CROSS-LINKING POLYPEPTIDE THAT INDUCES APOPTOSIS
20190085090 ยท 2019-03-21
Inventors
Cpc classification
C07K2317/32
CHEMISTRY; METALLURGY
C07K2317/60
CHEMISTRY; METALLURGY
C07K2317/24
CHEMISTRY; METALLURGY
C07K2317/73
CHEMISTRY; METALLURGY
C07K2317/34
CHEMISTRY; METALLURGY
C07K2317/62
CHEMISTRY; METALLURGY
International classification
Abstract
Described is a polypeptide comprising at least four domains specifically binding to a certain MHC peptide complex, the domains separated by linker amino acid sequences, thereby providing each domain with the capability to bind a separate MHC peptide complex, to a nucleic acid molecule encoding such a polypeptide, to a vector comprising such a nucleic acid molecule, to a host cell for expression of such a polypeptide, to a pharmaceutical composition comprising such a polypeptide, and to a kit of parts comprising at least two polypeptides of the disclosure.
Claims
1.-27. (canceled)
28. A single polypeptide chain comprising: at least four specific binding domains, the specific binding domains separated by linker amino acid sequences, wherein each specific binding domains comprises an immunoglobulin fragment.
29. The single polypeptide chain of claim 28, further comprising: a peptide that is not a linker and does not comprise an immunoglobulin fragment.
30. The single polypeptide chain of claim 28, having six specific binding domains.
31. The single polypeptide chain of claim 28, wherein the at least one specific binding domain is a V.sub.H.
32. The single polypeptide chain of claim 28, wherein the specific binding domains are able to bind to an MHC-I-peptide complex.
33. The single polypeptide chain of claim 32, wherein the MHC-I-peptide complex comprises a peptide derived from a tumor related antigen.
34. The single polypeptide chain of claim 31, wherein at least one of the specific binding domains comprises SEQ ID NO:11 or SEQ ID NO:12.
35. The single polypeptide chain of claim 31, wherein at least one of the linker amino acid sequences comprises SEQ ID NO:18, SEQ ID NO:19, SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO:23, or SEQ ID NO:24.
36. The single polypeptide chain of claim 31, wherein the specific binding domains are able to bind to the MHC-peptide complex, but not to the peptide itself, another MHC-peptide complex, or an empty MHC.
37. A method for producing the single polypeptide chain of claim 28, the method comprising: culturing a host cell comprising a polynucleotide encoding the polypeptide, allowing for expression of the polynucleotide, and harvesting the polypeptide.
38. A pharmaceutical composition comprising: the single polypeptide chain of claim 28, and a suitable diluent and/or excipient.
39. The pharmaceutical composition according to claim 38, further comprising a cytostatic and/or tumoricidal agent.
40. A conjugate of the single polypeptide chain of claim 28, and a cytostatic or tumoricidal agent.
41. A single polypeptide chain consisting of: six specific binding domains, the specific binding domains separated by linker amino acid sequences, wherein each specific binding domains comprises an immunoglobulin fragment; and a peptide that is not a linker and does not comprise an immunoglobulin fragment.
42. The single polypeptide chain of claim 41, wherein the specific binding domains are each V.sub.H.
43. A single polypeptide chain consisting of: six specific binding domains, the specific binding domains separated by linker amino acid sequences, wherein each specific binding domains comprises a V.sub.H; and a peptide that is not a linker and does not comprise a V.sub.H.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
[0041]
[0042]
[0043]
[0044]
[0045]
[0046]
[0047]
[0048]
[0049]
[0050]
[0051]
[0052]
[0053]
[0054]
DETAILED DESCRIPTION
[0055] As outlined in previous international patent application publication WO2007/073147, the desired specific and selective killing of aberrant cells via the apoptosis machinery can be achieved by contacting these cells with a multivalent mono-specific protein complex comprising multiple antigen-specific MHC-restricted single chain T-cell receptors (TCRs) and/or MHC-restricted antigen-specific antibodies, which antigen is expressed by the targeted aberrant cells and presented in the context of MHC molecules. This finding then, opened the possibility to selectively kill a population of cells that are positive for a certain MHC-peptide complex of interest, for example, tumor cells expressing HLA class I molecules in complex with peptides derived from tumor-associated antigens.
[0056] Without wishing to be bound by theory, and based on in the disclosure in this application, it is thought that a multivalent like, for example, a hexavalent mono-specific protein induces apoptosis via the clustering of a number of (identical) MHC-p complexes on the cell surface of a target cell. The data shown in the previous application WO2007/073147 suggest that clustering of three MHC-p complexes may not be sufficient for apoptosis induction, whereas a hexavalent complex is very efficient in inducing apoptosis. Thus, it is disclosed now that apoptosis induction requires the binding of at least four, preferably at least five, more preferably at least six MHC-p complexes by one multivalent single-chain protein.
[0057] The terms protein and polypeptide have roughly the same meaning throughout the text of this application and refer to a linear proteinaceous sequence comprising two or more linked amino acid residues. In the context of the proteins and protein complexes that specifically bind to MHC-p complexes, binding molecules and polypeptides have the same meaning as protein and protein complexes. The term apoptosis refers to the process of programmed cell death.
[0058] In one embodiment, a multivalent single-chain protein encompasses four, five, six, seven, eight, nine, ten, eleven or twelve domains or clusters of domains, each domain or cluster of domains capable of recognizing and binding to a specific MHC-peptide complex. In contrast to the known methods for apoptosis induction using anti-MHC antibodies, a multivalent single-chain monomeric protein, disclosed herein, can induce apoptosis itself and does not require any external post-translational cross-linking. The multiple domains or multiple clusters of domains are connected to form a linear sequence at the DNA level and thus connected into a linear single-chain monomeric polypeptide via regular peptide bonds at the protein level.
[0059] Disclosed is a multivalent single-chain protein comprising at least four and preferably six domains or clusters of domains capable of recognizing and binding to a specific MHC-peptide complex. At least four or preferably six domains or clusters of domains preferably recognize the same MHC-peptide complex, i.e., the preferred multivalent single-chain protein is mono-specific with respect to the MHC-p complex. The domains of the multivalent single-chain protein that specifically recognize and bind to a MHC-p complex can be TCR domains or a functional fragment thereof (together herein referred to as TCRs) and/or an antibody that mimics TCR specificity, for example, a genetically engineered antibody, such as a single-chain variable fragment (scFv) or the variable domain V of the heavy chain H of an antibody (referred to throughout the text as VH, Vh or V.sub.H). Also, a multivalent single-chain protein hereof may encompass TCR domains as well as MHC class-restricted antibody domains, provided that both types of domains recognize essentially the same MHC-peptide antigen. In the specification, MHC-peptide complex and MHC-peptide antigen have the same meaning. In the context of a peptide that is presented by an MHC molecule, forming an MHC-p complex, the terms peptide, peptidic antigen, antigenic epitope and antigenic peptide refer to the same peptide in the complex.
[0060] Multivalent TCR domain complexes and therapeutic applications thereof are known in the art. In international patent application publication WO2004/050705, a multivalent TCR domain complex comprising at least two TCR domains, linked by a non-proteinaceous polymer chain or a linker sequence composed of amino acid residues, is disclosed. The disclosed use of the TCR complex is in targeting cell delivery of therapeutic agents, such as cytotoxic drugs, which can be attached to the TCR complex. Di-, tri- and tetravalent TCR complexes are disclosed but divalent TCR complexes are preferred. Importantly, complexes of more than four TCRs are not described. Furthermore, WO2004/050705 focuses solely on the use of a multivalent TCR complex for the delivery of a therapeutic agent, e.g., a toxic moiety for cell killing, to a target cell. It does not teach or suggest the apoptosis-inducing capacity of a multivalent TCR complex itself. The antigen-specific MHC-restricted binding capacity of a multivalent monomeric single-chain protein hereof is sufficient to induce apoptosis of a target cell expressing the relevant antigen. Therefore, using the sole protein hereof only is sufficient for obtaining the desired effect. In, for example, application WO2004/050705, the additive use of an additional or attached cytotoxic agent or toxic moiety is, for example, required.
[0061] In the previous application WO2007/073147, it is disclosed that separate individual polypeptide monomers that together build up a multivalent complex of that application, be it antigen-specific MHC-restricted TCRs, TCR-like antibodies or combinations thereof, are post-translationally linked or connected to each other in any suitable manner, be it covalently or non-covalently using standard polypeptide linkage chemistry, in order to achieve the desired pro-apoptotic activity.
[0062] Any proteinaceous domain or cluster of domains capable of specifically recognizing and binding to an MHC-peptide complex, comprising either MHC class I or MHC class II proteins, is suitably used in a multivalent apoptosis-inducing single-chain protein. In one embodiment, this protein, comprises at least four, for example, six or even more domains or clusters of domains, connected through regular peptide bonds between the peptide backbone of the domains or clusters of domains building up the multivalent polypeptide, comprising amino acid sequences corresponding to the V.sub.H domains of human antibodies.
[0063] Exemplified is the generation of a hexavalent mono-specific single-chain monomeric protein, which is specific for a tumor antigen. This hexavalent single-chain protein has therapeutic value in the treatment of cancer. Moreover, the skilled person will appreciate that the disclosure is not limited to any type of antigen, and that hexavalent single-chain proteins are provided that can selectively kill target cells, like, for example, selected aberrant cells, expressing any antigen.
[0064] Preferably, the polypeptide is capable of specifically and efficiently recognizing and binding to a cancer-specific epitope or an epitope associated with autoimmune disorders or an epitope presented by any other aberrant cell, for all examples in the context of MHC. Cancer cells may express a group of antigens termed cancer testis antigens (CT). These CT are presented as antigenic peptides by MHC molecules (as MHC-p complexes) to CTLs. In fact, these CT are immunogenic in cancer patients as they may elicit anti-cancer responses. They exhibit highly tissue-restricted expression, and are considered promising target molecules for cancer vaccines and other immune intervention strategies.
[0065] To date, more than 44 CT gene families have been identified and their expression studied in numerous cancer types. For example, bladder cancer, non-small lung cancer, prostate cancer, melanoma and multiple myeloma express CT genes to a high level. Experiments have shown that expression of these CT genes was indeed testis restricted in healthy individuals. Other antigens that were shown to elicit immune responses in cancer patients include differentiation antigens, such as, for example, the melanoma antigens gp100, Mart-i, Tyrosinase, or antigens that are over-expressed in cancer cells, such as, for example, p53, Her-2/neu, WT-1. Both groups of antigens are not specific for these aberrant cells and are also expressed in healthy tissue, and may therefore elicit autoimmune disease when targeted. In a preferred embodiment, the hexavalent single-chain protein is capable of recognizing and binding to an MHC class I- or to an MHC class II-tumor antigen complex, in particular melanoma associated antigens (MAGE), specifically at tumor cells, leaving healthy cells and tissue essentially unaltered, NB: testis do not present antigens in the context of HLA. The antigen is, for example, a peptide from a member of the CT gene families. The antigen can also be selected from the series of tumor antigens and/or from the series of antigens expressed in the tissue or organ affected by cancer cells, for which it is known that their expression is not tumor specific or not specific for the tissue or organ bearing cancer cells, as is known, for example, for gp100, Mart-1, Tyrosinase, p53, Her-2/neu, WT-1. These antigens are selected as a therapeutic target when the risk for adverse effects is acceptable when related to the beneficial outcome of the treatment with hexavalent single-chain protein, which targets the antigenic peptide complexed with MHC. The general benefit of the disclosure is that, where up until now targets associated with cell surfaces were the predominant goal, intracellular targets now become available through presentation by MHC-1 and/or MHC-2. This means that a renewed survey of intracellular antigens will be carried out to identify intracellular antigens that are tumor specific enough to merit using them as targets in the disclosure. Such a screen has already been carried out in the context of tumor vaccination schemes. Targets that are valuable (because of sufficient specificity, not necessarily efficacy) as tumor vaccine candidates will also be valuable for the disclosure: MAGE-A1, -A2, -A3, -A4, -A5, -A6, -A7, -A8, -A9, -A10, -A11, -A12, -A12, MAGE-B, MAGE-C2, LAGE-1, PRAME, NY-ESO-1, PAGE, SSX-2, SSX-4, GAGE, TAG-1, TAG-2, and HERV-K-MEL.
[0066] Human tumor antigens presented by MHC class II molecules have been described, with nearly all of them being associated to multiple myeloma or malignant melanoma. The first antigenic peptide related to a melanoma-specific antigen found was a peptide derived from MAGE-1. Furthermore, three melanoma epitopes were found to originate from the MAGE family of proteins and presented by HLA-DR11 and HLA-DR13. Another set of melanoma antigens, known to contain also MHC class I tumor antigens, comprises Melan-A/MART-1, gp100 and tyrosinase. For an overview of T-cell epitopes that are of use for the disclosure, also see worldwide web at cancerimmunity.org/peptidedatabase/Tcellepitopes.htm.
[0067] The first discovered CT, belonging to the group of MAGE-A antigens, has an expression profile that is uniquely restricted to cancer cells and testis cells. However, testis cells are not targeted by the immune system, as they lack expression of MHC molecules. The MAGE-A antigens belong to a family of twelve genes that show high homology. Their expression has been associated with early events in malignant cell transformation and metastatic spread of cancer cells. In addition, down-regulation of MAGE-A expression may induce apoptosis in cancer cells. Within the MAGE-A genes several antigenic epitopes are known by the art. Antigenic peptides usually are presented as 8- or 9-mer amino acid peptides by MHC class I molecules. In addition, antigenic epitopes are known that are present in multiple MAGE-A genes due to the high homology between the different MAGE-A genes. These antigenic epitopes may be considered as multi-MAGE-A epitopes and are presented on cancer cells of various histologic origin. Therefore, they might serve as universal targets for anti-cancer therapy.
[0068] MHC molecules are also important as signal-transducing molecules, regulating immune responses. Cross-linking of MHC Class I molecules on B- and T-cells initiates signals that can result in either anergy, or apoptosis, or alternatively in cell proliferation and cytokine production. Several intracellular signaling pathways have been identified that are induced by MHC class I cross-linking. These include 1) phosphorylation of tyrosine kinases, leading to enhanced levels of intracellular calcium ions; 2) activation of the JAK/STAT pathway; and 3) inhibition of PI3K, resulting in the activation of JNK activation. Very high affinity antibodies against MHC that are internalized after binding may induce apoptosis. To be certain in the case of T cell and/or B cell derived tumors, the effect of the molecules may be tested in vitro before initiating therapy.
[0069] A further aspect relates to a method for providing the hexavalent single-chain monomeric protein hereof. As described herein above, it typically involves providing a nucleic acid encoding the desired hexavalent polypeptide. This nucleic acid molecule can be introduced, preferably via a plasmid or expression vector, into a prokaryotic host cell and/or in eukaryotic host cell capable of expressing the construct. In one embodiment, a method provides a hexavalent single-chain apoptosis inducing protein comprises the steps of providing a host cell with one or more nucleic acid(s) encoding the hexavalent protein capable of recognizing and binding to a specific MHC-peptide complex, and allowing the expression of the nucleic acids by the host cell.
[0070] Preferred host cells are bacteria, like, for example, bacterial strain BL21 or strain SE1, or mammalian host cells, more preferably human host cells. Suitable mammalian host cells include human embryonic kidney (HEK-293) cells or Chinese hamster ovary (CHO) cells, which can be commercially obtained. Insect cells, such as S2 or S9 cells, may also be used using baculovirus or insect cell expression vectors, although they are less suitable when the polypeptides include elements that involve glycosylation. The hexavalent single-chain polypeptides produced can be extracted or isolated from the host cell or, if they are secreted, from the culture medium of the host cell. Thus, in one embodiment, a method comprises providing a host cell with one or more nucleic acid(s) encoding the hexavalent single-chain polypeptide capable of recognizing and binding to a specific MHC-peptide complex, allowing the expression of the nucleic acids by the host cell. Methods for the recombinant expression of (mammalian) proteins in a (mammalian) host cell are well known in the art.
[0071] As will be clear, a hexavalent single-chain protein hereof finds its use in many therapeutic applications and non-therapeutic applications, e.g., diagnostics or scientific applications. Provided herein is a method for inducing ex vivo or in vivo apoptosis of a target cell, comprising contacting the cell with a hexavalent single-chain protein hereof in an amount that is effective to induce apoptosis. The target cells can be conveniently contacted with the culture medium of a host cell that is used for the recombinant production of the hexavalent single-chain protein. In one embodiment, it can be used for in vitro apoptosis studies, for instance studies directed at the elucidation of molecular pathways involved in MHC class I and class II induced apoptosis. Hexavalent single-chain proteins hereof may also be used for the detection of (circulating) tumor cells, for the target-cell-specific delivery of cytotoxic compounds or for the delivery of immune-stimulatory molecules.
[0072] Preferably, the hexavalent single-chain protein is used for triggering apoptosis of aberrant cells in a subject, more preferably a human subject. For therapeutic applications in humans it is preferred that a hexavalent single-chain protein does not contain amino acid sequences of non-mammalian origin. More preferred are hexavalent single-chain proteins, which only contain human amino acid sequences. Therefore, a therapeutically effective amount of a hexavalent single-chain protein capable of recognizing and binding to a disease-specific epitope can be administered to a patient to stimulate apoptosis of aberrant cells expressing the epitope without affecting the viability of (normal) cells not expressing the disease-specific epitope, e.g., a peptide antigen presented in the context of MHC. It is demonstrated herein that a method hereof allows for the killing of cells in an antigen-specific, MHC-restricted fashion. In a specific embodiment, the disease-specific epitope is a cancer-epitope, for example, a melanoma-specific epitope. The killing of aberrant (tumor) cells while minimizing or even totally avoiding the death of normal cells will generally improve the therapeutic outcome of a patient following administration of the hexavalent single-chain protein.
[0073] Also provided is a hexavalent single-chain protein hereof as medicament. In another aspect, provided is the use of a hexavalent single-chain protein for the manufacture of a medicament for the treatment of cancer. For example, a single-chain protein is advantageously used for the manufacture of a medicament for the treatment of melanoma.
[0074] Antibody fragments of human origin can be isolated from large antibody repertoires displayed by phages. One aspect hereof is the use of human antibody phage display libraries for the selection of human Fab fragments specific for MHC class I molecules presenting cancer testis antigenic peptides. Antibody Fab fragments specific for MHC class I, HLA-A0201 molecules presenting a multi-MAGE-A epitope have been selected (essentially as described in R. A. Willemsen et al., Cytometry A., 2008, 73:1093-1099) and shown to bind the relevant antigen only. As these antibody-Fab fragments usually display low affinity a method is provided that allows the generation of relatively high avidity antibody chains able to induce apoptosis in a MHC-restricted peptide specific way. An aspect of the disclosure is the development of a single-chain protein molecule comprising multiple antigen binding motifs to enhance MHC-peptide binding avidity, resulting in cross-linking of the MHC-peptide complexes and induction of apoptosis.
[0075] An MHC-p complex-specific polypeptide in a multivalent single-chain monomeric protein form hereof is, for example, an MHC-restricted antigen-specific TCR-like antibody (Ab) or functional fragment thereof, which is multimerized at the DNA level in order to obtain a single-chain polypeptide construct upon expression.
[0076] Human V.sub.H domains usually do not meet the standards for stability and efficient expression that are required by the field. They tend to be unstable and poorly expressed. A process called camelization may be used to convert human V.sub.H into more stable antibody fragments.
[0077] The human antibody germline region V.sub.H-3 displays high homology with antibody V.sub.H fragments of llamas. Llamas have two types of antibodies, those composed of heavy and light chains, and antibodies that only contain heavy chains. These heavy-chain only antibodies bind antigens similar to classical antibodies composed of heavy and light chains. The smallest functional llama antibody binding domain, the V.sub.HH domain, also called single domain antibodies (sdAb), has been shown to be expressed well and may bind antigen with high affinity. In addition, it has been shown that some of the characteristics, such as ease of expression and stability, of llama sdAb can be transferred to, e.g., human V.sub.H by replacing a few amino acids in the human V.sub.H for those of llama V.sub.H. High avidity antibody molecules can then be generated by ligation of several camelized human V.sub.H domains into one single molecule.
[0078] Preferred molecules may comprise up to six camelized or non-camelized human V.sub.H domains interspersed by short linkers providing flexibility between the V.sub.H domains, thus generating six essentially identical binding domains specific for a single epitope (see, for an example, SEQ ID NO:4 and SEQ ID NO: 13). For example, a hexavalent mono-specific protein is generated that is specific for the HLA-A0201 restricted multi-MAGE-A epitope within a single polypeptide, referred to as a single-chain protein or single-chain polypeptide or monomeric protein or monomeric polypeptide. See, for further details, the outlined Examples below. It may be appreciated that this technology allows for the generation of multivalent single-chain proteins that comprise any number of the same or different single domain antibodies. For several reasons (such as, ease of production) repeats are not always the best option. Thus, also contemplated is using different binding domains (essentially recognizing the same target) separated by several different linkers, as shown in
[0079] A hexavalent single-chain monomeric protein hereof, comprising six linearly linked human V.sub.H domains is used, for example, to induce apoptosis in cancer cells that express both the MAGE-A genes and HLA-A0201. Noteworthy, specificity for this MHC-peptide complex is provided in this way as cells that do not express HLA-A0201 or that do not express MAGE-A are not killed. See the Examples section for further details. Apoptosis in cancer cells is, for example, detected in vitro by several assays known to the art, including cytotoxicity assays, Tunnel assays and assays detecting active caspases. In animal studies, apoptosis is, for example, revealed by monitoring reduced tumor growth, detection of active caspases or performing a tunnel assay on isolated tumor material.
[0080] In literature, it is shown that a single nine amino acid (A.A.) peptide present in MAGE-A2, -A3, -A4, -A6, -A10, and -A12 is presented by HLA-A0201 on tumor cells, and can be recognized by cytotoxic T-lymphocytes..sup.(1) This nine A.A. peptide with sequence Y-L-E-Y-R-Q-V-P-G (SEQ ID NO:7) is almost identical to the HLA-A0201 presented MAGE-A1 peptide Y-L-E-Y-R-Q-V-P-D (SEQ ID NO:9), except for the anchor residue at position 9. Replacement of the anchor residue with Valine results in a 9 A.A. peptide with enhanced binding capacity to HLA-A0201 molecules..sup.(1) Human and mouse T-lymphocytes recognizing the Y-L-E-Y-R-Q-V-P-V (SEQ ID NO:10) peptide presented by HLA-0201 also recognize the original MAGE-A Y-L-E-Y-R-Q-V-P-G (SEQ ID NO:7) and Y-L-E-Y-R-Q-V-P-D (SEQ ID NO:9) peptides presented on tumors of distinct origin. As diverse tumors may each express at least one MAGE-A gene, targeting of this so-called multi-MAGE-A epitope includes the vast majority of tumors. As an example, MAGE-A expression in human prostate tumor cell lines and in human xenographs was analyzed and shown to be highly diverse, but in each individual sample tested at least one MAGE-A gene was expressed (Table 2), confirming that targeting this multi-MAGE-A epitope serves as an essentially universal HLA-A0201 restricted target for therapy.
[0081] Of course, several other multi mage or multi target epitopes may be designed. Contemplated are combinations of tumor-specific antigen derived MHC presented epitopes in different HLA restrictions of both MHC-I and MHC-II targeted by multimeric (>=4) binding domains to induce apoptosis in aberrant cell. A number of MHC-peptide combinations that can be targeted (but not limited to) are HLA-A0201/YLEYRQVPG/D (SEQ ID NO:7/9), HLA-CW7/EGDCAPEEK (SEQ ID NO:8), HLA-A24/TFPDLESEK (SEQ ID NO:26) or IMPKAGLLI (SEQ ID NO:27), and HLA-DP4 or HLA-DQ6/KKLLTQHFVQENYLEY (SEQ ID NO:28).
[0082] In one embodiment, human antibody fragments specific for the HLA-A0201 presented multi-MAGE-A epitope Y-L-E-Y-R-Q-V-P-V (SEQ ID NO:10) are identified and isolated from a human phage display library. The selected human antibody fragments are optimized regarding their specificity and avidity, and provide the amino acid sequences used for the design and production of hexavalent single-chain polypeptides specific for efficient binding of the HLA-A0201-MAGE-A Y-L-E-Y-R-Q-V-P-G (SEQ ID NO:7) complex, referred to as hexa-AH5. In another embodiment, hexa-AH5 is produced comprising a C-terminal human antibody Fc domain amino acid sequence, providing hexa-AH5Fc with essentially the same or comparable binding characteristics compared to hexa-AH5. In yet another embodiment, hexa-AH5 is produced comprising a C-terminal human serum albumin (HSA) amino acid sequence, providing hexa-AH5HSA with essentially the same or comparable binding characteristics compared to hexa-AH5.
[0083] In one embodiment, the hexa-AH5 and/or its equivalents hexa-H5Fc and/or hexa-H5HSA are used in the production of a pharmaceutical composition. In yet another embodiment, hexa-AH5 construct(s) is/are used for the production of a pharmaceutical composition for the treatment of a disease or a health problem related to the presence of aberrant cells exposing the epitope comprising the HLA-A0201-MAGE-A Y-L-E-Y-R-Q-V-P-G (SEQ ID NO:7) complex for hexa-AH5, hexa-AH5Fc and hexa-AH5HSA. The aberrant cells are, for example, tumor cells. In a further embodiment, hexa-AH5 and/or its equivalents hexa-AH5Fc and/or hexa-AH5HSA is used for the treatment of cancer. In yet another embodiment, hexa-AH5 and/or its equivalent, is used, for example, for the treatment of prostate cancer, breast cancer, multiple myelomas or melanomas.
[0084] The disclosure is exemplified by the Examples below.
ABBREVIATIONS USED
[0085] A.A., amino acid; Ab, antibody; ADA, anti-drug antibodies; AFP, alpha-fetoprotein; APC, antigen presenting cell; 2-M, 2-microglobulin; CDR, complementarity determining region; CEA, carcino-embryonic antigen; CHO, Chinese hamster ovary; CT, cancer testis antigens; CTL, cytotoxic T-lymphocyte; DC, dendritic cell; EBV, Epstein-Barr virus; ELISA, enzyme linked immunosorbent assay; HEK, human embryonic kidney; HLA, human leukocyte antigen; i.v., intravenously; kDa, kilo Dalton; MAGE, melanoma-associated antigen; MHC, major histocompatibility complex; MHC-p, MHC-peptide; PBSM, PBS containing 2% non-fat dry milk; sc-Fv, single-chain variable fragment; V.sub.HH or sdAb, single domain antibodies; TCR, T-cell receptor; VH, Vh or V.sub.H, variable amino acid sequence of an antibody heavy domain.
EXAMPLES
Example 1: Selection of Human Antibody Fragments Specific for HLA-A0201/Multi-MAGE-A
[0086] 1.1
[0087] To obtain human antibody fragments specific for the HLA-A0201 presented multi-MAGE-A epitope Y-L-E-Y-R-Q-V-P-G (SEQ ID NO:5) a Human Fab phage display library was constructed according to the procedure previously described by de Haard et al..sup.(2) and used for selections essentially as described by Chames et al..sup.(3) Human Fab phages (10.sup.13 colony-forming units) were first pre-incubated for 1 hour at room temperature in PBS containing 2% non-fat dry milk (PBSM). In parallel, 200 l Streptavidin-coated beads (Dynal) were equilibrated for 1 hour in PBSM. For subsequent rounds, 100 l beads were used. To deplete for pan-MHC binders, each selection round, 200 nM of biotinylated MHC class I-peptide (MHC-p) complexes containing an irrelevant peptide (Sanquin, the Netherlands) were added to the phages and incubated for 30 minutes under rotation. Equilibrated beads were added, and the mixture was incubated for 15 minutes under rotation. Beads were drawn to the side of the tube using magnetic force. To the depleted phage fraction, subsequently decreasing amounts of biotinylated MHC-p complexes (200 nM for the first round, and 20 nM for the second and third round) were added and incubated for 1 hour at room temperature, with continuous rotation. Simultaneously, a pan-MHC class I binding soluble Fab (D3) was added to the phage-MHC-p complex mixture (50, 10, and 5 g for rounds 1-3, respectively). Equilibrated streptavidin-coated beads were added, and the mixture was incubated for 15 minutes under rotation. Phages were selected by magnetic force. Non-bound phages were removed by five washing steps with PBSM, five steps with PBS containing 0.1% TWEEN, and five steps with PBS. Phages were eluted from the beads by 10 minutes incubation with 500 l freshly prepared tri-ethylamine (100 mM). The pH of the solution was neutralized by the addition of 500 l 1 M Tris (pH 7.5). The eluted phages were incubated with logarithmic growing E. Coli TG1 cells (OD.sub.600nm of 0.5) for 30 minutes at 37 C. Bacteria were grown overnight on 2TYAG plates. The next day, colonies were harvested, and a 10 l inoculum was used in 50 ml 2TYAG. Cells were grown until an OD.sub.600m of 0.5, and 5 ml of this suspension was infected with M13k07 helper phage (510.sup.11 colony-forming units). After 30 minutes incubation at 37 C., the cells were centrifuged, resuspended in 25 ml 2TYAK, and grown overnight at 30 C. Phages were collected from the culture supernatant, as described previously, and were used for the next round panning. After three selection rounds a 261-fold enrichment was obtained, and 46 out of 282 analyzed clones were shown to be specific for the HLA-A2-multi-MAGE-A complex (
1.2 Human Fab Specific for the HLA-A0201/Multi-MAGE-A Epitope Bind Antigen-Positive Cells
[0088] Selected Fab phages were then analyzed for their capacity to bind HLA-A0201-positive EBV-transformed B-LCL loaded with the multi-MAGE-A peptide Y-L-E-Y-R-Q-V-P-V (SEQ ID NO:10). The B-LCL line BSM (0.510.sup.6) was loaded with multi-MAGE-A peptide (10 g in 100 ptl PBS) for 30 minutes at 37 C., followed by incubation with the Fab phages AH5, CB1, CG1, BD5 and BC7 and analyzed by flow-cytometry. As shown in
[0089] Phages presenting AH5, CB1 and CG1, as well as the HLA-A0101/MAGE-A1-specific Fab phage G8.sup.(4) were then used to stain tumor cell lines of distinct histologic origin. To this end prostate cancer cells (LNCaP), multiple myeloma cells (MDN), melanoma cells (MZ2-MEL43 and G43), and breast cancer cells (MDA-MB157) were stained and analyzed by flow cytometry (
1.3 Fab AH5 Binds HLA-A0201/Multi-MAGE-A Complexes Only
[0090] ELISA using multiple peptide/MHC complexes then confirmed the specificity of Fab-AH5. To this end HLA-A0201 complexes presenting peptides multi-MAGE-A, gp100, JCV and MAGE-C2, as well as a HLA-A1/MAGE-A1 complex were immobilized on 96-well plates and incubated with phages displaying Fab AH5 and control Fab G8. As shown in
Example 2: Production of Hexameric Proteins Comprising Camelized Single Domains AH5 VH Domains
2.1 Design of Genes for Production of Hexameric AH5 VH Proteins
[0091] Human antibody germline gene VH3 demonstrates high homology to llama single domains VHH. Exchange of amino acids 44, 45 and 47 in the human VH3 genes by amino acids present in llama VHH at these positions has shown to enhance stability and expression of the human VH3 genes..sup.(5) The AH5 VH demonstrates a low homology to germline gene VH3-33*01 (71% as determined by IMGT homology search) however, its expression and stability might benefit from the exchange of amino acids 44, 45 and 47 by llama VHH amino acids, a process called camelization. In addition a gene was compiled that upon expression would comprise six AH5 VH domains. To this end, a gene called hexa-AH5 was designed comprising the pelB secretion signal, which was operatively linked to six codon-optimized, camelized AH5 VH domains with GSTSGS linkers between each AH5 VH domain (see hexa-AH5, see SEQ ID NO:1 for the DNA sequence and SEQ ID NO:4 for the amino acid sequence). This gene was synthesized by Geneart (Regensburg, Germany) and cloned into the pStaby 1.2 vector (Delphi genetics, Belgium) for expression in E. coli.
2.2 Production and Purification of Hexameric AH5 VH Protein
[0092] For expression of hexameric AH5 VH proteins (hexa-AH5, see SEQ ID NO: 1 for the DNA sequence and SEQ ID NO:4 for the amino acid sequence) the pStaby-Hexa-AH5 vectors were introduced via electroporation into SE1 bacteria. Positive clones were grown in the presence of 2% glucose at 30 C. until OD.sub.600=0.8. Bacterial TYAG medium was then replaced with TY medium containing 1 mM IPTG to induce expression. After overnight culture at 30 C. bacteria and medium were harvested. The periplasm fraction was collected after incubation of bacteria with PBS/EDTA/NaCl for 30 minutes on ice. Protein expression was then analyzed by SDS-PAGE. As shown in
[0093] Hexameric AH5 VH proteins were isolated from media and bacteria using Ni-affinity purification. To this end, medium was incubated with Ni-coupled Sepharose-beads and incubated overnight, while stirring gently. To obtain intracellular proteins bacteria were lysed and cellular debris removed by centrifugation. After overnight dialysis with PBS Hexameric AH5 VH proteins were purified with Ni-Sepharose. Purity of the Hexameric AH5 VH proteins was checked by SDS-PAGE and protein concentration determined by BCA protein assay (Pierce).
Example 3: Hexameric AH5 VH Proteins Induce Apoptosis in Diverse Tumor Cells
[0094] Cross-linking of MHC class I molecules by pan-MHC class-I and 32M-specific antibodies results in the induction of apoptosis..sup.(6) This process was shown to be caspase-9 dependent and results in the eradication of MHC class I-positive tumor cells in vitro and in vivo. The induction of apoptosis by pan-MHC class I antibodies and anti-2M-specific antibodies is not specific for tumors expressing tumor-specific antigens. In contrast, cross-linking of peptide/MHC molecules through the interaction of molecules that resemble T-cell receptors binding to specific peptide/MHC complexes will result in tumor-specific apoptosis induction. Efficient cross-linking will depend on the number of peptide/MHC complexes that are simultaneously bound by the therapeutic molecule.
3.1 Hexameric AH5 Protein Kills Diverse Tumor Cells
[0095] The hexameric AH5-VH proteins were analyzed for their capacity to induce apoptosis by incubation with diverse tumor cells, known to express both HLA-A0201 and MAGE-A genes. The cell-lines Daju, Mel 624 (melanoma), PC346C (prostate cancer), as well as MAGE-A-negative cells (911 and HEK293T) were incubated with 10 g/ml Hexa-AH5 protein (in DMEM medium, supplemented with pen/strep, Glutamine and non-essential amino acids). Four hours later, cells were visually inspected for classical signs of apoptosis, such as detachment of the cells from tissue culture plates and membrane blebbing. As shown in
3.2 Hexameric AH5 Protein Induces Active Caspase-3
[0096] A classical intra-cellular hallmark for apoptosis is the presence of active caspase-3. To determine whether or not the Hexameric AH5 proteins induce active caspase-3, Daju cells were incubated with 10 g/ml Hexa-AH5 protein. After four hours FAM-DEVD-FMK (SEQ ID NO:25), a fluorescently labeled inhibitor for caspase-3/7 was added to the tissue culture medium. This substrate can pass the cell-membrane and only when active caspase-3 is present, a bright fluorescent signal will be detected by, e.g., fluorescent microscopy.
[0097] As shown in
Example 4: Hexameric AH5 Protein Induces Apoptosis in a Transplantable Human Tumor Model
[0098] To demonstrate apoptotic activity of the Hexa-AH5 proteins in three-dimensional human tumors, an orthotopic prostate cancer model was used. To this end, human PC346C prostate cancer cells were injected into the prostate of male NOD-SCID mice and allowed to grow until tumors were detectable by ultrasound guided inspection.
4.1 Intra-Tumoral Injection of Hexa-AH5 Results Induces Apoptosis
[0099] The human PC346C prostate tumors in NOD-SCID mice were injected once directly with 10 g Hexa-AH5 protein (in 20 l total volume). The next day, tumors were removed, fixed and paraffin embedded. Slides were prepared from the paraffin-embedded tumors and stained with the Tunnel Universal Apoptosis Detection Kit (Genescript), an assay that detects fragmented DNA, a classical marker of apoptosis. In brief, slides were heated for 30 minutes at 60 C., washed three times with PBS, and incubated for one hour with proteinase K solution. Slides were then incubated with blocking solution (3% H.sub.2O.sub.2 in methanol) for 10 minutes, washed with PBS and incubated for one hour at 37 C. with Tunnel reaction mixture (equilibrium buffer, Biotin-11-dUTP, and TdT). After three washes slides were incubated with Streptavidin-HRP solution for 30 minutes at 37 C., and finally incubated with DAB-substrate (DAB-buffer, H.sub.2O.sub.2 in PBS).
[0100] Microscopic analysis of tumor material treated with Hexa-AH5 demonstrates large areas of apoptotic cells (see
4.2 Intravenous Injection of Hexa-AH5 Induces Apoptosis in Orthotopic Prostate Cancer Cells
[0101] 4.2.1 Prostate Tumor Cells Demonstrate Nicked DNA after i.v. Injection with Hexa-AH5
[0102] In a next experiment NOD-scid mice with the orthotopic human PC346C prostate tumor were injected once via tail vain with 25 g Hexa-AH5 (in 150 jil total volume). The next day, tumors were removed, paraffin embedded and tumor slides stained for Nicked DNA with the Tunnel assay.
[0103] As shown in
4.2.2 Prostate Tumor Cells Demonstrate Active Caspase after i.v. Injection with Hexa-AH5
[0104] NOD-scid mice with the orthotopic PC346C tumor were injected once via tail vain with 25 g Hexa-AH5 (in 150 l total volume). The next day, these mice received an injection with FLIVO (Immunohistochemistry Ltd.), a fluorescently labeled caspase inhibitor. This inhibitor was allowed to circulate and pass cellular membranes for one hour. Tumors were then removed, fixed and paraffin embedded.
[0105] Analysis of Hexa-AH5-treated tumors by fluorescence microscopy demonstrated the presence of numerous cells that stained positive for the caspase substrate (see
Example 5: Construction of Hexa-AH5 Genes to Improve Circulation and Tumor Penetration
[0106] The pharmacokinetic properties of therapeutic proteins, e.g., their distribution, metabolism and excretion are dependent on factors, such as shape, charge and size. Most small plasma molecules (MW<50-60 kDa) possess very short half-life, whereas larger plasma proteins, such as human serum albumin (HSA) and immunoglobulins (Ig) have very long half-lives (19 days for HSA, 1-4 weeks for Ig). Indeed, addition of IgG-Fc or Human serum albumin has shown to extend circulation time, tumor penetration and antitumor effects when linked to therapeutic proteins.
5.1 Construction of Hexameric AH5 with IgG1-Fc and Human Serum Albumin
[0107] The Hexameric AH5 construct was linked to the IgG1-Fc region or to human serum albumin, codon optimized for expression in eukaryotic cells and cloned into the pcDNA-3.1+ vector (Geneart, Regensburg, Germany) (see DNA sequence with SEQ ID NO:2 and amino acid sequence with SEQ ID NO:5 for hexa-AH5Fc, and see DNA sequence with SEQ ID NO:3 and amino acid sequence with SEQ ID NO:6 for hexa-AH5HSA, respectively).
5.2 Hexameric AH5-Fc and AH5-HSA Induce Active Caspase-3
[0108] The hexameric AH5-FC and AH5-HSA constructs, cloned into pcDNA-3.1+, were expressed in 293T cells. Supernatant obtained four days after transfection was used to induce apoptosis in melanoma 624 cells known to express HLA-A0201 and MAGE-A genes. To this end, melanoma 624 cells were seeded in 24-well plates (0.2510.sup.6 cells/well) and allowed to attach overnight. The next day, medium was replaced with medium obtained from transfected 293T cells. Results showed positive caspase-3 staining for 624 melanoma cells treated with both hexa-AH5-Fc and Hexa-AH5-HSA. No staining was observed for 624 cells incubated with plain medium or HLA-A0201 positive, MAGE-A-negative 911 cells (
5.3 Extended Survival of Mice and Delayed Tumor Growth of Mice Treated with Hexameric AH5
[0109] Mice inoculated with melanoma cell line Daju (HLA-A0201/MAGE-A positive) were treated with intravenous injections of hexameric AH5 protein (2.5 ug/2 times/week). Shown are 1) tumor free mice, and 2) tumor growth (
5.4 Enhanced Induction of Apoptosis by Dimeric Hexameric AH5CH1 and 11HCH1.
[0110] For expression in eukaryotic cells the AH5CH1 and 11HCH1 sequences were introduced into the pMSec SUMOSTAR vector (Hexameric AH5CH1 and 1lHCH1 were produced in supernatant of 293T cells after transfection with CaPO4. One hour after incubation of Daju and MEL624 melanoma cells with 293T supernatant (1:1 diluted in DMEM, 5% FCS) membrane blebbing and detachment of cells were observed.
5.5 Improved Expression of Hexameric AH5 at 25 C.
[0111] Expression of Hexameric AH5 in SE1 at 30 C. or 37 C. in shaking flasks was shown to result in many unwanted smaller products. Lowering the temperature during growth and production to 25 C. resulted in a marked improvement of production. Less, to no side products were obtained as well as a higher yield of the protein (
TABLE-US-00004 TABLE 1 Examples for the frequency of MAGE-A expression by human cancers Table 1: Examples for the frequency of MAGE-A expression by human cancers Frequency of expression (%) MAGE- MAGE- MAGE- MAGE- MAGE- MAGE- MAGE- cancer A1 A2 A3 A4 A6 A10 A11 Melanoma 16 E 36 E 64 E 74 Head and neck 25 42 33 8 N N N Bladder 21 30 35 33 15 N 9 Breast 6 19 10 13 5 N N Colorectal N 5 5 N 5 N N Lung 21 30 46 11 8 N N Gastric 30 22 57 N N N N Ovarian 55 32 20 E 20 N N osteosarcoma 62 75 62 12 62 N N hepatocarcinoma 68 30 68 N 30 30 30 Renal cell 22 16 76 30 N N N carcinoma E, expressed but the frequency is not known; N, expression by tumors has never been determined or observed
TABLE-US-00005 TABLE 1B Expression analysis of MAGE-A1-A6 genes detected by nested RT-PCR with common primers in squamous cell carcinoma of the head and neck. Primary site % of positive expression Larynx 72.7% (8/11) Hypopharynx 100% (2/2) Base of tongue 50% (1/2) Tonsil 100% (2/2) Total (n = 17) 76.5% (13/17)
Adapted from: ANTICANCER RESEARCH 26: 1513-1518 (2006)
TABLE-US-00006 TABLE 2 MAGE-A expression in human prostate cancer cell lines and prostate cancer xenografts. Table 2: MAGE-A expression in human prostate cancer cell lines and prostate cancer xenografts. Cellline/ MAGE- Xenograft A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 LNCaP + ++ ++ ++ + PC346C + ++ ++ + ++ + + ++ OVCAR + + + + JON ++ ++ ++ + + PNT 2 + + + + + C2 SD48 + + + + PC-3 + + + PC 374 + PC 346p + ++ ++ ++ + ++ + PC 82 + + PC 133 ++ + + PC 135 + PC 295 + PC 324 + + + PC 310 + ++ + ++ + PC 339 ++ ++ + ++ + + + Expression of the MAGE-A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11 and A12 genes in diverse prostate tumor cell lines and prostate xenografts was analyzed by RT-PCR. Shown are expression levels in individual samples tested. Blank = no expression, + = low expression, ++ = high expression. All cell lines/xenografts express at least one MAGE-A gene.
REFERENCES
[0112] (1) Stephanie Graff-Dubois, Olivier Faure, David-Alexandre Gross, Pedro Alves, Antonio Scardino, Salem Chouaib, Francois A. Lemonnier and Kostas Kosmatopoulos. Generation of CTL Recognizing an HLA-A*0201-Restricted Epitope Shared by MAGE-A1, -A2, -A3, -A4, -A6, -A10, and -A12 Tumor Antigens: Implication in a Broad-Spectrum Tumor Immunotherapy. The Journal of Immunology, 2002, 169:575-580. [0113] (2) Hans J. de Haard, Nicole van Neer, Anneke Reurs, Simon E. Hufton, Rob C. Roovers, Paula Henderikx, Adriaan P. de Bruine, Jan-Willem Arends, and Hennie R. Hoogenboom. A Large Non-immunized Human Fab Fragment Phage Library That Permits Rapid Isolation and Kinetic Analysis of High Affinity Antibodies. The Journal of Biological Chemistry, 1999, 274:18218-18230. [0114] (3) P. Chames, H. R. Hoogenboom, and P. Henderikx. Selection of antigens against biotinylated antigens. In Antibody phage display, methods and protocols, Edited by P. M. O'Brien and R. Aitken. Methods in Molecular Biology 2002, 178:147-159. [0115] (4) Patrick Chames, Simon E. Hufton, Pierre G. Coulie, Barbara Uchanska-Ziegler, Hennie R. Hoogenboom. Direct selection of a human antibody fragment directed against the tumor T-cell epitope HLA-A1-MAGE-A1 from a nonimmunized phage-Fab library. PNAS, 2000. 97:7969-7974. [0116] (5) Lutz Riechmann, Serge Muyldermans. Single domain antibodies: comparison of camel VH and camelized human VH domains. Journal of Immunological Methods 1999, 231:25-38. [0117] (6) Jing Yang, PhD and Qing Yi. Killing Tumor Cells Through Their Surface b2-Microglobulin or Major Histocompatibility Complex Class I Molecules. Cancer 2010. 116:1638-1645.