Hao1 Is Not a Pathogenic Factor for Ectopic Ossifications but Functions to Regulate the TCA Cycle In Vivo

Abstract

Ossification of the posterior longitudinal ligament (OPLL), a disease characterized by the ectopic ossification of a spinal ligament, promotes neurological disorders associated with spinal canal stenosis. While blocking ectopic ossification is mandatory to prevent OPLL development and progression, the mechanisms underlying the condition remain unknown. Here we show that expression of hydroxyacid oxidase 1 (Hao1), a gene identified in a previous genome-wide association study (GWAS) as an OPLL-associated candidate gene, specifically and significantly decreased in fibroblasts during osteoblast differentiation. We then newly established Hao1-deficient mice by generating Hao1-flox mice and crossing them with CAG-Cre mice to yield global Hao1-knockout (CAG-Cre/Hao1flox/flox; Hao1 KO) animals. Hao1 KO mice were born normally and exhibited no obvious phenotypes, including growth retardation. Moreover, Hao1 KO mice did not exhibit ectopic ossification or calcification. However, urinary levels of some metabolites of the tricarboxylic acid (TCA) cycle were significantly lower in Hao1 KO compared to control mice based on comprehensive metabolomic analysis. Our data indicate that Hao1 loss does not promote ectopic ossification, but rather that Hao1 functions to regulate the TCA cycle in vivo.

Keywords: ectopic ossification; hydroxyacid oxidase 1; ossification of the posterior longitudinal ligament; tricarboxylic acid cycle.

Figure 1

Hao1 expression was downregulated during osteoblastic differentiation of fibroblasts. Expression of Alp and eight candidate genes in mouse embryonic fibroblasts cultured in osteoblastic differentiation conditions (namely, in the presence or absence of 50 ng/mL of BMP2) for 24 or 48 h, as analyzed by real-time PCR. Data represent mean values of indicated parameters ± S.D. (* p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 2

Establishment and characterization of Hao1 conditional knockout models. (A–D) We generated a targeting vector in which Hao1 exon 2 was flanked by loxP sequences and used it to transduce ES cells by homologous recombination (A). Neomycin-resistant colonies were picked and homologous recombination at the target site confirmed by Southern blot analysis of Sca1-digested genomic DNA. Blots were hybridized to either a 5′ or 3′ probe, identifying respective 9.4 kb or 5.2 kb fragments (B). The neomycin-resistant cassette flanked by frt sequences was deleted using flippase enzyme (C,D), and resultant ES cells were implanted into surrogate mothers. Hao1-flox F1 chimeric mice were obtained and bred to generate homozygous Hao1-flox mice. (E) Aorta, kidney, bone and liver were collected from ten-week-old WT and Hao1-deleted CAG-Cre/Hao1^flox/flox (KO) mice, and Hao1 mRNA expression was analyzed by real-time PCR. Shown is relative Hao1 expression in indicated tissues compared with that in wild-type aorta ± S.D. (F) Liver was collected from ten-week-old WT and Hao1-deleted CAG-Cre/Hao1flox/flox mice and Hao1 protein expression was analyzed by Western blotting. Red arrow symbols means the direction of each primer.

Figure 3

Body weight, micro CT, bone morphology, and histological evaluation. Eight-week-old Hao1-deleted CAG-Cre/Hao1^flox/flox (KO) mice were fed a high-phosphate diet for 2 weeks, and sacrificed at 10 weeks of age. (A) Comparison of body weights of male and female WT and KO mice at 10 weeks of age. Data represent mean values of indicated parameters ± S.D. (* p < 0.05; *** p < 0.001). (B,C) Micro CT analysis of sagittal (B) and axial (C) slices of spinal column. Scale bar = 2 mm. (D) Micro CT analysis of sagittal slice of tibia. Scale bar = 2 mm. (E–L) Bone morphology analysis of distal femoral metaphysis. (M–O) Von Kossa-stained images of kidney (M) and aorta (N) and of HE-stained liver (O). Scale bars = 100 µm.

Figure 4

Metabolome analysis of urine from ten-week-old Hao1 KO and wild-type mice. Data represent mean values of indicated parameters ± S.D. (* p < 0.05; ** p < 0.01; *** p < 0.001).