Fig. S2

Comparison of FP expression in Tg(trβ2:XFP) lines with trβ2 mRNA expression across ages. (A) In situ hybridization (ISH) for trβ2 mRNA and expression of trβ2:MYFP enhanced by immunostaining with anti-GFP at 32 hpf, showing that both message and FP expression driven by the exogenous promoter are in progenitor cells at this age. D, dorsal; N, nasal; T, temporal; V, ventral. (B) Simultaneous labeling for trβ2 mRNA and tdTomato expression driven by the exogenous trβ2 promoter (anti-DsRed used to recover tdTomato signal). (Left) View of the excised retina from the back. Arrowhead indicates mitotic profiles. (Center) Magnified image of the mitotic profiles. (Right) Orthogonal view of the boxed region in the left image. Arrows indicate progenitor cells that colabel for message and tdTomato. Message appears to be reduced in cells at the apical surface (arrowheads). (C) In situ hybridization (Left) and trβ2:tdTomato expression (Right) at 4 dpf showing that both message and FP expression becomes restricted to cone photoreceptors. (D) En face view of 4-dpf photoreceptor layer showing both trβ2 mRNA and tdTomato expression driven by the exogenous trβ2 promoter. trβ2 mRNA was not observed in mitotic cells labeled with trβ2:tdTomato, indicating that trβ2 mRNA expression is likely to be regulated by the cell cycle, as observed previously in chick retina (10). We noticed that not all trβ2:tdTomato-expressing cells contained trβ2 mRNA signal (this was consistent across experiments). Because tdTomato is relatively stable, its fluorescence may not faithfully reflect temporal changes in trβ2 mRNA or protein expression. Indeed, in mice, TRβ2 protein distribution across the photoreceptor layer varies from mouse to mouse at around birth, in contrast to its more uniform distribution before and after this age (11). However, every cone in the 4-dpf zebrafish that expressed trβ2 message was labeled by trβ2:tdTomato.