Mutations in CHD7 are the major cause of CHARGE syndrome, an autosomal dominant disorder with an estimated prevalence of 1/15,000. We have little understanding of the disruptions in the developmental programme that underpin brain defects associated with this syndrome. Using mouse models, we show that Chd7 haploinsufficiency results in reduced Fgf8 expression in the isthmus organiser (IsO), an embryonic signalling centre that directs early cerebellar development. Consistent with this observation, Chd7 and Fgf8 loss-of-function alleles interact during cerebellar development. CHD7 associates with Otx2 and Gbx2 regulatory elements and altered expression of these homeobox genes implicates CHD7 in the maintenance of cerebellar identity during embryogenesis. Finally, we report cerebellar vermis hypoplasia in 35% of CHARGE syndrome patients with a proven CHD7 mutation. These observations provide key insights into the molecular aetiology of cerebellar defects in CHARGE syndrome and link reduced FGF signalling to cerebellar vermis hypoplasia in a human syndrome.
CHARGE syndrome is a rare genetic condition that causes various developmental abnormalities, including heart defects, deafness and neurological defects. In most cases, it is caused by mutations in a human gene called CHD7. CHD7 is known to control the expression of other genes during embryonic development, but the molecular mechanisms by which mutations in CHD7 lead to the neural defects found in CHARGE syndrome are unclear.
During embryonic development, the neural tube—the precursor to the nervous system—is divided into segments, which give rise to different neural structures. The r1 segment, for example, forms the cerebellum, and the secretion of a protein called FGF8 (short for fibroblast growth factor 8) by a nearby structure called the isthmus organiser has an important role in this process. Since a reduction in FGF8 causes defects similar to those found in CHARGE syndrome, Yu et al. decided to investigate if the FGF signalling pathway was involved in this syndrome.
Mice should have two working copies of the Chd7 gene, and mice that lack one of these suffer from symptoms similar to those of humans with CHARGE syndrome. Yu et al. examined the embryos of these mice and found that the isthmus organiser produced less FGF8. Embryos with no working copies of the gene completely lost the r1 segment. The loss of this segment appeared to be caused by changes in the expression of homeobox genes (the genes that determine the identity of brain segments).
Embryos that did not have any working copies of the Chd7 gene died early in development, which made further studies impossible. However, embryos that had one working copy of the Chd7 gene survived, and Yu et al. took advantage of this to study the effects of reduced FGF8 expression on these mice. These experiments showed that mice with just one working copy of the Fgf8 gene and one working copy of the Chd7 gene had a small cerebellar vermis. This part of the cerebellum is known to be very sensitive to changes in FGF8 signalling. Yu et al. then used an MRI scanner to look at the cerebellar vermis in patients with CHARGE syndrome, and found that more than half of the patients had abnormal cerebella.
In addition to confirming that studies on mouse embryos can provide insights into human disease, the work of Yu et al. add defects in the cerebellar vermis to the list of developmental abnormalities associated with CHARGE syndrome. The next step will be to test if any mutations in the human FGF8 gene can contribute to cerebellar defects in CHARGE syndrome, and to investigate if any other developmental defects in CHARGE syndrome are associated with abnormal FGF8 levels.