Covalently-linked complexes and methods for enhanced cytotoxicity and imaging

Description Submit all comments and votes

The present invention relates to covalently-linked complexes (CLC) having enhanced diagnostic or therapeutic properties and methods of using these complexes. The CLC of the present invention has three components: (1) a targeting protein; (2) a cytotoxic agent, such as a radioisotope, a drug or a toxin; and (3) one or more enhancing moieties capable of promoting CLC-target cell interaction.

BACKGROUND OF THE INVENTION

Immunoconjugates consisting of antibody joined to a cytotoxic agent have been used in attempts to achieve selective killing of particular target cells, such as tumor cells. In theory, immunoconjugates or targeting protein conjugates should effect specific cellular cytotoxicity. In practice, however, in vivo administration of immunoconjugates has proven less efficacious than anticipated.

Several disadvantages related to retention, internalization and translocation of immunoconjugates have been identified. For instance, optimal retention of isotope-antibody fragment conjugates within tumor tissue after in vivo administration has not been demonstrated. Additional problems associated with target cell internalization and translocation of immunoconjugates have been recognized, particularly in regards to translocation and internalization of A-chain (derived from plant or bacterial toxin) immunoconjugates.

Thus, there is a need in the art for improved: (1) retention of targeting protein conjugates (especially antibody fragment conjugates) at target cell plasma membranes; (2) internalization of targeting protein conjugates into target cell endosomic vesicles; and (3) translocation of targeting protein conjugates across target cell endosomic vesicular membranes into the cytoplasm. Enhancement of the interaction of targeting protein conjugates with plasma membranes and/or internal membranes of target cells may improve the cytotoxicity of targeting protein conjugates administered in vivo. The present invention fulfills this need and further provides other related advantages.

SUMMARY OF THE INVENTION

The present invention describes a covalently-linked complex (CLC) for targeting a defined population of cells, comprising a targeting protein; a cytotoxic agent; and an enhancing moiety, wherein the enhancing moiety is capable of promoting CLC-target cell interaction.

A method for enhancing in vivo cytotoxicity of a targeting protein conjugate comprising administering to a tumor-bearing patient a therapeutically effective amount of the covalently-linked complex of the present invention is also disclosed.

In addition, a method for enhanced in vivo imaging of a tumor comprising administering to a tumor-bearing patient a diagnostically effective amount of the claimed covalently-linked complex is disclosed.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates a helical net structure representing an advantageous spatial arrangement of amino acids present in a translocating moiety of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

Prior to setting forth the invention, it may be helpful to set forth definitions of certain terms to be used within the disclosure.

Targeting protein: A protein, peptide or non-proteinaceous molecule that binds to a defined population of cells. The targeting protein may bind a receptor, an enzymatic substrate, an antigenic determinant, or other binding site present on the target cell population. Hereinafter, the term "targeting protein" will be inclusive of targeting proteins, targeting peptides and non-proteinaceous targeting molecules.

Translocating/internalizing moiety: A moiety capable of insertion into membranes at acidic pH (typically pH 5.0-5.5), or capable of insertion into or across membranes at pH 6-8.

Anchoring peptide: A peptide capable of insertion into membranes at physiological pH (typically pH 6.8-7.5).

Accessory moiety: A proteinaceous or non-proteinaceous moiety that serves as a substrate for target cell enzymes, promotes membrane retention or translocation of one or more anchoring, translocating or internalizing moieties and/or promotes target cell retention of a CLC.

Intracellular retention moiety: A peptide or non-peptide molecule that binds to specific intracellular structures or organelles, and promotes intracellular retention of a covalently linked targeting protein conjugate.

Combination peptide: An elongated, synthetic peptide that sequentially incorporates two or more enhancing moieties.

Fusion protein: A hybrid protein generated by means of recombinant DNA technology. A fusion protein is translated from messenger RNA as one continuous polypeptide chain, with the protein or peptide components joined together by peptide bonds.

Conjugate: A two-component hybrid molecule wherein the components are joined by a covalent chemical linkage.

Targeting protein conjugate: A covalently-linked two-component conjugate wherein one component is antibody (i.e., an immunoconjugate) or, more generally, a targeting protein. Typically, the second component of a targeting protein conjugate is a cytotoxic agent, such as a drug, a toxin, a cytotoxic peptide or a radionuclide. In contrast to fusion proteins, recombinant DNA methods are not involved in the covalent linkage of targeting protein conjugate components.

Covalently-linked complex (CLC): A three-component complex comprising (1) a targeting protein; (2) a cytotoxic agent; and (3) an enhancing moiety; wherein the three components of the CLC are joined together by covalent bonds.

Enhancing moiety: A moiety capable of promoting membrane interaction. Enhancing moieties of the present invention include translocating/internalizing moieties, anchoring peptides, accessory peptides, membrane-soluble hydrophobic molecules and intracellular retention peptides. In addition, an enhancing moiety may be synthesized with others in a larger combination peptide, or be fused to one or more components of a fusion protein. One or more enhancing moieties may be covalently attached to a targeting protein conjugate to form a CLC having enhanced membrane interactive characteristics.

In general, three levels of targeting protein conjugate-membrane interaction have been identified that may be important for optimal in vivo diagnostic or therapeutic efficacy: (1) binding of the conjugate to the target cell plasma membrane; (2) internalization of the conjugate into endosomic vesicles; and (3) translocation of the conjugate from endosomic vesicles into the cytoplasm, which gives a targeting protein conjugate access to cytoplasmic or nuclear target sites. If any one of these targeting protein conjugate-membrane interactive steps becomes rate-limiting, targeting protein conjugate potency may be diminished.

Optimization of the three levels of target cell membrane interaction noted above (i.e., retention, translocation, internalization) may enhance the cytotoxicity of targeting protein conjugates. Different types of targeting protein conjugates (for instance, targeting protein conjugated to either a drug, toxin or radioisotope) may require different levels of targeting protein conjugate-membrane interaction in order to achieve optimal cytotoxicity in vivo.

More specifically, radioisotope-targeting protein conjugates require binding and prolonged retention of the conjugate, either within the tumor or at the tumor cell plasma membrane, for maximal cytotoxic efficacy. Drug-targeting protein conjugates that are active at the plasma membrane may require (1) binding of the targeting protein conjugate at the plasma membrane, and (2) expression of cytolytic activity at the plasma membrane. Drug-targeting protein conjugates that are not active at the target cell plasma membrane additionally require internalization of the drug for cytotoxicity. Drug conjugates of this latter type and toxin-targeting protein conjugates require three levels of membrane interaction for cytotoxicity: (1) binding of the targeting protein conjugate at the plasma membrane; (2) internalization of the conjugate within the target cell; and (3) translocation of the conjugate from endosomic vesicles into the cytoplasm.

The "targeting protein" component of the covalently-linked complex (CLC) of the present invention directs a covalently-attached cytotoxic agent to a target cell population, such as tumor cells. Preferred targeting proteins useful in this regard include antibody and antibody fragments; peptides, such as bombesin, gastrin-releasing peptide, cell adhesion peptides, substance P, neuromedin-B, neuromedin-C, and metenkephalin; and hormones, such as EGF, .alpha.- and .beta.-TGF, estradiol, neurotensin, melanocyte stimulating hormone, follicle stimulating hormone, luteinizing hormone, and human growth hormone. Biotin, avidin, proteins corresponding to known cell surface receptors (including low density lipoproteins, transferrin and insulin), fibrinolytic enzymes, and biological response modifiers (including interleukin, interferon, erythropoietin and colony-stimulating factor) are also preferred targeting proteins. Analogs of the above-listed targeting proteins that retain the capacity to bind to a defined target cell population may also be used within the claimed invention. In addition, synthetic targeting proteins and peptides may be designed. Antibody and antibody fragments are particularly preferred targeting proteins within the claimed invention.

Monoclonal antibodies have precise specificity for a particular epitope present on a target cell population. When a cytotoxic agent, such as a drug, toxin, cytotoxic peptide or radioisotope, is conjugated to a monoclonal antibody, increased amounts of the cytotoxic agent may be administered in vivo (as compared to the unconjugated form of the cytotoxic agent), due to the selective targeting properties of the monoclonal antibody component of the conjugate.

Types of cytotoxic agents useful herein include toxins, drugs, cytotoxic peptides and radionuclides. Several of the potent toxins useful within the present invention consist of an A and a B chain. The A chain is the cytotoxic portion and the B chain is the receptor-binding portion of the intact toxin molecule (holotoxin). Because toxin B chain may mediate non-target cell binding, it is often advantageous to conjugate only the toxin A chain to a targeting protein. However, while elimination of the toxin B chain decreases non-specific cytotoxicity, it also generally leads to decreased potency of the toxin A chain-targeting protein conjugate, as compared to the corresponding holotoxin-targeting protein conjugate.

One possible explanation for the decreased potency of A chain-targeting protein conjugates is that B chain is required for translocation of the A chain across endosomic membranes into the target cell cytoplasm. In the absence of translocation, the targeting protein conjugate remains in the interior of an endosome, and is ultimately transported to a lysosome. Within the lysosome, the targeting protein conjugate is degraded, and thus the A chain cytotoxic agent fails to reach its cytoplasmic target site. The decreased potency associated with toxin A chain-targeting protein conjugates also accompanies the use of ribosomal inactivating protein-targeting protein conjugates. Ribosomal inactivating proteins (RIPs) are naturally occurring protein synthesis inhibitors that lack translocating and cell-binding ability.

Within the present invention, preferred toxins include holotoxins, such as abrin, ricin, modeccin, Pseudomonas exotoxin A, Diphtheria toxin, pertussis toxin and Shiga toxin; and A chain or "A chain-like" molecules, such as ricin A chain, abrin A chain, modeccin A chain, the enzymatic portion of Pseudomonas exotoxin A, Diphtheria toxin A chain, the enzymatic portion of pertussis toxin, the enzymatic portion of Shiga toxin, gelonin, pokeweed antiviral protein, saporin, tritin, barley toxin and snake venom peptides.

Preferred drugs suitable for use herein include conventional chemotherapeutics, such as vinblastine, doxorubicin, bleomycin, methotrexate, 5-fluorouracil, 6-thioguanine, cytarabine, cyclophosphamide and cisplatinum, as well as other conventional chemotherapeutics as described in Cancer: Principles and Practice of Oncology, 2d ed., V. T. DeVita, Jr., S. Hellmann, S. A. Rosenberg, J. B. Lippincott Co., Philadelphia, Pa., 1985, Chapter 14. A particularly preferred drug within the present invention is a trichothecene.

Experimental drugs, such as mercaptopurine, N-methylformamide, 2-amino-1,3,4-thiadiazole, melphalan, hexamethylmelamine, gallium nitrate, 3% thymidine, dichloromethotrexate, mitoguazone, suramin, bromodeoxyuridine, iododeoxyuridine, semustine, 1-(2-chloroethyl)-3-(2,6-dioxo-3-piperidyl)-1-nitrosourea, N,N'-hexamethylene-bis-acetamide, azacitidine, dibromodulcitol, Erwinia asparaginase, ifosfamide, 2-mercaptoethane sulfonate, teniposide, taxol, 3-deazauridine, soluble Baker's antifol, homoharringtonine, cyclocytidine, acivicin, ICRF-187, spiromustine, levamisole, chlorozotocin, aziridinyl benzoquinone, spirogermanium, aclarubicin, pentostatin, PALA, carboplatin, amsacrine, caracemide, iproplatin, misonidazole, dihydro-5-azacytidine, 4'-deoxydoxorubicin, menogaril, triciribine phosphate, fazarabine, tiazofurin, teroxirone, ethiofos, N-(2-hydroxyethyl)-2-nitro-1H-imidazole-1-acetamide, mitoxantrone, acodazole, amonafide, fludarabine phosphate, pibenzimol, didemnin B, merbarone, dihydrolenperone, flavone-8-acetic acid, oxantrazole, ipomeanol, trimetrexate, deoxyspergualin, echinomycin, and dideoxycytidine (see NCI Investigational Drugs, Pharmaceutical Data 1987, NIH Publication No. 88-2141, Revised November 1987) are also preferred.

Radionuclides useful within the present invention include gamma-emitters, positron-emitters, Auger electron-emitters, X-ray emitters and fluorescence-emitters, with beta- or alpha-emitters preferred for therapeutic use. Radionuclides are well-known in the art and include .sup.123 I, .sup.125 I, .sup.130 I, .sup.131 I, .sup.133 I, .sup.135 I, .sup.47 Sc, .sup.72 As, .sup.72 Se, .sup.90 Y, .sup.88 Y, .sup.97 Ru, .sup.100 Pd, .sup.100.sub.Pd, .sup.101m Rh, .sup.119 Sb, .sup.128 Ba, .sup.197 Hg, .sup.211 At, .sup.212 Bi, .sup.212 Pb, .sup.109 Pd, .sup.111 In, .sup.67 Ga, .sup.68 Ga, .sup.67 Cu, .sup.75 Br, .sup.76 Br, .sup.77 Br, .sup.99m Tc, .sup.11 C, .sup.13 N, .sup.15 O and .sup.18 F. Preferred therapeutic radionuclides include .sup.188 Re, .sup.186 Re, .sup.203 Pb, .sup.212 Pb, .sup.212 Bi, .sup.109 Pd, .sup.64 Cu, .sup.67 Cu, .sup.90 Y, .sup.125 I, .sup.131 I, .sup.77 Br, .sup.211 At, .sup.97 Ru, .sup.105 Rh, .sup.198 Au and .sup.199 Ag.

As noted above, with toxin-targeting protein conjugates, the limiting membrane interactive events appear to be the rate of internalization and the rate of translocation. With respect to drug-targeting protein conjugates that are not active at the plasma membrane, internalization of the targeting protein conjugate is required, with release of the drug from endosomic vesicles before the conjugate reaches the lysosome (i.e., is degraded). With drug-targeting protein conjugates that are active at the plasma membrane, internalization is not required, but a strong, prolonged interaction of the drug conjugate at the plasma membrane is important for cytotoxic efficacy. For radionuclide-targeting protein conjugates, only plasma membrane interaction is required, but effective cytotoxicity requires prolonged retention of the conjugate at the target membrane.

The present invention discloses compositions and methods that promote interaction(s) of targeting protein conjugates with various target cells. More specifically, biochemical linkage of a targeting protein conjugate and one or more enhancing moieties capable of promoting membrane interaction (or construction of an analogous recombinant fusion protein) results in a "covalently-linked complex" (CLC) having improved membrane interactive properties. Further, increased cellular interaction(s) of the CLC targeting protein conjugate component may result from secondary binding of a peptide receptor ligand (i.e., enhancing moiety component) to corresponding cell surface receptors. This increased cellular interaction can facilitate subsequent internalization of the CLC in association with the bound cell surface receptor.

Enhancing moieties useful within the present invention may be subdivided into several categories. The first category of enhancing moieties is designated "translocating/internalizing moieties" and includes pH-dependent membrane-binding peptides and pH-independent translocating peptides. The second category is designated "anchoring peptides," and includes membrane soluble peptide sequences and analogs thereof. Anchoring peptides are capable of binding to target cell membranes at physiologic pHs. A third category, "accessory moieties," may be used in conjunction with translocating/internalizing moieties or anchoring peptides to enhance membrane retention, internalization and/or translocation. A fourth category of enhancing moieties includes membrane permeation enhancers, designated "membrane-soluble hydrophobic molecules", such as fatty acids and analogs thereof, bile acids, membrane anesthetics, phospholipids, medium chain glycerides and fusidic acid. A fifth category of enhancing moieties, "intracellular retention moieties", includes molecules that increase intracellular retention of CLCs through binding to specific intracellular structures or organelles after internalization of CLCs. The optimal number of enhancing moieties per targeting protein conjugate may vary depending upon the particular CLC components and target cell involved, but this number may be readily determined by one of ordinary skill in the art of targeting protein conjugates.

In one alternative embodiment, several enhancing moieties that act by different mechanisms may be synthesized together in a single peptide ("combination peptide"). Because each enhancing moiety would constitute a single domain within a longer peptide, two or more domains (i.e., enhancing moieties) may act synergistically, thereby enhancing the effects of the extended peptide.

In yet another alternative embodiment, one or more enhancing moieties may be included in a fusion protein. For generation of a fusion protein that contains an enhancing moiety, a first DNA sequence (corresponding to a targeting protein, a cytotoxic agent or an enhancing moiety) is joined at the DNA level through recombinant DNA technology to a similar or dissimilar second (third, fourth, etc.) DNA sequence. The resultant fused DNA sequences are transcribed and translated into a hybrid fusion protein. When an enhancing moiety is incorporated into a fusion protein, the resultant fusion protein possesses improved membrane interactive properties.

In general, according to the present invention, the targeting protein component of a covalently-linked complex recognizes a binding site at the target cell membrane surface. A primary target cell interaction mediated by the targeting protein component of the CLC is followed by a secondary interaction of the enhancing moiety component with the plasma membrane. This secondary interaction between enhancing moiety and membrane stabilizes the targeting protein at the membrane surface. In instances where an antibody (or antibody fragment) is the targeting protein, interaction of enhancing moiety and target cell membrane may also increase the affinity of an antibody for its antigen.

TRANSLOCATING/INTERNALIZING MOIETIES

The first category of enhancing moiety consists of translocating/internalizing moieties. One class of translocating/internalizing moieties exhibits pH-dependent membrane binding. For a translocating moiety that assumes a helical conformation at an acidic pH, the translocating moiety acquires the property of amphiphilicity, e.g., it has both hydrophobic and hydrophilic interfaces. More specifically, within a pH range of approximately 5.0-5.5, a translocating moiety forms an alpha-helical, amphiphilic structure that facilitates insertion of the moiety into a target membrane. An alpha-helix-inducing acidic pH environment may be found, for example, in the low pH environment present within cellular endosomes.

In aqueous solution at physiological pH, a translocating moiety is mainly unfolded (due to strong charge repulsion among charged amino acid side chains) and is unable to interact with membranes. Within the present invention, it may be advantageous to position amino acid residues within a translocating peptide sequence so that charged amino acid side chains will stack one above the other when the peptide folds into an amphiphilic alpha helix at reduced pH. FIG. 1 represents a helical net display that illustrates an advantageous spatial arrangement of the charged side chains.

Charged amino acids capable of stacking within a translocating peptide sequence include glutamate, aspartate and histidine. A preferred pH-dependent membrane-binding translocating peptide includes a high percentage of helix-forming residues, such as glutamate, methionine, alanine and leucine. In addition, a preferred translocating peptide sequence includes ionizable residues having pKa's within the range of pH 5-7, so that a sufficient uncharged membrane-binding domain will be present within the peptide at pH 5 to allow insertion into the target cell membrane.

A particularly preferred pH-dependent membrane-binding translocating peptide in this regard is aa1-aa2-aa3-EAALA(EALA).sub.4 -EALEALAA-amide, which represents a modification of the peptide sequence of Subbarao et al. (Biochemistry 26: 2964, 1987). Within this peptide sequence, the first amino acid residue (aa1) is preferably a unique residue, such as cysteine or lysine, that facilitates chemical conjugation of the translocating peptide to a targeting protein conjugate. Amino acid residues 2-3 may be selected to modulate the affinity of the translocating peptide for different membranes. For instance, if both residues 2 and 3 are lys or arg, the translocating peptide will have the capacity to bind to membranes or patches of lipids having a negative surface charge. If residues 2-3 are neutral amino acids, the translocating peptide will insert into neutral membranes.

Yet other preferred translocating moieties include peptides of apo-lipoprotein A-1 and B; peptide toxins, such as melittin, bombolittin, delta hemolysin and the pardaxins; antibiotic peptides, such as alamethicin; peptide hormones, such as calcitonin, corticotrophin releasing factor, beta endorphin, glucagon, parathyroid hormone, pancreatic polypeptide; and peptides corresponding to signal sequences of numerous secreted proteins. In addition, exemplary translocating peptides may be modified through attachment of substituents that enhance the alpha-helical character of the translocating peptide at acidic pH.

Yet another class of translocating moieties suitable for use within the present invention include hydrophobic domains that are "hidden" at physiological pH, but are exposed in the low pH environment of the target cell endosome. Upon pH-induced unfolding and exposure of the hydrophobic domain, the moiety binds to lipid bilayers and effects translocation of covalently linked components into the cell cytoplasm. Such translocating moieties may be modeled after sequences identified in Pseudomonas exotoxin A, clathrin, or Diphtheria toxin. In addition, anti-clathrin antibodies or fragments thereof may aid translocation of a targeting protein conjugate/CLC into target cell cytoplasm after binding to a clathrin-coated pit.

Pore-forming proteins or peptides may also serve as translocating moieties herein. Pore forming proteins or peptides may be obtained or derived from C9 complement protein, cytolytic T-cell molecules or NK-cell molecules. These moieties are capable of forming ring-like structures in membranes, thereby allowing transport of attached components through the membrane and into the cell interior.

An example of a modified translocating moiety having translocating activity at acidic pH is fully succinylated melittin. In this example, a peptide (melittin) that normally binds to membranes at physiological pH is converted to a pH-dependent translocating moiety through succinylation of lysines. Upon succinylation, the peptide displays an amphipathic character only at acidic pHs.

Within the present invention, translocating/internalizing peptides may be designed and synthesized to provide enhanced membrane interaction(s). For instance, translocating peptides conforming to the helical net structure depicted in FIG. 1 may be generated. More specifically, in a translocating moiety designed according to FIG. 1, stacked glutamates may occupy positions in the helical net that are designated as "E"; the remaining amino acid residues may consist (entirely or predominantly) of strong neutral helix formers, such as methionine, alanine or leucine.

Insertion of a translocating moiety into a target cell membrane may be enhanced through stabilization of the amphiphilic alpha helix. Helix stabilization may be achieved: (1) by adding repeating "EALA" units to form a longer peptide; (2) by placing an amide at the C-terminus of the peptide, in order to counteract the helical dipole; (3) by polymerizing the peptide; (4) by substituting a neutral helix-former for one or more of the stacked glutamates; or (5) by attaching the peptide to a targeting protein conjugate through use of a longer crosslinking agent, in order to provide sufficient distance between the translocating peptide and the targeting protein conjugate to allow the peptide to contact and interact with the target cell membrane.

The amino acid sequence of a pH-dependent, membrane-binding translocating moiety of the claimed invention may be selected to include all L-amino acids or all D-amino acids having a side chain pKa from 5.0 to 9.0. D-amino acids may be advantageously used to form non-proteolyzable peptides, because the D-amino acid peptide bonds are not metabolized within the cell. Further, translocating moieties of the present invention may include a combination of L- and D-amino acids, wherein D-amino acids are substituted for L-amino acids on either side of a proteolytic cleavage site. Yet another preferred non-cleavable translocating peptide incorporates peptide bond analogs that are not susceptible to proteolytic cleavage by cellular enzymes.

The pH-dependent, membrane-binding translocating moieties of the present invention may be attached singly or may be polymerized. Chemical linkage of one or more translocating moieties (singly or polymerized) may be accomplished either: (a) by direct attachment of a translocating moiety to a targeting protein conjugate; (b) by disulfide, thioether, reduced or non-reduced Schiff base or peptide bond formation; or (c) by use of a peptide spacer. Preferred peptide spacers in this regard correspond to two or more amino acid residues that allow the translocating peptide to assume an alpha-helical conformation independent of its interaction with the targeting protein conjugate, and may allow sufficient distance for translocating peptide access to the cell surface from the peptide attachment site on the targeting protein.

Polymerization of translocating peptides may be accomplished by placing a cysteine residue at each end of a translocating peptide, followed by oxidation using dissolved oxygen or other mild oxidizing agent, such as oxidized glutathione. The average length of the polymerized translocating peptide may be controlled by varying the polymerization reaction conditions.

Mere membrane intercalation of a translocating moiety may be sufficient for translocation of the moiety across endosomic membranes. However, translocation may be improved by attaching to the translocating moiety a substrate for intracellular enzymes (i.e., an accessory peptide, to be discussed in more detail in a following section). It is preferred that an accessory peptide be attached to a portion(s) of the translocating moiety that protrudes through the endosomic membrane to the cytoplasmic face.

An example of an internalizing moiety that is active at physiological pH is the 37-62 fragment of the HIV transactivator (TAT) protein. This protein appears to be divided into four domains (Kuppuswamy et al., Nucl. Acids Res. 17:3551-61, 1989). Purified TAT protein is taken up by cells in tissue culture (Frankel and Pabo, Cell 55:1189-93, 1989), and peptides such as the 37-62 fragment are rapidly taken up by cell in vitro (Green and Loewenstein, Cell 55:1179-88, 1989). The highly basic region mediates internalization and targeting of the internalizing moiety to the nucleus (Ruben et al., J. Virol. 63:1-8, 1989). Peptides or analogs that include a sequence present in the highly basic region, such as CFITKALGISYGRKKRRQRRRPPQGS, are conjugated to targeting protein conjugates to aid in internalization and targeting to the nucleus. Retro-inverso analogs of these peptides may be used to provide CLCs in which the internalizing moiety is resistant to proteolysis. Proteolysis resistance may be important for such peptides, since their potency appears to be diminished by proteolysis in lysosomes.

Another class of translocating/internalizing moieties includes molecules that promote cell surface receptor-mediated internalization. One type of such internalizing moiety is derived from a growth factor. Oncogene studies have determined that several oncogene-encoded proteins are growth factor receptors, and increased expression of these receptors on tumor cell surfaces has been observed. Therefore, peptides that bind to growth factor receptors, and whose sequences are derived from human growth factors, may be useful as internalizing moieties. The natural receptor-binding and internalization characteristics of growth factor-derived internalizing moieties will enhance internalization of CLCs. Further, the presence of growth factor-derived peptides in CLCs may provide additional targeting specificity due to increased numbers of growth factor receptors on tumor target cells.

Preferred growth factor-derived peptides include EGF (epidermal growth factor)-derived peptides, such as CMHIESLDSYTC and CMYIEALDKYAC; TGF-.beta. (transforming growth factor .beta.)-derived peptides; peptides derived from PDGF (platelet-derived growth factor) or PDGF-2; peptides derived from IGF-I (insulin-like growth factor) or IGF-II; and FGF (fibroblast growth factor)-derived peptides.

Another type of internalizing moiety includes non-peptide hormones or hormone analogs, such as the steroid estrogens (estradiol-17.beta. and estriol are exemplary) and anti-estrogens (tamoxifen and clomiphene are exemplary). Anti-estrogens may be preferred internalizing moieties when the CLC target cell population is tumor cells, since anti-estrogens exhibit an anti-proliferative effect, in contrast to the proliferative effect of estradiol on both tumor and normal cells.

Covalent linkage of a translocating/internalizing moiety and a targeting protein conjugate (forming a CLC) may provide enhanced retention of the conjugate (complex) at a target cell plasma membrane upon in vivo administration. In addition, a covalently-linked complex may exhibit more rapid and efficient internalization rates than the corresponding two-component targeting protein conjugate, due to secondary interaction(s) of the enhancing moiety with the plasma membrane. Inclusion of both anchoring (see following section) and translocating/internalizing peptides within a CLC may further facilitate initial binding and enhanced translocation/internalization of the CLC across endosomic membranes into the target cell cytoplasm.

ANCHORING PEPTIDES

A second category of enhancing moiety suitable for use within the present invention consists of anchoring peptides. Typically, anchoring peptides contain membrane soluble peptide sequences which are highly apolar and tend to form alpha helices when inserted into a membrane. When incorporated into a CLC, membrane insertion of the anchoring peptide component may help to secure the targeting protein conjugate component to a target cell membrane, and may further promote internalization of the targeting protein conjugate/CLC.

A model for anchoring peptide interaction with a plasma membrane is the opiate form of beta endorphin. Beta endorphin has one region that is responsible for receptor binding, and another region which can assume an amphiphilic helix (anchoring portion). The amphiphilic helix of beta endorphin is believed to be responsible for an initial membrane interaction, which is followed by diffusion of the hormone through the cell membrane. Diffusion through the membrane allows the receptor-binding region of the molecule to find its appropriate receptor (binding site). In theory, the reverse process may also occur--the receptor binding portion (targeting portion) of beta endorphin interacts with its receptor, followed by alpha-helix formation and membrane insertion of the anchoring portion of the molecule.

Anchoring peptides suitable for use within the present invention may be (i) chemically synthesized; (ii) made by recombinant DNA technology; or (iii) isolated from viral fusion proteins or other proteins. Viral fusion peptides, such as those described by Gallaher (Cell 50: 327-28, 1987), are exemplary of anchoring peptides of the claimed invention. Preferred viral fusion peptide sequences in this regard may be derived from viral proteins of measles virus, respiratory syncytial virus, Sendai vir