Protein delivery to membranes

Abstract

There is provided a phospholipid composition which is a bilayer or micelle comprising at least one embedded protein-polymer surfactant conjugate comprising an anchor protein, wherein the anchor protein is a cationised protein or an anionised protein, the composition characterised in that the anchor protein is: a) an active enzyme; or b) is a protein which does not comprise a —CH.sub.2C(O)NCH.sub.3(CH.sub.2).sub.3NCH.sub.3).sub.2H.sup.+ linker covalently bonded to an amino acid side chain.

Claims

1. A phospholipid bilayer or micelle comprising at least one embedded protein-polymer surfactant conjugate comprising an anchor protein conjugated to a surfactant, wherein the anchor protein is a cationised protein or an anionised protein, wherein the at least one protein-polymer surfactant conjugate is embedded within the phospholipid bilayer or micelle; and wherein the anchor protein is a protein which does not comprise a —CH.sub.2C(O)NCH.sub.3(CH.sub.2).sub.3N(CH.sub.3).sub.2H.sup.+ linker covalently bonded to an amino acid side chain.

2. The phospholipid bilayer or micelle according to claim 1, wherein the protein-polymer surfactant conjugate comprises a surfactant containing polyethylene glycol.

3. The phospholipid bilayer or micelle of claim 1, wherein the anchor protein is linked to a secondary molecule which is CshA, a portion of CshA comprising the fibronectin-binding portion of CshA (SEQ ID NO: 19) or is a functional variant having at least 90% sequence identity to SEQ ID NO:19.

4. The phospholipid bilayer or micelle according to claim 1, wherein the anchor protein is an enzyme.

5. A cell comprising a phospholipid bilayer according to claim 1 in its cell membrane.

6. The cell of claim 5, wherein the cell is a mesenchymal stem cell or a cardiomyocyte.

7. A pharmaceutical composition comprising the phospholipid bilayer or micelle according to claim 1, and further comprising a pharmaceutically acceptable carrier, diluent or vehicle.

8. A surgical composition comprising the phospholipid bilayer or micelle according to claim 1 and at least one surgically acceptable carrier, diluent or vehicle.

9. A tissue engineering scaffold comprising the phospholipid bilayer or micelle according to claim 1.

10. A polypeptide comprising: a) SEQ ID NOS: 11 or 40 or a functional variant having at least 97% sequence identity with SEQ ID NOS: 11 or 40; b) a fusion protein of a supercharged GFP and OpdA (SEQ ID NO: 13) or a functional variant having at least 60% sequence identity with SEQ ID NO: 13; c) a fusion protein of a supercharged GFP and SEQ ID NO: 14 or a functional variant having at least 80% sequence identity with SEQ ID NO: 14; d) a fusion protein of a supercharged GFP and P1GF.sub.(123-144) (SEQ ID NO: 15) or a functional variant having at least 93% sequence identity with SEQ ID NO: 15; or e) a fusion protein of a supercharged GFP and any of SEQ ID NOs: 19-39; or a supercharged GFP and a functional variant of any of SEQ ID NOs: 19-39 having at least 60% sequence identity thereto.

11. A polynucleotide encoding a polypeptide according to claim 10.

12. The polynucleotide of claim 11, wherein the polynucleotide comprises any of SEQ ID NOS: 2 or 4-7.

13. A method of making the phospholipid bilayer or micelle according to claim 1, comprising a) providing a protein-polymer surfactant conjugate comprising an anchor protein conjugated to a surfactant; and b) contacting a phospholipid bilayer or micelle with the protein-polymer surfactant conjugate to embed the protein-polymer surfactant conjugate within the phospholipid bilayer or micelle, wherein the anchor protein is a cationised protein or an anionised protein and is a protein which does not comprise a —CH.sub.2C(O)NCH.sub.3(CH.sub.2).sub.3N(CH.sub.3).sub.2H.sup.+ linker covalently bonded to an amino acid side chain.

14. The method of claim 13 wherein the anchor protein is a supercharged protein obtained by a method comprising expression of a recombinant DNA sequence encoding the supercharged protein.

15. The method of claim 14 wherein the recombinant DNA sequence further encodes a secondary molecule, such that the recombinant DNA sequence encodes a fusion protein comprising the supercharged protein and the secondary molecule, and wherein the secondary molecule comprises one or more of CshA, a portion of CshA comprising the fibronectin-binging portion of CshA (SEQ ID NO: 19), a functional variant having at least 90% sequence identity to SEQ ID NO: 19, OpdA (SEQ ID NO: 10 or SEQ ID NO: 39), thrombin, prothrombin, P1GF-2 (SEQ ID NO: 22), a portion of PIGF-2.sub.(123-144) (SEQ ID NO: 21), a functional variant of P1GF-2 having at least 90% sequence identity to SEQ ID NO: 21, a SpyCatcher polypeptide (SEQ ID NO: 23) or a SpyTag polypeptide (SEQ ID NO: 24), or comprises a functional variant of any of these having at least about 60% sequence identity to the non-variant sequence (SEQ ID NOS: 10, 19, 21-24, or 39).

16. The method of claim 13, wherein the anchor protein is an enzyme.

17. A method of labelling a cell with a protein label, comprising contacting a cell with a phospholipid bilayer produced by a method comprising: a) providing a protein-polymer surfactant conjugate comprising an anchor protein conjugated to the surfactant; and b) contacting a phospholipid bilayer with the protein-polymer surfactant conjugate to embed the protein-polymer surfactant conjugate within the phospholipid bilayer, wherein the anchor protein is a cationised protein or an anionised protein and is a protein which does not comprise a —CH.sub.2C(O)NCH.sub.3(CH.sub.2).sub.3N(CH.sub.3).sub.2H.sup.+ linker covalently bonded to an amino acid side chain, and wherein the phospholipid bilayer forms the external membrane of the cell and the protein-polymer surfactant conjugate comprises the protein label.

18. The method of claim 17, wherein the anchor protein is an enzyme.

19. A method of promoting tissue growth and/or healing in a subject in need thereof, comprising introducing the cell of claim 5 to a site where tissue is desired to grow and/or heal in the subject, wherein the anchor protein is a protein known to promote growth and/or healing of the tissue, or wherein the protein-polymer surfactant conjugate comprises a secondary molecule which is known to promote growth and/or healing of the tissue.

20. A method of targeting a cell to a tissue in a subject in need thereof, comprising contacting a tissue with the cell of claim 5, wherein the protein-polymer surfactant conjugate comprises a protein that specifically targets the tissue.

21. A method of treating myocardial infarction, cardiomyopathy and/or myocarditis in a human or animal subject in need thereof, comprising administering the cell of claim 5 to the human or animal subject, wherein the protein-polymer surfactant conjugate comprises a protein that specifically targets cardiac tissue.

22. A method of delivering a protein to the interior of a cell, comprising contacting the cell with a phospholipid bilayer or micelle comprising at least one embedded protein-polymer surfactant conjugate comprising an anchor protein conjugated to a surfactant, wherein the anchor protein is a cationised protein or an anionised protein, wherein the at least one protein-polymer surfactant conjugate is embedded within the phospholipid bilayer or micelle, and wherein the anchor protein is a protein which does not comprise a —CH.sub.2C(O)NCH.sub.3(CH.sub.2).sub.3N(CH.sub.3).sub.2H.sup.+ linker covalently bonded to an amino acid side chain, and wherein the protein-polymer surfactant conjugate comprises a molecule which promotes or inhibits the speed/rate of endocytosis.

23. The method of claim 22, wherein the anchor protein is an enzyme.

24. A method of decontaminating a sample comprising a poison, comprising contacting the sample with a phospholipid bilayer or micelle comprising at least one embedded protein-polymer surfactant conjugate comprising an anchor protein conjugated to a surfactant, wherein the anchor protein is a cationised or an anionised protein, wherein the at least one protein-polymer surfactant conjugate is embedded within the phospholipid bilayer or micelle, wherein the anchor protein is a protein which does not comprise a —CH.sub.2C(O)NCH.sub.3(CH.sub.2).sub.3N(CH.sub.3).sub.2H.sup.+ linker covalently bonded to an amino acid side chain, and wherein the anchor protein is an enzyme which can neutralise the poison or is linked to a secondary molecule which can bind to or neutralise the poison.

25. The method of claim 24 wherein the sample is a surface, land, a soil sample, or a fabric, wherein the anchor protein is an enzyme which can neutralize the poison, and wherein the enzyme is OpdA or a functional variant or portion thereof capable of degrading an organophosphorus compound.

26. The method of claim 24, wherein the sample is a surface, land, a soil sample, or a fabric, wherein the anchor protein is linked to a secondary molecule which can bind to or neutralise the poison, and wherein the secondary molecule is OpdA SEQ ID NO: 39) or a functional variant or portion thereof capable of degrading an organophosphorus compound.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

(2) FIG. 1 shows the reaction pathway for the covalent alteration of glutamic acid and aspartic acid residues via a nucleophilic addition-elimination mechanism;

(3) FIG. 2 is a diagram showing a protein-polymer surfactant conjugate (or “protein-surfactant bioconjugate”) inserted into a phospholipid bilayer;

(4) FIG. 3 is a diagram showing a protein-polymer surfactant conjugate comprising an anchor protein and a secondary molecule, with the secondary molecule positioned outside the membrane;

(5) FIG. 4 shows the chemical structure of various surfactants used in the work described herein (for S621 and S907 x=11-13, for S621 y=7-9, for S907 y=14-15);

(6) FIG. 5 shows the mass spectrum of scGFP-OpdA;

(7) FIG. 6 shows (a) the UV-vis absorbance of [scGFP-CshA] and [scGFP-CshA][S] and (b) the fluorescence excitation and emission of [scGFP-CshA] and [scGFP-CshA][S];

(8) FIG. 7 shows UV-vis absorbance of scGFP-OpdA;

(9) FIG. 8 shows excitation and emission properties of scGFP-OpdA;

(10) FIG. 9 shows mass spectra of thrombin and cationised thrombin at a charge number of 3;

(11) FIG. 10 shows the zeta potentiometries of OpdA, cOpdA and [cOpdA][S];

(12) FIG. 11 shows the change in zeta potential over time during cationisation of thrombin;

(13) FIG. 12 shows (a) the secondary structure composition of OpdA, cOpdA, and [cOpdA][S], and (b) the thermal denaturation of OpdA, cOpdA, and [cOpdA][S];

(14) FIG. 13 shows the radial probability distribution of OpdA, cOpdA, and [cOpdA][S];

(15) FIG. 14 shows (a) the circular dichroism of scGFP-CshA and (b) [scGFP-CshA][S] at 30, 60 and 90° C.;

(16) FIG. 15 shows the number of scGFP-CshA and [scGFP-CshA][S] molecules bound per hMSC;

(17) FIG. 16 shows the fluorescence of the primed hMSCs;

(18) FIG. 17 shows the cytotoxicity of scGFP-CshA and [scGFP-CshA][S];

(19) FIG. 18 shows the proliferation of hMSCs primed with [scGFP_CshA] and [scGFP_CshA][S];

(20) FIG. 19 shows the Michaelis-Menten parameters of OpdA, cOpdA, and [cOpdA][S];

(21) FIG. 20 shows the change in enzymatic activity over time during cationisation of thrombin;

(22) FIG. 21 shows a confocal micrograph of [Thrombin][S]-catalysed fibrin gel;

(23) FIG. 22 shows the cell adherence of unlabelled hMSCs, scGFP-CshA primed hMSCs and [scGFP-CshA][S] primed hMSCs on BSA-coated and Fn-coated plates; and

(24) FIG. 23 shows scanning electron microscopy images of explanted bovine articular cartilage with hMSCs and scGFP-PIGF.sub.(123-144)-primed hMSCs.

EXAMPLES

(25) General Methods

(26) Plasmid Preparation

(27) The OpdA gene was acquired in a pETMCSI vector, and required no further processing. The scOpdA gene was inserted into a pETMCSI vector via Gibson assembly, as described previously in Gibson et al. (Nat Methods. (2009) Apr. 12; 6(5):343-5). scGFP-CshA, scGFP-SpyCatcher, mCherry-SpyTag, CshA-SpyTag, scGFP-OpdA, and scGFP-PIGF genes were inserted into pOPINF vectors via the In-Fusion™ cloning system, according to manufacturer's instructions. The vectors were amplified via transformation into Stellar cells (Clontech, US) or Top10 cells (Thermo Fisher Scientific, US), followed by miniprep purification (Qiagen, Germany), each according to the manufacturer's instructions. DNA and amino acid sequences are listed in Table 5:

(28) TABLE-US-00005 TABLE 5 Protein gene and amino acid sequences utilised herein Protein/construct as Amino acid referred to herein Gene sequence sequence OpdA SEQ ID NO: 1 SEQ ID NO: 10; SEQ ID NO: 39 scOpdA SEQ ID NO: 2 SEQ ID NO: 11; SEQ ID NO: 40 scGFP SEQ ID NO: 3 SEQ ID NO: 12 scGFP-OpdA SEQ ID NO: 4 SEQ ID NO: 13 scGFP-CshA SEQ ID NO: 5 SEQ ID NO: 14 scGFP-PIGF.sub.(123-144) SEQ ID NO: 6 SEQ ID NO: 15 scGFP-SpyCatcher SEQ ID NO: 7 SEQ ID NO: 16 mCherry-SpyTag SEQ ID NO: 8 SEQ ID NO: 17 CshA-SpyTag SEQ ID NO: 9 SEQ ID NO: 18 Fibronectin binding domain — SEQ ID NO: 19 of CshA CshA — SEQ ID NO: 20 PIGF-2.sub.(123-144) — SEQ ID NO: 21 PIGF-2 — SEQ ID NO: 22 SpyCatcher — SEQ ID NO: 23 SpyTag — SEQ ID NO: 24 bovine prothrombin — SEQ ID NO: 25 human prothrombin — SEQ ID NO: 26

(29) Protein Expression

(30) OpdA, scOpdA, scGFP-CshA, scGFP-OpdA, and scGFP-PIGF.sub.(123-144) were obtained by expression in BL21(DE3) cells (New England Biolabs, USA), transformed with vectors containing their respective genes, using routine methods. Protein specific parameters are outlined in Table 6. Bovine thrombin and human fibrinogen were obtained from commercial sources (Sigma, Cat. No T7326 and F3879, respectively).

(31) TABLE-US-00006 TABLE 6 Protein expression parameters Protein Medium Temperature Induction OpdA Terrific broth, with 30° C. None 100 μM CoCl.sub.2 scOpdA Terrific broth, with 30° C. 1 mM IPTG when 100 μM CoCl.sub.2 Abs.sub.600 ≥ 0.6 scGFP-OpdA Terrific broth, with 37° C. 1 mM IPTG when 100 μM CoCl.sub.2, 10 Abs.sub.600 ≥ 0.6 g/L NaCl scGFP-CshA Terrific broth 37° C. 1 mM IPTG when Abs.sub.600 ≥ 0.6 scGFP-PIGF.sub.(123-144) Lysogeny broth 37° C. 1 mM IPTG when Abs.sub.600 ≥ 0.6 scGFP-SpyCatcher Terrific broth 37° C. 1 mM IPTG when Abs.sub.600 ≥ 0.6 mCherry-SpyTag Terrific broth 37° C. 1 mM IPTG when Abs.sub.600 ≥ 0.6 CshA-SpyTag Terrific broth 37° C. 1 mM IPTG when Abs.sub.600 ≥ 0.6

(32) Protein Purification

(33) Lysis buffer was added to cell pellets and lysed using pulse sonication, using routine methods. The protein was then purified using fast protein liquid chromatography (FPLC). Proteins were further purified using size exclusion chromatography, using routine methods.

(34) Protein specific purification steps are outlined in Table 7. No purification was required for the commercially purchased thrombin or fibrinogen.

(35) TABLE-US-00007 TABLE 7 Protein purification conditions Protein Method Lysis buffer Elution buffer OpdA Anion exchange 30 mM HEPES, 100 N/A (DEAE column) μM CoCl.sub.2, pH 8 scOpdA IMAC (Ni-NTA 30 mM HEPES, 1.5M 30 mM HEPES, 1.5M column) NaCl, 20 mM NaCl, 1M imidazole, imidazole, pH 8 pH 8 scGFP-OpdA IMAC (Ni-NTA 20 mM Sodium 20 mM Sodium column) phosphate, 1M NaCl, phosphate, 1M NaCl, 2 nM MgCl.sub.2, 50 mM 2 nM MgCl.sub.2, 500 mM imidazole, pH 8 imidazole, pH 8 scGFP-CshA IMAC (Ni-NTA 20 mM Tris-HCl, 1M 20 mM Tris-HCl, 1M column) NaCl, 20 mM NaCl, 500 mM imidazole, pH 7.5 imidazole, pH 7.5 scGFP-PIGF.sub.(123-144) IMAC (Ni-NTA 20 mM Tris-HCl, 1M 20 mM Tris-HCl, 1M column) NaCl, 20 mM NaCl, 500 mM imidazole, pH 7.5 imidazole, pH 7.5 scGFP-SpyCatcher IMAC (Ni-NTA 20 mM Tris-HCl, 1M 20 mM Tris-HCl, 1M column) NaCl, 20 mM NaCl, 500 mM imidazole, pH 7.5 imidazole, pH 7.5 mCherry-SpyTag IMAC (Ni-NTA 20 mM Tris-HCl, 1M 20 mM Tris-HCl, 1M column) NaCl, 20 mM NaCl, 500 mM imidazole, pH 7.5 imidazole, pH 7.5 CshA-SpyTag IMAC (Ni-NTA 20 mM Tris-HCl, 1M 20 mM Tris-HCl, 1M column) NaCl, 20 mM NaCl, 500 mM imidazole, pH 7.5 imidazole, pH 7.5

Synthesis of Glycolic Acid Ethoxylate 4-nonylphenyl ether (Oxidised IGEPAL-00890) Surfactant

(36) Surfactant was prepared as described in Armstrong et al. (Nat. Commun. (2015) Jun. 17; 6:7405). Briefly, 2 g IGEPAL CO-890 dissolved in 50 mL deionised-water was mixed with 30 mg 2,2,6,6,-tetramethyl-1-piperidinyloxyl (TEMPO), 50 mg NaBr, and 5 mL NaClO solution containing 10-15% available chlorine. The solution was periodically adjusted to pH 11 and stirred for 24 hours. The reaction was quenched with ethanol and adjusted to pH 1. Solvent extraction was performed with 3 washes of 80 mL aliquots of chloroform, then 3 washes with 80 mL aliquots of deionised water adjusted to pH 1. The resulting solution was dried under reduced pressure at 40° C. The remaining solid was redissolved in 40 mL ethanol, recrystallised at −20° C., the ethanol decanted, and the crystals dried under reduced pressure at 65° C.

(37) Protein-Surfactant Conjugation

(38) To form the conjugated constructs, glycolic acid ethoxylate 4-nonylphenyl ether was added to a solution of cationised protein or protein comprising supercharged GFP or OpdA (see below). Any excess surfactant may be removed via dialysis, using 14,000 MWCO tubing.

(39) The specific parameters are presented in Table 8.

(40) TABLE-US-00008 TABLE 8 Conjugation parameters Moles of Surfactant surfactant per Protein form cationic site Buffer Time Temperature cationised 10 mg/mL 1 30 mM HEPES, 100 1 hour 4° C. OpdA solution μM CoCl.sub.2, pH 8 cationised Solid 1.4 60 mM HEPES, pH 7 1 hour Room Thrombin temperature scGFP-CshA/ 25 mg/mL 1.4 20 mM Tris-HCl, pH Overnight 4° C. OpdA/ solution 7.5 PIGF.sub.(123-144)

(41) Mass Spectrometry

(42) Mass spectrometry was performed using a Bruker ultrafleXtreme MALDI-TOF/TOF mass spectrometer in linear positive mode. The matrix was a saturated solution of either sinapinic acid or α-hydroxycinnamic acid in a mixture of equal volumes acetonitrile and water, with a final concentration of 0.1% trifluoroacetic acid. 0.5 μL of 1:1 sample and matrix mixture was spotted on a ground steel plate for analysis.

(43) Dynamic Light Scattering and Zeta Potentiometry

(44) Dynamic light scattering (DLS) and zeta potentiometry analyses were performed on a Zetasizer Nano SP (Malvern Instruments, UK), and the data analysed using Zetasizer software (Malvern Instruments).

(45) Small Angle X-ray Scattering

(46) Small angle X-ray scattering was performed on the B21 beamline at the Diamond Light Source, Oxford. Samples were concentrated with 10,000 MWCO spin concentrators and flow-through retained for use as backgrounds. The samples were then spun through 1,000,000 MWCO spin concentrators to remove large contaminants. Samples were exposed for 18 frames of 10 seconds each. Data analyses were performed with the ScÅtter software package, using ATSAS plugins.

(47) Cell Culture

(48) Human mesenchymal stem cells (hMSCs) were harvested from the proximal femur bone marrow of osteoarthritic patients undergoing total hip replacement surgery, in full accordance with Bristol Southmead Hospital Research Ethics Committee guidelines (reference #078/01), and having received informed consent from all patients. Cells were cultured at 5% CO2, using low-glucose DMEM, supplemented with 10% fetal bovine serum, 2 mM GlutaMAX (Gibco, US), 100 μg/mL penicillin-streptomycin and 5 ng/mL freshly supplemented basic human Fibroblast Growth Factors (FGF) (Peprotech, USA).

(49) Cell Priming

(50) Cells were washed with PBS, and suspended with trypsin-EDTA solution (Sigma, UK). The protein solution was added to the suspended cells in phenol free DMEM, and left to shake and incubate at 37° C. for 15 minutes. The cells were then centrifuged at 500 g for 5 minutes, and the supernatant discarded. The cells were then resuspended for immediate use or to be plated.

(51) Alternatively, a protein solution was applied directly to plated cells. The cells were washed with PBS, and the protein solution added in an appropriate buffer for up to 30 minutes with shaking at 37° C. The cells were then washed with PBS again, and ready for use.

(52) Cell Cytotoxicity Assays

(53) The cytotoxicity of the constructs was assayed using either (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) (MTS) or alamarblue® according to the manufacturers' instructions. Briefly, hMSCs were plated in 96-well plates at a range of concentrations to produce a standard curve. A known quantity of cells was primed with a solution of the construct for 15 minutes then washed with PBS before incubation with either MTS or alamarBlue solution for 1-2 hours. Absorbance or fluorescence values were then collected using a plate reader, and the values compared against the standard curve to determine the percentage survival of primed cells.

(54) UV-Visible and Fluorescence Spectrophotometry

(55) UV-visible and fluorescence spectrophotometry were performed using routine methods.

(56) Bicinchoninic Acid Assay

(57) Bicinchoninic acid assays were performed according to the manufacturer's instructions. Briefly, 20 μL of samples were added to 200 μL of reagents A and B (Thermo Scientific, UK) mixed in a 50:1 ratio in a 96-well plate. The plate was then incubated for 30 minutes at 37° C., before measuring the absorbance at 530 nm using a plate reader. Absorbance values collected for analytes were compared against a standard curve of a protein at known concentrations to calculate the concentration of the analytes.

(58) Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE)

(59) SDS-PAGE analysis was performed using routine methods. Briefly, analytes were mixed 1:1 with sample application buffer comprising glycerol, SDS, EDTA, Tris, mercaptoethanol and bromophenol blue, and heated to 95° C. for 5 minutes. The samples were then loaded into Novex® 4-20% Tris-glycine pre-cast gels (Thermo Fisher Scientific). A voltage of 200 V was applied for 50 minutes, and the resultant gel stained with Coomassie Blue stain.

(60) Circular Dichroism

(61) Synchrotron-radiation circular dichroism was performed on the B23 beamline at the Diamond Light Source, Oxford. Samples were desalted into chloride-free buffers. Spectra were collected from 185 to 260 nm, using a cuvette with a pathlength of 200 μm. For thermal studies, data were collected from 20 to 90 to 20° C. at 5° C. intervals with 1 minute incubation time. Alternatively, lamp-radiation circular dichroism was performed on a J-1500 CD spectrophotometer (JASCO, Germany), using a 100 μm pathlength cuvette. Data deconvolution was performed using the BeStSel web service. (Micsonai et al. (2015) Proc. Natl. Acad. Sci. U.S.A. 112, E3095-3103).

(62) Fluorescence Microscopy

(63) Confocal microscopy was performed using a Leica TCS SP8 confocal laser scanning fluorescence microscope (Leica Microsystems, Germany), using routine methods. Widefield microscopy was performed using a Leica DMI6000 inverted epifluorescence microscope (Leica Microsystems, Germany), using routine methods. OpdA and thrombin were fluorescently tagged with either 5(6)-carboxyfluorescein N-hydroxysuccinimide (Sigma) or rhodamine N-hydroxysuccinimide (Thermo Scientific, Germany), according to the manufacturer's instructions, whereas scGFP-based constructs are inherently fluorescent. To observe localization of the complexes, proteins were added to cells plated in a glass-bottom dish for 10-30 minutes, washed with PBS, then imaged.

(64) Scanning Electron Microscopy

(65) Bovine articular surface samples were fixed with 2.5% glutaraldehyde for 1 hour, rinsed three times for 10 minutes with 100 mM sodium phosphate buffer pH 7.4, placed in 1% osmium tetroxide for one hour, washed three times for 10 minutes with 100 mM sodium phosphate buffer, then washed with water for 10 minutes. Dehydration steps were made with 25, 50, 70, 80, 90, 96, and 100% ethanol, changing concentration every 10 minutes, and then processed with a critical point dryer. The samples were sputter coated with palladium or chromium and imaged on an FEI field emission scanning electron microscope (Quanta 200).

(66) Proliferation

(67) Proliferation of tissue engineered hMSCs within [cThrombin][S] catalysed fibrin constructs was analysed by comparing the results of MTS assays (described above in ‘Cell cytotoxicity assays’) performed over time. The effect of priming hMSCs with [scGFP-CshA] and [scGFP-CshA][S] on their proliferation was analysed using a haemocytometer to count cells, and comparing them to their seeding number.

(68) Flow Cytometry

(69) hMSCs primed with protein complexes were harvested, washed in an initial wash step, and centrifuged at 1500 RPM for five minutes. The sediment was re-suspended in PBS containing a dead stain. Suspensions containing approximately 1,000,000 cells per mL were transferred to individual flow cytometry tubes, and analysed using a flow cytometer and associated software. The cell suspension was passed through the interrogation point at a rate of 100-300 events per second with a total of 20,000 whole cell events measured. The side scatter area (SSC-A), forward scatter area (FSC-A), forward scatter height (FSC-H), and experiment-specific fluorescence were measured, with unlabelled cells as a control group to define the gated areas used for all samples. The whole cell populations were defined by an FSC-A vs SSC-A gate firstly, with data outside this region excluded as cell debris. Following this, the whole cell populations were gated by FSC-A vs FSC-H defining the single cell populations. The single cell populations were further gated by defining an upper limit on the FSC-A vs the dead stain filter dot plot, and data above this limit were excluded as dead cells. The live cells were gated on a FSC-A vs. FITC-A plot, with data inside the region corresponding to scGFP positive labelled cells and data outside the region corresponding to non-fluorescent cells (priming hMSCs with scGFP-CshA constructs), or were gated on a PE-CF594-A vs. FITC-A plot, with data inside the region corresponding to scGFP positive (Q1 and Q2) and mCherry positive labelled cells (Q2 and Q4), and data outside the region corresponding to non-fluorescent cells (Q3) (cell-surface scGFP-SpyCatcher and mCherry SpyTag reaction). Experiment-specific parameters are given in Table 9.

(70) TABLE-US-00009 TABLE 9 Flow cytometry parameters Initial Instrument Filters for wash Dead Fixing and measuring Experiment step staining solution software fluorescence scGFP-CshA Phenol- 0.004 mg/mL No fixing NovoCyte, Qdot 605-A priming free Propidium NovoExpress (propidium DMEM iodide in PBS iodide), FITC-A (scGFP) scGFP- Phenol- 1% (v/v) 1% para- LSR Fortessa APC-Cy7-A SpyCatcher free Zombie NIR in formaldehyde X20, (Zombie NIR), and DMEM PBS for 15 FACSDiva FITC-A mCherry- then minutes at (scGFP), PE- SpyTag cell- PBS room CF594-A surface temperature, (mCherry) reaction then washed with PBS and fixed

(71) Cell Membrane Uptake Quantification

(72) hMSCs were primed for 15 minutes using protein (e.g. [scGFP-CshA]) and conjugate (e.g. [scGFP-CshA][S]) at a range of concentrations in phenol-free DMEM. The amount of protein bound to cell membranes could be calculated by subtracting the amount of protein in the supernatant, determined using UV-visible spectrophotometry at 487 nm, from the amount of protein added to the cells.

(73) Sedimentation velocity analytical ultracentrifugation (SV-AUC) SV-AUC experiments were performed on a Beckman Optima XL-I (Beckman Coulter, USA) using the UV/Visible absorption system at 280 nm and 487 nm, at 40,000 rpm and at 20° C. using two channel 12 mm Epon centerpieces. Buffer density and viscosity was determined using a Lovis 2000 rolling ball viscometer (Anton Paar, Austria). Sedimentation coefficients (S) were determined using the continuous distribution Lamm equation model (c(S)) and were converted to standard conditions (Sw (20, w)). Molecular weights were calculated directly from integrated c(s) peaks.

(74) Chemical Cationisation Methods

(75) Protein Cationisation

(76) Protein (OpdA pr thrombin) was cationised using a method derived from that described in Armstrong et al. (Nat. Commun. (2015) Jun. 17; 6:7405). Briefly, a solution of protein (OpdA or thrombin) in HEPES buffer was added to pH-neutralised N—N′-dimethyl-1,3-propanediamine (DMPA) at a given ratio, and the solution pH-adjusted with 6M HCl. N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC) was added either in a single addition or two half additions, and the solution was left to stir, then desalted using buffer exchange with 10K MWCO spin concentrators to end the reaction. Specific experimental parameters for each protein are presented in Table 10. Performing the method as reported by Armstrong et al. would lead to severe loss of enzyme activity, as the cationisation reaction leads to inactivation over time. The inventors have determined that either limiting the reaction time or performing size exclusion chromatography on the crude cationisation solution produces active enzyme.

(77) TABLE-US-00010 TABLE 10 Cationisation parameters. Ration Ratio DMPA:anionic EDC:anionic Reaction Reaction Reaction Protein Buffer sites sites pH time temp. Purification OpdA 30 mM 300:1 50:1 5.1 24 hours 4° C. Size exclusion HEPES, chromatography 100 μM CoCl.sub.2 Thrombin 60 mM 150:1 34:1 6.5 1 hour Room None HEPES temp.

(78) Recombinant Preparation of Supercharged Proteins

(79) Preparing Supercharged Fusion Proteins

(80) Supercharged GFP was as described in Lawrence et al. (J. Am. Chem. Soc. (2007) vol. 129 p. 10110-10112). For preparation of scGFP fusion proteins with the fibronectin-binding portion of CshA (SEQ ID NO:19), OpdA (SEQ ID NO:20), PIGF-2.sub.(123-144) (SEQ ID NO:21) and SpyCatcher (SEQ ID NO:23), a linker region was designed as outlined below. Subsequent steps were carried out as described above in the section headed “Plasmid preparation”.

(81) Linker Design

(82) The linking regions used to form the fusion proteins were designed using methods outlined in Chen et al. (Adv. Drug. Deliv. Rev. (2013) Sep. 29; 65,1357-69).

(83) Supercharging OpdA

(84) OpdA was supercharged to form scOpdA by mutation of 11 aspartic/glutamic acid residues to lysine residues, listed in Table 11 below (position numbering with reference to SEQ ID NO:10). The gene with mutated residues was ordered from Eurofins Genomics (Germany).

(85) TABLE-US-00011 TABLE 11 Mutations made to OpdA to produce scOpdA scOpdA modifications D76K D97K D109K E120K E121K E135K D136K D184K D211K D212K E239K

(86) Protein-Specific Assays

(87) Paraoxon Hydrolysis

(88) Proteins (OpdA-based constructs) were diluted to a working concentration in buffer. 100× paraoxon stocks were prepared in isopropanol. Formation of 4-nitrophenolate was measured at 405 nm, using an empirically determined extinction coefficient of ε.sub.405=12013 M.sup.−1.Math.cm.sup.−1. Non-linear regression was performed on initial-rate data to determine the Michaelis-Menten parameters.

(89) [cOpdA][S] Membrane Activity Assay

(90) 3′,6′-bis(diphenylphosphinyl) fluorescein (DPPF) was synthesised as a substrate for fluorescence imaging as described by Liguo An et al. (Chem. Eur. J. (2007), Feb. 2; 13:1411). DPPF was dissolved in DMSO to a stock concentration of 100 mM, and applied to cells at a final concentration of 1 mM for 30 minutes. Cells plated on glass bottom microwell dishes were labelled with 12 μM [cOpdA][S] after DPPF exposure. Images were collected using confocal microscopy. Acetylthiocholine was also used to assay the activity of membrane-bound [cOpdA][S] over 5 days. 450 μM acetylthiocholine and 300 μM 5,5′-dithiobis-(2-nitrobenzoic acid) was applied to hMSCs and hMSCs primed with 10 μM [cOpdA][S] plated in a 96-well plate at day 0, 1, and 5. The resulting absorbance was read at 412 nm over time, and an extinction coefficient of 14150 M.sup.−1.Math.cm.sup.−1 was used to calculate acetylthiocholine turnover from the initial rate.

(91) OpdA Neutron Reflectometry

(92) Neutron reflectometry was performed on the INTER beamline at the Isis facility, Oxford, and on the D17 beamline at the Institut Laue-Langevin, Grenoble. Floating bilayers of 4:1 2-oleoyl-1-palmitoyl-sn-glycero-3-phosphocholine (POPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (POPG) were assembled on a silicon block with a 1-Palmitoyl-2-[16-(acryloyloxy)hexadecanoyl]-sn-glycero-3-phosphorylcholine (al-PC) monolayer covalently bound to the silicon substrate using the Langmuir-Blodgett trough deposition method. OpdA constructs were loaded at concentrations of 0.2 and 5 μM, and loosely bound material washed with buffer. Data deconvolution was performed with the RasCal software package for MATLAB.

(93) cThrombin Fibrin Formation

(94) Thrombin-catalysed fibrin formation was measured through absorbance at 600 nm during the cationisation process. Briefly, 75 μL of 0.06 mg/mL thrombin was added to 125 μL of 5 mg/mL fibrinogen, shaken for 20 seconds, and the absorbance at 600 nm measured over time.

(95) Fibrin Constructs Catalysed by Membrane-Bound [cThrombin][S]

(96) [cThrombin][S]-primed hMSCs were added to a solution of 7.5 mg/mL human fibrinogen in wells precoated with agarose. Successful fibrin formation could be analysed using confocal microscopy with Alexa-594 tagged fibrinogen.

(97) Solution Coupling of SpyCatcher and SpyTag Constructs

(98) The coupling of [scGFP-SpyCatcher] or [scGFP-SpyCatcher][S] with either [mCherry-SpyTag] or [CshA-SpyTag] was investigated using SDS-PAGE. Equal volumes of either [scGFP_SpyCatcher] or [scGFP_SpyCatcher][S] and [mCherry_SpyTag] or [CshA_SpyTag] were mixed in a glass vial and agitated using a magnetic stirrer. At predetermined time points throughout the reaction, 10 uL of the resulting solution was removed and mixed with an equal volume of SDS sample application buffer for 5 minutes. The range of samples obtained in this method were applied to SDS-PAGE gels as previously described before subsequent staining and destaining.

(99) Static Adhesion Assay for scGFP-CshA and scGFP-PIGF.sub.(123-144)

(100) Cell-substrate adhesion was investigated using a CyQUANT® NF cell proliferation assay kit (Invitrogen, UK). Human fibronectin (Sigma Aldrich) was diluted to 10 μg/mL with PBS, collagen I (rat tail; Sigma Aldrich) and collagen II (bovine trachea; Sigma Aldrich) were diluted to 0.2 mg/mL. 100 μL of these solutions were used to coat each well of the non-tissue-culture-treated 96 well plate. The plates were then washed three times with PBS solution containing 10 mg/mL bovine serum albumin (BSA; Sigma Aldrich) to block the non-specific interactions. The wells treated with BSA were used as a control. Cells were primed with protein complexes and the cells were harvested and counted using a haemocytometer. Standard curve samples were established in expansion medium. After four hours of incubation, medium was removed from cells by gentle aspiration and 100 μL of dye binding solution was dispensed into each well. The plate was covered and incubated at 37° C. for 1 hour. The fluorescence intensity of each sample was measured using a plate reader with excitation at 485 nm and emission detection at 530 nm. Adhesive cell numbers were compared to control samples of untreated cells incubated with phenol-free DMEM.

(101) Flow Adhesion Assay

(102) Dynamic cell adhesion experiments were carried out with an ExiGo microfluidics pump (Cellix Ltd flowing through a Vena8 Fluoro+ biochip. The chip was coated overnight with 0.1 mg/mL collagen II (Sigma Aldrich) and unspecific sites were blocked with 10 μg/mL BSA (Sigma Aldrich). The channel was washed with phenol-free DMEM with no additives for 30 seconds at 40 μL/min. scGFP-PIGF.sub.(123-144)-primed and [scGFP-PIGF.sub.(123-144)][S]-primed hMSCs were resuspended at a density of 1 million cells per mL in phenol-free DMEM without additives. A 50 μL aliquot was added to the channel reservoir each time and the cells were withdrawn at flow rates of 6, 4, or 3 mL/minute.

(103) Adhesion to Bovine Articular Cartilage Explants

(104) Cartilage explants were harvested form the lateral and patellar groove of 6-8-week old calves, obtained 6-8 hours after death. The disks were delimited with an 8 mm biopsy punch and carefully detached with a surgical scalpel (size 22; Swann Morton). After dissection, the pieces were kept in DMEM with 10% FBS, 100 μg/mL penicillin-streptomycin. Cartilage discs were cut to 6 mm diameter with a biopsy punch and placed in a non-tissue culture treated 96 well plate (Fisher, UK) with 200 uL of phenol-free DMEM without supplements. Cells were primed with either scGFP-PIGF.sub.(123-144) or the corresponding conjugate and resuspended in phenol-free DMEM. Cells were added onto the cartilage and placed in incubator at 37° C. with 5% CO2 for 4 hours. The samples were then fixed for SEM imaging or histology analysis.

(105) In Vivo Transplantation of scGFP-CshA-Primed hMSCs in Mice

(106) Male 20-week-old FVB/N and BALB/c nude mice were purchased from the Animal Resource Centre (Perth, Western Australia). All animal procedures were approved by the Animal Ethics Committee of the University of Queensland and were carried out in accordance with Australian Code for the Care and Use of Animals for Scientific Purposes 8th edition. Mice were anaesthetized with isoflurane. Body temperature was controlled by placing mice on a heating pad set to 37° C. 150 μL of a suspension of [scGFP-CshA][S], 2×10.sup.6 untreated hMSCs, or [scGFP-CshA][S] modified hMSCs was injected with a 27 gauge needle either through a tail vein (intravenous injection) or through the chest wall into the left ventricle (intracardiac injection), respectively. Prior to the injection, the hMSCs were maintained at 4° C., and the cells were gently resuspended with a pipette to ensure no aggregation before the injection. The mice were sacrificed at 2 hours and 24 hours after the injection. Genomic DNA of the heart and lung were isolated using DNA Mini Kit (Qiagen, USA) and primers targeting the human Alu sequence according to the manufacturer's instructions. Droplet digital PCR (ddPCR) was then used to quantify the number of human cells in each tissue. Briefly, 20 μL of ddPCR reaction mix was separated into droplets with a QX200 Droplet Generator (BioRad, USA). The droplets were transferred into a 96-well PCR plate, sealed and incubated at following cycling conditions: one cycle of 95° C. for 5 minutes, 45 cycles of 95° C. for 30 seconds, 55° C. for 1 minute and one cycle of 4° C. for 5 minutes, 90° C. for 5 minutes and an infinite hold of 12° C. After thermal cycling, the PCR plate was transferred in QX200 Droplet Reader (read) and read in the FAM channel using QuantaSoft version 1.7.

(107) Results

(108) Protein Expression and Purification

(109) All proteins were confirmed to be expressed and purified using SDS-PAGE, mass spectrometry (MADLI-TOF), and activity assays. The mass spectrum for scGFP-OpdA is shown in FIG. 5. UV-visible spectrophotometry and fluorescence spectrophotometry confirmed the correct folding of the scGFP-fusion constructs (FIGS. 6-8).

(110) Cationisation

(111) The successful cationisation of OpdA and thrombin was confirmed using matrix-assisted laser-desorption-ionisation time-of-flight mass spectrometry (MALDI-TOF). OpdA was shown to increase in mass by approximately 1700 Da, corresponding to the addition of 20 DMPA molecules. Thrombin cationisation led to an increase in mass of approximately 3300 Da, equivalent to 39 DMPA molecules. The mass spectra collected at a charge number of 3 for thrombin are shown in FIG. 9.

(112) Zeta potentiometry was used to show the increased charge associated with cationisation. Cationisation increased the zeta potential of OpdA from −7 mV to +21 mV (see FIG. 10). The change in zeta potential during cationisation of thrombin is shown in FIG. 11.

(113) Structural changes associated with cationisation were assayed using dynamic light scattering (DLS), circular dichroism (CD), and small angle X-ray scattering (SAXS). DLS showed the cationisation of OpdA lead to an increase in size of 0.8 nm corresponding to the addition of DMPA molecules to surface residues, whilst CD showed minimal changes in secondary structure but an increase in thermal stability (FIG. 12). SAXS showed OpdA remained dimeric post-cationisation (FIG. 13).

(114) Conjugation

(115) Electrostatic grafting of the anionic headgroup of the surfactant to positively charged residues leads to a decrease in the surface charge of proteins, therefore the zeta potential is expected to decrease. scGFP-PIGF.sub.(123-144) was shown to have a zeta potential of +22 mV, while [scGFP-PIGF.sub.(123-144)][S] was −0.5 mV. scGFP-CshA had a zeta potential of +1 mV despite the highly anionic CshA region, and [scGFP-CshA][S] was −15 mV. cOpdA to [cOpdA][S] showed a reduction of 13 mV (FIG. 10).

(116) An increase in size corresponding to the addition of a surfactant corona is also expected. DLS showed an increase in hydrodynamic diameter of 1.9 nm, 2 nm and 2.9 nm for the conjugation of cOpdA, scGFP-CshA and scGFP-PIGF.sub.(123-144), respectively. scGFP-OpdA showed a 388 nm increase in size due to the formation of clusters. SV-AUC showed an increase in the sedimentation coefficient of [scGFP-CshA][S] from 4.1 to 4.8, indicating surfactant binding.

(117) Importantly, surfactant conjugation did not lead to denaturation. CD was used to assess the secondary structures of OpdA and scGFP-OpdA constructs. [cOpdA][S] showed minimal changes in secondary structure to each of OpdA and cOpdA, and retained the improved thermal stability of cOpdA (FIG. 12). Conjugation of scGFP-OpdA lead to an increase in thermal stability (FIG. 14). UV-visible spectrophotometry and fluorescence spectrophotometry confirmed that the fluorophore's structure was maintained (FIG. 6).

(118) Cell Loading

(119) The successful loading of the conjugate systems to membranes was confirmed through microscopy, spectrophotometry, flow cytometry, reflectometry and activity assays.

(120) The scGFP-based constructs are inherently fluorescent, and so were simply visualised using fluorescence microscopy. scGFP-CshA, [scGFP-CshA][S], scGFP-PIGF.sub.(123-144), [scGFP-PIGF.sub.(123-144)][S], scGFP-OpdA, and [scGFP-OpdA][S]-primed cells all displayed fluorescent membranes, but the conjugate species retained the membrane fluorescence for a longer time period: approximately 15 minutes for unconjugated protein versus more than 24 hours for conjugated protein, indicating that the unconjugated proteins are rapidly endocytosed, whilst conjugated protein is retained at the membrane.

(121) Thrombin and OpdA each had to be fluorescently tagged prior to imaging, as described in ‘Microscopy’ method section. cOpdA was rapidly internalised, whereas [cOpdA][S] remained at the cell membrane. [cThrombin][S] was observed at the cell membrane for up to 7 days. Native OpdA and thrombin did not interact with cells.

(122) UV-visible spectrophotometry may be used to determine the amount of protein bound to cell membranes. Approximately 0.7 billion scGFP-CshA complexes bound per cell, whereas 1 billion [scGFP-CshA][S] complexes bound per cell, after 15 minutes (FIG. 15). Similarly, flow cytometry showed significantly greater fluorescence for [scGFP-CshA][S]-loaded cells versus scGFP-CshA-loaded cells (FIG. 16). The amount of scGFP-CshA observable per cell is significantly reduced after longer time periods compared to the amount of [scGFP-CshA][S], due to rapid endocytosis of the unconjugated protein.

(123) Neutron reflectometry was used to assess the insertion of [cOpdA][S] into a model membrane. [cOpdA][S] was shown to insert into the lipid bilayer and, once the membrane was saturated with conjugate, forms a layer above the membrane. cOpdA penetrated the bilayer and disrupted the supporting monolayer, indicating that it was not embedded within the membrane.

(124) Importantly, the cells could be treated with each protein construct without significant cell cytotoxicity. The surfactant, glycolic acid ethoxylate 4-nonylphenyl ether, was also assayed. The maximum assayed loading concentrations below significant cytotoxicity are presented in Table 12 below. Asterisks mark the data where significant cytotoxicity was not observed. The data for scGFP-CshA and [scGFP-CshA][S] are shown in FIG. 17.

(125) Primed cells were also shown to proliferate readily. [cThrombin][S]-primed cells were shown to proliferate via an MTS assay over 22 days (FIG. 11). scGFP-CshA and [scGFP-CshA][S]-primed cells proliferated at the same rate as unprimed cells, as confirmed by direct cell counting (FIG. 18), indicating that process of priming the cells with the priming entity (scGFP-CshA or [scGFP-CshA][S]) had no impact on the ability of cells to proliferate.

(126) The differentiation potential of cells was also not affected by priming. [cThrombin][S]-primed cells treated with adipogenic or osteogenic media displayed characteristic fat droplets and extracellular calcium deposits, respectively, after 21 days, as observed with widefield microscopy.

(127) TABLE-US-00012 TABLE 12 Maximum loading concentrations for the conjugated proteins; * indicates significant cytotoxicity was not observed Maximum assayed loading concentration below Protein significant cytotoxicity Assay method [cOpdA][S] 15 μM alamarBlue [cThrombin][S] 5.2 μM MTS [scGFP-CshA][S] 8 μM MTS [scGFP-PIGF.sub.(123-144)][S] 12 μM MTS [scGFP-SpyCatcher][S] 14 μM MTS mCherry-SpyTag 14 μM MTS [scGFP-SpyCatcher][S]- 14 μM MTS [mCherry-SpyTag] Glycolic acid ethoxylate 25 mM* MTS 4-nonylphenyl ether

(128) Solution Activity of Constructs

(129) Post-modification, each protein maintained activity. Previous work by Brogan et al. (Nat. Commun. (2014) Oct. 10; 5:5058) with lipases from Rhizomucor miehei and Thermomyces lanuginosus reported a 98% and 85% reduction in substrate-turnover rate post-cationisation, and a further 55% and 40% reduction post-conjugation, respectively, therefore it is surprising that the inventors were able to maintain activity. The assays required to determine activity are specific to each protein.

(130) OpdA-based constructs were assayed for activity by measuring the hydrolysis rates of paraoxon. The Michaelis-Menten parameters of cOpdA were not significantly different to those of OpdA, however the K.sub.M was significantly decreased (25.0±4.49 vs. 45.6±8.80 μM) and the k.sub.cat significantly increased (92.4±3.63 vs. 75.4±3.68 s-1) for [cOpdA][S], leading to a 2.2-fold increase in the specificity constant (FIG. 19). scGFP-OpdA was shown to have a K.sub.M of 31 μM, and a k.sub.cat of 1.98 s.sup.−1.

(131) The activity of thrombin was assayed by measuring the absorbance at 600 nm corresponding to fibrin formation from fibrinogen cleavage. During cationisation, the activity of thrombin was retained up to 120 minutes of cationisation, although the activity was gradually decreased with increased cationisation duration, as shown in FIG. 20.

(132) The coupling of scGFP-SpyCatcher and [scGFP-SpyCatcher][S] to either mCherry-SpyTag or CshA-SpyTag was assayed using SDS-PAGE. The appearance of a band at a high molecular weight indicated the formation of the isopeptide bond between the SpyTag and SpyCatcher moieties, in each of the conjugated and non-conjugated samples, for each of the SpyTag constructs.

(133) Membrane Activity of Constructs

(134) The activity of [cOpdA][S] at cell membranes was followed via microscopy. Cells exposed to DPPF for 30 minutes then treated with [cOpdA][S] exhibited increased fluorescence at cell membranes. Furthermore, hMSCs primed with [cOpdA][S] were able to turn over more acetylthiocholine than unprimed cells over at least 5 days.

(135) [cThrombin][S] bound to cell membranes was able to cleave fibrinogen to form a fibrin gel, as confirmed with confocal microscopy, using Alexa-594-tagged fibrinogen (FIG. 21).

(136) scGFP-CshA, scGFP-PIGF.sub.(123-144), and their respective conjugates were assayed for their ability to bind to fibronectin (scGFP-CshA and scGFP-PIGF.sub.(123-144)) and collagen I and II (scGFP-PIGF.sub.(123-144)). scGFP-CshA-primed cells and [scGFP-CshA][S]-primed cells adhered in significantly greater numbers than unlabelled cells to fibronectin-treated plates (FIG. 22). Under 4 mL/min flow, hMSCs primed with [scGFP-PIGF.sub.(123-144)][S] adhered in significantly greater numbers to collagen II than unprimed hMSCs, as observed via widefield microscopy.

(137) Flow cytometry showed that hMSCs primed with scGFP-SpyCatcher or its corresponding conjugate were able to form covalent bonds with mCherry-SpyTag for up to 72 hours.

(138) [scGFP-PIGF.sub.(123-144)][S]-primed hMSCs were seen to adhere in greater numbers to explanted bovine articular cartilage than unprimed hMSCs, as seen in FIG. 23.

(139) In Vivo Activity of Constructs

(140) [scGFP-CshA][S]-primed hMSCs were transplanted into mice via intravenous and intracardiac injection. Upon harvesting the heart and lung tissue from the mice after 2 hours, 24 hours, and 4 weeks, the number of hMSCs in each tissue was determined using droplet digital PCR. The tissue:plasma distribution coefficient of [scGFP-CshA][S]-primed hMSCs in the heart was shown to have increased 2-fold relative to unprimed hMSCs at 2 hours and 24 hours.