Clonal model of vertebral column and skull development derived from genetically mosaic skeletons in allophenic mice

Abstract

A clonal model of the origins of the vertebral column and parts of the skull has been derived from an analysis of 35 genetically mosaic skeletons of four-parent allophenic mice of the C57BL/6⇌C3H strain combination, by comparing bone morphology with that of pure-strain controls. The number of attributes that could be reliably discriminated between control strains and evaluated in allophenic skeletons was greatly augmented by superimposing images in a microscope with suitable optics. There was considerable variation among allophenic individuals in the relative proportions and locations of the two cell strains, ranging from animals in which only one half-vertebra differed in strain-type from the rest of the axial skeleton to cases with finely interspersed strain differences scattered within and among individual vertebrae. Some left or right half-vertebrae were of indeterminate strain-type, from which cell strain admixture is inferred, and many other vertebrae showed left-right asymmetry and differences of strain-type from anterior and posterior neighbors. The composite picture obtained from the entire series is that there are at least four units of independent morphological strain-type variability, and therefore a minimum of four developmental cell lineages, in any vertebra. These appear to correspond to the four intersegmentally derived components: the two caudal sclerotomites from a pair of somites and the two cranial sclerotomites from the next succeeding pair of somites. The suggested archetype of vertebral development therefore is either that the caudal and cranial sclerotomites are the clonal units capable of separate gene control or, alternatively, that the entire sclerotome component from each somite is the clonal unit. The mouse body vertebral column (cervical to sacral regions inclusive—some 30 vertebrae) may thus be derived from a minimum, or possible actual number, of only 120 cell lineages that are complete clones or parts of clones, in which tissue-specific gene function may have begun in a somite-precursor stage. Allphenic skulls also showed indeterminate phenotypes within each midline or paired bone studied, indicating the origin of each from at least two cells. The occipital region displayed many abnormalities; these are consistent with the conclusion that the postotic part of the skull is also clonally derived from sclerotome parts which, because of strain differences in growth rates, failed to fuse in the normal manner. The C57BL/6 strain-type occurred more frequently in the skull and the C3H type in the vertebral column, especially in the lumbosacral region. This distribution probably reflects an autonomous, strain-specific selective advantage enjoyed by C57BL/6 sclerotome cells in the early forming (anterior) somites and by C3H in the later forming (posterior) somites. The clonal model of axial skeletal development, obtained from allophenic mice, appears to be applicable to single-genotype mice as well. A clonal basis seems indicated for many congenital skeletal defects due either to mutant genes or to teratogenic agents.