Small eye (Sey): a mouse model for the genetic analysis of craniofacial abnormalities

It will be clear to readers of this volume that the morphogenesis of the vertebrate head involves a complex series of reciprocal interactions between different cell populations (e.g. ectoderm, neurectoderm, cranial paraxial mesoderm and neural crest). These are brought together as a result of cell migration and the coordinated growth and fusion of facial primordia (see Wedden et al., this volume). A variety of biochemical, molecular and immunological techniques are being used to identify factors responsible for mediating the various subroutines of this overall programme. Another very powerful tool that can be brought into play is that of genetics, and a number of mouse mutants have been described which show inherited craniofacial abnormalities (Green, 1981). Some of these appear to be the result of primary defects in processes such as neural crest migration or the formation of cartilage, e.g. patch (Ph) and cartilage matrix deficiency (cmd) (Green, 1981). However, other mutations seem to have more specific effects on cell interactions during cranial morphogenesis. One of these is the dominant mutation known as Small eye (Sey) which has been mapped to mouse chromosome 2, about 5 cm proximal to the b-2-microglobulin locus (Hogan et al. 1986). Heterozygous Sey/+ mice have small vacuolated lenses and develop cataracts within a few weeks of age. Homozygous Sey/Sey embryos can be distinguished as early as 10·5 days post coitum (p.c.) by the complete failure of lens induction and the absence of nasal pits. Subsequently the optic vesicles become distorted and degenerate, and nasal cavities and olfactory bulbs do not develop. Homozygous embryos die soon after birth because newborn mice cannot breathe through their mouths. Apart from these defects in eye and nose development, Sey/Sey embryos appear quite normal, suggesting that the Sey mutation affects some specific aspect of craniofacial morphogenesis, rather than a process common to many embryonic tissue interactions.

In this paper, we describe in more detail the embryonic development of Sey/Sey and Sey/⁺ embryos, and speculate about the primary defect. We show that while Sey is allelic with Dickie small eye (now Sey^Dey) and Harwell small eye (Sey^H), it is not allelic with Coloboma (Cm), another mutation affecting eye development on chromosome 2 (Searle, 1966).

A colony of Sey/⁺ mice is maintained at the National Institute for Medical Research. They are currently at the 6th backcross to C57BL/10ScSn. Sey/⁺ mice are clearly distinguished soon after birth by the size of their eyes and develop cataracts after 3 weeks. On the C57BL background about 25 % (25/96) of the Sey/ + offspring of both sexes develop hydrocephaly and die by the about 8 weeks of age. Histological examination of fixed brains suggests bilateral hydrocephaly of the lateral ventricles, not involving the third ventricle. Cm/ + males were kindly provided by Dr J.-L. Guenet, Institut Pasteur. For the timing of embryos, noon on the day of the vaginal plug is 0·5 days p.c.

Previous studies showed that, in presumed Sey/Sey embryos, the optic vesicles grow out but lens induction fails to take place (Hogan et al. 1986). Mesodermal cells are present between the optic epithelium and the overlying ectoderm. However, it is not clear whether this is due to failure of mesodermal cells in this position to die, thus preventing intimate contact between optic vesicle and presumptive lens placode (Silver & Hughes, 1974 and see Bard et al., this volume), or whether the mesodermal cells migrate into the space made available as a result of the primary failure of the two epithelial tissues to make or maintain intimate contact.

Absence of contact between optic vesicle and ectoderm is apparent in embryos at 10·5 days p.c. (Hogan et al. 1986). By 11·5 days p.c. the optic vesicle has become very distorted, as shown in Fig. 1. The sections also show the absence of nasal pits in the homozygous embryos.

Scanning electron microscopy confirms our previous observations that the maxillary, mandibular and hyoid processes develop normally in presumed Sey/ Sey embryos (Fig. 2). At 11·5 days p.c., it can be seen that the maxillary processes fuse with small protrusions in the position of the frontonasal mass (Fig. 2B,D). These protrusions would normally be greatly enlarged and expanded by the growth of the nasal pits and the surrounding neural-crest-derived mesenchyme cells (Fig. 2A,C).

Fig. 3 shows a section through the eye of a presumed heterozygous Sey/+ embryo at 15·5 days p.c. The lens is vacuolated and is only about half the size observed in normal littermates. Probably because of the small size of the lens, there is infolding of the anterior margins of the eye cup and infiltration of mesodermal cells.

In previous studies, we have shown that the Sey mutation maps to chromosome 2, about 5 cm from the beta-2-microglobulin locus. Sey is allelic with two early prenatal lethal mutations, Sey^H and Sey^Dey (Dickie’s small eye), both of which may involve small deletions or DNA rearrangements at the Sey locus but extending into nearby genes which affect early development (Hogan et al. 1986; Hogan, Hetherington & Lyon, 1987).

Since the mutation Coloboma (Cm) has been mapped to mouse chromosome 2, proximal to nonagouti (a), between we and un (Searle, 1966; Davisson & Roderick, 1981), we tested for allelism between Sey and Cm. This was achieved by crossing Sey/+ heterozygous females with Cm/+ males. Homozygous Sey/Sey, Sey/Sey^Dey or Sey/Sey^H embryos have a characteristic phenotype, lacking both eyes and nose (Hogan et al. 1986; Hogan, Hetherington & Lyon, 1987). However, as shown in Table 1, only 2/68 embryos from the cross between Sey/+ and Cm/+ lacked eyes altogether, and these did not resemble Sey/Sey embryos in other respects. Rather, the results are compatible with Sey and Cm being nonallelic and with both Sey/ + and Cm/ + embryos having colobomatous eyes, so that they cannot be clearly distinguished by external observation at the stages examined, and Sey/Cm embryos being microphthalmic.

There are several possible explanations for the underlying defect caused by the Sey mutation. First, there may be a failure in the mechanism whereby regions of the ectoderm of the early embryo become specified as presumptive lens or nasal placodes, programmed to respond later to inductive signals from either the lens vesicle or the frontonasal mesenchyme, respectively. According to this hypothesis, in Sey/Sey embryos, no positional information is given at all, while in heterozygotes the areas instructed to become lens or nasal placode are abnormally small. Alternatively, the defect could be in the process of induction itself, attenuating the production, transfer or reception of the signal which presumably passes from one tissue to another. These models assume that the Sey mutation involves a dominant ‘loss of function’ rather than ‘gain of function’, but other scenarios cannot be excluded. The most fruitful way to distinguish between these models is to identify the normal Sey gene product by means of ‘reverse genetics’, i.e. cloning the gene, identifying the transcript and predicting the nature of the gene product. The fortuitous localization of the Sey mutation close to the beta-2-microglobulin locus and to other cloned loci (e.g. Woychik et al. 1985), and the existence of a number of presumed deletion mutants at the Sey locus may allow this goal to be achieved in the not-too-distant future.

We thank J.-L. Guenet, Institut Pasteur, Paris, for providing Cm/ + mice and for his enthusiastic interest in the project.