ZIC3, a GLI superfamily transcription factor, is involved in establishing normal embryonic left-right patterning. Multiple abnormalities in the central nervous system (CNS) and axial skeleton have also been observed in mice bearing a Zic3 null allele, mice with a Zic3 overexpression allele, and the majority of patients carrying ZIC3 mutations. Previous studies indicate that ZIC3 protein can bind to the GLI consensus binding site (GLIBS) and physically interact with GLI3, a transcription factor involved in multiple aspects of neural and skeletal development. We investigated in vitro interactions of ZIC3 with GLI3 and the effect of ZIC3 mutations identified in patients with either heterotaxy or isolated cardiovascular malformations. Electrophoresis mobility shift assay (EMSA) revealed that all five intact zinc finger (ZF) domains were necessary for binding of ZIC3 to GLIBS. Inclusion of GLIBS upstream of a basal TK promoter had no effect on the activation of the promoter by ZIC3 alone, but it enhanced the synergistic activation of ZIC3 and GLI3. Wild-type (WT) ZIC3 showed specific binding to GLI3 in GST-pull-down assays. Nonsense and frameshift ZIC3 mutants lacking one or more of the zinc finger domains did not physically interact with GST-GLI3; however, two missense mutants c.1213A>G (p.K405E, fifth ZF domain), and c.649C>G (p.P217A, conserved N-terminal domain) retained binding. Luciferase reporter assays indicated that both p.P217A and p.K405E mutants also retained coactivation with GLI3 of reporter gene expression activity, while all the GLI3-nonbinding ZIC3 mutants lacked this activity. Interestingly, no CNS or skeletal abnormalities were observed in patients bearing the p.P217A or p.K405E mutations.

Increased risks of objectively defined respiratory disease, which have been previously documented, were not seen. Nevertheless, the study suggested increased risk of respiratory symptoms with various occupational exposures as well as likely healthy worker effect.