Age-related reproductive performance and transcriptome profiling of testis in male blue catfish, Ictalurus furcatus

Abstract

Catfish farming is the largest aquaculture industry in the U.S., where hybrid catfish produced by channel catfish, Ictalurus punctatus ♀ × blue catfish, I. furcatus ♂, represent > 50% of the harvest. Previous studies indicated a high degree of variation in reproductive performance among individuals. Therefore, it is crucial to establish a connection between paternal age and reproductive success. This study investigated the reproduction of 103 blue catfish males aged 2 to 10 years. Morphometric data were collected, and blood was drawn to quantify testosterone (T), 11-ketotestosterone (11-KT), osmolality, and ions. Histological images of the testis assessed stages of spermatogenesis, and sperm were activated for kinematic analysis. The testis transcriptome was profiled across distinct ages, 2, 4, 7, and 9. For fish aged 2 to 10 years, those over 7 years had the largest size, with a rapid increase until age 7 and a slight increase afterwards. Absolute sperm production was affected by age, with the highest levels observed at age 6, followed by a decreasing trend. T and 11-KT did not increase after age 6, while sperm kinematics were not impacted by age. Transcriptomics identified 5220 differentially expressed genes (DEGs) in all comparisons. The most DEGs were identified at age 7 with 2261 down-regulated and 1824 up-regulated genes. Functional enrichment revealed significant changes in axoneme, motile cilium, and sperm flagellum at age 7. In contrast, immune processes were increased in the testis at age 9 with viral and bacterial defense responses. Results suggest that farmers should maintain consistent supplies of age 6 and 7 males to increase hatchery efficiency.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-025-12014-w.

Introduction

The demand for animal-source protein continues to increase due to the growing human population, which is expected to reach 9.7 billion by 2050 [1]. Combined, captured and farmed aquatic animals contribute ~ 15% of total animal protein consumed globally [2]. While capture fisheries production continuously declines, aquaculture food production compensates for an increasing share of human protein needs [3]. The U.S. catfish production is the leading domestic aquaculture industry, with the highest production by volume and economic value [4]. Mississippi, Alabama, and Arkansas tristate region accounted for > 93% of U.S. catfish production and 90% of the U.S. catfish production area in 2019 [5]. Hybrid catfish produced by crossing channel catfish (Ictalurus punctatus) females with blue catfish (I. furcatus) males have better production than their parental species and other catfish species [6], and about 53% of the water surface area in the tristate region was used for hybrid catfish production [5].

At present, a major challenge for the hybrid catfish industry is the need to sacrifice blue catfish males to collect sperm to be used to fertilize channel catfish eggs. This is because of the lack of natural, spontaneous hybridization in pond environments due to biological differences between these species [7]. As a result, sperm collection is a considerable investment due to the limited availability of milt, which can only be obtained once from each male after reaching sexual maturity between 4 and 7 years of age [8]. Thus, high-quality male gametes are one of the most important factors for improved fish production [9]. These cells are also critical for creating germplasm repositories for future breeding and fertilization needs [10]. Therefore, further understanding of the physiological and molecular mechanisms of sperm form and function associated with puberty and fish ageing is urgently needed to improve sperm collection and quality.

Puberty and sexual maturation are biological processes where an animal acquires the capacity for reproduction, which is induced and driven by a series of hormones of the brain-pituitary–gonadal axis [11, 12]. Gonadotropin-releasing hormone (GnRH) is released by the hypothalamus, which stimulates the release and secretion of the pituitary gonadotropins (luteinizing hormone and follicle-stimulating hormone), which further stimulate testis development and steroidogenesis in male fish [13, 14]. Testosterone (T) and 11-ketotestosterone (11-KT) are the main sex steroid hormones secreted by Leydig cells in the testis in response to gonadotropins [15, 16], act at lower levels of the hormonal cascade, and are involved in both genomic and non-genomic processes [17]. The androgen 11-KT is an oxygenated metabolite of T and is considered the most potent androgen responsible for paracrine action in Sertoli cells, leading to the completion of spermatogenesis in fish [11, 12]. Several research studies have indicated a positive relationship between circulating T and 11-KT levels with spermatogenesis in African catfish (Clarias gariepinus) [18], Japanese huchen (Hucho perryi) [19], and brill (Scophthalmus rhombus) [20]. Therefore, understanding the effects of fish ageing on circulating androgen levels provides insights into age-related changes in sperm quality and production.

Germ cells and somatic cells within the testis undergo complex and highly coordinated functional and histological changes once activated by hormones during different developmental stages [13, 21]. The resulting cell types are primarily defined as spermatogonia, spermatocytes, spermatids, and spermatozoa within spermatogenic tissue [22, 23]. Diploid spermatogonial stem cells have the capacity for self-renewal and initiate the process of spermatogenesis to differentiate into spermatogonia when dividing [24]. After the mitotic phase with different generations of spermatogonia, the developmental cycle transitions into the meiotic phase of spermatocytes. This is followed by the spermiogenic phase, where haploid spermatids are formed. Finally, spermatids transform into mature flagellated spermatozoa, which acquire motility [25]. As males progress through spermatogenesis, cells become more specialized and change in size and shape. The Spermatogenic Maturity Index (SMI) is a useful tool that assesses the stage of spermatogenesis by estimating the area fractions of various cell types of gamete development in histological sections of the testes [23]. Thus, quantitative histology can provide a more comprehensive picture regarding sperm production and ageing.

Sperm quality is defined by its capacity to fertilize an egg effectively and produce a viable offspring [9], which can also change as fish age. Motility and velocity are the critical swimming kinematic parameters commonly used as indicators of sperm quality [26, 27]. These parameters can be quantified with Computer Assisted Sperm Analysis (CASA) software [28]. Fish sperm remains immotile within the testis and seminal fluid and then becomes activated after release into the aquatic environment [27]. Sperm activation is due to changes in osmolality and ionic composition when exposed to water, and movement further depends on biochemical composition and energy contained within spermatozoa [29]. Besides the physiological differences between immature and mature fish, reproductive capacity and sperm kinematics can be affected by fish age. Accordingly, several studies have focused on the effects of ageing on sperm kinematics in rainbow trout (Oncorhynchus mykiss) [30], Arctic charr (Salvelinus alpinus) [31] and zebrafish (Danio rerio) [32]. Therefore, understanding sperm kinematics as an indicator of sperm quality for different age groups is critical for effective selection of broodstock.

Understanding molecular mechanisms related to reproduction is also critical for accurately predicting sperm form and function. Reproductive maturation in males involves a series of testis-related genes involved in the gonad development and production of functional germ cells. In recent years, RNA sequencing has commonly been used to profile gene expression, map potential pathways, and quantify the transcriptome [33]. The transcriptome acts as the key intermediate between the genome and proteome, reflecting active gene expression within a cell at a specific point in ontogeny [34]. Testicular tissue exhibits a high level of specific gene expression related to the reproductive system [35]. The testis transcriptome can highlight genes associated with testis development from immature to sexually mature that can produce active spermatozoa. Although an increasing number of studies have been focused on gonadal development in fish [36–38], little is known about age-related changes in the testis transcriptome.

Age selection is critical for identifying optimal broodstock for sperm collection in hybrid catfish production. However, the relationship between male age, sperm quality, and the underlying molecular mechanisms remains poorly studied in blue catfish males. To address this knowledge gap, we evaluated various parameters, including blood serum androgen levels, ions, testis histology, and sperm kinematics in blue catfish broodstock at 2 to 10 years of chronological age. We profiled the testis transcriptome across four distinct age groups (2, 4, 7, and 9 years), corresponding to immature, first sexual maturity, later sexual maturity (middle-aged) and older individuals. This combined physiological, histological, and transcriptomic approach will help identify age-related reproductive performance in blue catfish males which will aid in better management of commercial blue catfish populations for efficient mass production of hybrid catfish with higher growth potential than their parents.

Materials and methods

Ethical statement

All experimental animal protocols were approved by the Auburn University Institutional Animal Care and Use Committee (AU-IACUC) with protocol #2023–5229 and Warmwater Aquaculture Research Unit with IACUC protocol #FY23-001. All methods used in this experiment were performed by following relevant guidelines and regulations.

Blue catfish sampling

Fish were obtained from the United States Department of Agriculture, Agricultural Research Service (USDA-ARS) Warmwater Aquaculture Center in Stoneville, MS, U.S. In total, 103 blue catfish males were sampled from 14 to 18 May 2023 and 23 to 24 May 2024, representing 9 age classes ranging from 2 to 10 years of chronological age. Fish were tagged according to their age and reared in 0.4 ha earthen ponds before conducting the study. Dissolved oxygen and chlorides were maintained at > 5 ppm and > 100 ppm, respectively, in all broodstock ponds. The fish were fed with a 32% protein commercial catfish floating pellet (Delta Western, Indianola, MS). Fish were collected by seining and subsequently euthanized by blunt force trauma to the head following institutional animal care guidelines. The total weight, total length, head width, head circumference, and hump circumference were recorded. Blood (~ 5 to 10 mL) was immediately drawn from the caudal vein to determine serum levels of T, 11-KT, osmolality, and ions (Ca²⁺, Na⁺, K⁺, Mg²⁺, and Cl⁻) (see Sect. "Blood serum analyses"). Testes were dissected and carefully removed from the peritoneal cavity using forceps and surgical scissors and then weighed. Somatic weight was calculated (body weight − testes weight) to determine gonadosomatic index [GSI = (testes weight / somatic weight) × 100]. A small (~ 1 g) tissue sample was taken from the middle of the testis and preserved in 10% phosphate-buffered formalin for histological analysis (see Sect. "Testes histology"). The testicular tissue was stored in Hank’s Balanced Salt Solution (HBSS; reverse osmosis deionized water with 8 g/L NaCl, 0.4 g/L KCl, 0.16 g/L CaCl₂ × 2H₂O, 0.2 g/L MgSO₄ × 7H₂O, 0.12 g/L Na₂HPO₄ × 7H₂O, 0.06 g/L KH₂PO₄, 0.35 g/L NaHCO₃, 1 g/L glucose) [39] for sperm quality processing (see Sect. "Sperm collection and quality analyses"). An additional small testis sample (~ 0.5 g) was collected in triplicate from each male, snap-frozen in liquid nitrogen, and stored at − 80 °C for RNA extraction and transcriptomics (see Sect. "Transcriptomic analyses and validation").

Blood serum analyses

Sex-steroid hormones

Blood samples were stored at 4 °C for 12 to 24 h in 15 mL centrifuge tubes without anticoagulant to separate cell fractions and serum. Separated serum was then transferred to 1.7 mL microcentrifuge tubes and centrifuged at 12,000 rpm for 15 min at 4 °C. The supernatant was stored at − 80 °C until further analysis. Serum T and 11-KT levels were quantified using ELISA kits (11-KT #582,751 and T #582,701, Cayman Chemical, MI, U.S.) following the manufacturer's instructions. Optimal dilutions were determined to range from 1:2 to 1:9 for T and from 1:3 to 1:350 for 11-KT, indicating significant variation between immature and mature ages. Serum samples were analyzed in triplicate per male, yielding an intra-assay variation of 14.8% for T and 11.8% for 11-KT. Inter-assay variation was 11.5% for T and 9.8% for 11-KT. Assay sensitivity (80% B/B₀) for T was ~ 6 pg/mL, in the range of 3.9 to 500 pg/mL. Sensitivity (80% B/B0) for 11-KT was ~ 1.3 pg/mL, in the range of 0.78 to 100 pg/mL. Spectrophotometric readings were obtained using a Cytation 3 (Biotek, Winooski, VT, U.S.) instrument at 412 nm.

Osmolality and ion analyses

Frozen stored (-80 °C) serum samples were used to measure osmolality using a vapor pressure osmometer (Vapro 5520, Wescor Inc., Logan, UT, U.S.). The osmometer was calibrated every three samples. A serum sample (10 µL) was pipetted onto a pre-cut piece of filter paper, loaded in the osmometer and analyzed in duplicate. Between 5 to 15 serum samples were analyzed per age. Serum samples were shipped overnight on dry ice to the Waters Agricultural Laboratories, Inc. (Camila, GA, U.S.) for ion analyses, including Ca²⁺, Na⁺, K⁺ Mg²⁺, and Cl⁻. In brief, serum samples were digested with 6 mL of HCL, 4 mL of HNO₃, and then analyzed for Ca²⁺, Na⁺, K⁺, Mg²⁺ ions using iCAP™ PRO ICP-OES element analyzer (Thermo Fisher Scientific, Bremen, Germany). Dilutions ranged between 200 and 500, depending on sample volume. Cl⁻ was analyzed using Chloride QuanTab® Test Strips (Hach, Loveland, CO, U.S.) with dilutions ranging between 2–10, depending on the sample volume submitted. Between 5 to 14 serum samples were analyzed per age.

Testes histology

Histological analysis was conducted on fish from 2023, which included 7 age classes ranging from 2 to 9 years of age (except age 5). Preserved testes samples were processed at the Scott-Ritchey Research Center, Auburn University College of Veterinary Medicine, using a routine 12-h cycle program. In brief, tissue samples were dehydrated in ethanol solutions, gradually increasing from 70 to 100%, and then embedded in StatLab Parapro 360 paraffin (PP360, StatLab Medical Products, Inc., TX, U.S.). Each sample was step-sectioned at 4 µm with 32 µm between sections, and three sections were mounted on each slide. The slides were air-dried at room temperature and incubated at 60 °C for 10 min before being stained using hematoxylin and eosin. Digital images of tissue slides were captured using a Zeiss Imager.A2 microscope (Carl Zeiss Microscopy, LLC, White Plains, NY, U.S.) equipped with 40 × objective (Zeiss A- Plan 40 × / 0, 65 pH 2), Axio-cam 305 camera, and Zen Pro v.6.1 imaging software (Zeiss, Oberkochen, Germany). Between 4 to 6 testes samples were analyzed per age. For each sampled fish, three histological sections were placed on a slide, where 9 digital images were taken per slide. Digital images were analyzed by Image J (v. 1.53), with a 48-point grid placed on the image to estimate the area fraction of different cells within testicular tissue using ImageJ plugin “Gird” and “Cell counter”. The area fraction for each cell type was calculated by dividing the number of points classified within each cell type by the total number of grid points. Progression of spermatogenesis was then assessed using a SMI [23], where tissue was organized by cell type and area fractions per tissue category weighted by a factor ranging from 0 to 1, with testicular somatic cells (Ts) = 0, spermatogonia (Sg) = 0.25, spermatocytes (Sc) = 0.5, spermatids (St) = 0.75, and spermatozoa (Sz) = 1. An example of SMI estimation is provided in Supplementary Table 4.

Sperm collection and quality analyses

Sperm collection and counting

Blood vessels and peritoneum were manually removed from the testes with forceps to limit contamination. Testes were rinsed again with HBSS and macerated prior to straining through a 1 mm mesh into sterile 50 mL centrifuge tubes. Sperm density (cells/mL) was quantified in duplicate for each male using a Neubauer hemocytometer [40]. In brief, sperm were diluted 500-fold using HBSS, in duplicate, and homogenized for ~ 10 s. The prepared dilution (10 µL) was pipetted onto the hemocytometer and counted at 20 × mag. where sperm inside five squares (0.2 mm²; top right, top left, bottom right, bottom left, and center) were quantified. The individual cell counts were summed for two counting reps and averaged. This number was multiplied by 5 to estimate the cells in the 5 × 5 grid (1 mm²). The dilution factor was multiplied by the estimated sperm cells in the 5 × 5 grid and multiplied by 10,000 (to identify cells/mL). Sperm production per male was quantified by multiplying the sperm density by the total volume of sperm extracted from each male.

Computer-assisted sperm analysis (CASA)

Extracted sperm samples were stored at 4 °C for CASA within 4 h of collection. HBSS was added to standardize the sperm concentration to 1 × 10⁹ cells/mL for CASA analysis. CASA software (CEROS II software, Hamilton Thorne Biosciences, Beverly, MA, U.S.) was used to quantify sperm kinematic traits [40]. Samples and activation media were placed in 1.5 mL microcentrifuge tubes and kept at 4 °C in an Echotherm™ Chilling/Heating Dry Bath (Torrey Pines Scientific, Carlsbad, CA, U.S.) for the duration of the experiment. In brief, 0.1 µL of diluted sperm solution were pipetted in an 80 µm 2X-CEL chamber (Hamilton Thorne Biosciences, Beverly MA, U.S.) and covered with a coverslip, then activated with 12 μL DH₂O supplemented with 0.5% bovine serum albumin (Merck Millipore, Burlington, MA, U.S.) under a light microscope. Three technical replicate activations were carried out for each male. Sperm velocity (VCL; μm/s), average path velocity (VAP; μm/s), straight-line velocity (VSL; μm/s), and percent motility (%), were analyzed at 10 s and 20 s post-activation. All videos were manually checked for quality control. In brief, sperm tracks were excluded from analysis, if the software mistakenly split a single sperm track or incorrectly combined the crossing tracks of multiple sperm. Cell tracks were also removed if sperm were drifting across the field of vision rather than actively swimming.

Transcriptomic analyses and validation

RNA processing and extraction

Testes from immature (2 years of age, n = 4), sexually mature (4 years of age, n = 4), and physically mature (7 and 9 years of age, n = 8) males were randomly chosen for RNA extraction. Samples (~ 50 mg) preserved in 600 µL of DNA/RNA shield (Zymo Research, Irvine, CA, U.S.) were thoroughly homogenized using a Mini Bead Mill Homogenizer (VWR, Radnor, PA, U.S.). Bead-beating (2 × cycles) was run for 30 s followed by a 1 min cooldown. Total RNA was extracted using Quick-RNA Miniprep Plus Kit (Zymo Research) following the manufacturer’s instructions. Total RNA was treated with DNase I and resuspended in 50 µL of nuclease-free water. Concentration and purity (260/280) of RNA was measured using a Nanodrop One^C spectrophotometer (ThermoFisher Scientific, Waltham, MA, U.S.) and RNA integrity assessed using an Agilent TapeStation 4200 system (Agilent Technologies, Palo Alto, CA, U.S.).

Sequencing, RNA-seq QC, and genome alignments

Extracted RNA samples were sent to Azenta Life Sciences (South Plainfield, NJ, U.S.) for library preparation and Illumina sequencing. Ribosomal RNA (rRNA) was depleted from samples using QIAseq FastSelect—rRNA Fish Kit (Qiagen, Hilden, Germany). Sequencing libraries (n = 16) were prepared using a NEBNext Ultra II RNA Library Preparation Kit (NEB, Ipswich, MA, U.S.), following the manufacturer’s instructions. Sequencing was performed on the Illumina NovaSeq X Plus platform using a 2 × 150 bp paired-end (PE) configuration (Azenta Life Sciences, South Plainfield, NJ, U.S.). Raw, demultiplexed sequence files, at a minimum of 26 M PE reads/sample, were provided in FASTQ format. Raw reads obtained from sequencing were subjected to quality control and preprocessing. The quality of raw RNA sequencing reads was assessed by FastQC (V.0.11.9). Clean reads were obtained by removing low-quality reads and adapter sequences from the raw data using Trimmomatic (V.0.39). RNA-seq reads shorter than 36 bp in length were excluded from subsequent analyses. The genome of blue catfish (I. furcatus) was obtained from NCBI (Genome assembly Billie_1.0; RefSeq accession number: GCF_023375685.1), and remaining high-quality reads were aligned to the genome using STAR aligner (V.2.7.5). Post mapping quality control was performed using MultiQC (V.1.10.1). BAM files were then used to generate a gene-level count data matrix for all samples via HTSeq-count (V. 0.11.1).

Differential gene expression and gene ontology (GO) analyses

Differential gene expression analysis was performed with edgeR (V.4.4.2) in R-bioconductor to carry out pairwise comparisons among 2,4,7, and 9 age groups. The statistical significance of differentially expressed genes (DEGs) was calculated based on fold change (significance at |log₂FC|> 1) and P < 0.05 adjusted according to the Benjamini and Hochberg method [41]. To gain a better understanding of gene expression at the systems level and to identify the functional annotations of differentially expressed genes in testis tissues, Gene Set Enrichment Analysis (GSEA) and Fisher’s Exact Test (FET) were performed on DEG lists. Enrichment of the three major GO categories [Biological Process (BP), Molecular Function (MF), and Cellular Component (CC)] was determined. Data were analyzed using OmicsBox program (V.3.3), where significance was set at FDR < 0.05. The datasets generated for this study can be found in the NCBI Gene Expression Omnibus (GEO) repository and can be accessed under the accession number GSE290987.

RT-qPCR validation of DEGs

The extracted RNA samples (n = 16) were used for cDNA synthesis and RT-qPCR validation. Extracted testis RNA samples were diluted to 50 ng/μL before performing cDNA synthesis using an iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA, U.S.) according to the manufacturer’s instructions. Briefly, each 20 μL reaction contained 4 μL of 5 × iScript reaction mix, 1 uL of iScript reverse transcriptase, 5 μL of nuclease-free water, and 10 μL of template RNA. cDNA samples were synthesized in a PTC Tempo 96 Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, U.S.). Program conditions were set at 25 °C for 5 min, 46 °C for 20 min, and 95 °C for 1 min. After the reaction, the cDNA was stored at − 20 °C. Genes of interest (n = 6) were selected for RT-qPCR, including prok2, deptor, helz2a, bend7, ccnd2b, and numa1. The reference genes were actb and ef1a. Primer sequences of these genes were designed from GenBank accession numbers using NCBI primer designing tool with amplification size < 200 base pairs (Table 1). All primer sets (Eurofins Genomics LLC., Louisville, KY, U.S.) were evaluated using PCR followed by electrophoresis on 2% gel in advance to assess expected product size. All cDNA samples were diluted to 1.25 ng/μL and utilized for the RT-qPCR reaction using the SsoAdvanced Universal SYBR Green Super Mix (Bio-Rad, Hercules, CA, U.S.) on a CFX Opus 384 real-time PCR System (Bio-Rad, Singapore). The RT-qPCR reaction was conducted in a 10 µL mixture consisting of 5 μL of master mix, 0.5 μL each of forward and reverse primer (10 μM stock), and 4 μL of diluted sample cDNA. Each reaction was performed with two technical replicates, and cycling conditions were 2 min at 95 °C, followed by 15 s at 95 °C, 15 s at 60 °C, and 30 s at 72 °C for 40 cycles. Afterwards, a melt curve analysis was performed to verify the product specificity using the following protocol: 5 s holding starting at 65 °C followed by a 0.5 °C rise until it reached 95 °C. The relative expression of the target genes was calculated using the 2^−∆∆Ct method [42], normalizing with the geometric average of two reference genes (ef1a and actb).

Table 1.

Primer sequences for RT-qPCR validation in blue catfish (Ictalurus furcatus) testis

Gene	Forward primer	Reverse primer	Product size (bp)	Accession no
actb	CCGTGACCTGACTGAATACC	GCCCATCTCCTGCTCAAAG	139	DQ399027
ef1a	CTGCTGCGGAATAATCGCCAA	CCCATCTCAGCGGCTTCCTT	190	XM_053613177.1
ccnd2b	AGCCCTACATGAGGCGGGTA	TTTGGACGCCAGGAAGAGGC	179	XM_053650280.1
numa1	TCCCAGACAGAGTGGGGGAA	CACTCTACGATCGGCCCCAG	191	XM_053647639.1
bend7	TCTTCTTCCACACTTGGCAGGAT	CTGTCCGTGTCCATGTGCTC	200	XM_053622286.1
prok2	GCCATGAGGTCCAGCTACTCTC	CGCCACCACACTGCGAATCT	106	XM_053636366.1
deptor	CATCCAGGCTGTTGACCCCA	TGCTGCAGATTCCACCAAAAGC	199	XM_053626863.1
helz2a	GCCATGGCCACAGAACCAGA	TTTGTGGTCGCCCAAAAGCA	81	XM_053634912.1

Statistical analysis

All data were analyzed using SAS statistical analysis software (v.9.4), IBM SPSS (v.30), and R software (v.2024.09.1—Build 394). Alpha was set at 0.05. Graphs were generated using R software and SigmaPlot (v.15).

Comparison of reproductive traits

Two statistical approaches were used to analyze age-related male reproductive performance indices. In the first approach, a series of one-way ANOVA models was used to compare the effects of the nine sampling ages on means for each trait. Residuals were tested for normality (Shapiro–Wilk test) and homogeneity of variance (plot of residuals vs. predicted values). Where necessary, data were log₁₀ transformed or arcsine square-root transformed to meet the assumptions of normality and homoscedasticity. Post-hoc analyses were composed using Tukey’s HSD multiple comparisons procedure. Values are reported as least square means ± standard error. In the second approach, age-related changes in absolute body measurements, testis weight, GSI, absolute sperm production, and hormones were examined by fitting a quadratic regression model to the data. Residuals were tested for normality (Shapiro–Wilk test) and homogeneity of variance. Where necessary, data were log₁₀ transformed to meet the assumptions of normality and homoscedasticity. Relative head circumference and relative hump circumference were estimated from the residuals of the regression between these variables and log₁₀ body weight. Relative sperm production was estimated from the residuals of the regression between sperm production and log₁₀ testis weight. Relative head circumference and relative hump circumference were examined by fitting a linear regression model and relative sperm production was examined by fitting a quadratic regression model to the data.

Comparison of sperm kinematics

Principal components analysis (PCA) was used to summarize variation in the three sperm velocity metrics (VAP, VSL, and VCL) and sperm motility at 10 s and 20 s post-activation separately. Sperm from age two fish was excluded from the analysis since the group had only two males with sperm. One informative PC axis was extracted that explained 84.1% of the variation at 10 s and 99.2% of the variation at 20 s. PC1 scores were then used to analyze the significance of fish ageing on sperm kinematics using one-way ANOVA models.

Results

Body morphometric measurements

Males of each age were sampled to measure body weight and a series of morphometric traits. Results from one-way ANOVA showed that body weight (F_8,94 = 129.54, P < 0.001; Fig. 1C), and several morphometric traits increased with age, including total length (F_8,94 = 109.62, P < 0.001; Fig. 1A), absolute hump circumference (F_8,94 = 98.97, P < 0.001; Fig. 2A), and absolute head circumference (F_8,94 = 147.12, P < 0.001; Fig. 2E). One-way ANOVA also showed that age impacted relative head circumference (F_8,94 = 2.46, P = 0.018; Fig. 2G); however, post-hoc comparisons only showed nearly significant differences between ages 6 and 10 (P = 0.051). Relative hump circumference was not significant between ages (F_8,94 = 0.77, P = 0.628; Fig. 2C). For fish aged 2 to 10 years, mean (± SEM) body weight ranged from 0.67 ± 0.48 kg at 2 years age to 10.07 ± 0.62 kg at 10 years age, with no significant increase in weight after 7 years (P > 0.05). Furthermore, total length ranged from 40.37 ± 1.75 cm at 2 years to 92.89 ± 2.25 cm at 10 years, showing a similar pattern as body weight, with no significant increase after age 7 (P > 0.05).

Fig. 1.

Effect of age on the changes in total length and body weight in male blue catfish (Ictalurus furcatus). One-way ANOVA models were run to compare total length (A) and body weight (C) across ages. Least square means ± standard errors are represented. Bars with same letters were not significantly different (P < 0.05). Quadratic regressions were generated between age and total length (B) and body weight (D). Colored dots indicate 2023 (red) and 2024 (blue) sampling years

Fig. 2.

Effect of age on changes in body morphometrics and sperm production in male blue catfish (Ictalurus furcatus). One-way ANOVA models were run to compare absolute hump circumference (A), relative hump circumference (C), absolute head circumference (E), relative head circumference (G), absolute sperm production (I) and relative sperm production (K) across ages. Least square means ± standard errors are represented. Bars with same letters were not significantly different (P < 0.05). Quadratic regressions were generated between age and absolute hump circumference (B), absolute head circumference (F), absolute sperm production (J) and relative sperm production (L). Linear regressions were generated between age and relative hump circumference (D) and relative head circumference (H). Colored dots indicate 2023 (red) and 2024 (blue) sampling years

Quadratic regression models were then used to create relationships between body morphometric traits and age. Several body morphometric traits and body weight exhibited quadratic relationships with age including total length (R² = 0.83, P < 0.001, y = 15.949 + 15.657x − 0.842x²; Fig. 1B), body weight (R² = 0.87, P < 0.001, y =  − 0.262 + 0.297x − 0.017x²; Fig. 1D), absolute hump circumference (R² = 0.83, P < 0.001, y = 4.402 + 10.552x − 0.582x²; Fig. 2B) and absolute head circumference (R² = 0.88, P < 0.001, y = 1.737 + 9.790x − 0.499x²; Fig. 2F). Relative head circumference and relative hump circumference data were also used in linear regression models, exhibiting a positive relationship with age in relative head circumference (R² = 0.05, P = 0.017, y =  − 0.013 + 0.002x; Fig. 2H) and there was no significant relationship in relative hump circumference (P > 0.05).

Reproductive performance indices

Mean testis weight increased as age increased (F_8,91 = 18.81, P < 0.001; Fig. 3A), where testis weight ranged from 0.21 ± 1.01 g at 2 years to 11.36 ± 1.21 g at 10 years. Lower testis weights were reported at ages 2 (0.21 ± 1.01 g) and 3 (1.54 ± 1.01 g), while the highest was reported at age 6 (14.35 ± 1.63 g). Thereafter, testis weight remained relatively high and stabilized, with some variability between ages 7 and 10. Changes in GSI were also quantified and were significant across ages (F_8,91 = 15.31, P < 0.001; Fig. 3C), with the lowest mean GSI observed at age 2 (0.026 ± 0.02%), and the highest GSI at age 6 (0.25 ± 0.03%). There was a notable increase in GSI between ages 3 and 4 (P = 0.010). After age 6, a decreasing trend was observed, but the differences between ages were not significant. Quadratic regression analysis demonstrated a significant relationship between testis weight and age (R² = 0.72, P < 0.001, y =  − 0.623 + 0.407x − 0.025x²; Fig. 3B) and between GSI and age (R² = 0.20, P < 0.001, y =  − 0.032 + 0.027x − 0.002x²; Fig. 3D).

Fig. 3.

Effect of age on the changes in reproductive indices in male blue catfish (Ictalurus furcatus). One-way ANOVA models were run to compare testis weight (A) and gonadosomatic index (GSI) (C) across the ages. Least squares means ± standard errors are represented. Scatter plots with the same letters were not significantly different (P < 0.05). Quadratic regressions were generated between age and testis weight (B) and gonadosomatic index (GSI) (D). Colored dots indicate 2023 (red) and 2024 (blue) sampling years

Absolute sperm production was significantly different across ages (F_7,74 = 5.38, P < 0.001; Fig. I). Between ages 2 to 5, absolute sperm production was relatively low, with the lowest sperm production observed at age 2 (0.80 ± 6.27 × 10⁹ cells). The highest average absolute sperm production was reported at age 6 (22.39 ± 3.96 × 10⁹ cells), and gradual decline in absolute sperm production was observed after age 6, which reached the lowest observed value at age 10 (11.21 ± 2.97 × 10⁹ cells), but the differences among later ages (6–10 years) were not significant. In addition, relative sperm production was not significantly different between ages (F_7,74 = 1.66, P = 0.132; Fig. K). Quadratic regression analysis demonstrated a significant relationship between absolute sperm production and age (R² = 0.37, P < 0.001, y = 7.246 + 0.777x − 0.051x²; Fig. J). Relative sperm production data were also used in quadratic regression models, and there was a significant relationship between age and relative sperm production with a relatively low R² value (R² = 0.08, P = 0.033, y =  − 0.572 + 0.219x − 0.018x²; Fig. L).

Blood hormones, ions, and osmolality

Blood serum androgens, T and 11-KT measured in male blue catfish ranged from 31.69 to 1,247.47 pg/mL and 29.20 to 5,323.50 pg/mL, respectively (Fig. 4). Both T and 11-KT differed in blood serum levels across the ages (F_8,94 = 9.29, P < 0.001; Fig. 4A, F_8,94 = 10.88, P < 0.001 ; Fig. 4B, respectively). The lowest mean value for T was recorded at age 2 (76.77 ± 62.58 pg/mL) while the highest value was observed at age 6 (754.50 ± 108.39 pg/mL). A similar trend was also observed for 11-KT, showing the lowest mean value at age 2 (152.64 ± 329.39 pg/mL), and the highest mean value at age 6 (2,648.01 ± 570.53 pg/mL). There was a significant difference in both T and 11-KT levels between ages 3 and 4, which showed a marked increase in circulating androgen levels when sexual maturity was reached. Once the males achieved sexual maturity, higher variations in serum androgen levels were observed. Mean androgen levels were not significantly different between the older ages, particularly after age 6, suggesting no heightened T and 11-KT levels once the fish become sexually mature. Additionally, 11-KT was the most abundant androgen in fish serum across all ages. Quadratic regression models were then used to create relationships between blood androgen levels and age. There was a quadratic relationship between T concentration and age (R² = 0.33, P < 0.001, y = 1.034 + 0.480x − 0.036x²;Fig. 4B) and between 11-KT concentration and age (R² = 0.36, P < 0.001, y = 0.882 + 0.682x − 0.050x²; Fig 4D).

Fig. 4.

Effect of age on the changes in blood T and 11-KT levels in male blue catfish (Ictalurus furcatus). One-way ANOVA models were run to compare T (A) and 11-KT (C) across the ages. Least squares means ± standard errors are represented. Scatter plots with the same letters were not significantly different (P < 0.05). Quadratic regressions were generated between age and T (B) and 11-KT (D). Colored dots indicate 2023 (red) and 2024 (blue) sampling years

Blood osmolality and a series of ions, including Na⁺, K⁺, Ca²⁺, Mg²⁺, and Cl⁻ were quantified across various ages (Table 2). The dominant ions of blood serum constituents were Na⁺ and Cl⁻. Significant differences were observed across ages for osmolality (F_8,94 = 2.39, P = 0.021), K⁺ (F_8,90 = 2.49, P = 0.017), Mg²⁺ (F_8,90 = 2.28, P = 0.028), and Cl⁻ (F_8,79 = 2.82, P = 0.008), whereas there were no significant differences in Na⁺ and Ca⁺ levels. Notably, blood plasma Cl⁻ ion level was significantly lower at age 2, and both K⁺ and Mg²⁺ ion levels were significantly lower at age 10.

Table 2.

Effect of age on osmolality, K⁺, Na⁺, Ca²⁺, Mg²⁺ and Cl⁻ levels in blue catfish males (Ictalurus furcatus). Data were analyzed using a series of one-way ANOVA models. Within columns, values followed by different letters are significantly different based on Tukey pairwise comparisons. Results represent means ± SEM (P < 0.05)

Age (years) and P-value	Osmolality	K+	Na+	Ca2+	Mg2+	Cl−
	(mOsm/kg)	(ppm)	(ppm)	(ppm)	(ppm)	(ppm)
2	261.1 ± 4.2a	378.1 ± 44.8b	3210.1 ± 65.8a	983.3 ± 239.4a	204.5 ± 25.2b	2803.8 ± 196.3a
3	257.3 ± 4.4a	347.4 ± 41.4ab	3295.5 ± 61.0a	1250.9 ± 221.6a	166.0 ± 23.3ab	3267.6 ± 148.4ab
4	267.4 ± 4.5a	338.4 ± 43.0ab	3264.2 ± 63.3a	711.3 ± 230.0a	153.4 ± 24.2ab	3404.9 ± 160.3ab
5	268.8 ± 5.8a	272.5 ± 54.8ab	3344.5 ± 80.6a	702.5 ± 293.2a	131.8 ± 30.9ab	2943.8 ± 196.3ab
6	279.8 ± 7.3a	349.0 ± 69.3ab	3498.2 ± 102.0a	878.2 ± 370.9a	210.4 ± 39.0b	3840.0 ± 277.7ab
7	268.3 ± 4.7a	282.8 ± 44.8ab	3441.5 ± 65.8a	1139.6 ± 239.4a	140.0 ± 25.2ab	3606.8 ± 167.4b
8	280.9 ± 4.5b	362.3 ± 44.8ab	3469.5 ± 65.8a	738.3 ± 239.4a	148.4 ± 25.2ab	3500.0 ± 167.4ab
9	261.1 ± 4.4a	472.3 ± 41.4b	3296.4 ± 61.0a	768.4 ± 221.6a	150.3 ± 23.3ab	3459.6 ± 167.3ab
10	263.5 ± 5.5a	210.6 ± 51.7a	3372.8 ± 76.0a	550.8 ± 276.4a	82.0 ± 29.1a	2957.8 ± 185.1ab
P-value	0.021	0.017	0.055	0.156	0.028	0.008

Histology and SMI calculations

Histology stains and digital images were analyzed for 33 males (in triplicate for each male) across ages 2, 3, 4, 6, 7, 8, and 9 (Fig. 5). This enabled the precise identification of histological changes of individuals at immature, developing, and mature stages. SMI values were quantified to analyze the progression of the spermatogenic process and were varied across the ages (F_6,26 = 8.16, P < 0.001; Fig. 6A). Males sampled at age 2 had the lowest mean SMI (± SEM) values (i.e., males possessing lower mature cells) (0.27 ± 0.06) while the highest SMI value was observed at age 9 (0.75 ± 0.67). SMI steadily increased up to age 6, and a gradual plateau was observed with no significant differences between older ages. The changes in testis cell type ratios were also quantified across different ages (Fig. 6B). Spermatozoa (Sz) and spermatids (St) were equally predominant after age 4 and remained stable in later ages. Fully mature sperm cells comprised < 10% of the cells analyzed at ages 2 and 3, with the majority being somatic cells and spermatogonia.

Fig. 5.

Digital images (scale bar = 50 μ m) of histological samples of testis for male blue catfish (I. furcatus) over time. Panel shows histology images of age 2 (A), age 3 (B), age 4 (C), age 6 (D), age 7 (E) and age 9 (F). Arrows indicating cell development include testicular somatic cells (Ts), spermatogonia (Sg), spermatocytes (Sc), spermatids (St), and spermatozoa (Sz)

Fig. 6.

Effect of age on spermatogenic maturity index (SMI; A) and cell type ratio (B) for male blue catfish (Ictalurus furcatus). Cell type is represented graphically, where somatic cells = S, spermatogonia = Sg, spermatocytes = Sc, spermatids = St, and spermatozoa = Sz. One-way ANOVA models were run to compare SMI between ages. Least square means ± standard errors are represented. Scatter plot with same letters were not significantly different (P < 0.05)

PCA analysis on sperm kinematics traits

The values of three sperm velocity metrics (VAP, VSL, and VCL) and sperm motility were used in PCA analysis at 10 s and 20 s separately (Fig. 7). The first PC axis (PC1) explained 84.1% of the variation at 10 s in sperm kinematics. When sperm kinematics at 20 s were considered, PC1 explained 99.2% of the variation. The results from ANOVA analysis on PC1 scores revealed no significant difference between ages at 10 s (P = 0.385) and 20 s (P = 0.831).

Fig. 7.

Principal components analysis (PCA) was performed to summarize variation in the three sperm velocity metrics (VAP, VSL and VCL) and sperm motility at 10 s (A) and 20 s (B) for male blue catfish (Ictalurus furcatus). The PCA plot captures the variance in the dataset in terms of the principal components and displays the most significant of these on the axes. The PC1 axis explained 84.1% of the variation at 10 s and 99.2% of the variation at 20 s. Colored dots represent the respective age groups

Testis transcriptomic profile

To screen for the genes related to testis ageing, a total of sixteen libraries were constructed and sequenced using the Illumina NovaSeq platform. As a result, a total of 465,038,946 raw reads were generated, yielding an average of 29,064,934 reads per sample. Reads shorter than 36 bases, reads with adaptor contamination, and reads with low-quality regions were removed, leaving a total of 384,362,845 cleaned reads (82.7%) for all samples. Of these, a total of 345,308,532 cleaned reads were mapped to the blue catfish genome (Billie_1.0) for an average mapping efficiency of 89.9% (Table 3).

Table 3.

Summary of the Illumina sequencing and mapping of testis transcriptomes in blue catfish (Ictalurus furcatus)

Male no	Raw reads mRNA	Cleaned reads mRNA	Mapping efficiency
M1	26,779,710	22,670,821	90.4%
M3	26,637,481	21,809,332	90.3%
M6	28,681,806	23,977,105	90.6%
M8	26,592,707	22,076,418	89.4%
M17	27,292,676	23,424,607	89.2%
M18	27,834,738	23,436,878	88.2%
M19	29,151,303	24,430,380	89.9%
M20	30,487,601	25,448,930	89.3%
M28	29,069,375	23,564,834	90.1%
M29	31,548,881	26,048,297	90.0%
M30	29,018,129	23,610,912	88.7%
M31	31,456,407	25,028,153	89.8%
M39	31,169,663	24,668,066	89.6%
M40	31,622,817	26,396,928	89.9%
M41	30,567,941	25,360,865	90.7%
M42	27,127,711	22,410,319	91.5%
Total reads	465,038,946	384,362,845
Total average (± SD)	29,064,934.13 (± 1,818,911.95)	24,022,677.81 (± 1,354,506.56)	89.9%

Quantification of differentially expressed genes (DEGs)

Testis samples from 16 blue catfish males across four age groups were used to build and sequence mRNA libraries. DEG analysis was conducted to perform pairwise comparisons among the 2,4,7, and 9 age groups (FDR < 0.05; |log₂FC|> 1). A higher absolute value of log₂FC denotes a greater magnitude of gene expression changes and a lower FDR value indicates a higher statistical confidence in the observed differences in comparisons. Each group was analyzed against one another to identify significant DEGs among all age groups. Across all comparisons, DEGs were only found between age 2 vs. age 4, age 2 vs. age 7, age 2 vs. age 9, and age 4 vs. age 7. In total, 5220 significant DEGs were identified from all comparisons. Among those, 2870, 4085, and 2792 DEGs were significantly expressed at ages 4, 7, and 9, respectively, in the fish testis compared to age 2 (Supplementary Table 1). The highest number of significant DEGs were identified at age 7. Next, we compared up-regulated and down-regulated DEGs separately from each age group to age 2 (Fig. 8). Briefly, 1795, 2261, and 1658 genes were significantly down-regulated at age 4, 7, and 9, respectively. In contrast, 1075, 1824, and 1134 genes were significantly up-regulated at age 4, 7, and 9, respectively. Additionally, only 2 genes were up-regulated at age 7 when compared to age 4.

Fig. 8.

Differential expressions of DEGs in blue catfish (Ictalurus furcatus) testis. Separate volcano plots showed DEGs in age 4 (A), age 7 (B), and age 9 (C) compared to age 2. Colored dots indicate significantly up-regulated (orange red) and down-regulated (navy) genes (FDR < 0.05; |log₂FC|> 1), which are also indicated by number

A three-way Venn diagram analysis was performed to identify common and unique DEGs among the age groups (Fig. 9). Notably, the largest number of DEGs was identified at age 7, with 775 and 478 unique up-regulated and down-regulated genes, respectively. Additionally, the analysis showed that 521 genes were significantly up-regulated, and 1055 genes were significantly down-regulated across all age groups, regardless of age. Several genes associated with testis transcriptome, which exhibit higher fold change values and are responsible for reproductive-related functions, were identified and most of them were up-regulated at age 7 and low expression at age 9 (Table 4). Up-regulated genes included transcripts with log₂FC > 3 at age 7, which indicated an expression level 8 times higher than at age 2. These genes are potentially involved in spermatid development, flagellar development, gonad development, spermatogenesis, and cell energy metabolism, such as adgb, dnajb13, insl3, mak, spata18, and spef1. In contrast, gonad differentiation-related genes were significantly down-regulated after sexual maturation, such as gsdf and fstl5. Notably, several protein-coding genes, such as dtx3lb.2, loxl4, bhlhe23, foxp3b, and pax1a, were exclusively down-regulated at age 9, among the top down-regulated genes.

Fig. 9.

The three-way Venn diagram illustrates common and unique expression of genes for male blue catfish (Ictalurus furcatus) at ages 4, 7 and 9 compared to age 2. The findings indicate that 521 and 1055 genes were consistently up- or down-regulated, respectively, regardless of age (FDR < 0.05; |log₂FC|> 1)

Table 4.

Testis-associated genes for male blue catfish (Ictalurus furcatus) as differentially expressed (FDR adjusted P-value < 0.05) in at least one pairwise comparison

Gene	Accession number	Description	log2 (fold changes)
			Age 4	Age 7	Age 9
gsdf	XM_053648099.1	Gonadal somatic cell derived factor	− 3.95	− 4.02	− 1.97
fstl5	XM_053631786.1	Follistatin-like 5	− 2.60	− 3.17	− 3.57
adgb	XM_053647490.1	Androglobin, transcript variant X5	–	6.33	–
agpat2	XM_053648551.1	1-acylglycerol-3-phosphate O-acyltransferase 2 (lysophosphatidic acid acyltransferase, beta)	1.14	1.08	–
aqp10b	XM_053638041.1	Aquaporin 10b	1.88	2.42	2.93
dnajb13	XM_053644918.1	DnaJ heat shock protein family (Hsp40) member B13, transcript variant X2	2.27	3.26	2.12
dnali1	XM_053632962.1	Dynein, axonemal, light intermediate chain 1	–	1.42	–
eno3	XM_053643383.1	Enolase 3, (beta, muscle)	–	1.35	–
exosc1	XM_053615398.1	Exosome component 1, transcript variant X2	2.57	4.01	–
gapdhs	XM_053610993.1	Glyceraldehyde-3-phosphate dehydrogenase, spermatogenic, transcript variant X2	–	–	− 1.60
ghrhrb	XM_053636257.1	Growth hormone releasing hormone receptor b	–	1.80	–
hsd17b1	XM_053640790.1	Hydroxysteroid (17-beta) dehydrogenase 1	3.77	4.29	4.83
insl3	XM_053651006.1	Insulin-like 3 (Leydig cell)	3.16	4.55	4.22
lrp3	XM_053616514.1	Low density lipoprotein receptor-related protein 3	–	− 2.43	− 2.03
mak	XM_053611412.1	Male germ cell-associated kinase, transcript variant X3	2.17	3.04	2.37
rnf17	XM_053627884.1	Ring finger protein 17	− 1.98	− 2.32	− 2.37
spag8	XM_053618722.1	Sperm associated antigen 8, transcript variant X1	–	2.70	–
spata17	XM_053633252.1	Spermatogenesis associated 17, transcript variant X4	–	2.11	–
spata18	XM_053614709.1	Spermatogenesis associated 18, transcript variant X6	3.39	4.34	3.50
spef1	XM_053621576.1	Sperm flagellar 1, transcript variant X6	2.74	3.39	–
tcte1	XM_053614871.1	t-complex-associated-testis-expressed 1	1.36	1.51	–
tegt	XM_053653042.1	Testis enhanced gene transcript (BAX inhibitor 1)	–	1.02	–
tekt4	XM_053615540.1	Tektin 4, transcript variant X1	–	2.47	–
trh	XR_008385916.1	Thyrotropin-releasing hormone, transcript variant X3	3.40	3.69	–
wdr54	XM_053610400.1	WD repeat domain 54, transcript variant X3	–	2.43	–

Values indicate the log₂ (fold changes) was greater than or less than the threshold for up- (log₂FC > 1) or down-regulation (log₂FC <  − 1), respectively. A dash (" – ") indicates a feature that was not identified as a DEG for a given comparison(s)

Gene ontology (GO) analysis

GO analysis allows the classification of genes and further describes their role in biological processes (BP), molecular functions (MF), and cellular components (CC). In this study, GO analysis was performed on the significant genes underlying each of these age group comparisons previously listed. Genes were mapped to GO terms, and then Fisher’s exact test (FET) and gene set enrichment analysis (GSEA) were used to determine the significance of GO terms between age groups. Although 2 significant genes were found between age 4 vs. age 7 comparison, no significant GO terms were identified in FET or GSEA analyses. Among the most specific GO terms identified as enriched (GSEA) at age 4 (106 terms), age 7 (117 terms), and age 9 (39 terms), 140 terms were unique, and only 8 terms were common between all comparisons regardless of the age (Supplementary Table 2). The highest modulation of GO terms from GSEA was reported at age 7, and the least modulation of terms at age 9.

Using a normalized enrichment score (NES) based approach to GSEA analysis, highly significant up-regulated and down-regulated GO terms related to BP, MF, and CC were collectively plotted (Fig. 10). The significantly enriched terms were grouped into 64, 15, and 27 GO terms categorized as BP, MF, and CC, respectively at age 4. There were significant terms up-regulated at age 4 related to BP, such as translational initiation (GO:0006413) and spermatid development (GO:0007286). In addition, GO terms related to CP, such as ciliary plasm (GO:0097014) and ciliary basal body (GO:0036064) were significantly up-regulated at age 4. Translation initiation factor activity (GO:0003743) was the only MF term up-regulated at age 4. Considering the middle age group, there were 73, 26, and 18 enriched GO terms categorized as BP, MF, and CC, respectively at age 7. Significantly modulated GO terms were associated with a higher number of genes at age 7, compared to age 4 and age 9. There were unique significant terms up-regulated in the age 7 group related to biological processes, such as non-motile cilium assembly (GO:1905515), axoneme assembly (GO:0035082), cellularization (GO:0007349), and cilium movement involved in cell motility (GO:0060294). In addition, CC terms of the axoneme (GO:0005930) and ciliary basal body (GO:0036064), and those related to MF terms, such as translation factor activity and RNA binding (GO:0008135) were significantly up-regulated, showing their unique modulation at age 7. Considering the later age group, there were only 25, 8, and 6 enriched GO terms categorized as BP, MF, and CC, respectively at age 9. Interestingly, a higher number of GO terms were down-regulated within the enriched GO terms at age 9. There were significant terms down-regulated at age 9 related to biological processes, such as cellular component maintenance (GO:0043954), response to calcium ion (GO:0051592), and inorganic cation imports across the plasma membrane (GO:0098659). There were no CC terms up-regulated at age 9, and all the up-regulated BP terms were related to immune responses, such as immune response-activating cell surface receptor signaling pathway (GO:0002429), defense response to virus (GO:0051607), and defense response to bacterium (GO:0042742). In addition, only two MF terms of single-stranded RNA binding (GO:0003727) and pentosyl transferase activity (GO:0016763) were significantly up-regulated.

Fig. 10.

Bubble plot graphs illustrate gene ontology (GO) terms enriched using GSEA analysis for male blue catfish (Ictalurus furcatus) at ages 4, 7, 9. Larger circles have more genes associated with that GO term, and adjusted P-value is denoted by color. All listed GO terms were statistically significant (P < 0.05)

Overrepresented and underrepresented GO terms were detected via FET among gene transcripts differentially expressed at ages 4 and 9. A total of 120, 116, and 108 terms were represented (FET) at ages 4, 7, and 9, with 170 unique terms and 26 common terms regardless of age (Supplementary Table 3). Most significant GO terms from FET were plotted (Fig. 11). Thirteen of the 106 terms identified as enriched in age 4 via GSEA were also identified through FET at age 4. Similarly, 21 of the 117 terms identified for age 7 and 7 of the 39 identified for age 9 via GSEA were also identified in the same comparisons through FET. There were significantly up-regulated terms identified at age 4, such as ammonium ion metabolic process (GO:0097164), hormone binding (GO:0042562), and mesodermal cell differentiation GO:0048333). Among the significant terms related to sperm motility, the sperm flagellum (GO:0036126) GO term was only up-regulated at age 7, under CC terms. Similar to GSEA analysis, immune system-related GO terms were up-regulated at age 9, such as immune system process (GO:0002376) and chemotaxis (GO:0006935).

Fig. 11.

Bar graphs illustrate gene ontology (GO) terms enriched using Fisher’s Exact Test for male blue catfish (Ictalurus furcatus) age 4 (A), age 7 (B), age 9 (C). Bars represent the adjusted P-values and gene count is denoted by color. All listed GO terms are statistically significant (P < 0.05)

Validation of RNA-Seq results by RT-qPCR

To validate the RNA-Seq data, six DEGs were randomly selected and subjected to RT-qPCR analysis with two house-keeping genes, actb and ef1a, as references (Fig. 12). The six DEGs included three up-regulated genes (prok2, deptor, and helz2a), one down-regulated gene (bend7) and two non-differentially expressed genes (ccnd2b and numa1). Melt curve analysis for the genes obtained from RT-qPCR showed a single peak, indicating the presence of one product, and expression levels of the genes were consistent with the RNA-seq results, which confirmed that the transcriptomic data were reliable.

Fig. 12.

RT-qPCR validation of the expression levels for significantly up-regulated, down-regulated and non-significant genes across ages for male blue catfish (Ictalurus furcatus). Bar graphs show the RNA-seq RPKM values (left) and RT-qPCR relative quantification values (right) for significantly up-regulated; deptor (A), helz2a (B), prok2 (C), down-regulated; bend7 (D) and non-significant; ccnd2b (E), numa1 (F) genes. Means ± standard errors are represented. Bars with same letters were not significantly different (P < 0.05)

Discussion

Optimizing gamete collection would enhance production efficiency and improve profitability for commercial farms. Since a high degree of variation among blue catfish individuals was reported in our previous studies [10, 40, 43], it is essential to establish links between paternal age and reproductive success, as it is apparent that advanced paternal age leads to changes in sperm performance. In this study, the reproductive biology of blue catfish males was characterized, focusing on age-related changes in growth, blood hormones and ions, testicular histology, and sperm kinematics, which provides a more comprehensive picture of the species' reproductive biology and physiology. The testis transcriptome was also profiled to investigate the molecular mechanisms underlying fish ageing. This study revealed that paternal age had an impact on reproductive performance with several key findings: (1) for fish aged 2 to 10 years, those over 7 years had the largest size, with rapid increase in size until age 7 and a slight increase afterward; (2) absolute sperm production increased to its peak at age 6, then showed a decreasing trend; (3) T and 11-KT levels did not increase after age 6; (4) sperm kinematics traits were not impacted by age at 10 s and 20 s post-activation; and (5) testis-associated significant genes were highly expressed at age 7 and least at age 9.

Morphometric characteristics analyzed included total length, head circumference, and hump size in addition to total weight. Their combination effectively describes the overall size and shape of the fish and serves as indicator of fitness [44]. Some of these characteristics develop as secondary sexual traits with sexual maturity, which enhances the reproductive capability in male fish for gaining access to gravid females in natural environments [45, 46]. On the other hand, understanding the size of sexual maturity is vital for optimal broodstock management in hatchery settings [47]. Our analysis showed that for fish aged 2 to 10 years, those over 7 years had the largest size, with a rapid increase in size until age 7 and a little increase afterwards. Morphometric data suggest prioritization of somatic growth at early ages (2–4 years), then shifting the energy from growth to reproductive investment with testis development. Notably, previous studies have reported that larger males of poeciliid species produce a greater number of sperm, which serves as an important predictor of paternity [48]. Also, increasing the size of hump serves as a potential indicator of sexual maturity, which has also shown a positive relationship with circulating 11-KT levels in hooknose Chinook salmon (Oncorhynchus tshawytscha) [49]. Furthermore, Dolly Varden charr (Salvelinus malma) males with larger hump size had sperm with higher velocities, providing evidence of the relationship between hump size and sperm kinematics [50]. Although we found significantly higher absolute hump sizes at age 7, followed by a gradual plateau, there were no changes in relative hump size across ages, which we measured by standardizing to body weight. This result might be due to higher variation in individual body sizes and hump sizes in older ages than younger ages. Furthermore, unpublished results from our lab have reported, for the first time in blue catfish, a positive relationship between paternal broodstock morphometrics and the weight and survival of their offspring.

In addition to body measurement data, a series of gonadal metrics were used to describe testicular development and growth. Notably, we found that the advancement of spermatogenesis was well described by testis weight and GSI across ages. Previous studies showed that testis weight reflected the growth of gonads, serving as the preliminary physical indicator of sperm production, and accurately predicting the spawning capability of individuals [51, 52]. Furthermore, the GSI value reflects the relative measure of testis weight to body weight, further explaining the reproductive investment of individuals [53]. Testis weight and GSI in our study were influenced by age, with markedly lower weights in early ages and significantly higher values observed at later ages. However, quadratic regression analysis showed a decreasing trend in testis weight and GSI at older ages, highlighting no further increase in reproductive investment. This further describes the prioritization of somatic maintenance over effort in reproduction as fish get older. Additionally, previous studies have reported strain differences in testis weights of 5-year-old blue catfish males, highlighting significant differences between strains within the same age group [54]. This suggests a considerable proportion of testis development relies on genetic variation within the same species, even at the same age. Research further reported that males from all strains had higher GSI in May than in June, showing significant changes even within a month [54]. Our study was conducted in May to ensure the fish of all ages were at their maximum reproductive performance, following previous findings, thereby minimizing the effect of sampling time within the reproductive season.

Several studies have reported age-related changes in sperm production and concentration [55, 56], while sperm quality traits such as duration of sperm motility and percent motility remained unchanged in fish species [32, 55]. Our study reported similar findings, showing no changes in sperm kinematics across ages, but there was an age-dependent change in absolute sperm production. However, an age-dependent decline in sperm quality and function was reported in Turquoise killifish (Nothobranchius furzeri) [57]. One possible explanation is that a species like the Turquoise killifish, which has a shorter lifespan, has accumulated more deleterious genetic variants compared to longer-lived species [57, 58]. However, most reproductive studies, even in longer-lived fish species, are for a short time, which may not capture long-term changes, or a possible reduction in sperm production due to functional changes in gonads at older ages [59]. Our study reported that absolute sperm production at age 6 was nearly 9 times that of age 3. Similar results have also shown that mature males produced greater quantities of sperm [60, 61]. In our study, quadratic regression analysis showed a decreasing trend in absolute sperm production as fish aged. Similar results were also reported in cichlid fish, with older males having larger testes but fewer sperm than younger ones [59]. This may be due to the deposition of connective tissues in the gonads at older ages [62].

It is well known that spermatogenesis is regulated by androgen hormones, primarily T and 11-KT at lower levels of the hormonal cascade [13]. A significant increase in androgen levels occurs during puberty, associated with the onset of spermatogenesis, which is a common phenomenon that was observed in many studies [12, 51]. This increase is closely linked to the higher proliferation of Leydig cells at the onset of puberty, which are responsible for secreting androgens [63]. Additionally, seasonal changes in serum androgen levels have been documented in many species [64–66]; however, less effort has been made to understand their roles in spermatogenesis and testis development across ages. Our analysis revealed that circulating T and 11-KT levels were lower in ages 2 and 3, and then showed a marked increase with sexual maturation, particularly at age 4. These androgens followed similar trends, gradually increasing and reaching the highest value at age 6. Then the androgen levels had less variability across older ages (age 6 to 10) with a decreasing trend, but the levels were not significantly different. Thus, concentrations of T and 11-KT were influenced by the age of fish at younger ages, up until full sexual maturity, after which age had less impact on circulating serum levels. Similarly, a previous study on blue catfish males documented no significant difference in serum T levels between three sexually mature stages, defined by body characteristics [67]. Our results showed that 11-KT was the most abundant androgen in fish serum across all ages, highlighting its significant role in sperm production. Higher 11-KT levels observed in this study may be attributed to the sampling time, as this study was conducted during peak spawning. Generally, similar findings regarding higher 11-KT abundance compared to T values have been reported in many studies [49, 68, 69]. However, it is possible to observe the changes in the abundance of 11-KT and T levels across the months of the year rather than ages, as 11-KT is an oxygenated metabolite of T [15]. Ongoing research aims to evaluate seasonal variation in these serum androgens, to provide a detailed understanding of androgen effects on reproductive performance in blue catfish males across seasons.

Significant changes in blood ions and osmolality were reported across spawning seasons and ages in previous studies [70–72]. However, most studies related to fish reproduction have focused on ion and osmolality changes in seminal plasma or ovarian fluid rather than blood levels. Previous comparisons between blood and seminal plasma reported that K⁺ is the major cation which showed significant differences, while the other ions are similar or at rarely lower concentrations in seminal plasma [73, 74], which helps provide insight into seminal osmolality and ionic composition. Osmolality is the key to activating sperm, where freshwater species release their sperm in an environment with lower osmolality than body fluids (hypo-osmotic conditions) [75]. This is followed by a readjustment of internal ionic concentrations by the sperm membrane processes. Eventually, the internal ionic concentrations reach levels at which dynein-ATPase activity is optimal for acquiring sperm motility [76]. Blood osmolality progressively increases during the early developmental stages of fish [77]. These changes in blood osmolality reflect the underlying physiological changes in osmoregulation [77]. In the current study, osmolality was significantly higher at age 8, while there were no significant differences in osmolality among other ages. This result might be due to higher individual variation within the group during fish ageing; however, the osmolality was below 300 mOsm/kg, which was commonly reported in other catfish species [73, 78]. In our study K⁺, Mg²⁺, and Cl⁻ levels in blood were influenced by age; however, the concentrations of Na⁺ and Ca²⁺ did not show any significant changes across ages. Similar results were observed in a previous study on Chinese sturgeon (Acipenser sinensis), with no significant changes in Na⁺ and Ca²⁺ levels across different age classes [72]. Recent studies found that the K⁺ concentration affects the sperm quality traits in Common carp (Cyprinus carpio) [79], Rainbow trout [75] and European burbot (Lota lota) [80]. Sperm acquired motility due to a lower K⁺ content, which further induces the rise of intracellular Ca²⁺ concentration during sperm activation [81]. Ca²⁺ is critical for increased synthesis of cAMP, which initiates flagellar motion [81]. Our results also indicated a lower K⁺ in mature ages than immature fish at age 2, indicating a higher potential of sperm motility after reaching maturity. Na⁺ and Cl⁻ were the most abundant ions found in our study, consistent with other findings [78, 82]. In blue catfish, the lower levels of Cl⁻ at age 2 and higher levels at age 7 indicated clear differences dependent on maturity stage. Several studies have found that a large number of chloride cells, specialized ionocytes groups that actively transport Cl⁻, begin to appear on the gills during fish ageing [77]. Possible reasons may include reproductive hormonal changes with maturity that enhance metabolic demands, as Cl⁻ is an important contributing ion for osmoregulation in freshwater fish [83].

Several cell maturation stages are associated with spermatogenesis, and histological changes occur at different developmental stages, once activated by hormones [13]. The onset of puberty was characterized by the first appearance of spermatozoa in the lumen of fish testis [25]. Spermatogenic maturity index (SMI) provides detailed information about the maturity stage of fish, considering the proportional cell types within the testis at specific time points [23]. Our data showed that SMI values were significantly lower at ages 2 and 3, then markedly increased at age 4. During early ages, testis tissues consisted mainly of connective tissue with somatic cells and spermatogonia in the germinal epithelium. However, one fish at age 2 and two fish at age 3 produced spermatozoa, indicating early puberty within the age classes. Additionally, SMI values remained stable during the older ages without showing significant differences among ages. After germ cell maturation, spermatozoa were the major cell types found within the lumen as shown by previous studies [21]. This result aligned well with hormonal changes, showing a clear trend between circulating hormones and germ cell maturation [25]. In the current study no males showed a regression period after reaching maturation, as we conducted the research during the peak spawning season [54].

Other than physiological parameters, advances in bioinformatic tools have made RNA-seq technology an emerging field in modern science, where transcriptome profiling is used to better understand gonadal development in animal studies [84]. Using differential gene expression analysis, 5220 significant DEGs were obtained by comparing all ages against one another. Across all comparisons, DEGs were only found at ages 4, 7, and 9, when compared to age 2. Furthermore, only two significant genes were found at age 7 when compared to age 4, and no significant genes found at age 9 when compared to either ages 4 or 7. This reflects significant changes occurring at sexual maturation while testicular identity and function remain stable or less impacted by ageing in blue catfish males, preventing a complete loss of testicular function with ageing. Our findings are consistent with a more recent study in short-lived African turquoise killifish, Nothobranchius furzeri, which reported that testes were less impacted by ageing than ovaries [85]. Another study involving C57BL/6 strain mice, Mus musculus had similar findings, with testes showing only limited age-related gene expression changes; however, ovaries showed extensive gene expression changes with age [86].

The analysis of common DEGs showed that 521 genes were significantly up-regulated, and 1055 genes were significantly down-regulated. This indicated that nearly 30% of significant DEGs were consistently regulated across all mature groups, regardless of age, when compared to immature fish. Although no significant differences in expression were observed between the mature age groups (4, 7, and 9), our findings showed that nearly 70% of significant genes were differentially expressed when immature fish were compared to mature fish collectively at ages 4, 7, and 9. This is further supported by the highest number of significant genes being modulated at age 7, with 775 and 478 unique up-regulated and down-regulated genes, respectively. Similar results were also observed in previous studies, with the highest number of significant genes that were regulated in middle-aged testis [85]. Therefore, it is suggested that greater changes related to reproductive performance in males occur during middle years of life compared to early (age 4) or later mature ages (age 9). This is supported by our results on reproductive traits, specifically GSI, T, and 11-KT, which showed peak values in middle-aged fish, particularly at age 6. However, considering the annual cyclic events related to the spawning season, significant gene expression changes occur with a higher number of DEGs, according to their developmental stages of germ cell renewal within the same age [87]. This suggests that the impact of the spawning season is also considerable when assessing the reproductive performance in mature males. This finding is further confirmed by our SMI values, which indicated a higher proportion of spermatozoa during the peak spawning season, regardless of age after sexual maturation. In contrast, there was no evidence of identifying spermatozoa in the lumen or sperm ducts during early spermatogenesis and the regression phase, due to the seasonal effects, as reported in previous studies [87, 88].

Candidate genes were identified based on the literature related to spermatid development, gonadal development, sperm quality, secretion and hormone production among the DEGs across ages. Interestingly, some testis-associated genes were exclusively and significantly expressed at age 7, aligning with the physiological observations discussed above. The gene adgb was up-regulated with a log₂FC > 6 in the middle-aged testis compared to the immature testis, which indicated an expression level > 64 times higher at age 7. Interestingly, two transcript variants of adgb were found in our dataset, and no significant expression of one variant at age 4 and age 9. Notably, this gene was found in both vertebrates and invertebrates, confirming its primordial appearance, producing a protein having around 1500 amino acids [89]. Previous studies provided evidence regarding the higher expression of adgb in fertile males compared to infertile males, suggesting its role in reproduction, sperm quality, and fertility [90]. Another study which examined sperm from adgb knockout animals reported reduced sperm quality, with flagellar malformations and shortened flagella [81]. Testis-associated genes such as dnajb13, spata18, trh, exosc1, spef1, mak, insl3, and hsd17b1 were up-regulated with log₂FC > 3, indicating an expression level 8 times higher than in immature fish, also with a clear higher expression than age 4 fish. Most of these genes showed higher expression at age 7 compared to age 9. This would suggest evidence for a trend of a non-linear pattern of candidate genes along the life span related to the reproductive performance of blue catfish males. dnajb13 is expressed in large amounts in testis, which involves sperm tail function, and this gene is member of type II heat shock protein 40 family [91]. The expression of dnajb1 gene was previously confirmed as an axonemal component of mouse mature spermatozoa, indicating the significance of DNAJB13 protein in sperm flagellum development and motility of mature spermatozoa [91, 92]. We report that dnajb13 was highly expressed in all mature ages compared to immature testis and with clear higher fold change value in age 7 fish, highlighting the efficiency of middle-aged males in sperm maturation.

Spata17 and spata18 are members of the spermatogenesis-associated (SPATA) family. This is an important regulatory family which includes numerous genes that contribute to the spermatogenesis process [93]. spata17 was exclusively up-regulated at age 7 in blue catfish and showed an age-dependent relationship. Overexpression of the spata17 revealed its potential in increased cell apoptosis during spermatogenesis [94]. Cell apoptosis helps in normal spermatogenesis for controlling germ cell numbers and eliminating abnormal and damaged cells, thereby contributing to improve sperm quality and the production of functional sperm [95, 96]. Notably, the high apoptosis activity at age 7 provides evidence of a higher sperm production rate at this age. The increased apoptosis by up-regulated spata17 gene at age 7 might help in removing damaged cells caused by the higher production rate. The spata18 gene contributes to mitochondrial quality control, leading to mitophagy after DNA damage, which is essential for maintaining mitochondrial functions and energy production in cells [97]. Previous studies reported that spata18 was a testis-biased gene [98], up-regulated with spermatid development in mouse and expression remained high in adults [99]. We observed the same expression pattern, with higher expression in all adult ages and a slight increase at age 7, which highlighted its increased role for middle-aged blue catfish. We also report a significant up-regulation of the spef1 gene at ages 4 and 7, with clearly higher expression at age 7 than at age 4. The spef1 gene is involved in the cytoskeleton of developing spermatids and may play a role in the maturation of spermatozoa, which is necessary for flagellar functions [100, 101]. Sperm membranes are composed of various fatty acids that have been linked to sperm quality and reproductive success [102]. Several genes are involved in transferring fatty acids to phospholipids, which are associated with cell membrane function in various body tissues. AGPAT1 is an enzyme involved in phospholipid biosynthesis and is responsible for converting lysophosphatidic acid (LPA) to phosphatidic acid (PA), which is a key step in lipid metabolism [103]. A recent study reported reduced sperm quality with impaired sperm development in the agpat1 gene knockout mouse model, highlighting its role in male reproduction [104]. Interestingly, our results also indicated significantly higher expression of the agpat2 gene at ages 4 and 7, with no expression at age 9, which provided evidence of reduced sperm quality at older ages.

A higher number of specific functional GO terms related to reproductive performance were enriched at age 7 in the testis transcriptome, which further confirmed the involvement of a large number of DEGs in numerous functions at middle age, with the least enrichment at age 9. Interestingly, it highlights that functional enrichment is more pronounced in the testis at a fully mature middle age than the pubertal age. Also, due to the absence of enriched GO terms between the age 7 and age 9 comparison, it is difficult to conclude that substantial changes occurred in later ages compared to middle ages. Obviously, this may be due to the majority of testis related genes showing only slight changes that may not be significantly different and also represents the nature of ageing in males which is hard to detect experimentally as described in previous studies [105]. However, considering pairwise comparisons with the immature age group, it is important to note that spermiogenesis-related functions were observed at each age. Spermatid development (GO:0007286) is one of the important functions associated with sperm maturation [106], which was also up-regulated at ages 4 and 7 in our data set. Interestingly, both GSEA and Fisher’s exact test analysis revealed no up-regulation of the spermatid development GO term at age 9. The axoneme plays a key role in generating sperm motility and providing an inner core structure of cilia and flagella [107]. Genes associated with the organization of the axoneme were up-regulated with the enrichment of axoneme assembly (GO:0035082) exclusively at age 7. In addition, in the cellular component, motile cilium (GO:0031514), ciliary basal body (GO:0036064), and axoneme (GO:0005930) were among the top GO terms, exclusively up-regulated at age 7. Several protein coding genes are responsible for motile cilium (GO:0031514) enrichment, including tekt4 [108], which was also up-regulated in our study. The coordination of sperm flagellar motion relies on tekt4, and its deficiency can cause asthenozoospermia [108, 109]. Together, these results provide clear evidence of acquiring motility, which is significantly pronounced in the middle-aged testis. In contrast, immune processes were increased in the testis of ageing blue catfish males. BP GO terms associated with viral (GO:0051607) and bacterial (GO:0042742) defense responses were up-regulated at age 9. Naturally, the presence of infectious agents induces the immune system to inflammatory response [110]. However, blood-testis and blood-epididymis barrier provides immune privilege, which limits the body immune response to foreign agents, protecting the testis from damage [111]. However, this balance may be interrupted due to ageing as reported previously in mice (Mus musculus) [111]. Thus, this leads to an increase in the immune response that could affect the expression of immune related genes. Similar results were observed in zebrafish testes, where the immune cell population in old testis atlas was more diverse and proportionally larger than the germ cell population [62]. Interestingly, most of the GO terms related to spermatogenesis were not up-regulated at age 9, highlighting the reduced reproductive efficiency at older ages.

Together, these findings demonstrate age-related changes in reproductive performance throughout the life span of blue catfish males. Physiological data revealed better performance at age 6, while transcriptomics data indicated superior performance at age 7. Although we could not perform RNA sequencing for age 6 due to the initial experimental design, the combined data underscores the peak performance at middle ages. Based on the results of this study, we suggest farmers manage blue catfish populations to have consistent supplies of 6 and 7 years old blue males available in living gene banks (ponds) or cryo-banks for hybrid production. In addition, culling older fish, which passed peak performance, and showed a trend towards poorer sperm quantity and quality in this study, could improve the efficiency of hatchery operations and would benefit the overall economics of hybrid catfish production.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

11-KT

11-Ketotestosterone

T

Testosterone

DEGs

Differentially expressed genes

GnRH

Gonadotropin-releasing hormone

SMI

Spermatogenic Maturity Index

CASA

Computer assisted sperm analysis

IACUC

Institutional Animal Care and Use Committee

USDA-ARS

United States Department of Agriculture, Agricultural Research Service

HBSS

Hank’s balanced salt solution

GSI

Gonadosomatic index

VCL

Curvilinear velocity

VAP

Average path velocity

VSL

Straight-line velocity

rRNA

Ribosomal RNA

GSEA

Gene set enrichment analysis

FET

Fisher’s exact test

GO

Gene ontology

BP

Biological process

MF

Molecular function

CC

Cellular component

FDR

False discovery rate

NES

Normalized enrichment score

RT-qPCR

Reverse transcription quantitative polymerase chain reaction

PCA

Principle components analysis

Author contributions

Initial idea: IAEB. Data collection: IAEB, SSNL, BGB, KAM, KRW, AEN, TJB. Data analyses: IAEB, SSNL, JWA, NMS, MKL. Writing: IAEB, SSNL. Editing: BGB, KAM, KRW, AEL, JWA, NMS, BHB, TJB, MKL, RAD, LAR. Funding: IAEB, BHB, RAD, XW, LAR. All coauthors approved the final manuscript.

Funding

This project was supported by Agriculture and Food Research Initiative Competitive Grant no. 2020-67015-31874 and 2023-67016-39455 from the USDA National Institute of Food and Agriculture. Funding was also provided by the USDA ARS and USDA National Institute of Food and Agriculture, Hatch project 1013854 (IAEB).

Data availability

The datasets generated for this study can be found in the NCBI Gene Expression Omnibus (GEO) repository and can be accessed under the accession number GSE290987.

Declarations

Ethics approval and consent to participate

All experimental animal protocols were approved by the Auburn University Institutional Animal Care and Use Committee (AU-IACUC) with protocol #2023-5229 and Warmwater Aquaculture Research Unit with IACUC protocol #FY23-001. All methods used in this experiment were performed in accordance with relevant guidelines and regulations. Consent to participate: Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.