A loss of lunatic fringe is associated with female infertility

The Wilson-Rawls laboratory was the first to describe the dynamic expression pattern of genes of the Notch signaling pathway in mammalian ovarian follicles (Johnson et al.,2001). In a recent report (Hahn et al., 2005), we described a completely penetrant infertility phenotype in female mice that are homozygous null for a lunatic fringe(Lfng) mutant allele, previously described by Evrard et al.(Evrard et al., 1998). Furthermore, we characterized defects in follicle development and meiotic maturation that correlate with a loss of expression of the Notch downstream target effector genes in follicles (Hahn et al., 2005). These data are consistent with our conclusion that Lfng and the Notch signaling pathway have an important role to play in ovarian folliculogenesis.

Xu and colleagues now report anecdotal evidence that some mice homozygous for a Lfng mutant allele (Lfng^lacZ) generated in their laboratory are fertile and were used occasionally in the generation of more Lfng^lacZ/lacZ mutants, as well as for the generation of compound mutations of Lfng and other genes in the Notch pathway(Xu et al., 2006). We note that this breeding strategy is not described in papers published by the Gridley laboratory. In fact, where a breeding strategy was described, the authors have bred Lfng^lacZ/+ animals to generate homozygous Lfng^lacZ/lacZ mutants(Zhang and Gridley, 1998; Zhang et al., 2002). However,Xu et al. concede that not all of the Lfng^lacZ/lacZmatings are productive. In other words, consistent with our observations, a lack of lunatic fringe is associated with infertility. These observations do not dispute our central conclusion that Lfng is important in ovarian folliculogenesis, but instead provide new information regarding the penetrance of the infertility phenotype in the two mutant alleles of Lfng.

Genetic mutations in mice are powerful tools for examining the contribution of specific genes to complex biological processes, such as folliculogenesis. Interpreting specific gene mutations can be confounded by variations in the penetrance of phenotypic defects in the same genetic background, as well as in different genetic backgrounds (Beck et al.,2000; Chia et al.,2005). Furthermore, variations in the severity of the phenotype can occur between mutant alleles generated in different laboratories. In the case of Lfng, mice homozygous for either mutant allele experience radical disruptions in the organization of the axial skeleton owing to a failure of the proper segmentation of the somites(Evrard et al., 1998; Zhang and Gridley, 1998). Despite similarities in their gross morphology, there was a greater loss of somite epithelialization in the Lfng^-/- embryos when compared with the Lfng^lacZ/lacZ embryos(Evrard et al., 1998; Zhang and Gridley, 1998). Perhaps it is not surprising that additional studies with these mice are revealing additional differences. Indeed, a careful analysis of fertility and fecundity rates in the Lfng^lacZ mice would offer an excellent comparison and contrast to the strain of mice used in our studies. None of this, however, refutes our conclusion that Lfng and, by extension,Notch signaling are important regulators of folliculogenesis.