Figure 1.
Diagram of the blastemal time course experiment.
RNA was harvested from tissue samples and subjected to deep sequencing on the Illumina GA II platform (see also Figure S1).
Figure 2.
Expression patterns of RNA-seq and qPCR along the time course.
Plots of the log2 ratios of TPMs (transcripts per million) of each time point relative to the time zero control for RNA-seq data (blue), and the log2 ratios of the qPCR expression measure (red) relative to the zero hour control. The R2 Pearson correlation across the time course is shown for each gene as well. The expression patterns are similar between the RNA-seq and qPCR data (average Pearson correlation for the four genes shown = 0.839, average Pearson correlation for all 19 genes = 0.741). See Figure S2 for the 15 additional genes not shown in Figure 2, and Table S2 for all Log2 ratios of RNA-seq and qPCR data. See Materials and Methods for an explanation of determining the ratios and for the Pearson calculation.
Figure 3.
Two-way clustering of all differentially expressed (DE, FDR<0.05) transcription factors (TFs).
Differential expression is relative to the zero hour control. Upper right: an early upregulated cluster of TFs dominated by oncogenes (ATF3, KLF4, KLF2, JUN3, EGR1, NR4A2, and FOS are all oncogenes). Oncogenes are highlighted with red text. Lower right: a cluster of TFs upregulated later in the time course including genes involved in limb development such as HOXD10 and HOXD11 (highlighted in blue text). Clustering and viewing was done with Mev 4.8.1 The distance measure used was the Pearson uncentered correlation with average linkage.
Figure 4.
Bootstrapping of the sample clusters from Figure 3.
To assess the uncertainty in hierarchical cluster analysis over samples, bootstrap resampling (10,000 iterations) was applied via the R package Pvclust [26]. The uncentered Pearson correlation is used as the distance metric with average linkage. The numbers above each edge show the probability of nodes below that edge occurring as a cluster in resampled trees, via ordinary bootstrap resampling (BP, green) or multiscale bootstrap resampling (AU, red). See Materials and Methods for details on bootstrapping.
Figure 5.
Enrichment of oncogenes during the early time points.
All upregulated DE genes at each time point (when compared to the zero hour control) were interrogated for being oncogenes by their presence in the Memorial Sloan Kettering Cancer Center Oncogene set (http://cbio.mskcc.org/CancerGenes/Select.action) and then performing a Fisher's exact test to assess significance. The enrichment of oncogenes during the first day is highly statistically significant (P-value<10−5). Late in the time course, oncogenes are not significantly enriched in the upregulated DE gene sets.
Figure 6.
Gene ontology (GO) analysis of the time course.
GO enrichment of molecular function (MF) and biological process (BP) level 5 GO terms during the time course for all DE genes (relative to the zero hour control). The value displayed is the −log10(GO term FDR). GO terms with an FDR<0.01 are shown. The GO terms were clustered by their enrichment patterns using the Pearson correlation as the distance measure with average linkage. During the first day enriched GO terms include GO terms representing immune response, chemotaxis, regulation of leukocytes, blood vessel development, and angiogenesis. In the middle of the time course (3 d–14 d), GO terms for tissue development, limb morphogenesis, bone development, and forebrain development are enriched. An ectoderm development GO term is enriched at 28 days. (Detailed GO enrichment information for upregulated axolotl genes is available at www.axolomics.org). (See also Table S4 for the −log10(GO term FDR) values.)
Figure 7.
Expression of embryonic and adult stem cell genes in ES, iPS, FS, and blastemas.
All samples are ratioed to the time zero axolotl juvenile time course control. While ES and iPS cells show upregulation of a variety of stem cell genes including the key ES transcription factors POU5F1 and NANOG, the axolotl blastemas do not show upregulation of POU5F1, SOX2, and NANOG, but do show upregulation of adult stem cell markers (such as SALL4 and HMGA2). Note the burst of oncogenes (KLF4, MYC) early in the time course and the upregulation of adult stem cell genes later in the time course.
Figure 8.
Expression of SALL1, SALL3, and SALL4 in juvenile blastemas.
SALL4 is upregulated earlier than SALL1 or SALL3. “TPM” is the Transcripts Per Million expression measure.