Chung HA et al. (2004), Screening of FGF target genes in Xenopus by mic...

XB-ART-3179

Genes Cells 2004 Aug 01;98:749-61. doi: 10.1111/j.1356-9597.2004.00761.x.

Show Gene links Show Anatomy links

Screening of FGF target genes in Xenopus by microarray: temporal dissection of the signalling pathway using a chemical inhibitor.

Chung HA , Hyodo-Miura J , Kitayama A , Terasaka C , Nagamune T , Ueno N .

???displayArticle.abstract???
Microarray is a powerful tool for analysing gene expression patterns in genome-wide view and has greatly contributed to our understanding of spatiotemporal embryonic development at the molecular level. Members of FGF (fibroblast growth factor) family play important roles in embryogenesis, e.g. in organogenesis, proliferation, differentiation, cell migration, angiogenesis, and wound healing. To dissect spatiotemporally the versatile roles of FGF during embryogenesis, we profiled gene expression in Xenopus embryo explants treated with SU5402, a chemical inhibitor specific to FGF receptor 1 (FGFR1), by microarray. We identified 38 genes that were down-regulated and 5 that were up-regulated in response to SU5402 treatment from stage 10.5-11.5 and confirmed their FGF-dependent transcription with RT-PCR analysis and whole-mount in situ hybridization (WISH). Among the 43 genes, we identified 26 as encoding novel proteins and investigated their spatial expression pattern by WISH. Genes whose expression patterns were similar to FGFR1 were further analysed to test whether any of them represented functional FGF target molecules. Here, we report two interesting genes: one is a component of the canonical Ras-MAPK pathway, similar to mammalian mig6 (mitogen-inducible gene 6) acting in muscle differentiation; the other, similar to GPCR4 (G-protein coupled receptor 4), is a promising candidate for a gastrulation movement regulator. These results demonstrate that our approach is a promising strategy for scanning the genes that are essential for the regulation of a diverse array of developmental processes.

???displayArticle.pubmedLink??? 15298682
???displayArticle.link??? Genes Cells

Species referenced: Xenopus
Genes referenced: arl5a arl5c errfi1 esr-5 fgf4 fgfr1 foxb1 gdf3 gpr4l.2 gsc kdm5c mapk1 mmut myf5 myod1 not nradd rhob spry2 tbxt wnt11b xmc
???displayArticle.antibodies??? Notochord Ab2 Somite Ab1
???displayArticle.morpholinos??? errfi1 MO1 mgc69520 MO2

???attribute.lit??? ???displayArticles.show???

	Figure 1 We identified FGF target genes by microarray analysis and subclassified them based on their sequence. (A) The data set in the scatter plot, show the expression level of genes on a logarithmic scale as the median value of DMSO-exposed explants on the x-axis vs. the median value of SU5402-exposed explants on the y-axis. (B) Out of the 43 genes including 38 down-regulated genes and 5 up-regulated genes by combinatorial microarray. (C) Out of 43 genes, 26 (60%) had unknown function or were novel and 14 genes (33%) had known functions, of which 11 were previously identified FGF targets. Three (7%) were retrotransposon or reverse transcriptase related genes. (D) Nine transcription factors (21%), 4 transmembrane proteins (9%), 3 cytoskeleton interaction (7%), and 3 transport-related genes (7%), and 2 growth factors (5%); however, there were no clues to the functions of 17 genes (40%).
	Figure 2 Whole-mount in situ hybridization of several interesting genes. Among identified genes, we picked up six novel genes that are highly expressed around blastopore, which is showed similar expression pattern with FGFR1 expression in gastrula. From left to right, expression patterns of early gastrula, late gastrula, neurula, and tailbud stage are shown. All gastrula embryos are vegetal but late gastrula of ARL5. Late gastrula of ARL5 is lateral view and dorsal is right. All neurula embryos are dorsal view and anterior is left. For tailbud embryos, XL019m15 and XL027l07 are dorsal view but the others are lateral view. Anterior is left.
	Figure 3 Semiquantitative RT-PCR confirmed FGF-dependent regulation of the identified genes. The amplification of cDNAs obtained from eFGF (0.01 ng), HAVø, or XFD (0.5 ng)-injected Keller explants, and SU5402, or DMSO-exposed Keller explants was performed. Each ratio represents the relative intensity of the respective explants divided by the intensity of the uninjected explants. These results are representative of three experiments.
	Figure 4 How novel genes are related to other FGF components. (A) RT-PCR analysis of animal cap explants was performed. cDNAs from 10 animal caps that had received injections of Xbra, Xspry2 (1 ng) co-injected with eFGF (0.01 ng), and Xmc (1 ng) were amplified by primers to ARL5, GPCR4, mig6, and NRH. (B) Using a MEK inhibitor, U0126, we further examined and confirmed that mig6 is regulated in a Ras-MAPK dependent manner, whereas the other genes are not. RT-PCR analysis of Keller explants was carried out in the presence of U0126 or DMSO.
	Figure 5 Xmig6 is regulated in an eFGF, MAPK, and Xbradependent fashion. (A) To look for the relationship among Xmig6, eFGF (0.01 ng), and Xbra (1 ng), we performed RT-PCR analysis MPK-1 (0.5 ng) and Xbra-EnR (Xbra repressor, 1 ng) with animal cap assay. Animal cells that received injection were proceeded into RT-PCR (B) In the same condition as (A), we further confirmed that the relationship between Xmig6 and FGF-MAPK pathway. In the presence of U0126 (50 μm) or DMSO as a control, animal cells that received injection were proceed to RT-PCR analysis.
	Figure 6 Xmig6 depleted embryos showed muscle differentiation defect in a Xmyf 5-dependent manner. (A) Transcription inhibition by Xmig6 MO was confirmed by Western blotting. Protein expression was detected by a rabbit polyclonal anti-GFP antibody and anti-rabbit IgG-HRP antibody. (B) X-Gal-stained embryos that received injection of Xmig6 MO (16.8 ng) or Xmig6 RNA (0.15–0.25 ng) to the one dorsal cell of 4-cell-stage were used for WISH with a DIG-labelled Xmyf 5 anti-sense RNA probe. Vegetal view. Xmig6 MO introduced embryos caused the completely depleted Xmyf 5 expression, which was rescued by mig6 RNA co-injection. (C) In eFGF induced animal caps, Xmig6 MO (8.4–16.8 ng) specifically suppressed the expression of Xmyf 5 and slightly suppressed XmyoD. This suppression was rescued by Xmig6 RNA (0.2–0.25 ng). No changes were observed in Xbra and Xnot expression. (D) In later stages, embryos that received Xmig6 MO injection were stained by 12/101 antibody and showed muscle differentiation defect whereas control MO injected embryos showed well-developed somites. We counted the embryos that proceeded 12/101 antibody staining and six of 20embryos were failed to be stained by the antibody. Xmig6MO introduced embryos were no effect on notochord detected by MZ15 antibody.
	Figure 7 XGPCR4 is potentially a non-canonical FGF target gene. We investigated the role of XGPCR4 in gastrulation by injecting its mRNA or XGPCR4 MO into the two dorsal cells of 4-cell-stage embryos. (A) The gain-of-function effect showed dose-dependent A-P axis truncation or spina bifida but embryos given mRNA injection (n = 60) did not show any changes in the differentiation of the notochord and somites. (B) The specificity of XGPCR4 MO was confirmed by Western blotting, using venus-tagged constructs in animal caps. (C) In XGPCR4-MO injected embryos, neither the expression of mesodermal markers nor the formation of the organizer was affected compared with control MO-injected embryos. (D) Embryos given an injection of XGPCR4 MO (16.8 ng) (n = 45) showed incomplete closure of the dorsal blastopore and the spina bifida phenotype. However, no changes in the formation of the notochord and somites were observed. Using mut-GPCR4 (500 pg) construct, we partially rescued the spina bifida phenotype caused by the GPCR4 MO (n = 48). Mut-GPCR4 RNA itself (n = 45) exerted the same effects on embryos as intact GPCR4 RNA.
	Table 1 The results of sequence based BLAST search and spatiotemporal expression patterns of identified genes during gastrulation stage are described. The remarkable spatiotemporal expression patterns that we described were detected by whole-mount in situ hybridization. No putative protein means no homologue by BLAST searches. Access our database website, XDB, for more information
	arl5c (ADP-ribosylation factor-like 5C) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 11, vegetal view, dorsal up.
	arl5c (ADP-ribosylation factor-like 5C) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 26, lateral view, anterior left, dorsal up.
	errfi1 (ERBB receptor feedback inhibitor 1) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 26, lateral view, anterior left, dorsal up.
	Figure S1 The effects of SU5402 to Xenopus embryogenesis.
	Figure S2 (A) Novel genes that identified in this study showed FGF-dependent regulation. Embryos that received HAV?, or XFD (0.5 ng) injection with lineage tracer in one dorsal cell at 4-cell stage were proceed to WISH when the stage reached. (B) Novel genes are regulated by FGF-dependent manner and do not affect mesodermal genes and FGF component genes.
	Figure S3 It is shown that the amino acid sequences of human mig6, mouse mig6, and Xenopus mig6 and their phylogenic tree.