Introduction

Idiopathic scoliosis (IS) has been associated with several genetic loci in varying study populations, reflecting the disorder's genetic complexity. One region of interest is on chromosome 17, flanking regions linked to neurofibromatosis type 1 (NF1). This region is of particular relevance because the most common osseous manifestation in NF1 is scoliosis (10–30% of patients). This alludes to a potential genetic correlation within this region affecting spinal development or stability. The objective of this research is to identify candidate genes within this region that are statistically linked to IS.

Introduction: Familial idiopathic scoliosis (FIS) is a complex genetic disorder potentially resulting from multiple genetic interactions and variants. A previous genome wide screen in a large population of families with FIS followed by fine mapping utilizing STRP’s identified and narrowed critical regions on chromosomes 9 and 16. A high density SNP map was then designed across these regions. This array was then assayed within the same population in an effort to link and/or associate specific genetic intervals or candidate genes with the expressed phenotype.

Methods: A sample of families with IS (202 families, 1198 individuals) was recruited with IRB approval and underwent a genomic screen. Results were analysed by model-independent linkage analysis (SIBPAL). Following initial analyses, families were then stratified according to mode of inheritance. 101 families (550 individuals) represented an autosomal dominant mode of heritability and underwent fine mapping in the candidate regions.

Custom SNP pools were designed for the candidate regions at a density of 1 SNP/58Kb. DNA from 550 individuals (AD group) were genotyped using the Illumina platform. A total of 1536 SNP markers were attempted, of which 1324 were released; 519 SNPs were genotyped on 9q32-24 and 805 SNPs genotyped on 16p12-q22. The map was generated using NCBI dbSNP chromosome report on Build 34. Overall missing rate was 0.06%; the overall duplicate error rate was 0.05%.

FIS was analysed both as a qualitative trait with an arbitrary threshold, and as a quantitative trait, or the degree of lateral curvature. Model independent sib-pair linkage analysis was performed on the subsets (SIBPAL, S. A. G. E. v4.5).

Results:

Chromosome 9: Multipoint model-independent qualitative analysis (threshold at ten degrees) did not result in any p values of < 0.05. When the threshold was set at 30 degrees, several regions with p values of < 0.005 were observed. One region spanned 10 Mb, and coincides with the region found to be most suggestive of linkage at the 0.05 level for the quantitative analysis which was 6 Mb in length.

Discussion: The current work has significance in the stepwise confirmation and narrowing of genomic regions which are potentially meaningful in the aetiology of FIS. Stratification of the initial sample into subgroups, initially by heritability and then by threshold of disease resulted in heightened significance at specific markers demonstrating the heterogeneity of this disorder. Ultimately, the independent association of genetic loci and this disorder will enhance the ability to elucidate prognosis, counsel patients, and guide therapeutic plans.

Introduction: Classification systems in relation to scoliosis have been a hallmark for the clinician in the development of therapeutic options. The triple curve pattern with three distinct lateral curvatures of approximately equal severity has been recognised as distinct and, potentially, unique in its presentation. From a large population of families with FIS, a subpopulation of families with a triple curve pattern was evaluated in order to determine if this curve pattern is distinct on a genetic level.

Methods: With IRB approval, a sample of families with FIS (202 families, 1198 individuals) were recruited and underwent a genomic screen. The results were analysed using a model independent linkage analysis (SIBPAL). A subgroup of FIS families with at least one member having a triple curve was identified (six families, 32 individuals). After initial linkage analysis, the group underwent further fine mapping analysis utilising a battery of SNPs.

Results: Analysis of the data from the genomic screen on the triple curve subgroup revealed significant areas on chromosome 10 when analysed qualitatively and quantitatively in either a single-point or multipoint fashion.

Conclusion: The utilization of clinical data to discern potential relevance of specific genetic loci in the aetiology of FIS has resulted in an area on chromosome 10 that is significant (p < 0.01). The relatively small population of families within this subgroup coupled with the strength of the data suggests a unique genetic etiological factor associated with the formation of a triple curve in FIS.

Recent literature has reported multiple critical regions identified through linkage analyses to be potentially relevant in relationship to the aetiology of FIS. This is supportive of the concept that FIS is a complex genetic disorder resulting from multiple genetic interactions and variants. While these areas harbour multiple genes, the work to date has been crucial to our ability to focus and hopefully eliminate massive areas on the genome that are irrelevant to this disorder. As one reviews these genes, however, one should develop a potential algorithm for prioritization of candidate genes. Additionally, one should delve into potential biological mechanisms in relationship to the creation of a spinal deformity. If you were a gene causing scoliosis, what would you look like and how would you function?

One approach to prioritization of candidate genes may be based on the virtue of their direct potential as a biological basis for the deformity, such as genes that encode for a protein of known function, the function of homologous proteins, and the tissue expression pattern. Within the localised region of chromosome 9, one such gene is COL5A1, a precursor for collagen type V alpha chains, a fibrillar forming collagen ubiquitously distributed within the connective tissues. A second group of genes may be those genes encoding regulatory proteins of the extracellular matrix.

Transmembrane 4 superfamily, member 6 (TM4SF6) localised on the critical region on Xq22 is believed to span the cellular membrane with a role in cellular adhesion within the matrix. A third group of genes may maintain a temporal and/or spatial pattern of expression that may relate to the building of the axial skeleton itself. The Iriquois genes isolated on chromosome 5 play multiple roles in embryonic development including anterior/posterior and dorsal/ventral patterning of the central nervous system. Lastly, genes that do not have an intuitive relationship to scoliosis, but are localised within areas of strong linkage, will need to undergo analysis. Multiple examples exist within the reported critical regions within the literature to date.

Another approach to the review of candidate genes within the regions is to think of known genetic disorders in which, 1) scoliosis is recognised as an element of the phenotype, and, 2) the gene and the biological mechanism of the disorder is well known. Immediate potential examples that come to mind are that of known collagen disorders such as osteogenesis imperfecta. The assumption that scoliosis is solely a result of mechanical load imposed upon abnormal connective tissue may be more elementary than what is truly occurring. Another example may be that of neurofibromatosis (gene – NF1). While this particular gene is localised near one of the identified regions, unfortunately, the biological function of the gene in relationship to phenotypic findings is still unknown.

In conclusion, genetic research related to FIS to date has driven us to unbelievable expectations within a relatively short period of time. Further understanding of this complex disease will best be accomplished with thoughtful experimental, orderly design ultimately to have an impact in the therapeutic treatment of this disorder.