Population Genomics of Megalobrama Provides Insights into Evolutionary History and Dietary Adaptation

Abstract

Megalobrama, a genus of cyprinid fish, is an economically important freshwater fish widely distributed in major waters of China. Here, we report the genome resequencing of 180 Megalobrama fish including M. amblycephala, M. skolkovii, M. hoffmanni, and M. pellegrini. Population structure indicated that geographically divergent Megalobrama populations were separated into six subgroups. A phylogenetic tree showed that M. skolkovii was more closely related to M. pellegrini than other species and M. hoffmanni was clustered apart from other Megalobrama species, showing a high nucleotide diversity in geographic groups. Treemix validated gene flow from M. amblycephala to M. skolkovii, suggesting that introgression may provide an important source of genetic variation in the M. skolkovii populations. According to the demographic history analysis, it is speculated that Megalobrama might have been originally distributed in the Pearl River with some spread to Hainan Island and northern China due to lower sea levels during the glacial period. Whole-genome selective sweeps analysis demonstrated that M. amblycephala likely developed an enhanced energy metabolism mostly through fatty acid degradation pathways whereas M. hoffmanni possibly regulate lipid absorption via the cholesterol metabolism pathway. Taken together, this study provides a valuable genomic resource for future genetic investigations aiming to improve genome-assisted breeding of Megalobrama species.

Keywords: Megalobrama; demographic history; feeding habits; genome resequencing; population structure.

Figure 1

Geographic map indicating the distribution and feeding habits of the Megalobrama species in this study. Circles reflect the geographic regions where the samples were collected from. The blue lines represent the rivers.

Figure 2

Phylogenetic analysis of different Megalobrama geographical populations. (A) Population genetic structure of 180 Megalobrama fish. The length of each colored segment represents the proportion of the individual genome inferred from ancestral populations (K = 2–8). The species name is at the bottom. (B) Cross-validation (CV) error for varying values of K in the admixture analysis. The minimum of estimated CV error on K = 6 suggests the most suitable number of ancestral populations. (C) Maximum-likelihood-based phylogenetic tree constructed using no admixed individuals. The scale bar represents pairwise distances between different individuals. Different colors represent different Megalobrama species. (D) Principal component analysis (PCA) of Megalobrama populations. Eigenvector 1, 2, 3, and 4 explained 45.06%, 18.03%, 4.00%, and 1.73% of the total variance, respectively. MH refers to M. hoffmanni, MA to M. amblycephala, MS to M. skolkovii, MP to M. pellegrini.

Figure 3

Population genetic analysis of Megalobrama species. (A) Gene flow analysis of Megalobrama species. Arrows indicate migration events that occur between populations. The heat map indicates migration weight. (B) Nucleotide polymorphism (π), and differentiation index (Fst) of the four Megalobrama species. The largest circle represents the large π value, and the longer line segment represents the large Fst value. MH refers to M. hoffmanni, MA to M. amblycephala, MS to M. skolkovii, MP to M. pellegrini. (C) Decay of linkage disequilibrium (LD) patterns for the four Megalobrama species inferred by the phylogenetic trees.

Figure 4

Demographic history of Megalobrama species. (A) Relative cross coalescence rates (CCR) between Megalobrama populations. When the two populations are completely mixed, the CCR is close to one. When they are completely split, the CCR is close to zero. The dotted line indicates that the CCR is 0.5. MP, MS, MA, and MH refer to M. pellegrini, M. skolkovii, M. amblycephala, and M. hoffmanni. g (generation time) = 2.5 years; μ (neutral mutation rate per generation) = 0.14 × 10⁻⁸. (B) PSMC model estimates changes in the effective population size over time, representing variation in inferred Ne dynamics. The undulating broken line in the figure is the estimated effective population size of each population in the evolutionary history. The time axis is not divided into deciles. μ = 0.1 × 10⁻⁸. MH-HN, MH-GD, MP, MA, MS-FY, and MS-other refer to the Hainan population of M. hoffmanni, Pearl River populations of M. hoffmanni, M. pellegrini, M. amblycephala population, Fuyuan population of M. skolkovii, and other M. skolkovii populations. (C) Timeline of Quaternary glaciation. Mya = million years ago.

Figure 5

Genome-wide inference of selection sweeps on chromosomes during the diet adaptation of M. amblycephala (A) and M. hoffmanni (B). MA and MH refer to M. amblycephala and M. hoffmanni. Gbe1, Akt2, and Aldh3a2 were identified from the comparison between M. amblycephala and M. skolkovii, Acss3, and Acsbg2 from the comparison between M. amblycephala and M. pellegrini, Hadhb from the comparison between M. amblycephala and M. hoffmanni, and Cyp46a1, Tas1r1, and Baat from the comparison between M. hoffmanni and M. amblycephala. The black curve indicates the nucleotide polymorphism ratio (PiR) analysis, and red curve indicates extended haplotype homozygosity between populations (XP-EHH) analysis. The green and pink boxes represent the position of the selected genes on the chromosomes.

Figure 6

Metabolism pathways and expression pattern of candidate genes. (A) Candidate genes of M. amblycephala and M. hoffmanni enriched in the metabolism pathways. The lines and selected genes in M. amblycephala and M. hoffmanni are indicated in green and pink, respectively. The dashed lines used to connect KEGG pathways represent indirect relationships. (B–F) The expression pattern of candidate genes (Acss3, Aldh3a2, Gbe1, Hadhb, and Cyp46a1) in the liver tissue of M. amblycephala and M. hoffmanni. (b–f) The expression pattern of candidate genes (Acss3, Aldh3a2, Gbe1, Hadhb, and Cyp46a1) in the spleen tissue of M. amblycephala and M. hoffmanni. Different letters indicate a significant difference (p < 0.05).