Central IL-1 receptor signaling regulates bone growth and mass

Abstract

The proinflammatory cytokine IL-1, acting via the hypothalamic IL-1 receptor type 1 (IL-1RI), activates pathways known to suppress bone formation such as the hypothalamo pituitary-adrenocortical axis and the sympathetic nervous system. In addition, peripheral IL-1 has been implicated as a mediator of the bone loss induced by sex hormone depletion and TNF. Here, we report an unexpected low bone mass (LBM) phenotype, including impairment of bone growth, in IL-1RI-deficient mice (IL-1rKO mice). Targeted overexpression of human IL-1 receptor antagonist to the central nervous system using the murine glial fibrillary acidic protein promoter (IL-1raTG mice) resulted in a similar phenotype, implying that central IL-1RI silencing is the causative process in the LBM induction. Analysis of bone remodeling indicates that the process leading to the LBM in both IL-1rKO and IL-1raTG is characterized mainly by doubling the osteoclast number. Either genetic modification does not decrease testosterone or increase corticosterone serum levels, suggesting that systems other than the gonads and hypothalamo pituitary-adrenocortical axis mediate the central IL-1RI effect on bone. We further demonstrate that WT mice express mouse IL-1ra in bone but not in the hypothalamus. Because low levels of IL-1 are present in both tissues, it is suggested that skeletal IL-1 activity is normally suppressed, whereas central IL-1 produces a constant physiologic stimulation of IL-1RI signaling. Although the pathway connecting the central IL-1RI signaling to bone remodeling remains unknown, the outburst of osteoclastogenesis in its absence suggests that normally it controls bone growth and mass by tonically restraining bone resorption.

Fig. 1.

Skeletal growth impairment and low bone mass phenotype in IL-1rKO mice. (A and B) Qualitative (2D images) and quantitative μCT analysis showing reduction in femoral length and decreased trabecular bone volume density (BV/TV, volume of trabecular network as percentage of total reference volume) in distal femoral metaphysis (A) and L3 vertebra (B). (C) Body weight. Data are mean ± SE obtained in 10 animals per condition. ^*, t test; P < 0.05.

Fig. 2.

Central and skeletal expression of IL-1ra. RT-PCR analysis using primers for human (hIL-1ra) and mouse (mIL-1ra) cDNA. (A) WT mice; similar results were obtained in six animals. (B) IL-1raTG mice; similar results were obtained in six animals. Numbers indicate representative animals used for mRNA extraction.

Fig. 3.

Central IL-1RI signaling regulates longitudinal and radial bone growth. (A and B) Qualitative (3D) and quantitative μCT analysis showing reduction in femoral length (A), in diaphyseal diameter (Dia.Dia) and medullary cavity diameter (Med.Dia) (B) in IL-1raTG mice. Shown are images obtained from mice with median length (A) and diaphyseal diameter (B). Dashed lines in A indicate position of diaphyseal segments shown in B.(C) Body weight in 5-week-old (Upper) and 15-week-old (Lower) mice. Data are mean ± SE in five 5-week-old and twelve 15-week-old mice per condition. ^*, t test; P < 0.05.

Fig. 4.

Low trabecular bone mass phenotype in IL-1raTG mice. (A) Qualitative (3D) μCT analysis. (B and C) μCT analysis of trabecular bone volume density (BV/TV, volume of trabecular network as percentage of total reference volume) in L3 vertebra (B) and distal femoral metaphysis (C). Data are mean ± SE in five 5-week-old and twelve 15-week-old mice per condition. ^*, t test; P < 0.05.

Fig. 5.

Elevated osteoclast number in IL-1rKO mice. Shown is a histomorphometric analysis. (A) Osteoclast number, surrogate of bone resorption activity, calculated as the number of TRAP-positive osteoclasts per trabecular surface area. (B-D) Measurements based on fluorescent calcein labeling of newly formed bone. (B) BFR, surrogate of total bone formation activity, calculated as the arithmetic product of MAR and Min.Peri. (C) Min.Peri, surrogate of osteoblast number, calculated as the percentage of calcein-labeled trabecular surface. (D) MAR, surrogate of osteoblast activity, calculated as the average distance between two calcein-labeled lines divided by the time interval between the two calcein injections. Data are mean ± SE in nine animals per condition. ^*, t test; P < 0.05.

Fig. 6.

High trabecular bone turnover in IL-1raTG mice. Shown is a histomorphometric analysis in 15-week-old mice. (A) Osteoclast number, surrogate of bone resorption activity, calculated as the number of TRAP-positive osteoclasts per trabecular surface area. (B-D) Measurements based on fluorescent calcein labeling of newly formed bone. (B) BFR, surrogate of total bone formation activity, calculated as the arithmetic product of MAR and Min.Peri. (C) Min.Peri, surrogate of osteoblast number, calculated as the percent calcein-labeled trabecular surface. (D) MAR, surrogate of osteoblast activity, calculated as the average distance between two calcein-labeled lines divided by the time interval between the two calcein injections. Data are mean ± SE in five to six animals per condition. ^*, t test; P < 0.05.