DEV024620 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Sfrs1 expression during the G2−M phase of the cell cycle. (A) RNA ISH for Sfrs1 in the mouse brain at E11.5, with the magnified images of the olfactory epithelium shown in boxes on either side of the brain section. (B) RNA ISH for Cdc20, a marker of G2−M phase of the cell cycle in the olfactory epithelium at E11.5. (C) Immunofluorescence for Ki67 and phosphorylated histone H3 (PH3) in the olfactory epithelium at E11.5. (D) RNA ISH for Cdc20 and immunofluorescence for phosphorylated histone H3 in the retina at E11.5.

• Supplemental Figure S2 -

  Fig. S2. Proliferation in the Sfrs1-cKO retina. (A) Immunofluorescence detecting PH3 at E12.5 is shown for the wild-type and the mutant retina. The DAPI-stained images of these sections are shown next to the immunofluorescence images. (B) Quantification of the phosphorylated histone H3⁺ cells in wild-type (n=3) and mutant (n=3) retinae.

• Supplemental Figure S3 -

  Fig. S3. Cross-reactivity of mouse monoclonal antibodies to the inner plexiform layer of the Sfrs1-cKO retina. In Fig. 8F, the staining for rhodopsin in the Sfrs1-cKO showed an increased staining in the IPL, which was not observed in the wild-type retina. Since the rhodopsin antibody is a mouse monoclonal antibody, the IPL staining most likely reflects cross-reactivity of the mouse secondary antibody to the immune cells. This is supported by two observations. First, immunofluorescence with a rabbit polyclonal antibody against rhodopsin showed no staining in the IPL for rhodopsin in both the Sfrs1-cKO (A) and the wild-type (B) retina. Second, when the anti-mouse secondary antibody was used by itself, the IPL in the mutant retina showed a robust staining, which was not observed in the wild-type retina (C,D). In the wild-type retina, only the blood vessels were stained (white arrowhead in D).

• Supplemental Figure S4 -

  Fig. S4. Postnatal production of Müller glia in the Sfrs1-cKO retina. (A) Confocal slice of the image shown in Fig. 8O used to demonstrate the colocalization of glutamine synthetase with the nuclear GFP. (B) The same confocal slice shown in A stained with DAPI. (C) High-magnification image of the nuclear GFP⁺ cell along with DAPI highlighting the canonical shape of the Müller glia nuclei.

• Supplemental Figure S5 -

  Fig. S5. The rules of nonsense-mediated mRNA decay and the position of the stop codon in Sfrs1. (A) A typical stop codon is in the last exon, more than 50 nucleotides downstream from the last exon-exon junction. (B) However, there is an exception to this rule, such that the stop codon is not in the last exon. Instead, it is in the penultimate exon. In this case, 98% of these genes have the stop codon less than 50 nucleotides upstream of the 3′-most intron (Nagy and Maquat, 1998). (C) Interestingly, Sfrs1 falls into the latter category and is itself an exception to that rule such that the stop codon for Sfrs1 is in the penultimate exon (exon 4), but is 966 nucleotides upstream of the 3′-most intron. In the case of Sfrs1, the normal mRNA does not follow the rules of stop codon recognition yet is not subjected to NMD. (D) This raises the question of whether or not the Sfrs1a isoform, which includes intron 3, would be subjected to NMD.

  Reference

  Nagy, E. and Maquat, L. E. (1998). A rule for termination-codon position within intron-containing genes: when nonsense affects RNA abundance. Trends Biochem. Sci. 23, 198-199.