Additive manufacturing (3D printing) has enabled fabrication of geometrically complex and fully interconnected porous biomaterials with huge surface areas that could be used for biofunctionalization to achieve multifunctional biomaterials. Covering the huge surface area of such porous titanium with nanotubes has been already shown to result in improved bone regeneration performance and implant fixation. In this study, we loaded TiO2 nanotubes with silver antimicrobial agents to equip them with an additional biofunctionality, i.e., antimicrobial behavior. An optimized anodizing protocol was used to create nanotubes on the entire surface area of direct metal printed porous titanium scaffolds. The nanotubes were then loaded by soaking them in three different concentrations (i.e., 0.02, 0.1, and 0.5 M) of AgNO3 solution. The antimicrobial behavior and cell viability of the developed biomaterials were assessed. As far as the early time points (i.e., up to 1 day) are concerned, the biomaterials were found to be extremely effective in preventing biofilm formation and decreasing the number of planktonic bacteria particularly for the middle and high concentrations of silver ions. Interestingly, nanotubes not loaded with antimicrobial agents also showed significantly smaller numbers of adherent bacteria at day 1, which may be attributed to the bactericidal effect of high aspect ratio nanotopographies. The specimens with the highest concentrations of antimicrobial agents adversely affected cell viability at day 1, but this effect is expected to decrease or disappear in the following days as the rate of release of silver ions was observed to markedly decrease within the next few days. The antimicrobial effects of the biomaterials, particularly the ones with the middle and high concentrations of antimicrobial agents, continued until 2 weeks. The potency of the developed biomaterials in decreasing the number of planktonic bacteria and hindering the formation of biofilms make them promising candidates for combating peri-operative implant-associated infections.

No abstract

Treatment and reconstruction of large bone defects, delayed unions, and nonunions is challenging and has resulted in an ongoing search for novel tissue-engineered therapies. Bone morphogenetic protein-2 (BMP-2) gene therapy is a promising strategy to provide sustained production of BMP-2 locally. Alginate polymer-based nonviral gene therapy with BMP-2 plasmid DNA (pBMP-2) in constructs with multipotent mesenchymal stromal cells (MSCs) has resulted in prolonged gene expression and bone formation in vivo. To further translate this technology toward larger animal models, important issues remain to be investigated, such as the necessity of seeded cells as a target for gene therapy. For that purpose, a large animal-screening model in an orthotopic location, with fully separated chambers, was investigated. Four cylinder-shaped implants were placed in the iliac crests of ten goats. Polycaprolactone tubes around each implant allowed bone ingrowth from the underlying bone and bone marrow and ensured separation of the experimental conditions. An empty tube showed low levels of spontaneous bone ingrowth, and implantation of autologous bone indicated proper bone function with respect to remodeling and resorption. Control ceramic scaffolds were compared to scaffolds containing pBMP-2 either or not combined with seeded MSCs. Fluorochrome incorporation evaluated at 3, 6, and 9 weeks and histomorphometry at 12 weeks after implantation revealed clear differences between the groups, with pBMP-2 combined with MSCs being the most effective. The BMP-2 was demonstrated in a variety of bone-residing cells through immunohistochemistry. Further analysis indicated that multinucleated giant cells might have an important role in transgene expression. Taken together, this work introduces a large animal model for studying bone formation at multiple sites simultaneously in an orthotopic location. The model appeared robust, showed no neighboring effects, and demonstrated effectivity of combined cell and gene therapy.

To explore the influence of inflammatory processes on bone formation, we applied a new in vivo screening model. Confined biological pockets were first created in rabbits as a response to implanted bone cement discs. These biomembrane pockets were subsequently used to study the effects of inflammatory stimuli on ectopic bone formation within biphasic calcium phosphate (BCP) constructs loaded with TNF-α, lipopolysaccharide (LPS) or lipoteichoic acid (LTA), all with or without bone morphogenetic protein (BMP)-2. Analysis of bone formation after 12 weeks demonstrated that the inflammatory mediators were not bone-inductive in combination with the BCP alone, but inhibited or enhanced BMP-induced bone formation. LPS was associated with a strong inhibition of bone formation by BMP-2, while LTA and TNF-α showed a positive interaction with BMP-2. Since the biomembrane pockets did not interfere with bone formation and prevented the leakage of pro-inflammatory compounds to the surrounding tissue, the biomembrane model can be used for in vivo approaches to study local inflammation in conjunction with new bone formation. Using this model, it was shown that the modulation of the inflammatory response could be beneficial or detrimental to the subsequent bone formation process. The co-delivery of inflammatory factors and bone-related growth factors should be further explored as a strategy to enhance the bone-forming efficacy of bone substitutes.

To induce osteogenicity in bone graft substitutes, plasmid-based expression of BMP-2 (pBMP-2) has been successfully applied in gene activated matrices based on alginate polymer constructs. Here, we investigated whether cell seeding is necessary for non-viral BMP-2 gene expression in vivo . Furthermore, to gain insight in the role of BMP-producing cells, we compared inclusion of bone progenitor cells with non-osteogenic target cells in gene delivery constructs. Plasmid DNA encoding GFP (pGFP) was used to trace transfection of host tissue cells and seeded cells in a rat model. Transgene expression was followed in both cell-free alginate-ceramic constructs as well as constructs seeded with syngeneic fibroblasts or multipotent mesenchymal stromal cells (MSCs). Titration of pGFP revealed that the highest pGFP dose resulted in frequent presence of positive host cells in the constructs. Both cell-loaded groups were associated with transgene expression, most effectively in the MSC-loaded constructs. Subsequently, we investigated effectiveness of cell-free and cell-loaded alginate-ceramic constructs with pBMP-2 to induce bone formation. Local BMP-2 production was found in all groups containing BMP-2 plasmid DNA, and was most pronounced in the groups with MSCs transfected with high concentration pBMP-2. Bone formation was only apparent in the recombinant protein BMP-2 group. In conclusion, we show that non-viral gene delivery of BMP-2 is a potentially effective way to induce transgene expression in vivo , both in cell-seeded as well as cell-free conditions. However, alginate-based gene delivery of BMP-2 to host cells or seeded cells did not result in protein levels adequate for bone formation in this setting, calling for more reliable scaffold compatible transfection methods.

Non-viral gene delivery is a safe technique to release sustained physiologic dosages of bone morphogenetic protein (BMP). Co-delivery of multiple BMPs can result in the formation of more potent BMP heterodimers. In this study, non-viral co-delivery of BMP-2/6 and BMP-2/7, as a means to produce heterodimers, was assessed.Goat MSCs were non-virally transfected with plasmid DNA encoding BMP isoforms (pBMP) known to be relevant for osteogenesis: BMP-2, -6 or -7. As a result, BMP-2, -6 and -7 were produced and detectable for up to 14 d and their combined delivery (pBMP-2 with pBMP-6 or pBMP-7) was used to create BMP-2/6 and BM-2/7 heterodimers. Formation and secretion of the heterodimer proteins was validated by sandwich enzyme-linked immunosorbent assay (ELISA). Produced BMPs and heterodimers were biologically active, as confirmed by differentiation of reporter cells and MSCs. To assess bone formation, transfected MSCs were seeded on to ceramic scaffolds and implanted subcutaneously into nude mice. Bone formation was significantly enhanced in the pBMP-2/6 condition and a trend for more bone formation was observed in the pBMP-2/7 and pBMP-6 homodimer condition. No bone was found in the pBMP-2, pBMP-7 or control condition.In conclusion, simultaneous delivery of pBMP-2 with pBMP-6 or -7 resulted in the production of heterodimers that were beneficial for bone formation as compared to BMP homodimers. Combination of BMP sequences could reduce the need for high BMP protein dosages and might enhance prolonged availability of the growth factors.

Ex vivo nonviral gene delivery of bone inductive factors has the potential to heal bone defects. Due to their inherent role in new bone formation, multipotent stromal cells (MSCs) have been studied as the primary target cell for gene delivery in a preclinical setting. The relative contribution of autocrine and paracrine mechanisms, and the need of osteogenic cells, remains unclear. This study investigates the contribution of MSCs as producer of transgenic bone morphogenetic proteins (BMPs) and to what extent the seeded MSCs participate in actual osteogenesis. Rat-derived MSCs or fibroblasts (FBs) were cotransfected with pBMP-2 and pBMP-6 or pBMP-7 via nucleofection. The bioactivity of BMP products was shown through in vitro osteogenic differentiation assays. To investigate their role in new bone formation, transfected cells were seeded on ceramic scaffolds and implanted subcutaneously in rats. Bone formation was assessed by histomorphometry after 8 weeks. As a proof of principle, we also investigated the suitability of bone marrow-derived mononuclear cells and the stromal vascular fraction isolated from adipose tissue for a one-stage gene delivery strategy. Bone formation was induced in all conditions containing cells overexpressing BMP heterodimers. Constructs seeded with FBs transfected with BMP-2/6 and MSCs transfected with BMP-2/6 showed comparable bone volumes, both significantly higher than controls. Single-stage gene delivery proved possible and resulted in some bone formation. We conclude that bone formation as a result of ex vivo BMP gene delivery can be achieved even without direct osteogenic potential of the transfected cell type, suggesting that transfected cells mainly function as a production facility for osteoinductive proteins. In addition, single-stage transfection and reimplantation of cells appeared feasible, thus facilitating future clinical translation of the method.

The main challenge in BMP-2 based application lies in finding strategies that prolong its effective period, as it has a short biological half-life. Several BMP-2 release profiles have shown to enhance bone formation at various application sites. However, it remains to be determined which BMP-2 release profile best augments bone formation and whether this effect is location-dependent. Therefore, the aim of this study was to investigate the effect of BMP-2 release from oligo[(polyethylene glycol) fumarate] bis(2-(methacryloyloxy)ethyl) phosphate (OPF-BP) composites on the osteoinductive efficacy at ectopic versus orthotopic application. By varying the BMP-2 loading method, three different OPF-BP composites were created with varied release profiles. The composites were compared to unloaded OPF-BP as negative control, and to the clinically used Infuse® absorbable collagen sponge (ACS) as positive control. Bone formation was assessed by micro-computed tomography after 9 weeks of subcutaneous implantation and 3, 6, and 9 weeks of orthotopic implantation in rats (n=48). Whereas a BMP-2 burst release of >49% generated significantly more bone compared to sustained release (burst release <30%) at the subcutaneous implantation site, differential release did not affect bone formation at the orthotopic site. Furthermore, all BMP-2 containing OPF-BP composites showed significantly more bone formation compared to ACS in the orthotopic implantation site. In conclusion, this study clearly shows that the osteoinductive effect of different BMP-2 release profiles is location dependent. Additionally, more bone formation in OPF-BP compared to ACS at both application sites emphasizes the role of biomaterials as a scaffold to achieve proper bone tissue formation.