
Fig. 7. A model for the genetic pathway describing lens induction and development. The arrows indicate genetic interactions determined by these and previous analyses. Gray arrows indicate a possible rather than a demonstrated pathway element. At the apex of the hierarchy is the pre-placodal phase of Pax6 expression (Pax6pre-placode). It is understood that the later phase of Pax6 expression in the lens placode (Pax6placode) is dependent upon earlier function of Pax6. The genetic pathway described reflects this interaction. As we observe that Fgfr and Bmp7 signaling cooperate to maintain the placodal phase of Pax6 expression, it follows that their input to the pathway must be upstream of Pax6placode. Previous analysis has shown that the early phase of Pax6 expression is unaffected in the Bmp7-null mice and, thus, Fgfr and Bmp7 signaling must converge on the pathway downstream of Pax6pre-placode. Evidence that Foxe3 is downstream of Pax6placode and Fgfr and Bmp7 signaling includes (1) the reduced level of Foxe3 expression in Tfr7/Tfr7, Bmp7+/- embryos, (2) similar phenotypes in Tfr7/Tfr7, Bmp7+/- and dyl (Foxe3dyl/dyl) embryos, (3) the absence of Foxe3 expression in small eye embryos, and (4) the absence of Foxe3 expression in embryos carrying a targeted deletion of the Pax6 ectoderm enhancer (Dimanlig et al., 2001). Thus, Foxe3 expression must lie upstream of events such as lens lineage proliferation, vesicle closure and separation. While we currently do not understand the genetic relationship between Foxe3 and Sox2, it is clear from this and previous analyses performed that Sox2 lies downstream of Pax6placode. Because Sox2 (but not Pax6) expression is diminished in the Bmp4-null mice, Bmp4 signaling must contribute to the pathway between Pax6placode and Sox2.