WO2005110497A1

WO2005110497A1 - Imaging of tumors and metastases using a gelatinase targeting peptide comprising the structure ctthwgftlc.

Info

Publication number: WO2005110497A1
Application number: PCT/FI2005/050161
Authority: WO
Inventors: Oula PEÑATE MEDINA; Sami Kaukinen; Kalevi Kairemo
Original assignee: Ctt Cancer Targeting Technologies Oy
Priority date: 2004-05-14
Filing date: 2005-05-16
Publication date: 2005-11-24
Also published as: FI20040682A0

Abstract

The invention relates to the use of gelatinase targeting peptides comprising the structure CTTHWGFTLC (CTT) for early detection and imaging of tumors and metastases in mammals, including humans. A preferred gelatinase targeting peptide for use according to the invention is cyclic CTT, which has been labeled and/or conjugated with labeled liposomes. Preferably, the labeled CTT peptide is 125 I-labeled AAY-CTT. Further objects of the invention are a composition for diagnosing tumors and metastases, and a method for early detection and imaging tumors and metastases in mammals.

Description

Imaging of tumors and metastases using a gelatinase targeting peptide comprising the structure CTTHWGFTLC

Field of the invention

The present invention relates to the use of gelatinase targeting peptides for early detection and imaging of tumors and metastases in mammals, including humans. A preferred gelatinase targeting peptide for use according to the invention is cyclic CTTHWGFTLC (CTT), which has been labeled and/or conjugated with labeled liposomes. Further objects of the invention are a composition for diagnosing tumors and metastases, and a method for early detection and imaging tumors and metastases in mammals.

Background of the invention

Early detection of primary tumors and metastases would greatly benefit cancer patients. Many cancers are observed at a time point when they are spread, and therapeutic intervention is difficult. Only a few methods are available for early and reliable tumor diagnosis, which can be done in a larger population. These include mammography for detection of breast cancer and PSA test for detection of prostate cancer¹, but unfortunately these techniques are not cancer specific. Early detection of recurrent and metastatic disease may lead to better prognosis. Therefore, there is a need for general cancer specific markers, which could be utilized in early tumor detection, localization, grading, staging, and follow-up. One example fulfilling some of these tasks is an octapeptide called octreotide, which can be used for early detection, localization, and follow-up of both recurrent and metastatic disease, especially in neuroendocrine tumors-. This peptide with several different conjugates has already now wide applications in clinical oncology.

Agents that preferentially recognize the tumor tissue or are able to home to the tu- mor vasculature following an intravenous injection could provide a means for accurate tumor imaging. Several antibody conjugates have been examined in tumor therapy and imaging in animal models²*⁴. The in vivo phage display application was developed to search for small molecular weight peptides that can gain access to the tumor or other desired tissue from circulation⁵. Peptides recognizing the tumor vasculature- or tumor lymphatics- have been obtained by screening with random peptide libraries in mice. Recently, the first biopanning was carried out also in a human patient with a goal to identify clinically useful organ targeting peptides².

One of the peptides that shows tumor homing ability in the mouse model is CTTHWGFTLC (CTT), a selective inhibitor of MMP-2 and MMP-9 gelatinases². When the peptide is given to mice bearing subcutaneous tumors, the peptide can prevent tumor growth and prolong the survival of cancer-bearing mice. Several studies have shown that tumors express elevated levels of MMP-2 and MMP-9 and the enzymes are produced not only by the tumor cells— but also by angiogenic endothelial cells— , tumor infiltrating leukocytes— , and the stroma surrounding a tu- mor¹² . Thus, because of their total high expression levels in tumors, gelatinases can be considered potential targets to deliver therapeutic and imaging agents to tumors. A recent study showed that tumor imaging in mouse is achieved with a sensitive peptide substrate, which, after cleavage by MMP-2, gives a fluorescent label— . A more controversial issue is whether inhibition of all MMP enzyme activity is an appropriate strategy for cancer therapy. Many clinical trials with MMP inhibitors have failed— , and in some animal models an inhibition of MMP-9 can even lead to unwanted results with enhanced tumor angiogenesis and tumor growth due to suppression of angiostatin production— .

We have synthesized here derivatives of CTT that can be radiolabelled or bound to isotope-containing liposomes. Because of its amphifilic properties, being both water-soluble and hydrophobic, CTT readily binds to the surface of liposomes and can be used as a targeting agent to deliver liposomes to gelatinase expressing cells— . We have studied biodistribution of these potential tumor imaging agents in mice bearing tumor xenografts and found the ability of CTT to localize the tumor site in live mice. Summary of the invention

The present invention is directed to the use of MMP-2 or MMP-9 gelatinase targeting peptides for early detection and imaging of tumors and metastases in mammals, including humans. The gelatinase targeting peptide is preferably a peptide compound comprising the cyclic structure CTTHWGFTLC or a derivative thereof. For detection, the peptide is linked to a detectable label, such as a radioactive label, a magnetic particle, a fluorescent label, an affinity label or a luminescent label. Preferably the peptide is linked to one or more radionuclides. Preferred radionuclides are for example ¹²⁵I and ^99mTc.

In a preferred embodiment of the invention the peptide is conjugated with liposomes. Liposomes may for example be coated with a radiolabeled peptide. Preferably, liposomes are coated with a radiolabeled peptide and then encapsulated with radiolabeled albumin. The peptide is linked on the liposome surface for example by fatty acid tails, hydrophobic peptide anchors, or by direct coupling to phospholipid head groups or polyethylene glycol lipids.

A further object of the invention is a composition for diagnosing tumors and metas- tases in mammals, comprising the cyclic peptide motif CTTHWGFTLC or its derivative, together with one more detectable labels, and preferably suitable carriers and/or adjuvants. In a preferred embodiment of the invention the composition further comprises liposomes.

A still further object of the invention is a method for early detection and imaging of tumors and metastases in mammals.

Brief description of the drawings

Figure 1. Effect of peptides on MMP-2 activity was determined with a fluorogenic gelatinase substrate. The studied peptides were A) AACTT (100 μM), AAY-CTT (100 μM), CTT (100 μM and 25 μM, ^99mTc-CTT (25 μM) and B) AAY-CTT (100 μM), cltfgwhttc (100 μM), CTT (100 μM).

Figure 2. A) Representative solution structure of CTTHWGFTLC peptide and B) an example conformation of [cltfgwhttc] peptide. Short interproton distances and dihedral angle restraints obtained by NMR have been used to obtain the models by restrained torsion angle molecular dynamics. Carbon atoms are shown in green, oxygens in red, nitrogens in blue and sulfur atoms in the disulfides are depicted in yellow.

Figure 3. HT1080 cells were plated on Matrigel-coated Transwell filters in the presence or absence of the CTT, ctthwgftlc or cltfgwhttc peptides (each 200 μM), and allowed to migrate for 18 h. Cells that traversed to the lower site of the filter were stained and the cells were counted on a microscope. The data show mean ± SD; n = 3.

Figure 4. A) Tissue distribution of the ¹²⁵I- AAY-CTT peptide in normal mice. The biodistribution of the labeled peptide is shown 30 minutes following injection and was corrected for weight. Results are expressed as percentage of injected dose per 0.1 g tissue (% ID/O.lg). All values are indicated as mean ± SD of 5 mice. B) Autoradiography of tumor tissue sections of a mouse given ¹²⁵I- AAY-CTT intravenously. Skin appears most right and gluteal muscles most left.

Figure 5. Biodistribution of liposomes encapsulated with ¹²⁵I-BSA label was ex- amined in tumor-bearing mice. Liposomes had a CTT peptide-coating (A) or no coating as a control (B). Radioactivity accumulated in the organs at 2 h time point after an intravenous injection is shown. The values indicate mean ± SD of 5 mice.

Figure 6. Gamma imaging of live mice having experimental tumors. The figures show examples of animals after 24 h of intravenous injection with ^99m Tc-CTT and ¹²⁵I-BSA double-labeled liposomes (A) or ^99m Tc-CTT alone (B). The arrow indicates the tumor site. The injection site in the tail vein is on the left.

Figure 7 shows to fluoro-GRENYHGCTT biodistribution in OV-90 xenografts.

Figure 8 shows I-125-CTT-Peg-PE-micelle biodistribution.

Figure 9 shows tumor/blood ratios of two 6F-CTT2-epimers. Mice (3 ani- mals/peptide/time point) were injected with radio-iodinated peptides and blood and tumor were collected at three time points after injection. Tumor accumulation of both peptides is observed as a function of time.

Detailed description of the invention

Abbreviations

BSA Bovine serum albumin

CTT CTTHWGFTLC

NOESY nuclear Overhauser enhancement spectroscopy PSA Prostate specific antigen

RMSD Root mean square deviation

TOCSY Total correlation spectroscopy

HSQC Heteronuclear correlation spectroscopy

Our results show the capability of a gelatinase targeting peptide to localize the tumor site in a mouse model. MMP-2 and MMP-9 gelatinases appear to be a good choice for the purpose of tumor imaging as they are not only expressed by the tumor cells but also by host cells surrounding the tumor. Particularly, the expression by endothelial cells in the neovasculature^10,11,14,15 may significantly contribute to the success of targeting, as the tumor mass itself, which is often necrotic, may be less permeable to exogenous agents and more difficult to image. Earlier studies showed that phage displaying the CTT peptide homes to the tumor xenograft of mouse, indicating that vascular targeting is indeed possible⁹. As radiolabeling of a small molecular weight peptide can easily affect the peptide conformation and ac- tivity, we focused our studies on identification of reagents, which are gelatinase inhibitors also after radiolabeling. Two such reagents, ¹²⁵I- AAY-CTT and ^99m Tc- CTT, were developed, and these peptides showed tumor homing ability, either alone or in conjugation with liposomes.

The 3D structure of the peptide is highly relevant for activity of CTT peptide. It has been shown that the linear peptide does not have the same 3D structure and thus loses its bioactivity. This also makes NMR as an powerful tool in controlling the effects of modifications to overall 3D structure. Additionally, it shows that there are differences between small peptides as regards stability of conformation which is of extreme importance for further conclusions regarding the usability of different peptides on clinical studies. This had lead to different labelling strategies in various applications of these peptides. Antibodies have broadly been used for imaging purposes. Also peptides that occur naturally have been used for imaging like octreoti- de. However there have not been extensive studies where phage display peptide derived peptides have been used for tumour imaging. Especially gelatinase targeting peptide has not been used before.

The NMR studies showed that CTT peptide adopts a saddle-like structure, which is rigidified by the disulfide bond. The ^99m Tc label caused a small decrease in gelatinase inhibitory activity, which may be explained by an increased length between the cysteines due to technetium chelation. We found that ^99m Tc-CTT was most effective when coated on the surface of liposomes. This made it possible to image a tumor site by a gamma camera in a real time. ^99m Tc-CTT or liposomes alone did not give a tumor image within a 30 min time frame, indicating that the combined effect of the two was necessary. The success of tumor imaging is likely due to an additive effect that CTT causes to liposome particles, which themselves can also target the tumor apparently due to the leakiness of the tumor blood vessels. In our studies, the gamma imaging of live tissues may also have been efficient, because we used intensively labeled liposomes that contained ^99m Tc-CTT label on the surface and ¹²⁵I-BSA label inside of liposomes. The biodistribution data in healthy and normal mice show that the lifetime of the ¹²⁵I- AAY-CTT peptide is relatively short. In 24 hours almost all signal is diminished. The main secretion route is through kidneys but some liver uptake is also been seen. Radioactivity accumulation was seen in the thyroid apparently due to free iodine released from peptide degradation. Otherwise, localization of the radio- labeled peptide to normal tissues investigated was minimal and attributable to blood pool activity. These results suggest that MMP-2 and MMP-9 are not expressed at such high levels in normal tissues that it would cause accumulation of the CTT peptide. Circulating blood cells also did not concentrate CTT, and our recent studies show that resting T cells and macrophages extracted from blood do not express gelatinases, with which the CTT peptide would react (Stefanidakis M. and Koivunen E., unpublished data). The uptake of hydrophobic and low molecular weight compounds by the liver and the kidney presents often an obstacle for ther- apy studies. However, the minor accumulation of the AAY-CTT-peptide in normal tissues suggests that it may not cause toxicity problems in vivo. This result supports the idea that selective gelatinase targeting agents could be suitable for diagnostic or therapeutic applications.

CTT-liposomes accumulated also to some extent to the lungs in a 2-hour time frame. This is understandable when taking in consideration of the amount of blood flown through the lungs compared with other organs and especially the tumour and muscle tissue. However, it is notable also that in mice bearing human tumor xenografts, the lungs are principal sites for metastases. Thus, though we did not detect visible lung metastases in the mice, it is possible that CTT-liposomes recognized tiny metastases that were generating. Further experiments are needed to clarify whether metastases are possible to see by a sensitive gelatinase targeting method.

Liposomes might well be developed into a miniature microscope that can be in- jected into the circulation to observe abnormal vasculature patterns and tumor formation. The targeting peptide of choice, CTT or another compound, can be linked on the liposome surface by fatty acid tails, hydrophobic peptide anchors, or by direct coupling to phospholipids head groups or polyethylene glycol lipids. Liposomes may have specific lipid, carbohydrate and/or positively charged coating, which prolongs liposome circulation time in the blood and causes less immunologi- cal reactions, but improves the ability to fuse with the target cells. Finally, if the destination is to kill the target cell, liposomes may be encapsulated with doxorubi- cin— or other poison, or a hydrophobic cancer drug can be carried in the lipid bilayer. Our studies describe gelatinase-binding labels that could be starting points towards the goal of accurate imaging and follow up of tumor development.

Experimental

Reagents

All reagents, unless stated otherwise, were obtained from Sigma- Aldrich and culture media from Gibco Life Technologies (Paisley, Scotland).

Synthetic peptides

Peptides were synthesized on an Applied Biosystems 433 A (Foster City, CA) automatic synthesizer using Fmoc-chemistry. For disulfide generation, peptides were dissolved at 1 mg/ml in 0.05 M ammonium acetate (pH 8) and mixed with H₂O for 40 min at room temperature so that 0.5 ml of 3 % H O₂ was added per 100 mg peptide. Peptides were purified by reversed phase HPLC and the molecular weight was identified by mass spectrometry analysis.

NMR.

CTT and retro-inverse peptide samples for NMR spectroscopy were prepared by dissolving freeze-dried peptides to aqueous buffer to result in approximately 2 mM samples. All spectra were acquired by Varian Unity Inova 600 MHz NMR spectrometer, equipped with ^/^C/^N triple-resonance probehead with an actively shielded z-axis gradient system. The proton assignments were derived from phase- sensitive two-dimensional TOCSY— and NOESY— spectra acquired at 2 and 8°C to adjust rotational correlation time suitable for NOE experiments. TOCSY spectra were recorded with 30, 75, and 120 ms mixing times using DIPSI-2rc spin-lock²² , whereas NOESY spectra were measured using 300 ms mixing time. The spectra were recorded with 512 and 4512 complex points in

and F dimension, corresponding to acquisition times of 56.9 and 501.3 ms, respectively. In addition one- bond proton carbon correlation spectra, ¹³C-HSQC— were taken to ascertain the assignments. 256 and 4000 complex points were collected corresponding tol2.1 and 400 ms acquisition times in

and F , respectively. Short interproton distance restraints were extracted from two-dimensional NOE spectra and main chain torsion angles HN-H^α scalar couplings from proton spectrum. For the structure generation, refinement and quality assessment we employed DYANA— .

Cell Culture

KS1767 kaposi's sarcoma and HT 1080 fibrosarcoma cells were maintained in DMEM medium supplemented with 10% fetal bovine serum, Glutamax I, penicillin 100 U/ml and streptomycin 0.1 mg/ml. Cell invasion assay was conducted using Matrigel coated invasion chambers in the serum-containing medium as described².

MMP-2 activity assay

MMP-2 activity was measured using a fluorogenic peptide substrate MCA-Pro- Leu-Ala-Nva-Dpa-Ala-Arg-NH₂ (Calbiochem, San Diego, CA) on a MOS-250 fast reaction spectrofluorometer with a thermostated cuvette compartment (Bio-Logic SA, Claix, France). Exitation was at 330 nm and emission spectra were scanned from 340 to 500 nm. Shortly, 1.1 U of pro-MMP-2 (Roche) in 60 μl HEPES buffer was activated by 10 mM APMA for 30 min at 37 °C. In each measurement, 1 mU of activated MMP-2 was mixed with 2.5 μM substrate in Hepes buffer in a quartz cuvette of 100 μl volume. Measurements were done at the 0, 2, 4, 6, 8 and 10 min time points. Peptide inhibitors were preincubated for 10 minutes with MMP-2 before experiment.

Technetium -99m labelling of CTT peptide

One mg of CTT was dissolved in 1 ml of H₂O and kept on cold block. Twenty μl of CTT solution was transferred to a small vial and 50 μg of SnCl₂ was added. After a 15 min incubation, labelling was performed with 8-10 mCi of fresh 99m- technetium (Radioisotope Laboratory of Helsinki University Central Hospital, Finland) for 10 min. The ^99m Tc-CTT peptide was purified on a Sep-Pak CI 8 cartridge (Waters, Milford, MA) using 0.9 % sodium chloride / ethanol (60:40) as the elution solvent.

Iodination of AAY-CTT peptide

The AAY-CTT peptide was labeled with 12_5τ I using iodogen as a catalysator. 5 MBq of Na¹²⁵I (Amersham) in 0.5 ml PBS was added to a tube containing 10 μg dried iodogen and 27 μg AAYCTT-peptide. The mixture was incubated for 20 min at room temperature. The labeled peptide was bound on a Sep-Pak C18 cartridge and eluted with 50% acetonitrile. The solvent was evaporated at +55°C and the purified dry peptide was then dissolved in 500 μl of PBS. The activity of the peptide was determined in a gamma counter (Cobra II, Packard Instruments). The AACTT and CTT peptides not containing a tyrosine were iodinated by using of N-succinimidyl 3-(tri-n-butylstannyl)benzoate (ATE) method—.

Preparation of labeled liposomes

Iodination of bovine serum albumin (BSA) was performed by the iodogen 1, 3, 4, 6 -tetrachloro- 3, 6- diphenylglycoluril method. Briefly, 100 μl (10 mg/ml) of BSA and ¹²⁵I (50MBg) were mixed in a iodogen-coated vial, and ¹²⁵I-BSA was then separated from free label on a PD-10 gel filtration column (Pharmacia). For encapsulation of ¹²⁵I-BSA on liposomes, we used the protocol described earlier for doxorubicin entrapment¹¹. POPC/POPE lipid stock solutions was mixed in chloroform to obtain the 80:20 mol/mol composition. The solvent was removed under a gentle stream of nitrogen and the lipid residue was subsequently maintained under reduced pressure for at least 2 h. Multilamellar liposomes were formed by hydrat- ing the dry lipids at room temperature with one ml of PBS together with ¹²⁵I-BSA so as to yield a lipid concentration of one mM. Multilamellar liposomes were freeze/thawed five times to enhance encapsulation—. Large unilamellar vesicles were obtained by extruding²^ liposome dispersions 19 times through a 100 nm poresize polycarbonate membrane (Nucleapore, Pleasanton, CA, USA) with a Li- posoFast Pneumatic small- volume homogenizer (Avestin, Ottawa, Canada). The pressure used for extrusion of vesicles through the filters was 25 psi (-170 kPa). In some experiments, the ^99m Tc-CTT peptide was mixed with the vesicles— to bind the peptide on the liposome surface.

Radionuclide imaging of tumor xenografts

Subcutaneous KS1767 kaposi's sarcoma tumors were made by injecting one million cells per NMRI female nude mouse (Harlan, Netherlands). Tumors usually formed within three weeks. Approximately 1 - 5 MBq labeled reagent (¹²⁵I-AAY- CTT, ¹²⁵I-AACTT, ^99m Tc-CTT or radioactive liposomes) was injected into avertin- anesthetized mice via the tail vein in a volume of 50 μl - 200 μl. Gamma imaging was done at 30 min and 24 h following the injection using a Picker Prism 1500XP single-head gamma camera connected to an Odyssey computer (Picker International, Highland Heights, OH). ^99mTc-activity was recorded by using 140 keV en- ergy peak (20% window) and ¹²⁵I activity using 30 keV gamma energy peak (35% window). The mice were then sacrificed and blood samples taken for radioactivity measurements. Tumor and other organs were collected and the radioactivity counted. The tumor tissue was sliced to 100 μm thick frozen samples using cryotomy and approximately every tenth sample was imaged by autoradiography. The mice were taken care by the instructions of the animal facility, and the experiments were approved by an ethical committee of the Helsinki University. Biodistribution of CTT peptide in normal mice

Five anesthetized balb/c mice received a total of 27 μg ¹²⁵I-AAY-CTT peptide (1.05 MBq) via the tail vein injections. After 30 minutes, the mice were sacrified, blood was drawn and tissues were dissected to determine the biodistribution of the peptide. Blood sample, heart, liver, kidneys, lungs, muscle, bone, brain, spleen and thyroid were collected and weighed, and their radioactivity was measured in a gamma counter (Wallac, Turku, Finland).

Results

Effect of chemical modification and radiolabeling on the CTT peptide activity

After direct labeling with ¹²⁵I on its aromatic residues, the CTT dodecapeptide lost its gelatinase inhibitory activity and could not be used as a tumor targeting agent in mouse. A tyrosine residue added on the first N-terminal exocyclic position gave a peptide that could be efficiently iodinated without activity loss. The 13 amino acid long ¹²⁵I-labeled AAY-CTT peptide inhibited MMP-2 with the same potency as CTT (Fig. 1 A). MMP-2 activity was assessed with a fluorogenic peptide substrate. The sensitivity of CTT to amino acid modifications is illustrated by the AA-CTT peptide lacking a tyrosine. This peptide was not an inhibitor and surprisingly stimulated the peptide substrate hydrolysis by MMP-2. We also examined peptide labeling with technetium isotope ^99m Tc, which can chelate between two cysteine resi- dues—. The ^99m Tc-CTT retained the gelatinase inhibitory activity but was slightly less active than the parent peptide, possibly because of opening of the disulfide bond.

We determined the solution structure for the CTT peptide. It shows that the peptide adopts a well-defined saddle-shaped circular form with RMSD 0.6 ± 0.3 A computed from ten lowest energy structures out of 50 conformers. The side chains, in particular the aromatic His, Trp and Phe residues, approach each other to make a compact structure (Fig. 2A), heavy atom RMSD 1.0 ± 0.3 A. Iodination of the aromatic residues or increasing the distance between the cysteines can thus be assumed to affect the peptide conformation and cause activity loss. As a retro-inversion pep- tide can often mimic the structure of the parent peptide—, we also synthesized the CTT retro-inversion peptide cltfgwhttc, in which the amino acids (except the gly- cine) are in D-form. The NMR structure of cltfgwhttc has an overall shape similar to that of CTT as expected (Fig. 2B). The cltfgwhttc peptide has protruding side chains, heavy atom RMSD 3.3 ± 1.1 A, but the course of the backbone is not as well defined as that of CTT leading to a large dispersion in coordinates RMSD 2.0 ± 0.7 A. Though CTT and cltfgwhttc have structural similarities, the conformations are not identical and this is also indicated by the finding that an antibody made against the CTT peptide— did not recognize the cltfgwhttc peptide in dot blot and microtiter well assays (data not shown).

The cltfgwhttc retro-inversion peptide had a potent gelatinase inhibitory activity and inhibited MMP-2 even better than CTT (Fig. IB). Just changing all amino acids to D-form did not generate a gelatinase inhibitor as the ctthwgftlc peptide containing D-amino acids in the original order lacked activity (not shown). In a tumor cell invasion assay, ctthwgftlc was also without a notable activity whereas the retro-inversion peptide blocked efficiently like CTT does (Fig. 3).

Biodistribution of radiolabelled CTT peptide in mice

Though the cltfgwhttc retro-inversion peptide was active, it had a limited water- solubility and was therefore not a good candidate for in vivo imaging studies in mice. As indicate above, the iodinated AAY-CTT peptide retained its gelatinase inhibitory activity and was a safer choice for such experiments. First, the bio- distibution of ¹²⁵I-labeled AAY-CTT was examined in normal healthy mice follow- ing an intravenous injection via the tail vein. After a 30-min circulation time, organs were collected to determine peptide homing to different tissues (Fig. 4A). Specific CTT peptide accumulation is expected to depend on an expression level of MMP2 and MMP-9. The thyroid accumulated highest amount radioactivity. This accumulation is likely due to free ¹²⁵I-label released in vivo and not due to gelatinase binding in the thyroid. Kidney had the next highest amount of radioactivity, suggesting rapid clearance of the peptide by the kidney. Some uptake of the peptide can also be seen in the liver. Overall, the mouse organs accumulated little radioactivity when considering the amount that circulated in the blood. A gram of tissue contained 0.5 % or less of the total label injected. When the organs were collected after 24 h, almost all ¹²⁵I-CTT had disappeared or metabolized, and the main secre- tion route was through kidneys (data not shown).

Next we studied the tumor homing ability of ¹²⁵I- AAY-CTT in mice bearing KS1767 Kaposi's sarcoma xenografts. After an intravenous injection, the peptide achieved a serum concentration of 8.5 - 17 μM as calculated on the basis of radio- activity in a blood sample taken. Following a 30 min circulation time, the tumor and its adjacent tissues were sectioned into 100 μM thick frozen samples using cryotomy. Approximately every tenth sample was imaged by autoradiography. Fig 4B shows that ¹²⁵I- AAY-CTT locates in the tumor site but not in the skin (most right) or gluteal muscles (left). As a control, we carried out similar tumor targeting experiments with ¹²⁵I-labeled AACTT, in which the labelling abrogated the gelatinase inhibitory activity. No tumor targeting was seen with ¹²⁵I- AACTT (not shown).

Tumor imaging with CTT-targeted liposomes

We also examined the suitability of using CTT-coated liposomes for tumor imaging in the mouse model. Liposomes were encapsulated with ¹²⁵I-BSA label for easy detection. Uncoated ¹²⁵I-BSA-containing liposomes were used as a control to see where the CTT peptide can specifically target the liposomes. The CTT peptide clearly changed the biodistribution of liposomes as studied after 2 h circulation time. As expected, CTT enhanced liposome homing to the primary tumor. Whereas the tumor / muscle targeting ratio for liposomes in the absence of CTT was 0.5 (Fig.5B), in the presence of CTT it was 2.6 (Fig. 5A). We chose the muscle as the reference tissue, since the muscle is rarely metastasized by tumor cells. CTT affected liposome binding not only to the tumor but also to apparently normal tissues, where metastases, if present, were not visible. CTT-mediated liposome accumulation was seen particularly in the lungs of tumor-bearing mice. Imaging of the primary tumor in live mice was examined by gamma camera. To obtain high lipo- somal radiolabel intensity, ^99m Tc-labelled CTT was used for liposome coating and ¹²⁵I-labeled BSA for liposome encapsulation. Following a 24 h circulation time, CTT-mediated liposome accumulation in the primary tumor was observed and the tumor site could be visualized (arrow in Fig 6A). At the same time interval, the ^99m Tc-CTT peptide alone without liposomes was unable to image the primary tumor (Fig. 6B).

References

1. Duffy MJ. PSA as a marker for prostate cancer. Ann. Clin. Biochem 1996;33:511-19.

2. Anderson CJ, Dehdashti F, Cutler PD, Schwarz SW, Laforest R, Bass LA, Lewis JS, McCarthy DW. 64Cu-TETA-octreotide as a PET imaging agent for patients with neuroendocrine tumors. J. Nuclear Med 2001;42:213-21.

3- Pegraro MD, Reese DM. Combined biological therapy of breast cancer using monoclonal antibodies directed against HER2/neu protein and vascular endothe- Hal growth factor. Seminars Oncol 2002;29(Suppl. 11):29-37.

4. Chang SS, Bander NH, Heston WD. Monoclonal antibodies: will they become an integral part of the evaluation and treatment of prostate cancer—focus on prostate-specific membrane antigen? Curr. Opinion Urol 1999; 9:391-95.

5. Pasqualini R, Ruoslahti E. Organ targeting in vivo using phage display peptide libraries. Nature 1996;380:364-66.

6. Arap W, Pasqualini R, Ruoslahti E. Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model. Science 1998;279:377-80.

7. Laakkonen P, Porkka K, Hoffman JA, Ruoslahti E. A tumor-homing peptide with a targeting specificity related to lymphatic vessels. Nat. Med 2002;8:751 - 55.

8. Arap W, Kolonin MG, Trepel M, Lahdenranta J, Cardo-Vila M, Giordano RJ, Mintz PJ, Ardelt PU, Yao VJ, Vidal CI, Chen L, Flamm A, Valtanen H, Weav- ind LM, Hicks ME, Pollock RE, Botz GH, Bucana CD, Koivunen E, Cahill D, Troncoso P, Baggerly KA, Pentz RD, Do KA, Logothetis CJ, Pasqualini R. Steps toward mapping the human vasculature by phage display. Nat. Med 2002;8:121-27. 9. Koivunen, E., Arap, W., Valtanen, E., Rainisalo, A., Medina, O. P., Heikkila, P., Kantor, C, Gahmberg, C. G., Salo, T., Konttinen, Y. T., Sorsa, T., Ruoslahti, E., and Pasqualini, R. Tumor targeting with a selective gelatinase inhibitor. Nat. Bio- technol. 1999;17: 768-74.

10. Brooks PC, Stromblad S, Sanders LC, von Schalscha TL, Aimes RT, Stetler- Stevenson WG, Quigley JP, Cheresh DA. Localization of matrix metallopro- teinase MMP-2 to the surface of invasive cells by interaction with integrin D_vβ₃. Cell 1996;85:683-93.

11. Nguyen M, Arkell J, Jackson CJ. Human endothelial gelatinases and angio- genesis. Int. J. Biochem. Cell Biol 2001;33:960-70. 12. Swallow CJ, Murray MP, Guillem JG. Metastatic colorectal cancer cells induce matrix metalloproteinase release by human monocytes. Clin. Exp. Metastasis 1996;14:3-11. 13. Saad S, Gottlieb D J, Bradstock KF, Overall CM, Bendall L J. Cancer cell- associated fibronectin induces release of matrix metalloproteinase-2 from normal fibroblasts. Cancer Res 2002;62:283-89.

14. Bremer C, Tung CH, Weissleder R. In vivo molecular target assessment of ma- trix metalloproteinase inhibition. Nat. Med 2001 ;7:743-48.

15. Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat. Rev. Cancer 2002;2:161-74. 16. Pozzi A, LeVine WF, Gardner HA. Low plasma levels of matrix metalloproteinase 9 permit increased tumor angiogenesis. Oncogene 2002;21:272-82.

17. Medina OP, Sδderlund T, Laakkonen LJ, Tuominen EKJ, Koivunen E, Kin- nunen PKJ. Binding of Novel Peptide Inhibitors of Type IV Collagenases to Phospholipid Membranes and Use in Liposome Targeting to Tumor Cells in Vitro. Cancer Res 2001;61:3978-85.

18. Braunschweiler L, Ernst RR. Coherence Transfer by Isotropic Mixing - Application to Proton Correlation Spectroscopy. J. Magn. Reson 1983;53:521-28.

19. Kumar A, Ernst RR, Wϋthrich KA. Two-Dimensional Nuclear Overhauser Enhancement (2D NOE) Experiment for the Elucidation of Complete Proton- Proton Cross-Relaxation Networks in Biological Macromolecules. Biochem. Biophys. Res. Commun 1980;95:1-6.

20. Cavanagh J, Ranee M. Suppression of Cross-Relaxation Effects in TOCSY Spectra via a Modified DIPSI-2 Mixing Sequence. J. Magn. Reson 1992;96:670-78. 21. John BK, Plant, Hurd RE. Improved Proton-Detected Heteronuclear Correlation Using Gradient-Enhanced z and zz-Filters, J. Magn. Reson 1992;A101:113-17.

22. Gϋnthert P, Mumenthaler C, Wϋtrich K. Torsion Angle Dynamics for NMR Structure Calculation with the New Program DYANA. J. Mol. Biol 1997;273:283-98.

23. Zalutsky MR, Noska MA, Colapinto EV, Garg PK, Bigner DD. Enhanced tumor localization and in vivo stability of a monoclonal antibody radioiodinated using N-succinimidyl 3-(tri-n-butylstannyl)benzoate. Cancer Res 1989;49: 5543-49. 24. Clifford CJ, Warren EL, Richard TW, Pfeiffer DR. Factors affecting solute entrapment in phospholipid vesicles prepared by the freeze-thaw extrusion method: a possible general method for improving the efficiency of entrapment. Chem. Phys. Lipids 1990;55:73-83.

25. MacDonald RC, MacDonald RI, Menco BM, Takeshita K, Subbarao NK, Hu LR. Small- volume extrusion apparatus for preparation of large, unilamellar vesicles. Biochim. Biophys. Acta 1991;1061:297-303. 26. George AJT, Jamar F, Tai M, Heelan BT, Adams GP, McCartney JE, Houston LL, Weiner LM, Opeprmann H, Peters AM, Huston JS. Radiometal Labeling of Recombinant Proteins by a Genetically Engineered Minimal Chelation Site: Technetium-99m Coordination by Single-Chain Fv Antibody Fusion Proteins Through a C-Terminal Cysteinyl Peptide. Proc. Natl. Acad. Sci 1995;92:8358- 62.

27. Van Regenmortel MHV, Muller S. D-peptides as immunogens and diagnostic reagents. Curr. Opin. Biotechnol 1998; 9:377-82. 28. Bjδrklund M, Valtanen H, Savilahti H, Koivunen E. Use of intein-directed peptide biosynthesis to improve serum stability and bioactivity of a gelatinase inhibitory peptide. Comb. Chem. High Throughput Screen 2003;6:29-35. In press.

Claims

1. Use of a peptide comprising the structure CTTHWGFTLC or its derivative, for detection and imaging of tumors and metastases.

2. The use according to claim 1, wherein the peptide is cyclic CTTHWGFTLC or its derivative.

3. The use according to claim 1 or 2 wherein the peptide is coupled to a radioactive label, an affinity label, a magnetic particle, a fluorescent label or a luminescent label.

4. The use according to claim 3 wherein the peptide is coupled to a radionu- clide, such as ¹²⁵I or ^99mTc.

5. The use according to claim 4, wherein the peptide is ¹²⁵I-labeled AAY- CTTHWGFTLC peptide.

6. The use according to claim 5 in combination with the use of cytostatic agents.

7. The use according to any one of the preceding claims wherein the peptide is conjugated with liposomes.

8. The use according to claim 7 wherein the peptide is linked on the liposome surface by fatty acid tails, hydrophobic peptide anchors, or by direct coupling to phospholipid head groups or polyethylene glycol lipids.

9. The use according to claim 6, wherein the liposomes are coated with radio- labeled peptide and/or encapsulated with ¹²⁵I-albumin.

10. The use according to any one of claims 6 to 8 wherein the liposomes are en- capsulated with cytostatic agents.

11. A composition for diagnosing tumors and metastases in mammals, comprising the cyclic peptide motif CTTHWGFTLC or its derivative, one or more detectable labels and optionally suitable carriers and/or adjuvants.

12. The composition according to claim 10 comprising liposomes.

13. A method for early detection and imaging of tumors and metastases in mammals, comprising the steps of (a) coupling a detectable label with a peptide compound comprising the cyclic structure CTTHWGFTLC or its derivative; (b) administering the mixture obtained to a patient; (c) allowing the mixture to accumulate at the tumor or metastase; (d) detecting the accumulated mixture so as to image the tumor or me- tastase.