Patent Application
20150218213
The present application includes a Sequence Listing filed in electronic format. The Sequence Listing is entitled “2923-1292_ST25.txt” created on Feb. 25, 2015, and is 28,000 bytes in size. The information in the electronic format of the Sequence Listing is part of the present application and is incorporated herein by reference in its entirety.
The present invention relates to inhibiting agents of a α6 integrin/E-cadherin molecular complex for use as a medicament, particularly for the prevention or/and treatment of metastases of a primary cancer disease, and a method of determining the prognosis of metastatic homing of a primary cancer disease.
Cancer known medically as a malignant neoplasm, is a term for a large group of different diseases, all involving unregulated cell growth. In cancer, cells divide and grow uncontrollably, forming malignant tumors, and invade nearby parts of the body. The cancer may also spread to more distant parts of the body through the lymphatic system or bloodstream. When the area of cancer cells at the originating site becomes clinically detectable, it is called a primary tumor. Some cancer cells acquire the ability to penetrate and infiltrate surrounding normal tissues in the local area, forming a new tumor. The newly formed tumor within the tissue is called a local metastasis. Some cancer cells acquire the ability to penetrate the walls of lymphatic and/or blood vessels, after which they are able to circulate through the blood stream (circulating tumor cells) to other sites and tissues in the body. This process is known as lymphatic and hematogenous spread, respectively.
After the tumor cells come to rest at another site, they repenetrate through the vessel or walls, continue to multiply and eventually another clinically detectable tumor is formed. This new tumor is known as a metastatic (or secondary) tumor. The impact of secondary tumors is often more fatal than that of the primary tumor.
Advanced colorectal cancer (CRC) is a leading cause of cancer-related mortality, particularly due to the spreading of malignant tumor cells to the liver, among other organs. Despite increased public awareness and screening colonoscopy, up to 25% of patients diagnosed with CRC already have hepatic metastasis. In a further 30-40% of patients suffering from CRC, metastases will develop later in the course of the disease, usually within two years from the resection of the primary tumor (Parkin et al. 2005, CA Cancer J Clin 55:74-108). Patients with operable liver-confined metastases may be cured by resection, but surgical cures are relatively rare in this setting (Tomlinson et al. 2007, J Clin Oncol 25:4575-4580). Most patients with metastatic disease are candidates for systemic chemotherapy to palliate symptoms and, potentially, downstage unresectable tumors to a resectable status (Meric et al. 2000, Ann Surg Oncol, 7:490-495; Adam et al. 2004, Ann Surg, 240:644-657). Without treatment, the median survival of patients with hepatic metastases is 6-8 months, and 5-year survival rates are lower than 5% (Wagner et al. 1984, Ann Surg, 199:502-508). The introduction of novel therapeutic agents (targeted bio-drugs such as Bevacizumab, Cetuximab and Panitumumab), in combination with cytotoxic drugs (i.e. oxaliplatin and irinotecan), has prolonged the median survival expectancy up to 24 months, albeit cure remains anecdotal (Hurwitz et al. 2004. N Engl J Med 350:2335-2342; Giantonio et al. 2007, J Clin Oncol 25:1539-1544; Hochster et al. 2008, J Clin Oncol 26:3523-3529; Saltz, 2008, Gastrointest Cancer Res, 2:S20-22; Kopetz et al. J Clin Oncol, 28:453-459; Jonker et al. 2007, N Engl J Med, 357:2040-2048; Hecht et al. 2009, J Clin Oncol, 27:672-680; Saltz et al. 2006, Nat Rev Drug Discov, 5:987-988; van Cutsem et al. 2007, J Clin Oncol, 25:1658-1664; Barugel et al. 2009, Expert Rev Anticancer Ther, 9:1829-1847). Therefore, hepatic metastasis clearly remains the central clinical challenge in the management of CRC.
A pivotal contribution to metastatic colonization comes from components of the host tissue and stroma. Therefore, targeting cancer microenvironments provides a promising strategy for the prevention or/and treatment of metastases.
It has long been recognized that several proteins integrate their stepwise action during the natural history of the progression and metastasis of human CRC (Fearon et al. 1990, Cell, 61:759-767; Vogelstein et al. 1988, N Engl J Med, 319:525-532). Insights into the molecular mechanisms underlying this disease have also begun to emerge through genomics and proteomics (Lin et al. 2007, Oncol Rep, 17:1541-1549; Zeitoun et al. 2008. Anticancer Res, 28:3609-3612; Nibbe et al. 2009, Mol Cell Proteomics, 8:827-845; Koh et al. 2008. Oncology, 75:92-101; Nannini et al. 2009, Cancer Treat Rev, 35:201-209). However, the fact that mRNA levels are not necessarily correlated with protein levels confers limitations on the significance of gene expression analyses (Nie et al. 2007. Crit Rev Biotechnol, 27:63-75). For example, comparative studies on metastatic prostate cancers revealed a concordance between protein and mRNA levels as low as 48-64% (Varambally et al. 2005. Cancer Cell, 8:393-406; Taylor et al. 2006. Cancer Res, 66:5537-5539). Alternatively, classical proteomic approaches are extremely time-consuming and expensive, which render their routine use difficult.
Angiopoietin-like 6 is a secreted factor whose mRNA has been detected particularly in the liver of humans. Although this protein shares a common structure with other members of the angiopoietin family, and particularly a coiled-coil domain in the N-terminal portion and a fibrinogen-like domain in the C-terminal portion, it does not bind to the Tie1 or Tie2 receptor and is currently considered an orphan ligand (Kim et al. 2000, Biochem J, 346 Pt 3:603-610; Oike et al. 2003, Proc Natl Acad Sci USA, 100:9494-9499; Oike et al. 2004, Blood, 103:3760-3765). Angiopoietin-like 6 regulates angiogenesis by preventing endothelial cell apoptosis, inducing endothelial cell migration and vascular leakiness and enhancing blood flow (Kim et al; Oike et al; Urano et al. 2008, Arterioscler Thromb Vasc Biol, 28:827-834). Some evidence suggests that RGD-binding integrins might be involved in angiopoietin-like 6-mediated cell adhesion, spreading and migration, although a direct interaction with integrins has not been described thus far (Zhang et al. 2006, Biochem Biophys Res Commun, 347:100-108).
Integrin α6, complexed with either β1 or β4 subunit, is a receptor for laminin with an emerging role in regulating angiogenesis and cancer progression through both direct and indirect mechanisms (Humphries et al. 2006, J Cell Sci, 119:3901-3903; Primo et al. 2010, Cancer Res, 70:5759-5769, Lee et al. 2006, J Biol Chem, 281:40450-40460; Gonzalez et al. 2002, Proc Natl Acad Sci USA, 99:16075-16080, Rabinovitz et al. 2001, Mol Biol Cell, 12:4030-4043; Robertson et al. 2008, Curr Pharm Des, 14:296-305). First, cellular delocalization of α6β4 integrin from hemidesmosomes to the edge of lamellipodia and filopodia has been related to a functional switch from adhesion to the extracellular matrix to migration at the invading front (Lipscomb et al. 2005. Cancer Metastasis Rev, 24:413-423). Second, physical interaction of α6β4 integrin with different tyrosine-kinase receptors has been shown to amplify pro-invasive signals (Bertotti et al. 2005. Cancer Res 65:10674-10679, Yoon et al. 2006, Cancer Res, 66:2732-2739). Third, both α6β1 and α6β4 integrins seem to be involved in CRC cell binding to hepatocytes as well as extravasation during the onset of metastasis, although a molecular mechanism for these functions remains to be elucidated (Enns et al. 2004. J Gastrointest Surg, 8:1049-1060; Robertson et al. 2009, Clin Exp Metastasis, 26:769-780).
E-cadherin is a well-described oncosuppressor protein, whose expression in the primary tumor counteracts cell detachment and is therefore associated with a better outcome (Christofori, 2003, Embo J, 22:2318-2323). Decreased production of E-cadherin, on the contrary, is one of the central events underlying epithelial-mesenchymal transition and carcinoma progression, in response to different cellular events such as the acquisition of loss-of-function mutations and loss-of-heterozygosis for the mutant allele, transcriptional or epigenetic repression and aberrant cellular localization (Ilyas et al. 1997, Gut, 40:654-659; Natalwala et al. 2008, World J Gastroenterol, 14:3792-3797; Kwak et al. 2007, Dis Colon Rectum, 50:1873-1880; Elzagheid et al. 2006, World J Gastroenterol, 12:4304-4309). Conversely, the role of E-cadherin in late stages of cancer progression needs further characterization. Remarkably, different reports show that mRNA and protein expression is regained in metastases, particularly in a subset of liver metastases from CRC and prostate carcinomas and increased levels of E-cadherin were found in metastatic tissue, particularly in liver metastases (Elzagheid et al.; Wells et al. 2008, Clin Exp Metastasis, 25:621-628, Truant et al., 2008, J Surg Res, 150:212-218).
It was found by the inventors that angiopoietin-like 6 acts as a ligand for cells that express a receptor complex of α6 integrin and E-cadherin. The interaction between the angiopoietin-like 6 and the α6 integrin/E-cadherin complex is found to have significant influence in metastasis homing and colonization. Experimental results show that inhibition of the α6 integrin/E-cadherin molecular complex may inhibit/reduce metastases on different levels.
Thus, a first aspect of the present invention refers to an inhibiting agent of the α6 integrin/E-cadherin molecular complex for use as a medicament, particularly for the prevention or/and treatment of metastasis of a primary cancer disease such as colorectal, bone, brain, breast, cervix, colon, gastric, liver, lung, pancreas, exocrine pancreas, duodenum, ovarian, renal, prostate, stomach, soft tissue, bone marrow, esophagus, skin cancer or lymphoma, particularly colorectal, stomach and lung cancer, more particularly colorectal cancer. The inhibiting agent may be used e.g. in human or veterinary medicine.
The α6 integrin/E-cadherin molecular complex is formed by direct and/or indirect molecular interaction between the full length α6 integrin protein (140 kD, SEQ ID No 16) or a proteolytic fragment thereof and the full length E-cadherin protein (120 kD, SEQ ID No 17) or a proteolytic fragment thereof.
Proteolytic fragments of α6 integrin protein preferably have a molecular weight of 10 to 130 kDa, preferably 20 to 120 kDa. More preferably, the proteolytic fragments include the fragments of aa24-1073, aa24-899, aa903-1073, aa24-594, aa595-899 and/or aa595-1073 of the full-length α6 integrin protein (SEQ ID No 16). Full length or proteolytic fragments of the α6 integrin are described by Pawar et al., Exp Cell Res, 2007, 313, 6, 1080-9; Demetriou et al., Open Cancer J., 2008, 2:1-4; Pawar et al.; Int J Radiat Biol., 2007, 83(11-12):761-7); Sroka et al., Carcinogenesis, 2006, 27(9):1748-57; Demetriou et al., Exp Cell Res, 2004, 294(2):550-8 and Davis et al., Cell Growth Differ., 2002, 13(3):107-13, which are herein incorporated by reference.
Proteolytic fragments of E-cadherin protein preferably have a molecular weight of 20 to 100 kDa, preferably 30 to 97 kDa, particularly 30 kDa, 40 kDa, 80 kDa or 97 kDa. More preferably, the proteolytic fragments of E-Cadherin include amino acids aa36-882 of the full-length sequence (SEQ ID No 17). Full length or proteolytic fragments of E-cadherin are described by Solanas et al. Nat Cell Biol. 2011; Céspedes et al. Am J Pathol, 2010, 177, 4, 2067-79; Lynch et al., J Oncol, 2010, 2010:53074-5; Elston et al., J Clin Endocrinol Metab, 2009, 94, 4, 1436-42; Huguenin et al. PLoS One, 2008, 3, 5, e2153; Najy et al. J Biol Chem, 2008, 283, 26, 18393-401; Ferber et al. J Biol Chem, 2008, 283, 19, 12691-700 and Lee et al. Eur Surg Res, 2007, 39, 4, 208-15 which are herewith incorporated by reference.
The term “direct molecular interaction” means a covalent bond or non-covalent interactions, such as electrostatic or van-der-Waals interactions or hydrogen bonds, particularly van-der-Waals interactions. The term “indirect molecular interactions” refers to domains and/or regions where the complex partners α6 integrin as well as E-cadherin are accumulated, i.e. where the concentration of both complex partners (α6 integrin+E-cadherin) is increased as compared to the average concentration of (α6 integrin+E-cadherin).
The α6 integrin/E-cadherin molecular complex is preferably expressed by a plurality of tumor cells, preferably by metastatic tumor cells, preferably metastatic tumor cells of primary colorectal, bone, brain, breast, cervix, colon, gastric, liver, lung, pancreas, exocrine pancreas, duodenum, ovarian, renal, prostate, stomach, soft tissue, bone marrow, esophagus, skin cancer or metastatic tumor cells of lymphoma. Particularly the α6 integrin/E-cadherin molecular complex is expressed by metastatic tumor cells of primary colorectal, stomach and lung cancer, more particularly primary colorectal cancer.
In a preferred embodiment, the molecular complex is expressed on the surface of metastatic tumor cells.
The inhibiting agent of the α6 integrin/E-cadherin molecular complex may be selected from inhibitors acting on the protein level or on the nucleic acid level.
In a preferred embodiment of the invention, the complex inhibitor acts on the protein level. In a preferred embodiment, the inhibitor binds to the α6 integrin/E-cadherin complex. By binding of the inhibitor to the complex, the activity of the complex, particularly its binding activity, may be altered, particularly reduced.
In one embodiment the inhibitor may be selected from an antibody, an antibody fragment or an antigen binding fragment specific for α6 integrin or/and E-cadherin or/and E-cadherin/α6 integrin complex, preferably for E-cadherin/α6 integrin complex, or an aptamer directed against E-cadherin or/and α6 integrin or/and E-cadherin/α6 integrin complex, preferably an aptamer directed against E-cadherin/α6 integrin complex, or a scaffold compound, which interacts and/or binds with α6 integrin or/and E-cadherin or/and E-cadherin/α6 integrin complex.
Preferably, the inhibitor is an antibody. The antibody may be selected from a polyclonal antibody, a monoclonal antibody, a chimeric antibody, a humanized antibody, a human antibody, a recombinant antibody or a fragment thereof, preferably Fab′ fragments, F(ab′)2 fragments or single-chain Fv fragments.
For the production of antibodies, a host animal, e.g. a mouse or rabbit, may be immunized with E-cadherin or/and α6 integrin or/and E-cadherin/α6 integrin antigen, optionally together with an adjuvant to increase the immunological response. A monoclonal antibody may be prepared by using known techniques, including but not limited to the hybridoma technique developed by Köhler and Millstein. Chimeric antibodies may be obtained from monoclonal antibodies by replacing non-human constant regions by appropriate human constant regions. Humanized antibodies may be obtained by replacing non-human framework regions in the variable antibody domains by appropriate human sequences. Human antibodies may be obtained from host animals, e.g. mice, comprising a xenogenic human immune system. Recombinant antibodies may be obtained by phage display and affinity maturation of given antibody sequences. Recombinant antibodies may be single-chain antibodies, bispecific antibodies etc.
Antibody fragments, which contain at least one binding site for E-cadherin or/and α6 integrin or/and α6 integrin/E-cadherin complex may be selected from Fab fragments, Fab′ fragments, F(ab′)2 fragments or single-chain Fv fragments. Aptamers directed against E-cadherin or/and α6 integrin or/and α6 integrin/E-cadherin complex may be obtained by affinity selection of nucleic acid and/or peptidic sequences according to known protocols.
In a further preferred embodiment, the inhibitor is a scaffold compound which interacts and/or binds with α6 integrin or/and E-cadherin or/and E-cadherin/α6 integrin complex. The scaffold compound may be selected from adnectins based on human fibronectin III, affibodies based on Z-domain of protein A, anticalins derived from lipocalins, atrimers based on tetranectin proteins, avimers or cystein-rich knotting peptides, DARPins based on ankyrin domains, Kringle domain derived from plasminogen, Kunitz domain derived from trypsin inhibitors.
WO 2008/064910 discloses peptides capable of selectively binding to metastatic cells having a sequence motif LRS and a length of 6 to 100 amino acids. The peptides, if labeled, can be used for the detection of hepatic metastases already in pre-clinical stages. The authors further suggest conjugating the peptides with chemotherapeutic drugs for target therapy. WO 2008/064910, however, does not give any hint to use these peptides alone, i.e. in non-conjugated form, as a medicament. It has now surprisingly been found that such peptides—without conjugated drugs or diagnostic agents—effectively inhibit the E-cadherin/α6 integrin complex.
Thus, in a preferred embodiment of the invention, the inhibiting agent of the E-cadherin/α6 integrin complex is a peptide having the sequence motif LRS and a length of 6 to 100, preferably to 70, more preferably to 40, most preferably to 35, amino acids. In a preferred embodiment, such peptides are not conjugated, e.g. chemically or physically, to other active agents, such as drugs or diagnostic agents, which are preferably different from the inhibitors according to the invention.
The term “peptide” includes amino acid sequences constituted by at least one of the 20 common amino acids that can be found in natural proteins, modified, e.g. non genetically encoded, amino acids, amino acid mimetics known in the art or unusual amino acids such as Aad, 2-Aminoadipic acid; EtAsn, N-Ethylasparagine; Baad, 3-Aminoadipic acid, Hyl, Hydroxylysine; Bala, beta-alanine, beta-Amino-propionic acid; AHyI, allo-Hydroxylysine; Abu, 2-Aminobutyric acid; 3Hyp, 3-Hydroxyproline; 4Abu, 4-Aminobutyric acid, piperidinic acid; 4Hyp, 4-Hydroxyproline; Acp, 6-Aminocaproic acid, Ide, Isodesmosine; Ahe, 2-Aminoheptanoic acid; Alle, allo-Isoleucine; Aib, 2-Aminoisobutyric acid; MeGly, N-Methylglycine, sarcosine; Baib, 3-Aminoisobutyric acid; MeIle, N-Methylisoleucine; Apm, 2-Aminopimelic acid; MeLys, 6-N-Methyllysine; Dbu, 2,4-Diaminobutyric acid; 5 MeVal, N-Methylvaline; Des, Desmosine; Nva, Norvaline; Dpm, 2,2′-Diaminopimelic acid; Nle, Norleucine; Dpr, 2,3-Diaminopropionic acid; Orn, Ornithine; and EtGly, N-Ethylglycine. Also included are amino acids in D-configuration.
In particular embodiments, the amino acid sequence may include one or more non-amino acids. In particular embodiments, the sequence of a peptide of the present invention may be interrupted by one or more non-amino acids.
The peptides of the present invention may be linear or cyclic peptides, preferably linear.
In a preferred embodiment of the present invention, peptides inhibiting the E-cadherin/α6 integrin complex comprise an amino acid sequence selected from the group consisting of ARPGLRS (SEQ ID NO. 1), MRYALRS (SEQ ID NO. 2), LRPGLRS (SEQ ID NO. 3), LRSGSGS (SEQ ID NO. 4), GIYRLRS (SEQ ID NO. 5), GVYSLRS (SEQ ID NO. 6), LRSGRGS (SEQ ID NO. 7), RREGLRS (SEQ ID NO. 8), SWYTLRS (SEQ ID NO. 9), LAYRLRS (SEQ ID NO. 10), LTYRLRS (SEQ ID NO. 11), VRPGLRS (SEQ ID NO. 12), LRSGRGS (SEQ ID NO. 13), preferably GIYRLRS (SEQ ID NO. 5) and GVYSLRS (SEQ ID NO. 6).
In a preferred embodiment, the inhibiting agent is a peptide comprising the amino acid sequence CGIYRLRSC (SEQ ID NO. 14) and CGVYSLRSC (SEQ ID NO. 15).
Due to their relatively small size, the peptides according to the invention can be synthesized in solution or on solid supports, according to well known techniques. Short peptides, generally from about 6 to 35-40 amino acids, can be easily produced with these techniques. Alternatively, recombinant cDNA technology can be used, in which a nucleotidic sequence coding for a peptide of the invention is inserted in an expression vector, transformed or transfected in proper host cells, and cultured in conditions suitable for protein expression.
In another embodiment the inhibitor of E-cadherin/α6 integrin complex acts on the nucleic acid level, e.g. by inhibiting E-cadherin or/and α6 integrin or/and E-cadherin/α6 integrin complex transcription and/or translation.
The inhibitor of E-cadherin/α6 integrin complex nucleic acid may be an α6 integrin or/and an E-cadherin gene expression inhibitor, preferably selected from nucleic acid effector molecules directed against E-cadherin or/and α6 integrin mRNA, such as RNAi molecules or precursors or templates thereof, antisense molecules or ribozymes.
RNAi molecules are RNA molecules or RNA analogues capable of mediating an interference of a target mRNA molecule. RNAi molecules may be siRNA molecules (short-interfering RNA molecules), which are short, double-stranded RNA molecules with a length of preferably 18-30 nucleotides and optionally at least one 3′ overhang. Further RNAi molecules may be shRNA molecules (short hairpin RNA molecules) having a length of e.g. 14-50 nucleotides. Optionally, the RNAi molecules may comprise ribonucleotide analogues in order to enhance the stability against degradation. The invention also encompasses precursors of RNAi molecules, i.e. RNA molecules which are processed by cellular mechanisms into active RNAi molecules. Further, the invention encompasses DNA templates of RNAi molecules or precursors thereof, wherein the templates are operatively linked to an expression control sequence. The RNAi molecules have sufficient complementarity to the α6 integrin or/and E-cadherin mRNA to allow specific degradation thereof, thereby inhibiting α6 integrin or/and E-cadherin expression.
In a preferred embodiment, the siRNA molecule has a sense strand selected from
In a further embodiment, the nucleic acid inhibitor molecule may be an antisense molecule, i.e. an antisense RNA, DNA or nucleic acid analogue molecule, which blocks translation of α6 integrin or/and E-cadherin mRNA by binding thereto and preventing translation. Antisense molecules may be single-stranded and preferably have a length of 14-30 nucleotides. Antisense molecules directed against the translation initiation site of E-cadherin or/and α6 integrin mRNA are preferred.
In a further embodiment, the E-cadherin or/and α6 integrin nucleic acid inhibitor may be a ribozyme. Ribozymes are enzymatic RNA molecules which catalyze specific cleavage of RNA, e.g. hammerhead ribozymes.
The inhibiting agent of the present invention is used as a medicament, particularly as a medicament for the prevention or/and treatment of metastases of a primary cancer disease. The primary cancer disease may preferably selected from the group consisting of colorectal, bone, brain, breast, cervix, colon, gastric, liver, lung, pancreas, exocrine pancreas, duodenum, ovarian, renal, prostate, stomach, soft tissue, bone marrow, esophagus or skin cancer or lymphoma, particularly colorectal, stomach or lung cancer, preferably colorectal cancer.
In a preferred embodiment, the inhibiting agent is used to prevent or/and reduce metastases in liver tissue, breast tissue, lung tissue, lymph nodes, bone tissue or brain tissue, preferably in liver tissue.
In a particularly preferred embodiment of the invention, the inhibiting agent is used for the prevention or/and treatment of metastases deriving from primary colorectal cancer in liver tissue.
In other words, the inhibiting agent may be used to prevent or/and reduce secondary cancer, particularly in liver tissue, breast tissue, lung tissue, lymph nodes, brain tissue or bone tissue, preferably in liver tissue.
In another aspect, the inhibiting agent of the invention may be used in combination with another (other than the inhibiting agent) anti-cancer or/and anti-viral therapy, preferably anti-cancer therapy. The anti-cancer therapy may be selected from chemotherapy, radiation therapy, surgical intervention, immunotherapy, gene therapy, target therapy or combinations thereof.
The inhibiting agent is preferably used in combination with at least another additional chemotherapeutic or/and antiviral agent. The chemotherapeutic agent may be selected from antimetabolites, DNA-fragmenting agents, DNA-crosslinking agents, intercalating agents, protein synthesis inhibitors, Topoisomerase 1 and 2 inhibitors, microtubule-directed agents, kinase inhibitors, hormones and hormone antagonists, anti-tumor antibodies, or any combination thereof. Preferably, the anti-cancer agent is selected from platinum compounds (oxaliplatinum), fluoropyrimidines (inhibitors of the thymidylate synthetase, such as capecitabine and its derivative 5-fluorouracil), alkaloids (inhibitors of the topoisomerase I, such as campthotecin and its derivative irinotecan).
The anti-viral agent may be selected from a protease inhibitor, a polymerase inhibitor, an integrase inhibitor, an entry inhibitor, an assembly secretion inhibitor, a translation inhibitor, an immunostimulant or any combination thereof.
Preferably, the inhibiting agent may be co-administered with at least another chemotherapeutic or/and anti-viral agent. In another embodiment, the inhibiting agent and the chemotherapeutic or/and anti-viral agent may be administered separately.
A further aspect of the invention is a pharmaceutical composition or kit which comprises as an active agent at least one inhibiting agent of α6 integrin or/and E-cadherin or/and α6 integrin/E-cadherin complex as described above, together with a pharmaceutically acceptable carrier, diluent and/or adjuvant. The pharmaceutical composition is preferably for use in medicine, e.g. in human or veterinary medicine.
The term “pharmaceutically acceptable carrier” preferably includes sterile water, buffers or isotonic saline.
The term “diluent and adjuvant” preferably includes solvents such as ethanol, antioxidants and/or preservatives.
The pharmaceutical composition may be formulated e.g. as tablets, pills, capsules, liquids, sirups, slurries, suspensions, injectable solutions etc. Depending on the specific disorder to be treated, the composition may be administered systemically or locally. Suitable routes may e.g. include oral, rectal, transmucosal, intestinal, intranasal, intraocular or pulmonal administration or parenteral delivery, including intramuscular, subcutaneous, intrathecal, intravenous or intraperitoneal injection or infusion.
The pharmaceutical composition comprises the active agent in an effective dose, sufficient to achieve its intended purpose. Determination of an effective dose can be carried out by the skilled person. For example, the effective dose may be estimated from cell culture assays and animal models. Usual dosages for administration in human medicine may range from e.g. 0.01 to 2000 mg/day, commonly from 0.1 to 1000 mg/day and typically from 1 to 500 mg/day.
The pharmaceutical composition according to the present invention may further comprise at least one other active agent, such as an anti-cancer, e.g. a chemotherapeutic agent or/and an anti-viral agent. The anti-cancer agent may or/and the anti-viral agent may be as defined above.
Another aspect of the invention is directed to a method of screening for an inhibiting agent for the α6 integrin/E-cadherin molecular complex, comprising the steps of:
(i) incubating a α6 integrin/E-cadherin molecular complex with an agent under conditions suitable to induce binding of the agent to the complex,
(ii) detecting the binding of the agent to the complex,
(iii) comparing the result obtained in step (ii) with a predetermined binding score, and
(iv) evaluating the agent to be an inhibiting agent for α6 integrin/E-cadherin molecular complex.
In the method according to the invention, α6 integrin/E-cadherin molecular complex and cells expressing the α6 integrin/E-cadherin molecular complex, respectively, are incubated with a candidate agent. Incubation preferably takes place at 2-10° C., preferably 4-6° C., in phosphate buffer saline or in cell culture Hepes-buffered medium (such as Iscove's Modified Dulbecco's Minimal Essential Medium) for 0.5-4 hours, preferably 1.5-2.5 hours. The binding of the candidate agent to the α6 integrin/E-cadherin molecular complex is detected via phage displayed peptide binding assay, radio- or dye-labelled ligand binding or/and surface plasmon resonance assay, preferably by phage displayed peptide binding assay. These methods are known in the art and are described, e.g. by Maynard et al., Biotechnol J., 2009, 4(11):1542-58; Hulme et al., Br J Pharmacol., 2010, 161(6):1219-37 and Marchiò et al., Cancer Cell., 2004, 5(2):151-62, which are herein incorporated by reference.
The results obtained are compared to the extent of binding of a known agent to the molecular complex. The predetermined binding score is a quantitative parameter of the binding of a known substance (standard) to the complex. Preferably, the predetermined binding score is selected such that any agent which has the same or a higher binding score than that of the standard can be regarded as an inhibitor.
Preferred standard substances are e.g. laminin 332 or E-cadherin.
In another aspect, the present invention provides a method for determining the prognosis of metastatic homing of a primary cancer disease, in particular the aggressiveness of the metastatic potential of a primary cancer disease, comprising the steps of:
(i) providing a sample of a patient suffering or suspicious to suffer from metastasis of a primary cancer disease,
(ii) determining the expression or/and amount of the α6 integrin/E-cadherin molecular complex or/and the amount of angiopoietin like 6 in the sample, and
(iii) optionally classifying the results obtained in step (ii) in predetermined disease states.
In a preferred embodiment, a sample, e.g. a blood sample, tissue sample or lymph liquid from a patient suffering from metastases of, e.g. colorectal, bone, brain, breast, cervix, colon, gastric, liver, lung, pancreas, ovarian, renal, pancreas, prostate, stomach, soft tissue, bone marrow or skin primary cancer or lymphoma primary cancer, particularly colorectal primary cancer, is provided. The sample may be a blood sample, tissue sample or lymph liquid. Particularly, the sample may be a blood sample or a tissue sample of the organs affected by the primary cancer, e.g. a colorectum sample, or/and a tissue sample of the organ suspicious to suffer from a secondary cancer organ, e.g. liver tissue, breast tissue, lung tissue or lymph liquid, preferably liver tissue.
Determination of the expression or/and amount of the α6 integrin/E-cadherin molecular complex or/and angiopoietin-like 6 protein, is carried out by conventional assays as known in the art, e.g. immunofluorescence staining.
High amounts or upregulated expression of α6 integrin/E-cadherin molecular complex or/and angiopoietin-like 6 are usually associated with advanced metastasis homing and shorter disease-free survival.
The amounts or the expression of α6 integrin/E-cadherin molecular complex or/and angiopoietin like 6 may be classified in predetermined disease states.
(
(
(
(
The location of α6 integrin and E-cadherin in paraffin-embedded liver metastases from CRCs was evaluated as described in
The amounts and localization of angiopoietin-like 6 in livers from healthy donors (n=17) (
For protein quantification, 50 μg of total lysate was loaded on an 8% SDS-polyacrylamide gel and proteins resolved by electrophoresis were blotted onto a PVDF membrane. Membranes were stained with the following primary antibodies: mouse monoclonal anti-β4 integrin clone 7, goat polyclonal anti-α6 integrin N-19, mouse monoclonal anti-β1 integrin clone P4G11, mouse monoclonal anti-E-cadherin clone 36, mouse monoclonal anti-angiopoietin-like 6 clone Kairos-60, goat polyclonal anti-vinculin N-19. Vinculin was used as a loading control and as a normalizer for the densitometric quantification of band intensity (illustrated in the graphs). (
Paraffin-embedded normal tissue samples from healthy donors were cut in 5 μm sections and were stained with the rabbit polyclonal anti-angiopoietin-like 6 antibody. (
(
Total mRNA from CRC cell lines (
(
(
(
U293 cells stably overexpressing E-cadherin, α6 integrin or both were grown in complete (10% FCS) culture medium in the presence of either the control or CGIYRLRSC (SEQ ID NO: 14) peptide. At 24 hour time points, cells were fixed and stained with crystal violet; their numbers were estimated by spectrophotometric evaluation. Results are shown as mean±standard deviation for each experimental point in 2 independent experiments. ANOVA analysis of the data revealed no statistical significance.
HCT-116m, SW-48, DLD-1 and HT-29 cell lines with a decreased expression of either α6 integrin or E-cadherin (see
To obtain an in vivo model of metastatic CRC, we implanted the patient-derived HCCM-1544 tumor as well as different CRC cell lines (HCT-116m, SW-48, DLD-1 and LS-174T) intrasplenically into CD-1 nude mice (2×106/mouse). To evaluate the effect of CGIYRLRSC (SEQ ID NO: 14) on liver metastasis, we injected cells either in medium alone (vehicle) or in the presence of the soluble peptide (CGIYRLRSC) (SEQ ID NO: 14). At the indicated time points, mice were euthanized and their livers and spleens were explanted. Livers were photographed for the quantification of external metastatic areas. Representative pictures of whole livers from 2 mice/group are shown for macroscopic evaluation of tumor morphology; the indicated p-values are referred to statistical analysis performed either with Fisher's exact test (in black) or t-test (in red). Spleen (primary tumor) and liver (metastasis) samples were OCT-frozen, cut in 10-μm sections and immunostained with anti-α6 integrin and anti-E-cadherin antibodies, followed by imaging with a confocal microscope. As described in
Human cell lines derived both from primary CRCs (HCT-116 and its derivative HCT-116m, SW-48, HT-29, DLD-1, LS-174T) and from a liver metastasis secondary to CRC (NCI-H630) were grown in complete (10% FCS) or serum-deprived (0.5% FCS) culture medium for 48 hours, followed by staining with anti-α6 integrin and anti-E-cadherin antibodies. Results of the cytofluorimetric analyses (fluorescence intensity and percent of positive cells) are shown as mean±standard deviation for each experimental point in 3 independent experiments. Differences in the experimental points were evaluated for their statistical significance by the use of ANOVA followed by Bonferroni's post-test.
The location of α6 integrin and E-cadherin was evaluated on 5 μm sections of paraffin-embedded primary CRCs by staining with specific antibodies followed by confocal microscope imaging. For the quantification of the fluorescent signals (illustrated in
The location of α6 integrin and E-cadherin in paraffin-embedded liver metastases from different primary tumors was evaluated as described in
The location of α6 integrin and E-cadherin in paraffin-embedded lung metastases from CRCs was evaluated as described in
The presence of α6 integrin and E-cadherin was evaluated by immunostaining of 5-μm sections of a large panel of paraffin-embedded cancer tissues (
(
The amounts of angiopoietin-like 6 were evaluated by staining of 5 μm tissue sections of primary CRC and liver metastasis from the same patients (n=22). Pictures of 20 samples for each tissue type are shown.
Antibodies, Recombinant Proteins, and Synthetic Peptides.
Goat polyclonal anti-α6 integrin N-19 (used for immunoblot) (sc-6597) and anti-vinculin N-19 (sc-7649), rabbit polyclonal anti-β4 integrin H-101 (used for ELISA) (sc-9090), and horseradish peroxidase (HRP)-conjugated donkey anti-goat IgG (sc-2033) were from Santa Cruz Biotechnology (Santa Cruz, Calif.). Mouse monoclonal anti-α6 integrin clone BQ16 (used for immunoprecipitation) was from Calbiochem (San Diego, Calif.). Rat monoclonal anti-α6 integrin clone GoH3 (used for immunostaining) was from AbD Serotec (Raleigh, N.C.). Mouse monoclonal anti-β4 integrin clone 7 (used for immunoblot) and anti-E-cadherin clone 36 were from BD Transduction Laboratories (Franklin Lakes, N.J.). Mouse monoclonal anti-β1 integrin clone P4G11 was from Chemicon (Millipore, Billerica, Mass.). Rabbit polyclonal anti-fd bacteriophage (B-7786) was from Sigma (St. Louis, Mo.). Alexa Fluor 488 anti-rat IgG and 555 anti-mouse IgG were from Invitrogen (Carlsbad, Calif.). HRP-conjugated donkey anti-mouse IgG was from Jackson ImmunoResearch (West Grove, Pa.). Mouse monoclonal anti-CD31 clone JC70A and HRP-conjugated anti-rabbit EnVision were from DAKO (Glostrup, Denmark). Rabbit polyclonal (used for immunostaining) and mouse monoclonal clone Kairos-60 (used for immunoblot) anti-angiopoietin-like 6, and recombinant angiopoietin-like 6 were from Alexis Biochemicals (Enzo Life Sciences, Farmingdale, N.Y.). Rabbit polyclonal anti-PRL3 (62) was a gift of Dr. Alberto Bardelli (Institute for Cancer Research and Treatment, Candiolo, Italy). Laminin (L-2020) was from Sigma. Targeting (CGIYRLRSC) (SEQ ID NO: 14) and control (CARAC) (SEQ ID NO: 43) peptides were from New England Peptides (Gardner, Mass.).
Cell Lines and Human Samples.
SW620 (ATCC CCL-227), NCI-H630 (ATCC CRL-5833), HepG2 (ATCC HB-8065), NCI-N87 (ATCC CRL-5822), A549 (ATCC CCL-185), HCT-116 (ATCC CCL-247), HT-29 (ATCC HTB-38), DLD-1 (ATCC CCL-221), SW-48 (ATCC CCL-231), LS-174T (ATCC CL-188), and U293 (ATCC CRL-1573) cell lines were from LGC-Promochem (Sesto San Giovanni, Italy), and were cultured according to the purchaser's instructions. The HCCM-1544 human metastatic CRC and corresponding in vitro and in vivo manipulations have been previously described (Tibbetts et al. 1993. Cancer, 71:315-321). A variant of HCT-116, selected in vivo for its ability to metastasize to the liver in pseudo-orthotopic models (here named HCT-116m), was provided by Dr. Alberto Bardelli. Fresh (grossly normal livers from CRC patients, primary CRCs, liver metastases secondary to CRC) and paraffin-embedded (grossly normal livers from CRC patients, primary CRCs, liver metastases of various origins) human specimens were collected by the Units of Surgical Oncology and of Pathology at the Institute for Cancer Research and Treatment. Paraffin-embedded human specimens of normal liver from healthy donors, of lung metastasis secondary to CRC, and of different healthy tissues were collected by the Unit of Pathology at the Molinette Hospital (Turin, Italy). Snap-frozen samples of lung metastasis secondary to CRC and liver metastasis secondary to renal cancer were from San Luigi Gonzaga Hospital (Orbassano, Italy); snap-frozen samples of ovarian cancer and matched metastasis were from Mario Negri Institute (Milan, Italy). Collection and manipulation of human samples were approved by the Institutes' Ethical Committees, and written informed consent was obtained from all patients in accordance with the Declaration of Helsinki.
Biopanning of Human Samples with Phage Display.
Fresh tissue samples were dissected with a scalpel and digested with 0.025% collagenase A (Roche Diagnostics, Monza, Italy) in Iscove's Modified Dulbecco Minimum Essential Medium (IMDM) for 2 hours at 37° C. with vigorous shaking. The resulting suspension was passed through a 40 μm nylon cell strainer (BD Labware, Franklin Lakes, N.J.), and cells were resuspended in binding medium (IMDM supplemented with 2% Fetal Calf Serum, FCS). With this protocol, we did not select for tumor epithelial cells; instead, we aimed at retaining the original mixed cell population (consisting mostly of epithelial, endothelial, and hematopoietic cells, and fibroblasts), as well as the tissue stroma. To preserve their phenotype, we avoided cell culture and maintained these cells in suspension in binding medium at 4° C. for the duration of the experiments. 1010 transducing units (TU) of a CX7C, CX9C, or CX3CX3CX3C phage library was added to 5×105 liver metastasis cells in binding medium and cells were incubated overnight (first round). For successive rounds, phage was first pre-adsorbed on normal liver cells for 1 hour at 4° C. and was subsequently incubated with liver metastasis cells for 2 hours at 4° C. Cells were washed 5 times with binding medium, and bound phages were recovered and amplified by infection of K91 Kan Escherichia coli bacteria in log-phase. Purification of phage particles and DNA sequencing of phage-displayed inserts were performed as described (Scott et al. 1990. Science, 249:386-390; Smith et al. 1993. Methods Enzymol 217:228-257).
Bioinformatics.
The protein BLAST tool (www.ncbi.nlm.nih.gov/BLAST) was used to investigate similarities between MTS peptides and the human proteome. Due to the short length of such peptide sequences, consequent high numbers of false positive similarities would lead to result misinterpretations. The following constraints were introduced to our search: (i) at least 3 consequent amino acids of the MTS peptide motif should be identical to that of the BLAST match; (ii) each output protein should share 3 similarity matches (i.e. should be similar to at least 3 MTS peptides or to an MTS peptide in at least 3 different regions). Peptides in the reverse orientation are expected to have significantly lower probability of mimicking natural proteins; therefore, the MTS_REV dataset was used as a further internal control for the BLAST searches. GO annotations were retrieved through the DAVID Bioinformatics Resource Functional Annotation tool (david.abcc.ncifcrf.gov; Huang et al. 2009. Nat Protoc 4:44-57; Huang et al. 2009. Nucleic Acids Res 37:1-13), searching for GOTERM C_C_ALL, with default program settings. GO-CC-enriched categories were associated with corresponding statistical significance, represented as a Benjamini-Hochberg-corrected p-value.
Isolation and identification of MTS-targeted proteins. The following oligonucleotides were annealed and inserted into pGEX-4T.1 between BamHI and NotI sites to create the pGEX-4T.1-CGIYRLRSC (SEQ ID NO: 14) plasmid:
The corresponding fusion peptide was purified to homogeneity from BL-21 Escherichia coli cell lysates by affinity chromatography on glutathione-sepharose beads (GE Healthcare), according to the manufacturer's protocol. HepG2 and NCI-H630 cells were lysed in 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.1% NP-40, 10% glycerol, and a protease inhibitor cocktail (SIGMA-Aldrich). 10 milligrams of total protein was pre-cleared on GST-Sepharose (GE Healthcare) prior to incubation with CGIYRLRSC-GST-Sepharose (4 μg peptide/mg total protein) overnight at 4° C. Bound proteins were eluted from Sepharose beads, separated on a 10% SDS-polyacrylamide gel, and stained with BioSafe Coomassie blue (BioRad). Specific bands were analyzed by mass spectrometry as described (Paget, 1989. Lancet, 1:571-573). Matrix-assisted laser desorption/ionization (MALDI) mass spectra were recorded on an Applied Biosystems Voyager DE-PRO mass spectrometer equipped with a reflectron time-of-flight (TOF) analyzer and used in delayed extraction mode (Applied Biosystems, Foster City, Calif.). Raw data, reported as monoisotopic masses, were introduced into the MASCOT peptide mass fingerprinting search program (Matrix Science, Boston, Mass.) for protein identification. Liquid chromatography (LC)-mass spectrometry (MS)/MS analyses were performed on a CHIP MS Ion Trap XCT Ultra equipped with a 1100 high pressure liquid chromatography (HPLC) system and a chip cube (Agilent Technologies, Palo Alto, Calif.). Peptide analysis was performed by data-dependent acquisition of one MS scan (mass range from 400 to 2000 m/z) followed by MS/MS scans of the three most abundant ions in each MS scan. Raw data from nanoLC-MS/MS analyses were introduced into the MASCOT software to search the human proteome.
Validation of MTS-Phage Targets.
Binding of single phage clones on whole cells was performed with a 109 TU input of each phage on 5×105 suspended cells in binding medium as described (Chambers et al. 2002. Nat Rev Cancer, 2:563-572). For overlay binding experiments, 5×109 TU/ml of each phage was incubated with 10 μm tissue sections of OCT-frozen tissues and detected as described, with the EnVision system (DAKO) and 3-amino-9-ethylcarbazole (AEC) as substrate (Arap et al 2002, Nat Med, 8:121-127; Padua et al. 2008, Cell, 133:66-77). Phage overlay images were acquired with an EC3 Leica camera (Leica Microsystems, Milan, Italy).
For the characterization of proteins targeted by CGIYRLRSC (SEQ ID NO: 14), 5 mg of synthetic peptide was immobilized on column-packed diaminodipropylamine-agarose (CarboxyLink Kit, Pierce, Euroclone, Milan, Italy) according to the manufacturer's protocol. After equilibration in PBS, the column was loaded with 10 mg of total protein from 7 pooled human samples of liver metastasis secondary to CRC, allowing recirculation for 30 minutes at 4° C. The column was washed with 10 ml of column buffer (PBS, 1 mM CaCl2, 1 mM MgCl2, 50 mM β-octyl-D-glucosylpyranoside, 1 mM PMSF and protease inhibitor cocktail), followed by salt elution of unspecific proteins in column buffer supplemented with 50 mM NaCl. Control and target protein elution was obtained with 2 mM of the control and CGIYRLRSC (SEQ ID NO: 14) peptide, respectively, and the column was finally cleared with 0.1 M NaCl, 0.1 M Glycine pH 3.00. Protein amounts in collected fractions were followed by reading their OD at 280 nm, and selected fractions were concentrated by the use of Microcon centrifugal filter devices with cut-off 10,000 (Millipore) to remove residual synthetic peptides. Proteins were quantified with the Coomassie (Bradford) Protein Assay Kit (Pierce), and 500 ng of each sample was evaluated for the presence of specific integrin subunits with a standard ELISA assay. In parallel, the relative amount of targeted proteins was assessed by phage binding as described (66), on 2 μg of each sample and with an input of 109 TU of fd-tet or CGIYRLRSC- (SEQ ID NO: 14) phage. Binding to BSA-coated microwells was used for normalization.
Preparation of Cells Stably Overexpressing the MTS-Targeted Proteins.
The cDNA of FLAG-tagged human angiopoietin-like 6, inserted into pcDNA3.1(+). Neo vector (pcDNA3.ANGL6), was a gift from Dr. Y. Oike (Japan Science and Technology Agency, Japan (Minn et al., 2007, Proc Natl Acad Sci USA, 104:6740-6745); the cDNA of human E-cadherin, cloned into a pcDNA3.1(+). Neo vector (pcDNA3.CAD1) was a gift from Dr. C. Gottardi (North Western University Medical School, Chicago, Ill. (Bos, et al. 2009, Nature 459:1005-1009); the cDNA of 134 integrin, cloned into a PRK5 plasmid (pRK5.ITB4), was purchased from Addgene (Cambridge, Mass.); the cDNA of human α6A integrin, inserted in pLXSN plasmid (pLα6SN (Kuo et al. 1995, Proc Natl Acad Sci USA, 92:12085-12089), was a gift of Dr. A. Magrelli (La Sapienza University, Rome, Italy). The latter was PCR-amplified with the following primer pair:
and subcloned into a pcDNA3.1(+). Hygro vector between HindIII and ApaI sites to obtain the pcDNA3.ITA6A plasmid. To stably overexpress of α6β4 integrin, E-cadherin, and angiopoietin-like 6, U293 cells were transduced with the described plasmids by the use of a calcium phosphate transfection kit (Invitrogen), followed by selection of single cell clones in culture media supplemented by geneticin (500 μg/ml) and/or hygromicin (200 μg/ml).
Preparation of Cell Lines with Silenced Expression of the MTS-Targeted Genes.
A transient gene silencing approach was applied to cells used exclusively for short-term, in vitro experiments. NCI-H630 cells were transduced with ON-TARGETplus SmartPOOL siRNA for ITGA6 and CDH1, or with control siRNA (Dharmacon, Lafayette, Colo.). For each experimental point, 2×105 cells were reverse-transfected with either control, ITGA6, CDH1, or both siRNA pools, according to the manufacturer's protocol. To quantify the downmodulation of targeted genes, RNA and protein levels were evaluated after 24 and 72 hours, respectively. A stable gene silencing approach was preferred for cells used for long-term, in vivo experiments. For this purpose, 2×105 HCT-116m, SW-48, HT-29, or DLD-1 cells were transfected with shRNA plasmid pools targeting ITGA6 (sc-43129-SH) or CDH1 (sc-35242-SH), or with non-targeting control shRNA plasmid pool A (sc-108060) (all from Santa Cruz Biotechnologies), according to the manufacturer's protocol. Following selection in medium supplemented with 2.5 μg/ml puromycin, 6 clones for each experimental point were subjected to dotblot analysis to confirm selective protein down-regulation. For this purpose, cell lysates (1 μg each) were spotted onto PVDF membranes; after drying, membranes were subjected to specific antibody staining with standard procedures.
Retrotranscription, Real-Time PCR and End-Point PCR.
RNA was retrotranscribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems) and was amplified with the Power SYBR Green PCR Master Mix (Applied Biosystems). For quantification of residual transcripts in silenced cells, the following primer pairs were used for real-time PCR amplification of the cDNAs in an ABI PRISM 7700 instrument:
To evaluate the presence of different splicing forms of α6 integrin, cDNAs from cell lines and from biopsies of human liver metastases secondary to CRC were subjected to end-point PCR amplification with the following primer pair:
Immunostaining.
OCT-frozen tissues were cut in 10 μm sections, paraffin-embedded tissues in 5 μm sections. For immunostaining of cell lines, 104 cells were plated on a SuperFrost Plus glass slide (Menzel-Glaser, Braunschweig, Germany) and were grown for 24 hours followed by fixation in 4% paraformaldehyde in PBS for 5 minutes at room temperature. Immunostaining was performed according to standard protocols. Fluorescent images were acquired with either a DMIRE2 confocal microscope or with a DMI 3000D microscope equipped with a DFC 360FX digital camera (all from Leica. Visible images were acquired with either an EC3 Leica (immunostaining of frozen tissues) or a High-Performance IEEE 1394 FireWire Digital CCD Camera (QIMAGING, Surrey, BC, Canada) (immunostaining of paraffin-embedded tissues). For the quantification of specific fluorescent signals, 3 confocal images (1024×1024 pixels, equivalent to 375×375 μm) for each sample, acquired by keeping all the parameters constant, were divided in two 8-bit images corresponding to the red and green fluorescence channels. These image pairs were analyzed with the Image Processing and Analysis software in Java (ImageJ, version 1.44h, rsbweb.nih.gov/ij/, using the Colocalization Highlighter plugin to create a binary representation of co-localized pixels, and the Image Calculator option to derive the non-co-localized pixels (Dong et al. 1994, J Natl Cancer Inst 86:913-920).
Cytofluorimetric analyses were performed with the use of the Cytofix/Cytoperm™ Kit (BD Transduction Laboratories), following the manufacturer's protocol.
Immunoblot and Immunoprecipitation.
Cells were lysed in 4 pack cell volumes of phosphate buffered saline (PBS), 1 mM CaCl2, 1 mM MgCl2, 1 mM phenylmethylsulfonyl fluoride (PMSF), protease inhibitor cocktail, supplemented with either 50 mM β-octyl-D-glucosylpyranoside (interaction studies) or 0.1% NP-40 (expression studies). Tissues were homogenized in a Potter-Elvehjem grinder in the same buffer (˜1 ml/100 mg tissue). Homogenates were cleared by centrifugation followed by filtration through 0.45 μm pore filters. For immunoprecipitation, lysates were pre-cleared for 1 hour at 4° C. on Protein G-Sepharose (GE Healthcare, Chalfont St. Giles, UK), followed by incubation with specific antibodies for 1 hour at 4° C. and addition of Protein G-Sepharose for another 2 hours at 4° C. Proteins were separated on 10% SDS-polyacrylamide gels and were blotted onto polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, Mass.). For protein quantification, densitometric analysis of the detected bands was performed with the QuantityOne software (BioRad, Hercules, Calif.); values were normalized to the intensity of vinculin at each experimental point.
Adhesion, Proliferation, and Migration Assays.
All the described in vitro tests were performed at least in triplicate. For investigation of cell adhesion to peptides and proteins, 1 μg of each substrate was incubated per well of a 96 well-plate for 1 hour at 37° C. After a blocking step in IMDM, 2% FCS for 1 hour at 37° C., 104 cells were allowed to adhere for 1 hour at 37° C. Samples were washed gently in PBS, and cells were fixed in 8% glutaraldehyde and stained in 0.25% crystal violet in 10% methanol. For tissue adhesion, OCT-frozen grossly normal liver samples were cut into 10 μm sections. Tissues were blocked in IMDM, 2% FCS for 30 minutes at 37° C., followed by incubation with 5×104 cells in 5% CO2 at 37° C., for the indicated periods of time. Samples were washed 4 times in the same medium and once in PBS, fixed in 4% para-formaldehyde, and stained with hematoxylin (BioOptica). Adhered cells were counted manually under a light microscope.
To test the effects of peptides on cell proliferation, we seeded 2×104 cells per well in a 24-well plate, in the presence of either control or CGIYRLRSC (SEQ ID NO: 14) peptide (100 μM). At the indicated time points, cells were fixed in glutaraldehyde, stained in crystal violet, and solubilized in 10% acetic acid. Cell growth was evaluated by absorbance at 590 nm in a microplate reader (Perkin Elmerm Waltham, Mass.). A calibration curve was set up with known numbers of cells and a linear correlation between absorbance and cell counts was established up to 5×105 cells/well.
Cell migration was evaluated in 24-well plates by the use of PET track-etched membrane transwells (8 μm pores) previously equilibrated in PBS for 1 hour at 37° C. To produce a gradient of angiopoietin-like 6, 5×104 ligand-producing U293 cells were seeded into the lower chamber in 800 μl of complete culture medium. After 24 hours, transwells were inserted, control (mock-transfected) or receptor-expressing U293 cells were seeded into the upper chamber (1×105 cells in 200 μl of complete culture medium), and the plates were incubated at 37° C. in 5% CO2 for 48 hours. Transwells were subsequently removed, cells on their upper side were scraped off, and migrated cells were fixed in glutaraldehyde and stained with crystal violet (BioOptica). Migrated cells were counted manually under a light microscope.
Animal Models of Human Metastatic CRC.
Six-week female CD1-nude mice were purchased from Charles River (Lecco, Italy). Animals were subjected to intraperitoneal anaesthesia with a mixture composed by 0.75 mg/ml xylazine (Xilor®, BI098, Milan, Italy), 1 mg/ml tiletamine-1 mg/ml zolazepam (Zoletil®, Virbac, Milan, Italy), in physiological solution. After the mice were deeply asleep, a midline incision was performed and target organs were gently exposed. Two or five million suspended cells were injected in 50 μl of culture medium intrasplenically (Tibbetts et al. 1993. Cancer, 71:315-321) or into the median liver lobe (Grothey, et al. 2009. Nat Rev Clin Oncol, 6:507-518), respectively. To investigate a pharmacological intervention on liver homing and/or colonization of CRC cells, animals were divided in two arms, receiving medium alone (vehicle) or supplemented with 100 μM CGIYRLRSC (SEQ ID NO: 14) peptide. The wound was closed by a double suture and each animal was given 0.1 mg caprofen (Rymadil®, Pfizer, Milan, Italy) in a physiological solution to allow post-operative pain relief and rehydration. Mice were strictly monitored until completely awake, and oral ampicillin was administered for 5 days following the surgery. Mice were euthanized at the indicated time points, and organs were photographed with a PL-200 digital photocamera (Samsung Electronics, Milan, Italy). External metastatic areas were quantified using ImageJ software.
Statistical Analyses.
All the analyses were performed with Prism 5 software (GraphPad, La Jolla, Calif.): two-way analysis of variance (ANOVA) followed by Bonferroni's post-test was used to evaluate differences within treatments; t-test and Fisher's exact test (two-tailed) were used to compare selected experimental points; Chi-squared test was used to analyze contingency tables; survival curves were drawn as Kaplan-Meier Cumulative Proportion Surviving graphs and p-values were calculated by the use of the log-rank (Mantel-Cox) test. In all the graphs, unless differently specified, asterisks indicate the following p-value ranges: *=p<0.05, **=p<0.01, =p<0.001.
A protocol for the isolation of heterogeneous cell populations by tissue fractionation of human liver metastases immediately after surgical removal was designed and cells extracted from matching adjacent, grossly normal livers as negative controls were used. Three phage-displayed peptide libraries with the general arrangements CX7C, CX9C and CX3CX3CX3C (C=Cys and X=any residue) were screened on a panel of tumor/non-tumor paired samples in several independent biopanning experiments. In 13 out of 22 experiments (59%), enrichment of phage populations binding to liver metastases in comparison to the corresponding normal liver controls was observed (see
We therefore extracted from the cognate MTS_FW dataset all the genes coding for extracellular proteins and termed this sub-dataset “extracellular protein signature” of liver metastases secondary to CRCs (Table 1).
drosophila)
The enrichment for LRS sequences among the MTS peptide population was suggestive of a role as a relevant ligand motif within the microenvironment of particularly the liver metastasis. For an initial molecular analysis, we selected a panel of LRS-containing peptides (n=7): CLRSGRGSC (SEQ ID NO: 38), CLRPGLRSC (SEQ ID NO: 39), CGIYRLRSC (SEQ ID NO: 14), CMRYALRSC (SEQ ID NO: 40), CARPGLRSC (SEQ ID NO: 41), CLRSGSGSC (SEQ ID NO: 42) and CGVYSLRSC (SEQ ID NO: 15). We first evaluated binding of the cognate phage clones to a panel of human cell lines from different primary tumors and metastases (n=3) (
We produced soluble CGIYRLRSC (SEQ ID NO: 14) as a fusion peptide with Glutathione S-Transferase (CGIYRLRSC-GST), to exploit the interaction with NCI-H630 cell surfaces toward the identification of potential receptor(s). HepG2 cells, which do not bind the CGIYRLRSC- (SEQ ID NO: 14) phage specifically (
The location of α6 integrin and E-cadherin in NCI-H630 and HepG2 cells was evaluated by confocal microscopy imaging (
These results show that α6 integrin and E-cadherin are expressed and coincident in regions of i.a. human liver metastases, where they participate in a molecular complex.
A BLAST search specific for proteins similar to the closely-related CGIYRLRSC (SEQ ID NO: 14) and CGVYSLRSC (SEQ ID NO: 15) peptides revealed a set of extracellular proteins (n=54), 4 of which were listed in the extracellular protein signature as well (Table 1), i.e. angiopoietin-like 6, perlecan, laminin α2 and nyctalopin. In this analysis, angiopoietin-like 6 received the highest identification score, because it shares similarity with the targeting peptides in two different regions of its fibrinogen-like domain. Interestingly, angiopoietin-like 6 mRNA has been detected particularly in the liver in humans (Kim et al. 2000, Biochem J 346 Pt 3:603-610; Oike et al. 2003, Proc Natl Acad Sci USA 100:9494-9499). To investigate whether angiopoietin-like 6 could actually be a ligand for the hepatic homing of metastatic CRC cells, we evaluated the presence of this protein in several tissue types from healthy donors (
By confocal microscopy imaging we observed that angiopoietin-like 6 accumulates in hepatic blood vessels of metastatic CRC patients (
With the intention of designing a molecularly-defined cell model capable of reproducing these metastasis/host interactions, the nature of the integrin component in the receptor complex were dissected. The α6 integrin (i) can form heterodimers with either β1 or β4, depending on the cell type (Humphries et al. 2006 J Cell Sci 119:3901-3903; Hemler et al. 1988, J Biol Chem 264:6529-6535; Hemler et al. 1988, J Biol Chem 263:7660-7665) and (ii) is encoded by two splicing variants of a single mRNA, resulting in proteins with an alternative cytoplasmic tail, namely α6A and α6B (Tamura et al. 1991, Proc Natl Acad Sci USA 88:10183-10187; Hogervorst et al. 1991, Eur J Biochem 199:425-433; Hogervorst, et al. 1993, J Cell Biol 121:179-191). Lysates of human liver metastases were affinity-purified with column-immobilized CGIYRLRSC- (SEQ ID NO: 14) peptide, observing a substantial enrichment in the β4 subunit in peptide- and phage-targeted protein fractions in comparison with the controls (
These results demonstrate that angiopoietin-like 6 could represent a potential, thus far unrecognized ligand for metastatic cells that express the α6 integrin/E-cadherin receptor complex.
To evaluate a potential role for the α6 integrin/E-cadherin complex in the process of hepatic metastasis, the capacity of HepG2 and NCI-H630 cells was compared to adhere to normal livers. Cells were incubated on human liver sections for increasing periods of time, ranging from 30 minutes to 5 days. At all timepoints, HepG2 cells adhered weakly to normal liver and grew in separate, individual aggregates. In contrast, the adhesion of NCI-H630 cells was significantly higher; moreover, these cells could grow and integrate into the hepatic tissue (
For a molecular dissection of the ligand/receptor system, the capacity of silenced NCI-H630 cells to bind the CGIYRLRSC (SEQ ID NO: 14) motif and to interact with angiopoietin-like 6 was evaluated. NCI-H630 cells in which ITGA6 or both mRNAs were silenced lost the capacity to bind the CGIYRLRSC- (SEQ ID NO: 14) phage. Consistently, these cells also exhibited an impaired adherence to microwells coated with the CGIYRLRSC- (SEQ ID NO: 14) peptide; this effect was particularly pronounced when both α6 integrin and E-cadherin were downmodulated. In parallel assays, NCI-H630 cells in which both mRNAs were silenced exhibited significantly lower binding to microwells coated with recombinant angiopoietin-like 6 (
These results indicate that CGIYRLRSC- (SEQ ID NO: 14) mimicked ligand proteins, such as angiopoietin-like 6, can act as microenvironment addresses for metastatic cells that express α6 integrin/E-cadherin receptor complex.
Besides being accumulated in liver blood vessels, therefore contributing to the recognition of metastatic cells by the host tissue (
These results demonstrate that the presence of both α6 integrin and E-cadherin is necessary for target cells to respond to a chemotactic gradient of angiopoietin-like 6 protein, further confirming a functional role for this ligand/receptor system.
Since the CGIYRLRSC- (SEQ ID NO: 14) peptide mimics a liver-enriched ligand for the α6 integrin/E-cadherin receptor complex, it was investigated whether it could interfere with this interaction, thus inhibiting the adhesion of metastatic cells to normal liver. It was observed that when NCI-H630 cells were incubated on human liver sections from several patients (n=27), a significant decrease in cell adhesion occurred after treatment with CGIYRLRSC (SEQ ID NO: 14) (
These data show that CGIYRLRSC (SEQ ID NO: 14) specifically inhibits the adhesion of metastatic CRC cells to the liver, possibly through interference with the angiopoietin-like 6/α6 integrin/E-cadherin ligand/receptor system.
In sum, the in vitro data indicated that two pivotal steps for the onset of liver metastasis, i.e. tumor/host tissue recognition and metastatic cell attraction, could be driven by the 06 integrin/Ecadherin/angiopoietin-like 6 system. Accordingly, the interference with the described ligand/receptor pair was investigated to result in impaired liver colonization and homing in vivo. An animal model of hepatic colonization was established by direct injection of human CRC cells into the livers of CD-1 nude mice. For a first set of experiments LS-174T, a cell line derived from a primary CRC that exhibits high expression of the complex proteins (
These results suggest that the CGIYRLRSC- (SEQ ID NO: 14) peptide interferes with early steps of tumor/host tissue recognition, while not influencing successive tumor growth. Consistently, we observed that soluble CGIYRLRSC (SEQ ID NO: 14) has no effects on the proliferation of U293 cells transduced with either α6 integrin, Ecadherin or both (
These data demonstrate that even an incomplete depletion of only one receptor protein is sufficient to alter liver colonization by CRC cells in small animal models.
Also, the interference of CGIYRLRSC- (SEQ ID NO: 14) peptide with the homing of CRC cells to the liver was investigated. For this purpose, the human metastatic CRC tumor HCCM-1544 (Tibbetts et al. 1993, Cancer 71:315-321), as well as different CRC cell lines, i.e. HCT-116m, SW-48, DLD-1 and LS-174T, were implanted into the spleens of CD-1 nude mice. Cells were injected either in medium alone or in the presence of the soluble peptide. Mice were euthanized at different time points after cell injection, ranging from 20 days (HCT-116m) to 195 days (DLD-1). At the time of sacrifice, in all the tumor models a primary splenic mass and a variable number of liver metastases were present. The frequency of hepatic involvement in the vehicle arms varied from 11% (DLD-1) to 100% (LS-174T), reflecting the different aggressiveness of these CRC cells. Treatment with CGIYRLRSC (SEQ ID NO: 14) resulted in a diminished homing of CRC cells to the liver, which was significant in all the models investigated, with the exclusion of the poorly metastatic DLD-1 cell line (
Remarkably, in all the experimental settings including the HCCM-1544 human tumor, a substantial increase in the presence of both α6 integrin and E-cadherin was observed, with diffuse regions of co-localization, in metastatic hepatic tissue compared to primary spleen tumors (
The results obtained both in vitro and in animal models prompted to investigate the role of the described ligand/receptor system in a clinical context. By quantitative confocal imaging, the amounts of α6 integrin, E-cadherin and their complex were evaluated in the following tumoral settings: primary CRCs (Duke's stage IV) (n=22) (
A correlation analysis with the clinical outcome of CRC patients revealed that high levels of α6 integrin, E-cadherin and their complex in liver metastases were all associated with shorter disease-free survival (
These results indicate that α6 integrin, E-cadherin and angiopoietin-like 6 are correlated with the progression of metastasis and could therefore act as prognostic markers and/or specific molecular targets for the development of anti-metastatic approaches to be translated into the clinics.
| Number | Date | Country | Kind |
|---|---|---|---|
| 12182994.9 | Sep 2012 | EP | regional |
| 12198505.5 | Dec 2012 | EP | regional |
This application is a 35 U.S.C. 371 National Phase Entry Application from PCT/EP2013/068150, filed Sep. 3, 2013, which claims the benefit of European Patent Application Nos. 12182994.9 filed Sep. 4, 2012 and 12198505.5 filed on Dec. 2, 2012, the disclosures of which are incorporated herein in their entirety by reference.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2013/068150 | 9/3/2013 | WO | 00 |
