Foxk1 promotes bone formation through inducing aerobic glycolysis

Abstract

Transcription factor Foxk1 can regulate cell proliferation, differentiation, metabolism, and promote skeletal muscle regeneration and cardiogenesis. However, the roles of Foxk1 in bone formation is unknown. Here, we found that Foxk1 expression decreased in the bone tissue of aged mice and osteoporosis patients. Knockdown of Foxk1 in primary murine calvarial osteoblasts suppressed osteoblast differentiation and proliferation. Conditional knockout of Foxk1 in preosteoblasts and mature osteoblasts in mice exhibited decreased bone mass and mechanical strength due to reduced bone formation. Mechanistically, we identified Foxk1 targeted the promoter region of many genes of glycolytic enzyme by CUT&Tag analysis. Lacking of Foxk1 in primary murine calvarial osteoblasts resulted in reducing aerobic glycolysis. Inhibition of glycolysis by 2DG hindered osteoblast differentiation and proliferation induced by Foxk1 overexpression. Finally, specific overexpression of Foxk1 in preosteoblasts, driven by a preosteoblast specific osterix promoter, increased bone mass and bone mechanical strength of aged mice, which could be suppressed by inhibiting glycolysis. In summary, these findings reveal that Foxk1 plays a vital role in the osteoblast metabolism regulation and bone formation stimulation, offering a promising approach for preventing age-related bone loss.

Fig. 1. Knockdown of Foxk1 inhibited osteoblast differentiation and proliferation.

A qPCR analysis of mRNA levels of Foxk1 in femurs of 3 months (3 m) and 12m-old mice (n = 3 per group). B Western blot analysis of FOXK1, RUNX2 and OSTERIX protein from 3 m and 12m-old mice femur. C qPCR analysis of mRNA levels of Foxk1 in bone tissues of osteoporosis patients and controls (n = 6 per group). D Western blot analysis and quantification of FOXK1 from bone tissues of osteoporosis patients and controls. E qPCR analysis of mRNA levels of Foxk1 in primary murine calvarial osteoblasts at day 0, day 3 and day 7 during osteoblast differentiation (n = 3 per group). F–J qPCR was used to quantify Foxk1, Runx2, Osterix, Alp and Ocn mRNA levels after 7 days of siRNA transfection in primary murine calvarial osteoblasts (n = 3 per group). K Western blot analysis of FOXK1, RUNX2 and OSTERIX protein levels in primary murine calvarial osteoblasts after transfection with siNC or siFoxk1 for 7 days. L Representative images of ALP staining of primary murine calvarial osteoblasts after transfection with siRNA for 7 days. M Representative images of ARS staining of primary murine calvarial osteoblasts after transfection with siRNA for 14 days. N Representative images and quantification of Edu incorporation in primary murine calvarial osteoblasts after transfection with siRNA for 3 days (n = 3 per group). O Representative images and quantification of Ki67 IF staining in primary murine calvarial osteoblasts after transfection with siRNA for 3 days (n = 3 per group). *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 2. Loss of Foxk1 in preosteoblasts leads to low bone mass and strength.

Representative images of microCT analysis of male (A) and female (B) Osx-cre;Foxk1^fl/fl mice and corresponding controls including two-dimensional construction of distal femur and three-dimensional reconstruction of trabecular bone of distal femur at 4 weeks old. Quantification of microCT analyses of the femur distal end from male (C) and female (D) Osx-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (male: n = 7 for Osx-cre mice and n = 8 for Osx-cre;Foxk1^fl/fl mice; female: n = 8 per group). Biomechanical properties analysis of the left femur from male (E) and female (F) Osx-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (n = 7 per group). Dynamic osteogenic index of trabecular bone, including MAR and BFR, from the femoral metaphysis in male (G) and female (H) Osx-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (n = 3 per group). Representative images and quantification of OCN IHC staining in male (I) and female (J) Osx-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (n = 5 per group). *p < 0.05, **p < 0.01.

Fig. 3. Deletion of Foxk1 in mature osteoblasts results in low bone mass and strength.

Representative images of microCT analysis of male (A) and female (B) Ocn-cre;Foxk1^fl/fl mice and corresponding controls including two-dimensional construction of distal femur and three-dimensional reconstruction of trabecular bone of distal femur at 4 weeks old. Quantification of microCT analyses of the femur distal end from male (C) and female (D) Ocn-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (male: n = 7 per group; female: n = 8 per group). Biomechanical properties analysis of the left femur from male (E) and female (F) Ocn-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (n = 6 per group). Dynamic osteogenic index of trabecular bone, including MAR and BFR, from the femoral metaphysis in male (G) and female (H) Ocn-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (n = 3 per group). Representative images and quantification of OCN IHC staining in male (I) and female (J) Ocn-cre;Foxk1^fl/fl mice and corresponding controls at 4 weeks old (n = 5 per group). *p < 0.05, **p < 0.01.

Fig. 4. Foxk1 plays a critical role in regulating aerobic glycolysis of osteoblasts.

A KEGG analysis of direct target genes from CUT&Tag-seq of Foxk1 overexpressing primary murine calvarial osteoblasts. B IGV visual analyzed the FOXK1 biding regions of its targeted genes. C Glucose uptake is determined in Foxk1 knockdown versus control primary murine calvarial osteoblasts (n = 6 per group). D–F The extracellular acidification rate (ECAR) in primary murine calvarial osteoblasts was measured using Seahorse XF assay, in response to Foxk1 knockdown. Glycolysis (E) and glycolytic capacity (F) were analyzed (n = 6 per group). G Lactate levels of medium cultured for Foxk1 knockdown and control primary murine calvarial osteoblasts was examined (n = 4 per group). H–M Metabonomic analysis glycolytic intermediate metabolites (G6-P, F6-P, F1,6-P, 3-PG, pyruvate and lactate) in primary murine calvarial osteoblasts with Foxk1 knockdown was performed (n = 6 per group). N Mapping of Foxk1-induced metabolic genes and metabolites within glycolytic pathways. The factors decreased in primary murine calvarial osteoblasts with Foxk1 knockdown are marked in red font. *p < 0.05, **p < 0.01, ***p < 0.01.

Fig. 5. Foxk1 regulation of osteoblast differentiation and proliferation relies on glycolysis.

A–C The extracellular acidification rate (ECAR) in control (oeNC) and Foxk1 overexpressed (oeFoxk1) primary murine calvarial osteoblasts treated with PBS or 2DG was measured using Seahorse XF assay. Glycolysis (B) and glycolytic capacity (C) were analyzed (n = 5 per group). D Lactate levels of medium cultured for Foxk1 overexpressed primary murine calvarial osteoblasts and control treated with PBS or 2DG was examined (n = 6 per group). E–H mRNA levels of Runx2, Osterix, Alp, and Ocn in control and Foxk1 overexpressed primary murine calvarial osteoblasts treated with PBS or 2DG were quantified by qPCR (n = 3 per group). Representative images of ALP (I) and ARS (J) staining of control and Foxk1 overexpressed primary murine calvarial osteoblasts treated with PBS or 2DG. K, L Representative images and quantification of Edu incorporation in control and Foxk1 overexpressed primary murine calvarial osteoblasts treated with PBS or 2DG (n = 3 per group). *p < 0.05, **p < 0.01, ***p < 0.01.

Fig. 6. Foxk1 increases the bone mass of aged mice through the induction of glycolysis.

A Experimental design of the 12m-old mice treated with AAV9-Foxk1 and 2DG via tail vein injection. Femurs from different groups were collected for further analysis after 56 days of AAV9 treatment. B The IVIS-100 optical imaging system was used to monitor GFP expression in individual tissues. y-axis, radiant efficiency (p/s/cm²/sr/μW/cm²). C Representative IF staining images of GFP and Osterix in femur of 3 m and 12m-old mice treated with AAV9-GFP. D Representative images of FOXK1 IHC staining of the femurs from treated 3 m and 12m-old mice. E mRNA levels of Foxk1 in femurs from treated 3 m and 12m-old mice was quantified by qPCR (n = 3 per group). F Representative MicroCT images of two-dimensional image construction of distal femur and three-dimensional image reconstruction of trabecular bone of distal femur from treated 3 m and 12m-old mice. G–J Quantification of microCT analyses of the femur distal end from treated 3 m and 12m-old mice (n = 6 per group). K, L Biomechanical properties analysis of the left femur from treated 3 m and 12m-old mice (n = 6 per group). M–O Dynamic osteogenic index of trabecular bone, including MAR (N) and BFR (O), from the femoral metaphysis in treated 3 m and 12 m mice (n = 3 per group). *p < 0.05, **p < 0.01, ***p < 0.01.

Fig. 7. A model of Foxk1 in osteogenesis and age-related bone loss.

Foxk1 promotes transcription of several glycolytic genes and improve the level of glycolysis in osteoblasts, which induces osteoblast differentiation and proliferation and bone formation. In aged mice, the expression of Foxk1 is decreased in osteoblasts, resulting in suppression of glycolysis. Limited glycolysis weakens osteoblast differentiation and proliferation, which causes age-related bone loss.