DEV012088 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. si-RNA targeting sequences and specificity. (A) Sequence of the short hairpin RNA target sequences directed against non-overlapping regions of Meis1 or Meis2. (B) Efficiency of the RNAi targeting constructs as assessed by luciferase reporter assay. Hek293T cells were co-transfected with the RNAi targeting constructs M2si_I, M2_siII M1si_I, respectively, together with a psiCHECK-2 reporter construct that contained the coding regions of Meis1 or Meis2 fused to Renilla luciferase (M1-RL and M2-RL, respectively; Promega, Mannheim, Germany). Co-expression of the RNAi-targeting construct together with the specific reporter plasmids containing M1-RL or M2-RL leads to RNAi-mediated knock-down of the Renilla luciferase activity. Because psiCHECK-2 contains a second luciferase gene, firefly luciferase, expressed from a HSV-TK promoter, transfection efficiency was normalized to the firefly luciferase activity. Renilla luciferase activity of psiCHECK without co-transfection of an RNAi targeting construct was used as reference and was set at 100%. Error bars indicate s.d. (C-D′) In vivo analysis. Meis1_si_I (C,C′) or Meis2_si_I (D,D′) were electroporated into the neural tube of HH10-11 chick embryos at 1 µg/µl together with pMIWIII-GFP (0.8 µg/µl). The endogenous transcript levels of both genes were markedly reduced in the right eyes, which were electroporated with the si-RNA targeting constructs, compared with the left, non-electroporated eyes.

• Supplemental Figure S2 -

  Fig. S2. Quantification of the rate of progenitor cell proliferation in the early chick retina. BrdU was injected into the right eye of embryos of the developmental ages indicated. Labeling duration was 2 hours. The percentage of BrdU⁺ RPCs was determined by comparing BrdU⁺ nuclei, visualized using the BrdU Labeling and Detection Kit I (Roche, Germany), with cell nuclei visualized with DAPI. RPCs were recognized by the small size and elongated shape of their nuclei (compared with the larger and more round nuclei of RGCs). An average of 2000 cells per embryo were counted. (A) Quantification of cell proliferation between HH19 (late E2) and HH24 (E4) in vivo. The percentages of BrdU-labeled cells at different locations along the retina’s central-to-peripheral axis at late HH19, HH22 and HH24 were determined. Owing to the small size of HH19 eyes, only a central, a nasal-peripheral and a temporal-peripheral region of the retina were analyzed to ensure that sufficient distance between the surveyed areas was kept. (B) Relative location of the retinal regions that were surveyed. (C-E) The central-to-peripheral decline in the number of BrdU-labeled nuclei correlates well with the gradual loss of Meis2 expression in the retina. This parallel is most obvious when cells of the central retina are observed (orange bars). Here, the percentage of BrdU⁺ cells dropped between HH19 and HH22, but did not decrease further by HH24, consistent with the idea that the loss of Meis2 expression in the central retina, which occurs after HH19, is associated with reduced progenitor cell proliferation. Scale bars: 100 µm in C; 200 µm in D; 500 µm in E.

• Supplemental Figure S3 - Fig. S3. Regulation of Pax6 expression by Meis. (A-B′′) Expression of Pax6 at HH18 (green) following retroviral misexpression of MeisEnR (A-A′′) or an EnR-fusion to an unrelated homeodomain protein (B-B′′). Cells expressing the EnR-fusion proteins were visualized with an antibody against the Myc-epitope fused to EnR and shown in red. (A′′, B′′) Overlay of both channels. (C) Quantification of the results. Gray bars represent the percentage of Pax6 and Myc double-labeled cells, black bars the percentage of Pax6-labeled cells that are not Myc⁺; error bars represent s.d. The majority of cells expressing MeisEnR do not express Pax6, whereas cells expressing the control-EnR protein or Meis2 fused to a triple HAepitope (M2HA) co-express Pax6 (n=17 specimens for MeisEnR; n=19 for GFP; n=4 for M2HA). (D-G) Pax6 is not required for Meis2 expression in the mouse eye. (D) Expression of mouse Meis1 in a wild-type embryo at E10.5; (F) Pax6^sey/sey littermate. (E,G) Expression of mouse Meis2 in a wild-type embryo at E10.5 is gradually lost in the presumptive neural retina (E), but retained in the optic vesicle remnant of a Pax6^sey/sey littermate (G). Meis1/Meis2 expression in the chick retina was unchanged following Pax6EnR transfection (data not shown). (H) Aberrant location of MeisEnR-expressing retinal progenitor cells at HH22. MeisEnR-expressing clones were often displaced towards the vitreal surface of the retina and did not show signs of differentiated neurons. This is reminiscent of the displacement of Pax6 mutant cells in the retinae of Pax6^sey/sey/wild-type chimeric mice (Collinson et al., 2003). MeisEnR-expressing cells are labeled with an anti-Myc antibody and shown in red in the context of Pax6 expressing cells show in green. (I) Reporter analysis of an α-enhancer-driven reporter construct. The reporter construct 406B/X of Kammandel et al. (Kammandel et al., 1999) was co-transfected into Hek293T cells together with a panel of expression constructs carrying the canonical splicing isoforms of Pax6 (Pax6^can), Meis1, Meis2 or of the unrelated homeodomain protein SOHo1 (Schulte and Cepko, 2000). Reporter activity alone was set as 1. Transfection of SOHo1 did not elevate reporter activity, whereas expression of Pax6^can autostimulated the α-enhancer 2.7-fold. When Pax6^can was combined with Meis1 or Meis2, reporter gene expression was further elevated about 1.6-fold. Meis1 or Meis2 expression without Pax6 also stimulated α-enhancer-driven reporter activity, albeit less than when co-expressed with Pax6. Meis2-mediated reporter stimulation was effectively blocked by co-transfection of the Meis2-specific RNAi-targeting construct M2_si_I. (J) Nucleotide sequence of the retina-specific α-enhancer located in intron 4 of the murine Pax6 gene (NCBI# AF008213); the previously identified Pax2/6-binding site is shown in blue; two Meis consensus sites in red.

  Collinson, J. M., Quinn, J. C., Hill, R. E. and West J. D. (2003). The roles of Pax6 in the cornea, retina, and olfactory epithelium of the developing mouse embryo. Dev. Biol. 255, 303-312.

  Schulte, D. and Cepko, C. L. (2000). Two homeobox genes define the domain of EphA3 expression in the developing chick retina. Development 127, 5033-5045.