|
igure 1.
Xfz5 Is Expressed in the Developing Retina and Is Required for Normal Eye Development
Xfz5 is expressed in the early optic vesicle starting at stage 17 (A) and continues to be strongly expressed in the eye at stage 21 (B) and stage 31 (C). Retinal sections at stage 41 (D) show Xfz5 is expressed within the CMZ but not in the peripheral region (brackets). (E–I) Reduced eye size following Xfz5 MO (Xfz5 ATG) injection at the 8-cell stage. (E) Uninjected side. (F) Xfz5 MO-injected side. Embryo marked with an asterisk is shown in dorsal views in (G)–(I). (G) Brightfield view. (H) eGFP expression marks injected side. (I) merge of (G) and (H).
|
|
Figure 2.
Inhibition of Xfz5 Causes Reduced Cell Number and Decreased Mitosis in the Optic Vesicle
(A) Reduced cell number on the Xfz5 MO-injected side versus uninjected side (control). Cell counts from a single section through the central part of the distal optic vesicle at stage 23/24. n = 15 embryos. (B) All optic vesicle cells on the Xfz5 MO-injected side are labeled by a 4 hr pulse of BrDU (brown), as confirmed by Hoechst staining (C), since the BrDU stain quenches the fluorescent signal (blue). (D) Reduced number of phosphohistone H3 (HP3) labeled cells within the optic vesicle (stage 23/24) on the Xfz5 MO-injected side compared to the uninjected side (control). n = 17 embryos. (E) Retinal clone size is reduced by the Xfz5 MO. GFP RNA was injected into blastomere V1.2.1 either alone (GFP; n = 9) or together with Xfz5 ATG MO (n = 9), the Xfz5 ATG mismatch MO (n = 5), the Xfz5 UTR MO (n = 5), or the Xfz5 UTR MO plus 500 pg of Xfz5 RNA lacking the MO target sequence (n = 6). The number of GFP-labeled cells was counted in six central retinal sections at stage 41 for each embryo. In (A), (D), and (E), error bars represent SEM; *p < 0.001 by Studentâs t test compared to either uninjected control side in (A) and (D) or GFP alone in (E).
|
|
Figure 3.
Inhibition of Xfz5 Results in Normal Eye Patterning but Decreased Expression of Genes Involved in Retinal Neurogenesis
The Xfz5 MO results in a smaller expression domain for Rx (A), Pax6 (B), Six6 (C), Chx10 (D), and Xfz5 (E), but the relative levels of expression are unchanged compared to the control side. En expression (arrowhead in [A]) is unchanged. Expression of Xdelta-1 (F–H), X-Notch-1 (I–K), Xash3 (L–N), XNgnR-1 (O and P), Xath5 (Q and R), and XBrn3d (S and T) are suppressed on the Xfz5 MO-injected side. Arrow indicates injected side.
|
|
Figure 4.
Xfz5 Is Required for Normal Sox2 Expression in the Optic Vesicle
Sox2 expression in (A) the emerging optic vesicles (stage 20) and (B) the developing eye (stage 24). (C) In retinal sections (stage 41), Sox2 expression is restricted to the CMZ, but excluded from the peripheral region (bracket). (D–F) Sox2 expression is suppressed on the Xfz5 MO-injected side (arrow in [D] and [F]) compared to the uninjected control side (E). (D) Frontal view. (E and F) Lateral views.
|
|
Figure 5.
Inhibiting Sox2 Expression Has a Similar Effect on Eye Development as Inhibiting Xfz5 Expression
(A and B) Reduced eye size with Sox2 MO injection at the 8-cell stage. (C) Sox2 MO injection causes reduction in anti-phosphohistone H3 (HP3) staining within the optic vesicle at stage 23/24. Uninjected side is the control. n = 8 embryos. (D) Retinal clone size is reduced by the Sox2 MO. GFP RNA was coinjected, and the number of GFP-labeled cells was counted in six central retinal sections for each embryo (n = 9 embryos). In (C) and (D), error bars represent SEM; *p < 0.001 by Student’s t test. Sox2 MO does not affect the intensity of Sox2 (E), Xfz5 (F), Rx (G), Pax6 (H), Six6 (I), and Chx10 (J) expression within the optic vesicle. En expression (arrowhead in [G]) is unchanged. The Sox2 MO suppresses the retinal expression of Xash3 (K) and Xdelta-1 (L). Injected side is on the right (arrow).
|
|
Figure 6.
TCF/Lef-Dependent Transcription Is Active in the Developing Eye and Is Required for Normal Eye Development
The TOP:dGFP transgene (A) is expressed in the developing eye (arrowhead) at stage 20 (B), and stage 23 (C), as revealed by in situ hybridization. The midbrain/hindbrain boundary is marked by En expression (arrow in [B]) or reporter expression (arrow in [C]). (D) Rx promoter driving ΔN-Tcf3 tagged with GFP (Rx:ΔN-Tcf3-GFP). (E) GFP antibody staining reveals transgene expression in the developing eye (stage 23). (F) Disrupted eye development in transgenic embryos (stage 41). (G) Sections through the same embryo reveal loss of neural retina tissue, but remnants of RPE and lens (L). Section is stained with the nuclear dye Hoechst (blue).
|
|
Figure 7.
TCF/Lef-Dependent Transcription Is Required for Neural Potential of Retinal Progenitors
Rx:ÎN-Tcf3-GFP transgenic embryos (stage 23) show normal expression of Rx and En (B) and the retinal progenitor markers Pax6 (D), Six6 (F), and Chx10 (H) compared to control Rx:GFP transgenic embryos (A, C, E, and G). Rx:ÎN-Tcf3-GFP transgenic embryos show loss of both Sox2 and Xdelta-1 expression in the retinal domain (bracket in [J] and [L]) as compared to control transgenic embryos (Rx:GFP) (I and K).
|
|
Figure 8.
Inhibiting Xfz5 Expression in Retinal Progenitors Biases Cells to Adopt the Mu¨ller Glial Cell Fate
(A) Expression of dnXfz5 in retinal progenitors by in vivo lipofection caused a 4-fold increase in the representation of Mu¨ller glia as compared to GFP controls. n = 1706 cells from 26 embryos for GFP, n = 402 cells from 9 embryos for dnXfz5. (B) Retinal section (stage 41) showing a GFP-labeled cell (green) expressing dnXfz5 that is colabeled using antibodies for the Mu¨ller glial cell marker CRALBP (red). (C) Expression of dnLRP6 in retinal progenitors by in vivo lipofection also caused a significant increase in the representation of Mu¨ller glia as compared to GFP controls. n = 391 cells from 6 embryos for GFP, n = 252 cells from 5 embryos for dnLRP6. For (A) and (C), the percent representation of each cell type was calculated as a weighted average, and error bars represent SEM; **p < 0.001, *p < 0.05 by Studentâs t test. G, ganglion cells; H, horizontal cells; A, amacrine cells; B, bipolar cells; PR, photoreceptor cells; M, Mu¨ller glial cells.
|
|
Figure 9.
Xfz5 Governs Neural Potential during Retinal Development
(A) During normal development, Xfz5 regulates the expression of Sox2 in retinal progenitors, allowing progenitors to proliferate, express proneural genes, and ultimately differentiate into postmitotic retinal neurons. (B) When Xfz5 function is blocked, Sox2 expression is reduced in retinal progenitors, resulting in reduced proliferation, loss of proneural gene expression, and a bias toward the nonneural Mu¨ller glial fate.
|
|
Figure S1. Inhibition of Xfz5 Function Does Not Alter Neural Patterning, Six3, or Cyclin D1 Expression
The MO-injected side is on the right (arrow). (A) The XFz5 MO causes no change in the level of Cyclin D1 expression in the optic vesicle, although the size of the expression domain is reduced. (B) The Xfz5 MO causes no change in the eye field, as demonstrated by Rx expression, or in anterior-posterior patterning, as demonstrated by the midbrain-hindbrain marker Engrailed (En) (arrowhead) at stage 17. The Xfz5 MO (C) or Sox2 MO (D) causes a reduction in the size of the expression domain for Six3, but the relative levels of expression are unchanged when compared to the uninjected control side.
|
|
Figure S2. Inhibition of Xfz5 Function Is Dependent upon Morpholino Concentration
When 10 ng of Xfz5 MO is used, Xdelta-1 expression recovers by stage 25 (E) with expression levels similar to control (A). This dose results in a smaller eye at stage 41 (F). However, when 20 ng of Xfz5 MO is used, Xdelta-1 expression is still absent at stage 25 (G) and results in a near complete loss of the eye at stage 41 (H). (A–D) uninjected control; (E–H) Xfz5 MO-injected side.
|
|
Figure S3. Inhibiting Xfz5 Expression Causes Reduced Eye Size and Delayed Retinal Development
The eye size on the Xfz5 MO-injected side (A and I) is smaller than on the uninjected control side (E and M), although lamination of the neural retina is normal as demonstrated by Hoechst nuclear staining. Ganglion and amacrine cells (B and F; α-Pax6 antibody staining) as well as rod photoreceptors (J and N; α-rhodopsin antibody staining) are properly positioned within their respective layers on both the uninjected (F and N, respectively) and the injected side (B and J, respectively). eGFP expression (C and K) identifies the Xfz5 MO-injected side. (D), (H), (L), and (P) are the merged images of Hoescht nuclear staining, antibody staining, and eGFP expression. Note that the antibody staining on the uninjected side extends more peripherally than that on the injected side (compare brackets in [L] and [P]), suggesting that the retina on the injected side has a larger CMZ reminiscent of a more immature retina. (QâT) BrDU analysis on stage 41 retinas. (Q and R) BrDU labeling is restricted to the CMZ on the uninjected control side. (S and T) BrDU- labeled cells are detected in the central retina as well as the CMZ on the Xfz5 MO injected side, indicating that the central retina is not fully postmitotic and may be delayed in its development. The number of BrDU cells within the CMZ is similar for control versus the Xfz5 MO-injected side.
|
