Loss of FUT8 impairs embryonic development by reducing cAMP production in granulosa cells

Abstract

Purpose

In cases of impaired fertility, the quality of oocytes and their subsequent embryonic development following fertilization are critical concerns in clinical practice. Core fucosylation, catalyzed by fucosyltransferase 8 (FUT8), is a common N-glycosylation modification that plays a key role in cell proliferation and signal transduction. This study examines the role of core fucosylation in follicular regulation during oocyte development.

Methods

Patients were categorized into two groups based on their embryo formation rates. Core fucosylation levels, catalyzed by FUT8, were assessed in serum and follicular fluid (containing granulosa cells). To investigate the molecular effects of FUT8 on oocyte quality, FUT8-knockdown human granulosa cells (KGN-KD), and ovaries from Fut8 gene knockout (Fut8^−/−) mice were analyzed.

Results

Core fucosylation levels were lower in patients with reduced blastocyst formation rates. FUT8 depletion led to decreased cAMP production following follicle-stimulating hormone (FSH) stimulation in both KGN-KD and the ovaries of Fut8^−/− mice. Additionally, the expression of FIGLA (folliculogenesis-specific basic helix-loop-helix transcription factor) and other genes associated with embryonic development was downregulated in the ovaries of Fut8^−/− mice. Oocytes from Fut8^−/− mice exhibited abnormal zona pellucida formation and impaired embryonic development.

Conclusion

These findings indicate that FUT8 ablation may disrupt embryonic development post-fertilization by reducing FSH receptor (FSHR) signaling and downregulating FIGLA expression. The results suggest that core fucosylation is essential for female reproduction and oocyte quality.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-025-03536-z.

Introduction

A high-quality oocyte is essential for the formation of a healthy human embryo. Despite advancements in assisted reproductive technology (ART) and other embryo-related techniques, oocyte quality remains a key determinant of embryo development outcomes [1]. Many patients undergoing ART experience not only a limited number of mature follicles and oocytes but also embryo loss [2], which remains a significant clinical challenge. Although extensive research has been conducted on oocyte development, the factors influencing oocyte quality are still not fully understood.

The ovarian follicular microenvironment comprises the oocyte and its surrounding granulosa cells [3]. Since oocytes rely on granulosa cells for most of their energy metabolism and biosynthesis substrates, their communication is crucial [4]. Activation of the transmembrane receptor FSHR (follicle-stimulating hormone receptor) on granulosa cells by FSH is fundamental to follicular development, ensuring synchronized development of both granulosa cells and oocytes [5]. FSHR is part of the N-glycosylated G protein-coupled receptor family [6], and N-glycosylation modifications influence its function, playing a critical role in female fertility [7, 8]. Fucosyltransferase 8 (FUT8) catalyzes the addition of fucose to the innermost N-acetylglucosamine (GlcNAc) residue of N-glycans in the Golgi apparatus, facilitating core fucosylation (Fig. 1A). This process is essential for cell proliferation and development [9]. Without core fucosylation, transmembrane receptors cannot function properly, which in turn disrupts downstream signaling pathways and ultimately affects normal cellular activities and functions [10]. The role of core fucosylation in follicular development and oocyte quality has yet to be fully elucidated.

Fig. 1.

Core fucosylation modifications are down-regulated in sera from patients with low blastula formation rate. A, Biosynthesis of core fucosylation by Fut8. Blue box indicates the core N-glycan structure. B, The mRNA expression of FUTs by RT-qPCR. The mRNA expressions of glycosyltransferases between the ovarian granulosa cells pooled from ten patients in Group B or Group A were compared after normalization to GAPDH. Data are presented as mean ± SEM. **p < 0.01. C, Core fucosylation levels in follicular fluid. The follicular fluid samples from each patient in Group A (n = 12) and Group B (n = 10) were probed with AOL (1:8000). D, The significant differences of the core fucosylation levels in C compared to total protein levels using CCB staining as control are shown as mean ± SD triplicates. ***p < 0.001

Following FSH stimulation, the cyclic adenosine monophosphate (cAMP) signaling pathway downstream of FSHR is activated and is essential for follicular growth and development [11]. Granulosa cells serve as the primary source of cAMP in oocytes, transporting it through transzonal projections at the ends of transmembrane synapses [12]. As a key second messenger, cAMP regulates multiple signaling pathways and transcription factors. During oocyte development, gene expression is precisely regulated by a network of transcription factors that are active from the early stages of follicular development [13, 14]. Among these, the folliculogenesis-specific basic helix-loop-helix transcription factor (FIGLA) is crucial for oocyte differentiation and also regulates the expression of zona pellucida (ZP) proteins [15, 16]. Studies on FIGLA-knockout mice have shown a complete absence of the ZP and impaired embryonic development [17]. However, the regulatory mechanisms controlling FIGLA expression remain unclear. Previous findings have indicated that core fucosylation influences follicular development, as Fut8-knockout (Fut8^−/−) mice exhibit abnormal ZP formation in their oocytes. Further investigation is needed to determine whether FUT8 affects oocyte development and its potential for subsequent embryonic development.

In this study, Fut8^−/− mice and FUT8-knockdown human granulosa cells were used to investigate the effect of FUT8 on female infertility and its underlying mechanism. The findings indicate that FUT8 ablation leads to reduced oocyte quality. Core fucosylation plays a role in follicular-oocyte development by regulating the FSHR-cAMP signaling pathway and influencing transcription factors associated with oocyte quality.

Materials and methods

Clinical samples

Serum and follicular fluid (including granulosa cells) samples were collected from 40 women recruited from Dalian Municipal Women and Children’s Medical Center. Serum samples were obtained by centrifuging whole blood at 4500 rpm to remove blood cells and were then stored at − 20 °C. Oocyte retrieval was performed using transvaginal ultrasound, selecting only mature follicles 36 h after human chorionic gonadotropin (hCG) injection. Follicular fluid and granulosa cell samples were collected directly through aspiration and immediately centrifuged at 3000 rpm for 10 min. Follicular fluid samples were filtered and protein concentrations were measured. The cell pellet was resuspended in 0.01% phosphate-buffered saline (PBS), and granulosa cells were isolated using Ficoll gradient centrifugation, followed by digestion with hyaluronidase (Sigma) for 10 min at 37 °C. Red blood cells were removed using a 50% Percoll gradient and the remaining granulosa cells were further purified by centrifugation before being processed for protein or RNA extraction. To minimize inter-group variability, granulosa cells from 10 patients were pooled to create a composite sample for each group. All participants were between 30 and 40 years old, with other disorders and factors related to ovarian function or sperm function disorder, other potential factors related to chromosome dysgenesis and ovarian dysfunction excluded. Written informed consent was obtained from all patients. The study procedures and protocols were approved by Dalian Municipal Women and Children’s Medical Center (Approval No: 2020008).

Mice

Heterozygous Fut8^± mice on an ICR background were bred to obtain Fut8^+/+ and Fut8^−/− mice, which were used for experiments at 8 weeks of age. All mice were housed under a 12-h light/dark cycle (08:00–20:00) at a temperature of 24 °C ± 1 °C, with free access to food and water. The animals were maintained in a specific-pathogen-free (SPF) facility at Dalian Medical University. All experimental procedures and protocols were approved by the Ethics Committee of Dalian Medical University (Approval No: AEE17013).

Antibodies

Biotin-labeled Alocasia oryzae lectin (AOL) was obtained from Vector Laboratories, CA. Anti-ZP3 (G-1) (Cat No. sc-398359) was sourced from Santa Cruz, USA, while anti-FSHR Ab (Cat No. 22665–1-AP) was purchased from Proteintech, USA. Horseradish peroxidase (HRP)-conjugated streptavidin (for AOL) (Cat No. 434323) and Alexa Fluor 594-conjugated (Cat No. A-11012) or 488-conjugated (Cat No. A-11008) goat anti-rabbit IgG (H + L) were acquired from Thermo Fisher Scientific, USA.

Generation of FUT8 knockdown and rescue KGN cells

To establish a stable FUT8 knockdown granulosa cell line (KGN-KD), a lentiviral shRNA approach was employed (Supplementary Fig. 2). A FUT8-targeting shRNA sequence (sense: 5'-TCTCAGAATTGGCGCTATG-3') was cloned into the pLKO.1 backbone, and lentiviral particles were produced using the psi-LVRU6GP system (GeneCopoeia). Following viral infection, KGN cells were cultured in medium containing puromycin to select for successfully transduced cells. For restoration of FUT8 expression, KGN-KD cells were subsequently transduced with a lentiviral vector carrying the full-length FUT8 cDNA. Cells with re-expressed FUT8 (KGN-KD-Re) were isolated using hygromycin selection.

Overexpression of mutant FSHR

pEZ-M14 expression vectors were constructed to generate four specific FSHR mutants: FSHR-N191Q, FSHR-N199Q, FSHR-N293Q, and FSHR-N318Q, obtained from GeneCopoeia (China) (Supplementary Fig. 3A). These mutant vectors were transfected into 293 T cells using Lipofectamine 2000, following the manufacturer’s protocol.

The FSHR extracellular domain (FSHR-ECD) overexpression vector, pEZ-M35-FSHR (Supplementary Fig. 3B), was synthesized by GeneCopoeia (China) and transfected into KGN and KGN-KD cells using Lipofectamine 2000.

Western and lectin blotting

Cells or tissues were lysed in radioimmunoprecipitation assay (RIPA) buffer supplemented with an enzyme inhibitor cocktail (Santa Cruz, CA). Total protein concentrations were measured using the bicinchoninic acid (BCA) protein assay kit (Pierce, Rockford, IL). For mouse ovaries, proteins were extracted separately from each ovary of Fut8^+/+ and Fut8^−/− mice. In the case of human serum and follicular fluid, protein concentrations from each patient were measured directly before loading.

Protein samples (10 ug) were loaded and separated using 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). Electrophoresis was performed at 110 V for 120 min, followed by protein transfer onto PVDF membranes (0.2 μm, Millipore Corp., Belford, USA). Membranes were blocked and incubated overnight at 4 °C with the primary antibody or AOL. After washing, membranes were incubated with either an HRP-labeled secondary antibody or HRP-conjugated streptavidin at room temperature for 1 h. Protein signals were detected using an ECL system (Amersham, Sweden).

FSH-FSHR interaction

To purify the FSHR-ECD, 10 plates of cell extract were gathered and purified by Anti-Flag Affinity gel purified kit (Sangon, Shanghai). Protein was resolved by SDS-PAGE and western blotted. The FSHR-ECD overexpressed in KGN or KGN-KD cells was immobilized onto a CM5 Chip. Recombinant human FSH (rhFSH) was diluted in PBS at varying concentrations. The interaction was measured with a 120-s contact time at a flow rate of 30 μl/min. All assays were conducted at 25 °C. The binding affinity of rhFSH to FSHR-ECD was determined using Biacore T200 and expressed as the equilibrium dissociation constant (K_D). Experimental data were analyzed using the Biacore T200 instrument.

In vitro ovarian culture

Ovarian tissue fragments(1 mm³) from each ovary of Fut8^+/+ and Fut8^−/− mice were placed in culture plates containing 1 mL of minimum essential medium (MEM) supplemented with insulin (10 ng/ml), transferrin (5 ng/ml), selenium (5 ng/ml), glutamine (2 mM), hypoxanthine (2 mM), human serum albumin (HSA) (1.25 mg/ml), ascorbic acid (50 mg/ml), rhFSH (50 mg/ml), epidermal growth factor (50 ng/ml), penicillin (100 mg/ml), and streptomycin (100 mg/ml) [18, 19]. The ovarian fragments were cultured for 10 days, with half of the medium replaced every 2 days. The lysates from ovarian cells were used for cAMP assays or real-time quantitative polymerase chain reaction (RT-qPCR).

RNA-seq analysis

For mouse ovaries, RNA was extracted separately from each individual mouse (two ovaries per mouse). For human granulosa cells, due to the limited cell count, samples from 10 patients were pooled per group. Only degree A samples with RIN > 9 were used for RNA-seq analysis. The total RNA quantity and integrity were assessed using the Bioanalyzer 2100 system (Agilent Technologies, USA) with the RNA Nano 6000 Assay Kit. Differential expression analysis was performed using a log₂ (fold-change) threshold. KEGG enrichment analysis was conducted using the David 6.8 database (http://david.ncifcrf.gov/home.jsp), and the results were visualized using ImageGP (http://www.ehbio.com/ImageGP/index.php).

cAMP assay

The GloSensor cAMP Assay Kit (Promega, USA) was used to measure cAMP levels induced by rhFSH. Cells were seeded in 6-well plates for 24 h before testing. The culture medium was replaced with a serum-free medium for 12 h overnight. For cAMP inhibitor, 10 µM TDI-10229 were included in the buffer. To inhibit phosphodiesterase activity, 100 μM 3-isobutyl-1-methylxanthine (IBMX, Sigma) was added before rhFSH (100 IU/L) stimulation. cAMP levels were then measured according to the kit’s standard protocol. For mouse ovaries, tissue was cultured before treatment with 100 μM of IBMX for 20 min. cAMP levels in individual ovaries from Fut8^+/+ and Fut8^−/− mice were measured using the GloSensor cAMP Assay Kit following 500 IU/L rhFSH stimulation for 30 min.

Immunofluorescence and confocal microscopy

Oocytes from Fut8^+/+ and Fut8^−/− mice were washed gently in PBS containing 0.05% polyvinylpyrrolidone (PVP), permeabilized with 0.5% Triton X-100 (in PBS/PVP) for 5 min, and rapidly washed three times in PBS/PVP. They were then fixed in 3.7% paraformaldehyde (PFA) (in PBS/PVP) for 20 min and blocked with 1% bovine serum albumin (BSA) (in PBS/PVP) at room temperature for 1 h. After blocking, oocytes were incubated overnight at 4 °C with primary antibodies. After washing, oocytes were incubated with secondary antibodies and DAPI before imaging with a confocal laser scanning microscope (Leica).

RT-qPCR

Total RNA was extracted from ovaries of Fut8^+/+ and Fut8^−/− mice or cells using RNAiso Plus (Takara) following the manufacturer’s instructions. For mouse ovaries, RNA was extracted separately from each mouse (two ovaries per sample). Due to the limited number of human granulosa cells, samples from 10 patients were pooled per group. Reverse transcription of RNA into complementary DNA (cDNA) was performed using the PrimeScript RT Master Mix Perfect Real Time Kit (Takara). RT-qPCR was conducted to measure the expression levels of target genes. RT-qPCR primers were obtained from Primer Bank or designed using Primer 5 software. Primer sequences are provided in Supplementary Table 1.

Dual-luciferase reporter gene assay

The Dual-Luciferase Reporter assay was conducted following standard protocols using the Dual-Luciferase Assay Kit (Promega, Madison, WI, USA). FIGLA sequences, ZP3 promoter sequences, and their respective mutant forms lacking docking sites were constructed, subcloned into a luciferase enzyme reporter vector, and designated as pcDNA3.1-FIGLA, Pgl3-ZP3-WT, and Pgl3-ZP3-MT. The plasmids were verified through sequencing. Relative luciferase activity was measured and each experiment was repeated three times.

In vitrofertilization (IVF)

Mice were injected with 10 IU PMSG to stimulate follicular development and followed by 10 IU hCG after 48 h to induce superovulation. Cumulus-oocyte complexes were collected from the oviducts 14 h after hCG injection and treated with hyaluronidase for 2 min to remove cumulus cells.

Oocytes from Fut8^+/+ and Fut8^−/− mice were placed in 100-μl drops of G-IVF™ PLUS medium (Vitrolife), covered with paraffin oil. Sperm from Fut8^+/+ and Fut8^−/− male mice were prepared at a final concentration of 5 × 10⁴ sperm/mL and added to the drops containing oocytes. Embryo development was monitored daily, assessing progression to the 2-cell, 4–8-cell, morula, and blastocyst stages from day 1 to day 5 of culture.

Statistical analysis

Statistical analysis and comparisons of experimental data were performed using GraphPad Prism (GraphPad Software Inc., La Jolla, CA, USA). A non-parametric t-test was used to compare differences between two groups, while one-way ANOVA was applied for comparisons among multiple groups. Results are expressed as mean ± standard deviation (SD, n ≥ 3). A p-value of < 0.05 was considered statistically significant.

Results

Loss of FUT8 is associated with a reduced blastocyst formation rate

In this study, clinical patients who underwent IVF or intracytoplasmic sperm injection (ICSI) cycles were categorized into two groups: the normal blastocyst formation rate group (Group A) and the low blastocyst formation rate group (typically below 30%) (Group B). Compared to Group A, Group B exhibited fewer antral follicles and mature oocytes, along with a lower blastocyst formation rate. However, no significant differences were observed between the two groups in terms of age, body mass index, luteinizing hormone, anti-Müllerian hormone, or progesterone levels (Supplementary Fig. 1A). To explore the potential link between low blastocyst formation rate and glycosylation, we analyzed the mRNA expression of FUT family glycosyltransferases in follicular fluid (containing granulosa cells) from both groups. FUT8 expression was reduced in Group B, while the expression levels of other glycosyltransferases remained relatively unchanged (Fig. 1B and Supplementary Fig. 1B).

Core fucosylation, catalyzed by FUT8, is the most prevalent N-glycosylation modification. In Group B, serum core fucosylation were lower (Supplementary Fig. 1C), as demonstrated by lectin blotting using AOL, which specifically binds to core-fucosylated N-glycans. Since follicular fluid plays a crucial role in follicular development, it is notable that core fucosylation levels in Group B were significantly reduced compared to Group A (Fig. 1C and D). Previous research has shown that FUT8 deletion results in a decreased number of oocytes and severely impacts fertility. These findings further suggest that core fucosylation levels are associated with a low blastocyst formation rate.

FUT8 ablation reduces cAMP production following FSH stimulation

As a G protein-coupled receptor, FSHR is located on the surface of granulosa cells and contains four potential N-glycosylation sites at N¹⁹¹, N¹⁹⁹, N²⁹³, and N³¹⁸ [20, 21]. FSHR, a typical seven-transmembrane receptor, possesses a hormone-binding domain that activates several key downstream pathways, including the cAMP signaling pathway. To investigate the role of N-glycosylation in FSHR function, we generated four FSHR mutants (N¹⁹¹Q, N¹⁹⁹Q, N²⁹³Q, and N³¹⁸Q) along with FSHR-WT in 293 T cells (Fig. 2A and Supplementary Fig. 3A). The molecular weight of FSHR was reduced in FSHR-N¹⁹¹Q and FSHR-N²⁹³Q mutants compared to FSHR-WT, whereas FSHR-N199Q and FSHR-N318Q mutants showed no changes (Fig. 2B). This indicates that N¹⁹¹ and N²⁹³ undergo N-glycosylation modifications. To further examine the role of core fucosylation in FSH-FSHR interactions, the FSHR-ECD was overexpressed in KGN-KD and KGN cells (Fig. 2C and Supplementary Fig. 2, Supplementary Figs. 3B and C). The FSHR-ECD was purified using a Flag-tag column, and core fucosylation levels were assessed. Results showed that FSHR-ECD in KGN-KD cells lacked core fucosylation (Fig. 2D). The affinity between rhFSH and FSHR was evaluated using a Biacore assay, revealing that loss of FUT8 weakened the interaction between FSH and FSHR (Fig. 2E).

Fig. 2.

Loss of FUT8 decreases FSH-FSHR interaction. A, Immunofluorescence staining of FSHR-N¹⁹¹Q, FSHR-N¹⁹⁹Q, FSHR-N²⁹³Q, FSHR-N³¹⁸Q mutant-expressing 293 T cells were stained with anti-FSHR (red) (1:500) and DAPI (blue). B, Western blot. Cell lysates of mutant FSHR-expressing 293 T cells were probed with anti-FSHR (1:1000). GAPDH was used as the loading control. C, Generation of KGN, KGN-KD and KGN-KD-Re cells. Cell lysates were probed with AOL (1:8000). The membrane was stained with CBB as a loading control. D, Core fucosylation of FSHR-ED was suppressed in KGN-KD-FSHR cells. FSHR-ED protein was purified with an anti-Flag Affinity gel purification kit, equal amounts of the two purified protein groups were probed with FSHR (1:1000) and shown as controls above, and probed with AOL (1:5000) below, demonstrates the core fucosylation modification on the proteins. E, FSH-FSHR interaction. FSH-FSHR-ECD interaction was measured with a Biacore T200. FSH (ranging from 8.75 to 280 μM) was added and affinity for the de-fucosylated FSHR-ECD was compared to the FSHR-ECD. Data are presented as mean ± SEM triplicates, **p < 0.01

The cAMP signaling pathway is a key downstream pathway of FSHR, playing a crucial role in follicle development and oocyte maturation. Following FSH stimulation, cAMP levels were lower in KGN-KD cells compared to KGN cells, but this reduction was reversed in KGN-RE cells upon FUT8 re-expression (Fig. 3A). The absence of core fucosylation in FSHRs reduced their FSH-binding ability and cAMP production in granulosa cells. RNA-seq analysis revealed that the cAMP signaling pathway was downregulated in Group B compared to Group A (Fig. 3B). Additionally, cAMP levels were significantly lower in Fut8^−/− ovaries than in Fut8^+/+ ovaries (Fig. 3C). When Fut8^+/+ and Fut8^−/− ovaries were treated with 100 μM of IBMX (a phosphodiesterase inhibitor) for 20 min, followed by FSH stimulation, cAMP levels increased markedly in Fut8^+/+ ovaries, but not in Fut8^−/− ovaries (Fig. 3D). These findings suggest that core fucosylation modulates FSH-FSHR interactions, thereby influencing FSHR-dependent cAMP production and ultimately affecting the developmental competence of oocytes.

Fig. 3.

Loss of FUT8 attenuates the cAMP signaling pathway. A, FSH-induced cAMP production. The cAMP levels in KGN, KGN-KD and KGN-KD-Re cells were measured with a GloSensor cAMP Assay kit after stimulation with 100 IU/L of rhFSH for 5 min. B, Enriched KEGG terms or canonical pathways associated with the significantly downregulated transcripts. RNA was isolated from granulosa cells of ten patients per group with normal or low blastula formation rate. C, Enriched KEGG terms or canonical pathways associated with the significantly downregulated transcripts. RNA was isolated from of each individual mouse (two ovaries) from Fut8^+/+ or Fut8^−/− mice. D, FSH-induced cAMP production was measured. The cAMP levels in each ovary were measured with a GloSensor cAMP Assay kit. Data are presented as mean ± SEM triplicates, *p < 0.05, **p < 0.01, ***p < 0.001

FUT8 ablation reduces FIGLA gene expression

FIGLA is the first identified transcription factor in oocytes, playing a key role in ZP formation and oocyte gene regulation. cAMP is a crucial intracellular signaling molecule that influences various pathways involved in ovarian development, with granulosa cells serving as the primary cAMP source for oocytes [22]. Given that both FIGLA and cAMP are present in the early stages of follicular development, we examined their relationship. The FIGLA gene expression in Fut8^−/− mice was significantly lower than in Fut8^+/+ mice (Fig. 4A). Notably, in Fut8^+/+ mice (Supplementary Fig. 4A), FIGLA expression decreased to levels comparable to Fut8^−/− mice following treatment with the cAMP inhibitor TDI-0229 [23] (Fig. 4B). These findings suggest a strong correlation between cAMP levels and FIGLA gene expression.

Fig. 4.

Loss of FUT8 suppresses Figla expression. A, Quantitative real-time PCR. RNA was isolated from cultured ovaries of each individual mouse (two ovaries) from Fut8^+/+ or Fut8^−/− mice after rhFSH stimulation. The mRNA expression levels of Figla were compared after normalization to GAPDH. Data are presented as means ± SEM triplicates. B, Quantitative real-time PCR. RNA was isolated after rhFSH stimulation from cultured ovaries each individual mouse (two ovaries) from Fut8^+/+ or Fut8^−/− mice treated with or without the cAMP inhibitor TDI-0229 (5 h), and Figla mRNA expression levels were compared after normalization to GAPDH. Data are presented as means ± SEM triplicates. C, Dual-luciferase reporter gene assay. The regulatory effect of FIGLA on the ZP3 gene promoter was determined. ****p < 0.0001

FUT8 ablation downregulates ZP3 expression via FIGLA

The ZP is essential for follicular and oocyte development. In mice, the ZP consists of ZP1, ZP2, and ZP3, whereas in humans, it includes ZP1, ZP2, ZP3, and ZP4. Among these, ZP3 is a key component in both species and plays a crucial role in oocyte embryonic development [24, 25]. A Dual-Luciferase Reporter assay confirmed that the transcription factor FIGLA binds to the ZP3 promoter, initiating its transcription, as evidenced by a significant increase in luciferase activity (Fig. 4C). Compared to Fut8^+/+ oocytes, the ZP of Fut8^−/− oocytes was thinner and structurally altered (Fig. 5A). Immunofluorescence analysis further demonstrated a reduction in ZP3 expression in Fut8^−/− oocytes (Fig. 5B), as well as in Fut8^−/− ovaries (Fig. 5C and Supplementary Fig. 4B). As shown in Fig. 5D, mRNA expression levels of ZP3 and ZP2 were lower in Fut8^−/− mice. Additionally, IVF experiments revealed a unique ZP shedding phenomenon in Fut8^−/− oocytes (Supplementary Fig. 4C). These findings indicate that FUT8 ablation reduces cAMP levels, impairing ZP formation and oocyte quality.

Fig. 5.

Loss of FUT8 reduces ZP formation in Fut8^−/− mice. A, ZP thickness. The ZP of Fut8^+/+ and Fut8^−/− oocytes was imaged by confocal microscopy. Scale bar = 100 µm. Significant differences are shown as the means ± SD, n = 5. B, Immunofluorescence assay. Fut8^+/+ and Fut8^−/− mature oocytes were stained with AOL (green) (1:1000), anti-ZP3 (red) (1:500) and DAPI (blue). C, Western blot. Lysates of Fut8^+/+ and Fut8^−/− ovaries were probed with anti-ZP3 (1:2000). GAPDH was used as the loading control. D, Quantitative real-time PCR. RNA was isolated from ovaries of Fut8^+/+ and Fut8^−/− mice. The mRNA expression levels of Zp1, Zp2 and Zp3 were compared. The relative expression of mRNAs was normalized to GAPDH. Data are presented as the means ± SEM triplicates

In vitro-fertilized Fut8^−/− embryos arrest at the 4–8-cell stage

During primordial follicle development, the ZP is gradually secreted by oocytes into the extracellular space. In germinal vesicle (GV) stage oocytes, the ZP contains pseudopod-like structures, known as transzonal projections (TZPs), which extend from granulosa cells [26]. These TZPs facilitate the exchange of small molecules, such as cAMP, between granulosa cells and oocytes [3]. In Fut8^−/− oocytes, the number of TZPs was reduced, and the gap distance between the ZP and oocyte membrane was increased (Fig. 6A). These findings indicate that FUT8 loss impairs TZP formation, subsequently disrupting oocyte development.

Fig. 6.

Loss of FUT8 results in impaired embryonic development. A, Immunofluorescence staining of TZPs (red) with phalloidin (1:500) in GV stage oocytes. B, Quantitative real-time PCR. RNA was isolated from ovaries of Fut8^+/+ and Fut8^−/− mice. The mRNA expression levels of Sohlh1, Lhx8, Ddx, Nobox, and Kit were compared. The relative expression of mRNAs was normalized to GAPDH. Data are presented as the means ± SEM triplicates. C, IVF assay. Ovulated Fut8^+/+ oocytes (upper panel) and Fut8^−/− oocytes (lower panel) were cocultured with Fut8^+/+ sperm, and representative micrographs of 2-cell embryos, 4–8 cell embryos, morulae, and blastocysts were observed. Significant differences are shown as the means ± SD, n = 4. D. Model for FUT8 regulation of oocyte development in the follicular microenvironment. In ovaries from patients with low blastula formation rate, decreased fucosylation on FSHR reduces the interaction between FSH and FSHR in granulosa cells. The attenuated cAMP signaling pathway suppresses the expression of the FIGLA gene and ZP3 in oocytes, and subsequent abnormal follicle development

During oogenesis, gene expression in oocytes undergoes dynamic regulation by germ cell-specific transcription factors. Key transcription factors, including spermatogenesis- and oogenesis-specific basic helix-loop-helix 1 (SOHLH1), LIM homeobox 8 (LHX8), DEAD-box helicase 4 (DDX4), NOBOX oogenesis homeobox (NOBOX), receptor tyrosine kinase (KIT), and FIGLA, play a role in inducing stem cell differentiation into oocytes [27]. Since these transcription factors are critical for oocyte development, their expression levels were examined in mouse oocytes. Compared to Fut8^+/+ mice, the expression levels of Sohlh1, Lhx8, Ddx, Nobox, and Kit were reduced in Fut8^−/− mice, whereas Lhx8 expression was increased (Fig. 6B). IVF of Fut8^−/− oocytes resulted in abnormal embryonic development (Fig. 6C). While some fertilized embryos progressed to the 2-cell stage, none developed further to the 4–8-cell stage. In mating trials, Fut8^−/− female mice were completely sterile, as they failed to produce offspring. In contrast, Fut8^−/− male mice exhibited normal fertility, producing offspring comparable to controls (Supplementary Fig. 4D). These findings underscore the critical role of core fucosylation in female reproduction.

Discussion

Oocyte quality is a key determinant of embryonic development and female fertility, as abnormal oocyte development can result in embryonic death [28, 29]. N-glycosylation plays a crucial role in female reproductive health. For instance, DPAGT1, the enzyme responsible for the initial step of protein N-glycosylation, is essential for oocyte development [7]. Additionally, the role of N-glycosylation in FSH and its receptors has been documented [8, 30]. Specifically, in eel FSHR, the N-glycosylation site N¹⁹¹ is required for PKA signaling activation through FSHR [31]. Our previous study revealed that clinically poor ovarian response is associated with reduced core fucosylation [2]. In the present study, we found that loss of FUT8-mediated core fucosylation weakened the FSH-FSHR interaction and impaired cAMP signal transduction the FSHR in granulosa cells, ultimately leading to abnormal embryonic development (Fig. 6D).

The follicular microenvironment consists of granulosa cells and oocytes. In ART, patients receive injections of rhFSH [32]. FSH binds to the FSHR-ECD, which contains four potential N-glycan sites at N¹⁹¹, N¹⁹⁹, N²⁹³, and N³¹⁸ [21, 33, 34]. N-glycans are essential for FSHR expression on the granulosa cell surface and for high-affinity binding to FSH [35]. Core fucosylation, a common N-glycosylation modification, plays a critical role in cell growth and signal transduction [36]. Our previous study demonstrated that core-fucosylated N-glycans enhance the FSH-FSHR interaction without affecting FSHR expression [2]. Upon FSH stimulation, FSHR signaling is tightly regulated in granulosa cells. However, loss of core fucosylation weakens FSHR signal transduction and decreases cAMP production after FSH stimulation. cAMP, a key intracellular second messenger, regulates multiple signaling pathways, and its primary source in oocytes is granulosa cells [22]. Transcriptomic analysis further confirmed the reduction of cAMP signaling in patients with low blastocyst formation rates and in Fut8^−/− granulosa cells. Collectively, these findings indicate that core fucosylation primarily regulates the FSH-FSHR interaction and cAMP production, both of which are essential for granulosa cell function and follicular development.

Oocytes lack many of the enzymes required for survival, making them dependent on granulosa cells for metabolic processes, such as glycolysis and various signal transduction pathways [37]. Due to the close interaction between granulosa cells and oocytes, cAMP plays a crucial role in oocyte development. The reduction in cAMP levels in Fut8^−/− granulosa cells negatively affected follicular and oocyte development. In Fut8^−/− ovaries, FIGLA expression was downregulated following FSH stimulation. However, in FUT8-knockdown KGN cells, FIGLA mRNA expression remained unchanged after treatment with the cAMP inhibitor TDI-0229, suggesting that FUT8 regulates cAMP production through the FSHR pathway, which in turn influences FIGLA expression. A Dual-Luciferase Reporter assay confirmed that FIGLA binds to the ZP3 promoter, and in Fut8^−/− ovaries, the reduced FIGLA levels led to the suppression of ZP3 expression. The ZP separates oocytes from granulosa cells in the primary follicle [38]. In Fut8^−/− oocytes, the ZP was thinner and distorted, creating an abnormally large gap between the ZP and the oocyte membrane. Granulosa cell TZPs extend into the ZP and form connections with oocytes, allowing the exchange of small molecules, including cAMP [39]. However, in Fut8^−/− follicles, TZPs failed to reach the oocyte membrane, disrupting communication between granulosa cells and oocytes, ultimately impairing embryonic development. Given that ZP3 is indispensable for the oocyte-to-embryo development [25], ZP3 reduction in Fut8^−/− oocytes also provides a mechanistic explanation for the observed arrest of embryos derived from these oocytes.

Core fucosylation levels are reduced in the granulosa cells and follicular fluid mixture of patients with a low blastocyst formation rate. Although Fut8^−/− oocytes can undergo normal fertilization and develop to the two-cell stage, embryonic death occurs immediately afterward. To investigate the role of FUT8 in oocyte development, we analyzed the expression of transcription factors and genes involved in follicular embryogenesis in Fut8^−/− mice. Oocyte development occurs in two stages, each characterized by distinct gene expression patterns [13]. During the first stage, genes responsible for oocyte self-development are highly expressed and their expression is regulated by a network of transcription factors, including FIGLA, NOBOX, LHX8, SOHLH1, KIT, and ZP3. Proper gene expression in this stage is crucial for oocyte quality and embryonic developmental potential [40–42]. The second stage involves genes related to oocyte meiosis and ovulation, and their abnormal expression can lead to ovulatory defects or post-ovulatory oocyte immaturity. Among these transcription factors, FIGLA plays a central role in oocyte transcriptional regulation [43] and is essential for guiding oocytes through normal cellular processes [27]. In Fut8^−/− follicles, FIGLA expression was suppressed, along with reduced expression of Nobox, Lhx8, Sohlh1, Kit, and Zp3. This suggests that the loss of FUT8 disrupts FIGLA expression, impairs ZP formation, and ultimately affects embryonic development following fertilization.

In conclusion, our findings indicate that FUT8 regulates the FSHR signaling pathway and cAMP production in granulosa cells. The absence of FUT8 reduces cAMP production, leading to abnormal follicular development and decreased embryonic developmental potential of oocytes. These results highlight the essential role of core fucosylation in female fertility and reproductive health. Future studies with larger, more diverse cohorts will be necessary to validate and extend our findings.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

M.L., W.L. and T.W. conceived the project and designed the experiments. T.W., Z.Z., C.Q. and analyzed the data and performed the clinical sample collection and data analysis. M.L., T.W., S.C. and Y.S. analyzed and interpreted the data. All authors have read and approved the final manuscript.

Funding

This work is supported by the National Natural Science Foundation of China (32171279), Basic and Applied Basic Research Foundation of Guangdong Province (2024 A1515012796), and Provincial Science and Technology Innovation Strategy Special City and County Science and Technology Innovation Support Project (STKJ2023008).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Since clinical information for patients admitted to the hospital was provided anonymously by clinicians, patient consent was not obtained. This study was approved by Dalian Municipal Women and Children's Medical Center with the ethics code 2020008.

All animal experimental procedures and protocols were approved by the Ethics Committee at the Dalian Medical University (approval No. AEE17013), and conducted in accordance with the institutional guides for the care and use of laboratory animals.

Consent for publication

Not applicable. This work does not contain data from any individual person.

Conflict of interest

The authors declare no conflict of interest.