BMC Dev Biol
September 29, 2006;
Genomic analysis of Xenopus organizer function.
Studies of the Xenopus organizer
have laid the foundation for our understanding of the conserved signaling pathways that pattern vertebrate embryos during gastrulation. The two primary activities of the organizer
, BMP and Wnt inhibition, can regulate a spectrum of genes that pattern essentially all aspects of the embryo
during gastrulation. As our knowledge of organizer
signaling grows, it is imperative that we begin knitting together our gene-level knowledge into genome-level signaling models. The goal of this paper was to identify complete lists of genes regulated by different aspects of organizer
signaling, thereby providing a deeper understanding of the genomic mechanisms that underlie these complex and fundamental signaling events. To this end, we ectopically overexpress Noggin
-1, inhibitors of the BMP and Wnt pathways, respectively, within ventral
tissues. After isolating embryonic ventral
halves at early and late gastrulation, we analyze the transcriptional response to these molecules within the generated ectopic organizers using oligonucleotide microarrays. An efficient statistical analysis scheme, combined with a new Gene Ontology biological process annotation of the Xenopus genome, allows reliable and faithful clustering of molecules based upon their roles during gastrulation. From this data, we identify new organizer
-related expression patterns for 19 genes. Moreover, our data sub-divides organizer
genes into separate head
organizing groups, which each show distinct responses to Noggin
-1 activity during gastrulation. Our data provides a genomic view of the cohorts of genes that respond to Noggin
-1 activity, allowing us to separate the role of each in organizer
function. These patterns demonstrate a model where BMP inhibition plays a largely inductive role during early developmental stages, thereby initiating the suites of genes needed to pattern dorsal tissues. Meanwhile, Wnt inhibition acts later during gastrulation, and is essential for maintenance of organizer
gene expression throughout gastrulation, a role which may depend on its ability to block the expression of a host of ventral
, and lateral
BMC Dev Biol
[+] show captions
Figure 1. Generating tissue samples with different aspects of organizer activity. (A) shows an overview of injection and embryo sorting procedure used to produce samples for microarray analysis. (B) shows the four injection mixtures below their respective tailbud phenotypes. Embryos ventrally injected with these mixtures were bisected at either stg. 10 or stg. 11.5 to produce the tissue conditions in (C).
Figure 7. Relaxed statistical criteria select for additional genes with organizer-related expression patterns. The cluster results for the second set of RP criteria, which required correlated expression in the Nog+Dkk and Dor conditions, identifies additional unknown genes with organizer-related expression patterns. The hierarchical tree is on the far left, followed by the clustergram, and then the summary of the RP results at 10% FDR. Colors and columns are same as described in the Figure 3A. Black ticks between the cluster tree and the clustergram mark every tenth gene, allowing referencing to 3 for the gene identities. Genes in this list that lacked described gastrula stage expression patterns were analyzed by whole mount in situ hybridization. The top of the cluster contains genes that were repressed in the Nog+Dkk and Dor conditions; unknown genes within this group are excluded from organizer tissues (magenta box). The bottom of the cluster contains genes that were activated in the Nog+Dkk and Dor conditions; unknown genes within this group are enriched in organizer tissues (orange box). Each tested gene is labeled with its name and the Affymetrix probe set number. Genes marked "no pattern" showed no staining, or a non-specific staining pattern that was similar to sense controls. Genes marked "no gastrula pattern," showed no pattern during gastrula stages, but did show specific patterns at later stages that are not shown here. Each photo is labeled with the developmental stage of the embryo in the bottom left corner, and the orientation in the bottom right corner. veg: vegetal view, dorsal faces up. dor: dorsal view, anterior faces up. gene name was assigned by protein sequence similarity using Homologene.
Figure 8. Noggin and Dkk-1 regulate the expression of newly identified genes. 4-cell embryos were ventrally injected with the noggin and/or dkk-1 concentrations described in Figure 1, and tested by in situ hybridization for patterns of ectopic induction or repression. In each case the observed in situ patterns confirm the microarray patterns (A-D) LRIG2 shows ectopic ventral expression in both the Noggin and the Noggin+Dkk-1 overexpressing embryos, but not in the Dkk-1 embryos. (E-H) ARHGEF3 shows ectopic ventral expression only in the Noggin+Dkk-1 overexpressing embryos. (I-L) HES6 shows ectopic ventral repression in the Noggin+Dkk-1 and the Dkk-1 overexpressing embryos. (M-P) Frzb3 expression is ectopically repressed by Noggin or Noggin+Dkk-1 overexpression. (Q-T) Xl.3529.1.A1 expression is ectopically repressed by Noggin+Dkk-1 and Dkk-1 overexpression. (U-X) Gadd45g is ectopically induced only in the Noggin+Dkk-1 condition. All embryos are between stages 10.5 and 11 and are shown in vegetal view with dorsal side facing up.
Figure 9. The head and trunk sub-cluster genes show distinct responses to organizer signaling. Three genes were selected from each sub-cluster, and tested by in situ hybridization at stage 10.5 (top row) and 11.5 (bottom row), in embryos ventrally injected with noggin or noggin+dkk-1. Black arrowheads mark ectopic staining. Xlim-1 (A-F) and otx2 (M-R) expression are similarly induced at stage 10.5 by Noggin and Noggin+Dkk-1, but by stage 11.5 Noggin+Dkk-1 induction is clearly much stronger and more widespread. For Xlim-1, ectopic expression induced by Noggin+Dkk-1 is observed migrating away from the blastopore lip region, but never for Noggin alone. (M-R) Frzb-1 expression is ectopically induced only by the combination of Noggin+Dkk-1, not Noggin alone, and neither can sustain expression into late gastrulation. For the three trunk genes, FoxD5b (XFD-12') (S-X), FoxA4a (pintallavis) (Y-D'), and Xsox-2 (E'-J') ectopic induction is similar in both intensity and spread in Noggin and Noggin+Dkk-1 overexpressing embryos at stage 10.5 and 11.5. A-L and S-J' vegetal view; M-R animal view. Dorsal faces up in all pictures. (K') Otx2 expression was assayed by real-time RT-PCR, in stage 11.5 ventral tissues injected with the same mixtures used to create the microarray samples. Weak, but significant (p = 0.043 by one-sided t-test), induction in Dkk-1 overexpressing tissues are seen compared to ventral, supporting the weak Dkk-1 inductions seen on the microarray. Error bars show the standard error calculated from two biological replicates.
Figure 6. Clusters 3 and 4 faithfully predict expression patterns for unknown genes. Genes found in clusters 3 and 4 that lacked described gastrula stage expression patterns were analyzed by whole mount in situ hybridization. Unknown genes from cluster 3 that showed a specific pattern are enriched in organizer tissues (orange box), and unknown genes from cluster 4 that showed a specific pattern are excluded from organizer tissues (magenta box). Each tested gene is labeled with its name and the Affymetrix probe set number. Genes marked "no pattern" showed no staining, or a non-specific staining pattern that was similar to sense controls. Genes marked "no gastrula pattern," showed no pattern during gastrula stages, but did show specific pat- terns at later stages that are not shown here. Each photo is labeled with the developmental stage of the embryo in the bottom left corner, and the orientation in the bottom right corner. veg: vegetal view, dorsal faces up. dor: dorsal view, anterior faces up.
Figure 2. Expression variation among the samples. (A) and (B) show scatter plot comparisons of the mean Nog+Dkk or Dor stg. 10 log2 expression values vs the mean stg. 10 Ven log2expression values. Probe sets measuring two known organizer genes, otx2 (green) and gsc (red), are labeled within the plots (otx2 probe sets: Xl.1268.1.S1_at, Xl.3004.1.A1_at, Xl.11672.1.A1_at, and XlAffx.1.11.S1_at; gsc probe sets: Xl.801.1.A1_at, Xl.801.1.S1_at, and Xl.801.1.S1_s_at). (C) R-square regression value summaries for each experimental condition compared to the stage-matched ventral condition. (D) Scatter plot comparison of the mean Nog and Ven clutch-1 log2 expression values vs. the mean Nog and Ven clutch-2 log2 expression values. Note that the clutch variation clearly exceeds differences between the experimental conditions (A-C). Contributing to this variation, many genes show a shift in the average expression between the two clutches, but retain a similar pattern in response to the experimental conditions. As an example, log2 expression values for otx5-A (Xl.3452.1.A1_at) are shown in (E). After standardizing the average log2 expression of both clutches, on a gene-by-gene basis, the comparison from (D) was repeated in (F), eliminating most clutch variation.
Figure 3. Clustering genes regulated by organizer signaling. (A) Key to the hierarchical cluster format used throughout the paper. The clustergram shows the standardized expression intensity for the ten experimental conditions, after replicates have been averaged. To the right of the clustergram, the RP method results at 10% FDR are summarized in four columns representing the comparisons of Nog, Nog+Dkk, Dkk, or Dor to Ven. The colors found in the row for each gene represent the tests passed by that gene. Two colors in one column indicate that a gene passed the column's test at both stages. (B) Hierarchical cluster of all the selected genes. The far right shows the hierarchical cluster tree, followed by the clustergram, then RP results. Black ticks between the cluster tree and the clustergram mark every tenth gene, allowing referencing to Additional file 2 for the gene identities. The list break into four main clusters, labeled with red, yellow, orange, and magenta bars. Expanded views of clusters 3 and 4 can be found in Figures 4 and 5, respectively.
Figure 4. Cluster 3 is enriched for genes involved in organizer function. This figure shows an enlarged view of all of the genes in cluster 3, from Figure 3. Each row is annotated with the probe set number and matching gene name. Genes with names in blue have a described role in organizer function. Genes with names in red have no described function during gastrula stage development. Genes with names in green have a published role or expression pattern that is not organizer related. † gene name was assigned by protein sequence homology using the NCBI Homologene database.
Figure 5. Cluster 4 is enriched for genes involved in ventral, lateral, and posterior patterning. This figure shows an enlarged view of all the genes in cluster 4, from Figure 3. Each row is annotated with the probe set number and matching gene name. Genes with names in blue have a described role or specific expression pattern in ventral, lateral, or posterior tissues. Genes with names in red have no described function during gastrula stage development. Genes with names green have a published role or expression pattern that is not ventral, lateral, or posterior related. † gene name was assigned by protein sequence homology using the Homologene database.