WO2000042185A1

WO2000042185A1 - Methods for regulating the stability of recombinant proteins and products useful therein

Info

Publication number: WO2000042185A1
Application number: PCT/US2000/000558
Authority: WO
Inventors: Daniel G. Chain
Original assignee: Mindset Biopharmaceuticals (Usa), Inc.; Mcinnis, Patricia, A.
Priority date: 1999-01-11
Filing date: 2000-01-11
Publication date: 2000-07-20
Also published as: AU2605600A

Abstract

Proteins, such as single-chain antibodies which are stable in their free form, are produced intracellularly by expressing the single-chain antibodies with stabilizing residues. Thus, the proteins are stabilized against proteolysis in vivo.

Description

METHODS FOR REGULATING THE STABILITY OF RECOMBINANT PROTEINS AND PRODUCTS USEFUL THEREIN

Field of the Invention

The present invention is directed to a method for producing recombinant proteins which are stable in vivo and whose stability may be regulated. The present invention is directed not only to the method discussed above, but also to the novel fusion proteins produced by means of the present invention, the DNA encoding such fusion proteins, vectors containing such DNA and suitable for gene therapy, as well as cells in which such vectors have been introduced.

Background of the Invention

In both bacterial and eukaryotic cells, relatively long-lived proteins, whose half-lives are close to or exceed the cell generation time, coexist with proteins whose half- lives are less than one percent of the cell generation time. Rates of intracellular protein degradation are a function of the cell ' s physiological state and appear to be controlled differentially for individual proteins. In particular, damaged and otherwise abnormal proteins are metabolically unstable in vivo . Although the specific functions of selective protein degradation are, in most cases, still unknown, it is clear that many regulatory proteins are extremely short-lived in vivo . Metabolic instability of such proteins allows for rapid adjustment of their intracellular concentrations through regulated changes in rates of their synthesis or degradation.

Intracellular proteolysis functions to eliminate abnormal proteins, maintain amino acid pools in cells affected by stresses, such as starvation, and generate protein fragments that act as hormones, antigens or other effectors. The concentrations of some proteins must vary with time and alterations in the state of a cell, so proteolytic pathways selectively destroy these proteins at the proper time. Conditionally unstable proteins, whether long-lived or short-lived, depending on the state of a cell, are often deployed as components of control circuits. In addition, many proteins are long-lived as components of larger complexes, such as ribosomes and oligomeric proteins, but are metabolically unstable as free subunits.

The ATP-ubiquitin proteasome dependent pathway is the major non-lysosomal proteolytic pathway that functions constitutively to degrade abnormal or damaged proteins.

Ubiquitin-mediated proteolysis can also be regulated and is of widespread importance. Regulated proteolysis by the ubiquitin pathway has been implicated in control of the cell cycle, transcription activation, antigen presentation, cell fate and growth, and in the formation and storage of memory. In this multienzyme pathway, protein substrates marked for degradation by conjugation to ubiquitin (a 76-amino acid residue protein) are hydrolyzed by the 26S proteasome, a 2000 kDa proteolytic complex. The post-translational coupling of ubiquitin to other proteins is catalyzed by a family of ubiquitin-conjugating enzymes and involves formation of an isopeptide bond between the C-terminal Gly residue of ubiquitin and the epsilon-amino group of a Lys residue in an acceptor protein. The features of proteins that confer metabolic instability are called degradation signals, or degrons, cf .

Varshavsky (1996) . Several signals can target specific proteins for degradation. The N-end rule, for example, describes a hierarchy of amino acid residues that confer varying degrees of instability and susceptibility of degradation when positioned at the amino terminus of any given protein. The N-end rule proposes that the in vivo half-life of a protein is a function of the N-terminal residue. Varshavsky has described destabilizing residues on proteins, and has classified them as primary, secondary and tertiary. In the case of primary destabilizing residues, the destabilizing activity requires the physical binding of the N- terminal residue by a protein called N-recognin or E3. In eukaryotes, one binding site of N-recognin binds an N- terminal Arg, Lys or His. The type 2 site binds N-terminal Phe, Leu, Trp, Tyr or lie. Accordingly, the primary destabilizing residues, denoted N-d^p, are subdivided into type 1 (N-d^pl) and type 2 (N-d^p2) residues.

Secondary destabilizing residues, denoted N-d^s, are Asp, Glu and Cys in mammalian cells. The destabilizing activity of the d^s residues requires their accessibility to Arg-tRNA-protein transferase.

N-terminal Asn and Gin residues are tertiary destabilizing residues, denoted N-d'. The destabilizing activity of N-d^fc residues requires accessibility to N- terminal amidohydrolase . A stabilizing N-terminal residue is a default residue, i.e., it is stabilizing because targeting components of an N-end rule pathway do not bind to it or modify it efficiently enough even in the presence of other determinants of an N-degron. Gly, Val and Met are stabilizing residues.

Destabilizing N-terminal residues are present in several natural proteins, but their physiological role in regulating protein turnover is limited due to the fact that most naturally synthesized proteins are co-translationally covalently modified to inhibit proteolysis. Thus, while engineered proteins containing destabilizing N-terminal residues are rapidly degraded, the importance of the N-end rule as a regulatory process is questionable at least in mammalian cells, where there is evidence that proteins degraded by the ubiquitin system are recognized by different signals .

While the N-end rule has been shown to operate in many different systems using engineered protein substances, the physiological role of an N-end rule pathway remains obscure. Recently, mouse and human genes encoding the recognition component of the N-end rule have been identified and cloned, indicating that the pathway plays a role in normal activity of cells. In muscle tissue, moreover, the N- end rule has recently also been shown to be responsible for up to 60% of the ATP-dependent degradation of soluble proteins. In muscle, this process appears to be tightly regulated to maintain the balance between the rates of protein synthesis and degradation that determines its size and functional capacity. Increased proteolysis is the major cause of rapid muscle wasting seen in many pathological states including, for example, sepsis, cancer cachexia, metabolic acidosis, fasting and diabetes. It is not known, however, how substrates containing destabilizing residues at the N-terminus of the coding region are removed from the initiator methionine residue to expose the destabilizing residue. Presumably, this requires the action of a methionine amino-peptidase enzyme or similar cleaving enzyme. Evidently, during changes in catabolism, an endoproteolytic clipping or a novel chemical modification of the N-terminus could be regulated to initiate proteolysis. Recent advances with recombinant antibodies have permitted the manipulation of genes coding for specific antibodies, thus allowing their ectopic expression in a wide variety of non-lymphoid mammalian cells, as well as in plants and other organisms. Intracellular expression and targeting of recombinant antibodies can be used to block biological functions or confer new phenotypic traits, such as viral resistance. The treatment of cancer, denervation atrophy and infectious disease are among several possible therapeutic applications of ectopic antibody expression. Intracellular targets for inactivation by antibodies include retrovirally- infected cells, such as HIV- infected cells, where the targets are the virally-encoded protein; for example, one can use antibodies against structural proteins, such as the envelope glycoproteins and gag proteins. One could use an antibody cocktail (i.e., a mixture of antibodies) to target a variety of viral proteins. Other preferred targets include oncogenes, such as growth factor receptors, receptors, growth factors, and the like. Ectopic antibody expression can be used to test causes of disease by creating transgenic animals which generate such antibodies.

Following the demonstration that antibody secretion can occur efficiently in non-lymphoid cells of neuronal and glial origin, it was suggested that the ectopic expression of secreted antibodies could be used to perturb the function of selected extracellular antigens in the nervous system and in other tissues. This approach has now been taken even further to target antibodies to the cytosol or to other intracellular compartments of cells. This procedure has been described as intracellular immunization. This intracellular immunization can be achieved by incorporating specific localization signals into antibody chains or domains, conferring on the antibodies a new intracellular localization.

The constant domains of antibody chains perform functions that are neither needed nor necessarily exploited for intracellular immunization. Thus, simpler antibody forms, such as Fab or single-chain Fv (ScFv) fragments, can be used, as well as whole antibodies for intracellular targeting.

Antibodies are generally targeted for secretion by the N-terminal hydrophobic leader sequences that are cleaved off as the molecule traverses the endoplasmic reticulum (ER) . Removal of the hydrophobic leader sequence of the antibody chain, or its substitution with a hydrophilic sequence, prevents translocation of the antibody into the ER and restricts it to the cytosol. If a nuclear localization signal or mitochondrial targeting signal is incorporated, then antibodies can also be localized to the nucleus or the mitochondria .

A single-chain antibody or single-chain Fv (ScFv) incorporates the complete antibody binding portion of an antibody in a single polypeptide chain of minimal size, e.g., with an approximate molecular weight of 26,000. In antibodies, the antigen- combining site is part of the Fv region, which is composed of the V_H and V_L variable domains on separate heavy and light chains. Efforts over two decades have indicated that Fv fragments can only rarely be prepared from IgG and IgA antibodies by proteolytic dissection.

Since 1988, single-chain analogues of Fv fragments and their fusion proteins have been reliably generated by genetic engineering methods. The first step in this generation involves obtaining the genes encoding V_H and V_L domains with the desired binding properties. These V genes may be isolated from a specific hybridoma cell line, selected from a combinatorial V-gene library, made by V gene synthesis or generated by phage display. The single-chain Fv is formed by connecting the component V genes with an oligonucleotide that encodes an appropriately-designed linker peptide, such as Gly4-Ser3 (SEQ ID NO:l) . The linker bridges the C- terminus of the first V region and the N-terminus of the second, ordered as either V_H-linker-V_L or V_L-linker-V_H . The scFv binding site can faithfully replicate both the affinity and specificity of this parent antibody combining site.

The half-life of an antibody chain depends on the intracellular compartment in which it is located. Secreted antibodies and antibody domains are generally stable. ScFvs which have been retained in the endoplasmic reticulum are, however, rapidly degraded. Similarly, the half-life of antibodies in the cytosol appears to be rather short, particularly in the absence of the antigen. This instability presents significant limitations in the use of intracellular immunization to block antigen function, particularly when high levels of expression are required over a long period. Varshavsky and coworkers have investigated means for varying the half-lives of a variety of proteins. For example, Baker et al, 5,766,927, disclose inhibitors of protein degradation in living cells using dipeptides . The half-life of intracellular proteins is increased in living cells by contacting the cells with a dipeptide regulator having an amino-terminal amino acid residue which is the same or similar to the amino terminal residue of the intracellular protein. A DNA construct which includes a nucleotide sequence encoding the desired amino acid sequence of the regulator can be introduced into the cell in which inhibition of the degradation of a specific type or class of proteins is desired. It thus acts as a competitive inhibitor for the degron protein which carries degradation of the proteins to be protected.

Varshavsky et al, 5,763,212; Wu et al, 5,538,862; and Wu et al, 5,705,387, disclose a method for regulating degradation of a recombinant protein using a heat activated degron, i.e., a destabilizing N-terminal amino acid residue which becomes a substrate of the N-end rule pathway only at temperatures high enough to result in at least partial unfolding of the protein. The DNA encoding the heat- inducible N-degron module can be linked covalently at its 3' end to the 5' end of a DNA sequence encoding a protein or peptide of interest. When expressed in a cell in which the N-end rule of protein degradation is operative, the heat- inducible N-degron module and any protein or peptide linked to the C-terminus of the heat-inducible N-degron module are rapidly degraded by enzymatic components of the N-end rule proteolytic pathway only after the degron has been activated by heating.

Additionally, degradation of a protein bearing an N- degron can be inhibited by prebinding the protein with a low molecular mass ligand which binds to the protein with high affinity. Thus, another way to regulate degradation of a recombinant protein in a cell is to transform the cell with an expression construct encoding a fusion protein comprising the protein of interest linked at its N-terminus to an N- degron. The cell is then contacted with an inhibitor at a concentration sufficient to achieve a predetermined intracellular concentration. Without the inhibitor, degradation of the expressed fusion protein within the cell would result. When degradation is desired, the administration of inhibitor is stopped.

Baker et al patents 5,683,904; 5,494,818; and 5,212,058 disclose ubiquitin-specific proteases which specifically cleave at the C-terminus of the ubiquitin moiety in a ubiquitin fusion protein. A ubiquitin fusion protein can be a naturally-occurring fusion protein or a fusion protein produced by recombinant DNA technology. Bachmair et al, 5,646,017; 5,496,721; 5,196,321;

5,132,213; and 5,093,242 disclose methods of designing or modifying protein structure at the protein or genetic level to produce the specified amino-termini in vivo or in vi tro .

These methods can be used to alter the metabolic stability and other properties of the protein or to artificially generate authentic amino-termini in proteins produced through artificial means. Genes encoding the proteins can be made to encode an amino acid of the desired class at the amino- terminus so that the expressed protein exhibits a predetermined amino-terminal structure which renders it either metabolically stable or unstable with respect to the N-end rule pathway of proteolytic degradation. Conventional techniques of site-directed mutagenesis for addition or substitution of appropriate codons to the 5' end of an isolated or synthesized gene can be used to provide a desired amino-terminal structure for the encoded protein. So that the protein expressed has the desired amino acid at its amino- terminus, the appropriate codon for a stabilizing amino acid can be inserted or built into the amino-terminus of the protein-encoding sequence.

In the examples given in the patents cited immediately above, the protein having a predetermined amino- terminal amino acid residue is produced by expressing the protein or polypeptide in a host cell as a fusion protein wherein the amino terminus of the protein or polypeptide is fused to ubiquitin, and the fusion protein is specifically cleavable by a protease at the junction of ubiquitin with the amino-terminal amino acid residue of the protein or polypeptide. The fusion protein is contacted with an extract containing a protease which specifically cleaves the ubiquitin fusion protein at the junction of ubiquitin and the amino-terminal amino acid residue of the protein or polypeptide. In all of the above five patents, a fusion protein including the desired amino acid terminal sequence fused to ubiquitin is produced, and then a protease which cleaves at the ubiquitin site is used to cleave ubiquitin from the protein of interest .

Other methods for blocking the degradation of protein by the ubiquitin pathway include the use of specific inhibitors of components of the ubiquitin pathway. For example, lactacystin is an irreversible inhibitor that reacts selectively with active site threonine residues in proteasomes to block the bulk of protein degradation in mammalian cells. Carbobenzoxyl-leucinyl-leucinyl-leucinyl is a membrane permeant peptide aldehyde that binds reversibly to multiple active sites in proteasomes. A major impediment to the development of effective gene therapy using intracellular antibody expression is the inability to achieve a high level of expression in recipient cells. It would be desirable to have a method which can be used to achieve a high level of expression of an inhibitory antibody. In other cases, it would be desirable to be able to modulate the level of antibody expression and, in particular, to remove the antibody after it has bound to its target antigen.

Citation of any document herein is not intended as an admission that such document is pertinent prior art, or considered material to the patentability of any claim of the present application. Any statement as to content or a date of any document is based on the information available to applicant at the time of filing and does not constitute an admission as to the correctness of such a statement.

Summary of the Invention

It is an object of the present invention to overcome the aforementioned deficiencies in the prior art.

It is another object of the present invention to express intracellularly in a stable form, antibodies which are unstable in their free form. It is yet another object of the present invention to provide a method for modulating the degradability of recombinant proteins intended to be intracellularly expressed. It is a further object of the present invention to provide a method for modulating the degradability of recombinant antibody proteins intended to be intracellularly expressed.

It is still another object of the present invention to express intracellularly a single-chain antibody which includes stabilizing residues at the N-terminal sequence.

It is yet a further object of the present invention to provide single chain antibody fusion proteins having a single chain antibody linked by means of a proteolysis- sensitive linker to an N-terminal stabilizing peptide or amino acid residue whereby the protein is stable prior to proteolysis but becomes unstable upon proteolysis by a specified restriction protease.

It is still another object of the present invention to increase the concentration of recombinant single-chain antibodies expressed intracellularly by inhibiting the proteolytic pathway with specific inhibitors .

It is yet another object of the present invention to regulate the expression of single-chain antibodies by reversibly inhibiting the proteolytic pathway with specific inhibitors .

According to the present invention, artificially- engineered proteins, such as single-chain antibodies which are unstable in their free form, are produced intracellularly by expressing the single-chain antibodies with stabilizing residues. Thus, the proteins are stabilized against proteolysis . |

The present invention is intended to solve the problem of the instability of ectopic antibodies in cells. In the course of an investigation as to whether scFv's are intrinsically unstable, a search of protein sequence databases with immunoglobulin heavy and light variable chain sequences was conducted. This search surprisingly showed that the vast majority of such sequences possess destabilizing residues at the start of the variable region. The residues become destabilizing when they are exposed as the free N-terminal residues. Presumably, the initiator methionine residues of the genetically engineered antibodies are somehow removed intracellularly, perhaps by the action of a methionine aminopeptidase enzyme or similar cleaving enzymes. As the first residues of antibody heavy and light variable chain sequences after the initiator methionine are predominantly destabilizing residues, this means that single- chain antibodies have a special predisposition to proteolysis by the N-end rule once the initiator methionine has been metabolically removed. This finding has practical significance and can explain the difficulty of high level expression of intracellular antibodies. As the N-end rule degradation is linked to the ubiquitin pathway, this observation is particularly significant in tissues, such as the brain, which have high levels of ubiquitin activity or under conditions in which the ubiquitin pathways are up- regulated as, for example, in cancer cells and during starvation, denervation atrophy and activation of the immune system.

As noted above, the in vivo half-life of an intracellular protein is a function of its N-terminal amino acid residue. The present invention provides a technique for generating proteins or peptides with a specified N-terminus in vivo or in vitro .

The nature of the amino acid exposed at the N- terminus of an intracellular protein has been shown to be one crucial determinant that specifies whether a protein will be long- or short-lived in vivo, and particularly intracellularly. As noted above, individual amino acids can be categorized as either stabilizing or destabilizing amino acids with respect to the half-life they confer upon a protein when exposed at the protein's N-terminus. Destabilizing amino acid residues confer short half-lives, down to a few minutes for some of the destabilizing amino acids, while stabilizing amino acid residues confer long half-lives of many hours.

Based upon the N-end rule, the N-terminus of a single-chain antibody can be designed to increase the intracellular half-life of the single-chain antibody when it is expressed in the cell. Genes encoding single-chain antibodies are made to encode an amino acid of the desired class at the N-terminus, after the initiation methionine, so that the expressed single-chain antibody exhibits a predetermined amino-terminal structure which renders it metabolically stable with respect to the N-end rule of proteolytic degradation once the methionine is metabolically removed. Alternatively, the single chain antibody can also be stabilized by fusing it to a peptide sequence containing a stabilizing residue at the N-terminus, after the initiation methionine. The stabilizing fusion protein or peptide sequence, referred to as a "stabilon", is attached upstream of the native destabilizing residue at the N-terminus of the antibody region. The stabilon is preferably linked to the antibody through a protease-sensitive linker region that can be cleaved by specific restriction proteases. Restriction proteases have well defined recognition signals that are cleaved within the target substrates. Restriction proteases have been used in a number of ways to cleave fusion proteins, for example, in vector targeting. In this case a retroviral vector is fused with a ligand such as Epidermal Growth Factor that binds to receptors on human cells. The cells are not infected until the linker is cleaved by the restriction protease. Non-limiting examples of restriction proteases are Factor Xa (Ile-Glu-Gly-Arg) (SEQ ID NO:35); Enterokinase (Aps4-lys) (SEQ ID NO: 36) ; and matrix metalloproteases MMP (Pro-Leu-Gly-Leu-Trp-Ala) (SEQ ID NO: 37) . The restriction protease can be an endogenous protein in the cell in which the single-chain antibody is expressed or the product of an inducible gene that is co-transfected with the stabilon fusion antibody protein under the control of an inducible promoter. The removal of the stabilon moiety by the restriction protease renders the single chain antibody susceptible to proteolysis by the N-end rule pathway. The amino acids which produce stable N-terminals are glycine, methionine, valine and serine . As methionine may be metabolically removed intracellularly, it is preferred to use a stable N-terminal residue other than methionine. While the stabilon may be a single stable N-terminal amino acid residue, such as glycine, valine or serine, it is preferably a peptide sequence consisting of two or more stable residues, preferably no more than 10. While they may include glycine, valine, serine and methionine in any order, it is preferred that the peptide comprise multiple repeats of the same stable amino acid residue such as, for example, Gly₁₀ (SEQ ID NO:32), Val₄ (SEQ ID NO:33), Ser₇ (SEQ ID NO:34), etc.

Another way to modulate the half-life of an intracellular single chain antibody besides use of a proteolysis-sensitive linker is to use a stabilon comprising multiple repeats of methionine residue. Thus, for example, a single chain antibody whose N-terminal comprises ten methionine residues, rather than one, would be expected to have a half-life ten times as long as the half-life of a single chain antibody having only the initiator methionine. This is because the mechanism by which the initiator methionine is metabolically removed must be repeated ten- times before the destabilizing residue becomes freely available at the N-terminal.

While the present invention is preferably directed toward improving the stability of intracellularly expressed single chain antibodies, the concept of using a stabilon connected by a proteolysis-sensitive linker to a destabilizing sequence, so that degradation can be initiated upon induction of expression of a restriction protease by means of an inducible promoter, has more general applicability. It may be applicable to any protein, whether or not it is initially unstable. Thus, for example, if it is desired to intracellularly express a desired recombinant protein and subsequently have it removed from the cell at will, one can engineer the protein so as to fuse it to a destabilizing residue or sequence connected to a stabilon by means of a proteolysis-sensitive linker. As a stabilon is present at the N-terminus, the protein will be stable. Naturally, one would have to verify that the biological activity of the protein of interest is not affected by adding such a fusion sequence to the N-terminus thereof. When it is desired to have the protein removed, the inducible promoter of the restriction protease may be activated or an endogenous inducible restriction protease may be activated such that the stabilon will be removed, leaving a destabilizing sequence fused to the N-terminus of the protein; such a destabilizing sequence will cause the protein to be degraded by the N-end rule pathway.

Inducible promoters are well-known and used predominantly in transgenic animal technology and in regulating intracellular expression for in vi tro type of experiments. One non-limiting example is a tetracycline inducible promoter, such as the tetracycline responsive promoters taught in U.S. Patent 5,650,298. When tetracycline is administered to the patient in whom the genes of the present invention have been introduced by gene therapy, the tetracycline inducible promoter will be activated, releasing the restriction protease and commencing a chain of reactions which ultimately causes elimination of the genetically engineered protein.

Gene therapy can be used to treat illnesses and conditions which have a genetic or metabolic cause or which result from infection. In many of these conditions, cells are either deficient in a protein or produce a dysfunctional protein. Gene therapy treats these conditions by introducing into the appropriate cell DNA coding for the normal gene product or DNA coding for a factor that can neutralize or block the activity of an abnormal functioning molecule. Gene therapy can be affected by receptor-mediated gene delivery transkaryotic implantation, viral shuttle vectors, such as retroviral gene transfer, etc.

Viruses have been used to deliver DNA in gene therapy. Among the types of gene therapy in which viruses have been used for transfer are HSV-1 vector mediated transfer of BDNF into cerebellar granule cells, Alonso et al

(1996) ; gene delivery to the heart and to cardiac myocytes and vascular smooth muscle cells using herpes virus vectors, Coffin et al (1996) ; neurotropic virus for treatment of Parkinson's Disease, Fink et al (1997); and expression of the lacZ reporter gene in the rat basal forebrain, hippocampus and nigrostraital pathway using a non-replicating herpes simplex vector, Maidment et al (1996) .

Haynes et al, (1996) reported on nucleic acid immunization involving the direct in vivo administration of antigen-inducing plasmid DNA molecules which produce microbial antigens at the site of DNA delivery. Krisky et al

(1997) disclose that herpes simplex virus type 1 carries a large number of viral functions which can be replaced by foreign genes to create a vector for gene therapy applications .

Other workers have produced antibodies in cells in vi tro by transforming human cells to produce antibodies, cf .

Della-Favera et al, 5,223,417; Zanette et al , 5,658,762; Takahashi et al, 5,032,511; Marasco et al, PCT WO94/02610; and Risser et al, 5,244,656.

According to the present invention, a DNA sequence encoding the single chain antibody, or other protein of interest, is ligated to a DNA sequence encoding the amino acid sequence of the stabilon which provides N-terminal stability, and also optionally to DNA encoding the proteolysis-sensitive linker between the DNA encoding the stabilon and the DNA encoding the protein of interest. This DNA sequence is then operably linked to a promoter which will permit expression of the amino acid sequence encoded by the

DNA intracellularly into the cell or cells of interest. This DNA sequence is delivered to the cells of interest either by direct injection or, more preferably, by the use of a conventional vector which is selected so as to carry the DNA selectively into the cells of interest. Thereafter, the protein with the stabilizing N-terminal sequence is expressed intracellularly in the targeted cells. If the protein of interest is a single chain antibody, or an antibody binding region or other ligand binding polypeptide, the expressed protein will then bind to the target antigen or ligand within the cell. The protein bearing the N-terminal stabilon is deemed a "protected" protein for purposes of the present invention. The N-termini of the antibodies are stabilized by adding one or more of glycine, methionine, serine and valine. If the engineered protein which is delivered to the cell as discussed above contains a proteolysis-sensitive linker region, the vector carrying the engineered DNA is further engineered to carry a DNA segment encoding a second protein along with an associated promoter. The second protein will be the restriction protease for which the proteolysis-sensitive linker is designed. The promoter will be an inducible promoter which will cause the restriction protease to be expressed only upon being subjected to the predetermined induction signal. While this second DNA sequence is preferably located on the same vector, it may also be introduced on a separate vector, either simultaneously with the introduction of the initial vector or separately therefrom. Alternatively, the promoter for the restriction protease may be a constitutive promoter and the DNA encoding the restriction protease enzyme and the promoter DNA operably linked thereto may be introduced at the time that it is desired to remove the initially introduced protein.

An advantage of causing the single chain antibody to be removed at will by means of the N-end rule degradation pathway is that it is expected that not only will the antibody be degraded but also the antigen to which it has bound. Thus, it is expected that not only will free antibody be degraded by this pathway, but entire immunoconjugates, thus causing removal of predetermined antigens or proteins bearing predetermined antigens which one wishes to remove from the cellular milieu.

In one embodiment, the "antibody gene" of the antibody cassettes utilizes a cDNA encoding heavy chain variable (V_H) and light chain variable (V_L) domains of an antibody which can be connected at the DNA level by an appropriate oligonucleotide to bridge the two variable domains, which on translation produces a single polypeptide (referred to as a single-chain variable fragment (sFv) ) capable of binding to a target, such as a protein. The antibody gene does not encode an operable secretory sequence, and, thus, the expressed antibody remains within the cell. In certain embodiments, a nucleotide sequence encoding an intracellular localization amino acid sequence may be used.

While this method can be used to introduce only one protected antibody into a cell, it is also possible to introduce a combination of antibodies or target a variety of viral target proteins using this method. Thus, for example, antibodies which are useful in treating Alzheimer's Disease may be administered in this manner. Other antibodies can be introduced into a cell for intracellular expression also. Such antibodies may be those against structural proteins, such as envelope glycoprotein and gag protein, against tat, rev, nef, vpu and/or vpx regulatory proteins. Other targets include oncogenes, such as trans-membrane growth factor receptors, receptors, growth factors, membrane associated guanine nucleotide binding proteins, etc.

Detailed Description of the Invention A DNA sequence containing nucleotides coding for a protein or peptide of interest, as well as nucleotides which code for an amino acid sequence at the N-terminus of the protein or peptide selected to increase the half-life of the protein or peptide are operably linked to a promoter that will permit expression of the protein or peptide in the cell(s) of interest (protein or peptide cassette). The protein or peptide with the protected N-terminus is expressed intracellularly. To produce in vivo a single-chain antibody, a DNA sequence containing a sufficient number of nucleotides to code for the portion of an antibody capable of binding to a target, as well as nucleotides which code for an amino acid sequence at the N- terminus of the antibody portion selected to increase the half-life of the antibody portion, are operably linked to a promoter that will permit expression of the antibody in the cell(s) of interest (antibody cassette), and the cassette is delivered to a cell. Thereafter, the antibody with the protected N-terminus is expressed intracellularly and binds to the target, thereby disrupting the target from its normal actions.

Thus, one aspect the present invention provides a method of targeting a particular molecule (target molecule) , preferably a receptor site or an undesired protein. This method comprises the intracellular expression of a single- chain antibody which has been stabilized and which is capable of binding to the specific target (e.g., a target protein), wherein the vector encoding the antibody preferably does not contain sequences coding for its secretion. These antibodies bind the target intracellularly.

Alternatively, the single-chain antibody can be stabilized against proteolysis by inhibitors of the proteolytic pathway. While any substance that blocks the recognition of the substrate of other components of the ubiquitin pathway can be used, which substances are well known to those skilled in the art, the most common substances are those which block the proteasome, thus preventing the degradative final step of the ubiquitin pathway.

The preferred embodiment of the present invention is directed to a method and a vector for modulating the degradability of intracellularly targeted antibodies and fragments thereof, particularly single-chain antibodies, as described below in detail. However, the present invention is also generally applicable to a method and a vector for modulating the degradability of any intracellularly expressed protein or peptide as would be well appreciated by those in the art. For instance, a protein or peptide, which includes an antibody or a fragment thereof, that is unstable intracellularly with respect to the N-end rule of proteolytic degradation as reported by Varshavsky (1996) can be stabilized by inserting into the gene encoding the protein or peptide a nucleotide sequence encoding a stabilon sequence immediately after the N-terminal ATC encoding the initiation methionine residue. In this way, even if the N-terminal initiation methionine residue is removed, the presence of a stabilon sequence at the new N-terminus of the protein or peptide stabilizes the protein or peptide with respect to the N-end rule of proteolytic degradation.

The term "stabilon" or "stabilon sequence" is intended to mean an amino acid sequence of one or more amino acid residues, preferably two to ten amino acid residues, which are stabilizing N-terminal residues, such as Gly, Val, Ser and Met, with respect to the N-end rule of proteolytic degradation according to Varshavsky (1996) . When the stabilon is a single amino acid residue, it is preferably Gly, Val or Ser. For a stabilon sequence of two or more amino acid residues, the stabilizing residues Gly, Val, Ser and Met can be combined in any order, although it is preferred that the stabilon sequence include multiple repeats of the same stable amino acid residue, i.e., Gly₁₀ (SEQ ID NO:32), Val₄ (SEQ ID NO:33), Ser₇ (SEQ ID NO:34), etc. Multiple repeats of Met residues can also be used as the stabilon sequence. While it may be susceptible to successive removal of the N-terminal Met residue by the action of a methionine amino acid peptidase or a similar enzyme, it will still prolong the stability of the protein. Accordingly, multiple N-terminal repeats of Met residues are stabilizing to the extent that Met residues are still available to protect the N-terminus, and this is dependent on the rate at which single Met residues are successively and enzymatically removed from the N-terminus. In this regard, it may be advantageous to provide more than ten repeats of Met residues at the N-terminus.

The use of a stabilon is directed to modulating the degradability of an intracellularly expressed protein or peptide to improve its intracellular stability. However, there are instances where it is desirable at some point to remove the protein or peptide. For example, when a protein or peptide is an antibody or fragment thereof and it has had sufficient time to bind a target antigen intracellularly, then it may be desirable to degrade the antibody and thereby degrade the antigen as part of an antibody/antigen complex once the antibody has served its purpose. This is important when the antibody cannot be made inhibitory to by binding to the antigen as is often the case since the active sites of a molecule are often not accessible to the surface. That is, the antibody does not necessarily achieve its function simply by binding to the antigen. Because the active site of the antigen is usually buried and, therefore, is not immunogenic, most antibodies are not inhibitory. Degradation of the antibody/antigen complex is another way to block function of the antigen when the antibody itself is not inhibitory.

According to an embodiment the present invention, modulating the degradability of a protein or peptide includes the ability to later destabilize and degrade a previously stabilized or naturally stable protein or peptide. Thus, for example, if it is desired to intracellularly express a desired recombinant protein and subsequently have it removed from the cell at will, the protein or peptide can be engineered so as to fuse it to a destabilizing sequence. In the case where the protein or peptide is naturally stable, a destabilizing sequence can be joined to an N-terminal stabilon sequence by means of a protease-sensitive linker region. Similarly, if the protein or peptide naturally contains the destabilizing N-terminal sequence, then the protein's or peptide ' s own destabilizing sequence can be joined at the N-terminus to a stabilon sequence by means of protease-sensitive linker region. Such a protease-sensitive linker region can be cleaved by an appropriate restriction protease whose expression can be placed under the control of an inducible promoter. When it is desired to have the protein removed, the inducible promoter of the restriction protease or an endogenous inducible restriction protease may be activated so that the stabilon is removed by cleavage at the protease-sensitive linker region, thereby exposing the destabilizing sequence at the N-terminus where it will cause the protein to be degraded by the N-end rule pathway. The protease-sensitive linker region is intended to be a sequence which is recognized and cleaved by a restriction protease whereby the protease-sensitive linker region is removed to leave a destabilizing sequence at the N-terminal end of the cleavage site. Suitable restriction proteases and their recognition and cleavage sites/sequences that can be used according to the present invention are well known in the art. Non-limiting examples of restriction proteases are Factor Xa (Ile-Glu-Gly-Arg) (SEQ ID NO: 35) ; Enterokinase (Aps4-lys) (SEQ ID NO: 36) ; and matrix metalloproteases MMP (Pro-Leu-Gly- Leu-Trp-Ala) (SEQ ID NO: 37) .

As will be further appreciated by those of skill in the art, a protein or peptide that is otherwise naturally stable intracellularly can be altered by effectively inserting a stabilon sequence, joined to a destabilizing sequence by a protease-sensitive linker region, between the N-terminal initiation Met residue and the rest of the protein or peptide as a unit by recombinant DNA methodology. Such an alteration would provide the ability to remove the stabilon by cleavage at the protease-sensitive linker region with an appropriate restriction protease and expose an artificial destabilizing sequence, which is recognized by the N-end rule proteolytic pathway and positioned N-terminal to an otherwise stabilizing sequence of the naturally stable protein or peptide . The term "destabilizing sequence" is intended to mean one or more amino acid residues selected from Arg, Lys, His, Asn, Gin, Asp, Glu and Cys, preferably Arg, Lys and His, according to the destabilizing residues reported by Varshavsky (1996) .

The recombinant DNA molecules, where a promoter is operably linked to a gene encoding a desired protein or peptide for intracellular expression is altered so as to insert a sequence encoding (1) a stabilon, (2) a stabilon joined to a protease-sensitive linker region, or (3) a stabilon joined to a destabilizing sequence through a protease-sensitive linker region, can be readily constructed in view of the high level of skill of those in the art with guidance from a wealth of standard reference texts on recombinant DNA technology, such as Ausubel et al (1987- 1998) , Maniatis et al (1989) , Watson et al (1987) , Darnell et al (1985) , Lewin (1985) , Old et al (1981) , Berger et al (1987) , etc.

A DNA molecule is said to be "capable of expressing" a protein or peptide if it contains nucleotide sequences which contain transcriptional and translational regulatory information, and such sequences are "operably linked" to nucleotide sequences which encode the protein or peptide. An operable linkage is a linkage in which the regulatory DNA sequences and the DNA sequence sought to be expressed are connected in such a way as to permit gene expression. The regulatory regions needed for gene expression in general include a promoter region, as well as the DNA sequences which, when transcribed into RNA, signal the initiation of protein synthesis. These regions normally include those 5'-non-coding sequences involved with initiation of transcription and translation. A promoter region is, thus, operably linked to a DNA sequence if the promoter is capable of effecting transcription of that DNA sequences. As used herein, a "promoter sequence" is the sequence of the promoter which is found on the DNA and is transcribed by the RNA polymerase . Suitable promoters, which are intended to include inducible promoters, include, but are not limited to, the metallothionine promoter (Brinster et al, 1982) , the heat shock protein promoter, elastase I gene control region, which is active in pancreatic acinar cells (Swift et al, 1984; Ornitz et al, 1986; MacDonald, 1987), the insulin gene control region, which is active in pancreatic beta cells (Hanahan, 1985) , the immunoglobulin gene control region, which is active in lymphoid cells (Grosschedl et al, 1984; Adames et al, 1985; Alexander et al, 1987) , the mouse mammary tumor virus control region, which is active in testicular, breast, lymphoid and mast cells (Leder et al, 1986), the albumin gene control region, which is active in the liver (Pinkert et al, 1987) , the alpha-fetoprotein gene control region, which is active in the liver (Krumlauf et al, 1985; Hammer et al, 1987) , the alpha 1-antitrypsin gene control region, which is active in the liver (Kelsey et al, 1987) , the beta-globin gene control region, which is active in myeloid cells (Mogram et al, 1985; Kollias et al, 1986), the myelin basic protein gene control region, which is active in oligodendrocyte cells in the brain (Readhead et al, 1987) ; the myosin light chain-2 gene control region, which is active in skeletal muscle (Sani, 1985) , and the gonadotropic releasing hormone gene control region, which is active in the hypothalamus (Mason et al, 1986) .

Vectors for delivery of a gene encoding a desired intracellularly expressed protein or peptide operably linked to a promoter and a second gene which encodes a restriction protease under the control of an operably linked inducible promoter, either together on the same vector or separately, are well known in the art. These vectors include mammalian expression vectors, such as a herpes vector, an adenovirus vector, a pox vector, a retroviral vector and other vectors taught above as vectors for gene therapy.

When the desired intracellularly expressed protein or peptide is provided with a stabilon and a protease- sensitive linker region, the vector carrying a gene encoding such a desired protein or peptide is preferably also engineered to carry a gene encoding a restriction protease for which the protease-sensitive linker region is designed. Alternatively, the gene encoding the restriction protease may be located on a separate vector which may be introduced into the host cell simultaneously with the vector carrying the gene encoding the desired intracellularly expressed protein or peptide or separately therefrom. In order that the gene encoding the restriction protease is controlled so that the expression of the restriction protease occurs only at an appropriate time, an inducible promoter operably linked to the protease gene is used. Non-limiting examples of such an inducible promoter include mouse mammary tumor virus promoter (induced by dexamethasone) , metallothionine promoter (induced by zinc) , yeast gal4 promoter (induced by galactose) , heat shock gene promoter, steroid hormone-responsive promoter, and tetracycline-responsive promoters. Besides the prokaryotic tetracycline-responsive promoters that function in eukaryotes, as taught, for example, in U.S. Patent 5,650,298, the lac repressor/operator inducer system of E. coli has been reported to function in eukaryotic cells.

Three approaches using the lac repressor/operator inducer system are known in the art:

(i) prevention of transcription initiation by properly placed lac operators at promoter sites (Hu et al, 1987; Brown et al, 1987; Figge et al, 1988; Fuerst et al, 1989; Deutschle et al, 1989) ; (ii) blockage of transcribing RNA polymerase II during elongation by a lac repressor/operator complex (lac R/O) (Deutschle et al, 1990); and

(iii) activation of a promoter responsive to a fusion between lacR and the activating domain of virion protein 16 (VP16) of herpes simplex virus (HSV) (Labow et al, 1990) ; Bairn et al, 1991) .

A suitable inducible promoter can be selected by those of skill in the art, taking into account that the inducer or inducing event should not be significantly toxic or harmful to the host in which it is administered.

In one preferred embodiment, the "antibody gene" of the antibody cassette uses a cDNA encoding heavy chain variable and light chain variable domains of an antibody which are connected at the DNA level by an appropriate oligonucleotide as a bridge of the two variable domains, which on translation form a single polypeptide (e.g., a single-chain variable fragment (sFv) ) which is capable of binding to a target, such as a protein. The single-chain variable fragment is substituted at the N-terminus in order to increase the half-life of the single-chain variable fragment once the fragment has been formed in the cell . The term "antibody" refers to at least that portion of an immunoglobulin capable of selectively binding to a target, such as an antigen or other protein. The term includes single-chain antibodies. The antibody is expressed from a DNA sequence which contains a sufficient number of nucleotides coding for the portion of antibody capable of binding to the target, as well as an amino acid sequence which extends the half-life of the antibody, i.e., which protects the antibody. This DNA sequence is also referred to as the antibody gene. The antibody gene is operably linked to a promoter to permit expression of the protected antibody in the cells of interest. Promoters are well known in the art and can readily be selected depending upon the cell type to be targeted. For example, when the function of a target protein is a result of its overexpression, by "turning on the promoter" one can selectively cause expression of the antibody. The entire sequence of antibody gene and promoter is also referred to as an antibody cassette. The antibody cassette is delivered to the cell using an appropriate vector to carry the cassette into the cell(s) of interest. Once the cassette is introduced into the cell, the protected antibody is expressed in the cell. The expressed protected antibody can then bind to the target antigen. Because the antibody is protected, there is sufficient time to bind to the target antigen before proteolysis destroys the antibody.

In a similar manner, proteins and peptides that otherwise would be immediately subjected to proteolysis in the cytosol can be expressed in protected form so that they are available for a sufficient length of time to bind their target antigen.

Antibodies used in the present invention can be antibodies to almost any type of biological molecule. For example, the antibodies used can be antibodies to intermediate metabolites, sugars, lipids, autocoids and hormones, as well as macromolecules, such as complex carbohydrates, phospholipids, nucleic acids, such as RNA and DNA, and proteins. One skilled in the art can readily, without undue experimentation, generate antibodies that specifically bind to small molecules and macromolecules. The preferred target molecules for antibodies used in the present invention include proteins, RNA, DNA and haptens . More preferably, the targets are proteins, RNA and DNA. Still more preferably, the target is a protein. It should be understood, however, that the present invention is not directed to such antibodies per se, but to means for protecting them against prompt intracellular degradation, and the stabilized antibodies so produced.

The compositions of the present invention include a recombinant DNA molecule containing a gene for the antibody or molecule of interest along with a nucleotide sequence for the amino acid at the N-terminus which prolongs the half-life of the antibody or molecule of interest in vivo, in association with a means of gene delivery. Such an antibody may preferably be an antibody such as that disclosed in WO 98-44955. Such antibodies are useful, for example, as medicaments for treating diseases or conditions of the central nervous system, such as Alzheimer's Disease.

The present invention is particularly useful for producing protected single-chain antibodies. These single- chain antibodies can be single-chain composite polypeptides having end-specific peptide binding capability and comprising a pair of amino acid sequences homologous or analogous to the variable regions of an immunoglobulin light and heavy chain (linked V_H-V_L or single-chain Fv) , and including the amino acid(s) at the N-terminus which impart stability to the antibodies in vivo. Both V_H and V_L may copy natural monoclonal antibody sequences, or none or both of the chains may comprise a CDR-FR construct of the type described in U.S. Patent 5,091,513. The separate polypeptides analogous to the variable regions of the light and heavy chains are held together by a peptide linker. Methods of production of such single-chain antibodies, e.g., single-chain FV (ScFv), particularly where the DNA encoding the polypeptide structure of the V_H and V_L chains are characterized or can be readily ascertained by sequence analysis, may be accomplished in accordance with the methods described, for example, in U.S. Patents 4,946,778; 5,091,513; and 5,096,815.

In constructing the recombinant gene encoding the molecule of interest, nucleotide sequences are included in the gene which ensure that the antibody produced will have the entire N-terminus.

For example, the antigen binding or variable region is formed by the interaction of the variable heavy and variable light domains at the amino termini of the chains. The smallest fragment containing a complete binding site is referred to as Fv and is a heterodimer of the variable light and variable heavy domains. However, it is possible to obtain binding without a complete binding site. Antigen binding can be obtained using only a heavy chain binding domain. A gene coding for such an antibody can be used as long as the fragment retains sufficient binding ability as compared to the parent antibody.

The host ' s immune system is used to prepare the antibody which will bind to a specific molecule, such as a target protein, by standard immunological techniques. For example, the antibody is prepared using the protein, or an immunogenic fragment thereof, or a peptide chemically synthesized based upon such protein. Any of these sequences can optionally be conjugated to keyhole limpet hemocyanin and used to raise an antibody in another mammal, such as mice, rabbits, rats and hamsters. Thereafter, the animals are sacrificed, and their spleens are obtained. Monoclonal antibodies are produced using conventional fusion techniques for making hybridoma cells, typically by fusing an antibody- producing cell with an immortal cell line, such as a myeloma cell, to produce the hybrid cell.

Antibodies are also prepared by in vi tro immunization, such as using a culture of spleen cells, injecting an antigen, and then screening for an antibody produced to the antigen. For in vitro immunization, spleen cells are harvested and incubated in medium together with the desired antigen at a concentration typically of about 1 μg/mL. Thereafter, an adjuvant, depending upon the results of the filter immunoplaque assay, is added to the cell culture. The spleen cells are incubated with the antigen for about three to five days and then harvested.

Cell suspensions of the in vi tro immunized mouse spleen cells are then incubated. The antibodies produced are detected by using a label for the antibodies, such as a labelled second antibody, such as rabbit anti-mouse IgA, IgG or IgM. In determining the isotype of the secreted antibodies, biotinylated rabbit anti-mouse heavy chain specific antibodies are used, followed by a reagent, such as horseradish peroxidase-avidin.

The DNA encoding any antibody or active fragment that is specific for the antigen of interest can be used in generating the recombinant antibody-encoding DNA molecules for use according to the present invention. For instance, the C-terminal end-specific monoclonal antibodies disclosed in WO 96/25435 may be used to obtain the recombinant antibody-encoding DNA molecules according to the present invention. mRNA may be isolated from hybridomas producing monoclonal antibodies determined to be selective for the antigen of interest. From the isolated hybridoma mRNA, cDNA is synthesized, and the nucleotide sequence encoding the variable domains of the monoclonal antibody may then be cloned using the polymerase chain reaction (PCR) with primers based on the conserved sequences at each end of the nucleotide sequences encoding the V domains of immunoglobulin heavy chain and light chain. The presence of restriction sites incorporated into the sequence of the PCR primers facilitates the cloning of PCR-amplified products encoding the variable region of the appropriate chain.

A recombinant gene encoding a recombinant single- chain Fv antibody molecule is constructed, for example, by joining nucleotide sequences encoding the V_H and V_L domains with a nucleotide sequence encoding a peptide interchain linker (Biocca et al, 1993; Duan et al, 1994; Mhashilkar et al, 1995; Marasco et al, 1993; Richardson et al, 1995; U.S. Patents 4,946,7778; 5,091,513; 5,086,815) or by inserting the variable domain-encoding nucleotide sequences to replace the corresponding sequences encoding the variable domain in a human immunoglobulin gene to thereby encode for a recombinant chimeric antibody (Orlandi et al, 1989; U.S. Patent 4,816,567) .

Alternatively, a known antibody to the target protein can be used, the most preferred example being antisenilin (see WO 98-44955) . Thereafter, a gene to at least the antigen binding portion of the antibody is synthesized. The gene preferably does not contain the normal signal peptide sequences. In preferred embodiments, the gene also encodes an intracellular localization sequence, such as one for the endoplasmic «t:eticulum, nucleus, nucleolar, etc. To retain antibodies in a specific location, a localization sequence, such as the KDEL sequence, is used. In some embodiments the antibody gene preferably also does not encode functional secretory sequences.

Variable light and variable heavy domain genes can be constructed using any of the antibodies obtained as above. Alternatively, commercially available variable heavy and variable light chain domain libraries can be used. In referring to the antibody gene, this term can encompass genes for both the heavy chain and light chain regions of the antibody gene. In addition, the gene is operably linked to at least one promoter which results in its expression. Any well-known promoter that permits expression in mammalian cells can be used. These promoters are well known in the art and include promoters, such as CMV, a viral LTR, such as the rous sarcoma virus LTR, HIV-LTR, LTLV-1LTR, the SV40 early promoter, E. coli lac UV5 promoter, and the herpes simplex thymidine kinase (tk) promoter. The combination of the antibody DNA sequence and the promoter is referred to as the antibody cassette. In vi tro production of antibodies is also possible using Phage Display technology. The production of recombinant antibodies is much faster compared to conventional antibody production and they can be generated against an enormous number of antigens. In contrast, in the conventional method, many antigens prove to be non- immunogenic or extremely toxic, and therefore cannot be used to generate antibodies in animals. Moreover, affinity maturation (i.e., increasing the affinity and specificity) of recombinant antibodies is very simple and relatively fast. Finally, large numbers of different antibodies against a specific antigen can be generated in one selection procedure. To generate recombinant monoclonal antibodies one can use various methods all based on Phage Display libraries to generate a large pool of antibodies with different antigen recognition sites. Such a library can be made in several ways: One can generate a synthetic repertoire by cloning synthetic CDR3 regions in a pool of heavy chain germline genes and thus generating a large antibody repertoire, from which recombinant antibody fragments with various specificities can be selected. One can use the lymphocyte pool of humans as starting material for the construction of an antibody library. It is possible to construct naive repertoires of human IgM antibodies and thus create a human library of large diversity. This method has been widely used successfully to select a large number of antibodies against different antigens. Another possibility is to prepare so- called patient libraries. First, the sera of individuals (in our case autoimmune patients) are tested for the presence of specific antibodies directed to the antigen of interest. From the lymphocyte pool of positive individuals, IgG libraries can be generated, which will contain IgG antibodies of high specificity and with high affinity. Construction of specialized libraries can be used to generate antibodies from patients with, for example, certain tumors for selection of anti-tumor activity.

Overexpression of a number of oncogenes has been reported to be associated with malignant cellular transformation. For example, amplification of myc has been reported in COLO 320 colon carcinoma cell cultures, the SKBR3 breast carcinoma cell line and in lung carcinoma cell lines. Amplification of N-myc has been reported in neuroblastoma cell lines and retinoblastoma . Amplification of c-abl, c-myb and other oncogenes has also been reported to be associated with malignant transformation { cf . Chapter 12, Weiss et al,

1985) .

High levels of various oncogenes have also been reported to increase the risk of recurrence of the tumor.

For example, a correlation between the level of bneu/c-erB-2 and the cause and course of human breast cancer has been reported ( cf . Paterson et al, 1991) . High levels of myc, int- 2 and hist-1 have also been associated with breast cancer. Similarly, elevated levels of the receptor for EGF, EGF-R, have been shown to be associated with breast cancer { cf . Grimaux et al, 1990) . The overexpression of these and other oncogenes have also been reported as being associated with other cancers . Many oncogenes show some homology to genes involved in cell growth. For example, Table 1 compares oncogenes with homologous cellular genes. In this table, PDGF denotes platelet-derived growth factor, FHJF denotes fibroblast growth factor, EFG denotes epidermal growth factor, and M-CSF denotes mononuclear-phagocyte growth factor. Table l¹

Adapted from Druker et al (1989)

The family includes src, fgr, yes, lck, hck, fyn, lyn and tkl .

The subcellular location of these oncogene products is uncertain.

Antibodies to most of these oncogenes have been reported. In addition to overexpression of oncogenes (sometimes referred to as ones) , some oncogenes undergo a mutation from a proto-onc (normal gene for normal protein) to an one (gene whose protein can cause malignant transformation) which appears to result in malignant transformation of cells. For example, point mutations at the ras gene at the codons for the ras p21 at residue positions 12, 13 and 61 have resulted in mutant ras p21 proteins which are associated with various cancers. Antibodies specific to many of these ras mutants are known. Similarly, expression of viral proteins can lead to disease resulting in illness and even death. The virus can be either an RNA or a DNA virus. For example, one type of RNA virus, retroviruses , are typically classified as being part of one of three subfamilies, namely, oncoviruses, spumaviruses, and lentiviruses . Infection by an oncovirus is typically associated with malignant disorders. The viral proteins encoded include the gag, pol and envelope proteins. In some instances the virus contains oncogenes which encode a protein capable of malignant transformation of a cell in culture. Lentiviruses result in infection which is generally slow and cause chronic debilitating diseases after a long latency period. In addition to genes encoding the gag, pol and envelope structure proteins, they also encode a variety of regulatory proteins . The RNA and/or DNA of the virus can take over the cell machinery to produce the virally-encoded protein.

For example, HTLV-1 is a retrovirus which is the etiological agent of adult T-cell leukemia-lymphoma (ATLL) , an aggressive neoplasm of CD4⁺ T-cells. The viral proteins expressed by such a virus result in the transformation of the cell . The tax and rex gene and gene products appear to be significant with respect to tumorigenicity . Thus, they are a preferred grouping of target molecules. HIV comprises a family of lentiviruses, including

HIV-1 and HIV-2, which are the etiological agents of immunodeficiency diseases, such as the acquired immune deficiency syndrome (AIDS) and related disorders (Barre- Sinoussi et al, 1983; Gallo et al, 1984; Levy et al, 1984). The Epstein-Barr virus has been linked to a number of tumors, such as selected outbreaks of Burkitt ' s lymphoma, nasopharyngeal cancer, and B-lymphomas in immunosuppressed individuals (zur Hausen, 1991) . Hepatitis B virus has been linked to hepatocellular cancer (zur Hausen, 1991) . In particular, the X open reading frame of the virus seems to be involved. Accordingly, an antibody that targets this region or expression products from this region would be preferred in the present method.

Papillomaviruses have been linked to anogenital cancer. In these viruses, the E6 and E7 genes appear to be involved and are good targets for a stabilized antibody according to the present invention.

By intracellular binding to nucleic acid, such as a DNA provirus, one can prevent or inhibit the integration of the virus into the cell. By binding to the RNA of the virus, one can interfere with its expression of viral protein.

Anti-nucleotide antibodies have been extensively studied { cf . PCT WO94/02610) , and the antibodies have the same basic features .

These antibodies can be produced and/or screened by standard techniques, such as by using RNA, to create a library containing antibodies. Alternatively, one can select and/or design antibodies to target and interfere with an important nucleic acid binding site, for example, the TAR element of the primate immunodeficiency viruses. This nucleic acid sequence is present in the 5'LTR and is responsive to tat, resulting in enhanced expression of viral protein.

By intracellular binding to target proteins of these oncogenes and viruses, it is possible to disrupt the normal functioning of such proteins, reducing or avoiding the disruptive effect of the protein. Because the intracellularly-expressed proteins of the present invention are stabilized against rapid proteolysis, the proteins of the present invention are available for sufficient time in the desired location to perform their desired function.

For example, binding to a protein that must be further processed, such as a receptor protein, a viral envelope protein, e.g., HIV gpl60, can significantly reduce the cleavage of the protein into its active components. In another example, the capsid protein, e.g., the HIV capsid protein, is modified co-translationally by addition of the fatty acid myristic acid. It appears that myristic acid is involved in the attachment of the capsid precursor protein to the inner surface of cells. In HIV proviruses, which have been altered so that they are not capable of adding this myristic acid, the provirus is not infective. Studies of myristylation reveal a requirement for glycine at position two from the amino terminus and also at amino acid residues within six to ten amino acids from the site of myristylation. Thus, antibody binding to the protein at and near these sites can disrupt myristylation. In a similar fashion, binding to a protein that has a significant external domain can hinder the effect of the protein.

In another embodiment, by binding to a dysfunctional receptor protein, one can block the undesired interactions that can result in cellular dysfunction, such as malignant transformation.

For example, many proteins, such as surface receptors, transmembrane proteins, etc., are processed through the endoplasmic reticulum-Golgi apparatus. Examples of these proteins include neu and envelope glycoproteins, such as those of the primate lentiviruses, such as HIV-1 or

HIV-2. By using antibodies that can be delivered to such a region of the cell and be specific for a particular protein, one can disrupt the function of the protein without disrupting other cellular functions. For example, the PDGF- /2 and FGF-like factors produced by sis and int-2 pass through the endoplasmic reticulum. These factors are involved in many cancers. Thus, in addition to targeting the receptor, one can target the growth factors by using antibodies to them. Growth factors are also expressed by many other malignant cells, such as from carcinoid syndrome tumors, which are also a useful target for the stabilized antibodies of the present invention.

The present method of intracellularly generating stable antibodies can be used to disrupt a function that is undesirable at a particular time. For example, the MHC class I and class II molecules are important in the immune system's recognition of antigens. However, such immune recognition, particularly from MHC class II molecules, can cause problems, such as in organ transplants. Thus, by targeting class II molecules with organ transplants, the host immune response can be down-regulated. These molecules can preferably be targeted at different points in their processing pathway. Preferably, one would use an inducible promoter, such as those previously discussed above, for the stabilized antibody gene . Thus, by taking into account the particular target, many variations of this method can be designed by the skilled artisan without undue experimentation.

For instance, the HIV-1 envelope gene directs the synthesis of a precursor polyglycoprotein termed gpl60. This protein is modified by addition of multiple N- linked sugars as it enters the endoplasmic reticulum. The glycosylated envelope protein precursor is then cleaved within the Golgi apparatus to yield a mature envelope protein comprised of an exterior glycoprotein, gpl20, and a transmembrane protein, JP41. The envelope glycoprotein complex is anchored to the virion envelope and infects the cell membrane by gp41 through non-covalent interactions. Following binding of the gpl20 exterior glycoprotein to the CD4 receptor, the fusion of viral and host cell membrane allows virus entry. The fusogenic domain of the gpl20/gp41 complex is thought to reside at the amino terminus of gp41 because this region exhibits sequence homology with a fusogenic domain of other viral proteins and because mutations in this region inactivate the virus and prevent viral fusion. While the processed gpl20 and gp41 are transported to the cell surface and secreted as part of the virion of the virus, sometimes referred to as viral particles, the uncleaved gpl60 is delivered to lysosomes for degradation. The cleavage process normally is relatively inefficient. Therefore, the method of using stabilized intracellular antibodies to bind to the newly synthesized gpl60 in the lumen of the endoplasmic reticulum and inhibit its transport to the Golgi apparatus greatly reduces the amount of protein available for cleavage to gpl20 and gp41. Accordingly, the viral particles produced have greatly diminished amounts of gpl20 and gp41 on their surface, and these particles are not considered to be infectious. This discussion of the HIV-1 gpl60/gp41 proteins exemplifies use of the present invention for other envelope proteins and processed proteins. The same techniques used herein can be adapted by known techniques based upon the present disclosure without undue experimentation.

Additionally, the envelope protein of the immunodeficiency viruses has been implicated in other aspects of the disease. For example, HIV infection of cell cultures typically generates an acute and/or chronic infection. In both cases, virus is produced and is released by budding at the cellular membrane. An acute infection is typically characterized by a cytopathic effect manifested by vacuolization of cells and formation of syncytia and consequent cell lysis. In tissue cultures, the cytopathic effects of HIV-1 consist of multinucleated giant cell (syncytium) formation and the lysis of single cells. Syncytium formation is mediated solely by the HIV-1 envelope protein expressed on the infected cell surface. The envelope binds to the CD4 receptor present on adjacent cells and then, via a fusion reaction analogous to that involved in virus entry, the opposed cell membrane are fused, forming heterokaryons .

Single cell lysis also depends on effective membrane fusion induced by the envelope glycoprotein, as some mutations in the gp41 amino terminus result in replication competition viruses that are attenuated for both syncytium formation and single cell lysis. It has also been reported that amino acid changes in gpl20 which affect processing of the gpl60 precursor can decrease single cell lysis, and that single cell lysis requires adequate levels of CD4 expression, independent of the level of viral protein expression or viral DNA in the infected cell. In addition, the HIV envelope glycoprotein has been implicated by a number of other individuals in explaining the onset of the associated immunodeficiency infected individuals. Siliciano et al (1988) have shown that a subset of CD4⁺ gpl20-specific clones manifest cytolytic activity and lyse uninfected autologous CD4⁺ T-cells in the presence of gpl20 in a process that is strictly dependent upon CD4 mediated uptake of gpl20 by T-cells. Since gpl20 can be shed from infected cells, this CD4 -dependent autocytolytic mechanism can contribute to the profound depletion of CD4⁺ T- cells in AIDS patients. Others have shown that an autoimmune idiotypic network develops in HIV-1 infections which leads to the development of autoimmune antibodies that destroy CD4⁺ T- cells. This autoimmune mechanism develops because of the sequence homologies between gpl20 and class II MHC molecules. The immunosuppressive effects of gpl20 on the CD4⁺ T-cell proliferation to antigenic stimulus have been demonstrated, suggesting that immunodeficiency diseases, such as HIV-1, may affect major histocompatibility complex II restricted antigen recognition independent of CD4⁺ T-cell loss. In rodent neurons, gpl20 has been shown to cause an increase in intracellular calcium and neuronal toxicity, an effect which might be mediated by activation of the nuclear endonuclease. In addition, activation induced T-cell death, or apoptosis, has also been proposed as occurring in vivo and accounting for the progressive depletion of CD4⁺ T-cells that leads to AIDS. In vi tro and in vivo soluble gpl20 can interact with

CD4 receptors on uninfected cells, leading to an abortive cell activation and, thus, trigger apoptosis. It has also been proposed that the envelope glycoprotein can act as a superantigen, binding only the variable- β region of the T- cell antigen receptor, thereby inducing massive stimulation and expansion of such T-cells, followed by deletion of energy. Thus, by decreasing the amount of gpl20, effects associated with AIDS can be alleviated and retarded.

Intracellular expression of a stabilized antibody to its target, for example, an antibody to the envelope glycoprotein or an antibody to HIV tat protein, results in an antibody that binds the target, e.g., envelope glycoprotein or tat protein, respectively, in the cell and prevents further processing. The present method is highly specific and does not adversely affect cellular functioning. Thus, a mutant envelope protein that contains a single point mutation that abolishes the protein's ability to bind to this antibody will be processed normally in cells that constitutively express the protein. Similarly, single-chain antibodies to other proteins will not affect the processing of the envelope protein. Thus, the present method permits using an antibody specific to a particular portion and results in a process that can be tailored for specific diseases.

Additionally, this method can be used prophylactically. The antibody can be under the control of a promoter that is specifically activated by the target (e.g., an HIV LTR) , thereby only turning on the antibody when the target is present. Other types of inducible promoters are known in the art and can be selected and used based upon the present disclosure.

The use of the present stabilized antibodies does not affect processing of other proteins. For example, the antibody to the HIV envelope glycoprotein does not bind other envelope glycoproteins and does not prevent processing of such a protein. For example, the processing of an unrelated envelope glycoprotein, such as Bunyavirus envelope glycoprotein, will not be affected.

Numerous other sites can be targeted, such as the cytoplasmic side of a membrane receptor. It is through the cytoplasmic tail that signal transduction occurs. For example, using the neu/erbB-2 receptor or G protein receptor one can target the loop or cytoplasmic tail, thereby preventing such single transduction. Stabilized fragments of antibodies to activated receptors, such as to phosphorylated amino acids, can be used, thus reducing the pool of target receptors . The stabilized antibodies bind specifically to the target, e.g., a protein, and, thus, effectively compete with other molecules that also form complexes with the protein. To ensure that the antibodies of the present invention can compete successfully with other molecules, they must retain at least about 75% of the binding effectiveness of the complete antibody to the target, i.e., have constant as well as variable regions. More preferably, the fragments have at least 85% of the binding effectiveness of the complete antibody. Still more preferably, the fragments have at least 90% of the binding effectiveness of the complete antibody. Even more preferably, the fragments have at least 95% of the binding effectiveness of the complete antibody.

The method of the present invention is broadly applicable to a wide range of target molecules, including proteins, RNA, DNA, haptens, phospholipids, carbohydrates, etc. The target molecules can be present in a wide range of hosts, such as animals, birds and plants. Preferably, the target is found in animals, including humans. More preferably, the species is one that has commercial importance, such as fowl, pigs, cattle, cows, sheep, etc. Most preferably, the target molecule is found in humans.

Although antibodies are able to recognize an almost limitless number of foreign molecules, in nature, antibodies recognize structures exterior to the cell. Once synthesized, antibodies are secreted into the surrounding fluid or remain bound to the outer cell membrane. In the present invention, the stabilized antibodies are expressed and these antibodies retain the ability to specifically bind to a target intracellularly.

Specificity for a particular target can be obtained by using the immune system itself . The target or an antigenic portion thereof is used, or a hapten-carrier complex to generate an antibody. This can be accomplished by standard techniques.

For example, the antigen binding or variable region is formed by the interaction of the variable heavy (V_H) and variable light (V_L) domains at the amino termini of the chains . The smallest fragment containing a complete binding site is referred to as Fv and is a heterodimer of the V_H and V_L domains. However, it is possible to obtain binding without a complete binding site. For example, one can obtain antigen binding activity using only a heavy chain binding domain (dAbs, also referred to as single domain antibodies) . As noted above, in the present invention, one can use a gene coding for such a stabilized antibody fragment as long as the stabilized fragment retains sufficient binding ability compared to the parent antibody. Preferably, the stabilized fragment contains at least a V_H and V_L heterodimer.

Determination of the three-dimensional crystal structures of antibody fragments by X-ray crystallography has led to the realization that variable domains are each folded into a characteristic structure composed of nine strands of closely packed β-sheets. The structure is maintained despite sequence variation in the V_H and V_L domains. Analysis of antibody primary sequence data has established the existence of two classes of variable region sequences, hypervariable sequences and framework sequences . The framework sequences are responsible for the correct β-sheet folding of the V_H and V_L domains, and for the interchain interactions that bring the domains together. Each variable domain contains three hypervariable sequences which appear as loops. The six hypervariable sequences of the variable region, three from the V_H and three from the V_L, form the antigen binding site, and are referred to as a complementarity determining region. By cloning the variable region genes for both the

V_H and the V_L chain of interest, it is possible to express these proteins in bacteria and rapidly test their function. One method is by using hybridoma mRNA or splenic mRNA as a template for PCR amplification of such genes. Thus, one can readily screen a stabilized antibody fragment to ensure that it has a sufficient binding affinity for the antigen. The binding affinity (K_d) should be at least about 10^"7L/M, more preferably at least about 10^"8 L/M.

In one embodiment, the genes encoding the light chain and the heavy chain encode a linker to make a single- chain antibody (sFv) . The sFv will properly fold even under the reducing conditions sometimes encountered intracellularly. The sFv typically comprises a single peptide with the sequence V_H-linker-V_L or V_L-linker-V_H . The linker is chosen to permit the heavy chain and light chain to bind together in their proper conformational orientation.

Thus, the linker should be able to span the 3.5 nm distance between its points of fusion to the variable domains without distortion of the native Fv conformation. The amino acid residues constituting the linker must be such that it can span this distance and should be five amino acids or larger. The amino acids chosen also need to be selected so that the linker is hydrophilic so that it does not get buried in the antibody. Preferably, the linker should be at least about 10 residues in length. Still more preferably, it should be about 15 residues in length. While the linker should not be too short, it also should not be too long, as that can result in steric interference with the combining site. Thus, it preferably should be 24 residues or less. The linker (Gly- Gly-Gly-Gly-Ser)₃ (SEQ ID NO: 2) is a preferred linker that is widely applicable to many antibodies as it provides sufficient flexibility. Other linkers that can be used are listed in PCT WO94/02610, the entire contents of which are hereby incorporated by reference .

Alternatively, one can use a 15-mer, such as the SEQ ID NO: 2 although any sequence can be used through mutagenesis that randomizes the amino acids in the linker. After that, phage display vectors are used to pull out the antibodies with the different linkers and screen for the highest affinity single-chain antibody generated. Preferably, the gene does not encode the normal leader sequence for the variable chains as it is preferred that the antibody is not expressed with a leader sequence. The nucleotides coding for the binding portion of the antibody preferably do not encode the antibody's secretory sequences, i.e., the sequences that cause the antibody to be secreted from the cell. These sequences can be contained in the constant region. Preferably, one also does not use nucleotides encoding the entire constant region of the antibodies. More preferably, the gene encodes fewer than six amino acids of the constant region.

The immune system can prepare a stabilized antibody which will bind to a specific molecule, such as a target protein, by standard immunological techniques, such as by using the protein, or an immunogenic fragment thereof, or a peptide chemically synthesized based upon such protein. Any of these sequences can be conjugated, if desired, to keyhole limpet hemocyanin (KLH) and used to raise an antibody in animals, such as mice, rabbits, rats and hamsters. Thereafter, the animals are sacrificed and their spleens are obtained. Monoclonal antibodies are produced by using standard fusion techniques for making hybridoma cells. This typically involves fusing an antibody-producing cell (i.e., spleen) with an immortal cell line, such as a myeloma cell, to produce the hybrid cell.

Another method for preparing antibodies is by in vi tro immunization techniques, such as using spleen cells. For this, an antigen is injected into a culture of murine spleen cells, and then an antibody produced to the antigen is screened. With this method, as little as 0.1 μg of antigen can be used, although it is preferred to use about 1 μg/ml of antigen. For in vi tro immunization, spleen cells are harvested and incubated at the desired amount, e.g., 1 x 10⁷ cells/mL, in medium plus the desired antigen at a concentration typically around 1 μg/mL. Thereafter, one of several adjuvants, depending upon the results of the filter immunoplaque assay, is added to the cell culture. These adjuvants include N-acetylmuramyl-L-alanyl-D-isoglutamine, or Salmonella typhimurium mytogen, or T-cell condition, which can be produced by conventional techniques or obtained commercially. The spleen cells are incubated with the antigen for four days and then harvested.

Single cell suspension of the in vi tro immunized mouse spleen cells are then incubated, for example, on antigen-nitrocellulose membranes in microfilter plates. The antibodies produced are detected by using a label from the antibodies, such as horseradish peroxidase-labelled second antibody, such as rabbit anti-mouse IgA, IgG and IgM. In determining the isotype of the secreted antibodies, biotinylated rabbit anti-mouse heavy chain specific antibodies can be used, followed by a horseradish peroxidase- avidin reagent .

The insoluble products of the enzymatic reaction are visualized as blue plaques on the membrane. These plaques are counted, for example, by using 25 times magnification. The nitrocellulose microfilter membrane readily absorbs a variety of antigens, and the filtration unit used for the washing step is preferred because it facilitates the plaque assay. The antibodies are then screened by standard techniques to find the antibodies of interest. Cultures containing the antibodies of interest are grown and induced, and the supernatants are passed through a filter and then through a column, such as an antigen affinity column or an anti-tag peptide column. The binding affinity is tested using a mini-gel filtration technique. A radioimmunoassay can be used to screen the antibodies. In a preferred alternative, one can measure "on" rates and "off" rates using an analytical system. This latter technique is preferred over in vivo immunization because the in vivo method typically requires about 50 μg of antigen per mouse per injection, and there are usually two boosters following primary immunization for the in vivo method. Alternatively, one can use a known antibody to the target protein. Thus, one can obtain antibodies to the desired target protein. The reagent, a gene that encodes at least the antigen binding portion of the antibody, is synthesized, together with a nucleotide sequence encoding a stabilon sequence. The gene preferably does not contain the normal signal peptide sequences. In some preferred embodiments, it also encodes an intracellular localization sequence, such as one for the endoplasmic reticulum, nucleus, nucleolar, etc. To be expressed in the normal antibody secretory system, such as the endoplasmic reticulum (ER) and Golgi apparatus, a leader sequence should be used. To retain such antibodies at a specific place, a localization sequence, such as the KDEL sequence, can be used. In some embodiments, the antibody gene preferably also does not encode functional secretory sequences .

Using any of the antibodies chosen, one can construct V_H and V_L genes, such as by creating V_H and V_L libraries from murine spleen cells that have been immunized either by the above-described in vi tro immunization technique or by conventional in vivo immunization and from hybridoma cell lines that have already been produced or are commercially available. One can also use commercially available V_H and V_L libraries. One method involves upswing spleen cells to obtain mRNA which is used to synthesize cDNA. Double-stranded cDNA can be made using PCR to amplify the variable region with a derivative BN terminal V region primer and a J region primer or with V_H family specific primers, e.g., mouse-12, human-7.

For example, the genes of the V_H and V_L domains of a broadly neutralizing antibody to the envelope glycoprotein of HIV-1, such as F105, can be cloned and sequenced. The first strand cDNA can be synthesized from total RNA by using oligo dT priming and the Moloney murine leukemia virus reverse transcriptase according to known procedures . This first strand cDNA is then used to perform PCR reactions using typical PCR conditions to amplify the cDNA of the immunoglobulin genes. DNA sequence analysis is then performed. Heavy chain primer pairs consist of a forward V_H primer and a reverse J_H primer, each containing convenient restriction sites for cloning. The Kabat database on immunoglobulins could be used to analyze the amino acid and codon distribution found in the seven distinct human V_H families. From this, the 35 base pair universal 5' V_H primer is designed. One could use a primer, such as TTTGCGGCCGCTCAGGTGCA (G/A) CTGCTCGAGTC (T/C) GG (SEQ ID NO:3) which is degenerate for two different nucleotides at two positions and will anneal to the 5' end of FR1 sequences. A restriction site, such as the 5' Notl site (left-underlined) , can be introduced for cloning the amplified DNA and is located 5' to the first codon to the V_H gene. Similarly, a second restriction site, such as an internal Xhol site, can be introduced as well (right-underlined) .

Similarly, a 66 -base pair J_H region oligonucleotide can be designed for reverse priming at the 3' end of the heavy chain variable gene, e.g., AGATCCGCCGCCACCGTCCCACCACCTCCGGAGC- CACCGCCACCTGAGGTGACCGTGACC (A/G) (G/T) GGT (SEQ ID NO:4) . This primer additionally contains a 45 nucleotide sequence that encodes a linker, such as the (Gly-Gly-Gly-Gly-Ser) ₃ (SEQ ID NO: 2) interchange linker. This primer contains two degenerate positions with two nucleotides at each position based on the nucleotide sequence of the six human J_H region minigenes . Restriction sites can be used, for example, a BspEl site (left-underlined) is introduced into the interchange linker for cohesive end ligation with the overlapping forward V_kappa primer. An internal BsTEII site (right-underlined) is introduced as well for further linker exchange procedures .

A similar strategy using a 45 nucleotide interchange linker is incorporated into the design of the 69 nucleotide human V_kappa primer. There are four families of V_kappa genes. The 5' V_kappa primer

GGTGGCGGTGGCTCCGGAGGTGGTGGGAGCGGTGGCGGCGGA-

TCTGAGCTC (G/C) (T/A)G (A/C) TGACCCACTCTCCA (SEQ ID NO:5), which will anneal to the 5' end of the FR1 sequence, is degenerate at three positions (two nucleotides each) . The interchange linker portion can contain a BspEI site for cohesive end cloning with the reverse J_H primer, although other restriction sites can also be used. An internal Sad site (right-underlined) can be introduced as well to permit further linker exchange procedures.

The reverse 47 nucleotide C_kappa primer (Kabat positions 109-113) GGGTCTAGACTCGAGGATCCTTATTAACGCGTTGGTGCAGC- CACAGT (SEQ ID NO: 6) is designed to be complementary to the constant regions of kappa chains (Kabat positions 109-113). This primer anneals to the 5' most end of the kappa constant region. The primer contains an internal Mlul site (right- underlined) preceding two stop codons. In addition, multiple restriction sites, such as BamHI Xhol/Xbal (left-underlined) , can be introduced after the tandem stop codons . A similar reverse nucleotide C_kappa primer, such as a 59 nucleotide primer can also be designed that contains a signal for a particular intracellular site, such as a carboxy terminal endoplasmic reticulum retention signal, Ser-Glu-Lys-Asp-Glu- Leu (SEQ ID NO: 7) ,

GGGTCTAGACTCGAGGATCCTTATTACAGCTCGTCCTTTTCGCTTGGTGCAG- CCACAGT (SEQ ID NO : 8 ) . Similar multiple restriction sites can be introduced after the tandem stop codons . After the primary nucleotide sequence is determined for both the heavy and kappa chain genes and the germ line genes are determined, a PCR primer can then be designed, based on the leader sequence of the V_H 71-4 germ line gene. For example, the V_H 71-4 leader primer TTTACCATGGAACATCTGTGGTTC (SEQ ID NO: 9) contains a 5'NcoI site (underlined) . This leader primer (P-L) is used in conjunction with a second J_H primer for PCR amplification experiments. The 35 base pair J_H region oligonucleotide is designed to contain the same sequence for reverse priming at the 3' end of the heavy chain variable gene.

TTAGCGCGCTGAGGTGACCGTGACC (A/G) (G/T) GGT (SEQ ID NO: 10). This primer contains two degenerate positions with two nucleotides at each position. A BssHII site (left-underlined) 3' to and immediately adjacent to the codon determining the last amino acid of the J region, allows convenient cloning at the 3' end of the V_H gene. An internal BstEII site (right-underlined) is introduced as well. This sequence is used to amplify the V_L sequence. The fragments amplified by the P-L (leader primer) and P linker (reverse primer) and P-K (V2 primer) and P-CK primers (reverse CK primer) are then cloned into an expression vector, such as the pRc/CMV (Invitrogen) , and the resultant recombinant contains a signal peptide V_H interchain linker and V_L sequences under the control of a promoter, such as the CMV promoter. To this is then added the corresponding sequence for the stabilizing moiety (stabilon) . The skilled artisan can readily choose other promoters that express the gene in the cell system of choice, e.g., a mammalian cell, preferably a human cell.

This single-chain antibody can be prepared based upon the present disclosure by any of a number of known means. For example, the V_H/J_H-ICL and ICL-V_kappa/C_kappa PCR fragments are digested with Notl/BNspEI and BspEI/Xbal, respectively, and cloned into a plasmid, such as pSL1180, using SURE bacteria as hosts. The resulting sFGV is restriction enzyme digested and the Notl/Bglll fragment is cloned into the Notl/BamHI site that is located 3' to the pelB signal peptide in a pET expression vector. The resulting plasmid is then transformed into the appropriate host, such as BL21 (DE3) . Plasmid fragments are obtained after suitable times, for example, 2 to 4 hours after induction at 24° with 0.2 mM IPTG and tested for its ability to bind its target, e.g., gpl20 binding activity, by standard techniques, such as ELISA.

The V_H71-4 leader and a J_H-BssHII primer are used to PCR amplify an intronless fragment containing the leader peptide and rearranged heavy chain gene. The fragment is bound and cloned in the forward direction into an EcoRV site in a plasmid, for example, pSLllδO. Subsequently, a NcoI/BstEII fragment is obtained and combined with the BstEII/Sph I fragment of, e.g., F105sFv from pSL1180, in a three-piece ligation with Ncol/SpHI digested pSL1180 to produce the V_H 71-4/SCA. A V_H 71-4 SCA containing the carboxyl-terminated SEKDEL sequence can be constructed using an ICL-V_kappa-SEKDEL PCR product that is blunt and cloned in the forward direction into an EcoRV site in pSL1180. The fragment is removed by BspoEI/Xbal digestion and combined with the Ncol/BspEI fragment of V_H 71-4/SCA in a three-part ligation with Ncol/Xbal digested pSL1180 to produce V_H 71- 4/KDEL. Before cloning into pRC/CMV, an EcoRI to Hindlll conversion linker is introduced into EcoRI digested pSL1180 containing the two single-chain antibodies. Subsequently a Hindlll/Xbal fragment from both single-chain antibodies is obtained and cloned into Hindlll/Xbal digested pRC/CMV to produce pRC/SCA and pRC/KDEL.

Similar strategies can be used to prepare virtually any other antibodies. For example, using the combination of mRNA purification, single-strand cDNA synthesis and PCR amplification using the V_H and J_H degenerative primers discussed above, an approximately 350 bp product can be obtained from spleen cells immunized against tat and anti-tat hybridoma cell lines. Using the same techniques, as described for heavy chains, a 320 bp V_kappa gene product can be obtained from spleen cells immunized against tat and the anti-tat hybridoma cell lines using the V_kappa and J_appa degenerative primers discussed above. Once obtained, the V_H and V_L domains can be used to construct sFv, Fv or Fab fragments .

A preferred target is one processed by the endoplasmic reticulum, where proteins are typically made.

However, there are instances where a greater degree of intracellular specificity is desired, for example, with targeting nuclear proteins, RNA, DNA, or cellular proteins, or nucleic acids that are subsequently processed. For example, with virally-encoded proteins, such as lentiviruses, structural proteins are typically cytoplasmically expressed, whereas regulatory proteins can be expressed in or near the nucleus. Thus, one preferably uses localization sequences for such targets. The stabilized antibodies of the present invention can be delivered intracellularly and can be expressed there and bind to a target protein. Localization sequences have been divided into routing signals, sorting signals, retention or salvage signals, and membrane topology-stop transfer signals (Pugsley, 1989) .

Myristolation sequences can be used to direct the antibody to the plasma membrane. Table 2 sets forth the amino- terminal sequences for known N-myristoyl proteins and their subcellular location. In addition, as shown in Table 2 below, myristoylation sequences can be used to direct the antibodies to different subcellular locations, such as the nuclear region. Localization sequence may also be used to direct antibodies to organelles, such as the mitochondria and the Golgi apparatus. The sequences Met-Leu-Phe-Asn-Leu-Arg- Xaa-Leu-Asn-Asn-Ala-Ala-Phe-Arg-His-Gly-His-Asn-Phe-Met-Val- Arg-Asn-Phe-Arg-Cys-Gly-Gly-Pro-Leu-Xaa (SEQ ID NO: 11) can be used to direct the antibody to the mitochondrial matrix. See, e.g., Tang et al for localization of proteins to the Golgi apparatus .

Table 2

The standard single letter amino acid code is used in the Table, the amino acid sequences using the three-letter code are set out in the Sequence Listing.

Abbreviations are PM = Plasma Membrane; G = Golgi; N = Nuclear; C = Cytoskeleton; s = cytoplasm (soluble) ; M = Membrane

For example, it is known that tat is located in subnuclear and subnucleolar regions of infected cells. Thus, it is preferable that the tat antibody target the nuclear and/or nucleolar regions of the cell. Since this antibody is to be synthesized in the cytoplasm, it does not have a leader sequence. To target the nuclear and/or nucleolar regions, the antibody does need a localization sequence. Preferred nuclear targeting sequences are SV40 and preferred nucleolar targeting regions are tat nucleolar signals. For example, a tat antibody, a single-chain antibody, with SV40 nuclear localization signal, will bind to tat and can reduce tat activity by over 80% when compared to the antibody with an immunoglobulin leader sequence, which directs the antibody to a different cellular compartment, such as the ER. Preferably, with viruses such as HIV, the structural proteins are targeted in the cytoplasm, such as envelop and gag, whereas the regulatory proteins, such as tat and rev, are targeted in the nucleus and nucleolar regions. More preferably, one would target rev using the rev nucleolar sequence Arg-Gln-Ala-Arg-Arg-Asn-Arg-Arg-Arg-Arg-Trp-Arg-Glu- Arg-Gln-Arg (SEQ ID NO: 31) . For example, the tax protein of HTLV-1 or HTLV-2 is also preferably searched for in the nucleus or nucleolus . If possible, it is preferable to use the localization signals of the target protein to direct the antibody to the desired location. For example, HIV-12 tat protein has a nucleolar localization signal, which is preferably used.

As the term is used here, the gene for the antibody can encompass genes for the heavy chain and light chain regions, including a sequence for imparting a stabilizing amino acid at the N-terminal end of the antibody expressed. The gene is operably linked to a promoter or promoters which result in its expression. Promoters that permit expression in mammalian cells are well known and include a CMV, a viral LTR, such as the rous sarcoma virus LTR, HIV-LTR, HTLV-1 LTR, the SV40 early promoter, E. coli lac UV5 promoter, and the herpes simplex tk promoter. This DNA sequence is referred to as the antibody cassette.

The antibody cassette is delivered to the cell by any known means. For example, a cassette containing these antibody genes, such as the sFv gene, can be targeted to a particular cell by a number of techniques. Using a mammalian expression vector, such as a herpes vector, an adenovirus vector or a pox vector, a retroviral vector, a plasmid bound to an antibody, etc., cells can be transduced by standard techniques well known to the skilled artisan. Preferably, this cassette is introduced in the cell using an HIV viral vector, which is defective in packaging HIV sequences, but will preferentially target HIV susceptible cells. In addition, one can use a promoter that will differentially express the gene in the desired target cell. For example, using an HIV-LTR as a promoter where the target is HIV- infected cells, the HIV viral proteins in the cell, such as tat, can result in enhanced expression of the antibody when compared to uninfected cells. In another embodiment one can transduce cells that are at greater risk for viral infection, such as CD4 cells. Intracellular expression of the antibody permits it to bind the target. This disrupts the function of the target, e.g., a protein, including the undesired function. For example, expressing the sFv of a broadly neutralizing antibody to envelope glycoprotein can intracellularly block the transport and interaction with the CD4 molecules of the HIV-1 glycoprotein, as well as cleavage of the protein. This can be accomplished by cloning an sFv without any targeting signal and an sFv antibody with an endoplasmic reticulum retention signal and inserting these into mammalian cells using a mammalian cell expression vector. Another example uses an antibody specific for one which is targeted to breast tissue to keep the neu protein of the cell.

The presence of target protein also assists the antibody in folding to the correct conformational state. These antibody-ligand complexes prevent the target from operating in its typical manner. For instance, cytopathic fusion mediated by the HIV-1 gp 120/41 is inhibited in the cells. This is shown by co-transfecting CD4⁺ Hela cells with the HIV-1 glycoprotein expressing pSVIII env and sFv or sFv-

KDEL plasmid DNAs at a ratio of 1:5 or transfecting the transformed cells with pSVIII.

Thus, by producing antibodies which have been genetically engineered to last sufficiently long to affect the target, one can intervene in a viral infection, such as an HIV-1 infection to alleviate the undesirable effect of the virus. Using this basic strategy, one can intervene in other viral and metabolic diseases, such as infections by DNA viruses, such as herpes simplex, and RNA viruses, such as HTLV-1 and 2.

The present method permits a wide range of approaches, even against the same disease. Antibodies against reverse transcriptase can interfere with template binding functions of the protein. Antibodies to this protein are known and include C2003, which binds to a sequence in the C-terminal portion of the p66 component. This antibody also binds to HIV-2. Such antibodies can be screened for from patient sera and antibodies cloned as described above.

Another approach is to target a critical nucleic acid sequence in the virus, such as the TAR element. The TAR element, which is responsive to tat, is located at the 5' end of messenger viral RNA. Tat binding to this TAR element has been shown to result in a depression of tar inhibitor of translation in vi tro . In addition, the tar element increases transcription, initiation, and also acts as an anti- attenuator of transcription elongation. By directing an antibody against the tar sequence, inhibition of tat binding will occur and there will be a dramatic decrease in transcription efficiency. This will ultimately result in an inhibition or reduction of virus production. A similar approach can be used to produce antibodies against the rev responsive element. Rev controls the synthesis of viral structural proteins, including the capsid protein, replicative enzymes and the envelope glycoprotein. The rev protein controls virion protein expression by controlling the cytoplasmic accumulation of RNA species. In the absence of rev activity, small multiply- spliced viral RNA species accumulate; in the presence of rev, full-length and partially-spliced envelope glycoprotein messenger RNAs accumulate. Antibodies directed against the rev responsive element should inhibit rev binding to the rev responsive element and, therefore, inhibit the major biological effect of rev. The rev protein regulates the synthesis of capsid, replicative enzyme and envelope glycoprotein production by regulating the accumulation of messenger RNA species from which they are made. Structural protein messenger RNAs require binding of rev protein to the folded RNA structure called rev responsive element for translocation from the nucleus to the cytoplasm. Inhibition of rev binding by an anti-rev antibody should prevent virus expression from infected cells. Such antibodies which have sufficient life in the body to inhibit rev binding can be synthesized using known technology based upon the present disclosure. For example, one can screen an RNA library with an antibody to obtain the desired antibody.

It has been proposed that tumor formation and metastasis is dependent upon angiogenesis (Folkman et al, 1991) . For example, human melanoma has been found to produce several proteins with angiogenic activity, including fibroblast growth factor (gFGF) , transforming growth factor alpha (TGFa) , and transforming growth factor beta (TFGb) . By using such proteins as targets for intracellular antibodies, tumor formation and metastasis may be limited.

It has been suggested that alteration at positions 12, 13 or 61 of the ras p21 protein results in tumor formation. Using the mutant protein as a target for an intracellular antibody, which can distinguish the oncogenic ras from proto-ras, tumor formation will be limited.

Antibodies capable of such specific binding are known in the art .

It is preferred to use a "cocktail" approach, i.e., a mixture of antibodies, in dealing with undesired viral proteins, thereby targeting a variety of viral proteins at one time and making it more difficult for mutants to evolve and, thus, avoid the antibody. A cocktail of antibodies to at least envelope glycoprotein and tat is preferred. Other cocktails include antibodies to reverse transcriptase, TAR, RRE, etc. These cocktails can be administered together or by co-transfections . It is preferred that no more than about three proteins in this intracellular region are targeted, preferably no more than about two. For example, gpl60 and gp41 can be targeted at the endoplasmic reticulum. As long as another intracellular target is in a different cellular region, i.e., nucleus vs, endoplasmic reticulum, it can also be targeted without having a detrimental effect on antibody production. One preferred cocktail of antibodies would be antibodies for at least one structural viral protein, such as capsid or envelope, and one for regulatory proteins, such as HIV rev, tat, HTLV-1 or tax, or for a nucleic acid sequence, such as TAR or RRE. Another preferred cocktail is of antibodies to the same target but at various intracellular locations, which can be done using different localization sequences. Thus, if some target is not bound to the antibody at one location and, for instance, is further processed, it can be targeted at a subsequent location. For example, with the envelope glycoprotein one could use localization sequences to target the protein at a number of points in its processing path. Alternatively, one can use multiple antibodies to target different epitopes of molecules. For example, one can use one antibody to target the CD4 binding region of an envelope glycoprotein and a second antibody to target the fusogenic domain of gp41.

For HIV-encoded protein, one preferred vector is to have at least two antibodies to capsid or envelope proteins and at least one antibody to a regulatory protein, such as to gpl60, gp41, tat and rev. Another cocktail can include antibodies to both the viral mRNA and the protein it encodes.

Other preferred HIV-encoded target proteins are nef, vpr and, for HIV-1, vpu, and, for HIV-2, vpx, more preferably, nef and vpu. For example, the nef protein exists in the cytoplasm, as well as being attached to the inner surface of the plasma membrane. The protein is modified co- translationally by addition of myristic acid to the penultimate glycine residue of the amino terminus. The vpr protein has been found incorporated into the capsid of the virus. The vpu protein is located within the cytoplasm of cells and may be associated with sub-cellular organelles. Antibodies to these proteins can be made by the methodology described herein. Further, these proteins can be more specifically targeted by the skilled artisan based upon this disclosure by selection of appropriate localization sequences .

Thus, using the above-described methodology, one can treat mammals suffering from an ailment caused by the expression or overexpression of specific proteins. This method can be used to treat viral and metabolic disease, such as HIV, HTLV-1, HTLV-2 and herpes. Similarly, individuals having malignant tumors or who are susceptible to malignant cellular transformation caused by a high level of a protein or proteins, and altered protein or proteins or a combination thereof, can be treated. For example, one can target at least one of the antigens with an antibody that specifically binds to the antigen. Then, one delivers an effective amount of a gene capable of expressing the stabilized antibody under conditions which permit its intracellular expression to cells susceptible to expression of the undesired target antigen. This method can be used as a prophylactic treatment to prevent or make it more difficult for such cells to be adversely affected by the undesired antigen by preventing processing of the protein, interaction by the undesired protein with other proteins, integration by the virus into the host cells, etc. Where a number of targets exist, one preferred target is proteins that are processed by the endoplasmic reticulum. Intracellular delivery of any of the antibody genes can be accomplished by using gene therapy techniques, such as described above. The antibody can be any of the antibodies as discussed above.

HIV infects CD4 positive human lymphocytes and other immune cells. By targeting such cells with an antibody that binds to at least one HIV encoded target molecule, e.g., a protein, it is possible to treat an individual infected with the virus, slow and/or retard the spread of infection, or prophylactically treat such cells to make it more difficult for them to become infected.

Any of the known forms of gene therapy can be used to deliver genes to the target. For example, using a cell- specific gene transfer mechanism, which uses receptor- mediated endocytosis to carry RNA or DNA molecules into cells, a protein acting as a ligand is coupled to a poly-L- lysine, which then combines with RNA or DNA (the gene) to form soluble complexes by strong electrostatic interaction, whereby one can deliver the genes to the cells of interest. For example, using an antibody against gpl20 or CD4 as the ligand, one can specifically target such cells.

Antibodies used to target the cells can be coupled to the polylysme to form an antibody-polylysine conjugate by ligation through disulfide bonds after modification with a reagent, such as succinimidyl-3- (2-pyridyldithio) propionate (SPDP) . The antibody-polylysine-gene complexes are produced by mixing the antibody polylysme conjugates with a moiety carrying the antibody cassette. Preferably, the polylysme has an average chain length of about 60 to about 500 lysine monomers .

Conjugation with the antibodies can be accomplished using SPDP. First dithiopyridine groups are introduced into both antibody or polylysine by means of SPDP, and then the groups in the polylysme are reduced to give free sulfhydryl compounds, which upon mixing with the antibodies modified as described above, react to give the desired disulfide bond conjugates. These conjugates can be purified by conventional techniques, such as using cation exchange chromatography . These conjugates are then mixed with the antibody cassette under conditions that permit binding.

To treat the targeted cells, the vectors can be introduced to the cells in vi tro with the transduced cells injected into the mammalian host wherein it will bind with the CD4 cell and be taken up. To increase the efficiency of the gene expression in vivo, the antibody cassette can be part of an episomal mammalian expression vector, for example, a vector which contains the human Pappova virus (BK) origin of replication and the BK large T antigen for extra- chromosomal replication in mammalian cells, a vector which contains an Epstein-Barr (EB) virus origin of replication, and nuclear antigen (EBNA01) to allow high copy episomal replication. Other mammalian expression vectors, such as herpes virus expression vectors, or pox virus expression vectors, can also be used. These vectors are commercially available. The antibody cassette is inserted into the expression vectors using standard techniques. These expression vectors can be mixed with the antibody-polylysine conjugates and the resulting antibody-polylysine-expression vector containing antibody cassette complexes can readily be made based upon the present disclosure. Sufficient amounts of these vectors are injected to obtain a serum concentration ranging between about 0.01 μg/ml to 30 μg/ml of antibody conjugate. Because the stabilized antibodies of the present invention last longer in the body than conventionally prepared antibodies, the amount of antibody used in some instances can be reduced.

The vectors according to the present invention, which include vectors carrying an antibody cassette, can be administered by any conventional means, including parenteral injection, intraperitoneal injection, intravenous injection, intracranial injection or subcutaneous injection. Oral or other routes of administration can also be used.

The materials can be administered in any convenient form, such as mixed with an inert carrier, such as sucrose, lactose or starch. The material can be in the form of tablets, capsules or pills. For parenteral administration, the material is typically injected in a sterile aqueous or non-aqueous solution, suspension or emulsion in association with a pharmaceutically-acceptable parenteral carrier, such as physiological saline. All references cited herein, including journal articles or abstracts, published or unpublished U.S. or foreign patent applications, issued U.S. or foreign patens, or any other references are entirely incorporated by reference herein, including all data, tables, figures, and text present in the cited references. Additionally, the entire contents of the references cited within the references cited herein are also incorporated by reference in their entirety.

References to known method steps, conventional method steps, known methods or conventional methods is not in any way an admission that any aspect, description, or embodiment of the present invention is disclosed, taught, or suggested in the relevant art .

The foregoing description of the specific embodiments will so fully reveal the general nature of the invention that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without undue experimentation and without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. The means, materials, and steps for carrying out various disclosed functions may take a variety of alternative forms without departing from the invention. Thus the expressions "means to..." and "means for...", or any method step language, as may be found in the specification above and/or in the claims below, followed by a functional statement, are intended to define and cover whatever structural, physical, chemical or electrical element or structure, or whatever method step, which may now or in the future exist which carries out the recited function, whether or not precisely equivalent to Lhe embodiment or embodiments disclosed in the specification above, i.e., other means or steps for carrying out the same function can be used; and it is intended that such expressions be given their broadest interpretation.

REFERENCES :

Adames et al, Nature 318:533-538 (1985)

Aitken et al, FEBS Lett. 150:314-318 (1982)

Alexander et al, Mol. Cell. Biol. 7:1436-1444 (1987)

Alonso et al, NeuroReport 7(18):3105 (1996)

Ausubel et al (Eds) , Current Protocols in Molecular Biology, Greene Publishing Assoc. and Wiley Interscience (New York, 1987-1998)

Bairn et al , Proc. Natl. Acad. Sci. USA 88:5072-5076 (1991)

Barre-Sinoussi et al, Science 220:868-871 (1983)

Berger et al, Guide to Molecular Cloning Techniques. Methods of Enzymology, Vol. 152, Academic Press, Inc. (San Diego, 1987)

Biocca et al, Biochem. Biopohvs . Res. Commun. 197:422-427 (1993)

Brinster et al, Nature 296:39-42 (1982)

Brown et al, Cell 49:603-612 (1987)

Carr et al , Proc. Natl. Acad. Sci. USA 79:6128-6131 (1982)

Chow et al, Nature 327:482-486 (1987)

Coffin et al, Gene Therapy 3(7) :560 (1996)

Darnell et al , Molecular Cell Biology. Scientific American Books, Inc. (New York, 1985) Deutschle et al, Proc. Natl. Acad. Sci. USA 86:5400-5405 (1989)

Deutschle et al, Science 248:480-483 (1998)

Druker et al, N. Eng. J. of Mol. 321:1383-1392 (1989)

Duan et al, Proc. Natl. Acad. Sci. USA 91:5075-5079 (1994)

Figge et al, Cell 52:713-722 (1988)

Fink et al, Experimental Neurology 144(1) :103 (1997)

Folkman et al, Origins of Human Cancer: A Comprehensive

Review, Cold Spring Harbor Laboratory Press (Cold Harbor Springs, NY, 1991)

Fuerst et al, Proc. Natl. Acad. Sci. USA 86:2549-2533 (1989)

Gallo et al, Science 224: 500-5034, (1984)

Grimaux et al, Int. J. Cancer 45:255-262 (1990)

Grosschedl et al, Cell 38:647-658 (1984)

Hammer et al, Science 235:53-58 (1987)

Hanahan Nature 315:115-122 (1985)

Haynes et al, J. Biotechnol 44(1-3) :37 (1996)

Henderson et al, Proc. Natl. Acad. Sci. USA 80:339-343 (1987

Hu et al, Cell 48:555-566 (1987)

Kelsey et al, Genes and Devel. 1:161-171 (1987) Kollias et al, Cell 46:89-94 (1986)

Krisky et al, Gene Therapy 4(10):1120 (1997)

Krumlauf et al, Mol. Cell. Biol. 5:1639-1648 (1985)

Labow et al, Mol. Cell. Biol. 10:3343-3356 (1990)

Leder et al, Cell 45:485-495 (1986)

Levy et al, Science 224:497-500 (1984)

Lewin, Genes II, John Wiley & Sons (New York, 1985)

MacDonald, Hepatology 7:425-515 (1987)

Maidment et al, Experimental Neurology 139(1) :107 (1996)

Maniatis et al, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory (Cold Spring Harbor, NY, 1989)

Marasco et al, Proc. Natl. Acad. Sci. USA 90:7889-7893 (1993)

Marchildon et al, Proc. Natl. Acad. Sci. USA 81:7679-76-82 (1984)

Mason et al, Science 234:1372-1378 (1986)

Mhashilkar et al , EMBO J. 14:1542-1551 (1995)

Mogram et al, Nature 315:338-340 (1985)

Old et al, Principles of Gene Manipulation: An Introduction of Genetic Engineering, 2nd Ed. , University of California Press (Berkeley, CA, 1981)

Ootsuyama et al, Jpn J. Cancer Res. 76:1132-1135 (1985) Orlandi et al, Proc. Natl. Acad. Sci. USA 86:3833-3837 (1989)

Ornitz et al, Cold Spring Harbor Symp. Ouant. Biol. 50:399- 409 (1985)

Paterson et al, Cancer Research 51:556-567 (1991)

Paul et al, Proc. Natl. Acad. Sci. USA 84:7827-7831 (1987)

Persing et al, J. Virol. 61:1672-1677 (1987)

Pinkert et al, Genes and Devel . 1:268-276 (1987)

Pugsley, Protein Targeting, Academic Press, Inc. (1989)

Ratner et al, Nature 313:277-284 (1985)

Readhead et al, Cell 48:703-712 (1987)

Rhee et al, J. Virol. 61:1045-1053 (1987)

Richardson et al, Proc. Natl. Acad. Sci. USA 92:3137-3141 (1995)

Sani, Nature 314 : 283 -286 (1985)

Schultz et al, J. Virol. 46:355-361 (1983)

Schultz et al, J. Virol. 133:431-437 (1984)

Schultz et al, Science 227:427-429 (1985)

Schultz et al, Biochem. Biophys. Res. Commun. 146:1234-1239 (1987)

Siliciano et al, Cell 54:561-575 (1988) Streuli et al, Nature 326:619-622 (1987)

Swift et al, Cell 38:639-646 (1984)

Tang et al, J. Bio. Chem. 207:10122

Varshavsky, Proc. Natl. Acad. Sci. USA 93:12142-12149 (1996)

Voronova et al, Mol. Cell. Biol. 4:2705-2713 (1984)

Watson et al, Molecular Biology of the Gene, Vols. I and II, The Benjamin/Cummings Publishing Company, Inc. (Menlo Park, CA, 1987)

Weiss et al, Eds., RNA Tumor Viruses, Molecular Biology of Tumor Viruses , 2nd Ed., Chapter 12, "Human Oncogenes", Cold Spring Harbor Laboratory, 1985, pp. 487-543

zur Hausen, Science 254:1167-1173, 1991.

Claims

WHAT IS CLAIMED IS:

1. A method for modulating the degradability of a protein or peptide expressed intracellularly, comprising: altering a gene encoding a protein or peptide at the N-terminus so that the expressed protein or peptide exhibits an amino-terminal structure which renders the protein or peptide metabolically stable with respect to the N-end rule of proteolytic degradation; delivering said gene into a host cell to transform the host cell; and expressing a protein or peptide which is rendered metabolically stable by the amino-terminal structure.

2. The method according to claim 1, wherein the amino terminal structure is an N-terminal initiation methionine residue immediately followed by a stabilon sequence .

3. The method according to claim 1, wherein the stabilon sequence is selected from the group consisting of Gly, Val, Met, Ser and repeats and combinations thereof.

4. The method according to claim 3, wherein the stabilon sequence is a single amino acid residue selected from the group consisting of Gly, Val and Ser.

5. The method according to claim 3, wherein the stabilon sequence is a multiple repeat of a single amino acid residue selected from the group consisting of Gly, Val, Met and Ser.

6. The method according to claim 2, wherein the stabilon sequence has a length in a range of two to ten amino acid residues.

7. The method according to claim 2, wherein the protein or peptide comprises a destabilizing sequence joined by a protease-sensitive linker region to the C-terminus of the amino terminal structure.

8. The method according to claim 7, wherein the step of delivering said gene into a host cell also delivers a gene encoding a restriction protease which cleaves the protease-sensitive linker region and whose expression is under the control of an inducible promoter, and wherein the method further comprises a step of inducing the expression of the restriction protease to cleave the metabolically stable protein or peptide at such time it is desired that the metabolically stable protein or peptide be degraded by the N- end rule pathway.

9. The method according to claim 8, further comprising the step of delivering a gene encoding a restriction protease, which cleaves the protease-sensitive linker region, into the host cell previously transformed with the altered gene to cleave the metabolically stable protein or peptide so that the protein or peptide is degraded by the N-end rule pathway.

10. The method according to claim 8, wherein the protein or peptide before said altering step is metabolically stable .

11. The method according to claim 1, wherein the protein or peptide is an antibody.

12. The method according to claim 1, wherein the protein or peptide is a single-chain antibody.

13. A vector system for the intracellular expression of a protein or peptide whose degradability can be modulated, comprising a nucleotide sequence adapted to intracellular delivery and expression, wherein said nucleotide sequence contains a promoter operably linked to a gene encoding a protein or peptide, said protein or peptide comprising a stabilon sequence immediately following a N- terminal initiation methionine residue that renders the protein or peptide metabolically stable with respect to the N-end rule of proteolytic degradation.

14. The vector according to claim 13, wherein said gene encoding a protein or peptide which further comprises a protease-sensitive linker region immediately followed by a destabilizing sequence, wherein said protease-sensitive linker region is disposed immediately after said stabilon sequence .

15. The vector according to claim 14, wherein said nucleotide sequence further contains an inducible promoter operably linked to a gene encoding a restriction protease which cleaves the said protease-sensitive linker region.

16. The vector according to claim 13, wherein said gene encoding said protein or peptide further encodes an intracellular localization sequence.

17. The vector system of claim 16, wherein the nucleotide sequence contains genes encoding more than one antibody to the same target, wherein the antibodies have different intracellular localization sequences and target the antigen at different intracellular locations.

18. The vector according to claim 13, wherein said protein or peptide encodes an antibody or fragment thereof capable of binding to an intracellular target antigen.

19. The vector system of claim 18, wherein the target antigen is selected from the group of antigens consisting of intermediate metabolites, sugars, lipids, autocoids, hormones, complex carbohydrates, phospholipids, nucleic acids and proteins.

20. The vector system of claim 18, wherein the target antigen is a hapten, an RNA sequence, a DNA sequence or a protein.

21. The vector system of claim 18, wherein the target antigen is a virally-encoded protein, or a protein whose expression results in malignant cellular transformation.

22. The vector system of claim 21, wherein the target antigen results in malignant transformation as a result of overexpression of the protein, an HTLV-1 protein or an HIV viral encoded protein.

23. The vector system of claim 22, wherein the antibody is an antibody capable of binding to the envelope glycoprotein or the capsid protein.

24. The vector system of claim 23, wherein the target antigen is the envelope gpl60 or the envelope gp41.

25. The vector system of claim 20, wherein the target antigen is a TAR element or a RRE sequence.

26. The vector system of claim 18, wherein the nucleotide sequence contains genes encoding antibodies to more than one target antigen.

27. The vector system of claim 26, wherein the target antigens are virally-encoded proteins and the antibodies are to at least two different virally-encoded proteins .

28. The vector system of claim 27, wherein the virally-encoded proteins are HIV encoded proteins and the antibodies are to at least one structural protein and at least one regulatory protein.

29. The vector system of claim 28, wherein the structural protein is an envelope glycoprotein and the regulatory protein is either the tat or rev protein.

30. The vector system of claim 29, wherein the antibody is to that portion of the capsid protein involved in myristylation.

31. The method according to claim 1, wherein the protein or peptide is expressed under the control of an inducible promoter.