Integrated analysis of lncRNA-miRNA-mRNA networks reveals stage-specific molecular regulation of oogenesis in Chinese alligator (Alligator sinensis)

Abstract

Background

Reptiles, as the earliest vertebrate group to fully adapt to terrestrial environments, have reproductive strategies that hold critical significance for understanding biological evolution and ecological adaptation. However, the mechanisms of oogenesis in reptiles, particularly in the endangered Chinese alligator (Alligator sinensis), remain poorly understood.

Results

Here we integrated transcriptomic and non-coding RNA analyses to elucidate stage-specific lncRNA/miRNA-mRNA regulatory networks during oogenesis. Gonadal tissue samples were collected at three critical developmental stages: 1-day post-hatching (AH), 15-day post-hatching (BH), and 90-day post-hatching (CH). RNA sequencing revealed 568 upregulated and 222 downregulated genes during the early proliferation phase (from AH period to BH period), and 667 upregulated and 241 downregulated genes during the maturation phase (from BH period to CH period). Additionally, we identified 1,194 lncRNAs and 1,808 miRNAs dynamically expressed across these stages. During early proliferation, cell cycle genes (CCN3 - CCN Family Member 3, CDC45 - Cell Division Cycle 45) and signaling pathways (WNT - Wingless-Type MMTV Integration Site Family, FGF - Fibroblast Growth Factor) dominated, while metabolic genes (ALDOB-Aldolase B, Fructose-Bisphosphate, FABP4 - Fatty Acid Binding Protein 4) were suppressed. In the maturation phase, meiosis-related genes (STAG3 - Stromal Antigen 3, SPO11- Sporulation Protein 11 Homolog) and germ cell-specific factors (DAZL - Deleted in Azoospermia-Like, NOBOX - Newborn Ovary Homeobox) were upregulated, accompanied by miRNA-mediated suppression of non-essential functions. This study provides the first comprehensive roadmap of non-coding RNA-mediated oogenesis in reptiles, offering insights for conservation strategies in endangered species.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-026-12542-z.

Background

The Chinese alligator (Alligator sinensis), a member of Archosauria and one of the closest living relatives of birds, provides a unique evolutionary and physiological context for studying oogenesis [1]. Its ovarian morphology and vitellogenic features show strong similarity to birds, reflecting conserved reproductive traits inherited from a common ancestor. Unlike chickens, whose oocytes reach pachytene at hatching [2], pachytene oocytes and fully formed follicles appear only around three months post‑hatching in alligators [3, 4]. This delayed developmental pace offers an extended window for dissecting key processes such as oogonial proliferation, meiotic initiation, and yolk deposition. Constructing transcriptomic and lncRNA–miRNA–mRNA regulatory networks therefore enables the identification of temporally regulated genes and pathways—including Hippo and Notch signaling [5, 6]—that govern follicle formation and reproductive timing. These features make the Chinese alligator a valuable model for uncovering the mechanistic and evolutionary basis of vertebrate oogenesis [7].

The advent of RNA transcriptome sequencing technology has led to the recognition of the significant roles that non-coding RNAs play in development and metabolism. Non-coding RNAs include microRNAs (miRNAs), transfer RNAs (tRNAs), ribosomal RNAs (rRNAs), small interfering RNAs (siRNAs), and long non-coding RNAs (lncRNAs). miRNAs are a class of highly conserved, endogenous, single-stranded small non-coding RNA molecules, approximately 18–25 nucleotides in length, that regulate gene expression post-transcriptionally by inhibiting translation or causing the degradation of target mRNAs through binding to the 3’ untranslated region (3’ UTR). miRNAs are involved in various biological processes, including cell proliferation, differentiation, apoptosis, organogenesis, and disease pathogenesis. Studies have demonstrated that they play a crucial role in the maturation process of mammalian oocytes. For instance, miR-21, miR-29, and miR-181 significantly increase during oocyte maturation, and they promote normal oocyte development by targeting and regulating genes associated with oocyte development and maturation [8, 9]. Furthermore, miRNA also plays a role in regulating hormone levels within the ovary, affecting the development and maturation of follicles, and thus influencing the quality and quantity of oocytes [10, 11]. In fish, specific miRNAs, including miR-202 and the miR-200s, are expressed in the ovary and play regulatory roles in oocyte development and the ovulation process [12, 13]. Therefore, miRNAs play a significant role in the process of oogenesis.

Long non-coding RNAs (lncRNAs) are RNA molecules that exceed 200 nucleotides in length and do not code for proteins. In recent years, research has indicated that lncRNAs play significant roles in the regulation of gene expression, cell proliferation, differentiation, and various other biological processes. During the oogenesis of mammals, the role of lncRNAs has been confirmed by multiple studies, demonstrating their crucial involvement in ovarian development and oocyte maturation. For example, a substantial number of lncRNAs have been identified in pig ovaries, and their expression levels are significantly correlated with the maturity of oocytes. Certain lncRNAs regulate oocyte development and maturation by influencing gene expression within the ovary [14]. In the mouse model, lncRNA is associated with the quality and quantity of oocytes, affecting their development by regulating hormone levels and cell proliferation. For instance, a lncRNA named MSTRG. 3902. 1 participates in the development and maturation of oocytes by regulating the expression of the NR5A2 gene [15]. In the process of fish oogenesis, lncRNA also plays a key regulatory role, such as the lncRNA ORG-AS discovered in tongue sole (Cynoglossus semilaevis), which negatively regulates genes related to oogenesis [16]. In rainbow trout, lncRNA affects oocyte development by regulating signaling pathways [17]. Currently, there is no research on the involvement of lncRNA in the oogenesis of the Chinese alligator.

The transcriptome represents the complete set of transcripts present within a population of cells or a single cell, encompassing mRNA, miRNA, lncRNA, and various other types, it provides a snapshot of gene expression under specific physiological conditions or at particular developmental stages. Furthermore, RNA-seq is a cutting-edge sequencing technology and has been applied to study the mechanisms of sex determination and developmental regulation in Chinese alligators [18]. However, the stage-specific molecular regulatory analysis of oogenesis in Chinese alligators remains relatively limited.

Methods

Sample collection

Given the critically endangered status of the Chinese alligator (Alligator sinensis) in China, a limited number of individuals were used in this study. All experimental materials were provided by Wuhu Dajiang Farm, with procedures approved by animal ethics guidelines. On July 15, 2021, 20 eggs from the same clutch (laid on June 20, 2021; embryonic age ~ 25 days) were carefully transferred to a constant-temperature incubator (29 °C ± 0.1 °C; humidity 90% ± 0.5%) at the School of Life Sciences, Anhui Normal University. After a 70-day incubation period, 18 eggs successfully hatched. Hatchlings were morphologically identified via head scale patterns, and gonadal tissues were collected at three key developmental stages—1-day post-hatching (AH), 15-day post-hatching (BH), and 90-day post-hatching (CH)—representing oocyte, primordial follicle, and primary follicle phases, respectively (Table.S1). Six gonadal samples (two per stage) were flash-frozen in liquid nitrogen and stored at -80 °C for total RNA extraction and transcriptomic sequencing. Gonadal tissues collected at the AH (1-day post-hatching), BH (15-day post-hatching), and CH (90-day post-hatching) stages are complex and comprise a mixture of cell types. The predominant cellular constituents are expected to include: oocytes at various developmental stages (oogonia, primary oocytes), follicular (granulosa) cells surrounding the oocytes, theca cells, and stromal cells including germinal epithelial cells and connective tissue. While the specific proportions shift during development, our bulk transcriptome data represents an averaged signal from this composite cellular milieu.

Total RNA isolation and sequencing

Total RNA was extracted using TRIzol reagent (Invitrogen) following the manufacturer’s protocol. Genomic DNA was removed using RNase-Free DNase (Biomarker). RNA quality and purity were assessed using a Bioanalyzer 2100 with an RNA 6000 Nano LabChip Kit (Agilent), confirming RNA integrity numbers (RIN) > 8. 0. Ribosomal RNA was depleted using the Epicentre Ribo-Zero Gold Kit (Illumina). Polyadenylated (poly-A) mRNA was enriched using oligo(dT) magnetic beads. Fragmented RNA (~ 200 nt) was reverse-transcribed into cDNA using the mRNA-Seq Sample Preparation Kit (RS-121-2001; Illumina). Paired-end libraries (average insert size: 300 bp ± 50 bp) were sequenced on an Illumina NovaSeq platform.

RNA-Seq read alignment and transcriptome assembly

Adapter-contaminated, low-quality, and ambiguous reads were trimmed using Cutadapt. Sequence quality was verified with FastQC. Clean reads were aligned to the Alligator sinensis genome(whole-genome resequencing: CRA006388, CRA006341, CRA006322; Nanopore long reads: CRA006327; Hi-C: CRA006324; transcriptome: CRA006337) using Bowtie2 and TopHat2. Transcripts were assembled per sample using StringTie [19], followed by merging all samples into a unified transcriptome. Expression abundance (FPKM) was quantified with StringTie, and differential expression analysis (|log2(fold change)| ≥2, p < 0.05) was performed using edgeR and Ballgown [20].

Functional annotation of Differentially Expressed Genes (DEGs)

Gene annotations were derived from Swiss-Prot, KEGG, Nr, Nt, KOG, GO, and InterPro databases. GO enrichment analysis of DEGs was conducted using the GOseqR package. Differential expression analysis was performed with DESeq2 [21](|log2(fold change)| ≥2, p < 0.05).

LncRNA identification and differential expression analysis

Transcripts overlapping known mRNAs or shorter than 200 bp were excluded. Coding potential was assessed using CPC [22], CNCI [23] and FEELnc. Transcripts with CPC scores < -1 and CNCI scores < 0 were discarded. Intersection results from all three tools defined the final lncRNA set. Expression levels (FPKM) of mRNAs and lncRNAs were calculated via StringTie. Differentially expressed mRNAs and lncRNAs (|log2(fold change)| ≥2, p < 0.05) were identified using edgeR and Ballgown.

LncRNA target gene prediction

Cis-acting target genes of lncRNAs were predicted by selecting coding genes within 100 kb upstream or downstream of lncRNA loci. Functional analysis of target genes followed the same pipeline as for mRNAs.

MiRNA sequencing and analysis

Total RNA was size-selected (18–30 nt) via agarose gel electrophoresis. After adapter ligation, cDNA libraries were constructed via reverse transcription and PCR amplification. Libraries (140 bp) were quality-checked using an Agilent 2100 Bioanalyzer and qPCR, then sequenced on an Illumina NovaSeq X Plus. Raw reads were filtered to remove low-quality sequences, adapter-free reads, polyA-rich reads, and tags with low abundance (< 2 counts). rRNA, scRNA, snoRNA, snRNA, and tRNA contaminants were identified and removed using GenBank and Rfam 11. 0 (blastn, identity ≥ 97%).

MiRNA target prediction and functional interaction analysis

Target genes of miRNAs were predicted using TargetScan 7.0 and miRanda 3.3a, with intersections retained. Functional miRNA-mRNA interactions were analyzed via MMpred. Putative miRNA targets were defined as mRNAs showing negative expression correlation (p < 0.05) with differentially expressed miRNAs (DEMs). Differential expression analysis was performed using DESeq2 (|log2 fold change| ≥ 2, p < 0.05) to obtain the list of differentially expressed miRNAs.

Results

Correlation analysis of RNA abundance across samples

We analyzed RNA abundance correlations among six samples and observed strong consistency between biological replicates (Fig. 1). All correlation coefficients greater than 0.9 indicate that intra-group variation is minimal, and PCA analysis intuitively demonstrates that the three groups of samples are significantly separated based on their gene expression profiles (Fig.S1), making them suitable for subsequent differential analysis.

Fig. 1.

Correlation analysis of gene expression levels

Transcriptome sequencing

Gonadal tissue samples were used for transcriptome assembly and annotation. RNA extracted from these samples was used to construct six cDNA libraries (Table S1), which were sequenced. A total of 70,992,744,000 raw reads were generated from the Chinese alligator (Alligator sinensis) cDNA libraries. After removing low-quality sequences, 70,006,064,679 clean reads were retained for mRNA and lncRNA identification (Table S2).

Differential transcript abundance analysis

The number of differentially expressed genes at different stages of oogenesis is shown in Table S3, the AH period compared with the BH period has 568 genes were upregulated, and 222 genes were downregulated (Table S4A). Compared to the CH period, the BH period has 667 genes were upregulated, and 241 genes were downregulated (Table S4C). Compared to the CH period, the AH period had 609 upregulated genes and 565 downregulated genes (Table S4B).

Gene Ontology (GO) (Table S5)enrichment and KEGG pathway analyses (Table S6) were performed to elucidate the functions of differentially abundant genes (Figs. 2, 3). GO analysis revealed that the most enriched biological processes included developmental process, reproductive process, regulation of biological process, and metabolic process. Key molecular functions encompassed transcription regulator activity, binding, and molecular function regulator (Due to the inherent imperfection of the Chinese alligator genome annotation, 5,727 of its genes cannot be assigned to specific GO categories, accounting for 26.4% of the total number of genes.). KEGG pathway analysis highlighted oogenesis-related pathways such as steroid hormone biosynthesis, ovarian steroidogenesis, PPAR signaling pathway, MAPK/MPF signaling pathway, FoxO signaling pathway, and PI3K-Akt signaling pathway.

Fig. 2.

GO enrichment analysis of stage-differential genes

Fig. 3.

KEGG pathway enrichment analysis of stage-specific differentially expressed genes

Additionally, several differentially expressed genes identified in our dataset mapped to key regulatory pathways essential for follicle formation, including the Hippo and Notch signaling pathways. Specifically, WNT9B, WNT4, FGF family members, PIK3CA, FOXO1, and MAPK-related genes correspond to canonical upstream regulators of the Hippo pathway, while cell-cycle–associated genes such as CDK2, CCNA2/CCNB1, CDC45, and MCM4/5/6 represent downstream effectors. For the Notch-related network, genes involved in cell–cell communication and follicular coordination, including GJA3, E2F1, BCL2, BCL10, IL2RB, and IL7R, were also differentially expressed. These findings suggest that Hippo- and Notch-associated regulatory modules are active during the developmental transitions of oogenesis in the Chinese alligator.

Analysis of oogenesis-related differentially expressed genes

During the early proliferative stage of oogenesis (spanning from the AH period to the BH period), these genes experienced significant upregulation in various biological processes. Specifically, cell cycle genes such as CCN3, CDC45, and MCM6 - Minichromosome Maintenance Complex Component 6, as well as those involved in cytoskeletal dynamics and signal transduction, including FGF, WNT9B - Wnt Family Member 9B, and RGS4 - Regulator of G Protein Signaling 4, were markedly upregulated (Table S4D). Notably, HSPH1 - Heat Shock Protein Family H Member 1, STARD4 - StAR-Related Lipid Transfer Domain 4, and CYP2J2 - Cytochrome P450 Family 2 Subfamily J Member 2 exhibited significant expression changes. Conversely, downregulated genes, such as ALDOB and PYGM - Phosphorylase, Glycogen, Muscle, were enriched in metabolic pathways. Genes related to vitamins/cofactors, like HGD - Homogentisate 1,2-Dioxygenase and PCBD1 - Pterin-4a-Carbinolamine Dehydratase 1, and those involved in signaling/regulation, including IL2RB - Interleukin 2 Receptor Subunit Beta and IL7R - Interleukin 7 Receptor, also displayed altered expression patterns, influencing immune response, cell growth, proliferation, and differentiation (Table S4E).

During the growth phase of oogenesis (oogonia differentiate into primary oocytes, initiate meiosis but arrest at prophase I; the cell volume increases significantly and accumulates nutrients), which spans from the BH period to the CH period, there was a significant upregulation of certain genes, particularly those related to the cell cycle, meiosis, and germ cell development. These include STAG3, SPO11, ACVR1C - Activin A receptor type 1 C, PIK3CA - Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha, AK3 - Adenylate kinase 3, and FOXO1 - Forkhead Box O1(Table S4F). Conversely, genes that experienced downregulation were those involved in signal transduction, metabolism, proliferation, and differentiation(Table S4G), such as BCL10 ( B Cell CLL/Lymphoma 10) and CD200R1A (Cell surface glycoprotein CD200 receptor 1).

LncRNA identification and target gene prediction

Through rigorous bioinformatics analysis, we identified 1,194 stable long non-coding RNAs (lncRNAs) (Tables S7A, B), with 848 dynamically expressed across all three stages. Venn diagrams and heatmaps were constructed to visualize differentially expressed lncRNAs between stages (Fig. 4). The AH period compared with the BH period showed 64 upregulated and 90 downregulated lncRNAs. The BH period compared with the CH period revealed 94 upregulated and 84 downregulated lncRNAs. The AH period compared with the CH period exhibited 94 upregulated and 87 downregulated lncRNAs.

Fig. 4.

Comparative analysis of stage-specific lncRNAs. a Venn diagram of stage-specific lncRNAs. b Bar plot of stage-specific lncRNA counts. c Heatmap of differentially expressed lncRNAs between AH and CH. d Heatmap of differentially expressed lncRNAs between AH and BH. e Heatmap of differentially expressed lncRNAs between BH and CH

Statistical analysis of lncRNA and mRNA expression across six samples revealed consistent expression trends (see Supplementary Fig.S2). We predicted 772 lncRNA-mRNA target pairs (646 mRNAs, 536 lncRNAs; see Table S7C) and conducted GO and KEGG enrichment analyses on the target mRNAs (Fig. 5). The GO analysis indicated enrichment in intracellular signaling, cytoskeletal organization, protein synthesis/modification, and cell cycle regulation. The KEGG pathways included alanine, aspartate, and glutamate metabolism, ribosome biogenesis, phenylalanine, tyrosine, and tryptophan biosynthesis, nicotinate and nicotinamide metabolism, and oocyte meiosis.

Fig. 5.

GO enrichment analysis and KEGG pathway analysis of lncrna target genes. a: Bubble plot of GO enrichment analysis for differentially expressed lncRNA target genes between AH and BH. b: Bubble plot of GO enrichment analysis for differentially expressed lncRNA target genes between BH and CH. c: Bubble plot of GO enrichment analysis for differentially expressed lncRNA target genes between AH and CH. d: Bar plot of GO enrichment analysis for differentially expressed lncRNA target genes between AH and BH. e: Bar plot of GO enrichment analysis for differentially expressed lncRNA target genes between BH and CH. f: Bar plot of GO enrichment analysis for differentially expressed lncRNA target genes between AH and CH. g: Circos plot of KEGG pathway analysis for differentially expressed lncRNA target genes. h: Bar plot of KEGG pathway enrichment analysis for lncRNA target genes. i: Bubble plot of KEGG pathway enrichment analysis for lncRNA target genes

MiRNA Identification, functional interaction analysis, and construction of the lncRNA-miRNA-mRNA regulatory network

To analyze lncRNA-miRNA-mRNA targeting relationships, we constructed a negatively correlated miRNA-target functional network (Fig. 6). To underpin this network, six small RNA libraries were generated to profile miRNA expression during oogenesis. After filtering adapter dimers, low-complexity sequences, common RNAs (rRNA, tRNA, snRNA, snoRNA), and redundancies, ~ 63 million clean reads (18–35 nt) were retained (Fig. 7). Tag annotation (Supplementary Figures S3–S6) identified 539 known and 1,269 novel miRNAs (Fig. 8). Differential miRNA expression was observed: compared to the AH period, the BH period exhibited 82 miRNAs upregulated and 33 downregulated. The BH period, in comparison to the CH period, had 124 miRNAs upregulated and 193 downregulated. The AH period, when contrasted with the CH period, showed 151 miRNAs upregulated and 146 downregulated (Fig. 9). miRNA-mRNA interactions were predicted using TargetScan v7. 0 and miRanda v3. 3a, with the overlapping predictions retained. A total of 1,173,671 miRNA-mRNA pairs (1,808 miRNAs, 8,468 mRNAs) were identified (Supplementary Figure.7). GO and KEGG analyses of differentially expressed miRNA (DEM) targets were conducted (Fig. 10). Since miRNAs regulate their targets through mRNA cleavage, the expression of target genes generally exhibited negative correlations with their corresponding miRNAs. Key findings were as follows: From the AH period to the BH period, 67,797 miRNA-mRNA pairs were identified (115 DEMs, 7,503 genes). From the BH period to the CH period, 203,799 miRNA-mRNA pairs were discovered (317 DEMs, 8,180 genes).

Fig. 6.

Functional network diagram of negatively correlated miRNA and target gene pairs. Red: up regulated; light blue: down regulated; green: lncRNA

Fig. 7.

Overview of small RNA sequencing in the Chinese alligator gonadals. a Length distribution map of sRNA. b Size distribution of all miRNAs. c Length comparison of lncRNAs and miRNAs

Fig. 8.

Expression levels of miRNAs in different samples

Fig. 9.

Statistical chart of differential miRNAs between groups. a Scatter plot of differentially expressed miRNAs across stages. b Venn diagram of stage-specific miRNAs

Fig. 10.

GO enrichment analysis and KEGG pathway analysis of differentially expressed miRNA target genes across stages. a GO enrichment analysis. b KEGG pathway analysis

In the functional network diagram of miRNA-target pairs constructed by us (Fig. 6), highlighting strong interactions between meiosis/cell cycle-related genes (WEE2 -WEE1-Like Protein Kinase 2, MCM4 - Minichromosome Maintenance Complex Component 4, ACVR1C, TOP2A - Topoisomerase II Alpha, etc.) and miRNAs (Table 1). Finally, by integrating predicted lncRNA-mRNA and miRNA-mRNA interactions, we established a lncRNA-miRNA-mRNA regulatory network (Fig. 6). The miRNA-Gene pathway networks at different stages of oogenesis are depicted in Figs. 11 and 12.

Table 1.

Partial differentially Expressed Genes (DEGs) and their corresponding pathways, as well as differentially expressed miRNAs

Gene	Pathway	miRNA related differential expression
WNT4	Wnt signaling pathway	miR-424-x,mR-450-x,miR-503-x,mR-654-y,novel-m0085-5p,novel-m0376-3p,nove-m0456-5p,novel-m0457-5p,novel-m0458-5p,nove-m0663-5p,novel-m1074-5p
RAD51	DNA damage repair (homologous recombination)	miR-11987-z
CDC45	DNA replication	miR-200-z,miR-654-y,novel-m0541-5p
MCM5/MCM6	DNA replication	miR-1-z,miR-206-y,miR-206-z,miR-29-y,miR-424-x,novel-m0376-3p,novel-m1058-3p
CDK2	Cell cycle (G1/S phase),DNA repair	novel-m0011-5p,novel-m0012-3p,novel-m0376-3p,novel-m0663-5p,novel-m1058-3p
TOP2A	DNA replication/repair	novel-m0541-5p,novel-m0433-5p,novel-m0657-5p
NEK2	DNA replication/repair	mR-2184-y,mR-313-y,mR-654-y,novel-m0019-3p,novel-m0065-3p,novel-m0433-5p,nove-m0857-3p,novelm0858-3p,novelml058-3p
GREB1	Estrogen receptor signaling	miR-2184-y,miR-101-z,novel-m0433-5p
E2F1	Cell cycle(G1/S phase)	miR-29-y,miR-133-zmiR-151-x,miR-11987-zmiR-424-x,miR-450-x,miR-133-x,novel-m0019-3p,novelm0301-5p,novel-m0338-5p,novel-m0456-5p,novel-m0457-5p,novel-m0458-5p,,novel-m0574-3pnovel-m1074-5p,novel-m1058-3p
FADS2	Fatty Acid Metabolism	miR-133-zmiR-151-x,miR-11987-z,miR-509-y,miR-195-z,miR-313-y,novel-m0065-3p,novel-m0085-5p,novel-m0456-5p,novelm0457-5p,novel-m0458-5p,novel-m0574-3p,novel-m0708-5p
CENPU	Mitosis/Chromosome separation	miR-1-zmiR-29-y,miR-200-zmR-206-y,miR-424-x,miR-101-zmiR-206-Zznovelm0085-5p,novel-m0338-5p,novel-m0541-5p,novel-m0857-3p,novel-m0858-3p,novel-m1074-5p,novel-m1058-3p,novel-m0708-5p
WEE2	Progesterone-mediated oocyte maturation	miR-1397-x,miR-101-z,novel-m0663-5p
Lpinl	PPAR signaling pathway	miR-424-x,miR-195-z,miR-313-y,novel-m0065-3p,novel-m0085-5p
SLC2A2	Insulin signaling pathway	miR-133-z,miR-1397-x,miR-2184-y,miR-11987-z,miR-424-x,miR-101-z,miR-313-y,novel-m0019-3p,novel-m0065-3p,novel-m0085-5p,novel-m0437-sp,nove-m0438-5p,novel-m0301-5p,novel-m0376-3p,novel-m0338-5p,novelm0541-5p,novelm0433-5p,novel-m0657-5p,novel-m1074-5p,novelm1058-3p,novel-m0708-5p
PTGDS	Arachidonic acid metabolism	miR-424-x,miR-503-x,novel-m0019-3p
HSD11B1	Steroid hormone biosynthesis	miR-29-y,miR-1397-x,miR-151-x,miR-101-z,novel-m0065-3p,novel-m0857-3p,novel-m0858-3p
COL5A1/COL24A1	ECM-receptor interaction	miR-654-y,miR-2184-y,miR-509-y,novel-m0708-5p
GADD45B	p53 signaling pathway	miR-424-x,miR-450-x,miR-654-y,novel-m0356-3p,novel-m0857-3p,novel-m0858-3p,novel-m1074-5p,novel-m0663-5p,novel-m1058-3p
SLC2A3	PI3K-Akt signaling pathway	miR-133-z,miR-424-x,miR-450-x,miR-195-z,miR-313-y,novel-m0437-5p,novel-m0438-5p,novel-m0433-5p,novel-m0456-5p,novel-m0457-5p,novelm0458-5p,novel-m0574-3p,novel-m1074-5p,novel-m0708-5p
Has2	Hyaluronan metabolism	miR-133-z,miR-1388-x,miR-352-x,miR-200-z,miR-2184-y,miR-424-x,miR-313-y,miR-654-y,novel-m0085-5p,novel-m0376-3p,novel-m0708-5p
MSH4	Meiosis	miR-195-z
BOLL	Oocyte meiosis	miR-200-z,miR-133-x,novel-m0019-3p,novel-m0437-5p,novel-m0438-5p,novel-m0663-5p
SYCE2	Meiosis	miR-424-x,miR-184-x
STRA8	Retinol metabolism	miR-2995-z,novel-m0024-3p,novel-m0047-3p,novel-m0519-3p,novel-m0520-3p,novel-m0521-3p,novel-m0435-5p

Fig. 11.

Molecular regulatory model from AH to BH Stage. Red: up regulated; blue: down regulated; square: signaling pathway

Fig. 12.

Molecular regulatory model from BH to CH Stage. Red: up regulated; blue: down regulated; square: signaling pathway

Discussion

This study used the Chinese alligator (Alligator sinensis) as a model to achieve, for the first time in reptiles, an integrated transcriptomic analysis of the ovulation process. It systematically revealed the stage-specific gene expression dynamics during oogenesis and their molecular regulatory mechanisms mediated by non-coding RNAs, including long non-coding RNAs (lncRNAs) and microRNAs (miRNAs).

During the early proliferative phase of oogenesis (from stage AH to stage BH), CCN3 interacts with cyclin-dependent kinases (CDKs) to drive germ cells out of the quiescent phase [24]. Meanwhile, CDC45 and the MCM complex (MCM5, MCM6) collectively ensure the faithful initiation of DNA replication, laying the foundation for cell expansion. This proliferative process is accompanied by profound metabolic reprogramming, characterized by the downregulation of genes involved in carbohydrate, lipid, and amino acid metabolism, which redirects energy flow from basal maintenance toward supporting meiotic preparation and cytoplasmic maturation [25–27]. Concomitantly, this process is accompanied by the activation of signaling pathway genes (such as FGF and WNT9B), which constructs an intercellular communication network and guides the coordinated development of oocytes and surrounding somatic cells [28, 29]. During this period, metabolic-related genes are generally downregulated, suggesting that the energy utilization strategy shifts from basal maintenance to supporting specific anabolic processes such as meiotic preparation and cytoplasmic maturation.

During the maturation phase of oogenesis (from stage BH to stage CH), the regulatory core shifts to the reinitiation and precise execution of meiosis. Genes related to the cell cycle and meiosis (such as STAG3, SPO11, MSH4, MSH5) are significantly upregulated, marking the official initiation of meiosis. SPO11 induces DNA double-strand breaks to initiate homologous recombination; STAG3, as a core component of the synaptonemal complex, ensures the correct pairing and synapsis of homologous chromosomes; while the MSH4-MSH5 heterodimer stabilizes recombination intermediates and precisely regulates crossover events. This mechanism collectively ensures the high-fidelity segregation and diversity of genetic material [30, 31]. Increased expression of key factors for germ cell development (such as DAZL, NOBOX, FIGLA - folliculogenesis specific bHLH transcription factor) further promotes the oocytes’ acquisition of maturation and fertilization potential [32, 33]. During this stage, the upregulation of genes related to cell junction and communication (such as GJA3 - Gap Junction Protein Alpha 3, ZP1-3 - Zona Pellucida Glycoprotein 1、2、3) is also observed, which enhances the functional coupling between oocytes and follicular cells, while the downregulation of some signal transduction and metabolic genes reflects the preferential allocation of resources to core processes such as meiosis [34, 35].

Non-coding RNAs exhibit multilevel regulatory functions in this process. Identified lncRNAs, through co-expression with mRNA, may participate in the regulation of key biological processes such as cell cycle, apoptosis (such as BCL10, BCL2 ), and cytoskeletal dynamics (such as ENTHD1 - Ectonucleotide Pyrophosphatase/Phosphodiesterase 1) under a cis-acting mechanism [36]. Meanwhile, this study identifies a large number of miRNA-mRNA interaction relationships. Key miRNAs (such as miR-424-x, novel-mi1058-3p, miR-313-y, miR-101-z) extensively participate in signaling pathways including Wnt, p53, PI3K-Akt, PPAR, and insulin signaling pathways by targeting genes such as WNT4, RAD51 - RAD51 Recombinase, CDC45, and E2F1 - E2F Transcription Factor 1 [37–39], playing core coordinating roles in processes such as cell cycle, apoptosis, metabolism, and DNA repair. The further constructed lncRNA–miRNA–mRNA regulatory network highlights the hub role of key miRNAs in the cross-regulation of multiple pathways.

Regulation of the cell cycle and meiosis constitutes the core of oogenesis. In the early stage, genes such as CCN3, CDC45, and MCM6 drive DNA replication and cell cycle progression; during the meiotic initiation stage, CDK2 - Cyclin-Dependent Kinase 2 collaborates with cyclins CCNA2/CCNB1 to coordinately regulate the G2/M transition [40–42], while STAG3, SPO11, and other factors ensure proper chromosome synapsis and recombination [43, 44]. This process is simultaneously subjected to precise regulation by LH-triggered ERK/MAPK cascade, PPAR pathway-mediated lipid metabolic reprogramming, and miRNAs (e.g., miR-424-x maintains genomic stability through the E2F1-p53 axis) [45–47]. The upregulation of ZP family genes (ZP1-3) ensures the structural integrity of post-meiotic oocytes through zona pellucida formation [48]. Furthermore, the balance of reactive oxygen species (NOX4/GPX3 - Glutathione Peroxidase 3) [49–52] and gap junctional communication (GJA3) collectively maintain the microenvironmental homeostasis required for meiosis [34, 35].

In addition to the major pathways discussed above, our analysis also revealed the involvement of two evolutionarily conserved regulatory modules—Hippo and Notch signaling—both of which are known to play indispensable roles in follicle activation and oocyte–somatic cell communication [53–55]. Upstream regulators of the Hippo pathway, including WNT9B, WNT4, FGF, PIK3CA, and FOXO1, were differentially expressed between developmental stages, together with downstream cell-cycle effectors such as CDK2, CCNA2, CDC45, and MCM family members. These components collectively suggest that Hippo signaling may contribute to coordinating proliferation, oocyte growth, and meiotic competence in the Chinese alligator [56]. Similarly, the detection of Notch-associated genes such as GJA3, E2F1, BCL2, BCL10, IL2RB, and IL7R indicates that Notch-mediated communication between oocytes and follicular cells may also participate in stage-specific follicular differentiation. Incorporating these pathways provides additional mechanistic support for the regulatory landscape identified in our lncRNA–miRNA–mRNA network.

GO and KEGG enrichment analyses further confirmed the stage-specificity of the aforementioned processes: the early stage was enriched in pathways related to germ cell growth, metabolism, and cell cycle; the maturation stage shifted to cell junctions, phagosomes, PPAR signaling pathway, etc., reflecting the core biological requirements of different developmental stages.

It is important to note that our study utilized bulk RNA-seq from whole gonadal tissues, which contain a mixture of cell types including oocytes, granulosa cells, theca cells, and stromal cells. Consequently, the identified lncRNA-miRNA-mRNA networks represent averaged expression profiles across these constituent cells. While this approach successfully reveals global, stage-specific molecular changes during oogenesis, it cannot definitively assign the identified interactions to a single specific cell type. For instance, an observed negative correlation between a miRNA and an mRNA could occur within the same cell, or they could be expressed in different, neighboring cells (e.g., a miRNA in granulosa cells regulating an mRNA in oocytes via paracrine signaling). To strengthen our findings, we prioritized interactions where the predicted miRNA target sites were conserved and focused our biological interpretation on well-established, oogenesis-related pathways and genes with known roles in germ cells or somatic follicular cells.

In summary, oogenesis in the Chinese alligator is a complex biological process finely coordinated by dynamic changes in gene expression, regulation by non-coding RNA networks, hormonal signal guidance, and metabolic reprogramming. The precise spatiotemporal coordination of these mechanisms collectively ensures that germ cells efficiently and accurately develop into functional oocytes with fertilization potential. Through transcriptomic analysis, this study systematically reveals, for the first time, the key molecular events and regulatory networks in reptiles during this process, providing important insights into understanding the conserved and specific mechanisms of reproductive evolution and oogenesis in amniotes.

Conclusions

This study focuses on the molecular regulatory mechanisms during the oogenesis of the Chinese alligator (Alligator sinensis). The research emphasis is on the interactions between long-chain non-coding RNAs (lncRNAs), microRNAs (miRNAs), and messenger RNAs (mRNAs), which are considered to be key factors. Although extensive research has been conducted on the molecular mechanisms of oogenesis in mammals and various fish species, systematic analyses of oogenesis in Chinese alligator remain relatively limited [57, 58]. This study employed high-throughput sequencing technology and bioinformatics analysis methods to reveal the expression characteristics and interaction networks of long non-coding RNAs (lncRNAs), microRNAs (miRNAs), and messenger RNAs (mRNAs) during the oogenesis of different developmental stages in the Chinese alligator. The findings indicate that there are significant molecular characteristic differences at each stage of oogenesis, with specific lncRNAs, miRNAs, and mRNAs showing significant upregulation or downregulation trends at different stages. The study provides a new perspective for a deeper understanding of the reproductive biology of the Chinese alligator and, by constructing a lncRNA-miRNA-mRNA regulatory network, not only elucidates the complex molecular mechanisms involved in the process of oogenesis but also lays a theoretical foundation for subsequent conservation and breeding research [59–61].

This study revealed key molecules in the process of oogenesis in Chinese alligators through gene expression analysis. A total of 1,194 stable lncRNA expressions were identified, with 848 showing dynamic expression. During the early proliferation phase, lncRNA participates in the initiation of the cell cycle and inhibition of apoptosis through cis-regulation of adjacent genes (such as CDK10- Cyclin-Dependent Kinase 1, BCL10); in the mature phase, lncRNA (such as MSTRG. 3902. 1 targeting NR5A2) coordinates chromosome stability and microenvironment communication by regulating meiosis pathways (KEGG enriched in oocyte meiosis) and genes related to cell junctions (such as GJA3). The co-expression network of differentially expressed lncRNA and target genes further supports their temporal regulatory function during stage transitions. Additionally, we identified a total of 539 known miRNAs and 1,269 novel miRNAs, which exhibit significant stage-specific expression. We constructed a miRNA-mRNA negative regulatory network to predict important regulatory relationships between miRNA and mRNA. In the early proliferation phase, miR-424-x balances proliferation and apoptosis by inhibiting genes such as WNT4; in the mature phase, novel-m1058-3p coordinates DNA repair and energy metabolism by regulating genes such as RAD51. The negative correlation between miRNA and target genes (such as CDK2, TOP2A) highlights their fine regulation of cell cycle, meiosis, and metabolic pathways. Further research can explore the specific functions and evolutionary significance of these differentially expressed lncRNAs and mRNAs during oogenesis. In the interaction network analysis, the constructed lncRNA-miRNA-mRNA regulatory network revealed significant interaction relationships among these molecules. By constructing the lncRNA-miRNA-mRNA interaction network, key hub molecules (such as miR-424-x, lncRNA MSTRG.3902.1) were found to integrate cell cycle progression, metabolic adaptation, and genetic stability through cross-regulation of pathways such as Wnt, PI3K-Akt, and PPAR. For example, lncRNA affects DAZL expression through chromatin modification, while miRNA coordinates the restart of meiosis by inhibiting genes like E2F1. The expression of lncRNA is closely related to specific miRNAs and mRNAs, providing a new perspective for understanding the molecular regulation during oogenesis. This interaction network can not only help identify new biomarkers but may also provide a reference for the study of oogenesis in other species [62]. By delving into the specific mechanisms of action of various molecules within the network, the potential functions and biological significance of these molecules in oogenesis can be revealed. The results of pathway analysis indicate that multiple reproduction-related signaling pathways, such as the PI3K-Akt signaling pathway and the MAPK signaling pathway, are significantly activated during the process of oogenesis. The activation of these pathways provides the necessary molecular signals for oocyte development and may drive the developmental process of oocytes. Whether there are differences in the pathways activated at different stages and the interaction of these pathways with other biological processes are important directions for future research. Further exploration of the specific roles of these pathways in oogenesis will contribute to a comprehensive understanding of the molecular mechanisms of oocyte development. Finally, this study explores potential causal relationships and finds that certain lncRNAs have potential causal relationships with miRNAs and mRNAs, and that the strength of these relationships varies at different stages of oogenesis. This provides a theoretical basis for subsequent functional experiments. Future research can validate these potential causal relationships through experiments and explore their application value in other biological processes [63]. Further research will help to uncover the complex network regulating oogenesis, providing a new theoretical basis for the conservation and breeding of Chinese alligators.

Abbreviations

CCN3

CCN Family Member 3

CDC45

Cell Division Cycle 45

WNT

Wingless-Type MMTV Integration Site Family

FGF

Fibroblast Growth Factor

ALDOB

Aldolase B, Fructose-Bisphosphate

FABP4

Fatty Acid Binding Protein 4

STAG3

Stromal Antigen 3

SPO11

Sporulation Protein 11 Homolog

DAZL

Deleted in Azoospermia-Like

NOBOX

Newborn Ovary Homeobox

CDK10

Cyclin-Dependent Kinase 10

CD200R1A

Cell surface glycoprotein CD200 receptor 1

ACVR1C

Activin A receptor type 1 C

PIK3CA

Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha

AK3

Adenylate kinase 3

Cyp19A1

Cytochrome P450 Family 19 Subfamily A Member 1

BLIMP1

B Lymphocyte Induced Maturation Protein 1

PRDM14

PR Domain Zinc Finger Protein 14

OCT4

Octamer-Binding Transcription Factor 4

NANOG

Nanog Homeobox

BMP

Bone Morphogenetic Protein

FoxO1

Forkhead Box O1

FSHR

Follicle-Stimulating Hormone Receptor

MCM6

Minichromosome Maintenance Complex Component 6

RGS4

Regulator of G Protein Signaling 4

HSPH1

Heat Shock Protein Family H Member 1

STARD4

StAR-Related Lipid Transfer Domain 4

CYP2J2

Cytochrome P450 Family 2 Subfamily J Member 2

PYGM

Phosphorylase, Glycogen, Muscle

HGD

Homogentisate 1,2-Dioxygenase

PCBD1

Pterin-4a-Carbinolamine Dehydratase 1

IL2RB

Interleukin 2 Receptor Subunit Beta

IL7R

Interleukin 7 Receptor

BCL10

B Cell CLL/Lymphoma 10

WEE2

WEE1-Like Protein Kinase 2

MCM4

Minichromosome Maintenance Complex Component 4

MCM5

Minichromosome Maintenance Complex Component 5

MCM6

Minichromosome Maintenance Complex Component 6

BCL2

B Cell CLL/Lymphoma 2

CDK10

Cyclin-Dependent Kinase 10

CDK2

Cyclin-Dependent Kinase 2

ENTHD1

Ectonucleotide Pyrophosphatase/Phosphodiesterase 1

WNT4

Wingless-Type MMTV Integration Site Family Member 4

RAD51

RAD51 Recombinase

TOP2A

Topoisomerase II Alpha

NEK2

NIMA-Related Kinase 2

HAS2

Hyaluronan Synthase 2

GREB1

Growth Regulation By Estrogen In Breast Cancer 1

E2F1

E2F Transcription Factor 1

FADS2

Fatty Acid Desaturase 2

CENPU

Centromere Protein U

KIF18B

Kinesin Family Member 18B

KIF18A

Kinesin Family Member 18 A

KIF20B

Kinesin Family Member 20B

KIF23

Kinesin Family Member 23

KIF4

Kinesin Family Member 4

FIGLA

folliculogenesis specific bHLH transcription factor

GJA3

Gap Junction Protein Alpha 3

GPX3

Glutathione Peroxidase 3

SYCP1

Synaptonemal Complex Protein 1

ZP1

Zona Pellucida Glycoprotein 1

ZP2

Zona Pellucida Glycoprotein 2

ZP3

Zona Pellucida Glycoprotein 3

Authors’ contributions

PL conducted data analysis and visualization and drafted the manuscript. RL, FY, and YZ were responsible for writing review and editing. SL and XZ were in charge of sample collection and processing. HN and WX conceived the research and ensured funding. All authors assisted in reviewing the manuscript and agreed to its publication.

Funding

This research was supported by the National Natural Science Foundation of China(32370542,32470521).

Data availability

The raw sequencing data are available in the NCBI Sequence Read Archive (SRA) under BioProject accession numbers PRJNA1398788 and PRJNA1398664.

Declarations

Ethics approval and consent to participate

The animal protocols used in this study were in accordance with the Measures for the Administration of the Permit for experimental animals provided by the Ministry of Science and Technology of the People’s Republic of China (2nd ed. No. 593, 2001). Humane animal care and handling procedures were approved by the Guide for the Care and Use of Laboratory Animals prepared by the Ethics Committee of Anhui Normal University.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.