Spermidine enhances steroidogenesis by elevating eIF5A^hyp level in hierarchical follicular granulosa cells of the goose

Abstract

Imbalanced steroid hormone levels in female poultry can impair follicular development, thereby adversely affecting reproductive performance. Spermidine is reported to be involved in regulation of female reproduction and can provide aminobutyl as the only substrate required for the activation of eukaryotic translation initiation factor 5A (eIF5A^hyp). In this study, goose granulosa cells were treated with spermidine and interference/overexpression eIF5A with the aim of investigating the effects and mechanisms of spermidine on steroidogenesis. Estradiol (E2) and progesterone (P4) levels increased with spermidine concentration, peaking at 160 μmol/L (P < 0.01). Compared with control, eIF5A overexpression significantly increased E2 levels (P < 0.05), but had no effect on P4. Moreover, compared with the eIF5A overexpression, 160 μmol/L spermidine combined with eIF5A overexpression treatment further increased E2 level (P < 0.05). Mechanistically, we found that both spermidine and eIF5A overexpression treatments significantly increased eIF5A^hyp as well as StAR protein expression (P < 0.05). These results suggest that spermidine can enhance steroidogenic capacity of goose granulosa cells, and this effect is mainly realized by mediating eIF5A^hyp. Notably, protein functional annotation revealed that differentially expressed proteins were significantly enriched in steroid biosynthesis and metabolism pathways. HMOX1, COQ9, and SERPINE2 were preliminarily identified as key proteins for spermidine-eIF5A^hyp axis-regulated steroidogenesis in goose granulosa cells.

Introduction

Poultry follicular growth and development is an extremely complex and delicate process, such as selection of dominant follicle, follicular atresia and follicular wall reorganization, those process are regulated by hormones and cytokines (Wadood et al., 2021). Steroids are synthesized by cholesterol and further converted into steroid hormones such as estradiol (E2), progesterone (P4), and testosterone (Miller and Auchus, 2011). E2 and P4, key sex hormones that regulate female reproductive efficiency, can bind to their receptors and activate different signaling pathways to inhibit follicular atresia and improve egg production (Ru et al., 2024; Sechman, 2013).

Polyamines, primarily including putrescine, spermidine, and spermine, participate in multiple fundamental cellular processes and homeostasis, such as cell differentiation, apoptosis, and maintenance of protein synthesis. Among them, spermidine, a representative member of polyamines, has been the most extensively studied (Casero, et al., 2018; Pisani, et al., 2024). Spermidine, a natural anti-inflammatory metabolite and biologically active polycation, is found in nearly all organisms (Michael, 2016; Pegg, 2009). It easily combines with negatively-charged biomolecules, such as DNA, RNA and proteins, to help regulate their biological function, which is critical for cell proliferation and differentiation (Lenis et al., 2017; Wei et al., 2022). DFMO irreversibly inhibits ODC, the key rate-limiting enzyme in cellular polyamine synthesis. This inhibition reduces intracellular polyamines, downregulates steroidogenic acute regulatory protein (StAR) expression, and decreases serum E2 and P4 levels in rats, leading to follicular developmental stagnation or atresia (Bastida et al., 2005; Khan et al., 2021). Moreover, we previously reported a significant positive correlation between polyamines and steroid hormone levels and their receptor gene expression during the development of hierarchical follicles in geese (Jiang et al., 2023a), suggesting that polyamines could regulate follicular development in female animals by mediating steroidogenesis. Further studies have revealed that spermidine can also alleviate oxidative stress and inhibit apoptosis in goose granulosa cells, thereby maintaining the normal functions of granulosa cells (Jiang et al., 2023b). However, the exact mechanism by which spermidine regulates steroidogenesis has not been well explained in previous studies.

Eukaryotic translation initiation factor 5A (eIF5A) is highly conserved and universally expressed in eukaryotic cells, which plays an important regulatory role in the initiation and extension of protein translation (Lubas et al., 2018). Spermidine acts as the only aminobutyl provider required for the activation of eIF5A to participate in the regulation of the hypusination modification process of eIF5A, thereby converting the inactive precursor eIF5A into activated eIF5A (eIF5A^hyp) (Levasseur et al., 2019; Puleston et al., 2019). Menon et al. (2014) showed that eIF5A can regulate ovarian function by mediating the expression of ovarian luteinizing hormone receptor genes. In addition, Gulappa et al. (2017) found that GC7 (eIF5A^hyp inhibitor) could reduce luteinizing hormone/choriogonadotropin receptor mRNA-binding protein expression, thereby promoting the expression of luteinizing hormone/choriogonadotropin receptor in rats. These results suggest that eIF5A^hyp is involved in the regulation of reproduction-related hormones. However, it is not clear whether eIF5A^hyp influences steroidogenesis. eIF5A^hyp is likely a critical participant in the regulation of steroidogenesis by spermidine. Therefore, in this research, exogenous spermidine was supplemented into goose granulosa cells, and eIF5A interference/overexpression was combined to explore the functions and regulatory mechanisms of spermidine and eIF5A in goose ovarian steroidogenesis. This is the first time to investigate the regulatory mechanism of the spermidine-eIF5A^hyp axis on steroidogenesis in granulosa cells of goose follicles, and it can provide theoretical references for enhancing follicular development, decelerating follicular atresia, and augmenting egg production in geese.

Materials and methods

Cell culture

The F1-F4 hierarchical follicles of Sichuan white geese under the same feeding environment were collected in a beaker containing PBS and transferred to cell culture room to isolate granulosa cells. The vascular layer on the surface of the follicle was removed. Subsequently, the white spot on the follicle was observed, and two-thirds of the yolk was discharged at the back of the white spot. The membrane layer was positioned in a glass culture dish, and one end was gently agitated to dissociate the granulosa cell layer. The isolated granulosa cell layer was then transferred into a small beaker placed on an ultra-clean bench. After performing mechanical shredding using scissors for 8 to 10 minutes, collagenase II was added for a digestion period of 5 minutes. Digestion was terminated by adding complete medium, and cells were filtered through a 70-μm filter. The cells were washed three times with DMEM/F12, 10 minutes per wash, and then resuspended in complete medium. The cell density was subsequently adjusted, and the cells were finally cultured in a cell incubator at 37°C with 5 % CO₂.

Cell viability assays

The activity of granulosa cells in 96-well plates following the corresponding treatment was determined. DMEM was utilized to prepare the medium solution with 10 % CCK-8. Subsequently, the original culture medium in the 96-well plate was removed, and 100 μL of DMEM containing 10 % CCK-8 was added to each well. After gentle shaking and mixing, the cells were incubated in an incubator with 5 % CO₂ at 37°C for 1-2 hours. The absorbance of each well at a wavelength of 450 nm was measured.

Enzyme linked immunosorbent assay

The DHPS enzyme activity in granulosa cells was detected. Briefly, protein samples were collected using RIPA buffer (containing 1 % PMSF), and the supernatant was obtained after centrifugation for the measurement of DHPS enzyme activity. The DHPS enzyme activity assay kit was purchased from Shanghai Enzyme-linked Biotechnology Co., Ltd (Shanghai, China). According to the kit instructions, blank wells, standard wells, and sample wells were set up. Subsequent experimental procedures included sample loading, incubation, washing, addition of enzyme-labeled reagent, color development, and reaction termination. The OD value of each well was measured at 450 nm, using a microplate reader (Multiskan SkyHigh, Thermo Fisher Scientific). A standard curve was plotted with the standard concentration as the abscissa and the OD value as the ordinate; the DHPS concentration in the samples was calculated using the equation derived from the standard curve. Moreover, the levels of E2 and P4 hormones in the granulosa cell culture medium were also determined. In short, following the instructions from Shanghai Enzyme-linked Biotechnology Co., Ltd. (Shanghai, China), the blank well, dilution standard well, and sample well to be tested were prepared. After adding the samples, a series of test operations including incubation, washing, and enzyme addition were carried out in sequence. The absorbance of each well at the final wavelength of 450 nm was measured. A standard curve was constructed, and the obtained absorbance value data was substituted into the formula to calculate the sample concentration. By multiplying with the corresponding dilution factor, the actual concentration of the sample under test was determined.

Construction of eIF5A interference sequences and overexpression vectors

si-eIF5A was designed and synthesized by Shanghai GenePharma Company Co., Ltd. (Shanghai, China), and the interference sequences are listed in Table 1. The pEGFP-eIF5A and pEGFP-N1 fluorescent plasmids were designed and synthesized by Shanghai Shenggong Biological Co., Ltd. (Shanghai, China), and the glycerol bacteria were obtained and stored at -20°C. According to the instructions of OMEGA plasmid extraction kit, the plasmid required for transfection was extracted.

Table 1.

Interference sequence information.

siRNA name	Primer sequences (5ʹ to 3ʹ)	Purification method
si-NC	F: UUCUCCGAACGUGUCACGUTT	HPLC
	R: ACGUGACACGUUCGGAGAATT
si-eIF5A	F: ACAUCAAGCGCUGCGACUUTT	HPLC
	R: AAGUCGCAGCGCUUGAUGUUG

Note: eIF5A: Eukaryotic translation initiation factor 5A.

Cell transfection

When the granulosa cells reached 70-90 % density in the culture plate, according to the instructions of Lipofectamine 3000 (Thermo Fisher Scientific), the corresponding reagents should be diluted to the appropriate concentration, then mixed with siRNA/pEGFP-eIF5A which diluted to the appropriate concentration. The treated granulosa cells were replaced with a complete medium supplemented with 10 % fetal bovine serum and devoid of antibiotics. After removing the original medium, PBS was carefully added to each culture well along the wall. The cells were then gently shaken and washed twice, followed by the addition of the newly prepared medium. Subsequently, the mixed reagent was added to each culture plate following the instructions. After gentle shaking and mixing, the plates were placed in a 37°C cell culture incubator with 5 % CO₂, and quantitative reverse transcription PCR and western blot were used to detect the transfection efficiency.

Quantitative reverse transcription PCR

The total RNA was extracted from the ovary and granulosa cells using the traditional Trizol method (Takara, Dalian, China), and the degradation of RNA was examined by 1-1.5 % agarose gel electrophoresis. In accordance with the instructions of the reverse transcription kit (Takara, Dalian, China), cDNA was synthesized through reverse transcription from the isolated total RNA. The optimal temperature was determined by evaluating the amplification efficiency and the dissolution curve using the SYBR Green q-PCR kit (Takara, Dalian, China), and primer information is presented in Table 2. Based on the instructions of the TB Green® Premix Ex Taq™ II kit (Takara, Dalian, China), a 10 μL reaction system was constructed using cDNA as a template TB Green Premix Ex Taq II 5 μL, forward primer 0.2 μL, reverse primer 0.2 μL, cDNA 1 μL, RNase-Free dH₂O 3.6 μL to perform the RT-qPCR reaction. The reaction conditions were 95°C for 30 seconds; 60°C for 30 seconds; and 72°C for 30 seconds, for a total of 35 cycles. After amplification, the amplification curve and melting curve were generated, and the Ct values were calculated to analyze the results. The relative expression of target genes was calculated with GAPDH serving as an internal reference gene.

Table 2.

Primer information.

Gene	Primer sequences (5ʹ to 3ʹ)	Annealing temperature (°C)	Product length (bp)
eIF5A	F: CGGCTTCGTGGTGCTCAAGG	60.5	153
	R: TTGTGGGTGGAGGGGCAGATG
StAR	F: AGAATCTTGACCTCTTTGACGCTG	59.6	87
	R: GAGACGGTGGTGGATAACGGA
CYP11A1	F: CATGCACATCCTGGAGAACTTCAAG	60.0	105
	R: AGCGTCAGGTAGATGGGTTTGTC
CYP19A1	F: CTGGTCCTGGTCTCGTGCGTAT	61.0	139
	R: GATGTGTCAAGCATGATCCGTCTC
CYP17A1	F: CTCACTGACACCAGCATCGG	61.0	102
	R: GGGCTTGTCCCACTCCTT
3β-HSD	F: TGTGACGTTCCTGTACCGTG	60.7	153
	R: TTACAACGGGTACACGCCTC
17β-HSD	F: TCCTTGGGTGCTATTGTC	57.4	169
	R: TGCTCCCTTGAGACTCTATC
GAPDH	F: GTGGTGCAAGAGGCATTGCTGAC	65.0	86
	R: GCTGATGCTCCCATGTTCGTGAT

Note: eIF5A: Eukaryotic translation initiation factor 5A; StAR: Steroidogenic acute regulatory protein; CYP11A1: Cytochrome P450 family 11 subfamily a member 1; CYP19A1: Cytochrome P450 family 19 subfamily a member 1; CYP17A1: Cytochrome P450 family 17 subfamily a member 1; 3β-HSD: 3β-hydroxysteroid dehydrogenase; 17β-HSD: 17β-Hydroxysteroid dehydrogenases; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

Western blot

Tissue and cell samples were lysed with RIPA buffer (containing 0.1 % PMSF). After centrifugation, the supernatant was collected. Subsequently, the BCA detection kit (Beyotime, Shanghai, China) was employed to determine the protein concentration, and the samples were then diluted to 2 mg/μL. Next, 1/3 volume of 4 × protein loading buffer (with 10 % β-Mercaptoethanol) was added, and the mixture was well-blended and incubated in a metal bath for 10 minutes. A 10 % polyacrylamide gel electrophoresis was utilized to separate the target protein, which was then transferred onto a nitrocellulose membrane (Bio-Rad, Shanghai, China). The nitrocellulose membrane was blocked with 5 % skim milk at room temperature and then incubated with primary antibodies at 4°C overnight. The relevant primary antibodies and their dilution ratios were as follows: anti-eIF5A (1:1000, Proteintech, China), anti-StAR (1:2000, Abmart, China), anti-eIF5A^hyp (1:2000, Sigma, USA), and rabbit anti-β-Tubulin (1:1000, ABclonal, China). After washing with TBST (Beyotime, Shanghai, China), the membrane was incubated with goat anti-rabbit immunoglobulin (IgG) (1:1000, Beyotime, China). The membrane was washed again. Finally, the target bands were visualized by ECL (Beyotime, Shanghai, China) using a gel imaging system. The relative protein content was calculated after analyzing the optical density values using Image Lab (Bio-Rad, CA, USA).

Proteomic analysis

Proteomics sequencing was performed by Hangzhou Lianchuan Biotechnology Co., Ltd. (Hangzhou, China), 4 different treatment groups, 4 biological replicates in each group, and a total of 16 samples were sequenced. The samples were ground in liquid nitrogen, lysis buffer was added, each group of samples was lysed by ultrasonication, and the supernatant was collected after centrifugation. BCA method was used for concentration measure of the total protein, SDS-PAGE was used for quality control, trypsin was used for protein digestion. Then the separated peptides were ionized by a nano-ESI and transferred to Orbitrap Exploris™ 480 mass spectrometer (Thermo Fisher Scientific, San Jose, CA) for DIA (Data independent Acquisition) mode detection. The dried peptide samples were reconstituted with 0.1 % formic acid and centrifuged at 20000 g for 10 min. Separation was performed by a Thermo Scientific Vanquish Neo UHPLC liquid phase system. The liquid-phase separated peptides were transferred to a Thermo Astral mass spectrometer for acquisition in DIA mode. Then preprocessed the samples and used timsTOF Pro2 spectrometry platform to obtain raw data by DIA mode. DIA-NN was used to analyze the DIA data with library-free method. Using Gene Ontology (GO) and kyoto encyclopedia of genes and genomes (KEGG)KEGG databases, functional annotation and pathway analysis were performed for total and differential proteins. Differential proteins were subjected to GO and KEGG pathway enrichment analysis using R software (version 4.5.0). The t-test function in R was employed to calculate significant p-values for inter-sample and inter-group differences. Significantly differentially expressed proteins were defined by the criteria: P < 0.05 and fold change (FC) > 1.2 for upregulation, or P < 0.05 and FC < 0.80 for downregulation.

Data processing and statistical analysis

The experimental data used in this paper were sorted out by Excel, and the data were analyzed by one-way ANOVA analysis of variance using IBM SPSS Statistics 27 software. Experimental data visualization was accomplished with GraphPad Prism 8.0 software and Origin 2024 software. The test results were expressed as Mean ± SD., and P < 0.05 indicated a significant difference.

Results

Effect of spermidine on E2 and P4 levels in goose granulosa cells

In order to explore whether spermidine affects steroidogenesis in goose granulosa cells, cells were treated with varying doses of spermidine. Results show that different concentrations of spermidine had no significant effect on the cell viability (P > 0.05; Fig. 1A). Treatment with 40-100 μmol/L spermidine could markedly decrease the DHPS enzyme activity, and activity reached nadir upon treatment with 100 μmol/L spermidine (P < 0.05; Fig. 1B). Upon further elevation of the spermidine concentration, the DHPS enzyme activity exhibited a significant increase. Notably, the DHPS enzyme activity peaked when treated with 160 μmol/L spermidine (Fig. 1B, P < 0.05). The results regarding the hormone levels in the medium demonstrated that, compared with the control group, treatment with 100 μmol/L spermidine could remarkably enhance the hormone levels of E2 and P4 (P < 0.05; Fig. 1C and D). Moreover, when treated with 160 μmol/L spermidine, the concentrations of E2 and P4 in the medium attained their maxima (P < 0.05; Fig. 1C and D). Based on the above results, 100 and 160 μmol/L spermidine were selected for subsequent experiments.

Fig. 1.

Impacts of varied concentrations of spermidine on granulosa cell viability, DHPS enzyme activity, and E2 and P4 levels in the cell culture medium. (A) Granulosa cell activity. (B) DHPS enzyme activity. (C) and (D) The levels of E2 and P4 hormones in cell culture medium. The experimental data were represented as the Mean ± SD. Different letters indicate a significant difference between the corresponding group and the other groups (P < 0.05).

eIF5A participates in spermidine-mediated steroidogenesis

As shown in Fig. 2A, transfection of pEGFP-eIF5A for 24 hours and 48 hours had no significant effect on the cell viability of granulosa cells (P > 0.05), while transfection of pEGFP-eIF5A for 72 hours significantly increased the cell viability (P < 0.05). Transfection of pEGFP-eIF5A to cells for 24 hours, 48 hours and 72 hours significantly increased the gene expression of eIF5A (P < 0.05; Fig. 2B). As shown in Fig. 2C, transfection of pEGFP-eIF5A to granulosa cells for 48 hours and 72 hours could significantly increase the protein expression of eIF5A (P < 0.05). Subsequently, granulosa cells treated with pEGFP-eIF5A for 48 hours were selected for subsequent experiments. Furthermore, compared with control group, 100 μmol/L spermidine treatment could significantly increase P4 hormone level in medium (P < 0.05), but had no significant effect on E2 level (P > 0.05; Fig. 2D and E). Transfection of pEGFP-eIF5A significantly increased the level of E2 (P < 0.05; Fig. 2D and F). Compared with pEGFP-eIF5A transfection, combined spermidine treatment led to an increase in the levels of E2 and P4, and 160 μmol/L spermidine significantly increased E2 levels (P < 0.05; Fig. 2F).

Fig. 2.

Effect of spermidine in combination with eIF5A overexpression on steroidogenesis. (A) Granulosa cell activity. (B) Gene expression of eIF5A. (C) eIF5A protein expression. (D-G) The effect of spermidine combined with overexpression eIF5A on E2 and P4. The experimental data were represented as the mean ± SD. Different letters indicate a significant difference between the corresponding group and the other groups (P < 0.05).

The expression of eIF5A^hyp and StAR is regulated by spermidine

The western blot results show that, 100 μmol/L spermidine significantly increased eIF5A^hyp protein expression compared to the control (P < 0.05; Fig. 3A and C). pEGFP-eIF5A significantly increased the protein expression of eIF5A, eIF5A^hyp and StAR (P < 0.05; Fig. 3A–D). As shown in Fig. 3E–H, compared with the control group, 160 μmol/L spermidine treatment significantly increased the protein expression of eIF5A^hyp and StAR (P < 0.05). pEGFP-eIF5A significantly increased the protein expression of eIF5A, eIF5A^hyp and StAR (P < 0.05; Fig. 3E–H).

Fig. 3.

The effects of the combination of spermidine and eIF5A overexpression on the protein expression levels of eIF5A, eIF5A^hyp and StAR. (A-D) Effect of 100 μmol/L spermidine combined with overexpression eIF5A on eIF5A, eIF5A^hyp and StAR protein expression. (E-H) Effect of 160 μmol/L spermidine combined with overexpression eIF5A on eIF5A, eIF5A^hyp and StAR protein expression. The quantification of eIF5A was performed with β-Actin as an internal reference, while the quantification of eIF5A^hyp and StAR were performed with β-Tubulin as an internal reference. The experimental data were represented as the mean ± SD. Different letters indicate a significant difference between the corresponding group and the other groups (P < 0.05).

Effects of spermidine combined with eIF5A overexpression on the expression of steroidogenesis-related genes

The expression of steroidogenesis-related genes is essential for maintaining normal follicular development within the avian follicle. As shown in Fig. 4, compared with the control group, 100 μmol/L spermidine treatment significantly reduced the gene expression of CYP11A1 and CYP19A1 (P < 0.05; Fig. 4B and D), and significantly increased the gene expression of 3β-HSD and 17β-HSD (P < 0.05; Fig. 4E and F), but had no significant effect on StAR and CYP17A1 (P > 0.05; Fig. 4A and C). eIF5A overexpression significantly reduced the gene expression of StAR, CYP17A1 and CYP19A1 (P < 0.05; Fig. 4A, Fig. 4, Fig. 4), and significantly increased the gene expression of CYP11A1, 3β-HSD and 17β-HSD (P < 0.05; Fig. 4B, Fig. 4, Fig. 4). Compared with the pEGFP-eIF5A group, the combined treatment of spermidine significantly reduced the gene expression of StAR, CYP11A1, CYP19A1 and 3β-HSD (P < 0.05; Fig. 4A, B, Fig. 4, Fig. 4), and significantly increased the gene expression of CYP17A1 and 17β-HSD (P < 0.05; Fig. 4C and F). In addition, compared with the control group, 160 μmol/L spermidine treatment significantly reduced the gene expression of CYP11A1, CYP17A1, CYP19A1 and 3β-HSD (P < 0.05; Fig. 4H–K), and significantly increased the gene expression of StAR (P < 0.05; Fig. 4G), but had no significant effect on 17β-HSD (P > 0.05; Fig. 4L). pEGFP-eIF5A transfection significantly reduced the gene expression of StAR, CYP17A1 and CYP19A1 (P < 0.05; Fig. 4G, Fig. 4, Fig. 4), and significantly increased the gene expression of CYP11A1, 3β-HSD and 17β-HSD (P < 0.05; Fig. 4H, Fig. 4, Fig. 4). Compared with pEGFP-eIF5A transfection group, spermidine combined treatment could significantly reduce the gene expression of CYP11A1, CYP19A1, 3β-HSD and 17β-HSD (P < 0.05; Fig. 4H, J, Fig. 4, Fig. 4), and significantly increase the gene expression of StAR and CYP17A1 (P < 0.05; Fig. 4G and I).

Fig. 4.

Effects of spermidine combined with overexpression eIF5A on the expression of steroidogenesis-related genes. (A) and (G) Gene expression of StAR. (B) and (H) Gene expression of CYP11A1. (C) and (I) Gene expression of CYP17A1. (D) and (J) Gene expression of CYP19A1. (E) and (K) Gene expression of 3β-HSD. (F) and (L) Gene expression of 17β-HSD. The experimental data were represented as the mean ± SD. Different letters indicate a significant difference between the corresponding group and the other groups (P < 0.05).

Proteomics profiles of goose granulosa cells treated with spermidine combined with eIF5A gene interference

To further explore the key regulatory protein in the spermidine-mediated regulation of steroidogenesis by eIF5A, we carried out proteomics sequencing of goose granulosa cells following si-eIF5A. As shown in Fig. 5A–C, interfering with the eIF5A gene in goose granulosa cells significantly reduced the protein expression of eIF5A and eIF5A^hyp compared with the control group (P < 0.01). After combined treatment with spermidine, the protein expression of eIF5A and eIF5A^hyp was significantly restored (P < 0.05). On this basis, we performed proteomic sequencing on goose granulosa cells. As shown in Fig. 5D, the principal component analysis of protein expression indicated that the samples from each group were not clearly separated, suggesting that si-eIF5A and spermidine had little differential impact on proteins. Fig. 5E and F displayed the differentially expressed proteins in the spermidine vs control group and the spermidine+si-eIF5A vs spermidine group, respectively. The volcano plot of spermidine vs control group reveals that spermidine significantly up-regulated 43 proteins and significantly down-regulated 37 proteins in goose granulosa cells compared to the control group. The results of the comparison between the spermidine+si-eIF5A group and the spermidine group indicated that a total of 165 proteins were significantly up-regulated and 168 proteins were significantly down-regulated. Fig. 5G shows the common differentially expressed proteins between the spermidine vs control group and the spermidine+si-eIF5A vs spermidine group. A total of 20 proteins were screened out. As shown in Fig. 5H, among these key proteins, 10 proteins (PRKCH, TCF25, HMOX1, RAMAC, ZHX2, NCAPD2, COQ9, CCDC43, ROGDL, EDA2R) were significantly up-regulated after spermidine treatment but significantly down-regulated after combined interference. In contrast, 8 proteins (PRKD3, SERPINE2, ADAMTSL3, AK6, CENPN, TNPO1, LGALS3) showed the opposite trend. Through literature review, HMOX1, COQ9 and SERPINE2 have been further identified as being significantly associated with the regulation of steroid hormones and animal ovarian function (Ortega et al., 2017; Yang et al., 2025; Zhang et al., 2024).

Fig. 5.

Effects of spermidine combined with si-eIF5A on protein expression in goose granulosa cells. (A-C) The impact of spermidine and si-eIF5A on the protein expression of eIF5A and eIF5A^hyp. (D) Principal component analysis of proteins in goose granulosa cells. (E) and (F) Volcano plots of differentially expressed proteins for the spermidine vs control and spermidine+si-eIF5A vs spermidine groups, with red indicating up-regulation and blue indicating down-regulation. (G) Venn diagrams of differentially expressed proteins for the spermidine vs control and spermidine+si-eIF5A vs spermidine groups. (H) Clustering of Key differential protein expression patterns in each sample.

Functional annotation of differentially expressed proteins

Based on the results of differential protein analysis, we performed GO functional annotations. The GO annotation results showed that the most significantly enriched biological process for spermidine-induced differentially expressed proteins was regulation of T cell proliferation. Notably, three pathways directly related to steroids were significantly enriched: regulation of steroid biosynthetic process, regulation of steroid metabolic process, and regulation of cholesterol biosynthetic process (Fig. 6A). As shown in Fig. 6B, in the spermidine+si-eIF5A vs spermidine group, the most significantly enriched pathways were regulation of autophagy, regulation of intracellular estrogen receptor signaling pathway, estrogen receptor signaling pathway, and negative regulation of intracellular estrogen receptor signaling pathway—all of which are steroid-related pathways. This indirectly validated the critical role of spermidine-mediated eIF5A^hyp in regulating goose granulosa cells steroid hormones. Additionally, KEGG functional annotation of differentially expressed proteins revealed that proteins differentially expressed in the spermidine vs control group were significantly enriched in pathways such as proteoglycans in cancer, RNA polymerase, and polycomb repression complex (Fig. 6C), whereas the spermidine+si-eIF5A vs spermidine group showed significant enrichment in pathways including amyotrophic lateral sclerosis, autophagy, and tight junction (Fig. 6D).

Fig. 6.

Functional annotation of differential expressed proteins in granulosa cells after treatment with spermidine and si-eIF5A. (A) and (B) Functional annotation of differentially expressed proteins based on GO database. (C) and (D) Functional annotation of differentially expressed proteins based on KEGG database.

Discussion

Steroid hormones, such as E2 and P4, are essential for the reproductive process in poultry. During follicular growth and development in poultry, the level of E2 secretion in the membranous layer is high to promote follicular growth, follicular dominance, and follicular maturation in the early stages of growth (Lin et al., 2011). As follicular development nears maturity, the secretion of E2 in the membranous layer decreases while the secretion of P4 in the granular layer increases, peaking approximately 4 hours before ovulation. This promotes follicular volume increase and ultimately leads to the onset of ovulation (Lin, Jia and Zhang, 2011). Currently, it has been discovered that the production and secretion of ovarian steroid hormones are not solely regulated by the hypothalamic-pituitary-ovarian axis. Instead, they are also influenced by key genes involved in steroidogenesis within granulosa cells and membrane cells, along with related hormone receptors, metabolic enzymes, and other genetic elements and substances. These factors can regulate the production and secretion of follicular steroid hormones via autocrine and paracrine effects, as well as through associated signaling pathways. This regulation jointly controls follicular growth, follicular atresia, cell proliferation, and apoptosis (Wadood et al., 2021).

Our previous study found a significant positive correlation between polyamine content and steroid hormones in goose follicles, suggesting polyamines may regulate steroidogenesis (Jiang et al., 2023a). Furthermore, previous research has revealed that following continuous administration of putrescine for 5 days, the levels of E2 and P4 in the serum of mice were significantly elevated (Fernandes et al., 2017). In contrast, after inhibiting polyamine synthesis, the levels of E2 and P4 in the blood of mice were significantly reduced, and follicular development was blocked (Fashe et al., 2010). At the cellular level, we discovered that 48 hours after interfering with ODC, a key gene in polyamine anabolism, there was a significant decrease in the level of E2 (Niu et al., 2021). Moreover, after 72 hours, the level of P4 decreased significantly, and after overexpressing ODC for 48 hours, the P4 level was significantly higher than that of the control (Niu et al., 2021). In line with these studies, the findings of the present study further validate the significant role of polyamines in steroidogenesis. Our results indicate that the levels of E2 and P4 secreted by goose follicular granulosa cells increase in a dose-dependent manner with spermidine, and reach the highest level when treated with 160 μmol/L. Since 160 μmol/L had already exerted a significant effect on steroidogenesis, we refrained from exploring the effects of higher concentrations of spermidine. Instead, we delved deeper into the mechanism of how spermidine functions.

Spermidine serves as the only substrate for eIF5A^hyp, which provides an aminobutyl group that attaches to a specific lysine residue to activate eIF5A (Park et al., 1981; Park and Wolff, 2018). Activated eIF5A can participate in translation initiation, elongation, and termination, and plays an important role in cell proliferation and differentiation as well as the treatment of disease (Schuller et al., 2017). Research findings indicate that eIF5A can be involved in the modulation of ovarian function by mediating the expression of the ovarian LHR gene. Moreover, GC7 has been demonstrated to impede the influence of FSH and HCG on the expression of both the LH receptor binding protein gene and protein within the rat ovary. Notably, eIF5A can mediate the expression of the LH receptor by means of the cAMP/PKA signaling pathway, thereby playing a significant role in the regulation of ovarian function (Gulappa et al., 2017). Inhibition of the hypusination modification of eIF5A will lead to a decrease in the expression of LRBP, which will lead to a down-regulation of LHR, thereby affecting steroidogenesis (Gulappa et al., 2015; Menon et al., 2014). The present study revealed that upon overexpression of eIF5A, the secretion level of E2 hormone by granulosa cells exhibited a remarkable elevation, whereas the level of P4 hormone showed a significant decline. This finding implies that eIF5A might play a role in the biosynthesis of steroid hormones, potentially influencing the hormonal balance and related physiological processes within the cells. Addition of spermidine to eIF5A overexpression further enhanced the ability of steroidogenesis in goose granulosa cells, and the level of E2 was significantly increased by 160 μmol/L spermidine. These results further suggest the importance of eIF5A and spermidine in steroidogenesis.

StAR, which is a key factor in steroid hormone production and significantly correlated with oocyte quality, is the key rate-limiting enzyme in the initial and rate-limiting steps of steroidogenesis, namely the transfer of cholesterol to the mitochondrial inner membrane (Miller, 2017). This study found that 100 μmol/L spermidine treatment had no effect on the protein expression of StAR, while 160 μmol/L spermidine treatment increased the protein expression of StAR. In addition, eIF5A overexpression was also able to significantly increase the level of StAR protein expression. Notably, StAR was further significantly increased in the eIF5A overexpression group in combination with spermidine compared to the eIF5A overexpression group, suggesting that the promotion of StAR protein expression by spermidine is dependent on eIF5A. By examining the expression of eIF5A^hyp, we found that both eIF5A overexpression and spermidine treatment alone were able to significantly increase the expression of eIF5A^hyp protein, which also indicates that eIF5A^hyp is a necessity for the function of eIF5A.

ODC, as the key rate-limiting enzyme in polyamine synthesis, has been found through research that inhibiting ODC activity with DFMO, the gene expression levels of StAR, 3β-HSD and 17β-HSD were significantly reduced (López-García et al., 2008). In addition, Studies have shown that spermidine can increase the gene expression of 3β-HSD and 17β-HSD inhibited by electromagnetic radiation, thereby increasing the number and activity of spermatozoa (Shahin et al., 2019). In this study, it was found that 100 μmol/L spermidine treatment could increase the gene expression of 3β-HSD and 17β-HSD, but had no effect on the gene expression of StAR. On the contrary, 160 μmol/L spermidine treatment could increase StAR and decrease the gene expression of 3β-HSD, but had no effect on the gene expression of 17β-HSD. These results suggested that 100 μmol/L spermidine may affect steroidogenesis by increasing the gene expression of 3β-HSD and 17β-HSD, and accelerate the conversion of steroids to P4 in mitochondria. 160 μmol/L spermidine may affect steroidogenesis by increasing StAR and decreasing 3β-HSD gene expression. Studies have shown that polyamines can inhibit the binding of cortical iron oxide reducing protein and redox chaperone protein, prevent the formation of cortical redox protein/CYP11A1 complex, and then inhibit the process of CYP11A1 catalyzing cholesterol to pregnenolone (Bernhardt and Neunzig, 2021). In granulosa cells, the expression of CYP19A1 is positively correlated with E2 level. In a certain range, the higher the expression of CYP19A1, the more E2 is secreted, indicating that CYP19A1 can indirectly affect the development of follicles by regulating hormone levels (Liu et al., 2015). This study found that 100 μmol/L and 160 μmol/L spermidine treatment could reduce the gene expression of CYP11A1 and CYP19A1. 100 μmol/L spermidine treatment had no effect on the gene expression of CYP17A1, while 160 μmol/L spermidine could reduce the gene expression of CYP17A1. It is suggested that spermidine may regulate the steroidogenesis of granulosa cells by regulating the gene expression of CYP11A1, CYP17A1 and CYP19A1. FSH reduces the expression of LRBP by inhibiting eIF5A (Gulappa et al., 2017), and HCG-induced down-regulation of LHR expression requires the inhibition of eIF5A through the cAMP-PKA-ERK1/2 pathway (Gulappa et al., 2015). This study found that eIF5A overexpression reduced the gene expression of StAR, CYP17A1 and CYP19A1, and increased the gene expression of CYP11A1, 3β-HSD and 17β-HSD. Compared with the eIF5A overexpression group, 100 μmol/L spermidine combined with eIF5A overexpression treatment reduced the gene expression levels of StAR, CYP11A1, CYP19A1 and 3β-HSD, and increased the gene expression levels of CYP17A1 and 17β-HSD. However, 160 μmol/L spermidine combined with eIF5A overexpression treatment reduced the gene expression levels of CYP11A1, CYP19A1, 3β-HSD and 17β-HSD, and increased the gene expression levels of StAR and CYP17A1. In summary, these results suggested that spermidine-eIF5A^hyp axis may affect the expression of steroid hormone-related genes, and then participate in the steroidogenesis of goose granulosa cells.

In order to further explore the key node protein regulated by eIF5A and spermidine, we performed proteomics sequencing of goose granulosa cells interfering with eIF5A combined with spermidine treatment. After analyzing the sequencing data, we found that the abundance of 80 proteins, including 43 up-regulated proteins and 37 down-regulated proteins, was significantly altered by spermidine treatment compared to the control group. In addition, compared with the spermidine group, spermidine combined with si-eIF5A treatment altered the abundance of a total of 333 proteins, including 165 up-regulated proteins as well as 168 down-regulated proteins. The intersection of these differentially expressed proteins yielded a total of 20 common differentially expressed proteins. Remarkably, ten of these common differential proteins were significantly increased by spermidine and then significantly decreased after spermidine combined with si-eIF5A treatment, including PRKCH, TCF25, HMOX1, RAMAC, ZHX2, NCAPD2, COQ9, CCDC43, ROGDI, EDA2R. In contrast, 8 proteins showed the opposite trend, including PRKD3, SERPINE2, ADAMTSL3, AK6, CENPN, TNPO1, LGALS3. Among these key differentially expressed proteins, we found through reviewing existing literature that three proteins, HMOX1, COQ9, and SERPINE2, may be involved in spermidine-eIF5A-mediated steroidogenesis in goose granulosa cells. The HMOX1 is an inportant protein involved in ferroptosis, oxidative stress and oocyte maturation (Abudureyimu et al., 2024; Barragán et al., 2023; Li et al., 2024; Zhang et al., 2024); COQ9 was involved in the synthesis of coenzyme COQ10, thus affecting mitochondrial function(Ben-Meir et al., 2015). Research had shown that the mutation in COQ9 modifies caused AMH levels was disorder, then affect the fertility of Holstein cows (Ortega et al., 2017); SERPINE2 can affects the production of steroid hormones. Studies have shown that after using siRNA to knock down SERPINE2 protein expression, apoptotic related proteins such as Bax and Caspase 9 was significant increase, and Bcl-2 was significant decrease. At the same time, si-SERINE2 will caused the E2 level and the mRNA expression of CYP19A1 was significant decrease, overexpression SERINE2 showed an opposite result. Those results indicate that the deficiency of SERINE2 will restrain sterodiogenesis and promote cell apoptotic (Yang et al., 2025). In addition, it has been found that SERINE2 can participate in the regulation of the ERK signaling pathway (Bergeron et al., 2010). Study had shown that ERK signaling pathway plays a crucial role in the regulation of granulosa cell sterodiogenesis (Baddela et al., 2023; Seger et al., 2023). Inhibition of the ERK signaling pathway can in turn affects steroid hormone (Kau et al., 2012). These results suggest that SERINE2 may be a core protein involved in spermidine-mediated steroidogenesis.

In addition, the annotation results of differentially expressed proteins based on the GO database showed that the differentially expressed proteins of spermidine vs control group were significantly enriched in important pathways related to steroid hormone biosynthesis and metabolic processes. The key enriched proteins included APOB, ERLIN2 and NR3C1, which play important roles in lipid transport, cholesterol homeostasis, and immune-metabolic regulation, and are also one of the important proteins involved in the steroid metabolic pathway. Furthermore, the differentially expressed proteins of spermidine+si-eIF5A vs spermidine group were also significantly enriched in pathways directly related to steroid metabolism. The key enriched proteins included SRC, TP63, and VPS18, which participate in steroid metabolism and signaling pathways through mechanisms such as signal transduction, transcriptional regulation, and vesicle transport: SRC directly regulates enzyme activity and nuclear receptor function as a kinase and co-activator (Chaturvedi et al., 2008); TP63 is responsible for encoding the transcription factor p63, a protein that exerts diverse functions—ranging from regulating ectodermal development and preserving genomic stability to sustaining female reproductive capacity (Tucker et al., 2022); In cells, the VPS18 protein is mainly localized in the membrane system, especially in membrane structures related to vesicle transport. It interacts with other proteins to form HOPS and CORVET complexes, which play key roles in intracellular material transport and membrane fusion processes (Dong et al., 2024).

In summary, our results demonstrate that spermidine promotes steroidogenesis in goose granulosa cells, and that this function is largely dependent on the activation of eIF5A. Spermidine-eIF5A^hyp axis also has the capacity to promote steroidogenesis in goose granulosa cells by affecting the expression of genes and proteins related to steroidogenesis (including StAR, CYP11A1, CYP19A1, 3β-HSD, and 17β-HSD). In addition, the results of protein functional annotation showed that the differentially expressed proteins induced by spermidine and si-eIF5A were significantly enriched in steroid biosynthesis and metabolism-related pathways. HMOX1, COQ9 and SERPINE2 were preliminarily identified as important participants in the regulation of steroidogenesis in goose granulosa cells by the spermidine-eIF5A^hyp axis.

Declaration of competing interest

The authors declare that they do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted.

Acknowledgements

The study was supported by National Natural Science Foundation of China (No. 32172727) and Sichuan Science and Technology Program (No. 2024NSFSC0300).