Structural Abnormalities of Spermatozoa in Triploid Gynogenetic Crucian Carp (Carassius auratus)

The spermatozoa of triploid gynogenetic crucian carp (Carassius auratus) (3nDTCC) possess a spermatogenesis process with a normal genetic background. However, the genetic materials of these spermatozoa do not completely inherit gynogenetic progeny in general. Understanding the intrinsic mechanism may be helpful for developing breeding strategies of gynogenetic fishes. In this study, the spermatozoa ultrastructure was systematically studied in diploid red crucian carp and 3nDTCC to demonstrate their cytological structural differences. In addition, the artificial breeding tests of 3nDTCC(♀) with different ploidy spermatozoa were performed to verify the contributions of genetic materials from 3nDTCC spermatozoa to the gynogenesis progeny. Furthermore, the mRNA expression of centriole-related genes (i.e., cep57, cetn1, rootletin, and nek2) involved in spermatozoa packaging was also determined by quantitative real-time PCR (qPCR) to illustrate the molecular expression characteristics of the spermatozoa packaging process in 3nDTCC. The results reveal the adaptive features of spermatozoa in 3nDTCC, including the loose midpiece structure, abnormal head structure, and abnormal expression of centriole-related genes, which may influence the motility of spermatozoa and make it not involved normally in the genetic composition of the gynogenesis offspring.

Localization of RNA Pol II CTD (S5) and Transcriptome Analysis of Testis in Diploid and Tetraploid Hybrids of Red Crucian Carp (♀) × Common Carp (♂)

Polyploidy occurs naturally in fish; however, the appearance of these species is an occasional and gradual process, which makes it difficult to trace the changes in phenotypes, genotypes, and regulation of gene expression. The allotetraploid hybrids (4nAT) of red crucian carp (RCC; ♀) × common carp (CC; ♂) generated from interspecies crossing are a good model to investigate the initial changes after allopolyploidization. In the present study, we focused on the changes in the active sites of the testicular transcriptome of the allotetraploid by localization of RNA Pol II CTD YSPTSPS (phospho S5) using immunofluorescence and RNA-seq data via bioinformatic analysis. The results showed that there was no significant difference in signal counts of the RNA Pol II CTD (S5) between the different types of fish at the same stages, including RCC, CC, 2nF1, and 4nAT, which means that the number of transcriptionally active sites on germ cell chromosomes was not affected by the increase in chromosome number. Similarly, RNA-seq analysis indicated that in the levels of chromosomes and 10-kb regions in the genome, there were no significant changes in the highly active sites in RCC, 2nF1, and 4nAT. These findings suggest that at the beginning of tetraploid origin, the active transcriptome site of 4nAT in the testis was conserved in the regions of the genome compared to that in RCC and 2nF1. In conclusion, 4nAT shared a similar gene expression model in the regions of the genome with RCC and 2nF1 with significantly different expression levels.

Organization and Expression Analysis of 5S and 45S Ribosomal DNA Clusters in Autotetraploid Fish Derived From Carassius Auratus Red Var. (♀) × Megalobrama Amblycephala (♂)

Background: Autotetraploid fish (4n = 200, RRRR) (abbreviated as 4nRR) are derived from whole genome duplication of red crucian carp (2n = 100, RR) (abbreviated as RCC). rDNA is often used to study molecular evolution of repeated sequences because it has high copy rate and special conserved coding regions in genomes. In this study, we determined the sequences (5S, ITS1-5.8S-ITS2 region), structure, methylation level (NTS and IGS), and expression level (5S and 18S) of 5S and 45S rRNA genes in 4nRR and RCC in order to elucidate the effects of autotetraploidization on ribosomal DNA (rDNA) in fish. Results: Results showed that there was high sequence similarity of 5S, 5.8S and ITS1 region between 4nRR and RCC. This study also identified two different types of ITS2 region in 4nRR and predicted the secondary structure. It turns out that both secondary structures are functional. Compared with the diploid ancestor of RCC, there was no significant difference in NTS (5S) methylation level, but the expression level of 5S rRNA was lower in 4nRR, indicating that methylation had little effect on the expression level in 4nRR. IGS (45S) was hypermethylated in 4nRR compared to RCC, but the expression of 18S gene were no significantly different from that in RCC, indicating that methylation regulation affected gene expression in 4nRR. Conclusion: These results demonstrate the effects of related structure and expression of autotetraploidization on rDNA. In addition, this study provides reference for studying the effect of autopolyploid on the evolution of species.

Variations in the Mitochondrial Genome of a Goldfish-Like Hybrid [Koi Carp (♀) × Blunt Snout Bream (♂)] Indicate Paternal Leakage

Previously, a homodiploid goldfish-like fish (2n = 100; GF-L) was spontaneously generated by self-crossing a homodiploid red crucian carp-like fish (2n = 100; RCC-L), which was in turn produced via the distant hybridization of female koi carp (Cyprinus carpio haematopterus, KOC, 2n = 100) and male blunt snout bream (Megalobrama amblycephala, BSB, 2n = 48). The phenotypes and genotypes of RCC-L and GF-L differed from those of the parental species but were similar to diploid red crucian carp (2n = 100; RCC) and goldfish (2n = 100; GF), respectively. We sequenced the complete mitochondrial DNAs (mtDNAs) of the KOC, BSB, RCC-L, GF-L, and subsequent generations produced by self-crossing [the self-mating offspring of RCC-L (RCC-L-F2) to the self-mating offspring of RCC-L-F2 (RCC-L-F3) and the self-mating offspring of GF-L (GF-L-F2)]. Paternal mtDNA fragments were stably embedded in the mtDNAs of both lineages, forming chimeric DNA fragments. In addition to these chimeras, several nucleotide positions in the RCC-L and GF-L lineages differed from the parental bases, and were instead identical with RCC and GF, respectively. Moreover, RCC-L and GF-L mtDNA organization and nucleotide composition were more similar to those of RCC and GF, respectively, compared to parental mtDNA. Finally, phylogenetic analyses indicated that RCC-L and GF-L clustered with RCC and GF, not with the parental species. The molecular dating time shows that the divergence time of KOC and GF was about 21.26 Mya [95% highest posterior density (HPD): 24.41–16.67 Mya], which fell within the period of recent. The heritable chimeric DNA fragments and mutant loci identified in the mtDNA of the RCC-L and GF-L lineages provided important evidence that hybridizations might lead to changes in the mtDNA and the subsequent generation of new lineages. Our findings also demonstrated for the first time that the paternal mtDNA was transmitted into the mtDNA of homodiploid lineages (RCC-L and GF-L), which provided evidence that paternal DNA plays a role in inherited mtDNA. These evolutionary analyses in mtDNA suggest that GF might have diverged from RCC after RCC diverged from koi carp.

Genetic Variation in an Experimental Goldfish Derived From Hybridization

Owning to the extreme difficulty in identifying the primary generation (G0), the common ancestor of various twin-tail goldfish strains remains unclear. However, several authors have hypothesized that this ancestor may have been the crucian carp (Carassius auratus). Previously, we generated an experimental hybrid goldfish (EG) from the interspecific hybridization of red crucian carp (Carassius auratus ♀, RCC) × common carp (Cyprinus carpio ♂, CC). Unlike either parent, EG possessed twin caudal fins similar to those of natural goldfish (Carassius auratus, NG). The genetic characteristics of EG, as well as the mechanisms underlying its formation, are largely unknown. Here, we identified the genetic variation in the chordin gene that was associated with the formation of the twin-tail phenotype in EG: a stop codon mutation at the 127th amino acid. Furthermore, simple sequence repeat (SSR) genotyping indicated that, among the six alleles, all of the EG alleles were also present in female parent (RCC), but alleles specific to the male parent (CC) were completely lost. At some loci, EG and NG alleles differed, showing that these morphologically similar goldfish were genetically dissimilar. Collectively, our results demonstrated that genetic variations and differentiation contributed to the changes of morphological characteristics in hybrid offspring. This analysis of genetic variation in EG sheds new light on the common ancestor of NG, as well as on the role of hybridization and artificial breeding in NG speciation.