A Mendelian randomization study for drug repurposing reveals bezafibrate and fenofibric acid as potential osteoporosis treatments

Abstract

Background: Lipid pathways have been implicated in the pathogenesis of osteoporosis (OP). Lipid-lowering drugs may be used to prevent and treat OP. However, the causal interpretation of results from traditional observational designs is controversial by confounding. We aimed to investigate the causal association between genetically proxied lipid-lowering drugs and OP risk. Methods: We conducted two-step Mendelian randomization (MR) analyses to investigate the causal association of genetically proxied lipid-lowering drugs on the risk of OP. The first step MR was used to estimate the associations of drug target genes expression with low-density lipoprotein cholesterol (LDL-C) levels. The significant SNPs in the first step MR were used as instrumental variables in the second step MR to estimate the associations of LDL-C levels with forearm bone mineral density (FA-BMD), femoral neck BMD (FN-BMD), lumbar spine BMD (LS-BMD) and fracture. The significant lipid-lowering drugs after MR analyses were further evaluated for their effects on bone mineralization using a dexamethasone-induced OP zebrafish model. Results: The first step MR analysis found that the higher expression of four genes (HMGCR, NPC1L1, PCSK9 and PPARG) was significantly associated with a lower LDL-C level. The genetically decreased LDL-C level mediated by the PPARG was significantly associated with increased FN-BMD (BETA = -1.38, p = 0.001) and LS-BMD (BETA = -2.07, p = 3.35 × 10^-5) and was marginally significantly associated with FA-BMD (BETA = -2.36, p = 0.008) and reduced fracture risk (OR = 3.47, p = 0.008). Bezafibrate (BZF) and Fenofibric acid (FBA) act as PPARG agonists. Therefore genetically proxied BZF and FBA had significant protective effects on OP. The dexamethasone-induced OP zebrafish treated with BZF and FBA showed increased bone mineralization area and integrated optical density (IOD) with alizarin red staining. Conclusion: The present study provided evidence that BZF and FBA can increase BMD, suggesting their potential effects in preventing and treating OP. These findings potentially pave the way for future studies that may allow personalized selection of lipid-lowering drugs for those at risk of OP.

Keywords: Mendelian randomization; bezafibrate and fenofibric acid; drug repurposing; lipid-lowering drugs; osteoporosis.

FIGURE 1

Summary of Study Design. Note: eQTL, expression quantitative trait loci; SNP, single nucleotide polymorphism; MR, Mendelian randomization; BMD, bone mineral density; FA-BMD, forearm bone mineral density; FN-BMD, femoral neck bone mineral density; LS-BMD, lumbar spine bone mineral density.

FIGURE 2

Association of the genetically inferred LDL-C level with FA-BMD, FN-BMD and LS-BMD. Note: Mendelian randomization estimates for the effects of exposure to drug target modulation on Forearm bone mineral density (FA-BMD), Femoral neck bone mineral density (FN-BMD) and Lumbar spine bone mineral density (LS-BMD). SNP, single nucleotide polymorphism; CI, confidence interval.

FIGURE 3

Association of the genetically inferred LDL-C level with fracture. Note: Mendelian randomization estimates for the effects of exposure to drug target modulation on fracture. SNP, single nucleotide polymorphism; CI, confidence interval; OR, odds ratio.

FIGURE 4

The potential biological function pathways by gene set enrichment analysis using the BZF and fenofibrate -treatment gene expression profile in LINCS database.

FIGURE 5

Effect of BZF and FBA on osteoporosis induced by dexamethasone in zebrafish. Note: Zebrafish Alizarin Red Staining (A–F), (A) 0.04% Dimethylsulfoxide; (B) 10 µm Dexamethasone, (C, D) 10 µm Dexamethasone + BZF (1,10 μg/mL), (E, F) 10 µm Dexamethasone + FBA (1,10 μg/mL).

FIGURE 6

Effect of BZF and FBA on osteoporosis induced by dexamethasone in zebrafish. (A) Bone mineral area; (B) IOD bar chart; **Compared with the 0.04% DMSO group, p < 0.001; ++ Compared with the 10 µm Dexamethasone group, p < 0.001. ABS (SD): the absolute value of “±SD”.