Limbs move beyond the Radical fringe

Abstract

Zhang and Gridley reply — We too have created mice that have a large deletion in the Radical fringe gene and investigated the resulting effect on their development. Our results agree with those of Moran et al. in showing that Rfng is not an essential gene in mice, but they do not support the idea that the presence of a neo- selection cassette inserted into the Rfng locus invariably leads to embryonic lethality.

Main

We constructed and analysed a targeted mutation of the Rfng gene by using a targeting vector in which a section of the Rfng genomic sequence, encoding the region from amino acid 52 of the Rfng protein through to the end of the Rfng gene, was deleted. The vector contained a neo- selection cassette in the same transcriptional orientation as the Rfng gene. Mice homozygous for our Rfng mutant allele formed viable, morphologically normal adults at the expected mendelian frequencies (34 of 126 homozygotes (27%) on a mixed C57BL/6J×129/SvImJ background, and 3 of 16 homozygotes (19%) on an inbred 129/SvImJ background). The absence of wild-type Rfng transcripts in our Rfng homozygous-mutant adult mice was confirmed by northern blotting.

It is possible that a difference in the construction of the two targeted Rfng alleles could explain the different mouse phenotypes generated by Moran et al. — the orientations of the neo- selection cassette were not the same in the two targeted Rfng alleles, for instance. Reversing the orientation of the neo sequence for transcription creates a potent splice-acceptor site¹,².

In considering the implications of their finding for the interpretation of Lfng mutant phenotypes³,⁴, Moran et al. correctly point out that, in some instances, the presence of a neo- selection cassette in a targeted allele can affect the transcription of other genes, and hence the mutant phenotypes they generate. To confirm that the mutant phenotype is independent of the presence of a neo cassette in the two Lfng mutant alleles, a targeted Lfng allele constructed without a neo cassette needs to be tested.

However, we believe that it is unlikely that the neo cassette in the two Lfng mutant alleles is responsible for the alterations in Uncx4.1 transcription or for the defects in somite formation that we detected in Lfng homozygous-mutant embryos. The spontaneous mouse mutant pudgy (pu) is caused by a frameshift mutation in the Dll3 gene⁵, which encodes a ligand for the Notch family of proteins. This frameshift leads to premature truncation of the Dll3 protein⁵. As homozygous Dll3^pu /Dll3^pu mutant embryos show defects in somite formation very like those seen in Lfng mutant embryos, we compared Uncx4.1 transcription in both Dll3^pu /Dll3^pu and Lfng^lacZ /Lfng^lacZ mutant embryos and found that the Uncx4.1 expression pattern was similarly disorganized in both mutants. Given that the presence of a neo cassette does not alter Uncx4.1 expression in Dll3^pu /Dll3^pu embryos, the simplest explanation is that the alterations in Uncx4.1 expression and somite formation in Lfng^lacZ /Lfng^lacZ mutant embryos are caused by perturbations of the Notch signalling pathway, and not by the presence of a neo cassette in the Lfng^lacZ mutant allele.