Xia-Yun Jiang

Functional conservation and divergence of duplicated insulin-like growth factor 2 genes in grass carp (Ctenopharyngodon idellus).

Insulin-like growth factor 2 (IGF2) is a potent mitogenic and survival factor involved in the regulation of growth, development and reproduction in animals. Only one IGF2 gene exists in mammals. Recently, two igf2 genes have been identified in zebrafish, which presumably resulted from gene duplication. However, sequence information of duplicated igf2s and their functional regulation in other teleost fish is still unknown. Here, we report the identification of two igf2 cDNAs in grass carp, Ctenopharyngodon idellus. Like their human ortholog, grass carp igf2a and igf2b mRNAs encoded two structurally distinct mature IGF peptides. Both of them were detected by RT-PCR throughout embryogenesis. Ubiquitous expression of igf2b mRNAs was observed in embryos, whereas igf2a mRNAs were expressed mainly in the notochord and brain with in situ hybridization. In adult fish, igf2b mRNAs were transcribed in multiple tissues, whereas igf2a mRNAs were detected mainly in the liver. Hepatic levels of igf2a and igf2b transcripts were both up-regulated by growth hormone injection. Furthermore, the levels of hepatic igf2a and igf2b mRNAs decreased significantly during starvation and were rebounded rapidly after re-feeding. Our results suggest that duplicated igf2 genes have evolved divergent yet played an overlapping biological role in regulating grass carp growth and development.

Transcriptome Analysis of Blunt Snout Bream (Megalobrama amblycephala) Reveals Putative Differential Expression Genes Related to Growth and Hypoxia.

The blunt snout bream (Megalobrama amblycephala) is an important freshwater aquaculture species, but it is sensitive to hypoxia. No transcriptome data related to growth and hypoxia response are available for this species. In this study, we performed de novo transcriptome sequencing for the liver and gills of the fast-growth family and slow-growth family derived from ‘Pujiang No.1’ F10 blunt snout bream that were under hypoxic stress and normoxia, respectively. The fish were divided into the following 4 groups: fast-growth family under hypoxic stress, FH; slow-growth family under hypoxic stress, SH; fast-growth family under normoxia, FN; and slow-growth family under normoxia, SN. A total of 185 million high-quality reads were obtained from the normalized cDNA of the pooled samples, which were assembled into 465,582 contigs and 237,172 transcripts. A total of 31,338 transcripts from the same locus (unigenes) were annotated and assigned to 104 functional groups, and 23,103 unigenes were classified into seven main categories, including 45 secondary KEGG pathways. A total of 22,255 (71%) known putative unigenes were found to be shared across the genomes of five model fish species and mammals, and a substantial number (9.4%) of potentially novel genes were identified. When 6,639 unigenes were used in the analysis of differential expression (DE) genes, the number of putative DE genes related to growth pathways in FH, SH, SN and FN was 159, 118, 92 and 65 in both the liver and gills, respectively, and the number of DE genes related to hypoxic response was 57, 33, 23 and 21 in FH, FN, SH and SN, respectively. Our results suggest that growth performance of the fast-growth family should be due to complex mutual gene regulatory mechanisms of these putative DE genes between growth and hypoxia.

Speciation of polyploid Cyprinidae fish of common carp, crucian carp, and silver crucian carp derived from duplicated Hox genes

Recent studies on comparative genomics have suggested that a round of fish-specific whole genome duplication (3R) in ray-finned fishes might have occurred around 226-316 Mya. Additional genome duplication, specifically in cyprinids, may have occurred more recently after the divergence of the teleosts. The timing of this event, however, is unknown. To address this question, we sequenced four Hox genes from taxa representing the polyploid Cyprinidae fish, common carp (Cyprinus carpio, 2n=100), crucian carp (Carassius auratus auratus, 2n=100), and silver crucian carp (C. auratus gibelio, 2n=156), and then compared them with known sequences from the diploid Cyprinidae fish, blunt snout bream (Megalobrama amblycephala, 2n=48). Our results showed the presence of two distinct Hox duplicates in the genomes of common and crucian carp. Three distinct Hox sequences, one of them orthologous to a Hox gene in common carp and the other two orthologous to a Hox gene in crucian carp, were isolated in silver crucian carp, indicating a possible hybrid origin of silver crucian carp from crucian and common carp. The gene duplication resulting in the origin of the common ancestor of common and crucian carp likely occurred around 10.9-13.2 Mya. The speciations of common vs. crucian carp and silver crucian vs. crucian carp likely occurred around 8.1-11.4 and 2.3-3.0 Mya, respectively. Finally, nonfunctionalization resulting from point mutations in the coding region is a probable fate for some Hox duplicates. Taken together, these results suggested an evolutionary model for polyploidization in speciation and diversification of polyploid fish.

Molecular cloning and function analysis of insulin-like growth factor-binding protein 1a in blunt snout bream ( Megalobrama amblycephala )

Insulin-like growth factor-binding protein 1 (IGFBP-1), a hypoxia-induced protein, is a member of the IGFBP family that regulates vertebrate growth and development. In this study, full-length IGFBP-1a cDNA was cloned from a hypoxia-sensitive Cyprinidae fish species, the blunt snout bream (Megalobrama amblycephala). IGFBP-1a was expressed in various organs of adult blunt snout bream, including strongly in the liver and weakly in the gonads. Under hypoxia, IGFBP-1a mRNA levels increased sharply in the skin, liver, kidney, spleen, intestine and heart tissues of juvenile blunt snout bream, but recovered to normal levels after 24-hour exposure to normal dissolved oxygen. In blunt snout bream embryos, IGFBP-1a mRNA was expressed at very low levels at both four and eight hours post-fertilization, and strongly at later stages. Embryonic growth and development rates decreased significantly in embryos injected with IGFBP-1a mRNA. The average body length of IGFBP-1a-overexpressed embryos was 82.4% of that of the control group, and somite numbers decreased to 85.2%. These findings suggest that hypoxia-induced IGFBP-1a may inhibit growth in this species under hypoxic conditions.

Insertion efficiency of Tgf2 transposon in the genome of Megalobrama amblycephala: Insertion efficiency of Tgf2 transposon in the genome of Megalobrama amblycephala

To study insertion efficiency of goldfish Tgf2 transposon in the genome of Megalobrama amblycephala, we built Tgf2-Mlyz2-RFP donor plasmid with goldfish Tgf2 transposon left and right arms, and then co-injected with goldfish Tgf2 transposase mRNA into the 1-2 cell stage fertilized eggs of M. amblycephala. In 30 d- and 180 d-stage larval, RFP fluorescence can be observed in back and side muscle of the fish. The rate of RFP fluorescence expression was 48.1%. In adult fish, PCR results demonstrated that integration efficiency of goldfish Tgf2 transposition system was 31.5% in M. am-blycephala genome. RT-PCR analysis showed that RFP RNAs were highly transcribed among all the 12 tissues in three transgenic fishes, while it could be highly detected only in muscle, skin, and kidney in another two individuals. Our results suggested that RFP expression in tissues vaied among different M. amblycephala. By means of the inverse PCR, the copy numbers of Tgf2 transposon were at least 2 in transgenic M. amblycephala. The average copy number of each fish was about 5. Over 50% of flanking sequences at the insertion site have homologous sequence in other vertebrate species. Our data suggest that goldfish Tgf2 transposon can efficiently mediate gene insertion in M. amblycephala, which could been used in transgene and gene trap in M. amblycephala.

The goldfish hAT-family transposon Tgf2 is capable of autonomous excision in zebrafish embryos.

The goldfish (Carassius auratus) Tgf2 transposon is a vertebrate DNA transposon that belongs to the hAT transposon family. In this study, we constructed plasmids containing either the full-length Tgf2 transposon (pTgf2 plasmid) or a partially-deleted Tgf2 transposon (ΔpTgf2 plasmid), and microinjected these plasmids into fertilized zebrafish (Danio rerio) eggs at the one- to two-cell stage. DNA extracted from the embryos was analyzed by PCR to assess transient excision, if any, of the exogenous plasmid and to verify whether Tgf2 is an autonomous transposon. The results showed that excision-specific bands were not detected in embryos injected with the ΔpTgf2 plasmid, while bands of 300-500bp were detected in embryos injected with pTgf2, which indicated that the full-length Tgf2-containing plasmid could undergo autonomous excision in zebrafish embryos. DNA cloned from 24 embryos injected with pTgf2 was sequenced, and the results suggested that Tgf2 underwent self-excision in zebrafish embryos. Cloning and PCR analysis of DNA extracted from embryos co-injected with ΔpTgf2 and in vitro-transcribed transposase mRNA indicated that partially-deleted-Tgf2-containing ΔpTgf2 plasmid also underwent excision, in the presence of functional transposase mRNA. DNA cloned from 25 embryos co-injected with ΔpTgf2 and transposase mRNA was sequenced, and the results suggested that partially-deleted Tgf2 transposons plasmids were excised. These results demonstrated that excisions of Tgf2 transposons were mediated by the Tgf2 transposase, which in turn confirmed that Tgf2 is an autonomous transposon.

Two myostatin genes exhibit divergent and conserved functions in grass carp (Ctenopharyngodon idellus)

Myostatin (MSTN) is an important negative regulator of myogenesis, which inhibits myoblast proliferation and differentiation. Here, we report the isolation and characterization of two mstn genes in grass carp (Ctenopharyngodon idellus). Grass carp mstn-1 and mstn-2 cDNAs are highly divergent, sharing a relatively low amino acid sequence identity of 66%. In adult fish, both orthologs are expressed in numerous tissues and they are differentially regulated during a fasting/refeeding treatments. During embryogenesis, the mRNA levels of both mstn-1 and -2 were upregulated significantly at the beginning of somitogenesis, and maintained at high levels until hatching. Using in situ hybridization, grass carp mstn-1 mRNA was found to ubiquitously express at 12hpf, with strong signals in the notochord, and in the eyes, brain and tailbud at 24hpf, and in brain and notochord at 36hpf. In comparison, the mstn-2 mRNA can be detected in the eyes, brain and notochord at 24hpf, and in the notochord and hindbrain at 36hpf. Further overexpression of mstn-1 mRNA caused a strongly ventralized phenotype by inhibiting dorsal tissue development, while injection of mstn-2 mRNA resulted in obvious embryonic abnormalities in grass carp. These results provide some new insights into the functional conservation and divergence of mstn genes in teleost species.

The N-terminal zinc finger domain of Tgf2 transposase contributes to DNA binding and to transposition activity

Active Hobo/Activator/Tam3 (hAT) transposable elements are rarely found in vertebrates. Previously, goldfish Tgf2 was found to be an autonomously active vertebrate transposon that is efficient at gene-transfer in teleost fish. However, little is known about Tgf2 functional domains required for transposition. To explore this, we first predicted in silico a zinc finger domain in the N-terminus of full length Tgf2 transposase (L-Tgf2TPase). Two truncated recombinant Tgf2 transposases with deletions in the N-terminal zinc finger domain, S1- and S2-Tgf2TPase, were expressed in bacteria from goldfish cDNAs. Both truncated Tgf2TPases lost their DNA-binding ability in vitro, specifically at the ends of Tgf2 transposon than native L-Tgf2TPase. Consequently, S1- and S2-Tgf2TPases mediated gene transfer in the zebrafish genome in vivo at a significantly (p < 0.01) lower efficiency (21%–25%), in comparison with L-Tgf2TPase (56% efficiency). Compared to L-Tgf2TPase, truncated Tgf2TPases catalyzed imprecise excisions with partial deletion of TE ends and/or plasmid backbone insertion/deletion. The gene integration into the zebrafish genome mediated by truncated Tgf2TPases was imperfect, creating incomplete 8-bp target site duplications at the insertion sites. These results indicate that the zinc finger domain in Tgf2 transposase is involved in binding to Tgf2 terminal sequences, and loss of those domains has effects on TE transposition.

Two follistatin-like 1 homologs are differentially expressed in adult tissues and during embryogenesis in grass carp (Ctenopharyngodon idellus).

Follistatin-like 1 (Fstl1) peptides play important roles in inhibiting myoblast proliferation and differentiation. Here, we characterized and examined the expression patterns of fstl1a and -b in grass carp (Ctenopharyngodon idellus). These genes encode 314 aa and 310 aa peptides, respectively, sharing a sequence identity of 83%. Except for the existence of the follistatin-N-terminal (FOLN) and Kazal-type 2 serine protease inhibitor (Kazal 2) domains, grass carp Fstl1a and -b do not share amino acid sequence similarity with Fst1 and -b. Both fstl1a and -b mRNAs were widely expressed in adult tissues. During embryogenesis, grass carp fstl1a and -b mRNA was detected in the presomitic mesoderm and somites at 12h post fertilization (hpf). At 24hpf, fstl1a mRNA was expressed in the hindbrain, somites, notochord and tailbud, while fstl1b mRNA was only detected in the tailbud. At 36hpf, fstl1a mRNA was detected in the hindbrain and notochord, and fstl1b was also expressed in the notochord. Furthermore, fstl1a and -b were downregulated in brain and liver tissue following injection with 10 or 50μg hGH, while fstl1b was significantly up-regulated in muscle tissue after 10μg hGH treatment. Both fstl1a and -b were significantly up-regulated at 2, 4 or 6days of nutrient restriction, and fstl1a was still highly expressed in the liver and muscle after 3days of refeeding, as was fstl1b in the brain and muscle. The expression of these genes returned to near control levels following 6days of refeeding. Our findings suggest that the two fstls play important but divergent roles in embryonic development and tissue growth regulation in grass carp.

Functional conservation and divergence of duplicated fibroblast growth factor receptor 1 (fgfr1) genes in blunt snout bream (Megalobrama amblycephala).

Fgfr1 is a fibroblast growth factor receptor involved in regulating cell growth, proliferation, differentiation and migration. Here, we report the isolation and characterization of duplicated fgfr1 genes in blunt snout bream (Megalobrama amblycephala). Blunt snout bream fgfr1a and -1b cDNAs were found to share a relatively high sequence identity of 82%. During embryogenesis, both fgfr1a and -1b mRNAs were highly detected at zygotes but gradually decreased and then constantly expressed after 16hpf, besides a strong expression for the fgfr1b mRNA at 12hpf. Whole-mount in situ hybridization demonstrated that fgfr1a mRNA was transcribed at the eyes, mid-hindbrain boundary (MHB), brain, posterior somites and tailbud at 16hpf, while the fgfr1b mRNA was only detected at the eyes and posterior somites at the same period. At 28hpf embryos, both fgfr1a and -1b mRNAs were expressed in the eyes, brain, pharyngeal arches and tailbud, and in the eyes, brain, pharyngeal arches and notochord at 55hpf. In adult fish, fgfr1a mRNA was strongly expressed in the gill, gonad, brain and midgut, but examined relatively low in the skin and kidney. In contrast, the fgfr1b mRNA was highly detected in the brain and liver and quite low in the skin, gill and kidney. During starvation, both fgfr1a and -1b mRNAs were significantly up-regulated in the intestine and liver, but down-regulated in the brain. Moreover, duplicated fgfr1 mRNAs were differentially inhibited in tissues with exogenous recombinant hGH. Our results suggest that two fgfr1 genes play important roles in regulating growth and development in blunt snout bream.