Genome-wide Association Study For Osteoporosis Related Phenotypes

Osteoporosis (OP) is a systemic skeletal disorder disease characterized by low bone mass and micro-architectural deterioration in bone tissue, with a consequent increase in bone fragility and susceptibility to fracture, which mainly appeared in elderly people and impacted on a loss of mobility and quality of life. Osteoporosis is a common, complex disease which is influenced by both genetic and environmental factors, and genetic factors account for most of contributions. The pathogenesis of OP is complex. Previous studies found that OP was determined by many minor effect genes which commonly interact with each other to form a complex network. In addition, OP shared some genetic factors with other complex diseases, such as obesity, diabetes and so on. There are many risk factors for OP, such as bone mineral density (BMD) and bone size (BS) which are the main phenotypes used in the genetic study of OP. Based on linkage and association analyses, many genetic factors of OP have been identified, however many of them are still unknown. At present, the most useful method for genetic factors identifying is univariate genome wide association study (GWAS) which can detect the risk factors of diseases among the whole genome. However, univariate analysis ignores the gene-gene interactions and the pleiotropic effect of genes which play important roles in disease pathogenesis. To identify those factors, in the study, we used four methods (bivariate association analysis, pathway analysis, gene-gene interaction analysis and copy number variation association analysis) to analyze our GWAS date sets.1. Bivariate GWAS for BMD and BMI (body mass index). OP and obesity are two closely related diseases. They may share some genetic factors which had been little explored by studies. Genome wide bivariate association analysis is a recently developed method which can detect the association between one marker and two phenotypes. Here, based on this method, we performed the first bivariate GWAS in Chinese. We genotyped 1627 unrelated Chinese adults and conducted the bivariate GWAS for hip BMD and BMI using R software. A total of 86 SNPs represent 37 genes were found strongly associated with the phenotype pair of BMI and hip BMD reached the bivariate P value level of P<1×10-4. The top 5 SNPs showing the most significant signal are rs377046 (P=4.52×10-6), rs2166856 (P=5.49×10-6), rs4582677 (P=9.29×10-6), rs4593399 (P=9.52×10-6) and rs3862738 (P=9.97×10-6). Among them, rs377046, rs3862738 and rs4593399 are locating in three haplotype blocks respectively. We replicated the top 5 SNPs and another 31 SNPs among those three haplotype blocks in 2286 unrelated Caucasians and found 12 SNPs near the tumor necrosis factor (ligand) superfamily, member 11 (TNFSF11) gene were successfully replicated. Currently, some evidence suggested that the TNFSF11 gene was functionally related to both obesity and osteoporosis. This is the first to discover that the TNFSF11 gene maybe a pleiotropic gene for both OP and obesity.2. Pathway-based GWAS for BSAmong the whole genome, genes always interact with each other and constitute the pathways which play important roles in the pathogenesis of complex diseases. However, studies for pathway association analysis are rare. BS is an important OP-related phenotype and its genetic factors are largely unknown yet. In the study, we performed the first pathway-based GWAS for BS based on a newly developed pathway analysis method. After initial and replicate studies, two pathways (Apoptotic processes pathway and Regulation of growth pathway) were found significantly associated with spine BS in both Chinese and Caucasian. Some evidence suggested that the pathways and some pathway genes, such as Interferon-gamma gene (IFNG) and High-mobility group AT-hook 2 gene (HMGA2), were functionally related to bone metabolism.3. Pathway-based gene-gene interaction analysis for BMDRecently, little of study pay attention to gene-gene interaction, the phenomenon of two non-allelic co-regulate of the same phenotype, which is an important risk factor for the polygenic diseases. Here, we analyzed the gene-gene interaction for hip BMD based on the pathway analysis results. Three samples include 26 gene expression subjects,1627 Chinese and 2286 Caucasians were selected for the study. We first identified the pathway related to hip BMD, and then the gene-gene interactions were tested among the genes in the same identified pathway. After initial and replicate studies, only the interaction that between the solute carrier family 1, member 7 (SLC1A7) gene and the ubiquitin carboxyl-terminal hydrolase 24 (USP24) gene was found associated with hip BMD in both Chinese and Caucasians with the initial and replicate P values of 7.82×10-5 and 0.04 respectively. Both of above two genes are correlated with the nuclear factor kappa?which is a key factor in bone metabolism.4. Genome wide copy number variation association (CNV) study for height.Height, mainly determined by genetic factors, is an important covariate in genetic study of OP. CNV plays important roles in the pathogenesis of many diseases. In the study, we tested the association between human height and CNV at the first time in 618 unrelated Chinese genotyped by Affymetrix 500K SNP array, and found four CNVs were borderline association with height.In the study, using the newly developed methods, we found that TNFSF11 was co-regulated of OP and obesity, two pathways were association with BS, the interaction between SLC1A7 and USP24 was related to hip BMD, and 4 CNVs were correlated with height. The study improved our knowledge of OP and will benefit for future similar studies.