Leibovitz’s L-15 medium, α-Minimum Essential Medium (α-MEM), bovine serum albumin (BSA), HEPES buffer, insulin-transferrin-selenium-sodium pyruvate (ITS-A), L-glutamine, ascorbic acid, and CellROX Green Reagent were obtained from Thermo Fisher Scientific (Tokyo, Japan). Polyvinylpyrrolidone (PVP) was purchased from Merck (Darmstadt, Germany). Recombinant human TNF-α, IL-1β, and IL-6 were obtained from Proteintech (Rosemont, IL, USA), and recombinant human FSH was provided by the National Hormone and Peptide Program (Harbor-UCLA Medical Center, Torrance, CA, USA). Enzyme immunoassay (EIA) kits for estradiol and testosterone were purchased from Cayman Chemical (Ann Arbor, MI, USA). The Power SYBR™ Green Cells-to-CT™ Kit and Power SYBR™ Green PCR Master Mix were obtained from Ambion (Waltham, MA, USA) and Applied Biosystems (Foster City, CA, USA), respectively. Custom PCR primers were synthesized by Eurofins Genomics (Tokyo, Japan). Polyclonal anti-collagen type III (N-terminal, #22734-1-AP) antibody was obtained from Proteintech (Rosemont, IL). Alexa Fluor 488-conjugated donkey anti-rabbit IgG was purchased from Invitrogen (Waltham, MA). 4’,6-diamidino-2-phenylindole (DAPI) was obtained from Nacalai Tesque (Kyoto, Japan). The tissue-clearing reagent SCALEVIEW-S4 was acquired from FUJIFILM Wako Chemicals (Osaka, Japan).

All animal experiments were approved by the University of Fukui Committee on Animal Care (Approval Nos. R03059, R04047, R05046, R06034) and adhered to the NIH Guide for the Care and Use of Laboratory Animals. Female Sprague Dawley rats were purchased from Sankyo Labo Service Corporation (Tokyo, Japan) and maintained under controlled environmental conditions with unrestricted access to food and water.

Large preantral follicles were mechanically isolated from excised ovaries of 14-day-old rats in L-15 medium supplemented with 1 mg/mL BSA to maintain tissue viability. Follicles were dissected from the ovarian cortex using a 28.5-gauge insulin needle (Becton Dickinson, New Jersey, USA) under a stereomicroscope (SZX16, Olympus, Tokyo, Japan), ensuring that the basement membrane and TC layer were preserved. Only structurally intact follicles with clearly distinguished GC and TC layers were selected. Follicle diameter was measured along two perpendicular axes using an IX71 inverted microscope (Olympus). Follicles with a GC layer diameter of 130–160 µm, delimited by the basement membrane, were used for culture.

Due to the preserved optical transparency of isolated preantral follicles, two distinct morphological subtypes were readily distinguishable based on the thickness of their TC layers: single-layered TC follicles, characterized by a thin TC layer, and multi-layered TC follicles, with two to three concentric TC layers (Supplementary Figure 1). Preliminary experiments revealed that the TC layer was ≤5 µm in single-layered follicles and ranged from 6 µm to <15 µm in multi-layered ones. Based on these criteria, follicles were categorized for experimental use as follows: those with TC layer thickness ≤5 µm were classified as single-layered TC follicles, whereas those with 6–15 µm thickness were defined as multi-layered TC follicles (Supplementary Figure 2). Approximately ten follicles of each subtype—single−layered and multi−layered TC follicles—were isolated from a single juvenile female rat.

To ensure model consistency, most experiments were conducted using single-layered TC follicles. However, in the final set of analyses, both follicle types were included in a direct comparison to evaluate how TC structural complexity influences follicular responses to pro-inflammatory cytokines.

Preantral follicles were individually cultured in 96-well plates (Sarstedt, Newton, NC, USA) at 37 °C under 5% CO₂ in humidified air. The culture medium consisted of α-MEM supplemented with 2.6 mg/mL HEPES, 1 mg/mL BSA, 10 mg/mL ITS-A, 4 mg/mL L-glutamine, 5 µg/mL ascorbic acid, 100 U/mL penicillin, 100 µg/mL streptomycin, and 2% (w/v) PVP. The addition of PVP increased medium viscosity to maintain follicular spherical integrity and reduce mechanical stress.

The follicles were cultured for three days with or without 10 ng/mL of FSH and different concentrations of TNF-α, IL-1β, or IL-6 (10 or 100 ng/mL), as indicated. The FSH concentration was selected based on preliminary dose-response findings, and cytokine concentrations were based on previous in vitro studies investigating their effects on GC and TC (13, 15, 20–22).

Follicle growth was monitored noninvasively daily using bright-field imaging, and diameters were analyzed using cellSens Standard software (Olympus). GC volume was calculated from the distance between the basement membranes, and TC volume was calculated by subtracting GC volume from total follicle volume. Growth was quantified by comparing volume changes between days 0 and 3. Media were changed every other day, and conditioned media on day 3 were collected and stored at −20 °C for subsequent hormone assays. Data represent 12–40 follicles per group from three to six independent experiments and are expressed as mean ± standard error of the mean (SEM).

Estradiol and testosterone concentrations in day 3 culture conditioned media were quantified using commercial EIA kits, according to manufacturer protocols. Intra-assay coefficients of variation (CVs) were 10.9% for estradiol and 9.5% for testosterone; inter-assay CVs were 12.5% and 11.7%, respectively. Assay sensitivities were 10 pg/mL for estradiol and 6 pg/mL for testosterone. Results represent 4–5 follicles per group from three independent experiments and are expressed as mean ± SEM.

Following sonication-mediated cell lysis, total RNA extraction and cDNA synthesis from individual follicles on day 3 of culture were performed using the Power SYBR™ Green Cells-to-Ct™ Kit, in accordance with the manufacturer’s instructions. During RNA preparation, samples were treated with DNase I to eliminate potential genomic DNA contamination. Quantitative real-time PCR was performed using Power SYBR™ Green PCR Master Mix on a StepOnePlus™ Real-Time PCR System (Applied Biosystems). The thermal cycling conditions were as follows: an initial denaturation at 95 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 60 s.

Target genes included FSH receptor (Fshr), LH receptor (Lhcgr), cholesterol side-chain cleavage enzyme (Cyp11a1), 17α-hydroxylase/17,20-lyase (Cyp17a1), aromatase (Cyp19a1), anti-Müllerian hormone (Amh), and transforming growth factor beta 1 (Tgfb1). 18S rRNA was used as the internal reference gene. Relative gene expression levels were calculated using the 2^−ΔΔCt method. Primer sequences are provided in Supplementary Table 1.

Primers for quantitative PCR were designed in-house and synthesized by Eurofins Genomics (Tokyo, Japan). Prior to experimental use, each primer pair was rigorously validated by generating standard curves to determine the slope, coefficient of determination (R²), and amplification efficiency. Only primer sets that met the following criteria were used for analysis: slope values between −3.58 and −3.10, R² values ranging from 0.950 to 0.999, and amplification efficiencies between 90% and 110%. In addition, melt curve analysis was performed after each qPCR run to confirm the presence of a single specific amplification product. Primer sets that produced primer–dimer formation or nonspecific amplification products were excluded from the study. Data represent 4–5 follicles per group from three independent experiments and are presented as mean ± SEM.

To overcome the technical limitations of sectioning small follicles (diameter <200 μm), we developed a whole-mount immunofluorescence staining method. On day 3 of culture, follicles were fixed in 4% paraformaldehyde for 30 minutes, blocked with 2% horse serum, 1% BSA, and 0.1% Triton X-100, and then incubated overnight at 4 °C with rabbit anti-collagen III antibody (1:200) to detect a tissue fibrosis marker. Alexa Fluor 488-conjugated donkey anti-rabbit IgG antibody (1:1000, Invitrogen) was used as the secondary antibody, and nuclei were counterstained with DAPI. Follicles were cleared with the tissue-clearing reagent SCALEVIEW-S4 for 60 minutes and imaged with an FV1200 confocal laser scanning microscope (Olympus). This enabled visualization of the internal structure of the follicle, including the GC and TC layers, with spatial resolution. Fluorescence intensity within the TC layer was quantified using Fiji (ImageJ 2.9.0/1.53t; NIH, USA) to assess type III collagen expression. A 20−µm−wide band immediately adjacent to the outer surface of the follicular basement membrane was defined as the region of interest (ROI), and the mean fluorescence intensity within this ROI was measured. Background fluorescence was subtracted from each value to obtain corrected signal intensities. Representative images from 7–9 follicles per group are shown.

On day 1 of culture, follicles were incubated with CellROX™ Green Reagent for 30 minutes at 37 °C. After washing, they were washed, fixed in 4% paraformaldehyde in PBS (pH 7.4) for 15 minutes, and counterstained with DAPI. The follicles were then cleared using SCALEVIEW-S4 for 60 minutes and imaged with a confocal microscope (FV1200, Olympus). To assess oxidative stress in the GC layer, fluorescence intensity of the CellROX signal was quantified using Fiji (ImageJ). ROIs were defined as the area enclosed by the follicular basement membrane excluding the oocyte, and mean fluorescence intensities were measured. Background fluorescence was subtracted from each measurement to obtain corrected values. Results represent 6–11 follicles per group from three independent experiments and are expressed as mean ± SEM.

Statistical analyses were performed using GraphPad Prism 8.4.3 (GraphPad Software, La Jolla, CA, USA). Two-group comparisons were assessed using Student’s t-test. For three or more groups, one-way ANOVA followed by Tukey’s post-hoc test was applied. Homogeneity of variance was verified prior to analysis. A P-value < 0.05 was considered statistically significant.

To investigate how pro-inflammatory cytokines influence in vitro follicular development, large preantral follicles with a single TC layer (single-layered TC follicles) were cultured for three days with or without 10 ng/mL FSH and different concentrations of TNF-α, IL-1β, or IL-6 (10 or 100 ng/mL). Representative bright-field images are shown in Figure 1A.

In the absence of both cytokines and FSH, GC volumes remained largely unchanged (mean change: 0.08 ± 0.42 nL). TNF-α alone had no effect on GC volume. FSH significantly promoted GC proliferation (P < 0.01), but this FSH-induced proliferation was markedly inhibited by TNF-α (P < 0.01, Figure 1B, Supplementary Table 2). In contrast, IL-1β and IL-6 had no significant impact on GC volume, irrespective of the presence of FSH.

Regarding theca cells, follicles cultured without FSH or cytokines showed minimal change in TC volume (0.10 ± 0.15 nL). However, all three cytokines—TNF-α, IL-1β, and IL-6—induced TC expansion independently of FSH (P < 0.01, Figure 1C, Supplementary Table 2). FSH alone also increased TC volume (P < 0.01), and this effect was not significantly altered by co-treatment with any of the cytokines.

Steroidogenic activity was assessed by quantifying estradiol and testosterone levels in the culture medium. In the absence of FSH, TNF-α, IL-1β, and IL-6 did not significantly affect estradiol production (Figure 2A, Supplementary Table 3). FSH treatment led to a substantial increase in estradiol levels (P < 0.01), which was significantly attenuated by all three cytokines (P < 0.01).

Testosterone levels were similarly affected. TNF-α and IL-1β had no effect on basal testosterone production, while IL-6 slightly elevated testosterone in the absence of FSH (P < 0.01, Figure 2B, Supplementary Table 3). However, the robust FSH-induced increase in testosterone was significantly suppressed by TNF-α, IL-1β, and IL-6 (P < 0.01).

Quantitative RT-PCR was performed to evaluate the expression of GC-related genes (Fshr, Cyp19a1, and Amh). TNF-α significantly downregulated Fshr mRNA, irrespective of FSH treatment (P < 0.01, Figure 3A, Supplementary Table 4). IL-1β had no significant effect on Fshr, whereas IL-6 slightly increased Fshr in the absence of FSH but suppressed its expression in the presence of FSH (P < 0.01).

FSH robustly upregulated Cyp19a1 expression (P < 0.01), which was consistently suppressed by all three cytokines (P < 0.01, Figure 3B, Supplementary Table 4). For Amh, TNF-α and IL-1β significantly decreased expression (P < 0.01), while IL-6 induced a modest increase (P < 0.05, Figure 3C, Supplementary Table 4).

Expression of TC-related genes (Lhcgr, Cyp11a1, and Cyp17a1) was also analyzed. FSH significantly induced Lhcgr expression (P < 0.01), which was notably inhibited by TNF-α, IL-1β, and IL-6 (TNF-α and IL-6: P < 0.01, IL-1β: P < 0.05, Figure 4A, Supplementary Table 5). Similarly, Cyp11a1 expression was increased by FSH and attenuated by IL-1β and IL-6, while TNF-α had no significant effect (Figure 4B, Supplementary Table 5).

In the absence of FSH, IL-6 reduced Cyp17a1 expression (P < 0.01). FSH strongly induced Cyp17a1 levels (P < 0.01), but this upregulation was suppressed by all three cytokines (P < 0.01, Figure 4C, Supplementary Table 5).

Despite the cytokine-induced expansion of the TC layer (Figure 1C), all three cytokines significantly suppressed FSH-induced androgen production (Figures 2B, 4A–C). To explore this discrepancy, we examined the expression of Tgfb1, a key mediator of fibrotic signaling.

TNF-α, IL-1β, and IL-6 markedly upregulated Tgfb1 mRNA expression (TNF-α: P < 0.05, IL-1β and IL-6: P < 0.01, Figure 5A, Supplementary Table 6), suggesting activation of fibrosis-related pathways. Immunofluorescence staining of type III collagen further supported this hypothesis. IL-1β and IL-6 modestly increased type III collagen deposition within the TC layer (IL-1β: P < 0.05, IL-6: P < 0.01, Figures 5B, C, Supplementary Table 6), consistent with early fibrotic signaling.

To assess oxidative stress in preantral follicles, intracellular reactive oxygen species (ROS) were visualized using CellROX Green. Both TNF-α and IL-1β significantly elevated ROS levels in GC, as evidenced by increased fluorescence intensity (P < 0.05; Figures 6A, B, Supplementary Table 7). In contrast, IL-6 did not induce a significant rise in ROS generation in GC.

Two morphological subtypes of preantral follicles—single-layered and multi-layered TC follicles—were compared to determine whether TC structure influences cytokine sensitivity (Supplementary Figures 1, 2).

In the absence of FSH, TNF-α, IL-1β, or IL-6 did not altere GC volume in either single-layered or multi-layered TC follicles (Figures 1B, 7A). Under FSH stimulation, 100 ng/mL TNF-α suppressed GC proliferation in both subtypes. However, at a lower concentration (10 ng/mL), TNF-α significantly inhibited GC proliferation only in single-layered TC follicles, but not in multi-layered TC follicles (Figures 1B, 7A, Supplementary Table 8). These results suggest a protective role of multi-layered TC structure against low-dose TNF-α-induced GC growth inhibition.

TNF-α and IL-6 stimulated TC proliferation in single-layered TC follicles but not in multi-layered TC follicles (Figure 7B, Supplementary Table 8). In contrast, IL-1β increased TC volume in both subtypes (P < 0.01), indicating a distinct mechanism of action.

To assess oxidative stress, intracellular ROS levels in both follicle types were measured following TNF-α treatment. Fluorescence intensity, indicative of ROS accumulation, was significantly lower in multi-layered TC follicles compared to single-layered TC follicles (P < 0.05, Figure 7C, Supplementary Table 9). Furthermore, TNF-α suppressed Fshr expression in single-layered TC follicles (P < 0.01) but not in multi-layered TC follicles (Figure 7D, Supplementary Table 9). These results suggest that the multi-layered TC structure confers protection against TNF-α-induced oxidative stress and GC suppression in preantral follicles.