HA-cebpa mRNA rescued cebpa loss-of-function phenotype. (A) X tropicalis embryos injected with cebpaATG or control MO at one-cell stage (i) were allowed to develop to the 32-cell stage (ii), and then coinjected with HA-cebpa mRNA or eGFP mRNA and X-gal tracer mRNA into the 2 C1 blastomeres, which are fated to give rise to the aVBI. The embryos were fixed at stage 23 and stained for X-gal (iii-v) to assess the migration of the injected cells. (B) The extent of migration of X-gal⁺ cells was scored by placing a grid over the embryos and determining the percentage of the total area of each embryo that contained X-gal⁺ cells. Error bars represent SEM of 10 embryos. Statistical ANOVA was done using the SPSS software package. Compared with the control MO-injected embryos, the cebpaATG-injected morphants showed a significantly reduced extent of migration (P < .001), which was rescued by HA-cebpa mRNA to a level similar to that seen in the control embryos (P = .678). (C) Real-time PCR analysis on embryos at stage 23 injected with cebpaATG or control MO, coinjected with HA-cebpa mRNA or eGFP mRNA. Primitive myeloid markers mpo, lcp, mmp7, and spi1 were significantly reduced by cebpaATG MO (i-iv) and were rescued to normal or even higher levels by HA-cebpa. Expression levels were normalized relative to rpl8. Error bars represent SEM of 4 independent experiments. Statistical ANOVA was done using the SPSS software package. The experimental conditions that showed a significant difference in expression level relative to the control MO + gfp mRNA-injected embryos are labeled as *P < .05, **P < .01, or ***P < .001.