Fig. 2

Immature T lymphocytes undergo cell death in mcm5^−/− embryos.

A–C Analysis of the expression of Rag2:DsRed, C-myb:GFP and Coro1a:GFP at 5 dpf. The expression of C-myb:GFP almost did not overlap with Rag2:DsRed expression (A, B, yellow column, n = 3, 3%), and most Rag2:DsRed-positive cells overlapped with cells labeled with Coro1a:GFP (A, C, yellow column, n = 3, 82.4%). D, E No significant difference was observed for Coro1a:GFP cells in siblings (d1, n = 24) and mcm5^−/− embryos (d2, n = 11) at 3.5 dpf (d1-d2, E; n = 11), but at 4 dpf, the number of Coro1a:GFP cells was greatly decreased (d3-d4, E; n = 13). This phenotype can be rescued by injecting mcm5 mRNA at the 1- to 4-cell stage (d5-d6, E; n = 17) or heat shock-inducing mcm5 expression at 3 dpf (d7-d8, E; n = 18). Here we counted the Coro1a:GFP labeled cells in each single slice, then added them together as the whole T cell number. F TUNEL-positive Coro1a:GFP cells were increased in mcm5^−/− embryos (n = 13) compared with their siblings (n = 13). G There were fewer H3P-positive Coro1a:GFP cells in mcm5^−/− embryos (n = 10). Here H3P is a mitotic marker Histone H3 phosphorylated at Serine 10. Scale bars, 40 μm. Statistical analyses of the expression of Coro1a:GFP are shown on the right. For (B, C, E–G), the data were presented as means ± SD; The P-values (t-test; two-tailed); NS, not significant. “***” P < 0.001.