Comparative Proteomic Analysis of Gonadal Tissue in Solea senegalensis Reveals Reproductive Deregulation Associated with F1 Individuals

Abstract

Reproductive dysfunction in captive-bred Senegalese sole (Solea senegalensis) limits aquaculture production consolidation, particularly due to reduced fertility and poor sperm quality in F1 males. To elucidate the molecular mechanisms underlying this problem, a quantitative proteomic analysis was conducted using LC–MS/MS on gonadal tissues from wild and F1 males and females. A total of 2221 proteins were identified, of which 1797 were retained after quality filtering. Comparative analyses revealed clear segregation by origin (F1 [cultivated] and wild) and sex (male and female), and 86 proteins were differentially expressed between F1 and wild males. Functional enrichment showed significant downregulation of key reproductive processes in F1 males, including sperm–egg recognition, binding of sperm to zona pellucida, and acrosome reaction, suggesting impaired gamete interaction and fertilization ability. Conversely, F1 males displayed metabolic and proteolytic pathway enrichment, which is indicative of compensatory energy demands. Protein–protein interaction network analysis identified a reproductive subnetwork dominated by zona pellucida sperm-binding proteins, which exhibited reduced connectivity in F1 males. These results demonstrate a coordinated suppression of molecular components essential for sperm–egg communication and acrosomal exocytosis, providing proteomic evidence for the systemic deregulation of the reproductive machinery in F1 fish. This study identifies potential protein biomarkers linked to reproductive performance, offering molecular targets to improve broodstock management and fertilization success in S. senegalensis aquaculture.

1. Introduction

Solea senegalensis (Kaup 1858) is a flatfish belonging to the Soleidae family that is mainly distributed along the Mediterranean and Atlantic coasts of southeastern Europe and northeastern Africa [1]. S. senegalensis is a species of high commercial interest, with global aquaculture production reaching 2284 tonnes in 2023 [2]. For several years, it has been proposed as a species with strong potential for European aquaculture diversification due to its high market value and production capacity [3]. Reproduction control in aquaculture is essential to provide good quality gametes and thus obtain mass production of larvae and to implement improvement programs that preserve traits of commercial interest [4,5]. The primary bottleneck in S. senegalensis farming is the reproductive dysfunction of males born and reared in captivity (hereafter referred to as F1) [6]. This dysfunction is related to reproductive behavior, such as the lack of courtship by F1 males and mate selection by females [7].

Senegalese sole is oligospermic (produces < 130 μL of semen) and exhibits asynchronous and semicistic spermatogenesis, i.e., the differentiation of haploid spermatids into spermatozoa (spermiogenesis) occurs within the lumen of the seminiferous tubules [5,8]. However, sperm production and quality, measured in terms of sperm volume and motility, is lower in F1 sole than in wild specimens, which could be a reason for the low fertilization of the F1 [9].

Strategies aimed at addressing the reproductive problem of S. senegalensis, including hormonal therapies [5,10,11,12], in vitro fertilization [13], and sole diet supplementation with macroalgae [14], have been diverse in recent years. However, none of these interventions have yet provided a definitive solution to the problem.

Furthermore, coexistence experiments indicate that F1 broodstock exposed to wild breeders (either currently or during their juvenile stage) exhibit behavioral learning. This interaction facilitates reproductive success, including courtship and fertilization; however, these positive outcomes tend to diminish over time. Furthermore, it has been determined that, in these soles, there was minimal participation of breeders in spawning and fertilization (only one pair is responsible for the offspring), which leads to a loss of genetic variability in the offspring [15,16].

With the development of omics techniques, the first transcriptome assembly was performed by Benzekri et al. [17]. Subsequently, three assemblies of the sole genome have been made [18,19,20], identifying the follicle-stimulating hormone receptor, FSHR, as the sex-determining gene of S. senegalensis [20].

Recent studies of the methylome in gonadal samples of S. senegalensis males at different stages of maturity and of different origin (wild or reared in captivity) have shown that those reared in captivity (F1) showed higher levels of methylation than wild ones, with DMCpG (Differentially Methylated CpG) sites associated with genes related to sex differentiation and gonadal development [21,22]. Thus, methylation at CpG sites was negatively correlated with the expression of genes involved in calcium, TGF-beta, and MAPK signaling pathways, which were deregulated in F1 male sole. These pathways are crucial for cell differentiation, gonadal development, maturation and cell activation, which are related to reproductive processes [22]. Hypermethylation of the MAPK pathway and low expression of the fshr gene in F1 sole testes could alter the endocrine axis, resulting in reduced fertility [22]. Furthermore, genes associated with olfactory receptors and behavior in S. senegalensis were overexpressed in wild-type males compared to F1 [22,23].

The determination of proteomic profiles in gonadal tissues provides detailed insight into the biological processes regulating reproductive physiology in teleost fish, including sex-specific differences in gametogenesis, steroidogenesis, and fertilization-related mechanisms, as demonstrated in previous proteomic studies of ovaries and testes in several teleost species [24,25,26]. In particular, comparative proteomics has emerged as a powerful tool for assessing changes in protein expression associated with environmental variations and the selection of successive generations for commercially relevant species in aquaculture breeding programs [27].

Aquaculture breeding programs generate divergences in omic profiles, which can influence reproductive quality, gonadal maturation, and protein expression patterns associated with essential biological functions [28]. These differences in omic profiles between wild and F1 individuals can affect gamete production and, consequently, the success of restocking and breeding programs [29]. Previous studies [25,30] have identified key proteins involved in processes such as hormonal biosynthesis, cytoskeleton remodeling, and the synthesis of signal transduction factors during gonadal differentiation in flatfish. However, the molecular mechanisms associated with adaptation in S. senegalensis F1 individuals remain poorly understood.

Proteomic studies in the Senegalese sole using gonadal tissue are sparse and not recent, the first being carried out by Forné et al. [31] who studied changes in protein levels during spermatogenesis in two populations of Senegalese sole, one wild and one F1, finding reduced levels of proteins related to sperm motility in F1s or increased levels of proteins involved in maintaining the redox status in F1, leading to sperm malformation. Decreased levels of ferritin and keratins have also been observed in F1s, leading to impaired fertility in F1 [32].

In this context, the present study aims to compare the protein expression profiles in gonadal tissues of wild-type and F1 individuals of S. senegalensis using a quantitative proteomic approach based on mass spectrometry. This analysis is intended to identify significant molecular differences that can provide insights into how captive conditions and selection affect the mechanisms of protein regulation in the gonads of males and females, thereby contributing to the development of more efficient reproductive management strategies for this species.

2. Materials and Methods

2.1. Individuals and Sampling

Samples were grouped according to two factors: origin (wild-origin [Wild] and captive-born and reared [F1]) and sex (male and female). Wild-type refers to adult S. senegalensis individuals captured directly from the natural environment, whereas F1 denotes first-generation individuals born and reared entirely from wild broodstock under aquaculture conditions. A total of 18 gonadal samples were processed and divided into four experimental groups (

Table 1. Summary of samples used in the study, organized by experimental group (origin and sex) and biological replicate identifiers.

Group	Sex 2	Origin 1	Replicas	Samples
F1F	F	F1	5	F1F-1, F1F-2, F1F-3, F1F-4, F1F-5
F1M	M	F1	5	F1M-1, F1M-2, F1M-3, F1M-4, F1M-5
wtF	F	wt	5	wtF-1, wtF-2, wtF-3, wtF4, wtF-5
wtM	M	wt	3	wtM-1, wtM-2, wtM-3

¹ F1: cultivated; wt: wild-type. ² F: female; M: male.

). Individuals corresponded to sexually differentiated males and females, and the gonadal developmental stage was phenotypically determined during dissection. Samples were codified using the initial letter corresponding to the origin (wt or F1), followed by the letter representing the sex (M or F), and a number identifying the biological replicate.

2.2. Protein Extraction

Proteins were extracted from 0.5 mg of gonadal tissue using QIAzol™ Lysis Reagent (QIAGEN, Hilden, Germany), following the RY16 protocol developed by the manufacturer for protein extraction from lipid-rich tissues. The gonadal tissue was homogenized in QIAzol using a FastPrep-24™ Classic instrument (MP Biomedicals, Santa Ana, CA, USA). system, and phase separation was performed by chloroform addition. The organic/interphase fraction was recovered for protein precipitation after centrifugation. Absolute ethanol was added to remove residual nucleic acids, followed by centrifugation at 2000× g for 2 min at 4 °C. The resulting phenol–ethanol supernatant containing the protein fraction was transferred to a new tube, and the proteins were precipitated by the addition of isopropanol, incubated for 10 min at room temperature, and subsequently centrifuged at 12,000× g for 10 min at 4 °C. The protein pellet was washed three times with a guanidine–ethanol solution (0.3 M guanidine hydrochloride in 95% ethanol) and incubated at room temperature for 20 min for each wash, followed by a final wash with absolute ethanol. After air-drying, proteins were resuspended in a urea/DTT solution (10 M urea, 50 mM DTT) and incubated at room temperature for 1 h to facilitate solubilization. The samples were then heated at 95 °C for 3 min and sonicated on ice using short pulses until complete protein dissolution was achieved. Finally, the samples were centrifuged at 10,000× g for 10 min at room temperature to remove the insoluble material. The soluble protein fraction was recovered, and the protein concentration was determined by fluorometry using a Qubit™ 4.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) with the Qubit™ Protein and Protein Broad Range (BR) Assay Kits according to the manufacturer’s instructions. Protein extracts were stored at −20 °C until further analysis by LC–MS/MS.

2.3. Preparation of Samples for Proteomic Study

Protein extracts were cleaned-up in 1D SDS-PAGE at 10% of polyacrilamyde. Samples were loaded in a stacking gel, and 100 V was applied until the electrophoresis front reached the resolving gel. After protein extract were separated 1 cm in resolving gel, electrophoresis was completed, and the gel was stained with Commasie Blue. Protein bands were diced and kept in water until digestion. For the digestion, briefly, protein bands were firstly distained in 200 mM ammonium bicarbonate (AB)/50% acetonitrile for 15 min and 5 min in 100% Acetonitrile. Protein was reduced by addition of 20 mM dithiothreitol in 25 mM AB and incubated at 55 °C for 20 min. The mixture was cooled to room temperature, followed by alkylation of free thiols by addition of 40 mM iodoacetamide in 25 mM BA in the dark for 20 min. The protein bands were washed twice in 25 mM AB. Proteolytic digestion was performed by addition of Trypsin (Promega, Madison, WI, USA), 12.5 ng/ul of enzyme in 25 mM AB and incubated at 37 °C temperature overnight. Protein digestion was stopped by addition of trifluoroacetic acid at 1% final concentration. Digest samples were dried in speedvac.

2.4. nLC-MS2 Analysis

Briefly, nano LC was performed in Dionex Ultimate 3000 nano UPLC (Thermo Scientific, Waltham, MA, USA) with a C18 75 μm × 50 cm Acclaim Pepmam column (Thermo Scientific, Waltham, MA, USA). Previously, peptide mix was loaded in a 300-μm × 5 mm Acclaim Pepmap precolumn (Thermo Scientific, Waltham, MA, USA) in 2% acetonitrile/0.05% TFA for 5 min at 5 μL/min. Peptide separation was performed at 40 °C for all runs in. Mobile phase buffer A was composed of water, 0.1% formic acid. Mobile phase B was composed of 20% acetonitrile, 0.1% formic acid. Samples were separated at 300 nl/min. Mobile phase B increases to 4–45% B for 60 min; 45–90% B for 1 min, followed by a 5 min wash at 90% B and a 15 min re-equilibration at 4% B. Total time of chromatography was 85 min. Eluting peptide cations were converted to gas-phase ions by nano electrospray ionization and analyzed on a Thermo Orbitrap Fusion (Q-OT-qIT, Thermo Scientific, Waltham, MA, USA). Mass spectrometer was operated in positive mode. Survey scans of peptide precursors from 400 to 1500 m/z were performed at 120 K resolution (at 200 m/z) with a 5 × 10 ion count target. Tandem MS was performed by isolation at 1 Th with the quadrupole, CID fragmentation with normalized collision energy of 35, and rapid scan MS analysis in the ion trap. The AGC ion count target was set to 105 and the max injection time was 75 ms. Only those precursors with charge state 2–5 were sampled for MS2. The dynamic exclusion duration was set to 15 s with a 10-ppm tolerance around the selected precursor and its isotopes. Monoisotopic precursor selection was turned on. The instrument was run in top speed mode with 3 s cycles, meaning the instrument would continuously perform MS2 events until the list of non-excluded precursors diminishes to zero or 3 s, whichever is shorter. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (Cambridge, UK) via the PRIDE (1) partner repository with the dataset identifier PXD071796.

2.5. Data Analysis

The raw data were analyzed using PEAKS Studio ProX (version 10.6; Bioinformatics Solution Corp Inc., Waterloo, ON, Canada). The reference library was acquired from NCBI and corresponded to the S. senegalensis RefSeq protein database derived from the genome assembly IFAPA_SoseM_1 (assembly accession GCF_019176455.1). The raw data were analyzed with parent mass error tolerance set to 15 ppm and a fragment mass error tolerance of 0.05 Da. The following settings were used to account for post-translational modifications and chemical labeling: Carbamidomethylation of cysteine residues was set as fixed modification, methionine oxidation and Acetylation (Protein N-term) was set as variable modification. Protein unique peptides was set to larger than 1 and a high confidence score of −10IgP > 20 was applied to indicate an accurately identified protein.

2.6. Bioinformatic Study

Bioinformatic analysis was performed on a quantification matrix in the R environment (v4.3.1). Data were transformed into the log₂ scale and filtered to retain only proteins with valid values in at least three biological replicates. Differential expression analysis was conducted using the limma package (v3.56.2), applying linear models and contrasts between experimental groups. Proteins were considered differentially expressed (DEPs) if they met the criteria of a p-value < 0.05 and an absolute difference in log₂-fold change > 0.58.

For functional annotation, the protein sequences corresponding to the DEPs were subjected to a similarity analysis using BLASTp Version 2.15.0 against the Danio rerio proteome via the NCBI web platform (https://blast.ncbi.nlm.nih.gov). The homologous identifiers obtained were processed in R and subsequently used for functional enrichment analysis of Gene Ontology (GO) terms and KEGG pathways using the online tool g:Profiler (https://biit.cs.ut.ee/gprofiler, version 2024.05).

Protein–protein interaction (PPI) networks were constructed using the differentially expressed proteins (DEPs) that were downregulated in the F1M vs. wtM and F1F vs. F1M comparisons. The RefSeq identifiers were first converted to D. rerio identifiers through a BLASTp analysis and subsequently used as input for the STRING database (https://string-db.org/) version 12.0, applying a confidence score threshold of ≥0.7. The resulting network was exported in graphML format and imported into Cytoscape v3.10.0 for visualization and topological analysis. Centrality metrics (degree, betweenness, and closeness) were calculated. Functional annotation and node categorization were performed through the cross-importation of specific protein lists, highlighting key proteins with reproductive functions for subsequent analysis and interpretation.

3. Results

3.1. Identification and Quantification of Proteins

A total of 2221 proteins were identified across the 18 gonadal tissue samples of S. senegalensis (Table S1). After filtering, 1797 proteins were retained, specifically considering those detected in at least three replicates within the same experimental group (Figure 1). The analysis of shared proteins between groups revealed a core set of 955 proteins common to all conditions, alongside group-specific expression patterns. Specifically, 9 proteins were identified exclusively in the F1F group, 18 in the F1M group, 7 in the wtF group, and 3 in the wtM group (Table S2). Furthermore, shared subgroups were observed among specific combinations, such as 8 proteins common between wtF and wtM and 19 between F1F and F1M, suggesting internal similarities between groups sharing the same origin (Figure 1).

Exploratory Proteomic Analysis and Identification of Exclusive Proteins

After normalizing the intensities, the proteins detected in a minimum of two replicates within each experimental group were considered for subsequent analysis. This criterion, implemented using less restrictive thresholds, enabled the scope of detection to be expanded without compromising the biological validity of the resulting data. This approach facilitated the robust identification of proteins exhibiting group-specific expression within the dataset. A substantial number of unique proteins were detected, including 50 exclusives to wild-type individuals, 68 to F1 individuals, 385 to males, and 108 to females. This outcome clearly reveals differential expression patterns associated with the origin and sex of the analyzed specimens. The comparison of proteins exclusive to F1 males (F1M) vs. those of wild-type males (wtM) revealed a divergent protein expression profile (Table S3). Specifically, 33 proteins were found to be unique to wtM, and 23 were exclusive to F1M.

Exclusive proteins identified in the F1M group included, among others: arpin-like protein (XP_043873833.1), filamin-binding LIM protein 1 (FBLIM1; XP_043880017.1), cysteine and glycine-rich protein 3 (CSRP3; XP_043880807.1), tubulin polymerization-promoting protein family member 2 (TPPP2; XP_043883540.1), protein PAT1 homolog 1 (PATL1; XP_043884691.1), mitochondrial heat shock protein 75 kDa (Hsp75; XP_043906303.1|XP_043906302.1), and sarcoplasmic/endoplasmic reticulum calcium ATPase 2 (SERCA2; XP_043897940.1|XP_043897939.1). These proteins were implicated in the regulation of cellular morphology, cytoskeletal stability and cell motility, microtubule dynamics, RNA processing, and calcium homeostasis. Furthermore, they are considered fundamental for the maintenance of testicular function [33,34,35]. These findings suggest that the F1M group exhibits a distinct regulatory state of the cytoskeleton and cellular mobility, possibly reflecting an adaptation or disruption related to their origin. Conversely, the proteins exclusively detected in the wtM group included components strongly linked to sperm function and gamete interaction (Table S3). These specific proteins were cilia- and flagella-associated protein 53 (CFAP53; XP_043868067.1, XP_043868068.1), multiple isoforms of tubulin α/β (axonema), sperm acrosome membrane-associated protein 4-like (SPACA4) chain oocyte (XP_043892398.1), RIB43A-like with coiled-coils protein 2 (RIBC2; XP_043892598.1), and nucleoside diphosphate kinase 7 (NME 7; XP_043900902.1, XP_043900901.1) (Table S3).

This differential profile suggests that, while F1 males maintain conserved ‘core’ modules (e.g., cell dynamics, metabolism, and proteostasis), they are deficient in key motility and acrosomal markers that are distinctly present in the wild-type (wt) population. This finding is entirely consistent with the dysregulation of reproductive processes inferred from subsequent Gene Ontology (GO) and PPI analyses, which will be presented further in the paper.

3.2. Correlation and Clustering Analysis

The Pearson correlation coefficient (Figure S1a) demonstrated a clear segregation among the experimental groups, revealing a strong intragroup correlation and a lower intergroup correlation, particularly between the wild-type (wt) and F1 individuals, as well as between sexes. Additionally, the Principal Component Analysis (PCA) confirmed this structure by identifying a clear separation between males and females, and between the groups of different origin (F1 and wt), primarily along the first two principal components (Figure S1b).

Furthermore, the protein expression analysis, visualized via the heatmap (Figure 2), revealed differentiated clustering patterns of the samples based on both origin and sex. The replicates corresponding to F1 female individuals (F1F) demonstrated high internal consistency and were clearly grouped into a separate cluster. In contrast, the samples from males of different origins (F1M and wtM) also tended to cluster together, albeit with greater dispersion. These observations from the correlation and clustering analysis collectively suggest that both the origin (wild-type vs. F1) and sex significantly contribute to the overall variability of protein expression in the gonads, thereby constituting biologically relevant factors within the experimental model.

3.3. Differential Expression Analysis

The differentially expressed protein (DEP) analysis was conducted for four experimental comparisons between the groups: F1 males versus wild-type males (F1M vs. wtM), F1 females versus F1 males (F1F vs. F1M), F1 females versus wild-type females (F1F vs. wtF), and wild-type females versus wild-type males (wtF vs. wtM).

The comparison between males of different origins (F1M vs. wtM) (Figure 3a) showed a moderate number of differentially expressed proteins (DEPs), with 47 proteins overexpressed in the wild-type (wt) and 39 underexpressed in the F1 group, compared to 1608 proteins with no significant change in expression. Among the downregulated proteins in F1M were the enzyme glutathione peroxidase 1b (GPX1b) (XP_043879076.1), which prevents cellular damage caused by oxidative stress; keratin type II cytoskeletal 8 (KRT8) (XP_043892356.1), responsible for maintaining cellular integrity in response to stress and indirectly related to fertility, as its expression is linked to processes such as embryonic development; neuroblast differentiation-associated protein AHNAK (AHNAK) (XP_043896809.1), which regulates calcium channels, blood–brain barrier formation, embryonic development, and lipid metabolism, among other processes. Table S4 shows the five most differentially overexpressed and underexpressed proteins in each compared group, which are indicated in Figure 3. The largest number of proteins with significant changes was observed in the F1F vs. F1M comparison (Figure 3b): 524 proteins were overexpressed in males and 77 were underexpressed in females, while 1178 proteins showed no significant differences. These results demonstrated a marked divergence in the proteomic profile between females and males within the F1 group. In the F1F vs. wtF comparison, 90 proteins were overexpressed in wild-type females and 27 were underexpressed in the F1 females. The majority of proteins (1331) did not exhibit significant changes (Figure 3c). Finally, in the wtF vs. wtM comparison, 271 proteins were identified as overexpressed in males and 35 as underexpressed in females, while 1445 proteins showed notable differences (Figure 3d). The observed distribution indicates differential regulation of the proteome associated with sex, which is maintained under natural conditions even in wild-type individuals.

3.4. Functional Enrichment Analysis in Gene Ontology (GO)

To characterize the biological functions associated with DEPs, a functional enrichment analysis was performed for the four comparisons based on Gene Ontology (GO) annotations. The annotated DEPs were classified into three main GO categories: Biological Process (BP), Cellular Component (CC), and Molecular Function (MF). This classification allowed the interpretation of the functional context of the observed proteomic changes. The GO functional enrichment analysis revealed a consistent pattern of underexpression of proteins associated with reproductive functions in the comparisons involving F1 males. This finding suggests a possible compromise in reproductive competence mechanisms under culture conditions.

In the F1M vs. wtM comparison (Figure 4a), F1 males exhibited a repression of processes such as acrosome reaction (GO:0007340), egg coat formation (GO:0035803), and supramolecular complex (GO:0099080). This pattern suggests a systematic deregulation of functions essential for fertilization and proper sperm function. Consistently, in the F1F vs. F1M comparison (Figure 4b), the underexpressed proteins in F1 males showed significant enrichment in terms of reproductive functions, including binding of sperm to zona pellucida (GO:0007339), fertilization (GO:0009566), sperm–egg recognition (GO:0035036), and regulation of reproductive process (GO:2000241). These findings reflect a reduced abundance of proteins involved in sperm recognition, adhesion, and fusion events. Comparisons of F1F vs. wtF (Figure S2a) and wtF vs. wtM (Figure S2b) revealed distinct functional enrichment patterns. An increased representation of structural processes and cytoskeleton reorganization was observed in F1 females, whereas underexpressed proteins were mainly associated with ribonucleoprotein assembly and post-transcriptional regulation. In the wtF vs. wtM comparison, wild-type females showed enrichment of metabolic processes and protein maturation, whereas wild-type males exhibited a higher representation of terms related to lipid transport and lipoprotein metabolism.

GO Level 2 Term Enrichment in F1M vs. wtM and F1F vs. F1M Comparisons

To delve deeper into the Biological Process (BP) category, an additional analysis was conducted using GO Level 2 terms. This hierarchical level allows for the identification of broad yet functionally informative biological processes, thereby facilitating the comparison between experimental conditions that involved F1 male individuals. Consequently, the analysis focused solely on the comparisons between F1M and wtM and between F1F and F1M. This decision was made because these comparisons exhibited a greater functional diversity and biological relevance of the enriched terms, in contrast to the Cellular Component (CC) and Molecular Function (MF) categories, where the terms at this level presented less variability and interpretive contribution.

In the comparison between F1 males and wild-type males (F1M vs. wtM), downregulated proteins in F1 males (Figure 5a) showed a strong representation of terms associated with essential reproductive processes, including sperm–egg recognition (GO:0035036), binding of sperm to zona pellucida (GO:0007339), acrosome reaction (GO:0007340), egg coat formation (GO:0035803), and positive regulation of acrosome reaction (GO:2000345). As these processes are critical for sperm maturation and sperm–oocyte interaction during fertilization, their reduced representation suggests a deregulation of the reproductive molecular machinery in F1 males. In addition, a downregulation of proteins associated with muscle cell differentiation (GO:0042692) and muscle structure development (GO:0061061) was also observed. This may reflect alterations in flagellar architecture and sperm cytoskeletal dynamics, potentially affecting sperm motility and fertility competence.

Conversely, upregulated proteins in F1 males (Figure 5b) were mainly associated with metabolic and energy-related processes, including the phosphagen metabolic process (GO:0042400), phosphocreatine biosynthetic process (GO:0014862), glycolytic process (GO:0006096), and ADP catabolic process (GO:0046033). In addition, pathways related to purine ribonucleoside diphosphate catabolic process (GO:0009154) and intracellular iron ion homeostasis (GO:0006879) were also enriched. This pattern indicates increased bioenergetic activity and cellular maintenance in the gonads of F1 individuals, potentially reflecting metabolic compensation in response to culture conditions that may require higher energy expenditure to sustain tissue homeostasis. In the F1F vs. F1M comparison (Figure S3a), F1 males exhibited a significant negative regulation of proteins involved in key reproductive functions. The most prominent terms included sperm–egg recognition (GO:0035036), acrosome reaction (GO:0007340), egg activation (GO:0007343), and binding of sperm to zona pellucida (GO:0007339). These results point to a reduced expression of components essential for fertilization and sperm competence, which may be influenced by differences in sexual maturation between sexes and the captive environment [36]. Conversely, the upregulated proteins in F1 females (Figure S3b) were predominantly associated with intracellular transport processes. Among the most notable pathways were the positive regulation of protein localization to endosome (GO:1903217), early endosome to late endosome transport (GO:0042147), and protein localization to nuclear envelope (GO:1902594). This suggests that gonads have increased activity in mechanisms related to subcellular organization and vesicular dynamics. Overall, these results highlight a clear functional divergence, whereby the gonads of F1 males prioritize metabolic and energy-production pathways, whereas wild-type individuals maintain the expression of key components associated with fertilization and spermatogenic maturation. This evidence supports a functional reorientation of the gonadal proteome in F1 males, in which a central mechanism underlying the reduced reproductive capacity observed in cultured individuals may be an imbalance between metabolism and reproduction.

3.5. Functional Analysis in KEGG Pathways

The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis revealed a predominance of metabolic processes, intracellular transport, and cellular structures, whereas pathways directly related to reproduction were generally scarce. In the F1M vs. wtM comparison, downregulated pathways in F1 males included Phagosome, Adherens junction, and Pentose phosphate pathway, although with lower significance (Figure S4a), indicating subtle structural and functional modulation in F1 individuals. Conversely, several overexpressed pathways related to metabolism and proteolysis, including Glycolysis/Gluconeogenesis, Fatty acid degradation, Pyruvate metabolism, and Proteasome, were identified in F1 males, suggesting the activation of energetic and protein turnover pathways. Furthermore, the enrichment of motor proteins and Gap junction pathways indicates alterations in cytoskeletal dynamics and intercellular communication. In the F1F vs. F1M comparison (Figure S4b), downregulated proteins in F1 males (or upregulated proteins in F1 females) were significantly enriched in pathways such as Spliceosome, RNA degradation, and Ribosome biogenesis in eukaryotes, all of which are involved in post-transcriptional regulation and protein synthesis. In contrast, upregulated proteins in F1 males (or downregulated in F1 females) were associated with metabolic and signaling pathways, including Glycolysis/Gluconeogenesis, Apoptosis, Tight junction, Pyruvate metabolism, and Carbon metabolism, indicating bioenergetic and signaling differences between sexes. The F1F vs. wtF comparison (Figure S4c) showed a higher proportion of downregulated pathways in F1 females (i.e., enriched in wild-type females), notably Spliceosome, Ribosome biogenesis in eukaryotes, RNA degradation, and Citrate cycle (TCA cycle), suggesting a reduced RNA processing capacity and energy metabolism in F1 females. In contrast, the pathways overexpressed in F1 females included ECM–receptor interaction, Focal adhesion, Lysosome, and Ferroptosis, which potentially reflect enhanced cell–matrix interactions, intracellular degradation, and oxidative stress responses. Finally, in the wtF vs. wtM comparison, multiple pathways overexpressed in wild-type females were identified, primarily linked to protein degradation and metabolic processes, such as Carbon metabolism, Proteasome, Biosynthesis of amino acids, Valine, leucine and isoleucine degradation, Fatty acid degradation, and Glycolysis/Gluconeogenesis. Downregulated pathways included Nucleocytoplasmic transport and Protein processing in the endoplasmic reticulum, although with lower statistical significance (Figure S4d), indicating higher metabolic and proteolytic activity in WT females than in males. Collectively, KEGG pathway analysis highlights pronounced differences in metabolic and proteolytic activity between sexes and between F1 and wild-type individuals, whereas pathways classically associated with reproduction remain sparsely represented.

3.6. PPI Network Analysis of Differentially Expressed Proteins

The protein–protein interaction networks constructed from downregulated proteins in the F1M vs. wtM and F1F vs. F1M comparisons revealed dense and modular architectures, reflecting complex molecular coordination within gonadal tissue. In the F1M vs. wtM comparison (Figure 6a), the PPI network comprised 99 nodes and 318 edges and exhibited a relatively fragmented structure. Within this network, ZP3a.1 and ZP3a.2 formed an isolated 17-node subnetwork with high internal cohesion (degree = 16 and 12, respectively). This organization suggests a disruption of reproductive connectivity in F1 males, consistent with the downregulation of GO biological processes associated with sperm–egg binding (GO:0007339), acrosome reaction (GO:0007340), and fertilization (GO:0009566). In contrast, the PPI network corresponding to the F1F vs. F1M comparison (Figure 6b) included 204 nodes and 428 edges and displayed a dense and highly interconnected architecture with a prominent peripheral cluster. This cluster was composed mainly of zona pellucida (ZP) glycoproteins, including NP_001013289.1 (ZP3a.1), NP_001025291.1 (ZP3a.2), NP_001034972.2 (ZP2.1), NP_571405.2, NP_571903.2, NP_571904.2, and XP_002665864.2 (ZP4). These proteins showed high local connectivity (degree = 30–42) and multiple interactions with structural and regulatory nodes, suggesting a coordinated role in sperm–oocyte recognition, fertilization, and zona pellucida remodeling. Overall, these results indicate that the concerted deregulation of zona pellucida glycoproteins in F1 individuals alters the reproductive interactome network, potentially impairing sperm recognition capability and fertilization efficiency. This provides a robust molecular framework to explain the low reproductive capacity observed in S. senegalensis F1 individuals, supporting the idea that culture conditions induce systemic proteomic reprogramming affecting both structural organization and signaling pathways essential for reproductive success.

PPI Subnetwork Analysis

PPI subnetworks were constructed from the F1M vs. wtM and F1F vs. F1M comparisons to specifically examine functional connections among proteins associated with reproductive processes, considering only downregulated proteins linked to GO terms related to fertilization and sperm–egg binding.

Three main proteins—NP_001013289.1, NP_001025291.1, and NP_999937.2—were identified in the F1M vs. wtM subnetwork (Figure 7a), all associated with GO terms related to fertilization, including acrosome reaction (GO:0007340), sperm–egg recognition (GO:0035036), and binding of sperm to zona pellucida (GO:0007339). These proteins formed a peripheral but tightly connected module, indicating a reduced yet specific functional interaction that may be affected by the downregulation observed in F1 males. The loss of connectivity within this group suggests an alteration of molecular pathways involved in sperm competence and oocyte activation. In contrast, the F1F vs. F1M subnetwork (Figure 7b) revealed a cluster of seven key proteins (NP_001013289.1, NP_001025291.1, NP_001034972.2, NP_571405.2, NP_571903.2, NP_571904.2, and XP_002665864.2) forming a densely interconnected module. These proteins, homologous to Danio rerio zona pellucida sperm-binding proteins, were associated with GO terms such as binding of sperm to zona pellucida (GO:0007339), acrosome reaction (GO:0007340), and sperm–egg recognition (GO:0035036). NP_001013289.1 and NP_001025291.1 displayed the highest centrality values within this subnetwork, suggesting a regulatory role in sperm recognition and adhesion mechanisms. The topological organization of these proteins indicates that they act as coordinated functional nodes, whose underexpression could compromise the critical stages of oocyte maturation and fertilization.

Both subnetworks provide evidence of a convergent pattern of low expression in the zona pellucida (ZP) family proteins. This suggests that the reproductive dysfunction in S. senegalensis F1 individuals may be associated with a reorganization of the molecular interactions that sustain fertilization and gametic recognition events.

4. Discussion

This study performed a comparative proteomic analysis of gonadal tissue samples from F1 and wild-type S. senegalensis individuals to explore the molecular basis associated with the low reproductive rate observed in cultured fish. The identification of 2221 proteins across 18 samples reflected a robust coverage of the gonadal proteome, comparable to similar studies in fish such as Cynoglossus semilaevis [37] and Acipenser sinensis [38], where equivalent protein numbers have been reported following the application of LC-MS/MS strategies.

The observed pattern of both shared and unique proteins per group, along with the clustering demonstrated in the correlation analyses, indicated a strong influence of both sex and origin on the protein expression profile. This suggests a significant biological divergence between F1 and wild-type individuals, as well as between males and females, which is consistent with findings reported in other marine fish where substantial physiological differences have been observed [39].

The results of the functional analysis in the F1M vs. wtM comparison evidenced a significant repression of biological processes associated with reproduction in F1 males, reflected by the negative enrichment of Level 2 GO terms such as sperm–egg recognition (GO:0035036), binding of sperm to zona pellucida (GO:0007339), acrosome reaction (GO:0007340), egg coat formation (GO:0035803), and positive regulation of acrosome reaction (GO:2000345). These processes are essential for gamete interaction, as they regulate sperm–oocyte adhesion, acrosomal exocytosis, and penetration of the egg envelope—all indispensable events for successful fertilization [40,41]. The underexpression of proteins associated with these processes suggests a functional impairment of the factors involved in sperm competence and the fertilization capacity of F1 males. Specifically, the underexpression of proteins related to the acrosome reaction and sperm recognition indicates an alteration in the calcium-dependent exocytic machinery, which is necessary for releasing hydrolytic enzymes that enable zona pellucida penetration [42,43,44]. In flatfish, proteins homologous to the zona pellucida sperm-binding proteins (ZP2 and ZP3) and acrosomal vesicle proteins have been described as direct mediators of sperm binding and oocyte activation [45,46]. The observed reduction in these components in F1 males suggests that the processes of gamete adhesion and recognition may be partially compromised, thus impacting the efficiency of in vitro fertilization reported in cultured S. senegalensis individuals [3,11]. Furthermore, the enrichment of terms such as muscle cell differentiation (GO:0042692), muscle structure development (GO:0061061), and dendrite morphogenesis (GO:0048813) among the underexpressed proteins may reflect alterations in flagellar cytoskeletal organization, thereby affecting sperm motility. The axonemal structure of the flagellum relies on a coordinated assembly of microtubules (composed of tubulins) and motor proteins (such as dyneins). Dysregulation of this assembly has been associated with immotile or low-velocity spermatozoa in marine fish under culture conditions [47,48,49]. The term response to oxidative stress (GO:0006979) within this suppressed set reinforces the hypothesis that oxidative stress is a limiting factor for sperm quality in F1 males. GPX1b was one of the proteins observed to be downregulated in F1M. The overproduction of reactive oxygen species (ROS) in the gonads under environmental or metabolic stress can induce mitochondrial damage, lipid peroxidation, and sperm DNA fragmentation [50,51]. The simultaneous reduction in terms associated with antioxidant defense and muscle cell differentiation processes may reflect an accumulated physiological vulnerability in F1 individuals, manifested as a decline in sperm quality and functionality.

The differential expression analysis demonstrated that the F1F vs. F1M comparison yielded the highest number of DEPs, reflecting a pronounced gonadal molecular difference under culture conditions. This finding is consistent with what has been described in D. rerio, where sex-specific differences in gonadal protein expression are evident [25]. Comparisons between F1 and wild individuals (F1F vs. wtF and F1M vs. wtM) showed a more limited number of DEPs, which may suggest that, while alterations induced by the culture environment exist, their magnitude does not reach that observed between sexes within the same origin. Nevertheless, the wtF vs. wtM comparison also evidenced significant differential regulation, indicating the presence of proteomic sexual dimorphism in wild individuals, albeit with profiles distinct from those observed in the F1 generation.

The GO functional analysis revealed key findings. In the F1F vs. F1M comparison, overexpressed proteins in females were associated with cytoskeletal organization and intracellular trafficking, functions necessary for oocyte development and cytoplasmic reorganization, as previously indicated by studies on teleost oocytes [52,53]. Conversely, the underexpressed proteins in males included processes directly implicated in fertilization and gamete interaction, such as acrosome reaction and binding of sperm to zona pellucida. This could, at least in part, account for the low reproductive competence observed in F1 males. Similar reproductive impairments in cultured males have been reported in other cultured teleost species. For instance, in the greater amberjack (Seriola dumerili), male individuals exhibited a more pronounced gonadal regression and lower steroid activity than their wild counterparts [54]. Likewise, genomic signatures of selection that could be associated with adaptation to the culture environment and reproductive efficiency changes have also been identified in domesticated rainbow trout (Oncorhynchus mykiss) populations [55].

The KEGG pathway enrichment analysis indicated that the major differences between the groups were concentrated in metabolic and intracellular processing pathways. The limited representation of classic reproduction-associated pathways reinforces the idea that the observed effects may be mediated by indirect or post-transcriptional mechanisms rather than by the direct activation of reproductive pathways. Nevertheless, the overexpression of pathways such as Apoptosis, Tight junction, and ECM-receptor interaction could be related to gonadal remodeling mechanisms that have been reported as critical during teleost gametogenesis [56].

Finally, the PPI analysis allowed the extraction of functional subnetworks centered on proteins linked to reproductive processes, identifying a set of proteins homologous to zona pellucida (ZP) glycoproteins, including zp3a.1 (NP_001013289.1), zp3a.2 (NP_001025291.1), zp2.1 (NP_001034972.2), zp2.3 (NP_571903.2), zp2.5 (NP_571405.2), and zp4 (XP_002665864.2). These proteins are essential components of the zona pellucida envelope in teleost fish and play key roles in oocyte extracellular matrix formation and sperm–egg interaction [57]. In D. rerio, ZP proteins are sex-dependently expressed and localized in the ovarian follicle, directly participating in the attraction and specific binding of spermatozoa to mature oocytes [58]. Their presence has been extensively validated through transcriptomic, proteomic, and immunohistochemical studies [59]. Specifically, zp3a.1 and zp3a.2 act as ligands for sperm receptors, thereby determining the success of fertilization [60]. The underexpression of these proteins in F1 individuals, as observed in our PPI network and GO analysis, suggests a critical impairment in the zona pellucida architecture and functionality. Previous studies in fish have linked ZP genes suppression with stress-induced infertility or suboptimal culture conditions [61,62]. Furthermore, ZP proteins are not only structural but also modulators of the perivitelline environment, as they block polyspermy and activate signaling pathways after fertilization [63]. The peripheral yet highly connected localization of these proteins within the PPI subnetwork reinforces the hypothesis that their dysfunction could indirectly impact other functional networks associated with reproductive success, including vesicular transport mechanisms and ovarian matrix remodeling.

Collectively, our results suggest that the coordinated underexpression of a group of key reproductive proteins in F1 individuals may be linked to the low fertilization rates observed in culture, offering potential biomarkers for monitoring the reproductive status of S. senegalensis breeding programs. This approach has proven effective in teleosts such as Clarias magur, where PPI network analysis successfully identified genes associated with reproductive processes [64].

5. Conclusions

This comparative proteomic analysis provides a comprehensive overview of the gonadal proteome of S. senegalensis and demonstrates that protein expression patterns are strongly influenced by both sex and biological origin (wild-type versus cultured F1 individuals). The generated dataset offers robust coverage of gonadal proteins and highlights clear molecular differences associated with reproductive performance under aquaculture conditions.

A distinct testicular proteomic profile was identified in cultured F1 males compared with wild-type males. Specifically, F1 males exhibited a coordinated downregulation of proteins associated with key reproductive processes, including acrosome reaction, sperm–egg recognition, and oocyte activation. Functional enrichment analyses further revealed that this repression affects multiple biological pathways essential for successful fertilization, indicating a multifactorial reproductive dysfunction rather than the alteration of a single molecular mechanism.

Integration of GO enrichment and PPI network analyses reinforced these findings by identifying the underexpression of zona pellucida (ZP) homologous glycoproteins, such as zp3a.1 and zp3a.2, in F1 males. Given the central role of these proteins in sperm–oocyte interaction and fertilization, their reduced expression provides a plausible molecular basis for the low reproductive competence observed in cultured S. senegalensis. Collectively, these results indicate that captive rearing conditions induce a reprogramming of the gonadal proteome in F1 males, compromising reproductive efficiency and highlighting potential molecular targets for improving broodstock management in aquaculture.