The effects of GATA-1 and NF-E2 deficiency on bone biomechanical, biochemical, and mineral properties

Abstract

Mice deficient in GATA-1 or NF-E2, transcription factors required for normal megakaryocyte (MK) development, have increased numbers of MKs, reduced numbers of platelets, and a striking high bone mass phenotype. Here, we show the bone geometry, microarchitecture, biomechanical, biochemical, and mineral properties from these mutant mice. We found that the outer geometry of the mutant bones was similar to controls, but that both mutants had a striking increase in total bone area (up to a 35% increase) and trabecular bone area (up to a 19% increase). Interestingly, only the NF-E2 deficient mice had a significant increase in cortical bone area (21%) and cortical thickness (27%), which is consistent with the increase in bone mineral density (BMD) seen only in the NF-E2 deficient femurs. Both mutant femurs exhibited significant increases in several biomechanical properties including peak load (up to a 32% increase) and stiffness (up to a 13% increase). Importantly, the data also demonstrate differences between the two mutant mice. GATA-1 deficient femurs break in a ductile manner, whereas NF-E2 deficient femurs are brittle in nature. To better understand these differences, we examined the mineral properties of these bones. Although none of the parameters measured were different between the NF-E2 deficient and control mice, an increase in calcium (21%) and an increase in the mineral/matrix ratio (32%) was observed in GATA-1 deficient mice. These findings appear to contradict biomechanical findings, suggesting the need for further research into the mechanisms by which GATA-1 and NF-E2 deficiency alter the material properties of bone.

Fig. 1

Femoral geometry measured in GATA-1 deficient, NF-E2 deficient, and control mice. Mean values for midfemoral (A) total cross-sectional area inside the periosteal envelope (CSA, mm²), (B) cortical and trabecular bone area within this same envelope (BA, mm²), (C) cortical bone area (Cort BA, mm²), and (D) cortical thickness (Cort Th, μm) were assessed by μCT. Open boxes represent control bones and cross-hatched boxes represent mutant bones (GATA-1 or NF-E2 deficient). Error bars represent the SEM (n = 9–15). *Significant difference from respective control, P < 0.05.

Fig. 2

Femoral biomechanical properties for GATA-1 deficient, NF-E2 deficient, and control mice. Mean values for midfemoral (A) peak load (N), (B) stiffness (N/mm), and (C) maximum stress (MPa) were assessed by 3-point bending analysis. Open boxes represent control bones and cross-hatched boxes represent mutant bones. Error bars represent the SEM (n = 9–15). *Significant difference from respective control, P < 0.05.

Fig. 3

Load versus deflection curve for GATA-1 deficient, NF-E2 deficient, and control mice. The first point is at yield, the second point is at peak, and the final point is at failure.

Fig. 4

Mean whole body and femoral BMD in GATA-1 deficient, NF-E2 deficient, and control mice were assessed by peripheral DEXA. Open boxes represent control bones and cross-hatched boxes represent mutant bones. Error bars represent the SEM (n = 9–15). *Significant difference from respective control, P < 0.05.

Fig. 5

Representative X-ray diffractograms of GATA-1 deficient (A) and control (B) mouse bone powder (black trace) compared to a synthetic inorganic hydroxyapatite of known composition (red trace). Diffraction reflections from the powders correspond to that of a poorly crystalline apatite (reflections are broader than those of the synthetic hydroxyapatite).