Ethics statement

Ethical approval for this study was obtained from the Institutional Review Board of Kaohsiung Veterans General Hospital (Approval ID: KSVGH21-CT1-43). The study conformed to the ethical principles set forth in the Declaration of Helsinki, ensuring the protection of participants’ welfare, rights, and privacy. The review process included a thorough assessment of the study protocol, covering aspects such as participant recruitment, informed consent procedures, data management, and confidentiality safeguards. This comprehensive ethical review underscores our commitment to maintaining the highest standards of research integrity and participant protection throughout the study. All participants provided written informed consent. This study was registered on ClinicalTrials.gov (identifier: NCT06546774).

Participants and inclusion/exclusion criteria

The study recruited infertile women undergoing frozen embryo transfer (FET) cycles at our reproductive center. The inclusion criteria targeted patients between 35 and 45 years of age with a body mass index (BMI) of 18 to 35 kg/m². Exclusion criteria were comprehensive, excluding patients with a history of oophorectomy, primary ovarian insufficiency, congenital uterine anomalies, or severe intrauterine adhesions. Additionally, patients undergoing preimplantation genetic testing for aneuploidy (PGT-A), recipients of donated oocytes, cancer patients, and those who had used hormone treatments or supplements in the past three months were also excluded. A total of 161 women were included and assigned to either the melatonin or control groups. Participants in the melatonin group received a daily dose of 2 mg of melatonin (Somn Well XR Tablet® by Biofrontier Inc., Asia, Taiwan) for at least eight weeks before commencing their IVF cycles. The analysis was conducted separately for all participants aged ≥ 35 years and for a subset of older women aged ≥ 38 years to evaluate the effects of melatonin supplementation.

Protocols for controlled ovarian stimulation and frozen embryo transfer

In this study, all participants underwent a GnRH antagonist protocol for controlled ovarian stimulation. Initial evaluations included hormone screening and transvaginal ultrasound to assess antral follicle counts. Ovarian stimulation commenced within 5 days of the menstrual cycle using a combination of recombinant follicle-stimulating hormone and recombinant luteinizing hormone (Pergoveris, Merck Serono SA, Aubonne, Switzerland). Patient response was continuously monitored with serial transvaginal ultrasounds and hormone level assessments. Dosage of the stimulants was adjusted based on the ovarian response.

Daily GnRH antagonist injections (Cetrotide 0.25 mg, Pierre Fabre Medicament Production, Aquitaine Pharm International, Idron, France) were started when the leading follicle reached 12–14 mm in diameter and continued until the final oocyte maturation. A dual-trigger approach was employed, consisting of combined recombinant hCG (Ovidrel 250 µg, Merck Serono S.p.A., Modugno, Italy) and a GnRH agonist (Lupro 2 mg, Nang Kuang Pharmaceutical Co., Ltd., Tainan, Taiwan), when at least one follicle reached a mean diameter of 18 mm. Transvaginal ultrasound-guided oocyte retrieval was performed 36 h after trigger administration. Fertilization was carried out using either conventional IVF or intracytoplasmic sperm injection (ICSI), based on the results of semen analysis or previous fertilization outcomes.

Embryo quality on Day 3 was assessed according to standards established by the Istanbul consensus workshop, evaluating the degree of fragmentation, uniformity of blastomeres, and the presence of multinucleation. Embryos were categorized into Grades 1 to 3. In this study, top-quality Day 3 embryos were those with 6 to 10 cells and classified as Grade 1. This grade was characterized by less than 10% fragmentation, consistent blastomere morphology, and no evidence of multinucleation. For Day 5 embryos, quality was assessed using the Gardner and Schoolcraft scoring system, which includes evaluation of expansion degree (grades 1–6), inner cell mass morphology (grades A-C), and trophectoderm morphology (grades A-C). Top-quality Day 5 embryos were classified as Grade 3AA or better. All embryos were cryopreserved using the vitrification technique under a freeze-all strategy for subsequent FET.

Endometrial preparation was conducted using hormone replacement therapy (HRT). The process began on day 2 or day 3 of the menstrual cycle with daily oral estradiol doses of 6–8 mg (Ediol, Synmosa Biopharma Corporation, Hsinchu County, Taiwan) in combination with estradiol gel (Oestrogel gel, Besins, Drogenbos, Belgium). After 14 days of estrogen therapy, endometrial thickness was measured using a 2-dimensional vaginal ultrasound (Voluson E8, GE Healthcare, Chicago, USA). Once the endometrial thickness reached at least 7 mm, luteal phase support was initiated. This included daily intravaginal administration of 90 mg progesterone gel (Crinone 8% gel, Merck Serono, Hertfordshire, UK), daily oral administration of 30 mg dydrogesterone (Duphaston, Abbott, Olst, the Netherlands), and weekly intramuscular injections of 125 mg progesterone (Progeston Depot, Tafong Pharmaceutical Co., Ltd., Changhua City, Taiwan). For embryo transfer, cleavage-stage embryos were thawed and transferred on the fourth day following progesterone initiation, whereas blastocysts were thawed and transferred on the sixth day. The embryo transfer procedure was meticulously conducted under transabdominal ultrasound guidance to ensure precise placement. Upon confirmation of pregnancy, progesterone supplementation was maintained until the gestational period reached 10–12 weeks.

Outcomes measure for IVF-FET cycles

The primary endpoint of this study was the live birth rate, defined as the birth of a viable infant after 24 weeks of gestation. Secondary endpoints included clinical pregnancy rate, ongoing pregnancy rate, and miscarriage rate. Clinical pregnancy was identified through the presence of a fetal heartbeat detected via transvaginal sonography at 6–7 weeks of gestation. Ongoing pregnancy was characterized by the continuation of the pregnancy beyond 12 weeks of gestation. Miscarriage was defined as the loss of a pregnancy following the detection of fetal cardiac activity but occurring before 24 weeks of gestation.

Human cumulus cell collection

Following oocyte retrieval, the cumulus-oocyte complexes were collected and washed before being transferred into IVF medium under paraffin oil for further processing. To isolate the cumulus cells, they were initially treated with 40 IU/mL hyaluronidase (SynVitro™ Hyadase, Origo, Knardrupvej, Denmark) for 3 min, followed by thorough washing with phosphate-buffered saline (PBS). Subsequently, the cumulus cells underwent mechanical dissociation and were washed again. The isolated cumulus cells were then resuspended in Histopaque 1077 (Sigma-Aldrich, Waltham, MA, USA) and enriched with a culture medium containing 10% fetal bovine serum (Gibco, Thermo Fisher Scientific, Waltham, MA, USA), 5 mg/L insulin, 5 mg/L transferrin, 5 µg/L sodium selenite (ITS, Sigma), and 1.25 µM androstenedione (4-androstene-3, 17-dione, Sigma). The resuspended cumulus cells were plated into a 4-well plate at a concentration of 2 × 10⁴ viable cells per well. They were then incubated at 37.5 °C in a humidified environment with 5% CO₂ for up to 24 h to prepare for subsequent experiments.

Protocol for cell culture and treatment

The human ovarian granulosa cell line (HGL5) was obtained from Applied Biological Materials Inc. These cells were cultured under standard conditions in a humidified incubator, maintaining a temperature of 37 °C and 5% CO₂. The culture medium consisted of 10% fetal bovine serum, 1% penicillin/streptomycin, 2% Ultroser G (Pall Corp.), and 1% ITS Plus (Zen-Bio) to support optimal cell growth. The HGL5 cell line was utilized as a model to study cellular senescence, which was induced through serial passaging. In this research, cells that had undergone passages 59 and 113 were deemed senescent. To induce oxidative stress, HGL5 cells were treated with hydrogen peroxide (H₂O₂). To assess the potential protective effects of melatonin, cells were pretreated with 25 µM melatonin for 20 h before exposing them to 0.8 mM H₂O₂ for 4 h. The objective was to investigate how melatonin alleviates oxidative stress-induced damage in HGL5 cells by comparing the effects observed in melatonin-treated versus untreated cells.

Mitochondrial function measurement

Mitochondrial function was assessed using established assays as described previously [20]. Cells from each experimental condition were collected and subjected to a series of staining procedures with specific fluorescent probes designed to assess distinct aspects of mitochondrial function. The following probes were used: CellROX (5 µM) to quantify oxidative stress, DCFDA (5 µM) for general reactive oxygen species (ROS) detection, MitoTracker Green (10 nM) to measure mitochondrial mass, tetramethylrhodamine methyl ester (TMRM; 200 nM) to evaluate mitochondrial membrane potential, and MitoSOX (5 µM) to detect mitochondrial superoxide levels (all probes sourced from Molecular Probes). Staining was performed at 37 °C in a controlled incubator to maintain optimal conditions for probe performance. After staining, cells were thoroughly washed with PBS to eliminate any unbound fluorescent dye. The washed cell pellets were then subjected to centrifugation to concentrate the cells, which were subsequently resuspended in PBS for further analysis. Flow cytometry was employed to quantify the fluorescence intensity of the various probes, providing precise measurements of mitochondrial function and integrity.

Oxygen consumption rate measurement

Oxygen consumption rate (OCR) was measured as outlined in a previous study [26]. This analysis was conducted with an extracellular flux analyzer (Agilent Technologies, Santa Clara, CA, USA) utilizing the Seahorse XF HS Mini platform. This platform is designed to provide precise measurements of cellular respiration by monitoring OCR in real-time. For the assays, cells were cultured in specialized assay plates at a density of approximately 2,000 cells per well. During the analysis, OCR was continuously recorded to track dynamic changes in cellular respiration over time. To assess different facets of mitochondrial function, a sequential treatment approach was utilized. Initially, oligomycin was administered to inhibit ATP synthase, thereby measuring the basal respiration and ATP-linked respiration. Following this, FCCP (carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone) was introduced to uncouple oxidative phosphorylation and determine the maximal respiratory capacity of the mitochondria. Finally, antimycin A and rotenone were applied to block the electron transport chain, thereby quantifying the non-mitochondrial respiration and assessing the residual respiration due to other cellular activities. This multi-step approach provided a detailed profile of mitochondrial function, including basal respiration, maximal respiratory capacity, and non-mitochondrial respiration, contributing to a comprehensive understanding of cellular energy metabolism.

Western blotting

Western blotting was performed as previously described [27]. The analysis utilized primary antibodies specific for Total OXPHOS (Abcam, catalog number ab110411), NRF2 (Cell Signaling Technology, catalog number 33649), KEAP1 (Proteintech, catalog number 10503-2-AP), and alpha-tubulin (GeneTex, catalog number GTX628802. The Western blot procedure involved protein extraction, SDS-PAGE separation, and transfer to a nitrocellulose membrane. Following the transfer, the membrane was incubated with the aforementioned primary antibodies, and detection was achieved using appropriate secondary antibodies and chemiluminescent substrates. This methodology provided insights into the expression levels and interactions of the targeted proteins, contributing to a deeper understanding of their roles in cellular processes.

RNA extraction and real-time polymerase chain reaction

RNA was extracted using REzol (Protech Technology), a reagent known for its efficacy in isolating high-quality RNA from diverse sample types. Following extraction, mRNA expression levels were quantified using SYBR Green-based quantitative real-time PCR (qRT-PCR) performed on the StepOne system (Applied Biosystems), which is recognized for its precision in gene expression analysis. For accurate normalization of the qRT-PCR data, RNU6-1 was used as the reference gene. RNU6-1 is a small nuclear RNA commonly used as an internal control due to its stable expression across different sample conditions, which ensures consistency and reliability in the measurement of target gene expression. Detailed information regarding the primer sequences utilized for the qRT-PCR assays is provided in Supplementary Table S1.

Statistical analysis

For the analysis of patient data, the Kolmogorov–Smirnov test was employed to assess the normality of continuous variables. Quantitative data were compared using independent t-tests, whereas categorical variables were analyzed with either the Chi-square test or Fisher’s exact test, as appropriate. All statistical analyses were performed using SPSS version 20.0 (Statistical Package for the Social Sciences, Chicago, IL, USA), with a significance threshold set at P < 0.05. In the experimental study, data were derived from a minimum of three independent replicates, and results are presented as the mean ± standard error of the mean. Statistical analysis was performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA). Two-way ANOVA, followed by Tukey’s post hoc test, was used to determine significant differences between group means, with statistical significance set at P < 0.05.