Increasing incidence of bone defects due to trauma, disease, or tumor resection has given rise to a growing need for bone grafts
[1]. Currently, for the surgical treatment of bone defects, autologous bone grafts still represent the “gold standard” for bone repair
[2]. However, the clinical applications of autografts are restricted by the limited donor tissue availability and potential for severe postoperative complications
[3]. To overcome these limitations, intensive investigations in bone tissue engineering (BTE) and material science have been carried out to produce ideal bone substitute materials (BSMs) as an alternative to autografts
[4][5][6]. So far, various BSMs have been developed, and some have even entered clinical use, including metals and alloys (e.g., stainless steel, titanium alloy, and magnesium alloy)
[7], minerals (e.g., hydroxyapatite and tricalcium phosphate among, other calcium phosphate compounds)
[8], synthetic polymers (e.g., polyurethan)
[9], naturally derived polymers (e.g., collagens and gelatin)
[5], as well as a significant number of composites of these materials
[10][11]. However, there is much room for improvement in order for these biomaterials to become real autograft substitutes and achieve successful bone regeneration. Incomplete or defective regeneration can lead to loss of tissue mass or replacement with fibrotic scars, which are associated with impaired functional recovery
[12]. Angiogenesis is another big challenge for synthetic BSMs, and even BTE constructs, as an insufficient blood supply can lead to necrosis and ultimately, to failure of the bone replacement
[13]. Therefore, although clinical research in synthetic bone grafting has been conducted for more than a century, only a few of these methods are available for long-term clinical use. As the main component of bone matrix, collagen (type I) and its engineered forms plays an increasing important role as a component of bone substitute materials and in BTE because of its excellent biocompatibility, cell adhesion, and osteoconductivity
[4][14][15][16]. However, its shortcomings including poor mechanical properties, high degradability, and lack of osteoinductivity currently limited to date routine clinical applications
[4][14][17][18]. For this, many attempts have been made to improve collagen-based implants in bone tissue repair and engineering.
2. Collagen in Native Bone Tissues
Collagen (type I) is a major structural component of mammalian bone, constituting 90% of the organic components of the bone extracellular matrix (ECM)
[19]. Thus, it has been stated that the ideal synthetic bone grafts should mimic the ECM of autologous bones as much as possible, since the ECM found in natural tissues supports cell attachment, proliferation, and differentiation
[20]. Scaffolds should consist of appropriate ECM-like biochemistry and nano/micro-scale surface topographies in order to formulate favorable binding sites to actively regulate and control cell and tissue behavior
[17][21]. Therefore, understanding the synthesis, structure, and distribution of collagen in bone tissue is critical for bone regeneration
[17]. The process of collagen synthesis in bone tissue occurs mainly by fibroblasts and osteoblasts
[22]. Within the cells, polypeptides formed in the endoplasmic reticulum are the building blocks of collagen, called alpha chains
[23][24][25][26]. After translation and posttranslational modifications, three alpha chains are linked by disulfide bridges
[27], then twisted into a triple helical procollagen
[28]. The ability to form a triple helix is the most important property of collagen, which is mostly based on the repetitive Gly-X-Y triplet of alpha chains
[22]. Glycine, with the smallest volume, is located at the inside of the triple helix. X is usually a proline, but could also be other amino acids; and Y is often hydroxyproline. The alpha chains are twisted in a left-handed way, while the triple helixes are folded in a right-handed helix because of the presence of hydrogen bonds. The formed triple helical procollagen is excreted into ECM by transport vesicles. In the ECM, the globular N- and C-terminal propeptides are removed from procollagen by the N- and C- proteinases, which is the key step in the formation of mature collagen
[25]. The helix dimension is on the order of 1.25 nm wide and 300 nm long
[29], with a mass of ~285 kDa. The removal of the N- and C- terminal propeptides would trigger the self-assembly of supramolecular structures (collagen microfibrils, fibrils, and fibers). Finally, the covalent crosslinking and non-covalent bonds stabilize these supramolecular structures
[30].
In bone, groups of five triple helical collagen molecules form microfibrils. The microfibrils are spontaneously organized into fibrils and fibers that can be up to 1 cm in length and 1 mm in diameter and with the characteristic 67 nm banding feature, called the D-period
[31][32]. The hydroxyapatite (HA) nanocrystals are deposited by osteoblasts on the collagen fibrils, constituting the inorganic and organic phases of the bone matrix. The bone matrix is thus hierarchically structured, consisting of a 65% mineral phase, HA, a 35% organic phase (~90% type I collagen, 5% noncollagenous proteins, and 2% lipids by weight), and a residual amount of water (
Figure 1). All these scales play an important role in the remarkable mechanical properties of bone. HA provides the rigidity, and collagen fibers improve the toughness of bone. Thus, to better mimic the native bone tissue, collagens used in synthetic bone materials should maintain their natural structure as much as possible. This imposes high requirements on collagen sources and extraction techniques.
Figure 1. Hierarchical structure of type I collagen fiber and human bone.
3. Applications of Collagen in Bone Tissue Regeneration and Engineering
The earliest reports of the use of collagen in the biomedical field date back to the 19th century
[33]. Currently, collagens directly extracted from animal tissues, or produced as recombinant proteins, with or without further modifications according to crosslinking, polymerization, or fibrillization, have been widely applied in vitro as standardized 3D materials to investigate the influence of microstructural and mechanical features on cell behaviors, such as cell attachment, cell contraction, cell motility, and related gene expressions
[34][35]. Collagen-based materials in various forms, including membranes, sponges or matrices, hydrogel, and composite scaffolds, are also widely used in vivo to support bone tissue regeneration in different clinical applications
[18][36] (
Figure 2).
Figure 2. Widely used collagen-based medical devices for bone defect repair. (A) Porcine pericardium collagen membrane. (B) Lyophilized type I collagen sponge. (C) Scanning electron microscope (SEM) imaging of a cross-sectional view of a collagen membrane with a tightly packed smooth side at the bottom and a rough side on the top. (D) SEM image of lyophilized porous collagen sponge.
3.1. Collagen Membranes
Various kinds of membrane materials are applied in the clinical practice, such as in the field of dentistry, to achieve guided bone regeneration (GBR)
[37] (
Figure 3).
Figure 3. Clinical treatment of bone cysts. (A) Exposure of the mandible in the area of the cyst and opening. (B) Insertion of a xenogeneic bone substitute material, (C) covering with a pericardium-based collagen barrier membrane. (D–F) Condition after bony healing at 3 months.
Expanded polytetrafluoroethylene (ePTFE; Teflon) presented the first success for controlling the migration of soft tissue, and especially epithelial cells, to achieve GBR
[38]. However, with the wide use of ePTFE, this kind of non-resorbable membrane shows a high rate of wound dehiscence and infections, and often needs to be removed by a second operation, which causes a secondary damage to the wound
[39][40]. One promising alternative to overcome these concerns is the application of resorbable collagen membranes, which offer many advantages, such as a single-step surgical procedure, improved soft tissue healing, the incorporation of the membranes by the host tissues, and a quick resorption in case of exposure
[41]. However, the disadvantages of collagen membranes are related to their unfavorable mechanical properties and their unpredictable resorption rates, even with the use of native dermis-derived materials
[42] (
Figure 4).
Figure 4. Exemplary histological images of the tissue response to a cross-linked collagen membrane (CC) within the subcutaneous connective tissue (CT) at day 30 post implantation. Black arrows = macrophages, black arrowheads = multinucleated giant cells, green arrows = fibroblasts, red arrows = eosinophilic granulocytes, and yellow arrows = lymphocytes (HE-staining, 200× magnification, scalebar = 20 µm).
Thus, different approaches, such as the choice of other tissue sources and manufacturing methods, have been tested to overcome these issues
[43][44]. Interestingly, cross-linking based on ribose has been shown to be a favorable alternative for the production of collagen-based GBR membranes
[45]. Other methods to overcome these issues are the use of new manufacturing approaches, such as electrophoretic deposition, electrospinning, and 3D printing, to improve the mechanical and biological properties of collagen membranes
[46][47]. Recently, another potential material alternative, based on a resorbable magnesium grid for volume stability, combined with a collagen membrane or a pure magnesium membrane, have been developed for GBR procedures
[48][49].
3.2. Collagen Sponges or Matrices
Collagen sponge is one of the most useful biomaterials owing to its excellent function and properties, as well as its easy processing, sterilization, and preservation
[50]. Collagen sponges are generally formed using a freeze-drying process (also known as ice-crystal templating, lyophilization, or ice-segregation-induced self-assembly) (
Figure 5)
[51][52]. By altering the freezing conditions, such as freezing temperature, time, and molds, the pore size and shape of collagen sponge could be tailored
[53]. For the optimal bioactivity of tissue regeneration, the pores should be large enough to permit cell migration and nutrient diffusion, and small enough to promote cell attachment
[54][55].
Figure 5. Freeze-drying process of collagen sponge formation. Upon freezing, collagen molecules are entrapped within the developing ice crystals, which have formed into hexagonal structures.
FDA approved collagen sponge-based devices are mainly used as absorbable hemostatics in different clinical conditions (
Figure 6). For BTE, collagen sponges are mainly applied as basic scaffolds for carrying bioactive substances, including growth factors, cells, drugs, etc. Recently, a collagen sponge, combined with BMP-2 and a bridge protein, showed the improved safety of BMP-2 and effects on bone regeneration for spinal fusion
[56]. A gene delivery system encoding fibroblast growth factor (FGF-2) and BMP-2 embedded in collagen sponge showed optimal bone regeneration effects
[57]. Atelocollagen molecules self-assembled into collagen fibrils, then processed with freeze-drying and crosslinking to build up a collagen sponge scaffold, which showed improved osteogenic differentiation and bone formation in vitro and in vivo (
Figure 7)
[58]. In addition, collagen sponges play an important role for in vitro models. To mimic the complex hierarchical environment of the native ECM, experimental design tools and optimized freeze-casting systems present exciting opportunities for the tailored architectural design of ice-templated collagen sponge scaffolds
[59][60].
Figure 6. Tissue reactions and integration behavior of a collagen sponge for hemostasis in dental applications at day 15 post implantation within the subcutaneous connective tissue (CT) of Wistar rats. (A) Overview of the implantation bed. A peripheral region (PR and double arrows) in which a cellular migration was noticed was separable from a nearly cell-free central region (CR), showing the gradual integration pattern of the biomaterial (Azan-staining, “total scan,” 100× magnification, scalebar = 5mm). (B) Tissue reaction within the PR including mainly macrophages (black arrows), as well as lower numbers of eosinophils (blue arrows) in combination with collagen fiber apposition (asterisks) and blood vessel ingrowth (red arrows) (Giemsa-staining, 400× magnification, scalebar = 10 µm). (C) Tissue reactions within the CR, including single macrophages (black arrows) within the interspaces of the collagen fibers (asterisks) of the sponge (Azan-staining, 400× magnification, scalebar = 20 µm).
Figure 7. Collagen sponge scaffold built up with self-assembled atelocollagen fibrils, which improved preosteoblast proliferation and differentiation, and bone formation effects in vivo.
3.3. Collagen Hydrogels
Hydrogels are water-swellable polymer materials with a 3D network structure formed through crosslinking reactions
[61]. Naturally derived collagen hydrogels present more satisfying biocompatibility and biological activity compared to other synthetic polymers or hybrid polymers
[62]. Similarly, they exhibit low stiffness and rapid degradation, which hamper their application in animal models and clinical tests
[63]. To overcome these drawbacks, one approach is based on the self-assemble ability of collagen hydrogels, as they polymerize into a fibrillar structure at physiological pH and ionic strength and temperature following an entropy-driven process
[64][65]. Another efficient strategy is the production of composite hydrogels formed by the combination of collagen hydrogels and synthetic components, including functionalized polymers and organic/inorganic nanoparticles and ions
[61]. The hydrogel collagen nanocomposite, in combination with strontium and seeded mesenchymal stem cells, showed the highest radiographical and histological scores compared with only collagen hydrogel and other control groups in full-thickness bone defect regeneration in the rabbit model
[66]. The application of nanomaterials to improve the mechanical properties of hydrogel showed enhanced stem cell adhesion and provide valuable guidance for the design of hydrogel-based materials
[67]. Currently, a new generation of composite collagen hydrogels is widely applied as an injectable hydrogel scaffold for in situ bone tissue repair, flexible drug delivery systems (nanogels or microgels), and implanted bone tissue scaffolds using 3D-printing, electrospinning, or other techniques
[61].
3.4. Collagen-Based Composite Materials
3.4.1. Collagen Integrated with Organic or Inorganic Materials
In FDA and EU approved collagen-based composite material, collagen bone void fillers (CBVFs) are one of the protagonists, and they are used for the backfilling and structure stabilization of bone voids in orthopedic surgeries. CBVFs are commonly composed of inorganic materials, such as calcium sulfate or calcium phosphate augmented with hydroxyapatite, and a native tissue-derived collagen suspension or collagen hydrogels, providing improved biocompatibility, resorbability, and filling ability
[68]. In recent decades, CBVFs and other collagen-based scaffolds have been introduced into various materials, such as hyaluronic acid, alginate, glycosaminoglycans (GAGs), silk fibroin (SF), metals, bioactive glasses, and some novel carbon-based materials
[69][70][71][72][73][74]. For example, a mechanically robust, injectable, and thermoresponsive CBVF was formed with carboxylated single wall carbon nanotubes (COOH-SWCNTs), chitosan, and collagen, as reported by Kaur et al.
[75]. They found that the material showed improved mechanical properties and bioactivity with the addition of COOH-SWCNTs
[75].
3.4.2. Collagen-Based Composite Materials Loaded with Growth Factors, Cells, or Drugs
Although collagen type I fibers contain multiple biological cues that could directly interact with cells to modulate their adhesion, proliferation, and differentiation, largely improving the biocompatibility and osteoconductivity of bone substitute materials, collagen alone is not osteoinductive
[14]. Thus, to achieve effective bone replacement and sufficient tissue vascularization, collagen-based materials can be modified to include other bioactive substances, such as growth factors (e.g., BMPs, VEGFA, etc.), cells (e.g., BMSCs, osteoblasts, osteoclasts, HUVECs, etc.), and drugs (e.g., antibiotics). The first collagen bone graft loaded with recombinant human bone morphogenetic protein-2 (rhBMP-2) was approved by the FDA in 2002
[76]. However, after an initially promising start, concerns regarding safety and cost-effectiveness of BMPs have been raised
[77]. A solution may be the optimization of the delivery system of BMPs to decrease the need for high doses of BMPs and to prevent their applications
[78]. In addition to increasing the loading efficiency by these modifications, other attempts have been made by inhibiting the BMP antagonists and the combination of multiple growth factors
[79]. In addition, collagen materials could be integrated with drugs for antibacterial, osteogenesis, and angiogenesis activity. Nabavi et al. reported that collagen hydrogel loaded with tacrolimus, which can enhance osteogenic differentiation by activating BMP receptors, showed improved bone formation compared with that of other non-tacrolimus groups
[62]. Because of the interactions between the RGD (arginine-glycine-aspartate) sequence and stem cell integrin receptors, collagen-based materials can increase the adhesion, proliferation, and differentiation of stem cells, providing an ideal platform for cell delivery
[62]. Recent studies have reported some rational designs of materials by surface modification and stiffness adjustment with nanomaterials to tune the tether mobility and anisotropic nanoscale presentation of RGD
[67][80][81]. A collagen scaffold carried with dental pulp stem cells (DPSCs) showed a higher amount of calcification in the reconstructed defects
[82]. However, in another study, in vivo micro-CT analysis confirmed that the acellular scaffolds generated larger volumes of bone than the DPSCs seeded scaffolds
[83]. Some other attempts at creating collagen-based composites are also focusing on the integration of collagen with peptide delivery and gene therapy
[84][85][86][87]. These studies remain controversial and have yet to be explored. Another promising direction is the synergistic effects of multiple bioactive cations on vascularization and bone defect repair, which is a much safer and simpler option
[88].
4. Conclusions and Perspectives
Significant strides have been made in the field of collagen (type I) for bone tissue engineering. Advances in exploring various collagen sources, as well as extraction and purification processes, have made collagen preparation available with maximum native structure and minimum immunogenicity/antigenicity. Advances in collagen modifications have offered optimized collagen mechanical properties and biological resistance. For a wide variety of applications in bone grafting, collagens are often modified or combined with other materials to construct various bone substitute materials, such as collagen sponges, hydrogels, nanofibers/microfibers, or nanoparticles/microspheres. Collagen-based composites with bioceramic materials (e.g., HA, TCP, and BGs, etc.), carried with growth factors, peptides, cells, drugs, and genes, have made the multifunctionalized BSMs available with improved osteoconductivity, osteoinductivity, osteointegration, osteogenesis, and vascularization. Although many of them have shown great effects on bone regeneration in in vitro and in vivo studies, only a few of them are currently FDA- or EU- approved for clinical applications. These composites still face challenges regarding complete mechanical properties, biological stability and activity, immune response, regional vascularization, and other safety issues for human bone regeneration, which hinder their translations into clinical devices. However, in addition to the continuous development of new tools and technologies for material manufacturing, such as 3D printing and electrospinning, researchers are directing attention to patient-specific collagen-based materials. Rather than relying on the material alone, an integration of functional material manufacture and the innate regeneration potential of patients sheds new light on potential clinical translation within the next decade.