Our long term aim is to identify genes affecting atrial septal defect (ASD) and other congenital cardiac abnormalities in humans, and to study their mechanism of action. The genetic basis of idiopathic ASD, and the genetic modifiers of familial ASD, are unknown. We have extensively characterised a mouse model of human familial ASD due to mutations in the cardiac homeobox gene NKX2-5. As in human families, heterozygous mutations in murine Nkx2-5 confer ASD and conduction abnormalities, but with less penetrance and expressivity. Overt ASD occurs in 1 in 100 Nkx2-5 heterozygotes, although atrial septal dysmorphogenesis is evident as dramatically increased frequencies of patent foramen ovale (PFO) and septal aneurysm. While PFO in humans is in most cases clinically benign, it can nonetheless be regarded as a mild variant of ASD, and indeed our studies in mice strongly support the hypothesis that there is a pathological continuum between PFO and ASD. In the course of our studies, we recognised a pronounced variability in the frequency of PFO and other septal abnormalities between inbred strains of wildtype mice. These strain-specific effects also profoundly modify the incidence and severity of septal defects in Nkx2-5 heterozygotes. The Nkx2-5 model provides a powerful genetic resource to identify modifiers of familial ASD/PFO, as well as QTL underlying idiopathic ASD. Since ASD/PFO is just one of the manifestations of Nkx2-5 mutation in both humans and mice, we further hypothesise that modifiers of ASD/PFO will have a broader role in the genetics of congenital heart defects. We will use QTL analysis to identify chromosomal regions determining frequency of PFO in different inbred mouse strains. We have identified an anatomical parameter, the length of the atrial septum primum flap valve, which correlates near perfectly with the percentage of PFO in a particular strain. This parameter will form the main basis of our QTL analysis. Genes responsible for QTL of highest impact will be identified by candidate gene approaches aided by the mouse genome project and microarray analysis. Their relevance to incidence, penetrance and phentypic expression in familial and idiopathic ASDs in humans will be assessed using our patient DNA resource.
