Key words

bone tissue engineering (BTE); mineralization; hierarchical organization; nanoscale; biomaterials

5.1 Introduction

The possibility of applying nanotechnology approaches in biomaterials science is giving rise to new trends in the orthopedic domain. Specifically, the advent of sensitive techniques and a better understanding of important scientific phenomena, together with the motivation of meeting clinicians’ demands, are making bone tissue engineering (BTE) a challenge, with increasing interest for regenerative approaches instead of merely replacement solutions. Several scientific fields are being merged in order to fulfill the demanding complexity of such a task.

Being able to design nanostructured materials is of crucial significance, since cell interactions with biomaterials will occur at the nano-level. The challenge of bottom-up approaches able to mimic nature’s outstanding mineralized structures is facilitated by nanotechnology tools.

Advances in nanotechnology also allow the development of novel nanodevices that not only have better cytocompatibility and bioactive properties but can also behave as unique drug-delivery platforms. Moreover, these current approaches allow for the production of hierarchical architectures with organized multilevel structures, a key feature of natural materials.

Despite the bright future indicated by nanotechnology in BTE, important limitations still need to be overcome regarding the actual methodologies. Cytotoxicity of nanomaterials requires careful assessment. Additionally, technical difficulties, such as ensuring proper vascularization of scaffolds, also need to be addressed.

In the following sections, important concepts will be defined. Several examples of new approaches in BTE are presented as well as the rationale behind them. The aim of this chapter is not to provide in-depth insights into the different nanoscale designs inspired in the biomineralization process but to give an overview of possibilities and limitations when nanodevices are applied in BTE.

5.1.2 Bone structure and properties

Bone is a dynamic, highly vascularized tissue that is formed from a composite of 60% mineral (mostly nanoscale hydroxyapatite crystals), 30% organics (including collagen, glycoproteins, proteoglycans, and sialoproteins), and 10% water (Athanasiou et al., 2000; Salgado et al., 2004). Its complex cellular architecture continues to remodel throughout the lifetime of an individual, giving bone an innate ability to regenerate and heal injuries below a critical size, helped by local or recruited stem cells (Einhorn, 1995; Place et al., 2009). However, some injuries are beyond the limit that the body can self-repair, namely cases of severe fractures, bone tumor resections, age-related restrictions, scarring and inflammation processes. In these cases a human approach is required in order to assure proper healing (Einhorn, 1995; Place et al., 2009). However, mimicking bone’s natural structure and mechanical profile remains a challenge in BTE.

Toughness and flaw tolerance are generally associated with natural biomineralized composites. They are believed to be intimately related to an advantageous hierarchical arrangement of structural motifs at the nanoscale. Bone tissue also presents this hierarchical architecture and is organized in seven different levels, as summarized in Fig. 5.1 (Weiner and Wagner, 1998).

Beginning at the nanoscale, around the order of 1 nm are the amino acids that form the collagen molecules. The collagen molecule or tropocollagen is approximately 300 nm long and 1.5 nm in diameter and has a 3D polypeptide stranded structure – see Fig. 5.1 VII(a). The collagen molecules will associate in collagen fibrils with a diameter of 200 nm (Vanderrest and Garrone, 1991; Fong et al., 2012). Together with smaller quantities of various proteoglycans and glycoproteins these components represent the organic part of the bone (Mescher, 2009). The organic/inorganic association characteristic of this tissue is due to the presence of hydroxyapatite crystals aligned along the type I collagen fibrils’ c-axis. The dimensions of the crystal will be 50 × 25 nm in length and width and with 2–3 nm in thickness, presenting plate-shape morphology. The mineralized collagen fibrils will be arranged in lamellae (3–7 μm thick) (Athanasiou et al., 2000) of fibril array, each having a different pattern according to the fibril’s orientation. The concentric association of these fibril arrays results in a final collagen fiber with a diameter of 2 μm (Fong et al., 2012). The rotation of the crystal sublayers, together with the rotation of the collagen fibril bundles around their axis, enhances the isotropic properties of bone found at the macroscopic scale, giving it its strength (Luz and Mano, 2010; Ritchie, 2011). The lamellae are formed by the secondary osteons, interstitial lamellae and the inner and outer circumferential lamellae.

The mineral amount present in the bone tissue is also very important in determining its mechanical properties and function (Currey, 1999). This parameter will be defined by the cell activity in the bone. Osteoclasts release an enzyme that destroys the bone tissue, forming tunnels along the longitudinal axis of bone. Then, osteoblasts rebuild the secondary osteon cylindrical tubes, by secreting circular rings of lamellae that surround the vascular or Haversian canal in the center of the osteon or Haversian system with a diameter of 200 μm and length of 10–20 mm (Athanasiou et al., 2000; Fong et al., 2012). See Fig. 5.1 III. When osteoblasts are trapped in the newly synthesized organic osteoid that will soon mineralize, they become osteocytes. A supply of nutrient is facilitated to osteocytes through a microcirculating system called canaliculi. See Fig. 5.1 II(b) (Weiner and Wagner, 1998; Athanasiou et al., 2000; Ott et al., 2010).

The last and macroscopic level of the hierarchical architecture corresponds to the group of closely packed osteons that compose the cortical bone (Weiner and Wagner, 1998; Ott et al., 2010).

Mechanical support attributed to the skeletal system is essentially provided by cortical bone that, from a biomechanical perspective, behaves like a semibrittle, viscoelastic, and orientation-dependent material (Athanasiou et al., 2000; Fong et al., 2012). It is lamellar morphology of cortical bone that is responsible for the reduction of crack propagation and increase of toughness (Ott et al., 2010). Cancellous bone – see Fig. 5.1 II(a) – is a lighter, less dense form of osseous tissue consisting of trabecular plates and bars. It is found in the highly vascular inner parts of bone where hematopoiesis and ion exchange occur (Fong et al., 2012).

5.1.3 BTE

The term tissue engineering is defined by Langer and Vacanti (1993) as ‘an interdisciplinary field of research that applies the principles of engineering and the life sciences towards the development of biological substitutes that restore, maintain, or improve tissue function.’ Recently, tissue engineering has gained increasing support over traditional methods as a method to treat debilitating musculoskeletal disorders affecting bone, ligament, and cartilage such as osteoarthritis and osteoporosis, due to its interdisciplinary approach that focuses on tissue regeneration rather than on its replacement (Laurencin et al., 1999; Place et al., 2009).

The so-called first generation biomaterials (from the 1960s to early 1980s) were represented by prostheses, aiming only to match bone physical properties, and causing a minimal toxic reaction. See Fig. 5.2(a). They were based on almost inert materials. This type of material stimulates the host tissue to produce a nonadherent fibrous capsule around them with the purpose of isolating the foreign object. With time, this protection mechanism may cause the deterioration of the implant. These types of material do not regenerate bone, since they do not present specific bioactivity. They are used merely as bone replacement devices. Metals, ceramics, and polymers may be included in the nearly inert materials group (Barone et al., 2011).

Bioactive materials arrived during the 1980s and initiated the second generation of biomaterials. Their significance in bone tissue regeneration is related to the strong chemical bond between the host tissue and the material, to avoid the development of a fibrous capsule that will compromise the fixation of the implant, and the success of the intervention over time (Hench, 1998).

Regarding the BTE field, the concept of bioactivity is directly related to the ability of a material to bond chemically to the bone, through the formation of an apatitic layer in vivo, or to the ability of inducing the precipitation of hydroxyapatite when immersed in simulated body fluid (SBF) in vitro (Hench, 1988; Kokubo et al., 1990; Kokubo and Takadama, 2006). The degradation behavior of bioactive materials must be synchronized with the cellular events leading to new bone growth (Rezwan et al., 2006).

Bioactive materials are classified according to the time taken for more than 50% of the interface to bond to bone (t_0.5bb). The bioactivity index (I_B) is calculated using formula (5.1) (Hench, 1988):

${\mathit{I}}_{\mathrm{B}}=100/{\mathit{t}}_{0.5\mathrm{bb}}$ [5.1]

[5.1]

An I_B value greater than 8 (class A) means that the material will bond to both soft and hard tissue. Materials with an I_B value of less than 8 (class B), but greater than 0, will bond only to hard tissue.

This leads to the distinction of two different classes. Class A comprises bioactive glasses which exhibit rapid and strong bonding to bone by means of a series of chemical reactions at the bone tissue interface. They are defined as being osteoproductive, and osteoconductive (Hench, 1998). Moreover, bioactive glasses are also able to bond to soft connective tissue. Class B relates to materials which bond slowly and only to bone, like synthetic hydroxyapatite. They are classified as bioactive ceramics, and have only osteoconductive properties (Hench, 1998). When a bioactive material bonds to soft tissue, whether in vitro or in vivo, collagen fibers (which are in all soft tissues) become embedded and bonded within the growing apatitic layer on the bioactive material’s surface (Hench, 1988; Dubok, 2000).

Bioactive materials may be used as coating for prostheses or as scaffolds and fillers, as is depicted in Fig. 5.2(b). 3D, porous, degradable, polymeric scaffolds provide mechanical support while allowing the ingrowth of new bone as the scaffold degrades. The pore size of scaffolds needs to be greater than 30 μm to allow bone ingrowth (Yang et al., 2001).

More recent approaches include the extraction of cells from a patient and later transplantation after proper cell culture, and the culturing of the cells in a 3D scaffold for implantation in a defect (Cartmell et al., 2003; Peter et al., 2010a). With strategies combining both biological systems and engineered substrates, biomaterials are now on their third generation. The actual goal is to elicit specific cellular responses at the level of molecular biology. Such a shift from a materials and mechanics approach to biological-based tissue repair requires careful understanding of the application of molecular biology to bone regeneration (Hench et al., 2002).

Bone repair typically involves the use of autografts, allografts, xenografts, and synthetic materials (Laurencin et al., 1999; Kneser et al., 2006). Autologous bone grafting is normally the first choice for bone replacement, but its use is limited by its short supply and pain resulting from the harvest process (Stevens et al., 2005). However, obtaining tissue from another biological source implies a certain risk of rejection and disease transfer. Tissue engineering may be a suitable solution to these limitations (Laurencin et al., 1999; Venugopal et al., 2008). Figure 5.2 summarizes bone replacement and regeneration strategies.

There are still some important limitations in BTE that need to be overcome, namely cell necrosis occurring within the inner core of a mineralized scaffold (Cartmell et al., 2003) or the inadequacy of its structural properties to the functionality of the tissue. The solutions may lie in a BTE nanoscale approach regarding the development of new techniques and materials or combinations of both, in order to obtain the organized structure that gives bone its successful mechanical properties and cellular interactions. The design principles employed to develop hierarchical approaches aiming to mimic natural materials formation processes, can be extrapolated to the whole class of biomedical materials, including polymers, metals, ceramics, or hybrid combinations (Tan and Saltzman, 2004).

A BTE biomaterial-based device ideally would present a certain set of characteristics related to each one at the macro-, micro- and nanolevels. Beginning at the macrolevel, biocompatibility, biodegradability and mechanical properties are the essential criteria to be fulfilled. At the micro-level the key features for assuring the biodevice success are determined by tissue architecture, surface chemistry, surface stiffness, cell migration, nutrient delivery, and vascularization ability. Ultimately, at the nano-level, bioactive factors, cell adhesion, mineralization and gene expression will play a valuable role on defining the device fate (Santo et al., 2012).

Biomaterials are classified according to their response when implanted in the body. Therefore, they are described as bioactive, nearly inert, and resorbable. While the former concepts have been described above, resorbability is defined as the ability of a material to support bone growth during the healing process, and later gradually degrade in metabolizable residues. It is therefore a very desirable feature for a biomaterial (Barone et al., 2011). Both polymers and ceramics may have adjustable bioresorbability.

Bioceramics are special compositions of ceramic materials in the form of powders, coatings, or bulk devices. They are suitable to repair, augment, or replace diseased or damaged bony tissue (Hench, 1998). Other materials, such as metals and polymers, are also used in bone repair applications. However they are not naturally bioactive, meaning that they will strongly bond to bone in vivo or will not develop an apatitic layer when immersed in SBF (Kokubo et al., 1986; Hench, 1991).

Despite the challenge that relies on creating nanodevices and materials that are able to regenerate bone, some particular points must be considered. As in every other biomaterial, the influence of the topography on the host tissue response is one of the most important concerns during the design and manufacture of a biomaterial (Singhvi et al., 1994; Curtis and Wilkinson, 1997; Ito, 1999). Moreover, a biomaterial aiming to be applied to bone regeneration must be based on the following important features: biocompatibility with bone cells; bioactivity/promotion of hydroxyapatite mineralization and osseointegration; suitable mechanical properties/support of the formation of natural bone; biodegradability. They need also to provide adequate mechanical support regarding the function of the bone and respecting the overall hierarchy, and interactions between the natural stiff inorganic (mineral apatite) crystals and the soft organic (collagen) layers (Fratzl et al., 2004; Gupta et al., 2006; Ritchie, 2011; Fong et al., 2012).

Biomimetic mineralization relays on the use of organic molecules to prepare inorganic crystals with ordered structures seldom found in natural minerals (Sun et al., 2012). Some examples of minerals synthesized via biomimetic mineralization are: CaCO₃ (Sommerdijk and de With, 2008) BaSO₄, BaCO₃ (Zhu et al., 2009), and hydroxyapatite. Organic additives are used as templates. They are vital in biomimetic mineralization as they determine the morphologies and properties of mineral crystals (Sun et al., 2012).

In the next section several nanodevices developed for BTE applications, and somehow inspired by the mineralization process, will be presented, as well as the rationale behind their development.

5.2 Materials and techniques for nanoscale design

Whether a bone defect is going to be repaired with an orthopedic implant or with a tissue engineered construct, its success will always be determined by the biomaterial choice. This happens because, as already mentioned, the most important requirement on BTE is to assure an immediate interaction between the biomaterial and the host tissue (Christenson et al., 2007). Traditional materials such as ceramics, metals and polymers, already used in BTE, are being now reinvented at the nanoscale and adapted to meet specific targets and functionalities, such as delivery systems of drugs and proteins or matrices for bone cell regulation (Singh et al., 2012).

Commonly, the concept ‘nanomaterial’ refers to materials with a nanosized topography or composed of nano-sized basic units comprising dimensions in the scale range of 1–100 nm. These will include nanostructured materials, nanocrystals, nanocoatings, nanoparticles, and nanofibers, based respectively on basic components, grain sizes, individual layers, particles, and fibers within the range of 1–100 nm (Christenson et al., 2007).

Surface properties such as roughness, charge, chemistry, and wettability are crucial for healthy functionality of the general cellular system and will influence protein adsorption (Wilson et al., 2005; Christenson et al., 2007). Since nanoscale protein interactions are essential to control cell functions such as proliferation, migration, and extracellular matrix (ECM) production, it is advantageous to optimize these interactions by tailoring the biomaterial’s surface (Benoit and Anseth, 2005; Christenson et al., 2007).

Current techniques for producing hierarchical structures for mineralization at the macroscale include electrospinning (Zhang et al., 2008), directional freezing (Zhang et al., 2005; Deville et al., 2006; Wegst et al., 2010), and biotemplating (Sotiropoulou et al., 2008). However, the majority of these techniques fail to orientate the nucleation of hydroxyapatite.

New approaches and methodologies aiming to mimic bone are now arising. Several techniques are used to produce BTE nanostructures: vapor–liquid–solid method (Westwater et al., 1997), thermal evaporation (Pan et al., 2001), controlled precipitation (Hong et al., 2009c), sol–gel route, microemulsions (Sang et al., 2010), template and biomimetic synthesis (Cushing et al., 2004), flame spray (Mackovic et al., 2012), gas aggregation (Martínez, et al., 2012), dip coating, and spin coating (Limmer et al., 2002). The following strategies are used to produce micro/nanomaterials with multilevel structures: self-assembly (Murugan and Ramakrishna, 2005), template synthesis (Porter et al., 2009), Ostwald ripening, or the Kirkendall effect (Zhao and Jiang, 2009), and evaporation-induced self-assembly (EISA) (Luz and Mano, 2012a, b). Other strategies may be used to extend the efficiency of the engineered structures. For instances, layer-by-layer can be used to include functional components, such as bone morphogenic proteins (BMPs), into the developed nanodevices (Gribova et al., 2012; Hammond, 2012). Furthermore, recreation of the cellular microenvironment through the production of ECM by cells seeded on the implanted constructs is also possible (Thibault et al., 2011; Fong et al., 2012).

Sol–gel processing has an inherent flexibility that makes it a good technique for producing nanostructures typically in the order of a few nanometers in size. Besides the ability to set the stoichiometric chemical composition, the obtained compounds will have compositional homogeneity at a molecular level and a sol stability conferred by electrostatic stabilization (Brinker and Scherer, 1985; Limmer et al., 2002). Above all, sol–gel processing offers many advantages for the processing of materials such as organic–inorganic hybrids, nanocomposites, and coatings on complex patterned surfaces, making this process very popular for BTE applications (Limmer et al., 2002).

The multitude of properties resulting from the conjugation of the above-mentioned techniques result in complex structures that broaden their potential applications and successfully address the demanding BTE requirements.

5.3 Nanoparticles

The size and easy dispersability of nanoparticles mean they can be applied in a multitude of biomedical devices and strategies, enhancing mechanical properties and increasing or attributing the bioactive character of the material, or even both. Nanoparticles may also work as an independent nanodevice, in applications such as nanocarriers. Moreover, the wide variety of core materials available, coupled with tunable surface properties, make nanoparticles an excellent platform for a broad range of biomedical applications (De et al., 2008).

Bioactive nanoparticles are often based on ceramics, since these materials can induce hydroxyapatite precipitation. Several preparation methods are commonly followed to synthesize inorganic nanoparticles, namely controlled precipitation, sol–gel technique, microemulsions, template synthesis, and biomimetic synthesis (Cushing et al., 2004). Physical methods such as flame spray (Mackovic et al., 2012) and gas aggregation source are also used (Martínez et al., 2012).

It is extremely important that the chosen method allows for composition, shape, size, and aggregation control in order to assure that the developed particles meet the application requirements (Munoz-Espinoza et al., 2012). Commonly, solution-based synthesis are chosen to synthesize nanoparticles, since a more accurate control of the stoichiometry is achieved (Cushing et al., 2004).

Inorganic nanoparticles have been used in combination with polymeric matrixes to develop new biomaterials that provide a microenvironment that more closely mimics natural bone tissue physiology, namely nanofiber composites, layer-by-layer coatings, scaffolds, injectable materials, and imprinting towards creating of patterned bioactive surfaces (Alves et al., 2010; Kim et al., 2011). The resulting nanodevices are valuable alternatives to the original approach of using ceramic blocks to fill bone defects. Moreover, the development of polymer/inorganic hybrids has been recognized as a strategy to improve the mechanical behavior of ceramic-based materials. Compared with microsized bioactive ceramic particles, nanosized particles have a larger surface area and can form a tighter interface with polymer matrix in composites, resulting in better mechanical properties. Moreover, the high specific surface area of nanoceramics allows for not only a faster release of ions but also a higher protein adsorption, and thus bioactivity will also be increased (Suryanarayana, 1995; Han et al., 2005).

Silica is an abundant biocontaminant with high biocompatibility, commonly designed in nanoparticles. For instances, monodispersed, spherical silica nanoparticles may be synthesized at room temperature by the hydrolysis of tetramethyl orthosilicate (TMOS) in alcohol media under catalysis by ammonia (Xu et al., 2003). Xu et al. prepared polyethylene glycol (PEG)-coated silica nanoparticles with sizes ranging from about 50–350 nm in diameter using this methodology. Since these nanoparticles are able to encapsulate certain reagents in their matrixes, they are suitable for biomedical applications, namely in vivo diagnosis, analysis, and measurements (Xu et al., 2003).

Silica nanoparticles were reported as having inhibitory effects on osteoclasts and stimulatory effects on osteoblasts, in vitro (Beck et al., 2012). The mechanism of bioactivity is a consequence of an intrinsic capacity to antagonize activation of NF-κB, a signal transduction pathway required for osteoclastic bone resorption but inhibitory to osteoblastic bone formation. Therefore, silica nanoparticles were shown to promote a significant enhancement of bone mineral density in mice in vivo, providing evidence for the potential application of silica nanoparticles as a pharmacological agent to enhance bone mineral density and protect against bone fracture (Beck et al., 2012).

One popular example of a material successfully synthesized in the form of nanoparticles, also with a silica base, is bioactive glass. Nanoscale bioactive glasses have gained attention due to their superior osteoconductivity and cytocompatibility when compared to micro-bioactive glass materials (Boccaccini et al., 2010; Mackovic et al., 2012). Bioactive glasses are the gold standard material for bone regeneration. They consist of a silicate network incorporating sodium, calcium, and phosphorus in different relative proportions (Boccaccini et al., 2010). The therapeutic effect of bioactive glasses arises from the influence of soluble calcium and silicate species on the genetic expression of osteoprogenitor cells (Lin et al., 2010). Bioactive glass can promote the proliferation and activity of fibroblasts and accelerate the process of vascularization, facilitating the healing of skin wounds (Hong et al., 2010). Due to their high surface area, bioactive glass nanoparticles (BG-NPs) have enhanced dissolution. This could cause an undesirable increase in Ca²⁺ ions in culture medium, since the increased intracellular Ca²⁺ ion level may induce apoptosis of the cells. However, this drawback has not been reported so far, indicating that the BG-NPs are biocompatible (Webster et al., 2000; Hong et al., 2010; Luz and Mano, 2011). Preparation methods of nanosized bioactive glasses are similar to those used for the general inorganic nanoparticles.

The first work mentioning the preparation of BG-NPs was from Xia and Chang in 2007. They developed a quick alkali-mediated sol–gel method to obtain 20–40 nm sized particles, this being the size that is controlled through addition of ammonia solution. The obtained particles are depicted in Fig. 5.3. The gelation time was around 2 min, and decreased with an increase in concentration of the ammonia solution. Calcination of gel powders was achieved at 600 °C, the final powder being an amorphous glass.

In the meantime Vollenweider et al. (2007) also developed BG-NPs, but this time the 20–50 nm diameter particles were obtained by flame spray synthesis. The authors then treated demineralized human dentin with the obtained nanoparticles and compared the remineralization potential of the nanosized particles with a micrometer-sized, commercial reference material (PerioGlas). The substantially higher remineralization rate induced by nanometer-sized vs. micrometric bioactive glass particles confirmed the importance of particle size in clinical bioglass applications.

Misra et al. (2008) performed a study in which they compared the influence of using bioactive glass microparticles versus nanoparticles in a composite based on poly(3hydroxybutyrate) (P(3HB)). The microparticles were melt-derived while the nanoparticles were flame spray synthesized. Changes to the structural, thermal, and mechanical properties of P(3HB)/bioactive glass composites were investigated, and the results confirmed that the addition of nanosized bioactive glass particles had a more significant beneficial effect on the mechanical and structural properties of a composite system than did microparticles, as well as enhancing protein adsorption, two desirable effects for the application of the composites in tissue engineering.

In order to develop BG-NPs with improved dispersibility, Hong et al. (2009a) added an innovation to the sol–gel synthesis of BG-NPs. SiO₂-CaO-P₂O₅ ternary BG-NPs of 30–100 nm in diameter were obtained via the combination of sol–gel and coprecipitation methods. The precursors were hydrolyzed in acidic condition and subsequently condensed and precipitated in an alkaline solution. The great advantage of this method is that the agglomeration of nanoparticles via the linkage of H₂O molecules during the drying process is inhibited by lyophilization of the gel particles. After calcination, a well-dispersed bioactive glass nanoparticle can be obtained without grinding and sieving, thereby lowering production costs.

Nanoparticles are often used as nanofillers. However, their shape may influence the final mechanical properties of the materials. For instance, it was reported that needle-like or short-fiber inorganic particles can improve the mechanical performance of polymer–inorganic composites much more effectively than spherical fillers (Unal et al., 2004; Du et al., 2006). Therefore, Hong et al. (2009b) developed rice-shaped particles with the composition SiO₂-CaO-P₂O₅ ≈ 6:74:20 (mol). The size distribution of these nanoparticles is quite narrow, with most of the particles being around 70 nm in diameter and 215 nm in length. An in vitro investigation with MG-63 osteoblast-like cells revealed a high cytocompatibility of such BG-NPs compared with the microsized ones (Mackovic et al., 2012).

Nanofiller shape is not the only property that needs to be addressed. Several studies have been conducted in order to better understand the influence of characteristics as composition, preparation conditions and size of BG-NPs. Luz and Mano (2011) compared the bioactive behavior of BG-NPs from the binary (SiO₂-CaO) and ternary (SiO₂-CaO-P₂O₅) systems. They concluded that the best bioactivity results are not only related to the composition but also with the preparation conditions. BG-NPs from the ternary system have more bioactive character than BG-NPs from the binary one. Moreover, the sol–gel pH also influences the final bioactivity of the particles. Higher preparation pHs increase bioactivity in particles. Increased temperatures during the heat treatment will also enhance the final bioactivity of the particles. This is because amorphousness of the samples increases when submitted to these conditions. However, a higher crystallinity may lower the dissolution rate to ineffective values (Luz and Mano, 2011). Nevertheless, binary bioactive glasses with the composition SiO₂-CaO are still bioactive and possess desirable biological properties. Bioactive glasses offer the possibility of easily adapting their composition to meet specific needs. Some work has shown that doping bioactive glasses with different ions can add value to these materials (Gentleman et al., 2010). For instance, magnesium, one of the main substitutes for calcium in biological apatite (Gutowska et al., 2005), when included in the bioactive glass formulation, can enhance osteoblastic adhesion (Webster et al., 2002; Luz and Mano, 2012a). Other elements were already used to dope bioactive glass, namely Sr, which is known to enhance osteoblastic differentiation (Isaac et al., 2011), and Ag₂O which confers bacteriostatic and bactericidal properties to bioactive glass (Bellantone et al., 2002). Also the processes leading to the formation of different nanoparticles need to be fully understood in order to control their morphology and both chemistry and physical behavior. The impact of the nanosize of these inorganic structures is also being studied.

Hong et al. (2010) studied the effect of bioactive glass on the biomechanical properties of various mammalian cells. By using to atomic force microscopy (AFM) to measure the biomechanical properties of mammalian cells, they concluded that binary BG-NPs can significantly decrease the plasma membrane stiffness of bone marrow stem cells. However, when the study was conducted with bovine aortic endothelial cells, the stiffness was increased and the elongation of the cells was stimulated forming endothelial networks. These results indicate that the vascularization process may be facilitated due to the implantation of BG-NPs (Hong et al., 2010).

The optimization of BG-NPs is still being researched in order to overcome some limitations, mostly related to morphology control and aggregation issues. For instance, BG-NPs (Si:P:Ca = 29:13:58 weight ratio) of about 40 nm diameter were prepared via the sol–gel method and then lowmolecular-weight poly(L-lactic acid) (PLLA) was successfully grafted onto the surface of BG-NPs nanoparticles via the coupling of diisocyanate. The aim was to improve the phase compatibility between the polymer and the inorganic phase (Liu et al., 2008).

El-Kady et al. (2010) used a modified alkali-mediated sol–gel route to obtain BG-NPs. The modified sol–gel method resulted in a reduction of the gelation time to about a minute rather than days as in the traditional sol–gel process. Furthermore, fast gelation prevented the aggregation and growth of colloidal particles to sizes larger than 100 nm. The proposed method is thus capable of delivering nanoparticles of sizes less than 100 nm with minimum agglomeration.

The polyvalence of nanoparticles in BTE has already been mentioned. Nanoparticles can be used in a multitude of applications and, due to their reactivity resulting from the nanoscale dimensions, simple approaches may have great results.

Recently, Luz and Mano (2012b), inspired by colloidal crystals, used BG-NPs as building blocks for the construction of hierarchical organized structures by self-assembly. In a very simple strategy, drops of BG-NP aqueous suspensions were left to evaporate on biomimetic superhydrophobic surfaces. The crystallization degree of the structures was controlled by the evaporation rates taking place at room temperature or at 4 °C. Spherical aggregates were obtained with a hierarchical ordered morphology from nano- to micro- and macroscale. The crystallization degree of the structures influenced the Ca/P ratio of the apatitic film formed at their surface, after 7 days of immersion in SBF. This allowed the regulation of bioactive properties and the ability to release potential additives that could be also incorporated in such particles with a high efficiency. The impact of such technology is high, allowing the production of microspheres with biomedical applications using a highly competitive method.

Metals processed at the nanoscale may also be useful in biomedical applications (De et al., 2008). For instance, silver is known to have an antibacterial action, representing thereby a good solution to infections on the surface of an implant, one of the main problems in reaching a suitable level of osseointegration (Miranda et al., 2012). Hydroxyapatite/silver nanocomposites have already been designed for this purpose. The combination of the bioactivity of the ceramic matrix with the biocide activity of the silver nanoparticles gives these nanocomposites an interesting range of applications in BTE (Miranda et al., 2012).

Finally, still regarding nanoparticles for BTE, a new class of biomaterials with additional functionalities targeting the bone must be mentioned. Magnetic properties are being included in bone-related biomaterials in order to enhance the therapeutic potential of novel nanodevices. The combination of magnetic properties with biocompatible composition may open the door to new biofunctional nanomaterials with broader action on repairing bone injuries while healing them. Magnetic nanoparticles, normally iron oxide nanoparticles, are used for in vivo biomedical applications in areas related to therapeutic (hyperthermia and drug-targeting) and diagnostic applications (nuclear magnetic resonance imaging) (Tartaj et al., 2003; Ferrari, 2005; Corot et al., 2006). A remarkable advantage of a system comprising magnetic nanomaterials is the possibility of using external magnetic fields to guide drug carriers to precisely target an area of the body with minimal or non-invasive methods (Medeiros et al., 2011). Regarding cancer treatment, hyperthermia consists of heating tumors to temperatures between 43 and 47 °C. Within this interval, the malignant cells are selectively destroyed whereas the healthy ones undergo only small and/or reversible damage (Ruiz-Hernandez et al., 2006). Thus biomaterials can be used as implantable thermoseeds to focus heat on the target region without overheating the surrounding healthy tissues.

Specifically for bone repair and related cancer therapy, the materials and carriers should be biocompatible with bone, such as hydroxyapatite and bioactive glasses. Ruiz-Hernandez et al. (2006) were the first to run hyperthermia heating experiments as well as preliminary biocompatibility assays for bioactive glass implantable thermoseeds. They concluded that the presence of sol–gel glass modifies the magnetic properties, so improving the heating power. The ability to reach a hyperthermic temperature range together with the bioactive behavior makes bioactive glass a very promising candidate for bone cancer treatment.

Based on ceramic, polymeric, or metallic compositions, nanoparticles are playing an increasingly important role in nanomedicine and in BTE in particular. Efforts are already being made in order to move a step further regarding the optimization of these polyvalent nanodevices.

Nanoparticle surfaces are now being engineered in a mimicking strategy aiming at overcoming the body’s physiological barriers. Nanoparticles used for multimodal diagnostics and for target-specific drug/gene-delivery applications are covered with biomolecules that mimic the ones present on the cell membrane, such as proteins, peptides, and carbohydrates. These strategies assure that upon injection in the bloodstream or following oral administration, the nanoparticles will reach the intended target (Gong and Winnik, 2012).

Clinically, the use of nanoparticles in vivo is mainly related to bioimaging and therapy. In order for BTE nanodevices to be successfully implanted in vivo, aspects such as bioconjugation and nanotoxicity need to be explored (Bear et al., 2011). Nanoparticles can offer a multitude of functionalities through the flexibility of their dimensions, morphology and composition. Hence, they are good candidates for a promising avenue in BTE research.

5.4 Nanofibers and nanotubes

Nanofibers are often used in the preparation of scaffolds for BTE. The rationale for using nanofibers is related to the theory that cells attach and organize well around fibers with diameters smaller than the diameter of the cells (Laurencin et al., 1999). In fact, it has been reported that cells might migrate through fibrous matrices by pushing the fibers aside (Christenson et al., 2007). Therefore, materials offering low resistance to the amoeboid movement of the cells, such as nanoscale fibers, are more suitable for promoting cell migration (Rodriguez-Lorenzo et al., 2012). Moreover, nanofibers can mimic the physical structure of the major constructive elements in the native ECM (Kadler et al., 1996; Zhang et al., 2007).

Nanofibers are normally produced by electrospinning. Briefly, a high voltage is applied to the polymer solution to draw out nanofibers that are then collected on a ground plate in the form of a mesh. Nanofibers can be cross-linked in order to adjust solubility or mechanical properties. They can also be modified in order to increase their biocompatibility or bioactivity.

Nanofibers may be also produced by self-assembly. Collagen fibrils of bone itself are formed by self-assembly of the collagen triple helices and the hydroxyapatite crystals. The hydroxyapatite crystals grow within these fibrils in such a way that their c-axes are oriented along the long axes of the fibrils (Traub et al., 1989).

There are several examples in the literature of the application of self-assembly phenomena to the production of nanofibers resembling the ones existing in the ECM. Fischer et al. (2012) chose collagen and hyaluronic acid as polymers for constructing a nanofiber mesh-based scaffold because they are the main components of the ECM and have been utilized in electrospinning. The collagen/hyaluronic acid meshes were cross-linked to render them insoluble and conjugated with gold nanoparticles to promote biocompatibility. The results showed that the produced scaffolds were successful in promoting cellular attachment, being thereby a suitable choice for a tissue engineered solution to promote cell growth.

Hartgerink et al. (2001) described a self-assembly-based method for producing a nanostructured fibrous scaffold resembling the ECM. The work was based on the self-assembly of a peptide-amphiphile via a pH-controlled reversible mechanism. The design of this peptide-amphiphile allowed the nanofibers to be reversibly cross-linked to enhance or decrease their structural integrity. After cross-linking, they were capable of nucleating hydroxyapatite in an alignment similar to that observed at the lowest level of hierarchical organization of bone, in that the crystallographic c-axis of hydroxyapatite was oriented along the long axis of the organic fibers.

Injectable scaffolds with controllable release are also possible. Hosseinkhani et al. (2007) produced a 3D scaffold by mixing a peptide-amphiphile aqueous solution with BMP-2 suspension. A 3D network of nanofibers was formed with an extremely high aspect ratio and high surface areas. In vivo release profile of BMP-2 from 3D network of nanofibers was investigated. It was demonstrated that the subcutaneous injection of an aqueous solution of peptide-amphiphile together with BMP-2 in rats resulted in the formation of a transparent 3D hydrogel at the injected site and induced significant homogeneous ectopic bone formation around that area. This was in marked contrast to BMP-2 injection alone or peptide-amphiphile injection alone. This kind of strategy represents a promising procedure to improve tissue regeneration.

Electrospinning is also an effective way of producing controllable nanofibers for BTE (Prabhakaran et al., 2011). By adjusting both concentration and feeding rate, it is possible to influence bioactivity. This technique may be applied not only to pure polymeric solutions, but also to produce inorganic based nanofibers (Chen et al., 2012).

Chen et al. (2012) used electrospinning to prepare bioactive TiO₂ fiber films. An acetic acid/ethanol/tetrabutyltitanate/polyvinylpyrrolidone solvent system was used as a precursor for the electrospinning. The TiO₂ fiber structures (including its fiber diameter, morphology, and phase composition) could be controlled by changing feeding rate, precursor concentration, and sintering temperature, showing that electrospinning is an effective way to prepare bioactive TiO₂ fiber films.

Nanotubes may also be prepared by electrospinning using a polymeric sacrificial template that will later be destroyed by heat treatment after the material has been covered. Kim et al. (2010) used a nanofiber mesh of a polymer (polycaprolactone) as a template that was mineralized within solutions via a biomimetic process. Subsequent heat treatment (over 500 °C) completely eliminated the inner polymer, so preserving the surface mineral phase in the form of nanotubes with diameters of hundreds of nanometers with nonwoven mesh, replicating the initial nanofiber template. Polycaprolactone nanowires have been synthesized through template synthesis. The in vitro tests showed an enhanced biological response (Porter et al., 2009).

Researchers are now exploring nanofibers in a multitude of forms that can be used in bone-related applications. For instance, magnetic properties were attributed to hydroxyapatite nanotubes by embedding magnetic nanoparticles within the inorganic tubes (Singh et al., 2012). The magnetic hydroxyapatite nanotubes were produced using a template made of magnetite nanoparticles/polycaprolactone and later underwent surface mineralization and thermal treatment.

Zhao et al. (2005) used single-walled carbon nanotubes as a scaffold for the growth of artificial bone material. The tubes were chemically functionalized with phosphonates and poly(aminobenzene sulfonic acid). The negatively charged functional groups on the nanotubes attracted the calcium cations and led to self-assembly of hydroxyapatite. Well-aligned plate-shaped hydroxyapatite crystals with 3 nm thickness were obtained after 14 days of mineralization (Zhao et al., 2005).

As already mentioned, at the smallest length scale of bone, collagen triple helices spontaneously form nanoscale bundles, which act as a template for the crystallization of hydroxyapatite nanocrystals. Researchers are making an effort to produce nanofibers able to mimic bone structure while maintaining the ability to induce crystal nucleation and growth of hydroxyapatite. This strategy is a good approach to follow: despite the results already obtained, there are still good opportunities for optimization.

5.5 Nanopatterns

At the nanolevel, every aspect of the biointerface of the biomaterial surface will have a considerable effect on the desired cellular response. Therefore, it would be of great value if researchers could control the chemical and physical characteristics at the material’s surface. Nanopatterns have been created in order to control cell interactions with the surface by means of an imposed pattern with controlled geometry and periodicity. Specific cellular responses may then be targeted (Tan and Saltzman, 2004).

Despite its significance, just a few works regarding mineralized patterns have been produced to date. Ozawa and Yao (2002) were the first to create mineralizable patterns at the microscale. They presented two different methods for the formation of an apatite micropattern by a combination of a biomimetic process and transcription of a resist pattern. These two kinds of strategies for forming apatite micropatterns are applicable for the development of various smart biomaterials, such as cellular biosensing devices, by combining the bioaffinity of apatite with properties of other functional materials at the microscale. However, one can go even further and produce micropatterns based on bioactive nanoparticles.

Tan and Saltzman (2004), inspired by the natural mineralization process, created a material with a complex structural form in which a nanometer-scale mineral phase is organized in a controlled fashion on a micrometer-scale template that is preset by a controllable microfabrication process. The micrometer-scale structures were created using photolithography and reactive ion etching techniques. Acidic moieties were generated on the surface through silanization and succinylation. Later, a layer of nanostructured calcium phosphate was formed on the patterned surface in supersaturated calcium phosphate solution. The developed materials were biocompatible with bone cells, inducing a range of desirable cellular responses (Tan and Saltzman, 2004).

Shi et al. (2007) produced smart surfaces capable of controlling and triggering the occurrence of biomineralization in biodegradable substrates. The bioactive substrates were prepared from PLLA and reinforced with Bioglass®. A hollowed polycarbonate mask was used to expose only certain regions of the substrate surface to plasma treatment, allowing for the insertion of poly(N-isopropylacrylamide) (PNIPAAm) into specifically designated areas, creating a mineralizable pattern. It is found that such treatment, together with temperature, could trigger the formation of apatite on the biodegradable substrate upon immersion in SBF above the PNIPAAm lower critical solution temperature. On the other hand, no apatite is formed at room temperature. A control experiment on a material that is not subjected to surface treatment does not show any evidence of mineral deposition at the two analyzed temperatures. By patterning the surface, it was possible to merge the temperature switching and spatial control of biomimetic apatite formation. This concept could also extend to the biomimetic production of other minerals, where it would be triggered by another kind of stimulus (e.g., pH or ionic strength) in substrates with more complex geometries (Shi et al., 2007).

Luz et al. (2012) showed how bioactive nanoparticles may be easily patterned on a surface by micro-contact printing. This technique allowed the creation of a mineralizable pattern on a chitosan membrane using a poly(dimethylsiloxane) stamp inked in a BG-NPs pad. This membrane was then immersed in SBF and an apatitic pattern was created. Cells were also cultured on these membranes and results showed that L929 cells replicated the initial inorganic pattern preferring the environment created by the BG-NPs ionic release rather than migrating to chitosan. Depending on the nanoparticles chemistry and on the pattern created, cellular response can be directed and studied at the nanoscale. In addition to the opportunity to study cell behavior when exposed to mineralizable nanoparticles, this work suggests new possibilities of potential applications, not only in BTE, but also in guided tissue regeneration for skin and osteochondral areas, and also in angiogenesis.

It is already known that nanometer features tightly control osteoblast behavior (Lamers et al., 2010). Therefore, it would also be desirable that non-mineralizable substrates could be also successfully patterned at the nanoscale for BTE applications. Lamers et al. (2010) evaluated the role of different nanometer features, like (an)isotropy, pattern depth, width and spacing on (initial) cellular behavior, by using a very high throughput biochip. They observed that isotropic nanosquares do not induce morphological changes on osteoblasts, but they specifically enhance motility up to a maximum at a pattern spacing of 400 nm.

Puckett and coworkers (Puckett and Webster, 2007; Puckett et al., 2007) were also able to control the osteoblastic alignment on nanopatterned titanium. They showed that the orientation of the nanosurface coating of metal implants is also an important parameter when trying to enhance the osteoblastic adhesion. Osteoblast functions were studied on nanopatterned titanium substrates created by electron beam evaporation as a means of controlling the direction of bone growth. These patterns appear to promote bone cell functions more similar to long bones of the body. As a result of mimicking the structure and properties of bone, initial formation of anisotropic bone upon implantation could occur (Puckett et al., 2007). Other works show that grafting titanium bone implants with nanoparticles containing Arg-Gly-Asp-Cys peptide (RGDC) improves the adhesion behavior of cells seeded on these materials (Minh Ngoc et al., 2012).

Li et al. (2008) developed a novel method that allows the easy deposition of a wide variety of predetermined topographical geometries of nanoparticles of a bioactive material on both metallic and non-metallic surfaces. Using different mesh sizes and geometries of a gold template, hydroxyapatite nanoparticles suspended in ethanol have been electrohydrodynamically sprayed on titanium and glass substrates under carefully designed electric field conditions. Thus, different topographies, e.g. hexagonal, line and square, from hydroxyapatite nanoparticles were created on these substrates. The thickness of the topography could be controlled by varying the spraying time.

By precisely controlling material structures on both micrometer and nanometer length scale by using patterning techniques, researchers have proved that it is possible to create new materials with direct application in bone tissue engineering. Moreover, the developed techniques can be extrapolated to all the classes of biomedical materials, namely polymers, metals, ceramics, or hybrid combinations. Such research can aid the further development of smart surfaces that control cell behavior and account for improved osseointegration around orthopedic and dental implants (Lamers et al., 2010).

5.6 Drug-delivery systems

Bacterial infection commonly occurs after orthopedic surgeries. The administration of antibiotics by oral or intravenous via presents several drawbacks, such as systemic toxicity and limited bioavailability. Targeted drug-delivery systems for local delivery at the site of the implantation offer great advantage in the orthopedic field (Popat et al., 2007).

Nanostructures, in particular nanoparticles, may act as suitable and promising drug-delivery systems due to their ability of enhancing endocytosis of drugs by targeting cells and also by facilitating capillary penetration (De et al., 2008; Yang et al., 2010).

The efficacy of drug-delivery systems based on nanostructures may address some issues related to the properties of currently used drugs, namely solubility, in vivo stability, pharmacokinetics, and biodistribution (Jong and Borm, 2008). Nevertheless, drug-delivery systems have made great progress in the past few years. Researchers are developing ways to use internal stimuli to control drug release instead of external ones in order to avoid cell damage (Mal et al., 2003; Hernandez et al., 2004; Nguyen et al., 2007; Lu et al., 2008; Park et al., 2009). For instance, these kinds of system allow the release of drugs when exposed to the acidic pHs of cancer cell lysosomes, integrating targeted drug-delivery with internal stimulus induced self-release.

Although drug-delivery has been a polymer-dominated field, the blossoming of nanotechnology means that ceramic materials are now showing much promise for numerous drug delivery applications (Yang et al., 2010). Ceramic nanodevices present some useful characteristics not shared by their polymer-based homologous alternatives. First, they usually have longer biodegradation times and are slowly degradable, meaning that it is easier to control drug release kinetics and also to retain drugs for longer times after administration. Also, the small swelling ratios of ceramics prevent the release of a high amount of drugs as is commonly seen in polymeric hydrogels, such as poly(2-hydroxyethyl methacrylate) drug-delivery systems (Yang et al., 2010). Since ceramics are normally bioactive, they can be of an extraordinary value when approaching drug delivery especially targeted for bone.

Merging drug-delivery science with BTE may bring outstanding results for this field. Mineralization at the nanoscale can be used as an ingenious way to control both particle structure by reinforcement and the drug release by controlling the dissolution. Some examples found in the literature indicate that drug-delivery is being associated with mineralization and bone regeneration.

Min et al. (2012) developed micelles with a core–shell-corona structure that in the aqueous phase provided the three distinct functional domains: a hydrated PEG outer corona for prolonged circulation, the anionic poly(L-aspartic acid) middle shell for calcium phosphate mineralization, and the hydrophobic poly(L-phenylalanine) inner core for doxorubicin loading. The doxorubicin release from the doxorubicin-loaded mineralized micelles at physiological pH was efficiently inhibited, whereas at an endosomal pH (pH 4.5), doxorubicin release was facilitated due to the rapid dissolution of the calcium phosphate mineral layers in the middle shell domains. The calcium phosphate mineralization on assembled nanoparticles may serve as a useful guide for enhancing the antitumor therapeutic efficacy of various polymer micelles and nanoaggregates.

Another in situ biomineralization approach lead to the production of poly(N-isopropylacrylamide) and calcium phosphates hybrid nanocomposites (Shi et al., 2012). Biomimetic self-assembly enabled the interaction between PAA and Ca²⁺, leading to the formation of a homogeneous and robust nanocomposite. Smart drug release was possible since the nanocomposites responded to pH and heat. The introduction of calcium phosphates nanocrystallines decreases the permeation of the encapsulated drug effectively.

In another work (Yao and Webster, 2009), titanium was anodized to possess nanotubular surface structures suitable for drug delivery. The nanotubes were 200 nm deep and had 80 nm of inner diameter. These surfaces were able to promote bone cell functions (such as adhesion and differentiation) in vitro and in vivo compared with unanodized titanium. In order to test local drug delivery, anodized titanium with nanotubular structures were loaded with penicillin-based antibiotics using a co-precipitation method in which drug molecules were mixed in SBF to collectively precipitate with calcium phosphate crystals. Results showed for the first time that such co-precipitated coatings on anodized nanotubular titanium could release drug molecules for up to three weeks whereas previous studies have demonstrated only a 150 min release of antibiotics through simple physical adsorption. These findings represent a promising surface treatment for titanium that could be used for local drug delivery for improving orthopedic applications (Yao and Webster, 2009).

Also, TiO₂ nanotubes filled with gentamicin were able to reduce bacterial adhesion on their surface while still being able to enhance osteoblast differentiation (Popat et al., 2007). Fig. 5.4 presents the results of this work.

A plethora of nanodevices may be used for drug delivery sytems in BTE, namely nanoparticles, nanofibers, nanoscaffolds, liposomes, dendrimers, and nanogels. For instance, a nanogel of cholesterol-bearing pullulan in combination with prostaglandin E₂ showed efficacy in inducing new bone formation (Kato et al., 2007). Nanodevices have become excellent platforms to design targeted drug-delivery systems for biomedical applications in general. For BTE in particular when bioactive materials like ceramics are chosen, several advantages may be listed, such as their ability to modulate drug release kinetics, incorporate multifunctional molecules and target specific focus sites. Furthermore, the potential and development of hybrid or composite ceramic-polymer drug-delivery systems that incorporate the benefits from other types of materials should not be neglected (Yang et al., 2010).

Although some challenges remain, mostly regarding toxicity issues (Motskin et al., 2009), it is important to continue the effort of understanding the metabolism and elimination routes from the body of drug-delivery nanoplatforms in order to guarantee promising avenues to diagnose, understand and treat numerous bone diseases through drug-delivery while regenerating the tissue.

5.7 Nanocomposites

Natural mineralized composites, such as bone, tooth, and shells, result from the mixture of two or more phases, where at least one of them is in the nanometer size range. They exhibit ordered, complex, hierarchical microstructures that current synthetic composite materials cannot achieve (Ji and Gao, 2010; Luz and Mano, 2010).

Experiments show that the improved strength and toughness of natural biocomposites is caused by a structural and functional organization at different length scales, including a nanometer scale (Mayer, 2005; Younis et al., 2012). These characteristics should be taken into account when fabricating bioinspired advanced materials (Deville et al., 2006).

It has already been shown that ceramics at the nanoscale stimulate interactions between materials and cells (Webster et al., 2000). Inorganic biomaterials present high brittleness. It is therefore difficult to obtain porous scaffolds based solely on pure inorganic materials (such as bioceramics or bioglasses). However, the combination of inorganic nanoparticles or nanofibers with polymeric systems enables the reinforcement of nanocomposites improving their mechanical properties. Hence, these systems possess the potential to be used in a series of orthopedic applications, including BTE and regeneration (Boccaccini et al., 2010). Biopolymers are biodegradable and also contain structural groups similar to natural extracellular components, being thus suitable for implantation in living bodies (Peter et al., 2010a). Besides biodegradability and biocompatibility, nanocomposites for BTE also require an appropriate porous structure and the ability to induce hydroxyapatite precipitation in SBF solution in order to obtain tissue regeneration from both a chemical and a biological activity (Rocha de Oliveira et al., 2012).

The choice of the technique used in the production process of the BTE device will be important to direct the final characteristics of the nanocomposites. Techniques such as thermally induced phase separation (Hong et al., 2008), sol–gel, solvent casting, electrospinning or combinations of more than one technique, such as solid–liquid phase separation combined with solvent extraction (El-Kady et al., 2010), are normally used to produce BTE nanocomposites.

Through freeze-drying, adequate microporosity is assured for cellular migration (Webster et al., 2000; Peter et al., 2009, 2010b). Cross-linking can be also used to combine different materials and improve their mechanical properties. For instances, Wang et al. (2006) synthesized a sol–gel derived bioactive nanocomposite containing BG-NPs and phosphatidylserine through cross-linking of collagen and hyaluronic acid by using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide and N-hydroxysuccinimide.

Self-assembly is a very suitable technique for BTE, since bone itself is based on self-assembly of hierarchical structures in collagen matrix, and involves the spontaneous nucleation of Ca/P ions and the oriented array of formed hydroxyapatite crystals on collagen template (Wang et al., 2012). Collagen allows the most successful mimicking approaches. It has been shown to align hydroxyapatite in vitro on the nanoscale (Deshpande and Beniash, 2008) and has been recently mineralized with macroscopic domains of oriented hydroxyapatite by utilizing its liquid-crystalline behavior (Nassif et al., 2010).

Recently, Wang et al. (2012) produced a hydroxyapatite/collagen biomimetic nanocomposite prepared through self-assembly. Hydroxyapatite nanocrystals formed as preferentially oriented needles, 50–100 nm in length, on the collagen fiber matrix. The nanocomposite obtained is analogous in both composition and nanostructured architecture to native bone. Longer aging time promotes the growth and purification of nano-hydroxyapatite on collagen based on a chemical bonding.

In such a biomimetic physiological environment, calcium phosphates mineralize and nucleate directly on the collagen molecules templates through self-assembly. Hydroxyapatite crystals are oriented in the c-axis direction parallel to the long axis of the collagen fibril via electrostatic interactions between the lateral COO⁻ in the triple helix structure of collagen and Ca²⁺ in the surface of hydroxyapatite (Wang et al., 2012).

Inorganic materials used to reinforce nanocomposites are normally: hydroxyapatite, β-tricalcium phosphate, calcium phosphate and montmorillonite (Peter et al., 2010a). These materials will also give a bioactive character to the nanocomposites.

BG-NPs are a popular nanofiller. To avoid possible aggregation of nanoparticles in the polymeric matrix due to incompatibility, and negative consequences for the mechanical properties, the surface of the particles must be modified. For instance, low-molecular-weight PLLA was grafted onto the surface of the sol–gel-derived BG-NPs by diisocyanate and the ring-opening polymerization of the L-lactide (Liu et al., 2009). By grafting organic molecules on BG-NPs, their dispersion was improved in the polymeric matrix and thus the mechanical properties did not decrease.

In general, the inclusion of ceramic nanoparticles in a polymeric matrix has positive consequences: alteration of water uptake and consequently the degradation rate and increase in biocompatibility, bioactivity, and cytocompatibility. Furthermore, the induction of a nanostructured topography on the surface of the composites improves protein adsorption in comparison to the unfilled polymer and the composites containing micron-sized bioactive glass particles (Misra et al., 2008).

Besides three-dimensional porous scaffold structures, nanocomposite membranes have been also applied in the field of bone tissue regeneration. Specifically, in guided tissue/bone regeneration, membranes are used as barriers to prevent the faster growing soft tissue cells from entering the defect space and to regenerate periodontal ligament, cementum, and bone. Membranes are important for guided tissue regeneration since they can improve healing of surrounding tissues. By directing soft tissue growth, infection will be prevented and repair will be facilitated (Fong et al., 2012).

Chitosan/BG-NP composite membranes were developed to be used as barrier membranes to soft tissues with possible applications in periodontal regeneration (Mota et al., 2012). It is possible to target the nanocomposites properties by adjusting the formulations of the inorganic part. Luz and Mano (2012a) showed this by preparing chitosan nanocomposite membranes using two distinct BG-NPs systems, namely SiO₂:CaO:P₂O₅ (mol.%) = 55:40:5 and SiO₂:CaO:P₂O₅:MgO (mol.%) = 64:26:5:5. The chitosan/magnesium-based nanoparticles composite presented moderate bioactive character. Also, their higher hydrophilicity was found to stimulate a better osteoblastic response towards cellular differentiation and mineralization. Finally, nanofibers dispersed on a biodegradable polymeric matrix have been also used to reinforce BTE composites (Chen et al., 2011).

Some conclusions can be drawn regarding the multitude of works that develop nanocomposites for BTE. In general, the addition of nanoparticles has a significant stiffening effect on the composite modulus. Surface modification is also useful to establish deeper bonding between inorganic and organic material, which will also improve the nanocomposite mechanical properties (Khaled et al., 2007; Liu et al., 2008, 2009; Misra et al., 2008). Interconnected pores (Zheng et al., 2007) with pore sizes in the range 150–500 μm (Peter et al., 2010b) have been obtained. In BTE applications, interconnected pores with interconnection sizes above 300 μm are reported to be necessary to avoid hypoxic conditions and favor osteogenesis (Karageorgiou and Kaplan, 2005).

Nanocomposites represent a viable solution in BTE strategies. By varying parameters such as the polymer cross-link or the size and composition of the nanofillers, it is possible to adapt the composite chemical and physical characteristics to a specific clinical need.

5.8 Nanogels and injectable systems

Nanomedicine devices and techniques allow the improvement of minimally invasive procedures that can bring great benefit for the orthopedic field. Injectable matrixes may also deliver encapsulated cells and bioactive agents.

To target the orthopedic reconstructive and regenerative medicine in particular, chitosan-β-glycerophosphate salt formulations were mixed with BG-NPs in order to conceive injectable thermo-responsive hydrogels with rheological properties and gelation points adequate for intracorporal injection. In vitro bioactivity tests, using incubation protocols in SBF, allowed the observation of bone-like apatite formation in the hydrogel formulations containing bioactive nanoparticles (Couto et al., 2009).

Besides the inclusion of nanoparticles in organic/inorganic nanocomposites, nanoparticles can also work as fillers in a variety of moldable and injectable systems, as bone cements. Calcium phosphate cements possess the advantage of self-hardening to form hydroxyapatite in the bone cavity, thereby avoiding the problem of sintered hydroxyapatite implants that require a perfect match between implant shape and bone defect (Miyamoto et al., 1995; Laurencin et al., 1999). Although calcium phosphate-based cements present very high osteoconductivity, their brittleness and low strength limit their use to only non-stress-bearing locations (Xu et al., 2004). Nanoscaled fillers inclusion on these systems may help to improve the composite’s strength until they match that of cortical bone.

Since bone cements under stress can develop cracks, some modifications may be performed in order to precisely tune the mechanical properties of the bone cements. For instance, Xu et al. fused silica particles with silicon carbide whiskers to roughen the whisker surfaces for enhanced retention in the resin matrix (Xu et al., 2004). The strength of the nanocomposites was three times higher than the strength achieved in previous studies for conventional bioactive composites containing hydroxyapatite particles in this type of resins. The mechanical properties of the final composite nearly matched those of cortical bone and trabecular bone.

Stevens et al. (2005) injected calcium cross-linked alginate gels or modified hyaluronic acid gels into an artificial space between the tibia and the periosteum in rabbit. This stimulated bone and cartilage formation from resident progenitor cells in the inner layer of the periosteum. Injectable systems show that complex tissues can be generated from relatively simple materials by using the body as a bioreactor.

BTE approaches using injectable systems have been developed as a good scaffolding solution for irregularly shaped bone defects or with difficult access. In addition, such minimally invasive injectable systems offer the possibility of being combined with living cells or therapeutic drugs. Hence, they represent a very complete tissue engineering solution.

The development of targeted gels is also important for tissue regeneration because matrix elasticity is sufficient to induce lineage-specific differentiation of progenitor cells (Engler et al., 2006). For instance, Engler et al. (2004, 2006) showed that the differentiation of human mesenchymal stem cells is dependent on the 2D matrix elasticity of collagen-coated polyacrylamide substrates on which they were cultured. The cells grown on moderately firm substrates, with E of 10 kPa, similar to that of muscle, exhibited upregulated myogenic markers, whereas the others proceeded down an osteogenic pathway when cultured on a stiffer substrate (E of 35 kPa, similar to that of cross-linked collagen of osteoids).

Owing to their tissue-like elasticity, PEG-based hydrogels are highly desirable as 3D scaffolds for tissue regeneration. Also, their high permeability can mimic the native ECM. However, their applications are limited by their poor mechanical properties and bioinert nature, which restrict cell adhesion and spreading. To overcome these limitations, a novel hierarchical nanocomposite hydrogel composed of PEG diacrylate and hydroxyl mesoporous silica nanoparticles was developed via in situ free-radical polymerization. Structural and physicochemical characterization shows that hydroxyl mesoporous silica nanoparticles act as both reinforcing agents and adhesion sites in the hydrogel system, which significantly enhances mechanical properties and cellular affinity. Due to the anchoring effect of the hydroxyl mesoporous silica nanoparticles, the hybrid hydrogels possessed more greatly enhanced mechanical properties than the pure PEG hydrogels. Furthermore, the introduction of hydroxyl mesoporous silica nanoparticles into hydrogel systems may be beneficial for osteogenesis, making these nanocomposite hydrogels a promising scaffold for BTE (Yang et al., 2012).

By using injectable systems cells, drugs and molecular signals can be delivered to irregularly shaped bone defects in a minimally invasive manner. These injectable materials can also work as a scaffold, especially in locations that are not easily accessible.

5.9 Surface functionalization and templating

Many studies claim that the processes of biomineralization initiation and control are based on collagen templates and on the recognition of Ca²⁺ by a variety of noncollagenous biomolecules (Bradt et al., 1999), such as acidic proteins rich in aspartate, glutamate and its derivative γ-carboxyglutamate, phosphorylated residues such as phosphoserine (Wu et al., 2003), and also acidic glycosaminoglycans (GAGs) and other polysaccharides, which are significant bone constituents (Robey, 1996; Bradley et al., 2010). Soluble macromolecules interact with the matrix to control mineral deposition, crystal habit, and orientation within a structural framework that acts a template for nucleation and growth of hydroxyapatite (Olszta et al., 2007; Newcomb et al., 2012). While biopolymers provide a base for the initial organization of mineral ions into a crystallographic layer, biomolecules are also presumed to play an inhibitory role in mineralization by binding to specific crystal surfaces, thus preventing further growth of that surface (Shaw and Ferris, 2008; Bradley et al., 2010). In the light of this knowledge, the controlled organization of well-defined, bioinspired architectures based on organic–inorganic hybrid materials is of great interest for applications in tissue regeneration (Newcomb et al., 2012).

In an effort to mimic the composition of mineralized tissue in artificial systems, several organic templates have been investigated for their ability to nucleate biological minerals on the nanoscale, including self-assembled monolayers (Dey et al., 2010), biopolymers, phospholipids (Collier and Messersmith, 2001), and poly(amino acids) (Bradley et al., 2010). Most of these systems fail to mimic the hierarchical structure of bone, since they are not able to induce the nanoscale crystallographic alignment of hydroxyapatite.

Medical devices able to induce osseointegration regardless of the implantation site, or both bone quantity or quality, need to exhibit the right functional groups for apatite nucleation on their surfaces (Alves et al., 2010). Some functional groups have been shown to induce bone-like apatite formation, namely Ti-OH, Zr-OH, Nb-OH, Ta-OH, —COOH, Si-OH and PO₄H₂ (Alves et al., 2010). These groups have in common a negatively charged character, given by the neutral isoelectric points at pH values much lower than 7. They will attract Ca²⁺, ${{\mathrm{PO}}_4}^{3-}$ and ${{\mathrm{CO}}_3}^{2-}$ from the environment, forming an amorphous layer that will later crystalize in hydroxyl carbonate apatite with oriented apatite crystals (Dey et al., 2010; Hench and Thompson, 2010).

Since the ECM, which is responsible for templating hydroxyapatite in mineralized tissues, contains 1D, fiber-like nanostructures, nanodevices mimicking these structures are gaining attention. For instance, peptide amphiphiles containing a hydrophobic alkyl tail covalently attached to a peptide sequence were recently used to study biomimetic mineralization (Newcomb et al., 2012). In this study the authors showed that the ability to nucleate oriented hydroxyapatite using a fibrous supramolecular template depends strongly on the details of its nanoscale architecture. Minerals oriented relative to the principal axis of the fibers, as it occurs in mammalian bone, were only nucleated on calcium binding nanostructures with curved, cylindrical architectures, and not on chemically similar ones with flat surfaces. When cylindrical nanostructures were part of a hierarchically aligned monodomain gel containing bundles of nanofibers, the ability to nucleate oriented crystals over multiple length scales was maintained. The templates described in this study may also be useful to investigate the role of alignment in cell signaling and preparation of therapeutic constructs to promote in vivo regeneration of mineralized tissues (Newcomb et al., 2012).

Besides the peptide-based self-assembly approach to designing templates for biomineralization, cylindrical assemblies of filamentous bacteriophage bearing acidic coat proteins have already demonstrated oriented hydroxyapatite mineralization (He et al., 2010; Wang et al., 2010).

Surfaces can also be functionalized targeting the control of cell behavior, and indirectly the cellular mineralization process. Rezania and Healy (1999) functionalized solid materials surfaces with peptide sequences incorporating both cell- and heparin-adhesive motifs. They were able to enhance the degree of cell surface interactions and, hence, influence the long-term formation of mineralized ECM in vitro.

In order to design new BTE devices regarding biomineralization at the nanoscale, it is important to understand the main mechanisms responsible for apatite induction. Surface functionalization may be a very useful tool to attribute bioactivity to materials not able to induce apatitic deposition for themselves. Moreover, mimicking nature’s organic templates towards the biomineralization process initiation may be the key to obtaining hierarchical organized nanostructures similar to the ones that can be found in bone.

5.10 Conclusions and future trends

The BTE field is evolving and presents a plethora of innovative features aimed at meeting the challenging demands of clinicians. Materials for bone replacement no longer need to be inert. Nowadays the challenge of BTE is the design of a matrix that mimics the natural properties of bone while providing a temporary scaffold for tissue regeneration, and at the same time maintaining the resorbable, bioactive, and biocompatible characteristics. A new class of smart materials is being created not only to replace bone, but also to regenerate the damaged tissue based on external or even internal stimulus.

Bone has been an inspirational source for tissue engineering, since it presents an extraordinary hierarchical architecture with notable biological and mechanical properties. It is known that its structure is based on aligned collagen bundles embedded with nanometer-sized inorganic hydroxyapatite crystals; however, the relationship between morphology of the organic matrix and orientation of mineral is poorly understood. Unraveling the underlying mineralization process and how that knowledge can lead to the development of new devices and materials sharing the same remarkable mechanical properties of this natural nanocomposite is the main goal of BTE.

The central goal of this chapter was to bring forward some of the recent advances of nanoscale design in biomineralization towards the development of new biomaterials for BTE in the field of nanomedicine applied to bone. In Section 5.1, some important concepts related to the issues being discussed were presented, as well as a brief review of the physiological and anatomical characteristics of bone tissue. Section 5.2 described the latest nanotechnology trends applied to the physical and chemical mimicking of bone environment using synthetic materials.

Hybrid organic–inorganic structures mimicking the composition of mineralized tissue for functional bone scaffolds were analyzed. The combination of biodegradable polymers (synthetic and natural) with nanoscale bioactive ceramics or fibers is emerging as a powerful approach toward third generation bioactive materials comprising injectable osteoconductive biomaterials, thin coatings and films or self-assembling osteoconductive nanobiomaterials. The impressive possibilities are closely linked to the broad range of properties they offer.

Nanostructures have been recognized as being a key point for the successful achievement of bone regeneration. These devices must possess the ability to sense biological demands and also to adapt to the mutability of the biological environment. In order to mimic the complex hierarchical structure of bone, new methodologies are arising or being adapted to meet nanoscale design challenges. Modern materials and devices are being created, based on a dynamic interaction with the biological environment, starting at the nanoscale, thus allowing bone tissue regeneration to be targeted at the cellular signal level. These materials need to be able to communicate with cells and direct them to adhere, proliferate, migrate, or differentiate.

A number of issues still need to be addressed. Some parameters of modern BTE devices lack optimization. Degradation rate and mechanical properties are required to match with new bone formation and also bone vascularization needs to be improved in order to avoid necrosis of the host tissue surrounding the implant. This is an important point as it limits the upscale of bioartificial devices. However, the long-term health effects of nanomaterials are still not fully understood. The development of scalable and reproducible manufacturing methods is also important. In the future, polyvalent devices comprising abilities as support, therapeutics, and bioactivity will be implanted in bone tissues able to respond perfectly to the demanding in vivo environment. These materials will be able to release drugs to specific organs. Efficient combination of growth factors with the materials will also improve their efficacy in regenerating healthy bone. However, this increasing complexity of nanodevices may compromise their financial viability regarding translation to clinical use owing to the level of chemical and physical complexity necessary for these materials to be able to influence cell behavior.

5.11 Acknowledgement

This work was supported by the Portuguese Foundation for Science and Technology (FCT), through project PTDC/CTM-BPC/112774/2009 and the PhD grant SFRH/BD/45777/2008.