HEPARAN SULPHATES
20170043053 · 2017-02-16
Inventors
Cpc classification
A61L15/60
HUMAN NECESSITIES
C08B37/0075
CHEMISTRY; METALLURGY
A61P17/02
HUMAN NECESSITIES
A61K8/735
HUMAN NECESSITIES
A61P19/08
HUMAN NECESSITIES
A61L27/3834
HUMAN NECESSITIES
A01N1/126
HUMAN NECESSITIES
A61L15/60
HUMAN NECESSITIES
A61P19/04
HUMAN NECESSITIES
A61L27/54
HUMAN NECESSITIES
International classification
A61L27/54
HUMAN NECESSITIES
C08B37/00
CHEMISTRY; METALLURGY
Abstract
A heparan sulphate that binds TGP1 is disclosed.
Claims
1. Heparan sulphate HS16.
2. Heparan sulphate HS16 in isolated or substantially purified form.
3. Heparan sulphate HS16 according to claim 1 or 2 wherein the HS16 is capable of binding a peptide or polypeptide having, or consisting of, the amino acid sequence RKDLGWKWIHEPKGYH (SEQ ID NO:1).
4. Isolated or substantially purified heparan sulphate HS16 according to any one of claims 1 to 3, wherein following digestion with heparin lyases I, II and III and then subjecting the resulting disaccharide fragments to HPLC analysis the heparan sulphate HS16 has a disaccharide composition comprising: TABLE-US-00005 Disaccharide Normalised weight percentage UA,2S-GlcNS,6S 14.75 3.0 UA,2S-GlcNS 4.58 2.0 UA-GlcNS,6S 12.98 3.0 UA-GlcNS 22.24 3.0 UA,2S-GlcNAc 0.56 0.5 UA-GlcNAc,6S 12.63 3.0 UA-GlcNAc 32.26 3.0
5. Isolated or substantially purified heparan sulphate HS16 according to any one of claims 1 to 3, wherein following digestion with heparin lyases I, II and III and then subjecting the resulting disaccharide fragments to HPLC analysis the heparan sulphate HS16 has a disaccharide composition comprising: TABLE-US-00006 Disaccharide Normalised weight percentage UA,2S-GlcNS,6S 14.75 1.0 UA,2S-GlcNS 4.58 0.4 UA-GlcNS,6S 12.98 1.0 UA-GlcNS 22.24 1.6 UA,2S-GlcNAc 0.56 0.4 UA-GlcNAc,6S 12.63 1.0 UA-GlcNAc 32.26 1.6
6. Heparan sulphate HS16, or heparan sulphate HS16 in isolated or substantially purified form, according to any one of claims 1 to 5 obtained by a method comprising: (i) providing a solid support having polypeptide molecules adhered to the support, wherein the polypeptide comprises a heparin-binding domain having the amino acid sequence RKDLGWKWIHEPKGYH; (ii) contacting the solid support with a mixture comprising glycosaminoglycan, preferably a heparan sulphate preparation, such that polypeptide-glycosaminoglycan complexes are allowed to form; (iii) partitioning polypeptide-glycosaminoglycan complexes from the remainder of the mixture; (iv) dissociating glycosaminoglycans, preferably heparan sulphate species, from the polypeptide-glycosaminoglycan complexes; (v) collecting the dissociated glycosaminoglycans, preferably one or more heparan sulphate species.
7. The method of claim 6 wherein the polypeptide has, or consists of, the amino acid sequence selected from RKDLGWKWIHEPKGYH (SEQ ID NO:1).
8. The method of claim 6 or 7 wherein the mixture comprising glycosaminoglycans is a heparan sulphate preparation obtained from porcine mucosa (HS.sup.PM).
9. A composition comprising heparan sulphate HS16 according to any one of claims 1 to 8.
10. The composition of claim 9, further comprising a growth factor, preferably TGF1.
11. A cosmetic composition, pharmaceutical composition or medicament comprising heparan sulphate HS16 according to any one of claims 1 to 8.
12. The pharmaceutical composition or medicament of claim 11 wherein the pharmaceutical composition or medicament further comprises TGF1 protein and/or mesenchymal stem cells.
13. The pharmaceutical composition or medicament of claim 11 or 12 for use in a method of medical treatment.
14. Heparan sulphate HS16 according to any one of claims 1 to 8 for use in a method of medical treatment.
15. Heparan sulphate HS16 according to claim 14 wherein the method of medical treatment comprises a method of wound healing in vivo.
16. Heparan sulphate HS16 according to claim 14 wherein the method of medical treatment comprises the repair and/or regeneration of tissue, preferably connective tissue.
17. Use of Heparan sulphate HS16 according to any one of claims 1 to 8 in the manufacture of a medicament for the treatment of a disease, condition or injury to tissue, wherein the method involves the repair and/or regeneration of tissue, preferably connective tissue.
18. A method of treating a disease, condition or injury to tissue in a patient, the method comprising administration of a therapeutically effective amount of heparan sulphate HS16 to the patient leading to repair and/or regeneration of the tissue, preferably connective tissue.
19. The method of claim 18 wherein the method comprises administering heparan sulphate HS16 to tissue at or surrounding a wound or location on the patient's body at which regeneration or repair of tissue is required.
20. The method of claim 18 or 19 wherein the method further comprises administering TGF1 protein to the patient.
21. A method of treating a disease, condition or injury to tissue in a patient, the method comprising surgically implanting a biocompatible implant or prosthesis, which implant or prosthesis comprises a biomaterial and heparan sulphate HS16 according to any one of claims 1 to 8, into tissue of the patient at or surrounding the site of the disease, condition or injury leading to repair and/or regeneration of the tissues.
22. A biocompatible implant or prosthesis comprising a biomaterial and heparan sulphate HS16 according to any one of claims 1 to 8.
23. A method of forming a biocompatible implant or prosthesis, the method comprising the step of coating or impregnating a biomaterial with heparan sulphate HS16 according to any one of claims 1 to 8.
24. A method of culturing stem cells in vitro, the method comprising culturing stem cells in vitro in contact with heparan sulphate HS16 according to any one of claims 1 to 8.
25. Culture media comprising heparan sulphate HS16 according to any of claims 1 to 8.
26. The culture media of claim 25, further comprising TGF1.
27. A kit of parts, the kit comprising a predetermined amount of heparan sulphate HS16 according to any one of claims 1 to 8 and a predetermined amount of TGF1.
28. Products containing therapeutically effective amounts of: (i) Heparan sulphate HS16 according to any one of claims 1 to 8; and one or both of (ii) TGF1 protein; (iii) Mesenchymal stem cells, or cells in the fibroblastic lineage, for simultaneous, separate or sequential use in a method of medical treatment.
29. A method of increasing the stability of a growth factor, preferably TGF1, the method comprising contacting a growth factor, preferably TGF1, with heparan sulphate HS16 according to any one of claims 1 to 8.
30. A cosmetic method comprising administering heparan sulphate HS16 according to any one of claims 1 to 8 to a subject.
31. A preparation comprising a blood derived product and a predetermined quantity of heparan sulphate HS16.
32. The preparation of claim 31, wherein the preparation is a platelet preparation.
33. A method of preserving biological material, the method comprising contacting biological material with a predetermined quantity of heparan sulphate HS16.
Description
BRIEF DESCRIPTION OF THE FIGURES
[0293] Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:
[0294]
[0295]
[0296]
[0297]
[0298]
[0299]
[0300]
[0301]
[0302]
[0303]
[0304]
[0305]
[0306]
[0307]
[0308]
[0309]
[0310]
[0311]
[0312]
[0313]
EXAMPLES
Example 1
Structural Requirements for Heparin/Heparan Sulfate-Transforming Growth Factor-1 Interactions and Signal Potentiation
[0314] Background: Heparin is able to bind to and potentiate transforming growth factor-1 (TGF-1) signaling.
Results: The molecular determinants of the interaction of heparin/heparan sulfate (HS) and TGF-1 were identified.
Conclusion: There are defined structural requirements for the interaction of TGF-1 with heparin/HS which influence TGF-1 signal potentiation.
Significance: An understanding of HS-TGF-1 interactions can guide TGF-1 therapy development.
Abstract
[0315] Transforming growth factor-1 (TGF-1) is a heparin binding protein that has been implicated in a number of physiological processes, including the initiation of chondrogenesis by human mesenchymal stem cells (hMSCs). Here we show that heparin can bind to and potentiate TGF-1 signaling for hMSCs. This potentiation occurs through the modulation of the TGF-1 pathway via TGF- receptors and leads to the upregulation of early chondrogenic genes. Molecular interaction and cell-based assays also demonstrated that heparin chains that are 18-22 saccharides (dp18-22) in length and lack 2-O-sulfation are optimal for binding TGF-1. Interrogation of the interaction between TGF-1 and heparin through structural proteomics allowed the identification of novel lysine residues on TGF-1 involved in heparin binding. With this information we isolated a sub-fraction of porcine mucosal heparan sulfate (HS) that had an increased affinity for TGF-1. This TGF-1-binding HS was able to better bind to and potentiate the activity of both TGF-1 and latent TGF-1 compared to the original starting HS. This study is the first to report on the structural requirements for the interaction of heparin with TGF-1. It also lays the foundation for the development of an HS-based strategy to modulate TGF-1 signaling for cartilage repair, where exogenous protein doses could be either reduced or dispensed with.
Introduction
[0316] The glycosaminoglycans (GAGs) heparan sulfate (HS).sup.1 and heparin are structurally related, linear polysaccharides that are known to bind numerous extracellular proteins and growth factors and modulate their functions (1). Transforming growth factor-1 (TGF-1) is a potent heparin-binding growth factor (2-5) that has been shown to play roles in fibrosis (6,7), skin healing (8), cancer metastasis (9,10) and chondrogenesis (11-15). This ability of TGF-1 to drive the chondrogenic differentiation of human mesenchymal stem cells (hMSCs) and maintain the chondrogenic phenotype has made it of particular interest in the development of cartilage repair strategies (13, 15-18).
[0317] While appearing successful initially, such approaches face significant barriers in their translation into the clinic, as supraphysiological doses of TGF-1 are often employed to overcome clearance, and even modest doses have been shown to produce undesirable outcomes, such as synovial inflammation (19,20). Apart from the problem of non-physiological doses, there is also the ongoing need to localize the growth factor to the site of treatment to prevent it from triggering systemic side effects, including fibrosis and oncogenesis (9, 10, 21). Additionally, sensitivity to TGF-1 decreases with age (22), so adequate TGF-1 dosing presents even more risk for aged patients. In response to these challenges, new strategies are being developed that reduce or completely remove the need for exogenous growth factors, better localize and control the delivery of the growth factor at the site of treatment, and boost either cellular sensitivity to the growth factor or the factor's signaling efficiency. Some groups have already addressed the first two hurdles through the use of self-assembling peptide amphiphiles (23,24), and have demonstrated that endogenous levels of TGF-1 are sufficient to drive local MSC differentiation (25). However, synthetic peptide amphiphiles pose significant immunogenic risk, and fail to address the need to enhance signaling activity within the desired cellular targets. An ideal therapy would act to enhance TGF-1 signaling without exogenous TGF-1 application.
[0318] Our group has previously shown that HS GAGs are able to modulate the effects of a number of clinically relevant growth factors (26-29). Here we examine the mechanism of action of heparin and HS association with TGF-1, and their potentiation of signaling within hMSCs. We demonstrate that the binding of heparin to TGF-1 potentiates its activity via the TGF- type I receptor-SMAD2/3 pathway, and that there are specific constraints on the structural requirements for such binding. Additionally, we utilize this information to isolate a TGF-1-binding population of HS that is compositionally different from, and more effective than porcine mucosal HS (HS.sup.PM) in potentiating TGF-1 signaling. The work here paves the way for further studies of TGF-1-HS interaction, and aids the development of HS-based strategies to regulate hMSC behavior for tissue repair.
Experimental Procedures
Human MSC Isolation and Cell Culture
[0319] Primary hMSCs (Lonza) were isolated from the bone marrow mononuclear cells of a young healthy adult human donor by plastic adherence and characterized as previously described (30,31). The adherent cells were maintained in a basal media consisting of DMEM-low glucose (1000 mg/l, DMEM-LG) supplemented with 10% fetal calf serum, 100 U/ml penicillin, 100 g/ml streptomycin and 2 mM L-glutamine, and cultured under standard conditions at 37 C. and 5% CO.sub.2 in a humidified atmosphere. Media replacement was every three days. Cells were detached with 0.125% trypsin/Versene (pH 7.0) upon reaching 75-80% confluence, and re-plated at a density of 3,000 cells/cm.sup.2 under the same culture conditions. All experiments were carried out with cells at passage 5.
Chondrogenic Differentiation
[0320] Chondrogenic differentiation was carried out using a modified micromass culture system as described by Zhang et al. (32). Briefly, passage 4 hMSCs were harvested and resuspended in chemically defined chondrogenic media (PT-3003, Lonza) at 210.sup.7 cells/ml. Droplets of 12.5 l were then seeded into the middle of each well in a 24-well plate and left to adhere at 37 C. for 2 h, after which, 500 l of chondrogenic media supplemented with either 10 ng/ml of TGF-1 (100-21C, PeproTech) alone (Media) or TGF-1 with 10 g/ml of heparin (Sigma-Aldrich) (Media+Hep) was added to each well. The cell droplets coalesced into spherical masses after 24 h and the micromasses harvested on day 3.
Surface Plasmon Resonance (SPR)-Based Analysis of TGF-1-GAG Interactions
[0321] Biotinylated heparin was prepared based on the protocol reported by Hernaiz et al. (33). Briefly, 20 mg of heparin was filter-sterilized (0.22 m) in 1 ml of water and incubated with 8.6 mol of N-hydroxysuccinimide-biotin (NHS-biotin) (Pierce) in 20 l of dimethyl sulfoxide (DMSO) for 2 h at 4 C. The biotinylated heparin was then extensively dialyzed (7000 MWCO) to remove unreacted biotin. Immobilization of the biotinylated heparin onto a streptavidin (SA) sensor chip (GE Healthcare) was carried out using the immobilization wizard on the Biacore T100 (GE Healthcare) with a targeted immobilization level of approximately 40 response units (RUs). HBS-EP running buffer (10 mM HEPES, 150 mM NaCl, 3.0 mM EDTA, 0.05% (v/v) Tween 20, pH 7.4) was used for the immobilization.
[0322] TGF-1-heparin interactions were effected by preparing a series of TGF-1 protein samples (50 to 800 nM final concentration) diluted in HBS-EP-0.1 running buffer (0.1% instead of 0.05% (v/v) Tween 20). For competitive binding experiments, a final concentration of 200 nM TGF-1 in HBS-EP-0.1 was mixed with either 5 or 10 g of one of the following GAGs: heparin (Hep); size-fractionated heparin (degree of polymerization dp4, 6, 8, 10, 12, 14, 16, 18, 20, 22 and 24) (Iduron); selectively desulfated heparin (2-O-desulfated, 6-O-desulfated and N-desulfated) (Iduron); HS.sup.PM (HO-03103, Celsus Laboratories); affinity-isolated TGF-1-binding HS (HS16.sup.+ve); or TGF-1-non-binding HS (HS16.sup.ve). The sample solutions were then injected over the heparin-coated chip at a flow rate of 30 l/min for 120 s, with HBS-EP-0.1 being subsequently passed over the chip for a further 1200 s to monitor TGF-1 dissociation. After dissociation, the sensor surface on the chip was regenerated by 2 washes of 2 M NaCl injected at 30 l/min for 60 s. The response was measured as a function of time (sensogram) at 25 C. The maximum binding response for each condition was normalized to the response obtained from TGF-1 alone.
GAG-Binding Plate Assay
[0323] To determine the ability of TGF-1 to bind to heparin, we utilized positively-charged GAG-binding plates (Iduron) as a capture substrate. GAGs were immobilized in each well and then challenged with TGF-1 according to the manufacturer's instructions. Briefly, triplicate wells were first pre-coated with 5 g/ml of full length heparin, size-fractionated heparin (dp14, 16, 18, 20, 22 and 24) or selectively de3ed heparin prepared in standard assay buffer (SAB: 100 mM NaCl, 50 mM sodium acetate, 0.2% v/v Tween 20, pH 7.2), and then incubated overnight at room temperature. The plates were next washed carefully three times with SAB, blocked with 250 l of blocking solution (0.4% w/v fish skin gelatine, Sigma-Aldrich, in SAB) and incubated for 1 h at 37 C. TGF-1 was then dissolved in blocking solution at a concentration of 100, 200, or 400 ng/ml. The plates were washed three times with SAB and each dilution of protein (200 l) was dispensed into triplicate wells and incubated for 2 h at 37 C., rinsed with SAB and 200 l of 750 ng/ml monoclonal mouse anti-TGF-1 antibody (MAB2401, R&D Systems) added in blocking solution. Plates were then incubated for 1 h at 37 C., washed with SAB, and 200 l of 1 g/ml polyclonal goat anti-mouse biotinylated antibody (ab6788, Abcam) added in blocking solution. Again, plates were incubated for 1 h at 37 C., washed with SAB, and 200 l of 220 ng/ml ExtrAvidin AP (Sigma-Aldrich) was added in blocking solution, incubated for 30 min at 37 C., and then rinsed with SAB. Finally, 200 l of development reagent (SigmaFAST p-Nitrophenyl phosphate, Sigma-Aldrich) was added, incubated at 37 C. for 40 min and read at 405 nm within 1 h.
Differential Scanning Fluorimetry (DSF)
[0324] DSF was performed on a 7500 Fast Real PCR System (software version 1.4, Applied Biosystems), as described previously (34,35). TGF-1 (2.5 M) was tested with or without heparin (25 M). To facilitate the melting of TGF-1, 10 mM of dithiothreitol (DTT) was added to the reaction mix. Experiments were run as previously described (34). First derivatives of the melting curves were calculated using Origin 7 (OriginLab Corp.) to determine the melting temperature of TGF-1 under the various conditions. Experiments were run in triplicate, but for the purposes of clarity, the data presented here only shows the average of the replicates.
Cell Lysis and Western Blotting
[0325] Human MSCs were cultured in 6-well plates in basal media for 24 h at a density of 10,000 cells/cm.sup.2. TGF-1 treatments were then prepared at either 1 ng/ml or 5 ng/ml alone or in the presence of either 10 g/ml or 40 g/ml of full length heparin, or 1 ng/ml of TGF-1 with 10 g/ml size-fractionated or selectively desulfated heparin, HS.sup.PM, HS16+ or HS16, and incubated at room temperature for 10 min before being added to the cells. Latent TGF-1 (LTGF-1) treatments were similarly prepared at 3.3 ng/ml alone or with 10 g/ml of the various GAGs described above. For inhibitor studies, the cells were pre-treated for 30 min with 10 M SB431542 (Sigma-Aldrich) or DMSO before treatment with TGF-1. The cells were then subjected to the various TGF-1 treatments for 1, 6 or 24 h and lysed in 2 Laemmli buffer, before being resolved on a 4-12% SDS-PAGE gel. Samples were then immunoblotted with antibodies against SMAD2/3 (#3102, Cell Signaling), phosphorylated SMAD2 (pSMAD2, #3108, Cell Signaling), phosphorylated SMAD3 (pSMAD3, #9520, Cell Signaling) and actin (MAB1501R, Millipore). Densitometry was carried out using Quantity One software (version 4.6.6, Bio-Rad).
Reverse Transcription and Quantitative PCR (qPCR)
[0326] Total RNA was isolated from chondrogenic micromass pellets using TRIZOL reagent (Invitrogen, Life Technologies) according to the manufacturer's protocol. Reverse transcription was carried out on 1 g RNA using the SuperScript VILO cDNA Synthesis Kit (Invitrogen, Life Technologies) following the manufacturer's instructions, with the incubation at 42 C. being carried out for 2 h instead of 1 h. Each qPCR contained 40 ng cDNA, 1 l TaqMan primer-probe mix per gene, and 10 l Taqman Fast Universal PCR Master Mix (Applied Biosystems, Life Technologies) in a final volume of 20 l. Thermal cycling conditions were 95 C. for 20 s, followed by 45 cycles of 95 C. for 3 s and 60 C. for 30 s. Each qPCR was run in duplicate and gene expression was normalized to HPRT1 expression to obtain the Ct value. The average value of biological triplicates was taken. Chondrogenic micromass pellets cultured in media without heparin (Media) were used as controls (Ct). Relative expression levels for each primer set were expressed as fold changes by the 2.sup.Ct method (36). The following TaqMan primer-probe assays (Applied Biosystems, Life technologies) were used: HPRT1 (Assay ID: Hs01003267_m1), SOX9 (Assay ID: Hs00165814_m1) and COMP (Assay ID: Hs00164359_m1).
Protect and Label
[0327] The heparin-binding sites on TGF-1 were identified by the Protect and Label approach, as described by Ori et al. for FGF-2 (37), except that 1 nmol of TGF-1 protein and 0.1% (w/v) RapiGest SF Surfactant (Waters Corporation) was used to elute the protein from the mini-column. Digested and biotinylated peptides were purified on a C18 ZipTip (Millipore) and then analyzed by tandem mass spectrometry (MS). Up to 2 g of the biotinylated peptides were injected into an LTQ Velos instrument (Thermo) using an EASY-nLC (Proxeon). Peptides were separated on a PicoFrit column (HALO, C18, 90 , 2.7 m, 75 m (ID)100 mm length) (New Objectives) using a 60 min linear gradient (2-40% (v/v) acetonitrile in 0.1% formic acid). Data acquisition was performed using a TOP-10 strategy where survey MS scans were acquired in the dual pressure linear ion trap. MS scans ranging from 310 to 1400 m/z, AGC target 3e4 and maximum injection time of 10 ms. The 10 most intense ions with an ion intensity above 1000 and a charge state excluding one were sequentially isolated to a maximum AGC target value of 4e4 for a maximal 100 ms and fragmented by Collision Induced Dissociation (CID) using a normalized collision energy of 30%. A dynamic exclusion list was applied using an exclusion list size of 500, one repeat count, repeat duration of 45 s, exclusion duration of 30 s as well as a mass width of 1.0 low and 1.5 high. Expiration was disabled.
[0328] Data analysis was performed using Mascot search (version 2.3, Matrix Science) using the ipi.HUMAN.v3.86.decoy database (183,568 sequences) and applying the following parameters: digest, chymotrypsin (FWYUP); maximum missed cleavages, 2; Fixed modifications, carbamidomethyl (Cys); possible modifications, acetyl (Lys), acetyl (Protein N-term), biotin (Lys), oxidation (Met); parental ion tolerance, 2 Da; fragment ion tolerance, 0.8 Da. Biotinylated peptides with a Mascot score higher than 20 were manually validated.
.SUP.3.H-Heparin-Binding Assay
[0329] To determine the heparin-binding ability of the TGF-1-derived peptide (sequence -RKDLGWKWIHEPKGYH-AHX-K(Biotin) [AHX=6-aminohexanoic acid]; [SEQ ID NO:7]), 0.5 mg of the peptide was reconstituted in 1 ml of phosphate buffered saline (PBS). The peptide was then adsorbed onto a nitrocellulose disc (6 mm diameter) by incubating the disc in 1 ml of the reconstituted peptide at room temperature for 1 h with constant shaking. Discs incubated in PBS alone served as negative controls. After adsorption, the discs were dried in a vacuum oven at 80 C. and 10 in Hg for 45 min, washed 3 times with PBS, and then incubated with 1 ml of 0.1 Ci/ml .sup.3H-heparin for 16 h at room temperature with constant shaking. The discs were then washed 4 times with PBS and the amount of .sup.3H-heparin bound measured with a scintillation counter.
Affinity Isolation of HS16.SUP.+ve
[0330] Isolation of HS16.sup.+ve was carried out as previously described (29) using the TGF-1 peptide sequence described above.
[0331] Briefly, 3 mg of the peptide was coupled to a HiTrap streptavidin HP column (GE Healthcare, Buckinghamshire, UK), which was then used for affinity chromatography with commercially available porcine mucosal HS (HSPM, Celsus Laboratories Inc, Ohio, USA). HSPM was dissolved at 1 mg/mL in low-salt buffer (20 mM phosphate, 150 mM NaCl, pH 7.2), loaded at a flow rate of 0.2 mL/min and the column washed in the same buffer until the baseline absorbance at 232 nm (A232) reached zero. Bound HS was eluted in a single step with high-salt buffer (20 mM Phosphate, 1.5 M NaCl, pH 7.2), peak fractions were monitored at A232, collected, and the column re-equilibrated with low-salt buffer. The eluted (HS16+ve) and flow-through (HS16ve) peaks were collected separately, freeze dried, desalted on a HiPrep 26/10 desalting column (GE Healthcare, Buckinghamshire, UK) at a flow rate of 10 mL/min, freeze-dried again and stored at 20 C.
Proton NMR Spectroscopy
[0332] HS.sup.PM, HS16.sup.+ve and HS16.sup.ve samples were pooled and exchanged in D.sub.2O three times (three passes of dissolution of the dried powder in D.sub.2O (0.5 to 1 mL) and freeze drying until fully lyophilized) and the dry weight determined. NMR analysis was carried out at 30 C. in 5 mm tubes as D.sub.2O solutions and included tBuOH (0.2 mg/mL) as an internal standard. The optimum concentration for comprehensive data-sets was 15 mg/mL (.sup.1H), albeit the HS preparations were approximately 3 mg/mL. Proton (500 MHz) NMR spectra were recorded on a three channel Bruker AvanceIII500. The probe was a Bruker two channel 5 mm broadband Nuclei Probe (31P-109Ag) equipped with actively shielded 50 G/cm Z-axis Pulsed Field Gradients. The NMR spectra were phase corrected as required and were reference to tBuOH (.sup.1H 1.24 ppm; .sup.13C (methyl) 30.29 ppm). Assignments for signals were based on those reported by Guerrini et al. (38).
Alcian Blue/Silver Stain of GAG in Native PAGE
[0333] To examine the size distribution of polysaccharide chains in HSPM, HS16+ve and HS16ve samples, 2 g of each GAG was run on a 12% native PAGE gel that had been pre-run at 80 V for 30 min to remove residual ammonium persulfate and tetramethylethylenediamine (TEMED). Samples were prepared in a final volume of 25 L with 4 electrophoretic mobility shift assay (EMSA) buffer (40 mM Tris-HCl, pH 8.0, 40% (v/v) ultrapure glycerol, 0.4% (v/v) NP40 and 400 mM KCl) diluted to 1 with Tris-Glycine buffer (25 mM Tris, 192 mM Glycine). A molecular weight ladder and BSA were used as molecular weight markers, while heparin was used as a positive control for the Alcian Blue staining. The gel was then stained with 0.5% (w/v) Alcian Blue in 2% (v/v) acetic acid for 45 min, destained in 2% (v/v) acetic acid for 15 min and washed in MilliQ water overnight to remove excess stain. Subsequently, the gel was silver stained to visualise the protein markers and enhance the contrast of the Alcian Blue-stained GAGs.
HPLC-Size Exclusion Chromatography-Refractive Index (HPLC-SEC-RI) of Affinity Isolated HS
[0334] HPLC-SEC-RI chromatograms were obtained using a TSK gel G4000PWXL (7.8 mm30 cm) and a TSK gel G3000PWXL (7.8 mm30 cm) (TOSOH Corp.) in series on a Waters 2690 Alliance system with a Waters 2410 refractive index monitor (range 64). The dn/dc for quantification from the RI was set at 0.129 (39). Samples were injected (50 g) and eluted with 50 mM ammonium acetate with a flow rate of 0.5 ml/min, at room temperature. Data was collected and analyzed using DAWN Astra software (Version 4.73.04, Wyatt Technology Corp.). The elution volumes of molecular weight (MW) standards were based on the elution volumes of heparin oligosaccharides (Iduron and Dextra Laboratories) run under the same conditions. Run times for these columns were 100 min in both cases. All GAG samples were at a concentration of 1 mg/ml in water.
Digestion of HS Samples with Heparin Lyase Enzymes
[0335] HS.sup.PM, HS16.sup.+ve and HS16.sup.ve samples were solubilized in water (1100 l) and filtered (Minisart RC15, 0.2 m syringe filter unit, Sartorius Stedim, #17761) to remove any particulate matter. As a further clean-up step, the filtered solution was passed through a 2000 MWCO membrane (Vivaspin 2, Hydrosart, Sartorius Stedim, #VS02H91, 2000 MWCO HY membrane, 2 mL ultrafiltration spin column) by centrifugation (4000 rpm, 1 h, 15 C.). The retentate was washed with water (31 ml), recovered from the filter and lyophilized. The purified HS samples were solubilized in water (1 mg/ml) and aliquots (21 ml) of each freeze-dried sample were taken for analysis. The HS samples were digested to di- and oligosaccharides by the sequential addition of heparin lyase enzymes (Heparin lyase I, II and III, Ibex Technologies) based on the method of Brickman et al. (40), but with some modifications. The dry HS samples were re-solubilized in digestion buffer (500 l; 50 mM sodium phosphate buffer, pH 7.0) and heparin lyase I (5 l; 5 mIU) was added to each sample. The samples were incubated (37 C., 2 h) with gentle mixing on a rotating wheel (9 rpm). Heparin lyase III (5 l; 5 mIU) was added to the digests and incubated for a further 1 h (as above). Heparin lyase 11 (5 l; 5 mIU) was added and the digests were incubated as above, for 18 h. Finally, aliquots (5 l; 5 mIU) of all three heparin lyases were added simultaneously and the digests were incubated for a further 24 h. The enzyme digestion was terminated by heating (100 C., 5 min). All three HS samples were digested in duplicate and analysed by HPLC with UV detection (232 nm).
HPLC-SEC-RI of Digested HS Samples
[0336] The HPLC-SEC chromatograms were obtained using two Superdex Peptide 10/300 GL columns (30010 mm, GE Healthcare) in series, on a Waters 2690 Alliance system with a Waters 2410 refractive index detector (range 64). The dn/dc for quantification from the RI was set at 0.129 (39). Samples (2 mg/ml) were injected (50 l; 100 g) and eluted with 50 mM ammonium acetate (0.5 ml/min) at room temperature. Heparin oligosaccharide standards (Iduron and Dextra Laboratories) were run under the same conditions. Run times for these columns were 120 min. Data was collected and analysed using DAWN Astra software (Version 4.73.04, Wyatt Technology Corp).
Disaccharide Compositional Analysis by HPLC
[0337] Twelve disaccharide standards, derived from the digestion of high-grade porcine heparin by bacterial heparinases, were purchased from Iduron. A stock solution of each disaccharide standard was prepared by dissolving the disaccharide in water (1 mg/ml). To determine the calibration curves for the disaccharide standards, a standard mix containing 20 g/ml of each of the disaccharides was prepared from the stock solutions. From this twelve disaccharide standard mix a dilution series containing 20, 10, 5, 2.5, 1.25, 0.625 and 0.3125 g/ml of each disaccharide was prepared. The HS.sup.PM, HS16.sup.+ve and HS16.sup.ve digests (2 mg/ml) were diluted with water to give 100 g/ml solutions and then filtered using hydrophilic PTFE disposable syringe filter units (0.2 m, 13 mm, Advantec). The HPLC separation conditions were based on those of Skidmore et al. (41). The analyses were performed on an Agilent 1260 Infinity liquid chromatography system (Agilent Technologies) with an Agilent 1260 MWD VL detector monitored at 232 nm. HS-derived disaccharides were separated on a ProPac PA1 column (Thermo Scientific, 4 mm250 mm) with a guard column. Gradient elution was performed using a binary solvent system. Eluent A was water at pH 3.5 (adjusted using HCl), and eluent B was 2 M NaCl at pH 3.5 (adjusted with HCl). The gradient program was as follows: 100% A from 0-1 min, then 0-35% B from 1-32 min, then 35-65% B from 32-47 min, then 100% B from 47-57 min, then 100% A from 57-60 min. The injection volume was 50 l. The column was eluted at a flow rate of 1.0 ml/min and maintained at 40 C. Disaccharides present in the HS digests were identified from their elution times by comparison with the elution times of the disaccharides in the twelve disaccharide standard mixes. HS16+ and HS16 digests were injected twice per duplicate digest (4 injections in total), while HS.sup.PM samples were injected once per duplicate digest (2 injections in total).
Plasmin Digestion
[0338] In order to determine the ability of the various GAGs to protect TGF-1 from proteolytic digestion, TGF-1 (100 ng) was pre-incubated with either 10 g of Hep, HS.sup.PM, HS16.sup.+ve or HS16.sup.ve or alone in PBS at room temperature for 10 min. Plasmin digestion was carried out by adding 0.5 mU of plasmin to the TGF-1 samples and incubating them at 37 C. for 1.5 h. Samples were subsequently run on a 4-12% SDS-PAGE gel and visualized by silver staining. All samples were made up to a final volume of 10 l in PBS.
Alkaline Phosphatase (ALP) Assay
[0339] To determine the effect of HS16+ve on BMP-2 activity, C2C12 mouse myoblasts were seeded in duplicate at 20,000 cells/cm2 in complete C2C12 medium (DMEM-LG, 10% (v/v) FCS, 100 U/mL penicillin and 100 g/mL streptomycin) and allowed to attach for 24 h. The complete medium was then replaced with treatment medium (DMEM-LG, 5% (v/v) FCS, 100 U/mL penicillin and 100 g/mL streptomycin) with or without 100 ng/mL BMP-2 and/or 5 g/mL HS16+ve, HS3+ve or HSPM and the cells incubated for 3 days. Total cell lysate was then collected in RIPA buffer containing a protease inhibitor cocktail (Calbiochem, Merck Millipore, MA, USA) and protein content determined using a BCA protein assay kit (Thermo Fisher Scientific). ALP activity was measured by incubating 5 g of protein with p-nitrophenyl phosphate (Sigma-Aldrich) for 1 h at 37 C. and reading the change in absorbance at 405 nm. RIPA buffer alone and 1 L (10,000 U/mL) of calf intestinal phosphatase (New England Biolabs Ltd, Ontario, Canada) were used as negative and positive controls respectively. Each sample was read in duplicate to give a total of 4 readings per treatment group.
Heparinase Treatment of hMSCs
[0340] To assess the influence of endogenous HS on TGF-1 signalling, hMSCs were seeded at a density of 7,500 cells/cm.sup.2 in 12-well plates in basal medium and allowed to attach overnight. A combination of heparinase I, II and III (1.2 mIU/mL of each) was then added to the media in each well and incubated for 24 h. The cells were then exposed for 6 h to TGF-1 treatments, prepared at either 1 ng/mL or 5 ng/mL alone or pre-incubated for 10 min at room temperature with either 10 g/mL or 40 g/mL of full length heparin, and then lysed for immunoblotting in 2 Laemmli buffer. Treatments were prepared in serum-free medium (DMEM-LG supplemented with 100 U/mL penicillin, 100 g/mL streptomycin and 2 mM L-glutamine) to avoid increasing the background levels of pSMAD seen when fresh serum is added to cells.
Immunofluorescence Staining of Heparinase-Digested Cells
[0341] To ensure that the heparinase treatment effectively removed endogenous HS chains from hMSCs, cells were seeded in 8-well chamber slides at a density of 3,500 cells/cm.sup.2 in basal medium. Cells were allowed to attach overnight before being treated with a combination of heparinase I, II and III (1.2 mIU/mL each), added directly to each well, for 24 h. Subsequently, cells were fixed in 4% (w/v) paraformaldehyde in PBS for 10 min at room temperature, blocked with 3% (w/v) bovine serum albumin (BSA) in PBS for 30 min at room temperature and then incubated with the anti-HS 10E4 antibody (1:25 dilution in 0.3% (w/v) BSA-PBS) (AMS Biotechnology, Abingdon, UK) for 3 h at room temperature. Cells were then incubated with an anti-mouse IgM-FITC antibody (1:500 dilution in 0.3% (w/v) BSAPBS) (BD Pharmingen, Becton, Dickinson and Company, NJ, USA) for 45 min at room temperature and the nuclei stained with Hoechst 33342 (2 g/mL in PBS) (Life Technologies) for 10 min at room temperature. Samples were imaged on an Olympus IX-81.
Results
Heparin Binds to TGF-1 and Potentiates its Activity
[0342] In order to determine the effects of heparin on TGF-1 signaling, we first set out to ensure that heparin was able to bind to TGF-1. Both SPR and GAG-binding plate assays demonstrated that TGF-1 bound in a dose-dependent manner to heparin that had been either immobilized in a 96-well plate (
[0343] The next step was to determine if the interaction between heparin and TGF-1 was influencing the TGF-1 signaling pathway. To accomplish this, hMSCs were treated with varying amounts of heparin and TGF-1, and protein harvested at 1, 6 and 24 h post-treatment in order to examine the levels of pSMAD2 and pSMAD3 (
[0344] To further investigate this effect, we proceeded to examine the transcript levels of TGF-1 target genes expressed during the early stages of the chondrogenic differentiation of hMSCs. Chondrogenic differentiation of the hMSCs was carried out in the presence of either Media (TGF-1 alone) or Media+Hep (TGF-1+Heparin). After 3 days of culture in chondrogenic media, the micromass pellets cultured in Media+Hep displayed 5-fold higher levels of both SOX9 and COMP mRNA transcripts (
Heparin Length Requirements for TGF-1 Binding and Activity
[0345] We next sought to determine the minimum length of heparin needed to bind to TGF-1. The ability of soluble size-fractionated heparin fragments (dp4 to dp24) to competitively inhibit the binding of TGF-1 to a heparin coated SA-chip increased in proportion to their length (
Heparin Sulfation Requirements for TGF-1 Binding and Activity
[0346] Given the predominantly ionic nature of heparin-protein interactions, we next set out to examine the influence that the various sulfate groups in heparin have over the interaction between it and TGF-1. Biacore competition assays indicated that the loss of 2-O sulfate groups from heparin (2-O-de) had a minimal effect on its ability to competitively inhibit TGF-1 binding to immobilized heparin (
[0347] Collectively, the data indicate that the loss of 2-O-sulfation from heparin, and to a lesser degree 6-O-sulfation, actually improves the ability of heparin to potentiate the TGF-1 signal, suggesting that the relationship between binding strength and bioactivity is not linear.
Identification of TGF-1 Heparin Binding Sites
[0348] As heparin is known to interact with configurations of basic residues present in numerous, susceptible growth factors, our next goal was to identify the actual heparin-binding site(s) within TGF-1. Previous studies that identified putative heparin-binding sites on TGF-1 did so through the identification of heparin-binding motifs present in the linear protein sequence (2,4). However, such an approach fails to take into consideration the full 3-dimensional (3D) conformational nature of proteins, and thus fails to identify heparin-binding sites that may only be apparent from the tertiary structure of the protein. As such, we decided to employ the Protect-and-Label strategy developed by Ori et al. (37) to determine if such 3D sites were present within TGF-1. Our analysis identified 8 lysines (K13, K26, K31, K37, K60, K95, K97 and K110) that appear to be involved in the binding of TGF-1 to heparin (
[0349] All but two of the identified lysines, K13 and K110, have been previously identified as part of TGF-1's heparin-binding domains (
Isolation of Affinity Selected TGF-1-Binding HS (HS16.SUP.+ve.)
[0350] Having identified the structural features and requirements of heparin-TGF-1 interactions, our next goal was to isolate a TGF-1-binding fraction of HS from the heterogeneous pool that constitutes commercially available HS.sup.PM preparations. To do so, we first designed a heparin-binding peptide derived from TGF-1 (
TABLE-US-00003 TABLE1 SummaryofpeptidesidentifiedbyProtectandLabelstructureproteomics. LabeledpeptideswereidentifiedbytandemmassspectrometryandanalyzedbyMascotsearchVersion 2.3(MatrixScience).Here,asummaryofthepeptidesinvolvedintheheparin-bindingsitesand thelabeledpositionisprovided. SEQID Peptide Sequence Residues.sup.a NO. 1 C(carbamidomethyl)FSSTEK(biotin)NC(carbamidomethyl)C(carbamidomethyl)VRQLY 7-21 8 2 IDFRK(biotin)DLGW 22-30 9 3 RK(biotin)DLGWK(acetyl)W 25-32 10 4 RK(acetyl)DLGWK(biotin)W 25-32 10 5 IHEPK(biotin)GY 33-39 11 6 SLDTQYSK(biotin)VL 53-62 12 7 YVGRK(biotin)PK(acetyl)VEQL 91-101 13 8 YVGRK(acetyl)PK(biotin)VEQL 91-101 13 9 SNMIVRSC(carbamidomethyl)K(biotin)C(carbamidomethyl)S 102-112 14 .sup.aResidue numbering according to FIG. 5A
[0351] Proton NMR, HPLC-SEC-RI and disaccharide compositional analyses were then carried out to determine if there were any systematic differences between HS.sup.PM, HS16.sup.+ve and HS16.sup.ve. NMR analysis of the three HS samples revealed several subtle differences (arrows,
[0352] The most notable difference in the NMR spectra of the three HS samples was the slight decrease in signal intensity at 5.4 ppm of HS16+ve (arrow,
[0353] Based on the elution times of several heparin size standards (dp8, 12, 20 and 26), our data also indicate that HS16.sup.+ve is composed of HS chains that are longer than 26 saccharides. Finally, disaccharide compositional analysis of the three HS sample digests showed that although HS.sup.PM and HS16.sup.ve were similar, HS16.sup.+ve was enriched in UA-GlcNS,6S and UA,2S-GlcNS,6S and contained less UA-GlcNAc, UA-GlcNS, UA,2S-GlcNAc and UA,2S-GlcNS (
[0354] HS chains are known to vary greatly in terms of chain length [Esko, J. D., K. Kimata, and U. Lindahl, Proteoglycans and Sulfated Glycosaminoglycans, in Essentials of Glycobiology, A. Varki, et al., Editors. 2009, Cold Spring Harbor Laboratory Press: Cold Spring Harbor, N.Y.], which partially explains their ability to bind to such a plethora of proteins. In order to enhance the relative specificity of an HS preparation for a given protein, it is necessary to reduce this variation. We examined the size distribution of polysaccharide chains within the three HS samples. Resolution of the three samples and heparin by native PAGE showed that HS16+ve was predominantly composed of longer HS chains compared to HS16ve and HSPM. Heparin, which is comparatively more homogenous than HS, was used to give an appreciation of the high heterogeneity present in HS preparations. To validate these findings, size exclusion chromatography (HPLC-SEC-RI) was carried out. The chromatograms from SEC indicated that HS16+ve consists of a poly-disperse subset of the HSPM from approximately dp8 to >dp26 (
[0355] The data indicate that the pool of HS that makes up HS16+ve is markedly different from HS16ve and HSPM with respect to both size distribution and composition. Additionally, the relative reduction in UA,2S-GlcNAc and UA,2SGlcNS seen in HS16+ve corroborated the earlier observation that the loss of 2-O sulfate from heparin actually served to increase its bioactivity towards TGF-1 (
TABLE-US-00004 TABLE 2 Disaccharide composition of heparin lyase digested HS samples. HS samples were digested with heparin lyase I, II and II and the resulting disaccharides were separated via HPLC. Disaccharides were identified by comparing their elution times with those of known disaccharide standards and their proportions in each HS sample were calculated with several calibration curves. % Disaccharide UA- UA- UA- UA,2S- UA- UA,2S- UA.2S- GlcNAc GlcNS GlcNAc,6S GlcNAc GlcNS,6S GlcNS GlcNS,6S HS.sup.PM 35.11 25.62 12.92 0.62 10.36 5.76 9.60 HS16+ 32.26 22.24 12.63 0.56 12.98 4.58 14.75 HS16 35.44 26.77 12.79 0.64 9.83 6.08 8.45
HS16.SUP.+ve .Binds to and Potentiates TGF-1 Signaling
[0356] Given the difference in the composition of HS16.sup.+ve compared to HS16.sup.ve and HS.sup.PM, we next set out to investigate if these differences resulted in any functional consequences. Examination of the ability of these HS fractions to bind to TGF-1 in the Biacore competition assay indicated that HS16.sup.+ve was able to bind to TGF-1 with a much higher affinity than HS16.sup.ve or HS.sup.PM (
[0357] As most TGF-1 in vivo is found in an inactive form, known as latent TGF-1 (LTGF-1), and HS16.sup.+ve was isolated from a pool of HS.sup.PM, we next sought to explore the effects that HS16.sup.+ve might have on LTGF-1. Our data demonstrated that again HS16.sup.+ve was able to potentiate LTGF-1-induced pSMAD signals more significantly than heparin (Hep), HS.sup.PM and HS16.sup.ve (
Isolation and Characterisation of hMSCs
[0358] In order to examine the biological effects of the heparin-TGF-1 interaction, it was necessary to isolate primary hMSCs. Commercially available bone marrow cells (Lonza) were purchased and hMSCs isolated via plastic adherence. Passage 0 cells were subsequently expanded and frozen in batches of 1106 cells per vial. Cells were screened by FACS for MSC surface marker expression at passage 5 as described by Dominici et al. [Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy, 2006. 8(4): p. 315-317]. More than 95% of the isolated hMSCs expressed CD73, CD90 and CD105, while CD14, CD19, CD34, CD45 and HLA-DR were not expressed. The isolated cells were also able to differentiate into osteoblasts, adipocytes and chondroblasts in vitro.
[0359] Thus the cells that had been isolated by plastic adherence were deemed to satisfy the minimal criteria characteristics of hMSCs.
Effects of Heparin on TGF-1 Signalling in hMSCs
[0360] Having isolated primary hMSCs, we next examined if the interaction between heparin and TGF-1 could influence the cellular response to TGF-1 signalling, which was measured by way of the downstream phosphorylation of SMAD2 and SMAD3. Heparin has been shown to potentiate the effects of TGF-1 in primary rat and bovine smooth muscle cells (SMCs) and the CCL64 mink lung epithelial cell line, but not in primary human saphenous vein (SMCs) [McCaffrey, T. A., et al., Transforming growth factor-beta activity is potentiated by heparin via dissociation of the transforming growth factorbeta/alpha 2-macroglobulin inactive complex. J Cell Biol, 1989. 109(1): p. 441-44; McCaffrey, T. A., et al., Protection of transforming growth factor activity by heparin and fucoidan. J Cell Physiol, 1994. 159(1): p. 51-59]. We therefore postulated that should heparin potentiate TGF-1 activity in hMSCs, the effects would only be seen as the pSMAD2 and pSMAD3 signals from TGF-1 alone started to subside. Thus the time points chosen for our initial TGF-1 dosing experiment were 6, 12, 24 and 48 h. Passage 5 cells were treated with a range of TGF-1 doses and total cell lysate collected at 6, 12, 24 and 48 h post treatment. Lysate samples were then resolved on 4-12% (w/v gradient) SDS-PAGE gels, transferred onto a nitrocellulose membrane and probed for phospho-SMAD2 (pSMAD2) (138D4, Cell Signaling Technology), phospho-SMAD3 (pSMAD3) (C25A9, Cell Signaling Technology), total SMAD2/3 (Cell Signaling Technology) and actin (Clone C4, Merck Millipore) by Western blotting. Our results showed that without TGF-1 (0 ng/mL), there was a low background level of pSMAD2 and pSMAD3 signalling. With 1 ng/mL, the pSMAD2 signal was quite intense at 6 h post treatment, and started to subside from 12 h onwards. The pSMAD3 signal mirrored that of the pSMAD2 signal, albeit at a lower intensity. With both 5 ng/mL and 10 ng/mL TGF-1, the pSMAD2 signal was seen to remain saturated across all the time points tested, but the pSMAD3 signal returned to background levels by 24 h. The 6 h time point was thus chosen for all subsequent experiments.
[0361] Our next goal was to determine the dose of heparin to use for our experiments. As McCaffrey et al. (supra)] have previously reported the effective TGF-1-potentiating dose of heparin to be between 1-100 g/mL, we chose to use doses within this range. Cells that were treated with 1 ng/mL of TGF-1 pre-incubated with 10 g/mL of heparin maintained a stronger pSMAD2 and pSMAD3 signal compared to cells treated with the same dose of TGF-1 alone (
[0362] We next sought to examine the influence that cell surface HS might have on TGF-1-driven SMAD signalling. To do so, passage 5 cells were treated with heparinase I, II and III (1.2 mIU/mL each) for 24 h in hMSC culture medium, before being treated with TGF-1 with or without heparin in serum-free medium. The 6 h time point used for Western blotting necessitated the use of serum-free media after heparinase treatment, in order to avoid the effects that the growth factors in the serum would have on background levels of SMAD2 and 3. Immunofluorescence staining of cell surface HS with the anti-HS 10E4 antibody showed that after 24 h treatment, nearly all the cell surface HS had been removed. However, the removal of cell surface HS did not appear to affect the pSMAD signals produced when the cells were treated with TGF-1. Our results suggest that the role played by heparin in potentiating TGF-1 signalling is different from the role it plays in FGF-2 signalling [Schlessinger, J., et al., Crystal Structure of a Ternary FGF-FGFR-Heparin Complex Reveals a Dual Role for Heparin in FGFR Binding and Dimerization. Molecular Cell, 2000. 6(3): p. 743-750.].
Comparison of HS16+ve and BMP-2-Binding HS (HS3+ve)
[0363] Having shown that HS16+ve enhances TGF-1 signalling, it was of interest to determine if it might similarly enhance the activity of other members of the TGF- superfamily. Our group has previously reported the affinity isolation of an HS fraction, HS3+ve, which enhances the activity of BMP-2 [Murali, S., et al., Affinity-selected heparan sulfate for bone repair. Biomaterials, 2013. 34(22): p. 5594-5605; WO2010/030244]. The structural similarity between the two proteins warranted a comparison of the composition of HS16+ve and HS3+ve (
[0364] This observed difference in the compositions of HS16+ve and HS3+ve had functional consequences for their activity. Surprisingly, HS16+ve was found to bind to BMP-2 better that HS3+ve in SPR-based binding competition assay (
[0365] Collectively, the data show that HS that is affinity purified using a TGF-1 peptide is compositionally different from the original HS preparation, will bind to the full length protein and potentiate its activity both in its active and latent forms. Also, the HS purified with the TGF-1 peptide is different from that purified with a BMP-2 peptide and this difference is sufficient to alter or tune its activity towards TGF-1 and reduce the heterogeneous effects of unfractionated HS.sup.PM.
Discussion
[0366] In this study we have shown that heparin is able to bind to TGF-1, and in so doing, enhance the thermal stability of the growth factor. This stabilization appears to prolong the half-life of TGF-1 signaling activity in hMSCs. Under chondrogenic differentiation conditions, this heparin-mediated potentiation enhanced the expression of early chondrogenic genes. Our findings support the idea that the potentiating effect of heparin on this chondrogenic gene expression occurs as a result of heparin acting via the TGF-1 signaling pathway, and that a GAG chain between 18-22 saccharides long is required to optimally bind to TGF-1 and potentiate its signal. Examination of the ternary TGF-1 ligand-receptor complex indicated that longer heparin chains might interfere with the binding of the TGF-6 type II receptor (TRII) with TGF-1 during complex formation (43). The loss of 2-O sulfation, and to a lesser extent 6-O sulfation, actually improves the ability of heparin to potentiate the TGF-1 signal despite a reduction in binding affinity. Together with our earlier findings on the influence of heparin chain length, these results provide compelling evidence for the current sugar code hypothesis of HS (44-47) and reinforce the idea of a non-linear relationship between binding strength and bioactivity (48). We have also identified K13 as a new residue on the TGF-1 monomer involved in heparin binding. Guided by that data we proceeded to isolate a population of HS that preferentially binds to TGF-1 (HS16.sup.+ve) from the heterogeneous mix obtained from porcine mucosal preparations. Characterization of HS16.sup.+ve demonstrated that it was compositionally different from HS.sup.PM and HS16.sup.ve, and that this difference enabled it to better bind to and potentiate both TGF-1 and LTGF-1 signaling in hMSCs.
[0367] Previous studies have shown that heparin will bind to TGF-1 and protect it from protease activity and circulatory clearance by a 2-macroglobulin, thereby potentiating its signal (3,5). This has led to its use as either a TGF- carrier or as a scaffold material in some cartilage repair studies (49,50). Some studies have even utilized heparin to control the release of growth factors during the chondrogenic induction of murine MSCs (51). However, this sugar is unlikely to see widespread adoption as a therapeutic agent for tissue repair or growth factor modulation, not only because of the risk of uncontrolled bleeding and thrombocytopenia (52), but because the hypersulfation of heparin means that is capable of promiscuously binding to over 200 different extracellular proteins, collectively known as the heparin interactome or heparanome (53,54). This implies that heparin does not possess sufficient specificity to be used for targeted growth factor modulation in an in vivo system, where growth factor production and localization cannot easily be controlled. Nonetheless, our in vitro data indicate that heparin can be used to prolong TGF-1 activity during hMSC chondrogenic differentiation, suggesting that there is a therapeutic potential to be realized if these inherent limitations can be overcome.
[0368] The use of HS instead of heparin could theoretically overcome these hurdles, as HS does not possess the anticoagulant activity of heparin (55). However, HS does give rise to its own set of problems, including the hypervariability of raw preparations. There are, however, techniques that have been developed to surmount these difficulties (26, 27, 29) and, by utilizing one of these techniques (29), we were able to demonstrate that selective sub-populations of HS that preferentially bind specific growth factors can be isolated. It was noteworthy that while HS16+ was isolated using a TGF-1-derived peptide, HS16.sup.+ve is also able to modulate the effects of LTGF-1, the more physiologically abundant form of TGF-1 in vivo. Interestingly, heparin has been reported to inhibit the activation of LTGF-1 (56), whereas HS16.sup.+ve does not. This raises interesting questions about the in vivo synthesis of TGF-1-binding HS by cells during normal and altered physiological states.
[0369] The work reported here lays the groundwork for future TGF-1-HS studies and builds on the binding model proposed by Lyon et al. (2). Our data identifies K13 as a novel residue that appears to influence the binding of heparin to TGF-1 through a spatial orienting of the polysaccharide through the groove that runs between the protein monomers (
[0370] TGF-1 is synthesized first as a pro-protein that is cleaved intracellularly to yield the small latent complex (SLC). Mature SLC consists of the TGF-1 dimer, noncovalently linked to the dimeric latency-associated peptide (LAP). For the majority of cell types so far studied SLC is released with latent TGF-1-binding protein-1 (LTBP-1) together, so forming the large latent complex (LLC) (61). LTBP-1 pushes latent TGF-1 into the extracellular matrix (ECM) by interacting with a variety of adhesive proteins (62), so creating deposits of latent TGF-1 that can be made available upon cell-mediated activation. Although the LAP has a structure considered to be stable, two regions of the molecule can be unfolded (60) in such a way that it traps TGF-1 in the SLC. When the conformations of these regions are mechanically forced fully open, active TGF-1 is released from the LAP (58). This simultaneous unfolding of both domains, an all-or-nothing snap mechanism necessary for full TGF-1 release, is possible only when LAP is bound to LTBP-1. Whether and how HS is involved in either the generation or release of this mechanical force is an interesting question.
[0371] All these considerations point to the need for extensive in silico modelling of the heparin-TGF-1 interaction to validate our results, and the development of computational tools to decipher the domain organization of HS preparations like HS16.sup.+ve (63). Such studies would open up the possibility of refining our affinity-based isolation of HS or even the synthesis of chemically defined TGF-1-specific HS molecules (64).
[0372] In conclusion, we show that heparin, and affinity isolated HS, can be used to modulate TGF-1 signaling on hMSCs. We are also the first to report on the structural requirements for heparin-TGF-1 interactions. Taken together, the data reiterates the importance of how an understanding of the structural interaction between these molecules can guide therapy development. This holds promise for the development of a novel therapeutic strategy for cartilage repair, which utilizes carbohydrate molecules to modulate TGF-1 activity to drive the chondrogenic differentiation of hMSCs. Such a strategy could also be extended to other tissue repair strategies that involve the use of growth factors.
[0373] In our study, a peptide containing the heparin binding site of TGF-1 was used to isolate a TGF-1-binding fraction of HS from porcine mucosal HS (HSPM) by affinity purification. The isolated TGF-1-peptide binding HS, termed HS16+ve, was found to be compositionally different from the non-bindng HS fraction, termed HS16ve, and the original HSPM starting material. This variance in composition enhanced the ability of HS16+ve to bind to, and modulate the activity of TGF-1 relative to HS16ve and HSPM. Surprisingly, HS16+ve was also able to modulate the activity of LTGF, the inactive, storage form of TGF-1. When compared with HS3+ve, an HS variant developed to enhance BMP-2 activity [Murali, S., et al., Affinity-selected heparan sulfate for bone repair. Biomaterials, 2013. 34(22): p. 5594-5605], HS16+ve was found to possess compositional differences, which altered its ability to potentiate BMP-2 activity compared to HSPM and HS3+ve.
[0374] In this work, the peptide used was 16 amino acids in length, while full length mature TGF-1 is 112 amino acids in length. Peptides in solution are known to adopt conformations different from those assumed when part of a full protein, and structure predictions of the TGF-1 peptide used for HS16+ve isolation, using PEPFOLD [Maupetit, J., P. Derreumaux, and P. Tuffery, PEP-FOLD: an online resource for de novo peptide structure prediction. Nucleic Acids Research, 2009. 37(suppl 2): p. W498-W503; Maupetit, J., P. Derreumaux, and P. Tuffery, A fast method for large-scale DeNovo peptide and miniprotein structure prediction. Journal of Computational Chemistry, 2010. 31(4): p. 726-738; and 216. Thvenet, P., et al., PEPFOLD: an updated de novo structure prediction server for both linear and disulfide bonded cyclic peptides. Nucleic Acids Research, 2012. 40(W1): p. W288-W293], do not match its native structure in TGF-1. This raises the question of the mechanism that drives our peptide-based affinity purification, as it is hard to conceive that a single stretch of peptide will be able to recreate the spatial organisation of the TGF-1 heparin-binding domain. One could argue that the interaction between the peptide and HS is primarily driven by ionic interactions, which is almost certainly true for peptide-heparin interactions (unpublished data from our group), but it does not appear to be the case here as the use of a peptide from a different protein (BMP-2) alters the profile of the isolated HS fractions.
[0375] The extensively investigated consensus sequence of basic residues involved in heparin-binding is proposed to adopt one of two motifs of basic residues: -X-B-B-BX-X-B-X-X- or -X-B-B-X-B-X-X- (where X is any neutral or acidic amino acid and B is a basic residue) [Cardin, A. D. and N. J. Weintraub, Molecular modeling of protein glycosaminoglycan interactions. Arteriosclerosis, Thrombosis, and Vascular Biology, 1989. 9(1): p. 21-32.]. A third motif has been proposed to exist in TGF-1: -X-BX-X-B-X-X-B-X-X-B-X- [McCaffrey, T. A., D. J. Falcone, and B. Du, Transforming growth factor-1 is a heparin-binding protein: Identification of putative heparin-binding regions and isolation of heparins with varying affinity for TGF-1. J Cell Physiol, 1992. 152(2): p. 430-440.]. Intriguingly, Pace and Scholtz [Nick Pace, C. and J. Martin Scholtz, A Helix Propensity Scale Based on Experimental Studies of Peptides and Proteins. Biophysical Journal, 1998. 75(1): p. 422-427] have reported that basic residues have a high propensity to form -helices in solution. Given the organisation of basic residues in these proposed motifs and that of the -helix (3.6 amino acid residues per turn), it would not be surprising to find that these motifs adopt a helical structure in solution with their basic residues arrayed along the same plane. If true, such organisation might confer some degree of selectivity to the peptides.
[0376] Sizing and compositional analysis of HS16+ve showed that it was enriched for longer polysaccharide chains, less heterogeneous, in terms of chain size distribution, and enriched for 6-O- and N-sulphated disaccharides relative to both HS16ve and HSPM. This corroborated our findings from our study of heparin-TGF-1 interactions, where we identified the need for heparin chains to be at least equivalent to a dp22 and possess 6-O- and N-sulphate groups in order to effectively bind to and modulate TGF-1 activity. Enrichment for longer chains of HS can be explained by the need for at least 22 saccharide units to bridge the two heparin/HS binding sites on the TGF-1 homodimer. Such chains would also have to satisfy the sulphate distribution criteria to effectively interact with TGF-1, which would further narrow the range of HS chains selected by our purification.
[0377] HS16+ve was able to potentiate TGF-1-driven SMAD signalling to a similar degree as heparin. Unexpectedly, its effect on LTGF-1 was more pronounced than that of heparin with LTGF-1. As LTGF-1 is the predominant form of TGF-1 in vivo, this raises interesting questions about the synthesis of, and physiological role played by HS in TGF-1 signalling. The latency-associated peptide (LAP) portion of LTGF has a structure considered to be stable, although two regions of the molecule can be unfolded [Shi, M., et al., Latent TGF- structure and activation. Nature, 2011. 474(7351): p. 343-349] in such a way that it traps TGF-1 in the LTGF-1 complex. When the conformations of these regions are mechanically forced fully open, active TGF-1 is released from the LAP [Buscemi, L., et al., The Single-Molecule Mechanics of the Latent TGF-1 Complex. Curr Biol, 2011. 21(24): p. 2046-2054]. This simultaneous unfolding of both domains, an all-or-nothing snap mechanism necessary for full TGF-1 release, is possible only when LAP is bound to LTGF-binding protein-1 (LTBP-1). It is interesting to note that both LAP and LTBP-1 have been reported to interact with heparin [Lee, M. J., Heparin Inhibits Activation of Latent Transforming Growth Factor-1. Pharmacology, 2013. 92(5-6): p. 238-244; Chen, Q., et al., Potential Role for Heparan Sulfate Proteoglycans in Regulation of Transforming Growth Factor- (TGF-) by Modulating Assembly of Latent TGF--binding Protein-1. J Biol Chem, 2007. 282(36): p. 26418-26430; and Parsi, M. K., et al., LTBP-2 has multiple heparin/heparan sulfate binding sites. Matrix Biology, 2010. 29(5): p. 393-401]. Whether and how HS is involved in either the generation or release of this mechanical force is a pertinent question, though the evidence suggests that the HS synthesised in vivo may be tuned to activate LTGF-1.
[0378] Comparison of the composition of HS16+ve with that of HS3+ve revealed differences in their compositions. The differences observed in the ability of either HS variant to bind to and modulate BMP-2 activity is probably a result of a combination of the differences in composition and consequently the disaccharide sequences embodied in the HS chains. Modelling of the BMP-2-heparin interaction [Gandhi, N. S. and R. L. Mancera, Prediction of heparin binding sites in bone morphogenetic proteins (BMPs). Biochimica et Biophysica Acta (BBA)Proteins and Proteomics, 2012. 1824(12): p. 1374-1381] suggests that the association varies significantly from that of heparin with TGF-1, which would explain the differences in the compositions of HS16+ve and HS3+ve, and hints at the diversification of heparin/HS binding sites within the TGF- superfamily [Rider, C. C., Heparin/heparan sulphate binding in the TGF- cytokine superfamily. Biochem Soc Trans, 2006. 34(Pt 3): p. 458-460]. The modest increase observed in the ability of HS16+ve to potentiate BMP-2 driven ALP expression relative to HS.sup.PM might be attributable to either the enrichment of sulphated HS chains in HS16+ve, making it more similar to heparin than HSPM, or an overlap in the composition of HS chains in HS16+ve and HS3+ve preparations.
[0379] While HS does not appear to be encoded with the same absolute sequence specificity seen in nucleic acids and proteins, our results provide compelling evidence for the current HS sugar code hypothesis [Gama, C. I., et al., Sulfation patterns of glycosaminoglycans encode molecular recognition and activity. Nat Chem Blot, 2006. 2(9): p. 467-473; Duchesne, L., et al., Transport of Fibroblast Growth Factor 2 in the Pericellular Matrix Is Controlled by the Spatial Distribution of Its Binding Sites in Heparan Sulfate. PLoS Blot, 2012. 10(7): p. e1001361; Chang, Z., et al., Differential ability of heparan sulfate proteoglycans to assemble the fibroblast growth factor receptor complex in situ. FASEB J, 2000. 14(1): p. 137-144; and Jastrebova, N., et al., Heparan sulfate domain organization and sulfation modulate FGF2 induced cell signaling. J Blot Chem, 2010] and reinforce the idea of a non-linear relationship between binding strength and bioactivity [Rudd, T. R., et al., Comparable stabilisation, structural changes and activities can be induced in FGF by a variety of HS and non-GAG analogues: implications for sequence-activity relationships. Org Biomol Chem, 2010. 8(23): p. 5390-5397.]. One of the major questions that remain to be answered is the level of stringency required for a given HS chain to interact with a given protein. It is hoped that the development of improved computational tools to decode the organisation of GAGs will aid in this endeavour [Spencer, J. L., et al., A Computational Approach for Deciphering the Organization of Glycosaminoglycans. PLoS ONE, 2010. 5(2): p. e9389].
REFERENCES FOR EXAMPLE 1
[0380] 1. Ori, A., Wilkinson, M. C., and Fernig, D. G. (2011) J Biol Chem [0381] 2. Lyon, M., Rushton, G., and Gallagher, J. T. (1997) J Biol Chem 272, 18000-18006 [0382] 3. McCaffrey, T. A., Falcone, D. J., Brayton, C. F., Agarwal, L. A., Welt, F. G., and Weksler, B. B. (1989) J Cell Biol 109, 441-448 [0383] 4. McCaffrey, T. A., Falcone, D. J., and Du, B. (1992) J Cell Physiol 152, 430-440 [0384] 5. McCaffrey, T. A., Falcone, D. J., Vicente, D., Du, B., Consigli, S., and Borth, W. (1994) J Cell Physiol 159, 51-59 [0385] 6. Lee, J.-H., Lee, H., Joung, Y. K., Jung, K. H., Choi, J.-H., Lee, D.-H., Park, K. D., and Hong, S.-S. (2011) Biomaterials 32, 1438-1445 [0386] 7. Watson, R. S., Gouze, E., Levings, P. P., Bush, M. L., Kay, J. D., Jorgensen, M. S., Dacanay, E. A., Reith, J. W., Wright, T. W., and Ghivizzani, S. C. (2010) Lab Invest 90, 1615-1627 [0387] 8. Siebert, N., Xu, W., Grambow, E., Zechner, D., and Vollmar, B. (2011) Lab Invest 91, 1753-1765 [0388] 9. Jakowlew, S. (2006) Cancer Metastasis Rev 25, 435-457 [0389] 10. Yang, Y.-a., Dukhanina, O., Tang, B., Mamura, M., Letterio, J. J., MacGregor, J., Patel, S. C., Khozin, S., Liu, Z.-y., Green, J., Anver, M. R., Merlino, G., and Wakefield, L. M. (2002) J Clin Invest 109, 1607-1615 [0390] 11. Grimaud, E., Heymann, D., and Rdini, F. (2002) Cytokine Growth Factor Rev 13, 241-257 [0391] 12. Rosen, F., McCabe, G., Quach, J., Solan, J., Terkeltaub, R., Seegmiller, J. E., and Lotz, M. (1997) Arthritis Rheum 40, 1275-1281 [0392] 13. Toh, W. S., Liu, H., Heng, B. C., Rufaihah, A. J., Ye, C. P., and Cao, T. (2005) Growth Factors 23, 313-321 [0393] 14. van der Kraan, P. M., Blaney Davidson, E. N., Blom, A., and van den Berg, W. B. (2009) Osteoarthritis Cartilage 17, 1539-1545 [0394] 15. Yang, X., Chen, L., Xu, X., Li, C., Huang, C., and Deng, C.-X. (2001) J Cell Biol 153, 35-46 [0395] 16. Miura, Y., Parvizi, J., Fitzsimmons, J. S., and O'Driscoll, S. W. (2002) J Bone Joint Surg Am 84, 793-799 [0396] 17. Sato, M., Ishihara, M., Ishihara, M., Kaneshiro, N., Mitani, G., Nagai, T., Kutsuna, T., Asazuma, T., Kikuchi, M., and Mochida, J. (2007) J Biomed Mater Res B Appl Biomater 83B, 181-188 [0397] 18. Wang, W., Li, B., Li, Y., Jiang, Y., Ouyang, H., and Gao, C. (2010) Biomaterials 31, 5953-5965 [0398] 19. Allen, J. B., Manthey, C. L., Hand, A. R., Ohura, K., Ellingsworth, L., and Wahl, S. M. (1990) J Exp Med 171, 231-247 [0399] 20. Leah, E. (2013) Nat Rev Rheumatol 9, 382-382 [0400] 21. Bakker, A. C., van de Loo, F. A. J., van Beuningen, H. M., Sime, P., van Lent, P. L. E. M., van der Kraan, P. M., Richards, C. D., and van den Berg, W. B. (2001) Osteoarthritis Cartilage 9, 128-136 [0401] 22. Blaney Davidson, E. N., Scharstuhl, A., Vitters, E. L., van der Kraan, P. M., and van den Berg, W. B. (2005) Arthritis Res Ther 7, R1338-R1347 [0402] 23. Kopesky, P. W., Vanderploeg, E. J., Kisiday, J. D., Frisbie, D. D., Sandy, J. D., and Grodzinsky, A. J. (2011) Tissue Eng Part A 17, 83-92 [0403] 24. Shah, R. N., Shah, N. A., Del Rosario Urn, M. M., Hsieh, C., Nuber, G., and Stupp, S. I. (2010) Proc Natl Acad Sci USA 107, 3293-3298 [0404] 25. Coupes, B. M., Williams, S., Roberts, I. S. D., Short, C. D., and Brenchley, P. E. C. (2001) Nephrol Dial Transplant 16, 361-367 [0405] 26. Bramono, D. S., Murali, S., Rai, B., Ling, L., Poh, W. T., Lim, Z. X., Stein, G. S., Nurcombe, V., van Wijnen, A. J., and Cool, S. M. (2012) Bone 50, 954-964 [0406] 27. Helledie, T., Dombrowski, C., Rai, B., Lim, Z. X. H., Hin, I. L. H., Rider, D. A., Stein, G. S., Hong, W., van Wijnen, A. J., Hui, J. H., Nurcombe, V., and Cool, S. M. (2012) Stem Cells Dev 21, 1897-1910 [0407] 28. Murali, S., Leong, D. F. M., Lee, J. J. L., Cool, S. M., and Nurcombe, V. (2011) J Biol Chem 286, 17755-17765 [0408] 29. Murali, S., Rai, B., Dombrowski, C., Lee, J. L. J., Lim, Z. X. H., Bramono, D. S., Ling, L., Bell, T., Hinkley, S., Nathan, S. S., Hui, J. H., Wong, H. K., Nurcombe, V., and Cool, S. M. (2013) Biomaterials 34, 5594-5605 [0409] 30. Samsonraj, R. M., Raghunath, M., Hui, J. H., Ling, L., Nurcombe, V., and Cool, S. M. (2013) Gene 519, 348-355 [0410] 31. Rider, D. A., Dombrowski, C., Sawyer, A. A., Ng, G. H. B., Leong, D., Hutmacher, D. W., Nurcombe, V., and Cool, S. M. (2008) Stem Cells 26, 1598-1608 [0411] 32. Zhang, L., Su, P., Xu, C., Yang, J., Yu, W., and Huang, D. (2010) Biotechnol Lett 32, 1339-1346 [0412] 33. Hernaiz, M., Liu, J., Rosenberg, R. D., and Linhardt, R. J. (2000) Biochem Biophys Res Commun 276, 292-297 [0413] 34. Uniewicz, K. A., Ori, A., Xu, R., Ahmed, Y., Wilkinson, M. C., Fernig, D. G., and Yates, E. A. (2010) Anal Chem 82, 3796-3802 [0414] 35. Xu, R., Ori, A., Rudd, T. R., Uniewicz, K. A., Ahmed, Y. A., Guimond, S. E., Skidmore, M. A., Siligardi, G., Yates, E. A., and Fernig, D. G. (2012) J Biol Chem 287, 40061-40073 [0415] 36. Livak, K. J., and Schmittgen, T. D. (2001) Methods 25, 402-408 [0416] 37. Ori, A., Free, P., County, J., Wilkinson, M. C., and Fernig, D. G. (2009) Mol Cell Proteomics 8, 2256-2265 [0417] 38. Guerrini, M., Naggi, A., Guglieri, S., Santarsiero, R., and Torn, G. (2005) Anal Biochem 337, 35-47 [0418] 39. Knobloch, J. E., and Shaklee, P. N. (1997) Anal Biochem 245, 231-241 [0419] 40. Brickman, Y. G., Ford, M. D., Gallagher, J. T., Nurcombe, V., Bartlett, P. F., and Turnbull, J. E. (1998) J Biol Chem 273, 4350-4359 [0420] 41. Skidmore, M. A., Guimond, S. E., Dumax-Vorzet, A. F., Yates, E. A., and Turnbull, J. E. (2010) Nat Protoco 5, 1983-1992 [0421] 42. Inman, G. J., Nicols, F. J., Callahan, J. F., Harling, J. D., Gaster, L. M., Reith, A. D., Laping, N. J., and Hill, C. S. (2002) Mol Pharmacol 62, 65-74 [0422] 43. Radaev, S., Zou, Z., Huang, T., Lafer, E. M., Hinck, A. P., and Sun, P. D. (2010) J Biol Chem 285, 14806-14814 [0423] 44. Gama, C. I., Tully, S. E., Sotogaku, N., Clark, P. M., Rawat, M., Vaidehi, N., Goddard, W. A., Nishi, A., and Hsieh-Wilson, L. C. (2006) Nat Chem Biol 2, 467-473 [0424] 45. Chang, Z., Meyer, K., Rapraeger, A. C., and Friedl, A. (2000) FASEB J 14, 137-144 [0425] 46. Duchesne, L., Octeau, V., Bearon, R. N., Beckett, A., Prior, I. A., Lounis, B., and Fernig, D. G. (2012) PLoS Biol 10, e1001361 [0426] 47. Jastrebova, N., Vanwildemeersch, M., Lindahl, U., and Spillmann, D. (2010) J Biol Chem [0427] 48. Rudd, T. R., Uniewicz, K. A., Ori, A., Guimond, S. E., Skidmore, M. A., Gaudesi, D., Xu, R., Turnbull, J. E., Guerrini, M., Torn, G., Siligardi, G., Wilkinson, M. C., Fernig, D. G., and Yates, E. A. (2010) Org Biomol Chem 8, 5390-5397 [0428] 49. Kim, M., Kim, S. E., Kang, S. S., Kim, Y. H., and Tae, G. (2011) Biomaterials 32, 7883-7896 [0429] 50. Park, J. S., Woo, D. G., Yang, H. N., Lim, H. J., Chung, H.-M., and Park, K.-H. (2008) Transplantation 85, 589-596 [0430] 51. Xu, X., Jha, A. K., Duncan, R. L., and Jia, X. (2011) Acta Biomater 7, 3050-3059 [0431] 52. Kelton, J. G., Arnold, D. M., and Bates, S. M. (2013) N Engl J Med 368, 737-744 [0432] 53. Lamanna, W. C., Kalus, I., Padva, M., Baldwin, R. J., Merry, C. L. R., and Dierks, T. (2007) J Biotechnol 129, 290-307 [0433] 54. Ori, A., Wilkinson, M. C., and Fernig, D. G. (2008) Front Biosci 13, 4309-4338 [0434] 55. Naimy, H., Leymarie, N., and Zaia, J. (2010) Biochemistry 49, 3743-3752 [0435] 56. Lee, M. J. (2013) Pharmacology 92, 238-244 [0436] 57. Khan, S., Gor, J., Mulloy, B., and Perkins, S. J. (2010) J Mol Biol 395, 504-521 [0437] 58. Buscemi, L., Ramonet, D., Klingberg, F., Formey, A., Smith-Clerc, J., Meister, J.-J., and Hinz, B. (2011) Curr Biol 21, 2046-2054 [0438] 59. Chen, Q., Sivakumar, P., Barley, C., Peters, D. M., Gomes, R. R., Farach-Carson, M. C., and Dallas, S. L. (2007) J Biol Chem 282, 26418-26430 [0439] 60. Shi, M., Zhu, J., Wang, R., Chen, X., Mi, L., Walz, T., and Springer, T. A. (2011) Nature 474, 343-349 [0440] 61. Annes, J. P., Munger, J. S., and Rifkin, D. B. (2003) Journal of Cell Science 116, 217-224 [0441] 62. Ramirez, F., and Rifkin, D. B. (2009) Current Opinion in Cell Biology 21, 616-622 [0442] 63. Spencer, J. L., Bernanke, J. A., Buczek-Thomas, J. A., and Nugent, M. A. (2010) PLoS ONE 5, e9389 [0443] 64. Chappell, E. P., and Liu, J. (2013) Bioorg Med Chem 21, 4786-4792
Example 2
Effects of HS16+ve on Chondrogenic Differentiation of MSCs In Vitro and In Vivo
Introduction
[0444] Articular cartilage is a tissue found at the ends of long bones that serves as both a shock absorber and lubricant in our joints. As a consequence of its avascular nature, injuries sustained by articular cartilage often fail to heal. Current research into the development of cartilage repair strategies has focused on stimulating the response from either endogenous or transplanted cells, through the use of activated biomaterials or provision of inductive cues to the cells [145. Guo, X., et al., Repair of osteochondral defects with biodegradable hydrogel composites encapsulating marrow mesenchymal stem cells in a rabbit model. Acta Biomaterialia, 2010. 6(1): p. 39-47; Chu, C. R., M. Szczodry, and S. Bruno, Animal Models for Cartilage Regeneration and Repair. Tissue Engineering Part B: Reviews, 2009. 16(1): p. 105-115; Fritz, J., et al., Articular cartilage defects in the kneebasics, therapies and results. Injury, 2008. 39(1, Supplement): p. 50-57; Hunziker, E. B., Articular cartilage repair: basic science and clinical progress. A review of the current status and prospects. Osteoarthritis Cartilage, 2002. 10(6): p. 432-463; Gille, J., et al., Cell-Laden and Cell-Free Matrix-Induced Chondrogenesis versus Microfracture for the Treatment of Articular Cartilage Defects: A Histological and Biomechanical Study in Sheep. Cartilage, 2010. 1(1): p. 29-42; Haleem, A. M., et al., The Clinical Use of Human Culture-Expanded Autologous Bone Marrow Mesenchymal Stem Cells Transplanted on Platelet-Rich Fibrin Glue in the Treatment of Articular Cartilage Defects. Cartilage, 2010. 1(4): p. 253-261; Moran, C. J., et al., Restoration of Articular Cartilage. J Bone Joint Surg Am, 2014. 96(4): p. 336-344; Dani{hacek over (s)}ovi{hacek over (c)}, Lu., et al., The tissue engineering of articular cartilage: cells, scaffolds and stimulating factors. Exp Biol Med, 2012. 237(1): p. 10-17]. TGF-1 has emerged as a key player for the induction of cartilage repair because of its ability to stimulate the chondrogenic differentiation of MSCs [Buxton, A. N., et al., Temporal exposure to chondrogenic factors modulates human mesenchymal stem cell chondrogenesis in hydrogels. Tissue Eng Part A, 2011. 17(3-4): p. 371-80; 233. Bosnakovski, D., et al., Chondrogenic differentiation of bovine bone marrow mesenchymal stem cells in pellet cultural system. Experimental Hematology, 2004. 32(5): p. 502-509; Ng, F., et al., PDGF, TGF-, and FGF signaling is important for differentiation and growth of mesenchymal stem cells (MSCs): transcriptional profiling can identify markers and signaling pathways important in differentiation of MSCs into adipogenic, chondrogenic, and osteogenic lineages. Blood, 2008. 112(2): p. 295-307], and drive the expression of cartilage ECM molecules [Li, H., et al., Comparative analysis with collagen type II distinguishes cartilage oligomeric matrix protein as a primary TGF-responsive gene. Osteoarthritis Cartilage, 2011. 19(10): p. 1246-1253; Iqbal, J., et al., Age-Related Effects of TGF- on Proteoglycan Synthesis in Equine Articular Cartilage. Biochem Biophys Res Commun, 2000. 274(2): p. 467-471; Grimaud, E., D. Heymann, and F. Rdini, Recent advances in TGF- effects on chondrocyte metabolism: Potential therapeutic roles of TGF- in cartilage disorders. Cytokine Growth Factor Rev, 2002. 13(3): p. 241-257; Serra, R., et al., Expression of a Truncated, Kinase-Defective TGF- Type II Receptor in Mouse Skeletal Tissue Promotes Terminal Chondrocyte Differentiation and Osteoarthritis. J Cell Biol, 1997. 139(2): p. 541-552; Blaney Davidson, E., et al., TGF beta-induced cartilage repair is maintained but fibrosis is blocked in the presence of Smad7. Arthritis Res Ther, 2006. 8(3): p. R65]. As such, most studies to date have employed the use of exogenous TGF-1, either alone or in combination with other growth factors, to drive hMSC chondrogenic differentiation [Blaney Davidson, E. N., P. M. van der Kraan, and W. B. van den Berg, TGF- and osteoarthritis. Osteoarthritis Cartilage, 2007. 15(6): p. 597-604; Park, J. S., et al., Heparin-Bound Transforming Growth Factor-3 Enhances Neocartilage Formation by Rabbit Mesenchymal Stem Cells. Transplantation, 2008. 85(4): p. 589-596; Goepfert, C., et al., Cartilage Engineering from Mesenchymal Stem Cells, in Bioreactor Systems for Tissue Engineering II, C. Kasper, M. van Griensven, and R. Prtner, Editors. 2010, Springer Berlin/Heidelberg. p. 163-200; Mara, C S., et al., Regulation of Chondrogenesis by Transforming Growth Factor-3 and Insulin-like Growth Factor-1 from Human Mesenchymal Umbilical Cord Blood Cells. J Rheumatol, 2010. 37(7): p. 1519-1526].
[0445] While appearing successful initially, such approaches face significant barriers in their translation into the clinic, as supraphysiological doses of TGF-1 are often employed, and even 20 ng doses have been shown to produce undesirable outcomes, such as synovial inflammation [Leah, E., Osteoarthritis: TGF- overload at bones of cartilage degeneration. Nat Rev Rheumatol, 2013. 9(7): p. 382-382; Allen, J. B., et al., Rapid onset synovial inflammation and hyperplasia induced by transforming growth factor beta. J Exp Med, 1990. 171(1): p. 231-247]. Apart from the problem of non-physiological doses, there is also the need to localise the growth factor to the site of treatment to prevent it from triggering systemic side effects, such as fibrosis and oncogenesis [Jakowlew, S., Transforming growth factor- in cancer and metastasis. Cancer Metastasis Rev, 2006. 25(3): p. 435-457; Bakker, A. C., et al., Overexpression of active TGF-beta-1 in the murine knee joint: evidence for synovial-layer-dependent chondro-osteophyte formation. Osteoarthritis Cartilage, 2001. 9(2): p. 128-136; Yang, Y.-a., et al., Lifetime exposure to a soluble TGF- antagonist protects mice against metastasis without adverse side effects. J Clin Invest, 2002. 109(12): p. 1607-1615.]. Additionally, sensitivity to TGF-1 decreases with age [Blaney Davidson, E. N., et al., Reduced transforming growth factor-beta signaling in cartilage of old mice: role in impaired repair capacity. Arthritis Res Ther, 2005. 7(6): p. R1338-R1347], so adequate TGF-1 dosing presents even more risk for aged patients. In response to these challenges, new strategies are being considered that reduce, or completely remove the need for exogenous growth factors; better localise and control the delivery of the growth factor at the site of treatment; and boost either cellular sensitivity to the growth factor or to the factors signalling efficiency.
[0446] Having described the development of HS16+ve, and its ability to potentiate the TGF-1-driven SMAD response in MSC monolayer cultures, we hypothesised that it could be used to improve cartilage healing by sequestering endogenous TGF-1. We thus sought to examine its effects on (1) the chondrogenic differentiation of hMSCs in vitro; and (2) the cartilage-healing response in a rabbit model.
[0447] This Example describes the use of a modified micromass pellet culture system for the examination of the in vitro effects of HS16+ve on the chondrogenic differentiation of hMSCs. It then moves on to examine the effects of the previously described HS variants, when used in conjunction with the current clinical standard of care, within a full depth, osteochondral defect of the trochlea groove in a rabbit model for cartilage repair.
Materials and Methods
[0448] Reverse Transcription and Quantitative PCR (qPCR)
[0449] Total RNA was isolated from chondrogenic micromass pellets using TRIZOL reagent (Life Technologies, CA, USA) according to the manufacturer's protocol. Reverse transcription was carried out on 1 g RNA using the SuperScript VILO cDNA Synthesis Kit (Life Technologies) following the manufacturer's instructions, with the incubation at 42 C. being carried out for 2 h instead of 1 h. Each qPCR contained 40 ng cDNA, 1 L TaqMan primer-probe mix per gene, and 10 L Taqman Fast Universal PCR Master Mix (Life Technologies) in a final volume of 20 L. Thermal cycling conditions were 95 C. for 20 s, followed by 45 cycles of 95 C. for 1 s and 60 C. for 20 s. Each qPCR was run in duplicate and gene expression was normalised to HPRT1 expression to obtain the Ct value. The average value of biological triplicates was taken. Chondrogenic micromass pellets cultured in media without GAG or TGF-1 were used as controls (Ct). Relative expression levels for each primer set were expressed as fold changes by the 2-Ct method [Livak, K. J. and T. D. Schmittgen, Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2-CT Method. Methods, 2001. 25(4): p. 402-408]. The following TaqMan primer-probe assays (Life technologies) were used:
In Vivo Study Design
[0450] Twenty-two skeletally mature, male New Zealand White rabbits (average age 9 months and body weight 3.9 kg) were used for this study. All rabbits received bilateral osteochondral defects in the femoral trochlea groove and each defect randomly assigned to one of four treatment groups: (1) Gel alone, (2) Gel+HSPM, (3) Gel+HS16+ve, and (4) Gel+HS16ve. Every defect received 60 L of a hyaluronic acid-based hydrogel (Gel) (AuxiGel, Termira AB, Stockholm, Sweden) [Bergman, K., et al., Injectable cell-free template for bone-tissue formation. Journal of Biomedical Materials Research Part A, 2009. 91A(4): p. 1111-1118.] alone or with 10 g of HSPM, HS16+ve or HS16ve. Two rabbits died from gastric stasis postsurgery and were not included in the analysis.
Defect Creation and Gel Injection
[0451] The research protocol used for this study was approved by the Institutional Animal Care and Use Committee, A*STAR Singapore, and followed all appropriate guidelines. All surgical procedures were carried out under general anaesthesia, consisting of a combination of ketamine (35 mg/kg) and xylazine (5 mg/kg) injections and isoflurane via a face mask, and aseptic conditions. A medial para-patellar skin incision of 15-20 mm was made and the patella dislocated laterally. One full thickness, critically-sized osteochondral defect (4 mm diameter, 2 mm depth) was made in the centre of each femoral trochlea groove with complete debridement of the calcified cartilage. Subsequently, 3 microfractures (0.8 mm diameter, 2 mm depth) were made in each defect using an orthopaedic drill and direct pressure applied with surgical gauze to ensure all bleeding had stopped prior to the application of the designated treatment. Treatments were applied with a 200 L pipette and allowed to set. All defects were observed to fill with blood while the gel carrier was setting.
[0452] Once the gel carrier had set, the patella was repositioned and the joint flexed 15 times to ensure the treatment remained in place before the incision was closed in layers, and rabbits allowed to weight-bear. The wound site was further sealed with Vetbond tissue adhesive (3M, MN, USA). Prophylactic antibiotics (Enrofloxacin, 10 mg/kg) and analgesics (Buprenorphin, 0.1 mg/kg) were administered subcutaneously for 5 days postsurgery. At 12 weeks all rabbits were euthanized with pentobarbital (150 mg/kg) after sedation. Distal femurs were harvested and imaged macroscopically before being processed for histological and immunohistochemistry (IHC) analysis.
Gross Pathologic Observation of Joints
[0453] Images of the joints were examined and scored by a blinded observer unaware of the treatment groups. Macroscopic scoring was based on the International Cartilage Repair Society (ICRS) Visual Assessment Scale (ICRS I scoring system) [Brittberg, M. and L. Peterson, Introduction of an articular cartilage classification. ICRS Newsletter, 1998. 1: p. 5-8.].
Histology Analyses
[0454] Harvested distal femoral heads were fixed in 10% (v/v) neutral-buffered formalin (NBF) for 1 week under vacuum and decalcified in 5% (v/v) formic acid at room temperature for an average of 6-7 days. The samples were subsequently embedded in paraffin wax and sectioned (5 m) across the middle of the defect. Sections were deparaffinised and stained with Masson's trichrome, Alcian blue (pH 1, counterstained with neutral red) and Safranin-O.
Immunohistochemistry (IHC) Analyses
[0455] IHC staining was carried out using either the Leica Bond-III or the Leica Bond-Max Autostainer (Leica Nussloch GmbH, Germany) and the Bond Refine Detection Kit (Leica). Sections were deparaffinised with Bond Dewax solution (Leica) and antigen retrieval carried out by incubating with Proteinase K (20 g/mL) (Sigma-Aldrich) for 15 min at room temperature. Endogenous peroxidase activity blocked by incubating with 3-4% (v/v) H.sub.2O.sub.2 for 15 min. Sections were then blocked in 10% (v/v) goat serum for 30 min before incubation with primary antibody (Collagen Type I (1:1000), Novus Biologicals, CO, USA and Collagen Type 11 (1:2000), Acris Antibodies, Inc., CA, USA) diluted in Bond Primary Antibody
[0456] Diluent (Leica) for 30 min at room temperature. Detection of staining was carried out as described in the Bond Refine Detection Kit and nuclei were counterstained with haematoxylin for 5 min. All washes were carried out with 1 Bond Wash Solution (Leica).
Histological Scoring
[0457] Examination and scoring of stained sections was carried out by a masked observer unaware of the treatment groups. Scoring was based on the O'Driscoll [O'Driscoll, S. W., F. W. Keeley, and R. B. Salter, The chondrogenic potential of free autogenous periosteal grafts for biological resurfacing of major full thickness defects in joint surfaces under the influence of continuous passive motion. An experimental investigation in the rabbit. J Bone Joint Surg Am, 1986. 68(7): p. 1017-1035] and ICRS II [Mainil-Varlet, P., et al., A New Histology Scoring System for the Assessment of the Quality of Human Cartilage Repair: ICRS II. The American Journal of Sports Medicine, 2010] scoring systems, and tissue filling was determined by quantifying the percentage of each tissue type (i.e. bony tissue, fibrous tissue, fibrocartilage, hybrid cartilage and hyaline cartilage) within the chondral and sub-chondral space.
Results
[0458] Effects of HS16+ve on Chondrogenic Differentiation of hMSCs In Vitro
[0459] In order to assess the effects of HS16+ve on the chondrogenic differentiation of hMSCs in vitro, it was first necessary to establish the effects of TGF-1 alone. Chondrogenic differentiation was carried out using a modified micromass culture system. [Zhang, L., et al., Chondrogenic differentiation of human mesenchymal stem cells: a comparison between micromass and pellet culture systems. Biotechnol Lett, 2010. 32(9): p. 1339-1346.]. Briefly, passage 4 hMSCs were harvested and resuspended in chemically defined chondrogenic media (PT-3003, Lonza, MD, USA) at 210.sup.7 cells/mL. Droplets of 12.5 L were then seeded into the middle of each well in a 24-well plate and left to adhere at 37 C. for 2 h, after which, 500 L of chondrogenic media supplemented with either 1 or 10 ng/mL of TGF-1 (100-210, PeproTech) alone or with either 5 or 10 g/mL of heparin (Sigma-Aldrich), porcine mucosal HS (HSPM) (Celsus laboratories), HS16+ve or HS16-ve was added to each well. The cell droplets coalesced into spherical masses after 24 h. Media was changed every 3 days and the micromasses harvested on days 3, 7, 14 and 21.
[0460] Wet weights of the resulting control (Ctrl) and TGF-1-treated (10 ng/mL) (TGF-1) micromass pellets were taken at days 3, 7, 14 and 21 (
[0461] Gene expression analysis of the micromass pellets at days 3 (SOX9 and COMP only), 7, 14 and 21 revealed that treatment with 10 ng/mL TGF-1 (TGF-1) consistently led to the increased expression of the chondrogenic markers SOX9 (
[0462] Histological examination of the paraffin-embedded pellets at day 21 by Alcian blue staining, which stains GAGs blue, revealed that TGF-1-treated pellets contained more GAGs then control pellets.
[0463] Having established the effects of TGF-1 on the chondrogenic differentiation of hMSCs, we next sought to examine the effects that heparin has, before embarking on the investigation of the effects of HS16+ve. In Example 1 we showed that heparin was able to potentiate the TGF-1-driven pSMAD signal in hMSCs at 6 h post treatment. Since medium was changed every third day during the chondrogenic differentiation process, we chose to use a day 3 time point to examine the effects of heparin on the expression of SOX9 and COMP, both early chondrogenic markers [Barry, F., et al., Chondrogenic Differentiation of Mesenchymal Stem Cells from Bone Marrow: Differentiation-Dependent Gene Expression of Matrix Components. Exp Cell Res, 2001. 268(2): p. 189-200; Li, H., et al., Comparative analysis with collagen type II distinguishes cartilage oligomeric matrix protein as a primary TGF-responsive gene. Osteoarthritis Cartilage, 2011. 19(10): p. 1246-1253; Huang, A. H., A. Stein, and R. L. Mauck, Evaluation of the Complex Transcriptional Topography of Mesenchymal Stem Cell Chondrogenesis for Cartilage Tissue Engineering. Tissue Eng Part A, 2010. 16(9): p. 2699-708; Zaucke, F., et al., Cartilage oligomeric matrix protein (COMP) and collagen IX are sensitive markers for the differentiation state of articular primary chondrocytes. Biochem J, 2001. 358(1): p. 17-24.]. We also reasoned that if heparin were to potentiate the effects of the TGF-1, further increases in the response to the recommended dose of 10 ng/mL might not be detectable. Therefore, a lower dose of TGF-1 was used in parallel with the recommended dose. Our data show that 5 g/mL of heparin did not significantly alter the level of SOX9 expression regardless of the amount of TGF-1 used (
[0464] A dose of 10 g/mL of GAG was selected to be used in conjunction with a 1 ng/mL dose of TGF-1. Histological analysis of pellets cultured with either 1 ng/mL TGF-1, 1 ng/mL TGF-1 and 10 g/mL heparin or 10 ng/mL TGF-1 for 21 days showed that the higher dose of TGF-1 lead to an increase in GAG production and deposition, based on Alcian blue staining. The use of heparin with TGF-1 led to a slight increase in GAG deposition relative to 1 ng/mL of TGF-1 alone, but this increase was still less than that seen with 10 ng/mL of TGF-1.
[0465] Next, hMSCs were differentiated for 21 days in the presence of 1 ng/mL TGF-1 alone (1 TGF-1) or in combination with 10 g/mL of GAG (heparin, HSPM, HS16+ve or HS16ve) or with 10 ng/mL TGF-1 (10 TGF-1) as a positive control.
[0466] Analysis of SOX9 mRNA expression at 21 days showed that all the sugars used did not produce significant changes (
[0467] Aggrecan expression was similar with low and high doses of TGF-1 and heparin and HS16+ve at day 21 (
[0468] Histological examination of the pellets by Alcian blue staining at day 21 did not indicate significant differences between the pellets cultured with the various sugars, relative to low TGF-1. High TGF-1 did however induce a modest increase in GAG production compared to low TGF-1.
Effects of HS16 on Chondrogenic Differentiation of MSCs In Vivo
[0469] Skeletally mature adult New Zealand white rabbits were chosen for our in vivo trial based both on their extensive use in cartilage repair studies, and to avoid the spontaneous healing observed in juveniles. We opted to trial our compound in a model comprising a full depth osteochondral defect in the femoral trochlea groove, where microfracture is used with a commercially available hyaluronic acid-based hydrogel (AuxiGel, Termira AB) [Bergman, K., et al., Injectable cell-free template for bone-tissue formation. Journal of Biomedical Materials Research Part A, 2009. 91A(4): p. 1111-1118], based on guidelines outlined by Reinholz et al. [Reinholz, G. G., et al., Animal models for cartilage reconstruction. Biomaterials, 2004. 25(9): p. 1511-1521], the ICRS [Hurtig, M. B., et al., Preclinical Studies for Cartilage Repair: Recommendations from the International Cartilage Repair Society. Cartilage, 2011. 2(2): p. 137-152] and current standard-of-care practices in hospitals [Fritz, J., et al., Articular cartilage defects in the knee-basics, therapies and results. Injury, 2008. 39(1, Supplement): p. 50-57; Hunziker, E. B., Articular cartilage repair: basic science and clinical progress. A review of the current status and prospects. Osteoarthritis Cartilage, 2002. 10(6): p. 432-463].
[0470] Sequence alignment of rabbit and human TGF-1 revealed that TGF-1 was not only highly conserved across both species, but the identified heparin-binding domain was nearly identical (
[0471] A 12-week study was performed comparing the following groups: (1) control group treated with 60 L of hydrogel (Gel alone) per defect: (2) defects treated with 60 L of hydrogel and 10 g of HSPM (HSPM); (3) defects treated with 60 L hydrogel and 10 g of HS16+ve (HS16+ve), and (4) defects treated with 60 L hydrogel and 10 g of HS16ve (HS16ve). Based on our earlier work, 10 g/mL of GAG was determined to be optimal for enhancing the effects of 1 ng/mL of TGF-1. As such, it was decided that a dose of 10 g of GAG would be sufficient to achieve this optimal concentration within the defect, even after accounting for possible diffusion within the synovial cavity. Defects were created as described above. Two rabbits died of gastric stasis before the end of the trial and were therefore excluded from the analysis.
[0472] At the end of the trial, whole femurs were harvested from the rabbits, fixed in 10% (v/v) NBF and imaged macroscopically before being decalcified and processed for histology. Macroscopic observation of defects after 12 weeks revealed a slight difference between the control (Gel alone) and treatment groups with regard to tissue filling. While there was an equal amount of variation in tissue filling within each treatment group, more defects in the control group were incompletely filled relative to those in the other groups. The median scores of the groups treated with HSPM, HS16+ve and HS16ve were higher than those of Gel alone (
[0473] Histological scores from the O'Driscoll and ICRS II scoring systems showed no significant differences between the treatment groups. Tissue filling of the defect was determined by first identifying the borders of the chondral and sub-chondral spaces, superimposing a grid over the imaged sections and then measuring the amount of space filled for each histological space. In terms of tissue filling, nearly all the samples exhibited complete subchondral filling and high levels of chondral filling. There were no statistical differences between the tissue filling scores for all the treatment groups. The median percentage of sub-chondral filling in all samples was similar. All three compounds (HSPM, HS16ve and HS16+ve) had median tissue filling scores that were higher than control samples (Gel), which is the current clinical standard-of-care treatment.
[0474] The in vitro data indicate that HS16+ve was able to enhance the TGF-1-induced expression of a number of chondrogenic markers, relative to HSPM and HS16ve. This suggests that it may be necessary to pre-load, e.g. coat or impregnate, gel-constructs with HS and TGF1 prior to implantation.
[0475] The in vivo data show that treatment of full depth osteochondral defects with a single dose of sugar, in conjunction with microfracture and hydrogel implantation, is at least as good as the current standard-of-care treatment, and does not produce undesired side effects. HS16+ve had median scores in the in vivo data that were higher than control samples (Gel), which is the current clinical standard-of-care treatment.