Advanced glycation end products mediate biomineralization disorder in diabetic bone disease

Abstract

Patients with diabetes often experience fragile fractures despite normal or higher bone mineral density (BMD), a phenomenon termed the diabetic bone paradox (DBP). The pathogenesis and therapeutics opinions for diabetic bone disease (DBD) are not fully explored. In this study, we utilize two preclinical diabetic models, the leptin receptor-deficient db/db mice (DB) mouse model and the streptozotocin-induced diabetes (STZ) mouse model. These models demonstrate higher BMD and lower mechanical strength, mirroring clinical observations in diabetic patients. Advanced glycation end products (AGEs) accumulate in diabetic bones, causing higher non-enzymatic crosslinking within collagen fibrils. This inhibits intrafibrillar mineralization and leads to disordered mineral deposition on collagen fibrils, ultimately reducing bone strength. Guanidines, inhibiting AGE formation, significantly improve the microstructure and biomechanical strength of diabetic bone and enhance bone fracture healing. Therefore, targeting AGEs may offer a strategy to regulate bone mineralization and microstructure, potentially preventing the onset of DBD.

Keywords: AGEs; advanced glycation end products; aminoguandine; biomineralization; bone fracture; bone quality; collagen mineralization; diabetes bone disease; metformin; type 2 diabetes mellitus.

Graphical abstract

Figure 1

Diabetic mice show higher bone mineral content and lower bone strength (A) Representative μCT reconstruction images of the right femurs from 8- and 20-week-old WT and DB groups with pseudo color (present BMD from 0 to 1); bar, 1 mm; WT, wild-type mice; DB, db/db mice. (B) BMD of the femur shafts from WT and DB groups; n = 5 for each group; BMD, bone mineral density. (C) BV/TV of the femur shafts from WT and DB groups; n = 5 for each group; BV, bone volume; TV, tissue volume. (D) Po (tot) of the cortical femur shafts from WT and DB groups; n = 5 for each group; Po (tot), total porosity (percent). (E–H) Quantitative analysis of the right femurs from WT and DB groups in three-point bending test; n = 3 for each group. Data were represented as mean ± SD (error bars) from biological replicates; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, by two-way ANOVA with Tukey’s post-test. Those not marked showed statistically no significant differences. See also Figures S1–S5.

Figure 2

Diabetic bones show mineralization disorder (A) Representative SEM images with lower resolution of cross-section of cortical bone samples from WT 8W, DB 8W, WT 20W, and DB 20W groups; yellow hollow circles mark cortical bone pores; bar, 1 μm; WT, wild-type mice; DB, db/db mice; W, week. (B) Representative SEM images with higher resolution of cross-section of cortical bone samples from WT 8W, DB 8W, WT 20W, and DB 20W groups; yellow arrows represent the broken ends of collagen fibrils; bar, 100 nm. (C) Representative SEM images of longitudinal section of cortical bone samples from WT 8W, DB 8W, WT 20W, and DB 20W groups; both white and yellow lines mark the orientation of adjacent mineralized collagen in circumferential lamellae of cortical bone; bar, 10 μm. (D) Representative BSHG images of femur slices from WT CON, DB CON, DB AG, and DB MET groups; the dashed lines mark the edges of the cortical bone; bar, 100 μm. BL, bone lacunae. (E–G) Quantitative analysis of the BSHG images in (D); n = 5 for each group; Conn.Dn, connected density; Col.V, collagen volume; TV, tissue volume. Data were represented as mean ± SD (error bars) from biological replicates; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, by two-way ANOVA with Tukey’s post-test. Those not marked showed statistically no significant differences. See also Figures S6 and S7.

Figure 3

AGE accumulation analysis in the cortical bones of diabetic human and mice (A) Representative IHC-staining images of AGEs in human cortical bone fragments from HC and DM fracture patients; Areas stained brown represent locations of positive staining; bar, 100 μm. HC, healthy control; DM, diabetes mellitus; AGEs, advanced glycation end products. (B) Positive area of the AGEs stained in the IHC images as quantitative measurements; n = 5 for each group. (C) AGE ELISA analysis of proteins from in human cortical bone fragments of HC and DM fracture patients. n = 5 for each group. (D) Representative IHC-staining images of AGEs in right femur slices from WT and DB groups. Areas stained brown represent locations of positive staining; bar, 50 μm; WT, wild-type mice; DB, db/db mice; W, weeks. (E) Positive area of the AGEs stained in the IHC images in (D) as quantitative measurements; n = 5 for each group. (F) AGE ELISA analysis of proteins from WT and DB groups; n = 5 for each group. Data were represented as mean ± SD (error bars) from biological replicates; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, by non-paired Student’s t test for B and C and by two-way ANOVA with Tukey’s post-test for E and F. See also Figure S8.

Figure 4

AGE accumulations in HG lead to lower collagen intrafibrillar mineralization, and inhibition of AGE accumulations by guanidines increases collagen intrafibrillar mineralization (A) Schematic diagram of in vitro collagen mineralization; COL1, collagen I; LG, low glucose concentration; HG, high glucose concentration; AG, aminoguanidine; MET, metformin; d, day(s); TEM, transmission electron microscope. (B) Representative TEM images of collagen after mineralization for 3 and 7 days under LG or HG; bar, 1 μm. (C) Representative SIM images of collagen after mineralization for 7 days; bar, 100 nm; SIM, structure illumination microscopy.

Figure 5

Guanidines improve bone microstructure and mechanical properties in DB mice (A) Representative μCT reconstruction images of the right femurs from WT CON, DB CON, DB AG, and DB MET groups with pseudo color (present BMD from 0 to 1); bar, 1 mm; WT, wild-type mice; DB, db/db mice; CON, control; AG, aminoguanidine; MET, metformin. (B) BV/TV of the whole right femur shafts from WT CON, DB CON, DB AG, and DB MET groups; n = 5 for each group; BV, bone volume; TV, tissue volume. (C) BMD of the cortical femur shafts from WT CON, DB CON, DB AG, and DB MET groups; n = 5 for each group; BMD, bone mineral density. (D) Po (tot) of the cortical femur shafts from WT CON, DB CON, DB AG, and DB MET groups; n = 5 for each group; Po (tot), total porosity (percent). (E–H) Quantitative analysis of right femurs from WT CON, DB CON, DB AG, and DB MET groups in three-point bending test; n = 3 for each group. Data were represented as mean ± SD (error bars) from biological replicates; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, by one-way ANOVA with Sidak’s post-test. Multiple comparisons included “WT CON vs. DB CON,” “DB CON vs. DB AG,” “DB CON vs. DB MET,” and “DB AG vs. DB MET.” Those not marked showed statistically no significant differences. See also Figures S9 and S10.

Figure 6

Guanidines reduce AGE accumulation and improve bone mineralization and collagen arrangement in diabetic bone (A) Representative SEM images with lower resolution of cross-section of cortical bone samples from WT CON, DB CON, DB AG, and DB MET groups. Yellow hollow circles mark cortical bone pores; bar, 1 μm; WT, wild-type mice; DB, db/db mice; CON, control; AG, aminoguanidine; MET, metformin. (B) Representative SEM images with higher resolution of cross-section of cortical bone samples from WT CON, DB CON, DB AG, and DB MET groups; yellow arrows represent the broken ends of collagen fibrils; bar, 100 nm. (C) Representative SEM images of longitudinal section of cortical bone samples from WT CON, DB CON, DB AG, and DB MET groups; both white and yellow lines mark the orientation of adjacent mineralized collagen in circumferential lamellae of cortical bone; bar, 10 μm. (D) Representative BSHG images of femur slices from WT CON, DB CON, DB AG, and DB MET groups; The dashed lines mark the edges of the cortical bone; bar, 100 μm. n = 5 for each group; BL, bone lacunae. (E) Representative IHC-staining images of AGEs; Areas stained brown represent locations of positive staining; bar, 50 μm. (F–H) Quantitative analysis of the BSHG images in (D); n = 5 for each group; Conn.Dn, connected density; Col.V, collagen volume; TV, tissue volume. (I) Positive area of the AGEs stained in the IHC images in (E) as quantitative measurements; n = 5 for each group. Data were represented as mean ± SD (error bars) from biological replicates; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, by one-way ANOVA with Sidak’s post-test. Multiple comparisons included “WT CON vs. DB CON,” “DB CON vs. DB AG,” “DB CON vs. DB MET,” and “DB AG vs. DB MET.” Those not marked showed statistically no significant differences. See also Figures S11–S13.

Figure 7

Guanidines promote bone fracture healing in diabetic mice (A) Schematic diagram of fracture modeling and drug treatment patterns in mice models. (B and C) Quantitative analysis of the callus of fractured right femurs from WT CON, DB CON, DB AG, and DB MET groups; n = 5 for each group; WT, wild-type mice; DB, db/db mice; CON, control; AG, aminoguanidine; MET, metformin. (D–F) Quantitative analysis of torsion test of healed fracture right femurs from WT CON, DB CON, DB AG, and DB MET groups; n = 3 for each group. (G) Elastic modulus analysis of three-point bending test of healed fracture right femurs from WT CON, DB CON, DB AG, and DB MET groups; n = 3 for each group. (H) Positive area of the AGEs stained in the IHC images in (D) as quantitative measurements; n = 5 for each group; AGEs, advanced glycation end products. (I) Representative μCT reconstruction images of healed fracture right femurs from WT CON, DB CON, DB AG, and DB MET groups; bar, 10 mm. (J) H&E staining images of healed fracture femur slices from WT CON, DB CON, DB AG, and DB MET groups; n = 5 for each group; bar, 10 mm (above) and 1 mm (down). (K) Masson staining images of healed fracture femur slices from WT CON, DB CON, DB AG, and DB MET groups; n = 5 for each group; bar, 1 mm. (L) Representative IHC-staining images of AGEs of healed fracture femur slices from WT CON, DB CON, DB AG, and DB MET groups; areas stained brown represent locations of positive staining; bar, 1 mm. Data were represented as mean ± SD (error bars) from biological replicates; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.000, by one-way ANOVA with Sidak’s post-test. Multiple comparisons included “WT CON vs. DB CON,” “DB CON vs. DB AG,” “DB CON vs. DB MET,” and “DB AG vs. DB MET.” Those not marked showed statistically no significant differences. See also Figures S14–S16.