Design and Use of Chimeric Proteins Containing a Collagen-Binding Domain for Wound Healing and Bone Regeneration

Abstract

Collagen-based biomaterials are widely used in the field of tissue engineering; they can be loaded with biomolecules such as growth factors (GFs) to modulate the biological response of the host and thus improve its potential for regeneration. Recombinant chimeric GFs fused to a collagen-binding domain (CBD) have been reported to improve their bioavailability and the host response, especially when combined with an appropriate collagen-based biomaterial. This review first provides an extensive description of the various CBDs that have been fused to proteins, with a focus on the need for accurate characterization of their interaction with collagen. The second part of the review highlights the benefits of various CBD/GF fusion proteins that have been designed for wound healing and bone regeneration.

Introduction

Collagen has attracted a lot of attention in the field of tissue engineering given its excellent biocompatibility and biodegradability, along with high safety and weak antigenicity.¹ Collagen-based biomaterials have been commonly used in various formulations, including hydrogels, sponges, and microparticles. Most of them are made of denatured collagen, that is, gelatin,² and some are already approved by the Food and Drug Administration (FDA).³ Interestingly, the mechanical properties of collagen scaffolds can be tuned by crosslinking, resulting in higher tensile strength and proteolytic resistance^4,5 and allowing for their use in a wide scope of applications. Moreover, these biomaterials can been loaded with various biomolecules to be released over time, be it small molecules such as steroids, antibiotics, and chemotherapy agents, proteins such as growth factors (GFs) and antibodies, or liposomes.⁴ The potential benefits of using biomaterials as drug carriers are tremendous when compared to bolus injections: the half-life of drugs is usually increased and they concentrate locally, which improves the tissue response and prevents the undesirable side effects that arise with systemic distribution.⁶

Most of the collagen-derived biomaterials that have been developed for tissue engineering and regenerative medicine purposes simply trap the drugs in their network.⁶ Drug release is thus governed by diffusion through the pores of the scaffold, which makes sustained delivery over a long period difficult to achieve. In particular, there is generally an initial burst release, which causes high drug loss and possible side effects.^7,8 For example, a burst of bone morphogenetic protein-2 (BMP2) was shown to promote inflammation and osteoclastic activity.⁷ Although extended release may be attained by reducing the size of the pores, this strategy hinders nutrient transport and severely affects cell viability.⁹

GFs are frequently used in therapeutic strategies such as bone and cartilage engineering,¹⁰ myocardial regeneration,¹¹ and wound healing¹² because they can promote cell adhesion, migration, differentiation, and proliferation. Their efficacy, however, depends highly on their spatiotemporal delivery, an essential trait that can affect cell fate.¹³ Although no GF binding site has been reported in collagen yet,¹⁴ some GFs, for example, transforming growth factor-β1 (TGF-β1), basic fibroblast growth factor (bFGF), and BMP2, naturally present a strong affinity for collagen and can bind to collagen-based scaffolds via ionic interactions.^7,15,16 Their capture, however, depends on the isoelectric point of the GF ^7,17 and the release profile usually features an initial burst,⁷ which makes this approach unsuitable for controlled release.

A variety of techniques have been developed for the covalent chemical conjugation of GFs to collagen scaffolds so as to provide a highly localized and long-lasting signaling. The random covalent grafting of GFs has proved worthwhile,^18,19 in particular by photoirradiation²⁰ or through the NHS/EDC coupling chemistry that targets free amine groups on the GF.²¹ The attachment of the proteins to the substrate may however mask or alter their receptor binding moieties and, consequently, negatively impact their biological activity.^22,23 The covalent binding of GFs in an oriented manner with the help of a cysteine tag has also been investigated.²⁴ The addition of a cysteine may, however, hamper protein folding via the formation of unwanted disulfide bridges, hence also resulting in a loss of activity. Importantly, covalent binding may prevent the endocytosis of the GF, which can be crucial for signaling.²⁵

To circumvent these limitations, numerous research groups have focused on the tethering of GFs in a stable but noncovalent manner using specific affinity pairs. Strategies that have been reported involve biotin and streptavidin,²⁶ two complementary coil peptides,^27–29 DOPA-containing peptides,³⁰ or specific binding domains that target heparin/heparan sulfate,³¹ laminin,³² fibrin,³³ or fibronectin (FN).³⁴ In particular, fusion proteins with a collagen-binding domain (CBD) have been the subject of intense research to functionalize collagen-based biomaterials. Interestingly, type I collagen is the most abundant protein in the human body³⁵ and is found in the extracellular matrix (ECM) of bones, skin, tendons, cornea, artery walls, as well as internal organs. Hence, a recombinant collagen-binding GF tethered to a resorbable collagen biomaterial would also bear the potential to bind to the extracellular collagenous matrix after resorption of the implant, thus exhibiting a significantly prolonged bioavailability.

The purpose of this review is (1) to examine the various CBDs that have been characterized and discuss how their origin and size affect their affinity and specificity for collagen and (2) to assess to which extent collagen biomaterials functionalized with CBD fusion proteins have emerged as powerful tools in the fields of wound healing and bone regeneration.

Collagen-Binding Domains

The term CBD encompasses very distinct polypeptide domains, which are either engineered or derived from native collagen-binding proteins such as FN, the von Willebrand factor (vWF), and several collagenases. This diversity of origin results in a great heterogeneity of sizes, from a seven amino acid-long peptide to a 42-kDa protein domain (Table 1). Bearing in mind that these domains will be fused to GFs and that controlling their release is of paramount importance for tissue engineering and regenerative medicine applications, it is crucial to assess how their sequence and their length affect their interaction with collagen, especially in terms of stability.

Table 1.

Reported K_D and Calculated EC₅₀ for Various Collagen-Binding Polypeptides

CBD origin			Reported K_D		Calculated EC₅₀
Protein	Subunit	Substrate	SPR (nM)	ELISA (nM)	EC₅₀ (nM)	R²	n	Reference
Human FN	Full-length FN	Col I α1		13	70	0.975	15	⁴⁹
		Col I α1			340	0.999	6	³⁸
		Col II		58	75	0.991	6	⁵⁰
		Denatured Col II		2	3	0.980	6	⁵⁰
		Type A gelatin	0.08					¹⁴⁰
		Type B gelatin	2					¹⁴⁰
		Col IX	3					¹⁴¹
	I₆-II₁-II₂	Col I α1	31,000					¹⁴²
	I₆-II₁-II₂-I₇	Col I			410	0.981	10	⁵⁵
		Denatured Col I			580	0.985	10	⁵⁵
	I₆-II₁-II₂-I₇-HGF	Col I			76^a	0.983	6	¹⁴³
	I₆-II₁-II₂-I₇-I₈-I₉	Col I α1		420	610	0.995	13	⁴⁹
		Col I		800	750	0.941	10	⁵⁸
		Col I α2		1800	1200	0.995	28	³⁸
		(POG)₁₀		150,000	120,000	0.986	90	³⁸
		Gelatin		650	800	0.973	14	⁵³
	CQDSETRTFY	Col I			900,000	0.997	6	⁶⁰
Human vWF	Full-length vWF	Col I		1.8	2	0.962	10	⁶²
		Col III	3.4		0.8	0.985	8	⁶³
		Col I			8	0.992	8	¹⁴⁴
		Col VI		8	10	0.998	5	⁶⁴
		Col VI			3	0.988	8	¹⁴⁴
		Col III			20	0.984	5	⁷³
		Col III	2					³⁶
		Col I	30					³⁶
		Col III peptide	2					⁶⁹
	A1 domain	Col I	6					⁴⁰
		Col III	10	400	430	0.932	6	⁴⁰
	A3 domain	Col I		1800	1800	0.973	7	⁶⁷
		Col III	8					³⁶
		Col III peptide	1800					⁶⁹
		Col III		1600	2000	0.997	6	⁴⁰
	EGF-A3	Col I			580	0.998	5	¹⁴⁵
	21 kDa fragment	Col I			780	0.971	9	³⁷
	WREPSFCALS	Col I			29,000^a	0.999	5	³⁷
	WREPSFCALS-bFGF	Col I			1300^a	0.995	8	¹⁴⁶
		Col I			670	0.992	7	¹³²
PlGF2	PlGF2_(123–144)	Col I		126	110	0.996	7	¹⁴
	BMP2-PlGF2_(123–144)	Col I		110	100	0.986	7	¹⁴
	PDGF BB-PlGF2_(123–144)	Col I		96	100	0.991	7	¹⁴
	VEGF A-PlGF2_(123–144)	Col I		124	120	0.994	7	¹⁴
Clostridium histolyticum collagenase G	S3b	(POG)₈	547,000					⁸⁷
		Col I			12,000^a	0.915	10	¹⁰²
	S3aS3b	(POG)₈	63,000					⁸⁷
		Col I			5200^a	0.976	10	¹⁰²
C. histolyticum collagenase H	Full-length ColH	Col I		99.5	100	0.963	15	⁸⁵
	S3	Col I		15,900	22,000^a	0.990	17	⁸⁵
		Col I			12,000^a	0.985	9	¹⁰²
	S2bS3	Col I		339	2000	0.980	17	⁸⁵
		Col I			2100	0.934	9	¹⁰²
		(POG)₈	3					¹⁴⁷
	FGF1-S2bS3	(POG)₈	8					¹⁴⁷
Staphylococcus aureus CNA	N1-N2-N3	Col I		2100	1800	0.979	7	¹⁴⁸
	CNA55	Col I	1300	210	290	0.997	27	¹⁴⁹
		Col I		2200				⁹⁴
	N1-N2	Col I		200				⁹⁴
	CNA35	DBS4	3					⁹⁴
		(GPO)₁₁	140					⁹⁴
		(GPP)₁₁	7.5					⁹⁴
		Col I		500	260	0.992	11	⁹⁵
		Col I		21	26	0.991	11	¹⁴⁸
		Col I	91		17	0.988	6	⁹³
Vibrio mimicus metalloprotease	Full-length VMC	Col I		2100			5	⁹⁷
	CBD	Col I		4000			4	⁹⁷
Decorin	Full-length decorin	Col I		6	4	0.977	8	⁸⁹
		Col I	21					⁹¹
		Col VI	39					⁹¹
	LRR 4–5	Col I			7400	0.977	8	⁸⁹
Engineered peptide	TKKTLRT	Col I			2300	0.980	7	¹⁰⁰
	TKKTLRT-EGF	Col I		239	1600	0.937	6	¹⁵⁰
	TKKTLRT-VEGF	Col I		430	2400	0.963	7	¹⁵¹
	TKKTLRT-bFGF	Col I			270	0.991	7	¹³²
	TKKTLRT-BMP2	Col I		270	140	0.999	4	¹⁵²
		Col I			6600	0.979	4	¹²³
	TKKTLRT-NGF	Col I		510	4500	0.992	4	⁴⁷
		Col I		510	6500	0.951	6	¹⁵³
	TKKTLRT-NT3	Col I		350	850	0.973	7	¹⁵⁴
	TKKTLRT-PDGF	Col I		5.5	1100	0.971	7	⁴⁴
		Col I		92	1100	0.991	7	⁴³
	TKKTLRT-PTH	Col I		292	1200	0.996	7	¹⁵⁵
	TKKTLRT-BDNF	Col I		470	1100	0.995	8	¹⁵⁶
		Col I		420	1900	0.944	4	⁴⁶
	TKKTLRT-EphA4LBD	Col I		430	1500	0.990	7	¹⁵⁷
	TKKTLRT-PlexinB1LBD	Col I		360	2900	0.995	7	¹⁵⁷
	TKKTLRT-EGFR Ab	Col I		460	2800	0.955	8	¹⁵⁸
		Col I		210	1600	0.968	7	¹⁵⁹

See Quantitative Characterization of CBD-Collagen Interactions section for methodological approach.

The plateau (saturation) has not been reached during the experiment.

bFGF, basic fibroblast growth factor; BMP, bone morphogenetic protein; CBD, collagen-binding domain; ELISA, enzyme-linked immunosorbent assay; FN, fibronectin; LRR, leucine-rich repeat; PDGF, platelet-derived growth factor; PlGF, placental growth factor; SPR, surface plasmon resonance; VEGF, vascular endothelial growth factor; vWF, von Willebrand factor.

Quantitative characterization of CBD-collagen interactions

For most of the CBDs under study, either an apparent affinity (thermodynamic association constant, K_A, expressed in M⁻¹), or its inverse, the thermodynamic dissociation constant (K_D, expressed in M), has been reported to characterize the strength of their interaction with collagen. Given the diversity of their origin and of the types of collagen, it is unsurprising that the dissociation constant values broadly span from the low nanomolar to the micromolar (Table 1). Such a range of affinities for CBD-collagen interactions could be exploited to target certain types of collagen with high specificity and/or to finely tune the release rate of a protein of interest, by fusing it to an appropriate CBD. However, to avoid any misinterpretation of the data presented in Table 1, one may analyze them with caution.

On the one hand, many techniques have indeed been used by the research community to determine the affinities of the various CBDs for the different types of collagen. Those include surface plasmon resonance (SPR)-based biosensor assays,³⁶ enzyme-linked immunosorbent assays (ELISA),¹⁴ radioactivity assays by using iodinated proteins,³⁷ as well as fluorescence titration assays by measuring changes in anisotropy of fluorescein-labeled collagen chains.³⁸ All of these methods have their own merits and limitations to assess an apparent dissociation constant.³⁹ Interestingly, large differences can be pinpointed between the values derived from distinct techniques within the same report. For example, when characterizing the interactions between the von Willebrand A1 domain for type III collagen, Morales et al. determined an apparent K_D of 8 nM by SPR and of 400 nM by ELISA.⁴⁰ Such a discrepancy can result from several biases inherent to the techniques at hand. For instance, in most SPR assays, the collagen is covalently attached to the biosensor surface, whereas in ELISA, the substrate is nonspecifically adsorbed at the bottom of the well. Both approaches may thus lead to distinct alterations of the three-dimensional conformation of the triple helix of collagen and, as a result, affect its interaction with CBDs.⁴⁰ Other artifacts related to inappropriate assay conditions, as those documented in specialized reports,^41,42 may also be invoked to explain these discrepancies.

On the other hand, it is not uncommon that distinct research groups report different apparent K_D for the same interaction, even when assessed with the same assay (Table 1). For example, Sun et al. reported an apparent K_D of 5.5 nM for the capture of the fusion protein TKKTLRT-PDGF (platelet-derived growth factor) by a type I collagen membrane, whereas Lin et al. reported an apparent K_D of 97 nM for the same protein incubated on type I collagen-coated wells.^43,44 As much as the techniques, the procedure for data processing and K_D determination may therefore be worth questioning. In most of the articles here reviewed, the apparent K_D values were indeed derived by linearizing the experimental data via a Scatchard plot for instance. Such an approach is prone to introduce errors and biases,⁴⁵ especially when the plot deviates from linearity and when data are omitted.^46,47

To overcome these limitations, we here propose a rationalized method for data processing. By curve fitting the complete data set with the GraphPad Prism 6 Software (GraphPad Prism Software, Inc., San Diego, CA), using a Hill-type equation that assumes a 1:1 interaction, the half-maximal effective concentration (EC₅₀) of a CBD binding to collagen can be calculated: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} Y = { \frac { { Y_ { max } } \; } { 1 + { \frac { E { C_ { 50 } } } { \left[ { CBD } \right] } } } } . \end{align*} \end{document}

where [CBD] and Y correspond to the incubated CBD concentration and the corresponding response (signal minus blank), respectively. Y_max, the theoretical maximal response, and EC₅₀, the half-maximal effective concentration, are outputs. Note that the Hill slope was fixed at 1.0 for the sake of comparison.

This curve-fitting method was applied to all the CBD-collagen interaction studies when raw results were available, by extracting the data from the research article figures using the WebPlotDigitizer 3.9 software.⁴⁸ When applicable, the EC₅₀ values derived from the fits, the coefficient of determination (R²), as well as the number of experimental data points extracted from the figures (n) are given in Table 1, in addition to the apparent dissociation constant value (K_D) that was reported by the authors. Although all R² values were close to 1, which is indicative of a good fit between the model and the data, some research groups have used a low number of CBD concentrations in their investigation. Given that the reliability of any curve fitting is highly dependent on the number of data points, both the K_D values they reported and the EC₅₀ values we calculated need to be treated with caution. Furthermore, several cases of biphasic interactions were encountered, for which we only provided the lowest EC₅₀ (corresponding to the highest apparent affinity).

With this analytical approach, we found that the interaction of TKKTLRT-PDGF for type I collagen, with reported K_D values of 5.5 and 97 nM by Sun et al. and Lin et al., respectively, was best depicted by an EC₅₀ of 1100 nM in both cases.^43,44 This calculated value is plausible, given that the majority of the TKKTLRT fusion proteins feature a dissociation constant in the low micromolar range (Table 1).

Fibronectin

FN is a 440-kDa dimeric protein of the ECM. It binds to many biological partners such as collagen, fibrin, heparin, and integrin (Fig. 1A). FN has a very high apparent affinity for type I and type II collagens (K_D values of 13 and 58 nM, respectively).^49,50 FN binds even more strongly to denatured collagen, that is, gelatin (K_D = 2 nM),^50,51 which suggests that its binding sites within collagen are, at least partially, masked in the native triple helix.^49,52

FIG. 1.

Structure of several proteins containing a CBD: human FN (A), von Willebrand factor (B), placental growth factor (C), Clostridium histolyticum collagenases ColG and ColH (D), decorin (E), and Staphylococcus aureus adhesin (F). CBD, collagen-binding domain; FN, fibronectin.

The CBD of FN is a 42-kDa domain composed of four type I and two type II modules (I₆-II₁-II₂-I₇-I₈-I₉, see Fig. 1A). All six modules are required for full affinity,⁵³ supporting a cooperative mode of interaction where they simultaneously participate.^54,55 The FN type II modules are found in many other collagen-binding proteins, including the mannose receptor,⁵⁶ the factor XII,⁵⁷ and the matrix metalloproteinase 2 (MMP2).⁵⁵ Similarly to the full-length protein, the CBD of FN displays a higher affinity for gelatin than for properly folded collagens. It is, however, monomeric and has a moderate affinity for type I collagen (K_D = 420–800 nM),^49,58 hence highlighting the importance of FN dimerization (and of the resulting avidity) for optimal binding to collagen. Numerous subdomains containing three or more modules have been studied, including a short peptide sequence extracted from the module I₉, CQDSETRTFY, which binding to type I collagen proved to be very weak (EC₅₀ = 900,000 nM, Table 1).^59,60

von Willebrand factor

The vWF is a multimeric blood glycoprotein involved in hemostasis and formed from 275-kDa subunits crosslinked via disulfide bonds.⁶¹ The mature vWF has a very high affinity for type I, type III, and type VI collagens (K_D values of 1.8, 3.4, and 8 nM, respectively)^62–64 but poorly binds to gelatin, in stark contrast to FN.^65,66 Several moieties of the vWF came under scrutiny for their ability to bind to collagen (Fig. 1B), in particular, the 20.3-kDa A1 domain and the 19.3-kDa A3 domain.⁶⁷ Although conflicting results exist in the literature, the A1 domain of vWF does not seem to play a significant role in the binding of vWF to type I and type III collagens^68,69 but is crucial for its interaction with the type IV and type VI collagens.^64,70,71 In contrast, the A3 domain is considered to be the main collagen-binding site of vWF^72,73 and has but a moderate affinity for the type I and type III collagens (K_D values of 1800 and 1600 nM, respectively).^40,67 The A3 domain is thought to interact with the RGQAGVMGF/RGEOGNIGF sequences of the α1/α2 chains of type I collagen and with the RGQOGVMGF sequence in type II and III collagens.⁶⁹

Shorter domains of vWF have also been studied for their ability to bind to collagens, in particular, WREPSFCALS and WREPGRCELN that were derived from the bovine and human vWF, respectively (Fig. 1B).³⁷ To avoid unwanted homodimerization or misfolding,^74,75 two variants bearing a cysteine-to-methionine substitution in position 7, that is, WREPSFMALS and WREPGRMELN,^76,77 have also been studied. The WREPSFCALS peptide is derived from a 21-kDa sequence (F⁵⁷⁰-K⁶⁸²) of the bovine vWF precursor (NP_001192237.1); this sequence, first described by Takagi et al., is located at the end of the propeptide⁷⁸ that is cleaved during posttranslational modifications.⁷⁹ As opposed to the A3 domain, the affinity of the WREPSFCALS peptide for type I collagen is very weak (EC₅₀ = 29,000 nM, Table 1).³⁷ Nonetheless, it binds to all collagen types from I to V, in addition to gelatin.³⁷ Its 21-kDa parent fragment possesses the same characteristics but displays a higher affinity for type I collagen (EC₅₀ = 780 nM, Table 1).³⁷

Placental growth factor

The placental growth factor (PlGF), a member of the vascular endothelial growth factor (VEGF) family, is a glycoprotein essential for angiogenesis, in particular, over the bone marrow-derived cells.⁸⁰ The PlGF gene undergoes alternative splicing and gives rise to several homodimeric variants between 33.4 and 49.4 kDa. The sequence analysis of two splice variants that interact differently with the ECM, namely PlGF1 and PlGF2, enabled the identification of a 2.8-kDa amino acid sequence that binds to collagen (RRRPKGRGKRRREKQRPTDCHL, Fig. 1C).¹⁴ While initially considered as a heparin-binding domain,⁸¹ this sequence binds very strongly to type I collagen (K_D = 126 nM) and even more to FN, vitronectin, heparan sulfate, and other ECM proteins.¹⁴ Although it has a very high isoelectric point (pI = 12.0), a scrambled version of the peptide did not bind to type I collagen, thus indicating that the interaction of this CBD with collagen is sequence dependent and not merely electrostatic.¹⁴ Conversely, a C-terminal cysteine-to-serine substitution had little to no influence on the interaction.¹⁴ To our knowledge, no reports have yet been made regarding the specificity of this CBD toward other collagen types.

Clostridium histolyticum collagenases

ColH (116 kDa) and ColG (126 kDa) are two collagenases from the pathogenic C. histolyticum (Fig. 1D).⁸² They feature a broad substrate specificity and target various types of collagen as well as gelatin.^83,84 For instance, the full-length ColH enzyme displays a strong affinity for type I collagen (K_D = 99.5 nM).⁸⁵ Their CBDs, corresponding to the S3 domain (13 kDa) for ColH and the S3a+S3b domains (26 kDa) for ColG, are often associated with the polycystic kidney disease-like domain that precedes them, that is, S2a+S2b (20 kDa) for ColH and S2 (10 kDa) for ColG (Fig. 1D).

Interestingly, the S2b+S3 fragment of ColH binds to type I collagen in a biphasic manner, with a strong affinity interaction (K_D = 339 nM) followed by a moderate affinity interaction at higher concentrations (K_D = 2110 nM),⁸⁵ whereas the S3 domain alone poorly binds to it (K_D = 15,900 nM).⁸⁵

Another interesting CBD is the short fragment of ColG corresponding to the S3a+S3b domains (Fig. 1D), which binds to all collagen types from I to IV. This domain even binds to the collagen-like peptide (POG)_n, provided that the number of triad repeats, n, is large enough for the peptide to adopt a triple-helical conformation,⁸⁶ although weakly (K_D = 63,000 nM).⁸⁷ Likewise, it does not bind to gelatin, further suggesting that this CBD recognizes the triple-helical structure of collagen.^84,87

Decorin

Decorin is a small proteoglycan that interacts with collagen fibrils in all connective tissues.⁸⁸ Similarly to biglycan and fibromodulin, its core protein is mainly constituted of leucine-rich repeats (LRR). Although the glycosaminoglycan chains play a role in the binding of decorin to collagen, the core protein alone also binds to all types of collagen (I–VI) with strong affinity (e.g., K_D = 6 nM for type I collagen).^89–91 Major collagen binding sites were identified in a CBD composed of LRR 5 and 6 of the protein (NP_001911.1, Fig. 1E) that binds to type I collagen with moderate affinity (EC₅₀ = 7400 nM, Table 1).^88,89 Moreover, an 11-residue subdomain of this CBD, SYIRIADTNIT, showed promise as a short CBD, given its capacity to inhibit the interaction between decorin and type I collagen (K_i = 4000 nM).⁸⁹

Staphylococcus aureus adhesin

S. aureus is a very common opportunistic pathogen that persistently colonizes about 20% of the human population.⁹² As a gram-positive bacterium, S. aureus is covered with adhesive proteins, among which are adhesins.⁹³ CNA35 is a 35-kDa fragment of the S. aureus adhesin (CNA) that hooks the bacteria to collagen. This CBD is composed of two domains: N1 and N2, which can entwine the collagen triple helix (Fig. 1F).⁹⁴ The interaction of CNA35 with the type I collagen shows a biphasic behavior characterized by a high-affinity binding mode (K_D = 500 nM) and a low-affinity mode (our analysis indicated an EC₅₀ > 300,000 nM).⁹⁵ To a lower extent, this CBD also binds to collagen II, III, and IV, but neither to collagen V nor to collagen VI.⁹⁵

Vibrio mimicus metalloprotease

V. mimicus is a pathogenic bacterium that is responsible for certain cases of gastroenteritis. It expresses a metalloprotease named VMC and that binds to type I, II, and III collagens, in addition to gelatin.⁹⁶ A 33-residue-long CBD was identified, LVLSRPGQFAQWAQTVKNLGEQYNAEFAVWLDT (3.8 kDa), which has a moderate affinity for type I collagen (K_D = 4000 nM).⁹⁷ Interestingly, this CBD contains two FAXWXXT repeats, which were shown to be very important for its collagen-binding capacity. The second repeat in particular is well conserved across several species of Vibrio.⁹⁷

Engineered collagen-binding peptides

Several collagen-binding peptides were identified by phage display⁹⁸ or by ribosome display, an in vitro method for selection and evolution of peptides (unpublished results).⁹⁹ Another strategy to engineer a CBD based on the concept of sense–antisense peptide recognition, from which TKKTLRT ensued. This peptide was engineered as the antisense peptide of the collagenase cleavage site within the α2 chain of type I collagen. Interestingly, TKKLRT has almost the same hydrophilicity plot as the SQNPVQP and SSNPIQP sequences, both of which are found in a collagenase (the matrix metalloproteinase 1 and the neutrophil collagenase, respectively).¹⁰⁰ TKKTLRT can bind to gelatin,¹⁰¹ but has a moderate affinity for type I collagen (our analysis indicated an EC₅₀ of 2300 nM, Table 1).¹⁰⁰ In addition, Fukata et al. demonstrated that its maximal binding capacity to type I collagen is significantly weaker than that of the ColH-derived CBD that comprises the S2b+S3 domains.¹⁰² Similarly to the PlGF-derived CBD, TKKTLRT features a very high isoelectric point (pI = 11.2), and a scrambled counterpart that did not bind to collagen, namely LTTKKR, was identified.¹⁰⁰ The data thus indicated that the TKKTLRT-collagen interaction is sequence dependent and not solely electrostatic.

Collagen-mimetic peptides

The collagen-mimetic peptides (CMPs), also known as collagen hybridizing peptides, are composed of 6–10 repeats of the GPO or GPP triads, where O is the hydroxyproline.¹⁰³ These peptides mimic the triple helix conformation of collagen and have a high propensity to hybridize with collagen, both in vitro and in vivo.¹⁰³ More specifically, CMPs can interact with the unfolded strands of denatured collagen to reform a triple-helical structure.¹⁰⁴ On account of their capacity to target unfolded collagen, CMPs are of prime interest for the treatment of pathological conditions that cause collagen degradation, such as cancer, atherosclerosis, arthritis, and fibrosis.⁹⁸ CMPs have demonstrated high specificity and high affinity for type I–V collagens, with dissociation constants as low as 10 nM, depending of the number of triad repeats.¹⁰⁵ However, since they spontaneously self-assemble into helical homotrimers, CMPs require heating prior use so as to favor a monomeric state and thus to enable their interaction with denatured collagen. To circumvent this issue, Wahyudi et al. have modified a (GPO)_n sequence with a photocleavable group that prevents multimer formation, until the group is removed using UV-irradiation.⁹⁸

Applications of GFs Fused to CBDs

Numerous CBDs have been either discovered or engineered over the last decades, which provides the research community with a remarkable toolbox to target collagen for therapeutic applications. This collection of peptide sequences has been efficiently exploited for the stable attachment of proteins and cells to collagen, be it in scaffolds or in vivo. In fact, many recombinant collagen-binding proteins have been reported in the literature, most of which are GFs or cell-binding domains.

Regarding their production, bacterial expression has been the most common platform for these recombinant proteins because of the high yields that can be reached, even though bacteria produce nonglycosylated proteins. Indeed, all of the previously described CBDs, except those derived from FN and decorin, are naturally unglycosylated and can therefore be expressed in Escherichia coli without any loss of collagen-binding activity.^55,67,106 In the specific case of FN, it is not clearly established whether the glycosylation of the Asn511 residue within the module I₈ significantly impacts its affinity for collagen.^107–109 Altogether, bacterial expression is generally suitable as far as the collagen-binding moiety is concerned. It is, however, crucial to consider whether the glycosylation state of the fused protein of interest affects its biological activity to select the appropriate expression platform.

Clinical applications

In the fields of regenerative medicine and tissue engineering, GFs are of prime interest to direct cell fate and control tissue formation. Their very short half-life in vivo is, however, a major drawback that limits the efficiency of bolus injections,¹¹⁰ a problem that is sometimes addressed by their daily administration.¹¹¹ In an effort to prevent the diffusion of the GFs, while maintaining their bioactivity, chimeric proteins corresponding to GFs fused to a CBD have been developed (Fig. 2A). Among other applications, these fusion proteins can be administered to accelerate and extend the repair of cardiovascular wounds and skin wounds, such as ulcers and diabetic wounds. Accordingly, they can also benefit chirurgical reconstructions, such as the uterus, bladder, extrahepatic bile duct, urethra, and abdominal wall, as described in Table 2. Moreover, CBD-GF fusion proteins can be used to induce bone formation and thus to aid fracture healing or to improve the success of a bone graft. They can also be used in the treatment of critical-sized bone defects. For example, mandibular and calvarial or to induce vertical bone regeneration, critical in dentistry (Table 3).

FIG. 2.

Biological effect of various chimeric proteins containing a CBD: (A) collagen functionalization with growth factors or cytokines to stimulate cell proliferation or differentiation, (B) cells being engineered with CBDs at their surface to provide collagen anchorage, and (C) collagen decoration with cell-binding domains to promote cell adhesion.

Table 2.

Reported Chimeric Collagen-Binding Growth Factors Designed to Improve Wound Healing

Fusion protein (origin of the CBD)	Application, animal model	Significant results, comparison with pristine GF
EGF-CBD (S2b+S3 domains)¹⁶⁰	Subcutaneous injection in mice	✓ Higher retention at the site of injection (up to 10 days)
		✗ No mitogenic effect in vivo
EGF-CBD (WREPSFMALS)⁷⁶	Administration by enema in a nude mouse model of experimental colitis	✓ Significant accumulation at the areas of exposed collagen on the luminal surface of the colon
		✓ Better regeneration of the intestinal crypts
CBD-EGF (I₆-II₁-II₂-I₇-I₈-I₉)^161,162	Implantation combined to a collagen sponge/hydrogel in a full-thickness diabetic mouse wound	✓ Higher retention within the implant (up to 4 days)
		✓ Improved granulation tissue formation
		✗ Do not bind to a large extent to the wound when applied without carrier
	Intravascular administration in a rat/rabbit model of injured carotid artery	✓ Direct binding to injured vessels (without carrier)
		✓ Higher retention at the site of injury (even after 24 h of recirculated blood)
EGF-CBD (V. mimicus VMC)¹⁶³	Functionalization of collagen-coated culture plates	✓ Enable A-431/HaCaT cell lines to proliferate in vitro without soluble GF
		✗ Mitigated responses in terms of EGF bioactivity depending on the structure of the fusion protein
bFGF-CBD (S2b+S3 domains)¹⁶⁰	Subcutaneous injection in mice	✓ Improvement of the cell proliferation at the site of injection (up to 7 days after injection)
CBD-bFGF (WREPSFMALS)¹⁴⁶	Topical application within a collagen hydrogel in both normal or diabetic rat wounds	✓ Significant reduction of the healing time of wounds (7 vs. 9 days for bFGF)
		✓ Delivering the protein within a collagen hydrogel accelerates the healing even more
CBD-bFGF (TKKTLRT)¹³²	Implantation of a collagen membrane functionalized with CBD-bFGF	✓ Higher promotion of the cellularization and neovascularization of the implant
		✓ TKKTLRT-bFGF provides better results than WREPSFMALS-bFGF
CBD-bFGF (TKKTLRT)^164–167	Uterine horn reconstruction, bladder regeneration, full-thickness abdominal wall defect repair (all in rat), extrahepatic bile duct regeneration (in pig)	✓ Higher pregnancy rate after 90 days (87% vs. 60% for bFGF)
		✓ Better urodynamics bladder compliance after 90 days
		✓ Higher integration of myofibers in the implant, leading to a better mechanical strength after 90 days
		✓ Good regeneration of the tissues (✗ no bFGF control)
CBD-VEGF₁₂₁ (I₆-II₁-II₂-I₇-I₈-I₉)⁵²	Intramuscular injection in mice	✓ Higher retention at the injection site (up to 24 h)
		✓ Better interstitial cell proliferation (compared to VEGF₁₂₁ and VEGF₁₆₅)
		✓ No observed mobilization of EPCs in the blood after 4 days (no systemic side effect)
VEGF₁₂₁-CBD and PDGF-CBD (PlGF-2)¹⁴	Topical application in diabetic mouse full-thickness wounds	✓ Significant acceleration of the wound closure
		✓ Better granulation tissue formation and neovascularization
		✓ Drastic reduction of the vascular permeability side effect, compared to native VEGF₁₆₅
CBD-VEGF₁₂₁ (TKKTLRT)^{151,168–170}	Implantation of a collagen membrane functionalized with CBD-VEGF or injection of this protein in a rat infarcted myocardium	✓ Better neovascularization of the myocardium
		✓ Reduction of the scar size
		✓ Enhancement of the capillary growth
		✓ Improvement of the cardiac functions
		✓ CBD-VEGF was almost undetectable in the serum 3–6 h postinjection, in stark contrast with native VEGF₁₂₁
CBD-VEGF₁₂₁ (TKKTLRT) ^171,172	Full-thickness injury uterus (rat), extensive urethral defect (dog)	✓ Improvement of the pregnancy rate at the scar site (50 vs. 19% for VEGF)
		✓ Good regeneration of the urethra (✗ no VEGF control)
VEGF₁₆₄-CBD (S3 domain)^173–175	Subcutaneous injection in a rat dorsal-skin flap model	✓ Decrease in the necrosis rate and promotion of the neoangiogenesis
		✗ Controversial study (concerning the lack of bioactivity of native VEGF)
CBD-PDGF (TKKTLRT)^43,44	Subcutaneous implantation of a functionalized collagen membrane or administration of the membrane in a rabbit dermal ischemic ulcer	✓ Better promotion of the cellularization and neovascularization of the implant
		✓ Improved re-epithelialization of the wound
		✓ Better granulation tissue formation and neovascularization
CBD-HGF (I₆-II₁-II₂-I₇)^143,176	Subcutaneous implantation of a functionalized collagen sponge in rat	✓ Promotion of neovascularization in a dose-dependent manner
		✗ CBD-HGF do not remain longer than HGF in a mouse dorsal wound when administered without carrier
CBD-HGF (I₆-II₁-II₂-I₇)¹⁷⁷	Functionalized cardiac patch as treatment for myocardial defect in pig	✓ Significant improvement in the recovery of the myocardium (mechanical and electrophysiological functions)
CBD-HGF (I₆-II₁-II₂-I₇)¹²⁹	Intravascular administration in a rat model of injured carotid artery	✓ Accelerate the re-endothelialization of the artery
		✗ Significant aggravation of the intimal hyperplasia of the artery

The tick and cross symbols indicate whether the fusion of a CBD domain to the protein under study was beneficial or detrimental when used instead of the wild-type protein, respectively.

EPC, endothelial progenitor cell; GF, growth factor.

Table 3.

Reported Chimeric Collagen-Binding Growth Factors Designed to Improve Bone Regeneration

Fusion protein (origin of the CBD)	Application, animal model	Significant results, comparison with pristine GF
CBD-bFGF (TKKTLRT)¹⁷⁸	3D scaffold made of chondroitin sulfate and gelatin loaded with native TGF-β and CBD-bFGF	✓ Higher retention in the scaffold (half-life of 5 days vs. 1 day for bFGF)
		✓ Better cell niche for promoting the in vitro differentiation of MSCs into chondrocytes
CBD-bFGF (S2b+S3 domains)^179,180	Administration combined to a collagen powder in a mouse femur fracture/graft model	✓ Faster bone union
		✓ Induction of a hard callus bridge
		✓ bFGF-S2b+S3 markedly enhance bone formation in comparison with bFGF-S3(133)
CBD-bFGF (S2b+S3 domains)^181–183	Administration combined to a collagen sheet, demineralized bone powder, DBM, or (POG)₁₀	✓ Strong induction of bone formation at the site of implantation
		✓ The fusion protein can be administered combined to a large panel of collagen carriers
CBD-TGFβ1 (WREPSFMALS)^184–186	In vitro differentiation of MSC in a collagen gel	✓ Better chemotactic and mitogenic activity
		✓ Induction of osteogenic differentiation
		✓ BMP2 can be added to complete osteoinduction(186)
CBD-BMP2 (WREPSFCALS) or (TKKTLRT)^152,187	Subcutaneous implantation of functionalized DBM	✓ Formation of ectopic bone, both in the center and on the outskirts of the implant
		■ DBM is known to have osteoinductive properties by itself
CBD-BMP2 (WREPSFMALS)⁷⁵	Implantation of a functionalized collagen sponge in the dorsal muscle of rats	✓ Formation of a high quantity of mature bone, with trabeculae, medullar cavities, and blood vessels
		✓ Surprisingly, 400 ng of protein was sufficient to induce bone formation, on the contrary to native BMP2(131)
CBD-BMP2 (TKKTLRT)¹⁵²	Treatment of critical-sized rabbit mandible defect in combination with DBM	✓ Neoformed bone fills almost all of the defect site (80% vs. 60% for BMP2)
		✓ DBM does not have sufficient mechanical compression strength for weight-bearing bone application
CBD-BMP2 (TKKTLRT)¹²³	Subcutaneous implantation of functionalized MBM in rat	✓ Formation of bone-like tissue, with high ALP activity
		✓ MBM has similar mechanical properties as untreated cancellous bone
CBD-BMP2 (TKKTLRT)¹⁸⁸	Titanium cylinders, filled with MBM and covered with a functionalized collagen sheet, fixed on top of rabbit skulls	✓ Induction of bone formation within the MBM in contact with the functionalized collagen membrane: the membrane-induced bone formation within the cylinder (vertical bone regeneration)
BMP2-CBD and PDGF-CBD (PlGF-2)¹⁴	Topical delivery in a rat critical-sized calvarial defect	✓ This CBD displays a very strong binding for various proteins of the ECM
		✓ Improved bone coverage after 4 weeks (74% vs. 45% for pristine GFs)
		✓ Higher number of MSCs/pericytes
		✓ Very good results without carrier, and even better ones in the case of a delivery in a fibrin matrix
CBD-BMP3 (WREPSMALS)¹⁸⁹	Subcutaneous implantation of a functionalized collagen sponge in rat	✓ Induction of calcium deposition in a dose-dependent manner
		✗ No sign of bone formation nor increase in the ALP activity, both markers of complete osteoblastic differentiation¹⁹⁰
	Treatment of a rat calvarial defect with a functionalized collagen sponge or a collagen-coated ceramic block	✓ Promotion of bone repair: improvement of the bone coverage, displaying marked osteoinductive differentiation
CBD-BMP4 (I₆-II₁-II₂-I₇-I₈-I₉)¹⁹¹	Subcutaneous implantation of a collagen-PLGA hybrid scaffold, functionalized with CBD-BMP4, and seeded with MSCs	✓ Improved calcium deposition
		✓ Higher expression of osteogenic markers
CBD-BMP4 (I₆-II₁-II₂-I₇-I₈-I₉)¹⁹²	Administration of CBD-BMP4 as treatment for a mouse cranial bone defect	✓ Significant improvement in the ossification area of the defect (58% vs. 37% for BMP4)
		✓ The CBD of FN itself had a small positive effect on induction of bone formation
CBD-BMP4 (I₆-II₁-II₂-I₇-I₈-I₉)¹⁹³	Treatment of osteochondral defect in rabbit with functionalized photocrosslinkable gelatin scaffold, loaded with stromal cells	✓ Reparation of the subchondral bone via the generation of de novo articular cartilage-like tissue
PTH_1–33–CBD (S3 domain)^134,194	Intraperitoneal or intravascular injection in a mouse model of osteoporosis	✓ A single dose was sufficient to improve the mouse bone mineral density and the blood ALP activity for weeks
		■ The PTH-S2b+S3 fusion protein was not as effective as the PTH-S3 fusion protein
PTH_1–33–CBD (S3 domain)¹²⁸	Monthly subcutaneous injection in ovariectomized rat	✓ Significant increase in the bone mineral density of the rats
		✓ No observed PTH side effects, such as hypercalcemia

The tick, cross and square symbols indicate whether the fusion of a CBD domain to the protein under study was beneficial, detrimental or had no impact when used instead of the wild-type protein, respectively.

DBM, demineralized bone matrix; ECM, extracellular matrix; MBM, mineralized bone matrix; MSC, mesenchymal stem cell; TGF, transforming growth factor.

GFs fused to a CBD have generally provided better clinical results when compared to their native counterparts, which can be attributed to the increased bioavailability enabled by the collagen-binding activity. In most cases, the delivery of the fusion proteins is improved when they are combined with a proper carrier, which sustains their retention at the site of administration for several days. Most of the carriers are collagen derivatives that are approved by the FDA, and they can meet most clinical requirements given their variety of forms: sponge, hydrogel, membrane, powder, demineralized bone matrix (DBM) or mineralized bone matrix (MBM), and even the CMP (POG)₁₀ (Tables 2 –4). Interestingly, no carrier seems to be required in the specific case of vascular applications (Table 2).

Table 4.

Reported Collagen-Binding Fusion Proteins Designed to Improve Cell Adhesion

Fusion protein (origin of the CBD)	Application/animal model	Significant results, comparison with negative control
A3-GPI (A3 domain of vWF)¹³⁸	Infusion in a mouse injured artery of surface-functionalized bone marrow-derived EPC	✓ Innovative concept: the cells are surface-functionalized thanks to GPI fusion proteins
		✓ Numerous cells incorporate at the site of injury, leading to an improved re-endothelialization of the artery, in comparison with the control cells, in which the extracellular A3 domain was cleaved
CBD-EPLQLKM (TKKTLRT)¹⁹⁵	Implantation of functionalized collagen membranes in pig full-thickness wounds	✓ EPLQLKM peptide displays a high specific affinity for bone marrow-derived MSC
		✓ Promotion of a higher healing rate, with numerous infiltrating cells and high blood vessel density, compared to the collagen membrane alone
CBD-RGD (WREPSFMALS)¹⁹⁶	Intramuscular implantation in rats of collagen sponges, functionalized with BMP2 and the peptide CBD-RGD	✓ Promotion of the osteogenesis within the implant, displaying more bone trabeculae and mineralized matrix, in comparison with a collagen sponge loaded with either BMP2 or the peptide alone.
		✓ BMP2 was used at a subfunctional dose
CBD-OC-FN_9–10 (TKKTLRT)¹⁹⁷	Treatment of critically sized rat calvarial bone defects with functionalized collagen-coated PCL scaffolds	✓ The fusion protein is composed of the CBD, the OC protein, and the fragment FN_9–10 of FN (which includes the RGD sequence)
		✓ Improvement of the coverage of the bone defect (50% vs. 30% for the collagen-coated PCL scaffold alone)
CBD-PPFLMLLKGSTR (A3 domain of vWF)¹⁹⁸	In vitro culture of neural stem cells in a functionalized collagen hydrogel	✓ PPFLMLLKGSTR is derived from the G3 domain of laminin and binds to the α3β1 integrin¹⁹⁹
		✓ The survival of neural stem cells is increased after 48 h, suggesting an antiapoptotic effect, in comparison with the collagen hydrogel alone
CBD-G3-FNTPSIEKP (SYIRIADTNIT)¹³⁹	Implantation in a rat brain of a functionalized collagen hydrogel, seeded with neural stem cells	✓ The fusion protein is composed of the CBD, the G3 domain (from laminin α1), and the peptide FNTPSIEKP (from laminin γ1)
		✓ A significant higher number of transplanted cells survive and proliferate in comparison with the collagen hydrogel alone
CBD-GRILARGEINFK (A3 domain of vWF)²⁰⁰	In vitro culture of hippocampus neurons on functionalized type I collagen	✓ The GRILARGEINFK peptide is derived from neural cell adhesion molecule (NCAM)
		✓ Enhancement of the neuron adhesion and the neurite extension, in comparison with either a fusion protein without CBD, or the addition of soluble GRILARGEINFK peptide in the medium

GPI, glycosylphosphatidylinositol; OC, osteocalcin; PCL, polycaprolactone.

It is here worth mentioning that several GFs, such as the high-molecular-weight isoforms of VEGF, naturally possess a heparin-binding domain.¹¹² This feature can be used to improve their retention, be it on a heparinized biomaterial or via their capture by endogenous heparan sulfate.¹¹³ Nevertheless, in most studies, the administration of CBD-VEGF fusions yielded better results than that of native VEGF₁₆₅ (Table 2). Similarly, bFGF naturally binds to collagen and gelatin¹¹⁴ but CBD-bFGF fusions provided better outcomes (Tables 2 and 3).

Chimeric collagen-binding GFs for wound healing

Wound healing is a complex process comprising (1) an initial inflammatory phase that involves the clotting cascade and the recruitment of neutrophils and macrophages, (2) a secondary proliferative phase that is characterized by a strong angiogenesis, fibroplasia, formation of granulation tissue, and collagen deposition, and (3) a remodeling phase.¹¹⁵ Throughout this process, GFs play crucial roles such as chemoattractants (EGF, FGF-1/2, PDGF, TGF-β, SDF-1α),^116,117 mitogens (EGF, FGF-1/2, HGF, PDGF),^116,118 and promoters of angiogenesis (FGF-1/2, NGF-β, PDGF, TGF-β, VEGF).^111,119 All of these factors are released in a highly controlled manner for an optimal spatiotemporal distribution leading to the appropriate modulation of cell response.¹¹⁸ Hence, providing exogenous GFs may accelerate or improve the quality of the wound healing process, although the success of the approach greatly depends on the carrier used for their topical delivery.¹²⁰ For example, it has been long known that the topical application of EGF may accelerate the rate of epidermal regeneration in the cases of partial-thickness skin wounds or chronic wounds.^12,121 Several designs of collagen-binding GFs are reported in Table 2, which highlights their typical application and the advantages of their use over native GFs.

Chimeric collagen-binding GFs for bone regeneration

Bone regeneration is an intricate process regulated by numerous cytokines and GFs. Similarly to wound healing, these bioactive molecules play different roles, from chemoattraction (BMP, FGF, PDGF, VEGF, TGF-β)¹²² and cell proliferation (FGF, IGF, PDGF, TGF-β)¹²² to osteoinduction, that is, differentiation of mesenchymal stem cells (MSC) into mature bone cells (BMP, IL-11, TGF-β).¹²² To promote osteogenesis, exogenous GFs can thus be used to functionalize scaffolds such as DBM (made of spongy bone treated with an acid) or MBM (made of spongy bone treated with an oxidizing process).^33,123 It is here worth noticing that some of these GFs, such as BMP2, along with several resorbable biomaterials based on type I collagen, are already approved by the FDA for bone regeneration purposes.^124–126 In this context, collagen-binding fusion proteins are very appealing to better control the localization of GFs, and numerous chimeras were shown to improve bone regeneration when compared to their native counterparts (Table 3).

Side effects of chimeric collagen-binding GFs

Diffusible GFs feature a short half-life in vivo due to their sensitivity to denaturation, oxidation, or proteolysis.²³ Their translation as clinical tools thus requires high doses, sometimes repeated on a daily basis, to be provided to the site of action to induce the desired effect.¹¹¹ Supraphysiological concentrations of native GFs are, however, known to trigger unwanted side effects, both locally and systemically, limiting the amount that can be administered.^14,127 For instance, heterotopic bone formation and hypercalcemia were observed with BMP³³ and PTH¹²⁸ injections, respectively, and both GFs increased the risks of cancer. Similarly, intimal hyperplasia can arise when using HGF,¹²⁹ whereas VEGF is prone to cause hypotension and edema, see heart failure.¹³⁰

Given their biology, the use of recombinant CBD-GF fusions may trigger similar side effects. Nonetheless, their binding to collagen fibers hinders both their diffusion and degradation, thus allowing for higher local concentrations, lower systemic distribution, and prolonged activity than their native counterparts. Lower doses can thus be administered to obtain a similar effect. For instance, Visser et al. were able to trigger bone formation in a rat model with 400 ng of a CBD-BMP2 fusion protein,⁷⁵ while it is generally admitted that 460 ng is the required minimum.¹³¹ More interestingly, the fusion of a GF with a CBD may even abrogate systemic effects, as observed with VEGF and PTH (Tables 2 and 3).

Influence of the CBD

The collagen-binding ability of a fusion protein greatly depends on which CBD is fused to the GF. Several studies have shown that using a CBD that features a strong affinity for collagen, that is, a low K_D, will remain longer at the site of administration, hence providing better healing potential. For example, TKKTLRT-bFGF was shown to promote a stronger neovascularization than WREPSFMALS-bFGF,¹³² in good agreement with TKKTLRT displaying a higher affinity for collagen than WREPSFMALS (Table 1). Similarly, bFGF fused to the S2b+S3 domains of ColH markedly enhanced bone formation in comparison with bFGF fused to the S3 domain alone,¹³³ which has a lower affinity for collagen (Table 1).

Notwithstanding, opposite results were obtained with CBD-PTH chimeras: the PTH-S2b+S3 fusion protein was not as effective as the PTH-S3 fusion protein in improving the bone mineral density of mice. As an explanation, the authors suggested that either the fusion protein was cleaved in vivo within the S2b domain, thus causing its early release, or that the increase in collagen-binding activity caused its sequestration on collagen, thus reducing the bioavailability of PTH to its receptor.¹³⁴

Applications of Other Proteins Fused to CBDs

For regenerative medicine or tissue engineering applications, notwithstanding the fact that GFs can promote cell growth, proliferation, and differentiation, the efficient recruitment of the cells to the tissue may require an additional stimulus. In that endeavor, cell adhesion peptides can be fused to a CBD and administered to promote the colonization of a collagen-based biomaterial. Indeed, the migration of cells such as MSC is enhanced when their integrins bind to partners such as the RGD sequence^135,136 and various laminin domains.¹³⁷ Interestingly, besides providing anchorage, this type of interactions can also promote cell survival, differentiation, and proliferation.¹³⁶

As opposed to the recruitment of endogenous cells toward an implant, cell therapy focuses on the administration of exogenous cells. One of the main challenges in the field is to deliver those cells at a specific site, which proves to be challenging when the cells are administered intravascularly. Consequently, providing the cells with a collagen-binding ability could allow for their guidance into the bloodstream toward exposed endogenous collagen, for example, at the site of a vascular injury to improve re-endothelialization. Such a feat could be achieved thanks to a fusion protein that comprised a CBD and a glycosylphosphatidylinositol (GPI) anchor (Table 4). More specifically, when cells are incubated with a CBD-GPI chimera, the latter penetrates their lipid bilayer, thus providing them with a collagen-binding capacity, as depicted in Figure 2B.¹³⁸ Interestingly, CBD-adhesion peptide fusions can also be used for cell therapy when combined with a collagen-based carrier, as a mean to increase the survival rate of transplanted cells, for example, in the case of neural stem cell implantation in the brain.¹³⁹

Conclusion

In this review, we have highlighted the therapeutic potential of fusion proteins comprising a CBD, with a focus on GFs combined with collagen-based or collagen-coated scaffolds, for wound healing and bone regeneration applications. Numerous in vivo studies have indeed indicated that the fusion of a given GF to a CBD led to a drastic improvement of its therapeutic effect when compared to that of the native GF administered in its soluble form. This enhancement is most likely attributable to a prolonged bioavailability of the GF, be it due to the fact that the GF is preserved from degradation or that it is present for a longer period of time and at a higher local concentration in the vicinity of its cell surface receptors. Of interest, the use of a chimeric CBD-GF may also limit the occurrence of the systemic side effects that are observed with several diffusible GFs, which can again be explained by their efficient sequestration at the site of injury. The best example of this beneficial effect is arguably VEGF, which fusion to a CBD enabled the uncoupling of its angiogenic role from its hyperpermeability activity.

In most of the references collated in this review, researchers report the benefits of a given CBD-GF fusion protein when added to a single collagen-based scaffold. Despite all the promising results presented therein, these data remain hard to interpret as a whole to establish the importance of the biophysical traits of the interaction between the CBD and the collagen-based scaffold (e.g., GF loading capacity, binding affinity, and stability) on therapeutic outcomes.

Indeed, the amount of the chimeric protein bound to the scaffold or the impact of varying this quantity has seldom been reported. On the same note, the effect of replacing a CBD by another is merely addressed. A significant amount of work thus remains to be done to better characterize the interactions occurring between the various types of collagen and the different CBDs identified so far. These experiments are of prime importance, not only to resolve the discrepancies between several of the values listed in Table 1 but also to provide the research community with a broader knowledge of these interactions. We believe that such a characterization, if exhaustive, would guide the design of novel tailored CBD-GF chimeras for optimal therapeutic effects.

Footnotes

Acknowledgments

We apologize to authors whose relevant articles were not cited due to space limitations. This work was supported by the Canada Research Chair on Protein-Enhanced Biomaterials, by the Natural Sciences and Engineering Research Council of Canada, and by the MEDITIS training program.

Disclosure Statement

No competing financial interests exist.

References

Lee

C.H.

, Singla

, and Lee

Biomedical applications of collagen. Int J Pharm, 221, 1, 2001.

Hoffman

A.S.

Hydrogels for biomedical applications. Adv Drug Deliv Rev, 64, Supplement, 18, 2012.

Yano

, Tsuyuki

, Watanabe

, Kasanuki

, and Yamato

The regulation of allogeneic human cells and tissue products as biomaterials. Biomaterials, 34, 3165, 2013.

Friess

Collagen–biomaterial for drug delivery. Eur J Pharm Biopharm, 45, 113, 1998.

Young

, Wong

, Tabata

, and Mikos

A.G.

Gelatin as a delivery vehicle for the controlled release of bioactive molecules. J Control Release, 109, 256, 2005.

Ruszczak

, and Friess

Collagen as a carrier for on-site delivery of antibacterial drugs. Adv Drug Deliv Rev, 55, 1679, 2003.

King

W.J.

, and Krebsbach

P.H.

Growth factor delivery: how surface interactions modulate release in vitro and in vivo. Adv Drug Deliv Rev, 64, 1239, 2012.

Wallace

D.G.

, and Rosenblatt

Collagen gel systems for sustained delivery and tissue engineering. Adv Drug Deliv Rev, 55, 1631, 2003.

Pang

, and Greisler

H.P.

Using a type 1 collagen-based system to understand cell-scaffold interactions and to deliver chimeric collagen-binding growth factors for vascular tissue engineering. J Investig Med, 58, 845, 2010.

10.

Lee

S.-H.

, and Shin

Matrices and scaffolds for delivery of bioactive molecules in bone and cartilage tissue engineering. Adv Drug Deliv Rev, 59, 339, 2007.

11.

Miyagawa

, Sawa

, Taketani

, Kawaguchi

, Nakamura

, Matsuura

, et al. Myocardial regeneration therapy for heart failure hepatocyte growth factor enhances the effect of cellular cardiomyoplasty. Circulation, 105, 2556, 2002.

12.

Brown

G.L.

, Nanney

L.B.

, Griffen

, Cramer

A.B.

, Yancey

J.M.

, Curtsinger

L.J.

, et al. Enhancement of wound healing by topical treatment with epidermal growth factor. N Engl J Med, 321, 76, 1989.

13.

Samorezov

J.E.

, and Alsberg

Spatial regulation of controlled bioactive factor delivery for bone tissue engineering. Adv Drug Deliv Rev, 84, 45, 2015.

14.

Martino

M.M.

, Briquez

P.S.

, Güç

, Tortelli

, Kilarski

W.W.

, Metzger

, et al. Growth factors engineered for super-affinity to the extracellular matrix enhance tissue healing. Science, 343, 885, 2014.

15.

Yamamoto

, Ikada

, and Tabata

Controlled release of growth factors based on biodegradation of gelatin hydrogel. J Biomater Sci Polym Ed, 12, 77, 2001.

16.

Tabata

, and Ikada

Protein release from gelatin matrices. Adv Drug Deliv Rev, 31, 287, 1998.

17.

Muniruzzaman

, Tabata

, and Ikada

Complexation of basic fibroblast growth factor with gelatin. J Biomater Sci Polym Ed, 9, 459, 1998.

18.

Miyagi

, Chiu

L.L.Y.

, Cimini

, Weisel

R.D.

, Radisic

, and Li

R.-K.

Biodegradable collagen patch with covalently immobilized VEGF for myocardial repair. Biomaterials, 32, 1280, 2011.

19.

Koch

, Yao

, Grieb

, Prével

, Noah

E.M.

, and Steffens

G.C.M.

Enhancing angiogenesis in collagen matrices by covalent incorporation of VEGF. J Mater Sci Mater Med, 17, 735, 2006.

20.

Ito

, Hasuda

, Terai

, and Kitajima

Culture of human umbilical vein endothelial cells on immobilized vascular endothelial growth factor. J Biomed Mater Res A, 74, 659, 2005.

21.

Staros

J.V.

, Wright

R.W.

, and Swingle

D.M.

Enhancement by N-hydroxysulfosuccinimide of water-soluble carbodiimide-mediated coupling reactions. Anal Biochem, 156, 220, 1986.

22.

Boucher

, St-Laurent

, Loignon

, Jolicoeur

, De Crescenzo

, and Durocher

The bioactivity and receptor affinity of recombinant tagged EGF designed for tissue engineering applications is defined by the nature and position of the tags. Tissue Eng Part A, 14, 2069, 2008.

23.

Lee

, Silva

E.A.

, and Mooney

D.J.

Growth factor delivery-based tissue engineering: general approaches and a review of recent developments. J R Soc Interface, 8, 153, 2011.

24.

Backer

M.V.

, Patel

, Jehning

B.T.

, Claffey

K.P.

, and Backer

J.M.

Surface immobilization of active vascular endothelial growth factor via a cysteine-containing tag. Biomaterials, 27, 5452, 2006.

25.

Dobrowolski

, and De Robertis

E.M.

Endocytic control of growth factor signalling: multivesicular bodies as signalling organelles. Nat Rev Mol Cell Biol, 13, 53, 2012.

26.

Davis

M.E.

, Hsieh

P.C.H.

, Takahashi

, Song

, Zhang

, Kamm

R.D.

, et al. Local myocardial insulin-like growth factor 1 (IGF-1) delivery with biotinylated peptide nanofibers improves cell therapy for myocardial infarction. Proc Natl Acad Sci U S A, 103, 8155, 2006.

27.

Boucher

, Liberelle

, Jolicoeur

, Durocher

, and De Crescenzo

Epidermal growth factor tethered through coiled-coil interactions induces cell surface receptor phosphorylation. Bioconjug Chem, 20, 1569, 2009.

28.

Murschel

, Liberelle

, St-Laurent

, Jolicoeur

, Durocher

, and De Crescenzo

Coiled-coil-mediated grafting of bioactive vascular endothelial growth factor. Acta Biomater, 9, 6806, 2013.

29.

Lequoy

, Murschel

, Liberelle

, Lerouge

, and De Crescenzo

Controlled co-immobilization of EGF and VEGF to optimize vascular cell survival. Acta Biomater, 29, 239, 2016.

30.

Anderson

T.H.

, Yu

, Estrada

, Hammer

M.U.

, Waite

J.H.

, and Israelachvili

J.N.

The contribution of DOPA to substrate–peptide adhesion and internal cohesion of mussel-inspired synthetic peptide films. Adv Funct Mater, 20, 4196, 2010.

31.

Wissink

M.J.B.

, Beernink

, Pieper

J.S.

, Poot

A.A.

, Engbers

G.H.M.

, Beugeling

, et al. Binding and release of basic fibroblast growth factor from heparinized collagen matrices. Biomaterials, 22, 2291, 2001.

32.

Cao

, Sun

, Zhao

, Xiao

, Chen

, Gao

, et al. The use of laminin modified linear ordered collagen scaffolds loaded with laminin-binding ciliary neurotrophic factor for sciatic nerve regeneration in rats. Biomaterials, 32, 3939, 2011.

33.

Martino

M.M.

, Briquez

P.S.

, Maruyama

, and Hubbell

J.A.

Extracellular matrix-inspired growth factor delivery systems for bone regeneration. Adv Drug Deliv Rev, 94, 41, 2015.

34.

Kumra

, and Reinhardt

D.P.

Fibronectin-targeted drug delivery in cancer. Adv Drug Deliv Rev, 97, 101, 2016.

35.

Di Lullo

G.A.

, Sweeney

S.M.

, Korkko

, Ala-Kokko

, and San Antonio

J.D.

Mapping the ligand-binding sites and disease-associated mutations on the most abundant protein in the human, type I collagen. J Biol Chem, 277, 4223, 2002.

36.

, Moake

J.L.

, and McIntire

L.V.

Characterization of von Willebrand factor interaction with collagens in real time using surface plasmon resonance. Ann Biomed Eng, 30, 1107, 2002.

37.

Takagi

, Asai

, and Saito

A collagen/gelatin-binding decapeptide derived from bovine propolypeptide of von Willebrand factor. Biochemistry (Mosc.), 31, 8530, 1992.

38.

Ingham

K.C.

, Brew

S.A.

, and Migliorini

Type I collagen contains at least 14 cryptic fibronectin binding sites of similar affinity. Arch Biochem Biophys, 407, 217, 2002.

39.

Heinrich

, Tissot

, Hartmann

D.J.

, and Cohen

Comparison of the results obtained by ELISA and surface plasmon resonance for the determination of antibody affinity. J Immunol Methods, 352, 13, 2010.

40.

Morales

L.D.

, Martin

, and Cruz

M.A.

The interaction of von Willebrand factor-A1 domain with collagen: mutation G1324S (type 2 M von Willebrand disease) impairs the conformational change in A1 domain induced by collagen. J Thromb Haemost, 4, 417, 2006.

41.

Underwood

P.A.

Problems and pitfalls with measurement of antibody affinity using solid phase binding in the ELISA. J Immunol Methods, 164, 119, 1993.

42.

Nieba

, Krebber

, and Plückthun

Competition BIAcore for measuring true affinities: large differences from values determined from binding kinetics. Anal Biochem, 234, 155, 1996.

43.

Lin

, Chen

, Sun

, Zhao

, and Dai

The effect of collagen-targeting platelet-derived growth factor on cellularization and vascularization of collagen scaffolds. Biomaterials, 27, 5708, 2006.

44.

Sun

, Lin

, Xie

, Chen

, Zhao

, Han

, et al. Collagen membranes loaded with collagen-binding human PDGF-BB accelerate wound healing in a rabbit dermal ischemic ulcer model. Growth Factors Chur Switz, 25, 309, 2007.

45.

Motulsky

, and Christopoulos

Fitting Models to Biological Data Using Linear and Nonlinear Regression: A Practical Guide to Curve Fitting. Oxford, UK: Oxford University Press, 2004.

46.

Han

, Sun

, Lin

, Zhao

, Gao

, Zhao

, et al. Linear ordered collagen scaffolds loaded with collagen-binding brain-derived neurotrophic factor improve the recovery of spinal cord injury in rats. Tissue Eng Part A, 15, 2927, 2009.

47.

Sun

, Lin

, Chen

, Zhao

, and Dai

Promotion of peripheral nerve growth by collagen scaffolds loaded with collagen-targeting human nerve growth factor-beta. J Biomed Mater Res A, 83, 1054, 2007.

48.

Rohatgi

WebPlotDigitizer Version 3.9. 2015. Available at: http://arohatgi.info/WebPlotDigitizer (last accessed Mar 31, 2016).

49.

Ingham

K.C.

, Brew

S.A.

, and Isaacs

B.S.

Interaction of fibronectin and its gelatin-binding domains with fluorescent-labeled chains of type I collagen. J Biol Chem, 263, 4624, 1988.

50.

, Abbonante

, Yigit

, Balduini

, Kaplan

D.L.

, and Brodsky

Definition of the native and denatured type II collagen binding site for fibronectin using a recombinant collagen system. J Biol Chem, 289, 4941, 2014.

51.

Engvall

, and Ruoslahti

Binding of soluble form of fibroblast surface protein, fibronectin, to collagen. Int J Cancer, 20, 1, 1977.

52.

Ishikawa

, Eguchi

, Wada

, Iwami

, Tono

, Iwaguro

, et al. Establishment of a functionally active collagen-binding vascular endothelial growth factor fusion protein in situ. Arterioscler Thromb Vasc Biol, 26, 1998, 2006.

53.

Katagiri

, Brew

S.A.

, and Ingham

K.C.

All six modules of the gelatin-binding domain of fibronectin are required for full affinity. J Biol Chem, 278, 11897, 2003.

54.

, Lin

Y.-S.

, and Brodsky

Collagen interactions: drug design and delivery. Adv Drug Deliv Rev, 97, 69, 2016.

55.

Steffensen

, Xu

, Martin

P.A.

, and Zardeneta

Human fibronectin and MMP-2 collagen binding domains compete for collagen binding sites and modify cellular activation of MMP-2. Matrix Biol, 21, 399, 2002.

56.

Napper

C.E.

, Drickamer

, and Taylor

M.E.

Collagen binding by the mannose receptor mediated through the fibronectin type II domain. Biochem J, 395, 579, 2006.

57.

Citarella

, te Velthuis

, Helmer-Citterich

, and Hack

C.E.

Identification of a putative binding site for negatively charged surfaces in the fibronectin type II domain of human factor XII—an immunochemical and homology modeling approach. Thromb Haemost, 84, 1057, 2000.

58.

Ingham

K.C.

, Brew

S.A.

, and Migliorini

M.M.

Further localization of the gelatin-binding determinants within fibronectin. Active fragments devoid of type II homologous repeat modules. J Biol Chem, 264, 16977, 1989.

59.

Sistiabudi

, and Ivanisevic

Collagen-binding peptide interaction with retinal tissue surfaces. Langmuir, 24, 1591, 2008.

60.

Sistiabudi

, Paderi

, Panitch

, and Ivanisevic

Modification of native collagen with cell-adhesive peptide to promote RPE cell attachment on Bruch's membrane. Biotechnol Bioeng, 102, 1723, 2009.

61.

Sadler

J.E.

Biochemistry and genetics of Von Willebrand factor. Annu Rev Biochem, 67, 395, 1998.

62.

Pareti

F.I.

, Fujimura

, Dent

J.A.

, Holland

L.Z.

, Zimmerman

T.S.

, and Ruggeri

Z.M.

Isolation and characterization of a collagen binding domain in human von Willebrand factor. J Biol Chem, 261, 15310, 1986.

63.

van der Plas

R.M.

, Gomes

, Marquart

J.A.

, Vink

, Meijers

J.C.

, de Groot

P.G.

, et al. Binding of von Willebrand factor to collagen type III: role of specific amino acids in the collagen binding domain of vWF and effects of neighboring domains. Thromb Haemost, 84, 1005, 2000.

64.

Hoylaerts

M.F.

, Yamamoto

, Nuyts

, Vreys

, Deckmyn

, and Vermylen

von Willebrand factor binds to native collagen VI primarily via its A1 domain. Biochem J, 324(Pt 1), 185, 1997.

65.

Bockenstedt

, McDonagh

, and Handin

R.I.

Binding and covalent cross-linking of purified von Willebrand factor to native monomeric collagen. J Clin Invest, 78, 551, 1986.

66.

Morton

L.F.

, Griffin

, Pepper

D.S.

, and Barnes

M.J.

The interaction between collagens and factor VIII/von Willebrand factor: investigation of the structural requirements for interaction. Thromb Res, 32, 545, 1983.

67.

Cruz

M.A.

, Yuan

, Lee

J.R.

, Wise

R.J.

, and Handin

R.I.

Interaction of the von Willebrand factor (vWF) with collagen. Localization of the primary collagen-binding site by analysis of recombinant vWF a domain polypeptides. J Biol Chem, 270, 10822, 1995.

68.

Cruz

M.A.

, Handin

R.I.

, and Wise

R.J.

The interaction of the von Willebrand factor-A1 domain with platelet glycoprotein Ib/IX. The role of glycosylation and disulfide bonding in a monomeric recombinant A1 domain protein. J Biol Chem, 268, 21238, 1993.

69.

Lisman

, Raynal

, Groeneveld

, Maddox

, Peachey

A.R.

, Huizinga

E.G.

, et al. A single high-affinity binding site for von Willebrand factor in collagen III, identified using synthetic triple-helical peptides. Blood, 108, 3753, 2006.

70.

Flood

V.H.

, Schlauderaff

A.C.

, Haberichter

S.L.

, Slobodianuk

T.L.

, Jacobi

P.M.

, Bellissimo

D.B.

, et al. Crucial role for the VWF A1 domain in binding to type IV collagen. Blood, 125, 2297, 2015.

71.

Flood

V.H.

, Gill

J.C.

, Christopherson

P.A.

, Bellissimo

D.B.

, Friedman

K.D.

, Haberichter

S.L.

, et al. Critical von Willebrand factor A1 domain residues influence type VI collagen binding. J Thromb Haemost, 10, 1417, 2012.

72.

Sixma

J.J.

, Schiphorst

M.E.

, Verweij

C.L.

, and Pannekoek

Effect of deletion of the A1 domain of von Willebrand factor on its binding to heparin, collagen and platelets in the presence of ristocetin. Eur J Biochem, 196, 369, 1991.

73.

Lankhof

, Van Hoeij

, Schiphorst

M.E.

, Bracke

, Wu

Y.-P.

, Ijsseldijk

M.J.W.

, et al. A3 domain is essential for interaction of von Willebrand factor with collagen type III. Thromb Haemost, 75, 950, 1996.

74.

Tuan

T.L.

, Cheung

D.T.

, Wu

L.T.

, Yee

, Gabriel

, Han

, et al. Engineering, expression and renaturation of targeted TGF-beta fusion proteins. Connect Tissue Res, 34, 1, 1996.

75.

Visser

, Arrabal

P.M.

, Becerra

, Rinas

, and Cifuentes

The effect of an rhBMP-2 absorbable collagen sponge-targeted system on bone formation in vivo. Biomaterials, 30, 2032, 2009.

76.

Hall

F.L.

, Kaiser

, Liu

, Chen

Z.H.

, Hu

, Nimni

M.E.

, et al. Design, expression, and renaturation of a lesion-targeted recombinant epidermal growth factor-von Willebrand factor fusion protein: efficacy in an animal model of experimental colitis. Int J Mol Med, 6, 635, 2000.

77.

, Prescott

M.F.

, Liu

P.X.

, Chen

Z.H.

, Liau

, Gordon

E.M.

, et al. Long term inhibition of neointima formation in balloon-injured rat arteries by intraluminal instillation of a matrix-targeted retroviral vector bearing a cytocidal mutant cyclin G1 construct. Int J Mol Med, 8, 19, 2001.

78.

Takagi

, Fujisawa

, Sekiya

, and Saito

Collagen-binding domain within bovine propolypeptide of von Willebrand factor. J Biol Chem, 266, 5575, 1991.

79.

Wise

R.J.

, Pittman

D.D.

, Handin

R.I.

, Kaufman

R.J.

, and Orkin

S.H.

The propeptide of von Willebrand factor independently mediates the assembly of von Willebrand multimers. Cell, 52, 229, 1988.

80.

Carmeliet

, Moons

, Luttun

, Vincenti

, Compernolle

, De Mol

, et al. Synergism between vascular endothelial growth factor and placental growth factor contributes to angiogenesis and plasma extravasation in pathological conditions. Nat Med, 7, 575, 2001.

81.

Hauser

, and Weich

H.A.

A heparin-binding form of placenta growth factor (PlGF-2) is expressed in human umbilical vein endothelial cells and in placenta. Growth Factors, 9, 259, 1993.

82.

Eckhard

, Schönauer

, and Brandstetter

Structural basis for activity regulation and substrate preference of clostridial collagenases, G., H, and T. J Biol Chem, 288, 20184, 2013.

83.

Matsushita

, Jung

C.-M.

, Katayama

, Minami

, Takahashi

, and Okabe

Gene duplication and multiplicity of collagenases in Clostridium histolyticum. J Bacteriol, 181, 923, 1999.

84.

Toyoshima

, Matsushita

, Minami

, Nishi

, Okabe

, and Itano

Collagen-binding domain of a Clostridium histolyticum collagenase exhibits a broad substrate spectrum both in vitro and in vivo. Connect Tissue Res, 42, 281, 2001.

85.

Matsushita

, Jung

C.-M.

, Minami

, Katayama

, Nishi

, and Okabe

A study of the collagen-binding domain of a 116-kDa Clostridium histolyticum collagenase. J Biol Chem, 273, 3643, 1998.

86.

Bella

, Eaton

, Brodsky

, and Berman

H.M.

Crystal and molecular structure of a collagen-like peptide at 1.9 A resolution. Science, 266, 75, 1994.

87.

Matsushita

, Koide

, Kobayashi

, Nagata

, and Okabe

Substrate recognition by the collagen-binding domain of Clostridium histolyticum class I collagenase. J Biol Chem, 276, 8761, 2001.

88.

Svensson

, Heinegård

, and Oldberg

Decorin-binding sites for collagen type I are mainly located in leucine-rich repeats 4–5. J Biol Chem, 270, 20712, 1995.

89.

Kalamajski

, Aspberg

, and Oldberg

The decorin sequence SYIRIADTNIT binds collagen type I. J Biol Chem, 282, 16062, 2007.

90.

Bidanset

D.J.

, Guidry

, Rosenberg

L.C.

, Choi

H.U.

, Timpl

, and Hook

Binding of the proteoglycan decorin to collagen type VI. J Biol Chem, 267, 5250, 1992.

91.

Nareyeck

, Seidler

D.G.

, Troyer

, Rauterberg

, Kresse

, and Schönherr

Differential interactions of decorin and decorin mutants with type I and type VI collagens. Eur J Biochem, 271, 3389, 2004.

92.

Foster

T.J.

, Geoghegan

J.A.

, Ganesh

V.K.

, and Höök

Adhesion, invasion and evasion: the many functions of the surface proteins of Staphylococcus aureus. Nat Rev Microbiol, 12, 49, 2014.

93.

Kang

H.J.

, and Baker

E.N.

Intramolecular isopeptide bonds: protein crosslinks built for stress? Trends Biochem Sci, 36, 229, 2011.

94.

Zong

, Xu

, Liang

, Keene

D.R.

, Höök

, Gurusiddappa

, et al. A “Collagen Hug” model for Staphylococcus aureus CNA binding to collagen. EMBO J, 24, 4224, 2005.

95.

Krahn

K.N.

, Bouten

C.V.C.

, van Tuijl

, van Zandvoort

M.A.M.J.

, and Merkx

Fluorescently labeled collagen binding proteins allow specific visualization of collagen in tissues and live cell culture. Anal Biochem, 350, 177, 2006.

96.

Lee

J.-H.

, Ahn

S.-H.

, Lee

E.-M.

, Kim

Y.-O.

, Lee

S.-J.

, and Kong

I.-S.

Characterization of the enzyme activity of an extracellular metalloprotease (VMC) from Vibrio mimicus and its C-terminal deletions. FEMS Microbiol Lett, 223, 293, 2003.

97.

Lee

J.-H.

, Ahn

S.-H.

, Lee

E.-M.

, Jeong

S.-H.

, Kim

Y.-O.

, Lee

S.-J.

, et al. The FAXWXXT motif in the carboxyl terminus of Vibrio mimicus metalloprotease is involved in binding to collagen. FEBS Lett, 579, 2507, 2005.

98.

Wahyudi

, Reynolds

A.A.

, Li

, Owen

S.C.

, and Yu

S.M.

Targeting collagen for diagnostic imaging and therapeutic delivery. J Control Release, 240, 323, 2016.

99.

Tada

, Kitajima

, and Ito

Design and synthesis of binding growth factors. Int J Mol Sci, 13, 6053, 2012.

100.

de Souza

S.J.

, and Brentani

Collagen binding site in collagenase can be determined using the concept of sense-antisense peptide interactions. J Biol Chem, 267, 13763, 1992.

101.

, Chen

, Meng

, Liu

, Zheng

, Zhang

, et al. 3D bioprinting of BMSC-laden methacrylamide gelatin scaffolds with CBD-BMP2-collagen microfibers. Biofabrication, 7, 44104, 2015.

102.

Fukata

, Itoh

, Nonaka

, Ogawa

, Nakamura

, Matsushita

, et al. Direct cytocidal effect of galectin-9 localized on collagen matrices on human immune cell lines. Biochim Biophys Acta, 1840, 1892, 2014.

103.

S.M.

, Li

, and Kim

Collagen mimetic peptides: progress towards functional applications. Soft Matter, 7, 7927, 2011.

104.

Wang

A.Y.

, Mo

, Chen

C.S.

, and Yu

S.M.

Facile modification of collagen directed by collagen mimetic peptides. J Am Chem Soc, 127, 4130, 2005.

105.

Wang

A.Y.

, Foss

C.A.

, Leong

, Mo

, Pomper

M.G.

, and Yu

S.M.

Spatio-temporal modification of collagen scaffolds mediated by triple helical propensity. Biomacromolecules, 9, 1755, 2008.

106.

Boutet

, Lieberherr

, Tognolli

, Schneider

, Bansal

, Bridge

A.J.

, et al. UniProtKB/Swiss-Prot, the manually annotated section of the UniProt KnowledgeBase: how to use the entry view. Methods Mol Biol, 1374, 23, 2016.

107.

Millard

C.J.

, Campbell

I.D.

, and Pickford

A.R.

Gelatin binding to the 8F19F1 module pair of human fibronectin requires site-specific N-glycosylation. FEBS Lett, 579, 4529, 2005.

108.

Delannoy

, and Montreuil

Deglycosylation of the chymotryptic collagen-binding fragment of human plasma fibronectin does not modify its affinity to denatured collagen. FEBS Lett, 247, 25, 1989.

109.

Millard

C.J.

Structural and functional characterisation of the collagen binding domain of fibronectin. [PhD thesis]. Oxford (UK): University of Oxford; 2007.

110.

Kawai

, Suzuki

, Tabata

, Ikada

, and Nishimura

Accelerated tissue regeneration through incorporation of basic fibroblast growth factor-impregnated gelatin microspheres into artificial dermis. Biomaterials, 21, 489, 2000.

111.

Graiani

, Emanueli

, Desortes

, Linthout

S.V.

, Pinna

, Figueroa

C.D.

, et al. Nerve growth factor promotes reparative angiogenesis and inhibits endothelial apoptosis in cutaneous wounds of type 1 diabetic mice. Diabetologia, 47, 1047, 2004.

112.

Chan

T.R.

, Stahl

P.J.

, and Yu

S.M.

Matrix-bound VEGF mimetic peptides: design and endothelial-cell activation in collagen scaffolds. Adv Funct Mater, 21, 4252, 2011.

113.

Singh

, Wu

B.M.

, and Dunn

J.C.Y.

The enhancement of VEGF-mediated angiogenesis by polycaprolactone scaffolds with surface cross-linked heparin. Biomaterials, 32, 2059, 2011.

114.

Tabata

, Nagano

, Muniruzzaman

, and Ikada

In vitro sorption and desorption of basic fibroblast growth factor from biodegradable hydrogels. Biomaterials, 19, 1781, 1998.

115.

Stadelmann

W.K.

, Digenis

A.G.

, and Tobin

G.R.

Physiology and healing dynamics of chronic cutaneous wounds. Am J Surg, 176(2, Supplement 1), 26S, 1998.

116.

Steenfos

H.H.

Growth factors and wound healing. Scand J Plast Reconstr Surg Hand Surg, 28, 95, 1994.

117.

Sun

, Zhao

, Li

, Chen

, Hou

, Xiao

, et al. Controlled release of collagen-binding SDF-1α improves cardiac function after myocardial infarction by recruiting endogenous stem cells. Sci Rep, 6, 26683, 2016.

118.

Bennett

N.T.

, and Schultz

G.S.

Growth factors and wound healing: biochemical properties of growth factors and their receptors. Am J Surg, 165, 728, 1993.

119.

Carmeliet

Angiogenesis in health and disease. Nat Med, 9, 653, 2003.

120.

Robson

M.C.

, Mustoe

T.A.

, and Hunt

T.K.

The future of recombinant growth factors in wound healing. Am J Surg, 176(2, Supplement 1), 80S, 1998.

121.

Brown

G.L.

, Curtsinger

, Jurkiewicz

M.J.

, Nahai

, and Schultz

Stimulation of healing of chronic wounds by epidermal growth factor. Plast Reconstr Surg, 88, 189, 1991.

122.

Kempen

D.H.R.

, Creemers

L.B.

, Alblas

, Lu

, Verbout

A.J.

, Yaszemski

M.J.

, et al. Growth factor interactions in bone regeneration. Tissue Eng Part B Rev, 16, 551, 2010.

123.

Zhao

, Chen

, Lin

, Sun

, Zhao

, Zhang

, et al. The bone-derived collagen containing mineralized matrix for the loading of collagen-binding bone morphogenetic protein-2. J Biomed Mater Res A, 88, 725, 2009.

124.

Hankenson

K.D.

, Gagne

, and Shaughnessy

Extracellular signaling molecules to promote fracture healing and bone regeneration. Adv Drug Deliv Rev, 94, 3, 2015.

125.

Ferreira

A.M.

, Gentile

, Chiono

, and Ciardelli

Collagen for bone tissue regeneration. Acta Biomater, 8, 3191, 2012.

126.

El-Jawhari

J.J.

, Sanjurjo-Rodríguez

, Jones

, and Giannoudis

P.V.

Collagen-containing scaffolds enhance attachment and proliferation of non-cultured bone marrow multipotential stromal cells. J Orthop Res, 34, 597, 2016.

127.

Simons

, and Ware

J.A.

Therapeutic angiogenesis in cardiovascular disease. Nat Rev Drug Discov, 2, 863, 2003.

128.

Ponnapakkam

, Katikaneni

, Sakon

, Stratford

, and Gensure

R.C.

Treating osteoporosis by targeting parathyroid hormone to bone. Drug Discov Today, 19, 204, 2014.

129.

Ohkawara

, Ueda

, Shinozaki

, Kitajima

, Ito

, Asaoka

, et al. Hepatocyte growth factor fusion protein having collagen-binding activity (CBD-HGF) accelerates re-endothelialization and intimal hyperplasia in balloon-injured rat carotid artery. J Atheroscler Thromb, 14, 185, 2007.

130.

Matsuno

, Kozawa

, Yoshimi

, Akamatsu

, Hara

, Mori

, et al. Lack of alpha2-antiplasmin promotes pulmonary heart failure via overrelease of VEGF after acute myocardial infarction. Blood, 100, 2487, 2002.

131.

Wang

E.A.

, Rosen

, D'Alessandro

J.S.

, Bauduy

, Cordes

, Harada

, et al. Recombinant human bone morphogenetic protein induces bone formation. Proc Natl Acad Sci U S A, 87, 2220, 1990.

132.

Zhao

, Chen

, Li

, Lin

, Sun

, Zhao

, et al. Vascularization and cellularization of collagen scaffolds incorporated with two different collagen-targeting human basic fibroblast growth factors. J Biomed Mater Res A, 82, 630, 2007.

133.

Uchida

, Matsushita

, Nishi

, Inoue

, Horikawa

, and Takaso

Enhancement of periosteal bone formation by basic fibroblast-derived growth factor containing polycystic kidney disease and collagen-binding domains from Clostridium histolyticum collagenase. J Tissue Eng Regen Med, 2015 [Epub ahead of print]; DOI: 10.1002/term.2019.

134.

Ponnapakkam

, Katikaneni

, Miller

, Ponnapakkam

, Hirofumi

, Miyata

, et al. Monthly administration of a novel PTH-collagen binding domain fusion protein is anabolic in mice. Calcif Tissue Int, 88, 511, 2011.

135.

D'Souza

S.E.

, Ginsberg

M.H.

, and Plow

E.F.

Arginyl-glycyl-aspartic acid (RGD): a cell adhesion motif. Trends Biochem Sci, 16, 246, 1991.

136.

Ruoslahti

, and Pierschbacher

M.D.

New perspectives in cell adhesion: RGD and integrins. Science, 238, 491, 1987.

137.

Sephel

G.C.

, Burrous

B.A.

, and Kleinman

H.K.

Laminin neural activity and binding proteins. Dev Neurosci, 11, 313, 1989.

138.

Tan

, Song

, Jin

, Zhao

, Chen

, Qing

, et al. vwF A3-GPI modification of EPCs accelerates reendothelialization of injured vessels via collagen targeting in mice. J Drug Target, 1, 2016.

139.

Nakaji-Hirabayashi

, Kato

, and Iwata

In vivo study on the survival of neural stem cells transplanted into the rat brain with a collagen hydrogel that incorporates laminin-derived polypeptides. Bioconjug Chem, 24, 1798, 2013.

140.

Van Vlierberghe

, Vanderleyden

, Dubruel

, De Vos

, and Schacht

Affinity study of novel gelatin cell carriers for fibronectin. Macromol Biosci, 9, 1105, 2009.

141.

Parsons

, Gilbert

S.J.

, Vaughan-Thomas

, Sorrell

D.A.

, Notman

, Bishop

, et al. Type IX collagen interacts with fibronectin providing an important molecular bridge in articular cartilage. J Biol Chem, 286, 34986, 2011.

142.

Pickford

A.R.

, Smith

S.P.

, Staunton

, Boyd

, and Campbell

I.D.

The hairpin structure of the 6F11F22F2 fragment from human fibronectin enhances gelatin binding. EMBO J, 20, 1519, 2001.

143.

Kitajima

, Terai

, and Ito

A fusion protein of hepatocyte growth factor for immobilization to collagen. Biomaterials, 28, 1989, 2007.

144.

Vanhoorelbeke

, van der Plas

R.M.

, Vandecasteele

, Vauterin

, Huizinga

E.G.

, Sixma

J.J.

, et al. Sequence alignment between vWF and peptides inhibiting the vWF-collagen interaction does not result in the identification of a collagen-binding site in vWF. Thromb Haemost, 84, 621, 2000.

145.

Kato

, Sato

, and Iwata

Ultrastructural study on the specific binding of genetically engineered epidermal growth factor to type I collagen fibrils. Bioconjug Chem, 18, 2137, 2007.

146.

Andrades

J.A.

, Wu

L.T.

, Hall

F.L.

, Nimni

M.E.

, and Becerra

Engineering, expression, and renaturation of a collagen-targeted human bFGF fusion protein. Growth Factors Chur Switz, 18, 261, 2001.

147.

Brewster

L.P.

, Washington

, Brey

E.M.

, Gassman

, Subramanian

, Calceterra

, et al. Construction and characterization of a thrombin-resistant designer FGF-based collagen binding domain angiogen. Biomaterials, 29, 327, 2008.

148.

, Rivas

J.M.

, Brown

E.L.

, Liang

, and Höök

Virulence potential of the staphylococcal adhesin CNA in experimental arthritis is determined by its affinity for collagen. J Infect Dis, 189, 2323, 2004.

149.

Rich

R.L.

, Deivanayagam

C.C.S.

, Owens

R.T.

, Carson

, Höök

, Moore

, et al. Trench-shaped binding sites promote multiple classes of interactions between collagen and the adherence receptors, α1β1 integrin and Staphylococcus aureus Cna MSCRAMM. J Biol Chem, 274, 24906, 1999.

150.

Yang

, Zhao

, Chen

, Han

, Sun

, Xiao

, et al. Collagen-binding human epidermal growth factor promotes cellularization of collagen scaffolds. Tissue Eng Part A, 15, 3589, 2009.

151.

Zhang

, Ding

, Zhao

, Sun

, Chen

, Lin

, et al. Collagen-targeting vascular endothelial growth factor improves cardiac performance after myocardial infarction. Circulation, 119, 1776, 2009.

152.

Chen

, Lin

, Wang

, Zhao

, Wang

, Zhao

, et al. Homogeneous osteogenesis and bone regeneration by demineralized bone matrix loading with collagen-targeting bone morphogenetic protein-2. Biomaterials, 28, 1027, 2007.

153.

Sun

, Lin

, Chen

, Zhao

, Xiao

, et al. Collagen scaffolds loaded with collagen-binding NGF-beta accelerate ulcer healing. J Biomed Mater Res A, 92, 887, 2010.

154.

Fan

, Xiao

, Zhang

, Chen

, Tang

, Hou

, et al. Linear ordered collagen scaffolds loaded with collagen-binding neurotrophin-3 promote axonal regeneration and partial functional recovery after complete spinal cord transection. J Neurotrauma, 27, 1671, 2010.

155.

X.-C.

, Huang

, Wang

, Li

C.-Q.

, and Zhou

Collagen-targeting parathyroid hormone-related peptide promotes collagen binding and in vitro chondrogenesis in bone marrow-derived MSCs. Int J Mol Med, 31, 430, 2013.

156.

Liang

, Han

, Jin

, Xiao

, Huang

, Ni

, et al. The promotion of neurological recovery in the rat spinal cord crushed injury model by collagen-binding BDNF. Biomaterials, 31, 8634, 2010.

157.

, Han

, Zhao

, Ding

, Wei

, Li

, et al. Functionalized collagen scaffold implantation and cAMP administration collectively facilitate spinal cord regeneration. Acta Biomater, 30, 233, 2016.

158.

Liang

, Li

, Chen

, Wang

, Zhao

, Zhuang

, et al. A collagen-binding EGFR single-chain Fv antibody fragment for the targeted cancer therapy. J Control Release, 209, 101, 2015.

159.

Liang

, Li

, Wang

, Chen

, Zhao

, Sun

, et al. A collagen-binding EGFR antibody fragment targeting tumors with a collagen-rich extracellular matrix. Sci Rep, 6, 18205, 2016.

160.

Nishi

, Matsushita

, Yuube

, Miyanaka

, Okabe

, and Wada

Collagen-binding growth factors: production and characterization of functional fusion proteins having a collagen-binding domain. Proc Natl Acad Sci U S A, 95, 7018, 1998.

161.

Ishikawa

, Terai

, and Kitajima

Production of a biologically active epidermal growth factor fusion protein with high collagen affinity. J Biochem (Tokyo), 129, 627, 2001.

162.

Ishikawa

, Terai

, Yamamoto

, Harada

, and Kitajima

Delivery of a growth factor fusion protein having collagen-binding activity to wound tissues. Artif Organs, 27, 147, 2003.

163.

Kim

D.-G.

, Kim

E.-Y.

, Kim

Y.-R.

, and Kong

I.-S.

Construction of chimeric human epidermal growth factor containing short collagen-binding domain moieties for use as a wound tissue healing agent. J Microbiol Biotechnol, 25, 119, 2015.

164.

Chen

, Shi

, Yi

, Chen

, Zhang

, Fang

, et al. Bladder regeneration by collagen scaffolds with collagen binding human basic fibroblast growth factor. J Urol, 183, 2432, 2010.

165.

Shi

, Chen

, Zhao

, Chen

, Xiao

, Wei

, et al. Regeneration of full-thickness abdominal wall defects in rats using collagen scaffolds loaded with collagen-binding basic fibroblast growth factor. Biomaterials, 32, 753, 2011.

166.

, Sun

, Lin

, Hou

, Wang

, Zhou

, et al. Regeneration of uterine horns in rats by collagen scaffolds loaded with collagen-binding human basic fibroblast growth factor. Biomaterials, 32, 8172, 2011.

167.

, Tao

, Chen

, Ren

, Hou

, Zhou

, et al. Extrahepatic bile duct regeneration in pigs using collagen scaffolds loaded with human collagen-binding bFGF. Biomaterials, 33, 4298, 2012.

168.

Gao

, Liu

, Gao

, Wang

, Zhao

, Chen

, et al. A myocardial patch made of collagen membranes loaded with collagen-binding human vascular endothelial growth factor accelerates healing of the injured rabbit heart. Tissue Eng Part A, 17, 2739, 2011.

169.

Yan

, Chen

, Lin

, Li

, Xiao

, Hou

, et al. Acceleration of diabetic wound healing by collagen-binding vascular endothelial growth factor in diabetic rat model. Diabetes Res Clin Pract, 90, 66, 2010.

170.

Tan

, Chen

, Yan

, Lin

, Xiao

, Hou

, et al. Promotion of diabetic wound healing by collagen scaffold with collagen-binding vascular endothelial growth factor in a diabetic rat model. J Tissue Eng Regen Med, 8, 195, 2014.

171.

Jia

, Tang

, Wu

, Hou

, Chen

, et al. Urethral tissue regeneration using collagen scaffold modified with collagen binding VEGF in a beagle model. Biomaterials, 69, 45, 2015.

172.

Lin

, Li

, Song

, Wang

, Meng

, Yang

, et al. The effect of collagen-binding vascular endothelial growth factor on the remodeling of scarred rat uterus following full-thickness injury. Biomaterials, 33, 1801, 2012.

173.

Akimoto

, Takeda

, Matsushita

, Inoue

, Sakamoto

, Hattori

, et al. Effects of CB-VEGF-A injection in rat flap models for improved survival. Plast Reconstr Surg, 131, 717, 2013.

174.

Lineaweaver

W.C.

, and Zhang

Letters. Effects of CB-VEGF-A injection in rat flap models for improved survival. Plast Reconstr Surg, 133, 423e, 2014.

175.

Fang

, Lineaweaver

W.C.

, Chen

M.B.

, Kisner

, and Zhang

Effects of vascular endothelial growth factor on survival of surgical flaps: a review of experimental studies. J Reconstr Microsurg, 30, 1, 2014.

176.

Okiyama

, Kitajima

, Ito

, Yokozeki

, Miyasaka

, and Kohsaka

Addition of the collagen binding domain of fibronectin potentiates the biochemical availability of hepatocyte growth factor for cutaneous wound healing. J Dermatol Sci, 61, 215, 2011.

177.

Ota

, Gilbert

T.W.

, Schwartzman

, McTiernan

C.F.

, Kitajima

, Ito

, et al. A fusion protein of hepatocyte growth factor enhances reconstruction of myocardium in a cardiac patch derived from porcine urinary bladder matrix. J Thorac Cardiovasc Surg, 136, 1309, 2008.

178.

, Liang

, Mou

, Li

, Sun

, Zhuang

, et al. Regulation of human mesenchymal stem cells differentiation into chondrocytes in extracellular matrix-based hydrogel scaffolds. Colloids Surf B Biointerfaces, 114, 316, 2014.

179.

Saito

, Uchida

, Ueno

, Matsushita

, Inoue

, Nishi

, et al. Acceleration of bone formation during fracture healing by injectable collagen powder and human basic fibroblast growth factor containing a collagen-binding domain from Clostridium histolyticum collagenase. J Biomed Mater Res A, 102, 3049, 2014.

180.

Ueno

, Uchida

, Saito

, Matsushita

, Yogoro

, Nishi

, et al. Acceleration of bone union after structural bone grafts with a collagen-binding basic fibroblast growth factor anchored-collagen sheet for critical-size bone defects. Biomed Mater, 9, 35014, 2014.

181.

Uchida

, Matsushita

, Naruse

, Mima

, Nishi

, Hattori

, et al. Acceleration of periosteal bone formation by human basic fibroblast growth factor containing a collagen-binding domain from Clostridium histolyticum collagenase. J Biomed Mater Res A, 102, 1737, 2014.

182.

Saito

, Uchida

, Matsushita

, Inoue

, Sekiguchi

, Aikawa

, et al. Acceleration of callus formation during fracture healing using basic fibroblast growth factor-kidney disease domain-collagen-binding domain fusion protein combined with allogenic demineralized bone powder. J Orthop Surg, 10, 59, 2015.

183.

Sekiguchi

, Uchida

, Inoue

, Matsushita

, Saito

, Aikawa

, et al. Acceleration of bone formation during fracture healing by poly(Pro-Hyp-Gly)10 and basic fibroblast growth factor containing polycystic kidney disease and collagen-binding domains from Clostridium histolyticum collagenase. J Biomed Mater Res A, 104, 1372, 2016.

184.

Andrades

J.A.

, Han

, Becerra

, Sorgente

, Hall

F.L.

, and Nimni

M.E.

A recombinant human TGF-beta1 fusion protein with collagen-binding domain promotes migration, growth, and differentiation of bone marrow mesenchymal cells. Exp Cell Res, 250, 485, 1999.

185.

Becerra

, Guerado

, Claros

, Alonso

, Bertrand

M.L.

, González

, et al. Autologous human-derived bone marrow cells exposed to a novel TGF-β1 fusion protein for the treatment of critically sized tibial defect. Regen Med, 1, 267, 2006.

186.

Claros

, Rico-Llanos

G.A.

, Becerra

, and Andrades

J.A.

A novel human TGF-β1 fusion protein in combination with rhBMP-2 increases chondro-osteogenic differentiation of bone marrow mesenchymal stem cells. Int J Mol Sci, 15, 11255, 2014.

187.

Chen

, Lin

, Zhao

, Wang

, Zhao

, Liu

, et al. Activation of demineralized bone matrix by genetically engineered human bone morphogenetic protein-2 with a collagen binding domain derived from von Willebrand factor propolypeptide. J Biomed Mater Res A, 80, 428, 2007.

188.

Lai

C.-H.

, Zhou

, Wang

Z.-L.

, Lu

H.-B.

, and Gao

Use of a collagen membrane loaded with recombinant human bone morphogenetic protein-2 with collagen-binding domain for vertical guided bone regeneration. J Periodontol, 84, 950, 2013.

189.

Han

, Perelman

, Tang

, Hall

, Shors

E.C.

, and Nimni

M.E.

Collagen-targeted BMP3 fusion proteins arrayed on collagen matrices or porous ceramics impregnated with type I collagen enhance osteogenesis in a rat cranial defect model. J Orthop Res, 20, 747, 2002.

190.

Yang

, Shi

, Tu

, Wang

, Liu

, Fan

, et al. Bone morphogenetic proteins: relationship between molecular structure and their osteogenic activity. Food Sci Hum Wellness, 3, 127, 2014.

191.

, Kawazoe

, Kitajima

, Myoken

, Tomita

, Umezawa

, et al. Spatial immobilization of bone morphogenetic protein-4 in a collagen-PLGA hybrid scaffold for enhanced osteoinductivity. Biomaterials, 33, 6140, 2012.

192.

Shiozaki

, Kitajima

, Mazaki

, Yoshida

, Tanaka

, Umezawa

, et al. Enhanced in vivo osteogenesis by nanocarrier-fused bone morphogenetic protein-4. Int J Nanomed, 8, 1349, 2013.

193.

Mazaki

, Shiozaki

, Yamane

, Yoshida

, Nakamura

, Yoshida

, et al. A novel, visible light-induced, rapidly cross-linkable gelatin scaffold for osteochondral tissue engineering. Sci Rep, 4, 2014. Available at: www.nature.com/articles/srep04457 (last accessed April 14, 2016).

194.

Ponnapakkam

, Katikaneni

, Nichols

, Tobin

, Sakon

, Matsushita

, et al. Prevention of chemotherapy-induced osteoporosis by cyclophosphamide with a long-acting form of parathyroid hormone. J Endocrinol Invest, 34, e392, 2011.

195.

Shao

, Zhang

, Pi

, Wang

, Jia

, Zhu

, et al. Polycaprolactone electrospun mesh conjugated with an MSC affinity peptide for MSC homing in vivo. Biomaterials, 33, 3375, 2012.

196.

Visser

, Arrabal

P.M.

, Santos-Ruiz

, Fernandez-Barranco

, Becerra

, and Cifuentes

A collagen-targeted biomimetic RGD peptide to promote osteogenesis. Tissue Eng Part A, 20, 34, 2014.

197.

Won

J.-E.

, Yun

Y.-R.

, Jang

J.-H.

, Yang

S.-H.

, Kim

J.-H.

, Chrzanowski

, et al. Multifunctional and stable bone mimic proteinaceous matrix for bone tissue engineering. Biomaterials, 56, 46, 2015.

198.

Hiraoka

, Kato

, Nakaji-Hirabayashi

, and Iwata

Enhanced survival of neural cells embedded in hydrogels composed of collagen and laminin-derived cell adhesive peptide. Bioconjug Chem, 20, 976, 2009.

199.

Kim

J.-M.

, Park

W.H.

, and Min

B.-M.

The PPFLMLLKGSTR motif in globular domain 3 of the human laminin-5 alpha3 chain is crucial for integrin alpha3beta1 binding and cell adhesion. Exp Cell Res, 304, 317, 2005.

200.

Miyazaki

, Kato

, Teramura

, and Iwata

A collagen-binding mimetic of neural cell adhesion molecule. Bioconjug Chem, 19, 1119, 2008.