WO2007030012A2 - Extracellular matrix imaging - Google Patents

Abstract

The present invention relates to labelled conjugates, for example alpha 1 intergrin A-domain or collagen adhesin of S. aureus (CNA 35), for imaging extra cellular matrix components. It also relates to compositions and kits comprising these labelled conjugates and to imaging methods in which the conjugates are used. The conjugates of the invention comprise a peptide sequence which binds to an extra cellular matrix component, i.e. to an extra cellular matrix protein or to proteoglycans, and which is coupled to an imaging agent. In particular conjugates which bind to elastin, collagen or to a glycosaminoglycan are envisaged. The conjugates according to the invention have some important advantages over existing tools for imaging extra cellular matrix components, in particular for applications in tissue-engineering experiments or MRI. The use of protein conjugates of the invention allow visualization of very small extra cellular matrix components, such as newly formed collagen fibrils.

Description

EXTRACELLULAR MATRIX IMAGING

Field of the invention

The present invention relates to the field of peptide conjugates used in visualization. In particular peptide conjugates which can be used for extra cellular matrix imaging, e.g. in the field of tissue engineering.

Background of the invention

In mammalian tissue the extra cellular matrix is composed of a mixture of proteins and proteoglycans. The proteins can be distinguished into structural proteins (collagens and elastin), which account for most of the mechanical properties of tissue, and the specialized proteins, such as laminin, fibronectin and fibril in. Proteoglycans are composed of long polysaccharide chains (glycosaminoglycans or GAGs) attached to a protein core. Examples of GAGs are: heparan sulphate, keratan sulphate, hyaluronic acid, chondroitin sulphate, dermatan sulphate. GAGs are important because they regulate the compressive properties and lubrication of tissue. Engineering tissue in vitro for human implantation requires highly controlled extra-cellular matrix (ECM) production and remodelling in order to replicate the mechanical properties of native tissue. When replicating and monitoring the mechanical properties of native or engineered tissues, visualization of the GAGs, collagen and elastin is most important. As an example, implantable tissue-engineered heart valves have been generated that are promising for use in the pulmonary valve position, but do not have sufficient mechanical properties to withstand the higher mechanical loads at the aortic valve position (Hoerstrup et al, Circulation 102 (19 Suppl 3) (2000) III 44-9. Collagen is the protein primarily responsible for the load-bearing properties of heart valves and other tissues, and collagen organization rather than collagen content dictates these mechanical properties (Billiar et al., J. Biomechanical Engineering 122 (2000) 23-30). Adequate collagen remodelling studies require monitoring of changes in local collagen structures in time. For this purpose, standard histological techniques requiring cell fixation and biochemical assays requiring destruction of tissue constructs are unsuitable. In order to monitor long term growth and changes in orientation and structure of collagen in vitro, a probe is needed that allows visualization over time of collagen fibers surrounding living cells, without sacrifice or fixation. The many types of collagen are grouped into families according to their functions. Types I, II, III, V and XI belong to the most abundant family, the fibril- forming collagens. Each type is distributed differently in different tissues and has unique amino acid sequences, but in all types of collagen three protein chains align to form a single triple helical molecule "Collagen. Primer in Structure, processing and Assembly, Brinckmann", J., Notbohm, H., Muller, P. K., Eds., Topics in Current Chemistry vol 247 Springer- Ver lag: New York, 2004. Collagen has intrinsic properties that allow enhanced microscopic visualization in tissues undisturbed by the addition of labelling agents or fixatives: birefringence under polarized light, autofluorescence and second harmonic generation. In addition, fine images have been produced of collagen fibers and fibrils under physiological conditions using confocal reflection microscopy, and differential interference contrast microscopy. However none of these techniques has proved to be sensitive and specific enough to visualize the small newly formed fibrils of growing tissues, such as in tissue-engineered constructs. Methods for enhancing visualization of individual collagen fibers, and especially the much smaller fibrils, are currently quite limited. Most histological and immunohistological techniques have been developed for use on fixed tissue, requiring sacrifice of tissue samples. The collagen stains commonly used such as picro-sirius red, Von Gieson picro-fuchsin and Masson's trichrome, are unsuitable for live cells because they rely on the affinity of strong acid dyes for the cationic groups of the collagen proteins. Fluorescent antibodies could be used under physiological conditions, but tissue engineering and remodelling experiments would require large amounts of such antibodies, which would be expensive. Another potential problem with antibodies is that their relatively large size could cause limited diffusion into dense tissues, and their high binding affinity to collagen may affect tissue function and development. The use of dichlorotriazinyl amino fluorescein (DTAF) has been reported as a small fluorescent probe for visualizing collagen fibers in live tissue, both for in vivo healing experiments and for in situ mechanical studies. However, DTAF is a non-specific amine-reactive fluorescent dye that is also known to non-specifically label carbohydrates at near neutral pH. Therefore, specificity for collagen is expected to be low (Blakeslee & Baines (1976) J. of Immunological Methods 13: 305-320). Another drawback of DTAF is that it forms a covalent bond, a modification that is irreversible and may affect the collagen properties. For use in tissue-engineering experiments a collagen specific probe would preferably show reversible binding, be specific for certain types of collagen, small enough to readily diffuse into tissues, inexpensive to produce and not affect cell viability or collagen properties.

Short description of the figures

Figure 1. Solid-phase binding of CNA35-OG488 (solid symbols) and αiI-GST-OG488

(open symbols) to human collagen type 1 (circles) and BSA (squares) detected by the emission at 538 nm (excitation wavelength: 485 nm). The inset shows a magnification of the titration curve at low probe concentrations. The probes were in PBSM (PBS (phosphate buffered saline) containing 2 mM MgCl₂. PBS contained 0.1 M phosphate buffer, 2.7 μM KCl and 137 mM NaCl, pH 7.4) buffer containing 1 mg/mL BSA and all assays were done in triplicate. Bars indicate plus and minus one standard deviation.

Figure 2. Solid phase binding assay comparing relative binding specificity of protein- based probes and DTAF for a variety of ECM proteins. Figure 2A depicts the results of the assay as performed using the following ECM proteins: human collagen I, laminin, elastin, fibrinogen, fibrin, fibronectin, BSA. Figure 2 B depicts the results of the assay as performed using the following ECM proteins: human collagen I, rat collagen I, bovine collagen I, BSA. Figure 2 C depicts the results of the assay as performed using the following ECM proteins: human collagen I, human collagen II, human collagen III, human collagen IV, human collagen V, human collagen VI, BSA. Binding experiments were done with 1 μM of CNA35-OG488 (black bars), 5 μM of αiI-GST-OG488 (white bars), or 200 μM of DTAF (gray bars) in PBSM buffer with 1 mg/mL BSA. Fluorescence intensities are shown relative to the intensity each probe shows with human collagen type I. All assays were done in triplicate. Bars indicate plus and minus one standard deviation.

Figure 3. Confocal Laser Scanning Microscopy images of live HVS cells incubated with CellTracker Orange (grey channel) and (A) 0.5 μM CNA35-OG488, (B) 2.5 μM αiI-GST-OG488 or (C) 7.5 μM DTAF (dark grey channel). Bar = 50 μm. Figure 4. Amino acid sequence of HIS tagged CNA 35. The part containing the HIS tag is underlined and the part representing the CNA35 sequence is shown in bold.

Figure 5. Amino acid sequence of αil-GST. The collagen binding domain from alpha 1 is shown in bold.

Figure 6. Amino acid sequence of the alpha 1 fragment obtained after thrombine cleavage of αil-GST.

Figure 7. Detailed image of the collagen architecture within a mouse carotid artery at a certain depth both for (A) SHG signal of collagen (grey) and (B) fluorescence signal of CNA35- OG488 probe (light grey). (A) and (B) are recorded at the same position. Picture size is 512x512 pixels and pixel size is 0.2 x 0.2 μm.

Figure 8. 1 - VI: subsequent images through a mouse carotid artery (3D stack) using CNA35-OG488. Cell nuclei (DNA/RNA) are shown in dark grey (blue), collagen in light grey (yellow) and elastin in medium grey (red). Image resolution is 1024 xlO24 pixels and the pixel size is 0.2 x 0.2μm.

Figure 9. Representative image of an engineered cardiovascular construct after 4 weeks of static culture using CNA35-OG488. Cells are shown in medium grey (blue, some cells are designated by arrows annotated with 2), collagen in light grey (green) and scaffold is shown in dark grey and is indicated by arrows annotated with 1 (purple (combination of blue and red)). Signal intensity of cells and scaffold are improved after imaging. Picture size is 1024 x 1024 pixels and the pixel size is 0.20 x 0.20 μm.

Figure 10. Images I - VI show in vitro collagen production of cells in monolayer culture, studied over time (0 - 52 hours). The cell cytosol is shown in light grey

(red) and collagen is shown in dark grey (green). The time from the start of the experiment is shown in the upper left corner. Image resolution is 512 x 512 pixels and the pixel size is 0.40 x 0.40 μm.

Figure 11. (A) High magnification image of one single myofibroblast cell in monolayer culture, 1 day after plating. The cell cytosol is shown in medium grey (red) and collagen is shown in dark grey (green). Image resolution is 922 x 922 pixels and the pixel size is 0.10 x 0.10 μm. (B) Detailed image of a monolayer culture which was stained with CN A35- OG488 48 hours after seeding and with CNA35-RRX 72 hours after seeding, medium grey (green) and medium grey (red) respectively. In the intervening period the cells were kept in medium without any probe. Dark grey (yellow) represents area where the two probes colocalize. This figure is also present as a colour image to better distinguish between CNA35-OG488 and CNA-RXX bindings.

Figure 12. Fluorescence of wells used for MRI measurements. A) Milk powder- blocked and incubated with CNA35-liposomes; B) Collagen-coated and incubated with bare liposomes; C and D) Collagen-coated and incubated with CNA35-liposomes; E) Not coated and not incubated.

Figure 13. Ti-weighted MR image of 5 wells of a 96-wells plate. A) Milk powder- blocked and incubated with CNA35-liposomes; B) Collagen-coated and incubated with bare liposomes; C and D) Collagen-coated and incubated with CNA35-liposomes; E) Not coated and not incubated.

Figure 14. Solid-phase binding assay of CNA35-liposomes to rat tail collagen type I (solid squares). Liposome binding was monitored by measuring the fluorescence of the rhodamine lipids at 620 nm using an excitation of 578 nm Control experiments using non-modified liposomes incubated on human collagen type I (open triangle) and CNA35-liposomes incubated on milk-powder blocked well without collagen (open squares) are also shown for comparison. The solid line represents a fit to a 1 : 1 binding model using a Kd of 3 nM. Detailed description

Labelled conjugate

The present invention relates to a labelled conjugate for imaging extra cellular matrix components. It also relates to a composition comprising the labelled conjugate and to imaging methods in which the conjugate is used.

The conjugates according to the invention have some important advantages over existing techniques for imaging extra cellular matrix components, in particular for applications in tissue-engineering experiments. Unlike the label-free microscopic techniques that depend on auto fluorescence, birefringence effects under polarized light, second harmonic generation, confocal reflection or differential interference contrast, the use of protein conjugates of the invention allow visualization of much smaller extra cellular matrix components, such as newly formed collagen fibrils. In addition, the conjugates of the present invention can be used to distinguish between different types of components. For instance, in the case of collagen, it can be used to distinguish between different collagen species, e.g. between type I and type III, or between type III and type XI, by using collagen-binding domains with different binding specificity. The conjugates, in particular a CNA35-based conjugate, has also several advantages compared to the use of fluorescently labelled antibodies. Collagen binding protein domains from CNA and related proteins typically bind collagen with a K_A of around 10^"7-10^"6 M, which ensures that binding is sufficiently strong but not so tight that it becomes essentially irreversible. The latter property is important when such a probe is used to monitor collagen formation and remodelling in tissue-engineering experiments, where a probe should not affect the structural organization of the collagen that is formed and continuously modified. Finally, CNA35 can be recombinant Iy expressed in E. coli and purified in a single step in high yields, which is an advantage for applications that require relatively large amounts of probes, where the use of fluorescently labelled monoclonal antibodies would become cost-prohibitive.

In one aspect, the invention relates to a labelled conjugate for imaging extra cellular matrix compounds in a biological tissue, wherein the conjugate comprises a peptide sequence which binds to an extra cellular matrix component, i.e. to an extra cellular matrix protein or to proteoglycans, and which is coupled to an imaging agent. In one embodiment the conjugate binds to any one of elastin, collagen or a GAG, such as heparan sulphate, keratan sulphate, hyaluronic acid, chondroitin sulphate or dermatan sulphate. In another embodiment, separate conjugates binding specifically to one of these components, are combined to form a composite conjugate which specifically binds more than one of these components.

Examples of suitable collagen binding peptide sequences include sequences which comprise or consist of:

(1) all or part of alpha 1 integrin, in particular the A-domain SEQ ID NO:3, preferably the A-domain of alpha 1 integrin (SEQ ID NO:3) fused to another molecule, more preferably SEQ ID NO. 2 comprising the A-domain of alpha 1 integrin fused to a glutathione S-transferase (GST));

(2) bacterial adhesion protein such as collagen adhesin of Staphyloccocus aureus (CNA), preferably the truncated form (CNA35), more preferably the truncated form CNA35 fused to another molecule, and most preferably the truncated form CNA35 fused to an His-tag (see SEQ ID NO:1) .

Many other collagen-binding protein domains present in integrins, bacterial adhesion proteins or von Willebrand factor are known in the art, e.g. from House et al, Methods: A Companion to Methods in Enzymology 6 (1994) 134-142;. Dickeson et al., Cell. MoI. Life Sci. 54 (1998) 556-566; Bienkowska et al., J. Biol. Chem. 272 (1997) 25162- 25167. These different collagen-binding molecules can be used to develop conjugates with differing specificities.

Collagen binding peptide sequences, but also other biopolymers, such as DNA and PNA, may be identified using methods known in the art. Such methods include e.g. screening libraries containing antibody fragments or peptide libraries, and particularly useful, phage display libraries. These libraries may be screened for collagen binding molecules and positive hits can then be recombinant Iy expressed, labelled and analysed in a manner similar as described here for CNA35 and alpha 1 integrin. In one embodiment, the peptide sequence in the labelled conjugate comprises amino acids 30-344 of Staphyloccocus aureus collagen adhesin (CNA35).

In another embodiment, the peptide sequence in the labelled conjugate comprises amino acids 123-336 of mature integrin subunit alpha 1 or SEQ ID NO:3 . Examples of suitable elastin binding peptide sequences include sequences which comprise or consist of proteins known to be associated with elastin in microfibrils such as fibrilines, MAGP's, emilins and fibulins, see for example Kielty et al.(2000) J Cell ScL 115: 2817-2828, or elastin binding domains identified in bacterial adhesins such as EbpS, FnBPA and FnBPB from Staphyloccocus aureus, see for example Park et al, J. Biol. Chem. (1999) 274, 2845-2850; Roche et al., J. Biol. Chem. (2004) 279:38433- 38440. Examples of suitable glucosaminoglycan (GAG) binding peptide sequences include sequences which comprise or consist of GAG binding domains found in many viruses (Herpes, Picorna, HIV, Dengue, Japanese Enceptalitis virus), GAG binding domains present in integrines, or GAG binding domains present in other ECM proteins such as fibronectin, vitronectin and laminin. See fore example Wu, Suh-Chin et al. Biomacromolecules (2004), 5: 2160-2164.

According to one preferred embodiment, the labelled conjugate of the invention, comprises a peptide sequence able to bind collagen, elastin or GAG, wherein said peptide sequence is optionally fused to another molecule and wherein the peptide sequence is selected from the following list:

(1) an alpha 1 integrin A-domain, preferably SEQ ID NO:3,

(2) a bacterial adhesion protein or bacterial adhesin, (3) all or part of (1) or (2), or

(4) peptide sequences homologous with (1), (2) or (3).

More preferably, the labelled conjugate comprises as bacterial adhesion protein (2) the collagen adhesin of S. aureus (CNA). Even more preferably, the labelled conjugate comprises a truncated form of the collagen adhesin of S. aureus (preferably CNA35, which is represented by part of the amino acid sequence of SEQ ID NO:1 : CNA35 starts at amino acid 13 of SEQ ID NO:1), even more preferably the truncated form of CNA35 fused to another molecule, and most preferably the truncated form CNA35 fused to an His-tag (SEQ ID NO:1) or a peptide homologue thereof. According to another even more preferred embodiment, the labelled conjugate comprises the truncated form known as CNA 19 of the collagen adhesin of S. aureus, even more preferably the truncated form CNA 19 fused to another molecule, and most preferably the truncated form CNAl 9 fused to an His-tag or a peptide homologue thereof. CNAl 9 is a peptide comprising amino acids 151 till 318 of the collagen adhesin of S. aureus (Zhong Y. et al, (2005), EMBO J., 24:4224-4236).

According to another more preferred embodiment, the labelled conjugate comprises as peptide sequence (1), the peptide sequence SEQ ID NO:3 fused to another molecule, even more preferably the peptide sequence SEQ ID NO:3 fused to a GST (SEQ ID. No.2) or a peptide homologue thereof.

By "all or part of in (3) is meant any peptide sequence variant derived from a given peptide sequence as defined in (1) or (2) as long as it still exhibits its biological activity, which is to be able to bind collagen, elastin or GAG. A peptide sequence is said to exhibit a biological activity, when its Kd for collagen, elastin or GAG is preferably lOOμM or lower. More preferably, the Kd is 50 μM or lower. Even more preferably, the Kd is 1 μM or lower. Kd is preferably assessed by solid phase binding assay as described in example 2.

The skilled person will understand that peptide sequences may be modified, for instance to facilitate purification. Suitable modifications are fusions to His-tags or glutathione S-transferase. Thus, in one preferred embodiment, amino acids 30-344 of Staphyloccocus aureus collagen adhesin (CNA35) were fused to an N-terminal His-tag (SEQ ID No.l). In another preferred embodiment, amino acids 151-318 of Staphyloccocus aureus collagen adhesin (CNA 19) were fused to an N-terminal His-tag. In another embodiment, amino acids 123-336 of mature integrin subunit alpha 1 were fused to glutathione S-transferase (SEQ ID No. 2).

Peptides for conjugates of the invention may be isolated as such from natural human, animal or plant sources. Alternatively, they may be prepared recombinantly, e.g. by cloning, using techniques known in the art. Yet alternatively, peptide sequences may be prepared synthetically, e.g. by solid phase peptide synthesis according to the established method of Merrifield (Solid Phase Peptide Synthesis, 2nd ed., pp. 53-123, 1984 The Pierce Chemical Co., Rockford, IL.) Merrifield RB. 1963. Solid phase peptide synthesis. I. The synthesis of a tetrapeptide.J Am Chem Soc 85, 2149-2154 Chan, et al. Eds., Fmoc Solid Phase Peptide Synthesis: A Practical Approach, Oxford University Press, New York, New York (2000).

By "peptide sequences homologous" in (4) is preferably meant peptide sequences which differ only by a conservative amino acid substitution from the amino acid sequences of known collagen, elastin or GAG binding peptides as defined in (1), (2) or (3), or by one or more non-conservative amino acid substitutions, deletions, or insertions located at positions which do not destroy the biological activity of the peptide (in this case, the ability of the peptide to bind collagen, elastin or GAG), are also encompassed by the present invention. The assessment of the biological activity of the peptide is preferably performed as described above. A peptide homologue may also include, as part or all of its sequence, one or more amino acid analogues, molecules which mimic the structure of amino acids. Using collagen as an example, suitable modifications include the introduction of cysteine residues, e.g. by site-directed mutagenesis at positions remote from the collagen binding site of CNA35 to allow better control over the position of the label.

Homologous peptide sequences may (also) have at least 50% identity with the amino acid sequence of SEQ ID NO: 3 (alpha 1 integrin A-domain) or with the amino acid sequence starting from amino acid 13 of SEQ ID NO:1 (CNA35). Preferably, the homologous peptide sequences has at least 55%, more preferably at least 60%, even more preferably at least 65%, even more preferably at least 70%, even more preferably at least 75%, even more preferably at least 80%, even more preferably at least 85%, even more preferably at least 90%, even more preferably at least 95%, even more preferably at least 97%, and even more preferably at least 99% identity with SEQ ID NO: 3 or SEQ ID NO:4. Most preferably, the homologous peptide sequence is identical with SEQ ID NO: 3 or SEQ ID NO:4. Percentage of identity is determined as the number of identical amino acid residues between aligned sequences divided by the length of the aligned sequences minus the length of all the gaps. Multiple sequence alignment was performed using DNAman 4.0 optimal alignment program using default settings. Homologous peptide sequences may (also) have enhanced biological activity compared to the biological activity of the peptide sequence they derive from, which is the corresponding sequence among the peptide sequences as defined in (1), (2) or (3)). Biological activity is preferably assessed as described above. Biological activity is enhanced when the Kd of the peptide for collagen, elastin or GAG is preferably 20 nM or less. More preferably, the Kd is 1OnM or less. Even more preferably, the Kd is InM or less. Kd is preferably assessed as defined earlier. The skilled person will understand that it might be attractive to increase the biological activity of the peptide. However, the binding of the peptide to collagen, elastin or GAG should stay reversible. For example, this peptide when present in a labelled conjugate of the invention can be attractively used in a method of the invention, preferably in a method, wherein two labelled conjugates of the invention are being used sequentially (dual staining) as defined later in the description. Alternatively, homologous peptide sequences may have decreased biological activity compared to the biological activity of the peptide sequence they derive from. Biological activity is preferably assessed as described above. Biological activity is decreased when the Kd of the peptide for collagen, elastin or GAG is preferably comprised between about 50μM and 100 μM. More preferably, the Kd is comprised between about 70μM and 100 μM. Even more preferably, the Kd is comprised between about 80μM and 100 μM. Kd is preferably assessed as defined earlier. The skilled person will understand that it might be attractive to decrease the biological activity of the peptide. However, the binding of the peptide to collagen, elastin or GAG should be high enough to be specific and detectable.

SEQ ID NO: 3 is the peptide sequence of the human alpha 1 integrin A-domain. Homologous peptide sequences may be isolated from other organims than human. Depending on the organism wherein the labelled conjugate will be used, the skilled person will choose the most suited organism. It is also encompassed by the present invention to isolate several peptide sequences exhibiting binding to collagen, elastin or GAG from one single organism. Therefore, according to one first preferred embodiment, the homologous peptide originates from human and is homologous with SEQ ID NO:3. Concerning the bacterial adhesion protein or bacterial adhesin, any bacterial suspected to express a functional adhesion protein or adhesin can be used as source of homologous amino acid sequence. Preferably, the bacteria is a Staphyloccocus species. More preferably, a Staphylococcus aureus strain.

All homologous peptides defined herein may be obtained using state of the art molecular biology techniques.

The imaging agent of the labelled conjugate may be any detectable label which is known in the art. An imaging agent is any agent able to give or to enhance a contrast in an imaging method, leading to the formation of an image. In the context of the invention, an image is formed with the aim to obtain a representation of structures, features or properties of relevance for objective clinical diagnosis, prognosis, screening and evaluation.

Suitable labels include radioactive labels and non-radioactive labels, such as paramagnetic labels, fluorescent labels (including organic dyes), quantum dots and immunolabels. Preferred fluorescent labels are amine-reactive fluorescein derivatives. Several amine-reactive fluorescein derivatives are commercially available (see for example Invitrogen (http://probes.invitrogen.com/handb)). Preferred amine-reactive fluorescein derivatives are Oregon Green® 488-maleimide (Oregon Green® 488- maleimide 5-isomer Invitrogen, catalogus no. 06034), Oregon Green® 488-succinimide (Oregon Green® 488 carboxylic acid succinimidyl ester 5-isomer, Invitrogen, catalogus no: 06147, )Alexa® 488 succinimide (Alexa® fluor 488, Invitrogen, catalogues No: A20000), and Rhodamine Red™-X (Rhodamine Red™-X, succinimidyl ester 5-isomer, Invitrogen, catalogus no: R6160).

Preferred labelled conjugates In a preferred embodiment, the labelled conjugates comprise as peptide sequence SEQ

ID NO:1 or SEQ ID NO:2.

More preferred labelled conjugates of the invention are selected from the following group:

SEQ ID NO:1 (CNA35 fused to His tag) coupled to an amine-reactive fluorescein derivative. Preferably, the amine-reactive fluorescein derivative is Oregon Green®

488-maleimide, Oregon Green® 488-succinimide, Alexa® 488 succinimide, or

Rhodamine Red™-X, and SEQ ID NO:2 (alpha 1 integrin A-domain fused to GST) coupled to an amine-reactive fluorescein derivative. Preferably, the amine-reactive fluorescein derivative is Oregon Green® 488-maleimide, Oregon Green® 488-succinimide, Alexa® 488 succinimide or Rhodamine Red™-X. Even more preferred labelled conjugates of the invention are selected from the following group:

SEQ ID NO:1 (CNA35 fused to His tag) coupled to Oregon Green® 488-maleimide (CNA35-OG488), SEQ ID NO:1 (CNA35 fused to His tag) coupled to Rhodamine Red™-X (CNA35- RRX),

SEQ ID NO:1 (CNA35 fused to His tag) coupled to Alexa® 488 succinimide (CNA35-AF488), and

SEQ ID NO:2 (alpha 1 integrin A-domain fused to GST) coupled to Oregon Green® 488-maleimide (cciI-GST-OG488).

Labelled conjugates for Magnetic Resonance Imaging (MRI)

In another preferred embodiment, the imaging agent is a liposome containing a MRI contrast agent. These types of labelled conjugates are specifically attractive for carrying out MRI as described in the section entitled MRI. Any known MRI contrast agent may be used. Preferably, the MRI contrast agent consists or comprises gadolinium. Other well known contrast agents are iron oxide nanoparticles.

The preparation of these labelled conjugates containing as imaging agent a liposome containing a MRI contrast agent may be carried out using two alternative conjugation techniques. The first one was described in Mulder et al (Mulder, W. J et al. (2004) A liposomal system for contrast-enhanced magnetic resonance imaging of molecular targets. Bioconjug Chem 15, 799-806). This first strategy involved modification of lysine residues on CNA35 using the SATA (7V-Succinimidyl S-acetylthioacetate) coupling ligand. After deprotection of SATA, the resulting thiol group was reacted with maleimide-functionalized PEG-DSPE (l,2-distearoyl-sn-glycero-3- phosphoethanolamine-7V-[methoxy(poly(ethylene glycol)]) incorporated into the liposome. The second strategy involved the use of Native Chemical Ligation (NCL) (as described in example 11). This second conjugation method was also described for another protein in Grogan et al (Grogan, M. J. et al. (2005) Synthesis of lipidated green fluorescent protein and its incorporation in supported lipid bilayers. JAm Chem Soc 127, 14383-14387.). Briefly, in this second conjugation method, CNA35 was expressed as a fusion protein with an intein-domain using the commercially available IMPACT (Novagen) system. After affinity purification on a chitin resin, CNA35 was cleaved from the intein domain by treatment with MESNA (sodium 2-sulfanylethanesulfonate) yielding CNA35 with a C-terminal thioester. Subsequently, CNA35 containing a thioester was reacted with liposomes containing cysteine-functionalized pegylated phospholipids forming a peptide bond between the pegylated phospholipids and the C- terminus of CNA35. The resulting liposomes contained on average 100 copies of the protein per liposome. These liposomes are specific for collagen and have an improved affinity for it as demonstrated in the example. These liposomes are highly suited to be used in MRI imaging.

Kit In another aspect, the invention relates to a kit comprising a labelled conjugate of the invention. The kit comprises in separate containers (i) a labelled conjugate of the invention, and (ii) additional components selected from the group consisting of: a control binding substrate such as collagen, elastin or GAG, a cellular probe, a molecular probe, a negative control, a solvent, a buffer, a mineral oil, a medium, and a mounting agent.

The kit may comprise several labelled conjugates of the invention. Preferably, the kit comprises at least one of the preferred labelled conjugates of the invention as defined in the section entitled "preferred labelled conjugates" and/or "labelled conjugates for MRI". A cellular or molecular probe can be any probe known to be specific for components of interest within a biological tissue or a cell. For example: SYTO44, eosin, Cell Tracker Blue CMAC (CTB), Cell Tracker Orange CMRA (CTO) probes (all from Molecular Probes, The Netherlands) which are specific probes for DNA/RNA, elastin, and the last two for cell cytoplasm, respectively. A negative control is preferably a labelled conjugate as defined in the invention, except is comprises a peptide sequence, which exhibits no detectable binding to collagen, elastin or GAG. A preferred negative control is a labelled conjugate having for peptide sequence an inactive mutant form of CNA35. This mutant form has a specific point mutation at tyrosine 175 of the collagen adhesin of S. aureus (CNA) (Symersky J, et al, (1997), NATURE STRUCTURAL BIOLOGY 4 (10): 833-838). A medium can be a sterile medium used for cell culture. Depending on the type of cellular application envisaged, the skilled person would know which medium is the most suited to be present in the kit. A medium can also be an embedding medium or mounting medium. An embedding medium is a firm medium used to fixate tissue specimens or cells medium. It can be any known embedding medium, such as paraffin, a hydrogel or plastic,. A mounting medium is used to mount microscopic objects or tissue specimens, usually between glass slides, for storage and microscopic inspection. For example entellan from DABCO can be used A mounting agent is an agent known to delay fluorescence fading or bleaching. Any known mounting agent may be present. Examples of mounting agents are: PVA-DABCO or PVA-NPD (SigmaAldrich or Fluka). The kit is preferably used for carrying out any method of the invention as defined later in the description.

Composition

In another aspect, the invention relates to a composition comprising a labelled conjugate according to the invention. Preferably, the composition comprises at least one of the preferred labelled conjugates of the invention as defined in the section entitled "preferred labelled conjugates" and/or "labelled conjugate for MRI". The composition is preferably used for carrying out any method of the invention as defined later in the description. Compositions of the invention may be used for in vivo or ex vivo, such as in vitro, imaging of extra cellular matrix components from human or animal origin, such as from man, rat, mouse, chicken, dog, rabbit or bovine origin. For in vivo or ex vivo use, the composition of the invention is preferably a pharmaceutical composition comprising an effective amount (e.g. an amount effective to enhance image contrast in imaging) of the labelled conjugate of the invention together with at least one pharmaceutically effective carrier or excipient. The composition may comprise as additional component together with a label conjugate of the invention at least one of the following components selected from the group consisting of: a cellular probe, a molecular probe, a negative control, a serum, a solvent, a salt solution, a buffer, a mineral oil, a medium, and a mounting agent. All these components have already been defined in the section entitled " kit". Additionally according to a preferred embodiment, the composition comprises a labelled conjugate of the invention with a cell and/or a biological tissue. Preferred types of cells and biological tissue are defined in the section entitled "method". According to another preferred embodiment, the composition comprises a labelled conjugate of the invention with a support material of synthetic or natural origin. Any support material suited for cell growth and/or tissue culture may be used. According to another preferred embodiment, the composition comprises a labelled conjugate of the invention with a material of synthetic or natural origin to facilitate or guide the transport and delivering of the labelled conjugate to a given place. Any bioresorbable material approved for clinical application may be used, e.g. PGA, PLA, PGLA, or natural materials like fibrin, gelatin.

Extra cellular matrix components which may be visualised using these conjugates are collagen, in particular type I and type III; elastin and glucosaminoglycans, such as heparan sulphate, keratan sulphate, hyaluronic acid, chondroitin acid sulphate and dermatan sulphate. (Newly formed) fibrils of collagen type I and III are extra cellular matrix components which are preferably visualised using the labelled conjugates of the invention. Accordingly according to a preferred embodiment, the labelled conjugate of the invention, binds to collagen type I or III, preferably from human origin. According to a more preferred embodiment, the labelled conjugate of the invention, binds to (newly formed) fibrils of collagen type I or III.

Method

Therefore, in yet another aspect, the invention relates to a method for in vivo or in vitro imaging of extra cellular matrix components in a biological tissue or cell culture. The method comprises:

(i) contacting the biological tissue or cell culture with a labelled conjugate or kit or composition according to the invention; and then

(ii) visualising the imaging agent of the labelled conjugate so that an image is obtained.

The biological tissue or cell culture is tissue or cell culture in or derived from the human or the animal body. Suitable tissues or cell cultures include cardiovascular tissue, such as heart valve, myocardium, pericardium, blood vessels; connective tissue, such as bone, cartilage, tendon, ligament and fascia; intervertebral disk; lung tissue, muscle tissue, skin, stem cells, fibroblasts, skeletal muscle cells, cardiac muscle cells, blood vessel cells, blood derived and blood forming cells, bone marrow derived cells, umbilical cord cells, skin cells, nerve cells cartilage cells, bone cells, cells of embryonic origin. Preferably, the biological tissue is cardiovascular.

According to a preferred embodiment, the method of the invention is ex vivo. Ex vivo means the method of the invention is carried out on cells or tissues that are first extracted from the human or animal body.

According to another preferred embodiment, the method is performed in an engineered biological tissue.

According to another preferred embodiment, the method is performed real time. It means using a labelled conjugate of the invention, one can follow de novo formation of ECM components in cells. Preferably, the real time imaging method is performed using a labelled conjugate, which is photostable to prevent its rapid bleaching, and whose collagen binding activity is stable. Photostability is preferably assayed by incubating the labelled conjugate of the invention in a given biological sample for 24 hours. The fluorescence intensity is measured at the onset of the incubation and after 24 hours. According to a preferred embodiment, if the fluorescence intensity measured after 24 hours incubation is still between about 80 to 90% of the initial fluorescence intensity measured, the labelled conjugate is said to be photostable in a given biological sample. More preferably, the fluorescence intensity measured after 24 hours is still between about 85 and 90% of the initial fluorescence intensity measured. Preferred imaging agent is Alexa® 488 succinimide. More preferably, the labelled conjugate CNA35- AF488 is used. The peptide stability is monitored via the assessment of the binding activity of the peptide present in the labelled conjugate as defined earlier. A stable labelled conjugate is preferably a labelled conjugate exhibiting an enhanced biological activity. It preferably means that the Kd of the peptide present in the labelled conjugate for collagen, elastin or GAG is preferably 20 nM or less. More preferably, the Kd is 1OnM or less. Even more preferably, the Kd is InM or less. Kd is preferably assessed as defined earlier.

According to a preferred embodiment, the method is performed, so that images are obtained within a single cell. According to another preferred embodiment, the biological tissue or cell culture is (sequentially) contacted with two labelled conjugates of the invention (dual staining). Imaging of dual staining is attractive for visualizing de novo formation of fibrils in case of tissue engineering experiments (active remodelling). Preferably, for this application, one of the labelled conjugates is a photostable labelled conjugate as defined in the former paragraph.

According to a preferred embodiment, the method allows the imaging of newly formed collagen fibrils as extra cellular matrix components. Preferably, the collagen fibrils are from type I or III. This is an attractive application of the present invention, since it is the first time labelled conjugates are available that allows this type of imaging. Preferably, at least one of the preferred labelled conjugates of the invention as defined in the section entitled "preferred labelled conjugates" is used in this method.

In all the methods described above in this section, in a preferred embodiment, at least one of the preferred labelled conjugates presented in the section entitled " preferred labelled conjugate" and/or labelled conjugate for RMI" is used.

MRI

According to a preferred embodiment, the imaging method wherein the labelled conjugate of the invention is used is MRI. More preferably, the labelled conjugate used in a MRI method is at least one of those described in the section entitled "labelled conjugate for MRI". MRI is increasingly used in clinical diagnostics, for a rapidly growing number of indications. The MRI technique is non-invasive and can provide information on the anatomy, function and metabolism of tissues in vivo. MRI scans of tissue anatomy and function make use of the two hydrogen atoms in water to generate the image. Apart from differences in the local water content, the basic contrast in the MR image mainly results from regional differences in the intrinsic relaxation times Tl and T2, each of which can be independently chosen to dominate image contrast. However, the intrinsic contrast provided by the water Tl and T2 and changes in their values brought about by tissue pathology are often too limited to enable a sensitive and specific diagnosis. For that reason increasing use is made of MRI contrast agents that alter the image contrast following intravenous injection. The degree and location of the contrast changes provide substantial diagnostic information. Certain contrast agents are predominantly used to shorten the Tl relaxation time and these are mainly based on low-molecular weight chelates of the gadolinium ion (Gd3+). The most widely used T2 shortening agents are based on iron oxide (FeO) particles.

MRI is an emerging technology, in which non-invasive MR imaging, nano techno logy and genomics are combined for the in vivo visualization of cellular and molecular processes. Whereas conventional diagnostic imaging detects the anatomical outcome of a disease, molecular imaging is going to provide an opportunity to monitor the development of disease processes in an early stage, possibly before the onset of pathology.

Recent advances in molecular and cell biology and the increased knowledge of the mammalian genome have led to a growing understanding of the cellular mechanisms that modulate or mediate biological processes and diseases. These cellular mechanisms can be revealed by the use of specific ligands that can serve as markers for the underlying disease, and that provide opportunity to monitor progress of the disease or progress of treatment. The emerging approach of molecular imaging aims to use such molecules as markers for in vivo imaging, which should allow new powerful non invasive diagnosis based on the combination of existing imaging methods with the modern molecular description of disease. MRI is an imaging modality with a good spatial resolution, but for MRI to become suitable as a molecular imaging modality the inherently low sensitivity has to be dealt with. This may be realized by using contrast agents with a very high relaxivity, e.g. by using nanoparticles such as liposomes containing a high payload Gd-complexes or using iron oxide particles with a high payload of iron (Massoud, T. F et al, (2003) Molecular imaging in living subjects: seeing fundamental biological processes in a new light. Genes Dev 17, 545-580).

Use of a conjugate

Yet another aspect of the invention is the use of a conjugate according to the invention or a kit or a composition according to the invention in tissue engineering, diagnosis, protein or drug targeting or for the coating of surfaces.

Tissue engineering aims at the development of living tissues to repair or replace damaged, diseased or non- functional tissues and organs using in vitro, or a combination of in vitro and in vivo, techniques (Vacanti, J.P., R. Langer (1999). Tissue Engineering: The Design and Fabrication of Living Replacement Devices for Surgical Reconstruction and Transplantation. The Lancet 354(Suppl 1): 32-34). As a multidisciplinary field it converges knowledge from materials sciences, engineering, ^ (cell)biology and medicine to fabricate new tissues from living cells and a scaffolding within a simulated physiological environment.

In its most fundamental paradigm, tissue engineering entails the seeding of living cells, harvested from a donor, on a pre-shaped biodegradable support material, or scaffold, of a synthetic or natural origin. This cell-scaffold construct is generally cultured in a so-

I^ called bioreactor under conditions that favour cell expansion, tissue growth and tissue functioning. This includes the application of biological (e.g. hormones) and or biophysical stimuli (e.g. forces, electric pulses) relevant for tissue development and functioning. The scaffold provides initial anchorage and support for the cells, until they have produced and reorganised their own environment, also referred to as the extra ^ cellular matrix, to form a tissue. Ideally, tissue formation and scaffold degradation should go hand in hand to ensure and maintain the mechanical stability of the tissue. In one embodiment of the invention, conjugates or kits or compositions of the invention are used for imaging collagen, in particular collagen formation and the appearance of newly formed fibrils. In another embodiment, the conjugates or kits or compositions of 0 the invention are used for distinguishing between different types of collagen by using collagen binding peptides with different binding specificity.

In yet another embodiment of the invention, conjugates or kits or compositions of the invention are used to deliver therapeutic, diagnostic, cosmetic or other compounds to extra cellular matrix. For example the extend and type of collagen-specific diseases 5 (e.g. Ehlers-Danlos Syndrome) or fibrillin-related disorders (Marfan' s Syndrome) can be diagnosed and monitored with time when visualizing the imaging agent of the labelled conjugate in vivo. According to a preferred embodiment, the conjugate of the invention or a composition or a kit comprising it is used for the preparation of a diagnostic agent for the imaging of extra cellular matrix components.

30

In yet another embodiment, conjugates or kits or compositions of the invention are used to coat a surface to promote the attachment of extra cellular matrix components thereto. Examples are the coating or patterned coating of tissue engineering scaffolds or implantable materials or prostheses, such as vascular grafts or stents, to promote tissue ingrowth and attachment or to control collagen formation and organization into specific structures.

In another preferred use of the invention, a labelled conjugate of the invention is used in MRI. More preferably, the labelled conjugate is as presented in the section entitled "labelled conjugate for MRI".

In any of the uses described in this section, in a preferred embodiment, at least one of the preferred labelled conjugates presented in the section entitled " preferred labelled conjugate" and/or at least one of the labelled conjugate of the section entitled "labelled conjugate for RMI" is used.

In this document and in its claims, the verb "to comprise" and its conjugations is used in its non-limiting sense to mean that items following the word are included, but items not specifically mentioned are not excluded. In addition, reference to an element by the indefinite article "a" or "an" does not exclude the possibility that more than one of the element is present, unless the context clearly requires that there be one and only one of the elements. The indefinite article "a" or "an" thus usually means "at least one".

The present invention is further described by the following examples which should not be construed as limiting the scope of the invention.

Examples

Materials & Methods

General

The following proteins were purchased from Sigma-Aldrich (Zwijndrecht, the Netherlands): human placenta collagen type I (C7774), human placenta collagen type III (C4407), human placenta collagen type IV (C7521), human placenta collagen type V (C3657), rat tail collagen type I (C7661), calf skin collagen type I (C3511), bovine plasma fibrinogen (F8630), laminin from Engelbreth-Holm-Swarm murine sarcoma (basement membrane) (L2020), human aorta elastin (E6902), human plasma fibronectin (F0895), bovine serum albumin (BSA) (A3912), anti-collagen type I monoclonal antibody (clone COL-I) produced in mouse ascites fluid (C2456), and IgGl subtype anti-collagen type III monoclonal antibody (clone FH-7A) produced in mouse ascites fluid (C7805). Human joint cartilage collagen type II (ab7534) and human placenta collagen type VI (ab7538) were from Abeam (Cambridge, UK). Human Venous Saphena (HVS) cells were provided by Dr. Simon Hoerstrup (University Hospital, Zurich Switzerland). Dichlorotriazinyl fluorescein (DTAF), Oregon Green 488 carboxylic acid succinimidyl ester 5-isomer, Oregon Green 488- maleimide 5-isomer, Rhodamine Red-X succinimidyl ester 5-isomer, and CellTracker Orange CMRA were purchased from Molecular Probes (Leiden, the Netherlands). Phosphate buffered saline (PBS) contained 0.1 M phosphate buffer, 2.7 μM KCl and 137 mM NaCl, pH 7.4. PBSM was PBS containing 2 mM MgCl₂. Tris-buffered saline (TBS) contained 20 mM Tris-HCl, 150 mM NaCl, pH 7.5. Protein concentrations were determined at 280 nm using 8280nm = 33176 Cm^-1M^"1 for CNA35 and 8280nm = 57706 cm^{" 1}M^"1 for CCiI-GST. Labelling ratios were determined by measuring the absorbance at 496 nm for Oregon Green® 488-maleimide (Oregon Green® 488-maleimide 5-isomer Invitrogen, catalogus no. 06034), (ε₄₉6nm = 82800 Cm^-1M^"1) and Oregon Green® 488- succinimide (Oregon Green® 488 carboxylic acid succinimidyl ester 5-isomer, Invitrogen, catalogus no: 06147, ) (84₉6nm = 72000 Cm^-1M^"1), or the absorbance at 570 nm for Rhodamine Red™-X (Rhodamine Red™-X, succinimidyl ester 5-isomer, Invitrogen, catalogus no: R6160 (£57o_nm = 122400 Cm^-1M^"1). For proteins labelled with Oregon Green®, the absorbance at 280 nm was corrected for the contribution by the dye by subtraction of 0.12 x A4₉6nm, and for those labelled with Rhodamine Red™-X by subtraction of 0.17 x A57o_nm-

Expression, purification and fluorescent labelling ofCNΛ35

Vector pQE30CNA35 coding for the collagen binding part of the A-domain of Staphylococcus aureus collagen adhesin (amino acids 30-344) fused to an N-terminal His-tag was a kind gift from Dr. Magnus Hook (Texas A&M University, USA). The plasmid was transformed into E. coli BL21(DE3) (Novagen, Nottingham, UK). Recombinant expression and purification were performed essentially as described in Rich et al, J. Biol. Chem. 274 (1999) 24906-24913. 0.5 L LB medium containing 0.1 g/L ampicillin was inoculated with 25 mL overnight culture of E. coli BL21(DE3)/ pQE30CNA35 and grown at 37 °C/250 rpm. Protein expression was induced at OD600=0.1 by addition of 1 mM isopropyl-β-D-thiogalactosidase (IPTG). Bacteria were collected by centrifugation 6 h after induction. The cells were lysed with 30 mL BugBuster and 30 μL Benzonase (Novagen, Nottingham, UK) according to manufacturer's directions. After removal of insoluble cell debris by centrifugation at 40,000 g for 1 hr at 4 ⁰C, the supernatant was applied on a His-Bind Ni-affinity column (Novagen, Nottingham, UK) using the manufacturer's directions for washing and elution of the His-tagged protein. Protein-containing eluted fractions were pooled and buffer exchanged into 0.1 M sodium bicarbonate buffer, pH 8.4 by repeated concentration and dilution steps using an Amicon stirred ultrafiltration cell with a 10,000 MWCO membrane. The CNA35 protein was further concentrated to a final concentration of 300 μM using an Amicon Ultra-4 centrifugal filter device equipped with a 10,000 MWCO membrane. Oregon Green® 488-succinimide (15x molar excess) or Rhodamine Red™-X (7x molar excess) were added from concentrated stock solutions in DMF and the reaction was incubated for 1 h at room temperature. Another buffer exchange by repeated concentration and dilution was performed in a centrifugal filter device as above to remove unbound dye and replace buffer with PBS. The final concentration and degree of labelling were determined by measuring the absorbance spectrum of the labelled protein.

Expression, purification and labelling of human recombinant aJ-GST fusion protein.

The expression plasmid pGEX-4T-3-αi-A-dom (a kind gift from Dr. Sue Elizabeth Craig, University of Manchester, UK) encoding for an N-terminal glutathione ^-transferase (GST) fusion protein of the I domain of human integrin subunit αi (amino acids 123-336 of the mature polypeptide; see Calderwood et al, J. Biol. Chem. 272 (1997) 12311-12317) was transformed into E. coli BL21(DE3). Protein expression was done essentially as described in Tuckwell et al.(1995) J. Cell Sci. 108: 1629-1637. 0.5 L LB medium with 0.1 mg/mL ampicillin was inoculated with an overnight culture of E. coli BL21(DE3)/ pGEX-4T-3-αi-A-dom and grown at 37 °C/250 rpm. Protein expression was induced at OD600=0.6 by addition of 1 mM IPTG. Bacteria were collected by centrifugation after overnight expression. The cells were lysed using 30 mL BugBuster and 30 μL Benzonase according to manufacturer's directions and 2 mM dithiothreitol (DTT) was added. After removal of insoluble cell debris by centrifugation at 40,000 g for 1 hr at 4 ⁰C, the supernatant was loaded onto a 5-mL glutathione Sepharose column (Amersham, Roosendaal, the Netherlands) using PBS, pH 7.3 as binding buffer. Washing and elution were done according to the manufacturer's directions. Additional DTT was added (2 mM) and the purified protein was stored at - 80 ⁰C. Just before labelling, the protein was thawed and the buffer containing glutathione and DTT was replaced by PBS by repeated concentration and dilution steps using an Amicon stirred ultrafiltration cell with a 10,000 MWCO membrane (Millipore, Amsterdam, the Netherlands). The protein was concentrated to 10 - 80 μM using an Amicon Ultra-4 centrifugal filter device equipped with a 10,000 MWCO membrane. A concentrated stock of Oregon Green 488-maleimide in DMSO was added to provide a 10x molar excess of label over protein. The labelling reaction was incubated overnight at 4 ⁰C, and unreacted dye was removed by repeated concentration and dilution in a centrifugal filter device as above. The final concentration and degree of labelling were determined by measuring the absorbance spectrum of the labelled protein.

Solid-phase binding assays

All proteins were filtered through 0.2 μm polysulfone filters prior to coating onto the plates. Clear, flat-bottom Costar 96-well protein-binding plates (Corning, Schiphol-Rijk, the Netherlands) were incubated overnight on a rotating table at 4° C with 50 μL of 30 μg/mL protein in PBS. Fibrin-coated plates were prepared by adding 25 μL of 60 μg/mL fibrinogen in PBS to a well containing 25 μL of 0.015 LVmL thrombin in PBS, followed by a similar overnight incubation at 4° C. Proteins were then aspirated from the wells using an automated Wellwash AC plate washer (Thermo, Breda, the Netherlands) and the wells were rinsed 3 times with 250 μL PBSBM (1 mg/mL BSA in PBSM) at 21° C. Wells were blocked with 100 μL of 50 mg/mL BSA in PBS for 3 h at 37° C, aspirated and rinsed 3 times with 300 μL PBSBM. 50 μL of a solution of the probe in PBSM was added to each well and incubated 3-4 h at 37° C. Wells were aspirated and washed 3 times with 350 μL PBSBM, after which the fluorescence was read with an automated Fluoroskan Ascent FL plate reader (Thermo) (excitation: 485 nm; emission: 538 nm). Each binding experiment was done at least in triplicate. The average fluorescence reading of empty wells was subtracted from the raw fluorescence data. Normalized intensities were calculated by dividing the intensity measured for a probe on a particular target by the average intensity measured for that probe on 3 human collagen type I control wells present in each lane of the 96-well plate.

Collagen visualization in cell culture

HVS cells, grown to passage 5 with regular cell culture methods (Schnell et al., Thorac. Cardiovasc. Surg. 49 (2001) 221-225), were plated on 25-mm #1 round glass coverslips (Menzel-Glaser, Braunschweig, Germany) in 6-well Nunclon Delta Surface cell-culture disposable plates (Nunc 140675 Roskilde, Denmark) in Advanced Dulbecco's Modified Eagle Medium with 10% fetal bovine serum, 1% Glutamax, and 0.1% gentamycin. Nine days after plating, cells were incubated 45 min in medium containing 10 μM CellTracker Orange CMRA, rinsed 2 times with medium, incubated for 90 min in medium containing 0.5 μM Oregon Green 488-labeled CNA35 (CNA35- OG488), 2.5 μM Oregon Green 488-labeled αil-GST (αiI-GST-OG488), or 7.5 μM DTAF, and finally rinsed once with medium. Cells were analyzed using a Zeiss LSM 510 confocal laser scanning microscope (CLSM) using a 63x oil immersion objective and the following excitation/emission wavelengths: 488 nm/Bandpass filter 505-530 nm for Oregon Green-labelled probes and DTAF; and 543 nm/Longpass filter 585nm for CellTracker Orange.

Collagen visualization in tissue

4 μm frozen sections of rat tibialis myotendonous junction on Polysine microscope slides (BDH, Dorset, UK) were incubated at room temperature in PBS containing a final concentration of 1 μM CNA35-RRX and 20 μM DTAF for 2 hr, then rinsed 3 times with equal volumes of PBS. PVA-DABCO mounting medium (Fluka, 10981 Buchs, Switzerland) was applied to sections, cover slip was applied, and the tissues were allowed to dry 10-15 min before analysis by CLSM as described above. For histology, similar sections were air dried for 20 min and permeabilized for 5 min using acetone with subsequent drying for 5 min. A mix of 1 μM CNA35-OG488 and the collagen type I antibody (1:2000 diluted) or the collagen type III antibody (1:4000 diluted) was prepared in PBS. Sections were incubated with this mix for 45 min at room temperature, washed three times in PBS, incubated with a secondary antibody (GaM-IgG- Alexa 555, Molecular Probes, diluted 1:400 in PBS) for 45 min at room temperature and after a final wash with PBS, embedded in Mowiol mounting medium (Calbiochem, Nottingham, UK) and covered with a glass coverslip. Images were obtained using a NIKON E800 fluorescence microscope (UVIKON, Bunnik, the Netherlands) coupled to a Basler AlOlC progressive scan color CCD camera.

The methodology of each of four different in vivo applications is described below in the following sequence: (I) mouse carotid artery (figures 7, 8) , (II) engineered cardiovascular constructs (figure 9), (III) time- lapse monolayer (figure 10) and (IV) dual staining monolayer (figure 11). Results are given in examples 6 and following.

Tissue and cell preparation.

(I)A Swiss mouse was euthanized by a mixture of O₂/ CO₂ gas. Carotid arteries were isolated and mounted on two glass micropipettes in a perfusion chamber filled with 10 ml phosphate buffered saline (PBS; Sigma, USA) containing the fluorescent probe(s) at the desired concentration. A transmural pressure of 80 mmHg was applied in order to mimic physiological pressure. Experiments were approved by the local ethics committee on the use of laboratory animals. Procedures followed were in accordance with the institutional guidelines.

(II) Cell culture: Human Venous Saphena (HVS) myofibroblasts were obtained from patients and expanded using regular cell culture methods (Schnell et al, 2001). The medium used to culture these cells consists of advanced Dulbecco's Modified Eagle Medium

(DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Biochrom,

Germany), 1% L-glutamax (Gibco, USA) and 0.1% gentamycin (Biochrom, Germany). The medium used for tissue culture contained 0.3% gentamycin and additional L-ascorbic acid 2-phosphate (0.25 mg/1 ; Sigma, USA).

Scaffold preparation and sterilization: Rectangular shaped (0.5 x 2.5 x 1.0 mm³) non-woven polyglycolic acid (PGA) scaffolds (density 72.76 mg/cm³; Cellon, Luxembourg) were coated with a 1% (w/v) poly-4-hydroxybutyrate (P4HB; Symetis Inc., Switzerland) in tetrahydrofuran (THF; Merck, Germany) (Hoerstrup et al., 2000). After vacuum drying for 48 hours the scaffolds were placed in 70% ethanol for 4-5h to obtain sterility. The ethanol was allowed to evaporate overnight and the scaffolds were washed three times in phosphate buffered saline (PBS, Sigma, USA). The scaffolds were placed in tissue culture medium overnight to facilitate cell attachment.

Cell seeding and tissue culture: myofibroblast cells at passage 7 were seeded on these scaffolds with fibrin gel as a cell carrier, which resulted in cardiovascular tissue constructs (MoI et al, 2005). The cells were seeded at a concentration of approximately 40 million cells per cm³ and the constructs were subsequently cultured in tissue culture medium (at 37° C and 5% CO₂). Medium was changed every 3-4 days.

(Ill and IV) For the time-lapse experiment cells were grown to passage 6 and plated on round glass coverslips (Menzel-Glaser, Germany) at a concentration of 10000 cells/cm². The cells were cultured in regular cell culture medium.

Staining procedure:

(I) SYTO44, eosin (Molecular Probes, the Netherlands), and CNA35-OG488 were used as specific fluorescent markers for DNA/RNA, elastin, and collagen, respectively. All probes are excitable with two-photon microscopy and exhibit broad emission spectra with maxima at 480nm (SYTO44), 520nm (collagen probe), and 560nm (eosin), respectively. Labelling solutions were CNA35-OG488 [1.0 μM] or a mixture of SYTO44 [1.5 μM], collagen probe [1.0 μM], and eosin [0.25 μM] in 10 ml of PBS. Labelling solutions were applied luminally and abluminally. (H) Cell Tracker Blue CMAC (CTB; Molecular Probes, the Netherlands), and CNA35-OG488 were used as specific fluorescent markers for cell cytoplasm and collagen, respectively. CTB is excitable with two-photon microscopy and exhibits a broad emission spectrum at 466nm. The PGA scaffold exhibits a broad autofiuorescence emission spectrum above 560nm and below 500nm. Staining solutions were as follows: CTB [15.0 μM] and CNA35-OG488 [3.0 μM] in 3D culture medium. The CTB solution was applied for 5 hours and CNA35-OG488 for 24 hours. The samples were then rinsed for approximately 12 hours in 3D culture medium.

(Ill) Cell Tracker Orange CMRA (CTO; Molecular Probes, the Netherlands), and CNA35- AF488 (CNA Alexa Fluor 488) were used as specific fluorescent markers for cell cytoplasm and collagen, respectively. CTO and CNA35-AF488 exhibit emission spectra at 572nm and 520nm, respectively. 24 hours after seeding cells were incubated at 37 ⁰C for 45 min in medium with CTO [4 μM]. The coverslip with cells attached was transferred to a bioreactor which enables control of CO₂ (5%) and temperature (37 ⁰C) and which fits on the confocal microscope. The bioreactor was filled with cell culture medium containing CNA35-AF488 [0.475 μM] and to minimize evaporation the medium was covered with a coating of mineral oil (Sigma, STAD).

(IV) CNA35-OG488 was used to stain the collagen two days after plating the cells, whereas CNA35-RRX was used as specific fluorescent marker for collagen present three days after plating. CNA35-RRX exhibits a broad emission spectrum at 580nm. 48 hours after seeding the monolayer culture was incubated with culture medium containing CNA35-AF488

[0.475 μM]. 96 hours after seeding the monolayer culture was incubated with culture medium containing CNA35-RRX [0.475 μM]. The coverslip with cells attached was cultured at 5% CO₂ and 37⁰C.

Microscopy setup:

(I) For imaging of the collagen probe in a mouse carotid artery, two-photon laser scanning mode (TPLSM) was used (Van Zandvoort et al, 2004). A pulsed Ti: Sapphire laser was tuned and mode-locked at either 800nm (fluorescence) or 840nm for second harmonic generation. A 6Ox water dipping objective with a 2.0 mm working distance was used (numerical aperture 1.0, Nikon). Three photo multiplier tubes (PMT) were used to detect the fluorescent signal. For the SHG experiment, PMT 1 was set at 400-500nm in order to detect SHG and PMT 2 was set to 500-560nm for detecting the CNA35-OG488. For imaging of the used combination of three fluorescent markers, the PMTs were tuned at (1) 470 - 480 nm, (2) 500 - 520 nm, (3) 590 - 610 nm, corresponding to (parts of) the emission spectra of the used fluorescent markers. Separate images were obtained from each PMT (coded blue, green, and red, respectively) and combined into a single image. No additional image processing was performed.

(II) A similar setup was used for the imaging of the tissue engineered constructs was used. PMT 1 was set at 450-500nm in order to detect CTB signal and the autofluorescence from the scaffold, PMT 2 was set at 510-530nm for detection of the CNA35-OG488 signal and PMT 3 was set to detect all wavelengths greater than 600nm to detect autofluorescence from the scaffold. The signals detected from the scaffold and cells were relatively weak and consequently were slightly improved. (III) For both the time- lapse (figure 10) and the high magnification (figure 11) experiments, cells were analyzed using a LSM510 confocal laser scanning microscope (Carl Zeiss, Germany). A 63x oil immersion objective (numerical aperture 1.4; Carl Zeiss, Germany) was used. The tissue was excited in multi-track mode at the wavelengths 488nm (Helium-Neon laser) and 543nm (Argon laser). Two PMT's were used to detect the fluorescent signals. The filters for PMT 1 were set at 505-530nm to detect the signal originating from CNA35-AF488 and PMT 2 was set to detect all wavelengths larger than 585nm for the signal originating from CTO. The monolayer culture was followed for approximately 52 hours at the same focal position.

(IV) For the dual staining experiment, the same setup was used as in HI. The filters for PMT 1 were set at 505-530nm to detect the signal originating from CNA35-OG488 and PMT 2 was set to detect all wavelengths larger than 560nm for the signal originating from CNA35-RRX.

Below, the preparation of labelled conjugate of the invention suited for MRI is described. This labelled conjugate is subsequently characterized.

Preparation of a labelled conjugate for MRI imaging (example 11 and following) General. Unless stated otherwise, all reagents and chemicals were obtained from commercial sources and used without further purification. Dichloromethane (DCM) was obtained by distillation from P2O5. l,2-distearoyl-sn-glycero-3- phosphoethanolamine-7V-[amino(polyethylene glyco 1)2000] (NH₂-PEG-DSPE), 1,2- distearoyl-sft-glycero-3-phosphocholine (DSPC), 1 ,2-distearoyl-sn-glycero-3- phosphoethanolamine-7V-[methoxy(polyethylene glyco 1)2000] (PEG-DSPE) and 1,2- dipalmitoyl-sft-glycero-3-phosphoethanolamine-Λ/-(lissamine rhodamine B sulfonyl) (rhodamine-DPPE) were purchased from Avanti Polar Lipids (Albaster, USA). Gd- DTPA-bis(stearylamide) was purchased from Gateway Chemical Technology (St. Louis, MO, USA). Trityl-protected cysteine (Tr-Cys(Tr)-OH) was obtained from Bachem (Bubendorf, Switserland). Trityl-protected succinimidyl-activated cysteine (Tr-Cys(Tr)-OSu) was prepared according to a literature procedure (van Baal I et al, (2005), Angew Chem Int Ed Engl 44, 5052-5057). UV- Vis spectra were recorded on a Shimadzu Multispec 1501 spectrometer. Fluorescence spectra were obtained on an Edinburgh Instruments FS920 double-monochromator spectrometer. Primers used for all the cloning procedures were supplied by MWG (Ebersberg, Germany). Synthesis of cysteine- functionalized l,2-distearoyl-sn-glycero-3- phosphoethanolamine-iV- [amino(polyethylene glycol)2000] (Cys-PEG-DSPE).

NH₂-PEG-DSPE (1) (100 mg, 35.8 μmol) was dissolved in DCM (1 mL) under an atmosphere of argon. Tr-Cys(Tr)-OSu (30 mg, 43 μmol) and triethylamine (10 μL, 71 μmol) were added to the solution. The reaction was proceeded overnight at room temperature. The solution was concentrated under reduced pressure and the crude product was dissolved in CHCl₃ (0.5 mL). The crude product was purified by column chromatography (silica, CHCVMeOH, 19:1 v/v —» 9:1 v/v) and trityl-protected Cys- PEG-DSPE (2) (44 mg, 13 μmol) was obtained in 36% yield. Compound 2 was dissolved in triethyl silane (25 μL, 158 mmol). Subsequently, a solution of trifluoroacetic acid (1 mL) and DCM (1 mL) was added. The obtained solution was stirred for 2 hours at room temperature. The solution was concentrated under reduced pressure and the crude product was precipitated in diethyl ether. The product was filtrated and dried under reduced pressure to give Cys-PEG-DSPE lipid (3) (34 mg, 12 μmol, 92%). Compounds 1-3 were analyzed in detail with RP-HPLC and MALDI-TOF (see Supporting Info).

Plasmid constructs.

The EYFP gene was amplified by PCR from vector pE YFP-Nl (Clontech) using the primers 5'-GTG GTC ATA TGG TGA GCA AGG GCG AG-3' and 5'-GTG GTG

AAT TCC TTG TAC AGC TCG TCC ATG C-3' . The CNA35 gene was amplified from pQE30CNA35 (a kind gift from Dr. Magnus Hook , Texas A& M University,

USA) using the primers 5'-GTG GTC ATA TGG GAT CCG CAC GAG ATA TTT C-

3' and 5'-GTG GTT GCT CTT CCG CAT GCC TTG GTA TCT TTA TCC TGT TTT AAA AC-3'. The PCR products and the pTXBl vector were double-digested with the restriction endonucleases Nde I and Sap I (CNA35) or Nde I and EcoR I (EYFP) followed by ligation of the amplified DNA fragments in the open plasmids. DNA sequencing using T7 promoter and intein specific reversed primers (New England

Bio labs) confirmed the correct in- frame fusion of the proteins with the intein sequence.

Protein production.

The expression plasmids were transformed into E. coli BL21 (DE3) cells. Bacteria were grown in LB medium containing 100 μg/mL ampicillin at 37 °C and 250 rpm to an optical density (OD_OOO m) between 0.6-0.8. Protein expression was induced with 0.5 mM IPTG and the cultures were incubated for 4 hours at 37 °C. Cells were harvested by centrifugation for 5 minutes at 10,000 g at 4 ⁰C. The supernatant was removed and the cell pellet was resuspended using the BugBuster protocol (Novagen). After incubation for 20 minutes at room temperature, the cell suspension was centrifuged at 16,000 g for 20 minutes at 4 ⁰C. The supernatant was directly applied to a column of chitin beads, equilibrated with 10 column volumes of column buffer (20 mM sodium phosphate, 0.1 mM EDTA, 0.5 M NaCl, pH 8). The column was washed with 10 volumes of column buffer after which the column was quickly flushed with 3 column volumes of cleavage buffer (20 mM sodium phosphate, 0.1 mM EDTA, 0.5 M NaCl, pH 6) containing 50 mM sodium 2-mercaptoethanesulfonate (MESNA) and incubated overnight at room temperature. Elution fractions were collected and pooled after which the cleavage step was repeated to gain more thioester terminated proteins. The proteins were buffer-exchanged into 10 mM HEPES, 135 mM NaCl, pH 8.0 (HBS) using Amicon ultra centrifuge tubes (MWCO 10 kDa). The EYFP protein with a C-terminal MESNA thioester (EYFP-MESNA) and CNA35 protein with a C-terminal MESNA thioester (CNA35-MESNA) concentrations were determined by UV-vis using 8514 _nm = 84.000 M^-1Cm^"1 and ε₂₈o nm = 33167 M^-1Cm^"1 respectively. (Patterson, G.; et al (2001) Fluorescent protein spectra. J Cell Sci 114, 837-838). 1 L is. coli culture typically yielded 20 mg of pure EYFP-MESNA and 40 mg of CNA35-MESNA.

Liposome preparation.

Liposomes were prepared by lipid film hydration as described previously (Mulder, W. J. et al. (2004) A liposomal system for contrast-enhanced magnetic resonance imaging of molecular targets. Bioconjug Chem 15, 799-806).A mixture of DSPC (37 μmol), Gd-DTP A-bis(stearylamide) (25 μmol), cholesterol (33 μmol), PEG-DSPE (2.5 μmol), rhodamine-DPPE (0.1 μmol) and Cys-PEG-DSPE (2.5 μmol) were dissolved in CHClβ/MeOH 1 : 1 (v/v) and concentrated under reduced pressure at room temperature. The obtained lipid film was hydrated in HBS buffer (4 mL). This dispersion was extruded five times at 65 ⁰C through polycarbonate membrane filters with pores of 100 or 200 nm. Phospholipid concentrations were determined by phosphate analysis according to Rouser (Rouser, G. et al.(1970) Two dimensional then layer chromatographic separation of polar lipids and determination of phospholipids by phosphorus analysis of spots. Lipids 5, 494-496). The amount of lipids per liposome was calculated using a lipid surface area of 0.6 nm² (Strijkers, G. J. et al. (2005) Relaxivity of liposomal paramagnetic MRI contrast agents. Magma 18, 186-192)

and assuming unilamellar liposomes.

Ligation of CNA35 to cysteine-liposomes.

Native chemical ligation of cysteine-liposomes (180 μM Cys-PEG-DSPE) with CNA35 thioester (50 μM) was performed in HBS pH 8, 100 mM MESNA for 24 hours at 20 ⁰C. Ultracentrifugation of CNA35-liposomes was performed in a Kontron Centrikon T-2060 ultracentrifuge with a TFT 70.38 rotor for 1 hour at 270.000 g and 20 ⁰C. The obtained liposomal pellet was resuspended in HBS. Pellet and supernatant were analyzed using SDS-PAGE. Protein concentration was determined with the Quant-iT Protein Assay Kit (Invitrogen) according to manufacturer instructions.

Collagen binding assay (figure 14).

96 wells Corning EIA/RIA Microplates were coated overnight at 4 °C with 2.5 μg/ well (45 μl) rat tail collagen type I (Sigma, C7661) in TBS (50 mM Tris, 150 mM NaCl, pH 7.5). After overnight incubation the plates were blocked with 100 μl TBS containing 5 % (w/v) skim milk powder for 2 hour at room temperature. After washing the plates 3 times with 300 μl TBS, the plates were incubated with CNA35-liposomes, or non modified liposomes in HBS supplemented with 5% (w/v) skim milk powder for 3 hour at room temperature. Plates were washed 5 times with 50 mM Tris, 500 mM NaCl, pH 7.5, 0.1 % (v/v) Tween-20 and subsequently washed 2 times with TBS with 0.1 % (v/v) Tween-20. The fluorescence of the rhodamine-containing liposomes was measured in triplo on a Thermo Fluoroskan Ascent FL plate reader.

Fluorescence of wells used for MRI measurements (figure 12)

Wells in a 96-wells plate were coated with collagen or blocked with milk powder and subsequently incubated with a liposome suspension. Wells C and D were coated with collagen and incubated with CNA35-liposomes. Wells A, B and E were negative controls. Well A was blocked with milk powder and incubated with CNA35-liposomes. These should not bind to the well, because no collagen was present. Well B was coated with collagen, but incubated with bare liposomes. These liposomes should not bind to the collagen, because they did not contain any CNA35 proteins on their surface to bind collagen. A final negative control was well E, which was not coated or blocked and not incubated with a liposome suspension. It was just filled with HBS. After washing the unbound liposomes away, collagen binding by the liposomes was measured by fluorescence.

Tl weighted MR image (figure 13)

A Ti weighted MR image was prepared of all five wells by using an inversion recovery sequence and fitting equation 1.1 through the data. An inversion recovery sequence with 10 different inversion times ranging from 15 ms to 5000 ms and a repetition time of 15000 ms was used to generate a Ti-weighted image. Pictures of wells C and D show a decrease of Ti near the edges of the wells, caused by the selective binding of Gd³⁺-containing CNA35-liposomes. The Gd³⁺-ions cause the Ti of the surrounding water protons to decrease. The decrease of the Ti near the edges of the negative control wells was smaller and the distance of the effect was much smaller.

Example 1 Labelling of collagen binding peptides

CNA35 with His-tag (334 amino acids, SEQ ID NO. 1) and CCiI-GST (440 amino acids, SEQ ID NO. 2) were recombinant Iy expressed in E. coli and purified in one step using the His-tag attached to CNA35 or the GST-tag attached to OC₁I. We obtained ~ 100 mg of CNA35 and ~ 30 mg of CCiI-GST from 1 L of E. coli culture. SDS-PAGE analysis showed a purity of >95% and mass spectrometry (ESI-MS) confirmed the expected molecular weight for these fusion proteins. Because the CNA35 protein does not have cysteines that are available for conjugation with fluorescent dyes, we used amine-reactive succinimidyl ester dyes. Labelling CNA35 with Oregon Green 488 resulted in close to 3 fluorescent labels per protein, whereas rhodamine-labelled CNA35 (CNA35-RRX) contained on average 1.4 labels per protein. Preliminary solid-phase collagen binding assays using Oregon Green 488- labeled CCiI showed much lower signal intensities than those of fluorescent Iy labelled CNA35 or cciI-GST. We therefore used the cciI-GST fusion protein. Reaction with Oregon Green 488 maleimide yielded a protein with an average of 3 fluorescent labels per protein, which is due to the presence of additional cysteines in the GST domain. The yield was greater than 90 % after the labelling reaction and removal of unbound dye.

Example 2 Binding properties of labelled conjugates

To determine whether the fluorescently labelled proteins were still able to bind to collagen, binding to human collagen type I was studied using a solid-phase binding assay (Figure 1). CNA35-OG488 shows biphasic binding behavior consisting of high affinity binding with an apparent K_A of ~ 0.5 μM followed by additional binding at higher probe concentrations. Binding of the CCiI-GST-OG488 probe to human collagen type I was considerably weaker with an apparent dissociation constant of ~ 50 μM. Multiple binding sites have previously been observed for binding of CNA to both type I and type II collagens (Rich et al, J. Biol. Chem. 274 (1999) 24906-24913). Apparent Kd S ranging from 20 nM to 30 μM have been reported, depending on the specific type of collagen. We therefore conclude that fluorescent labelling of CNA35 using amine- reactive dyes does not perturb its affinity for collagen in a major way. The apparent K_A observed for binding of CCiI-GST-OG488 to human type I collagen is somewhat higher than reported by other studies for this domain, which may be due to steric interactions.

Example 3 Relative specificity of protein conjugates and DTAF

Next we determined the binding specifities of both protein-based probes and DTAF for a variety of collagens and other ECM proteins by comparing the fluorescence intensities in solid-phase assays at a fixed probe concentration (Figure 2). These measurements allow one to compare the relative specifities of the three probes. Because the coating efficiency may be different for different ECM proteins, these assays cannot be used to directly determine the relative affinity of a single probe for different ECM proteins. The ECM proteins tested here were chosen because they are the major proteins likely to be present in the ECM of tissue-engineered samples for which these probes were developed. In addition, fibrin is a common material used for seeding HVS cells on scaffolds for artificial heart-valves. Thus, it is crucial that any probe developed for this purpose does not bind to fibrin. CNA35-OG488 bound relatively well to all types of collagen that were tested, but did not bind to any of the other ECM proteins. αiI-GST-OG488 also showed specific binding to collagens, although low amounts of aspecific binding were observed for some of the other ECM proteins and BSA. As expected for an amine-reactive dye, DTAF reacted with all proteins tested (including BSA) and did not show any specificity for collagen. Both CNA35-OG488 and αiI-GST-OG488 showed slightly stronger binding to human collagen type I than to rat or bovine collagen type I, although it cannot be ruled out that the coating was more efficient for human collagen type I. Among the human collagens, CNA35-OG488 demonstrated a higher comparative specificity for type I than did αil- GST-OG488, which also showed relatively strong binding to many of the other human collagens. Again, DTAF did not show any specificity.

Example 4 Visualisation of collagen formation in cell culture

The performance of the probes in monitoring collagen formation around cultured cells was studied using Human Venous Saphena (HVS) cells. These cells are often used in tissue-engineering of heart valves and are known to produce human collagen type I . CLSM pictures of live HVS cells incubated with 0.5 μM of CNA35- OG488 very clearly showed abundant formation of collagen fibrils around each cell in the monolayer (Fig. 3A). Some fluorescence was also observed in the presence of 2.5 μM αiI-GST-OG488, but the intensity was much less than with CNA35-OG488 and its distribution also suggested less specific binding (Fig. 3B). The DTAF showed very little extra cellular fluorescence at all, showing that DTAF is unsuitable for detection of these early events in collagen formation (Fig. 3C).

Example 5 Visualisation of collagen formation in tissue sample

To compare the performance of these probes in tissue samples, we prepared frozen slices of rat myotendonous junction, as these are known to contain both type I and type III collagens, as well as myo fibers that do not contain intracellular collagen.

Tissue was incubated with a mixture of CNA35-RRX and DTAF to directly compare their specificities. DTAF staining of non-collagenous structures within muscle cells was seen, with a much lower signal inside the collagen-rich tendon. Tendon and the areas interlaced between muscle cells represents endomysium and basal lamina and were stained for CNA35. No intracellular fluorescence was obtained. These areas are known to contain substantial amounts collagen type I and III. Similar slices incubated with αiI-GST-OG488 showed large amounts of intracellular fluorescence. In next examples, several applications of among other the collagen specific probe CNA35- OG488 are illustrated. These examples demonstrate the great potential of this probe with regard to visualization of collagen architecture, synthesis and remodelling in a variety of culture systems. The probe shows improved properties over existing techniques such as SHG and enables real-time monitoring of collagen fibrillogenesis.

Example 6 Visualisation of collagen fibers in mouse carotid artery

In order to compare the probe with existing three-dimensional imaging techniques for (viable) tissue, 2-photon microscopy was used to image a mouse carotid artery. The artery was stained to visualize the cells, the elastin network and the collagen network. Figure 7A and 7B show that the CNA35-OG488 probe reveals much more detail than the backscattered SHG signal, which suggests a superior resolution of this collagen probe compared to the SHG signal. Furthermore, the perfect alignment of the SHG and probe signal confirms the specificity of the CNA35-OG488 probe. In addition, the collagen CNA35-OG488 probe showed a stronger signal than SHG with increasing depth in the tissue (figure 8). The first scans of figure 8 show the adventitia (outer layer), which consists primarily of collagen fibrils (light grey or yellow) and fibroblast nuclei ( dark grey or blue). Then the media (middle layer) is scanned, consisting mainly of elastin ( middle grey or red) and smooth muscle cell nuclei ( dark grey or blue). The last scans show the intima (luminal layer), which consists mainly of endothelial cells ( dark grey or blue) oriented perpendicular to the longitudinal axis of the artery (references). Surprisingly, in figure 8, small collagen fibrils are observed in between the endothelial cells in the intima with the CNA35-OG488 probe. These observed collagen fibers in between the endothelial cells were not observed previously with the use of SHG (Zoumi et al, 2004; Boulesteix et al, 2005; Schenke-Layland et al., 2005; Schenke-Layland et al., 2006). This possibly indicates that these fibers consist of a different type of collagen and do not exhibit SHG or that these fibers do not exhibit a very strong SHG signal. Previous studies have shown that forward scattered SHG signal is a stronger signal than the backward scattered SHG signal (Williams et al, 2005; Cox et al., 2003; Campagnola et al., 2002). Forward scattered SHG enables detailed visualization of collagen architecture within various tissues (Williams et al., 2005; Konig et al., 2005), however, with increasing tissue thickness (>0.5mm) imaging of the matrix structures relies predominantly on the backscattered signal (absorption of forward scattered SHG signal). A possible drawback to the use of fluorophores is the possibility of photo-bleaching, whereas intrinsic tissue properties like autofluorescence or SHG do not exhibit photobleaching.

Example 7 Visualisation of collagen fibrils in tissue engineered construct

The same probe was also applied on four weeks old tissue engineered cardiovascular constructs in order to look at developing collagen fibrils within live young tissues. It's known from biochemical assays and histology that the tissue engineered samples exhibit high amounts of collagen after approximately 3-4 weeks of culture and in these samples polyglycolic acid (PGA) scaffold is still present (MoI et al, 2005). The PGA scaffold exhibits a distinct and broad autofluorescence spectrum, which enabled us to distinguish between the three predominant constituents (i.e. cells, collagen and scaffold). Figure 9 shows a detailed view of the constituents of the engineered tissue constructs: cells (blue), collagen fibrils (green) and PGA/P4HB scaffold in purple. The collagen fibrils are closely associated with the surface of the cells. Figure 9 shows that the collagen fibrils are relatively thin compared to fibrils in mature native artery (figure 6 versus figure 7) and that these fibrils exhibit a very diffuse architecture compared to native fibrils. Mature native fibrils are bundled in fibers rather than individual fibrils. Collagen fibrils typically have a diameter in between 10 - 500 nm (Fratzl, 2003; Ushiki, 2002) and bearing in mind that the visualized fibrils are very young, these fibrils most likely have a diameter that is smaller than the optical resolution of 0.2 - 0.3 μm. The benefit of the fluorescent probe used here is that it allows subresolution collagen fibers to be imaged. Several successful attempts have been performed to study ECM remodelling in time by using confocal reflection microscopy in 3D fibrin lattices and fibroblast populated collagen lattices (Hartmann et al., 2005; Brightman et al., 2005; Voytik-Harbin et al, 2001; Wolf and Friedl, 2005). However, with confocal reflection microscopy alone it is not possible to distinguish between different constituents, whereas with multiple fluorescent labels and laser scanning microscopy this is possible. Furthermore, collagen assembly by cells has been quantified by using FTTC labelled rat-tail collagen (Johnson and Galis, 2003). This technique can be used to monitor the remodelling and structure of exogenous collagen, however, it can not be used to study newly synthesized endogenous collagen. Example 8 Real time visualisation of collagen fibril formation in vivo

The same CNA35-OG488 probe also enables real-time visualization of collagen fibrillogenesis. This was demonstrated by seeding myofibroblast cells on top of a glass coverslip and studying these cells over time in the presence of the probe. The CNA35 peptide fused to His-tag was coupled to a more stable fluorescent dye (Alexa® 488 succinimide (Alexa® fluor 488, Invitrogen, catalogues No: A20000) to prevent rapid bleaching of the probe (CNA35-AF488). The coupling to the Alexa dye was carried out the same way as for the other dyes.

Images of monolayer culture of myofibroblast cells using the CNA35-AF488 were studied in monolayer culture to demonstrate the possibilities of the CNA35-AF488 probe in vital imaging of collagen fibril formation. The monolayer culture was left untouched for 24 hours after plating and was studied subsequently for 52 hours (76 hours in total). The microscope settings were based on a 48 hours old culture, to prevent saturation of the signal with increase in collagen content and to detect sufficient detail early on in the culture. Figure 10 shows five scans at the same spatial and focal position selected from the period of 52 hours. The images show that the cells are moving around and actively secreting collagen into their surroundings. The first image shows a limited amount of collagen present, which is either newly formed or attached to the cell and left over from the previous culture. Over time the collagen content increases and fiber like structures start to appear. Some cells move around with the collagen fibers closely attached to their surface and form larger collagen aggregates, other cells release the collagen into the surroundings. Most importantly the images show the ability to monitor changes in collagen architecture over time.

The images presented in figure 10 show that the cells are actively secreting collagen into their surroundings and over tune the collagen content increases. One of the drawbacks of this technique compared to e.g. SHG, an intrinsic tissue property, is that the probe binds directly to the collagen. By binding to the collagen molecule or fibril, the probe potentially prevents proper fibril formation. Still the probe enables us to study collagen secretion in time. Tune lapse confocal reflection microscopy has been used to study collagen fibrillogenesis in vitro. Nevertheless, these experiments were performed on reconstituted collagen matrices (no de novo synthesis), whereas our technique enables real time visualization of collagen fibrillogenesis in both 2D and 3D cultures. Real time elastin fibrillogenesis and remodelling has been studied extensively with confocal laser scanning microscopy, using GFP and timer transfected cells (Kozel et al, 2006; Czirok et al, 2006). The probe presented here possesses similar properties with respect to real-time visualization of collagen fibrillogenesis.

Example 9 Visualisation of in vivo collagen fibril formation in one single cell

When we zoom in to a single cell (Figure 1 IA), which is stained for the cytosol and the surrounding collagen, we can see very detailed microstructures originating from the cell. In medium grey (red) the cell cytosol is visible, which clearly outlines the cell nucleus and the cell organelles. In dark grey (green) very small collagen fibers and fiber networks are distributed over the entire imaging plane, however, some fibrils are very closely associated with the cell surface and seem to originate from the cell surface. Furthermore, the probe is also attached to very small round structures within the cytosol. This shows that the probe enables detailed visualization of collagen structures at high magnifications and provides new opportunities to visualize collagen fibrillogenesis at the cellular level. Using even higher magnification lenses enables an even more detailed visualization of collagenous structures in and around the cell.

Example 10 Dual staining monolayer

Figure 1 IB shows a detailed image of a monolayer culture which was stained with CNA35-OG488 48 hours after seeding and with CNA35-RRX 72 hours after seeding, medium grey both. A color image was submitted wherein both staining are seen as green and red respectively. In the intervening period, the cells were kept in medium without any probe. The image shows that the probes predominantly colocalize (dark grey), even though they are applied at different time points, and that at certain locations medium grey (either green or red) dominates. The images required after this procedurejequired somewhat higher power for the 488 nm laser, which indicates a loss of fluorescence (possibly by dissociation).

In case of the double staining procedure, ideally the green probe has stained the early collagen, whereas the red probe stains the newly formed collagen. Ideally dual staining will result in similar studies for collagen fibrillogenesis as were performed for elastin (Czirok et al., 2006). However, the binding between the CNA probe and the collagen is an equilibrium and when medium without any probe is applied a new equilibrium will form. This will effectively decrease the concentration of probe bound to collagen. When the red probe is added a new equilibrium will form, however, some of the green probe will be replaced with red probe. In case of active remodelling processes the reversible binding of the probe is not desirable, as we cannot be sure if the collagen that is stained red is entirely newly formed. However, the experiment confirms that the probe binds relatively weak and reversible to the collagen fibrils, which reduces the risk of preventing proper collagen fibril formation and makes the probe more suitable for real-time monitoring of collagen fibrillogenesis.

Example 11 CNA35-liposomes specifically bind to collagen (figure 14)

The rhodamine fluorescence of the CNA35-functionalized liposomes was used to study their binding to a collagen-coated 96 well plate. Figure 14 shows specific binding of these liposomes to rat tail collagen type I at low nanomolar concentrations. Control experiments using non-functionalized liposomes or plates without collagen did not show a similar signal. A fit of the binding curve using a simple 1 : 1 binding model yielded a Kd of 3 ± 1 nM, representing a 150-fold increase in affinity compared to the protein itself. These results show that CNA35 conjugated to liposomes via native chemical ligation is fully active in binding collagen. Whether the observed increase in collagen affinity is due to multiple, simultaneous interactions between probe and collagen or mainly a statistical effect due to the presence of -100 copies of protein per liposome remains to be investigated.

The collagen binding of the CNA35-liposomes in all these five wells was subsequently visualized by MRI.

Example 12 use of CNA35-liposomes in MRI (figures 12 and 13)

Solid phase binding assays using a fluorescent plate reader were carried out using the CNA35-liposome conjugates bound specifically to collagen coated 96-well plates. The results are presented in figure 12 and show that the CNA35-liposomes had selectively bound to the collagen-coated wells, but the fluorescent signal of the negative controls A and B was significantly higher than of the well in which no liposomes had been present. These images show a clear decrease in Tl relaxation time near the collagen surface in wells incubated with CNA35 -collagen, whereas no effects were observed in control experiments with non-functionalized liposomes. Although the effect of the bound liposomes could not be observed over a large distance, a small layer of water with a decreased Ti could be clearly seen. We therefore conclude that MRI successfully visualized the selective binding of CNA35-liposomes to collagen.

References

K.L. Billiar and M.S. Sacks, Biaxial mechanical properties of the native and glutaraldehyde- treated aortic valve cusp: Part H-A structural constitutive model. Journal of Biomechanical Engineering, 122 (2000), 327-35.

Dahl S.L., Rucker R.B. andNiklason L.E., Effects of copper and cross-linking on the extra cellular matrix of tissue-engineered arteries. Cell Transplant, 14 (2005), 861 - 868.

De Campos Vidal B., Image analysis of tendon helical superstructure using interference and polarized light microscopy. Micron, 34 (2003), 423 - 432.

Petroll W.M., Ma L., Jester J. V., Direct correlation of collagen matrix deformation with focal adhesion dynamics in living corneal fibroblasts. Journal of Cell Science. 116 (2003), 1481 - 1491.

Staszyk C, Gasse H., A simple fluorescence labelling method to visualize the three- dimensional arrangement of collagen fibers in the equine periodontal ligament. Ann Anat. 2004 Apr; 186(2): 149-52.

Voytik-Harbin, S.L., Rajwa, B., Robinson, J.P., Three-dimensional imaging of extra cellular matrix and extra cellular matrix-cell interactions. Methods in Cell Biology, 63 (2001), 583 - 97. Zipfel, W.R., Williams, R.M., Christie, R., Nikitin, A.Y., Hyman, B.T., Webb, W. W., Live tissue intrinsic emission microscopy using multiphoton-excited native fluorescence and second harmonic generation. Proceedings of the National Academy of Sciences U.S A. 100 (2003), 7075-7080.

Brightman, A.O., Rajwa B.P., Sturgis J.E., McCalh'ster M.E., Robinson J.P., Voytik-Harbin S. L., Tune-lapse confocal reflection microscopy of collagen fibrillogenesis and extra cellular matrix assembly in vitro. Biopolymers. 54 (2000), 222 - 234.

Wolf, K., Friedl, P., Functional imaging of pericellular proteolysis in cancer cell invasion. Biochimie, 87 (2005), 315 - 320.

Cox, G., Kable, E., Jones, A., Fraser, L, Manconi, F., Gorrell, M.D., 3-dimensional imaging of collagen using second harmonic generation, Journal of Structural Biology. 141 (2003), 53 62.

Campagnola, P.J., Millard, A.C., Terasaki, M., Hoppe, P.E., Malone, C.J., Mohler, WA., Three-dimensional high-resolution second-harmonic generation imaging of endogenous structural proteins in biological tissues. Biophysical Journal. 82 (2002), 493 - 508.

Zoumi, A., Yeh, A., Tromberg, B. J., Imaging cells and extra cellular matrix hi vivo by using second-harmonic generation and two-photon excited fluorescence. Proceedings of the National Academy of Sciences U.SA. 99 (2002), 11014 - 11009.

Richards-Kortum, R., Sevick-Muraca, E., Quantitative optical spectroscopy for tissue diagnosis. Annual Reviews of Physical Chemistry. 47 (1996), 555 - 606. Williams, R.M., Zipfel, W.R., Webb, W. W., Interpreting second-harmonic generation images of collagen I fibrils. Biophysical Journal. 88 (2005), 1377 - 1386.

Schnell, A.M., Hoerstrup, S.P., Zund, G., KoIb, S., Sodian, R., Visjager, J.F., Grunenfelder, J., Suter, A., Turina, M., Optimal cell source for cardiovascular tissue engineering: venous vs. aortic human myofibroblasts. The Journal of Thoracic and Cardiovascular Surgery. 49 (2001),

221 - 225.

Sodian, R., Hoerstrup, S.P., Sperling, J.S., Daebritz, S., Martin, D.P., Moran, A.M., Kim, B. S., Schoen, FJ., Vacanti, J.P., Mayer, J.E. Jr., Early in vivo experience with tissue- engineered trileaflet heart valves. Circulation. 102 (2000), IU22 - UJ9.

MoI, A., Driessen, NJ., Rutten, M.C., Hoerstrup, S.P., Bouten, CV. C, Baaijens, F.P.T., Tissue engineering of human heart valve leaflets: a novel bioreactor for a strain-based conditioning approach. Annals of Biomedical Engineering. 33 (2005), 1778 - 1788.

van Zandvoort, M., Engels, W., Douma, K., Beckers, L., Oude Egbrink, M., Daemen, M., Slaaf, D.W., Two-photon microscopy for imaging of the (atherosclerotic) vascular wall: a proof of concept study. Journal of Vascular Research. 41 (2004), 54-63. Zoumi, A., Lu, X., Kassab, G.S., Tromberg, BJ., Imaging coronary artery microstructure using second-harmonic and two-photon fluorescence microscopy. Biophysical Journal. 87 (2004), 2778-2786.

Boulesteix, T., Pena, A.M., Pages, N., Godeau, G., Sauviat, M.P., Beaurepaire, E., Schanne- Klein, M.C., Micrometer scale ex vivo multiphoton imaging of unstained arterial wall structure. Cytometry A. 69 (2006), 20 - 26.

Konig, K., Riemann, L, High-resolution multiphoton tomography of human skin with subcellular spatial resolution and picosecond time resolution. Journal of Biomedical Optics. 8 (2003), 432-439.

Schenke-Layland, K., Madershahian, N., Riemann, L, Starcher, B., Halbhuber, KJ., Konig, K., Stock, U. A., Impact of cryopreservation on extra cellular matrix structures of heart valve leaflets. The Annals of Thoracic Surgery. 81(2006), 918 - 926. Schenke-Layland, K., Riemann, L, Stock, U. A., Konig, K., Imaging of cardiovascular structures using near-infrared femtosecond multiphoton laser scanning microscopy. Journal of Biomedical Optics. 10 (2005), 024017.

Konig, K., Schenke-Layland, K., Riemann, L, Stock, U.A., Multiphoton autofluorescence imaging of intratissue elastic fibers. Biomaterials. 26 (2005), 495 - 500.

Ushiki, T., Collagen fibers, reticular fibers and elastic fibers. A comprehensive understanding from a morphological viewpoint. Archives of Histology and Cytology. 65 (2002), 109- 12Fratzl, P., Cellulose and collagen: from fibres to tissues, Current Opinion in Colloid and Interface Science, 8 (2003), 32 - 39.

Hartmann, A., Boukamp, P., Friedl, P., Confocal reflection imaging of 3D fibrin polymers. Blood, Cells, Molecules and Diseases. 36 (2006), 191 - 193.

Johnson, C, Galis, Z.S., Quantitative assessment of collagen assembly by live cells. Journal of Biomedical Materials Research A. 67 (2003), 775 - 784.

Czirok, A., Zach, J., Kozel, BA., Mecham, R.P., Davis, E.G., Rongish, B.J., Elastic fiber macro-assembly is a hierarchical, cell motion-mediated process. Journal of Cell Physiology. 207 (2006), 97 - 106.

Kozel, BA., Rongish, B.J., Czirok, A., Zach, J., Little, CD., Davis, E.G., Knutsen, R.H., Wagenseil, J.E., Levy, MA., Mecham, R.P. Elastic fiber formation: a dynamic view of extra cellular matrix assembly using timer reporters. Journal of Cell Physiology. 207 (2006), 87- 96.

Claims

1. A labelled conjugate for imaging extra cellular matrix components in a biological tissue or cell culture wherein the conjugate comprises a peptide sequence which binds to collagen, elastin or glycosaminoglycan and which is coupled to an imaging agent.

2. The labelled conjugate according to claim 1, wherein the peptide sequence comprises a peptide sequence able to bind collagen, elastin or GAG, wherein said peptide sequence is optionally fused to another molecule and wherein the peptide sequence is selected from the following list:

(1) an alpha 1 integrin A-domain, preferably SEQ ID NO:3,

(2) a bacterial adhesion protein or a bacterial adhesin,

(3) all or part of (l) or (2), or

(4) peptide sequences homologous with (1), (2) or (3),

3. The labelled conjugate according to claim 2, wherein the bacterial adhesion protein or bacterial adhesin (2) is the collagen adhesin of S. aureus (CNA).

4. The labelled conjugate according to claim 3, wherein the peptide sequence comprises a truncated form of the collagen adhesin of S. aureus (CNA35), more preferably the truncated form (CNA35) fused to another molecule, even more preferably the truncated form (CNA-35) fused to an His-tag (SEQ ID NO:1) or a peptide homologue thereof.

5. The labelled conjugate according to claim 2 wherein the peptide sequence (1) is fused to another molecule, preferably the peptide sequence is SEQ ID. No.2 or a peptide homologue thereof.

6. The labelled conjugate according to any one of claims 1 to 5 wherein the imaging agent is a fluorescent label, preferably an amine-reactive fluorescein derivative.

7. The labelled conjugate according to any one of claims 1 to 5, wherein the imaging agent is a liposome containing a MRI contrast agent, preferably a liposome containing gadolinium.

8. The labelled conjugate according to any one of claims 1 to 7 wherein the conjugate binds to collagen type I or collagen type III, preferably from human origin.

9. A kit comprising a labelled conjugate according to any one of claims 1 to 8.

10. A composition comprising a labelled conjugate according to any one of claims 1 to

8.

11. A method for imaging extra cellular matrix components in a biological tissue or cell culture, which method comprises: (i) contacting the biological tissue or cell culture with a labelled conjugate according to any one of claims 1 to 8 or a kit according to claim 9, or a composition according to claim 10; (ii) visualising the imaging agent of the labelled conjugate so that an image is obtained.

12. The method according to claim 11, wherein the biological tissue or cell culture is ex vivo.

13. The method according to claim 11 or 12, wherein imaging is performed in an engineered biological tissue.

14. The method according to any one of claims 10 to 12, wherein the biological tissue or cell culture comprises cardiovascular tissue, connective tissue, intervertebral disk; lung tissue, muscle tissue, skin tissue, stem cells, fibroblasts, skeletal muscle cells, cardiac muscle cells, blood vessel cells, blood derived and blood forming cells, bone marrow derived cells, umbilical cord cells, skin cells, nerve cells cartilage cells, bone cells, cells of embryonic origin.

15. The method according to claim 14, wherein the biological tissue is cardiovascular.

16. The method according to any one of claims 11 to 15, wherein imaging is performed real time.

17. The method according to any one of claims 11 to 16, wherein imaging is performed within a single cell.

18. The method according to any one of claims 11 to 17 , wherein the biological tissue or cell culture is contacted with two labelled conjugates as defined in any one of claims 1 to 7.

19. The method according to any one of claims 11 to 18, wherein the extra cellular matrix components visualised are newly formed collagen fibrils.

20. The method according to any one of claims 11 to 19, wherein the imaging method is RMI and preferably wherein the labelled conjugate of claim 7 is being used.

21. Use of a conjugate according to any one of claims 1 to 8 or a kit according to claim 9, or a composition according to claim 10 in tissue engineering.

22. Use of a conjugate according to any one of claims 1 to 8 or a kit according to claim 9, or a composition according to claim 10 for the preparation of a diagnostic agent for the imaging of extra cellular matrix components.

23. Use of the conjugate as defined in claim 7 in RMI.