Manic fringe is not required for embryonic development, and fringe family members do not exhibit redundant functions in the axial skeleton, limb, or hindbrain

Abstract

Tight regulation of Notch pathway signaling is important in many aspects of embryonic development. Notch signaling can be modulated by expression of fringe genes, encoding glycosyltransferases that modify EGF repeats in the Notch receptor. Although Lunatic fringe (Lfng) has been shown to play important roles in vertebrate segmentation, comparatively little is known regarding the developmental functions of the other vertebrate fringe genes, Radical fringe (Rfng) and Manic fringe (Mfng). Here we report that Mfng expression is not required for embryonic development. Further, we find that despite significant overlap in expression patterns, we detect no obvious synergistic defects in mice in the absence of two, or all three, fringe genes during development of the axial skeleton, limbs, hindbrain, and cranial nerves.

Figure 1

Targeted deletion of Mfng. A) The Mfng endogenous locus, (numbered boxes = exons), the targeting vector and the structure of the targeted locus before and after cre recombination are shown. Locations of probe (solid lines) and PCR primers used in genotyping (arrows) are indicated. In the final Mfng^Δ1 allele, 899 bp surrounding exon 1 are replaced with a loxP site. B) After mating to EIIA cre mice, Mfng^Δ1 mice were genotyped by Southern blot using the 5′ probe to detect a 13 kb wildtype EcoRI band (arrow) or a 7 kb mutant EcoRI band (arrowhead). A representative PCR genotype is shown Arrow: endogenous band. Arrowhead: targeted band. C) Northern blot analysis of polyA+ mRNA isolated from brains of mice of the indicated genotypes. Blots were probed with a full length open reading frame cDNA probe to detect a 1.8kb Mfng mRNA. The blot was additionally probed with Rpl32 as a loading control. A 2.0 kb Mfng band is visible in Mfng^+/Δ1 and Mfng^Δ1/Δ1 RNA. D) The 5′ end of the Mfng locus is shown with numbered exons shown as boxes, and he 5′ end of the Mfng mRNA is shown. The longest identified 5′RACE product from Mfng^Δ1 mice is shown below with the spliced intro sequences indicated as a grey box. E) RT-PCR using indicated primers (small arrows) demonstrates that this splicing event can be detected in wild type RNA, as well as Mfng^Δ1 RNA.

Figure 2

No synergistic effects on skeletal development are observed in mice with deletions of multiple fringe genes. A) Skeletal preparations of mice of indicated genotypes stained with Alizarin red and Alcian Blue. Ventral (a-e) and dorsal (f-j) views of the ribs and dorsal views of the sacral spine and tail (k-o) are shown. Similar levels of skeletal disorganization are observed in all Lfng null genotypes, regardless of the genotype of other fringe family members. As no differences were observed between homozygous wildtype and heterozygous embryos, these were pooled and referred to as either “+” or “+/Δ”. B) The total number of rib abnormalities (left) and the total number of tail vertebrae (right) were quantified in Rfng+/?;Mfng−/−;Lfng+/? (n= 7), Rfng+/?;Mfng+/?;Lfng−/− (n=8) Rfng−/−;Mfng+/?;Lfng−/− (n=5), Rfng+/?;Mfng−/−;Lfng−/− (n=5) and Rfng−/−;Mfng−/−;Lfng−/− (n=14) animals. Results are shown as bar and whisker graphs (solid horizontal dash indicates the mean). In both analyses, all Lfng−/− genotypes were highly significantly different from the Rfng+/?;Mfng−/−;Lfng+/? group (p<0.001, ANOVA followed by Tukey post hoc).

Figure 3

Loss of multiple fringe genes does not perturb limb development, and Rfng expression is not regulated by En1. A) Alizarin red/ Alcian blue stained preparations of forelimbs of neonates of the indicated genotypes do not reveal overt limb patterning defects in the absence of two or all three fringe genes. C-F) ßgal staining of Rfng;En1 forelimbs of embryos of the indicated genotypes, with dorsal to the top. B,C, and E are 70um sections, while D and F are whole mount limb buds. B) Rfng expression is unchanged in 9.5 d.p.c. limb buds regardless of En1 genotype. C,D) Rfng is expressed in the dorsal ectoderm, ventral ectoderm and mesoderm of the limb bud at 10.5 d.p.c. AERs are visible as thickenings of the ectoderm at the limb apex (C) and along the distal margin of limb buds (D) in En1^+/+ and En1^+/− limbs. In En1^−/− limbs the bud is broader and the AER is ventrally expanded and flattened. No phenotypic differences are observed between Rfng^lacZ/+;En1^−/− and Rfng^lacZ/lacZ;En1^−/− embryos. E,F) At 11.5 d.p.c. En1^−/− limbs are wider than controls and exhibit flattened, expanded AERs, with frequent bifurcations, as well as nubs on the ventral anterior surface (*). Again, no phenotypic differences are observed between Rfng^lacZ/+;En1^−/− and Rfng^lacZ/lacZ;En1^−/− embryos. G) Fgf8 expression reveals that Rfngl^acZ/+;En1^+/+ and Rfng^lacZ/lacZ;En1^+/− , limb buds exhibit a single rim of Fgf8 staining, while both Rfng^lacZ/+;En1^−/− and Rfng^lacZ/lacZ;En1^−/− embryos exhibit an additional ventral rim of Fgf8 staining (arrow).

Figure 4

Fringe genes are not required for mouse hindbrain segmentation. A) Whole mount in situ hybridization with a probe against Krox20 was performed on 8.5 d.p.c. embryos of the indicated genotypes. Krox20 expression was observed in rhombomeres 3 and 5, and no overt differences were seen among the genotypes (n= 8 Rfng+/+;Mfng+/+;Lfng+/+, n=7 Rfng+/?;Mfng−/−;Lfng+/+, n=6 Rfng+/?;Mfng−/−;Lfng−/−, and n= 4 Rfng−/−;Mfng−/−;Lfng−/−). B) Whole mount immunohistochemistry was performed with a neurofilament antibody (2H3, Developmental Studies Hybridoma Bank) on 10.5 d.p.c. embryos of the indicated genotypes. Cranial nerve organization and positioning appears unaffected regardless of genotype. In the trunk irregular axon projections are observed in all Lfng−/− mice, as expected. (n= 20 Rfng+/+;Mfng+/+;Lfng+/+, n=15 Rfng+/?;Mfng−/−;Lfng+/+, n=2 Rfng+/?;Mfng−/−;Lfng−/−, and n= 6 Rfng−/−;Mfng−/−;Lfng−/−)