Characterization of a novel insertional mouse mutation, kkt: A closely linked modifier of Pax1

We describe a novel transgene insertional mouse mutant with skeletal abnormalities characterized by a kinked tail and severe curvature of the spine. The disrupted locus is designated kkt for "kyphoscoliosis kinked tail." Malformed vertebrae including bilateral ossification centers and premature fusion of the vertebral body to the pedicles are observed along the vertebral column, and the lower thoracic and lumbar vertebrae are the most affected. Some of the homozygous kkt neonates displayed two backward-pointing transverse processes in the sixth lumbar vertebra (L6) that resembled the first sacral vertebra, and some displayed one forward- and one backward-pointing transverse process in L6. The fourth and fifth sternebrae were also fused, and the acromion process of the scapula was missing in kkt mice. The skeletal abnormalities are similar to those observed in the mouse mutant undulated (un). The transgene is integrated at the distal end of chromosome 2 close to the Pax1 gene, as revealed by FISH analysis. However, mutation of the Pax1 gene is responsible for the un phenotype, but the Pax1 gene in the kkt mice is not rearranged or deleted. Pax1 is expressed normally in kkt embryos and in the thymus of mature animals, and there is no mutation in its coding sequence. Thus, the skeletal abnormalities observed in the kkt mutant are not due to a lack of functional Pax1. Mouse genomic sequences flanking the transgene and PAC clones spanning the wild-type kkt locus have been isolated, and reverse Northern analysis showed that the PACs contain transcribed sequence. Compound heterozygotes between un and kkt (un(+/-)/kkt(+/-)) display skeletal abnormalities similar to those of un or kkt homozygotes, but they have multiple lumbar vertebrae with a split vertebral body that is more severe than in homozygous un or kkt neonates. Furthermore, the sternebrae are not fused and no backward-pointing transverse processes are detected in L6. It is therefore apparent that these two mutations do not fully complement each other, and we propose that a gene in the kkt locus possesses a unique role that functions in concert with Pax1 during skeletal development.