D-amino acid inhibits biofilm but not new bone formation in an ovine model

Abstract

Background: Infectious complications of musculoskeletal trauma are an important factor contributing to patient morbidity. Biofilm-dispersive bone grafts augmented with D-amino acids (D-AAs) prevent biofilm formation in vitro and in vivo, but the effects of D-AAs on osteocompatibility and new bone formation have not been investigated.

Questions/purposes: We asked: (1) Do D-AAs hinder osteoblast and osteoclast differentiation in vitro? (2) Does local delivery of D-AAs from low-viscosity bone grafts inhibit new bone formation in a large-animal model?

Methods: Methicillin-sensitive Staphylococcus aureus and methicillin-resistant S aureus clinical isolates, mouse bone marrow stromal cells, and osteoclast precursor cells were treated with an equal mass (1:1:1) mixture of D-Pro:D-Met:D-Phe. The effects of the D-AA dose on biofilm inhibition (n = 4), biofilm dispersion (n = 4), and bone marrow stromal cell proliferation (n = 3) were quantitatively measured by crystal violet staining. Osteoblast differentiation was quantitatively assessed by alkaline phosphatase staining, von Kossa staining, and quantitative reverse transcription for the osteogenic factors a1Col1 and Ocn (n = 3). Osteoclast differentiation was quantitatively measured by tartrate-resistant acid phosphatase staining (n = 3). Bone grafts augmented with 0 or 200 mmol/L D-AAs were injected in ovine femoral condyle defects in four sheep. New bone formation was evaluated by μCT and histology 4 months later. An a priori power analysis indicated that a sample size of four would detect a 7.5% difference of bone volume/total volume between groups assuming a mean and SD of 30% and 5%, respectively, with a power of 80% and an alpha level of 0.05 using a two-tailed t-test between the means of two independent samples.

Results: Bone marrow stromal cell proliferation, osteoblast differentiation, and osteoclast differentiation were inhibited at D-AAs concentrations of 27 mmol/L or greater in a dose-responsive manner in vitro (p < 0.05). In methicillin-sensitive and methicillin-resistant S aureus clinical isolates, D-AAs inhibited biofilm formation at concentrations of 13.5 mmol/L or greater in vitro (p < 0.05). Local delivery of D-AAs from low-viscosity grafts did not inhibit new bone formation in a large-animal model pilot study (0 mmol/L D-AAs: bone volume/total volume = 26.9% ± 4.1%; 200 mmol/L D-AAs: bone volume/total volume = 28.3% ± 15.4%; mean difference with 95% CI = -1.4; p = 0.13).

Conclusions: D-AAs inhibit biofilm formation, bone marrow stromal cell proliferation, osteoblast differentiation, and osteoclast differentiation in vitro in a dose-responsive manner. Local delivery of D-AAs from bone grafts did not inhibit new bone formation in vivo at clinically relevant doses.

Clinical relevance: Local delivery of D-AAs is an effective antibiofilm strategy that does not appear to inhibit bone repair. Longitudinal studies investigating bacterial burden, bone formation, and bone remodeling in contaminated defects as a function of D-AA dose are required to further support the use of D-AAs in the clinical management of infected open fractures.

Fig. 1

A schematic of a femoral condyle plug defect is shown.

Fig. 2A–F

The images show inhibition and dispersion of bacterial biofilm formation of a representative clinical isolate of methicillin-resistant Staphylococcus aureus (MRSA) (strain SAMMC 41) after exposure to (A) 0 mmol/L d-AAs and (B) 6.75 mmol/L or more d-AAs. Quantification of inhibition of established biofilms from clinical isolates of (C) methicillin-susceptible S aureus (MSSA) and (D) MRSA after exposure to d-AAs shows dose-responsive behavior in a range of 0 to 81 mmol/L. Quantification of dispersion of established biofilms from clinical isolates of (E) MSSA and (F) MRSA after exposure to d-AAs shows dose-responsive behavior with a range of 0 to 81 mmol/L. The values are representative of the mean ± SD of four samples. **, ***, and **** = significant differences compared with the control (p < 0.01, p < 0.001, and p < 0.0001, respectively) as determined by one-way ANOVA and Tukey’s multiple comparison test, post hoc. CV = crystal violet; OD = optical density.

Fig. 3A–H

The dose-dependent inhibitory effect of d-amino acids (d-AAs) on bone marrow osteoblast differentiation was shown by the numbers of (A) alkaline phosphatase-positive colony forming units (CFU-ALP) and (B) osteoblast colony-forming units (CFU-OB) in osteogenic medium supplemented with d-AA concentrations from 0 to 81 mmol/L (values are reported as mean ± SD from three independent repeats with duplicates per each repeat). The osteoblast-specific gene expression of (C) Col1a1 and (D) Ocn (values are reported as mean ± SD from three independent repeats) is shown. The dose-dependent inhibitory effect of d-AAs on bone marrow stromal cell (BMSC) proliferation was shown by (E) cell number (OD₅₇₀) of BMSCs treated with d-AAs (0–81 mmol/L) for 6 days (reported as mean ± SD from five independent repeats); and (F) bromodeoxyuridine (BrdU) incorporation (determined by OD₄₅₀) of BMSCs treated with a series of d-AAs (reported as mean ± SD from three independent repeats with duplicates per each repeat). The dose-dependent inhibitory effect of d-AAs on osteoclast differentiation was shown by (G) representative images and (H) corresponding percentage tartrate resistant acid phosphatase (TRAP)-positive surface area of osteoclast cultures treated with d-AA concentrations 0 to 81 mmol/L (reported as mean ± SD from four independent repeats with four images per treatment taken for each repeat). *, **, ***, and **** = significant differences compared with the control (p < 0.05, p < 0.01, p < 0.001, and p < 0.0001, respectively) as determined by one-way ANOVA and Tukey’s multiple comparison test, post hoc. Ctr = control; OD = optical density; Col1a1 = collagen, type I, alpha 1; HPRT = hypoxanthine phosphoribosyl transferase; Ocn = osteocalcin.

Fig. 4A–G

Representative μCT three-dimensional reconstructions (scale bars = 2 mm) for the (A) low-viscosity (LV) graft before in vivo implantation and defects filled with (B) LV or (C) LV + d-amino acids (d-AAs) at 16 weeks show similar morphologic features throughout the defect site and increased density at the graft-host bone interface. The morphologic parameters, (D) bone volume/total volume (BV/TV), (E) trabecular separation (Tb.Sp), (F) trabecular number (Tb.N), and (G) trabecular thickness (Tb.Th) plotted versus the mean radius of the region of interest (R_m) show minimal differences between defects filled with low viscosity graft or low viscosity + d-AAs. Plotted controls included the mean of each parameter for native femur trabecular bone and low-viscosity graft before in vivo implantation (T0). Values are mean ± SD of four samples. Statistical significance was determined by two-way ANOVA and Tukey’s multiple comparison test, post hoc.

Fig. 5A–F

Half-view (of entire slide analyzed) low- (×2) magnification images of histologic cross-sections of defects filled with (A) low-viscosity (LV) or (B) LV+ d-amino acids (d-AAs) at 16 weeks show active remodeling. Sections were stained with Stevenel’s Blue and Van Gieson. Corresponding high- (×40) magnification images of highlighted portions of defects filled with (C) LV or (D) LV+ d-AAs show residual MASTERGRAFT^® (MG) ceramic particles, new bone (NB), and vascular development (arrows). Area % (E) new bone and (F) MG at four regions in the defect measured by histomorphometric analysis show minimal differences in new bone formation or MASTERGRAFT^® degradation between LV and LV+ d-AAs. Values are mean ± SD of eight samples per region. Statistical significance was determined by two-way ANOVA and Tukey’s multiple comparison test, post hoc.