CA2236036A1 - Signal-regulated, cleavage-mediated toxins - Google Patents

Abstract

Disclosed is a novel class of molecules referred to herein as sitoxins (signalrelated, cleavage-mediated toxins. A sitoxin combines several functional domains to produce a therapeutically effective molecule. More specifically, a sitoxin is comprised of an effector domain; a domain bearing an intracellular signalling moiety; and a domain located between the effector domain and the domain bearing the intracellular signalling moiety which specifies a cleavage site for a predetermined protease. Also disclosed are methods for selectively killing a target cell which is known to contain a predetermined protease. Such methods involve the introduction of a sitoxin into a target cell. The sitoxin can be introduced directly, or through the intermediacy of an expression vector. Following the introduction of the sitoxin into the target cell, cleavage by the predetermined protease separates the effector domain of the sitoxin from the intracellular signalling moiety, resulting either in a longerlived (and therefore more toxic) effector domain or in an effector domain that moves from a cellular compartment where the domain is nontoxic to a cellular compartment where the domain is able to exert its effect.

Description

W O 98/03538 PCTrUS97/10941 SIGNAL-REGULATED, CLEAVAGE-MEDIATED TOXINS

Background of the Invention Virus-infected cells differ from their uninfected counterparts in particular by the presence of virus-specific proteins. The levels of certain cellular proteins are also altered as a result of viral infection.
Most of the relatively few proteins that are encoded by a viral genome have functional counterparts in cells that a virus infects. In part for this reason, effective antiviral drugs remain, by and large, a goal to be reached.
A variety of antiviral therapies are currently in use. At the present time, the sole highly efficacious antiviral therapy is based on vaccination against specific viruses. In many (but not all) cases this strategy works well, in that an individual vaccinated against a specific virus becomes immune to a subsequent infection by this virus. Unfortunately, the high mutation rate of many viruses often renders an otherwise effective vaccine ineffective against new variants of a virus. Also, and crucially, while vaccines can prevent a viral infection, in most cases they cannot eliminate an established infection.
The existing therapies against an ongoing viral infection include the following strategies:
(1) activation of natural (immune system-mediated) antiviral responses through the administration of specific cytokines such as, for example, interferons;

(2) inhibition of the spread of an infection to uninfected cells through the blocking of receptors for a virus on the surface of susceptible cells, or through the blocking of receptor-binding proteins on the surface of virions; and t (3~ disruption of a viral intracellular reproduction cycle, comprising the steps of:
(a) administering a drug that kills infected cells early enough in the viral reproduction cycle to s prevent or reduce the formation of mature (infectious) virions; or (b) administering a drug that does not necessarily kill an infected cell but acts by directly suppressing the formation of infectious virions through the inhibition of a step in the viral reproduction cycle.
An example of the strategy outlined in item 3a is the treatment of herpesvirus-infected cells with ganciclovir (a purine nucleoside analog). Thymidine kinase (TK) of a herpesvirus phosphorylates ganciclovir more efficiently than the cellular TK. The resulting ganciclovir monophosphate is then converted by cellular kinases into ganciclovir triphosphate, which inhibits viral DNA replication and also (less strongly) cellular DNA replication; the latter effect results in death of a herpesvirus-infected cell at a pre-virion stage of the viral reproduction cycle. Thus, cells that are infected with a herpesvirus such as, for example, herpes simplex virus or cytomegalovirus, are more sensitive to killing by ganciclovir than their uninfected counterparts.
Treatments with ganciclovir or its analogs are among the few relatively efficacious antiviral therapies available at the present time. Unfortunately, these therapies are limited to herpesviruses and closely related viruses.
Moreover, since the cellular TK enzymes also phosphorylate ganciclovir (albeit less efficiently than herpesviral TK), a systemic treatment with ganciclovir or analogous drugs is accompanied by a multitude of side effects that confine this class of treatments largely (though not exclusively) to herpesviral infections of skin and mucosal surfaces.

W098/03538 PCT~S97/1~41 An example of the strategy outlined in item 3b is the treatment of cells infected by a retrovirus such as, for example, the human immunodeficiency virus (HIV), with azidothymidine (AZT). This analog of thymidine is phosphoryIated ln vivo by cellular enzymes; and in the form of a triphosphate preferentially inhibits viral reverse transcriptase. In addition, AZT that becomes incorporated into a growing DNA chain terminates further chain growth, thereby enhancing the overall inhibitory effect of AZT on the viral reproduction cycle.
Unfortunately, AZT also interferes with normal cellular functions, including cellular DNA replication, which limits its utility as an antiviral drug, and in most cases (including AIDS) renders AZT incapable of effecting a cure at doses that do not cause unacceptable side effects.
Another example of the strategy outlined in item 3b is the inhibition of the HIV processing protease (P~).
By preventing proteolytic processing of viral polyproteins, the recently developed PR inhibitors efficiently suppress the formation of infectious HIV
virions at doses that do not cause unacceptable side effects. This otherwise promising therapy does not fully address the problem of HIV variants that are selected in the course of anti-PR treatment and express inhibitor-resistant mutant PRs.
Thus, a major limitation of present-day therapies against an ongoing viral infection is an insufficient selectivity of a drug, or the emergence of drug-resistant viral mutants or (most often) both - it is these difficulties which account in part for the fact that there are no efficacious, curative drugs against a majority of pathogenic human or animal viruses, including many deadly ones. The relatively few antiviral therapies that do exist (for example, ganciclovir-based anti-herpesvirus therapies or AZT-based anti-HIV

W098/03538 PCT~S97/10941 therapies) are in most cases not curative in the presence of a systemic viral infection.

SummarY of the Invention The present invention relates to a novel class of molecules referred to herein as sitoxins (signal-related, cleavage-mediated toxins ), and to nucleic acids encoding the molecules. A sitoxin combines several functional domains to produce a therapeutically effective molecule.
More specifically, a sitoxin is comprised of an effector domain; a domain bearing an intracellular signalling moiety; and a domain located between the effector domain and the domain bearing the intracellular signalling moiety which specifies a cleavage site for a predetermined protease.
In another aspect, the invention relates to methods for selectively killing a target cell which is known to contain a predetermined protease. Such methods involve the introduction of a sitoxin into the target cell. The sitoxin can be introduced directly, or through the intermediacy of an expression vector. Following the introduction of the sitoxin into the target cell, cleavage by the predetermined protease activates the effector domain of the sitoxin.
Brief Description of Drawin~s Figure l. Degron-based sitoxins: (A) A degron-based sitoxin contains a toxic effector domain, an N-terminal domain bearing a degradation signal D (any degradation signal that results in the processive intracellular degradation of sitoxin), and the "target"
domain Tl containing a site that can be recognized and cleaved by a viral processing protease Pl. (B) The same design as in A but with the N-degron as the degradation signal.

W O 98t03538 PCT~US97/10941 Figure 2. The nuclear localization signal (NLS)-based sitoxin is analogous to the construct in Fig.
lA but contains an NLS instead of degron in the N-terminal part of the fusion.
.
Detailed DescriPtion of the Invention The present invention is based on the discovery of a new and generally applicable strategy for eliminating, or modifying, cells of a multicellular organism that are known to contain a predetermined protease (e.g., virus-infected cells) in the absence of a significant damage to unintended targets that do not contain the predetermined protease (e.g., cells not infected by the aforementioned virus).
The main idea of this strategy is illustrated, for example, in Fig. 1, which shows a tripartite fusion protein comprising: 1) an intracellular signalling domain (a degradation signal (degron) is illustrated); 2) a protease cleavage site; and 3) an effector domain (a toxic domain is illustrated). As discussed more fully below, a degron, in particular the N-degron, is an intracellular signalling domain which targets a protein for rapid destruction. Thus, the tripartite fusion protein of Fig. 1 is rapidly degraded, and is therefore nontoxic to a cell, provided that a degron remains linked to the toxic domain.
However, if the tripartite fusion protein is introduced into a cell which contains a protease that specifically recognizes and cleaves the fusion protein at the protease cleavage site, the degradation signal and the effector domain become unlinked. As a result, the cleavage by the protease greatly extends the intracellular half-life of the effector domain. The effect of ext~n~ing the half-life of a toxic effector domain, such as, for example, the ricin A-chain, is death of the cell which carries it. Conditionally toxic protein reagents of this new class will be referred to as sitoYin~ (signal-regulated, cleavage-mediated toxin~).

W O 98/03538 PCT~US97/10941 Sitoxins can be produced by any of the known methods for producing an amino acid copolymer having a predetermined sequence identity. However, the preferred method for constructing and producing a sitoxin employs recombinant DNA techniques. Through the use of such conventional techniques, DNA encoding the required sitoxin elements is isolated from a biological source or sources and modified as necessary using standard techniques such as site-directed mutagenesis.
Preferably, the minimum number of nucleotides required to encode an amino acid sequence (e.g., a peptide, polypeptide or protein (these terms will be used interchangeably in the present application)) that confers the required function is employed. Since each of the required elements of the sitoxin molecule is easily assayable, it is a matter of routine experimentation to determine the minimum length of a DNA fragment which will encode a functional sitoxin element. The resulting DNA
fragments are linked together, using standard recombinant DNA techniques, yielding an open reading frame which is translated into a fusion protein comprising all of the required sitoxin elements. Over the last decade, many amino acid sequences have been shown to be usable as interdomain linkers - sequences that form a hydrophilic and flexible segment of the polypeptide chain relatively resistant to endoproteases present in the bloodstream, intercellular spaces and inside the cells. A
particularly extensive collection of suitable linkers emerged from designs of single-chain antibodies, in which a linker sequence connects a light-chain antigen-binding domain to an analogous heavy-chain domain.
The open reading frame, prepared as described above, is inserted into a DNA expression vector which includes a transcriptional promoter and other sequences required for expression. The choice of expression vectors from among the many available options is largely dependent upon the cell type in which expression is desired. As discussed W098/03538 PCT~S97/10941 more fully below, eukaryotic expression vectors are preferred for many applications.
Alternatively, a sitoxin can also be produced by expressing it, through the intermediacy of prokaryotic expression vectors, in bacteria (e.g., E. coli), purifying the resulting overexpressed protein, and contacting the purified protein with target cells directly. When produced for use in this manner, a sitoxin should also bear an additional domain that enables its translocation into the cell's cytosol. (For purposes of clarity it is noted that the term "cytoplasm"
denotes the interior of a cell outside of its nucleus, while "cytosol" is the cytoplasmic milieu outside of many membrane-enclosed compartments (including the nucleus) that reside in the cytoplasm.) The use of such "translocation" domains, present in a variety of natural toxins such as, for example, whole ricin, whole Pseudomonas exotoxin A, and whole diphtheria toxin, is widely reported in the prior art. Typically, such reports relate to conventional, present-day chimeric toxins that recognize cell surface markers (Vitetta et al ., Imm . Today l~: 252 (1993); Pastan et al ., Annu. ~ev.
Biochem. 61: 331 (1992)). The fundamental, qualitative difference between the current chimeric toxins and sitoxins of the present invention is the sensitivity of sitoxins to intracellular ~as distinguished from cell surface) molecular targets, in contrast to the present-day chimeric toxins whose limitations stem in part from the confinement of their selectivity to the surface of a cell - from their inability, upon entering the cell, to "adjust" their toxicity in response to the intracellular protein composition.
An effector domain is preferably a protein or polypeptide that is able to exert a specific effect (e.g., to cause the death of a cell, or to disrupt a viral replication cycle) when the effector is delivered to a predetermined intracellular location. In W098/03538 PCT~S97/10941 particular, the effector domain of a sitoxin can be derived from a protein or polypeptide which acts as a toxin when delivered to a cell. Many such toxins are known in the art, including, for example, the A-chain of s ricin (and analogous plant toxins), the toxic domain of the Pseudomonas exotoxin and the toxic domain of diphtheria toxin. As discussed previously, the intracellular compartment specificity of certain toxic proteins or polypeptides can be exploited in connection with sitoxin designs in order to alter the selectivity of a sitoxin whose substrates are located in the cytosol but not in the nucleus. Such proteins or polypeptides include, for example, the diphtheria toxin and the Pseudomonas exotoxin A, both of which inhibit protein synthesis by ADP-ribosylating (and thereby inactivating) elongation factor 2 (EF2). Since the bulk of EF2 is cytosolic, the translocation of sitoxin containing a Pseudomonas-type toxic domain from the cytosol to the nucleus would physically separate a toxin from its substrate.
An effector domain can be derived from any protein or polypeptide the introduction of which into a predetermined cellular location would cause cell death.
For example, a deoxyribonuclease would act as a toxic effector domain if introduced into the nucleus (but not into the cytosol) of a target cell. Indeed, ECoRI has been shown to cleave nuclear DNA in ~ivo ( in the yeast Saccaromyces cerevisiae), killing the cells (Barnes and Rine, Proc. Natl . Acad . sci. USA 82: 1353 (1985)).
Examples of toxic domains whose substrates are present in both the cytosol and the nucleus, include, in addition to the A-chain of ricin, ribonucleases such as RNAase A and barnase, which have been used to produce conventional (cell surface-recognizing) chimeric toxins (Rybak et al., ~. Biol . Chem. 266: 21202 (l99l); Prior et al., Cell 64 :
1017 (l99l)).

W098/03538 PCT~S97/10941 _g_ Since a major aspect of the present invention relates to antiviral therapy, the goal of which is to eliminate an ongoing viral infection, the set of useful effector domains is not confined to cytotoxic proteins.
For example, an effector domain of a degron-based sitoxin can be an enzyme whose activity perturbs the viral reproduction cycle while not perturbing the viability of a cell to a similar extent. In an uninfected cell, a degron-based sitoxin of this type would be (by definition) short-lived and therefore nontoxic to the cell, whereas in a virus-infected cell the effector domain of a sitoxin would be separated from the sitoxin's degron (Figure lA), rendering the effector domain long-lived and toxic to a virus; whether the same effector domain would also be toxic to the infected cell would be determined by the intrinsic selectivity of the effector domain. Cytotoxic effector domains of a sitoxin are emphasized in the present invention because they provide a sufficient solution to the problem of stopping the spread of an ongoing viral infection. Indeed, since the processing proteases of most viruses are produced early in the viral reproduction cycle, an antiviral sitoxin bearing a conditional cytotoxic domain (Figures lA and 2) would be activated by a viral processing protease early in the infection cycle. As a result, an infected cell would be killed before the bulk of mature (infectious) virions could form. In addition to the use of protein or peptide effector domains, sitoxins may also employ small molecular weight compounds as cytotoxic effectors. An example of such a compound, methotrexate, in the context of a sitoxin, is discussed below.
With regard to the intracellular signal component, a variety of such signals have been described in the literature. For example, a short in vivo half-life can be conferred on a protein by one or more of distinct degradation signals, or degrons (Varshavsky, Cell 64: 13 (1992); Hershko & Ciechanover, Ann. Rev. Biochem. 61: 761 (1992)). The best understood intracellular degradation W O 98/03538 PCT~US97/10941 signal is called the N-degron (Varshavsky, Cell 64: 13 (1991)). This signal comprises a destabilizing N-terminal residue and an internal lysine (or lysines) of a protein substrate. A set of N-degrons bearing different destabilizing residues is referred to as the N-end rule -a relation between the in vivo half-life of a protein and the identity of its N-terminal residue. The lysine residue of an N-end rule substrate is the site of formation of a multiubiquitin chain, which is required for the substrate's degradation.
Ubiquitin is a protein whose covalent conjugation to other proteins plays a role in a number of processes, primarily through routes that involve protein degradation. The recognition of an N-end rule substrate is mediated by a targeting complex whose components include a ubiquitin-conjugating enzyme (one of several such enzymes in a cell) and a protein called N-recognin or E3. A targeted, multiubiquitinated substrate is processively degraded by the 26S proteasome - a multicatalytic, multisubunit protease. Aspects of the N-end rule, ubiquitin fusions and related technologies are the subject of a number of U.S. Patents issued to Varshavsky et al., including U.S. Patent Nos. 5,132,213;
5,212,058; 5,122,463; 5,093,242 and 5,196,321, the disclosures of which are incorporated herein by reference.
The in vivo half-life of a protein bearing a strongly destabilizing N-terminal residue such as arginine can be as short as 1 minute, whereas an identically expressed and otherwise identical protein bearing a stabilizing N-terminal residue such as valine has a half-life of more than 20 hours, resulting in a greater that 1,000-fold difference between the steady-state concentrations of these proteins in a cell.
Ubiquitin-dependent proteolytic systems (including the N-end rule pathway) share many components of the 26S
proteasome. Differences among these systems encompass W O 98/03538 PCTrUS97/10941 their distinct targeting complexes, whose recognins bind to degradation signals other than N-degrons. In addition to the N-degron, the amino acid sequences that function as degradation signals and are transplan~able to other proteins including, for example, the "destruction boxes"
of short-lived proteins called cyclins (Glotzer et al., Nature 349: 132 (1991); Ciechanover, Cell 79: 13, 1994)) and two specific regions of Mat~2, the short-lived transcriptional regulator of S. cerevisiae (Hochstrasser and Varshavsky, Cell 61: 697 (1990)).
The spectrum of intracellular signals that are useful in the design of sitoxins is not confined to degradation signals. For reasons analogous to those considered above for degrons, the signals that confer on a protein the ability to enter membrane-enclosed compartments (Schatz, Protein sci. 2: 141 (1993); Osborne and Silver, Annu . Rev . Biochem . 62: 219 (1993); and Dingwall and Lasky, Trends Biochem. sci. 16: 478 (1991)) can also be employed. The discussion herein is confined to nuclear localization signals (NLSs) but is also relevant to sitoxins bearing other portable signals, in particular other translocation signals.
Proteins smaller than -60 kD can enter the nucleus by diffusing through the nuclear pores, but the pore-mediated transport of a larger protein requires thepresence of at least one NLS accessible to components of the nuclear translocation system. NLSs are short sequences (10-20 residues) rich in lysine and arginine;
their steric accessibility in a target protein appears to be sufficient for their activity as nuclear translocation signals.
An NLS-based sitoxin is a fusion that includes (1) a toxic domain that exerts its effect exclusively or preferentially in the cytosol (see below); (2) an NLS-bearing domain; and (3) a recognition/cleavage site for aspecific viral processing protease inserted between the toxic domain and the NLS-bearing domain of the fusion (Figure 2). The mechanism of an NLS-based sitoxin is as follows. The introduction of an NLS-based sitoxin into an uninfected cell would result in a rapid, NLS-mediated translocation of the NLS-sitoxin into the nucleus, where the sitoxin's toxic domain, being cytosol-specific, would be unable to exert its toxic effect. However, in a virus-infected cell the same NLS-bearing sitoxin would be cleaved in the cytosol by a viral processing protease (at the recognition/cleavage site present in the sitoxin), resulting in the physical uncoupling of the NLS and the toxic domain of the initial protein fusion. ~aving lost the NLS, the toxic domain of sitoxin would remain in the cell's cytosol and exert its toxic effect, killing the cell. The net result of the predominantly cytosolic localization of an NLS-based sitoxin in virus-infected (but not in uninfected cells) would be its selective toxicity against virus-infected cells (Figure 2).
It should be noted that the selectivity of an NLS-based sitoxin can readily be "inverted" by employing a toxic domain that is active in the nucleus but not in the cytosol. Such a sitoxin would kill cells that lack a protease that cleaves a sitoxin between its NLS and toxic domain, but would spare cells that contain such a protease.
~he region (domain) of a sitoxin between its effector domain and the domain bearing the intracellular signaling moiety specifies the cleavage site for a predetermined protease. It is this protease-specific cleavage site which enables the constructs (sitoxins) of the present invention to become activated in a cell bearing the predetermined protease and to remain relatively inactive in a cell lacking such a protease. A
particularly well-defined group of proteases that are relevant to the present invention are virus-encoded processing proteases, whose main (and often the only) normal function is to convert a viral polyprotein - the initial translation product - into mature viral proteins, in particular the enzymes for viral replication, as well W098/03538 PCT~S97/10941 as structural proteins of the virions. While not all of the known animal and plant viruses depend on the proteolytic processing of polyprotein precursors for infectivity, a majority of them do (Krausslich and Wimmer, Annu. Rev. Biochem. 57: 701 (1988); Dougherty and Semler, ~icrobioi. Rev. 57: 781 (1993)).
Although the processing proteases of different viruses form a family that is diverse both structurally and mechanistically, most of these enzymes share at least the following properties:
(1) relatively small size (e.g., about 10-50 kD);
(2) proteolytic activity which does not require the presence of ATP or other high-energy cofactors;
(3) a viral protease is often encoded as a part of a po~yprotein precursor, from which the protease excises itself through the same pathway of sequence-specific cleavages that yields other mature protein products of the polyprotein;

(4) at any given cleavage site in a polyprotein, a viral protease typically makes just a single, sequence-specific cut, the exact location of which is dictated by short (e.g., approximately 10 residues) amino acid sequences on both sides of (and immediately adjacent to) the cleavage site; and (5) the relative efficiency of the protease-mediated cleavage at a given site is determined by the specific amino acid sequences flanking the site, and also by the extent of steric shielding of the site within a polyprotein - as a result, different mature proteins can be excised by the protease from the polyprotein precursor at different, physiologically relevant intrinsic rates.
For the purpose of the present invention, the crucial property of a viral processing protease is its ability to recognize and cleave within an amino acid sequence motif that is sufficiently short (e.g., approximately 20 residu~s) to be readily porta~le W098/03S38 PCT~S97/10941 (transplantable) to artificial (engineered) protease substrates and, at the same time, sufficiently long to be a unique site, in the sense that such a site would be left uncleaved by other intracellular (cytosolic) proteases in an uninfected cell that lacks the viral protease. Indeed, the extensive body of evidence for many viruses, including medically relevant ones, directly shows that either viral polyproteins or engineered substrates of specific viral proteases are not cleaved in uninfected cells that lack such a protease, while they are efficiently cleaved (processed) in virus-infected cells (reviewed by Krausslich and Wimmer, Annu. Rev.
Biochem. 57: 701 (1988); Dougherty and Semler, Microbiol.
Rev. 57: 781 (1993)). The crucial idea of the present invention is to combine the use of degrons or translocation signals for changing the state of a cytotoxic protein in a cell with the possibility of making this change of state conditional on the presence of a virus-specific processing protease. The latter is achieved through the utilization of the proteases's ability to cleave a protein bearing an appropriate, protease-specific cleavage site.
The most important factor considered in selecting an appropriate viral protease recognition sequence is the identity of virus of interest. Once the identity of the virus of interest is determined, one of skill in the art can consult the literature to determine an appropriate protease recognition sequence to be included as a component of the sitoxin. Two recent review articles containing such information, as well as citations to additional sources of such information, are Krausslich and Wimmer (Annu. Rev. Biochem. 57: 701 (1988)) and Dougherty and Semler (Microbiol. Rev. 57: 781 (1993)).
The specific protease recognition sequence which can be incorporated in the sitoxin mo~ecule is not limited, however, to those specifically discussed in these references or the citations contained therein.

W098/03538 PCT~S97/10941 The typical in vivo application of a sitoxin is in a therapeutic regimen designed to eliminate cells known to contain a predetermined protease. Consider, for example, a situation wherein a viral infection had been diagnosed in a patient, and a sitoxin whose protease cleavage site is recognized by a protease encoded by the virus has been designed as described above. (A large fraction of pathogenic human and animal viruses utilize polyproteins and a virus-specific protease as an essential part of their life cycle.) Two alternative delivery modes for this sitoxin are (l) direct delivery of sitoxin as a pre-made protein; or (2) deli~ery of sitoxin through the intermediacy of an expression vector that encodes this sitoxin. The mechanistic routes of comtoxin delivery can be either intravascular (for both protein-based and vector-based forms of sitoxin) or any other route (for example, a skin or mucosal surface application) that would serve to deliver a sitoxin to the bulk of virus-infected cells.
For a direct (protein-based) delivery of sitoxin via, for example, the intravascular route (Gilman et al., Eds., The Pharmacological Basis of Therapeutics (Pergamon Press, New York, l990)), the latter should possess not only a toxic domain but also a domain that mediates the translocation of a protein fusion from the cell surface to the cytosol. This aspect of sitoxins is confined to direct-delivery (as distinguished from expression-based) strategies, and is similar to the analogous aspects of current chimeric toxins discussed above (Vitetta et al., Imm. Today 14: 252 (1993); Pastan et al., Annu. Review Biochem. 61: (1992); and Olsnes et al., Som. Cell Biol.
2: 7 (1991)). If necessary, the initial step of a sitoxin's delivery as a protein can be made partially cell type-selective by fusing the sitoxin to a domain such as, for example, an antibody that binds to a surface marker on target cells. Crucially, and in contrast to the situation with~urrent chimeric toxins, the surface marker can be present on more that just target cells W O 98/03538 PCT~US97/10941 without significantly increasing the sitoxin's nonspecific toxicity. Indeed, the conditional toxicity of a sitoxin is decided by the environment it encounters after entering the cell; therefore, a sitoxin against a specific virus would not affect uninfected cells, which lack the virus-specific intracellular protease characteristic of cells infected by the virus.
The advantage of the direct-delivery strategy is that it bypasses potential problems associated with the vector-mediated delivery of a sitoxin. One potential drawback of the direct delivery is a larger size of a multidomain sitoxin (due to the presence of additional domains), in comparison to an otherwise identical sitoxin that is delivered through the intermediacy of an expression vector. Since the testing of sitoxins using either of these delivery strategies is technically straightforward, involves exclusively the existing technologies, and can be assessed directly and objectively, a prudent experimental approach would be to use both strategies in evaluating a given sitoxin, and to compare the results.
Both the direct (protein-based) and indirect (vector-based) delivery approaches are a part of ongoing efforts to improve bioavailability of protein drugs used in medical interventions, from cytotoxic treatments to gene therapies. The problem of insufficient selectivity is common to all of the current cytotoxic strategies:
once the effector reaches its intended intracellular compartment, the cell is likely to be killed irrespective of whether it was a target or an innocent bystander. For example, one difficulty with the current chimeric toxins is their nonspecific toxicity - largely, but not only, to the liver. This toxicity, which imposes a limit on both duration and intensity of treatments, stems in part from the clearance of an intravenously administered immunotoxin by cells of the reticuloendothelial system that are killed as a result of the toxin~s eintry into WOg8/03538 PCT~S97tlO941 these normal cells (Vitetta et al ., Imm. ~oday 14 : 252 (1993)).
By contrast, even nonselective delivery of the sitoxins of the present invention would not affect most nontarget cells. This feature of sitoxins will yield a much higher therapeutic index (i.e., a much higher tolerated intensity and duration of treatments).
Sitoxins designed for delivery by an expression vector would lack the "compartment-crossing" domain lo required for their directly delivered counterparts. The vector can be either a retroviral vector, adenoviral vector, another viral vector, or simply naked DNA within a gene delivery system, for example, a liposome-based delivery system. Both viral vectors and liposome-based lS gene delivery systems have been successfully used in approaches to gene expression in whole animals. Several types of vectors for gene therapy and other applications are already available (reviewed by Yee et al., Proc.
Natl . Acad . Sci . USA 91 : 9564 (1994); see also Mulligan, Science 260: 926 (1993) and Anderson, Science 256: 808 (1992)). These vectors can be used for the delivery of sitoxins in cell cultures, in whole animals, and, with appropriate preliminary testing, in human patients as well. Recent advances in the design of viral and plasmid-based vectors (see e.g., ~abel et al., Proc.
Natl. Acad. sci. USA 90: 11307 (1993) and Mulligan, Science 260: 926 (1993)) resulted in tailor-made, nonreplicating vectors that can transfect both growing and quiescent cells, and are either specific for cells that bear a predetermined surface marker or almost nonselective. Such vectors, in use for gene therapy and other applications (for example, Mulligan, Science 260:
926 (1993) and Anderson, Science 256: 808 (1992)), are powerful vehicles for the delivery of sitoxins.
Sitoxins can also be used in a variety of in vitro (cell culture-level) applications. A common feature of these applications is based on the ability of sitoxins to selectively eliminate virus-infected cells from a cell CA 02236036 l998-04-27 W O 98/03538 PCTrUS97/10941 population that contains both infected and uninfected cells. For example, in a commercial production setting, sitoxins can be employed to purge a valuable cell culture from cells that have been infected with a specific virus.
Since many viruses, including retroviruses, grow slowly and do not kill (or do not kill quickly) an infected cell, the ability of a sitoxin to selectively kill cells infected with a specific virus should make them valuable reagents for cell culture production.
An analogous but clinical application of sitoxins that is likely to prove especially valuable is to use them for selective elimination of virus-infected cells from, for example, the patient's bone marrow ~ for subsequent reinfusion of the virus-free marrow into the same or another (immunologically compatible) patient whose own bone marrow was deliberately destroyed by a marrow-ablating chemotherapy. (This approach is frequently used in the contemporary cancer therapy, and is often applied to patients who, in addition to a specific cancer, may also harbor a viral infection.) The sitoxin-mediated purging of either bone marrow or another medically relevant subpopulation of patient's cells from specific viruses should be of value both in the cases where the marrow is known in advance to be compromised by a specific viral infection (e.g., the AIDS patients) and in the cases where a viral infection might be suspected but is difficult to demonstrate unambiguously. The crucial property of sitoxins - their ability to kill cells infected bearing a specific virus while sparing uninfected cells - makes possible such prophylactic purgings of cell populations that one would like to be definitely free of a virus prior to a "downstream"
utilization of the cells, especially if this utilization involves the infusion of cells into a patient.

W098/03538 PCT~S97/10941 One remarkable advantage of sitoxins is that the probability of emergence of a sitoxin-resistant mutant virus is expected to be negligible. Indeed, since the cleavage site of an antiviral sitoxin is functionally s indistinguishable from the natural cleavage sites in a viral polyprotein (Figs. l and 2), and since the processing of polyprotein is essential for infectivity of the virus, mutations of the viral PR (including those that make it resistant to PR inhibitors) cannot produce a sitoxin-resistant virus without rendering it uninfectious as well. Therefore the emergence of inhibitor-resistant proteases - a major problem with the present-day drugs (inhibitors) that target viral proteases such as HIV PR -cannot occur with sitoxins, for the reasons stated above.
The sitoxin concepts disclosed herein (Figures l and 2) are compatible with a low molecular mass design as well. For example, the effector domain of a sitoxin construct can be a low molecular mass cytotoxic compound such as, for example, methotrexate (MTX). This compound would be linked to a short polypeptide (less than -15 residues in length) that contains the recognition/cleavage site for a viral processing protease, such as, for example, the HIV protease. The cleavage site-containing peptide would be linked to MTX
at a position that renders MTX inactive for the binding to its target - the cellular enzyme dihydrofolate reductase (DHFR). Thus, in its initial state, this low molecular mass sitoxin construct would be nontoxic. In a cell which lacks the relevant viral processing protease this sitoxin construct would remain nontoxic. By contrast, in a cell containing the viral protease, the above construct will be cleaved, resulting in MTX bearing a shorter peptide (or peptide-like) extension that corresponds to the C-terminal part of the initial recognition/cleavage site. The additional re~uirements for this design (this requirement is unnecessary for the designs of degron- or NLS-based sitoxins) is that the W098/03538 PCT~S97110941 remainino peptide (or peptide-like~ extension attached to MTX after the above cleavage event must be readily degradable by one of the ubiquitous exopeptidases normally present in the cell's cytosol. This degradation would release free (and toxic) MTX in cells containing the viral protease but not in cells lacking the protease, because the initial MTX-linked peptide (or peptide-like) extension is constructed to be insensitive to cytosolic exopeptidases.
EXAMPLES

ConstrUction and Testin~ of Sitoxin Against Human ImmunodeficiencY Virus (HIV) The initial testing of efficacy and selectivity of sitoxins will take place not in the whole animal but in the technically less complex setting of a mammalian (human) cell culture, and before that in the even simpler, better controlled setting of a yeast (Saccaromyces cerevisiae) cell culture. once a sitoxin that successfully performs in a cell culture has been designed, it can be tested in experimental animals, and subsequently in humans. With cell cultures, the vector-mediated delivery of sitoxin is technically straightforward, because several high-efficiency expression vectors and protocols for transfecting mammalian cells (including human cells) are available.
In yeast, this set of approaches and procedures has been developed to an even higher level of refinement and ease of use.
Briefly, the initial testing of a sitoxin specific for a predetermined virus would ask whether this sitoxin would kill cells expressing the viral processing protease while sparing otherwise identical cells that lack the protease. At this stage of the testing, the yeast-based cell system, with its powerful "reverse-genetics"
techniques, will be ur ~. Once a working si~oxin design has been established a the level of yeast cell culture, W098/03538 PCT~S97/10941 the analogous tests will be carried out with human cells, as described below.
The sitoxin to be designed in this example is a multidomain protein fusion whose C-terminal domain is the A-chain of ricin, linked to the nearest upstream domain Tl (Figure lB) by a short (5 to lO residues) linker sequence. The domain Tl contains a cleavage site for a processing protease of a specific virus, as described below. Simple-sequence, relatively hydrophilic linkers (see, e.g., Johnson and Bird, Neth. Enzymol . 203 : 88 (l99l)) will be used to join together individual domains of sitoxin. Finally, the N-degron, whose design and construction are well known in the art (Varshavsky, Cell 69: 725 (1992)), will be positioned at the N-terminus of a sitoxin, as described below.
The nucleic acids which encode the various elements of the sitoxin molecule will be isolated from naturally occurring sources (or engineered and publicly available sources) and assembled using standard recombinant DNA
techniques and either yeast or mammalian expression vectors described below. The toxic domain to be employed will be the ricin A-chain (other cytotoxic effectors, for example, the toxic domain of the Pseudomonas exotoxin or the toxic domain of the diphtheria toxin, can also be used to construct a sitoxin). The deduced amino acid sequence of the A-chain of ricin is provided, for example, in Funatsu et al., Biochimie 73: 1157 (l99l).
In this construction, no effort will be made to reduce the size of the toxic domain by deletion. Specifically, a cDNA-containing fragment encoding the complete ricin A-chain will be isolated from the plasmid pRAP229 (Ready et al., Proteins 10: 270 (l99l)) using conventional te~-hniques.
The degradation signals (degrons) to be positioned upstream of both the toxic domain and the Tl (cleavage site-containing) domain (Figure l) can be chosen from among several presently known degrons. The degradation signal to be employed in the first construct will be the W O 98/03538 PCTrUS97/10941 N-degron (see above), which has been dissected biochemically and genetically, and is understood in considerable detail (Varshavsky, Cell 69: 725 (1992) ) .
Several portable variants of the N-degron have been 5 described and analyzed (Varshavsky, Cell 69: 725 (1992)).
The actual portable N-degron employed in the construction below will be the one described by Bachmair and Varshavsky, Cell 56: 1019 (1989). This N-degron comprises a destabilizing N-terminal residue such as lo arginine (Arg), followed by the 45-residue sequence derived from E. coli Lac repressor. The corresponding procedures, sequences and plasmids, together with their restriction maps, are described in detail by Bachmair and Varshavsky, Cell 56: 1019 (1989). The construction of the T1 (cleavage site-containing) domain between the sitoxin's C-terminal toxic domain and the N-degron is described below.
Among the many possible choices of a specific virus for the first anti-viral sitoxin, initial experiments focus on a sitoxin specific for the human immunodeficiency virus (HIV). First, HIV is a classic example of a virus whose reproduction cycle involves the production of intracellular (cytosolic) polyprotein precursors of the mature HIV proteins and the cleavage of these precursors by the HIV protease (PR), which itself, initially, is a part of the precursor polyprotein (Tomasselli and Heinrikson, Meth. Enzymol. 241: 279 (1994)). Second, HIV is an ongoing, unsolved medical problem of major proportions, where a successful sitoxin design can be utilized immediately after its initial testing with cell cultures.
The A-chain of ricin as a toxic domain of the HIV-specific sitoxin is described above. Also described above are the design and sources of the N-degron that will be used to confer conditional instability of the HIV-specific sitoxin. ~hese two elements are generic elements of any N-degron-, ricin-based sitoxin. The HIV
specificity of this design will be conferred by its T1 W O 98/03538 PCTrUS97/10941 region (Figure lB), which will contain a cleavage site for the HIV protease. The T1 region can be as short as 20 residues, with the cleavage site in the middle of the region, the region itself being joined to the N-degron upstream of T1 and to the A-chain of ricin downstream of T1. The actual T1 region to be used can be chosen from several natural HIV protease cleavage sites in the HIV
polyproteins (Dougherty and Semler, Microbiol. Rev. 57:
781 (1993)). Many of these sites are cleaved at similar rates by the HIV protease. Initial experiments will employ site No. 6, the one between HIV protease and reverse transcriptase in the HIV precursor polyprotein (Debonck, AIDR Res. Human Retroviruses 8: 153 (1992)), but other sites of comparable efficiency can also be used to design the T1 region of HIV-specific sitoxin. The amino acid sequence abutting the HIV cleavage site No. 6 (specified in Dougherty and Semler, Microbiol. Rev. 57:
781 (1993)) (with the actual cleavage site between Phe and Pro) is sufficiently short to be easily constructed by synthesizing, using standard methods, an oligonucleotide encoding the specified sequence, and a complementary oligonucleotide. Annealing these oligonucleotides to form the DNA double helix one strand of which encodes the HIV cleavage site No. 6, and ligating the resulting short DNA fragments with other DNA
fragments that encode other components of the HIV-specific sitoxin described above would yield the desired protein fusion - the HIV-specific sitoxin ready for testing. None of these procedures and designs use anything but standard steps of the recombinant DNA
technology, well known to those skilled in the art.
Since an N-degron contains a destabilizing N-terminal residue, the ubiquitin fusion techniques can be used to produce such a residue at the N-terminus of HIV-specific sitoxin (Varshavsky, Cell ~9: 725 (1992)).
This technology is the subject of numerous U.S. Patents W098/03538 PCT~S97/10941 to Varshavsky et al. that are referred to above and are specifically incorporated by reference herein.
The fully assembled DNA open reading frame of the HIV-specific sitoxin will encode the following protein regions or domains, beginning from the N-terminus:
(l) u~iquitin;
(2) the N-degron described by Bachmair and Varshavsky, Cell 56: lOl9 (l989);
(3) the Tl region containing the HIV protease cleavage site described above and linked to the upstream and downstream regions of sitoxin through the linker sequences also described above; and (4) the toxic domain (A-chain) of ricin.
For expression in the yeast S. cerevisiae, the above open reading frame can be cloned into any of a number of readily available yeast expression vectors that allow regulated expression of the HIV-specific sitoxin.
Specifically, the pUB23 plasmid will be used (Bachmair et al., Science 234: 179 (1986)), which contains a galactose-inducible promoter and allows a tight regulation of the expression of HIV-specific sitoxin (no expression in the presence of glucose in the medium, and active expression upon the substitution of glucose with galactose). Two otherwise identical yeast strains will be used, one of which (the "control" strain) is one of the many "wild-type" S. cerevisiae strains, for example, S228c, while the other (the "test" strain) is an otherwise identical strain that constitutively expresses the HIV protease. This latter strain can be constructed using standard methods, well known to those skilled in the art. Specifically, the open reading frame encoding the HIV protease can be subcloned from the plasmid PRO4 (Strickler et al., Proteins 6: 139 (1989)). This open reading is placed downstream of a moderately active yeast promoter such as the CUPl promoter of the expression vector pRS314 (Sikorski and Hieter, Genetics 122: l9 (1989)). This construct, upon its introduction into the control yeast strain, will yield the test strain.

W098/03538 PCT~S97/10941 The plasmid encoding the HIV-specific sitoxin described above will then be introduced (using standard transformation techniques (Rothstein, R., Meth. Enzymol.
194: 281 (l99l)) into both the control and test yeast strains in the presence of glucose in the medium. Under these conditions the galactose-inducible promoter of the plasmid is repressed, resulting in negligible levels of HIV-specific sitoxin in the cells. At this point, both the control and test strains are transferred to a galactose-containing medium to induce the expression of the HIV-specific sitoxin. In both strains, the ubiquitin moiety at the N-terminus of the nascent HIV-specific sitoxin is rapidly (nearly cotranslationally) cleaved off by ubiquitin-specific proteases, which are present in all eukaryotic cells (Varshavsky, Cell 69: 725 (1992)), yielding the mature sitoxin (Figure lB) that bears a destabilizing residue such as, for example, Arg at the newly formed N-terminus. The presence of a destabilizing N-terminal residue and other elements of the N-degron 20 (see above) in the HIV-specific sitoxin renders it extremely short-lived in vivo, unless the N-degron is separated from the rest of the HIV-specific sitoxin in the test strain (but not in the control strain) through the cleavage at the Tl region of sitoxin (Figure lB) by 2 5 the HIV protease (note that the test strain, but not the otherwise identical control strain, expresses the HIV
protease).
As stated above, the A-chains of ricin and diphtheria toxin are toxic to both S. cerevisiae and mammalian cells, enabling a simpler yeast setting to be employed first. In a technically straightforward test, the control ("wild-type") S. cerevisiae strain will be compared with an otherwise identical strain that has been transformed with the above-described plasmid expressing HIV protease. Specifically, both strains will be also transformed with another of the above-described plasmids enroAing the N-degron-mediated sitoxin construct, both strains being grown on glucose-containing media, where W098/03538 PCT~S97/10941 the galactose-inducible, glucose-repressible promoter of the sitoxin-encoding plasmid is inactive. Thereafter both strains will be transferred to a galactose-containing medium, which induces the promoter and allows the expression of the sitoxin construct. In the HIV
protease-lacking strain, the sitoxin will be short-lived, owing to the presence of the N-degron, resulting in a low or negligible overall toxicity. By contrast, in the otherwise identical strain expressing HIV protease at least a fraction of the newly formed sitoxin will be cleaved by HIV protease, resulting in separation of the sitoxin's toxic domain and the N-degron (Figure l). The resulting "free" toxic domain will be a relatively long-lived protein (in comparison to the same domain bearing the N-degron), accumulating in a cell to a higher steady-state level and as a result causing a much greater overall toxicity. Experimentally, this difference will be manifested in the death of cells that express both the sitoxin and HIV protease versus survival of otherwise identical cells that express the sitoxin alone. The fate of cell populations in these experiments can be followed using several well-known and technically straightforward assays, for example, by determining growth curves of the corresponding cultures after their transfer to a galactose-containing medium (which induces sitoxin's expression), or, independently, by comparing the numbers of colonies formed by the above two strains under the same conditions. The in vivo half-life of sitoxin can be followed explicitly in such settings, if necessary, through the use of a standard pulse-chase assay.
Finally, if a modification of the sitoxin construct proves desirable, it can readily be accomplished using exclusively st~n~Ard recombinant DNA t~c~n;ques cited above.
Once the sitoxin design has been tested in the S.
cere~isiae setting, it will be used for a similar test in a human cell culture. Since an overexpression of HIV
protease has been shown to be toxic to mammalian cells W098/03538 PCT~S97/10941 (Vaishnav and Wong-Staal, Annu. Rev. Biochem. 60: 577 (l99l)), a stable human cell line will be constructed in which HIV protease is expressed from a tetracycline-repressible promoter, using the system developed by Gossen and Bujard (Proc. Natl. Acad. Sci. USA 89: 5547 (1992)). HeLa cells derivatives expressing the tetracycline-repressible transactivator protein (Gossen and Bujard (Proc. Natl . Acad. Sci . USA 89: 5547 (1992)), will be employed in these experiments. The use of Bujard's system, which is, by now, a standard method in the field of mammalian gene expression, makes possible a titration of the in vivo concentration of HIV protease to an optimal level for the tests with protease-expressing and protease-lacking cell lines - the tests closely analogous to those described above for S. cerevisiae.
These experiments will utilize transfections with plasmids encoding the corresponding proteins (HIV
protease and sitoxin; the sources of the corresponding open reading frames were given above, in the section describing experiments with S. cerevisiae.~ Among the many mammalian expression vectors suitable for these experiments is the pSG5 vector, sold by Stratagene Inc.
This vector is a 4.l kb E. coli.-~mm~lian "shuttle"
plasmid containing the SV40 early promoter and other relevant features for expression of inserted genes in mammalian cells. High-efficiency transfection of mammalian cells with plasmid-based vectors is a well-established art. Specifically, the Lipofectamine-mediated transfection protocol provided by Gibco/BRL Inc.
will be employed. This protocol results in the introduction of a transfecting plasmid into nearly every cell in culture without significant toxicity or cell death.
As with the yeast experiments described above, the sitoxin tests in human cells will determine the extent of killing by sitoxin of cells that lack HIV protease versus otherwise identical cells that express HIV protease. The W098~35~& PCT~S97/10941 conceptua.l and technical similarities between the yeast-and human cell-based tests of sitoxin extend to the adjustment protocols as well (if such adjustment procedures prove necessary at all). In particular, the half-lives of sitoxin constructs can be measured in human cells using a standard pulse-chase assay, and modifications of the open reading frames encoding the relevant constructs can also be carried out as described above.
Finally, a sitoxin that works as described in Figure l with human cells expressing or lacking HIV protease will be tested further, in a cell culture setting, using technically straightforward methods described above, for a differential toxicity of sitoxin against human cells infected with HIV versus uninfected human cells.
Positive results of this test would immediately enable at least one application of anti-viral sitoxins in general and the anti-HIV sitoxin in particular, namely their use to selectively kill virus-infected cells in human cell cultures that must be introduced (or reintroduced) into a patient, for example, in medical treatments that involve bone marrow transplantation.

Claims (26)

1. A signal-regulated, cleavage-mediated toxin, comprising:
a) an effector domain;
b) a domain bearing an intracellular signalling moiety; and c) a domain located between the effector domain and the domain bearing the intracellular signalling moiety which specifies a cleavage site for a predetermined protease.

2. The signal-regulated, cleavage-mediated toxin of Claim 1 wherein the predetermined protease is encoded by viral DNA.

3. The signal-regulated, cleavage-mediated toxin of Claim 1 wherein the intracellular signalling moiety is a degradation signal.

4. The signal-regulated, cleavage-mediated toxin of Claim 1 wherein the intracellular signalling moiety is a translocation signal.

5. The signal-regulated, cleavage-mediated toxin of Claim 4 wherein the translocation signal is a nuclear localization signal.

6. A DNA expression construct which encodes a signal-regulated, cleavage-mediated toxin, the signal-regulated, cleavage-mediated toxin comprising:
a) an effector domain;
b) a domain bearing an intracellular signalling moiety; and c) a domain located between the effector domain and the domain bearing the intracellular signalling moiety which specifies a cleavage site for a predetermined protease.

7. The DNA expression construct of Claim 6 wherein the predetermined protease is encoded by viral DNA.

8. The DNA expression construct of Claim 6 wherein the intracellular signalling moiety is a degradation signal.

9. The DNA expression construct of Claim 6 wherein the intracellular signalling moiety is a translocation signal.

10. The DNA expression construct of Claim 9 wherein the translocation signal is a nuclear localization signal.

11. A method for selectively killing a target cell known to contain a predetermined protease, the method comprising:
a) providing a DNA expression construct which encodes a signal-regulated, cleavage-mediated toxin, the signal-regulated, cleavage-mediated toxin comprising:
i) an effector domain;
ii) a domain bearing an intracellular signalling moiety; and iii) a domain located between the effector domain and the domain bearing the intracellular signalling moiety which specifies a cleavage site for the predetermined protease; and b) introducing the DNA construct from step a) into the target cell under conditions appropriate for expression of the signal-regulated, cleavage-mediated toxin.

12. The method of Claim 11 wherein the cell known to contain a predetermined protease is a prokaryotic cell.

13. The method of Claim 11 wherein the cell known to contain a predetermined protease is a eukaryotic cell.

14. The method of Claim 11 wherein the predetermined protease is encoded by viral DNA.

15. The method of Claim 11 wherein the intracellular signalling moiety is a degradation signal.

16. The method of Claim 11 wherein the intracellular signalling moiety is a translocation signal.

17. The method of Claim 16 wherein the translocation signal is a nuclear localization signal.

18. The method of Claim 9 wherein the cell known to contain a predetermined protease is a virus-infected cell.

19. A method for selectively killing a target cell known to contain a predetermined protease, the method comprising:
a) providing a multidomain fusion protein comprising:
i) an effector domain;
ii) a domain bearing an intracellular signalling moiety; and iii) a domain located between the effector domain and the domain bearing the intracellular signalling moiety which specifies a cleavage site for a predetermined protease; and b) introducing the multidomain fusion protein of step a) into the target cell.

20. The method of Claim 19 wherein the cell known to contain a predetermined protease is a prokaryotic cell.

21. The method of Claim 19 wherein the cell known to contain a predetermined protease is a eukaryotic cell.

22. The method of Claim 19 wherein the predetermined protease is encoded by viral DNA.

23. The method of Claim 19 wherein the intracellular signalling moiety is a degradation signal.

24. The method of Claim 19 wherein the intracellular signalling moiety is a translocation signal.

25. The method of Claim 24 wherein the translocation signal is a nuclear localization signal.

26. The method of Claim 19 wherein the cell known to contain a predetermined protease is a virus-infected cell.