Apoptosis-inducing human-origin Fcepsilon-Bak chimeric proteins for targeted elimination of mast cells and basophils: a new approach for allergy treatment

During the past few years, many chimeric proteins have been developed to specifically target and kill cells expressing specific surface molecules. Generally these molecules carry a bacterial or plant toxin to destroy the unwanted cells. The major obstacle regarding these molecules in their clinical application is the immunogenicity and nonspecific toxicity associated with bacterial or plant toxins. We lately reported a new approach for construction of chimeric proteins: we successfully replaced bacterial or plant toxins with human apoptosis-inducing proteins. The resulting chimeras were shown to specifically induce apoptosis in the target cells. Taking advantage of the human apoptosis inducing proteins Bak and Bax as novel killing components, we have now constructed new chimeric proteins targeted against the human FcepsilonRI, expressed mainly on mast cells and basophils. These cells are the main effectors of the allergic response. Treatment of the target cells with the new chimeric proteins, termed Fcepsilon-Bak/Bax, had a dramatic effect on cell survival, causing apoptosis. The effect was specific to cells expressing the FcepsilonRI of both human and, very unexpectedly, also of mouse origin. Moreover, interaction of the chimeric proteins with the mast cells did not cause degranulation. Fcepsilon-Bak/Bax are new chimeric proteins of human origin and, as such, are expected to be both less immunogenic and less toxic and, thus, may be specific and efficient reagents for the treatment of allergic diseases.     

GnRH-Bik/Bax/Bak chimeric proteins target and kill adenocarcinoma cells; the general use of pro-apoptotic proteins of the Bcl-2 family as novel killing components of targeting chimeric proteins

In recent years chimeric proteins carrying bacterial toxins as their killing moiety, have been developed to selectively recognize and kill cell populations expressing speciific receptors. The involvement of Gonadotropin releasing hormone (GnRH) has been demonstrated in several adenocarcinomas and a GnRH-bacterial toxin chimeric protein (GnRH-PE66) was thus developed and found to specifically target and kill adenocarcinoma cells both in vitro and in vivo. Because of the immunogenicity and the non-specific toxicity of the bacterial toxins, we have developed new chimeric proteins, introducing apoptosis inducing proteins of the Bcl-2 family as novel killing components. Sequences encoding the human Bik, Bak or Bax proteins were fused to the GnRH coding sequence at the DNA level and were expressed in E. coli. GnRH-Bik, GnRH-Bak and GnRH-Bax new chimeric proteins efficiently and specifically inhibited the cell growth of adenocarcinoma cell lines and eventually led to cell death. All three Bcl2-proteins-based chimeric proteins seem to induce apoptosis within the target cells, without any additional cell death stimulus. Apoptosis-inducing-proteins of the Bcl-2 family targeted by the GnRH are novel potential therapeutic reagents for adenocarcinoma treatment in humans. This novel approach could be widely applied, using any molecule that binds a specific cell type, fused to an apoptosis-inducing protein.     

Eliminating the six N-terminal amino acids of the caspase 3 large subunit improved production of a biologically active IL2-Caspase3 chimeric protein

Designing a chimeric protein and developing a procedure for its stable production as a biologically active protein, are key steps in its potential application to clinical trails. IL2-Caspase3 chimeric protein designed to target activated T lymphocytes was found to be a promising molecule for targeted treatment, however was found to be difficult to produce as a biological active molecule. Thus, we designed a new version of the molecule, IL2-Caspase3s, in which six amino acids (aa 29-34) from the N-terminus of the large subunit of caspase 3 were excluded. Repeated expressions, productions, and partial purifications of the IL2-Caspase3s yielded reproducible batches with consistent results. We found that IL2-Caspase3s causes cell death in a specific, dose-, and time-dependent manner. Cell death due to IL2-Caspase3s is caused by apoptosis. This improved and biologically stable IL2-Caspase3s chimeric protein may be developed in the future for clinical trails as a promising therapy for several pathologies involving activated T-cells. Moreover, this truncated caspase 3 sequence, lacking the N-terminal six amino acids of its large subunit, may be used in other caspase 3-based chimeric proteins targeted against various human diseases, using the appropriate targeting moiety.     

Cytotoxicity of IL6-PE40 and derivatives on tumor cells expressing a range of interleukin 6 receptor levels

The chimeric toxin IL6-PE40, which is composed of interleukin 6 (IL6) fused to a mutant form of Pseudomonas exotoxin (PE) devoid of its native cell recognition domain, can kill myeloma and hepatoma cells which express high levels of IL6 receptors. To enhance the usefulness of IL6-PE40 on potential target cells, we have attempted to develop more potent IL6-PE derivatives. We have developed nine new IL6-PE derivatives and assessed their cytotoxicity on human myeloma cells. Two of these new forms, IL6-domain II-PE40 and IL6-PE664Glu were more toxic to myeloma cells bearing IL6 receptors than was IL6-PE40. These two chimeric toxins were compared with IL6-PE40 for cytotoxicity toward a variety of tumor cell lines. We found that most tumor cell lines which are sensitive to IL6-PE40 are more sensitive to IL6-domain II-PE40 and IL6-PE664Glu. Cells with as few as 200-600 IL6 receptors/cell could be killed. The specificity of these chimeric toxins was shown through competition with recombinant IL6. Toxicity studies in mice demonstrated that the two new molecules had an LD50 of 10-20 micrograms/mouse. This compares to an IL6-PE40 LD50 of 20 micrograms/mouse. The new IL6-toxins could be detected in the serum up to 8 h after intraperitoneal administration with a peak at 1 h. These data suggest that IL6-domain II-PE40 and IL6-PE664Glu may be more useful than IL6-PE40 in killing IL6 receptor-bearing tumor cells in animals.     

Cell fusion activity of hepatitis C virus envelope proteins

To examine the cell fusion activity of hepatitis C virus (HCV) envelope proteins (E1 and E2), we have established a sensitive cell fusion assay based on the activation of a reporter gene as described previously (O. Nussbaum, C. C. Broder, and E. A. Berger, J. Virol. 68:5411-5422, 1994). The chimeric HCV E1 and E2 proteins, each consisting of the ectodomain of the E1 and E2 envelope protein and the transmembrane and cytoplasmic domains of the vesicular stomatitis virus G glycoprotein, were expressed on the cell surface. Cells expressing the chimeric envelope proteins and T7 RNA polymerase were cocultured with the various target cell lines transfected with a reporter plasmid encoding the luciferase gene under the control of the T7 promoter. After cocultivation, the cell fusion activity was determined by the expression of luciferase in the cocultured cells. The induction of cell fusion requires both the chimeric E1 and E2 proteins and occurs in a low-pH-dependent manner. Although it has been shown that HCV E2 protein binds human CD81 (P. Pileri, Y. Uematsu, S. Campagnoli, G. Galli, F. Falugi, R. Petracca, A. J. Weiner, M. Houghton, D. Rosa, G. Grandi, and S. Abrignani, Science 282:938-941, 1998), the expression of human CD81 alone is not sufficient to confer susceptibility to cell fusion in the mouse cell line. Treatment of the target cells with pronase, heparinase, or heparitinase reduced the cell fusion activity induced by the chimeric envelope proteins. These results suggest (i) that both HCV E1 and E2 proteins are responsible for fusion with the endosomal membrane after endocytosis and (ii) that certain protein molecules other than human CD81 and some glycosaminoglycans on the cell surface are also involved in the cell fusion induced by HCV.     

IL2-PE664Glu, a new chimeric protein cytotoxic to human-activated T lymphocytes

To produce a molecule that will kill activated T cells as well as lymphomas and leukemias expressing interleukin 2 (IL2) receptors, we have created a recombinant chimeric protein in which IL2 is attached in peptide linkage to a truncated mutant form of Pseudomonas exotoxin (PE) (Lorberboum-Galski, H., FitzGerald, D.J.P., Chandhary, V.K., Adhya, S., and Pastan, I. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 1922-1926). Although this molecule was very active on rodent cells, it had lower activity on some human cell types. A new chimeric protein termed IL2-PE664Glu has been constructed that is extremely toxic to both phytohemagglutinin blasts and mixed leukocyte reaction blasts prepared from monkey and human lymphocytes. The chimeric gene encoding this protein was constructed by fusing a cDNA clone for human interleukin 2 to the 5' end of a mutated cDNA encoding a full-length PE molecule. Four amino acids in domain I of PE were changed thus decreasing its nonspecific toxicity. IL2-PE664Glu is a much more active cytotoxic molecule for primate and human-activated T cells than IL2-PE40 which is a chimeric protein that was found to be an effective immunosuppressive agent in rodent models. Our results indicate that IL2-PE664Glu should be evaluated as an immunosuppressive agent for the treatment of human immune disorders in which activated T cells expressing the IL2 receptor are prominent.     

Expression vectors for streptavidin-containing chimeric proteins

We have constructed expression vectors for streptavidin-containing chimeric proteins. These vectors carry the DNA sequence corresponding to the core region of the streptavidin molecule, and have several unique cloning sites which facilitate construction of gene fusions of streptavidin with a target protein. A chimeric protein of streptavidin and the target protein should be expressible in Escherichia coli by using the T7 expression system. Because of the strong and specific biotin-binding affinity of the streptavidin moiety, such streptavidin-containing chimeric proteins should extensively expand the applications of the streptavidin-biotin system, and offer a variety of applications as new biological tools.     

Cytotoxic properties of a recombinant hybrid of the A protein of Staphylococcus aureus with a fragment of exotoxin A of Pseudomonas aeruginosa]

By gene-engineering technique a chimeric protein made up of fragments of Staphylococcus aureus protein A and .Pseudomonas aeruginosa exotoxin A has been constructed. The chimeric protein was shown to preserve features characteristic of its both constituents--it ADP-ribosylates elongation factor 2 and binds to Ig. Cytotoxic properties of the chimeric protein were studied in two model systems. Treatment of target cells in both systems was performed successively with antibodies against corresponding antigens and after washing--with recombinant chimeric toxin which bound to antibodies on the surface of target cells. In the first model system human B-lymphoma cells (Daudi line) carrying Ig molecules on their surface were treated with polyclonal antibodies against human Ig L-chains. In the other system, human T-lymphoma cells (Jurkat line) were treated successively with monoclonal antibodies against cell surface CD5 antigen and further on--with polyclonal antibodies against mouse Ig. In both systems, only a slight inhibition of the target cells' growth was registered. The probable reasons of low cytotoxic activity of the chimeric protein and prospects of increasing it are discussed.     

Interleukin-2-collagen chimeric protein which liberates interleukin-2 upon collagenolysis

Interleukin-2 (IL-2) is a potent activator of cellular immunity and has been utilized as an immunotherapeutic agent. We stably immobilized human IL-2 to collagen by covalently binding it to the N-terminus of human type III collagen (3A1) as IL2-3A1 chimeric protein using recombinant technology. The present study was aimed at liberating IL-2 from the immobilized chimeric protein by treating the chimera with bacterial collagenase. These IL2-3A1 chimeras were synthesized in insect cells which had been infected with baculovirus vectors carrying IL2-3A1 cDNA. The IL2-3A1 protein produced was shown to be in a pepsin-resistant triple helical structure and exhibited IL-2 activity to a similar extent as IL-2 itself. IL2-3A1 could be immobilized on the surface of plastic dishes by incubating it in the dishes. The IL-2 region of the immobilized IL2-3A1 was liberated to culture media by collagenase treatment and this freed IL-2 stimulated the growth of lined T cells. Thus, IL2-3A1 chimeric protein could be utilized as an IL-2 deliverer whose T cell mitogenic activity can be liberated by a collagenolytic environment.     

Redirecting therapeutic T cells against myelin-specific T lymphocytes using a humanized myelin basic protein-HLA-DR2-zeta chimeric receptor

Therapies that Ag-specifically target pathologic T lymphocytes responsible for multiple sclerosis (MS) and other autoimmune diseases would be expected to have improved therapeutic indices compared with Ag-nonspecific therapies. We have developed a cellular immunotherapy that uses chimeric receptors to selectively redirect therapeutic T cells against myelin basic protein (MBP)-specific T lymphocytes implicated in MS. We generated two heterodimeric receptors that genetically link the human MBP84-102 epitope to HLA-DR2 and either incorporate or lack a TCRzeta signaling domain. The Ag-MHC domain serves as a bait, binding the TCR of MBP-specific target cells. The zeta signaling region stimulates the therapeutic cell after cognate T cell engagement. Both receptors were well expressed on primary T cells or T hybridomas using a tricistronic (alpha, beta, green fluorescent protein) retroviral expression system. MBP-DR2-zeta-, but not MBP-DR2, modified CTL were specifically stimulated by cognate MBP-specific T cells, proliferating, producing cytokine, and killing the MBP-specific target cells. The receptor-modified therapeutic cells were active in vivo as well, eliminating Ag-specific T cells in a humanized mouse model system. Finally, the chimeric receptor-modified CTL ameliorated or blocked experimental allergic encephalomyelitis (EAE) disease mediated by MBP84-102/DR2-specific T lymphocytes. These results provide support for the further development of redirected therapeutic T cells able to counteract pathologic, self-specific T lymphocytes, and specifically validate humanized MBP-DR2-zeta chimeric receptors as a potential therapeutic in MS.     

Gene transfer using recombinant rabbit hemorrhagic disease virus capsids with genetically modified DNA encapsidation capacity by addition of packaging sequences from the L1 or L2 protein of human papillomavirus type 16

The aim of this study was to produce gene transfer vectors consisting of plasmid DNA packaged into virus-like particles (VLPs) with different cell tropisms. For this purpose, we have fused the N-terminally truncated VP60 capsid protein of the rabbit hemorrhagic disease virus (RHDV) with sequences which are expected to be sufficient to confer DNA packaging and gene transfer properties to the chimeric VLPs. Each of the two putative DNA-binding sequences of major L1 and minor L2 capsid proteins of human papillomavirus type 16 (HPV-16) were fused at the N terminus of the truncated VP60 protein. The two recombinant chimeric proteins expressed in insect cells self-assembled into VLPs similar in size and appearance to authentic RHDV virions. The chimeric proteins had acquired the ability to bind DNA. The two chimeric VLPs were therefore able to package plasmid DNA. However, only the chimeric VLPs containing the DNA packaging signal of the L1 protein were able efficiently to transfer genes into Cos-7 cells at a rate similar to that observed with papillomavirus L1 VLPs. It was possible to transfect only a very limited number of RK13 rabbit cells with the chimeric RHDV capsids containing the L2-binding sequence. The chimeric RHDV capsids containing the L1-binding sequence transfer genes into rabbit and hare cells at a higher rate than do HPV-16 L1 VLPs. However, no gene transfer was observed in human cell lines. The findings of this study demonstrate that the insertion of a DNA packaging sequence into a VLP which is not able to encapsidate DNA transforms this capsid into an artificial virus that could be used as a gene transfer vector. This possibility opens the way to designing new vectors with different cell tropisms by inserting such DNA packaging sequences into the major capsid proteins of other viruses.     

Inhibition of human NK cell-mediated killing by CD1 molecules

It is now well established that NK cells recognize classical and nonclassical MHC class I molecules and that such recognition typically results in the inhibition of target cell lysis. Given the known structural similarities between MHC class I and non-MHC-encoded CD1 molecules, we investigated the possibility that human CD1a, -b, and -c proteins might also function as specific target structures for NK cell receptors. Here we report that expression of CD1a, -b, or -c can partially inhibits target cell lysis by freshly isolated human NK cells and cultured NK lines. The inhibitory effects of CD1 molecules on NK cell could be shown upon expression of individual CD1 proteins in transfected NK-sensitive target cells, and these effects could be reversed by incubation of the target cells with mAbs specific for the expressed form of CD1. Inhibitory effects of CD1 expression on NK-mediated lysis could also be shown for cultured human dendritic cells, which represent a cell type that prominently expresses the various CD1 proteins in vivo. In addition, the bacterial glycolipid Ags known to be bound and presented by CD1 proteins could significantly augment the observed inhibitory effects on target cell lysis by NK cells.     

A new RNase-based immunoconjugate selectively cytotoxic for ErbB2-overexpressing cells

We report a new tumor-directed immunoRNase, a chimeric protein made up of an antibody fragment (single-chain Fv fragment) directed to ErbB2, a cell surface receptor, and a non-toxic, human ribonuclease, which upon cell internalization becomes cytotoxic. The immunoRNase is active as a ribonuclease, specifically binds and selectively kills ErbB2-positive cells. ErbB2 is one of the most specific tumor-associated antigens identified so far, overexpressed on tumor cells of different origin. Its choice as target antigen and that of a non-toxic, human RNase as the killer moiety makes this immunoRNase a new, potentially attractive anticancer agent.     

Costimulation of T-cell proliferation by a chimeric B7-antibody fusion protein

 T cells require at least two signals for activation and clonal expansion. The first signal conferring specificity is initiated by interaction of the T cell receptor with peptide-bearing MHC molecules. The second, costimulatory signal can be provided by cell-surface molecules on antigen-presenting cells such as B7-1 (CD80) and B7-2 (CD86), which interact with CD28 on T cells. To direct the costimulatory B7-2 molecule to the surface of tumor cells we have constructed a chimeric fusion protein, which consists of the extracellular domain of human B7-2 fused to a single-chain antibody domain (scFv) specific for the ErbB2 protein, a type I growth factor receptor overexpressed in a high percentage of human adenocarcinomas. This B7-2₂₂₅-scFv(FRP5) molecule, expressed in the yeast Pichia pastoris and purified from culture supernatants, binds to B7 counter-receptors and to ErbB2. B7-2_225-scFv(FRP5) localizes specifically to the surface of ErbB2-expressing target cells, thereby providing a costimulatory signal, which results in enhanced proliferation of syngeneic T cells. Accepted: 14 October 1997     

Rational immunotherapy with ribonuclease chimeras. An approach toward humanizing immunotoxins.

Members of the pancreatic ribonuclease (RNase) family have diverse activities toward RNA that could cause them to function during host defense and physiological cell death pathways. This activity could be harnessed by coupling RNases to cell binding ligands for the purpose of engineering them into cell-type specific cytotoxins. Therefore, the cytotoxic potential of RNase was explored by linking bovine pancreatic ribonuclease A via a disulfide bond to human transferrin or antibodies to the transferrin receptor. The RNase hybrid proteins were cytotoxic to K562 human erythroleukemia cells in vitro with an IC50 around 10(-7) M, whereas > 10(-4) M of native RNase was required to inhibit protein synthesis. Cytotoxicity required both components of the conjugate since excess transferrin or ribonuclease inhibitors added to the medium protected the cells from the transferrin-RNase toxicity. Importantly, the RNase conjugates were found to have potent antitumor effects in vivo. Chimeric RNase fusion proteins were also developed. F(ab')2-like antibody-enzyme fusions were prepared by linking the gene for human RNase to a chimeric antitransferrin receptor heavy chain gene. The antibody enzyme fusion gene was introduced into a transfectoma that secreted the chimeric light chain of the same antibody, and cell lines were cloned that synthesized and secreted the antibody-enzyme fusion protein of the expected size at a concentration of 1-5 ng/mL. Culture supernatants from clones secreting the fusion protein caused inhibition of growth and protein synthesis toward K562 cells that express the human transferrin receptor but not toward a nonhuman derived cell line. Since human ribonucleases coupled to antibodies also exhibited receptor mediated toxicities, a new approach to selective cell killing is provided. This may allow the development of new therapeutics for cancer treatment that exhibit less systemic toxicity and, importantly, less immunogenicity than the currently employed ligand-toxin conjugates.     

Characterization of diphtheria fusion proteins targeted to the human interleukin-3 receptor

Diphtheria fusion proteins are chimeric proteins consisting of the catalytic and translocation domains of diphtheria toxin (DT(388)) linked through an amide bond to one of a variety of peptide ligands. The ligand targets the molecule to cells and the toxin enters the cell, inactivates protein synthesis and induces cell death. Diphtheria fusion proteins directed to human myeloid leukemic blasts are a novel class of therapeutics for patients with chemotherapy refractory myeloid leukemia. Because of the presence of interleukin-3 (IL3) receptors on myeloid leukemic progenitors and its absence from mature myeloid cells, we synthesized four bacterial expression vectors encoding DT(388) fused to human IL3. Different molecules were engineered to assess the effects of modifications on yield, purity and potency of product. The constructs differed in the size of the linker peptide between the DT(388) and IL3 domains and in the presence or absence of an oligohistidine tag on the N- or C-terminus. Escherichia coli were transformed and recombinant protein induced and purified from inclusion bodies. Similar final yields of 3-6 mg of purified protein per liter of bacterial culture were obtained with each of the four molecules. Purity ranged from 70 to 90% after partial purification by anion-exchange, size-exclusion chromatography and/or nickel affinity chromatography. Proteins were soluble and stable at 4 degrees C and -80 degrees C in phosphate-buffered saline at 0.03-0.5 mg/ml. The fusion proteins showed predicted molecular weights by SDS-PAGE, HPLC and tandem mass spectrometry and had full ADP-ribosylating activities. Each was immunoreactive with antibodies to DT(388) and IL3. Each of the fusion proteins with the exception of the one with an N-terminal oligohistidine tag showed full IL3 receptor binding affinity (K:(d) = 3 nM) and potent and selective cytotoxicity to IL3 receptor positive human myeloid leukemia cell lines (IC(50) = 5-10 pM). In contrast, the N-terminal histidine-tagged fusion protein bound IL3 receptor with a 10-fold lower affinity and was 10-fold less cytotoxic to IL3 receptor positive blasts. Thus, we report a series of novel, biologically active DT(388)IL3 fusion proteins for potential therapy of patients with receptor positive myeloid leukemias.     

Targeted induction of apoptosis by chimeric granzyme B fusion proteins carrying antibody and growth factor domains for cell recognition

The serine protease granzyme B (GrB) of cytotoxic lymphocytes efficiently induces apoptosis by direct activation of caspases and cleavage of central caspase substrates. We employed human GrB as an effector function in chimeric fusion proteins that also contain the EGFR ligand TGFalpha or an ErbB2-specific single-chain antibody fragment (scFv) for selective targeting to tumor cells. GrB-TGFalpha (GrB-T) and GrB-scFv(FRP5) (GrB-5) molecules expressed in the yeast Pichia pastoris were bifunctional, cleaving synthetic and natural GrB substrates, and binding specifically to cells expressing EGFR or ErbB2 target receptors. Upon cell binding the chimeric molecules were internalized into intracellular vesicles, but could be released into the cytosol by the endosomolytic reagent chloroquine. Treatment with picomolar to nanomolar concentrations of GrB-5 and GrB-T resulted in selective and rapid tumor cell killing, accompanied by clear signs of apoptosis such as chromatin condensation, membrane blebbing, formation of apoptotic bodies and activation of endogenous initiator and effector caspases.     

Induction of HPV16 capsid protein-specific human T cell responses by virus-like particles

It has been postulated that upon binding to a cell surface receptor, papilloma virus-like particles (VLPs) gain entry into the cytosol of infected cells and the capsid proteins L1 and L2 can be processed in the MHC class I presentation pathway. Vaccination of mice with human papilloma virus-like particles consisting of capsid proteins L1 and L2 induced a CD8-mediated and perforin dependent protective immune response against a tumor challenge with human papilloma virus transformed tumor cells, which express only minute amounts of L1 protein. Here we show that HPV16 capsid proteins stimulate a MHC class I restricted CTL response with human peripheral blood lymphocytes (PBL) in vitro. The vigorous response was specific for VLP-infected target cells and was MHC class I restricted. Moreover we show the presence of at least one HLA-A*0201 restricted CTL epitope within the HPV-16 capsid proteins by using a VLP-'infected' HLA-A*0201 transfected human cell line as target cells. These results demonstrated that VLPs can induce a HPV16 capsid protein-specific immune response in humans, allowing the monitoring of immune responses induced by vaccines based on chimeric VLPs carrying additional immunogenic peptides or proteins in therapeutical applications in human patients.     

Single base discrimination for ribonuclease H-dependent antisense effects within intact human leukaemia cells.

We have previously demonstrated, in vitro, that phosphodiester and phosphorothioate antisense oligodeoxynucleotides could direct ribonuclease H to cleave non-target RNA sites and that chimeric methylphosphonodiester/phosphodiester analogue structures were substantially more specific. In this report we show that such chimeric molecules can promote point mutation-specific scission of target mRNA by both Escherichia coli and human RNases H in vitro. Intact human leukaemia cells 'biochemically microinjected' with antisense effectors demonstrated efficient suppression of target mRNA expression. It was noted that the chimeric methylphosphonodiester/phosphodiester structures showed single base discrimination, whereas neither the phosphodiester nor phosphorothioate compounds were as stringent. Finally, we show that the antisense effects obtained in intact cells were due to endogenous RNase H activity.     

A chimera of interleukin 2 and a binding variant of aerolysin is selectively toxic to cells displaying the interleukin 2 receptor

Aerolysin is a bacterial toxin that binds to glycosylphosphatidylinositol-anchored proteins (GPI-AP) on mammalian cells and oligomerizes, inserting into the target membranes and forming channels that cause cell death. We have made a variant of aerolysin, R336A, that has greatly reduced the ability to bind to GPI-AP, and as a result it is only very weakly active. Fusion of interleukin 2 (IL2) to the N terminus of R336A-aerolysin results in a hybrid that has little or no activity against cells that do not have an IL2 receptor because it cannot bind to the GPI-AP on the cells. Strikingly, the presence of the IL2 moiety allows this hybrid to bind to cells displaying high affinity IL2 receptors. Once bound, the hybrid molecules form insertion-competent oligomers. Cell death occurs at picomolar concentrations of the hybrid, whereas the same cells are insensitive to much higher concentrations of R336A-aerolysin lacking the IL2 domain. The targeted channel-forming hybrid protein may have important advantages as a therapeutic agent.