Identification of PDXDC1 as a novel pleiotropic susceptibility locus shared between lumbar spine bone mineral density and birth weight

Abstract

An increasing number of epidemiological studies have suggested that birth weight (BW) may be a determinant of bone health later in life, although the underlying genetic mechanism remains unclear. Here, we applied a pleiotropic conditional false discovery rate (cFDR) approach to the genome-wide association study (GWAS) summary statistics for lumbar spine bone mineral density (LS BMD) and BW, aiming to identify novel susceptibility variants shared between these two traits. We detected 5 novel potential pleiotropic loci which are located at or near 7 different genes (NTAN1, PDXDC1, CACNA1G, JAG1, FAT1P1, CCDC170, ESR1), among which PDXDC1 and FAT1P1 have not previously been linked to these phenotypes. To partially validate the findings, we demonstrated that the expression of PDXDC1 was dramatically reduced in ovariectomized (OVX) mice in comparison with sham-operated (SHAM) mice in both the growth plate and trabecula bone. Furthermore, immunohistochemistry assay with serial sections showed that both osteoclasts and osteoblasts express PDXDC1, supporting its potential role in bone metabolism. In conclusion, our study provides insights into some shared genetic mechanisms for BMD and BW as well as a novel potential therapeutic target for the prevention of OP in the early stages of the disease development. KEY MESSAGES : We investigated pleiotropy-informed enrichment between LS BMD and BW. We identified genetic variants related to both LS BMD and BW by utilizing a cFDR approach. PDXDC1 is a novel pleiotropic gene which may be related to both LS BMD and BW. Elevated expression of PDXDC1 is related to higher BMD and lower ratio n-6/n-3 PUFA indicating a bone protective effect of PDXDC1.

Keywords: Birth weight; Bone mineral density; Conditional false discovery rate; Osteoporosis; Pleiotropy.

Fig. 1

Q-Q plots. Stratified QQ plots of nominal versus empirical − log10 p values for a LS BMD as a function of significance of the association with BW, and b BW as a function of significance of the association with LS BMD. The level of − log10(p) > 0, − log10(p) > 1, − log10(p) > 2, − log10(p) > 3, − log10(p) > 4 correspond to p < 1, p < 0.1, p < 0.01, p < 0.001, p < 0.0001, respectively

Fig. 2

Manhattan plots. a Conditional Manhattan plot of conditional − log₁₀ FDR values for LS BMD given BW (LS BMD|BW). b Conditional Manhattan plot of conditional − log₁₀ FDR values for BW given LS BMD (BW|LS BMD). c Conjunction Manhattan plot of conjunction − log₁₀ FDR values for LS BMD and BW. The red line corresponds to conditional − log₁₀ FDR value of 1.3 (cFDR < 0.05)

Fig. 3

Protein–protein interaction network. Protein–protein interaction network for cFDR-significant pleiotropic susceptibility genes. Connections were based on evidence with an interaction score ≥ 0.40. Network nodes represent proteins produced by the corresponding genes, edges between nodes indicate protein–protein associations, edge color indicates the type of interaction and is specified on the bottom of the figure

Fig. 4

Visualization of posterior probability plots. We present a posterior probability of each SNP, b Z-score for BW, c Z-score for LS BMD, and d LD matrix

Fig. 5

Partially validation of PDXDC1. a Micro-computed tomography scanner assessment of trabecular bone microstructural parameters. Three-dimensional microstructural of trabecular bone of SHAM group and OVX group by micro-CT scanner at 16 weeks after operation. Scale bar, 100 μm. b Morphological analysis between SHAM group and OVX group of BV/TV, Tb. Th, Tb. N, and Tb. Sp. n = 4 per group. Data represent mean ± SD. *p < 0.05, **p < 0.01. c Immunostaining and quantification analysis of PDXDC1 between SHAM group and OVX group at growth plate and trabecula bone. Higher expression of PDXDC1 is related to higher BMD. Scale bar, 20 μm. n = 4 per group. Data represent mean ± SD. ****p < 0.0001. d, e Immunohistochemistry assay with serial sections of SHAM group. Both osteoclasts and osteoblasts express PDXDC1. Scale bar of d, 50 μm. Scale bar of e, 20 μm. f Immunostaining of PDXDC1 of transgenic (TG) mice and wild-type (WT) mice. Higher expression of PDXDC1 is related to lower ratio n-6/n-3 PUFA. Scale bar, 50 μm

Fig. 6

fat-1 TG mice identification. PCR genotyping identification with fat-1 fragment-specific primers. Lines 2, 3, and 4 indicated fat-1 gene expression; lines 1 and 5 indicated wild-type (WT) gene expression