Genetic Diversity of Four Consecutive Selective Breeding Generations in Channel Catfish, Ictalurus punctatus

Abstract

To elucidate the temporal dynamics of genetic diversity across successive breeding generations of channel catfish (Ictalurus punctatus) and enhance subsequent breeding efficiency, we systematically evaluated the genetic variation in four consecutive generations using ten highly polymorphic microsatellite loci. The number of alleles (Na), effective alleles (Ne), and Shannon’s index (I) all declined with increasing generations. The mean expected heterozygosity (He) decreased gradually from 0.822 to 0.805 but remained above 0.80, indicating that all generations maintained relatively high genetic diversity. Allele frequency analysis revealed the progressive fixation of alleles potentially linked to target traits, while some rare alleles were gradually lost. Analysis of molecular variance (AMOVA) demonstrated that 98% of the genetic variation occurred within generations, with weak differentiation among generations (Fst = 0.016). UPGMA clustering further indicated that later generations diverged from the base stock, whereas genetic distances among adjacent generations progressively narrowed, suggesting increasing convergence and stabilization of genetic structure. These findings provide both theoretical insights and practical guidance for the continuous selective breeding and germplasm conservation of channel catfish.

1. Introduction

Artificial selective breeding is recognized as an effective strategy to improve production performance, develop elite strains, and enhance aquaculture efficiency [1]. At present, genetic improvement in aquaculture species largely relies on family selection, mass selection, and hybridization [2]. Nevertheless, long-term artificial selection often results in reduced genetic diversity, elevated inbreeding risk, and diminished adaptability and resilience [3,4]. Preserving genetic diversity in breeding populations is therefore essential for the sustainable utilization of germplasm resources and stable improvement of high-performance strains. Continuous monitoring and scientific evaluation of genetic variation during the breeding process are thus imperative. Studies in multiple fish species have documented progressive declines in effective allele numbers, heterozygosity, and overall genetic diversity during successive selective breeding generations [5,6,7,8].

Microsatellite DNA (SSR), a highly polymorphic and codominant genetic marker, has been widely applied in aquaculture for analyses of population structure [9,10], genetic diversity assessment [11,12], and parentage identification [13,14]. Due to their sensitivity and operational simplicity, microsatellites are an effective tool for monitoring genetic variation in selectively bred populations. In recent years, SSR markers have played a vital role in genetic improvement studies of shrimp [15], mollusks [16], and fish [17,18], providing essential support for genetic management in aquaculture breeding programs.

Channel catfish (Ictalurus punctatus), also known as American catfish or channel catfish, is native to North America and represents the most technologically advanced and highest-yielding freshwater aquaculture species in the United States. Due to its strong environmental adaptability, superior flesh quality, and ease of processing, it has become an important global freshwater aquaculture species [19]. Since its introduction into China in the 1980s, technical barriers in breeding, reproduction, feed formulation, processing, and export have been successively overcome, establishing a complete industrial chain. Production increased from several thousand tons initially to approximately 500,000 tons annually [20]. China has thus become the world’s leading producer and consumer of channel catfish, with the industry emerging as a distinctive and rapidly expanding sector in national freshwater aquaculture.

However, due to the long-term absence of systematic selective breeding in China, combined with germplasm restrictions imposed by the United States, domesticated channel catfish have exhibited signs of germplasm degeneration, including reduced growth rates, uneven body size, color variation, loss of body spots, and shortened morphology [21]. These problems have severely impaired product quality, yield, and profitability. Moreover, the incidence of disease has been rising, with outbreaks of epidemic infections such as hemorrhagic septicemia causing substantial economic losses for farmers [22]. To address germplasm deterioration, the Ministry of Agriculture established the National Genetic Breeding Center for Channel Catfish at the Freshwater Fisheries Research Institute of Jiangsu Province. After more than a decade of research and development, significant progress has been made in genetic improvement. The newly bred strain “Jiangfeng No. 1” exhibits accelerated growth and enhanced disease resistance, and has been successfully promoted nationwide, generating considerable social and economic benefits [23].

Against this background, we employed ten highly polymorphic microsatellite loci to systematically evaluate genetic diversity and stock structure across four consecutive generations of channel catfish. By comparing allelic composition, heterozygosity, genetic distance, and molecular variance among generations, we aimed to elucidate patterns of genetic variation under artificial selection. This study provides a theoretical foundation for genetic improvement and strain development in channel catfish, while offering insights for sustainable breeding and germplasm conservation in aquaculture species.

2. Materials and Methods

2.1. Sample Collection

All channel catfish samples used in this study were obtained from the Lukou and Yangzhong experimental bases of the Jiangsu Freshwater Fisheries Research Institute. Between 2008 and 2009, a base breeding stock established with 405 individuals from these two culture facilities. The original stock was maintained by interbreeding with introduced stocks from Texas (1997), Arkansas (1999, 2003, 2004), and Mississippi (2001). Successive family-based breeding and seedling propagation were carried out from this base stock, generating four consecutive generations: G₀ (2008, 2009), G₁ (2011–2014), G₂ (2015–2018), and G₃ (2019–2022). The numbers of male and female broodstock used to establish the four generations are shown in Table S1. All channel catfish were reared in separate ponds for each generation, and samples were collected at 540 dah (days after hatching). The average body weights of catfish from the four generations were 947.28 ± 216.54 g, 966.72 ± 243.88 g, 1023.15 ± 277.36 g, and 1009.47 ± 235.92 g, respectively. Approximately 100 individuals were randomly sampled from each generation, and the detailed sample numbers for each generation are listed in Table 1.

Table 1. Genetic diversity of four channel catfish breeding generations.

Generation	Sample Number (n)	Observed Number of Alleles (Na)	Effective Number of Alleles (Ne)	Shannon Diversity Index (I)	Observed Heterozygosity (Ho)	Expected Heterozygosity (He)	Unbiased Expected Heterozygosity (uHe)
G0	103	11.10	6.448	1.987	0.828	0.822	0.826
G1	97	11.20	6.291	1.988	0.813	0.826	0.831
G2	100	10.90	6.155	1.957	0.853	0.822	0.826
G3	103	9.200	5.599	1.878	0.817	0.805	0.809
Mean	100.75	10.60	6.123	1.952	0.828	0.819	0.823

2.2. Genomic DNA Extraction

Tail fin tissue was clipped, preserved in 95% ethanol, and stored at –20 °C. Genomic DNA was extracted using the FastPure Tissue DNA Isolation Mini Kit (Vazyme, Nanjing, China) following the manufacturer’s protocol. DNA quality and concentration were assessed using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and working solutions were diluted to 100 ng/μL.

2.3. Selection of Polymorphic Microsatellite Markers and PCR Optimization

Ten highly polymorphic microsatellite loci isolated from the channel catfish [13] was selected for analysis (Table S2). Forward primers were synthesized with a 5′ FAM fluorescent label. Optimized PCR reactions were performed according to our previous report [13]. Condition of electrophoresis and allele call followed the previous study [13] and listed in the Supplemental File.

2.4. Data Analysis

Genotypic data were analyzed using GenAlEx 6.51 to calculate allele frequency, observed number of alleles (Na), effective number of alleles (Ne), observed heterozygosity (Ho), expected heterozygosity (He), unbiased expected heterozygosity (uHe), Shannon’s diversity index (I), Nei’s genetic identity, and Nei’s genetic distance. Analysis of molecular variance (AMOVA) was performed to partition total variance within and among stocks and estimate fixation indices (Fst). To evaluate genetic differentiation among consecutive breeding generations, pairwise Fst values were calculated according to Nei’s definition across ten microsatellite loci. The significance of each pairwise Fst was assessed using permutation tests (2000 random label permutations) to generate empirical p values under the null hypothesis of no differentiation. Multiple comparisons were adjusted using the Bonferroni correction. UPGMA clustering was conducted using PHYLIP 3.69 based on Nei’s standard genetic distance matrices, with bootstrap support assessed through 1000 replicates. The resulting phylogenetic tree was visualized with MEGA7 software.

3. Results

3.1. Genetic Diversity Analysis

Amplification with ten pairs of microsatellite primers across four breeding generations yielded 119 alleles, with 8–17 alleles per locus (Table S3). The total number of alleles per generation showed a decreasing trend: 111 (G₀), 112 (G₁), 109 (G₂), and 92 (G₃). Both the mean number of alleles (Na) and effective number of alleles (Ne) declined across generations, ranging from 9.2–11.2 and 5.599–6.448, respectively. The mean expected heterozygosity (He) ranged from 0.805 to 0.826, highest in G₁ and lowest in G₃. Unbiased expected heterozygosity (uHe) exhibited the same trend, ranging from 0.809 to 0.831. Mean observed heterozygosity (Ho) ranged from 0.813 to 0.853, with G₂ showing the highest and G₁ the lowest values. Shannon’s diversity index (I) ranged from 1.878 to 1.988, mirroring the order of He.

He is a key metric for assessing population polymorphism. Values between 0.5 and 0.8 indicate high genetic diversity, while values above 0.8 reflect very high genetic diversity. In this study, although He decreased gradually with successive generations, all four generations maintained values above 0.8, suggesting that genetic diversity remained at a very high level despite multiple rounds of selection.

3.2. Allelic Frequency Changes Among Generations

To examine changes in stock genetic structure during breeding, allele frequencies were calculated for all loci across the four generations (Figure 1; Table S3). The dynamics of allele frequencies could be broadly categorized into four patterns: (1) progressive increase, as in IP1-115, IP4-398, IP11-226, IP12-357, IP14-182, IP16-313, IP19-231, and IP20-293, which were initially rare but gradually became dominant; (2) progressive decrease, as in IP1-142, IP13-125, IP14-176, IP18-149, and IP19-228, which were common in G0 but declined sharply or disappeared in later generations; (3) relative stability across generations, such as IP4-392, IP11-241, IP13-113, IP16-301, and IP20-309; and (4) random fluctuations across generations, such as IP1-118, IP4-380, IP12-342, IP14-185, and IP20-305. In addition, many low-frequency alleles, such as IP14-188, IP14-206, IP4-386, IP13-133, IP19-228, and IP20-329, were gradually lost during the breeding process.

Figure 1. Gene frequency changes at 10 microsatellite loci across four selected generations, with generations on the vertical axis and allele frequencies on the horizontal axis.

3.3. Breeding Generations Structure Analysis

AMOVA of the four generations (Table 2) revealed highly significant differences among stocks (p = 0.001). However, the vast majority of genetic variation (98%) was attributable to within-generation differences, while only 2% was among generations. This indicates that within-generation variation is the primary source of genetic diversity in channel catfish. Pairwise Fst comparisons among the four generations showed very low overall differentiation (Fst = 0.002–0.009). Permutation-based post hoc tests revealed that all comparisons involving the base stock (G₀) and the later generations were statistically significant after Bonferroni adjustment (adjusted p < 0.01), whereas the difference between G₁ and G₂ was not significant (p = 0.76) (Table S4). These findings confirm that genetic divergence from the base population occurred gradually across generations, although the magnitude of differentiation remained minimal. The weak but significant differentiation likely reflects progressive allele frequency shifts associated with continuous selection and mild genetic drift, while the negligible difference between G₁ and G₂ suggests effective gene flow and stable breeding management. These results are consistent with the AMOVA outcome, indicating that most genetic variation (≈98%) resides within generations and that the overall population structure remains well-conserved during selective breeding.

Table 2. Analysis of molecular variance (AMOVA) for channel catfish breeding generations.

Source of Variation	Degrees of Freedom (df)	Sum of Squares (SSD)	Mean Squared Deviation (MSE)	Est. var	Percentage of Total Variance (P/%)
Among generations	3	66.698	22.233	0.138	2%	Fst = 0.008
Within generations	399	3319.617	8.320	8.320	98%	Nm = 29.9
Total	402	3386.315		8.458	100%	p = 0.001

3.4. Genetic Differentiation Among Generations

To further evaluate genetic differentiation, Nei’s genetic identity and genetic distance were calculated among the four generations (Table 3). Values of genetic identity ranged from 0.920 to 0.980, while genetic distances ranged from 0.012 to 0.061. The largest distance was between G₀ and G₃, which also showed the lowest genetic identity, whereas the smallest distance was between G₂ and G₃, which showed the highest identity. These results suggest more frequent gene flow between G₁ and G₂ compared with other generation pairs.

Table 3. Nei’s genetic identity (below diagonal) and Nei’s genetic distance (above diagonal) between the channel catfish breeding generations.

Generation	G0	G1	G2	G3
G0	-	0.038	0.045	0.061
G1	0.949	-	0.023	0.029
G2	0.940	0.935	-	0.012
G3	0.920	0.921	0.980	-

The UPGMA tree based on Nei’s genetic distances (Figure 2) showed that G₀ formed a separate branch, while G₃ first clustered with G₂, then grouped with G₁, and finally clustered with G₀. With successive breeding, later generations diverged progressively from the base stock, as indicated by increasing genetic distances and decreasing genetic similarity to G₀. Meanwhile, genetic distances among adjacent generations gradually narrowed, decreasing from 0.038 to 0.023 and finally to 0.012.

Figure 2. UPGMA clustering tree based on Nei’s genetic distance.

4. Discussion

Analysis of ten highly polymorphic microsatellite loci over generations of the channel catfish revealed a downward trend in the number of alleles, effective alleles, and Shannon’s diversity index with successive generations. Although the overall magnitude of decline was moderate, these findings nevertheless reflect a process of gradual genetic homogenization. Similar phenomena have been reported in other domesticated and selectively bred fish species [8,24,25], where limited stock size and sustained selection pressures often result in allele loss and reduced polymorphism.

It is noteworthy that despite these declines, the expected heterozygosity (He) in all generations remained above 0.80, indicating consistently high genetic diversity. An He greater than 0.8 is generally considered to reflect an extremely high level of genetic variation [26]. This outcome is closely related to the construction of the base stock, which was established using germplasm from multiple U.S. stocks, thereby providing a broad initial genetic background and ample variation for subsequent selection [27]. Such a foundation has been critical for minimizing inbreeding and maintaining stock vitality.

Allele frequency dynamics further demonstrated that certain alleles increased in frequency and became dominant, suggesting possible linkage with target traits and progressive fixation under selection. Conversely, some rare alleles were lost, reflecting the combined influence of genetic drift and selection, which together contract the gene pool. While this stabilization promotes the development of uniform strains with desirable traits, it also underscores the need to guard against excessive erosion of diversity, which could compromise adaptability and resilience [3,4].

The AMOVA results showed that the vast majority of genetic variation (98%) occurred within generations, with weak differentiation among generations (Fst = 0.016). This indicates that considerable genetic exchange and consistency have been maintained across generations, echoing findings from shrimp breeding studies [28], where effective management helped to control divergence and mitigate inbreeding depression. Moreover, the UPGMA clustering revealed that later generations gradually diverged from the base stock while genetic distances among adjacent generations decreased, pointing to a process of convergence and stabilization of stock structure. Such patterns are consistent with evolutionary dynamics in selectively bred stocks, where genetic variation tends to aggregate around traits under strong selection [29].

In addition to the genetic evaluation of domesticated channel catfish populations in China, it is important to consider breeding progress in other major producing regions. In the United States, where channel catfish breeding programs have been established for decades, several improved strains have been successfully developed, such as “Delta Select” and “NWAC103”, which exhibit faster growth, better feed conversion, and higher disease resistance [30,31]. These programs have been supported by systematic family selection and genomic-assisted breeding strategies, providing valuable reference models for genetic improvement. In contrast, channel catfish aquaculture in Southeast and South Asia is still largely based on unselected or imported stocks. Systematic breeding programs for channel catfish have not yet been initiated, and production mainly relies on commercial seed supply without genetic evaluation or improvement. Therefore, the continuous evaluation of genetic diversity, as conducted in this study, is essential to ensure sustainable selective breeding and prevent the erosion of germplasm resources worldwide.

5. Conclusions

Our study demonstrates that although the genetic diversity in channel catfish declined moderately across four generations of artificial selection, it remained at a high level overall, and the stock structure became progressively stable. These changes favor the fixation of desirable traits and the establishment of improved strains. Nonetheless, future breeding programs must strengthen genetic management. Strategies such as introducing external germplasm, establishing core breeding stocks, and applying molecular marker-assisted selection (MAS) or genomic selection (GS) can improve breeding efficiency while minimizing diversity loss, thereby ensuring the long-term sustainability of channel catfish improvement programs.