WO1990004414A1 - Conjugates of soluble t4 proteins and toxins and methods for treating or preventing aids, arc and hiv infection - Google Patents

Abstract

This invention relates to compositions and methods useful for the treatment and prevention of acquired immunodeficiency syndrome (AIDS), AIDS related complex (ARC), and human immunodeficiency virus (HIV) injection. More particularly, this invention relates to immunotoxins formed by conjugating a toxin to a soluble T4 protein, an HIV virus receptor that binds to the HIV virus or to target cells carrying the gp120/160 HIV marker. This invention also relates to methods for secreting soluble T4-toxin fusion protein conjugates through the cell membrane of a host. The compositions and methods of this invention advantageously provide a highly specific, site-directed delivery system for delivery of toxin to HIV infected cells and to circulating HIV virus.

Description

CONJUGATES OF SOLUBLE T4 PROTEINS AND

TOXINS AND METHODS FOR TREATING

OR PREVENTING AIDS. ARC AND HIV INFECTION

TECHNICAL FIELD OF INVENTION

This invention relates to compositions and methods useful for the treatment and prevention of acquired immunodeficiency syndrome (AIDS) , AIDS related complex (ARC) , and human immunodeficiency virus (HIV) infection. More particularly, this invention relates to immunotoxins formed by conjugating a toxin to a soluble T4 protein, an HIV virus receptor that binds to the HIV virus or to target cells carrying the gpl20/160 HIV marker. This invention also relates to methods for secreting soluble T4-toxin fusion protein conjugates through the cell membrane of a host. The compositions and methods of this invention advantageously provide a highly specific, site-directed system for delivery of toxin to HIV infected cells and to circulating HIV virus.

BACKGROUND OF THE INVENTION

In immunocompetent individuals, T4 lymphocytes interact with other specialized cell types of the immune system to confer immunity to or to defend against infection [E. L. Reinherz and S. F. Schlossman, "The Differentiation Function Of Human T-Cells", Cell. 19. pp. 821-27 (1980)]. More specifically, T4 lymphocytes stimulate the production of growth factors which are critical to a functioning immune system. For example, they act to stimulate B cells, the descendants of hemopoietic stem cells, which promote the production of defensive antibodies. They also activate macrophages ("killer cells") to attack infected or otherwise abnormal host cells, and they induce onocytes ("scavenger cells") to encompass and destroy invading microbes. The T4 surface protein of T4 lymphocytes is the primary target of certain infective agents, such as viruses and retroviruses. When T4 lymphocytes are exposed to such viruses, they are functionally impaired. As a result, the host's complex immune defense system is suppressed, and the host becomes susceptible to a wide range of opportunistic in¬ fections.

Such immunosuppression is seen in patients suffering from AIDS. AIDS is a disease characterized by severe or complete immunosuppression and attendant host susceptibility to a wide range of opportunistic infections and malignancies. In some cases, AIDS infection is accompanied by central nervous system disorders. Complete clinical manifestation of AIDS is usually preceded by ARC, a syndrome accompanied by symptoms such as persistent σeneralized lymph- adenopathy, fever and weight loss. The HIV retrovirus is thought to be the etiological agent responsible for AIDS infection and its precursor, ARC [M.G. Sargadharan et al., "Detection, Isolation And Continuous Production Of Cytopathic Retroviruses (HTLV-III) From Patients With AIDS And Pre-AIDS", Science. 224. pp. 497-508 (1984) ] .*

The genome of retroviruses, such as HIV, contains three regions encoding structural proteins. The σa region encodes the core proteins of the virion. The pol region encodes the virion RNA-dependent DNA polymerase (reverse transcriptase) . The env region encodes the major glycoprotein found in both the membrane envelope of the virus and the cytoplasmic membrane of infected cells. The capacity of the virus to attach to target cell receptors and to cause fusion of cell membranes (syncytia) are two HIV properties controlled by the env gene. These properties are believed to play a fundamental role in the pathogenesis of the virus.

The HIV env proteins arise from a precursor polypeptide — gpl20/160 — that, in mature form, is cleaved into a large, heavily glycosylated exterior membrane protein of about 481 amino acids — gpl20 —, and a smaller transmembrane protein of about 345 amino acids, which may be glycosylated — gp41 — [L. Ratner et al., "Complete Nucleotide Sequence Of The AIDS Virus, HTLV-III", Nature, 313, PP- 277-84 (1985)]. For convenience, in this application, the HIV exterior mem- brane protein will be referred to in terms of the precursor polypeptide, as — gpl20/160 —.

The host range of HIV is associated with cells which bear the T4 surface glycoprotein. Such

* In this application, human immunodeficiency virus ("HIV") , the generic term adopted by the Human Retro- virus Subcommittee of the International Committee On Taxonomy Of Viruses to refer to independent isolates from AIDS patients, including human T cell lymphotropic virus type III ("HTLV-III"), lymphadenopathy- associated virus ("LAV") , human immunodeficiency virus type 1 ("HIV-1") and AIDS-associated retrovirus ("ARV") will be used. cells include T4 lymphocytes and brain cells [P. J. Maddon et al. , "The T4 Gene Encodes The AIDS Virus Receptor And Is Expressed In The Immune System And The Brain", Cell. 47., PP- 333-48 (1986)]. The infection of a host by the HIV virus results in the depletion of functional T4 lymphocytes. Accordingly, the progression of AIDS/ARC syndromes can be corre¬ lated with the depletion of T4 lymphocytes which display the T4 surface glycoprotein (T4^* lymphocytes) . This T-cell depletion, with ensuing immunological compromise, may be attributable both to recurrent cycles of infection and to lytic growth, from cell- mediated spread of the virus. In addition, clinical observations suggest that HIV is directly responsible for the central nervous system disorders seen in many AIDS patients. The tropism of HIV for these cells is believed to be attributable to the role of the T4 protein as a membrane-anchored virus receptor.

Because the T4 protein behaves as the HIV receptor, its extracellular sequence probably plays a direct role in binding HIV. More specifically, it is believed that the gpl20/160 HIV envelope protein selectively binds to the T4 epitope(s), using this interaction to initiate entry into the host cell [A. G. Dalgelish et al., "The CD4 (T4) Antigen Is An Essential Component Of The Receptor For The AIDS Retrovirus", Nature, 312. pp. 763-67 (1984); D. Klatzmann et al. , "T-Lymphocyte T4 Molecule Behaves As The Receptor For Human Retrovirus LAV", Nature. 312. pp. 767-68 (1984)]. Accordingly, cellular expression of the T4 protein is believed to be sufficient for HIV binding, with the T4 protein serving as a receptor for HIV.

This T4 tropism of HIV has been demonstrated in vitro. When HIV isolated from AIDS patients is cultured together with T helper lymphocytes pre¬ selected for T4 protein presentation, the lymphocytes are efficiently infected, display cytopathic effects, including multinuclear syncytia formation, and are killed by lytic growth [D. Klatzmann et al., "Selective Tropism Of Lymphadenopathy Associated Virus (LAV) For Helper-Inducer T Lymphocytes", Science. 225. pp. 59-63 (1984) ; F. Wong-Staal and R. C. Gallo, "Human T- Lymphotropic Retroviruses", Nature. 317. pp. 395-403 (1985)]. It has also been demonstrated that a cloned cDNA version of a human T4 protein, when expressed on the surface of transfected cells from non-T cell lineages, including murine and fibroblastoid cells, endows those cells with the ability to bind HIV [P. J. Maddon et al., "The T4 Gene Encodes The AIDS Virus Receptor And Is Expressed In The Immune System And The Brain", Cell. 47, pp. 333-48 (1986)].

Therapeutic uses of soluble T4 proteins have been proposed for the prevention and treatment of the HIV-related infections AIDS and ARC. The nucleotide sequence and a deduced amino acid sequence for a DNA that purportedly encodes the full length human T4 protein have been reported [P. J. Maddon et al., "The Isolation And Nucleotide Sequence Of A cDNA Encoding The T Cell Surface Protein T4: A New Member Of The Immunoglobulin Gene Family", Cell, 42. pp. 9?-104 (1985)]. Based upon its deduced primary structure, the T4 protein can be functionally divided as follows: Amino Acid

Structure/Proposed Location Coordinates

Hydrophobic/Secretory Signal -23 to -1

Homology to V-Regions/ +1 to +94 Extracellular

Homology to J-Regions/ +95 to +109 Extracellular Glycosylated Region/ +110 to +374 Extracellular

Hydrophobic/Transmembrane +375 to +395 Sequence

Very Hydrophilic/ +396 to +435 Intracytoplasmic

Soluble T4 proteins have been constructed by truncating the full length T4 protein at amino acid 375, to eliminate the transmembrane and cytoplasmic domains. Such proteins have been produced by recombinant DNA techniques [R. A. Fisher et al., "HIV Infection Is Blocked In Vitro By Recombinant Soluble CD4", Nature. 331. pp. 76=78 (1988)]. These soluble T4 proteins advantageously interfere with the T4^* lymphocyte/HIV interaction by blocking or competitive binding mechanisms which inhibit HIV infection of cells expressing the T4 protein. By acting as soluble virus receptors, .soluble T4 proteins are useful as antiviral therapeutics to inhibit HIV binding to T4^* lymphocytes and virally induced syncytia formation.

Other methods for treating AIDS and ARC have included the administration of antiviral drugs, such as HPA-23, phosphonofor ate, sura in, ribavirin, azidothymidine ("AZT") and dideoxycytidine, which - apparently interfere with replication of HIV through reverse transcriptase inhibition. Although each of these drugs exhibits activity against HIV in vitro, only AZT has demonstrated potential benefits in clin- ical trials. AZT administration in effective amounts, however, has been accompanied by undesirable and debilitating side effects, such as bone marrow depression. It is likely, therefore, that hematologic toxicity will be a major limiting factor in the long term use of AZT.

Other proposed methods for treating AIDS have focused on the development of agents having activity against steps in the viral replicative cycle other than reverse transcription. Such methods include the administration of interferons or the application of hybridoma technology. Most of these treatment strategies are expected to require the co-admin¬ istration of immunomodulators, such as interleukin-2. To date, therefore, the need exists for the development of immunotherapeutic agents, methods and strategies for the treatment or prevention of AIDS, ARC, HIV infection and other immunodeficiencies caused by T-lymphocyte depletion or abnormalities.

DISCLOSURE OF THE INVENTION

The present invention solves the problems referred to above by providing compositions and methods for the treatment or prevention of AIDS, ARC and HIV infections. The compositions and methods of this invention are characterized by cytotoxic soluble T4 protein-toxin conjugates or "immunotoxins" comprising toxins conjugated to soluble T4 proteins. These immunotoxins utilize soluble T4 proteins as soluble virus receptors for the HIV envelope protein gpl20/160. This results in a toxin delivery system characterized by a high affinity for HIV and cells expressing the'HIV envelope protein. Advantageously, the immunotoxins of this invention inhibit HIV binding to T4^* lymphocytes by virtue of their competitive binding characteristics. and they actively destroy HIV and HIV infected cells expressing the gpl20/160 protein and producing the HIV virus.

More specifically, the soluble T4 protein portion of the immunotoxins of this invention binds to HIV or an HIV infected cell which presents the gpl20/160 surface marker, and the immunotoxin, or a portion thereof, is internalized or endocytosed into the cell. The toxin then intoxicates and kills..the infected cell, thus eliminating the source or reser¬ voir of the virus. The high target specificity of the soluble T4 protein portion of the immunotoxin results in delivery of the toxin directly to the target HIV or HIV-infected cells, enabling a high kill ratio with minimal toxin dosage, while minimizing any damage to non-HIV infected cells.

According to one embodiment of this inven¬ tion, recombinant DNA techniques are employed to construct immunotoxins which are fusion protein con- jugates comprising a toxin and a soluble T4 protein ("soluble T4-toxin fusion protein conjugates") . Another embodiment of this invention provides a method for secreting these soluble T4-toxin fusion protein conjugates through the cell membrane of a host. In a further embodiment of this invention, a soluble T4 protein and a toxin are chemically fused to form an immunotoxin.

The compositions and methods of this invention are useful in a variety of immunotherapeutic compositions and methods for treating immunodeficient patients suffering from diseases caused by infective agents whose primary targets are T4^* lymphocytes, particularly humans having AIDS, ARC, HIV infection or antibodies to HIV. These compositions and methods may also be used for treating AIDS-like diseases caused by retroviruses, such as simian immunodeficiency viruses, in mammals, including humans.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 depicts the nucleotide sequence and the derived amino acid sequence of the T4 cDNA of plasmid pl70.2.

Figure 2 depicts the nucleotide sequence and the derived amino acid sequence of plasmid pBG391. The amino acid sequence listed at line a is the reading frame which can be transcribed to produce a soluble T4 protein or portion thereof.

Figure 3 depicts the nucleotide sequence of the entire plasmid defined by pl99.7 (P_L mutet.rsT4) and its rsT4.2 insert and the amino acid sequence deduced from the rsT4 sequence. This includes the Clal-Clal cassette which defines the Met perfect rsT4.2 coding sequence.

In Figure 1, the T4 protein translation start (AA __) is located at the methionine at nucleotides 1199-1201 and the mature N-terminus is located at the lysine (AA_) at nucleotides 1274-1276.

In Figure 2, the T4 protein translation start (AA___) is located at the methionine at nucleoides 1207-1209 and the mature N-terminus is located at the lysine (AA-) at nucleotides 1282-1285.

In Figure 3, the T4 protein translation start (AA_.) is located at the methionine at nucleotides 64- 66.

Figure 4 is a schematic outline of the construction of plasmids pEX2 and pEX3.

Figure 5 is a schematic outline of the construction of plasmids pEX15 and pEX27.

Figure 6 is a schematic outline of the construction of plasmids pEX5 and pEXll. Figure 7 is a schematic outline of the construction of plasmid pEX39.

Figure 8 depicts the nucleotide sequence of a portion of pSA307 and the amino acid sequence of the regions of that portion coding for a streptavidin-like polypeptide. The amino acid sequence listed at line c is the reading frame which can be transcribed to produce a streptavidin-like polypeptide.

Figures 9A and 9B are schematic outlines of the construction of plasmids pEX45, pEX46, and pEX56.

Figure 10 illustrates the competition for gpl20/160 binding and cell killing activity among the competitor, rsT4.3, and the toxin, Pseudomonas^' Exotoxin A, or the immunotoxins, PEx45 and PEx46. Figure 11 depicts the amino acid sequence of the immunotoxin PEx45.

Figure 12 depicts the amino acid sequence of the immunotoxin PEx46.

Figure 13 depicts the amino acid sequence of Pseudomonas Exotoxin A.

Figure 14 illustrates the effects of contacting the immunotoxins of the present invention with C-NAD^* and elongation factor 2 (EF-2) .

Figure 15 depicts selective killing of a cell line expressing gpl20/160 on its surface by Pseudomonas Exotoxin A, or the immunotoxins, PEx45 and PEx46. The viability of the cells is then measured by the incorporation of ³H-labelled thymidine.

Figure 16 depicts the amino acid sequence of angiogenin.

Figures 17A and 17B are schematic outlines of the production of plasmids pTANGll and pTANG12.

Figure 18 depicts the amino acid sequence of TANG11, a fusion protein comprising a soluble T4 protein (having the formula AA -AA.., of Figure 2) , a portion of Pseudomonas Exotoxin A comprising the translocation domain, a portion of the ADP ribosylation domain and angiogenin.

Figure 19 depicts the amino acid sequence of TANG12, a fusion protein comprising a soluble T4 protein (having the formula AA..-AA of Figure 2) , a portion of Pseudomonas Exotoxin A comprising the translocation domain, a portion of the ADP ribosylation domain, the angiogenin signal sequence, and angiogenin. Figure 20 is a schematic outline of the construction of the plasmids pTANG13 and pTANG14.

Figure 21 depicts the amino acid sequence of TANG13, a fusion protein comprising a soluble T4 protein (having a formula AA.-AA of Figure 2) and mature angiogenin.

Figure 22 depicts the amino acid sequence of TANG14, a fusion protein comprising a soluble T4 protein (having a formula AA₃-AA₁₈₃ of Figure 2) , the angiogenin signal sequence, and angiogenin. Figure 23 illustrates the spatial relationship of the domains of Pseudomonas Exotoxin A and various immunotoxins according to this invention.

Figure 24 illustrates the plating efficiency cell assay. Figure 25 illustrates the gpl20 cell binding assay.

In the Figures, the amino acids are represented by single letter codes as follows:

* = position at which a stop codon is present. DETAILED DESCRIPTION OF THE INVENTION

This invention relates to immunotoxins comprising soluble T4 proteins conjugated to toxins, useful in methods and compositions for treating AIDS, ARC and HIV infections. The affinity of the soluble T4 protein component of the immunotoxins of this invention for the gpl20/160 HIV surface marker advantageously permits highly specific targeting of the toxin to HIV and to HIV infected cells. The soluble T4 proteins useful in producing the immunotoxins of this invention include all proteins, polypeptides and peptides which are natural or recombinant soluble T4 proteins, or portions thereof, and which are characterized by the immuno- therapeutic (antiretroviral) or immunogenic activity of soluble T4 proteins. They include soluble T4-like compounds from a variety of sources, such as soluble T4 protein derived from natural sources, recombinant soluble T4 protein and synthetic or semi-synthetic soluble T4 protein. Such soluble T4-like compounds advantageously interfere with the T4/HIV interaction by blocking or competitive binding mechanisms which inhibit HIV infection of cells expressing the T4 surface protein. The soluble T4 proteins used in the immunotoxins of this invention are all characterized by an affinity for the gpl20/160 HIV surface markers. Soluble T4 proteins include polypeptides selected from the group consisting of a polypeptide of the formula AA_₂ -A _3g2 of Figure 2, a polypeptide of the formula AA -AA_₆₂ of Figure 2, a polypeptide of the formula Met-AA _ of Figure 2, a polypeptide of the formula AA -AA,-. of Figure 2, a polypeptide of the formula A ^- A--_^. of Figure 2, a polypeptide of the formula Met-AA, __. of Figure 2, a polypeptide of the formula A₁~ A₃₇₇ of Figure 2, a polypeptide of the formula Met-AA. --- of Figure 2, a polypeptide of the formula AA -.-AA--- of Figure 2, or portions thereof.

Other soluble T4 proteins include poly¬ peptides selected from the group consisting of a poly- peptide of the formula AA __-AA.₈₂ of Figure 2, a polypeptide of the formula Met-AA. _₂ of Figure 2, a polypeptide of the formula AA.-AA.g₂ of Figure 2, a polypeptide of the formula AA.-AA._g- of Figure 2, a polypeptide of the formula AA_₂--AA₁₈₂ of Figure 2, followed by the amino acids asparagine-leucine-glut- amine-histidine-serine-leucine, a polypeptide of the formula AA.-AA.₈₂ of Figure 2, followed by the amino acids asparagine-leucine-glutamine-histidine-serine- leucine, a polypeptide of the formula Met-AAl.~l_nθ_o£_n of Figure 2, followed by the amino acids asparagine- leucine-glutamine-histidine-serine-leucine, a poly¬ peptide of the formula AA .--AA of Figure 2, a polypeptide of the formula AA.-AA... of Figure 2, a polypeptide of the formula Met-AA . of Figure 2, a polypeptide of the formula AA -.-AA..- of Figure 2, a polypeptide of the formula AA.-AA. _ of Figure 2, a polypeptide of the formula Met-AA. -.- of Figure 2, a polypeptide of the formula AA_₂_-AA__. of Figure 2, a polypeptide of the formula AA -AA.-. of Figure 2, a polypeptide of the formula Met-AA. ,-. of Figure 2, a polypeptide of formula AA__ -AA.... of Figure 2, a polypeptide of the formula AA.-AA.₄₅ of Figure 2, a polypeptide of the formula Met-AA._₁₄₅ of Figure 2, a polypeptide of the formula AA "ώ-J_-AAl. _coo, of Figure 2, a polypeptide of the formula AA.-AA.₆₆ of Figure 2, a polypeptide of the formula Met-AA. _gβ of Figure 2, or portions thereof.

Additionally, soluble T4 proteins include polypeptides selected from the group consisting of a polypeptide of the formula AA 1-AAJ-O--,i. of mature T4 protein, a polypeptide of the formula Met-AA __g of mature T4 protein, a polypeptide of the formula AA -AA_₇₄ of mature T4 protein, a polypeptide of the formula Met-AA. --. of mature T4 protein, a polypep- tide of the formula AA.-AA-__ of mature T4 protein, a polypeptide of the formula Met-AA, -„ of mature T4 protein, a polypeptide of the formula AA ---AA-- of mature T4 protein, a polypeptide of the formula AA __- AA-_₇ of mature T4 protein, or portions thereof. Soluble T4 proteins also include polypeptides selected from the group consisting of a polypeptide of the formula AA -AA._g2 of mature T4 protein, a polypeptide of the formula AA -AA ₂ of mature T4 protein, a polypeptide of the formula Met-AA. .__ of mature T4 protein, a polypeptide of the formula

AA ---AA.-- of mature T4 protein, followed by the amino acids asparagine-leucine-glutamine-histidine-serine- leucine, a polypeptide of the formula AA 1.-AA l.o_Q__i of mature T4 protein, followed by the amino acids asparagine-leucine-glutamine-histidine-serine-leucine, a polypeptide of the formula Met-AA. . of mature T4 protein, followed by the amino acids asparagine- leucine-glutamine-histidine-se ine-leucine , a polypeptide of the formula AA ---AA..- of mature T4 protein, a polypeptide of the formula AA.-AA ._ of mature T4 protein, a polypeptide of the formula Met-AA. ,.- of mature T4 protein, a polypeptide of the formula AA_₂--AA of mature T4 protein, a polypeptide of the formula AA.-AA . of mature T4 protein, a polypeptide of the formula Met-AA. .. of mature T4 protein, a polypeptide of the formula AA ---AA.-.-of mature T4 protein, a polypeptide of the formula AA -AA of mature T4 protein, a polypeptide of the formula Met-AA. .-. of mature T4 protein, a polypeptide of the formula AA_₂₃-AA.₄₅ of mature T4 protein, a polypeptide of the formula AA -AA.₄₅ of mature T4 protein, a polypeptide of the formula Met-AA.__14c ^of mature T4 protein, a polypeptide of the formula AA — Δ.m,>-AAl,O,O. of mature T4 protein, a polypeptide of the formula AA.-AA.gg of mature T4 protein, a polypeptide of the formula Met-AA, __gg of mature T4 protein, or portions thereof.

The amino-terminal amino acid of mature T4 protein isolated from T cells begins at lysine, the third amino acid of the sequence depicted in Figure 2. Accordingly, soluble T4 proteins also include polypeptides of the formula AA_-AA,₇₇ of Figure 2, or portions thereof. Such polypeptides include polypeptides selected from the group consisting of a polypeptide of the formula AA.-AA-_g2 of Figure 2, a polypeptide of the formula AA -AA_₇ of Figure 2, a polypeptide of the formula AA -AA.₈_ of Figure 2, a polypeptide of the formula AA_-AA.₈₂ of Figure 2, a polypeptide of the formula AA.-AA..- of Figure 2, a polypeptide of the formula AA.-AA.-. of Figure 2, a polypeptide of the formula AA_-AA.₄₅ of Figure 2, a polypeptide of the formula AA--AA._gg of Figure 2, and a polypeptide of the formula AA.-AA... of Figure 2. Soluble T4 proteins also include the above-recited polypeptides preceded by an N-terminal methionine group.

Soluble T4 proteins useful in the immuno¬ toxins and methods of this invention may be produced in a variety of ways. We have depicted in Figure 1 the nucleotide sequence of full-length T4 cDNA obtained from pl70.2 and the amino acid sequence deduced therefrom. The T4 cDNA of pl70.2 is almost identical to the approximately 1,700 bp sequence reported by Maddon et al.. supra. The T4 cDNA of pl70.2, however, contains three nucleotide substitutions which, in the translation product of this cDNA, produce a protein containing three different amino acids compared to the sequence reported by Maddon et al. These differences are at amino acid position 3, where the asparagine of Maddon et al. is replaced with lysine; position 64, where the tryptophan of Maddon et al. is replaced with arginine and at position 231, where the phenylalanine of Maddon et al. is replaced with serine. The asparagine reported at position 3 in Maddon et al. instead of lysine was the result of a DNA sequencing error [D. R. Littman et al., "Corrected CD4 Sequence", Cell. 55. p. 541 (1988)].

Soluble T4 protein constructs or DNA inserts may be produced by truncating the full length T4 protein sequence at various positions to remove the coding regions for the transmembrane and intracyto- plasmic domains, while retaining the extracellular region believed to be responsible for HIV binding. More particularly, soluble T4 proteins may be produced by conventional techniques of oligonucleotide directed mutagenesis and restriction digestion, followed by insertion of linkers, or by chewing back full-length T4 protein with enzymes.

Prior to such constructions, the cDNA coding sequence of a full length T4 clone, such as pl70.2, may be modified in sequential steps of site-directed mutagenesis and restriction fragment substitution to modify the amino acids at positions 64 and 231. For example, one may employ oligonucleotide-directed mutagenesis to modify amino acid 64. Subsequently, restriction fragment substitution with a fragment including the serine 231 codon of a partial T4 cDNA isolated from a T4 positive lymphocyte cell line [O. Acuto et al., Cell. 34. pp. 717-26 (1983)] library in λgt 11 may be used to modify the amino acid at position 231 [R. A. Fisher et al., Nature, supra] .

DNA sequences coding for soluble T4 proteins may be used to transform eukaryotic and prokaryotic host cells by conventional recombinant DNA techniques to produce recombinant soluble T4 (rsT4) proteins in clinically and commercially useful amounts. Such soluble T4 proteins include those produced according to- the processes set forth in copending, commonly assigned United States patent applications Serial No. 094,322, filed September 4, 1987 and Serial No. 141,649, filed January 7, 1988, and PCT patent application Serial No. PCT/US88/02940, filed September 1, 1988, the disclosures of which are hereby incorporated by reference.

Microorganisms and recombinant DNA molecules characterized by DNA sequences coding for soluble T4 proteins are exemplified by cultures deposited in the In Vitro International, Inc. culture collection, in Linthicum, Maryland, on September 2, 1987 and identified as:

EC100: E.coli JM83/pEC100 - IVI 10146 BG377: E.coli MC1061/pBG377 - IVI 10147 BG380: E.coli MC1061/pBG380 - IVI 10148 BG381: E.coli MC1061/pBG381 - IVI 10149.

Such microorganisms and recombinant DNA molecules are also exemplified by cultures deposited in the In Vitro International, Inc. culture collection on January 6, 1988 and identified as: BG-391: E.COli MC1061/pBG391 - IVI 10151 BG-392: E.coli MC1061/pBG392 - IVI 10152 BG-393: E.coli MC1061/pBG393 - IVI 10153 BG-394: E.COli MC1061/pBG394 - IVI 10154 BG-396: E.coli MC1061/pBG396 - IVI 10155 203-5 : E.coli SG936/p203-5 - IVI 10156. Additionally, such microorganisms and recombinant DNA molecules are exemplified by cultures deposited in the In Vitro International, Inc. culture collection on August 24, 1988 and identified as: 211-11: E.coli A89/pBG211-ll - IVI 10183 214-10: E.coli A89/pBG214-10 - IVI 10184 215-7 : E.COli A89/pBG215-7 - IVI 10185.

Alternatively, soluble T4 proteins may be chemically synthesized by conventional peptide syn- thesis techniques, such as solid phase synthesis.

[R. B. Merrifield, "Solid Phase Peptide Synthesis. I. The Synthesis Of A Tetrapeptide", J. Am. Chem. Soc. , 83, pp. 2149-54 (1963)].

By coupling that portion of a DNA sequence coding for a toxin with a DNA sequence coding for a soluble T4 protein, immunotoxins which primarily affect target HIV or HIV-infected cells may be produced in accordance with this invention. The term "toxin" as used in this application includes any molecule which kills or down regulates cells either upon contact or when introduced intracellularly. Toxins useful in this invention include, but are not limited to, proteins such as ricin, abrin, angiogenin, Pseudomonas Exo¬ toxin A, pokeweed antiviral protein, saporin, gelonin and diptheria toxin, or toxic portions thereof.

Many toxins exert their cytotoxic effects via a three step mechanism which includes: 1) binding of the toxin to the cellular membrane, 2) translocation of the toxin through the cellular membrane into the cytosol*, and 3) target cell inactivation. Such toxins

* Like other molecules that bind to receptors on the cell surface, these toxins are internalized by the pathway of receptor mediated endocytosis. In the first step of this process, a ligand binds to its receptor. The receptor-ligand complex then diffuses laterally within the cell membrane until it encounters a clathrin each comprise several different functional regions or "domains", which mediate one of these steps. Typically, one domain contains information for cell recognition and binding, a second domain translocates the toxin across the cell membrane to the cytosol — where the targets of most toxins reside, and a third domain catalyzes the reaction responsible for cell killing [I. Pastan et al., "Immunotoxins", Cell. 47, pp. 641-648 (1986)]. Although the third domain is the portion of the toxin which actually inactivates the target cell, the other two domains are normally critical to overall toxicity, because the toxin usually must be internalized into the cytosol to have a high activity. When such toxins are used to produce the immunotoxins of this invention, the translocation and cell inactivating domains of the toxin are kept substantially intact, while the binding domain is inactivated or deleted, being replaced with a soluble T4 protein which will specifically bind to HIV or an HIV-infected cell. In this application, the translocation and cell inactivating domains of the toxin are the "toxic portion" of the toxin.

Ricin, abrin, saporin and gelonin are plant ribosome-inactivating proteins [I. Pastan et al.. Cell, supra] ; angiogenin is a human blood vessel inducing protein which is also ribosome inactivating [D. K. St. Clair et al., "Angiogenin Abolishes Cell- Free Protein Synthesis By Specific Ribonucleolytic Inactivation Of Ribosomes", Proc. Natl. Acad. Sci. USA. 84. pp. 8330-34 (1987)]; while diphtheria toxin and

coated pit. The coated pit then gives rise to an endocytic vesicle, which moves away from the cell surface by saltatory motion. Once in the cytosol, the toxin then exerts its lethal effect. Pseudomonas Exotoxin A are bacterial toxins that inhibit protein synthesis by catalyzing the transfer of the ADP ribosyl moiety of oxidized NAD onto elongation factor-2 [G. Gray et al., "Cloning, Nucleotide 5 Sequence, And Expression In Escherichia Coli Of The Exotoxin A Structural Gene Of Pseudomonas Aeruqinosa", Proc. Natl. Acad. Sci. USA. 81. pp. 2645-49 (1984)]. Conjugates of toxins and cell-reactive antibodies directed against determinants on microorganisms,

10 neoplastic cells and virally infected cells have been used to form cell-specific immunotoxins [E. S. Vitetta, "Redesigning Nature's Poisons To Create Anti-Tumor Re¬ agents", Science, 238. pp. 1098-1104 (1987); I. Pastan et al. , Cell, supra1. However, only recently have

15. advances in recombinant DNA technology enabled the production of immunotoxins characterized by highly specific binding to a target cell antigen.

A preferred toxin useful in the compositions and methods of the present, invention to produce soluble

20 T4-toxin fusion protein conjugates is Pseudomonas

Exotoxin A, and portions thereof. Pseudomonas Exotoxin A (PE) is an extremely active, monomeric protein (molecular weight 66 kd) secreted by Pseudomonas aerucrinosa. It comprises a single polypeptide chain of

25 613 amino acids [Allured et al., Proc. Natl. Acad. Sci. USA. 83. pp. 1320-324 (1986)]. Pseudomonas Exotoxin A inhibits protein synthesis in eukaryotic cells through the inactivation of elongation factor-2 (EF-2) , by catalyzing the transfer of the ADP ribosyl moiety of

30 oxidized NAD onto EF-2.

The specific intoxication process caused by PE is believed to proceed by the following steps. First, PE binds to cells through a specific receptor on the cell surface. Next, the PE-receptor complex is

35 internalized into the cell. Finally, PE is transferred to the cytosol, where it enzymatically inhibits protein synthesis. Because cellular intoxication is prevented by weak bases, such as NH₄ ⁺, which raise the pH in acidic vesicles, PE is believed to be translocated across the cell membrane when it reaches an acidic environment. Upon exposure to acidic conditions, the hydrophobic domain of PE enters into the cell membrane, resulting in the formation of a channel through which the enzymatic domain, in extended form, passes across the membrane of the endocytic vesicle into the cytosol. The DNA sequence coding for the PE protein has been mapped and cloned [G. Gray et al., Proc. Natl. Acad. Sci. USA, supra] . The PE gene is characterized by three structurally distinct functional domains which mediate the toxic effects of PE [J. Hwang et al.,

"Functional Domains Of Pseudomonas Exotoxin Identified By Deletion Analysis Of The Gene Expressed In E.coli". Cell. 4._/ PP. 129-36 (1987)]:

— domain I, or the binding domain (B) ) which mediates binding of the toxin to the cell membrane. The binding domain is characterized by its ability to bind to the mammalian cell surface.

This domain comprises AA -AA. of PE (domain la) and AA3.6-5.-AA4.0.4. of PE (domain lb) ; — domain II, or the translocation domain (T) , which mediates translocation of the toxin through the cell membrane. The translocation domain is responsible for processing the exotoxin out of the endosome through the cell membrane into the cytosol. This domain comprises

^253^364 °^{f PE; and}

— domain III, or the ADP ribosylation domain (A) , which mediates killing of the cell by inactivating elongation factor 2. The ADP ribosylation domain comprises AA_4Q5-AA_g of PE. A boundary region has also been identified on the Pseudomonas Exotoxin A peptide backbone. This region corresponds to a 26 amino acid residue fragment extending adjacent from the translocation domain into the binding domain. The amino acid sequence of

Pseudomonas Exotoxin A is depicted in Figure 13 and the spatial relationships of the particular domains of PE are illustrated in Figure 23.

By deleting portions of the PE gene sequence, it is possible to identify the domains responsible for each function. This procedure is also useful to develop shortened versions of the PE gene sequence which no longer code for the binding function of domain I, but still retain the functions of domain II and III. Such shortened versions of the PE gene sequence which code for the toxic portion of PE, may be conjugated to a soluble T4 protein to form the immuno¬ toxins of this invention. The inclusion of the boundary region of PE in several soluble T -toxin fusion proteins of this invention leads to an unexpected increase in the cytotoxic effects of these proteins.

We employed shortened versions of the PE gene sequence to express immunotoxins including portions of PE without the binding region. One such PE fragment, identified as PE40, contains only the translocation domain and the ADP ribosylation cell inactivating domain of PE. In addition, we prepared various immunotoxins having a PE fragment, known as PE43, which includes the boundary region located at the juncture of the binding and translocation domains, together with the translocation domain and the ADP ribosylation cell inactivating domain. The boundary region comprises a 26 amino acid fragment located between the binding domain and the translocation domain. These immunotoxins are depicted in Figure 23.

Another preferred toxin useful in the compositions and methods of the present invention is angiogenin, and toxic portions thereof, used alone or in combination with Pseudomonas Exotoxin A. Angiogenin is a human serum protein comprising 125 amino acids, without the signal peptide (see Figure 16) . The ability of angiogenin to inhibit or inactivate protein synthesis [D. St. Clair et al., "Angiogenin Abolishes Cell-free Protein Synthesis by Specific Ribonucleolytic Inactivation of 40S Riboso es", Biochem.. 27, pp. 7263- 68 (1988) ] constitutes an activity in common with all toxins known to date. As previously stated, angiogenin is a human protein; whereas all toxins known to be useful as immunotoxins have been derived from plants or bacteria. Thus, angiogenin is likely to exhibit less of an immunogenic response than its plant or bacterial counterparts. Moreover, soluble T4 fusion proteins comprising angiogenin should be capable of demonstrating efficacy for prolonged periods of treatment.

In addition, angiogenin is resistant to acidic conditions (low pH) as well as to proteolytic degradation. It is known that toxins are typically internalized into the cell in an acidic environment. Although angiogenin has not typically been considered a "toxin", we believe that its ability to inhibit or inactivate protein synthesis in a target cell or to down regulate a cell renders it useful in the fusion protein conjugates of this invention in place of conventional toxins ^'to achieve comparable, if not enhanced, toxic effects. Accordingly, the use of the term "toxin" to describe angiogenin and soluble T4 protein-angiogenin fusion proteins, described herein, will also be referred to as immunotoxins. Further, the term "immunotoxin", as used herein, is meant to include all toxins that act through an immunological pathway or a directed biochemical pathway to induce their toxic response.

The immunotoxins of this invention are useful in a variety of immunotherapeutic compositions and methods. Advantageously, the immunotoxins of this invention inhibit HIV binding to T4⁺ lymphocytes by virtue of their competitive binding characteristics, and they actively destroy HIV infected cells express¬ ing the gpl20/160 protein and producing HIV. Accord¬ ingly, the immunotoxins of this invention may be used in pharmaceutical compositions and methods to treat humans having AIDS, ARC, HIV infection, or antibodies to HIV. In addition, these immunotoxins and methods may be used for treating AIDS-like diseases caused by retroviruses, such as simian immunodeficiency viruses, in mammals, including humans. The immunotoxins of this invention may be produced by conjugating a soluble T4 protein to a toxin. According to one embodiment of this invention, toxins may be conjugated to soluble T4 proteins using a chemical coupling method. In this method, the toxin is chemically coupled directly to the soluble T4 protein using, for example, techniques which have been used to couple toxins to antibody molecules via either disulfide or thioether linkages. More specifically, disulfide bonds may be created by introducing sulfhydryl groups into the toxin's amine groups by treating the toxin with iminothiolane [D.J.P. Fitzgerald, "Construction of Immunotoxins Using Pseudomonas Exotoxin A", Methods Enzymol.. 151, pp. 139-45 (1987)], or by using N-succinimidyl-3-(2- pyridyldithio)-propionate [J. A. Cumber et al. , "Preparation Of Antibody-Toxin Conjugates", Methods Enzymol. , 112, pp. 207-25 (1985)]. Alternatively, thioether bonds may be created by introducing thiol groups into the protein and an alkylating function into the toxin [M. J. Bjorn et al. , "Antibodv-Pseudomonas Exotoxin A Conjugates Cytotoxic To Human Breast Cancer Cells In Vitro". Cancer Res.. 46. pp. 3262-67 (1986)]. Example 4 illustrates the production of a soluble T4 protein-toxin conjugate by chemical coupling using disulfide linkages.

In another embodiment of this invention, toxins may be conjugated to soluble T4 proteins using a genetic fusion method. In this method, a hybrid DNA sequence is constructed from a DNA sequence coding for the toxin and a DNA sequence coding, for a soluble T4 protein using recombinant DNA techniques. Expression of the hybrid DNA sequence in an appropriate host results in a fusion protein comprising the soluble T4 protein and the toxin ("a soluble T4-toxin fusion protein conjugate") [Lorberboum-Galski et al.,

"Cytotoxic Activity Of An Interleukin -2-Pseudomonas Exotoxin Chimeric Protein Produced In Escherichia Coli" , Proc. Natl. Acad. Sci. USA. 85. pp. 1922-26 (1988); Hwang et al.. Cell, supra]. According to a preferred embodiment of the present invention, various soluble T4 protein- Pseudo onas Exotoxin A immunotoxins which are toxic to HIV or HIV-infected cells may be prepared. The soluble T4 portion of these fusion proteins -may comprise any soluble T4 polypeptide capable of binding HIV cells. The preferred soluble T4 polypeptide is characterized by the formula AA__₂₃-AA_37? of Figure 2, or fragments thereof. Particularly preferred are soluble T4 polypeptides represented by the formula AA —- Z__>-AA1.8_2_ or AA.-AA.₈₂ of Figure 2. Most preferred is a soluble T4 polypeptide represented by the formula AA.J-AA.„o_ -i of

Figure 2.

The Pseudomonas Exotoxin A portion of the immunotoxins of this invention preferably comprises the entire protein toxin, less a substantial portion of the binding domain (la) . A particularly preferred immunotoxin of this invention comprises a fragment of Pseudomonas Exotoxin A having the translocation domain (II) and the ADP ribosylation cell inactivating domain (III) fused to the soluble T4 protein. Another particularly preferred immunotoxin is a fragment of Pseudomonas Exotoxin A having the boundary region, the translocation domain and the ADP ribosylation cell inactivating domain substantially intact and fused to a soluble T4 protein. Most preferred is an immunotoxin comprising a fragment of Pseudomonas Exotoxin A comprising the boundary region, the translocation domain, and the ADP ribosylation cell inactivating domain fused to a soluble T4 polypeptide of the formula AA₃-AA._g3 of Figure 2.

Another particularly preferred embodiment of the present invention includes soluble T4 protein- angiogenin immunotoxins which exhibit toxic effects on HIV or HIV-infected cells. These immunotoxins may be produced by conjugation using chemical coupling procedures suitable for linking toxins to antibodies. These procedures include, but are not limited to, the creation of disulfide bonds by introducing sulfhydryl side chains into an amine functionality of angiogenin through iminothiolane treatment rFitzσerald. supra] .

Another mode of chemical construction employs thioether linkage incorporation through the introduction of sulfhydryl functionalities to the angiogenin followed by alkylation with a suitable alkylating reagent [BLQrn, supra1. Alternatively, genetic fusion techniques may be utilized to produce the soluble T4 protein- angiogenin immunotoxins of the present invention. We have expressed such proteins from E.coli using a hybrid DNA sequence constructed from the DNA sequence coding for angiogenin and a DNA sequence coding for a soluble T4 polypeptide using recombinant DNA technology. The hybrid DNA sequence, when expressed in an appropriate host, encodes an immunotoxin comprising a soluble T4 polypeptide and angiogenin.

Further, an immunotoxin protein comprising a soluble T4 protein capable of binding HIV cells fused to the translocation domain — AA 2-5-.-3-AA-36.4. — of

Pseudomonas Exotoxin A and angiogenin may also be prepared by genetic fusion techniques. In such immunotoxins, the soluble T4 portion of the protein preferably comprises a soluble T4 polypeptide of the formula AA₃-AA._a3 of Figure 2. A hybrid DNA sequence comprising a DNA sequence ^'coding for angiogenin with or without its signal sequence, a DNA sequence coding for the soluble T4 polypeptide of the formula AA -.i-AA1.8_Q_ of

Figure 2, and a DNA sequence coding for the translocation domain — AA 2_5_E3,-AA3-6.4. — of Pseudomonas

Exotoxin A may be prepared by genetic fusion techniques. Alternatively, such immunotoxins may also comprise the ADP ribosylation domain of Pseudomonas

Exotoxin A.

Similarly, a hybrid DNA sequence comprising a

DNA sequence coding for angiogenin with or without its signal sequence and a DNA sequence coding for the soluble T4 polypeptide of the formula AA- _>-AA.l₀o_j of

Figure 2 may be prepared by recombinant techniques.

This invention also relates to immunotoxins, as described above, in which any soluble receptor, rather than soluble T4 protein, may be coupled to a toxin. In addition, antibodies, or portions thereof, such as Fab' fragments, variable regions, hypervariable regions, and the like, may also be fused or conjugated to toxins to form immunotoxins according to this invention.

As is well known in the art, for expression of the DNA sequences of this invention, the DNA sequence should be operatively linked to an expression control sequence in an appropriate expression vector and employed in that expression vector to transform an appropriate unicellular host.

Such operative linking of a DNA sequence of this invention to an expression control sequence, of course, includes the provision of a translation start signal in the correct reading frame upstream of the DNA sequence. If a particular DNA sequence being expressed does 'not begin with a methionine, the start signal will result in an additional amino acid — methionine — being located at the N-terminus of the product. While such methionyl-containing product may be employed directly in the compositions and methods of this invention, it is usually more desirable to remove the methionine before use. Methods are available in the art to remove such N-terminal methionines from polypeptides expressed with them. For example, certain hosts and fermentation conditions permit removal of substantially all of the N-terminal methionine in vivo. Other hosts require in vitro removal of the N-terminal methionine. However, such in vivo and in vitro methods are well known in the art.

A wide variety of host/expression vector combinations may be employed in expressing the DNA sequences of this invention. Useful expression vectors, for example, may consist of segments of chromosomal, non-chromosomal and synthetic DNA sequences, such as various known derivatives of SV40 and known bacterial plasmids, e.g., plasmids from E.coli including colEl, pCRl, pBR322, pMB9 and their derivatives, wider host range plasmids, e.g., RP4, phage DNAs, e.g., the numerous derivatives of phage λ, e.g., NM989, and other DNA phages, e.g., M13 and filamenteous single stranded DNA phages, yeast plas¬ mids, such as the 2μ plasmid or derivatives thereof, and vectors derived from combinations of plasmids and phage DNAs, such as plasmids which have been modified to employ phage DNA or other expression control sequences.

In addition, any of a wide variety of expres¬ sion control sequences — sequences that control the expression of a DNA sequence when operatively linked to it — may be used in these vectors to express the DNA sequence of this invention. Such useful expression control sequences, include, for example, the early and late promoters of SV40 or the adenovirus, the lac system, the trp system, the TAC or TRC system, the major operator and promoter regions of phage λ, the control regions of fd coat protein, the promoter for 3- phosphoglycerate kinase or other glycolytic enzymes, the promoters of acid phosphatase, e.g., Pho5, the pro- moters of the yeast α-mating factors, the polyhedron promoter of the baculovirus system and other sequences known to control the expression of genes of prokaryotic or eukaryotic cells or their viruses, and various combinations thereof. A wide variety of unicellular host cells are also useful in expressing the DNA sequences of this invention. These hosts may include well known eukaryotic and prokaryotic hosts, such as strains of E.coli. Pseudomonas. Bacillus, Streptomvces. fungi, such as yeasts, and animal cells, such as CHO and mouse cells, African green monkey cells, such as COS 1, COS 7, BSC 1, BSC 40, and BMT 10, insect cells, and human cells and plant cells in tissue culture. For animal cell expression, we prefer CHO cells and COS 7 cells.

It should of course be understood that not all vectors and expression control sequences will function equally well to express the DNA sequences of this invention. Neither will all hosts function equally well with the same expression system. How¬ ever, one of skill in the art may make a selection among these vectors, expression control sequences, and hosts without undue experimentation and without departing from the scope of this invention. For example, in selecting a vector, the host must be considered because the vector must replicate in it. The vector's copy number, the ability to control that copy number, and the expression of any other proteins encoded by the vector, such as antibiotic markers, should also be considered.

In selecting an expression control sequence, a variety of factors should also be considered. These include, for example, the relative strength of the system, its controllability, and its compatibility with the particular DNA sequence of this invention, parti¬ cularly as regards potential secondary structures. Unicellular hosts should be selected by consideration of their compatibility with the chosen vector, the toxicity of the product coded for on expression by the DNA sequences of this invention to them, their secretion characteristics, their ability to fold proteins correctly, their fermentation requirements, and the ease of purification of the products coded on expression by the DNA sequences of this invention. Within these parameters, one of skill in the art may select various vector/expression control system/host combinations that will express the DNA sequences of this invention on fermentation or in large scale animal culture, e.g., CHO cells or COS 7 cells. The polypeptides produced on expression of the DNA sequences of this invention may be isolated from the fermentation or animal cell cultures and purified using any of a variety of conventional methods. One of skill in the art may select the most appropriate isolation and purification techniques without departing from the scope of this invention. Examples 1-3 and 5-8 of this application illustrate the production of immunotoxins of this invention using genetic fusion techniques.

According to another embodiment of this invention, a hybrid DNA sequence comprising a DNA sequence coding for a toxin and a DNA sequence coding for soluble T4 protein may further comprise a signal sequence coding for a streptavidin-like polypeptide. Expression of this hybrid DNA sequence in an appro¬ priate host results in the secretion of the soluble T4- toxin fusion protein conjugate through the membrane of the cell. This greatly facilitates the production and purification of the soluble T4-toxin fusion protein conjugate, because the secreted soluble T4-toxin fusion protein is in its biologically active form.

Streptavidin is an antibiotic produced by the bacteria Strepto vces avidinii and other Streptomvces species [E. O. Stapley et al.. Antimicrobial Agents And Chemotherapy 1963. pp. 20-27 (J. C. Sylvester ed. 1964)]. It occurs naturally as a tetramer with a molecular weight of about 60,000 daltons. Streptavidin is characterized by its strong affinity for biotin and biotin derivatives and analogues. Each of the four identical subunits of steptavidin has a single biotin binding site [K. Hoffman et al., Proc. Natl. Acad. Sci. USA. 77, pp. 4666-68 (1980)].

As used in this application, "streptavidin- like polypeptide" refers to a polypeptide which displays the biological or immunological activity of native streptavidin and which is able to bind to biotin or biotin derivatives or analogues. This polypeptide may contain amino acids which are not part of native streptavidin or may contain only a portion of native streptavidin. The polypeptide may also not be iden¬ tical to native streptavidin because the host in which it is made may lack appropriate enzymes which may be required to transform the host-produced polypeptide to the structure of native streptavidin. The DNA and amino acid sequences corresponding to streptavidin- like polypeptides have been published in PCT patent application WO 86/02077, and are depicted in Figure 8 of the present application^'. A signal DNA sequence is that portion of a

DNA sequence coding for a polypeptide or protein which encodes, as a template for mRNA, a sequence of hydrophobic amino acids at the amino terminus of the polypeptide or protein, i.e., a "signal sequence" or "hydrophobic leader sequence" of the polypeptide or protein. A signal DNA sequence is located in a gene for a polypeptide or protein immediately before the DNA sequence coding for the mature protein or polypeptide, and after the translational start signal (ATG) of the gene. Expression of the signal DNA sequence together with the DNA sequence encoding a polypeptide or protein, produces a precursor of the polypeptide or protein.

A precursor of a protein or polypeptide is a polypeptide or protein as synthesized within a host cell with a signal sequence, e.g., prestreptavidin. A mature polypeptide or protein is secreted through a host's cell membrane with the attendant loss or clipping (i.e., maturation) of the signal sequence of its precursor.

It is believed that only a portion of a signal sequence of a precursor of a protein or poly¬ peptide is essential for the precursor to be trans¬ ported through the cell membrane of a host and for the occurrence of proper clipping of the precursor's signal sequence to form the mature protein or polypeptide during secretion. Hence, the term "signal DNA sequence" means the DNA sequence which codes for the portion of the signal sequence essential to secretion of a precursor of a protein, polypeptide or peptide, produced within a host cell.

A soluble T4-toxin fusion protein conjugate of this invention may be secreted through the membrane of the host cell, upon expression in a suitable host, of a hybrid DNA sequence comprising a DNA signal sequence coding for a sufficient portion of a pre- streptavidin-like polypeptide to cause the resulting fusion protein to be secreted through the membrane of the host cell, and a DNA sequence coding for the soluble T4-toxin fusion protein conjugate. When expressed, the soluble T4-toxin fusion protein con¬ jugate is secreted through the membrane of the host cell transformed by the hybrid DNA sequence.

The hybrid DNA sequences of this invention may additionally code for a sufficient portion of a streptavidin-like polypeptide to allow binding of the resulting soluble T4-toxin fusion protein conjugate to biotin or its derivatives or analogues. Upon expression of this hybrid DNA sequence, the streptavidin moiety of the soluble T -toxin fusion protein conjugate may be bound to biotin or to one of its derivatives or analogues. Other secreted proteins or contaminants which do not bind to biotin may then be washed away and the soluble T4-toxin fusion protein conjugate eluted from the biotin. This embodiment of the present invention advantageously provides a method for producing a soluble T4-toxin fusion protein con¬ jugate that may be easily purified due to the binding affinity of streptavidin-like polypeptides for biotin and its derivatives or analogues.

Example 3 of this application illustrates the production and subsequent secretion of a soluble T4- toxin fusion protein conjugate using a streptavidin- like polypeptide as a signal sequence. The immt otoxins of this invention may also be used in combination with other therapeutics employed in the treatment of AIDS, ARC and HIV infection. For example, these immunotoxins may be used in combination with antiretroviral agents that block reverse transcriptase, such as AZT, HPA-23, phosphonoformate, suramin, ribavirin and dideoxycytidine. Additionally, these immunotoxins may be used^* with antiviral agents such as interferons, including alpha interferon, beta interferon and gamma interferon, or glucosidase inhibitors, such as castanospermine. Further, the immunotoxins and fusion proteins of the present invention may be administered with other rsT4- immunotoxins or toxins as described herein, which may result in an additive or synergistic effect. Such combination therapies advantageously utilize lower dosages of those agents, thus avoiding side effects which may be associated with high dosage regimens of those agents alone.

The pharmaceutical compositions of this invention typically comprise an immunotherapeutically effective amount of an immunotoxin of this invention and a pharmaceutically acceptable carrier. Therapeutic methods of this invention comprise the step of treating patients in a pharmaceutically acceptable manner with those compositions. An immunotherapeutically effective amount of an immunotoxin of this invention is that which is sufficient to lessen the immunocompromising effects of HIV infection, to prevent the spread of HIV infection or to kill HIV or HIV-infected cells. The pharmaceutical compositions of this invention may be in a variety of forms. These include, for example, solid, semi-solid and liquid dosage forms, such as tablets, pills, powders, liquid solutions or suspensions, liposomes, suppositories, injectable and infusable solutions. The preferred form depends on the intended mode of administration and therapeutic application. The compositions also preferably include conventional pharmaceutically acceptable carriers and adjuvants which are known to those of skill in the art. Generally, the pharmaceutical compositions of the present invention may be formulated and administered using methods and compositions similar to those used for pharmaceutically important polypeptides such as, for example, alpha interferon. Thus, the immunotoxins of this invention may be stored in lyophilized form, reconstituted with sterile water just prior to administration, and administered by conventional routes of administration such as parenteral, subcutaneous, intravenous, intramuscular or intralesional routes. An effective dosage may be in the range of about 0.5 to 0.25 mg/kg body weight/day, it being recognized that lower and higher doses may also be useful.

In order that this invention may be better understood, the following examples are set forth. These examples are for the purposes of illustration only, and are not to be construed as limiting the scope of the invention in any manner.

Examples 1-3 illustrate the production of soluble T4-toxin fusion protein conjugates using modified Pseudomonas Exotoxin A and recombinant soluble T4 protein (rsT4) . Example 3 further illustrates the construction and the secretion of a soluble T4 protein- toxin fusion protein conjugate using a streptavidin- like polypeptide as a signal sequence. Example 4 illustrates a soluble T4 protein-toxin conjugate produced via chemical coupling techniques. Example 5 illustrates the production of a soluble T4-toxin fusion protein conjugate comprising a soluble T4 polypeptide having the formula AA..-AA _g3 of Figure 2 and a portion of Pseudomonas Exotoxin A comprising domains II and III. Example 6 illustrates the^" production of a soluble T4-toxin fusion protein conjugate comprising a soluble T4 polypeptide having the formula AA₃-AA._g of Figure 2, a portion of Pseudomonas Exotoxin A comprising domains II and III, and the boundary region of Pseudomonas Exotoxin A. Example 7 illustrates the production of a soluble T4-toxin fusion protein conjugate comprising a soluble T4 polypeptide having the formula AA₃-AA._g3 of Figure 2, a portion of

Pseudomonas Exotoxin A comprising domains II and III, and the boundary region of Pseudomonas Exotoxin A. This protein was expressed from a modified nucleotide sequence which facilitated its purification. Example 8 illustrates the expression, isolation, and purification of the immunotoxins, PEx45, PEx46, and PEx56. Examples 9 and 10 illustrate the .in vitro activities of the immunotoxins of this invention. Example 11 illustrates the expression, isolation, and purification of TANG11 and TANG12. Example 12 illustrates the expression, isolation, and purification of TANG13 and TANG14.

EXAMPLES

In these examples, we conjugated soluble T4 proteins to toxins to form the immunotoxins of this invention. The soluble T4 protein used was recombinant soluble T4 protein ("rsT4") supplied by Biogen Research Corp. (Cambridge, Massachusetts) . The rsT4 was derived from a Chinese hamster ovary cell transfected with pBG391, which is characterized by DNA coding for AA_₂₃ to AA_₇₇ of T4 protein, a full-length soluble T4 protein, as depicted in Figure 2 [R. A. Fisher et al., "HIV Infection Is Blocked In Vitro By Recombinant Soluble CD4", Nature. 331. pp. 76-78 (1988)]. In the examples that follow, the molecular biology techniques employed, such as cloning, cutting with restriction enzymes, isolating DNA fragments, treating with Calf Intestinal alkaline Phosphotase (CIP) , ligating, and transforming E.coli. are standard protocol exemplified and further described in

T. Maniatis et al.. Molecular Cloning. Cold Spring Harbor Laboratory (1982) .

Example 1

BACTERIAL CELL EXPRESSION As depicted in Figure 4, we isolated DNA from

PA103, a culture of the natural producer of Exotoxin A, Pseudomonas aeruginosa. [American Tissue Culture Collection (ATCC) , accession no. 29,260], using standard procedures. We then constructed a cosmid library by cloning completely digested Pseudomonas DNA into an EcoRI-digested pHC79 plasmid (Boehringer Mannheim, Cat. No. 567,795). We then synthesized an oligonucleotide probe (PA85) , corresponding to nucleotides 827-840 of the published sequence of PE [G. Gray et al., Proc. Natl. Acad. Sci. USA, 81, pp. 2645-49 (1984)], using standard procedures and a 380A DNA synthesizer (Applied Bio- systems Inc.). We probed the cosmid library with PA85, and isolated cosmids which contained a 6Kb EcoRI fragment. Based upon the published sequence of PE and a Southern Blot, we determined that the 6Kb EcoRI frag- ment contained the complete PE sequence. One of the cosmids which tested positive for the fragment, designated pEXl, was then selected for further work.

Based upon the published sequence of the gene coding for PE, we knew that domain II of the gene begins just downstream of the BgJ.II site at 1500bp [J. Huang et al., Cell, 48, pp. 129-36 (1987)], and that domain II and the domain for ADP-ribosylation codes through the C-terminal portion of the PE sequence. We also knew that the PE structural gene could be subcloned as a 2.7Kb EcoRI-Pstl fragment. Therefore, we cut pEXl with EcoRI and Pstl, and subcloned the resulting fragment into EcoRI/Pstl- digested pUC19 (Bethesda Research Laboratory, Cat. No. 5364SA) , to produce pEX2. Several expression vectors utilize a Clal site downstream of the promoter for insertion of a gene of interest. Because the gene coding for PE may be conveniently employed as a BgLII-Clal fragment, a Clal site was introduced into the EcoRI site of pEX2 using oligonucleotide linkers, and the resulting 1.2Kb Bglll- Clal fragment was subcloned into Clal/Bglll-digested pHC79, to yield pEX3.

The 1.2Kb Bglll-Clal fragment of pEX3 contains domains II, lb and III of the Pseudomonas Exotoxin A gene [J. Hwang et al., Cell_r supra]. That fragment may be linked to DNA sequences coding for soluble T4 proteins to form a hybrid DNA sequence with Bglll sites engineered into the coding sequence. Upon expression, such hybrid DNA sequences code for an immunotoxin of this invention: a soluble T4-toxin fusion protein conjugate.

Accordingly, as depicted in Figure 5, we constructed a shortened version of rsT4 containing a 124 amino acid segment and a convenient Bglll site as follows. Plasmid pl99.7 [IVI #10191], which contains full length soluble T4 protein, was cut with BstNI. Klenowed, and a Bglll linker (NEB#1036) was ligated to the blunt ended DNA. Following inactivation of the ligase at 65°C, the mixture was digested with Clal and Bglll. Plasmid pHC79 was also cut with Clal and Bglll and the 2.7Kb fragment containing the ampicillin (Amp) marker and ori was electroeluted. This fragment was ligated with the pl99.7 digest and the mixture used to transform E.coli strain DH5 (Bethesda Research Laboratories, Cat. No. 8265SA) . Ampicillin resistant colonies were then probed using a T4-6 oligo probe, corresponding to a region of T4 protein (^AA ₉ _₁₀₃^ ' synthesized from the published sequence (Maddon et al., Cell, supra) . A clone (pEX15) was identified which carried an amino acid sequence extending 124 amino acids downstream of the initiating Met of the T4 protein sequence (AA ,-AA.₂₄, Figure 3). By sequencing analysis, it was established that the Bglll site was correctly inserted at the BstNI site at amino acid 124 (Figure 3) . The pEX15 plasmid was then digested with Clal and Bglll. and the 380 bp fragment containing the 124 amino acid rsT4 (AA_ -AA.₂₄, Figure 3), was electroeluted. pEX3 was cut with Clal and Bglll and the 1.2kb Clal-BgLII fragment was also electroeluted. The above-described 124 amino acid rsT4 and the segment containing Pseudomonas Exotoxin A domains II and III from pEX3 were then joined by ligating the Clal-Bglll fragments into the Clal site of Clal-cut expression vector pl97.12, to yield plasmid pEX27. Portions of pEX27 were sequenced to establish that they contained the soluble T4-toxin fusion protein coding sequence. All junction regions were found to be correct. We then tested for expression of pEX27 as follows. SG936, an E.coli Ion htpr double mutant [S. Goff and A. Goldberg, "ATP-Dependent Protein Degradation In E.Coli". Maximizing Gene Expression. W. Reznikoff and L. Gold (eds.) (1986)] was transformed with pEX27 by conventional procedures [T. Maniatis et al. , Molecular Cloning, p. 187, Cold Spring Harbor Laboratory (1982)]. Induction showed Western positive material which ran at a molecular weight of about 60 kd, confirming the presence of the soluble T4-toxin fusion protein conjugate.

The selectivity and specificity of the rsT4- PE fusion protein conjugate may be assessed on uninfected and HIV-infected H9 cells as follows. Twenty-four well cluster plates are seeded with 200,000 non-HIV infected H9 cells per well [M. Popovic et al., "Detection, Isolation And Continuous Production Of Cytopathic Retrovirus (HTLV-III) From Patients With AIDS And Pre-AIDS", Science. 224. pp. 497-500 (1984)], various concentrations of rsT4-PE fusion protein conjugate added, and the cells incubated at 37°C overnight. The cells are then pulse-labelled with ( H)-leucine at 1 microCurie/ml for two hours to evaluate protein synthesis, as assessed by incor¬ poration of leucine into trichloroacetic acid- precipitable radioactivity. Alternatively, HIV infected cells are incubated for four days with the rsT4-PE fusion protein conjugate at various concen¬ trations, (³H)-thymidine is added to 1 microCurie/ml, the cells are then incubated overnight at 37°C, and cell viability assessed by incorporation of thymidine into trichloroacetic acid-precipitable radioactivity.

Example 2

ANIMAL CELL EXPRESSION rsT4-PE immunotoxins according to this invention may also be produced in eukaryotic tissue culture. Accordingly, as depicted in Figure 6, we linked the 1.2kb PE fragment obtained as described in Example 1 to a shortened 182 amino acid version of soluble T4 protein as follows. Plasmid pBG391 [IVI #10193], which contains full length soluble rsT4, was cut at the Stul site corresponding to amino acid 184 of the T4 protein sequence (bp 1825, Figure 2) . A Bglll linker (NEB #1052) , which puts the site in the correct reading frame, was ligated to the cut DNA. The resulting plasmid, pEX5, contained the necessary expression machinery for animal cells to direct the expression of a 182 amino acid rsT4 (AA₃-AA₁₈₄) •

We then modified pEX3 (as described in Example 1) so that domains II and III of the PE gene were located on a Bglll fragment that could be ligated into the pEX5 plasmid. To do this, we changed the Clal site on pEX3 to a Bglll site. The resulting plasmid, pEX4, contained the II, lb, and III regions of the gene coding for Pseudomonas Exotoxin A on a Bglll fragment.

This 1.2kb fragment was electroeluted and ligated into pEX5 to yield pEXll. This plasmid was used for transient expression of an immunotoxin of this invention in COS 7 cells [derived from African Green Monkey cells] . As a positive control for soluble T4 protein, the full length soluble construct, pBG391, was used. pBG393, a plasmid which produces a truncated 182 amino acid version of rsT4, was also used as a positive control for soluble T4 protein. This vector was constructed by placing a translational stop signal at the Stul site at bp 1825 of pBG391 (Figure 2) . The truncated rsT4 produced by pBG393 is comparable to full length rsT4 in its ability to bind OKT4A. Media from transient expression using pBG391, pBG393 and pEXll was harvested after 48 hours. The media was immunoprecipitated using OKT4A [Ortho Diagnostics #7142]. No soluble T4 protein specific material was precipitated from pEXll transformants, whereas both the pBG391 and pBG393 transformants produced soluble T4 protein.

Northern analysis of the transient cell lines was carried out to establish RNA levels specific to soluble T4 protein. Both pBG391 and pBG393 transient lines gave strong signals hybridizing to a T4 protein probe. There was no T4 protein specific signal in the pEXll transients.

The Northern Blot result was negative, because the cells transformed the with immunotoxin construct had died. We believe that the toxin moved into the cytosol during secretion, thus killing the cells. Nonetheless, we believe that expression may be achieved in eukaryotic cell lines that are resistant to the toxin. We also believe that.-the immunotoxin con- structs of this invention may be expressed in yeast cell cultures.

Example 3

SECRETION OF SOLUBLE T4-TOXIN USING A STREPTAVIDIN SIGNAL SEQUENCE In this example, we used a streptavidin-like polypeptide signal sequence, to secrete a soluble T4- toxin fusion protein conjugate into the periplasm of E.coli, as well as into the culture media. This greatly facilitated the production and purification of the fusion protein, because the protein was secreted in its mature, correctly folded form. The streptavidin gene has been cloned and used successfully to secrete other heterologous proteins (PCT patent application WO 86/02077) . The starting plasmid pSA307 and the coding sequence for streptavidin and streptavidin-like polypeptides are depicted in Figure 8.

As illustrated in Figure 7, the expression plasmid p211.11 [IVI #10190], which contains an amino acid sequence extending 113 amino acids downstream of the initiating Met (AA .-AA..-, Figure 3), was first transformed into the host strain DH5, which methylates the upstream Clal site, making it resistant to cutting. The plasmid was then isolated from the transformed DH5, cut with Clal. at its downstream site, Klenowed and transformed into the methylation negative strain GM2929 [IVI #10187]. The resulting plasmid pSA400 contained only the Clal site upstream of the 113aa rsT4 coding sequence. Plasmid pSA400 was then cut with Clal. followed by limited digestion with Mung Bean Nuclease to remove the 5' CG overhang of the Clal site.

Meanwhile plasmid pSA307 [IVI #10194] was digested with Hindi and Alul and electroeluted. The resulting 429 bp fragment included the Hindi site at bp 194, up to the Alul site at bp623 (Figure 8). It contained the promoter, the translational start codon ATG, the signal sequence and 12 amino acids of the mature streptavidin protein.

This 429 bp HincII-AluI fragment was then ligated into the blunted pSA400 Clal site. The resulting plasmid, pSA410, contained the SA signal (Figure 7) genetically linked to the 113aa rsT4 (AA ,-AA _, Figure 3). Plasmid pSA410 was then cut with PflMl and Hindlll, and the resulting 612 bp frag- ment was ligated into PflMl and Hindlll digested plasmid pSA313 [IVI #10195]. The resulting plasmid, pSA415, contained the SA signal sequence linked to the

rsT4 segment at the Hind III site (bp 249,

Figure 3) . The remainder of the I24aa rsT4 coding sequence and the toxin was derived from pEX27. Plasmid pEX27, obtained according to Example 1, was digested with EcoRI. Klenowed, and ligated with Hindlll linkers, followed by digestion with Hindlll. The resulting 1.45kb fragment containing the soluble T4-toxin fusion protein conjugate was electroeluted and ligated into the Hindlll site of pSA415. The resulting plasmid, pEX39, contained the lac promoter directing expression of a hybrid protein comprising streptavidin's signal sequence and 12aa of the mature streptavidin protein, fused to a sequence containing a 124aa rsT4 (AA_ -AA-₂₄, Figure 3), which was fused to domains II and III of the Exotoxin A segment.

Plasmid pEX39 was then transformed into the E.coli strain HB101 IQ [IVI #10188]. The transformant HB101IQ/pEX39 was grown at 25°C in the presence of 5 mM IPTG. We tested for the expression and secretion of the hybrid protein into the cell periplasm and the extracellular medium using Western Blot and standard osmotic shock procedures.

We detected hybrid protein both in the periplasm and in the medium, suggesting that the hybrid protein was secreted out of the cell. To determine if the protein was properly folded, 0KT4A was used to immunoprecipitate the osmotic shock fraction. Upon analysis, it was shown that the hybrid secreted soluble T4-toxin fusion protein conjugate is precipitated by 0KT4A, suggesting that the protein is properly folded upon secretion. Additionally, both the osmotic shock fraction and the media containing the immunotoxin tested positive for ADP-ribosylation activity [H. Hwang et al., Cell, supra] further confirming the activity of the immunotoxin.

We believe that angiogenin, or toxic portions thereof, fused independently or in conjunction with other proteins which exhibit toxic effects may also be particularly well-suited to benefit from the use of the streptavidin signal sequence to assist in the secretion of these angiogenin comprising fusion proteins.

Example 4

CHEMICAL CONJUGATION

In this example, a ricin A chain is con¬ jugated to a full length amino acid, soluble T4 protein obtained according to R. A. Fisher et al., Nature. 331. pp. 76-78 (1988) . Highly purified ricin A chain is commercially available [XOMA Corporation, San Francisco] . The purity of the ricin A chain can be verified free of contaminating B chains by a whole cell toxicity assay. The full length amino acid soluble T4 pro¬ tein and purified ricin A chain are conjugated with the heterobifunctional crosslinker N-succinimidyl-3-(2- pyridyldithio)-propionate [SPDP, Pharmacia, Piscataway, New Jersey] according to standard protocols [A. J. Cumber et al., "Preparation Of Antibody-Toxin

Conjugates", Meth. Enzvmol. , 112 pp. 207-25 (1985)], as follows.

SPDP, dissolved at 6 mg/ml in dry dimethyl- formamide, is added to a solution of the soluble T4 protein at 10 mg/ml in Dulbecco's calcium and magnesium-free phosphate buffered saline (PBS) ; 10 microfilters SPDP per ml of the soluble T4 protein. The mixture is incubated for 30 minutes at room temperature, then applied to a column of Sephadex G25 equilibrated with PBS to separate the rsT4-SPDP conjugate from unreacted SPDP.

Ricin A chain is then fully reduced with dithiothreitol (DTT) as follows. DTT is added to a solution of ricin A at 10 mg/ml in PBS to 0.1 M final concentration. The solution is incubated for one hour at room temperature. The fully reduced ricin A chain is then separated from excess DTT by gel filtration on Sephadex G25 in PBS. The fully reduced ricin A chain is imme¬ diately incubated with soluble T4-protein-SPDP, for two hoμrs at room temperature, then for two days at 4°C, to effect coupling. The resulting soluble T4 protein ricin A conjugate is separated from uncoupled soluble T4 protein and ricin A chain by sequential chromato¬ graphy on an affinity column of Cibacron blue F3GA- Sapharose [P. P. Knowles and P. E. Thorpe, "Purifi¬ cation Of Immunotoxins Containing Ricin A Chain And Abrin A Chain Using Blue Sepharose CL-68", Analytical Biochemistry. 160. pp. 440-43 (1987)], to remove soluble T4 protein; and in an immunoaffinity column containing T4 protein specific antibody [R. A. Fisher et al., "HIV Infection Is Blocked In Vitro By Recom¬ binant Soluble CD4", Nature, supra1. to remove excess ricin A chain.

The conjugation of soluble T4 protein and ricin A is confirmed by sodium dodecyl sulfate- polyacrylamide gel electrophoresis and by radio- immunoassay with the use of rabbit anti-mouse Ig- coated microtiter plates and 125I labeled rabbit anti- ricin A chain antiserum [M. Okhuma and J. Poole, "Fluorescence Probe Measurement Of The Intralysomal pH In Living Cells And The Perturbation Of PH By Various Agents," Proc. Natl. Acad. Sci. USA. 75, p. 3327 (1978) ] .

Each molecule of soluble T4 protein is thus associated with at least one molecule of ricin A chain. These conjugates can be stabilized with carrier protein (reduced and alkylated human IgG) , and stored at -70°C.

Example 5

SOLUBLE T4 - PSEUDOMONAS EXOTOXIN A FUSION PROTEIN fPEx45

We then prepared PEx45, a soluble T4- Pseudo onas Exotoxin A fusion protein containing the boundary region of PE linked to the translocation and ADP ribosylation domains of PE. As depicted in Figure 9A, we cut plasmid pEXll (as prepared i.. Example 2) with Hindlll and EcoRI, and electroeluted the resulting restriction fragment encoding containing the soluble T4-PE43 fusion protein. We next cut the T4 expression plasmid p218* with EcoRI and Hindlll and electroeluted the 241 bp fragment encoding the N-terminus of T4. We then isolated the vector promoter portion of pEX27 (as prepared in Example 1) by digesting it with EcoRI and electroeluted the vector. We ligated the EcoRI-Hindlll fragment carrying the N-terminus of T4 and the HindiII-EcoRI fragment from pEXll to the vector promoter fragment isolated from pEx27 to yield pEX45. The pEX45 plasmid contains DNA encoding a polypeptide of the formula AA₃-AA._g3 of Figure 2 linked

* The expression vector p218 is prepared from λ pL DC197 containing a soluble T4 protein of the formula AA. -

A ^****.A**._g2 o -- Λf F-Cli.!g-r,u.-r_e— 2->. 1 at AA__, to a portion of Pseudomonas Exotoxin A comprising AA₂₂₄-AA₆-_ of that protein (as illustrated in Figures 11 and 13) . We transformed the E.coli strain A89 with pEx45. E. coli A89 is a tetracycline sensitive derivative of E. coli SG936. E.coli A89 had been isolated from E.coli SG396 according to the method • of S.R. Maloy and W.D. Num , "Selection For Loss of Tetrachycline Resistance By Escherichia coli". J. Bact.. 145. pp. 110-12 (1981) . We then grew the A89/pEX45 transformant under conditions of temperature shift induction.

Using standard techniques, we renatured and purified the PEx45 fusion protein after temperature shift induction. We then used the resulting fusion protein as an immunotoxin to selectively kill a CHO cell line expressing gpl20/160 on its surface. The killing level of 50% was determined to be at a concentration of 5 x 10^~10 M as further described in Example 8 and depicted in Figure 15. Figure 11 depicts the amino acid sequence of the immunotoxin product of hosts transformed with pEX45.

Example 6

SOLUBLE T4 - PSEUDOMONAS EXOTOXIN A FUSION PROTEIN fPEx46)

We next prepared PEx46, a soluble T4- Pseudomonas Exotoxin A fusion protein in which all of the binding region of PE was removed and which did not include the boundary region of PE (See Figures 9A, 9B, and 13) . As depicted in Figure 9B, we cut pEX3 (as prepared in Example 1) with Bglll and Aval. We then ligated the Bglll-Aval fragment to an oligonucleotide adapter having the sequence GAT CAT CAT CGA TCT CCA TGG GGA AGA TCT ACC to place a Bglll site immediately upstream of the Aval site in PE, yielding plasmid pEX31.

We then cut pEX31 with Bglll and EcoRI, and electroeluted the 1.2 kb fragment encoding PE40. We also cut pEX27 (as prepared in Example 1) with EcoRI and Bglll and electroeluted the resulting fragment encoding the N-terminus of T4. We ligated these two fragments into a EcoRI vector promoter segment which had previously been isolated from pEX27 (as described in Example 5) . The resulting plasmid, pEX44, contained DNA encoding soluble T4 protein linked to PE40 which corresponds to PE at AA₂₅₃-AA_gl3 (See Figures 13 and 23) .

As shown in Figure 9A, we then cut pEX44 with EcoRI and Bglll, and again electroeluted the fragment encoding PE40. We next cut pEX45 (as prepared in Example 5) with EcoRI and Bglll. and electroeluted the 615 bp fragment encoding the N-terminus of T4. We then isolated the vector promoter segments by cutting pEX45 with EcoRI and isolated the resulting 2.7 kb fragment. The three fragments — EcoRI-BgJLII from pEX45, the BglH-EcoRI from pEX44, and the vector promoter EcoRI fragment from pEX45 — were ligated to produce plasmid pEX46. That plasmid encodes a soluble T4 polypeptide of the formula AA.-AA.„- of Figure 2 linked to PE40. This construct was then transformed into the E.coli strain A89. Following temperature shift induction, we renatured and isolated the protein. We found that the fusion protein selectively killed CHO cells expressing gpl20/160 on their surface (See Figure 15) . In addition, PEx46 also killed HIV infected H9 cells. Figure 12 depicts the amino acid sequence of the immunotoxin product of hosts transformed with pEX46. In addition, we also constructed a pEX60 plasmid which contained a soluble T4 protein (having the formula AA..-AA of Figure 2) fused to only the ADP ribosylation cell inactivation domain of PE (without the translocation domain of PE) (See Figure 23) . This fusion protein failed to exert a toxic effect on cells expressing gpl20/160 on their surface. Accordingly, we believe that the translocation domain of PE is necessary to permit the fusion proteins of this invention to enter the cytoplasm and therefore inactivate protein synthesis in target cells.

Example 7

SOLUBLE T-4 PSEUDOMONAS EXOTOXIN A FUSION PROTEIN fPEx56)

The DNA sequences surrounding AA_nyo_c of T4 permit the initiation of translation of a second, shortened version of a soluble T4-Pseudomonas Exotoxin A fusion protein without the binding domain of PE. This shortened fusion protein interferes with the purification of the desired soluble T4-Pseudomonas Exotoxin A fusion protein in which the entire binding domain of PE is removed.

To avoid such interference, we prepared another soluble T4-Pseudomonas Exotoxin A fusion protein in which the entire the binding region of PE was removed (See Figure 23) . Specifically, we mutagenized plasmid pEX46 (see Figure 9A) with oligonucleotide T4-176 to form plasmid pEX56. Oligonucleotide T4-176 has the following bp sequence: GGA GGA CCA GAA AGA AGA AGT TCA GCT GCT GGT TTT CGG ATT GAC T. In the resulting plasmid pEX56, the initiation of protein translation does not occur at AA_Q,. of the T4 portion of this fusion protein. The amino acid sequence of PEx56 is identical to that of PEx46 (see Figure 12) . Example 8

EXPRESSION OF PEx45. PEX46 and PEX56

We expressed each of PEx45, PEx46, and PEx56 in E.coli and achieved induction of the lambda pL promoter by temperature shift. In each case, we found the expression level of the protein of interest to be approximately 15-20% of total cell protein as determined by densitometric scanning of a protein gel.

Renaturation and Immunoaffinity Purification of PEx45, PEx46 and PEx56

We suspended 10 g of E. coli cell pellet in 40 ml of 50 mM Tris-HCl, adjusted to about pH 8.0, containing 2 mM EDTA at 0°C. We then ruptured the suspended cells in a French press and pelleted the inclusion bodies at 10,000g for a period of about 30 min. at about 4°C. We washed the pellet by resuspending it into 50 ml of 50 mM Tris-HCl, adjusted to about pH 8.0, containing 2 mM EDTA and the inclusion bodies were subsequently repelleted at 10,000g for a period of about 30 min. at about 4°C.

We next dissolved the washed pellet in 50 mM Tris, 7 M guanidine-HCl, 2 mM EDTA, 10 mM DTT, adjusted to about pH 8.0 with HC1 at about 4°C. After about 1 hour of gentle agitation at about 4°C, we clarified the solution by centrifugation at 10,000g for a period of about 40 min. at about 4°C. We then renatured the fusion protein by dropwise addition, at the rate of 1 ml/min, of the fusion protein containing sample into a well stirred buffer of 50 mM Tris-HCl, adjusted to about pH 8.0, having a dilution ratio of 1:100. We allowed renaturation to continue for a period of about 24 hours at about 4°C under rapid stirring. This was performed in an open container, or alternatively, in a vessel under the positive influence of air or oxygen to facilitate oxidation.

We clarified the solution by filtering it through a 5 μm and 0.2 μ low protein binding filter (Millpore, Inc., Massachusetts). We then pumped the solution at a rate of about 10 ml/hr through a 10 ml immunoaffinity column at about 4°C. As the stationary phase of the column, we used a monoclonal antibody 6C6 (a gift of Drs. Blake Pepinsky and Werner Meier, Biogen, Inc. , Cambridge, Massachusetts) coupled to BrCN activated Sepharose 4B (Sigma, Inc., St. Louis, Missouri) . This antibody is known to interfere with gpl20/160 and OKT4 binding.

After loading the column with the protein containing sample, we washed the column with about 10 times the column volume with 50 mM Tris-HCl, pH 8.0. We eluted antibody bound proteins with 50 mM Tris, 3M KSCN, adjusted to pH 8.0. Immediately after elution, we exchanged the buffer to Dulbecco's PBS, without Ca^2* and Mg^7*, by repeated dialysis with stirring at about 4°C.

We verified the structure of the purified fusion proteins by one or more of the following identification methods: N-terminal sequence analysis, amino acid analysis, and bioactivity.

Without optimizing our purification condi¬ tions, we recovered at least about 0.5 - 1.0 mg/g E.coli of the fusion proteins PEx45, PEx46, and PEx56. We measured the purity of each of these fusion proteins to be > 90% by conventional SDS-PAGE (12% polyacryl- amide gel under reducing and non-reducing conditions) and HPLC size-exclusion chromotography (TSK 2000 SW, Toya Soda Corp., Japan).

We determined each of the fusion proteins, PEx45, PEx46, and PEx56, to have a monomeric physical form by using non-reducing SDS-PAGE and HPLC size- exclusion chromotography in phosphate buffer, adjusted to about pH 7.0.

Example 9

IN VITRO ACTIVITY OF THE IMMUNOTOXINS OF THIS INVENTION

The activities of PEx27, PEx44, PEx45, PEx46 and PEx56 immunotoxins were assessed in vitro by measuring gpl20 binding activity by direct competition with rsT4.3 in a solid phase receptor binding assay. The toxicity of PEx45 and PEx46 was then determined by an in vitro ADP ribosylation assay using elongation factor 2 as a substrate.

Without being bound by theory, we believe that the rsT4-PE immunotoxin fusion is taken up by endocytosis. Once in the endosome or lysosome, the fusion protein is then exposed to low pH conditions thereby inducing a conformational change in its translocation domain. This structural change of the translocation domain is likely to be the cause of the permeation of the fusion protein or portions thereof into the cytoplasm. Finally, the cell inactivating domain of the fusion protein performs the ADP ribosylation of elongation factor 2 which results in the blocking of protein translation and eventuating the death of the invaded cell. Only cells bearing gpl20/160 on their surface should be affected by this immunotoxin cell incorporation"process. gpl20 Binding Activity We coated Immulon II ELISA plates (Dynatec) with purified recombinant gpl20 by incubating each well with a solution of 50 μl of gpl20 (OD_2g_ = 0.005). After we removed the gpl20 by washing the plate with Dulbecco's PBS, without Ca^2* and Mg^2*, we incubated the gpl20-coated plate with samples (25 μl/well) of fusion proteins PEX27, PEX44, PEX45, PEX46 or PEX56 at dilutions ranging from about 1000 μM to 0.5 μM. We diluted the constructs in Dulbecco's PBS in the presence of milk proteins in order to block non¬ specific binding. The negative control used was rsT4.3, the soluble T4 polypeptide expression product of pBG381 (IVI 10149) . After a period of at least about 1 hour, we added horseradish peroxidase conjugated to rsT4.3 (HRP-rsT4) at a concentration of about 60 ng/ml (25 μl/well) as a competitor.

We continued incubation for at least about 2 hours. After we had washed the plates with Dulbecco's PBS, without Ca²⁺ and Mg⁺, containing about 0.05% Tween 20 (Sigma, Inc., St. Louis, Missouri), we detected the gpl20-HRP-rST4 complex by reacting 0-phenylenediamine (OPD, Sigma, Inc., St. Louis, Missouri) and H„0 with the horseradish peroxidase substrate and then scanned the ELISA plate at 490 nm (as depicted in Figure 25) . Fusion proteins containing 124 amino acids of T4 (PEx27 and PEx44) demonstrated an affinity to gpl20, while the proteins comprising 181 amino acids of T4 (PEx45 and PEx46) demonstrated a higher binding affinity to gpi20. The ribosylation activity was determined to be about the same for PE40 (PEx44 and PEx46) and PE43 (PEx27 and PEx45) containing proteins.

ADP Ribosylation of EF-2

The toxicity of Pseudomonas Exotoxin A is based on its enzymatic activity which carries out ADP ribosylation of elongation factor 2 (EF-2) thereby inhibiting protein translation of a target cell. This ADP ribosylation activity may be assayed by measuring the transfer of radioactively labeled substrate ("C- NAD^*) to EF-2. The fusion protein becomes bound with unpurified EF-2 on one domain and onto another domain with ^UC and NAD^* which bridges the substrates together to yield ADP on EF-2. As a positive control in this assay, we used PE [List Laboratories, California]; rsT4.3 served as a negative control. We also used a rabbit reticulocyte lysate as the source for EF-2.

As illustrated in Figure 14, we observed no significant difference in the ADP ribosylation activity of the PE40 and PE43 containing proteins (PEx45 and PEx46) . Moreover, the activity did not change when each of the PE43 containing fusion proteins were reduced prior to the assay (see for example PEx46 in Figure 14) . This indicates that the status of the fusion protein's disulfide bridges, located in the translocation domain and in the gpl20 binding domain, may not exert any influence on the ADP ribosylation activity.

Example 10

ADDITIONAL ASSAYS OF THE ACTIVITIES OF THE IMMUNOTOXINS OF THIS INVENTION

According to the design of the rsT4-PE fusion protein, the fusion protein should bind to cells which express gpl20/160 on the surface of HIV infected cells. The bound fusion protein should be taken up by endocytosis as with PE. Once in the endosome or lysoso e, the fusion protein is then exposed to low pH which should induce a conformational change in the translocation domain. Thus, presumably hydrophobic stretches are exposed to the surface thereby interacting with the membrane of the transport vesicle. This should result in a translocation of the ADP ribosylation domain of PE into the cytoplasm.

The enzymatically active domain of the fusion PE portion of the immunotoxins of this invention carries out the ADP ribosylation of EF-2, facilitating the blocking of protein translation. Eventually, cell death occurs. Cells which do not bear gpl20 on the surface should not be affected by the hybrid toxin. We tested in cell assays the potency of immunotoxins, PEx45 and PEx46, to kill gpl20/160 bearing cells. In the first assay, we used gpl60 expressing Chinese Hamster Ovary (CHO) cells as target cells; as control cells, we used MIS expressing CHO cells. In the second assay, we used HIV-infected H9 cells and uninfected H9 cells as controls. H9 specifies a CD4 positive human cell line. We used

H-thymidine incorporation as a measure of viability. We refer to this type of assay as a proliferation assay.

In a third assay, we used CHO^gp160 cells as a target cell line and CHO^IS cells as a control cell line. We determined the effect of the toxin by plating out and counting the number of surviving cells after they had been treated with either PEx45 or PEx46. We refer to this type of cell assay as a plating efficiency assay. The design of the rsT4-PE immunotoxin should permit binding to cells which express gpl20/160 on the surface of the cell lines in the above-described cell assays.

Proliferation Assay

We used a gpl60 expressing CHO cell line (CHO^SB,6°) as target cells and MIS-'-(Mullerian inhibiting substance) expressing CHO cell line (CHO^*ls) served as control cells in a cell proliferation assay. We treated a defined number of cells with a sample of immunotoxin for a period of about 24 - 48 hours in α- MEM, supplemented with GLN and 10% FBS. We then added H-labelled thymidine to the medium and the cells were allowed to incubate for a period of about another 24 hours. We harvested the cells and then washed them on a standard glass-fiber filter paper. We then counted the cell-associated radioactivity and used the obtained value as a direct measure in determining the viability of the cells to incorporate labelled thymidine. As illustrated in Figure 15, the effects of Pseudomonas Exotoxin A (PE) , PEx45, and PEx46 on CHO⁹⁰'⁶⁰ and CHO^IS cells after treatment for a period of about 48 hours is presented by plotting "% thymidine incorporation" versus the concentration of the protein.

It is evident from Figure 15, that PE itself does not discriminate between gpl20/160 expressing cells and MIS expressing cells. Both cell lines show a strongly reduced ability to incorporate ³H-thymidine. PEx45 has similar cell killing potential to PE when it is used on expressing CHO cells. This demonstrates that PEx45 actually binds to the cells, and is then translocated into the cytoplasm thereby inactivating protein translation. Cells which are missing the T4 specific receptor gpl60 (CH0^*IS) provide the "therapeutic window" of this drug (illustrated by the light shadowed bar in Figure 10) . As depicted in Figure 15, PEx46 shows less toxicity on CHO""⁶⁰ cells than PEx45. The nonspecific toxicity of this fusion protein is also reduced which results in approximately the same therapeutic window as for PEx45 (illustrated by the dark shadowed bar in Figure 10) . These results are listed below in Table 1. Table 1

* ³H-thymidine incorporation of untreated cells is 100%

The concentrations at which a 50% ³H-thymidine incorporation is observed are compared in Table 1. As shown by Table 1, PEx45 shows a 10 fold higher potency than either PEx46 or PEx56 and thus resembles the potency profile of native PE. Both PEx45 and PEx46 show a difference between specific and nonspecific killing of about 5 x 10² M. As illustrated in Table 1, when cells are treated for shorter time periods, the "therapeutic window" for PEx46 increased to at least about 2.5 x 10³ M. This increase may be due to the inactivation of the toxin during the longer treatment, resulting in a partial recovery of cell viability.

We demonstrated the necessity of the presence of a freely accessible receptor (gpl20/160) on the surface of cells for the activity of the fusion proteins in a rsT4.3 blocking experiment. We treated CHO^gp16C cells with PE, PEx45, or PEx46 in the absence or presence of rsT4.3 at a concentration of about 5.4 x 10^"7 M. In the absence of rsT4.3, PE and each of the immunotoxins, PEx45 and PEx46, demonstrated similar killing abilities (see Figure 15) . As shown in Figure 10, rsT4.3 reduced the killing activity of PEx46 by blocking the receptor. The activity of PE was not influenced by the presence of rsT4. These results demonstrate that PEx45 and PEx46 enter the cells via binding to gpl20/160 on the cell surface.

• HIV-infected H9 cells were similarly treated with PEx45 and PEx46. We used HIV strain IIIB. Uninfected H9 cells served as a control in this cell assay. H9 cells are human CD4 positive T-lymphocyte cells. We have shown that these HIV-infected H9 cells express about 32,000 molecules of gpl20 on the cell surface. We demonstrated this by using ¹²⁵I-labelled anti-gpl20 antibody (DURDA, Dupont, Delaware) . Comparable results to those described above with the CHO""⁶⁰ cell line were obtained in this assay.

Plating Efficiency Assay

As depicted in Figure 24, a small number of dispersed cells, 1000 CHO""⁶⁰ cells or 300 CHO^MIS cells/21 cm² plate, were incubated for a period of about 12-24 hours and were treated with the protein sample for a period of about 24 hours. Cells were then dispersed and incubated in fresh medium for seven days. After fixing the cells with paraformaldehyde and staining with hematoxylin, the number of surviving cell colonies were counted. This assay may determine the "killing" ability of the toxin more directly than the proliferation assay. Preliminary data from this assay indicated that treatment of CHO""⁶⁰ cells with the rsT4- PE fusion protein results not only in a temporary suppression of ³H-thymidine incorporation, but also in cell death. Example 11

SOLUBLE T4-PE - ANGIOGENIN FUSION PROTEINS (TANG11 AND TANG12)

We cloned and characterized the DNA sequence encoding angiogenin. We then genetically fused angiogenin to a soluble T4-PE polypeptide to prepare additional immunotoxin fusion proteins according to this invention. To test for angiogenin's ability to function as a toxin, we cut the ADP ribosylation domain of the PE portion found in the soluble T4-PE fusion plasmid, pEX46, corresponding to AA₄_₄ of PE. We replaced the remaining portion of PE (AA_4g5-AA_{g 3}) with angiogenin, with and without its signal sequence.

As illustrated in Figures 17A and 17B, we screened a λ EMBL3 library constructed from DNA of human mutant fibroblast line 49, XXXXX GM5009 (a gift of Dr. Mark Pasek, Biogen, Inc., Cambridge, Massachusetts) with an oligonucleotide probe, ANG-86, having the following nucleotide sequence: pATA GTG CTG GGT CAG GAA GTG TGT GTA CCT GGA GTT ATC. This sequence is based on the amino acid sequence of angiogenin illustrated in Figure 16. We chose one of several clones identified — EMBL Angl — for further study. Based upon the published sequence of angiogenin [Kurachi, et al. Biochemistry. 24. pp. 5494-499 (1985)], we knew that that the structural gene for the protein is found on a 736 bp Bglll to EcoRV fragment. We cut EMBL Angl with Bglll and EcoRV to obtain the 736 bp fragment containing the angiogenin gene. We then ligated the Bglll-EcoRV fragment into pHC79 [Boehringer Mannheim, Indianapolis, Indiana] which, had been previously cut with Bglll and EcoRV and screened with ANG-86. The resulting plasmid, pANGl, contained angiogenin in the Bglll to EcoRV sites of pHC79. In order to express the mature form of angiogenin in E.coli, we introduced a Clal site immediately upstream of the coding sequence of mature angiogenin. To accomplish this, we mutagenized pANGl with the oligonucleotide ANGIO-1 to yield pANG3.

ANGI0-1 has the following nucleotide sequence: CTG GGT CTG GGT GGA TCC AAA TCG ATT TGG ATG CAG CAG GAT AAC TCC. We then cut the plasmid pANG3, having angiogenin flanked by Clal sites, with Clal. We then cloned the resulting Clal fragment into the Clal site of the lambda pL expression vector pl97.12, which had been previously cut with Clal and transformed E.coli A89. We selected for tetracycline resistant colonies and screened with ANG-86. As a result, we isolated plasmid pANG4 which contained mature angiogenin under the lambda pL promoter. We then transformed pANG4 into the E.coli strain A89. The resulting strain was induced and the angiogenin purified by conventional methods as described by P. Denefle et al., "Chemical Synthesis of a Gene Coding for Human Angiogenin, Its Expression in E.coli and Conversion of Its Product into Its Active Form", Gene. 5_6, pp. 61-70 (1987).

We next engineered the angiogenin gene such that it could be genetically fused to other proteins, such as soluble T4 protein and Pseudomonas Exotoxin A. We designed two different versions — one with and one without the angiogenin signal sequence.

The plasmid pTANGll contains parts of three different proteins. It carries a soluble T4 polypeptide of the formula AA--AA-_g3 of Figure 2 linked to a portion of PE comprising the translocation domain and a portion of the ADP ribosylation domain (AA₂₅₃~ AA₄_₄) of Pseudomonas Exotoxin A, followed by the gene for mature angiogenin. The fusion protein TANG11 comprises rsT4 (AA₁~AA _g ) , the translocation domain of PE (AA₁₈₄-AA₃₀₀) , the binding domain (lb) of PE (AA₃₀ -AA_34Q) , a linker from the ADP ribosylation domain of PE (AA₃₄ -AA.._g) , and angiogenin (AA₄₃ -AA₅₅₉) (as depicted in Figure 19) . We obtained pTANGll as illustrated in

Figure 17B. We mutagenized pANGl with the oligonucleotide ANGIO-2 to yield pANG7 which has a Bglll site immediately upstream of the mature angiogenin sequence. ANGIO-2 has the following nucleotide sequence: GGT CTG GAT CTG ACC AAG ATC TTG GGG GGG CCG GGG CAG GAT AAC TCC ACG. We then introduced a BamHI site downstream of the angiogenin gene by cutting pANG7 with EcoRV and subsequently ligating in a BamHI linker to yield pANG9. The plasmid pANG9 has mature angiogenin flanked by a Bglll site upstream and a BamHI site downstream.

We also digested plasmid pEX46 (as prepared in Example 6) with BamHI which cuts at AA₄_ of PE. This site is within the ADP ribosylation domain of the PE gene and thus should remove the ADP ribosylation function of pEX46. We then digested pANG9 with Bglll and BamHI to yield a 600 bp fragment containing angiogenin. We purified the fragment and ligated it into the BamHI-digested pEX46, to produce pTANGll. We obtained pTANG12, which contains angiogenin with its signal sequence, by introducing a Bglll site immediately upstream of the intiating Met of the angiogenin signal peptide (see Figure 17B) . Specifically, we mutagenized pANGl with oligonucleotide ANGIO-3 in order to produce pANG8. ANGIO-3 has a nucleotide sequence: CCG CAG GAG CCT GTG TAG ATC TTT ATG GTG ATG GGC CTG. We then introduced a BamHI site downstream of the angiogenin gene by cutting pANG8 with EcoRV and ligating in a BamHI linker to produce plasmid pANGlO. We then digested pANGlO with Bglll and BamHI to yield a 600 bp fragment containing angiogenin. This fragment was purified and then ligated into the BamHI digested pEX46, to produce pTANG12.

The plasmid pTANG12 contains angiogenin, complete with signal peptide fragment, substituted for the ADP ribosylation cell inactivating domain of PE. We transformed the plasmids pTANGll and pTANG12 into E.coli strain A89. We can then renature and purify the protein after temperature shift induction of the promoter. The fusion protein TANG12 comprises rsT4 (AA.-AA.₀_) , the translocation domain of PE (AA - -AA₃₀ ) , the binding domain (lb) of PE (AA₃ -AA ) , a linker from the cell inactivation domain of PE (AA -AA _g) , the angiogenin signal sequence (AA -AA ) , and angiogenin (AA -AA ) (as depicted in Figure 18) .

Example 12

SOLUBLE T4-ANGIOGENIN FUSION PROTEINS (TANG13 AND TANG14) In order to test for the ability of soluble

T4-angiogenin fusion proteins to function as toxins without a portion of PE, we linked angiogenin directly to the C terminal end of a soluble T4 protein having the formula AAJ--AAl.o_J- of Figure 2. As illustrated in Figure 20, we obtained plasmid pTANG13 by cutting pEX46 (as prepared in Example 6) with EcoRI and electroeluted the resulting expression vector. We also cut pEX46 with EcoRI and BgJ.II, and electroeluted the 615 bp fragment containing the rsT4 gene. We next cut plasmid pANG9 (as prepared in

Example 11) with EcoRI and Bglll. and electroeluted the 740 bp fragment containing the angiogenin gene. We then linked that fragment with the expression vector fragment and the 615 bp T4 fragment previously isolated from pEX46. Following transformation into E.coli A89, we identified the pTANG13 plasmid so produced.

The plasmid pTANG13 comprises DNA encoding a soluble T4 polypeptide of the formula AA.-AA.₀_ of Figure 2 linked to mature angiogenin. The protein may be purified and renatured after temperature shift induction of the promoter. It may then be tested for T4 binding, selective killing of gpl60 producing cells (as described in Examples 9 and 10) , and RNase activity. The fusion protein TANG13 comprises rsT4 (AA.-AA ) and angiogenin (AA._g.-AA- ) (as depicted in Figure 21) .

We also produced a soluble T4-angiogenin fusion protein in which angiogenin carries its signal sequence. We digested the plasmid pANGlO (as described in Example 11) with EcoRI and Bglll - and electroeluted the 800 bp angiogenin-containing fragment. We then ligated that fragment with an expression vector fragment and the 615 bp fragment previously isolated from pEX46 to produce plasmid pTANG14.

The plasmid pTANG14 comprises DNA encoding a soluble T4-polypeptide of the formula AA₃-AA._g3 of Figure 2 linked to angiogenin and the angiogenin signal sequence. We transformed E.coli A89 with pTANG14. The resulting protein may be renatured and purified after temperature shift induction of the promoter. Fusion protein TANG14 comprises rsT4 (AA.-AA.. ) , the angiogenin signal sequence

' ^an~~ angiogenin (AA₂..-AA₃₃₄) (as depicted in Figure 22) . TANG14 can then be tested for its ability to selectively kill gpl60 cells.

Microorganisms and recombinant DNA molecules prepared by the processes of this invention are exem¬ plified by cultures deposited in the In Vitro International, Inc. culture collection, in Linthicum, Maryland, on September 29, 1988, and identified as:

DH5 : E.coli DH5 IVI 10186 GM2929 : E.COli GM2929 IVI 10187 HB101IQ : E.coli HB101I IVI 10188

211.11 : E.COli DH5/p211.11 IVI 10190 199.7 : E.COli DH5/pl99.7 IVI 10191

197.12 : E.coli DH5/pl97.12 IVI 10192 BG391 : E.COli DH5/pBG391 IVI 10193 SA307 : E.coli DH5/pSA307 IVI 10194 SA313 : E.COli DH5/pSA313 IVI 10195 SA400 : E.COli DH5/pSA400 IVI 10196 EX27 : E.coli DH5/pEX27 IVI 10197 EX39 : E.coli DH5/pEX39 IVI 10198.

Such microorganisms and recombinant DNA molecules are exemplified by cultures deposited in the In Vitro International, Inc. culture collection on October 16, 1989 and identified as:

EX45 E.coli DH5/pEX45

EX46 E.coli DH5/pEX46

EX56 E.coli DH5/pEX56

ANG4 E.coli DH5/pANG4

ANG9 E.COli DH5/pANG9

ANG10 E.coli DH5/pANG10

TANG11 E.coli DH5/pTANGll

A89 E.coli A89

While we have hereinbefore described a number of embodiments of this invention, it is apparent that our basic constructions can be altered to provide other embodiments which utilize the processes and compositions of this invention. Therefore, it will be appreciated that the scope of this invention includes all alternative embodiments and variations which are defined in the foregoing specification and by the claims appended hereto; and the invention is not to be limited by the specific embodiments which have been presented herein by way of example.

Claims

CLAIMS We Claim:

1. An immunotoxin comprising a soluble T4 protein conjugated with a toxin.

2. The immunotoxin according to claim 1, wherein the toxin is selected from the group consisting of ricin, abrin, angiogenin, Pseudomonas Exotoxin A, pokeweed antiviral protein, saporin, gelonin and diptheria toxin, and toxic portions thereof.

3. The immunotoxin according to claim 1, wherein the toxin is Pseudomonas Exotoxin A or a toxic portion thereof.

4. The immunotoxin according to claim 1, wherein the toxin comprises the translocation domain and the ADP ribosylation cell inactivating domain of Pseudomonas Exotoxin A.

5. The immunotoxin according to claim 4, wherein the toxin further comprises the boundary region of Pseudomonas Exotoxin A.

6. The immunotoxin according to claim 1, wherein the toxin is angiogenin or a portion thereof.

7. The immunotoxin..according to claim 6, further comprising the translocation domain and the ADP ribosylation cell inactivating domain of Pseudomonas Exotoxin A.

8. The immunotoxin according to claim 1, wherein the soluble T4 protein is selected from the group consisting of a polypeptide of the formula AA_ ₂₃~AA₃₆₂ of Figure 2, a polypeptide of the formula AA.- AA of Figure 2, a polypeptide of the formula Met- AA1-AA.J o.z. of Figure 2, a polypeptide of the formula AA —

₂₃~AA₃₇₄ of Figure 2, a polypeptide of the formula AA - AA₃₇₄ of Figure 2, a polypeptide of the formula Met- AA, -.^ of Figure 2, a polypeptide of the formula AA.- ₃₇₇ of Figure 2, a polypeptide of the formula Met- AA. _₇₇ of Figure 2, a polypeptide of the formula AA_ -AA₃₇₇ of Figure 2, or portions thereof.

9. The immunotoxin according to claim 1, wherein the soluble T4 protein is selected from the group consisting of a polypeptide of the formula AA_ _ -AA.₀. of Figure 2, a polypeptide of the formula Met- AA 1.-AAl.o₀_ of Figure 2, a polypeptide of the formula

AA 1-AA.o_Q2„ of Figure 2, a polypep^'tide of the formula

AA -AA.₀_ of Figure 2, a polypeptide of the formula AA -__j-AA. X.₀Q --i of Figure 2, followed by the amino acids asparagine-leucine-glutamine-histidine-serine-leucine, a polypeptide of the formula AA.- AA₁₈₂ ^of Figure 2, followed by the amino acids asparagine-leucine- gluta ine-histidine-serine-leucine, a polypeptide of the formula Met-AA, _lg2 of Figure 2, followed by the amino acids asparagine-leucme-glutamine-histidine- serine-leucine, a polypeptide of the formula AA _₃- AA of Figure 2, a polypeptide of the formula AA.- AA... of Figure 2, a polypeptide of the formula Met- AA.-AA... of Figure 2, a polypeptide of the formula AA_ ₂₃-AA₁₁₃ of Figure 2, a polypeptide of the formula AA.- AA. of Figure 2, a polypeptide of the formula Met- AA of Figure 2, a polypeptide of the formula

AA -AA . of Figure 2, a polypeptide of the formula AA -AA.₃ of Figure 2, a polypeptide of the formula Met-AA. ... of Figure 2, a polypeptide of formula AA_₂₃~AA ₅ of Figure 2, a polypeptide of the formula AA.-AA..₅ of Figure 2, a polypeptide of the formula Met-AA - A₁₄₅ of Figure 2, a polypeptide of the formula AA_₂₃-AA._gg of Figure 2, a polypeptide of the formula AA.-AA. -₆ of Figure 2, a polypeptide of the formula Met-AA, 1-AAl. _coo _c of Figure 2, or portions thereof.

10. The immunotoxin according to claim 1, wherein the soluble T4 protein is selected from the group consisting of a polypeptide of the formula

of mature T4 protein, a polypeptide of the formula Met-AA.-AA_3g2 of mature T4 protein, a polypep¬ tide of the formula AA.-AA₃₇₄ of mature T4 protein, a polypeptide of the formula

Met-AA -AA_₇₄ of mature T4 protein, a polypeptide of the formula AA.-AA ₇₇ of mature T4 protein, a polypeptide of the formula Met-AA -AA„ of mature T4 protein, a polypeptide of the formula AA_₂₃-AA₃₇₄ of mature T4 protein, a polypeptide of the formula AA__-- AA-₇₇ of mature T4 protein, or portions thereof.

11. The immunotoxin according to claim 1, wherein the soluble T4 protein is selected from the group consisting of a polypeptide of the formula AA_

Δ_._J-AA.Q₀Δ of mature T4 protein, a polypeptide of the formula AA₃-AA._g3of mature T4 protein, a polypeptide of the formula AA.-AA._g2 of mature T4 protein, a polypeptide of the formula Met-AA.-AA._g2 of mature T4 protein, a polypeptide of the formula AA_₂₃-AA._g2 of mature T4 protein, followed by the amino acids asparagine-leucine-glutamine-histidine-serine-leucine, a polypeptide of the formula AA -AA._g2 of mature T4 protein, followed by the amino acids asparagine- leucine-glutamine-histidine-serine-leucine, a polypeptide of the formula Met-AA -AA _a_ of mature T4 protein, followed by the amino acids asparagine-leucine-glutamine-histidine- serine-leucine, a polypeptide of the formula AA _ - AA .. of mature T4 protein, a polypeptide of the formula AA.-AA... of mature T4 protein, a polypeptide of the formula Met-AA.-AA.__ of mature T4 protein, a polypeptide of the formula AA - -AA... of mature T4 protein, a polypeptide of the formula AA.-AA... of mature T4 protein, a polypeptide of the formula Met-AA -AA . of mature T4 protein, a polypeptide of the formula AA -AA ₃ of mature T4 protein, a poly¬ peptide of the formula AA -AA of mature T4 protein, a polypeptide of the formula Met-AA -AA.-. of mature T4 protein, a polypeptide of the formula AA_₂₃~AA _ of mature T4 protein, a polypeptide of the formula AA - AA of mature T4 protein, a polypeptide of the formula Met-AA -AA _g of mature T4 protein, a polypep¬ tide of the formula AA ₃-AA._gg of mature T4 protein, a polypeptide of the formula AA.-AA of mature T4 protein, a polypeptide of the formula Met-AA. 1-AAl. _o._o. of mature T4 protein, or portions thereof.

12. A hybrid DNA sequence coding for an immunotoxin, said DNA sequence comprising a first DNA sequence coding on expression for a soluble T4 protein and a second DNA sequence coding on expression for a toxin.

13. The hybrid DNA sequence according to claim 12, wherein said second DNA sequence codes on expression for a toxin selected from the group con¬ sisting of ricin, abrin, angiogenin, Pseudomonas Exotoxin A, pokeweed antiviral protein, saporin, gelonin and diptheria toxin, and toxic portions thereof.

14. The hybrid DNA sequence according to claim 12, wherein the first DNA sequence is selected from the group consisting of:

(a) the DNA inserts of pl99.7, pBG391 and p211.ll;

(b) DNA sequences which hybridize to one or more of the foregoing DNA inserts and which code on expression for a soluble T4-like polypeptide; and

(c) DNA sequences which code on expression for a soluble T4-like polypeptide coded for on expression by any of the foregoing DNA inserts and sequences.

15. The hybrid DNA sequence according to claim 13, wherein said second DNA sequence codes on expression for Pseudomonas Exotoxin A or a toxic portion thereof.

16. The hybrid DNA sequence according to claim 15, wherein said second DNA sequence codes on expression for the translocation domain of Pseudomonas Exotoxin A.

17. The hybrid DNA sequence according to claim 15, wherein said second DNA sequence codes on expression for the translocation domain and the ADP ribosylation cell inactivating domain of Pseudomonas Exotoxin A.

18. The hybrid DNA sequence according to claim 16, wherein said second DNA sequence codes on expression for the translocation domain of Pseudomonas Exotoxin A, the ADP ribosylation cell inactivating domain of Pseudomonas Exotoxin A and the boundary region of Pseudomonas Exotoxin A.

19. The hybrid DNA sequence according to claim 12, wherein said first DNA sequence codes on expression for a soluble T4 polypeptide of the formula AA₃-AA ₃ of Figure 2 and said second DNA sequence codes on expression for the translocation domain of Pseudomonas Exotoxin A.

20. The hybrid DNA sequence according to claim 12, wherein said first DNA sequence codes on expression for a soluble T4 polypeptide of the formula AA- 3-AA1.8__ of Figure 2 and said second DNA sequence codes on expression for the translocation domain of Pseudomonas Exotoxin A, and the ADP ribosylation cell inactivating domain of Pseudomonas Exotoxin A.

21. The hybrid DNA sequence according to claim 12, wherein said first DNA sequence codes on expression for a soluble T4 polypeptide of the formula AA--AA ₃ of Figure 2 and said second DNA sequence codes on expression for the translocation domain of Pseudomonas Exotoxin A, the ADP ribosylation cell inactivating domain of Pseudomonas Exotoxin A and the boundary region of Pseudomonas Exotoxin A.

22. The hybrid DNA sequence according to claim 12, wherein said first DNA sequence codes on expression for a soluble T4 polypeptide of the formula AA₃-AA ₃ of Figure 2 and said second DNA sequence codes on expression for angiogenin.

23. The hybrid DNA sequence according to claim 21, said second DNA sequence further comprising a third DNA sequence which codes on expression for the translocation domain of Pseudomonas Exotoxin A.

24. A recombinant DNA molecule comprising a hybrid DNA sequence selected from the group consisting of the hybrid DNA sequences of any one of claims 12- 23, wherein said DNA sequence is operatively linked to an expression control sequence in the recombinant DNA molecule.

25. The recombinant DNA molecule according to claim 24, said molecule being selected from the group consisting of p218, pEXll, pEX27, pEX44, pEX45, pEX46, pEX56, pTANGll, pTANG12, pTANG13 and pTANG14.

26. An immunotoxin coded for on expression by a recombinant DNA molecule according to claim 24.

27. A method for producing an immunotoxin comprising the step of culturing a host transformed with a recombinant DNA molecule according to claim 24.

28. A host transformed with a recombinant DNA molecule according to claim 24.

29. The host according to claim 28, said host being selected from the group consisting of strains of E.coli, Pseudomonas. Bacillus. Streptomyces, fungi, such as yeasts, and animal cells, such as CHO and mouse cells, African green monkey cells, such as COS1, COS7, BSC1, BSC40, and BMT10, insect cells, and human cells and plant cells in tissue culture.

30. The immunotoxin according to claim 1, wherein the soluble T4 protein is chemically fused to the toxin.

31. The immunotoxin according to claim 30, wherein the soluble T4 protein is a polypeptide of the formula AA.-AA.,₄ of Figure 3 and the toxin is ricin A.

32. The immunotoxin according to claim 1, wherein the soluble T4 protein is genetically fused to the toxin.

33. The immunotoxin according to claim 26, wherein the soluble T4 protein is a full length soluble T4 protein, and the toxin is Pseudomonas Exotoxin A, or a toxic portion thereof.

34. The immunotoxin according to claim 26, wherein the soluble T4 protein is a soluble T4 polypeptide of the formula AA -AA._a3 of Figure 2 and the toxin is Pseudomonas Exotoxin A, or a toxic portion thereof.

35. A hybrid DNA sequence coding on expression for an immunotoxin, said DNA sequence comprising a first DNA sequence coding for a streptavidin-like polypeptide, a second DNA sequence coding on expression for a soluble T4 protein, and a third DNA sequence coding on expression for a toxin.

36. The hybrid DNA sequence according to claim 35, wherein the first DNA sequence is selected from the group consisting of:

(a) pSA307; (b) DNA sequences encoding polypep¬ tides which hybridize to the foregoing DNA sequence, and wnich code on expression for the polypeptide; and

(c) DNA sequences which code on expression for a polypeptide coded for on expression of any of the foregoing DNA sequences.

37. The hybrid DNA sequence according to claim 35, said DNA sequence comprising a sufficient portion of a signal DNA sequence to cause, upon expression of the DNA sequence, secretion of the immunotoxin coded for by the hybrid DNA sequence across the cell membrane of a host transformed with said hybrid DNA sequence.

38. The hybrid DNA sequence according to claim 35, wherein said third DNA sequence codes on expression for a toxin selected from the group con¬ sisting of ricin, abrin, angiogenin, Pseudomonas Exotoxin A, pokeweed antiviral protein, saporin, gelonin and diptheria toxin, and toxic portions thereof.

39. The hybrid DNA sequence according to claim 35, wherein said second DNA sequence is selected from the group consisting of:

(a) the DNA inserts of pl99.7, pBG391 and p211.11;

(b) DNA sequences which hybridize to one or more of the foregoing DNA inserts and which code on expression for a soluble T4-like polypeptide; and

(c) DNA sequences which code on expression for a soluble T4-like polypeptide coded for on expression by any of the foregoing DNA inserts and sequences.

40. The hybrid DNA sequence according to claim 38, wherein said third DNA sequence codes on expression for Pseudomonas Exotoxin A or a toxic portion thereof.

41. The hybrid DNA sequence according to claim 40, wherein said third DNA sequence codes on expression for the translocation domain and the ADP ribosylation cell inactivating domain of Pseudomonas Exotoxin A.

42. The hybrid DNA sequence according to claim 41, wherein said third DNA sequence codes on expression for the translocation domain, the ADP ribosylation cell inactivating domain, and the boundary region of Pseudomonas Exotoxin A.

43. The hybrid DNA sequence according to claim 38, wherein said third DNA sequence codes on expression for angiogenin, or a toxic portion thereof.

44. The hybrid DNA sequence according to claim 43, wherein said third DNA sequence further comprises a fourth DNA sequence which codes on expression for the translocation domain of Pseudomonas Exotoxin A.

45. A recombinant DNA molecule comprising a hybrid DNA sequence selected from the group consisting of the hybrid DNA sequences of any one of claims 36- 44, wherein the hybrid DNA sequence is operatively linked to an expression control sequence in the recombinant DNA molecule.

46. The recombinant DNA molecule according to claim 45, said molecule being pEX39.

47. An immunotoxin coded for on expression by a recombinant DNA molecule according to claim 45.

48. A host transformed with a recombinant DNA molecule according to claim 45.

49. The host according to claim 46, said host being selected from the group consisting of strains of E.coli, Pseudomonas, Bacillus. Streptomyces, fungi, such as yeasts, and animal cells, such as CHO and mouse cells, African green monkey cells, such as COS1, COS7, BSC1, BSC40, and BMT10, insect cells, and human cells and plant cells in tissue culture.

50. A method for producing an immunotoxin comprising the step of culturing a host transformed with a recombinant DNA molecule according to claim 24 or 45.

51. A pharmaceutically acceptable compo¬ sition comprising an immunotherapeutic or immuno- suppressive amount of an immunotoxin according to any one of claims 1, 30 or 47, and a pharmaceutically acceptable carrier.

52. A method for treating or preventing AIDS, ARC or HIV infection comprising the steps of administering to a patient, in a pharmaceutically acceptable manner, a pharmaceutical composition according to claim 51.