Collection of granulosa cells

Six healthy Merino lambs (aged 6 weeks) were selected from the Sheep Research Center, XinJiang Academy of Animal Sciences to extract follicular GCs. Lambs were treated with or without FSH to obtain two groups: 3 control lambs (CLs) and 3 lambs treated with FSH (FLs). For the FL group, lambs were intraperitoneally injected with 40 IU FSH four times every 12 h. Ovaries were collected under anaesthesia with xylidinothiazoline (2 mg/kg body weight), and follicular fluid was then obtained. GCs were harvested from follicular fluid and centrifuged at 1200 × g for 8 min.

Experimental procedures were carried out in full accordance with the Guide for the Care and Use of Animals of XinJiang Academy of Animal Sciences. All experimental procedures were approved by the Ethics Committee of the XinJiang Academy of Animal Sciences.

RNA sequencing

Total RNA, including mRNA and miRNA, was extracted from GCs in 3 CLs and 3 FLs using TRIzol Reagent (Invitrogen, USA). The quality and purity of the RNA were assessed using a NanoDrop ND-2000 (Thermo Fisher Scientific, USA) and Agilent 2100 Bioanalyzer (Agilent Technologies). For mRNA sequencing, the NEBNext® Ultra™ Directional RNA Library Prep Kit for Illumina® (NEB, USA) was used to generate the sequencing library, strictly following the manufacturer's instructions. For miRNA, the sequencing library was generated using the NEBNext® Small RNA Library Prep Set for Illumina® (NEB). The constructed libraries were then subjected to high-throughput sequencing on the Illumina HiSeq 2500 platform.

Differential expression and enrichment analysis

The quality control of sequencing data was performed using the FastQC tool. For mRNAs, reads were aligned to the sheep genome using HISAT2 software. Then, we used featureCounts to count the number of reads for each gene. For miRNAs, reads were aligned to the miRBase database to determine the expression of miRNAs. The analysis of differentially expressed mRNAs (DE-mRNAs) and miRNAs (DE-miRNAs) between CLs and FLs was performed using the DESeq2 software package, with a threshold set at p value < 0.05 and |log2FC (fold change)|> 1.

GO (Gene Ontology) and KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment analyses were performed for DE-mRNAs using the clusterProfiler R package. A p value < 0.05 was set as the significance criterion.

Construction of regulatory networks for miRNAs and mRNAs

Target mRNAs for DE-miRNAs were predicted using the miRanda database [11] and RNAhybrid database [12]. Identification of intersections with differentially expressed ferroptosis-related genes in the predicted miRNA target mRNAs was performed. The regulatory network was constructed based on the regulatory relationships between miRNA and its target mRNAs. The network was visualized using Cytoscape software.

Culture, transfection and treatment of granulosa cells

GCs were harvested from small follicles (1–2 mm in diameter) and large follicles (≥ 3 mm in diameter) in control lambs. GCs were resuspended in DMEM/F12 culture medium containing 10% foetal bovine serum and 1% antibiotic antifungal agent. Then, GCs were cultured at 37 ℃ and 5% CO2 in culture plates containing 2 × 105 cells per well.

The oar-miR-134-3p mimics, inhibitor, and NC (normal control) were purchased from Gemma Medical Technology Co., Ltd. (Shanghai, China). GCs from large follicles were seeded in a 24-well plate and transfected with oar-miR-134-3p mimics (100 nmol/μl), oar-miR-134-3p inhibitor (100 nmol/μl), and oar-miR-134-3p NC (100 nmol/μl) using Lipofectamine 3000 (Invitrogen, Shanghai, China) for 48 h. Subsequently, GCs were treated with or without erastin (ferroptosis inducer; Beyotime, Shanghai, China) for 24 h.

Cell proliferation analysis

Cells were seeded into a 96-well plate at approximately 5 × 103 cells per well. Then, the cells were further cultured to allow them to attach to the plate. Cell proliferation was assessed using a CCK-8 kit (Cell Counting Kit-8; Beyotime). First, 10% CCK-8 solution was added to each well, and then, the plate was returned to the incubator for an additional 24 h. Absorbance was measured at 450 nm using a microplate reader.

Dual-luciferase reporter assays

The region of HMOX1's 3'UTR containing the oar-miR-134-3p target site was cloned and inserted into the pmirGLO luciferase miRNA expression reporter vector (Gemma), creating the WT-HMOX1 reporter plasmid. Moreover, we mutated the target site to create the MUT-HMOX1 reporter plasmid. Subsequently, HEK293T cells were seeded into 24-well plates at approximately 5 × 104 cells per well. Once the cells adhered to the well, we transfected them using Lipofectamine™ 2000 (Invitrogen) following the manufacturer's instructions. Each well was transfected with 50 ng of either the WT-HMOX1 or MUT-HMOX1 reporter plasmid and 10 nM oar-miR-134-3p mimic or NC. After 48 h of transfection, we measured luciferase activity using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer's instructions.

ELISAs

Cell culture medium was collected and used to perform ELISAs for MDA and GSH according to the manufacturer's instructions (Biosharp, Wuhan, China).

Analysis of ROS and mitochondrial membrane potential

GCs were seeded in a 24-well plate and stained with dihydroethidium (DHE, 3 μM, Biosharp), followed by incubation at 37 °C for 20 min. With a fluorescence microscope, red fluorescence was observed in the TRITC channels. The mean red fluorescence was determined as the value of ROS.

GCs were seeded in a 24-well plate and stained with JC-1 dye (5 μM, Biosharp), followed by incubation at 37 °C for 20 min. With a fluorescence microscope, green and red fluorescence were observed under FITC and TRITC channels, respectively. The ratio of red to green fluorescence was determined as the value of the mitochondrial membrane potential.

qRT‒PCR analysis

Total RNA was extracted from GCs using TRIzol Reagent (Invitrogen). One microgram of total RNA was used for reverse transcription reactions, with the miScript II RT Kit (QIAGEN, China) used for miRNAs and the PrimeScript™ RT reagent Kit (TaKaRa, Japan) for mRNAs, following the manufacturer's instructions. For miRNAs, the miScript SYBR Green PCR Kit (QIAGEN) was used, and for mRNAs, SYBR Premix Ex Taq™ II (TaKaRa) was used to perform qRT‒PCR detection. Each reaction had a total volume of 20 μl, including 10 μl of SYBR Green PCR Master Mix, 0.8 μl of forward and reverse primers (10 μM), 2 μl of cDNA and 7.2 μl of RNase-free water. The qRT‒PCR was run on a StepOnePlus™ Real-Time PCR System (Applied Biosystems), with the reaction conditions set at 95 °C predenaturation for 5 min, followed by 40 cycles of 95 °C denaturation for 10 s and 60 °C annealing and extension for 30 s. Each sample was run with three technical replicates. U6 and B2M were used as the internal reference genes for miRNAs and mRNAs, respectively. The relative expression levels were calculated using the 2−ΔΔCt method. Primers are shown in Table S1.

Western blot

Proteins were extracted from tissue samples using RIPA protein extraction buffer. The protein concentration was determined using the BCA method. The extracted proteins were mixed with SDS sample buffer and denatured by heating at 100 °C. Equal amounts of denatured protein samples were loaded onto the gel sample slots and subjected to SDS‒PAGE electrophoresis. PVDF was activated by soaking in methanol or transfer buffer. Protein transfer to the membrane was performed using a wet transfer system for 2 h. The membrane was blocked with 5% nonfat milk to prevent nonspecific binding. The membrane was incubated with primary antibodies specific to HMOX1 (ABclonal, Wuhan, China; No. 3522110806; 1:2000) and β-actin (ABclonal; No. 9100; 1:2000) at 4 °C for 8 h. After the membrane was washed, it was incubated with a fluorescent or enzyme-labelled secondary antibody at room temperature for 2 h. The chemiluminescent substrate was used to develop a detectable signal from the labelled secondary antibody (Abcam; No. GR3422216-4; 1:5000). The membrane signal was detected using a chemiluminescence imaging system. The detected signal was quantitatively analysed using ImageJ software (NIH, MD, USA). The signal intensity of HMOX1 was compared to that of the internal reference protein (β-actin) to calculate the relative expression level of HMOX1.

Statistical analysis

Every experiment was performed with at least three biological replicates, and the data are presented as the means ± standard errors of three replicates. Statistical analysis was performed using GraphPad Prism v.9 (GraphPad, CA, USA). Student’s t test (two-tailed) or one-way analysis of variance was used for comparisons. A p value < 0.05 was considered to indicate statistical significance.