Digitoxin-induced apoptosis in ovarian granulosa cells disrupts follicular development and impairs reproductive performance

Abstract

Background

Digitoxin, a cardiac glycoside from Digitalis spp., is clinically employed for heart failure and atrial fibrillation, and yet its reproductive toxicity is not well understood. We have therefore assessed digitoxin′s influence on follicular development and reproductive outcomes in both cell-culture models and animal models.

Methods

The in vitro culture of porcine follicles was employed to assess gene expression levels related to proliferation and apoptosis, whereas an in vivo mouse model was utilized to evaluate hormone levels (via ELISA), ovarian morphology (H&E staining), apoptosis (TUNEL assay), and reproductive performance.

Results

Transcriptomic analysis revealed that, in the digitoxin-treated group, 1,577 genes were upregulated, whereas 4,359 genes were downregulated compared with the control group. KEGG enrichment analysis demonstrated significant involvement of these differentially expressed genes in pathways associated with amino acid metabolism, steroid biosynthesis, and oocyte maturation. Additionally, GO enrichment analysis underscored their role in regulating cellular functions and metabolic processes. GSEA further indicated that digitoxin influences pathways related to DNA function, meiotic progression, and steroid biosynthesis.

Moreover, treatment with 100 nM digitoxin for 48 h significantly reduced granulosa cell (GC) viability to ~ 15%, decreased proliferation to 3.02% (p < 0.05), and increased apoptosis to 34.9% (p < 0.05). Digitoxin treatment also downregulated PCNA, CDK4, and MCL1, while upregulating CASP3, CASP7, CASP8, and CASP9. In vivo, digitoxin administration resulted in lower serum levels of estradiol (E2), follicle-stimulating hormone (FSH), and luteinizing hormone (LH) (p < 0.001), an increase in antral follicle counts, a decrease in corpus luteum numbers (p < 0.05), a delayed onset of first estrus, and a reduction in litter size (p < 0.05).

Conclusion

Our study elucidates the multifaceted effects of digitoxin on GC function and female reproductive health, highlighting significant implications for fertility and reproductive toxicity. Whereas digitoxin is beneficial in treating cardiovascular diseases and certain tumors, the careful selection and monitoring of therapeutic doses are crucial, particularly for women of reproductive age requiring prolonged treatment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13048-026-01965-7.

Introduction

Follicular development is a crucial process that ensures the proper functioning of the reproductive system in mammals, directly influencing female fertility and reproductive efficiency. Granulosa cells (GCs) within the follicle not only provide essential nutrients and signaling support to oocytes, but also regulate follicle proliferation and maturation through the secretion of hormones and growth factors. The dynamic balance between the proliferation and apoptosis of GCs is critical for follicular growth and for preventing excessive atresia [1, 2]. In multiparous animals such as pigs, functional abnormalities of GCs are closely associated with impaired follicular development, which consequently affects estrous frequency, pregnancy rate, and litter size [3, 4].

Digitoxin, a cardiac glycoside primarily derived from the foxglove plant, specifically Digitalis purpurea and Digitalis lanata, is utilized in the treatment of cardiovascular diseases, particularly heart failure and atrial fibrillation [5, 6]. In addition, digitoxin exhibits promising anticancer activity, in particular by inducing cell cycle arrest and apoptosis selectively in cancer cells [7, 8]. Digitoxin, while showing promising applications in the treatment of cardiovascular diseases, cancers, inflammation, and viral infections, has a narrow therapeutic index, which limits its clinical use [5, 6]. This narrow margin between effective and toxic doses raises concerns about safety. Additionally, digitoxin is associated with liver toxicity and other adverse effects that can complicate treatment regimens. Patients may experience side effects such as gastrointestinal disturbances, cardiac issues, and neurological symptoms [9–11]. Despite known toxicities, the potential reproductive risks of digitoxin have received little investigation.

Evidence suggests that cardiac glycosides indirectly influence ovarian function by disrupting hormone secretion, oxidative stress responses, and various signaling pathways [12, 13]. Fluctuations in the intra-ovarian hormonal environment, particularly in levels of estrogen (E2), follicle-stimulating hormone (FSH), and luteinizing hormone (LH), play a vital role in regulating follicular development and maturation [14, 15]. Therefore, exploration of the impact of digitoxin on these hormones and the mechanisms involved might offer valuable insights into its effects on ovarian function. Previous studies have shown that digitoxin can affect the proliferation and apoptosis of various cell types [16, 17]. However, its specific effects on GCs during follicular development remain poorly understood.

The foxglove is a beautiful plant commonly used in garden decoration, although all parts, especially the roots, are highly toxic [18]. In regions such as China, India, Brazil, and parts of Africa, livestock (e.g., pigs, sheep, and cattle) are often raised in free‑range systems, increasing their exposure to toxic plants such as foxgloves [19, 20]. Humans can also be poisoned by ingesting the plant following its misidentification, its use in herbal medicines/tea, it being confused with other plants, and exposure to its pollen [21–26]. Additionally, the pervasive presence of pharmaceutical residues and agrochemicals in the environment has heightened the importance of investigating the potential reproductive toxicity of such compounds. As a plant-derived cardiac glycoside, digitoxin has the capacity to accumulate in the environment and to enter the food chain, potentially posing unknown reproductive risks [27–29]. Therefore, investigation of the reproductive toxicology of digitoxin is essential not only for evaluating its safety, but also for providing a scientific foundation for the development of effective animal husbandry practices and drug use policies.

In this study, we have employed a systematic approach that integrates in vitro cellular experiments with in vivo animal models to investigate the regulatory effects of digitoxin on follicular development and ovarian function. By elucidating the mechanisms through which digitoxin influences these processes, we aim to enhance our understanding of its potential adverse effects on reproductive performance in livestock and reproductive health in humans.

Materials and methods

Animal experiments

21-day-old female C57BL/6J mice were obtained from the Cancer Center of Sun Yat-sen Memorial Hospital (Guangzhou, China). Mice were housed individually in ventilated filter‑top cages and given ad libitum access to autoclaved food and water. They were maintained under specific pathogen‑free conditions at 25 °C and 60% relative humidity, with a 12‑h light/dark cycle. During the experimental period, body weight, estrous cycle stage, and clinical signs of illness were recorded daily. At the end of treatment, mice were euthanized by CO₂ inhalation, followed by decapitation to confirm death, in accordance with institutional animal care and use guidelines.

Intraperitoneal doses of cardiac glycosides reported in mice vary widely, from approximately 0.3 µg/kg to 2 mg/kg [30]. Daily doses of 1–2 mg/kg are commonly used in anticancer and inflammatory models, whereas lower doses are typically selected for chronic studies to reduce toxicity [16]. Accordingly, the dosages in the present study were chosen with reference to these prior reports.

Digitoxin (HY-B1357; MedChemExpress, Monmouth Junction, NJ, USA) was dissolved in dimethyl sulfoxide (DMSO; HY-YO320; MedChemExpress) to prepare a 5 mg/mL stock solution and stored at − 20 °C. For injections, 10 µL digitoxin solution in DMSO was mixed with 90 µL sterile physiological saline to a final volume of 100 µL (final DMSO 10% v/v), producing concentrations corresponding to doses of 1.0, 1.5, 2.0, 2.5, or 3.0 mg/kg. Control animals received an identical vehicle (10 µL DMSO without digitoxin + 90 µL saline). Intraperitoneal administration was selected to provide more precise control of dose and timing than oral delivery, thereby reducing variability attributable to differences in food and water intake [31]. Intraperitoneal injections were administered slowly to minimize local irritation, osmotic shock, and other potential side effects.

For the ovarian development assay, twelve 21-day-old female C57BL/6J mice were randomly divided into two groups (n = 6 per group): a vehicle control group receiving DMSO and a digitoxin-treated group receiving 2.5 mg/kg digitoxin. For the fertility study, another set of twelve 21-day-old female C57BL/6J mice were similarly assigned to two groups (n = 6 per group), with one group receiving digitoxin (2.5 mg/kg) and the other receiving an equivalent volume of DMSO vehicle. Intraperitoneal injections were administered every other day (between 9:00 and 10:00 am) for a total duration of three weeks. For downstream analyses, 3–4 mice per group were randomly selected and euthanized to obtain representative samples while conserving resources and minimizing variability.

At 42 days of age, digitoxin treatment was discontinued, and the female mice were co‑housed with male mice for mating. Age at puberty and the sizes of the first and second litters were recorded to assess the longer‑term effects of digitoxin on reproductive performance.

Isolation and culture of GCs

Porcine ovaries were collected from healthy Landrace sows (180-day-old sows in the early stages of estrus; approximately 90 kg) at a certified slaughterhouse in Guangzhou, China and transported to the laboratory on ice for immediate processing. Primary porcine GCs were isolated from antral follicles with diameters ranging from 3 to 5 mm under sterile conditions in a laminar flow hood [32]. The isolated cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM)/F12 (Hyclone, Logan, UT, USA) medium supplemented with 10% fetal bovine serum (FBS; Hyclone) at 37 °C in a humidified atmosphere containing 5% CO₂. When cultures reached ~ 70–80% confluence, GCs were treated with 100 nM digitoxin or an equivalent volume of DMSO (vehicle control) for 48 h. Cells were then harvested and total RNA and protein were extracted for downstream analyses.

Follicle isolation and culture

Porcine ovaries were collected from healthy Landrace sows (180-day-old sows in the early stages of estrus; approximately 90 kg) at a certified slaughterhouse in Guangzhou, China and transported to the laboratory on ice for immediate processing. After being thoroughly rinsed with 70% ethanol (MERYER, Shanghai, China) and phosphate-buffered saline (PBS; Biosharp, Guangzhou, China), antral follicles measuring 3 to 5 mm in diameter were carefully dissected, washed twice with PBS containing 1% penicillin-streptomycin (Thermo Fisher Scientific, Waltham, MA, USA), and cultured in serum-free DMEM medium under standard conditions (37 °C, 5% CO₂) [33]. Isolated porcine follicles were cultured in a medium containing either 100 nM digitoxin (in DMSO) or an equivalent volume of DMSO (vehicle control). The morphology and development of follicles were recorded on days 1, 3, and 5 of the culture period.

Cell viability assay

The viability of GCs was evaluated using the Cell Counting Kit-8 (CCK-8; Beyotime, Shanghai, China) [32]. After 24 h, 36 h, and 48 h of drug treatment, 100 µL medium containing 10% CCK-8 reagent was added to each well. The cells were subsequently incubated at 37 °C for 1 h in the dark. Absorbance was measured at 450 nm in a Bio-Rad^® iMark™ microplate reader (Bio-Rad, Hercules, CA, USA). The cell viability was calculated using the following formula: Cell viability (%) = [(A_sample - A_blank) / (A_control - A_blank)] ×100%. To ensure the reliability of our findings, all experiments were performed in triplicate.

EdU assay for cell proliferation

The 5-ethynyl-2’-deoxyuridine (EdU) assay, utilizing the Cell-Light™ EdU Kit (RiboBio, Guangdong, China), was employed to evaluate the proliferation of GCs [32]. Initially, GCs were plated in 48-well plates and cultured until they reached approximately 50% to 80% confluence. Each well was then treated with 200 µL diluted EdU medium (50 µM) for a duration of 2 h, after which 200 µL fixation solution (80% acetone diluted in PBS) was added to each well, followed by incubation at room temperature for 30 min. The wells were subsequently washed three times with PBS, with each wash step lasting 3 min.

Next, 200 µL permeabilization solution (0.5% Triton X-100 (Beyotime Biotech, Shanghai, China) in PBS) was added to each well, and the cells were incubated for 10 min to facilitate permeabilization. After another round of PBS washes, the cells were incubated with 200 µL 1× Apollo^® staining reaction solution for 30 min in the dark. Following this step, the cells were washed with PBS again, and the permeabilization step was repeated. Finally, 200 µL of 10% DAPI staining solution was added to each well, and the cells were incubated at room temperature in the dark for 30 min. Fluorescent images were captured using a Nikon ECLIPSE Ti2 fluorescence microscope (Nikon, Tokyo, Japan). To quantify proliferation, EdU-positive cells were manually counted, and the results were expressed as a proportion of total Hoechst-stained nuclei in a blinded fashion, from three randomly selected fields per well. To ensure the reliability of our findings, all experiments were performed in triplicate.

Cell apoptosis assay

Apoptosis levels were evaluated using an Annexin V apoptosis kit (Lianke Biotech, Shanghai, China), as described in our previous study [32]. Briefly, GCs were seeded in 6-well plates and treated with digitoxin at concentrations of 100 nM, 200 nM, and 300 nM for 24 and 48 h. After treatment, the cells were harvested, washed with PBS, and gently resuspended in 500 µL of 1× Annexin V binding buffer. Subsequently, 5 µL Annexin V-Fluorescein Isothiocyanate (FITC) and 5 µL propidium iodide (PI) staining solution were added to each sample, and the cells were incubated in the dark for 15 min. Apoptosis rates were then analyzed using a BD FACS Calibur Flow Cytometer (BD Biosciences, Franklin Lakes, NJ, USA) and FlowJo software (FlowJo, Ashland, OR, USA). To ensure the reliability of our findings, all experiments were performed in triplicate.

Hematoxylin and eosin (H&E) staining

Tissue processing and sectioning were performed according to the methods described in our previous study [32]. Left ovaries from three mice (selected across the control and experimental groups) were harvested for H&E staining and Terminal deoxynucleotidyl transferase dUTP Nick End Labeling (TUNEL) staining.

Serial paraffin Sect. (3 μm thick) were prepared using a Leica RM2016 microtome (Leica, Shanghai, China). The sections were dewaxed sequentially with Eco-friendly Dewaxing Solution I (Servicebio, Wuhan, China) for 20 min, followed by Dewaxing Solution II (Servicebio) for an additional 20 min. They were then rehydrated in absolute ethanol (100092683, SCRC, Shanghai, China) for 5 min, immersed in 75% ethanol for another 5 min, and rinsed under running tap water.

Sections were then stained with hematoxylin for 3 min, followed by sequential treatments with tap water, differentiation solution, bluing agent, and another rinse. The sections were dehydrated in 95% ethanol for 1 min, counterstained in eosin for 15 s, and then cleared sequentially in absolute ethanol, n-butanol, and xylene until transparent. Slides were examined using a NIKON ECLIPSE E100 microscope (Nikon), and images were captured with a NIKON DS-U3 imaging system (Nikon). Follicles of different stages were counted manually under a microscope by observers blinded to group allocation. Healthy follicles exhibited uniform morphology, intact cellular architecture, and appropriate developmental staging, whereas unhealthy follicles showed degeneration, irregular contours, and disrupted cellular organization.

TUNEL assay

TUNEL was conducted using a TUNEL apoptosis detection kit (Servicebio, Wuhan, China) [32]. Briefly, paraffin sections were dewaxed using an eco-friendly dewaxing solution and rinsed three times with distilled water, with each rinse lasting 5 min. The sections were then incubated with proteinase K for 20 min, followed by three washes with PBS. Next, a permeabilization solution was applied for 20 min, followed by three additional PBS washes. After the application of an equilibration buffer, the sections were incubated at room temperature for 10 min and then treated with the TUNEL reaction mixture for 1 h. Nuclei were counterstained with DAPI. Finally, the sections were mounted with an anti-fade mounting medium.

Fluorescent signals were visualized using a Nikon Eclipse C1 fluorescence microscope, and images were acquired with a Pannoramic MIDI automatic digital slide scanner (3DHISTECH, Budapest, Hungary). Quantitative analysis of fluorescence was performed using the ImageJ program (V1.8.0.112; https://imagej.net/ij/).

RNA extraction and quantitative real-time PCR

Total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific) following the manufacturer’s instructions [32]. The extracted RNA was subsequently reverse-transcribed into complementary DNA (cDNA) by using a cDNA synthesis kit (Yisheng, Shanghai, China). Quantitative real-time PCR (qRT-PCR) was conducted using the SYBR Green qPCR Master Mix on a Bio-Rad CFX96 Touch Real-Time PCR System (Bio-Rad). Gene expression levels were quantified using the 2^−ΔΔCt method. GAPDH served as the internal control for normalization in the RT‑PCR assays. To ensure the reliability of our findings, the experiments were performed in triplicate. Specific primer sequences utilized in the assay are provided in Table 1.

Table 1.

Sequences of primers utilized in qRT-PCR

Gene	Accession Number	Primer sequence (5’ to 3’)	Amplicon Length (bp)
PCNA	NC_000068	F: GGTTACTGAGGGCGAGAAGC	98
		R: GACCGGCTGAGACTTGCGTA
STAR	NM_011485	F: GCTCTCTACTCGGTTCTCGG	277
		R: TTCCACTCCCCCATTGCTTC
P53	NM_001127233	F: ATGGAGGAGTCACAGTCGGA	120
		R: ATCGTCCATGCAGTGAGGTG
Casp3	NM_009810	F: CCAAAGATCATACATGGAAGCG	185
		R: CTGAATGTTTCCCTGAGGTTTG
Casp9	NM_015733	F: GGCTGGTGGAAGAGCTGC	100
		R: GAGCCTGCCCGCTGGA
Casp8	NM_009812	F: GTCTGTACCTTTCTGGCGGA	300
		R: CACAACTCCTCCCCTTTGCT
Casp7	NM_007611	F: CCGTCCACAATGACTGCTCTTG	155
		R: CCCGTAAATCAGGTCCTCTTCC
CDK4	NM_009870	F: CATACCTGGACAAAGCACCTCC	135
		R: GAATGTTCTCTGGCTTCAGGTCC
CDK1	NM_007659	F: CATGGACCTCAAGAAGTACCTGG	136
		R: CAAGTCTCTGTGAAGAACTCGCC
MCL1	NM_008562	F: AGCTTCATCGAACCATTAGCAGAA	125
		R: CCTTCTAGGTCCTGTACGTGGA
CREB1	NM_009952	F: CACAGACCACTGATGGACAGCA	179
		R: AGGACGCCATAACAACTCCAGG
CCNB1	NM_172301	F: AGAGGTGGAACTTGCTGAGCCT	126
		R: GCACATCCAGATGTTTCCATCGG
SP1	NM_013672	F: CTCCAGACCATTAACCTCAGTGC	291
		R: CACCACCAGATCCATGAAGACC
GAPDH	NM_008084	F: CATCACTGCCACCCAGAAGACTG	153
		R: ATGCCAGTGAGCTTCCCGTTCAG

RNA sequencing and bioinformatics analysis

High-throughput RNA sequencing and subsequent bioinformatics analysis were conducted by Gene Denovo Biotechnology (Guangzhou, China). Total RNA was extracted from the samples by using TRIzol reagent (Invitrogen, Waltham, MA, USA), in accordance with the manufacturer’s protocol.

The concentration and purity of the RNA were assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific), ensuring that the Optical density (OD) 260/280 ratio was ≥ 1.8, and that the OD260/230 ratio was ≥ 2.0. RNA integrity was verified using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), with samples exhibiting RNA Integrity Numbers (RIN) ≥ 7.0 being considered acceptable for sequencing. mRNA was isolated and fragmented into short segments using a fragmentation buffer, followed by reverse transcription into complementary DNA (cDNA) using the NEBNext^® Ultra™ RNA Library Prep Kit for Illumina (New England Biolabs, Ipswich, MA, USA).

The quality of library construction was evaluated using the Agilent DNA 1000 Kit, and the final cDNA libraries were sequenced on an Illumina NovaSeq 6000 platform (Illumina, San Diego, CA, USA), generating paired-end reads. Raw sequencing data were initially processed with FastQC for quality control. High-quality clean reads were aligned to the reference genome (GRCh38; https://www.ncbi.nlm.nih.gov/assembly/GCF_000001405.39/) by using the Hisat2 aligner [34]. Transcript reconstruction and quantification were performed with StringTie [35]. Differentially expressed genes (DEGs) were identified with the DESeq2 package [36], using thresholds set at |log₂ fold change| ≥ 1 and a false discovery rate (FDR) < 0.05. Functional enrichment analyses of DEGs were conducted based on Gene Ontology (GO; http://geneontology.org/), Kyoto Encyclopedia of Genes and Genomes (KEGG; https://www.genome.jp/kegg/), and Gene Set Enrichment Analysis (GSEA) [37] to elucidate the biological processes and pathways involved. To ensure the reliability of our findings, the experiments were performed in triplicate.

Enzyme-linked immunosorbent assay

The concentrations of E2, FSH, and LH in mouse serum were quantified using commercial ELISA kits specific for mouse E2, FSH, and LH (Enzyme-linked Biotechnology, Shanghai, China) [32]. Kit catalog numbers and assay precision were as follows: Mouse E2 (MM-0566M1) - intra-assay coefficient of variation (CV) 6.2%, inter-assay CV 8.5%; Mouse LH (MM-44039M1) - intra-assay CV 5.7%, inter-assay CV 7.8%; Mouse FSH (MM-45654M1) - intra-assay CV 7.3%, inter-assay CV 9.4%. All assays were performed according to the manufacturer′s instructions.

At the conclusion of the experiment, blood was obtained from mice by retro-orbital bleeding and immediately transferred to tubes without anticoagulant. After the blood had been allowed to clot at room temperature for 20–30 min, samples were centrifuged at 3,000 rpm for 10 min, and the resulting serum was carefully transferred to fresh tubes for analysis.

For ELISA, a complex of horseradish peroxidase (HRP)-conjugated antibody and the corresponding hormone antigen was formed to create an antibody–antigen–enzyme conjugate. Standard solutions of known concentrations, together with experimental samples, were added to the designated wells of pre-coated ELISA plates and incubated at 37 °C for 30 min. After the plates had been washed, 100 µL conjugate solution was added to each well, and the samples were incubated at 37 °C for an additional 30 min. Following this incubation, the plates were washed again, and 50 µL substrate solution A and 50 µL substrate solution B were added to each well. The reaction was allowed to proceed in the dark at 37 °C for 10 min. Finally, 50 µL stop solution was added to each well, and the optical density (OD) at 450 nm was measured with an iMark™ Microplate Absorbance Reader (Bio-Rad). Concentrations of hormones were determined by interpolation from standard curves generated using known hormone standards. To ensure the reliability of our findings, the experiments were performed in triplicate.

Western blot analysis

Western blotting of ovarian tissue, isolated follicles, and GCs was performed as previously described [38]. Ovarian tissues and follicles were snap-frozen in liquid nitrogen immediately after isolation, transferred to cryogenic tubes, and stored at − 80 °C until analysis. For analysis, frozen samples were pulverized on dry ice and lysed in ice-cold radioimmunoprecipitation assay (RIPA) buffer (Bestbio, Shanghai, China) containing protease inhibitors (Biosharp). Samples were homogenized by using Bio-Gen PRO200 homogenizer (PRO Scientific, Oxford, CT, USA), incubated on ice for 20 min, and centrifuged at 14,000×g for 15 min at 4 °C. Supernatants were collected, and their protein concentrations were determined by bicinchoninic acid (BCA) protein concentration assay kit (Biosharp); aliquots were mixed with Laemmli buffer (Biosharp), heated at 95 °C for 5 min, and stored at − 80 °C until use. GCs were washed twice with cold PBS, lysed in RIPA buffer containing protease inhibitors (Biosharp) on ice for 30 min, clarified by centrifugation (14,000 × g, 15 min, 4 °C), and quantified by BCA protein concentration assay kit (Biosharp).

After denaturation, protein samples were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). A total of 15 µg protein per sample was loaded for Western blotting. Following electrophoresis, proteins were transferred onto polyvinylidene difluoride (PVDF) membranes (Bio-Rad). The membranes were then blocked with non-fat milk and incubated overnight at 4 °C with the following primary antibodies: Caspase-3 (CASP3; 66470-2-Ig, Proteintech, Rosemont, IL, USA; 1:1000), Caspase-9 (CASP9;10380-1-AP, Proteintech, 1:1000), proliferating cell nuclear antigen (PCNA;10205-2-AP, Proteintech;1:20000), Myeloid cell leukemia-1(MCL1; AF5311; Affinity Bioscience, Cincinnati, OH, USA; 1:1000), and α-Tubulin (11224-1-AP; Proteintech; 1:8000). After primary antibody incubation, the membranes were incubated with an HRP-conjugated goat anti-rabbit IgG secondary antibody (3032; SAB, Nanjing, China; 1:10000) at room temperature for 1 h. Signal detection was performed using Luminol substrate (Thermo Fisher Scientific) for chemiluminescence. Images were captured using the ChemiScope 6200 chemiluminescence imaging system (Clinx Science Instruments, Shanghai, China). α‑Tubulin was used as the loading control (reference protein) for normalization. Band intensities were subsequently analyzed and quantified by using the ImageJ program. To ensure the reliability of our findings, all experiments were performed in triplicate.

Statistical analysis

The numbers of independent biological samples and experimental replicates are indicated in the figures and legends. Data are expressed as mean ± standard error of the mean (s.e.m.). To evaluate differences between groups, Student′s t-test was utilized, with corresponding p-values indicated in the figures. A p-value < 0.05 (*) was considered significant between groups, p < 0.01 (**) highly significant, and p < 0.001 (***) extremely significant.

Results

Effect of digitoxin on the transcriptome of porcine GCs

Porcine GCs were treated with digitoxin at concentrations of 100 nM, 200 nM, and 300 nM for durations of 12, 24, 36, and 48 h (Fig. 1A). The results indicated that cell viability was lowest following treatment with 100 nM digitoxin for 48 h (Fig. 1B). Consequently, this concentration and time point (100 nM for 48 h) were selected for subsequent experiments.

Fig. 1.

Effect of digitoxin on viability and transcriptome of porcine GCs. Data are from n = 3 biological replicates. A Bar graph showing cell viability assessed by CCK-8 assay. B Line graph showing cell viability over time assessed by CCK-8 assay. C Bar chart showing the number of DEGs in GCs between the digitoxin-treated group and the DMSO control group. D KEGG pathway enrichment bubble plot of DEGs. E GO pathway enrichment bubble plot of DEGs. F GSEA showing enriched pathways of gene sets, with blue bars representing downregulated genes, and red bars representing upregulated genes. Data are presented as mean ± SEM. A p-value < 0.05 (*) was considered significant between groups, p < 0.01 (**) highly significant, and p < 0.001 (***) extremely significant. “ns” denotes no significant difference

Porcine GCs were treated with digitoxin, and total RNA was extracted for transcriptome sequencing. The analysis revealed that, compared with the control group, 1,577 genes were upregulated, whereas 4,359 genes were downregulated in the digitoxin-treated group (Fig. 1C). KEGG enrichment analysis showed that these DEGs were significantly enriched in signaling pathways related to amino acid metabolism, steroid biosynthesis, and oocyte maturation (Fig. 1D). Moreover, GO enrichment analysis revealed that these genes were involved in regulating cellular functions and metabolic processes (Fig. 1E). Furthermore, GSEA demonstrated that digitoxin probably influences signaling pathways associated with DNA function, meiotic progression, and steroid biosynthesis (Fig. 1F).

Effect of digitoxin on porcine GC proliferation and apoptosis

Results from the EdU assay demonstrated that porcine GC proliferation was significantly reduced after a 48-h treatment with digitoxin (100 nM) compared with the control group (p < 0.05, Fig. 2A). Additionally, qRT-PCR analysis revealed that digitoxin significantly downregulated the mRNA expression of key proliferation-related genes, including CDK4 (p < 0.001) and PCNA (p < 0.001; Fig. 2B). Western blot analysis further confirmed that digitoxin significantly suppressed the protein expression levels of PCNA (p < 0.05; Fig. 2C).

Fig. 2.

Effect of 100 nM digitoxin on proliferation and apoptosis of porcine GCs. Data are from n = 3 biological replicates. A EdU assay showing the effect of 24 h treatment with DMSO or 100 nM digitoxin on GC proliferation. Scale bar: 100 μm. B Effect of digitoxin on mRNA expression of key genes (CDK4 and PCNA) involved in the cell proliferation pathway. C Effect of digitoxin on protein expression of PCNA, which is involved in the cell proliferation pathway. D Effect of 24-h digitoxin treatment on GC apoptosis. E Effect of 48-h digitoxin treatment on GC apoptosis. F Line graph showing the effect of digitoxin on GC apoptosis. G Effect of digitoxin on mRNA expression of key genes in the apoptosis pathway. H Effect of digitoxin on protein expression of a key gene (CASP3) in the apoptosis pathway. Data are presented as mean ± SEM. A p-value < 0.05 (*) was considered significant between groups, p < 0.01 (**) highly significant, and p < 0.001 (***) extremely significant

The Annexin V-FITC assay results indicated that treatment with the various concentrations of digitoxin for 24 and 48 h significantly enhanced apoptosis in GCs (p < 0.05; Figs. 2D, 2E and 2F). Furthermore, qRT-PCR analysis revealed that digitoxin markedly upregulated the mRNA expression of key apoptosis-related genes, including CASP3 (p < 0.001), CASP7 (p < 0.01), CASP8 (p < 0.01), CASP9 (p < 0.001), BAX (p < 0.001), and P53 (p < 0.01) (Fig. 2G). Additionally, Western blot analysis confirmed a significant increase in CASP3 protein expression levels following digitoxin treatment (p < 0.05; Fig. 2H). Collectively, these findings suggest that digitoxin promotes apoptosis in porcine GCs.

Effect of digitoxin on porcine follicles

Isolated porcine follicles were cultured in a medium containing either 100 nM digitoxin or DMSO as a control. The results indicated that porcine follicles cultured in DMSO exhibited a gradual increase in vascularization and clear follicular fluid. In contrast, porcine follicles treated with 100 nM digitoxin showed a rapid loss of blood vessels and a turbid appearance of follicular fluid (Fig. 3A). qRT-PCR analysis showed that digitoxin significantly suppressed the mRNA expression of key genes involved in cell proliferation, including PCNA (p < 0.001), CDK1 (p < 0.01), MCL1 (p < 0.001), and STAR (p < 0.001) in the follicles. Concurrently, digitoxin promoted the mRNA levels of key apoptosis-related genes, such as CASP8 (p < 0.001), CASP9 (p < 0.001), and P53 (p < 0.001) (Fig. 3B). Western blot analysis (Fig. 3C) showed that 100 nM digitoxin significantly decreased MCL1 protein expression and upregulated CASP9 protein expression compared with the DMSO control (both p < 0.05). Taken together, these findings indicate that digitoxin inhibits porcine follicular development by promoting apoptosis in GCs.

Fig. 3.

Effect of 100 nM digitoxin on porcine follicle morphology and expression of proliferation /apoptosis markers. A Representative images of follicles at 1, 3, and 5 days after treatment (n = 3 biological replicates). B Relative mRNA expression of key genes involved in cell proliferation and apoptosis in digitoxin‑treated follicles (n = 3 biological replicates). C Protein expression of selected proliferation‑ and apoptosis‑related markers in digitoxin‑treated follicles (n = 3 biological replicates). Data are presented as mean ± SEM. A p < 0.01 (**) indicates a highly significant difference and p < 0.001 (***) indicates an extremely significant difference, while “ns” denotes not significant

Effect of digitoxin on murine ovarian development

In the experimental design, four different concentrations of digitoxin were administered: 1.5 mg/kg, 2.5 mg/kg, 3 mg/kg, and 4 mg/kg. Starting from 22 days of age, female mice received intraperitoneal injections of digitoxin every other day until they reached 42 days of age. At the 4 mg/kg dose, 100% mortality occurred after two injections. After three weeks of treatment, the 2.5 mg/kg group showed the most significant reduction in serum estrogen levels (p < 0.001; Fig. 4A). Therefore, the concentration of 2.5 mg/kg was selected for subsequent experiments.

Fig. 4.

Effect of digitoxin on mouse ovarian development. Data are from n = 3 biological replicates (one histological section/sample per mouse ovary) as indicated and analyzed after 3 weeks. A Effect of various concentrations of digitoxin (1.5-3 mg/kg) or DMSO on E2 levels of mouse serum. (B)-(G) 21-day-old mice were treated with 2.5 mg/kg digitoxin for 3 weeks. B H&E staining of ovaries showing follicle development (PF, preantral follicle; AF, antral follicle; CL, corpus luteum). Representative images and quantification of ovarian follicles. Follicles were counted on one histological section per mouse ovary (n = 3). Scale bar: 500 μm. C TUNEL staining of ovarian sections demonstrating the occurrence of cell apoptosis in mice ovaries. Representative images and quantification of TUNEL intensity in ovarian follicles. TUNEL intensity was analyzed on one histological section per mouse ovary (n = 3). Scale bar: 500 μm. D-E Serum FSH and LH levels in mice (n = 3). F Relative mRNA expression of genes involved in ovarian proliferation and apoptosis pathways (qPCR; normalized to housekeeping gene; n = 3). G Protein expression of selected proliferation- and apoptosis‑related markers (representative Western blots and quantification). Data are presented as mean ± SEM. A p-value < 0.05 (*) was considered significant between groups, p < 0.01 (**) highly significant, p < 0.001 (***) extremely significant, and “ns” denotes not significant

The results indicated that, compared with the control group, the digitoxin (2.5 mg/kg)-treated mice exhibited a significant presence of antral follicles in their ovaries, suggesting that digitoxin did not adversely affect the progression of follicular development from the preantral to the antral stages. However, the number of corpora lutea in the 2.5 mg/kg digitoxin group was significantly lower than that in the control group (p < 0.05; Fig. 4B), indicating that follicle development was primarily arrested at the antral follicle stage. TUNEL staining further revealed that digitoxin promoted apoptosis in GCs within the murine ovaries (Fig. 4C). Additionally, serum levels of FSH, and LH were significantly decreased (p < 0.001; Figs. 4D and E).

qRT-PCR analysis demonstrated that digitoxin significantly suppressed the mRNA expression of key proliferation-related genes, including CREB1 (p < 0.001), CCNB1 (p < 0.001), PCNA (p < 0.01), and SP1 (p < 0.001), while promoting the expression of apoptosis-related genes such as CASP3 (p < 0.001), CASP8 (p < 0.001), CASP9 (p < 0.001), and BAX (p < 0.001) in the ovaries (Fig. 4F). Western blot analysis revealed that digitoxin significantly decreased PCNA protein levels (p < 0.05; Fig. 4G) and showed a trend toward increased CASP3 protein expression, although this was not statistically significant (p > 0.05; Fig. 4G). Taken together, these findings suggest that digitoxin inhibits ovarian development in mice.

Effect of digitoxin on mouse age at first estrus and litter size

Beginning at 22 days of age, female mice were administered intraperitoneal injections of digitoxin, and vaginal estrus was monitored daily. The findings revealed that the age at first estrus in the 2.5 mg/kg digitoxin group (34.5 ± 1.5 days) was significantly delayed compared with that of the control group (31.7 ± 2.3 days; p < 0.01; Fig. 5A). Body weight was recorded every other day, and after 2 days of treatment, we found that the mice in the 2.5 mg/kg digitoxin group exhibited significantly lower body weight than that of the control group, indicating a substantial effect of digitoxin on body weight (p < 0.001; Fig. 5B).

Fig. 5.

Effects of 2.5 mg/kg digitoxin on reproductive and growth outcomes in mice. Data are from n = 3 biological replicates (mice). A Age at first estrus onset. B Body weight of mice during the critical growth period (postnatal days 21–42). C Age at first delivery. D Age at second delivery. E-F Litter sizes of the first (E) and second (F) litters, as measures of reproductive performance. G Body weight of pregnant mice from days 44–72. Data are presented as mean ± SEM. A p-value < 0.05 (*) was considered significant between groups, p < 0.01 (**) highly significant, p < 0.001 (***) extremely significant, and “ns” denotes not significant

Digitoxin injections were discontinued at 42 days of age, after which the mice were housed for mating. The analysis revealed no significant difference in the age of first delivery between the 2.5 mg/kg digitoxin group (64.8 ± 1.2 days) and the control group (65.8 ± 0.8 days; p > 0.05; Fig. 5C). However, the litter size of the first litter in the 2.5 mg/kg group (4.2 ± 2.8 pups) was significantly smaller than that of the control group (7.7 ± 1.3 pups; p < 0.01; Fig. 5D). For the second litter, the age at delivery for the 2.5 mg/kg group (89.8 ± 3.2 days) was significantly later than that of the control group (86.3 ± 3.3 days; p < 0.05; Fig. 5E), and the litter size was also significantly reduced (4.8 ± 3.8 pups compared to 7.7 ± 1.7 pups in controls; p < 0.05; Fig. 5F). Additionally, after pregnancy, the body weight of the control group mice was significantly higher than that of the 2.5 mg/kg group (p < 0.05; Fig. 5G). In summary, the data suggest that digitoxin administration delays the onset of estrus in mice and reduces litter size.

Discussion

By utilizing both porcine in vitro culture and in vivo mouse models, our study is the first to elucidate systematically the multifaceted regulatory effects of the cardiac glycoside digitoxin on ovarian GC function and the female reproductive system. Our findings reveal that digitoxin may significantly impact ovarian physiological functions by inhibiting cell proliferation, promoting apoptosis, disrupting hormone secretion, and altering the dynamics of follicular development.

Regulation of GC proliferation and apoptosis by digitoxin

This study confirms that digitoxin significantly inhibits GC viability, probably through a dual mechanism. On one hand, it downregulates the expression of proliferation markers, such as PCNA and the cell cycle regulator CDK4, by approximately 40–60%. This finding is consistent with a previous study indicating that cardiac glycosides suppress CDK4 in breast cancer cells [39]. Moreover, MCL1, a critical survival regulator in GCs [40, 41], was reduced by about 50% in our study. This result agrees with findings from a previous study that documented the downregulation of MCL1 following cardiac glycoside treatment in ovarian cancer models [42]. The reduction of MCL1 might directly lead to changes in mitochondrial outer membrane permeability, which is consistent with the observed activation of the caspase cascade downstream.

Digitoxin elicited pro-apoptotic effects by activating both the intrinsic (CASP9-mediated mitochondrial pathway) and extrinsic (CASP8-mediated death receptor pathway) apoptosis pathways (Fig. 3). The upregulation of CASP3 and CASP7, which are common executioner caspases, indicates that apoptosis has progressed to an irreversible phase. This finding agrees with previously reported mechanisms of cardiac glycoside-induced apoptosis in cancer cells [43, 44]; however, our study is the first to demonstrate this effect in the primary cells of the reproductive system. Notably, the activation of CASP9 suggests that digitoxin initiates a DNA damage response, offering a novel perspective on its reproductive toxicity [45, 46].

Disruption of the hypothalamic-pituitary-gonadal (HPG) axis

Digitoxin treatment led to a significant decrease in serum E2 levels of approximately 35%, which is directly associated with impaired GC function. As the primary site of estrogen synthesis, the reduced activity of GCs inevitably results in diminished E2 production [47]. Additionally, concurrent reductions in FSH and LH levels, observed with a decrease of 20–25%, suggest that digitoxin also directly impacts the pituitary secretion of gonadotropins. This phenomenon might be linked to the inhibitory effect of cardiac glycosides on Na⁺/K⁺-ATPase, a pump that plays a crucial role in regulating the electrical activity of gonadotropin-releasing hormone (GnRH) neurons [48–51]. Furthermore, the delayed onset of first estrus observed in animal experiments (with an average delay of 3.5 days) might be linked to the suppression of hypothalamic Kisspeptin neuron activity [52, 53].

Altered follicular development dynamics and reduced fertility

The abnormal phenotype observed, characterized by a 30% increase in the number of antral follicles together with a 45% decrease in corpus luteum count, indicates that digitoxin disrupts the follicular dominance selection process. This disruption is associated with follicular atresia resulting from the increased apoptosis of GCs [54, 55], which might also involve abnormalities in the anti-Müllerian hormone signaling pathway [56]. Furthermore, the reduction in litter size (40% fewer pups in the first litter and 52% fewer in the second) suggests a cumulative reproductive toxicity of digitoxin. This toxicity might be linked to decreased oocyte quality or altered endometrial receptivity [57, 58].

Clinical implications and future research directions

The widespread use of cardiac glycosides in treating cardiovascular diseases [59, 60] presents a significant contradiction when considering the reproductive risks identified in this study. Digitoxin probably impairs female fertility through several mechanisms, including the inhibition of GC proliferation, the promotion of apoptosis, the reduction of E2 levels, and the disruption of the HPG axis.

Digitoxin and digoxin are cardiac glycosides that both inhibit the Na⁺/K⁺-ATPase, but they differ in their pharmacokinetics and potentially in their reproductive risk [61]. Digitoxin is more lipophilic, is hepatically metabolized, and has a longer half‑life, whereas digoxin is less lipophilic and is excreted renally [61]. Animal studies indicate both drugs can be embryotoxic or impair gonadal function at high doses, but direct comparative reproductive toxicity data are limited [62]. Because digitoxin tends to accumulate in tissues and provides prolonged exposure, it could theoretically pose greater fetal or gonadal risks than digoxin. Clinical human data are sparse; therefore, dose adjustment and targeted reproductive toxicity studies are needed to establish relative safety and to support informed clinical decision‑making [63].

A key limitation of the historical literature is that many reports of ″digitalis″ toxicity are based on botanical extracts (e.g., Digitalis spp., Nerium oleander) rather than from pharmacopeial, chemically purified digoxin or digitoxin [64]. Botanical preparations are complex mixtures containing multiple cardiac glycosides plus variable non-glycoside constituents, and they often show wide batch-to-batch variability in active-constituent content, bioavailability, and the presence of cofactors or contaminants that can alter toxicity [64]. In addition, dosing with herbal material is typically imprecise, and the pharmacokinetic behavior (absorption, distribution, and metabolism) of glycosides within an extract can differ from that of isolated pharmaceutical compounds. For these reasons, toxicity or reproductive effects reported after plant ingestion cannot be assumed to predict the risk profile of a pharmaceutical product of known purity and dose [65].

Wherever possible, controlled toxicology or clinical studies should utilize well-characterized, quantified materials (pure compounds or standardized extracts) and include analytical verification (e.g., high-performance liquid chromatography (HPLC)) of active constituent concentrations to facilitate meaningful comparisons.

Previous research has demonstrated that plant compounds resembling cardiac glycosides, such as TPL-Me-G, can adversely affect fertility in rats by lowering estrogen and progesterone levels [49]. Clinically, the long-term use of cardiac glycosides in women has been linked to an increased risk of breast cancer [66]. Therefore, for women of reproductive age who require prolonged digitoxin treatment (such as those with chronic heart failure), it is essential to enhance the monitoring of reproductive health. This should include regular assessments of the ovarian reserve, such as anti-Müllerian hormone testing and antral follicle count, together with hormone level evaluations. A comprehensive assessment should be undertaken to explore whether digitoxin can be substituted with alternative therapies that pose a lower risk to reproductive health.

On the other hand, the antiproliferative effect of digitoxin on GC highlights its potential utility in treating sex cord-stromal tumors, such as granulosa cell tumors (GCTs) [67]. Research has shown that proteins associated with cell proliferation, including PCNA, CDK4, and MCL1, are often overexpressed in GCT cells [68], with digitoxin significantly downregulating these molecules. Therefore, digitoxin might function as an adjuvant chemotherapeutic agent or might be useful in combination with paclitaxel or cisplatin to enhance therapeutic efficacy and mitigate resistance [69, 70]. Moreover, the employment of tumor-targeted delivery systems, such as nanoparticles or antibody-drug conjugates, might improve selectivity and minimize systemic toxicity [71].

This study has several limitations. First, the precise molecular mechanisms of digitoxin action remain incompletely defined [72]; proteomic analyses, affinity purification, and other biochemical approaches are needed to identify direct binding partners and to map downstream signaling pathways. Second, we did not assess a broader set of reproductive outcomes (e.g., ovulation rate, fertilization, implantation, and embryo development) [73]; the inclusion of these endpoints would allow for a more comprehensive evaluation of reproductive function. Third, our experiments used relatively high, acute doses; studies of chronic low‑dose exposure that better reflect potential environmental or dietary intake are required to assess realistic risk [74, 75]. Fourth, all in vivo work was conducted in mice; replication in larger animal models or nonhuman primates would strengthen the translational relevance of the findings [76].

Since cardiac glycoside compounds are naturally found in various plants, they might enter the human body through environmental pollution or the food chain. This study indicates that environmental contamination with cardiac glycosides may pose a potential threat to the reproductive health of both wildlife and humans, underscoring the need for further assessment of their ecological risks.

In conclusion, our study elucidates the multifaceted effects of digitoxin on GC function and female reproductive health, highlighting significant implications for fertility and reproductive toxicity. Whereas digitoxin is beneficial in treating cardiovascular diseases and certain tumors, the selection and monitoring of therapeutic doses is crucial, particularly for women of reproductive age requiring prolonged treatment. Furthermore, as a potential environmental pollutant, the reproductive toxicity of digitoxin should be considered for individuals of all ages, emphasizing the need for increased awareness and assessment of its ecological risks.

Abbreviations

BCA

Bicinchoninic acid

CCK-8

Cell Counting Kit-8

CDK4

Cyclin-dependent kinase 4

cDNA

Complementary DNA

CL

Corpus luteum

CV

Coefficient of variation

DEGs

Differentially expressed genes

DAPI

4',6-diamidino-2-phenylindole

DMEM

Dulbecco's Modified Eagle Medium

DMSO

Dimethyl sulfoxide

E2

Estradiol

EdU

5-ethynyl-2'-deoxyuridine

ELISA

Enzyme-linked immunosorbent assay

FACS

Fluorescence-activated cell sorting (flow cytometry)

FBS

Fetal bovine serum

FITC

Fluorescein isothiocyanate

FSH

Follicle-stimulating hormone

GC / GCs

Granulosa cell(s)

GCT

Granulosa cell tumor(s)

GSEA

Gene Set Enrichment Analysis

GO

Gene Ontology

GnRH

Gonadotropin-releasing hormone

H&E

Hematoxylin and eosin

HPLC

High-performance liquid chromatography

HPG

Hypothalamic-pituitary-gonadal

HRP

Horseradish peroxidase

KEGG

Kyoto Encyclopedia of Genes and Genomes

MCL1

Myeloid cell leukemia 1

OD

Optical density

PCNA

Proliferating cell nuclear antigen

PBS

Phosphate-buffered saline

PF

Preantral follicle

PI

Propidium iodide

PVDF

Polyvinylidene difluoride

qRT-PCR

Quantitative real-time polymerase chain reaction

RIN

RNA Integrity Number

RIPA

Radioimmunoprecipitation assay

RNA-seq

RNA sequencing

s.e.m. (SEM)

Standard error of the mean

SDS-PAGE

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis

TUNEL

Terminal deoxynucleotidyl transferase dUTP Nick End Labeling

Authors’ contributions

X.Y., J.W., and B.M. contributed to the study of design. Y.J., M.L., and Y. S. conducted experimental studies. Y.J. and Y.Z. performed data analysis. Y.J. drafted the manuscript, while X.Y. and B.M. completed the manuscript editing. All authors reviewed and approved the final version of the manuscript.

Funding

This work was supported by the Science and Technology Project of Guangzhou (2024B03J1305), the Guangdong Basic and Applied Basic Research Foundation (2024B1515020112, 2024A1515012999, and 2023A1515030054), and the Breed Industry Innovation Park of Guangdong Xiaoerhua Pig (2022-4408X1-43010402-0019).

Data availability

The raw data from RNA-seq experiments have been uploaded to the SRA database (submission: PRJNA1269214) and will be released upon acceptance of the manuscript. Additionally, all other original data related to this study can be obtained from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All animal procedures were conducted in accordance with the “Instructive Notions with Respect to Caring for Laboratory Animals” (Ministry of Science and Technology of the People’s Republic of China). Approval was obtained from Sun Yat-sen University Cancer Center (permit no. L025504202306002).

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.