MTX+Cur (n = 8)
MDA	14 ± 3.51	7.87 ± 2.2	19.66 ± 3.62B,C	8.5 ± 1.4
SOD	212.85 ± 21	326 ± 40	109.34 ± 13.4D,E	273.28 ± 33.08
CAT	199.14 ± 17.3	255 ± 28	278.71 ± 36.23	420.29 ± 51A,B

Values are given as mean ± s.d. (n = 8/group). ^Acompared to Control group (p < 0.05), ^Bcompared to Cur group (p < 0.05), ^Ccompared to MTX+Cur group (p < 0.05), ^Dcompared to Cur group (p < 0.01), ^Ecompared to MTX+Cur group (p < 0.05), MDA; malondialdehyde, SOD; superoxide dismutase, CAT; catalase.

Effect of Cur on serum AMH levels

Due to AMH measurements, one extremely low value in the control and curcumin groups, and one extremely high value in the MTX and MTX+Cur groups were excluded from evaluation as they disrupted the statistically normal distribution. When serum AMH levels of the subjects were compared, it was found that AMH levels in the MTX-treated groups showed a significant increase compared to the control and Cur groups (p < 0,01 for all comparisons) (Figure 7).

Figure 7.

Serum AMH levels in each group.

Values are given as mean ± s.d. (n = 7/group). *Different from control and Cur groups (p < 0,01 for all comparisons). AMH; anti-mullerian hormone.

Immunohistochemical findings

In our study, p53 (Figure 8A–D) and PARP-1 (Figure 9A–D) protein levels were evaluated immunohistochemically.

Figure 8.

Immunohistochemical expressions of p53 in all groups.

Weak p53 expression was observed in the control (A) and curcumin (B) groups. Strong p53 expression is observed in the MTX group (C). Similar to the control and curcumin groups, weak p53 expression was noted in the MTX+Cur group (D). Secondary antibody only control (E). Scale bars; A, B, C, D: 20 µm, E: 100 µm.

Figure 9.

Immunohistochemical expressions of PARP-1 in all groups.

Weak PARP-1 expression was observed in the control (A) and curcumin (B) groups. Strong PARP-1 expression is observed in the MTX group (C). Similar to the control and curcumin groups, weak PARP-1 expression was noted in the MTX+Cur group (D). Secondary antibody only control (E). Scale bars; A, B, C, D: 20 µm, E: 100 µm.

In the statistical analysis of the H-score data obtained from the staining performed to determine the p53 and PARP-1 immunoreactivity, a significant difference was found between the MTX group and the Control, Cur and MTX+Cur groups in terms of P53 and PARP-1 protein levels (p < 0,05 both) (Table 2).

Table 2.

H-score values of p53 and PARP-1 expressions in each group.

	Control (n = 8)	Cur (n = 8)	MTX (n = 8)	MTX+Cur (n = 8)
p53	0.5 (0.2–0.53)	0.25 (0.1–0.83)	2.17 (1.6–2.85) A,B,C	0.55 (0.07–1.2)
PARP-1	0.45 (0.1–0.99)	0.08 (0–0.9)	2.55 (2.44–2.72)A,B,C	0.28 (0.02–0.92)

Values are given as median (min-max). (n = 8/group). ^Acompared to the Control group (p < 0.05), ^Bcompared to the Cur group (p < 0.05), ^Ccompared to the MTX+Cur group (p < 0.05). PARP-1; poly ADP-ribose polymerase-1.

No staining was detected in the ovarian tissue after the ‘secondary antibody only control’ application (Figure 8E,9E).

Discussion

In this study, we investigated the impact of Cur on ovarian tissue damage and the impairment of folliculogenesis as a result of MTX exposure to rats. For this purpose, we performed follicle count to evaluate folliculogenesis, measurement of SOD, MDA and CAT enzyme levels to evaluate changes in oxidant-antioxidant balance, and immunohistochemical staining to show changes in PARP-1 and P53 expression. We also assessed hormonal changes by measuring serum AMH levels, which is considered an indicator of follicular reserve.

Methotrexate is a cytotoxic agent frequently used in the treatment of various autoimmune diseases and malignancies. The cytotoxic effect of methotrexate is a leading cause of patient non-compliance and treatment failure [41, 42]. There are many studies that link the curative effect of MTX to changing the oxidant-antioxidant balance in favor of oxidants. The increase in peroxide levels in neutrophils as a result of MTX exposure has been shown to be accompanied with a decrease in glutathione levels at the cellular level [43]. In addition, studies with human T-cell leukemia cell lines and human monocyte cell lines (U937) have shown that when MTX is added to the medium, T cells and monocytes are directed to apoptosis, whereas antioxidant molecules such as N-acetyl cysteine and glutathione suppress apoptotic pathways and increase viability [44].

Numerous studies have shown that increased oxidants in the ovarian microenvironment can trigger pathologies such as oocyte meiosis arrest, granulosa cell apoptosis, and corpus luteum dysfunction. It has been reported that while MTX increases MDA levels in the ovary, it decreases SOD and GSH levels, so it is a cause of oxidative stress for the ovarian tissue, and the antioxidant resveratrol can protect the ovarian tissue against this oxidative stress [45]. Coenzyme Q10 (Co Q10) is one of the key molecules of the electron transport chain and is also a powerful antioxidant. Coenzyme Q10 administration with MTX to Wistar albino rats decreased MDA and Myeloperoxidase levels while it increased glutathione levels in ovarian ­tissue [46].

In our study, MDA levels in the MTX+Cur group showed a significant decrease compared to the MTX group, while SOD and CAT levels were found to be significantly higher. In terms of oxidant-antioxidant parameters, the results of our study are similar to the literature listed above.

B-cell lymphoma 2 (Bcl-2) is a protein that has an antiapoptotic effect by maintaining the mitochondrial membrane potential. It has been reported that the increase in Bcl-2 expressions in S49.1 and WEH17.2 murine lymphoid cell lines creates resistance to MTX, 1-β-D-arabinofuranosylcytosine and Vincristine-mediated apoptosis [47]. In a study conducted in rat intestinal cell line (RIE-1), it was reported that there had been a significant decrease in the number of cells leading to apoptosis and caspase 2, 3 and 9 expressions in the presence of Z-VAD fluoromethyl ketone (a caspase inhibitor) added to the culture medium with MTX [48].

The PARP enzyme is encoded by the PARP gene in humans. PARP-1 is linked to various events including preservation of chromatin structure and transcription, chromosomal organisation, DNA methylation and imprinting. PARP-1 also plays a key role in the cellular response to oxidative stress and in the repair of DNA molecules damaged by free radicals. The increase in PARP-1 activity is a very important indicator of DNA damage. Activation of PARP-1 in response to mild DNA damage results in chromatin strand exposure and initiation of DNA repair through the Poly ADP ribosylation of histone proteins. However, excessive PARP-1 activation can also lead to cell death by depletion of energetics [49, 50]. It shows that the increase of mitochondria-induced ROS in T. Cruzi-infected cardiomyocytes increases PARP-1 activity, and PARP-1 triggers inflammation by subsequently activating NF-κB [51]. In addition, it has been suggested that PARP-1 activation is increased in apoptotic granulosa cells in pig ovaries, and consequently caspase-3-mediated PARP-1 activation may be the key mechanism in granulosa cell apoptosis [47].

In our study, PARP-1 expression in the treatment group showed a significant decrease compared to the MTX group. This finding is interpreted as Cur, with its strong antioxidant activity, reduces PARP-1 activation by preventing ROS-induced DNA damage.

DNA can be damaged by a variety of exogenous and endogenous effects, each of which causes different forms of damage, including chemicals, radiation, free radicals, and chemotherapeutics [52]. Reactive oxygen species (ROS) are a group of highly reactive oxygen-containing molecules that can induce DNA damage and affect the DNA damage response [53]. ROS accumulation also leads to strand breaks in mitochondrial DNA and causes degradation of mitochondrial DNA [54]. DNA damage activates mechanisms such as cell cycle arrest, activation of DNA repair mechanisms, and apoptotic cell death when the damage is too severe and irreparable. There are many studies showing that p53 is the key molecule in the center of these mechanisms triggered after DNA damage. While the level of p53 protein is undetectable in normal cells, it increases rapidly in response to various stress signals that cause genomic damage [55, 56]. MTX is a competitive inhibitor of dihydrofolate reductase, which causes disruption of various metabolic pathways including nucleic acid synthesis and can therefore lead to induction of apoptosis and cell death. It has shown that MTX-dependent ROS production is the main mechanism inducing apoptosis by increasing p53 and p21 expression [57, 58].

In our study, we found that p53 expressions were significantly increased in the MTX group compared to the control, Cur and MTX+Cur groups. This finding is interpreted as that MTX induces apoptosis by increasing p53 activation through ROS-induced oxidative damage and is consistent with previous studies.

Numerous studies have proved the negative effects of MTX exposure on folliculogenesis and ovarian reserve in mouse and rat ovaries. It has shown that MTX has a dose-dependent toxic effect in all follicular development stages, especially primordial follicles, and increases the number of atretic follicles, ultimately affecting the ovarian reserve negatively[59–61].

Several studies have also studied the effects of MTX exposure on AMH levels in rats [57, 61, 62]. In contrast, Benian et al. showed that exposure to MTX increases AMH levels [63].

As a result of our study, both primordial and primary follicle numbers were found to be higher in the MTX+Cur group compared to the MTX group. Furthermore, the number of atretic follicles was significantly higher in the MTX group than in the MTX+Cur group. Another striking finding of our study in terms of follicle numbers is that the number of secondary follicles in the MTX applied groups has been higher than the control and Cur groups.

When evaluated in terms of AMH levels, the results of our study were interesting. Parallel to the increase in the number of secondary follicles, AMH levels were observed to be higher in the MTX-treated groups compared to the control and Cur groups.

AMH is known to be released from granulosa cells of ovarian follicles, especially in preantral and antral follicle stages. In other follicular stages, AMH expression is either very low or completely lost [64]. Ovarian steroidogenesis begins with the conversion of cholesterol to androgens in theca interna cells and is completed by the conversion of these androgens to estrogens in the granulosa cells [65]. Mitochondria is a vital organelle for steroid hormone biosynthesis. Mitochondria in the adrenal glands, gonads, placenta, and steroidogenic cells of the brain contain the cholesterol side chain cleavage enzyme (P450scc) and its two electron transfer partners, ferredoxin reductase and ferredoxin. This enzyme system converts cholesterol into pregnenolone and determines the net steroidogenic capacity [66]. Mitochondria form high-energy phosphate bonds of ATP with the electrochemical proton gradient created by transferring electrons through a series of electron carriers embedded in the mitochondrial membrane. The transfer of electrons to molecular O₂ is a tightly controlled process, in which only 1–2% of the electrons escaping from the electron transport chain (ETC) react with O₂ to form ${\mathrm{O}}_2^{-}$. Thus, mitochondrial ETC itself is a source of ROS [67].

Manganese superoxide dismutase (MnSOD) localized in mitochondria is one of the major ROS detoxification enzymes. No threshold value has been defined for low and high levels of ROS in tissue. However, it has been observed that low ROS levels increase NF-κB transcription via tumor necrosis factor alpha (TNF-α) and TNF Receptor-1 (TNFR1), which triggers MnSOD expression, whereas high ROS levels direct cells to apoptosis via the c-Jun N-terminal kinase (JNK) pathway [68].

Conclusion

In light of the results of our study and the literature presented above, it can be said that secondary follicles have higher granulosa cell load, thus steroidogenesis activity and number of mitochondria compared to other follicular stages. The high number of secondary follicles in MTX-treated groups in our study may reflect the presence of outnumbered mitochondria and therefore intensive MnSOD activity in the granulosa cells of secondary follicles. The interpretation of this is the fact that MnSOD is triggered by an increase in ROS which may have put secondary follicles in an advantageous position against MTX-induced ROS formation (Graphical abstract). In determining the results of oxidative stress, endogenous antioxidant enzyme activity is as important as ROS level. The results of our study show that AMH may not be a reliable parameter in demonstrating follicular reserve in the MTX-induced ovarian damage model, considering the AMH levels and secondary follicle numbers obtained in the MTX groups. Further in vivo and in vitro studies with MTX (or other oxidant agents) applied with different doses and durations in tissues with intense steroidogenetic activity such as ovary, testis and adrenal gland will contribute to the demonstration of the response pattern against oxidative stress in these tissues.

Glossary

Abbreviations

MTX

methotrexate

Cur

curcumin

SOD

superoxide dismutase

CAT

catalase

MDA

malondialdehyde

AMH

antimullerian hormone

PARP-1

poly [ADP-ribose] polymerase 1

ROS

reactive oxygen species

GPx

glutathione peroxidase

Prx

peroxyredoxins

NQO1

NAD(P)H: quinone oxidoreductase

NF-κB

nuclear factor kappa-light-chain-enhancer of activated B cells

Co Q10

coenzyme Q10

Bcl-2

B-cell lymphoma 2

P450scc

cholesterol side chain cleavage enzyme

ETC

electron transport chain

MnSOD

manganese superoxide dismutase

TNF-α

tumor necrosis factor alpha

TNFR1

TNF receptor-1

JNK

c-Jun N-terminal kinase

Funding Statement

The present study was supported by the Scientific Research Projects Coordinatorship of Bülent Ecevit University (2021-98210206-01).

Ethical statement

Ethical approval was obtained from the Institutional Animal Ethics Committee of Bülent Ecevit University prior to the study (Zonguldak, Turkey; 2020-22-03/09).

Authors contributions

Conceived and designed the experiments: MK, NK, Performed the experiments: MK, NK. Analyzed the data: MK. Contributed reagents/materials/analysis tools: MK. Contributed to the writing of the manuscript: MK, NK. All authors have read and approved the manuscript and agree to be responsible for all aspects of the work.

Disclosure statement

The authors have nothing to disclose and declare that there are no conflicts of interest.

Data availability statement

The data supporting the findings of this study are available for viewing within the article’s figures and tables. Upon request to the corresponding author, the data will be made available in spreadsheet form.