Pathway-based genome-wide association analysis identified the importance of regulation-of-autophagy pathway for ultradistal radius BMD

Abstract

Wrist fracture is not only one of the most common osteoporotic fractures but also a predictor of future fractures at other sites. Wrist bone mineral density (BMD) is an important determinant of wrist fracture risk, with high heritability. Specific genes underlying wrist BMD variation are largely unknown. Most published genome-wide association studies (GWASs) have focused only on a few top-ranking single-nucleotide polymorphisms (SNPs)/genes and considered each of the identified SNPs/genes independently. To identify biologic pathways important to wrist BMD variation, we used a novel pathway-based analysis approach in our GWAS of wrist ultradistal radius (UD) BMD, examining approximately 500,000 SNPs genome-wide from 984 unrelated whites. A total of 963 biologic pathways/gene sets were analyzed. We identified the regulation-of-autophagy (ROA) pathway that achieved the most significant result (p = .005, q(fdr) = 0.043, p(fwer) = 0.016) for association with UD BMD. The ROA pathway also showed significant association with arm BMD in the Framingham Heart Study sample containing 2187 subjects, which further confirmed our findings in the discovery cohort. Earlier studies indicated that during endochondral ossification, autophagy occurs prior to apoptosis of hypertrophic chondrocytes, and it also has been shown that some genes in the ROA pathway (e.g., INFG) may play important roles in osteoblastogenesis or osteoclastogenesis. Our study supports the potential role of the ROA pathway in human wrist BMD variation and osteoporosis. Further functional evaluation of this pathway to determine the mechanism by which it regulates wrist BMD should be pursued to provide new insights into the pathogenesis of wrist osteoporosis.

Fig. 1

–Log10 p values of all the 963 pathways with UD BMD. The arrow points to the ROA pathway. It should to be noted that there are three pathways (marked with *) whose nominal p values were less than that of the ROA pathway. However, after multiple-testing correction, these three pathways do not meet the significance criteria, that is, both FDR and FWER ≤ 0.05.

Fig. 2

Running-sum plot for the ROA pathway, including the location of the maximum enrichment score (ES) and the leading-edge subset. The x axis is the rank of the 25 genes in the ROA pathway in the whole gene list generated by ranking all the genes by their association significance with wrist BMD from the largest to smallest. These 25 genes are, from left to right, IFNA14, ATG12, IFNA21, IFNA5, IFNA17, IFNA8, IFNA16, IFNA4, IFNA7, ATG7, IFNA13, PIK3C3, ATG5, IFNG, IFNA1, GABARAPL1, IFNA2, ULK2, ATG3, IFNA10, BECN1, PRKAA2, PIK3R4, IFNA6, and PRKAA1. The y axis represents the running enrichment score. ES is the maximum deviation from zero achieved in the running-sum walk. Owing to a large number of genes ranked at the top of the whole gene list genome-wide, this pathway achieved a high ES (ES = 0.557). The bar plot at the top of the figure also shows the distribution of the 25 genes from the ROA pathway in the whole gene list, with the leading-edge genes indicated.

Fig. 3

Autophagy and the functional interactions among the genes in the ROA pathway. Basically, the autophagic process can be divided into three stages: vesicle nucleation, vesicle elongation and completion, and vesicle breakdown and degradation. ATG1, ULK, ATG6 (Beclin1), PIK3R4 (VPS15), and lipid kinase PIK3C3 (VPS34) are key proteins that mediate vesicle nucleation. AMPK and IFNs are important upstream regulators of autophagy. Many other ATG proteins, such as ATG3, ATG5, ATG7, ATG8, and ATG12, play important roles in elongation of the phagophore and formation of the autophagosome.