Figure 1.
Workflow from retinectomy to de novo assembly.
A. Retinectomy. B. Sample collection, mRNA-seq and de novo assembly. Stage E-0: The RPE immediately after retinectomy. Stage E-1: Almost all RPE cells that have lost their epithelial characteristics and formed aggregates have entered the S-phase of the cell-cycle. Stage E-2: Partially depigmented cells are segregated into two rudiment layers (pro-NR and pro-RPE), which give rise to a new neural retina and the RPE layer itself. Under the current experimental conditions, regenerating retinas at Stage E-1 and E-2 are obtained at day-10 and -14 po [8].
Figure 2.
Comparisons between transcriptomes deduced by different assemblers.
Scatter plot graphs show the length distribution of IS-transcripts deduced by Trinity (blue), Trans-ABySS (magenta) and Velvet-Oases (green). As suggested by the total number and N50 length of IS-transcripts, Velvet-Oases tended to give long IS-transcripts at the expense of their total number, while Trans-ABySS deduces a large number of short IS-transcripts. Trinity was intermediate with respect to these parameters. The Venn diagram shows how much one assembler covers the IS-transcripts deduced by the other assemblers. The number of IS-transcripts is given in parentheses. Trinity covered almost all the IS-transcripts deduced by either Velvet-Oases or Trans-ABySS.
Figure 3.
Comparisons with other newt transcriptomes.
Black circles: IS-transcripts of this study. Blue circles: cDNAs and EST-contigs reported in C. pyrrhogaster. Red circles: cDNAs and EST-contigs reported in P. waltl. Green circles: cDNAs, EST-contigs and IS-transcripts reported in N. viridescens. The values in each circle (written in corresponding color) mean the number and ratio of IS-transcripts, cDNAs or EST-contigs.
Figure 4.
A. Summary of annotation. B. E-value-, similarity (% identity)- and species-distribution in NR annotation. Species-distribution indicates that many of the C. pyrrhogaster IS-transcripts are close to genes of amniotes as well as those of amphibians such as Xenopus (Silurana) and Xenopus laevis, rather than fishes (e.g., Danio rerio, 2.9%; Oryzias latipes, 1.9%; these are included in ‘other’). Interestingly, within amphibians, the newt seems to adhere to X. (Silurana) rather than to X. laevis.
Table 1.
GO classification.
Figure 5.
Blastx predicted CDSs in 71,511 IS-transcripts (for the length distribution, see the graph ‘Blast’; for protein/peptide sequences, see Table S5) while ESTscan predicted CDSs in 22,773 IS-transcripts (for the length distribution, see the graph ‘ESTScan’; for protein/peptide sequences, see Table S6).
Figure 6.
A. Workflow of sample preparations for qPCR. RPE-choroid tissues harvested from the right eyes (intact eye) of 5 retinectomized animals at day-10 or day-14 were used for the normal RPE, i.e., the day-0 (Stage E-0) sample. RPE-derived cells harvested together with the choroid from the left eyes (retinectomized eye) of the same animals were used for the day-10 (Stage E-1) or day-14 (Stage E-2) sample. 3 different day samples (day-0, -10 and -14) were randomly grouped as one set of samples for qPCR. For more details, see Methods. B. qPCR analysis. 20 selected genes which have been suggested or inferred to be regulated in an early phase of retinal regeneration were found in the current in silico transcriptomes. Changes in their relative expression level until day-14 po were examined. Results represent means and SE. The number in parenthesis indicates the number of independent sample sets (see Methods) except for Pax6 and Chx10, whose expression was detected in 2 of 8 sample sets and 4 of 5 sample sets, respectively at day-14 po only. Statistical significance based on Sheffe’s test following the Friedman test (*: p<0.05; **: p<0.01), except for Pax6 and Chx10.