Natural and synthetic peptide-based biomaterials for bone tissue engineering

Abstract Introduction Since the dawn of human civilisation, biomaterials used for bone tissue engineering have evolved dramatically. More than 5,000 years ago, the ancient Egyptians used carved shells to replace teeth, but today, a bewildering variety of biomaterials, including autografts, allografts, xenografts, demineralised matrices, ceramics, polymers and composites, are used to replace/repair/regenerate bone tissues. Within this potpourri, natural and synthetic peptide-based biomaterials are promising candidates that could facilitate bone regeneration. This owes much to two properties of these peptidic biomaterials: (i) their facile fabrication into two/three-dimensional scaffolds that mimic the extracellular matrix of the bone and (ii) their controlled functionalisation with chemical cues that direct bone tissue growth. This review examines how these two properties could lead to biomaterials that facilitate bone tissue growth. Conclusion In this review, we have shown how biomaterials used for bone regeneration and tissue engineering have progressed from the exploitation of biomaterials extracted from natural sources to the preparation of synthetic peptide-based scaffolds. While a great deal has been learned about fabricating a functional bone biomaterial, much also remains to be uncovered.


Introduction
Since the dawn of human civilisation, biomaterials used for bone tissue engineering have evolved dramatically.More than 5,000 years ago, the ancient Egyptians used carved shells to replace teeth 1 , but today, a bewildering variety of biomaterials, including autografts, allografts, xenografts, demineralised matrices, ceramics, polymers and composites, are used to replace/repair/regenerate bone tissues 2 .
Bone is a composite material comprising an organic matrix (mainly) of Type I collagen and inorganic minerals of calcium phosphate.While conventional biomaterials such as bioglass and bone cement are very important for sealing bone fractures 3 , they do not permit the bone tissue to regenerate on its own-an important property of healthy bone.As such, there is growing effort to explore biomaterials that could serve as templates for bone growth.One such class of biomaterials is based on natural and synthetic peptides 4 .Such natural and synthetic peptidebased biomaterials offer the valuable possibility to be intricately designed on the macro-and nanoscopic scales to mimic the extracellular matrix (ECM) of bone cells for optimal bone growth.These biomaterials can be categorised into two subsets, natural and synthetic ECM, which we discuss separately.Under natural ECM, the use of demineralised bone matrix, Matrigel, and collagen in growing bone tissue is explored.Under synthetic ECM, we first look at how important bone scaffold properties such as stiffness and scaffold pore size have to be taken into account to generate a functional bone biomaterial.Thereafter, we explore how the peptides can be utilised to precisely control the biomineralisation of the biomaterial.

Discussion
Natural ECM-harnessing the potential of natural materials Demineralised bone matrix, produced by acidic decalcification of natural bone, retains much of the proteinaceous components native to the bone.These proteinaceous components, including but not limited to growth factors and osteogenic agents, are likely to be responsible for the impressive osteoconductive and osteoinductive properties of the demineralised bone matrix observed by Urist 5 .The bone matrix may be dead, but it is capable of promoting cellular re-population of the scaffold within 3 weeks after decalcification.Indeed, much clinical success is achieved when demineralised bone matrix is surgically implanted into bone defects, leading to new bone formation and acceleration of bone healing.
Matrigel, which is derived from Engelbreth-Holm-Swarm mouse sarcoma cells, is a basement membrane extract that contains multiple extracellular components such as laminin, entactin and collagen.It is widely used as a substrate for culturing stem cells and can increase the osteogenic potential of adipose-derived mesenchymal stem cells (MSCs).For instance, Kang et al. used Matrigel as two-dimensional and three-dimensional (3D) scaffolds to culture tibia.Indeed, morphologically, the wider inserts (300 and 500 μm) were observed to have been remodelled to support bone marrow formation.Knychala et al. also studied the influence of pore geometry on neo-tissue formation by varying the pore width (200 to 600 μm) of long rectangular slots of bone cements and found that human bone marrow stromal cells grew faster into narrower pores 16 .
Material composition of bone scaffolds also plays an important role in the outcome of bone tissue engineering.Scaglione et al. loaded bone marrow-derived cells into two scaffolds made of hydroxyapatite and a hydroxyapatite-collagen composite respectively 17 .Both scaffolds were able to support the in-scaffold population with osteoblasts, but there were differences in osteoblast deposition.For the hydroxyapatite scaffold, bone deposition occurred in a layer-by-layer fashion, whereas for the composite scaffold, bone deposition occurred in a woven manner.In addition, ordered spatial organisation of various cell types such as adipocytes and hematopoietic cells were observed in the hydroxyapatite, but not the composite, scaffold.The differing outcomes of loading the bone cells into the two scaffolds are likely to be due to the different (material) surface chemical cues presented to the bone cells.This would indicate the importance of regulating the surface of the scaffold so that the biomaterial can integrate smoothly with the native bone for tissue regeneration.

Peptides are essential adjuvants for bone tissue growth
Besides possessing the requisite physical properties, the scaffold has to be equipped with the appropriate chemical cues to direct cell attachment and growth.In this regard, peptides have been shown to be conducive in this role.The tripeptide Arg-Gly-Asp (RGD) is known to facilitate the adhesion, spreading

Macroscopic property of bone scaffold directs bone tissue growth
Stiffness of a scaffold, on which cells are attached, is known to play an important role in influencing stem cell differentiation.Engler et al. have cultured MSCs on both soft and stiff polyacrylamide gels and observed that differentiation of MSCs into osteoblasts was favoured on stiff gels 11 .Kong et al. studied the cell-material interaction at the molecular level and found that material stiffness affects the focal adhesion of the cell to the material surface 12 .Consequently, the surface traction forces experienced by the cell are transmitted to the nucleus, thus influencing differentiation outcome.However, soft hydrogel scaffolds could also be a useful 3D model for osteogenesis 13 .When Mari-Buye et al. encapsulated mouse pre-osteoblasts in the soft peptide RADA16 hydrogel (commercially available as PuraMatrix™), the cells were able to communicate and self-organise themselves within the hydrogel.After 24 days, the cells mineralised the hydrogel, making the material mechanically suitable for bone tissue engineering.
High scaffold porosity is essential for angiogenesis and bone matrix deposition, processes that lead to healthy bone tissue growth.Murphy et al. studied the effect of pore size (ranging from 85 to 325 μm) of collagen-glycosaminoglycan scaffold on osteoblast adhesion and found that proliferation was most prolific in scaffolds with the largest pore size 14 .Chang et al. evaluated the extent of bone in-growth into porous cylindrical-type hydroxyapatite implants, with pore size varying from 50 to 500 μm, in rabbits 15 .After 8 weeks of implantation, the hydroxyapatite inserts were removed from the tibia of the rabbits and assessed for changes in compression strength and morphological structure.The 300 μm-wide insert displayed the greatest increase in compressive strength, suggesting bone in-growth from the adipose-derived MSCs and observed that cells cultivated with Matrigel showed greater attachment, proliferation and osteogenic differentiation than cells grown without Matrigel 6 .
Type I collagen, being the main organic component of bone, has naturally been utilised as a scaffold in bone tissue engineering.By immersing collagen scaffolds in simulated body fluid, Al-Munajjed et al. deposited a coating of hydroxyapatite on the scaffold 7 .The resultant collagen-hydroxyapatite composite is stiffer and supports cell growth better than pure collagen alone.A collagen-chitosan composite has also been used as a template for culturing human marrow stromal cells 8 .In the presence of osteogenic medium, the stromal cells secreted a functional ECM that could induce the differentiation of MSCs into osteogenic cells and promote nucleation of calcium phosphate.

Synthetic ECM-engineering on the macro-and nanoscopic scales
Although it is possible to derive ECM from natural sources, there are a number of disadvantages associated with its use experimentally and clinically: (i) the material is very expensive, (ii) its quality can vary batchwise and (iii) there may be deleterious agents that are intractable to eradicate.Hence, efforts to prepare synthetic scaffolds that are easily formulated and at low cost are gathering pace.Unsurprisingly, this synthetic approach has challenges to overcome as well, chief being replicating the physical properties of the structural scaffold of natural ECM-the importance of stiffness, pore size/geometry and material composition to tissue engineering outcomes is well-established 9 .Although it is a daunting task, it is achievable-as Kirkham et al. have shown, their peptide hydrogel could act as a scaffold for the remineralisation of teeth defects caused by dental caries 10 .pre-osteoblasts would initiate bone tissue growth and calcium binding of phosphate would initiate crystallisation of hydroxyapatite.Electron microscopic analysis of the hydroxyapatite formed along the nanofibres showed that it is indeed very similar to the calcium phosphate in mineralised ECM.Unsurprisingly, such peptide-amphiphiles can support the attachment, proliferation and differentiation of stem cells 24 .

Nanoscopic peptide engineering controls calcium phosphate mineralisation
An important aspect of bone tissue growth is the mineralisation of the scaffold by calcium phosphate, which can be controlled on short peptides.Nonoyama et al. have shown that it is possible to arrange molecules of the PEG-conjugated (leucineglutamate) 8 peptide anti-parallel to each other on a mica surface 25 .As the hydrophobic leucine interacts preferably with the mica surface, the hydrophilic glutamate is oriented toward the environment.This allows the carboxylate groups of glutamate to chelate calcium and, on treatment with phosphate, forms ordered arrays of calcium phosphate.This surface template of amphiphilic peptide clarifies that the mineralisation of calcium phosphate to either amorphous-calcium phosphate or crystalline hydroxyapatite is highly dependent on surface morphology, which how these scaffolds could be useful in bone regeneration and bone tissue engineering.
Instead of having multiple osteoactive peptides with different functions attached to the scaffold, it is possible to craft a scaffold out of a single multifunctional osteoactive peptide.In a nanoscopic structural tour de force, Hartgerink et al. designed a multifunctional peptide-amphiphile that features: (i) RGD for binding pre-osteoblasts, (ii) phosphorylated serine for binding calcium, (iii) flexible linker for structural flexibility, (iv) reversible crosslinking motif and (v) a hydrophobic tail for directing molecular self-assembly (Figure 1) 22 .On dissolution of the peptide-amphiphile in water, the hydrophobic tails would aggregate.However, due to electrostatic interactions between adjacent dipolar head groups, the peptide-amphiphile does not form micelles, but align in a parallel fashion to form nanofibres instead 23 .The mechanical strength of the fibre can be enhanced via oxidation of the thiol groups to form disulphide linkages between adjacent peptide-amphiphiles.This property can potentially allow the nanofibrils to conform to any arbitrary shape of bone defects before oxidation to lock down the form of the nanofibrils.The surface of the nanofibre is decorated with RGD and phosphate groups, which bind preosteoblasts and calcium, respectively.Population of the nanofibres with and, consequently, cytoskeletal organisation of osteoblasts.When RGD was incorporated into poly(ethylene glycol) (PEG) hydrogels containing osteoblasts, Burdick et al. observed a significant increase in the extent of hydrogel mineralisation, presumably due to enhanced attachment of osteoblasts in the hydrogel 18 .The bone morphogenetic protein (BMP) is one of the most important growth factors involved in bone regeneration.Lin et al. have shown that a BMP-analogue peptide, P24, can enhance the osteoblastic differentiation of bone marrow stromal cells in vitro 19 .When P24 was incorporated into a poly(lactateglycolate)-(PEG-aspartate) copolymer, the sustained release of P24 promoted bone formation in vivo.A similar analogue of the transforming growth factor, which is important in directing the 3D condensation of bone cells to form the 3D matrix 20 , should prove equally useful.
Osteoactive peptides are also tacked directly on the scaffold to exert their influence.The osteogenic growth peptide (ALK) is responsible for bone tissue formation, osteopontin (DGR) is involved in bone mineralisation and the RGD-based sequence (PGR) is involved in binding cells.When short segments of ALK, DGR and PGR were chemically coupled via solid-phase peptide synthesis to the RADA16 peptide, the resultant peptide-functionalised hydrogels not only promoted the proliferation of mouse pre-osteoblasts, but also their osteogenic differentiation 21 .Among the three peptides, the binding sequence PGR induced the highest degree of osteogenic differentiation.The presence of PGR in the RADA16 hydrogel can also increase cell proliferation up to three times compared to pure RADA16 hydrogel, as well as induce extensive cell migration within the hydrogel compared to pure RADA16 hydrogel.These observations highlight the importance of osteoblast attachment to the scaffold in both cellular processes and illustrate solution of tropocollagen monomers and calcium cations through a nanoporous polycarbonate track-etched membrane into a solution of phosphate anions, which produces fibrils of collagen, with thickness depending on the size of the nanopores.Without the nanoporous membrane, only aggregates of collagen were obtained, illustrating the essential role of the nanopores in directing nanofibril formation.Besides the surface of the collagen nanofibrils, calcium phosphate was also found to crystallise between the segments within nanofibrils, a pattern that resembles closely the native bone matrix.This process generates a scaffold material that is more than an order of magnitude stiffer than pure collagen and is capable of inducing the differentiation of human adipose-derived stem cells into osteoblasts.This approach certainly bodes well for the use of such calcium phosphate-infused nanofibrils to expedite bone healing 30 .

Conclusion
In this review, we have shown how biomaterials used for bone regeneration and tissue engineering have progressed from the exploitation of biomaterials extracted from natural sources to the preparation of synthetic peptide-based scaffolds.While a great deal has been learned about fabricating a functional bone biomaterial, much also remains to be uncovered.To modulate the bone remodelling process, the controlled release of various growth factors at the right time is essential to trigger tissue development processes, especially angiogenesis, in the right order.Angiogenesis is critical, as a constant blood supply is required to carry oxygen and nutrients into the bone matrix.The ability to stimulate blood vessel development in the scaffold is paramount to the biomaterial being able to support bone growth.Hence, besides macroscopic mechanical properties and nanoscopic chemical properties, the controlled temporal affects the equilibrium of calcium phosphate nucleation/dissolution and precipitate/crystal growth 26 .This underscores how important it is to control calcium phosphate mineralisation on the nanoscale in order to achieve the desired bone growth.Thus, understanding the hydroxyapatite formation process in detail would be required in the design of materials that assist bone regeneration.In this regard, efforts have also been devoted to the development of computer programs, such as RosettaSurface.Design, which can help to decipher how changes in peptide sequence can affect calcium phosphate mineralisation 27 .
To be useful in bone regeneration, the biomaterial should preferably be fluid enough to be injected and stiffen over time.Even in the form of hydrogel, the (leucine-glutamate) 8 peptide is able to direct calcium phosphate mineralisation to proceed in a controlled fashion on the peptide fibres; without the amphiphilic peptide, an aggregate of calcium phosphate would form instead 28 .This hydrogel is also thixotropic, i.e. on stretching, the hydrogel becomes fluid but reverts to a viscous state over time.This is a pertinent property as the hydrogel ought to remain in place to assist in bone regeneration and not spill into tissue crevices.As different bones have different calcium phosphate forms, e.g.amorphous-calcium phosphate and hydroxyapatite, it would be essential if the crystallisation can be controlled to produce either form.Indeed, while (leucineglutamate) 8 led to the formation of amorphous-calcium phosphate, the amphiphilic (valine-glutamatevaline-serine-valine-lysine-valineserine) 2 peptide led to the formation of hydroxyapatite under basic conditions.
It is also possible to prepare, on a large scale with a simple set-up, mineralised nanofibrils of collagen that resemble native bone structure 29 .This can be achieved via flowing a

Figure 1 :
Figure 1: Illustration of the functions corresponding to various sectors of the peptide-amphiphile.