Therapeutic Effects of HIF-1α on Bone Formation around Implants in Diabetic Mice Using Cell-Penetrating DNA-Binding Protein

Abstract

Patients with uncontrolled diabetes are susceptible to implant failure due to impaired bone metabolism. Hypoxia-inducible factor 1α (HIF-1α), a transcription factor that is up-regulated in response to reduced oxygen during bone repair, is known to mediate angiogenesis and osteogenesis. However, its function is inhibited under hyperglycemic conditions in diabetic patients. This study thus evaluates the effects of exogenous HIF-1α on bone formation around implants by applying HIF-1α to diabetic mice and normal mice via a protein transduction domain (PTD)-mediated DNA delivery system. Implants were placed in the both femurs of diabetic and normal mice. HIF-1α and placebo gels were injected to implant sites of the right and left femurs, respectively. We found that bone-to-implant contact (BIC) and bone volume (BV) were significantly greater in the HIF-1α treated group than placebo in diabetic mice (p < 0.05). Bioinformatic analysis showed that diabetic mice had 216 differentially expressed genes (DEGs) and 21 target genes. Among the target genes, NOS2, GPNMB, CCL2, CCL5, CXCL16, and TRIM63 were found to be associated with bone formation. Based on these results, we conclude that local administration of HIF-1α via PTD may boost bone formation around the implant and induce gene expression more favorable to bone formation in diabetic mice.

Keywords: angiogenesis; bone formation; diabetes mellitus; hypoxia-inducible factor-1α; osteogenesis; protein transduction domain.

Figure 1

Representative images of undecalcified specimens of four groups. (a) and (e): Abundant and well-developed new bone (NB) formation observed around the implant in the normal mice with HIF-1α gel (NH) group, with some soft tissue (ST) engagement observed; (b) and (f): Thin and well-defined new bone formation observed in the normal mice with placebo gel (NP) group; (c) and (g): Plentiful vascular sinusoids with red blood cells (white arrow) and newly-formed bone surrounding the implant in the diabetic mice with HIF-1α gel (DH) group; (d) and (h): Fibrotic and adipose tissue (asterisk) surrounding the implant in the diabetic mice with placebo gel (DP) group.

Figure 2

Histomorphometric analysis. (a): Linear percentage of direct BIC in the total surface of implants. (b): Percentage of newly formed bone area in the circumferential zone within 100 μm of the implant surface. Data represent mean ± SD. *: p-value < 0.05 vs all other groups; #: p < 0.05 vs NH group.

Figure 3

GAL4 DNA binding domain (G4D) was cloned into pRSETB plasmid expression vector containing Hph-1-PTD sequence at the N-terminus (Hph-1-G4D). The HIF-1α gene was inserted into pEGFP-N1 UAS plasmid containing five consensus GAL4 binding sites (HIF-1α-UAS plasmid). HIF-1α-UAS plasmid and Hph-1-G4D were mixed at a 1:50 mass ratio. The mixture and liquid form of Matrigel^® were blended at a 1:1 ratio just before application for bone regeneration.

Figure 4

Surgical procedure of the implant. (a) Anesthetized mouse, (b) skin preparation, (c) incision, (d) preparation of the implant site, (e) gel injection, (f) placement of the implant, (g) closure of the surgical site layer by layer, (h) post surgery.