Fig. 4

The methionine cycle is involved in the response to YD and its involvement is conserved. a Schematic representation of the methionine cycle. Gene symbols are followed by information about temporal DE. “Y” indicates up-regulation in YD; “S” indicates down-regulation in YD; a dash (−) indicates no DE. From left to right: 8, 24, 32, 48 hpf. See example in top left corner. Colors highlight whether genes are up-regulated in YD only (red), down-regulated in YD only (green) or up- and down regulated at different time-points (grey). *Two different DE mat genes are identified: mat1a at 8 hpf and mat2ab at 48 hpf. **trdmt1 is the only DE dnmt homologue in this study. Since it is a RNA methyl transferase, for clarity, only RNA is depicted as a substrate for methylation. ***Four DE creatine kinase genes are combined in this schematic figure. b DE levels of DE genes from (a), with identical color scheme. c DE of three genes confirmed by qRT PCR, RNA obtained from pooled embryos. d DE of mat2ab confirmed by qRT PCR, RNA obtained from individual embryos. e DE of bhmt in response to YD treatment (25 % yolk reduction), NaCl exposure and heatshock, data obtained from qRT PCR. Values shown are Cp (Treatment) – Cp(Control), normalized against Cp(actb2). f Methionine cycle genes that overlap with DE genes (red line) and identical-sized random samples from all genes in this study (open bars). g Seven out of 12 DE genes homologues that correspond to methionine cycle genes from this study overlap to DE genes in the liver of maternally starved fetal mice (red line). Open bars show the overlap with methionine cycle genes and 12 randomly sampled genes from the complete mouse liver gene set. Asterisks denote significant P-values according to the Student’s T-test in (c) and (d). In (f) and (g), the P-value represents the fraction of randomly sampled overlaps that is equal to or larger than the overlap with DE genes after 1000 permutations