br Materials and methods br Results br
Materials and methods
Discussion Structural analysis of dnmt3 sequences of different ploidy fish cloned in this study showed high similarities to zebrafish sequences, which also contained a unique N-terminal motif and a calponin homology (CH) domain, a conserved proline-tryptophan-tryptophan-proline (PWWP) domain, an ATRX-DNMT3-DNMT3L (ADD) domain, and a C-terminal catalytic domain (Goll and Halpern, 2011). These domains are also appeared in Dnmt3A and Dnmt3B in mammals (Chédin, 2011). Distant hybridization combines different genomes and promotes the formation of new species, thus changing the gene expression and sequence structures in the offspring (Grant et al., 2005). Previous studies reported the partial maternal genetic rules found in 4nAT by transcriptomic analysis and molecular marker technologies (Liu et al., 2016; Wang et al., 2015). By sequence analysis, here we found that dnmt3 gene of 3n fish and 4nAT exhibited more similarity with their original maternal parent than paternal parent, which showed partial maternal genetic rules. However, we could not exclude the possibility of the selective expression or silence of parental genes for lacking the genomic DNA sequences information of polyploidy fish. The causes of male and female sterility in 3n fish are different. In male, although chromosomal spreading of germ gap junction displayed the abnormal pairing of homologous chromosomes, forming 50 bivalents and 50 univalent (Zhang et al., 2004), transcriptomic analysis revealed that genes involved in sperm flagellar assembly and motility were weakly expressed, whereas those involved in meiosis had no change (Xu et al., 2015). Gene dnmt3 in this study displayed no change in testis of different ploidy fish (Fig. S2). In female, histologic sections in this study revealed ovarian development was arrested before the primary growth phase. Both previous study and our results revealed that expression of meiotic key genes, dmc1, msh4, sycp2 and sycp3, which are critical for synaptonemal complex formation in pachytene stage of meiosis I, decreased in ovary of 3n fish (Kneitz et al., 2000; Tao et al., 2008; Yang et al., 2006; Yuan et al., 2000), which indicated that there was impairment in pachytene stage of prophase I during meiotic progression of oocytes. Also, the result of chromosome spreading with ovary cells directly indicated the synapsis defects in female 3n fish (Zhou et al., 2018). Couple with the pachytene arrest, dnmt3 gene expression also decreased in the ovary of 3n fish. Moreover, spatial expression characterization revealed dnmt3 mainly expressed in the gonad, especially in the oocytes. Furthermore, we compared the follicles at different stages of RCC with ovaries of 3n fish, which could not isolated separate follicles from ovary for tight connection between follicles and other somatic support cells. We found that during the normal developmental process, dnmt3 gene expressed higher in follicles of different stages (from PV to FG) when compared with the sterile ovary of 3n fish (Fig. S3). Based on the above results, we speculated that dnmt3 might regulate meiotic progression to involve in sterility in female 3n fish. In agreement with our results, Lsh, which is involved in DNA methylation in mammals, was shown to play critical roles in chromosome synapsis in female germ cells, which mutant exhibited sterility (La Fuente et al., 2006; Zhu et al., 2006). To clarify the mechanism of dnmt3 involving in regulating pachytene impairment of prophase I, which directly lead to sterility in 3n fish, the upstream signal and the downstream target of dnmt3 were investigated. In fertile fish, for the tiny size, indistinguishable form and short time last from oogonia to primary oocytes with meiotic arrest at diplotene stage of prophase I, it was difficult to catch the proper sample to study the mechanism of pachytene checkpoint regulation. Therefore, we changed the thinking and chose the sterile 3n fish as the material to “rescue” for the following functional experiments. Recent studies suggested that DNMT expression was under hormonal control (García-Carpizo et al., 2011). HPG axis, containing the reproductive hormones, was pivotal for oocyte development, including meiosis progression and vitellogenesis. Previous studies reported that the HPG axis is abnormal in sterile 3n fish (Long et al., 2006; Long et al., 2009a). In this study, we found that dnmt3 expression in the ovary of allotriploid fish increased after E2 injection, but had no change after hCG treatment, which might be caused by the different role of the two hormones. HCG, as the gonadotropin, was mainly effected in oocyte maturation and ovulation, and had fewer receptor to act in sterile 3n fish, in which lhr (luteinizing hormone receptor) gene displayed lower expression (Zhou et al., 2014). Study in mammals also observed that E2 increased Dnmt3a and Dnmt3b expression in dorsal hippocampus for enhancer memory consolidation (Zhao and Mcewen, 2010). Furthermore, the expression of dnmt3 is increased in follicles of PV to MV stage, and decreased in follicles of FG stage, which is consistent with the E2 concentration during oocyte developmental process (Fig. S3).