Mar Biotechnol (2015) 17:162–167 DOI 10.1007/s10126-014-9604-x
ORIGINAL ARTICLE
Zebrafish fgf10b has a Complementary Function to fgf10a in Liver and Pancreas Development Chuan Yan & Weiling Zheng & Zhiyuan Gong
Received: 15 July 2014 / Accepted: 29 September 2014 / Published online: 18 October 2014 # Springer Science+Business Media New York 2014
Abstract Fgf10 is a critical growth factor in mammals for development of endodermal organs such as the liver, pancreas, lung, and gut. Due to whole genome duplication, the zebrafish has two fgf10 orthologs, fgf10a and fgf10b. While fgf10a has a role in development of the esophagus and swimbladder, we found in the present study that fgf10b had a complementary expression pattern in the liver, pancreas, and gut. Morpholino knockdown of Fgf10b further confirmed its essential role in the normal development of liver and pancreas. Thus, our data provide another example of functional partition of two duplicated othologous genes during evolution. Keywords Fgf10 . Liver . Pancreas . Zebrafish
Introduction It has been well documented that following two rounds of whole genome duplication during evolution from invertebrates to vertebrates, there was an additional round of genome duplication of about 370 million years ago in the ray-finned fish lineage, which resulted in more genes in these fishes including the zebrafish than in most other vertebrate species (Postlethwait et al. 2004). Therefore, for at least 20 % of human genes, the zebrafish has two co-orthologs (Postlethwait et al. 2004). Although this presents a challenge to the application of functional genomics, the teleost genome Chuan Yan and Weiling Zheng authors contributed equally C. Yan : W. Zheng : Z. Gong (*) Department of Biological Sciences, National University of Singapore, Singapore, Singapore e-mail: [email protected] C. Yan : Z. Gong National University of Singapore Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore, Singapore
duplication presents previously unanticipated advantages for the analysis of gene function, because the subfunction partitioning in teleosts can facilitate the identification of gene functions obscured in mammals by either pleiotropy or haploinsufficiency. Fgf10 is a critical growth factor for endodermal development. The expression and function of fgf10a in zebrafish endodermal development has been well studied (Korzh et al. 2008; Winata et al. 2009). The zebrafish fgf10b paralog has not been subjected to any experimental studies until very recently when Maulding et al. reported its early developmental role in otic and epibranchial placodes (Maulding et al. 2014). In this study, we examined the expression and function of fgf10b during endodermal organ development. We showed that the expression of fgf10a and fgf10b are expressed in different domains in the endodermal organs, which provides a good example of spatial separation of gene function after the genome duplication. We further showed that fgf10b has a function in both positioning and growth of the liver and pancreas.
Materials and Methods Zebrafish Wild Type, Transgenic, and Mutant Strains Wi l d t y p e z e b r a f i s h a n d L i P a n t r a n s g e n i c l i n e [Tg(fabp10a:DsRed; elaA:efgp)] (Korzh et al. 2008) were maintained in the DBS zebrafish facility. All experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of National University of Singapore (Protocol 079/07). Reverse Transcription Quantitative Real-Time PCR Total RNAs were isolated from pooled embryos/fry (n>100 at each time point) by using RNeasy kit (Qiagen), and cDNA
Mar Biotechnol (2015) 17:162–167
synthesis was conducted using SuperScriptTM II First-Strand Synthesis System (Invitrogen). Real-time PCR was conducted using LightCycler DNA amplification kit SYBR Green I (Roche Applied Science) according to the manufacturer’s instruction manual with minor modifications and following the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines (Bustin et al. 2009). Whole Mount In Situ Hybridization and Histology Whole mount in situ hybridization with Digoxygenin-labeled riboprobes was performed using a standard protocol as described previously (Korzh et al. 1998). Morpholino Knockdown Fgf10b knockdown was performed using antisense morpholino oligonucleotides targeted against the 5′ UTR (starting from the 9th nucleotide upstream of the first ATG
Fig. 1 Sequence analyses of zebrafish fgf10a and fgf10b genes. a Alignment of Fgf10a and Fgf10b protein sequences. Similar amino acids are shaded in light blue and identical amino acids are in dark blue. Amino acid positions are labeled at the right end of each row. Alignment was generated by DNAMAN. b Phylogenetic analyses of fgf10 gene family in selected vertebrate species. The phylogenetic tree indicates that the two zebrafish fgf10 genes were evolved after species separation from other
163
codon) with the following sequence: 5′ TCATGGGACATG CTGCTCAACCAAT 3′. A standard control morpholino, (5′-CCTCTTACCTCAGTTACAATTTATA-3′) (Gene Tools, Philomath, OR), that targets human beta-globin intron, was used as the negative control. MOs were synthesized by Gene Tools (USA) and kept as 1 mM stock in Danieau’s solution. Wild type and transgenic embryos of one- to twocell stage were injected (200–250 embryos each group). For control, 1 ng of standard control morpholino was injected. For Fgf10b morphants, either 6 ng (low dose) or 12 ng (high dose) were injected.
Microscopy and Image Analysis Twenty larvae of each group were randomly chosen for imaging. The larvae were anesthetized in 0.08 % tricaine (Sigma, E10521) and immobilized in 3 % methylcellulose (Sigma, M0521). Observations and photography of live embryos were performed using a compound fluorescent microscope (Zeiss,
vertebrate species. c Syntenic relationship of the two zebrafish fgf10 genes with human FGF10. Zebrafish fgf10a and fgf10b are located in chromosome 5 (Dre5) and chromosome 21 (Dre21), respectively, and human FGF10 is in human chromosome 5 (Hsa5). The putative othologous genes between human and zebrafish are indicated with dashed lines
164
Germany). 2D measurement of liver sizes was performed using ImageJ as described previously (Huang et al. 2012). Statistics Analysis Statistical significance between different larvae groups was evaluated by two-tailed unpaired Student t test using InStat version 5.0 software for Windows (GraphPad, San Diego, CA). Statistical data are presented as mean values±standard error deviation (SED).
Results and Discussion Existence of Two Copies of fgf10 Genes in the Zebrafish Genome In an effort to identify members of FGF superfamily in zebrafish, we found an uncharacterized fgf sequence which was closely related to zebrafish fgf10a and annotated as fgf10b. The predicted Fgf10b amino acid sequence share 55.2 % identity with that of Fgf10a, and the two protein sequences could be aligned very well except for the first 30– 50 N-terminal amino acids (Fig. 1a). Phylogentical analyses indicated the two zebrafish genes were clustered together before they clustered with Xenopus counterparts and others Fig. 2 Temporal and spatial expression of fgf10b and fgf10a in different developmental stages. a RT-qPCR analyses of fgf10b and fgf10a expression from 1 cell stage to 96 hpf. Fgf10b was maternally transcribed, and the transcript was persistent beyond 96 hpf. Fgf10a transcript was detected starting at 12 hpf. Gapdh was used as the housekeeping gene. b, c Expression of fgf10b in endodermal organs at 72 hpf by whole mount in situ hybridization (n>30). Fgf10b was expressed in the liver, pancreas, and esophagus at 72 hpf. Red dotted circle indicates swimbladder. L liver, p pancreas, g gut, e esophagus, sb swimbladder. Scale bars 250 μm
Mar Biotechnol (2015) 17:162–167
(Fig. 1b), indicating that this pair of genes was duplicated within the fish species. By searching their genomic loci, we found that fgf10a was located in chromosome 12 while fgf10b in chromosome 5 (Fig. 1c). Previous synteny study showed that conserved gene order was observed at the human FGF10 and zebrafish fgf10a loci, indicating that fgf10a is a zebrafish ortholog of human FGF10 (Itoh and Konishi 2007). To investigate the potential synteny between zebrafish fgf10b and human FGF10, human FGF10 gene was compared with both zebrafish fgf10a and fgf10b with online synteny database (http://syntenydb.uoregon.edu/synteny_db/) (Catchen et al. 2009). By using 25 gene sliding window, zebrafish fgf10a showed a quite highly conserved gene cluster (11 genes) with human FGF10, in consistence with the previous synteny study (Itoh and Konishi 2007). In contrast, only two genes (paip1 and C5orf34) are present between zebrafish fgf10b and human FGF10 regions, suggesting the zebrafish fgf10b loci had went through quite dramatic changes during evolution. Temporal and Spatial Expression of fgf10b We first examined the transcript level of fgf10b at different developmental stages by reverse transcription quantitative real-time PCR (RT-qPCR). Fgf10b mRNA showed maternal expression, and the expression level was peaked at 12 hpf and maintained up to at least 120 hpf (Fig. 2a). In comparison,
Mar Biotechnol (2015) 17:162–167
fgf10a mRNA was detected only from 12 hpf. To further investigate the expression pattern of fgf10b in zebrafish embryos, whole mount in situ hybridization was carried out. The early expression of Fgf10b has been recently reported, and it is initiated in the mesoderm near the otic/epibranchial domain and tailbud and plays an important role in induction of otic and epibranchial placodes (Maulding et al. 2014), However, the expression and role of fgf10b in late development has not been reported. Here, we found that at 48 hpf, fgf10b expression was detected in several endodermal organs, including the liver, exocrine pancreas, and gut (data not shown). The expression became prominent at 72 hpf (Fig. 2b), but no fgf10b expression was observed in the swimbladder. In contrast, fgf10a expression was detected in the swimbladder as well as in the foregut (Fig. 2c). Developmental Defects of Liver and Pancreas in Fgf10b Morphants In order to reveal the potential function of fgf10b during zebrafish embryonic development, we used a translational blocking morpholino to knockdown Fgf10b expression; 6 or
Fig. 3 Impairment of liver and pancreas development in Fgf10b morphants. a–f Representative images of Fgf10b morphants at 5 dpf. Mopholino oligonucleotides were injected into embryos from the Lipan transgenic line, in which the liver and exocrine pancreas were labeled with dsRed and GFP fluorescence, respectively. Standard control was injected with standard control morpholino and low and high dose groups were injected with Fgf10b morpholino at 6 ng and 12 ng, respectively. Panels a, c, e are lateral view and panels b, d, f are ventral view with indication of left and right sides. i Quantification of delayed gastrulation in Fgf10b morphants. Fgf10b morphants displayed a delayed development judging from the decreased percentage of embryos entering
165
12 ng of morpholino was injected into embryos at 1-cell stage. The morpholino did not produce any obvious early effect up to gastrulation. Dose-dependent developmental delay was observed in the morphants in a dose-dependent manner. At 6 hpf, 46 % of morphants went through gastrulation in the low dose group while 23 % of morphants in the high dose group went through gastrulation; in comparison, 96 % control larvae injected with standard control morpholino completed gastrulation by 6 hpf. Using LiPan transgenic embyros which have RFP fluorescence in the liver and GFP fluorescence in the exocrine pancreas (Korzh et al. 2008), we found that the size of the liver and exocrine pancreas was obviously affected in Fgf10b morphants (Fig. 3a–f), and the effect was dose-dependent (Fig. 3j–l). In zebrafish, the liver is in the fast growth phase between 50 and 96 hpf (Field et al. 2003b; Korzh et al. 2008). At 5 dpf, the liver was observed to touch the pericardial cavity and to lie on top of the remaining yolk from the left lateral view (Fig. 3a, b). As shown in Fig. 3, in the fgf10b morphants, the liver growth was obviously impaired. The liver size was increasingly reduced with the increase of injected morpholino (Fig. 3j). At the same time (5 dpf), the exocrine pancreas in the
gastrulation at 6 hpf. j Quantification of liver size in Fgf10b morphants. Liver size was measured based on 2D images of the DsRed-labeled livers as described previously (Huang et al. 2012). At 5 dpf, significant decrease in liver size was measured in both low and high Fgf10b morpholinoinjected larvae as compared to control (*p
ORIGINAL ARTICLE
Zebrafish fgf10b has a Complementary Function to fgf10a in Liver and Pancreas Development Chuan Yan & Weiling Zheng & Zhiyuan Gong
Received: 15 July 2014 / Accepted: 29 September 2014 / Published online: 18 October 2014 # Springer Science+Business Media New York 2014
Abstract Fgf10 is a critical growth factor in mammals for development of endodermal organs such as the liver, pancreas, lung, and gut. Due to whole genome duplication, the zebrafish has two fgf10 orthologs, fgf10a and fgf10b. While fgf10a has a role in development of the esophagus and swimbladder, we found in the present study that fgf10b had a complementary expression pattern in the liver, pancreas, and gut. Morpholino knockdown of Fgf10b further confirmed its essential role in the normal development of liver and pancreas. Thus, our data provide another example of functional partition of two duplicated othologous genes during evolution. Keywords Fgf10 . Liver . Pancreas . Zebrafish
Introduction It has been well documented that following two rounds of whole genome duplication during evolution from invertebrates to vertebrates, there was an additional round of genome duplication of about 370 million years ago in the ray-finned fish lineage, which resulted in more genes in these fishes including the zebrafish than in most other vertebrate species (Postlethwait et al. 2004). Therefore, for at least 20 % of human genes, the zebrafish has two co-orthologs (Postlethwait et al. 2004). Although this presents a challenge to the application of functional genomics, the teleost genome Chuan Yan and Weiling Zheng authors contributed equally C. Yan : W. Zheng : Z. Gong (*) Department of Biological Sciences, National University of Singapore, Singapore, Singapore e-mail: [email protected] C. Yan : Z. Gong National University of Singapore Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore, Singapore
duplication presents previously unanticipated advantages for the analysis of gene function, because the subfunction partitioning in teleosts can facilitate the identification of gene functions obscured in mammals by either pleiotropy or haploinsufficiency. Fgf10 is a critical growth factor for endodermal development. The expression and function of fgf10a in zebrafish endodermal development has been well studied (Korzh et al. 2008; Winata et al. 2009). The zebrafish fgf10b paralog has not been subjected to any experimental studies until very recently when Maulding et al. reported its early developmental role in otic and epibranchial placodes (Maulding et al. 2014). In this study, we examined the expression and function of fgf10b during endodermal organ development. We showed that the expression of fgf10a and fgf10b are expressed in different domains in the endodermal organs, which provides a good example of spatial separation of gene function after the genome duplication. We further showed that fgf10b has a function in both positioning and growth of the liver and pancreas.
Materials and Methods Zebrafish Wild Type, Transgenic, and Mutant Strains Wi l d t y p e z e b r a f i s h a n d L i P a n t r a n s g e n i c l i n e [Tg(fabp10a:DsRed; elaA:efgp)] (Korzh et al. 2008) were maintained in the DBS zebrafish facility. All experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of National University of Singapore (Protocol 079/07). Reverse Transcription Quantitative Real-Time PCR Total RNAs were isolated from pooled embryos/fry (n>100 at each time point) by using RNeasy kit (Qiagen), and cDNA
Mar Biotechnol (2015) 17:162–167
synthesis was conducted using SuperScriptTM II First-Strand Synthesis System (Invitrogen). Real-time PCR was conducted using LightCycler DNA amplification kit SYBR Green I (Roche Applied Science) according to the manufacturer’s instruction manual with minor modifications and following the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines (Bustin et al. 2009). Whole Mount In Situ Hybridization and Histology Whole mount in situ hybridization with Digoxygenin-labeled riboprobes was performed using a standard protocol as described previously (Korzh et al. 1998). Morpholino Knockdown Fgf10b knockdown was performed using antisense morpholino oligonucleotides targeted against the 5′ UTR (starting from the 9th nucleotide upstream of the first ATG
Fig. 1 Sequence analyses of zebrafish fgf10a and fgf10b genes. a Alignment of Fgf10a and Fgf10b protein sequences. Similar amino acids are shaded in light blue and identical amino acids are in dark blue. Amino acid positions are labeled at the right end of each row. Alignment was generated by DNAMAN. b Phylogenetic analyses of fgf10 gene family in selected vertebrate species. The phylogenetic tree indicates that the two zebrafish fgf10 genes were evolved after species separation from other
163
codon) with the following sequence: 5′ TCATGGGACATG CTGCTCAACCAAT 3′. A standard control morpholino, (5′-CCTCTTACCTCAGTTACAATTTATA-3′) (Gene Tools, Philomath, OR), that targets human beta-globin intron, was used as the negative control. MOs were synthesized by Gene Tools (USA) and kept as 1 mM stock in Danieau’s solution. Wild type and transgenic embryos of one- to twocell stage were injected (200–250 embryos each group). For control, 1 ng of standard control morpholino was injected. For Fgf10b morphants, either 6 ng (low dose) or 12 ng (high dose) were injected.
Microscopy and Image Analysis Twenty larvae of each group were randomly chosen for imaging. The larvae were anesthetized in 0.08 % tricaine (Sigma, E10521) and immobilized in 3 % methylcellulose (Sigma, M0521). Observations and photography of live embryos were performed using a compound fluorescent microscope (Zeiss,
vertebrate species. c Syntenic relationship of the two zebrafish fgf10 genes with human FGF10. Zebrafish fgf10a and fgf10b are located in chromosome 5 (Dre5) and chromosome 21 (Dre21), respectively, and human FGF10 is in human chromosome 5 (Hsa5). The putative othologous genes between human and zebrafish are indicated with dashed lines
164
Germany). 2D measurement of liver sizes was performed using ImageJ as described previously (Huang et al. 2012). Statistics Analysis Statistical significance between different larvae groups was evaluated by two-tailed unpaired Student t test using InStat version 5.0 software for Windows (GraphPad, San Diego, CA). Statistical data are presented as mean values±standard error deviation (SED).
Results and Discussion Existence of Two Copies of fgf10 Genes in the Zebrafish Genome In an effort to identify members of FGF superfamily in zebrafish, we found an uncharacterized fgf sequence which was closely related to zebrafish fgf10a and annotated as fgf10b. The predicted Fgf10b amino acid sequence share 55.2 % identity with that of Fgf10a, and the two protein sequences could be aligned very well except for the first 30– 50 N-terminal amino acids (Fig. 1a). Phylogentical analyses indicated the two zebrafish genes were clustered together before they clustered with Xenopus counterparts and others Fig. 2 Temporal and spatial expression of fgf10b and fgf10a in different developmental stages. a RT-qPCR analyses of fgf10b and fgf10a expression from 1 cell stage to 96 hpf. Fgf10b was maternally transcribed, and the transcript was persistent beyond 96 hpf. Fgf10a transcript was detected starting at 12 hpf. Gapdh was used as the housekeeping gene. b, c Expression of fgf10b in endodermal organs at 72 hpf by whole mount in situ hybridization (n>30). Fgf10b was expressed in the liver, pancreas, and esophagus at 72 hpf. Red dotted circle indicates swimbladder. L liver, p pancreas, g gut, e esophagus, sb swimbladder. Scale bars 250 μm
Mar Biotechnol (2015) 17:162–167
(Fig. 1b), indicating that this pair of genes was duplicated within the fish species. By searching their genomic loci, we found that fgf10a was located in chromosome 12 while fgf10b in chromosome 5 (Fig. 1c). Previous synteny study showed that conserved gene order was observed at the human FGF10 and zebrafish fgf10a loci, indicating that fgf10a is a zebrafish ortholog of human FGF10 (Itoh and Konishi 2007). To investigate the potential synteny between zebrafish fgf10b and human FGF10, human FGF10 gene was compared with both zebrafish fgf10a and fgf10b with online synteny database (http://syntenydb.uoregon.edu/synteny_db/) (Catchen et al. 2009). By using 25 gene sliding window, zebrafish fgf10a showed a quite highly conserved gene cluster (11 genes) with human FGF10, in consistence with the previous synteny study (Itoh and Konishi 2007). In contrast, only two genes (paip1 and C5orf34) are present between zebrafish fgf10b and human FGF10 regions, suggesting the zebrafish fgf10b loci had went through quite dramatic changes during evolution. Temporal and Spatial Expression of fgf10b We first examined the transcript level of fgf10b at different developmental stages by reverse transcription quantitative real-time PCR (RT-qPCR). Fgf10b mRNA showed maternal expression, and the expression level was peaked at 12 hpf and maintained up to at least 120 hpf (Fig. 2a). In comparison,
Mar Biotechnol (2015) 17:162–167
fgf10a mRNA was detected only from 12 hpf. To further investigate the expression pattern of fgf10b in zebrafish embryos, whole mount in situ hybridization was carried out. The early expression of Fgf10b has been recently reported, and it is initiated in the mesoderm near the otic/epibranchial domain and tailbud and plays an important role in induction of otic and epibranchial placodes (Maulding et al. 2014), However, the expression and role of fgf10b in late development has not been reported. Here, we found that at 48 hpf, fgf10b expression was detected in several endodermal organs, including the liver, exocrine pancreas, and gut (data not shown). The expression became prominent at 72 hpf (Fig. 2b), but no fgf10b expression was observed in the swimbladder. In contrast, fgf10a expression was detected in the swimbladder as well as in the foregut (Fig. 2c). Developmental Defects of Liver and Pancreas in Fgf10b Morphants In order to reveal the potential function of fgf10b during zebrafish embryonic development, we used a translational blocking morpholino to knockdown Fgf10b expression; 6 or
Fig. 3 Impairment of liver and pancreas development in Fgf10b morphants. a–f Representative images of Fgf10b morphants at 5 dpf. Mopholino oligonucleotides were injected into embryos from the Lipan transgenic line, in which the liver and exocrine pancreas were labeled with dsRed and GFP fluorescence, respectively. Standard control was injected with standard control morpholino and low and high dose groups were injected with Fgf10b morpholino at 6 ng and 12 ng, respectively. Panels a, c, e are lateral view and panels b, d, f are ventral view with indication of left and right sides. i Quantification of delayed gastrulation in Fgf10b morphants. Fgf10b morphants displayed a delayed development judging from the decreased percentage of embryos entering
165
12 ng of morpholino was injected into embryos at 1-cell stage. The morpholino did not produce any obvious early effect up to gastrulation. Dose-dependent developmental delay was observed in the morphants in a dose-dependent manner. At 6 hpf, 46 % of morphants went through gastrulation in the low dose group while 23 % of morphants in the high dose group went through gastrulation; in comparison, 96 % control larvae injected with standard control morpholino completed gastrulation by 6 hpf. Using LiPan transgenic embyros which have RFP fluorescence in the liver and GFP fluorescence in the exocrine pancreas (Korzh et al. 2008), we found that the size of the liver and exocrine pancreas was obviously affected in Fgf10b morphants (Fig. 3a–f), and the effect was dose-dependent (Fig. 3j–l). In zebrafish, the liver is in the fast growth phase between 50 and 96 hpf (Field et al. 2003b; Korzh et al. 2008). At 5 dpf, the liver was observed to touch the pericardial cavity and to lie on top of the remaining yolk from the left lateral view (Fig. 3a, b). As shown in Fig. 3, in the fgf10b morphants, the liver growth was obviously impaired. The liver size was increasingly reduced with the increase of injected morpholino (Fig. 3j). At the same time (5 dpf), the exocrine pancreas in the
gastrulation at 6 hpf. j Quantification of liver size in Fgf10b morphants. Liver size was measured based on 2D images of the DsRed-labeled livers as described previously (Huang et al. 2012). At 5 dpf, significant decrease in liver size was measured in both low and high Fgf10b morpholinoinjected larvae as compared to control (*p
