IN VITRO MODELS TO STUDY BONE TISSUE ENGINEERING

Abstract

The placement of orthopaedic, as well as dental, oral and craniofacial implants, are common practices in medicine and denstistry today. Challenges to the successful outcome of such implants include loosening of the device and inadequate filling of bone defects. The engineering of bone tissue is a recent strategy to provide new solutions to such problems. Since skeletal tissue regeneration requires three components, i.e. cells, growth and differentiation factors, and extracellular matrix, the approach of bone engineering is to mimic the biological process by delivering to the injured site: 1. cells capable of differentiating into osteoblasts, 2. inductive factors, and 3. a scaffold, biodegradable or not, to support cells. Prior to experimental and clinical application of the innovative surfaces or scaffolds, the three components have to be tested in the Labs using reliable in vitro methods.

1. Cells. The source of cells is a key point: osteoblast is the differentiated cell able to form bone in vivo and in vitro, and should be used, but primary human osteoblasts (hOB) are seldom available to the Labs, whereas osteoblast-like cell lines and bone cells from animals are an easy source, but may give different responses. An additional aspect which cannot be disregarded is the source of the bone cells, since the age and gender of the donor, as well as the site of retrieval and the method of isolation, have been shown to affect the yield of cells, the proliferation rate and their ability to form bone in vitro.

Stromal cells from bone marrow (MSC), and other sites, have been shown to be a promising source of cells with high replicative and bone-forming potential. The same drawbacks outlined for osteoblasts apply to MSC.

In our lab human osteoblasts are mainly obtained from trabecular bone fragments and stromal cells from bone marrow of patients undergoing surgical revision of hip implants. HOB are usually isolated by seeding minced bone chips in culture plates to get outgrowth of single cells from fragments, as the isolation technique (mechanical vs enzymatic) appeared to have no effect on the differentiation process. Confluence of the cell layer is reached in approximately four weeks (14–40 dd) and the bone phenotype is assessed by alkaline phosphatase (ALP) cytochemistry and morphology, as well as mineralization after addition of ascorbic acid and b-glycerophosphate. MSC are isolated by gradient centrifugation and adherence to culture plastic; their replicative potential is evaluated by the colony forming assay, and ALP staining provides the test for differentiation toward bone-forming cells.

Preliminary evaluation of our cell isolates from orthopaedic patients showed that there is no direct correlation between the age of donor and the yield of hOB in terms of proliferation rate and ALP activity. As far as MSC are concerned, the addition of dexamethasone during cell expansion stimulated only a small increase in the number of colonies and ALP positive staining.

2. Inductive factors include growth factors, cytokines, peptide sequences and angiogenetic factors. The experience of our Lab will be given in a different presentation.

3. Specifically tailored biomaterials are crucial tools in tissue engineering: our experience is concerning in vitro testing of artificial materials developed by material scientists to replace bone.

Such materials have to provide biocompatibility, i.e. no inflammatory reaction or immunorejection, controlled biodegradation if necessary, and biomechanical features to comply with the anatomical requirements.

From a methodological point of view, the ‘engineered’ biomaterials can be classifieded as bi-dimensional (2D) materials or three-dimensional (3D) scaffolds.

2D surfaces are often well known materials already in clinical use, but innovations concern the ‘biomimetic approach’ applied to their surface. This means to recreate the ‘nanotopography’ of natural tissues, by modifying the roughness, or by mimicking the extracellular matrix (ECM) on the surface: both strategies aim to recruit bone cells and to promote bone formation.

In the framework of a national research project both 2D and 3D materials were assayed in our Lab.

Two types of titanium with different surfaces were tested with human osteoblasts, and compared to a commercial titanium with smooth surface. At 4 hours from seeding onto surfaces, hOB on smooth Ti were elongated, with evident spreading. On the rougher surfaces small focal contact patches were evident, and hOB showed a more rounded morphology whereas stitching to the irregular surface. By prolonging the culture time, all the surfaces were covered by cells, and differences were less evident. Therefore early osteoblast adhesion seems to be different on micro-rough and smooth titanium, but then hOB exhibited a similar proliferation rate. Our results show that surface roughness is not always increasing cell adhesion, and primary cells do require specific micro or nano-topography to spread and proliferate, unlike continuous cell lines which are easily growing on any substrate.

A second approach to control cell adhesion and spreading onto surfaces is the deposition of RGD sequence (Arg-Gly-Asp), the cell-binding domain shared by a number of bone related proteins, including collagen, fibronectin, bone sialoprotein, thrombospondin, vitronectin, etc. The process for immobilization of peptides on the surfaces is crucial, and the amount and pattern of immobilized peptide has to be controlled, as adhesion sites should have a specific spatial arrangement to be recognized by cell adhesion molecules. Inadequate distribution of such binding motifs has also been shown to promote apoptosis of cells, instead of enhanced adhesion.

In our lab polymers with irradiation treatment and RGD-addition were tested using human osteoblasts. In comparison with smooth surface, irradiated surfaces were found to promote cell adhesion and RGD immobilization was further increasing the number of cells highly spread, with well defined cytoskeleton, and evident stress fibers along the cell body. Therefore, RGD immobilization onto surfaces, if adequately tailored, is a powerful tool to recruit cells and to stimulate their function. Further improvements will make use of sequences which specifically bind osteoblasts to the functionalized surface.

3D scaffolds are conceived as bone substitutes for large bone defects: therefore they have to be able to host bone cells, to promote bone formation and to be replaced gradually by regenerated bone. They are mostly approved polymers which are modulated in terms of cristallinity, porosity, interconnections, etc. to get a controlled degradation rate, and often added with bone-like components (hydroxyapatite or b-tricalciumphosphate) to improve osteoconduction. Moreover, the scaffold can be loaded with cells or growth factors (BMPs), to fasten tissue regeneration, or with drugs for treatment of infection, cancer therapy, and so on. Naturally derived polymers, including the recent ‘bioscaffolds’, besides difficulty in preparation, suffer from poor control of enzymatic degradation and weak mechanical performance: therefore many research groups rely on synthetic polymers.

Poly-e-caprolactone (PCL) matrices, with micro- or macro-porosity, and with or without hydroxyapatite (HA) particles, have been extensively assayed in our Lab for their ability to support osteoblast growth and activity. In our hands the presence of HA particles within and onto the PCL scaffold was found to increase osteoblast adhesion and function. We have been able to detect surface colonization by continuous and primary bone cells, and also mineral formation after 3–4 weeks with proper additives, but the presence of viable cells in the ‘core’ of the scaffold is still a matter of debate. The employment of a spinner flask for cell seeding into matrices has been found to improve ‘conditioning’ of the scaffold, but not definitely cell entrance in depth. Confocal microscopy is to some extent faded by the autofluorescence of the polymer matrix, and light microscopy suffer from poor resolution. Results from our experience with hOB and MSC seeded on different 3D PCL scaffolds are presented.

Hydrophilic and hydrophobic polyurethane-based scaffolds (PU) were assayed in our lab, too. Despite high hydrophilicity and addition of the polymeric matrix with HA and b-tricalciumphosphate (TCP), hOB were not able to adhere and grow to confluence onto such porous polymers.

In summary, in vitro models with osteoblasts are a powerful tool to analyse biological compatibility of innovative surfaces or scaffolds, even if they are unable to model physiological function in vivo. Actually, that these models can work in the body has to be demonstrated in experimental in vivo testing, prior to clinical trial. However, the design and improvement of materials rely on the understanding of how cells basically respond to surfaces.

In conclusion, the challenge in bone engineering is to link clinical needs to material technology. In vitro and in vivo studies have demonstrated that the ability of materials to support bone formation can be enhanced by modifying the physical, chemical and biological characteristics of the surface, and surface micropatterning is a powerful tool for constructing elaborate intelligent bio-materials. But the biological response of bone cells and bone tissue, and therefore the orthopaedic research, is a critical step in material research and bone engineering.