Patent Application
20240033402
The instant application is a 35 U.S.C. § 371 filing of International Patent Application No. PCT/EP2020/074370, filed Sep. 1, 2020, the entire content of which is incorporated herein by reference for all purposes.
The instant application contains a Sequence Listing which has been submitted electronically in ASCII format and is hereby incorporated by reference in its entirety. Said ASCII copy, created on February 27, 2023, is named 740336_AWA9-003US_ST25.txt and is 12,118 bytes in size.
The present invention is concerned with bone regeneration, and more specifically with a method for producing macroporous scaffolds which are useful in bone regeneration. The macroporous scaffolds comprise silk and calcium phosphate.
Calcium phosphates (CaPs) are widely used for bone regeneration due to their resemblance to bone mineral phase. CaP is a family of mineral salts containing calcium and orthophosphate ions. They can be divided into two major categories depending on the processing temperature of the synthesis routes: high temperature CaP and low temperature CaP. High temperature CaP are produced by sintering at temperatures above 800° C. They are similar in composition with the inorganic phase of calcified tissues and have been shown to be osteoconductive and bioresorbable. Alpha- and Beta-tricalcium phosphate (α-TCP and β-TCP) are two types of high temperature CaPs. Although the mechanical properties of CaPs can be improved upon sintering, nanostructures are lost, resulting from coalescence of the nanocrystals at high temperatures. Low temperature CaPs include amorphous calcium phosphate (ACP), octacalcium phosphate (OCP), hydroxyapatite (HA) or calcium deficient hydroxyapatite (CDHA), which are naturally occurring in the bones and highly relevant for bone graft substitutes.
A simple, straight-forward method for producing macroporous scaffolds may be foaming by mechanical stirring with addition of a small amount of surface-active molecules to the liquid phase, followed by mixing with the calcium phosphate powder. The liquid foam may be stabilized by incorporation of the powder, which reduces the gravitational drainage and air diffusion between the bubbles. The porous structures may be maintained through solidification of the cement. The foaming agents should meet the essential requirement of being water soluble and biocompatible in order to be used as the foaming template for macroporous calcium phosphate scaffolds (CPS). A low molecular weight, non-ionic surfactant, Polysorbate or namely Tween 80, is biocompatible and has been approved by the FDA as drug additives for parenteral administration. Macroporous CPSs have been successfully produced in previous studies using Tween 80 as the foaming agent, with interconnected pores and tunable pore size distribution by varying the surfactant concentration (Montufar, E. B., et al. Acta Biomaterialia 6.3 (2010), Diez Escudero, Anna, et al. Acta Biomaterialia 60 (2017), Barba, Albert, et al. ACS Applied Materials & Interfaces (2017):acsami.7b14175, Pastorino, David, C. Canal , and M. P. Ginebra Acta Biomaterialia 12 (2015)). The CPSs obtained were proved to be injectable, maintaining the macroporous structure after injection. High resorption rate was found to be promoted by incorporation of macroporosity during both in vitro and in vivo studies. Accelerated differentiation of rat mesenchymal stem cells (rMSCs) into osteoblasts was shown in vitro, resulting from the spherical macropores that offered the niches for osteogenic differentiation of the MSCs. Such macroporous CPSs foamed with Tween 80 have been studied as drug delivery systems, with tunable drug release profile through tailored macroporosity (Pastorino, David, C. Canal, and M. P. Ginebra, Acta Biomaterialia 12 (2015)). Another type of foaming agents which have been widely studied are amphiphilic proteins such as albumin and gelatin. These biomolecules may not only offer desirable foaming capacity, but also biological recognition in vivo, thereby promoting cell adhesion and bone ingrowth.
Polymers have also shown promising performance as three-dimensional matrices for bone regeneration purposes due to their biocompatibility, biodegradability and tunable properties. Polymers can be categorized as synthetic polymers such as polydioxanone (PDS), polycaprolactone (PCL), polyglycolic acid (PGA), polylactic acid (PLA), and poly (lactic-co-glycolic acid) (PLGA), and naturally derived polymers including chitosan, collagen and silk fibroin.
In spite of its highly desirable properties, a biopolymer alone is insufficient for repair of larger bone defects. Polymers lack compressive strength in regard of bone repair, and are more commonly used in composites with ceramics. Enhanced mechanical behavior and bone mimetic topography can be achieved by incorporation of mineral components, for instance bioceramics. Considerable attempts have been made to develop polymer-ceramic composites, taking advantage of the unique properties of the two materials to improve both the mechanical and biological behavior of scaffolds for bone tissue engineering. Polymer-ceramic composites are considered to be a promising candidate for bone graft substitute and regeneration applications. However, they lack interconnective porosity on the macro scale which plays a vital role in facilitating cell ingrowth, material associated-osteoinduction and replacement of the scaffold by newly-formed bone.
Silk fibroins are produced by numerous species of spiders and by worms from various insects such as bees, butterflies, and moths. Silks produced by silkworms (typically Bombyx mon) and orb-weaving spiders have desirable mechanical properties, environmental stability, biocompatibility, and tunable degradation.
WO 2014/066884 and McNamara et al, Biomaterials 35(25): 6941-6953 (2014) disclose ceramic materials comprising calcium phosphate material and silk, and methods for preparing the ceramic materials. The method comprises preparing a composition comprising silk and calcium phosphate (CaP), forming a green body from the composition and sintering at 1300° C. or above of the green body to form the porous ceramic material.
WO 2016/100721 provides a silk ceramic material having enzymatically cross-linked amino acid side chains to generate injectable and flexible foam ceramics. Provided are compositions and methods of producing soft, flexible ceramic foam with silk polymeric crosslinking to serve as binders. The silk proteins are denatured by harsh solvents and require enzymatic formation of intermolecular covalent bonds to produce macrostructures.
Although various attempts have been made to introduce macroporosity to polymer-ceramic composites such as leaching, foaming, emulsion, replication, and rapid prototyping, there is still a great a need to be able to produce high strength, complex bio-active scaffolds, which are biocompatible, biodegradable, osteoconductive and even osteoinductive.
It is an object of the present invention to provide a simplified method for producing a macroporous scaffold, suitable for use in bone tissue generation.
It is an object of the present invention to provide a method for producing a macroporous scaffold which does not require the use of non-biocompatible components.
For these and other objects that will be evident from the following description, the present invention provides according to a first aspect a method for producing a macroporous scaffold according to the appended claims.
List of Appended Sequences
Calcium phosphate (CaP) is a family of mineral salts containing calcium and orthophosphate ions. They can be divided into two major categories depending on the processing temperature of the synthesis routes: high temperature CaP and low temperature CaP. High temperature CaP are produced by sintering at temperatures 800° C. They are similar in composition with the inorganic phase of calcified tissues and have shown to be osteoconductive and bioresorbable. Alpha- and Beta-tricalcium phosphate (α-TCP and β-TCP) are two types of high temperature CaPs. Although the mechanical properties of CaPs can be improved upon sintering, nanostructures are lost, resulting from coalescence of the nanocrystals at high temperatures. Low temperature CaPs include amorphous calcium phosphate (ACP), octacalcium phosphate (OCP), hydroxyapatite (HA) or calcium deficient hydroxyapatite (CDHA), which are naturally occurring in the bones and highly relevant for bone graft substitutes
Calcium phosphate scaffolds (CPSs) are an unique type of low termperature CaP which have attracted great attention as bone graft substitute. CPSs may be obtained through mixing a CaP-containing powder with an aqueous phase, upon which a plastic paste is formed that can be injected and easily molded into various shapes.
Physicochemical, mechanical and biological behavior of the CPSs may be tailored by varying several factors concerning the manufacture process, such as additives, doping, liquid-to-powder (L/P) ratio and particle size of the starting materials.
Here, we design a facilitated method for producing macroporous scaffolds which are useful in bone regeneration. This method combines non-denatured silk protein with a-tricalcium phosphate (α-TCP) with a hydroxyapatite (HA) seed to produce macroporous scaffolds which are useful in bone regeneration.
The term “macroporous” refers to a porosity above 10 μm as it is considered to be generated from the foaming process, in contrast to the intrinsic microporosity (below 10 μm) of CPSs.
The macroporous scaffold is biocompatible and suitable for use in bone tissue generation. The resulting macroporosity is suitable for cell colonization and vascularization along gradual replacement of the implant by newly-formed bone. The production method is simple to perform and does not require the use of non-biocompatible components. Improved macroporosity is associated with enhanced vascularization, enhanced permeability, enhanced nutrient diffusion, and enhanced osteoinduction.
The term “biocompatible”, as used herein, is defined as the ability of a material to perform with an acceptable host response in the body. It refers to components that do not cause significant harm to living tissue when placed in contact with such tissue, e.g., in vivo. Components are biocompatible if they are not toxic to cells. In additon, materials are biocompatible if their addition to cells in vitro results in less and/or no cell death, and/or their administration in vivo does not induce significant inflammation or other such adverse effects.
The present invention provides according to a first aspect a method for producing a macroporous scaffold, comprising the steps:
In the first step (a), a liquid phase which is an aqueous solution of a non-denatured silk protein is subjected to mechanical foaming.
For avoidance of doubt, foaming is the continuous formation of bubbles which have sufficiently high surface tension to remain as bubbles beyond the disengaging surface. Foaming may be used in various applications, such as industrial cleaning and preparation of building materials, and can be achieved through mechanical work (mechanical foaming) such as whipping, blending and stirring, but also through surface active components (surfactants) that reduce surface tension.
The foamability of the silk may arise from the alternating hydrophobic and hydrophilic sections in the peptide sequence, which allows it to adsorb to the air/water interface around the bubbles and stabilize them by reducing drainage and lamellae rupture. Silk concentration and foaming time may be two determining parameters for its foamability, since a foam can only be created when sufficient proteins are adsorbed to the interface within the foaming time scale. Adsorption may be enhanced by subsequent partial unfolding (surface denaturation) as a consequence of increased contact points at the interface. Increasing protein concentration may lead to higher amount of adsorption and better foaming properties.
The non-denatured silk protein dissolved in the aqueous solution is capable of assembling into a water-insoluble macrostructure at a water/air interface.
Importantly, the silk protein is present in soluble and non-denatured form in the liquid medium. This implies that silk protein is present in its native state, wherein it is properly folded, operative and functional. In contrast to the silk form used in the method disclosed herein, the presence of harsh solvents such as LiBr in the liquid medium denatures and solubilizes the silk protein.
The silk assembly process refers to when the silk proteins predominantly form bonds with the surrounding water molecules, irreversible formation of ordered polymers (foam) with predominantly intra- and intermolecular bonds between the silk proteins.
The assembly of the silk protein into foam is promoted by the presence of shearing forces and/or an interface between two different phases e.g. between a solid and a liquid phase, typically between air and a liquid phase or at a hydrophobic/hydrophilic interface, e.g. a mineral oil-water interface. The presence of the resulting interface may stimulate polymerization at the interface or in the region surrounding the interface, which region extends into the liquid medium, such that said polymerizing initiates at said interface or in said interface region.
Recombinantly produced silk is useful as matrices for culture of mammalian cells. The inclusion of cell adhesion motifs derived from the extracellular matrix (ECM) into such materials increases cell attachment and proliferation by interaction with integrins on the cell surface. The integrins do not just confer the physical connection between cells and the surrounding, but also mediate signals controlling for example cell growth, polarity, proliferation and survival. Moreover, the integrins are essential for cell migration by acting as the cells' “feet”.
The silk protein is typically present in a concentration of 0.5 -5 mg/ml, preferably 2 mg/ml or higher, in the liquid phase. Increasing protein concentration should lead to higher amount of adsorption and better foaming properties, yet a threshold concentration of 2 mg/ml was observed (shown in
The silk protein is typically foamed for short time periods, shorter than 30 s, preferably shorter than 10 s. The porosities shown in the results were all obtained using an optimal foaming time of 5 s. Prolonged foaming time led to increased small bubbles and less stable foams, which caused a decrease in macroporosity of the CPSs.
In a certain embodiment the method for producing a macroporous scaffold as disclosed herein the silk protein is a recombinant silk protein.
In a specific embodiment the method for producing a macroporous scaffold as disclosed herein the silk protein is a spider silk protein.
The terms “spidroins” and “spider silk proteins” are used interchangeably throughout the description and encompass all known spider silk proteins, including major ampullate spider silk proteins which typically are abbreviated “MaSp”, or “ADF” in the case of Araneus diadematus. These major ampullate spider silk proteins are generally of two types, 1 and 2. These terms furthermore include non-natural proteins with a high degree of identity and/or similarity to the known spider silk proteins.
In some preferred embodiments of these and other aspects of the invention, the spider silk protein is comprising, or consisting of, the protein moieties REP and CT, wherein REP is a repetitive fragment of from 70 to 300 amino acid residues, selected from the group consisting of L(AG)nL, L(AG)nAL, L(GA)nL, and L(GA)nGL, wherein n is an integer from 2 to 10; each individual A segment is an amino acid sequence of from 8 to 18 amino acid residues, wherein from 0 to 3 of the amino acid residues are not Ala, and the remaining amino acid residues are Ala; each individual G segment is an amino acid sequence of from 12 to 30 amino acid residues, wherein at least 40% of the amino acid residues are Gly; and each individual L segment is a linker amino acid sequence of from 0 to 30 amino acid residues; and CT is a fragment of from 70 to 120 amino acid residues, having at least 70% identity to either of SEQ ID NO: 4-5.
Optionally, a functional moiety is arranged either terminally in the spider silk protein, or between the moieties, or within any of the moieties, preferably terminally in the spider silk protein.
The spider silk protein typically comprises from 140 to 2000 amino acid residues, such as from 140 to 1000 amino acid residues, such as from 140 to 600 amino acid residues, preferably from 140 to 500 amino acid residues, such as from 140 to 400 amino acid residues. The small size is advantageous because longer proteins containing spider silk protein fragments may form amorphous aggregates, which require use of harsh solvents for solubilisation and polymerisation.
The spider silk protein may contain one or more linker peptides, or L segments. The linker peptide(s) may be arranged between any moieties of the spider silk protein, e.g. between the REP and CT moieties, at either terminal end of the spider silk protein or between the spidroin fragment and the cell-binding motif. The linker(s) may provide a spacer between the functional units of the spider silk protein, but may also constitute a handle for identification and purification of the spider silk protein, e.g. a His and/or a Trx tag. If the spider silk protein contains two or more linker peptides for identification and purification of the spider silk protein, it is preferred that they are separated by a spacer sequence, e.g. His6-spacer-His6-. The linker may also constitute a signal peptide, such as a signal recognition particle, which directs the spider silk protein to the membrane and/or causes secretion of the spider silk protein from the host cell into the surrounding medium. The spider silk protein may also include a cleavage site in its amino acid sequence, which allows for cleavage and removal of the linker(s) and/or other relevant moieties. Various cleavage sites are known to the person skilled in the art, e.g. cleavage sites for chemical agents, such as CNBr after Met residues and hydroxylamine between Asn-Gly residues, cleavage sites for proteases, such as thrombin or protease 3C, and self-splicing sequences, such as intein self-splicing sequences.
The protein moiety REP is fragment with a repetitive character, alternating between alanine-rich stretches and glycine-rich stretches. The REP fragment generally contains more than 70, such as more than 140, and less than 300, preferably less than 240, such as less than 200, amino acid residues, and can itself be divided into several L (linker) segments, A (alanine-rich) segments and G (glycine-rich) segments, as will be explained in more detail below. Typically, said linker segments, which are optional, are located at the REP fragment terminals, while the remaining segments are in turn alanine-rich and glycine-rich. Thus, the REP fragment can generally have either of the following structures, wherein n is an integer:
L(AG)nL, such as LA1G1A2G2A3G3A4G4A5G5L;
L(AG)nAL, such as LA1G1A2G2A3G3A4G4A5G5A6L;
L(GA)nL, such as LG1A1G2A2G3A3G4A4G5A5L; or
L(GA)nGL, such as LG1A1G2A2G3A3G4A4G5A5G6L.
It follows that it is not critical whether an alanine-rich or a glycine-rich segment is adjacent to the N-terminal or C-terminal linker segments. It is preferred that n is an integer from 2 to 10, preferably from 2 to 8, also preferably from 4 to 8, more preferred from 4 to 6, i.e. n=4, n=5 or n=6.
In some embodiments, the alanine content of the REP fragment is above 20%, preferably above 25%, more preferably above 30%, and below 50%, preferably below 40%, more preferably below 35%. It is contemplated that a higher alanine content provides a stiffer and/or stronger and/or less extendible fiber.
In certain embodiments, the REP fragment is void of proline residues, i.e. there are no Pro residues in the REP fragment.
Turning now to the segments that constitute the REP fragment, it is emphasized that each segment is individual, i.e. any two A segments, any two G segments or any two L segments of a specific REP fragment may be identical or may not be identical. Thus, it is not a general feature of the spidroin that each type of segment is identical within a specific REP fragment. Rather, the following disclosure provides the skilled person with guidelines how to design individual segments and gather them into a REP fragment, which is a part of a functional spider silk protein useful in a cell scaffold material.
Each individual A segment is an amino acid sequence having from 8 to 18 amino acid residues. It is preferred that each individual A segment contains from 13 to 15 amino acid residues. It is also possible that a majority, or more than two, of the A segments contain from 13 to 15 amino acid residues, and that a minority, such as one or two, of the A segments contain from 8 to 18 amino acid residues, such as 8-12 or 16-18 amino acid residues. A vast majority of these amino acid residues are alanine residues. More specifically, from 0 to 3 of the amino acid residues are not alanine residues, and the remaining amino acid residues are alanine residues. Thus, all amino acid residues in each individual A segment are alanine residues, with no exception or with the exception of one, two or three amino acid residues, which can be any amino acid. It is preferred that the alanine-replacing amino acid(s) is (are) natural amino acids, preferably individually selected from the group of serine, glutamic acid, cysteine and glycine, more preferably serine. Of course, it is possible that one or more of the A segments are all-alanine segments, while the remaining A segments contain 1-3 non-alanine residues, such as serine, glutamic acid, cysteine or glycine.
In an embodiment, each A segment contains 13-15 amino acid residues, including 10-15 alanine residues and 0-3 non-alanine residues as described above. In a more preferred embodiment, each A segment contains 13-15 amino acid residues, including 12-15 alanine residues and 0-1 non-alanine residues as described above. Without wishing to be bound by any particular theory, it is envisaged that A segments according to the invention form helical structures or beta sheets.
Furthermore, it has been concluded from experimental data that each individual G segment is an amino acid sequence of from 12 to 30 amino acid residues. It is preferred that each individual G segment consists of from 14 to 23 amino acid residues. At least 40% of the amino acid residues of each G segment are glycine residues. Typically, the glycine content of each individual G segment is in the range of 40-60%.
In certain embodiments, the first two amino acid residues of each G segment are not -Gln-Gln-.
Each individual L segment represents an optional linker amino acid sequence, which may contain from 0 to 30 amino acid residues, such as from 0 to 20 amino acid residues. While this segment is optional and not critical for the function of the spider silk protein, its presence still allows for fully functional spider silk proteins and polymers thereof which form fibers, films, foams and other structures. In particular, the amino acid sequence of a linker segment may resemble any of the described A or G segments, but usually not sufficiently to meet their criteria as defined herein.
In one embodiment of the REP fragment, one of the L segments contains 0 amino acids, i.e. one of the L segments is void. In another embodiment of the REP fragment, both L segments contain 0 amino acids, i.e. both L segments are void. Thus, these embodiments of the REP fragments according to the invention may be schematically represented as follows: (AG)nL, (AG)nAL, (GA)nL, (GA)nGL; L(AG)n, L(AG)nA, L(GA)n, L(GA)nG; and (AG)n, (AG)nA, (GA)n, (GA)nG. Any of these REP fragments are suitable for use with any CT fragment as defined below.
The CT fragment of the spidroin in the cell scaffold material has a high degree of similarity to the C-terminal amino acid sequence of spider silk proteins. As shown in WO 2007/078239, this amino acid sequence is well conserved among various species and spider silk proteins, including MaSp1, MaSp2 and MiSp (minor ampullate spidroin).
It is not critical which specific CT fragment is present in the spider silk protein in the cell scaffold material. A representative CT fragment is the amino acid residues 166-263 of SEQ ID NO: 1 derived from Euprosthenops australis, i.e. SEQ ID NO. 4. Another representative CT fragment is the MiSp sequence SEQ ID NO: 5. Thus, in one embodiment, the CT fragment has at least 70%, such as at least 80%, such as at least 85%, preferably at least 90%, such as at least 95%, identity to SEQ ID NO: 4 or SEQ ID NO: 5. The CT fragment may be identical to SEQ ID NO: 4 or SEQ ID NO: 5
The CT fragment typically consists of from 70 to 120 amino acid residues. It is preferred that the CT fragment contains at least 70, or more than 80, preferably more than 90, amino acid residues. It is also preferred that the CT fragment contains at most 120, or less than 110 amino acid residues. A typical CT fragment contains approximately 100 amino acid residues.
The term “% identity”, as used herein, is calculated as follows. The query sequence is aligned to the target sequence using the CLUSTAL W algorithm (Thompson et al, Nucleic Acids Research, 22:4673-4680 (1994)). A comparison is made over the window corresponding to the shortest of the aligned sequences. The amino acid residues at each position are compared, and the percentage of positions in the query sequence that have identical correspondences in the target sequence is reported as % identity.
The term “% similarity”, as used herein, is calculated as described above for “% identity”, with the exception that the hydrophobic residues Ala, Val, Phe, Pro, Leu, Ile, Trp, Met and Cys are similar; the basic residues Lys, Arg and His are similar; the acidic residues Glu and Asp are similar; and the hydrophilic, uncharged residues Gln, Asn, Ser, Thr and Tyr are similar. The remaining natural amino acid Gly is not similar to any other amino acid in this context.
Throughout this description, alternative embodiments according to the invention fulfill, instead of the specified percentage of identity, the corresponding percentage of similarity. Other alternative embodiments fulfill the specified percentage of identity as well as another, higher percentage of similarity, selected from the group of preferred percentages of identity for each sequence. For example, a sequence may be 70% similar to another sequence; or it may be 70% identical to another sequence; or it may be 70% identical and 90% similar to another sequence.
In a preferred spider silk protein according to the invention, the REP-CT fragment has at least 70%, such as at least 80%, such as at least 85%, preferably at least 90%, such as at least 95% or even 100% identity to SEQ ID NO: 1 or SEQ ID NO: 3.
The silk proteins are advantageously further containing a functional moiety. This functional moiety may be protein moiety as exemplified herein or a non-protein moiety. The functional moiety is arranged either terminally in the silk protein or within the silk protein, preferably N-terminally or C-terminally in the silk protein.
Desired bioactivities may be introduced by various bioactive proteins and peptides, such as a fibronectin peptide motif, which enhances cell adhesion and proliferation on the silk coatings, and the antimicrobial peptide Magainin I. The potential to include functional moieties is crucial in biomaterial applications in order to optimize the acceptance of the implants in the body and to tackle infection issues, which are also challenging successful implantation. Furthermore, more advanced bioactivities can be introduced using silk proteins fused to functional moieties in the form of protein domains with fold-dependent functions, such as affinity domains (e.g. Z domain binding IgG), enzymes (e.g. xylanase) or growth factors (e.g. fibroblast growth factor, FGF).
The spidroin fragment and the functional moiety are linked directly or indirectly to one another. A direct linkage implies a direct covalent binding between the moieties without intervening sequences, such as linkers. An indirect linkage also implies that the moieties are linked by covalent bonds, but that there are intervening sequences, such as linkers and/or one or more further moieties, e.g. 1-2 NT moieties.
The functional moiety may be arranged internally or at either end of the spider silk protein, i.e. C-terminally arranged or N-terminally arranged. It is preferred that the functional moiety is arranged at the N-terminal end of the spider silk protein. If the spider silk protein contains one or more linker peptide(s) for identification and purification of the spider silk protein, e.g. a His or Trx tag(s), it is preferred that it is arranged at the N-terminal end of the spider silk protein.
A preferred spider silk protein has the form of an N-terminally arranged functional moiety, coupled by a linker peptide of 0-30 amino acid residues, such as 0-10 amino acid residues, to a REP moiety. Optionally, the spider silk protein has an N-terminal or C-terminal linker peptide, which may contain a purification tag, such as a His tag, and a cleavage site.
The silk protein optionally contains a cell-binding motif (CBM) as a functional moiety. The optional cell-binding motif is arranged either terminally in the silk protein or within the silk protein, preferably N-terminally or C-terminally in the silk protein.
In certain preferred embodiments of these and other aspects of the invention, the functional moiety is a cell-binding motif selected from RGD, IKVAV (SEQ ID NO: 6), YIGSR (SEQ ID NO: 7), EPDIM (SEQ ID NO: 8), NKDIL (SEQ ID NO: 9), GRKRK (SEQ ID NO: 10), KYGAASIKVAVSADR (SEQ ID NO: 11), NGEPRGDTYRAY (SEQ ID NO: 12), PQVTRGDVFTM (SEQ ID NO: 13), AVTGRGDSPASS (SEQ ID NO: 14), TGRGDSPA (SEQ ID NO: 15), CTGRGDSPAC (SEQ ID NO: 16) and FNCC (SEQ ID NO: 17); and preferably from FNCC, GRKRK, IKVAV, RGD and CTGRGDSPAC, more preferably FNCC and CTGRGDSPAC; wherein FNCC is C1X1X2RGDX3X4X5C2; wherein each of X1, X2, X3, X4 and X5 are independently selected from natural amino acid residues other than cysteine; and C1 and C2 are connected via a disulphide bond.
In its most general form, FNCC is C1X1X2RGDX3X4X5C2 (SEQ ID NO: 17); wherein each of X1, X2, X3, X4 and X5 are independently selected from natural amino acid residues other than cysteine; and C1 and C2 are connected via a disulphide bond. FNCC is a modified cell-binding motif that imitates the α5β1-specific RGD loop motif of fibronectin by positioning cysteines in precise positions adjacent to the RGD sequence to allow formation of a disulphide-bridge to constrain the chain into a similar type of turn loop. This cyclic RGD cell-binding motif increases the cell adhesion efficacy to a matrix made of a protein containing the cell-binding motif, such as a recombinantly produced spider silk protein. The term “cyclic” as used herein refers to a peptide wherein two amino acid residues are covalently bonded via their side chains, more specifically through a disulfide bond between two cysteine residues. The cyclic RGD cell-binding motif FNCC promotes both proliferation of and migration by primary cells. Human primary cells cultured on a cell scaffold material containing the cyclic RGD cell-binding motif show increased attachment, spreading, stress fiber formation and focal adhesions compared to the same material containing a linear RGD peptide.
In preferred embodiments of FNCC, each of X1, X2, X3, X4 and X5 are independently selected from the group of amino acid residues consisting of: G, A, V, S, T, D, E, M, P, N and Q. In other preferred embodiments of FNCC, each of X1 and X3 are independently selected from the group of amino acid residues consisting of: G, S, T, M, N and Q; and each of X2, X4 and X5 are independently selected from the group of amino acid residues consisting of: G, A, V, S, T, P, N and Q. In certain preferred embodiments of FNCC, X1 is selected from the group of amino acid residues consisting of: G, S, T, N and Q; X3 is selected from the group of amino acid residues consisting of: S, T and Q; and each of X2, X4 and X5 are independently selected from the group of amino acid residues consisting of: G, A, V, S, T, P and N. In some preferred embodiments of FNCC, X1 is S or T; X2 is G, A or V; preferably G or A; more preferably G; X3 is S or T; preferably S; X4 is G, A, V or P; preferably G or P; more preferably P; and X5 is G, A or V; preferably G or A; more preferably A.
In certain preferred embodiments of FNCC, the cell-binding motif is comprising the amino acid sequence CTGRGDSPAC (SEQ ID NO: 16). Further preferred cyclic RGD cell-binding motifs according to the invention display at least 60%, such as at least 70%, such as at least 80%, such as at least 90% identity to CTGRGDSPAC (SEQ ID NO: 16), with the proviso that position 1 and 10 are always 0; position 4 is always R; position 5 is always G; position 6 is always ID, and positions 2-3 and 7-9 are never cysteine. It is understood that the non-identical positions among positions 2-3 and 7-9 can be freely selected as set out above.
A preferred group of cell-binding motifs are FNCC, GRKRK, IKVAV, and RGD, and in particular FNCC, such as CTGRGDSPAC.
A preferred silk protein with an FNCC cell-binding motif has the amino acid sequence of SEQ ID NO: 3.
In the subsequent step (b), a mineral phase comprising a-tricalcium phosphate (α-TCP) with hydroxyapatite (HA) is blended into the foamed liquid phase from step (a) during continued mechanical foaming, thereby forming a macroporous paste;
It has surprisingly been found that the silk protein holds the potential as a foaming agent to produce macroporous CPSs for bone tissue engineering and bone regeneration. Silk-incorporated CPSs showed enhanced cohesion properties and silk-based CPS showed no negative effect on the final composition. Silk-foamed CPSs showed cohesiveness after only 2 h, which was faster than the specimens prepared with Tween 80 solution regardless of the consolidation process. Furthermore, silk was shown to have the potential to be used as a foaming template for macroporous CPSs, with tunable porosity by varing the concentration of silk. This is particularly surprising in view of the various methods that have been developed.
When combined with hydroxyapatite, the carbonyl groups (>C═O) in the peptide bonds of silk can serve as the crystal nucleation sites, favoring anisotropic crystal growth. Microstructures revealed by SEM showed a mixture of needle and plate-like crystals in the silk-foamed CPSs, while only plate-like structures were obtained in the specimens foamed with Tween 80, indicating that a preferential orientation of crystal growth was facilitated by silk. Moreover, a more compact structure was observed for silk-foamed CPSs, as small crystals tend to be tightly arranged. The underlying mechanism has been explained in previous studies as the calcium ions chelated by the carbonyl oxygens can further interact with the phosphate groups, allowing continuous growth of hydroxyapatite crystals along the c-axis. Consequently, higher nanoporosity and specific surface area are generated, which are critical to material-associated osteoinduction and bioresorbability in vivo.
A hydroxyapatite seed is present in the mineral phase to provide nucleation sites for hydroxyapatite crystals. The initial hydroxyapatite concentration in the mineral phase is typically 0.1-10 wt %, such as 1-5 wt %, e.g. about 2 wt %.
Up to present, two types of CPSs are commercially available: hydroxyapatite (HA, Ca9(HPO4)(PO4)5(OH), most stable above pH 4.2) and brushite (CaHPO4.2H2O, stable under pH 4.2). Apatite cements are considered to be relevant for bone regeneration due to that the inorganic phase of natural bones comprises substituents-containing HA. Biomimetic HA are typically calcium-deficient HA (CDHA, Ca10-x(HPO4)x(PO4)6-x(OH)2-x). HA or CDHA are stable under physiological condition and commonly prepared through hydrolysis of α-tricalcium phosphate (α-TCP):
3α-Ca3(PO4)2+H2O→Ca9(HPO4)(PO4)5(OH)
The incorporation of silk altered the nucleation and crystal growth mechanisms of the CPSs, as minor needle- and plate-like crystals were found on silk samples, compared to Tween dense and porous samples.
This aspect of the invention is advantageous in that the foaming agent—the silk protein—meets the essential requirement of being water-soluble and biocompatible, to be used as the foaming template for macroporous CPSs. The benefit of this method over using porogens, detergents and/or salts in order to generate macroporosity is that the biocompatibility and injectability of the CPSs are not undermined by the use of the silk in order to achieve interconnectivity of the pores.
Despite the advantages of CPSs of being injectable and able to self-harden under physiological conditions, disintegration and release of fragments in contact with blood or other body fluids leading to embolism is a severe drawback in clinical practice. Previous studies have pointed out that one of the approaches to enhance cohesion of CPSs is by addition of a viscosity-enhancing agent, a polymeric additive for instance. Another plausible mechanism is through interaction of calcium ions with functional groups such as carboxylic groups in the additives, which gives rise to enhanced inter-crystal attraction, resulting in improvement on cement cohesion. It was surprisingly found thas the effect of silk enhancing the cohesion of CPSs are likely to lie at the basis of both criteria. Furthermore, the present invention is also based on the herein disclosed, surprising insights that by excluding detergents such as Tween 80, the cohesion time of silk-foamed CPSs (2 h) was even shorter than that of the dense controls (2 h 30 min, Table 2) at a constant L/P ratio, implying that the improved cohesion behavior was indeed originated from the presence of silk, rather than the compromised reactivity resulting from the low macroporosity and surface area compared to the Tween 80-foamed specimens.
In a preferred embodiment said method for producing a macroporous scaffold a powder form of the mineral phase is blended into the foamed liquid phase in step b).
The liquid foam is stabilized by incorporation of powder, which reduces the gravitational drainage and air diffusion between the bubbles. The porous structures are maintained through solidification of the cement. Furthermore, the inventors surprisingly found that silk-incorporated CPSs showed a dependence of compressive strength on macroporosity. As a matter of fact, a significant decrease in compressive strength was observed for Tween 80-foamed CPSs compared to the silk-foamed specimens in the present study, given that a two-fold higher macroporosity was achieved for the former.
Specimens foamed with silk (with a concentration of 3 mg/ml) showed lower macroporosity compared to Tween 80 but higher than the dense controls at the same L/P ratio. Despite that all of the samples presented intrinsic microporosity, more compact structures were shown by silk-incorporated CPSs compared to both the Tween 80-foamed specimens and the dense controls. While CPSs prepared with surfactant exhibited basically plate-like crystal morphology, a mixture of needle and plate-like structures were observed for silk incorporated cements, suggesting that the nucleation process was regulated by silk.
Apart from the foaming agent, Liquid-to-Powder (L/P) ratio may be another parameter of vital importance concerning the macroporosity of the CPSs. The porosity and microstructure of CPSs may be closely related to the L/P ratio and the particle size of the starting powder. Typical L/P ratios are in the range of 0.1-1.0, preferably 0.4-0.7, e.g. 0.5 or 0.65. As dissolution and crystal precipitation occur initially at the surface of the original particles, CPSs prepared at high L/P ratio may tend to show increased porosity, owing to enlarged distance between powder particles. With increasing L/P ratio, less powder is added, resulting in less shear forces required for mixing, thereby limiting the disruption of the foam structure, giving higher macroporosity in the end.
In a preferred embodiment, the mineral phase is comprising a retardant, such as sodium pyrophosphate (SPP). The retardant delays the setting of the composition to ensure workability thereof. A suitable concentration of the retardant in the mineral phase is 1-10 wt %, preferably 3-7 wt %, e.g. approximately 5 wt %.
In one preferred embodiment, the mineral phase is comprising an accelerator, such as Na2HPO4.2H2O (NaP). The presence of the accelerator NaP in the mineral phase accelerated the reaction and surprisingly improved the porosity. A suitable concentration of the accelerator in the mineral phase is 0.5-5 wt %, preferably 2-3 wt %, e.g. approximately 2.5 wt %.
In the subsequent step (c), the macroporous paste is stabilized into a desired shape. This implies that cement cohesion of the paste occurs. The stabilized paste stays integrated and resistant to washout.
The pastes can e.g. be transferred into a silicon mold of desired shape and placed in a humid atmosphere at 37° C. for 16 h to ensure cohesion.
Stabilization of the paste occurs at temperatures at or slightly above room temperature, and increasing temperature may be used to accelerate the speed of the reaction to induce faster formation of provided compositions.
In one embodiment, step c) is performed at 15-40° C. A preferred temperature range is room temperature, such as 15-25° C. or 20-25° C. This facilitates the stabilizing step and allows for molding into various shapes. Another preferred temperature range is around body temperature, such as This is useful for in vivo applications and allows for molding into various shapes.
In the subsequent step (d), the stabilized, shaped porous paste is mineralized into a macroporous calcium-deficient hydroxyapatite (CDHA)-based scaffold.
Mineralization of the stabilized, shaped paste can e.g. be achieved by immersing it in phosphate buffered saline (PBS) for 7 days to allow complete setting.
Mineralization of the stabilized, shaped paste occurs at temperatures at or slightly above room temperature, and increasing temperature may be used to accelerate the speed of the reaction to induce faster formation of provided compositions.
In one embodiment, step d) is performed at 15-40° C. A preferred temperature range is room temperature, such as 15-25° C. or 20-25° C. This facilitates the mineralizing step and allows for fixation into various shapes. Another preferred temperature range is around body temperature, such as 35-40° C. This is useful for in vivo applications.
The method for producing a macroporous scaffold disclosed herein is advantageous in that it does not comprise any high-temperature sintering step, i.e. treatment at temperatures above 800° C., involving the silk protein.
The method for producing a macroporous scaffold disclosed herein is also advantageous in that it does not involve addition of any non-biocompatible components such as detergents, solubilizing agents, and/or synthetic surfactants. Avoiding non-biocompatible components is highly desirable not to cause significant harm to living tissue when placed in contact with such tissue, e.g. in vivo and/or cause cell death, induce inflammation and/or other such adverse effects when administered in vivo.
In one embodiment, said method for producing a macroporous scaffold as disclosed herein is an in vitro method.
Using in vitro methods allows for the possibilty to investigate the ability of spider silk protein to serve as a foaming agent to produce novel macroporous CPS scaffolds for bone regeneration and to further study the synergies between silk and CPSs.
Successful bone regenerative scaffolds are expected to have a high porosity with interconnected macropores which support cell migration and vascularization while the implants are progressively degraded and replaced by newly-formed tissue.
These biomolecules not only offer desirable foaming capacity, but also biological recognition in vivo, thereby promoting cell adhesion and bone ingrowth.
CPSs may have similar compositions to the mineral phase of naturals bones, and may thus be resorbable in vivo.
In embodiment of the method, the macroporous paste is implanted into a human or an animal prior to step (d).
In one embodiment of the method, the macroporous paste is implanted into a human or an animal prior to step (c).
In a certain embodiment of the method, the macroporous paste is injected into a human or an animal prior to step (c).
In a specific embodiment, steps (c) and (d) are performed in situ in a human or an animal.
These features are highly advantageous, making the macroporous scaffold extremely appealing as bone void-fillers as they allow for minimal invasive procedure, and are adaptable to irregular cavities, avoiding gaps between scaffolds and the host tissue. Once injected, the macroporous paste may self-harden in situ through an isothermic setting reaction at physiological temperature. The self-setting reaction may occur in a dissolution/precipitation process: 1) dissolution of the powder upon mixing with the liquid phase, leading to supersaturation; 2) nucleation induced by supersaturation; 3) subsequent growth and precipitation of entangled crystals, resulting in macroscopic hardening of the cement.
According to a second aspect of the invention, a macroporous paste for use in the treatment or surgery of the human or animal body is provided.
The inventors surprsiningly found that silk hold the potential to serve as the foaming template for macroporous CPSs. The handling properties of CPSs were improved by incorporation of silk, showing positive effect on the cement behavior. The compositions of CPSs were also maintained in the presence of silk. The results of this invention shed light on a potential porous scaffold material for bone tissue engineering.
The present invention will now be further illustrated by the following non-limiting examples.
Alpha-Tricalcium Phosphate (α-Ca3(PO4)2, α-TCP) was synthesized by heating the mixture of calcium hydrogen phosphate (CaHPO4, Sigma-Aldrich, USA) and calcium carbonate (CaCO3, Acros Organics, USA) with a 2:1 molar ratio at 1400° C. for 2-15 h followed by quenching in the air. The powder obtained was milled in a planetary ball mill (PM400, Retsch, Germany) for 15 min at 300 rpm. For a-TCP, 2 wt % of hydroxyapatite seed (HA, Merck, Germany) was added after milling to provide nucleation sites for hydroxyapatite crystals.
In addition, some preliminary studies were performed with and without the addition of 2.5 wt % Na2HPO4.2H2O (NaP) to modify the pH and accelerate the reaction. This powder was used as the solid phase for preparation of calcium phosphate cements (CPC).
A silk protein functionalized with a RGD-motif from fibronectin (FN-silk; (“FNCC-4RepCT”) having the amino acid sequence of SEQ ID NO: 3 was used.
Different parameters were applied to manufacture CPSs as following:
1) CDHA was prepared through hydrolysis of α-TCP.
2) two L/P ratio: 0.50 and 0.65 mL·g-1,
3) two types of foaming agents: Tween 80 (Sigma-Aldrich, USA) and silk. A consistent concentration of 1 wt % of Tween 80 was used as the control. For silk, three different concentrations were used to prepare macroporous CPSs: 1.5 mg/ml, 2 mg/ml, 3 mg/ml;
4) two consolidation methods: direct foaming for macroporous CPSs and manually mixing for dense specimens.
5) After mixing, cement pastes were transferred immediately into silicon molds with two dimensions: ∅12× height 1 mm and ∅6× height 1 mm. Samples were placed in a humid atmosphere, e.g. for 16 h, at 37° C. to ensure cohesion, and subsequently immersed in 1×PBS solution (Sigma-Aldrich, USA) for 7-10 days to allow complete setting.
The CPS scaffolds were characterized in terms of composition, cohesion time, porosity, morphology, specific surface area, and compressive strength.
Phase characterization was performed using a diffractometer (D5000 Th-Th, Siemens, Germany) equipped with a Cu Kα anode. Data were collected over 2θ range from 10° to 80° at steps of 0.02° per second.
Phase identification was achieved by comparing the patterns obtained with the standard of α-TCP (JCPDS 04-010-4348), β-TCP (JCPDS 04-008-8714), HA (JCPDS 01-074-0565) brushite (JCPDS 04-013-3344) and monetite (JCPDS 04-009-3755).
Semi-quasi quantitative analysis was done by DIFFRAC.EVA (Bruker, USA). Chemical composition was characterized by attenuated total reflection-fourier-transform infrared spectroscopy (ATR-FTIR, Tensor 27, Bruker, USA). Transmittance spectra were obtained in the range of 400-4000 cm−1 with 1024 scans and a resolution of 2 cm−1 for silk-incorporated CPSs. For silk-free samples, spectra were obtained in the same range using 64 scans and resolution of 4 cm−1.
Cohesion time of CPSs at 37° C. was determined by addition of 0.5 mL PBS solutions to cement paste filled into a well-plate using a spatula at a set time interval of 30 min.
Macroporosity of the scaffolds was assessed by microcomputed tomopography (micro-CT, Skyscan, Bruker, USA). Images were acquired with pixel size of 10 μm at a voltage of 100 kV and a current of 100 μA with an Al—Cu filter. Reconstruction was performed using NRecon (Bruker, USA) and calculations were done by CTAn software (Bruker, USA) to acquire macroporosity.
The morphology and microstructure of CPSs were studied by scanning electron microscope (SEM, LEO 1550 Gemini, Carl Zeiss Meditec AG, Germany). Samples were sputtered with a thin gold coating for 40 s prior to operation. For each sample, surface and fractured cross-section were analyzed. The specific surface area was analyzed using Brunauer-Emmett-Teller (BET) method (ASAP 2020, Micrometrics Instrument Corp., USA).
The compressive strength was measured according to the ISO 5833 standard. The mechanical testing machine (Shimadzu AGS-X, Japan) was operated with a 5 kN load cell at a testing speed of 1 mm/min until failure. Samples were molded into cylindrical silicon molds with the dimension of ∅6× height 12 mm and stored in PBS at 37° C. for 10 days prior to measurement.
The cohesion time of CPSs prepared with a constant L/P ratio of 0.65 ml/g are summarized in Table 1. The silk-foamed samples were prepared with a concentration of 3 mg/ml. Effect of consolidation method on cohesion of CPSs were studied between the dense and the Tween 80-foamed specimens. The dense controls were cohesive after 2 h 30 min of setting, which was considerably shorter compared to 3 h 30 min for the foamed ones due to the lower surface area and reduced reactivity. Regarding the influence of the foaming agent, silk was found to be able to promote cement cohesion, with silk-foamed CPSs showing cohesiveness after only 2 h, which was faster than the specimens prepared with Tween 80 solution regardless of the consolidation process.
The effect of foaming agent, L/P ratio, and consolidation method on the compositions of the CPSs were investigated by XRD and ATR-FTIR.
Phase analysis by XRD revealed that the final compositions were predominantly CDHA, in spite of the different preparation protocols (shown in
Infrared spectroscopy performed by ATR-FTIR confirmed that the chemical composition of CPSs was maintained consistent regardless of the foaming agent or L/P ratio used (shown in
The consolidation method did not have sound effect on the composition. Typical amide bands were not able to be identified for CPSs prepared with silk, due to the low amount of protein (3 mg/ml). Besides, significant decrease in intensity of the three amide peaks could be provoked by interaction of carbonyl groups (>C═O) in amide bonds with calcium ions.
Morphological analysis was performed by SEM (
For Tween 80-foamed CPSs, specimens with L/P ratio of 0.5 ml/g were less porous than 0.65 ml/g on macro-scale, which was later confirmed by micro-CT (shown in the Appendix). CPSs prepared at L/P=0.5 ml/g presented mainly large pores, while those prepared at L/P=0.65 ml/g had a mixture of pores with different sizes, showing a less compact architecture. Enlarged inter-aggregate spaces were found in the microstructures of the samples prepared at a higher L/P ratio. The nanostructure was not affected by the different L/P ratios. Few macropores could be observed for the dense controls, yet the microporosity remained, due to that CPSs are porous materials and the inherent porosity was independent on the consolidation process.
Macroporosity of the CPSs was assessed by micro-CT to investigate the performance of silk as macroporous template for bone cement scaffolds. All of the samples were prepared at L/P=0.65 ml/g, aiming for potentially higher porosity. As shown in
The specific surface area (SSA) of the CPSs prepared at L/P ratio 0.65 ml/g was determined by BET as summarized in Table 2. The silk-foamed specimens were foamed with a concentration of 2 mg/ml based on the macroporosity analysis. Similar SSA values were acquired among the three samples (˜25 m2/g), showing that neither the foaming agent nor the consolidation method had any fundamental effect on the overall nanostructure of the CPSs.
At the same L/P ratio, CPSs foamed with Tween-80 showed the highest macroporosity, while those foamed with silk presented fewer pores, showing slightly higher macroporosity than the dense controls. A more compact microstructure was observed in the presence of silk. In contrast, Tween 80-foamed CPSs and the dense controls displayed predominantly plate-like crystals, while silk-foamed CPSs had a mixture of needles and plates. For CPSs foamed with Tween 80, higher macroporosity and a wider pore size distribution was displayed at a higher L/P ratio. Increased inter-aggregate distance was found in the microstructure and L/P=0.65 ml/g. The nanostructure was not markedly affected. The dense controls showed compact structure in the macroscopic range, yet the textural properties were similar to the Tween 80-foamed specimens on micro/nanometric scale.
The porosity of the dense control was 0.53±0.26%, which was the lowest among all samples.
The compressive strength (CS) was measured for CPSs with L/P ratio of 0.65 ml/g, as shown in
The CS values were 0.33±0.17, 1.85±0.42, and 11.53±3.87 MPa respectively, which were in alignment with their macroporosity.
In a preliminary study, samples of silk-CPF with a silk concentration of 2 mg/ml in the liquid phase showed to provide the best stability of the macroporous template. Therefore, these samples were used as a reference when optimising the concentration of Tween in the Tween-CPF samples, with the aim to obtain two CPF materials with a similar distribution of macropores. In addition, the effect in the porosity of the presence of NaP in the solid phase was evaluated for both silk-CPF and tween-CPF. For this optimisation process, the porosity of the materials was analysed in terms of total porosity (uCT) and visualized by SEM.
The composition selected for the CPF samples was 2 mg/ml silk and w/v % Tween, both of them containing NaP in the solid phase.
The presence of NaP in the solid phase improved the surface porosity, as shown in
Mice mesenchymal (mMSC) cells cultured on CDHA were assayed for total lactate dehydrogenase activity to monitor cell growth (mean and standard deviation of 3 independent experiments). The results are shown in
Cross sections of silk CDHA scaffolds were prepared after 3 weeks of culture with mMSCs.
It was shown that cells survive and grow well when cultured on both the Silk CDHA and the control CDHAs (tween and dense).
The cells migrated from the top of the scaffolds and approximately half way through.
In vitro Tests
To investigate cell viability, CDHA CPSs were prepared at (A) L/P=0.5 ml/g; (B) L/P=0.65 ml/g seeded with 10×103 mice mesenchymal cells (mMSCs) per cm2. The concentration of silk used for silk-foamed CPSs was 2 mg/ml. The results are shown in
Fluorescence micrographs of (A) Tween 80-foamed CPSs, (B) silk-foamed CPSs, and (C) the dense controls with a L/P ratio of 0.5 ml/g at 3 h and 14 days were prepared. The micrographs are shown in
Early and late cell attachment, respectively, is shown in
Macroporosity of CPSs were prepared with different L/P ratios. The silk-foamed CPSs were prepared with a concentration of 1 mg/ml silk, see
SEM was performed and images of the macrostructures are shown in
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2020/074370 | 9/1/2020 | WO |
