One of the critical events during the development of an embryo is the formation of a transient structure called the neural crest. Cells from the neural crest give rise to a wide range of cell types in the body, including cartilage and bone cells, pigment cells, nerve cells in the gut and nerve-supporting cells (called glia). Abnormalities in neural crest cells can give rise to developmental syndromes in humans, such as Waardenburg syndrome (WS) and Hirschsprung disease (HD). In HD, sections of the gut fail to develop nerve cells, which can be life-threatening for undiagnosed infants and for which the only treatment option currently is surgery. Patients with WS suffer a range of symptoms, including hearing loss and defects in pigmentation of the hair, eyes and skin. WS is a rare disease and thus little is known about its health and economic burdens, or about its prevalence. However, in a single study, pooled cohorts from the US, UK and South Africa described 274 individuals as WS sufferers and over 50 cases have been reported in Europe alone. WS is divided into 4 sub-types, WS types 1 to 4, which differ in symptoms, with WS types 1 and 2 being the most common. WS type 4 is particularly debilitating as it is associated with HD. While first described in the Dutch population, with an estimated incidence of 1 in 42 000 people, WS occurs in many racial groups and an incidence of 1/20 000 people was reported in Kenyan Africans. The various sub-types of WS have been associated with mutations in several genes, mainly transcription factors involved in neural crest cell specification and pigmentation such as PAX3, MITF and SOX10. Of particular interest to this study are the mutations in SOX10 associated with WS types 2 and 4, as well as with a severe neurological variant of WS type 4 called PCWH (Peripheral demyelinating neuropathy–Central dysmyelinating leukodystrophy–Waardenburg syndrome–Hirschsprung disease).
SOX10 plays a key role in the production, movement and fate of neural crest cells. The role played by SOX10 in neural crest cells stems from its ability to bind to the DNA and regulate expression of its target genes. Interestingly, a number of patients with WS have been shown to have subtle mutations in SOX10 within the DNA binding domain, a region of the protein that is important for its transcription factor activity. When these mutations were tested using cell culture (in vitro) methods, the results did not match well to the severity of the symptoms of the WS patient from which the mutation was isolated (we say there is no genotype-phenotype correlation). This is likely due to the fact that it is very difficult to reproduce the complex environment of the neural crest using cell culture experiments. It was therefore clear that new tools were needed to better understand the genotype-phenotype relationship in neural crest disorders using a living organism (an in vivo context).
The SOX10mutants project proposed to address this need by using a novel zebrafish rescue assay to investigate the effects of different mutations in the SOX10 gene on the development of cell types deriving from the neural crest, to test in vivo the genotype-phenotype correlations with WS. Importantly, sox10 gene expression and function is extremely similar between fish and mammals and the rescue assay was based on the use of a zebrafish line in which sox10 is mutated and which lacks melanocytes (pigment cells) and enteric neurons (nerve cells in the gut), similar to WS symptoms. The objectives of this study were, firstly, to determine if expressing the human mutant SOX10 proteins in the mutant zebrafish embryos was able to rescue the lack of melanocytes and enteric neurons. Secondly, we wished to determine in which cellular compartments the mutant proteins were located and thirdly, to explore the details of how the mutant proteins might act at a molecular level.
Our initial approach to in vivo testing worked well for melanocytes, but it proved difficult to get enteric neuron rescue. However, we improved the assay so that we now get clear rescue of both these key aspects of the mutant with normal human SOX10 protein. Thus, our study has developed an excellent zebrafish rescue assay. Importantly, the mutant human proteins generally showed a reduced ability to rescue melanocytes and enteric neurons, compared to the normal (wild-type) human SOX10. The in vivo zebrafish assay therefore demonstrated an improved genotype-phenotype correlation over the in vitro cell culture assays, highlighting the potential of using zebrafish to study WS. The initial data obtained for the second objective provided the first evidence that the cellular localisation of selected mutants in vivo was similar to results published using cell culture experiments. The delay needed to improve the rescue assay, particularly for enteric neuron rescue, meant that we were unable to address the third objective. Overall, the outcomes of this study were very positive and provide excellent proof-of-concept data for the use of zebrafish as a tool for studying human neural crest disorders.
