January 1, 2008;
Convergence of a head-field selector Otx2 and Notch signaling: a mechanism for lens specification.
Xenopus is ideal for systematic decoding of cis-regulatory networks because its evolutionary position among vertebrates allows one to combine comparative genomics with efficient transgenic technology in one system. Here, we have identified and analyzed the major enhancer of FoxE3
), a gene essential for lens
formation that is activated in the presumptive lens ectoderm
(PLE) when commitment to the lens
fate occurs. Deletion and mutation analyses of the enhancer based on comparison of Xenopus and mammalian sequences and in vitro and in vivo binding assays identified two essential transcriptional regulators: Otx2
, a homeodomain protein expressed broadly in head ectoderm
including the PLE, and Su(H), a nuclear signal transducer of Notch
signaling. A Notch
, is expressed in the optic vesicle
adjacent to the PLE, and inhibition of its activity led to loss, or severe reduction, of FoxE3
expression followed by failure of placode formation. Ectopic activation of Notch
signaling induced FoxE3
expression within head ectoderm
, and additional misexpression of Otx2
in trunk ectoderm
extended the Notch
expression posteriorly. These data provide the first direct evidence of the involvement of Notch
signaling in lens
induction. The obligate integration of inputs of a field-selector (Otx2
) and localized signaling (Notch
) within target cis-regulatory elements might be a general mechanism of organ-field specification in vertebrates (as it is in Drosophila). This concept is also consistent with classical embryological studies of many organ systems involving a ;multiple-step induction''.
[+] show captions
Fig. 1. In vivo deletion analysis identifies a 901 bp enhancer that directs PLE-specific expression of FoxE3. (A) FoxE3 expression in X. tropicalis embryos (stage 23) detected by in situ hybridization. (B-G) GFP expression detected by in situ hybridization in representative transgenic embryos (stages 22-24) generated with the reporter constructs shown to the left. White and black arrows in A-G indicate the PLE. The arrowhead in B indicates ectopic GFP expression in the presumptive oral ectoderm. Numbers of embryos with GFP expression in the PLE and the total number of normally (or near normally) developing embryos injected with each construct are indicated to the right, along with the percentage of GFP-positive cases. The 901 bp element necessary for PLE-specific expression is boxed with a dotted red line. *The expression in D was positive in the PLE but very spotty and broad, as shown in the left-hand panel.
Fig. 4. Comparative expression analysis of Notch signaling components and FoxE3 in X. laevis embryos, showing that Notch2 and FoxE3 are expressed in PLE, while Delta1 and Delta2 are expressed in presumptive retina. Expression of Notch2 (A-B′), Delta1 (C-E′), Delta2 (F-H′), and FoxE3 (I-K′) was detected by in situ hybridization from neural plate stages to early tailbud stages. Regions circled with black dotted lines in C and F are the approximate presumptive retina fields, where neither Delta1 nor Delta2 is expressed. Arrows in B,B′, D-E′, G-H′ and J-K′ indicate expression of Notch2, Delta1, Delta2 and FoxE3, respectively. The black arrowhead in C indicates Delta1 expression in the anterior neural ridge, and white arrowheads in A and I indicate Notch2 and FoxE3 expression in the pre-placodal ectoderm, respectively. White lines in E,H,K indicate the planes of transverse eye sections shown in E′,H′,K′.
Fig. 5. Effects of manipulation of Notch signaling on FoxE3 expression and subsequent lens placode formation. (A,B) Frontal view of Xenopus embryos injected with mRNA encoding Delta2Tr (500 pg), fixed at stage 23, and subjected to lacZ staining (magenta) and in situ hybridization with FoxE3 or Rx probe (purple or deep purple staining). White and black arrowheads in A-H indicate in situ hybridization signals on injected and uninjected sides of embryos, respectively. (C,D,F,G) The injected and uninjected sides of embryos injected with Delta2Tr mRNA, fixed at stages 29/30, and hybridized with γ1-crystallin or Rx probe. (H) A transverse head section of the embryo shown in F,G. (E) The injected side of an embryo injected with both Delta2Tr mRNA (500 pg) and wild-type Delta2 mRNA (500 pg), fixed at stage 29, and hybridized with γ1-crystallin probe. (I) Summary of Delta2Tr mRNA injection experiments. GFP mRNA (1000 pg) was injected as a control. (J,K) The injected and uninjected sides, respectively, of an embryo injected with GR-Su(H)DBM mRNA (1000 pg) and induced with Dex. Arrows in J-O indicate endogenous FoxE3 expression in the PLE. (L) The injected side of an embryo injected with GR-Su(H)DBM but not induced with Dex. (M,N) The injected and uninjected sides, respectively, of an embryo injected with GR-Su(H)VP16 mRNA (1000 pg) and induced with Dex. Black and white arrowheads in M indicate ectopic FoxE3 expression in the ectoderm overlying the anterior brain and that surrounding the cement gland, respectively. (O) The injected side of an embryo injected with GR-Su(H)VP16 but not induced with Dex.
Fig. 6. Otx2 confers the ability to activate FoxE3 in response to Notch signaling. (A-C′) Expression of Otx2 in the anterior ectoderm detected by in situ hybridization. At the neural plate stages, Otx2 is expressed in the anterior ectoderm including the presumptive lens-fields, which are circled with white dotted lines in A. Arrows in B indicate broad expression in the ectoderm that overlies the optic vesicles (ov) and surrounds the cement gland primordium (cg). Arrows in C indicate the border of ectodermal Otx2 expression. White lines in C indicate the planes of transverse sections shown in C′ and C′. Arrows in C′ and C′, respectively, indicate expression in the PLE overlying the optic vesicle and in the ectoderm overlying the forebrain. (D-H) Notch-Otx2 combination activates FoxE3 in the trunk ectoderm. Xenopus embryos injected with the mRNAs indicated in each panel were induced with Dex, and then subjected to lacZ staining and in situ hybridization with FoxE3 probe. Ectopic FoxE3 expression was not detected in embryos injected with mRNA encoding GR-Su(H)VP16 (1000 pg) (D), Otx2-GR (250 pg) (E), or NICD (1000 pg) (G), whereas it was detected in embryos injected with both GR-Su(H)VP16 (750 pg) and Otx2-GR (250 pg) (F), or both NICD (750 pg) and Otx2-GR (250 pg) (H). Arrowheads in F and H indicate ectopic FoxE3 expression. Arrows indicate endogenous FoxE3 expression in the PLE. The white line in F indicates the plane of the transverse section shown in the inset. Black arrowheads in the inset indicate the overlap of FoxE3 expression and nuclear lacZ staining in the ectodermal cells, and white arrowheads indicate cells in the underlying mesoderm layer showing nuclear lacZ staining but no FoxE3 expression. (I,J) The injected and uninjected sides, respectively, of an embryo injected with mRNA encoding GR-Otx2-En (250 pg), induced with Dex from stage 18, and then subjected to lacZ staining and in situ hybridization with FoxE3 probe at stage 22. Arrows indicate the PLE. (K) Transgenic experiments using Otx-Su(H) reporter constructs. Numbers of embryos with GFP expression in the PLE and the total number of normally (or near normally) developing embryos injected with the constructs shown on the left are indicated on the right-hand side with percentages of the GFP-positive cases. Gray and red boxes indicate Otx- and Su(H)-binding motifs, respectively, in the constructs, and crosses indicate base-substitution mutations introduced there. (L,L′) A representative transgenic embryo generated with Otx-Su(H)-βGFP. Black and white arrows indicate GFP expression in the eye and spinal cord, respectively. The white line indicates the plane of the transverse eye section shown in L′. Black and white arrowheads indicate GFP expression in the PLE and optic vesicle, respectively.
Fig. 3. Mapping of regulatory motifs essential for PLE-specific activity of the FoxE3 enhancer. (A-C) Representative transgenic embryos (stages 22-24) generated with the GFP reporter constructs shown on the left. Black arrows indicate the PLE. Numbers of embryos with GFP expression in the PLE and the total number of normally (or near normally) developing embryos injected with each construct are indicated on the right-hand side with percentages of the GFP-positive cases. The white line in A indicates the plane of the transverse section shown in the inset. The black arrow in the inset indicates GFP expression in the PLE overlying the optic vesicle. The embryo shown in C was generated by co-transgenesis, i.e. co-injection of the 462 bp enhancer of Xenopus FoxE3 amplified by PCR along with the βGFP cassette. (D) Identification of transcription factor-binding motifs essential for PLE-specific expression by mutation analysis. wt is the construct used in Fig. 3A (Xt462-βGFP). mt1-mt9 were generated from wt/Xt462-βGFP by introducing a base-substitution mutation (cross) into each of the conserved transcription factor-binding motifs. The bar chart shows the percentage of the embryos that showed GFP expression in the PLE among total developed embryos injected with the constructs shown on the left. Actual numbers of GFP-positive cases and total numbers of scored embryos are indicated in parentheses.