PMC

Expression profiling of all 20 soft-shell turtle Wnt genes shows Wnt5a expression in the carapacial ridge. (a) Whole-mount in situ hybridization (ISH) was performed for all Wnt genes in the genome. Wnt5a (red outline) is specifically expressed in the carapacial ridge (red arrowheads), whereas most of the other genes show similar expression patterns to their known mouse and chicken counterparts. Scale bars, 0.5 mm. (b) Soft-shell turtle embryo at stage TK14. (c) Carapacial ridge expression of Wnt5a confirmed by ISH on a 6-μm paraffin transverse section (the sectioned level for ISH is indicated by the dashed line in b). The arrowhead indicates the carapacial ridge; the arrow indicates the body wall. NT, neural tube; NC, notochord. Scale bars in b and c, 0.5 mm.

Turtle phylogeny and divergence time estimation by molecular clock analysis. (a) Two genome-sequenced turtles, the soft-shell turtle (P. sinensis) and the green sea turtle (C. mydas). (b) Estimated divergence times of 12 vertebrate species calculated using the first and second codon positions of 1,113 single-copy coding genes ( and ). Tree topology is supported by 100% bootstrap values and further statistical assessment ( and –). The black ellipses on the nodes indicate the 95% credibility intervals of the estimated posterior distributions of the divergence times. The red circles indicate the fossil calibration times used for setting the upper and lower bounds of the estimates. MYA, million years ago.

The molecular divergence of turtle and chicken embryos follows the hourglass model with a maximally conserved vertebrate phylotypic period. (a) Distances of whole-embryo GXPs (from depth-controlled, TMM (trimmed mean of M values)-normalized data) for 11,602 orthologs in selected developmental stages of the soft-shell turtle and chicken (). Error bars, s.d. ANOVA P value under heteroscedasticity = 7 × 10⁻⁷. (b) The hypothetical model (nested hourglass) in which both an hourglass-like divergence and a recapitulation-like relationship between ontogeny and phylogeny can be justified. The model infers that the most conserved developmental stage changes depending on how distantly related the species are that are being compared. Comparisons within vertebrate embryos gives a vertebrate phylotype (blue arrow), and comparisons within amniotes gives an amniote-type stage (red arrow) that emerges later than the vertebrate phylotype stage. (c) An all-to-all comparison of turtle and chicken GXP distances (total Manhattan) indicates that the highest similarity occurs between turtle stage TK11 and chicken stage HH16 embryos (see for statistical assessment). HH16 is the stage previously identified as the vertebrate phylotypic period²¹, which does not coincide with the model in b. Error bars, s.d. (d) Morphological appearance of the soft-shell turtle (stage TK11) and chicken (stage HH16) embryos that showed the highest GXP similarity (). Scale bars, 1 mm.

Extensive expansion of olfactory receptor genes in turtles. (a) A neighbor-joining tree constructed with all the intact group α olfactory receptors from eight vertebrate species (soft-shell turtle, green sea turtle, chicken, zebra finch, anole lizard, human, dog and Western clawed frog), with the group β olfactory receptors as the outgroup. Bootstrap values (from 500 resamplings) are shown on the branches. The scale bar represents the number of amino-acid substitutions per site. (b) Expansion of group α olfactory receptor genes in the evolution of tetrapods. Numbers in boxes indicate the current number of intact group α olfactory receptor genes in each species. The number of group α olfactory receptor genes in an ancestral species is shown in an ellipse at each node, and the numbers of gene gains and losses are shown on each branch with plus and minus signs, respectively. For divergence times, we used the median values obtained from TimeTree³⁰. Note that the majority of the expansion of the group α olfactory receptor genes occurred independently in each turtle lineage. The same color code for species is used in a,b. (c,d) Genomic clusters of olfactory receptor genes in scaffolds 55 and 145 of the soft-shell turtle genome. Vertical red bars represent class I (c) and class II (d) olfactory receptor genes. Bars above and below the horizontal line indicate opposite directions of transcription. Long bars depict intact olfactory receptor genes, whereas short bars depict olfactory receptor pseudogenes or gene fragments.

Molecular characteristics of turtle embryogenesis during and after the phylotypic period. (a) Shared expression of developmental genes () in the phylotypic stages of turtle and chicken embryos. The log₁₀-transformed relative expression levels of 11,602 orthologous genes from mapped-10M reads (data set based on randomly selected 10M tags mapped to the genome; Online Methods), with TMM-normalized data, were graphed on a scatterplot. Essentially the same results were obtained from other data sets (all-read-data, RPKM (reads per kilobase per million mapped reads) and TMM normalizations). The results of a statistical test to determine the groups of genes that have more similar expression can be found in the . (b) Genes that showed a statistically significant increase in their expression level after the phylotypic period (IAP; Online Methods). Each line represents the mean expression level of each IAP (increased expression after the phylotype) gene calculated, with two biological replications, for each stage. The names of the genes with the top three highest expression levels in TK23 are shown. Consequently, 233 turtle IAP genes were found. See for the expression pattern of the chicken orthologs of the turtle IAP genes. (c) Over-represented GO annotations for 233 turtle IAP genes with read depth–controlled, TMM-normalized data. Only the results corroborated by all of the data sets (mapped-10M reads (Online Methods), all reads, RPKM normalization and TMM normalization) are shown. Shown are P values calculated by Fisher’s exact test. (d) High numbers of tissue-specific miRNAs were identified (also in the carapacial ridge) in the embryo after the phylotypic period.