Naproxen at environmentally relevant doses modulates ovarian folliculogenesis: Morphometric analysis and Bcl-2 immunohistochemical expression in Wistar rats after chronic exposure
Article information
Abstract
Objective
Naproxen has received considerable attention due to its environmental persistence and incomplete removal during wastewater treatment. This study evaluated the ovarian toxicity of naproxen at an environmentally relevant dose to assess its potential reproductive risks.
Methods
Female Wistar albino rats were exposed for 100 days to naproxen alone (98.39 µg/L) or to a mixture containing naproxen (98.39 µg/L), ibuprofen (117.14 µg/L), diclofenac (609.12 µg/L), and trifloxystrobin (0.04 mg/kg/day) at a 4:1:3:2 ratio. Ovarian tissues were analyzed for oxidative stress biomarkers, B-cell lymphoma 2 (Bcl-2) expression, and follicle counts. Serum was analyzed for interleukin 6 (IL-6) and progesterone concentrations.
Results
Compared with controls, both treated groups had a significantly higher ovary index, lower catalase activity, and lower progesterone levels, whereas body weight, malondialdehyde levels, and serum IL-6 levels did not differ significantly. Morphometric and immunohistochemical analyses demonstrated reduced preantral follicle counts and Bcl-2 expression in the treated groups.
Conclusion
Chronic exposure to naproxen, alone or in combination, at an environmentally relevant dose was associated with ovarian dysfunction, impaired progesterone production and antioxidant status, and disrupted ovarian follicle dynamics.
Introduction
Non-steroidal anti-inflammatory drugs (NSAIDs) are widely used worldwide. These substances are commonly prescribed as analgesics in human medicine and are also frequently available without a prescription as over-the-counter medications. Because of their extensive use, NSAIDs have become a major class of environmental contaminants. Their widespread environmental occurrence has raised concern about potential risks to human health [1]. Accordingly, NSAIDs are classified as emerging contaminants. These compounds are usually detected in the environment at parts-per-trillion levels (ng/L or ng/kg), far below therapeutic doses and typical exposure levels [2].
Naproxen sodium, an NSAID derived from propionic acid, is primarily prescribed for osteoarthritis and rheumatoid arthritis and is also used as an antipyretic and analgesic agent. It is among the most frequently detected pharmaceuticals in environmental matrices, including aquatic systems and agroecosystems, along with other compounds such as carbamazepine, sulfamethoxazole, ibuprofen, gemfibrozil, atenolol, diclofenac, erythromycin, and bezafibrate. Because of incomplete degradation, naproxen occurs in groundwater, surface water, and drinking water, posing a potential threat to aquatic organisms through chronic exposure and to humans who consume these organisms [3]. Naproxen is poorly metabolized by humans, and wastewater treatment plants lack microbial strains capable of degrading it efficiently. In addition, naproxen sodium readily undergoes photolysis, and its phototransformation products are often more toxic than the parent compound. As a result, both naproxen and its transformation products are released into the environment [4]. In aquatic organisms, exposure to naproxen at environmental concentrations has been associated with altered enzymatic activity and gene expression, histopathological changes, cellular damage, and reproductive abnormalities. Naproxen can also affect mRNA expression and cause gastrointestinal and renal effects in zebrafish. Beyond aquatic ecosystems, naproxen may pose risks to human health, including DNA damage, congenital abnormalities, gastric ulcers, chronic depressive disorders, respiratory impairment, and cardiovascular and renal diseases [5].
The combined presence of pharmaceuticals in the environment highlights the need to assess their interactive effects. Cleuvers [6] demonstrated that predicting the toxicity of pharmaceutical mixtures is critical because these substances rarely occur in isolation. Studies have shown that naproxen interacts with other drugs, such as carbamazepine and sulfamethoxazole, and may affect amphibian growth and development at environmental concentrations [7].
Given the limited toxicological evidence on the ovarian toxicity of naproxen sodium at environmentally relevant doses, this study aimed to elucidate how long-term exposure to this chemical, individually or in a realistic mixture, may disrupt ovarian function and integrity in female Wistar rats.
Methods
1. Chemicals
Commercial water-dispersible granules of trifloxystrobin (50% w/w; CAS No. 141517-21-7) were used in this study. NSAIDs, including naproxen sodium tablets (CAS No. 26159-34-2), diclofenac sodium (CAS No. 15307-79-6), and ibuprofen (CAS No. 15687-27-1), were obtained from commercial pharmaceutical formulations and used as experimental compounds.
2. Animals
Female albino Wistar rats weighing 200±10 g were obtained from the Algiers Pasteur Institute. The rats were housed under a 12-hour light/dark cycle at 50%±10% humidity and 22±2 °C, with free access to drinking water and commercially available pelleted feed. The study began after a 2 weeks acclimatization period.
3. Experimental design
All procedures followed the recommendations of the bioethics committee that approved the animal experiments. Ethics approval was obtained from the Consultative Ethics Committee of the Biotechnology Research Center, Constantine, Algeria (Decision No. 27-BS-05/2024). Rats (n=21) were randomly allocated to three equal groups: the control group, which received normal drinking water; the naproxen-treated group (Nap), which received naproxen sodium; and the mixture-treated group (Mix), which received a solution containing naproxen sodium, ibuprofen, diclofenac sodium, and trifloxystrobin at a 4:1:3:2 ratio. Treatments were administered ad libitum daily for 100 days (Figure 1).
Experimental protocol. After 15 days of acclimatization, female rats were randomly assigned to three groups (n=7 per group): control, naproxen (Nap), and mixture (Mix). The control group received normal drinking water, the Nap group received drinking water supplemented with naproxen (98.39 µg/L), and the Mix group received drinking water containing naproxen (98.39 µg/L), ibuprofen (117.14 µg/L), diclofenac (609.12 µg/L), and trifloxystrobin (0.04 mg/kg/day) for 100 days.
Test compounds were prepared weekly as stock solutions in drinking water, and appropriate dilutions were then prepared to obtain working solutions, as described previously [8]. Table 1 summarizes the environmental concentrations and stock solutions used in this study [9,10]. These doses were adjusted based on the rats’ average daily water intake, which was approximately 37 mL/day, and mean body weight. Estimated daily intake (EDI; mg/kg/day) was calculated as follows: EDI=(C×V)/BW, where C represents the compound concentration in drinking water (mg/L), V represents average daily water consumption (L/day), and BW represents average body weight (kg).
Concentrations of chemicals in the mixture for stock solution preparation in the chronic toxicity study
After the 100-day treatment period, vaginal cytology was performed to determine the proestrus stage. Initial and final body weights were recorded to calculate body weight gain (BWG; final body weight−initial body weight) [11]. Animals in proestrus were deeply anesthetized and euthanized by cervical dislocation. The ovaries were rapidly removed, stripped of fatty tissue, blotted, and weighed together, and the ovary index was calculated as follows: the combined weight of both ovaries/body weight×100 [11].
4. Semi-automatic measurement of follicle populations
After sacrifice, one ovary from each animal was washed in phosphate-buffered saline (pH 7.3), fixed in 4% neutral buffered formaldehyde, dehydrated in ethanol, and embedded in Paraplast (Leica Biosystems). Serial sections (5 μm) were obtained at 10 evenly spaced levels using systematic random sampling. Sections were stained with hematoxylin and eosin for follicular histomorphometric assessment (seven sections/ovary/animal). Serial ovarian sections were imaged using an OPTIKA digital camera and Proview software. Follicle populations (primordial, primary, secondary, antral, and atretic follicles) were then semiautomatically quantified in 5–10 fields/section/ovary/animal (n=5/group) by two pathologists blinded to group assignment, using the Trainable Weka Segmentation plugin in Fiji software (WEKA, ver. 3.7.6). Follicles were classified according to standard criteria [12]. To avoid double-counting, only follicles containing a visible oocyte were quantified. Data were expressed as the mean number of follicles per 10 sections (Table 2).
Sample size and analytical units for each endpoint
5. Assessment of ovarian malondialdehyde
Ovarian malondialdehyde (MDA) concentrations were measured according to the protocol described previously [13]. Optical density was measured at 535 nm using a microplate reader. MDA concentrations were calculated using the molar extinction coefficient for the MDA–thiobarbituric acid complex (ε532=1.56×10⁵ M⁻¹ cm⁻¹).
6. Assessment of ovarian catalase
Ovarian catalase (CAT) activity was measured using a previously described protocol [14]. This method is based on the decomposition of H₂O₂ by CAT; absorbance was monitored for 30 seconds at 240 nm using a spectrophotometer. CAT activity was calculated using the molar extinction coefficient for CAT (ε240=0.036 M⁻¹ cm⁻¹) [15].
7. Assessment of serum progesterone levels
Serum progesterone concentrations were measured using an electrochemiluminescence assay, as described previously [16], with a detection range of 0.05–60 ng/mL.
8. Assessment of interleukin 6
Serum interleukin 6 (IL-6) levels were quantified using an IL-6 enzyme-linked immunosorbent assay kit (DRG International Inc.) according to the manufacturer’s protocol, with an assay sensitivity of 2 pg/mL.
9. Assessment of Bcl-2 expression
To assess ovarian B-cell lymphoma 2 (Bcl-2) immunostaining, sections were deparaffinized, rehydrated, and subjected to antigen retrieval by heating at 96 °C in Tris/ethylenediaminetetraacetic acid (EDTA) buffer (pH 9) with EnVision FLEX Target Retrieval Solution (K8004; Dako-Agilent Technologies) for 30 minutes, followed by cooling at room temperature. Sections were then incubated with the primary anti-Bcl-2 antibody (clone 124; 1:200; Dako-Agilent Technologies). Immunoreactivity was visualized using the anti-mouse EnVision FLEX, High pH (Link) kit (K8000; Dako-Agilent Technologies), which produced a brown 3,3'-diaminobenzidine (DAB) chromogen, and sections were counterstained with hematoxylin. Bcl-2 immunoreactivity was evaluated using the color deconvolution plugin in Fiji software [17]. Briefly, stromal and antral follicle regions of interest were manually annotated by two trained pathologists blinded to group assignment. The images were then inverted to produce a light DAB signal on a dark background, and color intensities were converted into pixel intensity values by applying a threshold. Pixel intensity values ranged from 0 (lightest) to 255 (darkest). Based on these values, the intensity score was assigned as summarized in Table 3. Bcl-2 immunoreactivity was evaluated in 5–10 fields across three sections per animal, with five animals analyzed from each experimental group (Table 3).
Semi-quantitative scoring system for Bcl-2 immunohistochemical staining based on pixel intensity values
10. Statistical analysis
Based on our preliminary study, an a priori power analysis was performed using G*Power ver. 3.1.9.7 (University of Düsseldorf) to calculate the sample size required to detect differences among the three groups. An effect size of f=0.75, a significance level of α=0.05, and a power of 1−β=0.80 indicated that seven rats per group were required to achieve sufficient statistical power. Statistical analyses were performed using GraphPad Prism 9.0 (GraphPad Software Inc.). Differences among experimental groups were assessed using one-way analysis of variance followed by the Tukey post hoc test. A p-value <0.05 was considered statistically significant for all tests. Data are presented as mean±standard error of the mean.
Results
1. Effect of naproxen on BWG and ovary index
As shown in Figure 2A, BWG did not differ significantly among the groups (p>0.05). As illustrated in Figure 2B, compared with the control group, both the Nap and Mix groups exhibited a significantly higher ovarian index (p=0.0016 and p=0.0011, respectively), whereas no significant difference was observed between the Nap and Mix groups (p=0.98).
Effects of chronic exposure to environmental doses of naproxen (Nap) and its mixture (Mix) on ovarian outcomes. (A) Body weight gain (BWG), (B) ovary index, (C) malondialdehyde (MDA), (D) catalase (CAT) activity in ovarian tissue, (E) serum interleukin 6 (IL-6), and (F) serum progesterone (P4) levels were measured after 100 days of exposure. Data are presented as mean±standard error of the mean (n=7 rats per group). Statistical comparisons were performed using one-way analysis of variance followed by the Tukey multiple comparison test. UCAT, catalase activity unit; ns, not significant (p>0.05). a)p<0.05; b)p<0.01; c)p<0.001.
2. Effect of naproxen on oxidative stress biomarkers
As shown in Figure 2C, MDA levels were similar across groups, with no significant differences observed (p>0.05). In contrast, Figure 2D shows that CAT activity was significantly decreased in both the Nap group (p<0.0001) and the Mix group (p<0.0001) compared with the control group. No significant difference in CAT activity was observed between the Nap and Mix groups (p=0.98).
3. Effect of naproxen on serum IL-6 and progesterone levels
As shown in Figure 2E, serum IL-6 levels did not differ significantly among the groups (p>0.05). In contrast, serum progesterone levels were significantly decreased in the treated groups compared with the control group (p<0.05). The difference between the Mix and Nap groups was not statistically significant (p=0.68) (Figure 2F).
4. Effect of naproxen on follicle counts
As illustrated in Figure 3A, the Mix group had significantly fewer primary follicles than the control group (p<0.001) and the Nap group (p=0.01). In addition, the number of secondary follicles was significantly reduced in both the Mix group (p=0.007) and Nap group (p=0.002) compared with the control group, whereas no significant difference was observed between the treated groups (p=0.81). Statistical analysis indicated no significant differences among groups in the number of primordial follicles (p>0.05).
Effects of chronic exposure to environmental doses of naproxen (Nap) ans its mixture (Mix) on ovarian B-cell lymphoma 2 (Bcl-2) expression. Representative immunohistochemical photomicrographs showing Bcl-2 protein expression in ovarian tissue from the (A) control, (B) Nap, and (C) mix groups. Positive Bcl-2 imminoreactivity is identified by brown cytoplasmic staining. Red arrowheads indicate Bcl-2-imminopositive cells within ovarian follicles and stromal tissue. Bcl-2 expression was evaluated in granulosa cells (GC), theca interna (TI), follicular antrum (FA), and oocytes (OV). Original magnification ×400. Scale bar=200 µm.
5. Effect of naproxen on Bcl-2 expression
As shown in Figures 3 and 4, Bcl-2 expression differed significantly among groups in both stromal cells and antral follicles, with the highest Bcl-2 intensity scores observed in the control group (Figure. 4B). Bcl-2 expression in antral follicles was significantly reduced in both the Nap and Mix groups compared with the control group (p<0.0001). The reduction in Bcl-2 intensity was greater in the Nap group than in the Mix group (p=0.006). In stromal cells, the Mix group had the lowest Bcl-2 expression, with significantly lower expression than both the Nap group (p=0.008) and the control group (p<0.0001) (Figure 4B); a moderate reduction was also observed in the Nap group compared with the control group (p=0.005).
Effects of chronic exposure to environmental doses of naproxen (Nap) and its mixture (Mix) on ovarian follicle development and B-cell lymphoma 2 (BCL-2) expression. (A) Quantification of ovarian follicles, including preantral and antral stages, and (B) immunohistochemical analysis of anti-apoptotic Bcl-2 protein in granulosa cells and stroma from the control, Nap, and Mix groups after 100 days of exposure. Data are presented as mean±standard error of the mean (n=5 animals per group). Statistical comparisons were performed using one-way analysis of variance followed by the Tukey multiple comparison test. PF, primordial follicles; PrF, primary follicles; SF, secondary follicles; AF, antral follicles; ns, not significant. a)p<0.05; b)p<0.01; c)p<0.001; d)p<0.0001.
Discussion
Environmental pollution driven by industrial and pharmaceutical activities is a growing global concern. Compounds such as NSAIDs, pesticides, and personal care products are major contributors to environmental contamination. However, current chemical risk assessment strategies often fail to capture the complexity of real-life human exposure to mixtures of these substances [18]. Despite their widespread environmental presence, the potential adverse effects of NSAIDs, alone or in combination with other pollutants, on reproductive health at environmentally relevant concentrations remain underexplored. This gap underscores the need for toxicological studies that simulate real-world environmental exposure. Therefore, the present study assessed the effects of long-term exposure to naproxen alone or in a mixture with diclofenac, ibuprofen, and trifloxystrobin administered in drinking water at environmental concentrations on ovarian health, based on BWG, ovary index, progesterone, MDA, CAT, IL-6, Bcl-2 expression, and primordial and preantral follicle counts. To approximate environmentally relevant oral exposure, female rats were exposed to these contaminants through drinking water.
The observed increase in the ovary index after long-term exposure in both treated groups may represent an early biomarker of reproductive toxicity. Previous research has demonstrated that NSAIDs may impair normal ovarian function by disrupting prostaglandin synthesis, leading to tissue changes such as edema, inflammation, or ovarian hypertrophy [19]. This increase in ovary index may be attributable to oxidative stress or inflammation-driven mechanisms that promote tissue remodeling and structural alterations.
Oxidative stress is defined as an imbalance between pro-oxidant and antioxidant species that causes molecular and cellular damage [20]. In the present study, two oxidative stress biomarkers were assessed: MDA and CAT. Treatment had no significant effect on MDA levels, although previous studies of subchronic and chronic NSAID exposure have reported significantly decreased MDA levels [21,22].
Regarding CAT activity, a key first-line antioxidant defense against reactive oxygen species [23], our results showed a marked reduction in both treated groups, suggesting impaired antioxidant capacity and redox imbalance, consistent with previous research [24]. Such redox disturbances may contribute to cellular damage and interfere with endocrine signaling pathways, thereby potentially impairing reproductive hormone synthesis and regulation. In female reproduction, these hormones regulate folliculogenesis, oogenesis, estrous cycling, sexual behavior, and ovulation [25]. Our results demonstrated a significant decrease in serum progesterone levels in the treated groups. Consistent with these findings, NSAIDs have been shown to cause luteinized unruptured follicle syndrome, preventing corpus luteum formation and subsequent progesterone production [26,27]. Moreover, pesticides have been documented to act as endocrine disruptors and induce oxidative stress-mediated hormonal dysregulation [28].
IL-6 is a key proinflammatory cytokine that is often upregulated during tissue injury and oxidative stress. It has been implicated in the regulation of ovarian function, including follicular development, ovulation, and luteal maintenance [29,30]. In the present study, IL-6 levels did not differ significantly among groups, suggesting that chronic exposure to naproxen under the present experimental conditions did not elicit a pronounced inflammatory cytokine response. However, IL-6 expression is tightly regulated and may not increase unless a substantial inflammatory or immune trigger is present.
Female fertility depends on ovarian follicles, which have two main functions: generating mature oocytes and synthesizing steroid hormones [31]. Follicular pool development and follicular growth are essential for fertility and full reproductive function. In humans, follicles regularly leave dormancy and begin a developmental process that takes approximately 1 year. This growth occurs in two phases: a gonadotropin-independent stage (from primordial to secondary follicles) that occurs throughout life, and a gonadotropin-dependent stage (from antral to ovulatory follicles) that begins after puberty [32,33]. In the histological analysis, we observed a significant reduction in healthy preantral follicle numbers in both Nap without a significant effect on the primordial reserve. This decline indicates that chronic exposure to naproxen may impair normal folliculogenesis, potentially by interfering with ovarian follicle recruitment, growth, and maturation. These findings are consistent with earlier studies demonstrating that NSAIDs and other endocrine-disrupting chemicals can adversely affect ovarian function by inhibiting prostaglandin synthesis, which is essential for follicular development and ovulation [34,35]. These disturbances can provoke short-term menstrual or estrous-cycle irregularities and, over time, may deplete the ovarian reserve and reduce fertility [36]. Observational studies of women undergoing assisted reproduction are broadly consistent with these findings, linking higher body burdens of common endocrine-disrupting chemicals, including bisphenol A, phthalates, and per- and polyfluoroalkyl substances, to diminished ovarian reserve and poorer in vitro fertilization outcomes [37].
Bcl-2 is an anti-apoptotic protein in the Bcl-2 family, a group of essential regulators of the intrinsic, or mitochondrial, apoptotic pathway. It suppresses programmed cell death largely by preserving mitochondrial membrane integrity and inhibiting cytochrome c release, thereby favoring cell survival. The balance between anti-apoptotic proteins such as Bcl-2 and pro-apoptotic members of the same family plays a vital role in determining cell fate under normal and pathological conditions [38]. In the present study, Bcl-2 expression was evaluated in both antral follicles and ovarian stromal compartments. Bcl-2 immunoreactivity was significantly reduced in the treated groups, with a more pronounced decrease in the mixture-exposed group across both compartments, suggesting greater pro-apoptotic susceptibility associated with combined chemical exposure. Given the central role of Bcl-2 in granulosa cell survival, its downregulation indicates increased vulnerability of ovarian cells to apoptotic signaling, which may compromise follicular development and ovarian functional integrity. These observations are consistent with the histomorphometric findings of reduced follicle populations and increased follicular atresia [39]. Similar patterns have been reported after exposure to methoxychlor, in which suppression of Bcl-2 expression was associated with increased follicular atresia and apoptotic signaling [40]. In addition, alterations in stromal cell survival may further exacerbate follicular loss by disrupting the ovarian microenvironment required for follicular support. The effects of naproxen—alone or in combination—on cell survival may be mediated, at least in part, by cyclooxygenase inhibition. This may result in reduced prostaglandin synthesis and downstream modulation of anti-apoptotic signaling, including Bcl-2 expression.
Collectively, these data indicate that long-term exposure to environmentally relevant doses of naproxen, alone or in a mixture, disrupted ovarian homeostasis without altering systemic inflammation, lipid peroxidation, or body weight. Treated groups exhibited impaired antioxidant defenses, as indicated by reduced CAT activity, diminished progesterone levels, and marked declines in growing follicles and anti-apoptotic Bcl-2 expression. Notably, these adverse effects on follicular development and cellular survival were most pronounced in the mixture group, suggesting that combined exposures may exert additive or interactive effects that warrant further mechanistic investigation.
This study has several limitations that should be considered. Although progesterone was measured as an indicator of ovarian endocrine activity, other reproductive hormones, such as estradiol, follicle-stimulating hormone, luteinizing hormone, and anti-Müllerian hormone, were not evaluated. In the present study, progesterone was selected as the primary marker of luteal function and ovarian activity after chronic exposure. However, including additional hormonal parameters would provide a more comprehensive characterization of endocrine regulation during folliculogenesis. Furthermore, apoptosis was assessed by immunohistochemical analysis of Bcl-2 expression; however, complementary assays, such as caspase-3 activity or terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) staining, were not performed. Although these analyses were beyond the scope of the present study, they should be considered in future investigations to further elucidate the molecular mechanisms involved in ovarian toxicity.
Notes
Conflict of interest
No potential conflict of interest relevant to this article was reported.
Acknowledgments
We are grateful for the technical assistance provided by Centre de Recherche en Biotechnologie (CRBT) and the Laboratory of Biotechnology of Bioactive Molecules and Cellular Physiopathology (LBMBPC). We also thank Prof. Loucif Lotfi and Mr. Mohamed Aimene Mihoubi of LBMBPC for their assistance and contributions to this work.
Author contributions
Conceptualization: SB, YB. Methodology: AO, SB, YB, SMB, KY, KB, AC. Formal analysis: AO, SB. Data curation: AO, SB. Project administration: SB. Visualization: AO. Software: SB. Validation: SB. Supervision: SB. Writing-original draft: AO. Writing-review & editing: AO, SB. Approval of final manuscript: SB.