WO2011103668A1 - Cancer specific peptides and arrays for screening same - Google Patents

Abstract

Cancer targeting peptides and methods for the identification of peptide markers for cancer are described. The peptides may be used in the treatment and/or diagnosis of cancer and also in the identification of receptors on cancer cells. Also described are assay devices that can be used to detect peptides with the ability to bind to cancer cells. An array device including a series of peptides on a membrane and to be used in the detection of peptides that bind to cancer cells is also described.

Description

CANCER SPECIFIC PEPTIDES AND ARRAYS FOR SCREENING SAME

INVENTORS: Kamaljit Kaur and Sahar Ahmed

PRIORITY: This application claims priority over United States Provisional Patent Application No. 61/308,482 entitled "PEPTIDE ARRAYS FOR SCREENING CANCER SPECIFIC PEPTIDES", filed February 26, 2010.

TECHNICAL FIELD

The technical field is in the area of peptide markers for cancer including cancer targeting peptides and means for identifying same. The technical field further relates to the use of the peptides for the detection and treatment of cancer.

BACKGROUND Current cancer therapies have low specificity for tumor cells and may have serious toxic side effects. For instance, a major drawback of current chemotherapy is that the treatment enters normal tissue in the body with indiscriminate cytotoxicity and does not preferentially accumulate at tumor sites. The dose of treatment actually reaching the tumor may be as little as 5 - 10% of the doses accumulating in normal organs. In this way, the anticancer effect is decreased and toxic effect to normal cells is highly increased. Accordingly, fears of harming the patient can often lead to a reduction in the dose of anticancer treatment administered to the patient. Such lower- than-optimal doses elicit incomplete tumor responses which can raise other problems, such as, for example, disease relapse and drug resistance. In light of the foregoing, targeting drugs to the cancer cells may help improve the outcome of existing cancer therapies.

Several strategies have been used or proposed for use in the diagnosis of cancer and for targeting the delivery of chemotherapeutic drugs to tumor cells. These strategies include the use of aptamers, bispecific antibodies, engineered antibody fragments, and tumor-homing peptides. Among these approaches, the use of antibodies has received favorable attention for a number of years.

In contrast to cell-surface-targeting antibodies, peptides have shown promise regarding tumor targeting as a delivery agent. Because of their small size, short half- life, and lack of immunogenicity, peptides have been suggested as an alternative to antibodies. In addition, peptides are nearly invisible to the immune system and are not uptaken in the reticuloendithelial system as are antibodies, and thus are expected to cause minimal or reduced side effects to the patient's bone marrow, liver and spleen.

In recent years, techniques such as peptide phage display have uncovered a number of peptides that can be used to target different tumors and cell types. Among these peptides, tumor homing peptides with RGD and NGR sequences have received particular attention. The RGD and NGR peptides target the α_νβ integrin and aminopeptidase N receptors, respectively, in the tumor cells and vasculature. A dodecapeptide, GE11, has also been reported that binds to epidermal growth factor receptor (EGFR) over-expressed by tumors. A 12-mer peptide, pi 60, with a yet unidentified receptor, has been shown to bind the human cancer cell lines MDA-MB- 435, MCF-7, and WAC-2 strongly and specifically. In vivo biodistribution experiments in tumor-bearing mice have shown that the main uptake of pi 60 was in tumor cells as compared to cells in non-tumour organs. Accordingly, pi 60 may show promise in the development of targeted drug delivery systems.

For an agent to be useful as a targeting vehicle, the agent must show selective binding to the tissue of interest and display limited uptake by healthy tissues. Peptides identified by methods such as phage display may be chemically manipulated for better binding to tumour cells and metabolic stability. For instance, peptide arrays for screening a library of designed peptides may be used to complement random phage display screening. For example, short peptides covalently bound to a solid surface may display specific binding affinity to cells, which are passed over the peptides. Such peptide/protein arrays are being used for several applications in the biomedical and biotechnology fields.

There is a need for the identification and development of peptides having increased binding affinity to cancer cells, such as, for example, breast cancer cells, wherein the peptides are non-cytotoxic, are short and easily internalized, and proteolytically stable. There is also a need to develop a peptide array-whole cell- binding assay for screening peptides that have high purity and bind specifically to cancer cells. Further, there is a need to develop and utilize specific peptides having higher binding affinity for targeting cancer cells and for providing targeted drug delivery.

SUMMARY

Methods to study peptide array-whole cell interactions are described. These methods may allow for the identification and development of peptides, or analogues thereof, that are capable of targeting specific cancer cells, such as, for example, breast cancer cells. More specifically, the methods may provide for the identification and development of peptides, or analogues thereof, having increased binding affinity for cancer cells as compared to their binding affinity to healthy cells.

Assays for monitoring the binding of peptides to cancer cells are also described. The peptide array-cell binding assays may be useful in the identification of short, cancer targeting peptides having high purity, as well as in the generation of diagnostic tools for cancer. Peptide assays that may be used for the identification of peptides that bind to cancer cells and/or the identification of receptors are also described.

Therefore, according to one aspect of the present invention, there is provided at least one peptide, or analogue thereof, that is capable of specifically binding to cancer cells. More specifically, the at least one peptide or thereof is capable of binding to specific cancer cells with an increased binding affinity than its affinity to healthy cells. In one embodiment, the peptide or analogue thereof may comprise the sequence WXEAAYQRFL. In another embodiment, the peptide or analogue thereof may comprise the sequence RGDPAYQGRFL.

According to another aspect, there is provided a modified peptide or analogue thereof, wherein the modifications may provide an increase in binding affinity to cancer cells. In one embodiment, the modification may comprise a substitution of an alpha amino acid with a D-amino acid. According to yet another aspect, the modifications may comprise a peptide or analogue thereof having mixed α/β peptidomimetics, wherein the mixed peptide or analogue thereof provides increased binding affinity and selectivity to cancer cells and maintains proteolytic stability. In one embodiment, the mixed peptide or analogue thereof may comprise a substitution of an alpha amino acid residue with a synthesized β3 amino acid derived from L- aspartic acid bearing the same side chain into certain positions in the inner part of the peptide.

According to another aspect, there is provided a method of determining peptide sequences which bind to cancer cells comprising (a) synthesizing a series of peptides on a membrane; (b) spotting each peptide in a pre-determined pattern on the membrane such that the peptides are bound to the membrane; (c) creating a duplicate membrane; (d) passing healthy cells over the first membrane; (e) passing cancer cells over the duplicate membrane; (f) determining the binding of the healthy cells to the peptides; (g) determining the binding of the cancer cells to the peptides; (h) comparing the binding of the (f) and (g); and (i) assessing which peptides display higher binding to cancer cells as compared to healthy cells.

According to another aspect, there is provided a method for the treatment or diagnosis of cancer comprising administering a peptide as identified by the methods as described herein to a subject.

According to yet another aspect of the method, there is provided a method for the treatment or the diagnosis of cancer comprising administering a peptide or analogue thereof having the sequence WXEAAYQRFL to a subject or the sequence RGDPAYQGRFL to a subject. Also included is the use of these peptides/analogues in the identification of cancer cell receptors.

There is also provided an assay device comprising an array of peptide molecules or analogues thereof on a membrane, wherein the peptide molecules/analogues are capable of interacting with healthy and/or cancer cells, wherein the cells that are capable of interacting with the peptides/analogues can be monitored by fluorescent labeling, and wherein the pattern of interaction between the peptide molecules and the cancer cells is indicative of the peptide's ability to be used as a marker for cancer.

According to yet another aspect of the method, there is provided a method for diagnosis of cancer comprising: (a) obtaining a biological sample from a human or non-human subject; (b) synthesizing an array of peptides on a membrane; (c) passing the biological sample over the peptide array to allow the cells in the sample to bind to the membrane; (d) washing the membrane; (e) fluorescently labeling the cells bound to the membrane; and (f) assessing the pattern of fluorescent labeling.

BRIEF DESCRIPTION OF THE DRAWINGS

Table 1 shows the amino acid sequences and the cell adhesion capacity of the pi 60 based peptide array library.

Figure 1(A) is a schematic representation of a cellulose membrane (amino- PEG500-UC540) showing the functionalization on the surface for spot synthesis. The surface has a loading capacity of 400 nmol/cm2 delivering peptide concentration of 50 nmol/spot (12.56 mm2).

Figure 1(B) is an image of the peptide array at λ_εχ= 465 nm and _em= 535 nm showing 70 peptide spots in duplicate.

Figure 2 is a graph showing the net fluorescence intensity of HT- 1080 cells loaded with different dyes. Cells (103) were loaded directly to the cellulose membrane (light grey) or to the peptide (CVLNGRXEC) spot on the cellulose membrane (dark grey). Membrane was imaged using Kodak Imager at lex= 465 nm, lem= 535 nm for CyQUANT and CFSE, and at lex= 385 nm, lem= 460 nm for DAP1.

Figure 3(A) is a graph showing the average CyQUANT fluorescence intensity of the peptide bound to MDA-MB-435 cells. Figure 3(B) is a graph showing the average CyQUANT fluorescence intensity of the peptide bound to MCF-7 cells. Figure 3(C) is a graph showing the average CyQUANT fluorescence intensity of the peptide bound to HUVEC cells.

Figure 4 shows some of the amino acid sequences of peptides identified from the present peptide array that displayed preferential binding profile to cancer cells. The changes in the peptide sequence compared to the wild type pi 60 are highlighted in bold. Sequence of pi 60 is shown for comparison.

Figure 5 is a schematic representation of solid phase peptide synthesis of FITC-βΑ-Ι Ι or FITC-1 1.

Figure 6(A) shows a fluorescence microscopy image of MDA-MB-435 cells after incubation with the FITC-1 1 peptide. Figure 6(B) shows a fluorescence microscopy image of MDA-MB-435 cells after incubation with the FITC-18 peptide. Figure 6(C) shows a fluorescence microscopy image of MDA-MB-435 cells after incubation with the FITC-40 peptide. Each incubation lasted for 30 minutes at a peptide concentration of 10-5 mol/L. Cell nuclei were stained blue with DAPI. Figure 6(D) shows peptide uptake by the MDA-MB-435 cells, and figure 6(E) shows peptide uptake by the HUVEC cells (E) as measured by flow cytometry.

Figure 7 shows FACS analysis for the competitive binding of Peptide 18, showing (A) autofluorescence of MDA-MB-435 cells; (B) fluorescence of cells after incubation with FITC-18 (10^~5 mol/L); and (C) fluorescence of cells after incubation with FITC-18 ( 10^~5 mol/L) in the presence of 100-fold excess Peptide 18.

Figure 8(A) shows FACS analysis of the competitive binding of various peptides in MDA-MB-435 cells (Top left), after incubation with 10^"5 mol/L FITC-1 1 (Top right); after incubation with FITC-1 1 (10^~5 mol/L) in the presence of 100-fold excess 1 1 (middle left) or RGDfK (middle right). Figure 8(B) shows the fluorescence image of binding and internalization of the FITC-1 1 peptide by MDA- MB-435 cells incubated for 30 minutes with FITC-1 1 (10^~5 mol/L) alone. Figure 8(C) shows the fluorescence image of binding and internalization of the FITC-1 1 peptide in the presence of unlabelled pi 60 (10^"4 mol/L). Ten slices from the top to the middle of the cells were extracted using the Z-stack scan mode of confocal fluorescence microscope.

Figure 9 depicts the sequence, mass and HPLC analysis of the synthesized peptide 18 analogues.

Figure 10 is a FACS analysis for the comparative binding of the peptide analogues 18 - 1 through 18 -10 to difference breast cancer (MDA-435, MDA231 and MCF7) and normal primary cell lines (MCF-I OA and HUVEC) compared to that of autofluorescence of cells only, and to positive control peptide 18. Fluorescence of cells was examined after incubation with 10 5 mol/L FITC-labelled peptides for 30 mins.

Figure 1 1 demonstrates the amino acid sequences of peptides 18-4, 18-9, and 18-10, respectively, that are capable of specifically targeting breast cancer cells. The peptide sequence modifications compared to peptide 18 are highlighted in bold, and sequence listing for peptide 18 is shown for comparison. Figure 12 demonstrates a RP-HPLC analysis of the serum degradation solutions of mixed α/β-peptide 18-4 (RHS), compared to a-peptide PI 8 (LHS) after incubation for different time intervals, namely, 0 h, 30 min and 24 h. PI 8 (or 18) is degraded in less than 30 minutes whereas 18-4 is stable for more than 24 hours. Peptides elute around 21-24 minutes. All the other peaks are from the media.

Figure 13 shows the stability of peptides 18-4, 18-9 and 18-10 compared to PI 8 in human serum.

Figure 14 shows fluorescence microscopy images of MDA-MB-435 and HUVEC cells after incubation with FITC-18-4 (Figure 14A), FITC-18-9 (Figure 14B) for 30 minutes compared to cell only (Figure 14C) and at a peptide concentration of 10-5 mol/L. Cell nuclei were stained blue with DAPI.

Figure 15 shows optical sectioning using confocal laser microscopy showing intracellular distribution of peptide 18-9 in MDA-435 cells after incubation for 30 min at 37°C. The focus plane was changed from bottom to top, and representative photographs are shown.

Figure 16 shows the cytotoxicity of peptides 4, and 9 against MDA-435 breast cancer cell line 48 h incubation, in vitro. Doxorubicin was used as positive control. Cells were treated with 7 different concentrations of peptides. After 48 h incubation, cell viability was estimated by MTT assay and expressed as percentage of untreated controls. The data represent the mean ± S.D. of two independent experiments, and each concentration was done in triplicate.

DETAILED DESCRIPTION OF THE INVENTION DEFINITIONS

BOP - Benzotriazole-l-yl-oxy-tris-(dimethylamino)

phosphoniumhexafluorophosphate.

DIC - N,N'-diisopropylcarbodiimide.

DMF - N,N-dimethylformamide (DMF).

Fmoc -9H-fluoren-9-ylmethoxycarbonyl.

Fmoc-SPPS - 9H-fluoren-9-ylmethoxycarbonyl-solid phase peptide synthesis.

FITC - Fluorescein isothiocyanate. Healthy cells - This means cells or cell lines which do not exhibit the symptoms of cancer as those symptoms would be understood by a person skilled in the art, for example normal cells such as Human umbilical vein endothelial cells (HUVEC). HOBt - 1 -hydroxybenzotriazole.

HPLC - High performance liquid chromatography.

MALDI-TOF -matrix-assisted laser desorption ionization time-of-flight.

NMM - N-methyl morpholine.

PEG - polyethylene glycol.

Peptides - The synthesized peptides as herein defined, including any analogues thereof.

Preferential binding to cancer cells: This means that the peptides show higher binding to cancer cells or cancer cell lines as compared to normal or healthy cell lines when the binding ability is compared using a standard binding assay. For example, the assay may include synthesizing peptides on a membrane, passing both cancer cells and healthy cells over the membrane and comparing the fluorescence of cancer cells and healthy cells.

RP-HPLC - Reverse phase high performance liquid chromatography.

SPOT - a method of peptide synthesis done in spots.

TFA - trifluoroacetic acid.

DETAILED DESCRIPTION

The present invention relates to methods to screen peptides, or analogues thereof, that are capable of targeting specific cancer cells. The methods may be used to detect the presence of a specific receptor on cancer cells and may be used to assist in the identification of a specific biomarker for cancer. The peptide array-cell binding assay may also be used in the generation of diagnostic tools for cancer. The peptides identified by the techniques may be used in a method to treat and/or diagnosis cancer.

According to one embodiment of the method, a peptide array is prepared by synthesizing a series of peptides, or analogues thereof, on a cellulose membrane. The peptides may be an 8-mer to a 15-mer, for example. The membranes that may be used for synthesizing the peptides may be acid-hardened for improved stability, for example, at a pH range of 1 to 14. In addition, the cellulose membranes may be derivatized with a hydrophilic spacer such as polyethylene glycol. The spacer allows the peptide to be away from the surface of membrane, which may then enhance peptide-ligand binding. The surface of the membrane may also be derivatized with a free amino terminal group or the membrane may be purchased with an amino acid already attached. A membrane that may be suitable in this regard is the amino- PEG5000 cellulose membrane-UC540 (Invatis, Germany). As shown in Figure 1, a cellulose membrane that may be used in the instant invention is a membrane, which is derivatized with beta-alanine.

The peptides may be synthesized on the membrane using, for example, the known SPOT synthesis technique. SPOT synthesis is a technique for positionally addressable, parallel chemical synthesis on a continuous cellulose membrane support, which has been automated. The SPOT method uses standard Fmoc chemistry based on SPPS on cellulose membranes. The procedure is a step wise method of adding activated amino acids to the membrane to build a peptide. In this technique, Fmoc amino acids are activated and spotted on the membrane. For example, Fmoc amino acids may be activated with, for example, HOBt, DIC, BOP, HATU or HBTU. A skilled person would understand that the peptides may be synthesized using a robotic synthesizer and the purity of the amino acids analyzed using techniques such as MALDI mass spectroscopy and RP-HPLC. The membranes with the peptide sequences may then be exposed to normal or healthy cell lines and to cell lines which are tumorous or cancer cell lines. For example, two cancer cell lines which may be used in this regard are MDA-MB-435 and MCF cell lines. MDA-MB-435 is a melanoma cell line that was considered as a model breast cancer cell line for several years. MCF-7 is an adenocarcinoma human breast cell line. HUVEC is a normal cell line that may be used for comparison.

The cells may then be allowed to bind directly to the peptides bound to the membrane. The bound cells may then be labeled with a dye such as a fluorescent dye. For example, dye such as CyQUANT dye may be used. CyQUANT dye binds specifically to the nucleic acids in cells. CyQUANT dye has low intrinsic fluorescence and displays large fluorescence enhancement and high quantum yield upon nucleic acid binding, which enhances detection. Dyes, such as CFSE and DAPI, may also be used. However, these later dyes may not able to detect cells to the same degree of sensitivity as the CyQUANT dye.

The relative cell adhesion of the peptides may be estimated based on a comparison of the fluorescence of the peptides bound to the healthy cell line and the fluorescence of the peptides bound to the cancer cell lines. Peptides which show enhanced binding to cancer cells may be used for further investigation. For example, fluorescence based experiments, such as flow cytometry and confocal microscopy may be used to monitor the affinity and specificity of the selected peptides in in vitro cell binding experiments. FITC labeling may be used to label peptides which are incubated with MDA-MB-435 cancer cells or HUV.EC cells (as negative controls). Techniques such as fluorescence microscopy may be used to detennine the location of the binding of the peptides to the cells.

The peptides that may be synthesized on the membrane represent peptides that have various substitutions over a known or control peptide. For example, a peptide such as pi 60, which is known to show preferential binding to cancer cells may be used as a control peptide. Substitutions may be made in the p i 60 peptide to determine, for example, the residues that are important for binding to the cancer cells. For example, techniques such as alanine scanning may be used to insert Ala at various positions along the peptide. The binding efficiency of all the peptides can then be compared. Alternatively, peptides can be synthesized which have mutations in the N- or C-terminal residues. Other substitutions, such as the replacement of a C-terminal hydrophobic Phe or Leu with a basic Lys residue, may be made. Peptides may also be synthesized with amino acids such as norleucine.

The peptides identified as having preferential binding to cancer cells may be used in the identification of receptors on cancer cells and/or for the diagnosis of cancer. By "receptor", it is meant a molecule on a cancer or tumor cell which binds to or is capable of binding to, a peptide or series of peptides or to a peptide fusion protein. Any of the peptides identified as having preferential or enhanced binding to cancer cells as compound to normal cells may be used in the treatment and/or diagnosis of cancer. Any of the peptides may also be used as a diagnostic tool for individuals with a propensity to develop cancer.

The peptides identified by the methods described herein may be also used to target drugs and biomolecules to cancer cells. For example, the peptides may be conjugated to micelles for the targeted delivery of anticancer drugs to cancer cells. The present invention also extends to pharmaceutical compositions comprising any of the peptides identified by the methods as described herein. Such a composition may also contain one or more pharmaceutically acceptable carriers and/or diluents.

EXAMPLES

Example 1

By way of example only, a method is now described with reference to one particular embodiment. The following description is in no way intended to be limited to one particular embodiment.

Peptide Array Synthesis

A library of seventy peptides was synthesized in an array format on a cellulose membrane (provided by Invatis, Germany) using the SPOT synthesis. The surface of the cellulose membrane is derivatized with a PEG linker and a free amino terminal group to allow for synthesis of the peptide array. The technique for SPOT synthesis is known.

Figure 1(A) shows a schematic representation of a cellulose membrane (amino-PEG₅oo-UC540, obtainable by Invatis, Germany) showing the functionalization on the surface for SPOT synthesis. The cellulose membrane has a loading capacity of 400 nmol/cm². The peptides were synthesized in duplicate by covalent conjugation to the free amino functional group using a step-wise Fmoc-SPPS procedure. To remove an Fmoc from a growing peptide chain, basic conditions may be used (such as 20% piperidine in DMF). Removal of side-chain protecting groups and peptides from the resin may be achieved by incubating in trifluoroacetic acid (TFA), deionized water, and triisopropylsilane.

Each peptide was synthesized at a concentration of about 50 nmoles and was spread on the membrane in a spot with a diameter of about 4 mm. Accordingly, the peptide concentration per spot was 12.56 mm². Each spot was separated from the next peptide (spot) by 8 mm evenly distributing 140 peptide spots on the membrane surface as shown in Figure 1 (B). More specifically, figure I B shows an image of the peptide array at λ_εχ= 465 nm and _em= 535 nm (which refers to the excitation and emission wavelengths) showing 70 peptide spots in duplicate, for a total of 140 spots.

Peptide Library Design. Table 1 shows the sequence of the peptide library. Peptide 1 or p i 60 is a linear 12-mer peptide cairying a net charge of zero. A series of peptide sequences based on pi 60 sequence were designed to investigate the ability of the peptides to selectively bind to cancer cells. The designed peptide sequences ranged from 9-mer to 1 1-mers. It has been shown that the N -terminal Val and Pro residues do not contribute toward neuroblastoma WAC2 cell binding. Accordingly, in some of the peptides, the first two N-terminal amino acids (Val-Pro) of p i 60 sequence were deleted.

A summary of the peptides that were designed is as follows:

Peptides 1-5:

Peptide 1 is the control peptide, pi 60; Peptide 2 is pi 60 with Nle substituted for Met4; Peptides 3 and 4 are the negative controls where four important C-terminal residues from pi 0 have been deleted; Peptide 5 is the peptide with a tumor homing RGD motif.

Peptides 6-13:

Peptides 6-13 are fusion sequences containing RGD and/or NGR incorporated into the p i 0 sequence. The rationale behind this design strategy was to make the peptide more specific for the cancer cells by targeting multiple receptors overexpressed on the surface of the cancer cells. Peptides 6 and 7 have an RGD(E) in the N-terminal region, whereas, peptides 8 and 9 have a Q(N)GR motif introduced toward the C-terminal. Peptides 10-13 have both the RGD(E) and the Q(N)GR mollis present in the sequence.

Peptides 14-23: Peptide 14 is essentially pi 60 with the first two amino acids deleted and Met replaced with Nle.

Peptides 15-23 represent Ala scan where each amino acid is substituted with alanine at a time. Peptides 24-50:

The third set of peptides are 10-mers and represent the Ala scan Peptides 14- 23, except that in Peptides 24-50, the N- (Trpl or Nle2) or C-terminal (Phe9 or Leu 10) residues were replaced with different amino acids. For instance, Trp l was substituted with conservative mutations such as Leu, Gin, Tyr, Phe, 4-chloro-Phe, and D-Phe or orthogonal amino acids such as acidic Glu or basic Lys residues. Likewise, Nle2 was substituted with Leu, D-Leu, Gin, Thr, Glu or Lys.

Peptides 38-50:

In Peptides 38-50, the C-terminal Phe9 and LeulO were replaced. Peptides 51-70:

Peptides 51-70 were 9-mers having three N-terminal residues from pi 60 deleted. It has been shown previously that an 8-mer peptide (pi 60-8-1) with four N- terminal residues deleted from the pi 60 sequence shows better affinity than pi 60 for the cancer cells. Therefore, the four -terminal residues in pi 0 may not be critical for binding and may be removable. Peptides 52-70 were the same as Peptides 32-50, but with the N-terminal Trp deleted.

Peptide array- Whole cell binding assay. Peptides attached to the solid support cellulose membrane were directly screened for cell binding. The cellulose membranes (provided by Intavis, Germany) were acid hardened for improved stability (pH 1-14) and have a hydrophilic spacer. The hydrophilic spacer (PEG) reduces the background as the peptide is build away from the cellulose surface. These properties of the membrane make the peptide arrays synthesized on them suitable for direct cell binding assays. Further, the peptide array membrane can be regenerated after the cell binding assay, allowing for repetitive use of the same peptide array. Generally, peptide arrays synthesized on cellulose membranes show affinity toward different binding moieties and accordingly, the peptide arrays have been used mainly for studying peptide-protein and peptide-antibody interactions.

The peptides were screened for specific binding to cancer cells using two cancer cell lines, MDA-MB-435 and MCF-7, and a normal cell line, HUVEC cells. The cells were allowed to bind directly to the peptides and the bound cells were labeled with the CyQUANT dye. CyQUANT dye has low intrinsic fluorescence and displays large fluorescence enhancement and high quantum yield upon nucleic acid binding, both of which enhance the signal. Using a representative 9-mer peptide on the cellulose membrane, CyQUANT allowed detection of as low as 1000 cells when imaged using the Kodak imager.

As shown in Figure 2, other dyes, such as CFSE and DAPI, were not able to detect cells bound in this range or enhanced the intrinsic fluorescence of the peptide upon cell binding. In comparison, the CyQUANT dye showed minimal enhancement of the intrinsic fluorescence of the peptides upon cell binding.

The relative cell adhesion of the peptides was estimated based on the fluorescence of the bound cells as shown in Figure 3. In particular, Figure 3 shows the average CyQUANT fluorescence intensity of the peptide bound cells: MDA-MB- 435 (A), MCF-7 (B), and HUVEC (C) cells (75 x 10³ cells/mL). The cells were incubated with the peptide array for 4 hr at 37 °C, followed by labeling the cells with the CyQUANT dye. The fluorescence intensity was measured using a Kodak imager at 465 nm excitation and 535 nm emission. The results are presented as mean fluorescence intensity ± S.D. All the peptides were compared to the wild type pi 60.

Table 1 shows the relative cell adhesion ratio for each peptide binding to cancer or normal cells. The affinity of the peptides for the cancer cell lines MCF-7 (as shown in Figure 3A) and MDA-MB-435 (as shown in Figure 3B) were similar, whereas the HUVEC cells showed very little binding to the majority of the peptides (as shown in Figure 3C). Based on these cell adhesion ratios, several peptides that displayed better cell binding compared to the p i 60 were identified.

Five peptides (10, 1 1 , 1 8, 40, and 47) displayed a 1.7 - 2.7 fold increase in binding to cancer cells compared to pi 60. Among the fusion Peptides 6- 1 , Peptides 6, 7, 10, and 1 1 displayed increased binding to MDA-MB-435 cells as compared to HUVEC cells. Most of the fusion peptides (7-13) also showed higher cell adhesion to the MCF-7 cells. In this context, "higher" means that the binding which was quantified was greater than pi 60, or in other words, the relative adhesion ratio is greater than 1 for cancer cells or less than 1 for HUVEC cells. Peptide 1 1 was chosen for further investigation as it displayed a cell adhesion ratio of 0.7 for the HUVEC cells as compared to MDA-MB-435 cells, implying that Peptide 1 1 has a higher affinity for cancer cells.

Of the alanine scan peptides (14-23), Peptide 18 displayed the highest binding affinity to cancer cells relative to pi 60. The relative cell adhesion ratio was 2.2 and 2.7 for MBA-MB-435 and MCF-7 cells, respectively. However, Peptide 18 also showed high binding to the HUVEC cells, with a relative cell adhesion ratio of 1.5. In Peptide 18, Pro was substituted with Ala as compared to pi 60. However, it is possible that the Pro residue could be replaced with an alternative amino acid to produce a peptide with better binding ability to cancer cells as compared to pi 60. In this regard, it has been found, using phage display, that peptides (designated pi 61 and p25) which had sequences very similar to pi 60 but with Pro substituted with Gin or Met, bound to a variety of tumor cells, and showed little or no binding to the normal cells.

In the present Example, several peptides, such as Peptides 24-70, having mutations in the N- or C-terminal residues, with respect to pi 60, displayed enhanced binding to the cancer cells. The two peptides that showed the highest affinity (up to 2.2 - 2.8 fold) for the cancer cells compared to pi 0 in this category were Peptides 40 and 47. In Peptides 40 and 47, a C-terminal hydrophobic Phe or Leu was replaced with a basic Lys residue, which resulted in these analogues having a positive charge (net charge +1 ).

Peptides 32-36, where the N-terminal Me was substituted, showed 1.4 - 2.1 fold better binding compared to pi 60. The 9-mer Peptides 52-70 generally showed either the same or less binding to the cancer cells compared to p i 60. These results suggest that the N-terminal Trp residue may be important in binding to cancer cells. Only two 9-mer peptides, Peptides 60 and 67, showed higher binding (range 1.2-2.0 fold) than i 60. The results support that the increase in the net positive charge of the peptides in the C -terminus may increase cell adhesion. This may be due to electrostatic interactions between the peptides and the cancer cells. However, introducing a positive charge (basic residue) in the N-terminal region, such as was done in Peptides 26 and 35, does not increase binding affinity to the same extent as observed for Peptides 40 and 47.

Affinity and Specificity of Selected Peptides. Three peptides, 1 1 , 18, and 40, which showed preferential binding to the cancer cells, were further investigated. The amino acid sequences of these peptides are shown in Figure 4, with the changes in the peptide sequence compared to the wild type pi 60 highlighted in bold.

Fluorescence based experiments, such as, flow cytometry and confocal microscopy were used to monitor the affinity and specificity of the selected peptides during in vitro cell binding experiments. Peptides pi 60, 1 1 , 18, and 40 were labeled with F1TC in the N-terminus via a β-alanine linker as shown in Figure 5. FITC- labeled peptides were incubated with the DA-MB-435 cancer cells or the normal HUVEC cells (as negative controls) for 30 min at 37 °C to characterize the peptide- cell binding.

Figure 6 shows fluorescence microscopy images of MDA-MB-435 cells after incubation with (A) FITC-1 1 , (B) FITC- 18, and (C) FITC-40 for 30 minutes at a peptide concentration of 10^"5 mol/L. Cell nuclei were stained blue with DAPI. Peptide uptake by the MDA-MB-435 cells (D) or HUVEC cells (E) measured by flow cytometry. The peptides (10^"5 mol/L), FITC-1 (pi 60, solid gray), FITC-1 1 (black dotted), FITC-18 (black solid), and FITC-40 (gray dotted) were incubated with the cells for 30 min at 37 °C. Auto-fluorescence of the cells is shown in grey. As shown in Figures 6A-C, the fluorescence microscopy images of the MDA-MB-435 cells show that the peptide analogues surrounded the nuclei of the cells. The FITC-labeled peptides, FITC- 1 1 , FITC-18, and FITC-40, bound to the surface of the cells and were present in the cytoplasm. The binding of the peptides to the cells was further confirmed using flow cytometry. Figures 6D and 6E show the fluorescence for the different peptides bound to MDA-MB-435 and HUVEC cells, respectively. FITC- labeled pi 0 was used as a positive control. In general, MDA-MB-435 cells showed higher FITC-fluorescence than the HUVEC cells when bound to FITC-labeled peptides. This suggests selective binding of the peptides to the cancer cells. MDA- MB-435 cells showed the highest FITC-fluorescence when bound to peptide 18, followed by 1 1 , pi 60, with the lowest binding for 40. These results indicate that Peptides 18 and 1 1 have higher affinity for the MDA-MB-435 cells compared to pi 60. Peptide 40 showed weak binding compared to pi 60, in contrast to what was observed in the peptide array-cell binding assay (relative cell adhesion ratio 2.7). This may be due to the high auto-fluorescence of peptide 40 which was enhanced after cell binding. Figure 7 shows a competition experiment wherein a 100-fold excess of unlabeled 18 was used. This excess caused an up to 95% decrease in the FITC- fluorescence of the FITC-18 bound cells. Similar competitive experiments for Peptide 1 1 showed very different results. As shown in Figure 8A, only a slight decrease (3%) in the fluorescence was observed when the cells were incubated with FITC-1 1 in the presence of 100-fold excess unlabeled 1 1. However, a substantial (48%) decrease in FITC-fluorescence was observed when FITC-1 1 was incubated in the presence of unlabeled RGDfK (100 folds). The fluorescence was not completely wiped out in the presence of either excess 1 1 or RGDfK, suggesting multiple binding sites for 1 1 on the cancer cells. As Peptide 1 1 is a fusion peptide with RGD and QGR sequences inserted in the p i 60 sequence, it can bind to the α_νβ? integrin, aminopeptidase N (CD13), and the putative pl 60 receptor on the cancer cells. MDA- MB-435 and MCF-7 cancer cells are known to express α_νβ? integrin, however, expression of APN (CD 13) by these cells is minimal.

Figure 8(A) shows FACS analysis for the competitive binding of the peptides, showing autofluorescence of MDA-MB-435 cells (top left), fluorescence of cells after incubation with 10^"5 mol/L FITC- 1 1 (top right), and fluorescence of cells after incubation with FITC-1 1 (10^~5 mol/L) in the presence of 100-fold excess 1 1 (bottom left) or RGDfK (bottom right). Fluorescence images showing binding and internalization of FITC- 1 1 by MDA-MB-43 cells were also obtained (Figures 8B and 8C). The cells were incubated for 30 minutes with (Figure 8B) FITC-1 1 (10^{" 5} mol/L) alone or (Figure 8C) in the presence of unlabelled pl 60 (10 ⁴ mol/L). Cell nuclei were stained blue with DAPI (λ_εχ= 360 nm, X_em= 460 nm) and overlaid with FITC -peptide fluorescent images (λ_εχ= 494 nm, λ_ειη= 520 nm). Ten slices from the top to the middle of the cells were extracted using the Z-stack scan mode of confocal fluorescence microscope. The scale bar is 10 μιη.

Cancer cell binding and uptake was also confirmed by visualization using the confocal microscopy. Figure 8B shows a z-stack scan of peptide FITC-1 1 bound to the MDA-MB-435 cells. The scan shows that peptide is present at the surface as well as inside the cells. Figure 8C shows similar images of FITC-1 1 bound to cells in the presence of 10-fold excess unlabeled pi 0. A decrease in the fluorescence is observed when pi 60 is present as a competitor. This decrease in fluorescence is of the same order as observed with unlabeled 1 1 as a competitor using flow cytometry (Figure 8A). These results, along with the FACS experiments, confirm that Peptide 1 1 binds to multiple receptor sites. Thus, targeting multiple receptors with a fusion peptide such as Peptide 1 1 can lead to peptides with higher affinity.

The binding of Peptide 18 to the cancer cells is presumed to be very similar to the pi 60 peptide, as there is single mutation in Peptide 18 compared to pi 60. Peptide 18, like pi 60, showed almost complete removal of fluorescence due to cell bound FITC-18 in the presence of unlabeled 18, as shown in Figure 7. The receptor mediated specific binding of pi 60 by the MDA-MB-435 and MCF-7 cancer cells is known. A large decrease in cell binding was observed when ¹²⁵I-labeled pi 60 (10^~6 mol/L) was allowed to bind cancer cells in the presence of unlabeled pi 60 as a competitor. The radio-labeled and FITC-labeled pi 60, which was inhibited in the presence of unlabeled pi 60, was internalized in the cancer cells. Generally, peptides like Peptide 18 bind to cancer cells in the low micromolar range. This may seem to be high concentration range in comparison to the other targeting moieties that have been reported. The HER-2 and EGFR specific antibodies and the cancer-specific aptamers bind to the cells in the nanomolar range. However, the peptides used in the present experiments may be considered to be small (10-mers) compared to other known antibodies (58 amino acids) and aptamers (39-85 nucleotides). These peptides can behave like small molecule drugs and may present fewer in vivo problems such as immunogenicity and toxicity as compared to large molecules.

Peptides 1 1 and 18 may be conjugated to micelles for targeted delivery of anticancer drug to cancer cells in vitro and in a mouse model. It is known that micelles having been surface coated with RGD peptide show an increase in the therapeutic efficacy of doxorubicin for sensitive and resistant cancers.

Equipment

Peptide array was made using a semiautomatic robot AutoSpot ASP222 (Intavis , Germany). Solid phase synthesis of peptides on Wang resin was done using manual synthesis. HPLC purification and analysis were carried out on a Varian Prostar HPLC system (Walkersville, MD, USA) using Vydac CI 8 semi -preparative ( 1 x 25 cm, 5 μιτι) and analytical (0.46 x 25 cm, 5 μηι) columns. Peptides were detected by UV absorption at 220 nm. Mass spectra were recorded on a matrix- assisted laser desorption ionization time-of-flight (MALDI-TOF) Voyager spectrometer (Applied Biosystems) or Waters micromass ZQ. Imaging experiments were done using Kodak Image Station 4000M (USA), Carl Zeiss microscope (Gottingen, Germany), and confocal laser scanning microscopy using Zeiss 510 LSMNLO confocal microscope (Carl Zeiss Microscope systems, Jena, Germany). FACS experiments were performed on Becton- Dickinson Facsort and analyzed by DakoCytomation Summit software. Peptide Array Synthesis

Seventy peptide sequences in duplicates ranging in length from 9 to 12 amino acids were synthesized in an array format on a cellulose membrane using an AutoSpot robot. Peptide arrays were synthesized on an amino-PEG500 cellulose membrane- UC540 (Intavis, Germany). DIGEN software (Jerini Biotools GmbH, Berlin, Germany) provided with the instrument was used for designing the arrays. The surface of the membrane is derivatized with a polyethylene (PEG) linker and a free amino terminal group. The Spot synthesis method was used to build the peptide arrays on the free amino terminal group. Briefly, Fmoc amino acids activated with HOBt and DIC for 15 minutes were spotted on the membrane in 60 nL aliquots per spot by a robotic syringe. The concentration of each peptide (spot) was controlled by the amount of the liquid delivered during the reaction steps. By delivering 60 nL of the activated amino acids (0.25 mM/niL) at intervals of 8.0 mm, a loading of 0.4 μι^ηοΐ/cm² was achieved. After coupling of the Fmoc amino acid, the membrane was removed from the synthesizer and was treated with acetic anhydride (2%) to cap any free remaining amino groups. The C-terminal end of the peptide was anchored to the surface of the amino-PEG500 cellulose membrane through a β-alanine linker. The membrane was then washed and treated with 20% piperidine in DMF for the Fmoc group deprotection. After washing with DMF and IPA, membrane was air dried and carefully repositioned on the robotic synthesizer to repeat the next coupling cycle. These steps were repeated for each amino acid till the end of the sequence. At the end, all peptides were N-terminally acetylated. The final removal of side chain protecting groups was performed by treating the membrane with a cocktail of reagents comprised of 15 mL TFA, 15 mL dichloromethane (DCM), 0.9 mL triisopropylsilane, and 0.6 mL water. The membrane was allowed to react with the cocktail solution in a polypropylene box with a lid for about 3 h. After extensive washing with DCM, DMF, and ethanol, the membrane was dried with cold air and stored in a sealed bag at 4 °C until use.

The peptides synthesized on the cellulose membrane were characterized by preparing a small control library of 9-mer peptides on a β-alanine membrane with a cleavable linker (Intavis, Germany). After the synthesis, peptides were punched out in an ELISA plate using a puncher (Intavis, Germany). Peptides were cleaved from the membrane using a mixture of TFA/triisopropylsilane/H₂0. Each peptide was characterized by MALDI mass spectrometry as well as analytical RP-HPLC. Peptides were found to be >99% pure.

Cell Lines All cell lines were cultivated at 37 °C in a 5% C0₂ incubator. The human cancer cell line MDA-MB-435 was cultured in RPMI 1640 with Glutamax containing 10% FCS (Invitrogen, Karlsruhe, Germany), 100 IU/mL penicillin, and 100 IU/mL streptomycin. The human breast cancer cell line MCF-7 (American Type Culture Collection, Manassas, VA) was cultured in DMEM with Glutamax containing 10% FCS (Invitrogen). Human umbilical vein endothelial cells (HUVEC), from the laboratory of Sandra Davidge, University of Alberta, were cultivated using Endothelial Cell Growth Medium (EGM, LONZA) containing 20% FCS, 2 mmol/L glutamine, 100 IU/mL penicillin, 100 IU/mL streptomycin, and 2 ng/mL basic fibroblast growth factor (Roche Diagnostics, Mannheim, Germany). Peptide Array-Cell Binding Assay: Peptide Screening

The peptide array membrane was soaked in ethanol for 30 seconds to prevent any precipitation of hydrophobic peptides, followed by its incubation in sterile PBS (pH 7.4) for 30 min. The cells were seeded directly on culture dish (75 x 10³ cells/mL) containing the peptide array membrane for 4 hr in serum free media, in order to prevent the proteolytic effect of serum on the peptides. After washing the non-bound cells, the membrane was frozen at -80 °C for 2 hr. The membrane was thawed at room temperature followed by incubation with the CyQUANT dye for 30 min following the manufacturer protocol, it was washed three times with PBS, each for 5 min by shaking on automatic shaker. The membrane was scanned using Kodak imager at 465 nm excitation and 535 nm emission and the net fluorescence intensity of each peptide spot was quantified using Kodak Molecular Imaging Software Version 4.0. The binding affinity of the cells for each peptide (spot) was determined by subtracting the net fluorescent intensity of the peptide itself (autofluorescence). An external standard (set of peptides) was used to calibrate the fluorescence intensity between scans performed on the same day and on different days. After each cell- binding experiment, the bound cells were removed by first washing with ethaiiol for 5 min, followed by treatment with 0.1 N HC1 for 20 min. The peptide array membrane was regenerated by washing with DMF (4 x 20 min), ethanol (3 x 3 min), and finally drying in air. Each cell-binding experiment was repeated twice. The results are presented as average fluorescence intensity (± standard deviation) of two duplicate peptide spots, two scans, and two different experiments, as shown in Figure 3. The relative cell adhesion ratio for each peptide analogue was calculated as the ratio of the average fluorescence of the peptide analogue divided by that of the pi 60 peptide.

Synthesis of FITC-labeled Peptides The FITC-labeled peptides, FITC-pl60, FITC-1 1 , FITC-18, and FITC-40 and the unlabeled peptides, pi 60, 1 1, RGD, and NGR were synthesized by manual solid phase peptide synthesis (SPPS) using Fmoc coupling protocols on Wang resin (0.1 mmol) (Beleid et al. Chem Biol Drug Des 72 (2008) 436-443). After the peptide synthesis, (3-alanine (spacer) was conjugated to the ^'N -terminal amino group followed by fluorescein isothiocyanate (FITC) coupling. FITC was coupled in dark for 20 hr followed by extensive washing of the resin. FITC-labeled peptide was cleaved from the resin, along with the deprotection of the amino acid side chains by TFA treatment. The crude cleaved peptides were precipitated by cold diethyl ether followed by their purification using reversed-phase HPLC. The mass of the products was determined by MALD1-TOF mass spectrometry. Fluorescence Microscopy

MDA-MB-435 cells (50,000) were cultured on the top of cover slip at 37 °C for 24 hr. The medium was removed and replaced with 1 mL of fresh serum free medium, containing FITC-labeled peptides at a concentration of 10^"5 mol/L. The cells were incubated with the peptides for 30 min at 37 °C. After incubation, the medium was removed and the cells were washed thrice with 2 mL serum free medium. The cells were fixed on ice with 2% formaldehyde for 20 min. The formaldehyde was removed by washing with medium (three times). The cover slips were put on slides containing one drop of DAPI-Antifade (Molecular Probes) to stain the nucleus. The cells were imaged under the fluorescence microscope (Zeiss) using green and blue filters with 20x magnification.

The samples prepared for fluorescence microscopy were also used for visualization by confocal microscopy. Confocal laser scanning microscopy was performed with a Carl Zeiss inverted confocal microscope with a 1 OOx oil immersion lens. Confocal stacks were processed using the Carl Zeiss LSM 5 Image software, which also operates the confocal microscope. For the competitive binding, the same experiment was carried out in the presence of unlabeled p i 60 peptide (10^~4 mol/L) as a competitor.

Flow Cytometry Analysis. Fluorescence-activated cell-sorting (FACS) analysis was used to evaluate the binding of the FITC-labeled peptides to the cancer MDA-MB-435 and the normal HUVEC cells. The MDA-MB-435 and HUVEC cells were placed into 6-well plates at a density of 10⁶ and 3 x 10⁵, respecti vely, in 3 mL of culture medium at 37 °C for 24 h. The culture medium was replaced by 1 mL of fresh serum-free medium, containing FITC-labeled pi 60, 1 1 , 18, and 40 peptides at a concentration of 10 ⁵ mol/L^, Cells were incubated with the peptides for 30 min at 37 °C. The media was then removed and the cells were washed 3 times with ice-cold phosphate-buffered saline (PBS) to remove the unbound peptide. The cells were then scrapped from the wells using manual scrapper. The cells were transferred to centrifuge tubes and centrifuged at 1 ,000 rpm for 7 min. The pellet was resuspended in FACS buffer (PBS with 5% FCS, and 0.09% sodium azide), washed once more and then resuspended again in FACS buffer. Untreated cells were subjected to similar steps without any peptide treatment to detect autofluorescence of the cells. The samples were then subjected to the FACS instrument, Becton-Dickinson Facsort to acquire data. The data was analyzed by DakoCytomation Summit software.

Competitive peptide binding assays were performed by incubating MDA-MB-

435 cells with FITC-labeled peptide 1 1 in the presence of a 100-fold excess unlabeled 1 1 or RGDfK. After incubation for 30 min at 37 °C, the cells were washed with ice- cold PBS. Thereafter FACS analysis was performed as described above. Similar experiment was repeated for FITC-18 in the presence of excess unlabeled 18. All the experiments for binding assay were repeated 2-4 times.

Example 2

By way of example only, at least one peptide is now described with reference to one particular embodiment. The following description is in no way intended to be limited to one particular embodiment.

Background

Breast cancer is one of the most serious terminal diseases in women. Different treatment approaches are available which depends on the stage of the disease, including surgical procedures, radiation, chemotherapy and hormonal therapy, gene therapy and antiangiogenic therapy is currently under investigation. A major hurdle associated with current chemotherapy is that they enter normal tissues in the body with indiscriminate cytotoxicity and do not preferentially accumulate at tumor sites. To improve the specific uptake of therapeutic drugs to tumors, different strategies have been developed. One of the most effective strategies is to target anticancer drug preferentially to tumor using engineered antibodies tumor homing peptides and aptamers that have specific receptor on particular types of tumor cells. In contrast to cell-surface-targeting antibody, peptides have shown good promising results regarding tumor targeting as a delivery agent. Peptides are smaller so have excellent tissue penetration properties and easy synthesis as well as conjugation to drugs and oligonucleotides are more feasible. In addition, peptides are nearly invisible to the immune system and are not uptaken in the reticuloendithelial system as antibodies, so expected to cause minimal or no side effects to bone marrow, liver and spleen.

A number of peptides have been identified by peptide phage display for targeting breast cancer cell types. One of these is a 12 amino acid residues peptide identified through phage display has been referred to as peptide pi 60. This peptide displayed high specificity for the breast cancer cell lines MDA-MB-435 and MCF-7 in vitro. With very little binding to control cell lines such as primary endothelial HUVEC cells. Furthermore, in vivo biodistribution experiments in tumor-bearing mice, it showed better tumor uptake and retention after organ perfusion when compared to normal organs, and relative to the well known RGD peptide it showed high accumulation in tumor cells versus normal ones in vivo. Taken all these together, peptide pi 60, therefore, shows considerable promise in the development of targeted drug delivery. Example 1 demonstrates a peptide library screening of p i 60 peptide against breast cancer cell lines (MDA-435, and MCF7) using the present peptide array method. In that example, a peptide 18 analogue was identified that which displayed 3- 4 fold higher binding affinity to cancer cell lines in vitro with a ¾ of 38 μΜ, and showed negligible affinity to HUVEC primary cell line.

Notwithstanding the potential therapeutic use of peptide 18, alpha peptides in general as therapeutic drugs have been largely hampered by their instability toward proteases, which severely diminishes their bioavailability. Moreover, they generally demonstrate poor bioavailability in tissues and organs, thereby presenting a significant hurdle and major impediment toward drug development.

In an attempt to overcome these hurdles, the peptides need to be chemically modified, such that, for example, blood clearance of the peptides can be minimized in comparison with the rate of extravasation at the target sites. In this respect, the introduction of D-amino acids, or unnatural amino acids, and peptide cyclization are the most common strategies to increase peptide enzymatic stability. One such method is the solid phase peptide synthesis (SPPS) of β-amino acid building block derived from L-aspartic acid.

In the present example, two issues are addressed. First, the development of a potent and selective analogue of peptide 18 is addressed. In order to do so, different alpha-amino acids structure modifications on peptide 18 sequence were performed. These modifications were selected based on the previous pi 60 library screening results of Example 1 , and their vitro binding characteristics were investigated. Second, the synthesis of a mixed α/β peptide 18 analogues was addressed in order to increase peptide proteolytic stability, by substituting subsets of a amino acid residue with our in house synthesized β₃ amino acids derived from L-aspartic acid bearing the same side chain into certain positions in the inner part of the peptide. It is contemplated that peptides with these modifications might display the same, if not enhanced, binding properties compared to the original alpha peptide. Finally, the functional and biological consequences of backbone alterations were investigated, and the enzymatic stability in human serum was evaluated. To the best of our knowledge, the synthesis and enzymatic degradation of cancer targeting peptides having beta amino acids incorporated into backbone sequence is not known. Thus, this Example 2 demonstrates a proteolitically stable ₃-substituted cancer targeting peptide.

In the present Example, 10 analogues of peptide 18 were designed,

synthesized, and FITC-labeled (see Figure 9). Binding and internalization of the FITC-labeled peptides into cells were studied in vitro using confocal laser scanning microcopy as well as flow cytometry techniques compared to parent peptide 18. The enzymatic stability of the highly potent analogues was investigated in human serum. The data show that three of the tested peptides 18-4 analogue with D-amino acids substitutions in the labile sites, and 18-9,and 18-10 peptide analogues having 3 beta amino acids substitutions exhibited higher cancer cell binding and exceptional resistance against proteolytic degradation in human serum compared to prototype peptide 18. In addition to that, these peptides have shown increased selectivity to cancer cell lines versus normal cell lines, and impart no cytotoxicity against tested cell lines.

Design and synthesis of peptide 18 analogues

As discussed, peptide 18 was utilized as the starting point in this Example. Structurally, it is a linear 10-mer peptide with a net charge of zero. With respect to parent peptide pi 60, it has P4A, and M3Nle substitutions, and the first two N-terminal amino acids (VI -P2) have been deleted. It displayed 3-fold better binding to breast cancer cell lines (MDA-435, and MCF7) with a very low binding affinity to normal cell line (HUVEC). Before designing the analogues, a detailed structure characterization of peptides pi 60 and 18 using NMR spectroscopy was performed. Structural investigation using 2D-NMR was carried out in 90 % TF A/water mixture and in 100% water. NMR studies in TFE solvent which mimics membrane conditions revealed the presence of helical conformations for the segment 2-9, and it is completely unstructured in aqueous solution. It is known that deletion of this segment can decrease peptide binding affinity, and highlighted the significance of peptide secondary structure for binding affinity to its putative receptor. It is contemplated that the overall physicochemical properties of these helices, and not their precise amino acid sequences, may be responsible for the binding affinity. Accordingly, 2 sets of analogues were designed with an aim of maintaining or increasing the helicity of the peptides.

In the first set of alpha analogues (18-1 - 18-4), substitutions were selected from the pi 60 peptide array screening, whereby the substitutions were shown to increase the binding affinity on the membrane compared to pi 60. The combination of two substitutions at a time in peptide 18 sequences were also performed to examine their effect on the binding affinity. As shown in Figure 9, peptide 18-1 has TrplTyr replacement, peptide 18-2 has Nle2Glu substitution, 18-3 has Nle2Leu, and Phe9Tyr replacements. Peptide 18-4 has the two known labile sites, amino acids Nle2, and Arg8 replaced with D- amino acids to increase stability.

In the second set of analogues, mixed α/β peptides (18-5 - 18-10) were synthesized to improve their enzymatic stability (see Figure 9). In α/β peptides 18-5 and 18-6 the amino acids Nle2, Arg8 and Phe9 (that are known enzymatic cleavable sites) were replaced with β³ amino acids derived from L-aspartic acid. In contrast, α/β peptides 18-7- 18-10 each contain three a to β³ replacement to maximize resistance to proteolysis. These analogues where fabricated by following a known sequence based design strategy, wherein β and a amino acid residues may be distributed in a repeating heptad pattern (ααβαααβ). The replacement sites in the analogues were selected to sample different types of positions in the heptad repeat. The peptides were all synthesized manually on chlorotrityl resin following SPPS protocol, peptide 18 was used as positive control in our experiments. For the synthesis of mixed α/β analogues, beta amino acids were directly incorporated in the sequence, by coupling the elongated chain with the Fmoc aspartic acid-oallyl, and the allyl protection was selectively removed using a mixture of palidium triphenylsilane for three times under nitrogen each for 45 min, coupling of different amines were then carried out to give the corresponding beta amino acid, coupling was carried out for two times each for 5 hours. Coupling with fresh reagents for two consecutive times was better than coupling once for longer times with respect to the degree of peptide purity. Some amines needed longer times for coupling especially methylamine, tyrosine and isobutyl amine. After complete assembly of the peptide on the resin, and removal of N-terminal Fmoc group, the FITC labeling of the peptide in the N-terminus via a β- alanine linker was carried out by coupling FITC with resin bound peptide in presence of DIPEA for 18 hours. Peptides were obtained with high degree of purity almost 95% and purified using RP-HPLC to 99% purity.

In vitro cell binding, specificity and selectivity experiments

To evaluate peptide analogues (18-1 - 18-10) as potential agents for cell targeting, FITC-labeled peptides were screened for specific binding to three human breast cancer cell lines MDA-MB-435, MDA-231 and MCF-7 using flow cytometry. MCF-IOA and HUVEC were used as a control cell line. FITC-labeled peptides were incubated with the cell lines in serum free media for 30 min at 37 °C at a concentration of 10^"5 mol/L, then cell binding and internalization was tested using flow cytometry. No FCS was present in the medium during the incubation to avoid degradation by serum proteins and to allow analysis of the peptide characteristics without the influence of additional variables. FITC-labeled 18 was used as a positive control.

Flow cytometry results as presented in Figure 10 showed that all peptides analogues were showing significant binding to the three breast cancer cell lines (% of labeled cells was almost 80%) compared to cell only (untreated cells). The pattern of uptake was similar in all of the cell lines but slightly higher in MCF-7 at the concentrations tested. This correlates to the notion that pi 60 has around 7 folds higher binding to MCF7 relative to MDA-435. Generally, the L-amino acids alpha analogues 18-1 - 18-3 showed relatively lower binding with respect to mixed α/β peptides, which may be due to their susceptibility to enzymatic degradation. Peptide 18-1 showed 3 fold higher binding compared to peptide 18, replacing N-terminal tryptophan with tyrosine presumably increase the interaction with receptor so increased the binding. Peptide 18-2 having isoleucine 2 replaced with charged glutamic acid showed decrease in binding relative to parent 18. Peptide 18-4, having 2 D-amino acids substitutions at the two labile sits, showed remarkable increase in binding, which may be due to high enzymatic stability, and this reflects that configuration of this two amino acids did not alter the interaction with the receptor. In mixed α/β analogues, all of the analogues showed either equal binding affinity or higher compared to 18 (% fluorescent cells almost 100%). Variation in the position of β-residue in the peptide sequences was tolerated. Peptides 18-9, and 18-10 showed the highest mean fluorescence and highest affinity for cancer cell lines, compared to parent peptide 18. These peptides having tryptophan replaced with β-naphthyl group side chain and alanine replaced with β-alanine side chain would be more hydrophobicity which may increase the interaction with the receptor. It is known that pi 60 having alanine replaced with β-alanine has a more than 2-fold increased binding capacity to WAC 2 cells when compared with native p 160.

On the basis of this data, the three most promisingly binding peptides (18-4, 18-9, and 18-10) were selected for further investigation (see Figure 1 1). As shown in Figures 12 and 13, and discussed in more detail below, peptides 18-4, 18-9 and 18-10 are shown to be proteolytically stable, whereas 18 is not.

The binding selectivity and cell internalization for breast cancer cell line

MDA-MB-435 versus normal cell lines MCF-I OA and HUVEC were carried out using flow cytometry and fluorescence microscopy for each of peptides 18-4, 18-9 and 18-10. Flow cytometry results showed that the three peptides and their parent peptide 18 have significantly lower binding to normal cells (45% fluorescent cells for MCF- IOA, and 35% to HUVEC) versus strong preferential binding breast cancer cells (100%) after 30 min incubation with peptides, suggesting their binding is cancer specific (Figure 10). It seems that peptides have relatively higher binding to normal mammary cell line MCF-10 when compared to that of HUVEC, and this is most likely due to the presence of the same receptor on normal mammary cells but expressed in low level. This selectivity makes them excellent candidates as therapeutic targeting agents for breast cancer. Next, to demonstrate specific binding, the incubation in the presence of unlabeled 18-4, 18-5, and 18-6 peptides (10^-5 mol/L) was performed. Results show that 18-5, and 18-6 caused an up to 75%, and 80% decrease in the FITC-fluorescence of the FITC-18-5, and 18-6 bound cells, respectively (data not shown). Similar competitive experiments result was obtained for peptide 18-4 which showed 80% decrease in binding (data not shown).

In a parallel experiment the binding selectivity and cellular uptake of 18-4 and

18-9 peptides was further studied using fluorescence microscopy in MDA-MB-435 and HUVEC cells. The fluorescently labeled peptides were incubated with cells for 30 min at 37^°C and the distribution was examined. The FITC-labeled peptides FITC-18-4 and FITC-18-9 were found to be bound to the cell membrane of MDA-MB-435 cancer cells, and uniformly distributed inside the cells. In contrast, there was very few peptides bound control HUVEC cells under the same experimental conditions (see Figure 14). To further demonstrate that most peptide molecules were not surface bound but internalized, optical sectioning of MDA-435 cells treated with 18-9 were performed, as shown in Figure 15.

Proteolytic Stability

Having established that peptides 18-4, 18-9, and 18-10 analogues bind with high affinity and selectivity to breast cancer cell lines, the peptides were further examined to determine whether they would be recognized and processed by proteolytic enzymes. Accordingly, the susceptibility of these peptides to proteolytic digestion in human serum compared to alpha peptide 18 was examined. The in vitro stability of pi 60 in human serum was investigated through incubation of pi 60 in human serum and HPLC analysis of serum samples taken at different timesfor 24 h incubation at 37°C. Degradation studies were analyzed using HPLC as this procedure would enable us to isolate possible degradation products directly and analyze them by mass spectrometry. Serum were used in high specific concentrations (25% serum), to be the rate limiting, not the substrate concentration, such that an a -peptide should be degraded within 10 min. The results (see Figure 13) showed that the 3 peptides were completely stable for 24 h, which reveals that ααβαα β backbone, and D- amino acid substitutions at labile sites can confer substantial resistance to proteolytic degradation relative to a-peptide 18. Peptide 18 completely degraded within 30 min (see Figure 12) giving two hydrolysis products. The degraded fragments of 18 eluted earlier, between 14 and 18 minutes, than 18 in HPLC (Figure 12 LHS). The mass spectrometry of these degradation products revealed that the cleavage occurred at NLe2 and Phe9 amino acids. This is in accordance with previous serum stability studied for PI 60 peptide. The greater stability of mixed α/β peptides 9, and 10 likely results from the greater helical propensity of these peptides, as detected by CD, as well the propensity of mixed α/β backbones to protect neighboring amides from proteolytic cleavage.

Cytotoxicity

Peptides 18-4, 18-9, 18-10 and peptide 18 were also evaluated for their cytotoxic effects against MDA-MB-435 breast cancer cell line using MTT assay. Doxorubicin was used as positive control. Peptides were incubated with cells for 24 h. The percentage cell viability was plotted as a function of the peptide concentration, as shown in Figure 16. In the presence of Dox, the cell viability was dramatically decreased; there was complete cell inhibition at 15 Μ concentration. In contrast, all the tested peptides were practically non toxic up to the highest concentration tested (ΙΟΟμΜ). This is consistent with previous results which showed very low cytotoxicity of peptides synthesized from β³ amino acids derived from L-Asp.

Chemotherapy and hormonal therapy play important roles in breast cancer treatment. Nevertheless, emergence of drug resistance, and side effects of these therapeutic regimens necessitates the search for specific tumor targeting agents. In addition, the clinical success of monoclonal antibodies such as Herceptin, Zevalin, Rituxan in the treatment of human cancer has validated the cell surface targeting approach in cancer therapy. It is contemplated that peptides provide better cell surface targeting agents than antibodies, in particularly when used as carriers for cytotoxic payloads such as chemotherapy or radionuclides. This may be due, in part, to peptides being smaller (can penetrate easily), less likely to bind to the reticuloendothelial system such as liver, spleen, bone marrow, and easy to chemically conjugate. Additionally, they are probably not immunogenic if they are short, and administered without any immunoadjuvants. This emphasizes on the continuous demand for short peptide ligands that can serve as cancer specific diagnostic and therapeutic probes. PI 60 peptide and its potent analogue peptide 18 displayed good affinity for breast cancer targeting in vitro. However, despite these promising results, these applicability of these peptides are hampered by its fast metabolic degradation, which might affect its targeting ability. It is known that the stability of peptide pi 60 can be degraded by serum proteases within 2 minutes after injection. Further, it is known that in vivo I-labeled pi 60 released high radioactivity in the blood stream, which is thought to be as a result of enzymatic degradation of peptides leading to labelled peptide fragments that are unable to bind to the tumor but circulate in the bloodstream. The present experiment aimed to synthesize different peptide analogues to cancer targeting peptide 18, with increase in binding affinity to breast cancer as well as high enzymatic stability.

To fulfill this aim, two sets of analogues were synthesized. In the first set of a peptide analogues 18-1 - 18-4, it was found that hydrophobic substitutions were more likely to increase the binding affinity to different breast cancer cell lines relative to charged substitutions. In particular, replacement of amino acids at the labile sites with either D-amino acids or β3 amino acids greatly enhanced peptide stability, which would in turn increase binding affinity. In the second set of mixed α/β peptides, the heptad rule was followed because it generates a stripe of beta residues that runs along one side of the helix, which would maintain secondary structure stability. Different promising ligands have been generated using this rule.

Fluorescence microscopy as well as flow cytometry cellular binding studies demonstrated that the selected most potent analogues (18-4, 18-9, and 18-10) were highly specific for breast cancer cell lines. Among the panel of the studied cell lines, MCF7 was the highest in binding. Their interaction with cells was specific and likely mediated through a specific receptor. Furthemiore, evaluation of this selectivity showed that the present peptides have lower binding to control cell lines. The present Example demonstrates that the present peptides bind to breast cancer cells but minimally bind to primary endothelial HUVEC cells and MCF-10 cells. The binding and internalization of the peptides by normal cells was much lower within the tested incubation period. The uptake of the peptide was confirmed using confocal microscopy, and most of the peptide was internalized, which demonstrates the suitability of these peptides to deliver drugs or cytotoxic payloads to the inside of cells. Thus, the present Example demonstrates that peptides having three amino acids substitutions demonstrated the highest binding affinity, although beta amino acids increases the length of peptide backbone, but still it can structurally and functionally mimic their alpha counterpart 18 and interacts more efficiently with its putative receptor. Herein, a mixed α/β that manifests a favorable profile of properties, including high binding, and high specificity, selectivity and resistance to proteolytic cleavage was generated.

Enzymatic stability studies for the present selected peptides showed that they are more stable in human serum up to 24 h. Results with control peptides 18 showed that the control peptide 18 is completely degraded within 30 min. The insertion of β3 amino acid following the heptad pattern confer substantial resistance to proteolysis and each β-amino acid tends to protect nearby amide linkages from proteolysis, which is consistent with previous reports for isolated alpha to β insertions. Peptide with D- amino substitutions at cleavable sites showed comparable stability results.

Finally, cytotoxicity results of the selected peptides against MDA-MB-435 using MTT assay revealed that the peptides have no cytotoxicity effect up to 100 μΜ concentration, much higher concentration that the applied one for binding. It has been shown that peptides prepared solely from B3 amino acids may be non-toxic to cells.

In summary, the present Example demonstrates that D-amino acid substituted peptide 18 analogues and mixed α/β-peptide 18 analogues having 3β amino acid derived from L-Asp monomers substitutions increases substrate-target recognition, retains the proteolytic degradation resistance and reduces cytotoxicity. The present peptides have an increased degree of cell internalization, and thus may be useful, enzymatically stable lead peptides that can either be directly coupled to an anticancer drug or decorate a drug carrier that encapsulates the drug (e.g., liposomes, micelles, and polymeric nanoparticles), or can be conjugated with a diagnostic moiety such as a fluorophore, nonmetallic isotope, an optical reporter, a boron neutron absorber, a paramagnetic metal ion, a ferromagnetic metal, a gamma-emitting radioisotope, a positron-emitting radioisotope, or an x-ray absorber. Targeted therapy restricts the toxic effect of a drug to the malignant tissues, thereby increasing the efficacy and decreasing the undesired side effects of the drug. Experimental section

Materials

Chemicals and reagents All chemicals were commercial products of the best grade quality. Fmoc amino acids, Chlorotrityl resin, (2-(6-Chloro-lH-benzotriazole- l -yl)-l ,l ,3,3- tetramethylaminium hexafluorophosphate) (HCTU), 1 -hydro xybenzotriazole (HOBt), were purchased from NovaBiochem (San Diego, CA). The side chains of amino acids were protected as follows: t-butyl (tBu) for glutamic, and tyrosine, Oallyl (Oallyl) for aspatic, amine side chains used for β³ amino acid synthesis were as follows: isobutylamine (sigma) for isoleucine, benzylamine (sigma) for phenylanine, 1- napthalenemethylamine (Alfa Aesar) for tryptophan, methylamine(sigma) for alanine, t-Butyl-N-(4-aminobutyl) carbamate (TCI-EP) for arginine, 4-tert-butoxybenzylamine (otava) for tyrosine. Cyquant and FITC dyes were obtained from Invitrogen (Eugene, Oregon, USA). Piperidine, N,N diisopropyl ethylamine (D1PEA), N-methyl morpholine (NMM), trifluoroacetic acid (TFA) and all other reagents were purchased from Sigma-Aldrich. All solvents used in purification were HPLC grade.

Equipment

RP-HPLC purification and analysis were carried out on a Varian Prostar (210 USA) HPLC system using Vydac semi -preparative CI 8 (1 x 25 cm, 5 μηι), analytical C8 (0.46 x 25 cm, 5 μηι). Compounds were detected by UV absorption at 220 nm. Mass spectra were recorded on a MALDI Voyager time-of-flight (TOF) spectrometer (Voyager TM Elite), or on a Waters micromass ZQ. Absorbance of the purple formazan product observed during MTT assay was measured using VERSA max microplate reader (Molecular Devices, Sunnyvale, CA, USA). All the procedures regarding the cell culture maintenance and treatment of cells were carried out in a level II biosafety cabinet. Imaging experiments were done using Carl Zeiss microscope (Gottingen, Germany). FACS experiments were performed on Becton- Dickinson Facsort and analyzed by DakoCytomation Summit software. Peptide Synthesis

Non-labelled peptide analogues (18-1 - 18-10) were synthesized manually using solid phase peptide synthesis on 2-chlorotrityl-chloride resin (0.2 mmole). The first Fmoc-amino acid was coupled using DIPEA for 6 hours. Further amino acids were coupled at two-fold excess using (HCTU/HOBt/NMM) as activating mixture in Dimethylformamide (DMF). After 2 hours coupling time at room temperature, the ninhydrin test was performed to estimate the completeness of the reaction. In mixed α/β peptide analogues (18-5- 18-10), β³ amino acids were added to the backbone of the peptide following Fmoc/allyl combined solid phase synthesis. As follows, N- Fmoc L-Aspartic acid was activated using HCTU/HOBT (2equiv), and NMM (4.5 equiv.) then coupled to the growing peptide in DMF for 2 h at room temp. Deprotection of the allyl from carboxyl group was carried out with Pd (PP1 )4 (0.08 equiv) and PhSiFb (8 equiv) in DCM/DMF (45 mins x 3) under nitrogen, followed by the introduction of the corresponding amine using the same coupling reagents as mentioned above. Some amines required longer times and double coupling (eg. alanine, tyrosine, isoleucine side chains), concentration of these amine should not exceed 2 equivalent to prevent cleavage of Fmoc protecting group before complete coupling. Fmoc groups were removed by treatment with 20% piperidine in DMF two times each for 7 minutes. Peptide 18 and peptide pi 60 were synthesized using manual SPPS as described in Example 1.

Peptide FITC conjugation: After the peptide synthesis, β-alanine (spacer) was conjugated to the N-temiinal amino group followed by fluorescein isothiocyanate (FITC) coupling. FITC labelling of the peptides analogues and the control peptides were carried out on resin by mixing N-terminal unprotected peptides it (0.05mmole) with FITC (2 equiv, O.lmMole), and DIPEA (5 equiv, 0.25) in anhydrous DMF (4 ml), protected from light, followed by stirring at room temperature for 24 hours. The resin was drained and washed extensively with DMF, CH₂CL₂, and IPA. The crude peptide were cleaved from the solid support together with the removal of the side-chain protecting groups using a solution of 50:50% trifluroacetic acid/ dicloromethane (TFA/DCM) for 1 hour at room temperature. The obtained mixture underwent solvent evaporation followed by anhydrous ethyl ether precipitation to yield the final crude peptide. The precipitate was centrifuged, washed with ether (4 X 10). The crude product was purified to homogeneity by semipreperative HPLC and then freeze dried to give an orange powder of fluorescently labelled peptides. Peptides were purified on CI 8 semipreperative column, using a gradient 12-100% in 35 min IP A/water, flow rate 1.5 ml/min, and retention time was as shown in Figure 9. Peptides characterization was carried out using MALDI-TOF, which showed the expected molecular weights of the peptides. Purity of the peptides was analysed using analytical RP-HPLC, which showed purity of 95%.

Cell culture

All cancer cell lines and Human mammary epithelial cell line MCF10A were purchased from the American Type culture collection (ATCC) and additives from invitogen. Human breast cancer cell line MDA-MB-435 was cultured in RPMI 1640 media supplemented with 10% FCS, 100 IU/mL penicillin, and 100 IU/mL streptomycin. Human breast cancer cell lines MCF-7, and MDA-231 were cultured in DMEM media containing 10% FCS, 100 IU/mL penicillin, and 100 IU/mL streptomycin. Human mammary epithelial cell line (MCF10A) was cultured in minimal essential growth media MEGM (Lonza, cedarlane) supplemented with same additives as previously described. Human umbilical vein endothelial cells (HUVEC) from the laboratory of Sandra Davidge, University of Alberta, were cultivated using Endothelial Cell Growth Medium EGM, (Lonza, cederlane) containing 20% FCS, 2 mmol/L glutamine, 100 IU/mL penicillin, 100 IU/mL streptomycin, and 2 ng/niL basic fibroblast growth factor (Roche Diagnostics, Mannheim, Germany). All cell lines were cultivated at 37°C in a 5% C0₂-95% 0₂ incubator and growth media were replaced every 48 h.

Flow cytometry binding and internalization experiment

The binding of the synthesized analogues (18-1 - 18-10) was evaluated against three human breast cancer cell lines (MCF7, MDA-231 , MDA-MB-435) using flow cytometry. Cells were grown in T-75 culture flasks containing media supplemented with FBS and antibiotics 80% confluency. Cells was then washed twice with PBS and incubated with trypsin solution 37 C to detach the cells. Cells were centrifuged at 500 g for 5 min, re-suspended in media, counted by hemocytometer, and diluted to 10*7 ml media. Then they were seeded in 6 well tissue culture plate (3ml per well) at 37°C for 24 h. The following day, cells were washed by PBS and incubated in serum free media containing FITC-labelled peptides at concentration of 10^"5 Mole /litre for 30 min at 37°C. Then the cells were washed 3 times with PBS, trypsinized to remove any surface bound peptides, and centrifuged at 5000 G for 5 min. The pellet was resuspended in FACS solution (10% FBS in PBS), and the flow cytometry was performed. A total of 10 000 events were collected monitoring Floroscein. The autofluoresence of the cell only without treatment was measured to differentiate between the peptide bound labelled cells and autoflorescence of unlabeled cells. Fluorescence up to the measured intersect was called autofluoresence and represented cut-off point value. Cells in which fluorescence was higher than that value were considered labelled with FITC. FACS analysis was carried out by DakoCytomation Summit software. The selectivity of selected peptides was evaluated against 2 normal cell lines (MCF-IOA human breast cell line, and HUVEC human umbilical cord vein).

Fluorescence Microscopy

MDA-MB-43 cells or HUVEC cells (50,000) were cultured on the top of cover slip at 37°C for 24 hr. The medium was removed and replaced with 1 mL of fresh serum free medium, containing FITC-labelled peptides (18-4, 18-9) at a concentration of 10^"5 mol/L. The cells were incubated with the peptides for 30 min at 37 °C. After incubation, the medium was removed and the cells were washed thrice with 2 mL serum free medium. The cells were fixed on ice with 2% formaldehyde for 20 min. The formaldehyde was removed by washing with medium (three times). The cover slips were put on slides containing one 5 drop of DAPI-Antifade (Molecular Probes) to stain the nucleus. The cells were imaged under the fluorescence microscope (Zeiss) using green and blue filters with 20x magnification.

The samples prepared for fluorescence microscopy were also used for visualization by confocal microscopy to confirm internalization. Confocal laser scanning microscopy was performed with a Carl Zeiss inverted confocal microscope with a 40x oil immersion lens. Confocal stacks were processed using the Carl Zeiss LSM 5 Image software, which also operates the confocal microscope.

Serum stability

Peptides 18-4, 18-9, and 18-10 were tested for their serum stability and compared to their a counterpart peptide 18. Following the following procedure, 2 mg of peptide (mMole) were dissolved in 100% water, and then 100 μΐ of solution was added to 250 μΐ human serum, 650 μΐ RPMI media is added to mimic biological system in 1.5 ml eppendorf tube and temperature equilibrated at 37°C ±1 C for 15 min before adding the sample from the peptide stock solution. The initial time was recorded and at known time intervals, 1, 2, 3, 5 and 24 hours, ΙΟΟμΙ of reaction solution was removed and added to 200μ1 of methanol for precipitation of serum proteins present in human serum. The cloudy reaction produced is cooled to 4 °C for 15 min and then spun at 5000G for 15 min to pellet the peptide serum proteins. Then 50 μΐ from the supernatant was automatically injected in RP-HPLC on Vydac CI 8 column using autosampler to eliminate manual injection error, and the linear gradient from 12-100% IPA/water in 35 min, flow rate 1.5 ml/min and the absorbance was detected at 214 nm, concentration of peptides and degradable products was measured by integrating the area under the curve and the identity was confirmed using MALDI- TOF, correction for small interfering serum peaks that co-el ute with peptide was subtracted from background.

Cytotoxicity

The cytotoxicity of peptide analogues 18-4, and 18-9 were tested by measuring the cell growth inhibition using MTT assay. Breast cancer human cell line MDA-435 were seeded in 96 well plates (Corning Inc., MA, USA) at concentration IX 10⁴ cells /well per 200 μΐ RPMI media supplemented with 10% FBS and antibiotics (lOOU/ml penicillin, l OOug/ml streptomycin), and incubated at 37°Cin 5% C0₂ atmosphere. After 24 hours, the cells were treated with different concentrations of the peptides prepared in sterile water and incubated for 48 h. Doxorubicin was used as positive control. Untreated cells were used as negative control. Each plate was incubated for 48 hours. The culture media was discarded and replaced with 200μ1 MTT solution (5mg ml media), and cells were incubated for 3.5 hours. All experiments were done in triplicate, and the data in the form of mean is presented. Following incubation, the media was sucked out, and the purple formozan product precipitated in each well was solubilized in DMSO (150μί), after gentle shaking for 10 min at room temperature, absorbance was measured at 570 nm using microtitre reader with a reference wavelength of 650 nm to substrate back. The percentage cell viability was expressed as the absorbance ratio of cells treated with peptides to untreated cells dissolved in complete media.

The foregoing examples are provided by way of illustration and for purposes of clarity and understanding. It is readily apparent to those of ordinary skill in the art in light of the teachings that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims. Unless defined otherwise all technical and scientific terms used herein have the same meaning as commonly understood to one of ordinary skill and the art to which this invention belongs.

TABLE 1

Relative Cell Relative Cell

Peptide Amino Acid Adhesion Peptide Amino Acid Adhesion

# Sequence MB- MCF-7 # Sequence MB- MCF-7

435 435

1 (pi 60) VPW EPAYQRFL 1.0 1.0 38 WXEPAYQREL 1.2 2.1

2 VPWXEPAYQRFL 1.3 1.4 39 WXEPAYQRLL 1.3 1.4

3 VPWMEPAY 0.5 0.4 40 WXEPAYQRKL 2.7 2.7

4 VPWXEPAY 0.7 0.8 41 WXEPAYQRQ 1.4 1.5

5 GRGDS 1.1 1.1 42 WXEPAYQRYL 1.6 1.8

6 RGEPAYQRFL 1.7 1.5 43 WXEPAYQRF*L 0.7 1.7

7 RGDPAYQRFL 1.5 1.9 44 WXEPAYQRfL 1.1 1.9

8 WXEPAYQGRFL 1.0 1.6 45 WXEP YQRFE 1.6 1.5

9 WXEPAYNGRFL 1.0 1.8 46 WXEP YQRF1 1.7 2.0

10 RGEPAYQGRFL 1.7 1.9 47 WXEPAYQRFK 2.2 2.8

11 RGDPAYQGRFL 1.7 1.8 48 WXEPAYQRFQ 1.2 1.5

12 RGEPAYNGRFL 1.4 1.7 49 WXEPAYQRFT 0.8 1.5

13 RGDPAYNGRFL 1.0 2.1 50 WXEPAYQRFF 1.1 1.8

14 WXEPAYQRFL 1.0 1.7 51 XEPAYQRFL 1.1 1.2

15 AXEPAYQRFL 1.1 1.2 52 EEPAYQRFL 1.1 1.2

16 WAEPAYQRFL 0.9 1.3 53 LEPAYQRFL 1.2 1.2

17 WXAPAYQRFL 1.6 1.9 54 1EPAYQRFL 1.0 0.9

18 WXEAAYQRFL 2.2 2.7 55 KEPAYQRFL 0.9 1.6

19 WXEPAAQRFL 0.9 1.7 56 QEPAYQRYL 0.8 1.0

20 · WXEPAYARFL 1.1 1.8 57 TEPAYQRYL 0.8 1.0

21 WXEPAYQAFL 1.1 1.5 58 XEPAYQREL 1.0 1.1

22 WXEPAYQRAL 1.0 1.2 59 XEP YQRLL 0.9 1.0

23 WXEPAYQRFA 1.3 1.6 60 XEPAYQRKL 1.6 1.7

24 EXEPAYQRFL 1.6 1.1 61 XEPAYQRQL 0.2 0.95

25 LXEPAYQRFL 0.6 1.2 62 XEPAYQRYL 0.6 1.0

26 KXEPAYQRFL 1.2 1.9 63 XEP YQRF^L 0.9 1.4

27 QXEPAYQRFL 0.9 1.2 64 XEPAYQRfL 0.8 1.1

28 YXEPAYQRFL 1.5 1.8 65 XEPAYQRFE 0.9 1.4

29 FXEPAYQRFL 1.2 1.3 66 XEPAYQRF1 0.9 0.9

30 F_XEPAYQRFL 1.0 1.3 67 XEPAYQRFK 1.2 2.0

31 fXEPAYQRFL 0.7 1.8 68 XEPAYQRFQ 0.5 1.0

32 WEEPAYQRFL 1.5 2.1 69 XEPAYQRFT 0.6 0.7

33 WLEPAYQRFL 1.6 1.8 70 XEPAYQRFF 0.8 1.2

34 WIEPAYQRFL 1.4 1.5

35 KEPAYQRFL 1.6 1.8

36 WQEPAYQRFL 1.5 1.4

37 WTEPAYQRFL 0.7 1.6

The C-terminus of the peptides is covalently attached to the cellulose membrane. An amino acid replacement or insertion of an amino acid in the pi 60 sequence is highlighted in bold and underlined. X stands for norleucine, F* refers to 4-chlorophenylalanine, and lower case letter denotes D-amino acid. Relative cell adhesion is the average ratio of fluorescent intensity of a peptide divided by the fluorescence of peptide 1 (pi 60).

Claims

Claims

1. A method of determining peptide sequences which bind to cancer cells comprising:

(a) synthesizing a series of peptides on a membrane;

(b) spotting each peptide in a pre-determined pattern on the membrane such that the peptides are bound to the membrane;

(c) creating a duplicate membrane;

(d) passing healthy cells over the first membrane;

(e) passing cancer cells over the duplicate membrane;

(f) determining the binding of the healthy cells to the peptides;

(g) determining the binding of the cancer cells to the peptides;

(h) comparing the binding of the (f) and (g); and

(i) assessing which peptides display higher binding to cancer cells as compared to healthy cells.

2. The method of claim 1 wherein the cells which bind the peptides are fluorescently labeled following the binding, and the binding of the healthy cells and the cancer cells to the peptides is determined by a fluorescence-based assay.

3. The method of claim 1 further comprising:

(j) selecting the peptides that show higher binding to cancer cells;

(k) modifying the sequence of the selected peptides; and

(1) synthesizing peptides with a tag in the N-terminus

(m)repeating steps (f) to (i) as defined in claim 1.

4. The method of claim 1 , wherein the peptides identified as having higher binding to the cancer cells are used to target drugs to the cancer cells.

5. A method for the treatment or diagnosis of cancer comprising administering a peptide as identified by the method of claim 1 to a subject.

6. A peptide, or an analogue thereof, capable of binding cancer

7. The peptide, or an analogue thereof, as defined in claim 6, wherein the peptide or analogue thereof is capable of binding cancers cells with a higher affinity than healthy cells.

8. The use of a peptide, or analogue thereof, wherein the peptide having a higher binding affinity to the cancer cells is used to target drugs to the cancer cells.

9. The use of the peptide, or analogue thereof, defined in claim 8 for the treatment or diagnosis of cancer.

10. The use of a peptide, or analogue thereof, capable of binding cancer cells for the identi ication of receptors on cancer cells and/or for the diagnosis of cancer.

1 1. An assay device comprising an array of peptide molecules on a membrane, wherein the peptide molecules are capable of interacting with healthy and/or cancer cells, wherein the cells that are capable of interacting with peptides can be monitored by fluorescent labeling, and wherein the pattern of interaction between the peptide molecules and the cancer cells is indicative of the peptide's ability to be used as a marker for cancer.