Fig. S3

Timing of Specification and Domain Formation of pMNs and LFPs, Related to Figure 3 (A) Specified cells increase marker expression independently from Shh response. 2 representative tracks from a Cyclopamine treated embryo (Cell 1, Cyclopamine added at the start of imaging near 9hpf) and a control embryo (Cell 2, no Cyclopamine). Embryos are double transgenic for ptch2:Kaede and mnx1:gfp. A 405nm laser scan was performed before acquiring each time point to convert Kaede(green) to Kaede(red). 3 additional Cyclopamine tracks and 5 additional control tracks show similar behavior (data not shown).(B) Representative lateral 3D and cross-sectional views of Cyclopamine treated embryos. Early treatment does not fully remove GFP+ (mnx1:gfp) cells, a possible consequence of drug efficiency and/or early pMNs that can be specified in the absence of extracellular Shh (Lewis and Eisen, 2001; Chen et al., 2001). Cross-sections indicate that the specified cells still localize to correct positions. All scale bars: 10 μm. (C) Cross-sectional view of nkx2.2a:mgfp expression at different times. Before 13hpf potential mGFP+ cells could not be uncovered by imaging this transgenic line. At 13 and 14hpf GFP+ cells are found in variable locations and the left-right symmetric distribution of GFP+ cells in the LFP domain is not established. After 15hpf most GFP+ cells line up into the LFP domain on both sides of the triangular MFP cell, this position is indicative of LFP fate after 16hpf.(D) Nkx2.2a:mgfp+ cell distribution over time. Each mark represents the location of one mGFP+ cell at the corresponding time point. Note the narrowing of distribution from 14 to 16hpf and the population average (±s.d.) moving closer to the MFP/notochord.