Abstract
Introduction
Adolescent idiopathic scoliosis (AIS) is the most common form of spinal deformity. It occurs mainly in girls and progresses during pre-pubertal and pubertal growth, which is a crucial period for bone mass acquisition. The cause and molecular mechanisms of AIS are not clear; at present the consensus is that AIS has a multifactor cause, with many genetic factors. During the past 5 years, considerable effort has been devoted to identify a gene or genes that cause a predisposition to AIS. Many loci for this disorder have been mapped to different chromosome regions, but no genes have been clearly identified as being responsible for AIS, and, most importantly, the resulting protein defects remain to be shown. We aimed to identify the gene(s) that could be involved in AIS and to validate their involvement by both genetic and functional analyses.
Methods
A large multiplex AIS French family was chosen for this study on the basis of clinical and radiological data. Whole genome genotyping of the 20 members of this family led to the mapping of a dominant disease-causing gene to two critical genomic intervals (Edery and colleagues, Eur J Hum Genet, accepted [2011]), but the causative mutation remains to be identified. In parallel, gene expression profiling was investigated by microarray analysis in RNA samples isolated from osteoblasts derived from healthy individuals and those with AIS. RNA samples were extracted from osteoblasts, purified, fluorescently labelled, and then hybridised to gene expression microarrays with the Illumina expression BeadChips technology containing more than 46 000 probes for the human genome (HumanHT-12). Data analysis in R version 2.10.1 (Bioconductor packages oligo and limma) was done, and genes that had at least 1·5-fold change in expression were considered differentially regulated relative to controls. AIS candidate genes within the critical intervals were selected on the basis of their mRNA expression in AIS individuals and by their known functions. The coding regions of these candidate genes were then sequenced to identify potential mutations. The biological activity of mutant proteins is under evaluation by in-vivo functional studies in zebrafish.
Results
In the AIS family, a maximum LOD score of 3·01 was reached on two specific chromosomal regions. The interval lengths of these regions were 7cM and 12cM. These two regions contain several genes that might be responsible for AIS.
Microarray analysis showed many genes that are differentially regulated in AIS osteoblasts compared with control osteoblasts. We recorded that 2·6% of the 24000 genes examined were upregulated in AIS osteoblasts, whereas 2·16% of them were downregulated. We observed a roughly 3-fold increase or decrease in the transcripts of many genes in AIS osteoblasts. Some of the differentially regulated genes are located within the two chromosomal candidate regions. The sequencing of the candidate genes' coding sequences was done on the family members. Sequence analysis showed two rare SNPs located on the coding regions of a gene that we called CH5G1. These two SNPs are located on the C-terminal region of the CH5G1 protein and affect its structure and probably its cellular activity and biological process leading to the disease. The C-terminal region of this protein interacts with the mRNA of a gene whose defects cause scoliosis as a secondary phenotype. The pathogenic nature of these SNPs is being investigated in the zebrafish model. The results suggest that CH5G1 gene's defects could be associated with AIS in this family.
Conclusions
Identification of susceptibility genes for AIS will facilitate the understanding of underlying biochemical pathways (functional studies) and ultimately the development of specific therapies (pharmacological studies). This is likely to have important implications, since the cause of AIS is unknown.
Acknowledgments
This study is supported by the Fondation Yves Cotrel, Institut de France.
