CA2224075C - Vectors for the diagnosis and treatment of solid tumors including melanoma - Google Patents

Vectors for the diagnosis and treatment of solid tumors including melanoma Download PDF

Info

Publication number
CA2224075C
CA2224075C CA2224075A CA2224075A CA2224075C CA 2224075 C CA2224075 C CA 2224075C CA 2224075 A CA2224075 A CA 2224075A CA 2224075 A CA2224075 A CA 2224075A CA 2224075 C CA2224075 C CA 2224075C
Authority
CA
Canada
Prior art keywords
tumor
cells
salmonella
salmonella typhimurium
bacteria
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
CA2224075A
Other languages
French (fr)
Other versions
CA2224075A1 (en
Inventor
John M. Pawelek
David Bermudes
Kenneth B. Low
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yale University
Original Assignee
Yale University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yale University filed Critical Yale University
Publication of CA2224075A1 publication Critical patent/CA2224075A1/en
Application granted granted Critical
Publication of CA2224075C publication Critical patent/CA2224075C/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61KPREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
    • A61K39/00Medicinal preparations containing antigens or antibodies
    • A61K39/02Bacterial antigens
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61KPREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
    • A61K48/00Medicinal preparations containing genetic material which is inserted into cells of the living body to treat genetic diseases; Gene therapy
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61PSPECIFIC THERAPEUTIC ACTIVITY OF CHEMICAL COMPOUNDS OR MEDICINAL PREPARATIONS
    • A61P31/00Antiinfectives, i.e. antibiotics, antiseptics, chemotherapeutics
    • A61P31/04Antibacterial agents
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61PSPECIFIC THERAPEUTIC ACTIVITY OF CHEMICAL COMPOUNDS OR MEDICINAL PREPARATIONS
    • A61P35/00Antineoplastic agents
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61PSPECIFIC THERAPEUTIC ACTIVITY OF CHEMICAL COMPOUNDS OR MEDICINAL PREPARATIONS
    • A61P35/00Antineoplastic agents
    • A61P35/04Antineoplastic agents specific for metastasis
    • CCHEMISTRY; METALLURGY
    • C12BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12NMICROORGANISMS OR ENZYMES; COMPOSITIONS THEREOF; PROPAGATING, PRESERVING, OR MAINTAINING MICROORGANISMS; MUTATION OR GENETIC ENGINEERING; CULTURE MEDIA
    • C12N1/00Microorganisms, e.g. protozoa; Compositions thereof; Processes of propagating, maintaining or preserving microorganisms or compositions thereof; Processes of preparing or isolating a composition containing a microorganism; Culture media therefor
    • CCHEMISTRY; METALLURGY
    • C12BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12NMICROORGANISMS OR ENZYMES; COMPOSITIONS THEREOF; PROPAGATING, PRESERVING, OR MAINTAINING MICROORGANISMS; MUTATION OR GENETIC ENGINEERING; CULTURE MEDIA
    • C12N15/00Mutation or genetic engineering; DNA or RNA concerning genetic engineering, vectors, e.g. plasmids, or their isolation, preparation or purification; Use of hosts therefor
    • C12N15/09Recombinant DNA-technology
    • C12N15/87Introduction of foreign genetic material using processes not otherwise provided for, e.g. co-transformation
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/574Immunoassay; Biospecific binding assay; Materials therefor for cancer
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02ATECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A50/00TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
    • Y02A50/30Against vector-borne diseases, e.g. mosquito-borne, fly-borne, tick-borne or waterborne diseases whose impact is exacerbated by climate change

Abstract

The present invention is directed to the isolation and use of super-infective, tumor-specific vectors that are strains of parasites including, but not limited to, bacteria, fungi and protists. In certain embodiments, the parasites include, but are not limited to, the bacterium Salmonella spp., such as Salmonella typhimurium, the bacterium Mycobacterium avium and the protozoan Leishmania amazonensis. In other embodiments, the present invention is concerned with the isolation of super-infective, tumor-specific, suicide gene-containing strains of parasites for use in treatment of solid tumors.

Description

VECTORS FOR THE DIAGNOSIS AND
TREATMENT OF SOLID TUMORS INCLUDING MELANOMA
1. FIELD OF THE INVENTION
The present invention is concerned with the isolation and use of super-infective, tumor-specific, attenuated strains of parasites including, but not limited to, bacteria, fungi and protists. In certain embodiments the parasites include the bacterium Salmonella spp., such as Salmonella typhimurium, the bacterium Mycobacterium avium, and the protozoan Leishmania amazonensis, for the diagnosis and treatment of sarcomas, carcinomas, and other solid tumor cancers. In other embodiments, the present invention is concerned with the isolation and use of super-infective, tumor-specific, suicide gene-containing strains of parasites.
2. BACKGROUND OF THE INVENTION
A major problem in the chemotherapy of solid tumor cancers is the delivery of therapeutic agents, such as drugs, in sufficient concentrations to eradicate tumor cells while at the same time minimizing damage to normal cells. Thus, studies in many laboratories are directed toward the design of biological delivery systems, such as antibodies, cytokines, and viruses for targeted delivery of drugs, pro-drug converting enzymes, and/or genes into tumor cells.
Houghton and Colt, 1993, New Perspectives in Cancer Diagnosis and Management 1: 65-70; de Palazzo, et al., 1992a, Cell.
Immunol. 142:338-347; de Palazzo et al., 1992b, Cancer Res.

52: 5713-5719; Weiner, et al., 1993a, J. Immunotherapy 13:110-116; Weiner et al., 1993b, J. Immunol. 151:2877-2886;
Adams et al., 1993, Cancer Res. 53:4026-4034; Fanger et al., 1990, FASEB J. 4:2846-2849; Fanger et al., 1991, Immunol.
Today 12:51-54; Segal, et al., 1991, Ann N.Y. Acad. Sci.
636:288-294; Segal et al., 1992, Immunobiology 185:390-402;
Wunderlich et al., 1992; Intl. J. Clin. Lab. Res. 22:17-20;
George et al., 1994, J. Immunol. 152:1802-1811; Huston et al., 1993, Intl. Rev. Immunol. 10:195-217; Stafford et al., 1993, Cancer Res. 53:4026-4034; Haber et al., 1992, Ann. N.Y.
Acad. Sci. 667:365-381; Haber, 1992, Ann. N.Y. Acad. Sci.
667: 365-381; Feloner and Rhodes, 1991, Nature 349:351-352;
Sarver and Rossi, 1993, AIDS Research & Human Retroviruses 9:483-487; Levine and Friedmann, 1993, Am. J. Dis. Child 147:1167-1176; Friedmann, 1993, Mol. Genetic Med. 3:1-32;
Gilboa and Smith, 1994, Trends in Genetics 10:139-144; Saito et al., 1994, Cancer Res. 54:3516-3520; Li et al., 1994, Blood 83:3403-3408; Vieweg et al., 1994, Cancer Res. 54:1760-1765; Lin et al., 1994, Science 265:666-669; Lu et al., 1994, Human Gene Therapy 5:203-208; Gansbacher et al., 1992, Blood 80:2817-2825; Gastl et al., 1992, Cancer Res. 52:6229-6236.
Because of their biospecificity, such systems could in theory deliver therapeutic agents to tumors. However, it has become apparent that numerous barriers exist in the delivery of therapeutic agents to solid tumors that may compromise the effectiveness of antibodies, cytokines, and viruses as delivery systems. Jain, 1994, Scientific American 7:58-65 (Jain). For example, in order for chemotherapeutic agents to eradicate metastatic tumor cells, they must a) travel to the tumors via the vasculature;
b) extravasate from the small blood vessels supplying the tumor;
c) traverse through the tumor matrix to reach those tumor cells distal to the blood supply; and d) interact effectively with the target tumor cells (adherence, invasion, pro-drug activation, etc).

Although antibodies and viruses can express specific recognition sites for tumor cells, they are dependent solely upon the forces of diffusion and convection in order to reach these sites. According to Jain:
An agent that destroys cancers cells in a culture dish should, in theory, be able to kill such cells in the body....
Sadly, however, the existing pharmacopoeia has not markedly reduced the number of deaths caused by the most common solid tumors in adults, among them cancers of the lung, breast, colon, rectum, prostate, and brain.... Before a blood-borne drug can begin to attack malignant cells in a tumor, it must accomplish three critical tasks. It has to make its way into a microscopic blood vessel lying near malignant cells in the tumor, exit from the vessel into the surrounding matrix (the interstitium), and finally, migrate through the matrix to the cells. Unfortunately, tumors often develop in ways that hinder each of these steps.
Jain points out that blood vessels supplying tumors are irregular and convoluted in shape so that blood flow is frequently restricted compared to that in normally vascularized tissue. In addition, there is an unusually high interstitial pressure in many tumors that counteracts the blood flow. Jain further points out that the two chief forces governing the transport of agents to tumor cells via the circulatory system are convection (the transport of molecules by a stream of flowing fluid), and diffusion (the movement of molecules from an area of high concentration to an area of low concentration). Since tumors are often non-uniformly vascularized, many cells in the tumors receive nutrients through the process of diffusion through the matrix. Jain and coworkers obtained data suggesting that "a continuously supplied monoclonal antibody having a molecular weight of 150,000 daltons could take several months to reach a uniform concentration in a tumor that measured one centimeter in radius and had no blood supply in its center."
2.1. BACTERIAL INFECTIONS AND CANCER
Regarding bacteria and cancer, an historical review reveals a number of clinical observations in which cancers were reported to regress in patients with bacterial infections. Nauts et al., 1953, Acta Medica. Scandinavica 145:1-102, (Suppl. 276) state:
The treatment of cancer by injections of bacterial products is based on the fact that for over two hundred years neoplasms have been observed to regress following acute infections, principally streptococcal. If these cases were not too far advanced and the infections were of sufficient severity or duration, the tumors completely disappeared and the patients remained free from recurrence.
Shear, 1950, J. A.M.A. 142:383-390 (Shear), observed that 75%
of the spontaneous remissions in untreated leukemia in the Children's Hospital in Boston occurred following an acute episode of bacterial infection. Shear stated:
Are pathogenic and non-pathogenic organisms one of Nature's controls of microscopic foci of malignant disease, and in making progress in the control of infectious diseases, are we removing one of Nature's controls of cancer?
Subsequent evidence from a number of research laboratories indicated that at least some of the anti-cancer effects are mediated through stimulation of the host immune system, resulting in enhanced immuno-rejection of the cancer cells. For example, release of the lipopolysaccharide (LPS) endotoxin by Gram negative bacteria such as Salmonella triggers release of tumor necrosis factor, TNF, by cells of the host immune system, such as macrophages, Christ et al., 1995, Science 268:80-83. Elevated TNF levels in turn initiate a cascade of cytokine-mediated reactions which culminate in the death of tumor cells. In this regard, Carswell et al., 1975, Proc. Natl. Acad. Sci. USA 72:3666-3669, demonstrated that mice injected with bacillus Calmette-Guerin (BCG) have increased serum levels of TNF and that TNF-positive serum caused necrosis of the sarcoma Meth A and other transplanted tumors in mice. Further, Klimpel et al., 1990, J. Immunol. 145:711-717, showed that fibroblasts infected in vitro with Shigella or Salmonella had increased susceptibility to TNF.
As a result of such observations as described above, immunization of cancer patients with BCG injections is currently utilized in some cancer therapy protocols. See Sosnowski, 1994, Compr. Ther. 20:695-701; Barth and Morton, 1995, Cancer 75 (Suppl. 2):726-734; Friberg, 1993, Med.
Oncol. Tumor. Pharmacother. 10:31-36 for reviews of BCG
therapy.

2.2. PARASITES AND CANCER CELLS
Although the natural biospecificity and evolutionary adaptability of parasites has been recognized for some time and the use of their specialized systems as models for new therapeutic procedures has been suggested, there are few reports of, or proposals for, the actual use of parasites as vectors.
In this regard, Pidherney et al., 1993, Cancer Letters 72:91-98 (Pidherney et al.) and Alizadeh et al., 1994, Infect. Immun. 62:1298-1303 (Alizadeh et al.) have provided evidence that the pathogenic free-living amoeba, Acanthamoeba castellani, has tumorcidal capabilities toward human tumor cells, including melanoma, when added to tumor cells growing in culture or when injected directly into tumors in nude mice. Pidherney et al. conclude:
The feasibility of utilizing the tumorcidal properties of pathogenic/f ree-living amoebae and their cell-free products in the treatment of drug-resistant or radio-resistant tumors warrants further investigation.
However, Pidherney et al. also point out that such pathogenic/free living amoebae can exist either as free-living organisms feeding on bacteria or as opportunistic pathogens producing life-threatening meningoencephalitis or blinding keratitis.
Thus, it is readily apparent that for any parasite to be effective as a therapeutic vector, for example, for human tumors, the benefit of the parasite as a vector must outweigh its risk as a pathogen to the patient. Therefore, although Pidherney et al. and Alizadeh et al. demonstrated cytotoxicity of pathogenic amoebae toward tumor cells, and further suggested their use in the treatment of drug-resistant and radio-resistant tumors, they offered no solution for the inherent pathogenicity of these organisms once injected into cancer patients. Furthermore, they offered no method, e.g., genetic selection for isolating super-infective, tumor-specific strains of pathogenic amoebae nor did they suggest insertion into the amoebael genome of genetic constructs containing inducible genes for the synthesis and secretion of pro-drug converting enzymes and/or suicide gene products.
Likewise, Lee et al., 1992, Proc. Natl. Acad. Sci.
USA 89:1847-1851 (Lee et al.) and Jones et al., 1992, Infect.
Immun. 60:2475-2480 (Jones et al.) isolated mutants of Salmonella typhimurium that were able to invade HEp-2 (human epidermoid carcinoma) cells in vitro in significantly greater numbers than the wild type strain. The "hyperinvasive"
mutants were isolated under conditions of aerobic growth of the bacteria that normally repress the ability of wild type strains to invade HEp-2 animal cells. However, Lee et al.
and Jones et al. did not suggest the use of such mutants as therapeutic vectors, nor did they suggest the isolation of tumor-specific bacteria by selecting for mutants that show infection preference for melanoma or other cancers over normal cells of the body. Without tumor-specificity or other forms of attenuation, such hyperinvasive Salmonella typhimurium as described by Lee et al. and Jones et al. would likely be pan-invasive, causing wide-spread infection in the cancer patient. Further, without selection for tumor specificity or employment of other forms of attenuation, use of such bacteria as therapeutic vectors would increase the risk of pan-infection and septic shock to the cancer patient.
Pan et al., 1995, Nature Medicine 1:471-477 (Pan et al.) described the use of Listeria monocytogenes as a vaccine for the immunization of mice against lethal challenges with tumor cells expressing the same antigen expressed by the Listeria vaccine. In addition, they showed regression of established tumors when immunized after tumor development in an antigen specific T-cell-dependent manner. However, Pan et al. did not show that Listeria monocytogenes could be used as a tumor specific vector, which would target and amplify within the tumor. Rather, Pan et al. showed that recombinant Listeria monocytogenes has the ability to deliver a foreign antigen to the immune system and to involve cell-mediated immunity against the same antigen.
Sizemore et al., 1995, Science 270:299-302 (Sizemore et al.) described the use of attenuated Shigella bacteria as a DNA delivery vehicle for DNA-mediated immunization. Sizemore et al. showed that an attenuated strain of Shigella invaded mammalian cells in culture and delivered DNA plasmids containing foreign genes to the cytoplasm of the cells. Foreign protein was produced in the mammalian cells as a result of the procedure. The Shigella vector was designed to deliver DNA to colonic mucosa, providing a potential oral and mucosal DNA immunization procedure as well as other gene immunotherapy strategies.
However, Sizemore et al. did not suggest the use of such attenuated Shigella as tumor vectors in that they could be used to target tumors and thereby express genes within them.
Rather, Sizemore et al. envisioned its use in vaccination therapy following oral delivery and invasion of the mucosa.
Clostridium was previously investigated as a potential therapeutic vector for solid tumors. The propensity of spores of the obligate anaerobe Clostridium to germinate in necrotic tissues is well known. Tetanus and gas gangrene result from successful colonization of necrotic tissue by pathogenic members of this genus.
Parker et al., 1947, Proc. Soc. Exp. Biol. Med. pp.
461-467 first showed that direct injection of spores of Clostridium histolyticus into a transplantable sarcoma growing in a mouse caused oncolysis, i.e., liquification, as well as regression of the tumor. In general the process of Clostridium-mediated oncolysis was accompanied by acute toxicity and death of the mice. Malmgren and Flanigan, 1955, Cancer Res. 15:473 demonstrated that mice bearing mammary carcinomas, hepatomas, and other tumors died within 48 hrs of intravenous injection of Clostridium tetani spores, whereas control, non-tumor bearing animals were asymptomatic for 40 days. Mose and Mose, 1964, Cancer Res. 24:212-216 (Mdse and Mose) described the colonization and oncolysis of tumors by Clostridium butyricum, strain M-55, a non-pathogenic soil isolate. Mose and Mose established the lack of human pathogenicity of the M-55 strain by administering spores to themselves, as reported by Carey et al., 1967, Eur. J. Cancer 3:37-46. Using Clostridium butyricum strain M-55, Mose and Mose reported that intravenous injections of spores caused oncolysis of the mouse Erlich ascites tumor, growing experimentally as a solid tumor. Aerobic spore-forming organisms--e.g., Bacillus mesentericus, Bacillus subtilis, which were prepared in a similar manner, did not show any oncolysis under the same conditions. Mose and Mose concluded that the clostridial oncolysis was restricted to anaerobic areas of the tumors because of the anaerobic metabolic requirements of the bacteria.

Gericke and Engelbart, 1964, Cancer Res. 24:217-221 showed that intravenously injected spores of strain M-55 produced extensive lysis of a number of different tumors, but with shortened survival times of the Clostridium-treated, tumor-bearing animals compared to non-treated tumor-bearing animals. Further, they found that "metastases in organs or lymph nodes were unaffected by the spores unless the metastatic tumors had reached a considerable size."
Thiele et al., 1964, Cancer Res. 24:222-233 showed that intravenously injected spores of a number of species of nonpathogenic Clostridia, including M-55, localized and germinated in tumor tissue, but not in normal tissues of the mouse. Thiel et al., 1964, Cancer Res. 24:234-238 found that spore treatment produced no effect when administered early in the development of the tumor, i.e., when the tumors were of small size. While the spores caused oncolysis in tumors of sufficient size, there was no effect in smaller tumors or metastases. The animals regularly died during oncolysis.
Carey et al., 1967, Eur. J. Cancer 3:37-46, concluded that small tumors and metastases had been noted to be resistant to oncolysis whereas large neoplasms were particularly favorable. Thus, the qualitative differences in germination of spores were likely to be not a characteristic of neoplastic and normal tissues per se, but related to physiologic and biochemical conditions found within large tumor masses.
Recent molecular genetic studies have focused on anaerobic bacteria of the genus Clostridium as potential tumor vectors. Fox et al., 1996, Gene Therapy 3:173-178 using a Clostridium expression vector were able to transform the E. coli cytosine deaminase gene into Clostridium beijerincki, which resulted in increased cytosine deaminase activity in the growth medium supernatant and cell extracts of transformed clostridial bacteria. Such supernatants, when added to cultures of mouse EMT6 carcinoma made the cells sensitive to 5-fluorocytosine, presumably through its conversion to the toxic 5-fluorouracil. Similarly, Minton et al., 1996, FEMS Microbiol. Rev. 17:357-364 inserted the E.
coli nitroreductase gene into Clostridium beijerincki and were able to detect expression of the gene in an in vivo murine tumor model through the use of antibodies directed against the E. coli nitroreductase gene. The nitroreductase gene product activates CB1954, a potent alkylating agent.
Nothing in any of the above references (or any other references known to the present inventors) suggests the use of any microorganisms, other than the obligate anaerobe Clostridium, as a potential therapeutic vector for solid tumors.

2.3. ATTENUATED SALMONELLA SPP.
Bacon et al., 1950, Br. J. Exp. Path. 31:703-713;
Br. J. Exp. Path. 31:714-724; 1951, Br. J. Exp. Path. 32:85-96 demonstrated that attenuation of Salmonella for virulence in mice can be achieved through auxotrophic mutations, i.e., through the use of mutants which lack the ability to synthesize precursor molecules necessary for growth. More specifically, the authors showed that purine-requiring (Pur-) auxotrophs of Salmonella were attenuated in mice.
Hoiseth and Stocker, 1981, Nature 291: 238-239 showed that Salmonella typhimurium auxotrophic mutants with requirements for aromatic amino acids (Aro-) were attenuated for virulence in C57BL mice. Further, Su et al., 1992, Microbiol. Pathogenesis 13:465-476 showed that one such Aro-mutant, the attenuated antigen carrier strain of Salmonella typhimurium, SL3261, was useful as a vaccine. The Shiga toxin B-subunit/hemolysin A (C-terminus) fusion protein was expressed and underwent extracellular export resulting in antigen-specific immune responses in mice inoculated with these bacteria.
O'Callaghan et al., 1988, Infect. Immun. 56:419-423 characterized Salmonella typhimurium that were both Aro- and Pur- and found that although they were highly attenuated in BALB/c mice, they persisted for several weeks in the livers and spleens following i.v. injections. They were found to be ineffective as vaccines when administered either orally or i.v.
Johnson et al., 1991, Mol. Microbiol. 5:401-407 (Johnson et al.) demonstrated that attenuation in Salmonella virulence can be achieved through mutations in the heat shock inducible protein HtrA, a serine protease. Chabalgoity et al., 1996, Mol. Microbiol. 19:791-801, demonstrated that such attenuated htrA- Salmonella typhimurium were useful as live vaccines.
However, none of the references by Bacon et al., Hoiseth and Stocker, O'Callaghan et al., Johnson et al., Su et al. 1992, Chabalgoity et al. 1996, nor any of the studies referred to in Table 4, infra, suggest that such avirulent strains of Salmonella typhimurium would survive and proliferate within solid tumors, nor that such avirulent mutants might be used as vectors for solid tumor therapy.
2.4. OBJECTIVES OF THE INVENTION
The problems associated with the many physical barriers for delivery of therapeutic agents to solid tumors provide clear and difficult obstacles in the design of effective delivery systems. Thus, there has been a long felt need in the art to provide delivery systems which are able to overcome these obstacles.
It is an object of the present invention to use and to provide more advanced biological vectors such as parasites having several distinct advantages as a novel delivery system, some of which are listed below, as well as to meet the challenges of tumor therapy.
Antibiotic Sensitivity: It is an advantage for a tumor-specific parasitic vector to be sensitive to exogenously administered antibiotics. Parasites, such as bacteria, can be eradicated within their hosts by the administration of antibiotics. Such antibiotic sensitivity allows for the eradication of the parasite from the cancer patient's body upon completion of the therapeutic protocol.
Biospecificity: It is an advantage for a vector to express specificity for its target cell, e.g., a tumor cell.
The more specificity, of the vector for the tumor cell, the lower the inoculum necessary for effective therapy, thereby reducing the risk of septic shock or pan-infection to the cancer patient. Parasites show a great degree of natural biospecificity, having evolved to utilize a variety of specific recognition and invasion mechanisms. (For general discussions on biospecificity see: Falkow, 1991, Cell 65:1099-1102; Tumomanen, 1993, Am. Soc. Microbiol. 59:292-296).
Mutant Isolation and Genetic Manipulation: It is an advantage, in the design and isolation of a parasite as a tumor-specific, therapeutic vector, for the parasite to be amenable to genetic manipulation. Parasites with haploid genomes and short generation times, for example, bacteria such as Salmonella typhimurium and enteroinvasive Escherichia coli, can be readily subjected to mutagenesis followed by enrichment procedures for the isolation of strains with desired new characteristics (see generally, Neidhardt et al., (ed.) 1987, Escherichia coli and Salmonella typhimurium, Cellular and Molecular Biology. American Society for Microbiology, pp 990-1033. Furthermore, the methods for the genetic analysis and stable introduction of genetic constructs into these bacteria are well known to the science of molecular genetics.
Chemotaxis: A chemotactic response toward cancer cells is an advantage for a tumor-specific vector, for example as a stimulus for the vector to invade through a basement membrane matrix such as that produced by endothelial cells in the vasculature, or as a stimulus for the vector to seek out cancer cells surrounded by tumor matrix.
Chemotactic responses in parasites and commensalists or mutualists, particularly in bacteria such as Escherichia coli and Salmonella typhimurium, are well documented. For a review of chemotaxis see Macnab, 1992, Ann. Rev. Genet.
26:131-158.
Replication Within Target Cells: The ability to replicate within target cells is an advantage for a tumor-specific vector. Such an ability allows for amplification of the therapeutic vector number within the infected cancer cell, thus increasing the therapeutic effectiveness of the vector. Progeny of vectors within cancer cells further infect surrounding or distant cancer cells, thus amplifying the vector number within the tumor cell population.
Anaerobic and Aerobic Metabolism: The ability to express invasive and amplification capacities under either aerobic or anaerobic conditions is an advantage for a tumor-specific vector. Solid tumors generally contain vascularized, oxygen-rich areas as well as necrotic oxygen-poor areas. A vector that is functional in both such environments would be able to reach a larger portion of tumor cells than one that can function in only one environment, such as, for example, an obligate anaerobe or aerobe.

3. SUMLILRY OF THE INVENTION
The present invention provides compositions and methods for delivery of genes and/or gene products to and/or into target mammalian cells in vitro or in vivo. The genes and/or gene products are delivered by microorganism vectors, including bacteria, fungal and protozoan parasites, which are selected and/or genetically engineered to be specific to a particular type of target mammalian cell. In a preferred embodiment, the vectors function under both aerobic and anaerobic conditions, are super-infective, tumor-specific microorganisms useful for diagnosis or treatment of sarcomas, carcinomas, lymphomas or other solid tumor cancers, such as germ line tumors and tumors of the central nervous system, including, but not limited to, breast cancer, prostate cancer, cervical cancer, uterine cancer, lung cancer, ovarian cancer, testicular cancer, thyroid cancer, astrocytoma, glioma, pancreatic cancer, stomach cancer, liver cancer, colon cancer, and melanoma.
Vectors useful for the methods of the present invention include but are not limited to Borrelia burgdorferi, Brucella melitensis, Escherichia coli, enteroinvasive Escherichia coli, Legionella pneumophila, Salmonella typhi, Salmonella typhimurium, Shigella spp., Streptococcus spp., Treponema pallidum, Yersinia enterocohtica, Chlamydia trachomatis, Listeria monocytogenies, Mycobacterium avium, Mycobacterium bovis, Mycobacterium tuberculosis, BCG, Mycoplasma hominis, Rickettsiae quintana, Cryptococcus neoformans, Histoplasma capsulatum, Pneumocystis carnii, Eimeria acervulina, Neospora caninum, Plasmodium falciparum, Sarcocystis suihominis, Toxoplasma gondii, Leishmania amazonensis, Leishmania major, Leishmania mexacana, Leptomonas karyophilus, Phytomonas spp., Trypanasoma cruzi, Encephahtozoon cuniculi, Nosema helminthorum, Unikaryon legeri.
As used herein, Salmonella typhimurium encompasses all Salmonella species. It has long been recognized that the various "species" of the genus Salmonella are in fact a single species by all acceptable criteria of bacterial taxonomy. The single species is now designated "Salmonella enterica". F. Neidhardt (ed.), Escherichia coli and Salmonella, 1996, Volume I, pp. xx, ASM Press, Washington DC.
An embodiment of the present invention is to provide methods for the isolation of super-infective, attenuated, tumor-specific mutants of microorganisms such as bacterial, fungal and.protozoan parasites. Further, the present invention provides methods for use of these microorganisms in the diagnosis and treatment of malignant and/or metastatic solid tumor cancers, such as melanoma or colon cancer. Moreover, these mutant parasites may express specific gene products, some of which are secreted into the cytoplasm or vacuolar space of the infected cell.
The present invention provides methods for the isolation of super-infective target cell-specific microorganisms. In particular embodiments, the invention provides for the isolation and use of super-infective, tumor-specific strains of parasites such as the bacterium Salmonella spp., including S. typhimurium, the bacterium Mycobacterium avium, and the protozoan Leishmania amazonensis. The tumor-specific vectors can also contain suicide genes.
One embodiment of the present invention provides methods for the isolation of and compositions comprising super-infective, tumor-specific mutants of Salmonella spp., e.g., Salmonella typhimurium, and for their use in the diagnosis and treatment of sarcomas, carcinomas, melanomas, colon cancer, and other solid tumor cancers. Another embodiment of the present invention provides methods for the isolation of and compositions comprising super-infective, tumor-specific mutants of Salmonella spp. containing a suicide gene. In a specific embodiment, the suicide gene is thymidine kinase from Herpes simplex virus or cytosine deaminase from Escherichia coli or human microsomal p450 oxidoreductase.
Another embodiment of the present invention provides methods for the isolation of and compositions comprising super-infective, tumor-specific mutants of the protozoan, Leishmania amazonensis and for their use in the diagnosis and treatment of sarcomas, carcinomas, melanomas, colon cancer, and other solid tumor cancers.
Yet another embodiment of the present invention provides methods for the isolation of and compositions comprising super-infective, tumor-specific mutants of the bacterium Mycobacterium avium and for their use in the diagnosis and treatment of sarcomas, carcinomas, melanomas, colon cancer, and other solid tumor cancers.
Yet another embodiment of the present invention provides methods for attenuation of parasite vector toxicity so as to reduce the risk of septic shock or other complications in the host, i.e., the patient receiving vector-delivered gene therapy. Such methods include mutagenesis of parasites; isolation of parasite mutants with increased tumor specificity, increased specificity for suicide gene expression and concomitant reduced ability to infect normal host cells in the body; isolation of mutants with enhanced chemotactic abilities toward cancer cell secretory products; isolation of mutants with genetically altered lipopolysaccharide composition; and isolation of mutants with altered virulence genes so as to achieve specific survival of the parasitic vector in cancer cells as opposed to normal cells of the host body.
The present invention further encompasses use of microorganism vectors for diagnosis or treatment of solid tumor cancers.
The present invention may be understood more fully by reference to the following definitions, detailed description of the invention, illustrative examples of specific embodiments and the appended figures in which:
4. DEFINITIONS
Attenuation: Attenuation, in addition to its traditional definition in which a microorganism or vector is modified so that the microorganism or vector is less pathogenic, is intended to include also the modification of a microorganism or vector so that a lower titer of that microorganism or vector can be administered to a patient and still achieve comparable results as if one had administered a higher titer of the parental microorganism or vector. The end result of attenuation is that the risk of toxicity as well as other side-effects is decreased, when the microorganism or vector is administered to the patient.
Suicide gene: A suicide gene is defined as a gene that when delivered to a target cell and expressed by a. vector of the present invention causes the death of the target cell and/or the vector.
Super-infective: A super-infective vector is defined as a vector which is able to attach and/or infect a target cell more readily as compared to the wild type vector. Depending on the population density of the inoculum, the ratio between super-infective vectors and wild type vectors detectably infecting a target cell approaches 4:1, preferably 30:1, more preferably 90:1. Most preferably, one is able to reduce the inoculum size and infection time so that only the super-infective vectors have time to attach to and/or infect cancer cells growing in cell culture in vitro or as tumors in vivo.
Tumor-specific: A tumor-specific vector is defined as a vector which is able to distinguish between a cancerous target cell and the non-cancerous counterpart cell so that the vector preferentially attaches to, infects and/or remains viable in the cancerous target cell.
5. BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1. Figure 1 depicts a DNA cassette system for expressing pro-drug converting enzymes. Each of the components is generated by PCR using primers containing specific restriction endonuclease sites NotI, Nsil, Ncol, Sfii or Pacl that allow for simple interchange of individual components. For example, (A) is the coding sequence for pro-drug converting enzymes such as thymidine kinase, cytosine deaminase or human microsomal p450 oxidoreductase; (B) is a promoter, which is active in an inducible, constitutive or cell specific manner; (C) is a N-terminal secretion signal sequence, such as the 0-lactamase signal sequence; and (D) is a C-terminal secretion signal sequence, such as the enteroinvasive E. coli hemolysin A signal sequence.
Figure 2A-B. Figure 2A-B are photomicrographs of Salmonella typhimurium wild type strain ATCC No. 14028 infecting human melanoma cell line M2. A starting population of ATCC No. 14028 was subjected to 10 cycles of infection into and recovery from M2 melanoma cells before use in the infection assay shown in Figure 2A-B. Figure 2A. Light micrograph of an infected melanoma cell. Figure 2B. DAPI
staining of the cell showing cell nucleus, (n), and numerous bacteria inside the cell, (arrow).
Figure 3A-C. Figure 3A-C are photomicrographs of Salmonella typhimurium wild type strain ATCC No. 14028 during the process of internalization into human melanoma cell line M2. Figure 3A. Phase contrast micrograph of a host cell.
Figure 3B. DAPI staining of the host cell showing the position of the bacteria, (arrow), and the host cell nucleus, W. Figure 3C. Lysosomal glycoprotein LAMP-1 antibody staining of the host cell showing co-localization of the bacteria with lysosomes and/or melanosomes.
Figure 4A-4F relate to producing converting enzyme expression constructs and expression using the same.
Figure 4A-B. Expression of the Herpes simplex thymidine kinase gene containing a 0-lactamase secretory signal sequence in Salmonella typhimurium super-infective clone 72. Figure 4A. Immunoblot analysis of Salmonella typhimurium strains using an anti-TK monoclonal antibody.
Lane 1: bacteria containing only the plasmid vector p279;
Lane 2: strain 14028 wt (CDC6516-60) (MO) containing the cytoplasmicly expressed form of TK (pHETK2); Lane 3: strain 14028 clone 72 containing the cytoplasmicly expressed form of TK (pHETK2); Lane 4: strain 14028 wt (MO) containing the 13-lactamase fusion form of TK (p5-3); Lane 5: strain 14028 clone 72 containing the ,3-lactamase fusion form of TK (p5-3);
Lane 6: strain 14028 wt (MO) containing the ,C3-lactamase fusion form of TK (p2lA-2); Lane 7: strain 14028 clone 72 containing the a-lactamase fusion form of TK (p2lA-2).
Relative molecular mass x 103 is shown on the left. No antibody reactivity is seen in the "vector only" control (Lane 1). In each lane where the cytoplasmically expressed TK is present (Lanes 2 and 3) two major isoforms of the protein are seen; a higher molecular mass isoform containing a leader sequence and a lower molecular mass isoform wherein the leader sequence has been proteolytically cleaved off. In each lane where the bacteria express the TK gene (3-lactamase signal sequence fusion (Lanes 4 to 7) two major isoforms of the protein are also seen: a higher molecular mass form containing the signal sequence and a lower molecular mass isoform wherein the signal sequence has been proteolytically cleaved off, which is the same apparent molecular mass as the processed form of the cytoplasmic enzyme.
Figure 4B. Relative TK enzyme activity associated with each of the samples in Figure A. Enzyme activity is expressed as the total number of counts of 125IdC
phosphorylated in a standard assay, Summers and Summers, 1977, J. Virol. 24:314-318. A small background is present in a bacterial extract from the vector only sample (Lane 1).
Significantly higher levels of TK activity are observed in the wild type and the super-infective clone 72 containing the cytoplasmic form of TK pHETK2, Lanes 2 and 3. Similar levels are observed in both the wild type and super-infective clone 72 containing the (3-lactamase signal sequence fusion isoform of TK p5-3, Lanes 4 and 5. Lower levels are observed in both the wild type and super-infective clone 72 containing the /3-lactamase signal sequence fusion isoform of TK p2lA-2, Lanes 6 and 7.
Figure 4-C is a schematic of the different Herpes Simplex virus thymidine kinase secretion and expression constructs.
Figure 4-D is a schematic of the different human microsomal cytochrome p450 oxidoreductase expression constructs.
Figure 4-E is a schematic of the E. coli cytosine deaminase secretion and expression construct.
Figure 4-F is a graph showing the amount of 5-FC
converted to 5-FU by different bacteria.
Figure 5A-B. Figure 5A-B are photomicrographs of histologic sections from a Cloudman S91 melanoma/macrophage hybrid #48 growing subcutaneously in a DBA/2J mouse. The tumor was excised from a mouse that had been inoculated 2 days earlier with 3 x 105 c.f.u. Salmonella typhimurium super-infective clone #72 carrying the HSV TK gene, clone #725-3-2 Figure 5A. A section stained with hematoxylin and eosin shows tumor cells with a central area of necrosis, denoted by arrows. Figure 5B. A section stained with Brown-Brenn stain (tissue gram stain) shows gram negative bacteria in a necrotic area of the tumor, denoted by the arrow. When viewed under a light microscope, the bacteria stain pink/purple against a yellow background.
Figure 6. Figure 6 is an electron micrograph of a section of a Cloudman S91 melanoma/macrophage hybrid #48 tumor excised from a DBA/2J mouse that had been inoculated i.p. 42 hours earlier with 4 x 106 Salmonella typhimurium super-infective clone 72. Visible in the micrograph are two Salmonella typhimurium bacteria, denoted by arrows, along with numerous melanosomes (m), sub-cellular organelles characteristic of melanoma cells. The co-localization of Salmonella and such melanosomes indicates that the bacteria are present in the cytoplasm of the melanoma cell.
Magnification=21,000x.
Figure 7. Figure 7 is a photomicrograph of a histologic section from a B16F10 melanoma growing subcutaneously in a C57BL/6J mouse. The tumor was excised from a mouse that had been inoculated 42 hours earlier with 1.8 x 105 c.f.u. Salmonella typhimurium super-infective clone #72 carrying the HSV TK gene. The sections were from the same tumor examined with the electron microscope as detailed in Figure 8. The section was stained with Brown-Brenn stain (tissue gram stain) and shows gram negative bacteria in a necrotic area of the tumor, denoted by arrows. When viewed under a light microscope, the bacteria stain pink/purple against a yellow background.
Figure 8. Figure 8 is an electron micrograph of a section from a B16F10 melanoma tumor excised from a C57BL/6J
mouse that had been inoculated i.p. 42 hours earlier with 1.8 x 105 Salmonella typhimurium super-infective clone #72 carrying the HSV TK gene. The section was from the same tumor examined with the light microscope as detailed in Figure 7. The micrograph shows numerous Salmonella typhimurium in extracellular spaces, denoted by arrows, and in an area of necrosis. A single bacterium is also seen within the cytoplasm of a dying melanoma cell. The cytoplasm of the dying melanoma cell also contains numerous black melanosomes (m), characteristic of the B16F10 melanoma.
Magnification = 9,750x.
Figure 9A-D. Figures 9A-D depict growth of Cloudman S91 melanoma/macrophage hybrid #48 tumors in DBA/2J
mice under various treatment conditions. Mice were inoculated s.c. in the flank region with 3 x 105 melanoma cells. The tumors were palpable 8-10 days later, and some of the mice were then further inoculated with Salmonella typhimurium super-infective clone #72 carrying the HSV TK
gene. Twenty-four hours post inoculation with bacteria, some groups of the mice were further inoculated i.p. with 2.0 mg ganciclovir. Ganciclovir inoculations were repeated 6 times over a 5 day period. Points represent caliper measurements of tumors in 2 to 5 mice per treatment group at the days indicated. Measurements in mm were made of length, width, and height for each tumor and volumes were calculated in mm3.
Average tumor volumes for each group of mice were defined as 100% on day 0, the beginning day of treatment. (Figure 9A).
Control mice: no Salmonella; no ganciclovir (Figure 9B).
Ganciclovir only; (Figure 9C). Salmonella only; (Figure 9D). Salmonella + ganciclovir.
Figure 10A-B. Figure 10A shows a control mouse and Figure 10B shows a Salmonella typhimurium-infected (7B) DBA/2J mouse. The mice were inoculated (s.c.) with Cloudman S91 melanoma/macrophage hybrid #48 tumor cells. Upon the appearance of palpable tumors some of the mice were inoculated (i.p.) with 3 x 105 c.f.u. Salmonella typhimurium clone 72 containing the HSV TK gene (clone #725-3-2) , allowed to eat and drink ad libitum for 10 days, and then treated with Sulfatrim' antibiotic in their drinking water for several more days and photographed. Tumors in the depicted mice were representative of the general state of tumor progression in mice in Salmonella-treated and untreated cages.
Figure 11A-11H relate to the effects of gancicylovir on tumor cell growth, in vivo or in vitro.
Figure 11A-B. Figures 11A-B depict a control (Figure 11A) and a Salmonella typhimurium-infected (Figure 11B) DBA/2J mouse. The mice were inoculated (s.c.) with Cloudman S91 melanoma/macrophage hybrid #48 tumor cells.
Upon the appearance of palpable tumors, some of the mice were inoculated with 3 x 105 c.f.u. Salmonella typhimurium clone #72 containing the HSV TK gene (clone #725-3-2) . Control and Salmonella-infected mice were then injected (i.p.) with 2.0 mg ganciclovir a total of 5 times over a 4 day period. The mice were then treated with Sulfatrim' antibiotic in their drinking water for several more days and photographed. The depicted mice are representative of the general state of tumor progression in mice, either in Salmonella-treated and untreated cages.
Figures 11(C-E) show the effect of ganciclovir on the growth of B16F10 melanomas in mice with and without inoculation of Salmonella typhimurium clone YS7211 (Figure 11-1 A); clone YS7213 (Figure 11-1 B); and clone YS7212 (Figure 11-1 C).
Figure 11-F is a graph showing the growth of B16F10 melanoma cells in monolayer culture in the presence or absence of ganciclovir at 10 g/ml or 25 g/ml.
Figure 11-G is a graph showing the effect of ganciclovir on the growth of B16F10 melanomas in mice following inoculation of Salmonella typhimurium clone YS7211 carrying the HSV thymidine kinase gene, YS7211/p5-3, with and without ganciclovir.
Figure 11-H is a graph showing the effect of total amounts of ganciclovir on the growth of B16F10 melanomas in mice following inoculation with Salmonella typhimurium clone YS7211 carrying the HSV thymidine kinase gene, YS7211/p5-3.
Figure 12A-B are electron micrographs. Figure 12A
is an electron micrograph of a section from a HCT 116 human colon tumor excised from a BALB/c nu/nu mouse. The mouse had been inoculated i.p. 72 hours earlier with 2.8 x 105 c.f u.
Salmonella typhimurium super-infective clone #72 containing the HSV TK gene, clone #725-32. Shown in the micrograph are numerous Salmonella typhimurium within the cytoplasm of a neutrophil associated with the tumor. Some of the bacteria are undergoing division as denoted by arrows. The neutrophil or polymorphonucleoleukocyte is characterized by its multi-lobed nucleus (n), (Magnification=21,000x).
Figure 12-B is an electron micrograph showing numerous Salmonella typhimurium, denoted by arrows, in extracellular spaces as well as contained within a single cell, possibly a neutrophil, seen in the upper left. Also seen in the field are two unidentified cells that appear to be dying as indicated by the large intracellular space, along with cellular debris.
Figure 13. Figures 13A-B depict Leishmania amazonensis adhesion to human melanoma cell line M2. Figure 13A. Phase contrast micrograph showing parasites attached to cell, (arrow). Figure 13B. DAPI staining showing the parasite DNA, (arrows), and the host cell nucleus, (n).
Figure 14. Figures 14A-C are photomicrographs of Leishmania amazonensis during the process of internalization into human melanoma cell line M2. Figure 14A. Phase contrast of a Leishmania trypomastigote, (arrow), entering a host cell. Figure 14B. DAPI staining showing the position of the parasite, (arrow), and the host cell nucleus, (n).
Figure 14C. Lysosomal glycoprotein LAMP-1 antibody staining of the host cell showing co-localization of the bacteria and the lysosomes.
Figure 15 A-D. Figure 15 A-C are graphs showing growth of Salmonella typhimurium clone 72 and clone YS7212 in minimal Medium 56 supplemented with glucose only; Medium 56 with glucose plus adenine, vitamin B1, isoleucine, valine, and uracil; or Medium 56 with tumor extract (10%) only.
Figure 15-D is a graph depicting growth of Salmonella typhimurium clones 72 and YS7212 following invasion into human M2 melanoma cells in culture.
Figure 16 A-D are graphs showing growth of B16F10 melanomas in C57B6 mice with and without inoculation of Salmonella typhimurium strains YS721 (Figure 16-A); YS7213 (Figure 16-B); YS7211 (Figure 16-C); and YS7212 (Figure 16-D).
Figure 17 is a graph showing that combination of CD
and 5-fluorocytosine prolong the survival of animals bearing B16F10 lung metastases when the animals are infected with a tumor-specific vector carrying the cytosine deaminase expression construct YS7212/pCD-Secl.
Figure 18 is a graph showing TNF-a production by human macrophages incubated with lipopolysaccharide isolated from wild type and attenuated strains of Salmonella typhimurium.
Figure 19 is a bar graph demonstrating the effect of clones YS7211 and YS7212 expressing the HSV thymidine kinase gene, YS7211/p5-3 and YS7212/p5-3, respectively, on mice bearing metastatic B16F10 tumors with or without ganciclovir treatment.
6. DETAILED DESCRIPTION OF THE INVENTION
The present invention is directed to the isolation of novel therapeutic and diagnostic parasitic vectors for solid tumor cancers, such as sarcomas, carcinomas, lymphomas or other solid tumor cancers, for example, germ line tumors and tumors of the central nervous system, including, but not limited to, breast cancer, prostate cancer, cervical cancer, uterine cancer, lung cancer, ovarian cancer, testicular cancer, thyroid cancer, astrocytoma, glioma, pancreatic cancer, stomach cancer, liver cancer, colon cancer, melanoma and their use. Described, in detail below, are the novel intracellular parasite vectors; methods for the isolation of the novel vectors; genetic engineering of the isolated vectors; and methods for use of the novel vectors as well as other vectors in treatment or detection of solid malignant tumors, including metastatic tumors and tumor cells.

6.1. NOVEL VECTORS AND METHODS FOR THEIR ISOLATION
The isolated vectors, which are for example, bacteria, fungi or protista, are able to differentiate between cancerous cells and non-cancerous counterpart cells.
For example, the isolated vectors are able to differentiate melanoma cells from melanocytes or differentiate colon cancer cells from normal colon epithelial cells. Table I is a representative list, which is in no way meant to limit the present invention, of intracellular parasitic and pathogenic microorganisms which are useful as tumor-specific vectors for the present invention and/or for isolation of novel mutant strains which are super-infective and tumor-specific vectors for use in the present invention.
REPRESENTATIVE LIST OF ORGANISMS USEFUL AS VECTORS
Gram negative bacteria Borrelia burgdorferi Brucella melitensis Escherichia coli enteroinvasive Escherichia coli Legionella pneumophila Salmonella typhi Salmonella typhimurium Shigella spp.
Treponema pallidum Yersinia enterocohtica Gram positive bacteria BCG (Bacillus Calmette-Guerin) Chlamydia trachomatis Listeria monocytogenies Mycobacterium avium Mycobacterium bovis Mycobacterium tuberculosis Mycoplasma hominis Rickettsiae quintana Streptococcus spp.

Fungi Cryptococcus neoformans Histoplasma capsulatum Pneumocystis carnii Apicomplexans Eimeria acervulina Neospora caninum Plasmodium falciparum Sarcocystis suihominis Toxoplasma gondii Kinetoplastida Leishmania amazonensis Leishmania major Leishmania mexacana Leptomonas karyophilus Phytomonas spp.
Trypanasoma cruzi Microsporidians Encephahtozoon cuniculi Nosema helminthorum Unikaryon legeri The bacterium Salmonella typhimurium, the bacterium Mycobacterium avium, and the protozoan Leishmania amazonensis are each particularly useful vectors for the present invention, since each of these organisms shows natural preference for attachment to and penetration into certain solid tumor cancer cells in tissue culture, as opposed to non-cancerous counterpart cells. Since these vectors, such as Salmonella, have a natural ability to distinguish between cancerous cells and their non-cancerous counterpart cells they are directly applicable to the methods for diagnosis or treatment according to the present invention. However, this tumor-specific ability, as well as, the ability to be super-infective as compared to the "wild type" parent strain may be enhanced and selected for by using the methods of the present invention described in Sections 6.1.1-6.1.4., infra.

6.1.1. ISOLATION BY CYCLING THROUGH IN VITRO
TISSUE CULTURE
One embodiment of the present invention is to isolate the novel vectors of the present invention by cycling a microorganism through a pre-selected target cell, preferably a solid tumor cancer cell, with one or more cycles of infection in in vitro tissue culture so that the cycled population and/or clonal isolates therefrom demonstrate enhanced infectivity of the target tumor cell as compared to the starting microbial population and enhanced selectivity as compared to the non-cancerous counterpart cell. The method entails selecting a parasite or pathogen and adding the microorganism to an in vitro tissue culture system of the particular type of solid tumor that one wishes to use as a target cell. For example, if one desires to target melanoma tumors, the target cell may be M2 human melanoma cells.
After incubating the tumor cells and microorganisms together, which allows enough time for the microorganism to attach and/or infect the tumor cell, the tumor cell culture is washed with either buffer or medium which contains an antibiotic agent effective against the specific microorganism used. The antibiotic agent kills any microorganisms that have not attached to and infected the tumor cell. If desired, the infected tumor cell culture may be incubated further in medium containing antibiotic for varying times, depending on the type of population of microorganisms to be isolated. For example, for longer incubation times, the microorganism population isolated has enhanced survival and/or proliferative abilities inside the tumor cells as compared to the starting population of microorganisms.
Additionally, the isolated populations can be cultured to isolate single colony clones using standard techniques.
The infected animal cells are collected and lysed, thus freeing the internalized microorganisms. The microorganism can then be isolated, for example, by centrifugation (2000x g for 4 minutes) and resuspending in fresh medium. The isolated microbial population may then be used for additional cycles of infection into and isolation out of the target tumor cell. The isolated microbial population may be placed first in appropriate growth medium for 1-2 doubling times before being subjected to additional infection cycles to insure their viability. The isolated microbial population may also be cultured so as to isolate and collect single colony clones. The isolated microorganisms may also undergo known in vitro techniques to determine their relative infective and selective abilities as compared to the "wild type" parent strain which did not undergo in vitro selection. For example, in side by side comparisons one may test the relative infectiveness of the isolated microorganism as compared to the "wild type" by using assays designed to quantitate the number of microorganisms which have attached to or invaded the target tumor cell and/or their ability to distinguish between cancerous and non-cancerous cells. In addition, parameters such as microorganism population density, may be varied in these in vitro assays which assists in determining what effect the overall concentration of inoculum of the clone or population being tested has on the ability of the microorganisms to differentiate between the target tumor cells and their non-cancerous counterparts.
For an illustrative example of super-infective, tumor-specific vectors isolated by cycling through in vitro tissue culture, see Section 7, infra.

6.1.2. ISOLATION BY CYCLING THROUGH IN VIVO
SOLID TUMORS
Another embodiment of the present invention is to isolate the novel vectors of the present invention by cycling the microorganism through solid tumors in vivo. This procedure is performed using experimental tumor models in mammals such as, for example and not by way of limitation, B16 mouse melanoma cells which form melanoma tumors in C57B6 mice and HCT116 human colon carcinoma cell which form colon carcinomas in nu/nu and other immuno-compromised mice.
Additionally, fresh biopsies of tumor tissue which are obtained surgically from a cancer patient may be used to inoculate nu/nu, scid or other immuno-compromised mice.
These tumors in mice which have grown from inoculated cancer cells are used as in vivo targets for the isolation of super-infective and tumor-specific vectors in a similar manner as in vitro target cells. Any tumor growing in mice or any other animal may be used in the present invention as a target for the isolation of super-infective and tumor-specific vectors in vivo.
Once the tumor is established in the mouse, by, for example, inoculation of cancer cells sub-cutaneously or transplantation of a tumor mass, the selected microorganism is inoculated into the mouse. After a pre-determined infection time after inoculation in which the microorganism becomes co-localized with the tumor and/or infects the tumor cells, the mice are sacrificed, the tumors excised, weighed and homogenized. An aliquot may be diluted into the proper microorganism growth medium and incubated at the proper growth conditions for 1-2 population doublings to insure the recovery of viable microorganisms for successive inoculations into tumor bearing mice. Further, if the isolated population is to undergo successive inoculations in tumor bearing mice, upon each successive inoculation, the number of microorganisms in the inoculate and the time of infection may be reduced to increase the stringency of selection for tumor-specific isolates. Additionally, the isolated populations can be cultured to isolate single colony clones using standard techniques. The isolated microorganisms may be used also in in vitro assays to determine their relative infective and selective abilities as compared to the "wild type" parent strain which did not undergo an in vivo selection procedure.
For an illustrative example of super-infective, tumor-specific vectors isolated in vivo in tumor-bearing mice, see Section 9, infra.
6.1.3. ISOLATION BY IN VITRO CHEMOTAXIS USING
MEDIUM CONDITIONED BY THE TARGET TUMOR
CELL
Another embodiment of the present invention is to provide methods for isolating super-infective and/or tumor cell-specific vectors by chemotaxis so that the isolated microorganisms have increased chemotactic ability towards tumor cell secretory products. The method entails using capillary tubes which are loaded with either liquid control medium or medium that has been conditioned by the target tumor cell as described by Adler (1973, J. General Microbiology 74:77-91). Conditioned medium is medium in which the target cells have been grown and subsequently has been filtered to remove the cells. One end of the capillary is sealed in a flame; the capillary is then quickly passed several times through a flame and is immediately plunged open end down into a beaker containing either the conditioned or control medium. As the capillaries cool, the liquid is drawn up inside.
The loaded capillary tubes are inserted open end down into a centrifuge tube containing medium and a suspension of the pre-selected microorganism. After a pre-determined period of incubation at 37 C in which the microorganism'chemotacts into the capillary tubes, the capillary tubes are removed with forceps, the sealed ends are opened and the opened capillaries are transferred into centrifuge tubes containing nutrient medium appropriate for the particular microorganism. It is important that the upper tips of the capillary tubes are covered with an appropriate medium for the particular type of microorganism to assure quantitative recovery of the microorganism from the capillary tubes during centrifugation. The capillary tubes are centrifuged, for example, at 4000x g for 4 minutes, to force the microorganism out of the tube. The capillary tubes are removed, the microorganism resuspended, and an aliquot spread onto the appropriate medium in either solid or liquid form to allow for quantitation.
Significant increases in the number of microorganisms entering into the capillary tubes containing conditioned medium as compared to controls indicates a positive chemotactic response toward secreted products of the target cell found in the conditioned medium. The populations isolated by this in vitro technique can undergo successive chemotaxis assay isolation or be used to isolate single colony clones. These clones or populations can be compared to the "wild type" parent strain in their ability to distinguish between the target cancerous cell and the non-cancerous counterpart cell as well as for super-infective ability.
For an illustrative example of super-infective, tumor-specific vectors isolated by in vitro chemotaxis using tumor cell-conditioned medium, see Section 8, infra.

6.1.4. ISOLATION OF MUTAGENIZED VECTORS
In any of the above-described methods for isolating super-infective, tumor-specific microorganisms, the "wild type" parent microorganism can be subjected first to mutagenesis before the microorganism is subjected to any isolation or selection procedure of the present invention.
For example, bacteria are subjected to treatment with nitrosoguanidine and ultraviolet B irradiation so that the hereditary genetic material is modified resulting in the altered expression of genes, both qualitatively and quantitatively, in the microorganism. Other types of chemical and high-energy mutagenesis are well known in the art. For example, alkylating agents such as dimethyl nitrosamine or ethyl methane sulfonate, or intercalating agents, such as ethidium bromide. Other approaches include transposon mutagenesis to introduce genetic flocks or fusions of genes with new promoters. Any mutagen can be used in the present invention to create mutant strains of microorganisms which may then undergo any of the selection methods of the present invention.
For an illustrative example of mutagenesis, see Section 7.1, infra.

6.2. GENETIC MANIPULATION OF THE SELECTED VECTORS
FOR DELIVERY OF GENES AND/OR GENE PRODUCTS TO
THE TARGET SOLID TUMOR CELLS AS WELL AS FOR
ATTENUATION OF VIRULENCE

6.2.1. GENETIC MANIPULATION FOR DELIVERY OF
GENES AND/OR GENE PRODUCTS TO THE
TARGET SITE
After the selection processes described above in which one obtains a super-infective, tumor-specific vector, one can genetically engineer such vectors so that any desired gene or gene product is delivered to a target site, preferably the site of a solid tumor, more preferably, into the tumor cell itself, the necrotic areas of the tumor or into tumor-associated lymphocytes and macrophages.
Additionally, one can genetically alter naturally occurring microorganisms which have a natural ability to infect tumor cells and/or be tumor-cell specific. These vectors are genetically engineered by a wide variety of methods known in the art, for example, transformation or electroporation. In a preferred embodiment of the present invention the vectors are engineered to deliver suicide genes to the target tumor cells. These suicide genes include pro-drug converting enzymes, such as Herpes simplex thymidine kinase (TK) and bacterial cytosine deaminase (CD). TK phosphorylates the non-toxic substrates acyclovir and ganciclovir, rendering them toxic via their incorporation into genomic DNA. CD
converts the non-toxic 5-fluorocytosine (5-FC) into 5-fluorouracil (5-FU), which is toxic via its incorporation into RNA. Additional examples of pro-drug converting enzymes encompassed by the present invention include cytochrome p450 NADPH oxidoreductase which acts upon mitomycin C and porfiromycin (Murray et al., 1994, J. Pharmacol. Exp.
Therapeut. 270:645-649).
Prodrug converting enzymes are being widely employed for use in gene therapy of malignant cancers (Vile and Hart, 1993, Cancer Res. 53:3860-3864; Moolten and Wells, 1990, J. Natl. Cancer Inst. 82:297-300; Wagner, et al., 1981, Proc. Natl. Acad. Sci. USA 78:1441-1445; Mullen, 1994, Cancer Res. 54:1503-1506; Huber et al., 1993, Cancer Res. 53:4619-4625; Waldman et al., 1983, J. Biol. Chem. 258:11571-11575;
Mullen, et al., 1992, Proc. Natl. Acad. Sci. 89:33-37; Austin and Huber, 1993, Mol. Pharmacol. 43:380-387). Table 2 is an illustrative list of pro-drug converting enzymes (Bagshawe, 1995, Drug Dev. Res. 34:220-230).
Prodrug converting enzymes have been expressed in several bacteria. The Herpes simplex virus has been expressed in E. coli (Garapin, 1980, Proc. Natl. Acad. Sci.
USA 78:815-819; Waldman et al., 1983, J. Biol. Chem.
258:11571-11575). Similarly, Simula et al., 1993, Toxicology 82:3-20, expressed the prodrug converting enzyme cytochrome p450 oxidoreductase in Salmonella typhimurium which confered sensitivity to mitomycin.
REPRESENTATIVE PRO-DRUG CONVERTING ENZYMES FOR USE IN VECTOR
THERAPY
Enzyme Pro-drug Reference Carboxypeptidase G2 benzoic acid mustards Bashawe et al., 1988;
Springer et al., 1990 aniline mustards Davies et al., 1994 phenol mustards Springer et al.
Beta-glucuronidase p-hydroxyaniline mustard- Roffer et al., 1991 glucuronide epirubicin-glucuronide Halsma et al., 1992 Mitaku et al., 1994 Penicillin-V-amidase adriamycin-N Kerr et al., 1990 phenoxyaceryl Penicillin-G-amidase N-(4'-hydroxyphenyl Bignami et al., 1992 acetyl)-palytoxin doxorubicin melphalan Vrudhula et al., 1993 fl-lactamase nitrogen mustard- Alexander et at., 1991 cephalosporin fl-phenylenediamine vinblastine derivative-cephalosporin cephalosporin mustard Meyer et at., 1993 Svensson et al, 1993 1B-glucosidase cyanophenylmethyl-$-D- Rowlandson-Busza et al., gluco-pyranosiduronic 1991 acid Nitroreductase 5-(adaridin-1-yl-)2, 4- Knox et al., 1988;
dinitrobenzamide Somani and Wilman, Carboxypeptidase A methotrexate-alanine Haenseler et al., 1992 Bagshawe et at., 1988, Br. J. Cancer 58:700-703.
Springer et al., 1990, J. Med. Chem. 33:677-681.
Davies et al., 1994, Ann. Oncol. 5 (Suppl 5):73(abstr).
Springer et al., 1994, A novel bisiodo-phenol mustard in antibody-directed enzyme pro-drug therapy (ADEPT). In: Programme of Eleventh Hammersmith Conference.
Advances in the Application of Monoclonal Antibodies. London: Hammersmith Hospital (abstr).
Haisma et al., 1992(a), Cancer Immunol. Immunother. 34:343-348.
Roffler et al., 1991, Biochem. Pharmacol. 42:2062-2065.
Haisma et al., 1992(b), Br. J. Cancer 88:474-478.
Mitaku et al., 1994, Ann. Oncol. 5 (Suppl 5):76 (abstr).
Kerr et al., 1990, Cancer Immunol. Immunother. 31:202-206.
Bignami et al., 1992, Cancer Res. 52:5759-5764.
Vrudhula et al., 1993, J. Med. Chem. 38:919-923.
Alexander et al., 1991, Tetrahedron Lett. 32:3296-3272.
Meyer et al., 1993, Cancer Res. 53:3956-3963.
Svensson et al., 1992, Bioconj. Chem. 3:176-181.
Rowlandson-Busza et al., 1991, Cytotoxicity following specific activation of amygladin.
In: Monoclonal Antibodies, Epenetos AA (ed), London: Chapman & Hall, pp. 179-183.
Knox et al., 1988, Biochem. Pharmacol. 41:4661-4669.
Somani et al., 1994, Ann. Oncol. 5 (Suppl 5):73 (abstr).
Haenseler, E., Esswein, A., Vitols, K.S., Montejano, V., Mueller et al., 1992, Biochemistry 31:214-220.

However, pro-drug converting enzymes such as the TK
and CD enzymes, when synthesized in bacteria such as Salmonella or Escherichia coli, are not normally secreted from the bacteria. Accordingly, the expression construct is designed such that the microorganism-produced gene products are secreted by the microorganism. Thus, TK or CD are able to generate phosphorylated acyclovir, ganciclovir, or 5-FU, within the target tumor cell cytoplasm and interstitial spaces of the target tumor. Secretion of TK and CD is achieved by introducing into the expression construct a secretory signal sequence, for example, from the 1i-lactamase gene (Talmadge et al., 1980, Proc. Natl. Acad. Sci. USA
77:3369-3373).
Alternate signal sequences, in addition to f3-lactamase, are also encompassed by the present invention.
Bacteria, for example, are known to have several means for secretion into the periplasm and the outside media. The most typical secretion sequences are N-terminal signal sequences containing hydrophobic transmembrane spanning domains. These sequences serve to guide the protein through the membrane and are removed as- or after the protein crosses the membrane.
Prokaryotic and eukaryotic N-terminal signal sequences are similar and it has been shown that eukaryotic N-terminal signal sequences are capable of functioning as secretion sequences in bacteria. In a preferred embodiment, the gene encoding the enzyme 3-lactamase (penicillinase) is used as the source of the signal sequence. This signal sequence is a well studied example of a bacterial enzyme which is secreted both into the periplasm and into the external media.
Further, some bacterial proteins utilize a different secretion signal which is located at the C-terminus. The enteroinvasive E. coli hemolysin A (hlyA) is the best studied member of this group. It has been shown that the secretion signal is present in the last 60 amino acids of that protein and that transfer of this domain to other proteins can result in their direct secretion into the media when the accessory proteins from the hemolysin operon (hy1C, A, B, & D) are present (Su et al., 1992, Microbial Pathogen. 13:465-476). An illustrative list of secreted proteins reviewed by Pugsley is presented in Table 3, (Pugsley A. P., 1988, Protein secretion across the outer membrane of gram-negative bacteria. In: Protein Transfer and Organelle Biogenesis, R.C. Dand and P. W. Robbins (eds), Academic Press, Inc., Harcourt Brace Jovanovich, Publishers, San Diego, pp 607-652).
SOURCES OF SECRETION SIGNALS FOR PRO-CONVERTING ENZYMES
Protein Organism Location in Transfected E. coli Ref.
and type of signal No.

Chitinase Serratia released into medium N- 14 marcescens terminal signal a-Hemolysin E. coil released into medium C- 9 terminal signal Heat labile various E. coil Similar to cholera toxin 2 12, 7, enterotoxin I strains subunits (A&B); N-terminal 29 signal in both; primarily in periplasm Heat-stable various E. coil N-terminal signal peptide; 11, 28 enterotoxin I strains secreted into the media Heat-labile various E. co/i N-terminal signal peptide 17 enterotoxin II strains Pullulanase Kelbsie/la Release into the medium; N- 3, 27, pneumoniae terminal signal peptide 6 Serine protease S. marcencens Secreted into the medium; N- 31 terminal signal peptide Pectate lyase Erwinia Mainly in the periplasm 15, 5 chrysanthemi Pectate lyase E. carotovara Periplasm 18, 32 Protease E. Secreted into the medium 30, 1 chrysanthemi Aerolysin Aeromonas Periplasm, N-terminal signal 13 hydrophila sequence (processed) Phospholipase C Pseudomonas not secreted by E. coli N- 4, 19, aeruginosa terminal signal sequence 26 Exotoxin A P. aeruginosa not secreted by E. coli 10 Cholera toxin Vibrio cho/erae Mainly periplasmic; 2 subunits 25, 20 Hemolysin V. cho%rae Periplasm 21 DNase V. Cho%rae Periplasm 22, 8 Thermostable V. Periplasm N-terminal signal 24 Hemolysin parahaemolytic peptide us IgA protease Haemolphilis Periplasm 2 influenzae IgA protease Nisseria Secreted into the medium 16 gonorrhoeae Pertussis toxin Bordetella Periplasm; 5 subunits all with 23 pertussis N-terminal signal peptides 1. Barras et al., 1986, FEMS Microbiol. Lett. 34:343-348.
2. Bricker et al., 1983, Proc. Natl. Acad. Sci. USA 80:2681-2685.
3. Chapon et al., 1985, J. Bacteriol. 164:639-645.
4. Coleman et al., 1983, J. Bacteriol. 153:909-915.
5. Collmer et al., 1985, J. Bacteriol. 161:913-920.
6. d'Enfert et al., 1987, EMBO J. 6:3531-3538.
7. DaRas et al., 1980, Nature 288:499501.
8. Focareta et al., 1985, FEMS Mcrobiol. Lett. 29:161-166.
9. Goebel et al., 1984, Structure, function and regulation of the plasmid-encoded hemolysin determinant of E. co/i. In Plasmids in Bacteria, D. R. Hehnski, S.
N.
Cohen, D. B. Cloewell, D. A. Jackson, and A. Hollaender (eds), pp. 791-805, Plenum, New York.
10. Gray et al., 1984, Proc. Natl. Acad. Sci. USA 81:2645-2649.
11. Guzmin-Verduzco et al., 1983, J. Bacteriol. 154:146-151.
12. Hirst et al., 1984, Proc. Natl. Acad. Sci. USA 81:7752-7756.
13. Howard et al., 1986, Mol. Gen. Genet. 204:289-295.
14. Jones et al., 1986, EMBO J. 5:2377-2383.
15. Keen et al., 1984, J. Bacteriol. 159:825-831.
16. Koomey et al., 1982, Proc. NatI. Acad. Sci. USA 79:7881-7885.
17. Lee et al., 1983, Infect. Immun. 42:264-268.
18. Lei et al., 1985, Gene 35:63-70.
19. Lory et al., 1983, Gene 22:95-101.
20. Mekalanos et al., 1983, Nature 306:551-557.
21. Mercuric et al., 1985, Mol. Gen. Genet. 200:472475.
22. Newland et al., 1985, Infect. Immun. 47:691-696.
23. Nicosia et al., 1986, Proc. NatI. Acad. Sci. USA 83:4631-4635.
24. Nishibuchi et al., 1985, J. Bacteriol. 162:558-564.
25. Pearson et al., 1982, Proc. NatI. Acad. Sci. USA 79:2976-2980.

26. Pritchard et al., 1986, J. Bacteriol. 167:291-298.
27. Pugsley, A. P., 1988, Protein secretion across the outer membrane of gram-negative bacteria. In: Protein Transfer and Organelle Biogenesis, R.C. Dand and P.
W. Robbins (eds), Academic Press, Inc., Harcourt Brace Jovanovich, Publishers, San Diego, pp 607-652.
28. So et al., 1980, Proc. Natl. Acad. Sci. USA 77:4011-4015.
29. Spicer et al., 1982, J. Biol. Chem. 257:5716-5721.
30. Wandersman C. (unpublished results cited in Pugsley, 1988).
31. Yanigida et al., 1986, J. Bacteriol. 166:937-944.
32. Zink et al., 1985, Appl. Environ. Microbiol. 49:714-717.

In another embodiment of the present invention, the desired genes expressed from the expression constructs are under the specific regulatory control of certain types of promoters. These promoters may be either constitutive, in which the genes are continually expressed, inducible, in which the genes are expressed only upon the presence of an inducer molecule(s) or cell-type specific control, in which the genes, including but not limited to suicide genes, are expressed only in certain cell types. Further, expression of foreign genes including prodrug converting enzymes frequently alters the phenotype of the bacteria. Therefore, it would be an advantage to drive the expression of a prodrug converting enzyme under exogenous control. This would allow exploitation of the bacterial phenotypes such as tumor targeting and amplification, after which time it would be beneficial to express the prodrug enzyme. Inducible promoters drive gene expression under specific conditions.
Furthermore, exogenously inducible promoters respond to specific stimuli including chemical signals which can be artificially introduced. It would be an advantage to drive the expression of a prodrug enzyme using an exogenously introduced agent which is approved for use in humans. The "SOS" response of bacteria (Friedberg et al., In: DNA Repair and Mutagenesis, pp. 407-455, Am. Soc. Microbiol. Press, 1995) is a response inducible by numerous agents including chemotherapeutic alkylating agents such as mitomycin (Oda et al., 1985, Mutation Research 147:219-229; Nakamura et al., 1987, Mutation Res. 192:239-246; Shimda et al., 1994, Carcinogenesis 15:2523-2529) which is approved for use in humans. Promoter elements which belong to this group include umuC, sulA and others (Shinagawa et al., 1983, Gene 23:167-174; Schnarr et al., 1991, Biochemie 73:423-431). The sulA
promoter includes the ATG of the sulA gene and the following 27 nucleotides as well as 70 nucleotides upstream of the ATG
(Cole, 1983, Mol. Gen. Genet. 189:400-404). Therefore, it is useful both in expressing foreign genes and in creating gene fusions for sequences lacking initiating codons.
In one embodiment, for example, the expression of the gene is controlled by a bacterial promoter which is activated in specific target cells. In a preferred mode of this embodiment, the bacterial promoter is activated primarily in specific target cells. In another embodiment, for example, the expression of the gene is controlled by a bacterial promoter which is activated only in specific tumor cells. An illustrative example of an expression construct which expresses a gene under the control of a promoter with the necessary secretion signal sequence is diagrammed in Figure 1.
In a preferred embodiment of the present invention, the expression of the gene is under the control of a promoter which is active only in the target cell. Microorganism promoters that are specifically or preferentially active in a target tumor cell are isolated by a number of different methods. For example, one method is using IVET (in vivo expression technology) promoter trap procedure for isolating specifically induced genes. This procedure is carried out by taking, for example, a random pool of Salmonella typhimurium DNA insertions generated by Sau3A restriction enzyme and cloning the fragments into the promoter trap vector pIVET
(Slauch et al., 1994, Methods Enzymol. 235:481-492; Mahan et al., 1993, Science 259:686-688). The cloning site is at the position of the promoter for the purA gene which is required for the synthesis of cyclic AMP. This representative pool is transfected back into Salmonella typhimurium and an integration event is induced which results in replacement of the endogenous purA gene. The population of bacteria carrying an integrated IVET plasmid is allowed to infect an animal bearing a solid tumor of the cell type of choice and after 24 hours bacteria are isolated from the tumor. Only those bacteria that received a plasmid whose random piece of Sau3A restricted DNA acts as a promoter within the tumor cells is capable of surviving. In addition to controlling the transcription of purA, the Sau3A restricted DNA promoter also controls the expression of the Q-galactosidase gene.
Two types of promoters are isolated which allow the survival of bacteria within tumors, constitutive and regulated. The constitutive promoter continues to control the positive expression of both genes, inside and outside of the tumor.
Whereas the regulated promoter is no longer active in cells other than the target cell.
Another method for identifying promoters that are active in tumors is to identify tumor-specifically induced microbiological gene products using two dimensional gel electrophoresis. For example, to determine which gene products are specifically or preferentially expressed in melanoma cells rather than macrophages, the method entails three parallel infections which proceed in tandem: (1) 5 x 10' clonal microorganisms are allowed to infect 5 x 106 melanoma cells, (2) 5 x 10' clonal microorganisms are allowed to infect 5 x 106 macrophages, and (3) 5 x 10' clonal microorganisms are maintained in growth phase in LB broth. After a 30 minute infection the cells are washed with DMEM with 10 g/ml gentamicin (for melanoma cells), RPMI 1640 with 10 g/ml gentamicin (for macrophages) and LB without gentamicin (for free microorganisms). After two hours the cells are pretreated with 50 mg/ml cyclohexamide to inhibit host cell protein synthesis for 15 minutes. The cells are then washed and placed in labeling medium (minus methionine) containing 75 Ci/ml 35S-methionine for 30 minutes, followed by 1 hour in normal medium. The cells are then harvested, denatured in 7M
urea buffer and subjected to isoelectrofocusing (IEF) followed by sodium dodecylsulfate (SDS) polyacrylamide gel electrophoresis (PAGE) and analysis by autoradiography. Gene products specifically expressed in melanoma cells appear as protein spots from microorganism-infected melanomas but not from microorganism-infected macrophages or from free microorganisms. The microorganismal genes that are specifically expressed are cloned from a Xgtll expression library using antiserum prepared from proteins derived from preparative IEF and SDS-PAGE gels. Subsequent cloning from a cosmid library results in DNA fragments containing the promoter elements for the tumor-specific expressed gene product.
Yet another method for isolating promoters which are specifically or preferentially activated in the target tumor cells is transposon mutagenesis. Transposon mutagenesis results in a pool of random mutants which can be tested for their ability to survive in epithelial cells but not in target tumor cells. Mutants are first tested for their continued ability to persist in epithelial cells.
Mutants no longer able to survive will be selected against.
Surviving mutants are picked at random and placed in a numbered array using 96 well plates. The target tumor cells are grown in 96 well plates and individually infected with a microorganism at a microorganism to host ratio of about 10:1 for 30 minutes, followed by washing and treatment with 10 g/ml gentamicin. After 24 hours the plates are rinsed and stained with 0.4% trypan blue to determine the ratio of living cells(clear) to dead cells (blue) using a 96 well plate reader. Microorganisms which are unable to survive within the target tumor cell are recovered from the original numbered plate. The genes are then cloned using the transposon as a=genetic marker to isolate the DNA containing the tumor-specific expressed gene and its promoter.
The vectors which can express the various pro-drug or "suicide genes" when given to the host should not confer antibiotic resistance to the host and more importantly the bacteria should remain as sensitive to as many antibiotics as possible. Therefore, these vectors should not carry any antibiotic resistance markers. This can pose a problem in maintaining the expression vectors in the bacteria in absence of selective pressure. However, the are a number of methods in which the vectors can be stably maintained without resorting to antibiotic resistance. For example, one such method is the construction of chromosomally integrated vectors expressing pro-drug converting enzymes or other "suicide genes" as described by Donnenberg, 1991, Am. Soc.
Microbiol., Annual Meeting, Abstract B-111, p.4; Donnenberg and Kaper, 1991, Infect. Immun. 59:4310-4317; and Ried and Collmer, 1987, Gene 57:239-246. Another method is the construction of stable episomal plasmids encoding "suicide genes" or pro-drug converting enzymes using a balanced lethal system. Such balanced lethal systems are defined by the fact that the vector encodes for a function that compensates for a deficiency in the bacteria, such that the presence of the vector is essential for the survival of the bacteria. Such a system is described by Galan et al., 1990, Gene 94:29-35.
This system has the advantage over chromosomal integration in that the plasmids are multicopy and, therefore, achieve higher expression levels.

6.2.2. GENETIC MANIPULATION FOR ATTENUATION OF
VIRULENCE
Many of the microorganisms encompassed by the present invention are causative agents of diseases in humans and animals. For example, sepsis from gram negative bacteria is a serious problem because of the high mortality rate associated with the onset of septic shock (R.C. Bone, 1993, Clinical Microbiol. Revs. 6:57-68). Therefore, to allow the safe use of these vectors in both diagnostics and treatment of humans and animals, the microorganism vectors are attenuated in their virulence for causing disease. In the present invention, attenuation, in addition to its traditional definition in which a microorganism or vector is modified so that the microorganism or vector is less pathogenic, is intended to include also the modification of a microorganism or vector so that a lower titer of that derived microorganism or vector can be administered to a patient and still achieve comparable results as if one had administered a higher titer of the parental microorganism or vector. The end result is to reduce the risk of toxic shock or other side effects due to administration of the vector to the patient.
Such attenuated microorganisms are isolated in a number of techniques. Such methods include use of antibiotic-sensitive strains of microorganisms, mutagenesis of the microorganisms, selection for tumor-specific, super-infective microorganism mutants in culture or in tumor-bearing animals, selection for microorganism mutants that lack virulence factors necessary for survival in normal cells, including macrophages and neutrophils, and construction of new strains of microorganisms with altered cell wall lipopolysaccharides.
For example, in Section 6.1 et seq. where methods are described for the isolation of super-infective, tumor-specific vectors, these same methods are also methods for isolating attenuated vectors; super-infective, tumor cell-specific vectors are by definition attenuated. As the vectors are highly specific and super-infective, the difference between the number of infecting bacteria found at the target tumor cell as compared to the non-cancerous counterparts becomes larger and larger as the dilution of the microorganism culture is increased such that lower titers of microorganism vectors can be used with positive results.
Further, the microorganisms can be attenuated by the deletion or disruption of DNA sequences which encode for virulence factors which insure survival of the microorganisms in the host cell, especially macrophages and neutrophils, by, for example, homologous recombination techniques and chemical or transposon mutagenesis. For example, a number of these virulence factors have been identified in Salmonella. Many, but not all, of these studied virulence factors are associated with survival in macrophages such that these factors are specifically expressed within macrophages due to stress, for example, acidification, or are used to induced specific host cell responses, for example, macropinocytosis, Fields et al., 1986, Proc. Natl. Acad. Sci. USA 83:5189-5193.
Table 4 is an illustrative list of Salmonella virulence factors which, if deleted by homologous recombination techniques or chemical or transposon mutagenesis, result in attenuated Salmonella.
REPRESENTATIVE VIRULENCE FACTORS FOR SALMONELLA
TYPHIMURIUM AND OTHER BACTERIA

Virulence Factor or Loci, Specific Stress Reference Overcome or Stimulated Response Acidification Alpuche-Aranda et al., 5'-adenosine monophosphate Biochenko and Levashev, Cytolysin Libbey et al., 1994 Defensin resistance loci Fields et al., 1989 DNAK Buchmeier and Hefferon, Fimbriae Ernst et al., 1990 GroEL Buchmeier and Hefferon, Induced Macropinocytosis Alpuche Aranda, et al., Ginocchio et al., 1992 Jones et al., 1993 Inv loci Betts and Finlay, 1992 Galon and Curtis III
Ginocchio et al., 1992 Lipoprotein Stone et al., 1992 LPS Gianeiella et al., 1973 Stone et al., 1992 Lysosomal fusion inhibition Ishibashi et al., 1992 Macropage survival loci Fields et al., 1989 Oxidative stress (Sox; in E. cold Nunoshiba et al., 1993 PhoP and PhoQ Behlau and Miller, 1993 Groisman et al., 1993 Miller et al., 1989 Pho activated genes (pag; e.g., pagB and Abshire and Neidhardt, pagC) 1993 Hefferon et at., 1992 Miller et al., 1992 Miller et al., 1989 Pulkkinen and Miller, Stone et al., 1992 PhoP and PhoQ regulated genes (prg) Miller et al., 1989 Behlau and Miller, 1993;

Porins Tufano et al., 1988 Serum resistance peptide Hackett et al., 1987 Virulence factors Abshir and Neidhardt, Loos and Wassenaar, Mahan et al., 1995 Sansonetti, 1992 Virulence plasmid Gulig and Curtiss, 1987 Rhen et al., 1993 Riikonen et al., 1992 spvB (virulence plasmid) Fierer et al., 1993 traT (virulence plasmid) Rhen and Sukupolvi, 1988 ty2 Elsinghorst et al., 1989 Abshiro et al., 1993, J. Bacteriol. 175:3734-3743 Alpuche-Aranda et al., 1992, Proc. Natl. Acad. Sci. USA 89:10079-83 Alpuche-Aranda et al., 1994, J. Exp. Med. 179:601-6088 Baumler et al., 1994, Infect. Immun. 62:1623-1630 Behlau et al., 1993, J. Bacteriol. 175:4475-4484 Belden et al., 1994, Infect. Immun. 62:5095-5101 Betts et al., 1992, Can. J. Microbiol. 38:852-7 Boichenko et al., 1987, Bull. Eksp. Biol. Med. 103:190-2 Boichenko et at., 1988, Zh. Mikrobiol. Epidemiol. Immunobiol. 7:9-11 Boichenko et al., 1985, Zh. Mkrobiol. Epidemiol. Immunobiol. 12:67-9 Bowe et al., 1994, Methods Enzymol. 236:509-26 Buchmeier et al., 1989, Infect. Immun. 57:1-7 Buchmeier et al., 1990, Science 248:730-732 Buchmeier et al., 1995, J. CIdn. Invest. 95:1047-53 Buchmeier et al., 1993, Mol. Microbiol. 7:933-936 Dragunsky et al., 1989, J. Biol. Stand. 17:353-60 Emoto et al., 1993, J. Immunol. 150:3411-3420 Ernst et al., 1990, Infect. Immun. 58:2014-2016 Elsinghorst et al., 1989, Proc. Natl. Acad. Sci. USA 86:5173-5177 Fields et al., 1986, Proc. Natl. Acad. Sci. USA 83:5189-93 Fields et al.,1989, Science 243:1059-62 Fierer et al., 1993, Infect. Immun. 61:5231-5236 Gianella et al., 1973, J. Infect. Dis. 128:69-75 Galan et al., 1989, Microb. Pathog. 6:433-443 Galan et al., 1990, Infect. Immun. 58:1879-1885 Ginocchio et al., 1992, Proc. Natl. Acad. Sci. USA 89:5976-5980 Gulig et al., 1987, Infect. Immun. 55:2891-901 Hackett et al., 1987, J. Infect. Dis. 155:540-549 Heffernan et al., 1992, J. Bacteriol. 174:84-91 Ishibashi et al., 1992, Microb. Pathog. 13:317-323 Libby et al., 1994, Proc. Natl. Acad. Sci. USA 91:489-493 Loos et al., 1994, Immun. Infekt. 22:14-19 Mahan et al., 1995, Proc. NatI. Acad. Sci. USA 92:669-673 Miller et al., 1989, Proc. Natl. Acad. Sci. USA 86:5054-5058 Miller et al., 1992, Infect. Immun. 60:3763-3770 Nunoshiba et al., 1993, Proc. Natl. Acad. Sci. USA 90:9993-9997 Pollack et al., 1986, Nature 322:834-836 Pulkkinen and Miller, 1991, J. Bacteriol. 173:86-93 Rhen et al., 1993, Mol. Microbiol. 10:45-56 Rhen et al., 1988, Microb. Pathog. 5:275-285 Riikonen et al., 1992, Microb. Pathog. 13:281-291 Sansonetti, 1992, Rev. Prat. 42:2263-2267 Stone et al., 1992, J. Bacteriol. 174:3945-3952 Tufano et al., 1988, Eur. J. Epidemiol. 4:110-114 Yet another method for the attenuation of the isolated vectors is to modify substituents of the microorganism which are responsible for the toxicity of that microorganism. For example, lipopolysaccharide (LPS) or endotoxin is primarily responsible for the pathological effects of bacterial sepsis. The component of LPS which results in this response is lipid A (LA). Elimination or mitigation of the toxic effects of LA results in an attenuated bacteria since 1) the risk of septic shock in the patient would be reduced and 2) higher levels of the bacterial vector could be tolerated. Rhodobacter (Rhodopseudomonas) sphaeroides and Rhodobacter capsulatus each possess a monophosphoryl lipid A (MLA) which does not elicit a septic shock response in experimental animals and, further, is an endotoxin antagonist. Loppnow et al., 1990, Infect. Immun. 58:3743-3750; Takayma et al., 1989, Infect.
Immun. 57:1336-1338.
Known similarities in lipid metabolism and genetic organization of lipid metabolic genes between Rhodobacter sphaeroides and other gram negative bacteria and the ability of Rhodobacter genes to complement E. coli mutations (Benning and Somerville, 1992(A), J. Bacteriol. 174:6479-6487;
1992(B), J. Bacteriol. 174:2352-2360; Carty et al., 1994, FEMS Microbiol. Lett. 118(3):227-231) demonstrate that, for example, Salmonella and other bacteria can be genetically altered to produce MLA, thereby reducing its potential of inducing septic shock. A preferred embodiment of the present invention is a Salmonella spp. strain that expresses MLA
rather than LA and also expresses HSV TK under the control of a tumor-specific promoter.
As an illustrative example, the generation of MLA
producing Escherichia coli or Salmonella typhimurium entails constructing a DNA gene library composed of 10 kB fragments from Rhodobacter sphaeroides which is generated in Xgtll or pUC19 plasmids and transfected into E. coli. Clones which produce MLA are positively selected by using an antibody screening methodology to detect MLA, such as ELISA. In another example one generates a cosmid library composed of 40 kB DNA fragments from Rhodobacter sphaeroides in pSuperCos which is then transfected into Salmonella typhimurium.
Clones which produce MLA are positively selected by using an antibody screening methodology to detect MLA, such as ELISA.
Yet another example for altering the LPS of Salmonella involves the introduction of mutations in the LPS
biosynthetic pathway. Several enzymatic steps in LPS
biosynthesis and the genetic loci controlling them in both E.coli and Salmonella typhimurium have been identified (Raetz, 1993, J. Bacteriol. 175:5745-5753 and references therein). Several mutant strains of Salmonella typhimurium and E. coli have been isolated with genetic and enzymatic lesions in the LPS pathway. One such mutant, firA is a mutation within the gene that encodes the enzyme UDP-3-O(R-30 hydroxymyristoyl)-glycocyamine N-acyltransferase, that regulates the third step in endotoxin biosynthesis (Kelley et al., 1993, J. Biol. Chem. 268:19866-19874). Salmonella typhimurium and E.coli strains bearing this type of mutation produce a lipid A that differs from wild type lipid A in that it contains a seventh fatty acid, a hexadecanoic acid (Roy and Coleman, 1994, J. Bacteriol. 176:1639-1646). Roy and Coleman demonstrated that in addition to blocking the third step in endotoxin biosynthesis, the firA- mutation also decreases enzymatic activity of lipid A 4' kinase that regulates the sixth step of lipid A biosynthesis.
Once the strain has been attenuated by any of the methods known in the art, the stability of the attenuated phenotype is important such that the strains do not revert to a more virulent phenotype during the course of treatment of a patient. Such stability can be obtained, for example, by providing that the virulence gene is disrupted by deletion or other non-reverting mutations on the chromosomal level rather than epistatically or that the "suicide gene" is stably integrated into the bacterial chromosome.
Another method of insuring the attenuated phenotype is to engineer the bacteria such that it is attenuated in more than one manner, e.g., a mutation in the pathway for lipid A production, such as the firA- mutation (Hirvas et al., 1991, EMBO J. 10:1017-1023) and one or more mutations to auxotrophy for one or more nutrients or metabolites, such as uracil biosynthesis, purine biosynthesis, and arginine biosynthesis as described by Bochner, 1980, J. Bacteriol.
143:926-933. In a more prefered embodiment of the invention, the bacterial vector which selectively targets tumors and expresses a pro-drug converting enzyme is auxotrophic for uracil, aromatic amino acids, isoleucine and valine and synthesizes an altered lipid A.
6.3. IN VITRO CANCER DIAGNOSTICS AND IN VIVO
TREATMENT OF SOLID TUMORS USING ISOLATED
VECTORS AND OTHER VECTORS

6.3.1. IN VITRO DIAGNOSTICS
An embodiment of the present invention is to provide methods for use of the vectors of the present invention in in vitro diagnostic assays and diagnostic kits for the detection of solid tumor cancers, including but not limited to melanoma. Also, the kits may comprise tumor-specific non-attenuated vectors. The in vitro diagnostic assays and kits are based on the enhanced specificity towards a cancerous cell rather than its non-cancerous counterpart cell of a vector. For example, and not by way of limitation, a putative solid tumor is biopsied from a patient. The tumor biopsy is minced and digested to a suspension of single cells. Aliquots of the suspension and a non-cancerous counterpart or control cell are cultured and infected with a tumor-specific vector according to the present invention.
After an incubation period, the number of tumor-specific microorganisms which attached to and/or infected the biopsied cells as compared to the non-cancerous counterpart or control cells is determined by any method known to those skilled in the art. A higher number of vectors found associated with the target cell as compared to the non-cancerous counterpart or control cells indicates that the target cell is cancerous, for example, about 5-10 times as many vectors will infect a tumor cell compared to a non-cancerous control cell. A non-cancerous counterpart or control cell is the normal cell from which the tumor cell is derived, for example, for melanoma cells the non-cancerous counterpart or control cell is melanocyte cells, for colon cancer the counterpart cell is colon epithelial cells. In one embodiment the ratio is determined as the number of vectors/target cell. In another embodiment, after infection, the cells are fixed and treated with a stain or antibody which recognizes DNA so that the vector DNA present in the target cell cytoplasm is visualized. The presence of DNA in the target cell cytoplasm indicates that the biopsied target cells are cancerous. In one embodiment of the present invention the diagnostic method comprises exposing a sample of cells which are suspected of being cancer cells to a tumor-specific vector or microorganism. The method also comprises exposing a sample of non-cancerous counterpart cells to the tumor-specific vector or microorganism as a comparative control. After incubating for a time period in which the microorganisms can attach to and/or infect cancer cells, the infectivity of the microorganism or vector for the cells suspected of being cancerous and the non-cancerous counterpart control cells can be compared.
The diagnostic kits of the present invention comprise an effective amount of a tumor-specific vectors.
The kits can further comprise an appropriate amount of non-cancerous control cells. The vector and/or cells may be supplied either frozen, lyophilized or growing on solid or in liquid medium. The diagnostic kits can further comprise inert ingredients and other kit components such as vials, packaging components and the like, which are well known to those skilled in the art.
In certain embodiments, the vectors useful for the methods of diagnosis of the present invention can further comprise tumor-specific, attenuated or non-attenuated vectors. In other embodiments, the kits of the present invention can comprise tumor-specific, attenuated or non-attenuated vectors.
For illustrative examples of in vitro diagnostics of solid tumor cancers, including but not limited to melanoma, see Sections 22, 25 and 26.

6.3.2. IN VIVO TREATMENT OF SOLID TUMORS
The vectors for use in in vivo cancer treatment are a subset of the vectors of the present invention. The vectors for in vivo treatment have been attenuated such that, when administered to a host, the vector has been made less toxic to the host and easier to eradicate from the host's system. In a preferred embodiment, the vectors are super-infective, attenuated and specific for a target tumor cell.
In a more preferred embodiment, the vectors are also sensitive to a broad range of antibiotics.
In addition, the isolated vectors can encode "suicide genes", such as pro-drug converting enzymes or other genes, which are expressed and secreted by the vector in or near the target tumor. The gene can be under the control of either constitutive, inducible or cell-type specific promoters. In a preferred embodiment, a suicide gene is expressed and secreted only when a vector has invaded the cytoplasm of the target tumor cell, thereby limiting the effects due to expression of the suicide gene to the target site of the tumor.
In a preferred embodiment, the vector, administered to the host, expresses the HSV TK gene. Upon concurrent expression of the TK gene and administration of ganciclovir to the host, the ganciclovir is phosphorylated in the periplasm of the microorganism which is freely permeable to nucleotide triphosphates. The phosphorylated ganciclovir, a toxic false DNA precursor, readily passes out of the periplasm of the microorganism and into the cytoplasm and nucleus of the host cell where it incorporates into host cell DNA, thereby causing the death of the host cell.
Another embodiment of the present invention is to provide methods of treatment of solid tumor cancers with isolated attenuated vectors of the present invention. For example, a patient is diagnosed with a solid tumor cancer by any method known in the art, including the in vitro diagnostic methods of the present invention. The vector used in the treatment may already be isolated using the methods of the present invention with target cell lines or using model tumors in mice of the target tissue. In another embodiment, the biopsied tumor cells are used in the selection assay for isolating a vector which is super-infective and tumor-specific for the tumor of the patient. In a preferred embodiment the vector is genetically modified, for example, to lack virulence factors, express a suicide gene or both as described in Section 6.2.2. In addition, the isolated vector is analyzed for sensitivity to antibiotics to insure the eradication of the vector from the patient's body after successful treatment or if the patient experiences complications due to the administration of the isolated vector.
When administered to a patient, e.g., an animal for veterinary use or to a human for clinical use, the vectors can be used alone or may be combined with any physiological carrier such as water, an aqueous solution, normal saline, or other physiologically acceptable excipient. In general, the dosage would range from about 1 to 1x108 c.f.u./kg, preferably about 1 to 1x102/kg.
The vectors of the present invention can be administered by a number of routes, including but not limited to: orally, topically, injection including, but limited to intravenously, intraperitoneally, subcutaneously, intramuscularly, intratumorally, i.e., direct injection into the tumor, etc.
The following series of examples are presented by way of illustration and not by way of limitation on the scope of the invention.

7. EXAMPLE: ISOLATION OF SUPER-INFECTIVE, TUMOR-SPECIFIC SALMONELLA TYPHIMURIUM IN VITRO

7.1. MUTAGENESIS BEFORE ISOLATION OF SUPER-INFECTIVE, TUMOR-SPECIFIC CLONES
A culture of Salmonella typhimurium strain #14028 was grown exponentially at 37 C in minimal medium 56 plus glycerol (0.5%) to OD600= 0.3, then chilled on ice. An aliquot was removed so that the culture could be titered for colony forming units (c.f.u.) on LB agar plates. The culture was washed and resuspended in Na citrate (0.1M, pH 5.5), incubated with fresh nitrosoguanidine (NG, 50 g/ml, 20 minutes, 37 C), washed once by centrifugation, resuspended in medium 56, chilled, and again an aliquot was removed so that the culture could be titered for c.f.u. on LB agar plates.
Another aliquot of the NG treated bacteria was diluted (1:5) into LB broth and grown to stationary phase for storage frozen at -80 C in 12% glycerol.
The remaining bacteria were irradiated with ultraviolet light, dose = 50J/m2, X= 254nm). An aliquot was removed and the cells were then titered for c.f.u. on LB agar plates, with another aliquot diluted 1:4 into LB broth, grown to stationary phase, and stored frozen at -80 C in 12%
glycerol.
The mutagenesis procedures produced an increase in the number of mutations in the strain by four criteria: 1) decreased survival of the bacteria following mutagenesis (nitrosoguanidine=6-fold; ultraviolet B irradiation=400-fold); 2) increased frequency of auxotrophic (nutritional requiring) mutants to (2%); 3) increased frequency of maltose-mutants to (2%); 4) increased frequency of galactose- mutants to (0.5%).

7.2. ISOLATION OF SUPER-INFECTIVE SALMONELLA
TYPHIMURIUM CLONES #70 AND #71 SPECIFIC FOR
CANCER CELLS
A population of Salmonella typhimurium wild type strain #14028 was mutagenized as described in Section 7.1 with nitrosoguanidine and UV irradiation. Briefly, the bacteria were grown exponentially at 37 C in minimal medium 56 plus glycerol to OD600=0 3, chilled on ice, washed, resuspended in Na citrate with 50 g/ml nitrosoguanidine and incubated for 20 minutes at 37 C. The bacteria were washed once by centrifugation and resuspended in medium 56. The bacteria were then irradiated with W light at a dose of 50J/m2, X=254nm.
Prior to infection by Salmonella, human M2 melanoma cells were inoculated into Coming Tissue Culture flasks (25cm2) at approximately 2x105 cells/flask in 4 ml DMEM
cell culture medium containing penicillin (100 units/ml), and streptomycin (100 g/ml), and incubated overnight in a 37 C, gassed (5%C02), humidified incubator. The next day the cells were rinsed twice with prewarmed Dulbecco's Minimal Essential medium supplemented with 10% fetal bovine serum (DMEM/FBS) and no antibiotics.
The mutated population of Salmonella typhimurium was cultured on LB agar overnight at 37 C or in a liquid culture. The following day the bacteria were transferred with a platinum wire loop to LB broth or to DMEM/FBS, adjusted in concentration to OD600= 0.1 (approximately 2 x 108 c.f.u./ml), and subjected to further growth at 37 C on a rotator. Following growth to the desired population density (monitored at an optical density of 600nm) the bacteria were diluted to a concentration of 106 c.f.u./ml in DMEM/FBS, and incubated at 37 C an additional 20 minutes.
The mutagenized bacterial population was subjected to a single cycle of infection into- and isolation from human M2 melanoma cells in culture. Portions of the mutagenized population were grown clonally on agar and 20 clones of Salmonella typhimurium were separately isolated and tested for their individual infectivity toward human M2 melanoma cells. The bacteria were added to animal cell cultures in 25 cm2 Coming Tissue Culture flasks at 4ml/flask, and incubated with the animal cells in a gassed (5oC02/95% air), humidified incubator at 37 C. After a 15 minute incubation with the animal cells the bacteria-containing medium was poured off and the cultures were rinsed gently with warmed DMEM/FBS
(4 ml) containing gentamicin sulfate (20 g/ml), an antibiotic that kills extracellular but not intracellular bacteria. The gentamicin sulfate-containing medium was poured off, fresh DMEM/FBS/gentamicin sulfate medium was added, and the cells were incubated for 60 minutes at 37 C. Following the 60 minute incubation with gentamicin sulfate, the medium was poured off, the flasks were rinsed lx with DMEM/FBS (without gentamicin sulfate), and 1 mM EDTA or an EDTA/trypsin solution (Sigma Chemicals, lx) in Ca"/Mg" free physiological saline (4 ml) was added. After incubating with EDTA or EDTA/trypsin for 20 minutes at 37 C, the flasks were shaken to suspend the animal cells, and aliquots were removed for quantitation. Animal cells were quantitated in a Coulter Counter (Coulter Electronics, Inc.) and bacteria were quantitated by plating aliquots on LB agar, incubating at 37 C, and counting colonies. Quantitation was expressed as the number of infecting (gentamicin resistant) bacteria/106 animal cells.
Two clones, "70" and "71", were found to be super-infective of melanoma cells, with infection capacities 5-10-fold greater than the mutagenized wild type strain (data not shown). Clones 70 and 71 were also assessed for their relative specificity of the following human cells in culture:
M2 melanoma cells and normal human melanocytes; "CaCo" colon cancer cells and normal human colon epithelium #1790 as depicted in Table 4(A).

TABLE 4(A) SPECIFIC INVASION OF S. TYPHIMURIUM INTO MELANOMA VS
MELANOCYTES AND COLON CANCER VS COLON EPITHELIUM IN CELL
to CULTURE: CLONES "70" AND "71 "+

Human Infecting Salmonella/106 human cells:
Cell Line Clone 70 (ratio) Clone 71 (ratio) normal 1.4 0.2x106 1.2 0.3x106 melanocytes M2 melanoma 7.3 2.0x106 (5.2) 5.7 0.7x106 (4.8) colon 1.5 0.1x106 0.8 0.2x106 epithelium (#1790) colon 7.2 2.0x106 (4.8) 2.3 0.3x106 (2.9) carcinoma (CaCo) 'cancer cell:normal counterpart cell +Results represent averages SD for triplicate infections.

The bacterial clones #70 and #71 showed strong invasion preference for melanoma and colon cancer cells over that for normal melanocytes and normal colon epithelial cells.
7.3. ISOLATION OF SALMONELLA TYPHIMURII7M SUPER-INFECTIVE CLONE #72 BY CYCLING IN IN VITRO
CELL CULTURE
Salmonella wild type strain #14028 was mutagenized with nitrosoguanidine and ultraviolet B irradiation as described in Section 7.1. A starting population of 5 x 108 mutagenized bacteria was grown to OD600=.450, diluted in DMEM/FBS to a concentration of 5 x 10' c.f.u./ml, and allowed to infect human M2 melanoma cells for 15 minutes. Infecting bacteria were isolated from the melanoma cells, and again allowed to infect fresh, uninfected populations of melanoma cells. The 2nd round of infecting bacteria were again isolated and subjected to further cycles of infection into, and isolation from, human M2 melanoma cells. After the completion of 4 such cycles, the population of melanoma-cycled bacteria which is designated 14028pop-1 was then plated on agar and 100 individual clones were picked and tested for their ability, compared to wild type bacteria, to infect M2 melanoma cells. The results of the selection process on 14028p p'1 and selected population sub-clones are detailed in Table 5.
Additionally, an aliquot of 14028pop-1 was subjected to two further cyclings in M2 melanoma cells. This 6X-cycled population was then subjected to 7 cycles of negative selection against normal human melanocytes. The 6X-cycled population was added to a culture of normal human melanocytes and incubated for 15 minutes. The supernatant was collected and was then added back to a fresh culture of normal human melanocytes. This negative selection procedure was carried out 7 times. This 6X-7X cycled population was again added to M2 melanoma cells, allowed to infect the melanoma cells for 15 minutes, and the bacteria were then collected from the M2 cells. This 6X-7X-1X cycled population was designated 14028 pop' 2 .
The mixed population of 4 times cycled Salmonella typhimurium, designated 14028p0p'1, showed a 3-fold increased infectivity of melanoma cells over that of the starting mutagenized population of wild type bacteria. Of the 100 clones isolated from population 14028POp'1 of Salmonella, two clones, #6 and #72, were found to be significantly super-infective of melanoma cells. The remaining bacterial clones showed infectivity that was similar to or below that of the wild type strain. In the experiment presented in Table 5, clone #6 was about 25-fold, and clone #72 was about 55-fold more infective than the mutagenized wild type strain during a 15 minute infection period. Escherichia coli, strain K-12, #CSH 101, was at least two orders of magnitude less infective than wild type Salmonella typhimurium, thus, demonstrating the natural ability of S. typhimurium to infect certain animal cells.

SALMONELLA TYPHIMURIUM POPULATIONS IN CULTURE+
Salmonella Strain Infecting Bacteria/106 melanoma cells (%wild type) Wild type S. typhimurium 3.8 3.0 x 104 100 #14028 (mutagenized) #14028P0P-' 1.1 0.4 x 105 290 Clone #6 8.6 1.0 x 106 2260 Clone #72 2.1 0.2 x 106 5500 E. coli K-12 < 102 < 1 +Results represent averages SD for triplicate infections.

Over several such experiments shown in Table 5, the infectivity of clone #72 toward melanoma cells varied from 5-to 90-fold over that of the wild type strain. This variation seemed to depend on the bacterial growth density prior to infection of melanoma cells. Therefore, the effect of population density on relative infectivity between wild type and clone #72 was determined.
Wild type S. typhimurium and super-infective clone #72 were grown as a lawn on LB agar plates. Portions of the cultures were removed with a platinum loop and inoculated into LB broth at a concentration of approximately 2 x 108 c.f.u./ml (OD600=0.1). The cultures were then placed on a rotator at 37 C and optical densities were monitored as a function of population density. At the optical densities indicated, aliquots of bacteria were removed, diluted in melanoma growth medium (DMEM/10% FBS) to a density of 1 x 106 c.f.u./ml, and allowed to infect human M2 melanoma cells.
Infectivity assays were carried out as described. The results are shown in Table 6.

INFECTIVITY OF WILD TYPE S. TYPHIMURIUM AND SUPER-INFECTIVE
CLONE #72 TOWARDS HUMAN MELANOMA CELLS: EFFECT OF BACTERIAL
POPULATION DENSITY+

Optical Density: Salmonella/106 melanoma Infectivity ratio:
(600nm) Clone #72 Wild type (Clone 72:wild type) 0.200 -0- -0- (no infectivity) 0.300 9.0 x 103 -0- (infinite) 0.400 5.0 x 104 -0- (infinite) 0.500 4.5 x 105 5.0 x 103 90:1 0.600 1.2 x 106 3.7 x 104 32:1 0.700 2.3 x 106 2.5 x 105 9:1 0.800 3.2 x 106 5.6 x 105 6:1 0.900 3.6 x 106 7.3 x 105 5:1 +Results represent averages of duplicate experiments. Variations between duplicates were < 15%.

The results demonstrate that infectivity of both bacterial strains was highly dependent on bacterial population density prior to infection, however, clone #72 was proportionately more infective than the wild type strain at low population densities.
The results shown in Tables 5 and 6 were also confirmed by phase and light microscopy which revealed super-infectivity of a 10x melanoma-cycled population of Salmonella typhimurium designated "M10" as shown in Figures 2 and 3. It was also found that wild type strain 14028 and clone 72 infected human M2 melanoma cells equally well when grown under anaerobic conditions prior to infection. However, when the strains are grown under aerobic conditions, strain 14028 was strongly suppressed in infectivity, whereas clone 72 remained induced. Thus, clone 72 was infective under either anaerobic or aerobic growth conditions and superinfective compared to wild type under aerobic growth conditions.

7.4. PREFERENTIAL SELECTIVITY OF S. TYPHIMURIUM FOR
CANCER CELLS: WILD TYPE STRAIN VS. SUPER-INFECTIVE CLONE #72 Super-infective Salmonella typhimurium clone #72 isolated in Section 7.3 was compared to the non-mutagenized wild type strain #14028 for relative infectivity of human M2 melanoma cells, normal human melanocytes, colon cancer cells and normal colon epithelium. The results are shown in Table TUMOR SPECIFICITY OF WILD TYPE S. TYPHIMUR/UM AND SUPER-INFECTIVE CLONE #72 TOWARD VARIOUS NORMAL AND CANCEROUS
CELLS IN CULTURE+
Animal Infecting Salmonella/
106 animal cells Cell Line wild type (ratio)* Clone #72 (ratio)*
normal 1.2 0.7x104 2.7 0.4x105 melanocyte (foreskin, human) M2 human 2.5 0.6x104 (2.1) 1.7 1.1x106 (6.3) melanoma normal colon 6.6 0.8x103 5.2 3.0x105 epithelium (1790, human) colon cancer 3.0 2.0x104 (4.6) 9.5 3.0x 105 (1.8) (HTB 39, human) "normal" 1.8 1.5x104 5.5 1.4x105 fibroblast (3T3, mouse) transformed 2.4 0.6x104 (1.3) 4.6 0.8x106 (8.4) macrophage (J774, mouse) +Results represent averages SD for triplicate infections.
'cancer cell:normal counterpart cell Each of the two bacterial strains showed invasion preference for human cancer cells over normal cells. Clone #72 was super-infective in all cases when compared to the wild type strain. Further, clone #72 showed a significantly higher degree of invasion specificity for human melanoma cells over normal melanocytes than the wild type strain did.
7.5. INFECTIVITY OF SALMONELLA TYPHIMURIUM WILD

TOWARD VARIOUS HUMAN CARCINOMAS IN CULTURE
In another series of experiments, the relative infectivities of clone 72 and wild type strain 14028, toward a variety of human carcinomas growing in culture, was determined. The experimental protocol used is described in Section 7.2. Results are presented in Table 7(A).

TABLE 7(A) INFECTIVITY OF SALMONELLA TYPHIMURIUM WILD TYPE AND

CULTURE
Sa/mone//a/106 Human Cells Ratio Origin of Wildtype: Clone # 72:140 Cell Line Primary Tumor 14028 72 28 M2 melanoma 4.0+3.8 x 104 4.2+3.5 x 105 11:1 HTB57 lung 2.8 1.3 x 103 4.5+2.1 x 104 16:1 HTB183 lung 1.0 0.3 x 105 4.1 +1.8 x 105 4:1 HTB54 lung 2.1 0.7 x 104 1.7 0.2 x 105 8:1 A549 lung 3.7 5.6 x 104 4.5 4.9 x 105 12:1 CRL1740 prostate 2.3 0.4 x 105 1.8 0.2 x 106 8:1 CRL1611 kidney 3.2 1.2 x 105 1.8 0.3 x 105 6:1 HTB52 liver 1.8 + 0.3 x 105 2.6 + 0.8 x 105 1.4:1 MCF7 breast 7.3 + 2.6 x 104 3.6 + 0.9 x 105 5:1 Results represent averages SD for n = 3-9 separate infections.

Both the wild type strain 14028 and clone #72 were able to infect each of the human cancer cells tested in culture. In all cases, clone 72 was superinfective compared to the wild type strain.
However, human lung line HTB57 was significantly less receptive to Salmonella typhimurium infectivity when compared to other cancer cell lines tested. In yet another series of experiments, the human lung line HTB57 was implanted into mice. In 10 of 10 nu/nu mice implanted with 1 x 107 HTB57 cells, no tumor "takes" were observed, even after several months. Whether or not these cells were receptive to Salmonella infection when grown as tumors was not determined.
7.6. DISCUSSION
In summary, the results demonstrate the following:
a) infectivity of S. typhimurium is dependent upon population density of the bacteria and b) super-infective clone #72 differs from the wild type strain in its increased infectivity of melanoma cells at all bacterial population densities and especially at low population densities under aerobic growth conditions. The ability to infect at low bacterial population densities is an advantage in the use of clone #72 as a tumor-specific vector, since it would allow for a lower c.f.u. of bacteria inoculated into the cancer patient, thus reducing the risk of septic shock in the patient. Additionally, the results demonstrate methods for the isolation of super-infective, tumor-specific mutants of S. typhimurium. Such mutants are represented by clones #6, #70, #71 and #72 that were isolated via enrichment procedures for melanoma infectivity by the bacteria. The results further demonstrate that wild type S. typhimurium exhibits specificity for human cancer cells over normal human cells in culture. Further, although clone 72 was originally selected for superinfectivity toward human melanoma cell line M2, it was additionally found to be superinfective toward human colon cancer cells and transformed mouse macrophages, when compared to the wild type strain 14028 (see Table 7). The expression of super-infectivity and tumor-specificity of isolated mutant clones represent attenuation of the bacteria and present distinct advantages for the use of such Salmonella clones as tumor-specific vectors in the diagnosis and therapy of human cancer.
8. EXAMPLE: SELECTION FOR SALMONELLA TYPHIMURIUM
MUTANTS WITH CHEMOTACTIC ABILITIES TOWARD MELANOMA
SECRETORY PRODUCTS IN VITRO
The melanoma cells were an artificially-produced hybrid line isolated from a polyethylene glycol induced fusion between Cloudman S91 mouse melanoma cells and peritoneal macrophages from a DBA/2J mouse. The hybrid cell line used herein was termed Cloudman S91 melanoma/macrophage hybrid #48. The hybrid cell line formed rapidly growing metastasizing tumors in DBA/2J mice, Pawelek et al., 1995, J.
Invest. Dermatol. 104:605. 5 x 106 Cloudman S91 melanoma/macrophage hybrid #48 cells were cultured at 30 C in a gassed, humidified incubator in 75cm2 culture flasks in DMEM/FBS culture medium containing 10% fetal bovine serum and no antibiotics. Control flasks containing DMEM/FBS but no melanoma cells were incubated in parallel. After 72 hours, the media were removed, aseptically filtered through 0.45 filters, and stored at 4 C.
Salmonella typhimurium super-infective clone #72 described above was subjected to mutagenesis with nitrosoguanidine and UV. The mutagenesis procedures produced an increase in the number of mutations in Clone #72 similar to that shown earlier when the wild type strain #14028 was mutagenized. This mutagenized population of clone 72 (1172'" t") was further used to select for mutants with enhanced chemotactic abilities toward melanoma cell secretory products, i.e., melanoma-conditioned culture media.
Procedures for loading capillary tubes with potential chemotactic attractants were modified from Adler (Adler, 1973, J. General Microbiology 74:77-91). Control and melanoma-conditioned culture media, described above, were loaded into 2X capillary tubes ("Microcaps", Drummond Scientific Co.) as described by Adler. The capillaries were handled with forceps. One end was sealed in a flame; the capillary was then quickly passed several times through the flame and immediately plunged open end down into a 10 ml beaker containing 1 ml control or melanoma conditioned culture medium. As the capillary cooled (about 10 minutes), liquid was drawn in about 1 cm.
Salmonella typhimurium, growing at 37 C in LB were collected by centrifugation and resuspended in control DMEM/FBS culture medium containing a concentration of 108 c.f.u./ml. Aliquots (200 Al, 2 x 107 c.f.u.) were pipetted into 1.5 ml microfuge tubes. Loaded capillary tubes (described above) were then inserted open end down into the Beckman microfuge tubes containing the Salmonella typhimurium, and the assay was begun by incubating at 37 C.
After 30 to 60 minutes, the capillary tubes were removed with forceps, the sealed ends were broken off with wire cutters, and the capillaries were transferred to 15 ml conical centrifuge tubes containing 3 ml LB broth. It was important that the upper tips of capillary tubes were covered with LB
broth in order to assure quantitative recovery of the bacteria via the centrifugation step described as follows.
The capillaries within the centrifuge tubes were then centrifuged (1000x g for 4 minutes) to force the bacteria out of the capillaries. The bacteria were resuspended by vortexing, and aliquots were spread onto LB agar plates for quantitation. Significant increases in the number of bacteria entering the capillaries containing melanoma-conditioned media compared to control-conditioned media indicated a chemotactic response of the bacteria to melanoma-secreted products.
Aliquots of mutagenized super-infective Salmonella typhimurium, 1172"ut", described above were placed on a rotor at 37 C, grown to an optical density of 0.4-0.6 at a wavelength of 600nm, and subjected to the chemotaxis procedures described above. The chemotaxis cycling procedure was repeated 4 times through successive challenges with melanoma-conditioned culture medium. The population obtained after 4 cycles was designated #72p P-2. After the 4th cycling, aliquots of the mixed populations of bacteria were frozen in glycerol. Additional aliquots of the mixed population of Salmonella typhimurium obtained from the fourth cycling were then compared to an uncycled mixed population of mutagenized clone 72 (1172mut1l) for relative chemotactic abilities toward control and melanoma-conditioned culture medium. The results are shown in Table 8.

EVIDENCE FOR POSITIVE CHEMOTACTIC RESPONSES OF S. TYPHIMURIUM
TO CONDITIONED GROWTH MEDIUM OF CULTURED MELANOMA CELLS+
Salmonella Bacteria/Capillary Tube:

Strain Control Medium Conditioned Medium Ratio:
#72mut 1.2x103 0.2 4.4x103 2.7 3.7:1 (mutagenized, no cycling) #72P0 2 0.5x 103 0.1 1.8x103 0.4 3.6:1 (mutagenized, cycled 4x) +Results represent average S.D for quadruplicate capillary tubes.

Both populations of bacteria tested showed positive chemotactic responses to melanoma-conditioned culture medium over control-conditioned medium, displaying an approximate 4:1 preference for the melanoma-conditioned medium. Although the chemotactic response of population #72p,p-2 was not statistically significant as compared to the chemotactic response of population #72't for melanoma conditioned medium, the chemotactic response of population #72p,p-2 was significantly reduced as compared to the chemotactic response of population #72't for control medium. Thus, the propensity of population #72POP2 to enter capillary tubes containing control medium was significantly reduced. These results suggest that population #72p,p-2 is less efficient in motility generally, however, upon exposure to melanoma-conditioned medium, population #72p,p-2 showed a chemotactic response equivalent to the control population.
Whatever the mechanisms for the different chemotactic phenotypes expressed by the #72" and #72 pop-2 populations of bacteria in Table 8, the results demonstrate that the phenotypes can be altered via the selection procedure of exposing bacteria to successive challenges of melanoma-conditioned media. It is likely that the mixed populations of mutagenized, chemotactically cycled bacteria isolated in these experiments contain a number of diverse mutants expressing likewise diverse phenotypes for the chemotactic response to melanoma cell-conditioned medium.
9. EXAMPLE: ISOLATION OF TUMOR-SPECIFIC MUTANTS OF
SALMONELLA TYPHIMURIUM BY CYCLING IN VIVO IN TUMOR-BEARING MICE
Tumor cells inoculated into DBA/2J mice from Cloudman S91 melanoma/macrophage hybrid cell line #48 were used as the target tumor for the selection of attenuated, tumor-specific Salmonella typhimurium. Super-infective Salmonella typhimurium clone #72 was mutagenized with nitrosoguanidine and WB as described in Section 7.1 producing a mutagenized population derived from clone #72.
The mutagenesis procedures produced an increase in the number of mutations in clone #72 similar to that shown earlier when the wild type strain #14028 was mutagenized. Cloudman melanoma/macrophage hybrid #48 cells were inoculated (s.c.) into DBA/2J mice at a concentration of 106 cells in 0.1 ml saline/inoculated site and a total of 4 sites/mouse in the shoulder and flank regions. After 8-10 days, palpable tumors developed, and the mice were inoculated (i.p.) with the mutagenized Salmonella population derived from super-infective clone #72. After 2 hours of infection, the mice were sacrificed, the tumors removed, weighed, and homogenized in a teflon homogenizer in 5 vol (vol/wt) LB broth. An aliquot of the homogenate was then diluted about 1:4 in LB
broth, placed on a rotator at 37 C, and incubated through 1-2 population doublings, should be monitored at OD600, in order to ensure the recovery of viable bacteria for successive inoculations into tumor-bearing mice. The procedure was repeated through 4 cycles of infection into mice, followed by recovery from tumors. At the beginning of each cycle, the number of bacteria inoculated and the time of infection was reduced from the previous cycle in order to increase the stringency of selection for tumor-specific mutants. The resultant population recovered after 4 cycles was designated #72pop1. The results of this procedure are detailed in Table 9 below.

SELECTION FOR MELANOMA-SPECIFIC SALMONELLA TYPHIMURIUM IN
TUMOR-BEARING MICE

Infection Total # Bacteria Infection Total # Bacteria Cycle Inoculated/mouse Time Recovered in Tumors*
1 1x1010 120 min 2.1 x 107 2 1x109 80 min 1.6x 106 3 6x108 60 min 1.7 x 106 4 2x108 40 min 1.4 x 105 'Infecting Salmonella were pooled from 4-8 separate tumors for each cycle These results demonstrate that infecting bacteria can be recovered from tumors in vivo. These results also demonstrate that in vivo cycling results in an enriched population since fewer-bacteria were isolated than were inoculated.

10. EXAMPLE: PROLIFERATION OF SALMONELLA TYPHIMURIUM
WITHIN MELANOMA CELLS
Proliferation of a gene-delivering vector within target tissue can both amplify the gene within the target tissue as well as allow one to reduce the titer of inoculated vector, thus reducing the risk of septic shock in the host.
The following examples demonstrate that Salmonella typhimurium proliferates in melanoma cells.
10.1. PROLIFERATION WITHIN CULTURED HUMAN M2 MELANOMA CELLS
It was found that Salmonella typhimurium proliferated within human M2 melanoma cells in culture with doubling times of about 30 to 60 minutes as illustrated below. Wild type Salmonella strain #14028 and super-infective clone #72 were separately introduced into the culture media of human M2 melanoma cells 2 x 105 melanoma cells/25cm2 tissue culture flask at 106 bacterial c.f.u./ml culture medium. After 1 hour, gentamicin (20 g/ml) was added to kill external, but not internalized bacteria, and melanoma cells were harvested and assayed for the number of internalized bacteria at the time points indicated. The results are presented in Table 10.

PROLIFERATION OF SALMONELLA TYPHIMURIUM WILD TYPE STRAIN
#14028 AND CLONE #72 WITHIN CULTURED HUMAN M2 MELANOMA
CELLS+

Salmonella Strain Time (h) Salmonella/106 Fold Increase Melanoma Cells #14028 wild type 1 6.8 x 105 ---2 1.8x 106 2.6x 4 1.8 x 107 26x 6 5.4 x 107 79x Clone #72 1 5.8 x 106 ---2 8.0x 106 1.4x 4 3.2 x 107 5.5x 6 1.4 x 108 24x +The numbers represent averages for duplicate and triplicate determinations, with the variation between replicates < 25%.
10.2. PROLIFERATION WITHIN MELANOMA TUMORS
GROWN IN MICE
DBA/2J mice were inoculated s.c. in four areas (left and right shoulders and flanks) with 106 Cloudman S91 melanoma/macrophage hybrid #48 cells. After the appearance of palpable tumors (8-10 days) the mice were further inoculated (i.p.) with 2 x 108 Salmonella typhimurium. The Salmonella strains tested were wild type #14028 and super-infective clone #72. At 4 hours and 21 hours post-inoculation with bacteria, mice were bled orbitally, and then euthanized by anesthesia with metofane. Tumors and livers were removed aseptically, rinsed with sterile NaCl (0.9%), weighed, and homogenized in LB broth at a ratio of 5:1 (vol:tumor wt). Bacteria were quantitated by plating the homogenates onto LB plates, incubating overnight at 37 C, and counting bacterial colonies. Numbers represent averages S.D. The results for the 4 hour and 21 hour incubations of the bacteria in mice are detailed in Tables 11(A) and 11(B).

A. DISTRIBUTION OF SALMONELLA TYPHIMURIUM 4 HOURS
FOLLOWING INOCULATION (I.P.) INTO CLOUDMAN S91 Salmonella Salmonella/ Salmonella/ Salmonella/ Tumor/
Strain ml Blood /gm tumor (wet wt) gm liver (wet Liver wt wild type 6x105 8.9 2.5x104(n =4) 3.6x105 1:4 clone 72 2x105 3.5 3.3x104(n =4) 2.4x105 1:7 B. DISTRIBUTION OF SALMONELLA TYPHIMURIUM 21 HOURS
FOLLOWING INOCULATION (I.P.) INTO CLOUDMAN S91 Salmonella Salmonella/ Salmonella/ Salmonella/ Tumor/
Strain ml Blood gm tumor (wet wt) gm liver (wet Liver wt wild type 1.0x 104 1.3 0.8x 109(n = 4) 4.4x106 300:1 clone 72 6.7x 103 2.1 2.7x 109(n = 4) 5.2x 105 4000:1 At 4 hours post-inoculation of Salmonella, there were fewer bacteria in the tumors than in the blood stream and the liver for both wild type clone 14028 and clone 72.
However, by 21 hours, Salmonella were found in great abundance in the tumors so that the ratio of bacteria/g tissue in tumors was 4,000:1 over that in the liver for super-infective mutant clone 72. After 21 hours post-inoculation of bacteria, the number of Salmonella in the tumors was similar for both the wild type Salmonella strain and clone 72, and was far greater than the total number of Salmonella originally inoculated, indicating that both wild-type and clone 72 strains of bacteria proliferated within the tumors. Thus, the ability of Salmonella typhimurium to infect melanoma cells and proliferate within them was expressed both in cell culture as seen in Table 10 and in tumors growing in mice as seen in Tables 11A and 11B.
The wild-type strain 14028 showed higher infectivity in liver than did clone 72. The higher infectivity of liver by the wild-type Salmonella was consistent with the observed greater lethality of the wild type stain toward DBA/2J mice than that produced by clone 72 at high bacterial inocula (>l09 c.f.u./mouse, data not shown) Similar results were observed with C57BL\6J mice bearing B16F10 melanomas as seen in Table 18, Section 15.2.
Together, these results demonstrate that selection for strains of bacteria or other parasites with enhanced tumor specificity in vitro yields mutant strains with attenuated host toxicity in vivo.

10.3. DISTRIBUTION OF SALMONELLA TYPHIMURIUM IN
TUMOR-BEARING MICE
The following experiments demonstrate that Salmonella can localize to and proliferate within a tumor of an animal bearing either multiply-implanted subcutaneous melanoma tumors or naturally occurring metastases.
10.3.1. DISTRIBUTION OF SALMONELLA FOLLOWING

MELANOMA TUMORS
DBA/2J mice were inoculated s.c. in four areas (left and right shoulders and flanks) with 106 Cloudman S91 melanoma/macrophage hybrid site. Palpable tumors appeared 8-days post-inoculation, representative animals were selected, and 2 of the 4 tumors (right shoulder and left flank) were directly inoculated with Salmonella typhimurium super-infective clone #72 at c.f.u.'s of 7 x 104 or 7 x 106 10 bacteria/tumor. At 21 hours post-inoculation, mice were euthanized with metofane. Tumors and livers were removed aseptically, rinsed with sterile NaCl (0.911), weighed, and homogenized with NaCl at a ratio of 5:1 (vol:tumor wt).
Bacteria were quantitated by plating the homogenates onto LB
plates, incubating overnight at 37 C, and counting bacterial colonies. The results are shown in Table 12.

DIRECT INOCULATIONS INTO TUMORS

Inoculum/ Salmonella/ Salmonella/ Tumor/
tumor a tumor (wet wt) g liver (wet wt) Liver 7.2 x 104 Tumor 1' 1 x 109 3.0 x 105 3,300:1 Tumor 2 3 X 107 100:1 Tumor 3 1 x 108 330:1 7.2x106 Tumor 1 * 4 x 109 5.0 x 106 800:1 Tumor 2 3 x 109 600:1 Tumor 3 3 x 107 6:1 inoculated tumor In summary, two days post-inoculation of super-infective Salmonella typhimurium clone #72 directly into specificized tumors, the Salmonella could be found in distal, non-inoculated tumors. The amounts of Salmonella found in the tumors far exceeded the amounts of Salmonella inoculated into the mice, proving that the Salmonella proliferated within the tumors. The results demonstrate that Salmonella typhimurium can proliferate within a tumor, exit that tumor via the circulatory system, travel to a distant tumor, and proliferate within that distant tumor.

10.3.2. DISTRIBUTION OF SALMONELLA INTO CLOUDMAN

This experiment shows that the bacteria should be able to target naturally-occurring metastases of solid tumors.
1 x 105 Cloudman S91 melanoma cells were inoculated s.c. in the tail of a DBA/2J mouse. After approximately four weeks, a soft tissue metastasis (-0.5 g) developed with no visible evidence of a primary tumor in the tail.
S. typhimurium clone 72 at 2 x 105 c.f.u. was inoculated i.p.
The mouse was sacrificed 48 hours post- inoculation, and the liver and tumor were removed, homogenized in Luria broth, and quantitated for S. typhimurium by serial dilutions on LB agar plates.
The results shown in Table 12(A) demonstrate that Salmonella typhimurium clone 72 can target and proliferate within a metastatic tumor.

30.
TABLE 12(A) DBA/SJ MOUSE WITH A SOFT-TISSUE MELANOMA METASTASIS' Tissue Salmonella/g tissue Tumor/Liver Liver 3.1 x 106 ---Tumor 3.2 x 109 1000:1 'Results represent determinations from a single animal.

11. EXAMPLE: ANTIBIOTIC SENSITIVITY OF WILD TYPE

11.1. SENSITIVITY TESTED IN VITRO
Wild type Salmonella typhimurium strain 14028 and super-infective mutant clone 72 were tested for antibiotic susceptibility and were each found to be sensitive to 12 different antibiotics currently used in treating bacterial infections. The bacteria were tested according to the standard protocol to determine antibiotic sensitivity as seen in clinical laboratories, so that a patient is not given an antibiotic to which the microorganism is resistant. The bacteria were tested for antibiotic susceptibility by subjecting them to the Disc Diffusion Susceptibility Technique Kit (Remel Corp., Lenexa, Kansas). The data are presented in Table 13.
ANTIBIOTIC SENSITIVITY

Wild Type: Clone 72:
Ampicillin S S
Cefoperazone S S
Ceftazidime S S
Cefuroxime S S
Gentamicin S S
Mezlocillin S S
Cefazolin S S
Ciprofloxacin S S

Unasyn S S
Ceftriaxone S S
TMP/SMX S S

11.2. SENSITIVITY TESTED IN VIVO
Susceptibility of Salmonella typhimurium clone 72 to antibiotics was further tested by injecting mice i.p. with bacteria, treating half of the mice with the antibiotic enrofloxacin, and observing the effects of enrofloxacin, an active analog of ciprofloxacin, on the survival of the mice.
Six week old C57B6 female mice were inoculated i.p.
with 105 cfu Salmonella typhimurium clone 72. After four days following inoculation with bacteria, three of the mice were further inoculated i.p. with 100 g/0.1 ml BAYTRIL'"' (enrofloxacin), and their drinking water was supplemented with 25 g/ml BAYTRILT". After 4 days, all the mice were given fresh drinking water without BAYTRIL'. After a total of 21 days, all surviving mice were euthanized and the experiment was terminated. The results are shown below in Table 13(A).
TABLE 13(A) CLONE 72 BAYTRIL (enrofloxacin) IN DRINKING H2O

Conditions Avg. time of Death S.D.
no antibiotic 9.3 5 days enrofloxacin > 21 days (days 4-10 post-inoculum) Mice receiving bacteria only and no antibiotic died after an average of 9 days following inoculation. Mice receiving bacteria followed by antibiotic treatment survived at least 21 days and showed no symptoms of Salmonella toxicity when the experiment was terminated. Thus, the results clearly demonstrate that mice can be rescued from Salmonella-mediated death by treatment with the antibiotic enrofloxacin. These results are consistent with those presented in Table 13 demonstrating antibiotic sensitivity of Salmonella typhimurium strains 14028 and clone 72 by the Disc Diffusion Susceptibility Technique.
The results further underscore the advantage of using antibiotic-sensitive bacteria as vectors in human tumor therapy, since the bacteria can be eliminated by introduction of antibiotics when desired.

12. EXAMPLE: ENHANCED EXPRESSION OF BACTERIAL PROMOTERS
IN MELANOMA CELLS
In a preferred embodiment of the present invention an isolated super-infective vector, such as Salmonella typhimurium clone 725-3-2 which carries the HSV TK gene, the gene is specifically induced in cancerous target cells as opposed to normal cells in the host. body. It has been shown that there is a higher relative induction of several Salmonella promoter genes, including pagB and pagC, (Miller et al., 1989, Proc. Natl. Acad. Sci. USA 86:5054-5058;
Miller et al., 1992, Infect. Immun. 60:3763-3770; Alpuche Aranda et al., 1992, Proc. Natl. Acad. Sci. USA 89:10079-10083) when the bacteria invade macrophages as opposed to epithelial cells. In order to test whether these promoters are also activated when Salmonella invade melanoma cells, we used Salmonella-bearing promoter constructs fused to the 6-galactosidase reporter gene.
Human melanoma M2 cells, (Cunningham et al, 1992, Science, 255:325-327) human epithelial 1790 cells and mouse macrophage cell line J774 cells (American Type Culture Collection) were seeded at a density of 1 x 106 host cells in 25 cm2 Corning tissue culture flasks. The cells were infected with 5 x 107 Salmonella typhimurium #14028/ml DMEM culture medium for 1 hour, washed with fresh medium, and further cultured for 6 hours with 50 g/ml gentamicin added to the culture medium in order to kill the external but not the internalized Salmonella. The melanoma cells were then harvested by scraping them from the substratum in isotonic 1 mM EDTA solution. The cells were pelleted, resuspended in PBS and an aliquot was removed for quantitation of the bacteria found within the melanoma cells. The remainder of the melanoma cells were assayed for 0-galactosidase activity.
Three Salmonella typhimurium clones were used: i) strain 14028 in which (3-galactosidase was constitutively expressed; ii) strain 14028 in which /3-galactosidase was expressed through activation of the pagB promoter; iii) strain 14028 in which /3-galactosidase was expressed through activation of the pagC promoter. Thus, through measurements of f3-galactosidase activity, analyses of bacterial pagB and pagC promoter induction in melanoma cells were carried out.
The results are detailed in Table 14.
ENHANCED EXPRESSION OF BACTERIAL PROMOTERS PAGB AND PAGC IN
CULTURED HUMAN MELANOMA M2 CELLS' Promoter-Induced Activity: Constitute Activity Cell Line p gB pagC
human epithelial 1.3:1 7.2:1 mouse transformed 2.9:1 17:1 macrophage human melanoma 3.8:1 31:1 relative activation of pagB and pagC was assessed through expression of promoter-inducible $-galactosidase activity.

Both the pagB and pagC Salmonella promoters were induced in human melanoma cells. Levels of induction in melanoma cells were greater than seen in either the epithelial or macrophage cell lines. These data demonstrate that the pagB or pagC promoter could be used to express genes, such as HSV TK or E. coli cytosine deaminase, in a melanoma cell-specific manner.

13. EXAMPLE: CLONING AND EXPRESSION OF PRODRUG
CONVERTING ENZYMES
The following sections demonstrate useful systems for expression of prodrug-converting enzymes useful for the methods and compositions of the present invention.

13.1. CLONING AND EXPRESSION OF HERPES SIMPLEX
VIRUS THYMIDINE KINASE IN SALMONELLA
TYPHIMURIUM
Herpes simplex thymidine kinase (HSV TK) is known to be an effective pro-drug converting enzyme in the inhibition of melanoma tumor growth (Bonnekoh et al., 1995, J. Invest. Derm. 104:313-317). Accordingly, procedures were carried out to insert an HSV TK gene with the (3-lactamase signal sequence into both Salmonella typhimurium wild type strain 14028 and super-infective tumor-specific mutant clone 72 which is derived from the wild type strain.
Herpes simplex thymidine kinase cloning by PCR
Plasmid DNA of the vector pHETK2 (Garapin et al., 1981, Proc. Natl. Acad. Sci. USA 78:815-819) was prepared by alkaline lysis, phenol/chloroform extraction and ethanol precipitation. PCR primers based on the complete sequence for the Herpes simplex thymidine kinase (McKnight, 1980, Nuc.
A. Res. 8:5949-5964) were: forward 5'-GATCATGCATGGCTTCGTACCCCGGCC-3' (SEQ ID NO:1) and reverse 5'-CTAGATGCATCAGTGGCTATGGCAGGGC- 3', (SEQ ID NO:2) which corresponds to bases 310-328 (forward) and 1684-1701 (reverse) of the published sequence, with an added sequence of GATCATGCAT (portion of SEQ ID NO:1) or CTAGATGCAT (portion of SEQ ID NO:2) (NsiI site and spacer) at the S'end of each primer. Each reaction mixture contained 50 ng DNA template, 10 pmoles of each primer, 100 mM deoxynucleotide triphosphates, 1.5 mM Mg" and 0.5 units Taq polymerase (Perkin Elmer Cetus, Norwalk, CT). Amplification was performed by 35 cycles of 94 C for 1 minute; 50 C for 15 seconds; 55 C for 1 minute; and 72 C for 2 minutes. The amplified DNA was purified and was cloned into either pBluescript II KS+ and sequenced with T3 and T7 primers to confirm the correct DNA had been cloned or was cloned into p279 cut with Pstl which provides the Q-lactamase signal sequence (Talmadge et al., 1980, Proc. Natl. Acad. Sci. USA
77:3369-3373). Transformants were screened using a probe generated from the original template by random priming (Boehringer Mannheim, Indianapolis, IN) using [a-32P]dCTP.
Positive clones were further screened by immunoblot.
SDS-PAGE and Immunoblot Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed on bacterial lysates according to Weber and Osbom, 1975, Proteins and sodium dodecyl sulfate: Molecular mass determination on polyacrylamide gels and related procedures. In: H. Neurath and R. Hill (eds) The Proteins, Third Edition, vol. 1, Academic Press, New York pp. 179-223. Immunoblots were performed according to Towbin et al., 1979, Proc. Natl. Acad.
Sci. USA 76:4350-4354. Primary anti-TK antibodies were generally used at a 1:1000 dilution. Secondary anti-mouse antibodies were alkaline phosphatase-conjugates (Promega, Madison, WI) used at a 1:7,500 dilution, followed by nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-indolyl phosphate (BCIP) colorimetric detection (Promega).
Thymidine Kinase Assay Bacterial lysates were prepared by pelleting 1 ml of log-phase bacterial culture for 30 seconds at 12,000 x g in a microfuge centrifuge. The pellet and supernatant were retained separately and the supernatant was further cleared by centrifugation for 10 min at 12,000 x g. The pellet was further treated by resuspension in 100 pl of phosphate buffered saline containing 1 mg/ml lysozyme and 1% (v/v) Triton X-100TH and subjected to three cycles of rapid freezing and thawing. The resulting material was clarified by centrifugation at 12,000 x g for 2 minutes. Thymidine kinase activity was assayed using a modified version of the assay described by Summers and Summers, 1977, J. Virol. 24:314-318.
The reaction mix was incubated at 37 C for 1 hour and then bound to DE81 paper (WhatmanTM), washed, and the associated radioactivity determined in a gamma counter.
Salmonella transformation Transformation of Salmonella strains was performed by electroporation as described by O'Callaghan and Charbit, 1990, Mol. Gen. Genet. 223:156-158. Plasmids transfected into Salmonella included pHETK2 (Garapin et al., 1981, Proc.
Natl. Acad. Sci. USA 78:815-819) p279 (Talmadge et al., 1980, Proc. Natl. Acad. Sci. USA 77:3369-3373) and two independent isolates of fl-lactamase fusions, p5-3 and p2lA-2 (See Figure 4-C for a diagram of p5-3 and p2lA-2 where these plasmids are designated "pTK Sec 1.". Salmonella typhimurium strains transfected were the wild type 14028 and the super-infective clone 72.
Two independent 8-lactamase-TK gene fusion constructs were isolated and expressed in Salmonella typhimurium 14028 wild type and clone 72. An immunoblot analysis and corresponding enzyme activity assay are presented in Figures 4A and 4B. All three TK-containing vectors, the cytoplasmically expressed pHETK2 and the /3-lactamase fusions p5-3 and p2lA-2, were detectable by immunoblot and enzyme assay. Relatively little enzyme activity was recovered from the culture supernatants. Since the immunoblot analysis shows processing of the signal sequence, secretion into the periplasmic space of the Salmonella typhimurium is expected.
13.2. SYSTEMS FOR EXPRESSION OF HERPES SIMPLEX
THYMIDINE KINASE USING VARIOUS PROMOTERS
AND SECRETION SIGNALS
A number of constructs were made to express TK
using other promoters and other secretion signals.

13.2.1. EXPRESSION AS A STAPHYLOCOCCUS PROTEIN A
FUSION UNDER THE LACI PROMOTER
Herpes simplex thymidine kinase was amplified by PCR as described in Section 13.1 above and cloned into the Pstl site of pBluescript. This TK clone was subcloned from bluescript to the BamHI and Hindlll cite of the secretion vector pEZZ18 (Promega, Madison, WI; Nilsson and Abrahamsen, 1990, Methods in Enzymology 185:144-161). This resulted in an in-frame fusion with Staphylococcus protein A under the lacl promoter. This plasmid was designated pTK-Sec2 and is diagramed in Figure 4-C. Plasmid pTK-Sec2 expresses thymindine kinase as determined by an immunoblot.

13.2.2. CLONING OF THE SERRATIA MARCESENS
CHITINASE SIGNAL SEQUENCE AND PROMOTER
The promoter and signal sequence of Serratia marcesens chitinase I (Jones et al., 1986, EMBO J. 5:467-473) was cloned by PCR. The forward and reverse primers had the following sequence: CTAGACTAGTTTGTCAATAATGACAACACCC (forward) (SEQ ID NO:3) and GATCGGATCCTTGCCCGGCGCGGCGGCCTG (reverse) (SEQ ID NO:4) which contain SpeI and BamHI sites, respectively. The resulting product was cloned into pSP72 and confirmed by DNA sequencing. This plasmid was designated pSP-CHT and is also diagramed in Figure 4-C.

13.2.3. EXPRESSION AS A CHITINASE SIGNAL SEQUENCE
FUSION UNDER THE CONTROL OF THE CHITINASE
PROMOTER
Herpes simplex thymidine kinase in pBluescript was subcloned into the pSP-CHT vector using BamHI and Hindlll.
This results in an in-frame fusion with the chitinase signal sequence under the chitinase promoter. This plasmid was designated pTK-Sec3 and is also diagramed in Figure 4-C.
Plasmid pTK-Sec3 expresses thymindine kinase as determined by an immunoblot.

13.3. EXPRESSION OF P450 OXIDOREDUCTASE PRODRUG
CONVERTING ENZYME IN BACTERIA USING AN
EXOGENOUSLY INDUCIBLE PROMOTER
The sulA promoter element (GENBANK #V00358; Cole 1983) was cloned from E. coli genomic DNA by PCR using forward (CTAGAAGCTTATAAGGGTTGATCTTTGTTGTC) (SEQ ID NO:5) and reverse (GTACGATATCCAGAACGATGTGCATAGCCTG) (SEQ ID NO:6) primers which incorporate the Hindill and EcoRV restriction sites respectively. The PCR conditions were 35 cycles of 95 C, 1 minute; 55 C 1 minute; and 72 C for 1 minute. The product was cloned into pSP72 and sequenced with the T7 primer to confirm that the correct DNA had been obtained.
The cloned DNA fragment was 100% identical to the published sequence.
The NADPH-dependent cytochrome p450 oxidoreductase (p450 OR) cDNA clone in the EcoRI site of pBluescript that lacks the first initiating ATG (deletion of the first 11 nucleotides) of the cDNA described by Yamano et al., 1989, Molecular Pharmacol. 35:83-88, was fused with the sulA
promoter and initiating sequence by cloning the sulA promoter obtained as described above into the Hindlll and EcoRV site of the p450 oxidoreductase gene. The resulting fusion consists of the sulA promoter, including the sulA ATG
(methionine) and subsequent 9 amino acids (YTSGYAHRS) (SEQ ID
NO:7) as well as six amino acids which follow introduced from the DNA polylinker and PCR primers (SGYRIP) (SEQ ID NO:8) followed with the second amino acid of p450 OR, which is G.
This construct, pSP-SAD4-5 is diagramed in Figure 4-D.
13.4. EFFECT OF EXPRESSION OF P450 OXIDOREDUCTASE CONVERSION OF PRODRUG ON
BACTERIAL GROWTH
It has been previously shown (Shiba et al., 1959, Nature 183:1056-1057) that some strains of bacteria are sensitive to low levels of mitomycin. Therefore, to compare the sensitivity of a specific bacterial strain with and without the sulA::p450 OR expression plasmid the following experiment was performed. If the construct is functional, the presence of the p450 OR gene is expected to result in increased activation of mitomycin resulting in decreased growth of the bacteria. The sulA::p450 OR expression plasmid used in this experiment is similar to pSP-SAD4-5 (Section 13.3) except that it is in a pBluescript (pBS) backbone and has the Q-galactosidase transcription unit. This construct is also diagramed in Figure 4-D. The pBS plasmid, with and without the expression construct, was transfected into Escherichia coli DH5a by electroporation and clones containing the correct plasmids were obtained and confirmed by plasmid isolation and DNA restriction analysis. For each of the two plasmid-bearing strains, a fresh, 4 hr (late log) culture was diluted 1:100 into LB with 100 g/ml ampicillin to select for the presence of the plasmid and grown at 37 C
at 250 rpms. Mitomycin C was added to the cultures in amounts of 0.0, 0.1 and 0.5 g per ml.
Optical density was determined at 600 nm using a Perkin Elmer double beam spectrophotometer at 2 and 18 hour time points. The results are presented in Table 14(A).
GROWTH OF BACTERIAL CULTURES IN THE PRESENCE OF MITOMYCIN C
OD600 t = 2 hours post drug Amount of Mitomycin C (Ng/ml) added Plasmid 0 0.1 0.5 pBS 0.052 0.050 0.053 suiA::p450 OR 0.037 0.030 0.024 OD600t= 18 hours post drug Mitomycin C (erg/ml) added Plasmid 0 0.1 0.5 pBS 2.33 2.23 1.93 suiA::p450 OR 2.26 0.34 0.071 Comparison of the growth of the E. coli strain DH5u containing pBS and sulA::p450 OR in the absence of drug at the 2 and 18 hour time points shows that the presence of the construct partially inhibits the rate of growth but does not inhibit attaining a high final OD at 18 hours. These data also show that bacteria carrying the pBS backbone plasmid alone are only partially inhibited at the higher mitomycin concentration. However, those carrying the sulA::p450 construct show significant inhibition at both early and late time points at both mitomycin concentrations. These data indicate a strong dose response to mitomycin conferred by the presence of the sulA::p450 construct.

13.5. EXPRESSION OF CYTOSINE DEAMINASE IN
SALMONELLA TYPHIMURIUM
E. coli cytosine deaminase (CD) has been shown to be an effective prodrug-converting enzyme useful for gene therapy (Hirschowitz et al., 1995, Human Gene Therapy 6:1055-1063; Huber et al., 1993, Cancer Res. 53:4619-4626; Huber et al., 1994, Proc. Natl. Acad. Sci. USA 91:8302-8306; Moolten, 1994, Cancer Gene Ther. 1:279-287; Mullen et al., 1992, Proc.
Natl. Acad. Sci. USA 89:33-37; Mullen et al., 1994, Cancer Res. 54:1503-1506; Trihn et al., 1995, Cancer Res. 55:4808-4812). CD functions by converting the non-toxic 5-fluorocytosine (5-FC) to the toxic compound 5-fluorouracil (5-FU). Salmonella possess an endogenous CD, however, its expression is catabolite repressed (West and O'Donovan, 1982, J. Bacteriol. 149:1171-1174). A CD expression vector using the constitutively active f3-lactamase promoter to ensure expression of CD within tumors was cloned as described below.
Cloning and expression of CD
PCR primers based on the complete sequence for E.
coli cytosine deaminase (Huber et al., 1993, Cancer Res.
53:4619-4629) were forward: 5'-GATCATGCATGTGGAGGCTAACAGT-3' (SEQ ID NO:9) and reverse: 5'-CTAGATGCATCAGACAGCCGCTGCGAAGGC-3' (SEQ ID NO:10), corresponding to the published sequence, with the added sequence GATCATGCAT (portion of SEQ ID NO:9) or CTAGATGCAT (portion of SEQ ID NO:10) which is a NsiI site and spacer at the 5' end of each primer. Each 25 Al reaction mixture contained 50 ng DNA template, 10 pmoles of each primer, 100 mM deoxynucleotide triphosphates, 1.5 mM Mg and 0.5 units Taq polymerase (Perkin Elmer Cetus, Norwalk, CT).
Amplification was performed by 35 cycles of 94 C for 1 minute; 50 C for 15 seconds; 55 C for 1 minute; and 72 C for 2 minutes. The amplified DNA was purified on an agarose gel and the band of correct size was cloned into 1) pBluescript II KS+ and sequenced with T3 and T7 primers to confirm the correct DNA had been cloned and 2) p279 cut with Pstl which provides the 0-lactamase signal sequence and the constitutive 0-lactamase promoter. This second construct was designated pCD-Secl and is diagramed in Figure 4-E. Transformants were screened using a [a-32P]dCTP-labeled oligonucleotide probe.
Positive clones were further screened by immunoblot using anti-CD antibodies described below.
Primary Antibodies to Cytosine Deaminase CD was subcloned from pBluescript into pGEX I and expressed using IPTG. The expressed protein was found to be insoluble and present in inclusion bodies. CD-glutathione-S-transferase (GST) fusion protein was purified from inclusion bodies by washing in 0.1% w/v Triton X-100, repelleted and resuspended in SDS-PAGE sample buffer. The material was separated on a 3 mm preparative 10% polyacrylamide gel and excised after visualization with 3M potassium acetate at 4 C.
The purifed bands were homogenized and injected i.p. into DBA2J mice with Freund's complete (day 0) and incomplete (day 14) adjuvant. After 6 weeks the mice were bled and the ability of the serum antibodies to bind to cloned CD
confirmed by Immunoblot.
SDS-PAGE and Immunoblot SDS-polyacrylamide gel electrophoresis (SDS-PAGE) was performed on bacterial lysates according to Weber and Osbom, 1975, Proteins and sodium dodecyl sulfate: Molecular mass determination on polyacrylamide gels and related procedures. In:H. Neurath and R. Hill (eds) The Proteins, Third Ed., Vol. I, Academic Press, New York, pp. 179-223.
Immunoblots were performed according to Towbin et al., 1979, Proc. Natl. Acad. Sci. USA 76:4350-4354. Primary anti-CD
antibodies described above were generally used at a 1:500 dilution. Secondary anti-mouse antibodies were alkaline phosphatase-conjugates (Promega, Madison, WI) used at a 1:7,500 dilution, followed by nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-indolyl phosphate (BCIP) colorimetric detection (Promega, Madison, WI).
CD Enzyme Assay Bacterial lysates were prepared by pelleting 50 ml of overnight bacterial culture at 3000x g for 10 minutes and resuspending them in 2.5 ml of PBS. The cells were sonicated and the debris removed by pelleting in a microfuge at 12,000x g for 10 minutes. The enzyme assay performed was modified from Mullen et al., 1992, Proc. Natl. Acad. Sci. USA:89:33-37. 10 Al of cell extract was incubated with 1 Al [H3]-5FC (1 ACI/ l) a 37 C. 1 Al was spotted on a Kodak 13254 microcrystalline nitrocellulose TLC plate, Eastman Kodak, Rochester New York, and separated using 95:5 Butanol:water with unlabeled 5FC and 5FU markers. The plates were cut based upon separation of the marker lanes and quantified using a liquid scintillation counter.
Salmonella transformation Transformation of Salmonella strains was performed by electroporation as described by O'Callaghan and Charbit, 1990, Mol. Gen. Genet. 223:156-158. Plasmids transfected into Salmonella were p279 and pCD-Secl. Salmonella typhimurium strains transfected were strains YS721, YS7211, YS7212 and YS7213 which are described infra in Section 18.
E. coli strains transfected were strains DH5a and KL498 (Acod).
Biodistribution of Salmonella carrying the CD expression construct Salmonella typhimurium clone YS7212 carrying the CD
expression construct pCD-Secl was grown in LB media to an OD600 of 0.8. An aliquot of 1.0 x 106 bacteria were inoculated i.p. into C57/B6 mice which had been implanted with 2 x 105 B16 melanoma cells 16 days prior to the bacterial infection.
At two days post bacterial infection, mice were sacrificed and tumors and livers assayed for the presence of the bacteria by homogenization and plating of serial dilutions.
The expressed protein product of pCD-Secl bound to the anti-CD anti-serum by immunoblot analysis. When this clone was transferred to the E. coli strain KL498 which lacks CD, it was found to confer a high degree of enzyme expression as determined by the conversion of 5FC to 5FU, as shown in Figure 4-F. Figure 4-F also demonstrates that the cloned CD-expression plasmid gives higher levels of conversion than the E. coli strain MG1655 which expresses the wild type haploid cod gene which encodes for endogenous CD.

14. EXAMPLE: PROLIFERATION OF SALMONELLA TYPHIMURIUM
CLONE #725-3-2 IN MELANOMA TUMORS IN MICE
In a similar set of experiments as in Section 10.2, DBA/2J mice (approximately 10 weeks) were inoculated (s.c.) with 3x105 Cloudman S91 melanoma/macrophage hybrid #48 cells in each of 4 sites over the right and left shoulders and flanks. Tumors were palpable 10-12 days post inoculation of tumor cells from tissue culture into mice. After two weeks post-inoculation of tumor cells (s.c.), tumor-bearing mice were additionally inoculated (i.p.) with 2x105 c.f.u. of S.
typhimurium clone 72 containing the HSV TK gene with the 0-lactamase signal sequence which is designated 725-3-2. After 2 and 10 days of bacterial infection without antibiotic treatment, representative tumor-bearing animals were sacrificed and their tumors and livers were homogenized and quantitated for c.f.u. of Salmonella per gram of tissue. In addition, individual clones of bacteria were isolated from the liver and tumor homogenates 10 days post-infection and tested for the genetic markers xylneg (inability to metabolize xylose, characteristic of clone #72) and tetTe9 (resistance to the antibiotic tetracycline). The genotype of the inoculated Salmonella typhimurium clone #725-3-2 was xylneg and tetres The tetres clones of Salmonella were assumed to carry the HSV TK
gene, since the HSV TK gene was carried on a plasmid that carried the tetres marker.
The results are presented in Tables 15 and 16.
After 2 days of infection, the tumors contained an average of 1.5 x 109 Salmonella/g tumor and 2.0 x 105 Salmonella/g liver, with an average ratio of tumor:liver of about 7,500:1.
INOCULATION (I.P.) INTO CLOUDMAN S91 MELANOMA-BEARING DBA/2J
MICE
Tissue Salmonella/g Tissue Tumor/Liver Liver (n = 2) 2.0 x 105 ---Tumor (n = 4) 1.5 0.9 x 109 7,500:1 After 10 days of infection (Table 16), the tumors contained an average of 2.9 x 109 Salmonella/g tumor and 2.7 x 105 Salmonella/g liver, a ratio of 11,000:1 (tumor:liver), similar to the distribution of bacteria seen 1-2 days post-infection. These results demonstrate that once inoculated (i.p.) into tumor-bearing mice, Salmonella enter the circulatory system, infect the tumor cells, proliferate within the tumors, and exist there in a compartmentalized fashion.

INOCULATION (I.P.) INTO CLOUDMAN S91 MELANOMA-BEARING DBA/2J
MICE

Salmonella/9 Tissue Tissue Tumor/Liver xylose tetracycline Liver 2.7 x 105 --- neg 9/10 res Tumor #1 4.2 x 109 16,000:1 neg 3/7 res #2 3.1 x 109 12,000:1 neg 7/8 res #3 1.3 x 109 4,800:1 neg 8/8 res Average 2.9 x 109 11,000:1 The results further demonstrate that 10 days post-infection, all of the bacterial clones examined were xylne9, proving their genetic relationship to the inoculated clone #725-3-2 , and that 27/33 clones remained tetras, demonstrating high degree of retention (82%) of the HSV TK containing plasmid within the host bacteria. In experiments not shown here, the same plasmid was found to be 100% retained after 42 hours of infection in tumor-bearing mice.
In a continuation of the above experiments summarized in Tables 14, 15 and 16, the Salmonella infections in melanoma-bearing mice were continued for a total of 4 weeks. To alleviate the symptoms of Salmonella poisoning (shaking, matted hair) the animals were placed on antibiotics for the final two weeks. Such antibiotic treatments consisted of the inclusion in the mouse drinking water of SULFATRIMT"' Pediatric Suspension (Schein Pharmaceutical, Inc.;
sulfamethoxazole 40 mg/ml, trimethoprim 8 mg/ml) at a final concentration of 15 ml SULFATRIM'/500 ml drinking water. At termination of the experiment, the surviving mice were sacrificed by euthanasia, and the tumors and livers were removed. Portions of the tissues (1-2 mm3) were fixed in formalin and stained for histological examination. The remaining portions were weighed, homogenized in 5 ml LB
broth/g tissue, and the number of Salmonella were quantitated on LB agar plates. Results are presented in Table 17.

INOCULATION (I.P.) INTO MELANOMA-BEARING DBA/2J MICE

Tissue Salmonella/g tissue Tumor/Liver Liver 5.1 x 107 ---Tumor #1 2.2 x 109 43:1 #2 9.4 x 108 18:1 #3 2.3 x 109 45:1 #4 2.0 x 108 6:1 Average 1.4 1x109 28:1 The excised melanoma tumors averaged less than 1 gram in weight compared to 5-10 gm tumors in control animals at death (data not shown). It was found that these tumors contained an average of 1.4 x 109 Salmonella/g tumor, similar to the number of tumor-infecting bacteria seen at 1, 2, and days post-inoculation. However, the number of Salmonella in the liver increased during the 4 week infection, so that the average ratio of bacteria in tumor over liver was reduced to 28:1 compared to the ratios obtained with infection 10 periods of up to 10 days as seen in Tables 14, 15 and 16.
15. EXAMPLE: MICROSCOPIC DETECTION OF SALMONELLA
TYPHIMURIUM IN MELANOMAS IN VIVO

15.1. DETECTION OF SALMONELLA TYPHIMURIUM

In order to study the histopathology of Salmonella infection in the tumor-bearing mice, representative melanoma tumors were removed from euthanized mice with or without Salmonella infection. Portions of the tissues (1-2 mm3) were fixed in formalin, embedded and sectioned, and the sections stained with either hematoxylin and eosin, or tissue gram stain for histological examination. Results of these studies are shown in Figures 5A-B.
Figures 5A-B are photomicrographs of histological sections from a Cloudman S91 melanoma/macrophage hybrid #48 melanoma growing subcutaneously in a DBA/2J mouse. The tumor was excised from a mouse that had been inoculated 2 days earlier with 3 x 105 c.f.u. Salmonella typhimurium super-infective clone 72 carrying the HSV TK gene, 725-3-Z A
portion of the tumor was weighed, immersed in LB at 5ml/g tumor, homogenized with a ground glass homogenizer, and the tumor homogenate was plated onto LB-Agar culture plates at various dilutions in order to quantitate the amount of Salmonella typhimurium in the tumor. Quantitation of the bacteria revealed that the tumor contained 1.4 x 109 Salmonella/g. Figure 5A. A section stained with hematoxyn and eosin shows a cross-section of the tumor with an area of necrosis, denoted by the arrow. Figure 5B. A section stained with tissue gram stain shows gram negative bacteria in an area of necrosis area of the tumor. When viewed with the light microscope, the bacteria stain pink/purple against a yellow background. Salmonella-infected necrotic areas were surrounded with dead tumor cells that did not stain with tissue gram stain but which could be detected through melanin-containing melanosomes (see Figure 6). These results show that the necrotic areas of solid tumors are accessible to Salmonella when the bacteria are introduced into a tumor-bearing host via the circulatory system.
In an additional set of analyses, sections of Cloudman S91 melanoma/macrophage hybrid #48 melanoma tumors growing in a Salmonella-infected mouse were examined with the electron microscope. To initiate the experiment, a mouse was inoculated s.c. with 8 x 105 tumor cells. A palpable tumor mass was detected 11 days later, at which time the mouse was inoculated i.p. with 3.6 x 106 c.f.u. of S. typhimurium super-infective clone #72. Forty-two hours post-inoculation, the mouse was sacrificed by metofane anesthesia. The tumor was excised using aseptic techniques. Quantitation of the bacteria within the tumor revealed that the tumor contained approximately 7.5 x 109 S. typhimurium/g upon excision at 42 hours. In contrast, the concentration of S. typhimurium in the liver from the same mouse was approximately 2.0 x 107/g, a ratio of bacteria in tumor to liver of approximately 400:1.
A second portion of the tumor was cut into 1-2 mm3 pieces and fixed in 1/2 strength Karnovsky's fixative for 6 hours at 4 C, followed by washing in cacodylate buffer overnight. The tumor tissue was post-fixed with 1% Os04 and 1.5% potassium ferrocyanide in cacodylate buffer for 2 hours and embedded in Spurr's resin. Ultrathin sections were stained with uranyl acetate and lead citrate. They were viewed with a Zeiss 109 electron microscope.
Shown in Figure 6 is an electron micrograph of a field within a melanoma tumor that includes two separate S.
typhimurium along with numerous melanosomes, which are specialized subcellular organelles present in the cytoplasm of melanoma cells. The presence of bacteria along with the melanosomes provides proof that the S. typhimurium entered the cytoplasm of the melanoma cell via the bloodstream of the mouse. The S. typhimurium in the electron micrograph appear identical to those shown previously in intestinal epithelial cells following an experimental infection of the mouse, Takeuchi, 1967, Am. J. Pathol. 50:109-1361.
In summary, i) examination with the light microscope revealed that Salmonella typhimurium exists in the necrotic areas of Cloudman S91 melanomas growing in infected DBA/2J mice; and ii) examination with the electron microscope revealed that Salmonella typhimurium also exists within the cytoplasm of melanoma tumor cells.

15.2. DISTRIBUTION OF SALMONELLA TYPHIMURIUM

C57BL/6J mice (11-13 weeks old) were inoculated s.c. in two sites (shoulder and flank), with 3.5 x 105 B16F10 mouse melanoma cells per site. After the appearance of palpable tumors (approximately 2 weeks) the animals were further inoculated i.p. with about 105 bacteria of the following three strains: i) wild type Escherichia coli K-12 strain #CSH 101; ii) Salmonella typhimurium strain 14028; and iii) mutant Salmonella typhimurium super-infective clone 72 carrying the HSV thymidine kinase gene, 725-3-2. After about 2 days of infection, mice were euthanized by anesthesia with metofane. Tumors and livers were removed aseptically, rinsed with sterile 0.9% NaCl, weighed, and homogenized in LB broth at a ration of 5:1 (vol. broth:wt. tumor). Prior to homogenization, 1-2 mm3 pieces of tissue were removed from representative tumors, fixed with 1/2 strength Karnovsky's fixative, and processed for analysis with the electron microscope. Bacteria in the homogenates were quantitated by plating onto LB plates, incubating overnight at 37 C, and counting bacterial colonies.
Results were as follows: i) wild type E. coli were found in relatively low numbers in both the tumor and liver of the inoculated animals at concentrations averaging <l03/g tumor and <l02/g liver. ii) wild type S. typhimurium were found in significantly higher numbers than E. coli in both tumor and liver, with infecting bacteria ranging from 2 x 107 to 6 x 108 c.f.u./g tumor, and 4 x 106 c.f.u./g liver. One of the two C57BL/6J mice inoculated with the wild type S.
typhimurium strain died, possibly from septic shock. iii) S.
typhimurium super-infective clone 72 were also found in significantly higher numbers than E. coli in both tumor and liver, further, the number of clone 72 S. typhimurium/g liver was significantly lower than the number of wild type S.
typhimurium/g liver. The results are detailed below in Table 18.

DAYS FOLLOWING INOCULATION (I.P.) INTO C57B6 MICE BEARING B16F10 TUMORS
Bacteria/ Tumor/
Bacterial Strain Mouse Tissue gm Tissue Liver E. colt K-12 A Liver 355 (CSH #101) Tumor #1 1200 4:1 Tumor #2 50 1:7 A' Liver 100 Tumor #1 50 1:7 S. typhimurium B Liver 4.3 x 106 (14028 wild type) Tumor #1 2.3 x 107 5:1 Tumor #2 6.0 x 108 136:1 B' (dead) S. typhimurium C Liver 2.0 x 104 (clone # 725-3-2) Tumor #1 1.0 x 108 5,000:1 Tumor #2 1.2 x 105 6:1 C' Liver 8.5 x 104 Tumor #1 9.3 x 108 11,000:1 (2.5g) In summary, Salmonella typhimurium displays natural capabilities over Escherichia coli in its ability to infect and proliferate within B16F10 melanoma tumors. Furthermore, super-infective clone 725-3-2 displays superior qualities to its wild type parental strain 14028 in its reduced infection of liver in C57BL/6J mice, i.e., the wild-type strain 14028 showed greater infectivity toward liver than did clone 725-3-2.
The higher infectivity of liver by the wild-type Salmonella was consistent with the observed greater lethality of the wild type stain toward DBA/2J mice and the greater infectivity of liver in DBA/2J mice than that produced by clone 72 as seen in Table 11B. Together, the results in Tables 11B and 18 provide the first evidence that selection for strains of bacteria or other parasites with enhanced tumor specificity in vitro can yield mutant strains with attenuated host toxicity in vivo.

15.3. MICROSCOPIC DETECTION OF SALMONELLA

Representative B16F10 melanoma tumors were removed from euthanized mice with or without Salmonella infection.
Portions of the tissues (1-2 mm3) were fixed in formalin, embedded and sectioned, and the sections stained with either hematoxylin and eosin, or tissue gram stain for histological examination. Results of these studies are shown in Figures 7A-B. Figures 7A and 7B are light micrographs of histological sections from a B16F10 melanoma growing subcutaneously in a CS7BL/6J mouse. The tumor was excised from a mouse that had been inoculated 2 days earlier with 2 x 105 c.f.u. Salmonella typhimurium super-infective clone, 725-3-2 carrying the HSV TK gene. Quantitation of the bacteria within the tumor revealed that the tumor contained approximately 9 x 108 c.f.u. S. typhimurium/g upon excision 2 days post-infection with bacteria. In contrast, the concentration of S. typhimurium in the liver from the same mouse was approximately 2.0 x 105/g, a ratio of bacteria in tumor to liver of approximately 400:1. Figure 7: A section stained with tissue gram stain shows gram negative bacteria in an area of necrosis within the tumor. The infected necrotic area is surrounded by dead melanoma cells that do not stain with the tissue gram stain but which appear brown in color due to the presence of melanized melanosomes. When viewed with the light microscope, the bacteria stain pink/purple against a yellow background. The results show that necrotic areas of B16 melanoma tumors are accessible to Salmonella when the bacteria are introduced into a tumor--bearing host via the circulatory system.
A second portion of the above-described tumor was cut into 1 mm3 pieces and fixed in 1/2 strength Karnovsky's fixative for 6 hours at 4 C, followed by washing in cacodylate buffer overnight. The tumor tissue was post-fixed with 1% Os04 and 1.5% potassium ferrocyanide in cacodylate buffer for 2 hours, and embedded in Spurr's resin. Ultrathin sections were stained with uranyl acetate and lead citrate.
They were viewed with a Zeiss 109 electron microscope as depicted in Figure 8.
The electron micrograph in Figure 8 shows numerous Salmonella typhimurium in extracellular spaces, denoted by arrows, in an area of necrosis. A single bacterium is also seen within the cytoplasm of a dying melanoma cell. The cytoplasm of the dying melanoma cell also contains numerous black melanosomes, characteristic of the B16F10 melanoma.
The S. typhimurium in the electron micrograph appear identical to those shown previously in intestinal epithelial cells following an experimental infection of the mouse, Takeuchi, 1967, Am. J. Pathol. 50:109-136.
In summary, i) examination with the light microscope revealed that Salmonella typhimurium exist abundantly in the necrotic areas of B16F10 melanomas growing in infected B16F10 mice; and ii) examination with the electron microscope revealed that Salmonella typhimurium also exist within the cytoplasm of tumor cells. Salmonella were also observed in tumor-associated neutrophils.
16. EXAMPLE: USE OF SUPER-INFECTIVE TUMOR-SPECIFIC
GENE-DELIVERING SALMONELLA TYPHIMURIUM FOR
TREATMENT OF MICE BEARING MELANOMA TUMORS
16.1. TREATMENT OF CLOUDMAN 591 MELANOMA
Salmonella typhimurium super-infective mutant 725-3-2, constitutively expressing the Herpes simplex virus thymidine kinase gene with the 0-lactamase signal sequence, was used for gene therapy of melanoma in mice (see Figure 4-C). DBA/2J mice (approximately 10 weeks) were inoculated (s.c.) with 3 x 105 Cloudman S91 melanoma/macrophage hybrid cells in each of 4 sites over the right and left shoulders and flanks. Tumors were palpable 10-12 days post-inoculation of tumor cells.
After two weeks post-inoculation of tumor cells, tumor-bearing mice were further inoculated (i.p.) with 2 x 105 c.f.u. of S. typhimurium clone 72 containing the HSV
thymidine kinase gene with the /3-lactamase secretory signal sequence which is designated 725-3-2. Twelve hours after inoculation of the bacteria, some of the mice were further inoculated (i.p.) with 2.5 mg ganciclovir sodium (Cytovene, Syntex Laboratories, Palo Alto, Calf.) in isotonic saline.
These same mice received this dosage of ganciclovir four times over a 3 day period. Control tumor-bearing mice also received ganciclovir but no bacteria. Another set of tumor-bearing mice was inoculated with bacteria, but received no ganciclovir. At various times appropriate groups of mice, treated as above, were also given the antibiotic SulfatrimTM' Pediatric Suspension (Schein Pharmaceutical, Inc.;
sulfamethoxazole 40 mg/ml, trimethoprim 8 mg/ml) at a concentration of 15 ml SulfatrimTM/500 ml drinking water.
Results were as follows:
1) Control melanoma tumor-bearing mice, receiving ganciclovir and antibiotic treatment (SulfatrimTM in drinking water) but no bacteria, developed rapidly growing tumors that initially doubled in size every 3 to 4 days, determined by caliper measurements as shown in Figure 9. These animals exhibited little or no side-effects from the ganciclovir treatment, confirming previous reports on the minimal toxicity of the ganciclovir pro-drug in mice in the absence of a suitable thymidine kinase converting enzyme (Bonnekoh et al, 1995, J. Invest. Dermatol. 104:313-317). By 30 days post-inoculation with tumor cells, all mice in this group had formed massive subcutaneous tumors (5-10 gm) and had died from melanoma.
2) One group of tumor-bearing mice received bacteria for a total of 10 days without administration of antibiotics, and received no ganciclovir. These animals had tumors that were significantly reduced in size from the tumors in control mice (Figure 10). The effect of Salmonella alone on reducing tumor size became evident several days after the effect of Salmonella plus ganciclovir on tumors had been observed as described below. However, all the animals in the "Salmonella alone" group developed symptoms of Salmonella infection (shaking, matted hair) and 50% of these animals succumbed between 5-10 days post-infection. The remaining animals were treated with the antibiotic SulfatrimT' (Schein Pharmaceutical, Oral Suspension) at a concentration of 15 ml/500 ml drinking water. This treatment reduced the clinical symptoms of Salmonella infection in the mouse population within 24-48 hours. The surviving animals from this protocol had significantly smaller tumors than control animals and remained alive past the 30 day period, when all of the control animals had died from melanoma.
3) Another group of tumor-bearing mice received ganciclovir plus bacteria during a 4-day treatment period.
About 500 of the animals succumbed within 1-2 days of this treatment, apparently from the conversion of ganciclovir to its toxic, phosphorylated form by the HSV TK expressed by the Salmonella clone 725-3-2 within the body of the mouse. At this time ganciclovir treatment was discontinued and the surviving animals were placed on Sulfatrim'' antibiotic to control the Salmonella infection. The total time of exposure to Salmonella without antibiotic was 4 days. The survivors from this protocol had significantly smaller tumors than control animals and remained alive past the 30 day period when all the control animals had died from melanoma (Figure 11).
In a further set of experiments, tumor progression was measured with calipers in various treated and untreated tumor-bearing mice. Groups of mice bearing Cloudman S91 melanoma/macrophage hybrid #48 melanoma tumors as described above were inoculated (i.p.) with 3 x 105c.f.u. Salmonella typhimurium super-infective clone 72 carrying the Herpes simplex virus thymidine kinase gene, 72532. Twenty-four hours after inoculation with bacteria, the mice were further inoculated with ganciclovir at doses of 2.0 mg, with a total of 6 inoculations over a 5 day period. The mice were then subjected to antibiotic treatment with a combination of 15 ml/500 ml Sulfatrim' and 20 g/ml Baytril' (Miles) in their drinking water. Baytril' or enrofloxacin is 1-cyclopropyl-7-(4-ethyl-l-piperazinyl)-6-fluoro-1,4-dihydro-4-oxo-3-quinolinecarboxylic acid. Tumor growth was assessed with periodic caliper measurements of tumor length, width, and height, and computed as tumor volume in mm3 by techniques known to the science of tumor biology. Results were plotted on a semi-logarithmic scale and generation times, the time in hours for one doubling in volume, were calculated. The following formula was used:
Generation time = 0.69 (t)/In (T1/T2), where t equals the time in hours between the initial tumor volume (T1) and the final tumor volume (T2) over the linear portion of the curve.
The results are shown in Figure 9 and Table 19.
Mean doubling times of tumors in untreated control mice and mice treated with ganciclovir but no Salmonella infection were similar, being 83 and 94 hours respectively. Tumors in mice treated with Salmonella for 5 days but no ganciclovir doubled at a mean rate of 125 hours. Tumors in mice treated with Salmonella for 5 days as well as ganciclovir showed no growth over the 10 day measurement period, and in some cases regressed with the treatment.

EFFECTS OF HSV TK-CONTAINING SALMONELLA TYPHIMURIUM ON
THE GROWTH OF CLOUDMAN S91 MELANOMAS IN DBA/2J MICE t TREATMENT WITH GANCICLOVIR

Treatment Mean Tumor Doubling Time (hrs) none 83 ganciclovir 94 S. typhimurium 125 S. typhimurium no growth + ganciclovir In summary: a) Control tumor-bearing animals receiving ganciclovir and antibiotic treatment, but no Salmonella, succumbed from massive tumors within 30 days of inoculation of tumor cells; b) Animals receiving Salmonella alone followed by antibiotic treatment showed reduced tumor growth rate and prolonged survival over control animals;
c) Animals receiving a combination of ganciclovir and Salmonella followed by antibiotic treatment showed little or no tumor growth compared to control animals, and prolonged survival over control animals. The results indicate that Salmonella typhimurium expressing the Herpes simplex virus thymidine kinase gene was able to convert ganciclovir to its phosphorylated form within the melanoma tumors, thus reducing tumor size and prolonging survival of the mice.

16.2. TREATMENT OF B16F10 MELANOMA
C57B6 mice were inoculated s.c., left shoulder region, with 5 x 105 B16F10 melanoma cells from culture. At 8 days post-tumor implantation some of the mice were further inoculated i.p. with 2 x 106 c.f.u. attenuated Salmonella typhimurium strains YS721, YS7211, YS7212 or YS7213 (see Section 18, infra) each carrying the HSV TK gene. At 11 days post-tumor implantation, GCV (ganciclovir sodium, CYTOVENE'M, Syntex Laboratories, Palo Alto, CA.) was inoculated i.p. into groups of mice (n=5 or n=10) under the following treatment protocols: a) total dose = 7.5 mg/mouse (2.5 mg day 11, 1.25 mg day 12; 2.5 mg day 18, 1.25 mg day 19); b) total dose =
5.0 mg/mouse (2.5 mg day 11, 2.5 mg day 12); c) total dose =
3.75 mg/mouse (2.5 mg day 11, 1.25 mg day 12); d) total dose = 2.5 mg/mouse (1.25 mg day 11, 1.25 mg day 12); e) total dose = 1.25 mg/mouse (1.25 mg day 11). At 18 days post-tumor implantation (10 days post bacterial inoculation) all animals were given enrofloxacin antibiotic, 0.2 mg/ml, in their drinking water (BAYTRIL") and maintained with this antibiotic supplement for 2 weeks. Tumor growth was assessed by caliper measurements and computed as volume in mm3. Animals were euthanized and listed as dead when the sum of their tumor measurements, length + width + height, reached 60 mm, or when they became moribund (listless, cessation of drinking).
The results obtained are illustrated in Figure 11C-H in Table 19(A).
TABLE 19(A) TYPHIMURIUM HSV TK GENE: EFFECTS OF GANCICLOVIR

Time Treated/Control Strain (n =) GCV Days S.D. T/C
Control (10) -0- 25+0 1.0 (10) 3.75mg 27+1 1.1 YS7212 (10) -0- 42+2 1.7 YS7212 (10) 3.75mg 33+4 1.3 YS7212/p5-3 (10) -0- 45+3 1.8 YS7212/p5-3 (10) 3.75mg 40 4 1.6 YS7213 (10) -0- 34+2 1.3 YS7213 (10) 3.75mg 35+2 1.4 YS7213/p5-3 (10) -0- 29+2 1.2 YS7213/p5-3 (10) 3.75mg 33+2 1.3 YS7211 (10) -0- 40+4 1.6 YS721 1 (10) 3.75mg 35+4 1.4 YS721 1 /p5-3 (10) -0- 34+2 1.4 YS7211/p5-3 (5) 1.25mg 30 4 1.2 (5) 2.50mg 37 4 1.5 (5) 3.75mg 38+4 1.5 (5) 5.Omg 42+6 1.7 (5) 7.5mg 39+6 1.6 Time of death post tumor cell inoculation.

The results from the various treatment protocols for the B1GF10 melanoma-bearing mice were as follows:

1) Effects of GCV on tumor-bearing animals with no bacterial inoculation Mice receiving melanoma cells but no bacteria were treated with GCV on days 11-12 post inoculation with tumor cells at doses from 3.75 mg-10 mg/mouse, depending on the experiment. In all trials, mice treated with GCV but no bacteria showed a small reduction in tumor volume that was noticeable within 5 days of GCV treatment and which persisted through the duration of the experiment, as shown in Figure 11C-E. GCV also elicited small but reproducible increases in survival time compared to that of non-treated control animals, as outlined in Table 19(A). These effects of GCV in the absence of bacterial treatment were not dependent upon dosage over the range studied.
These results demonstrate that the B16F10 cells employed in the experiment might have had the capacity to convert GCV to its toxic, phosphorylated form. Consistent with such a notion, it was found that proliferation of the B16F10 melanoma cells in culture was significantly suppressed when GCV was supplemented to the culture medium at 25 g/ml, but not at 10 g/ml as shown in Figure 11-F. Similar effects of GCV on DBA/2J mice bearing Cloudman S91 melanoma x macrophage hybrid 48, but not inoculated with bacteria, are reported in the Table 19.

2) Effects of GCV on tumor-bearing animals treated with bacteria not containing an HSV TK-plasmid When tumor-bearing mice were inoculated with Salmonella strains YS7211, YS7212 and YS7213, none of which contained the HSV TK gene, and then treated with GVC, GCV-mediated suppression of tumor growth was evident. Tumor suppression achieved with GCV was significantly greater than that seen with the bacteria alone, even when the suppressive effect of GCV on B16F10 tumors in control animals was taken into account. This indicated that Salmonella typhimurium could convert GCV to its phosphorylated, toxic form without the HSV TK gene, perhaps through endogenous phosphotransferase enzymes (Littler, et al., 1992, Nature 358:160-162; Sullivan et al., 1992, Nature 358:362-364).
Consistent with this notion was the finding that in addition to suppressing tumor growth, some combinations of bacteria and GCV treatment were highly toxic, shortening survival times of the animals, shown in Table 19(A). Toxicity might have resulted from production of phosphorylated GCV by those bacteria located in normal tissues such as liver or bone marrow.
3) Effects of GCV on tumor-bearing animals treated with bacteria containing an HSV TK-plasmid Tumor-bearing mice inoculated with HSV TK plasmid-containing Salmonella clone YS7211 (YS7211/p5-3) showed suppression of tumor growth and prolonged survival even in the absence of GCV treatment as shown in Figure 11-G.
Further, animals bearing both tumors and YS7211/p5-3 and additionally treated with 3.75mg GCV showed significant suppression of tumor growth above that seen in the absence of GCV. Using YS7211/p5-3 as a vector, GCV-mediated tumor suppression was evident in a dose-responsive manner when measured 28 days post implantation of tumors as shown in Figure 11-H. Tumor suppression correlated with increased average survival times for some categories of GCV-treated, tumor-bearing mice when compared to those inoculated with YS7211/p5-3 but not receiving GCV.

In summary:
1) In tumor-bearing animals not inoculated with Salmonella, GCV had a small suppressive effect on tumor growth that correlated with a small prolongation of survival.
2) Tumor-bearing animals inoculated with Salmonella not containing the HSV TK plasmid showed marked tumor suppression in response to GCV, above that seen in animals not treated with bacteria. In addition, some combinations of GCV and bacterial treatment were highly toxic to the animals, possibly through conversion of GCV to its toxic form by bacteria in extra-tumoral tissues such as liver or bone marrow.
3) Tumor-bearing animals inoculated with Salmonella containing the HSV TK plasmid also showed strong tumor suppression in response to GCV. It was not possible in these experiments to evaluate the relative contributions of HSV TK
as compared to endogenous Salmonella enzymes in the phosphorylation of GCV. However, using as a vector clone YS7211 containing the HSV TK expression plasmid, GCV-mediated tumor suppression and prolonged survival was demonstrated in a dose-dependent manner, see Figure 11-H.

17. EXAMPLE: LOCALIZATION OF SALMONELLA TYPHIMURIUM
WITHIN HUMAN TUMORS GROWN IN NU/NU MICE
The following experiments demonstrate localization of Salmonella in human tumors in experimental animals.

17.1. LOCALIZATION OF SALMONELLA WITHIN HUMAN
COLON TUMORS
NU/NU (BALB C) mice (9-10 weeks old) were inoculated s.c. in two areas (shoulder and flank), each with 1.5 x 107 HCT 116 human colon carcinoma cells. After the appearance of palpable, vascularized tumors (approximately 2 weeks) the animals were further inoculated i.p. with 3 x 105 Salmonella typhimurium super-infective clone 72532 carrying the HSV thymidine kinase gene. After 3.5 hours, 21 hours, and 72 hours of infection, mice were euthanized by anesthesia with metofane. Tumors and livers were removed aseptically, rinsed with sterile NaCl (0.9%), weighed, and homogenized with LB broth at a ration of 5:1 (vol. broth:wt. tumor). At 72 hours, prior to homogenization, pieces (1-2 mm3) were removed from representative tumors, fixed with 1/2 strength Karnovsky's fixative, and processed for analysis with the electron microscope. Bacteria in the homogenates were quantitated by plating onto LB plates, incubating overnight at 37 C, and counting bacterial colonies.
Results were as follows: At 3.5 hours and 21 hours there were insignificant levels of bacteria in the tumors or livers, even when the homogenates were plated undiluted onto LB agar plates. However, after 3 days 3/6 animals displayed high levels of Salmonella in the colon tumors, with bacterial tumor:liver ratios ranging up to 36,000:1. Data for these animals are summarized below in Table 20.
INOCULATION (I.P.) INTO HUMAN COLON CARCINOMA-BEARING NU/NU
MICE
Salmonella/ Tumor/
Mouse A gm tissue Liver Liver 2.6 x 104 Tumor 6.9 x 108 26,500:1 Mouse D

Liver 1.6 x 106 Tumor 3.1 x 109 2,000:1 Mouse E
Liver 1.0 x 105 Tumor 3.6 x 109 36,000:1 Shown in Figure 12-A an electron micrograph of a section from the HCT colon tumor excised from mouse A (Table 20) in which the number of Salmonella found to be 6.9 x 108/g tumor, and the tumor:liver ratio of infecting bacteria was 26,500:1. Shown in the micrograph are numerous Salmonella typhimurium within a vacuole in the cytoplasm of a neutrophil associated with the tumor. Some of the bacteria are undergoing division as denoted by the arrow. The neutrophil or polymorphonucleoleukocyte is characterized by its multi--lobed nucleus (n). Salmonella in tumor-associated neutrophils was also seen in infected B16F10 melanomas as described herein. The presence of bacteria in both colon and melanoma tumor-associated neutrophils following infection of tumor-bearing mice suggests that the Salmonella may have stimulated a host cellular immune response to the tumor cells. Enhancement of tumor immunity is thus another potential advantage in the use of parasites as tumor-specific therapeutic vectors.
17.2. LOCALIZATION OF SALMONELLA WITHIN VARIOUS
HUMAN TUMORS
Nu/nu (BALB C) mice (9-12 weeks old) were inoculated s.c. in the left shoulder region with 1-1.5 x 107 cells of the human lung carcinoma A549, human colon carcinoma HCT 116, human renal carcinoma CRL 1611, or human hepatoma HTB 52 (American Type Culture Collection). When palpable tumors developed, the mice were inoculated further with 2-5 x 106 cfu Salmonella typhimurium clone 72 for animals bearing human lung, liver, and renal tumors, and clone 725-3-2 for animals bearing human colon tumors. Clone 7253-2 carries the HSV thymidine kinase transcription unit. After 66-96 hours the animals were sacrificed, and the tumors and livers were removed and weighed. The tumor was homogenized in 5 vol LB
broth/gram wet weight tissue. Homogenates were quantitated by serial dilution on LB agar plates for the number of bacteria. The results are presented in Table 20(A) and represent the average standard deviation for n=3-4 animals.

TABLE 20(A) MICE BEARING HUMAN CARCINOMAS OF THE LUNG, COLON, KIDNEY, AND LIVER

Salmonella/g tissue:
Primary Tumor Tumor:
Tumor Tumor Liver wt(mg) Liver lung 3.2 1.4 x 109 1.0 0.3 x 107 462 186 320:1 carcinoma colon 2.5 1.6 x 109 5.8 8.9 x 105 428 235 4300:1 carcinoma hepatoma 6.7 11 x 108 5.7 9.0 x 106 103 29 120:1 renal 1.4 1.8 x 108 6.0 3.0 x 105 103 99 230:1 carcinoma As shown in Table 20-A, when inoculated i.p. into nu/nu mice, Salmonella typhimurium clone 72 was able to target human carcinomas of the lung, colon, kidney, and liver, and proliferate within them, generally, but not always, reaching levels of 108-109/g tumor. In the BALB/c nu/nu mice used, the skin was hairless and translucent allowing it to be determined visually that all the tumors were vascularized. The ranges of wet weights of the Salmonella-infected tumors were lung carcinoma, 220-600 mg;
colon carcinoma, 160-600 mg; hepatoma, 70-120 mg; and renal carcinoma, 40-250 mg.
Bacterial colonies were picked randomly from liver and tumor homogenates obtained from renal carcinoma- and hepatoma-bearing nu/nu mice 96 hrs post-inoculation of clone 72 and tested for phenotype by replicate plating. In all homogenates tested, 50/50 colonies were found to be Ade- and Xyl eg, consistent with the clone 72 phenotype.
The results further support the notion that derivatives of Salmonella typhimurium are useful as therapeutic vectors for a broad range of solid tumors, independent of the tumor origin or size. In several studies Salmonella clone 72 and its derivatives targeted and amplified within highly vascularized tumors as small as 40-100 mg in the case of human tumors in nu/nu mice, as well tumors of 4-8 g with large necrotic areas in the case of B16F10 melanomas in C57B6 mice. The ability to target and amplify within small vascularized tumors presents a distinct advantage of Salmonella typhimurium as a therapeutic tumor vector.

17.3. LOCALIZATION BY ELECTRON MICROSCOPY OF
SALMONELLA TYPHIMURIUM WITHIN HUMAN LUNG

A mouse was inoculated s.c. in the left shoulder region with 5 x 106 A549 cells. After 6 weeks the tumor was palpable and the animal was inoculated i.p. with 3 x 106 Salmonella typhimurium clone 72, for 66 hours. The animal was sacrificed and a portion of the tumor was homogenized and found to contain 1.6 x 109 Salmonella typhimurium/g. The central portion of the tumor was prepared for electron microscopy as follows: The portion of the tumor was cut into 1-2 mm3 pieces and fixed in 1/2 strength Karnovsky's fixative for 6 hours at 4 C, followed by washing in cacodylate buffer overnight. The tumor tissue was post-fixed with 1% Os04 and 1.501 potassium ferrocyanide in cacodylate buffer for 2 hours and embedded in Spurr's resin. Ultrathin sections were stained with uranyl acetate and lead citrate. They were photographed through a Zeiss 109 electron microscope. It should be noted for comparison purposes that the S.
typhimurium shown in the electron micrograph of Figure 12-B
appear similar to those shown previously in intestinal epithelial cells following an experimental infection of the mouse, Takeuchi, 1967, Am. J. Pathol. 50:109-1361.
Shown in Figure 12-B are numerous Salmonella typhimurium, denoted by arrows, in extracellular spaces as well as contained within a single cell, possibly a neutrophil, seen in the upper left. Also seen in the field are two unidentified cells that appear to be dying as indicated by the large intracellular space, along with cellular debris.

18. EXAMPLE: ATTENUATION OF SALMONELLA TYPHIMURIUM BY
MUTATION TO AUXOTROPHY
The studies below demonstrate that the reduced virulence of clone 72 (see, e.g., Section 15.2 above) is due to a Pur- phenotype. Further described are analyses of avirulent derivatives of clone 72 that were isolated as additional auxotrophic mutants, expressing in different combinations the phenotypes of Ade-, Ilv-, Arg-, Aro-, and Ura-.

18.1. MUTATION TO AUXOTROPHY
Clone 72 was examined for auxotrophic mutations and was found to have growth requirements for both adenine and vitamin B1, indicating a mutation(s) in the purine biosynthetic pathway (Pur-). An experiment was designed to test whether the ade- mutation could account for the observed attenuation of clone 72 described above. Populations of both wild type strain 14028 and clone 72 were mutagenized with UV
radiation and nitrosoguanidine as described in Section 7.1.
From the population of mutagenized strain 14028, three separate Pur- auxotrophic mutant clones were isolated and designated clones N, Q, and T. From the population of mutagenized clone 72, three separate Pur' revertant clones were isolated and designated clones R, U, and W.
C57B/6 mice were injected i.p. with 2 x 106 c.f.u.
Salmonella typhimurium of each of the strains obtained. The mice were allowed to eat and drink ad libitum and the cages were monitored for dead or moribund mice. Moribund animals (listless, cessation of drinking) were euthanized and counted with the other dead. After 10 or 30 days post-injection with bacteria the surviving animals were euthanized.
The results are shown in Table 20(B).
TABLE 20(B) MUTANTS OF SALMONELLA TYPHIMURIUM

Time of Death Survivors Survivors Strain Phenotype (Days S.D.) > 10 days > 30 days 14028 wild type 3.0 0.5 n.a. n. a.
72 superinfective, ade- 5.8 1.4 n.a. n.a.
R 72, Pur+ 3.9 0.4 n.a. n.a.
U 72, Pur+ 3.9 1.3 n.a. n.a.
W 72, Pur+ 4.1 0.9 n.a. n.a.
T 14028, Pur 6.8 1.5 n.a. n.a.
N 14028, Pur n.a. 4/8 n.d.

Q 14028, Pur" n.a. 5/8 n.d.
YS721 72, Ilv n.a. 10/11 6/11 YS7211 72, Ilv-, Arg' n.a. 8/8 7/8 YS7213 72, Ilv, Aro- n.a. 8/8 8/8 YS7212 72, IIv , Ura- n.a. 8/8 6/8 Results are the average SD for n = 8-12 animals n.a., not applicable; n.d., not done.

As shown in Table 20(B), Clone 72 was less virulent than the wild type strain 14028. However, 3 of 3 Pur' revertants of clone 72 (U, W, and T) expressed virulence similar to 14028. Conversely, 3 of 3 Pur- auxotrophic mutants isolated from strain 14028 (T, N, and Q) were less virulent than either 14028 or clone 72.
Isolation of additional auxotrophs from clone 72 produced even less virulent strains. For example, clone YS721-is an isoleucine-valine requiring (Ilv-) derivative of clone 72, and clone YS721 was significantly less virulent than clone 72. Similarly, auxotrophic derivatives of clone YS721 such as clones YS7211 (Arg-), YS7212 (Ura-), and YS7213 (Aro-) were all significantly less virulent than YS721 itself.

18.2. EVIDENCE THAT THE SUPERINFECTIVE

DISTINCT FROM ITS AUXOTROPHIC PURINE
REQUIREMENT
The various Salmonella typhimurium Pur and Pur' strains described above in Section 18.1 were assayed for their ability to infect human M2 melanoma cells in culture.
The in vitro infection assay employed was as described in Section 18.1.
The results are described in Table 20(C).
TABLE 20(C) PURINE MUTANTS OF SALMONELLA TYPHIMURIUM

Infecting Salmonella/
Strain Phenotype 106 melanoma x wild type cells/15'( S.D.) 14028 wild type 1.0 0.2 x 105 1.Ox 72 superinfective, ade' 9.8 0.7 x 105 9.8x R 72, Pur+ 5.9 1.4 x 105 5.9x U 72,Pur+ 1.1 0.2x106 11x W 72, Pur+ 1.1 0.3 x 106 11X
N 14028, Pur 1.9 0.5 x 105 1.9x Q 14028, Pur 1.5 1.0 x 105 1.5x T 14028, Pur 1.1 0.4 x 105 1.5x Results are the average SD for triplicate infections. The bacteria were cultured in LB
broth to 0.D.6= .600 prior to their dilution and use in the infection assays.
As shown in Table 20(C), Clone 72 displayed superinfectivity toward human M2 melanoma cells compared to wild type strain 14028. None of the 14028 Pur- derivatives differed significantly in its infectivity from strain 14028 itself, and all of the clone 72 Pur' derivatives expressed superinfectivity similar to clone 72 itself. The results demonstrate that the purine requirement exhibited by clone 72 which accounts for the reduced virulence of clone 72 in mice, is genetically separate from the superinfective phenotype of clone 72. These results demonstrate that neither mutation to nor reversion from purine auxotrophy effects expression of the superinfective phenotype characteristic of clone 72.

18.3. RETENTION OF THE SUPERINFECTIVE PHENOTYPE
BY ATTENUATED DERIVATIVES OF SALMONELLA

In the experiments below, the infectivity of certain auxotrophic derivatives of clone 72 described above in Section 18.1 was assessed in vitro. The phenotypes of the clones of Salmonella evaluated are shown in Table 20(B) in Section 18.1 above. Infectivity assays described in Section 10.1 were employed.
The results are presented in Table 20(D).
TABLE 20(D) INFECTIVITY OF SALMONELLA TYPH/MURIUM AUXOTROPHS
TOWARD HUMAN MELANOMA CELLS IN CULTURE
Infecting Salmonella/
Strain 106 melanoma cells/15 min x wild tvae 14028 (wild type) 4.3 x 104 1.Ox clone 72 4.4 x 105 10x clone YS721 3.2 x 105 7.4x clone YS721 1 2.0 x 105 4.7x clone YS7212 1.7 x 105 4.Ox clone YS7213 1.3 x 103 0.03x The results represent the average SD for 10-19 separate infections. The bacteria were grown in LB broth to O. D.600= 0.5 before being diluted prior to their use in the infection assays Salmonella typhimurium clones YS721, YS7211, and YS7212, though each somewhat less infective of M2 melanoma cells than clone 72, were nonetheless superinfective when compared to wild type strain 14028, indicating their partial retention of the superinfective phenotype. In contrast, clone YS7213 (Ade-, Ilv-, Aro-) was found to have greatly reduced infectivity, being about 30-fold less infectious toward M2 melanoma cells than the wild type strain 14028.

18.4. GROWTH OF PUR- AND URA- MUTANTS OF
SALMONELLA TYPHIMURIUM WITH NUTRITIONAL

Tumor extracts were prepared in the following manner: B16F10 melanoma tumor cells (5 x 105) were implanted s.c. into 68 week female C57B6 mice. After 3-4 weeks, the mice were sacrificed and the tumors removed aseptically and rapidly frozen, -20 C. A total of 51 g of frozen pooled tumors was thawed at 4 C and vigorously homogenized in 255 ml (5 vol) H2O in a capped Virtis tissue homogenizer in the cold for 1 hour. The resulting homogenate was made 100i with trichloracetic acid (TCA), placed on ice for 15 minutes, and centrifuged in a Beckman J21 centrifuge at about 20,000x g for 15 minutes at 4 C. Further procedures were conduced at room temperature. The clear, colorless supernatant fraction (300 ml) was retained and extracted by manual shaking for 1 minute with 1 volume (300 ml) anhydrous ether. Between extractions, the mixtures were allowed to settle and the upper phase (containing ether, extracted TCA, as well as ether-soluble compounds from the tumor extract) was removed by aspiration and discarded through approved environmentally-protective procedures. During 5 such extraction cycles, the pH of the water phase rose from a starting value of about pH 1 to a final value of pH 4-5, similar to that of distilled H201 indicating that the TCA had been effectively removed.
The water phase was bubbled with a stream of nitrogen for about 15 minutes, at which time the odor of ether had disappeared.
The solution was then filtered through a 0.2 micron filter, divided into aliquots and either used directly in the assays herein, or stored at -20 C for further use.
Wild type strain 14028, and its auxotrophic derivatives clone 72 (Pur, vitamin B1-) , and YS7212 (Ade-, vitamin B1-, Ilv-, Ura-) were grown overnight on a slant in 5 ml Luria broth (LB) at 35 C. The next day 0.1 ml of each culture was diluted into 10 ml of Medium 56 (0.037 M KH2PO41 0.06 M Na2HPO4, 0.02% MgSO4-7H201 0.2% (NH4) 2SO41 0.00101 Ca(N03)2 and 0.00005% FeSO4-7H20) supplemented with 0.2 g/ml vitamin B1, 33 g/ml adenine, 50 g/ml uracil, 83 g/ml isoleucine, 83 g/ml valine and 0.3% glucose, and grown on a rotor overnight at 37 C. The next day the cultures were collected by centrifugation and resuspended in plain Medium 56 (1 ml culture plus 9 ml Medium 56). Aliquots (0.25 ml) of these suspensions were then added to Medium 56 containing various supplements in the following manner:
A. Medium 56 plus glucose;
B. Medium 56 plus glucose, vitamin Bi, adenine, isoleucine, valine, and uracil; and C. Medium 56 and tumor extract (10%).
The bacteria were placed in a swirling H2O bath, 37 C, and growth as a function of OD600 was followed with a spectrophotometer. The starting optical densities for all of the cultures ranged from 0.005-0.07.
As demonstrated in Figure 15A-C, wild type strain 14028 was able to proliferate at about the same rates in all three of the media tested, including the most basic of the three, Medium 56 plus glucose. Unlike the wild type strain, neither clone 72 nor clone YS7212 was able to grow in Medium 56 plus glucose, indicative of their nutritional requirements originally observed through replicated plating on agar. In contrast both clone 72 and clone YS7212 were able to grow in Medium 56 supplemented with 10% tumor extract. Liver extracts prepared in the same manner were also able to support the growth of clones 72 and YS7212.
Although the inventors do not wish to be limited to a specific mechanism of action, since the growth state of auxotrophic strains of Salmonella is dependent upon the availability of nutrients, such auxotrophs would seem to have advantages as tumor vectors since the environment of the tumor could in theory provide such nutrients, for example in necrotic spaces or within actively dividing cells of the tumor. Thus, mutation of organisms such as Salmonella to auxotrophy not only reduces their virulence in vivo but also may provide a potential mechanism for their selective population and amplification within solid tumors.

18.5. PROLIFERATION OF PUR- AND URA- MUTANTS OF

MELANOMA CELLS IN CULTURE
In this Section it is demonstrated that the internal milieu of cultured M2 melanoma cells also can also supply the auxotrophic requirements of these clones, since both clone 72 and clone YS7212-were able to undergo several rounds of division once they invaded M2 melanoma cells cultured under aerobic conditions.
Salmonella typhimurium clones 72 and YS7212 were grown to O.D.600=0.8, or about 109 c.f.u./ml. The two strains were then added at 106C.f.u./ml culture media of human M2 melanoma cells as described above in Section 7.2. 15 minutes after infection with Salmonella, the eukaryotic cell cultures were rinsed with fresh medium and medium containing gentamicin (10 g/ml) was added. At hourly intervals over a 6 hour period, cultures were processed as described in Section 7.2 for quantitation of Salmonella/106 melanoma cells.
In addition, control flasks without melanoma cells but with bacteria were processed side-by-side with the experimental flasks containing melanoma cells.
The results are shown in Figure 15-D. Control flasks with Salmonella but without melanoma cells showed no viable bacteria over the 6 hour period, demonstrating that the wash procedure coupled with gentamicin treatment successfully eliminated all viable bacteria not protected by location within animal cells. In contrast, in the presence of M2 melanoma cells, Salmonella typhimurium clones 72 and YS7212 each increased significantly in number over the 6 hour period with doubling times of about 2 hours for each strain.
Phase and electron microscope analyses demonstrated that M2 melanoma cells were able to compartmentalize infecting Salmonella within vacuoles. The results indicate that the net rate of growth of Salmonella within the melanoma cells was a steady-state function, reflecting the ability of the melanoma cells to both stimulate growth of the auxotrophs through the supply of nutritional requirements, as well as to suppress the growth of the auxotrophs through anti-bacterial mechanisms.
18.6. BIODISTRIBUTION OF AUTOTROPHIC ATTENUATED
STRAINS OF SALMONELLA TYPHIMURIUM IN

These studies demonstrate the ability of clones YS721, YS7213, YS7211 and YS7212 to target tumors and proliferate within the tumor in vivo.
C57B6 6-8 week old female mice were inoculated s.c.
(left flank) with 2.5-5.0 x 105 B16F10 mouse melanoma cells.
When the tumors reached about 0.5 g (14-16 days post-tumor inoculum), the animals were further inoculated i.p. with the indicated strains of S. typhimurium. The bacterial inoculum was 4 x 105 cfu/mouse for strains 14028 and 72 and 2-4 x 106 cfu/mouse for strains YS721, YS7211, YS7212 and YS7213.
After 40 and 96 hours post-inoculation of bacteria, the mice were sacrificed, the tumors and livers were removed aseptically, rinsed with sterile NaC1 (0.9%), weighed, and homogenized in LB broth at a ratio of 5:1 (vol:tumor wt).
Bacteria were quantitated by plating the homogenates onto LB
plates, incubating overnight at 37 C, and counting bacterial colonies. The results presented in Table 20(E) represent the average SD for n=4-7 animals.
TABLE 20(E) BIODISTRIBUTION OF WILD TYPE AND ATTENUATED STRAINS OF

TUMORS
Salmonella/g tissue:
Strain Tumor Liver Tumor: Liver A. 40 hrs post-inoculation of bacteria 14028 6.5 6.8 x 109 2.4 2.8 x 107 270:1 72 1.7 1.2 X 109 1.9 2.3 X 105 9000:1 YS721 8.7 3.1 x 108 4.2 3.6 x 106 210:1 YS7211 3.3 3.0 x 107 8.1 8.4 x 105 41:1 YS7212 3.9 7.3 x 107 1.1 0.8 x 106 35:1 YS7213 1.5 2.8 x 108 4.0 3.1 x 105 375:1 B. 96 hrs post-inoculation with bacteria 14028 moribund/dead 72 moribund/dead YS721 3.2 1.5 x 109 4.7 6.9 x 106 680:1 YS7211 1.6 2.2 x 109 6.3 9.9 x 106 253:1 YS7212 1.1 7.4 x 109 5.1 8.6 x 105 2200:1 YS7213 1.3 2.5 x 109 2.2 6.9 x 105 5900:1 Each of the strains tested was able to target the tumor tissue and replicate to varying degrees within the tumor, as evidenced by the finding that in all cases the tumors contained 10-1000 times more Salmonella typhimurium than were first inoculated. Further, in all cases the tumor:liver ratio of bacteria/g tissue was at least 35:1 and in some cases approached 104. The tumors analyzed in the studies presented in Table 20(E) ranged in weights from 0.5-2.0 g. Of all the conditions and strains tested, clone 72 exhibited the highest tumor:liver ratio 40 hours post inoculation. Further, Salmonella typhimurium strain 14028, as well as clone 72 and its derivatives were also able to target and amplify within larger B16F10 melanoma tumors of, for example, 4-8 g. In addition, as shown in Section 10.3.2 and Table 12A, clone 72 can target and amplify within human solid tumors as small as 40 mg.
However, both clone 72 and the wild type strain 14028 were highly virulent toward C57B6 mice, especially mice bearing tumors. For example, C57B/6 mice bearing B16F10 melanomas injected with strains 1428 and 72 had average survival times of 2.1 0.4 days (n=6) and 4.7 0.5 days (n=9) post-inoculation of bacteria respectively. The biodistribution of these strains was thus not measured at 96 hours. Likewise, clone YS721, though attenuated compared to 14028 and clone 72, was nonetheless virulent in melanoma-bearing mice. For example, B16F10 melanoma-bearing C57B6 mice injected with clone YS721 had an average survival time of 8.1 0.2 days (n=11). Salmonella clones YS7211, YS7212 and YS7213, the least virulent of those examined, each displayed densities of greater than 109 cfu bacteria/g tumor 96 hrs post-inoculation with tumor:liver ratios of 253:1, 2200:1, and 5900:1 respectively.

18.7. PHENOTYPIC STABILITY FOLLOWING INCUBATION
OF SALMONELLA TYPHIMURIUM AUXOTROPHS IN
TUMOR-BEARING MICE
Genetic reversion of an auxotrophic phenotype could in theory result in an increase in virulence of the previously attenuated bacteria. Therefore, the stabilities of the auxotrophic phenotypes of the strains YS7211, YS7212 and YS7213 were tested following incubation of the bacteria in tumor-bearing mice.
Salmonella typhimurium obtained from the homogenates of livers and tumors of animals 40 hours post-inoculation of either YS7211, YS7212 or YS7213 were picked from LB plates and replicate plated onto minimal media agar plates supplemented with nutritional additives in different combinations. The supplements were isoleucine, valine, adenine/vitamin B1, arginine, uracil, aromatic amino acids, and glucose. For each of the three strains, 50/50 of the bacterial clones recovered from the tumor and liver homogenates displayed the expected phenotypes of the strain originally inoculated, indicating that in this experiment the strains were genetically stable enough not to revert substantially under the conditions tested.
However, it should be noted that the auxotrophic strains employed were not absolutely stable throughout these studies. In some cases genetic revertants were observed, most notably in the YS7211 strain wherein revertants from Arg-to Arg' were observed. For example, in a tumor-bearing mouse inoculated 96 hours earlier with clone YS7211 bearing a thymidine kinase-containing plasmid, 50 out of 50 bacteria isolated from the liver were found to be Pur-, Ilv- and Arg+, indicating that reversion and selective growth of the reverted organisms had occurred within the mouse. The finding that the auxotrophic phenotypes of clones YS7211, YS7212 and YS7213 were relatively stable in mice was supported by the long term survival of mice inoculated with these strains.

18.8. SUPPRESSION OF TUMOR GROWTH AND INCREASED

INOCULATED WITH AUXOTROPHIC MUTANTS OF
SALMONELLA TYPHIMURIIJM
C57B6 female mice, 5-7 weeks old, were inoculated s.c. in the left shoulder region with 5 x 105 B16F10 melanoma cells grown in culture. On the 8th day following inoculation of tumor cells, the mice were further inoculated i.p. with 2-4 x 106 c.f.u. of Salmonella typhimurium strains YS721, YS7211, YS7212 or YS7213. Tumor growth was assessed with periodic caliper measurements of tumor length, width and height, and computed as tumor volume in mm3. Results of tumor growth, shown in Figure 16A-D represent the averages SD for 5 animals/group with 5/5 animals surviving. After the point at which one or more animals died within a group, the average tumor sizes of the surviving animals were no longer shown when the data were plotted as shown in Figure 16A-D.
All tumor measurements were stopped after 33 days post implantation of tumor cells, even though 5/5 tumor-bearing animals treated with clone YS7211 were still alive at this time. The animals were allowed to eat and drink add libitum. Twenty-three days (Experiment #1) or 10 days (Experiment #2) following inoculation of bacteria, both control and bacteria treated mice were given Baytril' (enrofloxacin, 0.2 mg/ml drinking water) and maintained on this antibiotic for a total of 2 weeks. In Experiment #1 the times at which the mice became moribund (listless, cessation of drinking) or died, were noted. The results are presented in Table 20(F) as the average survival SD for the conditions tested. In Experiment #2 animals were sacrificed when the tumor reached 4 g and listed with the other dead as described in Experiment #1. The two different methods for assessing survival accounted for a somewhat shorter survival time for control animals in Experiment #2 (26 days) as compared to Experiment #1 (28 days).

TABLE 20(F) WITH SALMONELLA TYPHIMURIUM

Time of death post tumor cell inoculation:
Treated/Control Strain (Days SD) (T/C) Control Expt 1 28 2 1.0 (no bacteria) 2 26 3 1.0 YS7211 1 36 9 1.3 2 41 10 1.6 YS7213 1 36 5 1.3 2 38 6 1.5 YS7212 1 51 7 1.8 2 55 3 2.1 3 5 The results represent the average SD for 5 animals.
Figure 16A-D shows the average SD tumor volumes (mm3) versus time post inoculation of 5 x 105 B16F10 melanoma cells s.c. into C57B/6 mice. All four clones of Salmonella, namely clones YS721, YS7211, YS7212 and YS7213, elicited suppression of tumor growth in the animals. Clone YS721, attenuated through Ade- and Ilv- auxotrophy, was nonetheless toxic to tumor-bearing mice and resulted in no prolongation of survival compared to control tumor-bearing animals receiving no bacteria. Whereas the death of control animals was clearly due to very large tumor masses (4-8 g), the death of tumor-bearing animals inoculated with clone YS721 appeared to be a result of bacterial toxicity since the tumor burden in these animals was quite small and not life-threatening in itself. The tumors ranged from non-palpable to less than 0.5 g at the time of death.
In contrast, treatment of tumor-bearing mice with clone YS7211, YS7212 and YS7213, each less virulent than clone YS721, resulted in significant enhancement of survival in addition to suppression of tumor growth. The degree of suppression of tumor growth by the individual Salmonella clones, as seen in Figure 16A-D, correlated with their abilities to elicit enhanced survival, as seen in Table 20(F). The average time for tumors to reach 1 g (1000 mm3) was about 18 days for control animals, 31 days for animals treated either with YS7213 and YS7211 and 45 days (extrapolated) for YS7212. This corresponded to average survival times for 26 days for control tumor-bearing animals, compared to 38, 41, and 55 days for animals treated with clones YS7213, YS7211 and YS7212.
Thus, among the attenuated strain of Salmonella tested, the order of efficacy for suppression of tumor growth and prolongation of survival was YS7212>YS7211>YS7213.
Earlier treatment with an antibiotic, enrofloxacin, i.e., 10 days as compared to 23 days post-inoculation of bacteria, increased the survival time for tumor-bearing animals inoculated with Salmonella, but not that of control animals.
18.9. ANTI-TUMOR ACTIVITY OF AUXOTROPHIC
SALMONELLA TYPHIMURIUM EXPRESSING
CYTOSINE DEAMINASE
Experimental metastasis model of B16F10 was set up by injecting 1 x 105 cells into C57B/6 mice via the lateral tail vein on Day 0. Aliquots of 0.2 ml bacterial suspension of YS7212 carrying the cytosine deaminase expression construct (see Figure 4E for the CD construct) (approximately 1 x 107 CFU/ml) were injected intraperitoneally into mice on Day 5. 5-Flourocytosine (5-FC), at 0.4 ml aliquots dissolved in PBS at 10 mg/ml (final dose: 200 mg/kg), was injected into mice intraperitoneally on Day 7. Death of animals was recorded daily. Results are presented in Figure 17.
Figure 17 clearly demonstrates that combination of CD and 5-fluorocytosine prolong the survival Salmonella expressing animals bearing B16F10 lung metastases.

19. EXAMPLE: ATTENUATION OF SALMONELLA TYPHIMURIUM
THROUGH MUTATION IN LIPOPOLYSACCHARIDE
BIOSYNTHESIS
Several mutant strains of Salmonella typhimurium and E. coli have been isolated with genetic and enzymatic lesions in the LPS pathway (Raetz, 1993, J. Bacteriol.
175:5745-5753). One such mutant, the firA- mutation is within the gene that encodes the enzyme UDP-3-O(R-30 hydroxymyristoyl)-glycocyamine N-acyltransferase, that regulates the third step in endotoxin biosynthesis (Kelley et al., 1993, J. Biol. Chem. 268:19866-19874). Salmonella typhimurium and E. coli strains bearing this type of mutation produce a lipid A that differs from wild type lipid A in that it contains a seventh fatty acid, a hexadecanoic acid (Roy and Coleman, 1994, J. Bacteriol. 176:1639-1646) and has decreased lipid A 4' kinase activity.
A firA- mutant was investigated for its ability to induce TNFa production by human monocytes as well as its ability to target solid tumors in mice.
19.1. ABILITY OF SALMONELLA TYPHIMURIUM firA- TO
INDUCE TNF-a PRODUCTION BY HUMAN BLOOD
MONOCYTES

Salmonella typhimurium strain SH5014 and its firA-derivative SH7622 are described in Hirvas et al., 1991, EMBO
J. 10:1017-1023. The genotypes of these strains are as follows:
strain SH5014 ilv-1178 thr-914 his-6116 metA22 metE551 trpB2 xyl-404 HI-b H2-e,n,x flaA66 rpsLl20 rfaJ4041;
strain SH7622 ilv-1178 thr-914 his-6116 metA22 metE551 trpB2 xyl-404 H1-b H2-e,n,x flaA66 rpsLl20 rfaJ4041, ssc-l(firA`s) A derivative of Salmonella typhimurium firA- strain SH7622 was picked, designated SH7622-64, and used as the firA-strain for the experiments in this section as well as in Section 19.2 below. SH7622-64 was selected for its supersensitivity to the antibiotic novobiocin and temperature-sensitive growth, characteristics of the firA-SH7622 strain.
LPS was extracted from Salmonella typhimurium strain 14028 and its derivatives clone 72, clone YS7212, and clone YS7213; as well as strain SH5014 and its firA-derivative, clone SH7622-64, as follows: The bacteria were grown in 500 ml LB broth to O.D.600=0.9 or about 2 x 109 cfu/ml. They were collected by centrifugation, and the pellets, containing about 109 cfu/ml. They were collected by centrifugation, and the pellets, containing 1012 bacteria, were drained and stored frozen at -20 C. To extract LPS, the pellets were resuspended in 18.3 ml H20, and 15 ml redistilled phenol was added (H20:phenol, 55:45, vol/vol). The mixtures were placed in a shaking water bath at 69-70 C, for 1 hour producing a monophasic mixture, and then cooled on ice. On cooling the mixture separated into a phenol phase containing mainly proteins, and a water phase containing lipopolysaccharide and nucleic acid (Galanos, C., Luderitz, 0., and Westphal, 0., 1969). The water phase was lyophilized to dryness and the white fluffy lyophilized material was used as the source of LPS. The LPS was weighed and dissolved in H2O at 1 mg/ml, as stock for dilution in the incubations with human macrophages described below.
Human macrophages were prepared as follows and all procedures were at room temperature: Blood (60 ml) was collected from a healthy human volunteer into a heparinized syringe. The blood was layered in 7 ml aliquots over 4 ml of Isolymph' (density - 1.077 g/ml; 9.0 g sodium diatrizoate and 5.7 g Ficoll 400x"/100 ml H20; Pharmacia Fine Chemicals, A.B.
Uppsala, Sweden) in 15 ml Corning Plastic Centrifuge tubes, centrifuged at 2000x g for 45 minutes. The red blood cells pelleted through the Isolymph', neutrophils and other cells sedimented in a discrete band above this interface, and above the lymphocyte/macrophage band was serum, visible by its yellow color. The serum from each tube was removed by pipette, pooled in a total volume of about 30 ml and saved for supplementation into the culture media as described below. The lymphocyte/macrophage bands were pooled in a total volume of about 15 ml diluted with 40 ml RPMI 1640 culture medium, and centrifuged at 1000x g for 5 minutes.
The cloudy supernatant was discarded and about 0.2 ml of pelleted white cells was obtained. The cells were resuspended with 50 ml RPMI 1640 culture medium supplemented with 15% human serum (described above), penicillin (100 units/ml) and streptomycin (100 g/ml). The recovery of viable lymphocytes and macrophages from 60 ml whole blood was determined by hemocytometer counting to be about 7 x 107 cells. Together the cells, lymphocytes and monocytes were distributed into 24 well Corning Tissue Culture Plates at 0.5 ml/well, and incubated in a gassed humidified incubator at 37 C for 15 hours.
The next day the cultures were rinsed twice with serum-free RPMI 1640 containing antibiotics. Between each rinse, the cultures were incubated about 1 hour in the 37 C
incubator to facilitate removal of lymphocytes and other non-adherent cells. Adherent cells were found to be mostly, if not all derived from blood monocytes, i.e., macrophages that had differentiated from their blood monocyte state by virtue of attachment to the culture dish. For example, in a histochemical assay to determine the percentage of macrophages in the adherent population of cells 48 hours post-plating into culture, the population was found to be essentially 100% positive for expression of the enzyme non-specific esterase, a marker commonly used to distinguish monocytes and macrophages from lymphocytes and other cell types. Results indicated that most if not all of the cells employed in the LPS challenge described below were of monocyte origin.
After the second rinse described above, serum-free, antibiotic containing RPMI 1640 supplemented with LPS at the concentrations indicated was added to the cells, and the cultures were placed in a gassed, humidified incubator at 37 C overnight. After 20 hours, the well plate cultures were centrifuged in a Beckman GS-15 centrifuge at 8000x g for 10 minutes, and the supernatants were removed and assayed for TNF-a content using the QUANTIKINE'm Human TNF-a Immunoassay Kit #DTA50 (R&D Systems, Minneapolis, MN). TNF-a production as pg/ml by human macrophages is plotted as a function of pg/ml bacterial LPS added to the culture medium and shown in Figure 18.
Strain 14028 and its derivatives clone 72, clone YS7212, and clone YS7213, as well as strain SH5014, all induced TNF-a production by human macrophages at concentrations of LPS in a dose-dependent fashion.
Concentrations of LPS from each of these strains as low as 100 pg/ml (0.1 ng/ml) were stimulatory to TNF-a production, and increasingly stimulatory at concentrations of 103 pg/ml and 10 pg/ml, inducing TNF-a production by macrophages to levels of 600-800 pg/ml. The levels of TNF-a induced by the LPS were similar to the circulating levels of TNF-a found in patients with septic shock syndrome as well as in human volunteers injected with E. coli LPS (Morrison et al., 1994, ASM News 60:479-484). In contrast, LPS from firA- strain SH7622-64 was far less stimulatory to TNF-a production by the macrophages, and was detected only at concentrations of 104 pg/ml. Thus, on a dose response comparison, LPS from strain SH7622-64 was only about 11i as effective in stimulating macrophage TNF-a production when compared to LPS
from the fix-A' parental strain SH5014. Furthermore, the results demonstrate that strains YS7212 and YS7213 each produced LPS similar to wild type strain 14028 LPS, as assessed by stimulation of human macrophages.

19.2. TUMOR TARGETING BY SALMONELLA TYPHIMURIUM
BEARING THE firA- MUTATION
M27 mouse lung tumor cells or B16F10 mouse melanoma cells (5 x 105) were implanted s.c. in C57B6 mice. When the tumors were palpable, SH7622-64 grown in LB broth at 37 C to a density of about 109 cfu/ml (0D600=0.8) . Aliquots of 5-10 x 106 cfu were removed and inoculated into tumor bearing mice.
At 48 hrs (M27 lung) and 96 hrs (B16F10 melanoma) post-inoculation of bacteria the animals were sacrificed, the tumors and livers removed, weighed and homogenized in LB
broth at a ratio of 5 ml/g tissue. Homogenates were quantitated for bacteria by serial dilutions on LB agar plates. Results are presented in Table 20(G).

TABLE 20(G) TUMOR LOCATION BY SALMONELLA TYPHIMURIUM BEARING THE firA-MUTATION FOR LIPOPOLYSACCHARIDE BIOSYNTHESIS

Salmonella/g tissue:
Primary Tumor Tumor Liver Tumor: Liver M27 lung 2.9 x 106 -0- n.a.
B16 3.2 x 105 1 x 102 3200:1 The results are derived from single animals.
As shown in Table 20(G), strain SH7622-64 was able to locate within both the B16F10 melanoma and the M27 lung tumors when'inoculated i.p. into mice. These results, in combination with those in Section 19.1 which show that LPS
from this particular firA- strain was greatly suppressed in its ability to induce TNF-a in human macrophages, demonstrate that Salmonella attenuated through a mutation in endotoxin biosynthesis can be useful as tumor vectors in vivo.

20. EXAMPLE: TUMOR-SPECIFIC ACCUMULATION OF CLONES

CARCINOMA
This example demonstrates that auxotrophic mutant Salmonella clones YS721 and YS7211 locate to lung carcinoma.
The experimental model of the Lewis lung carcinoma was set up by injecting 5x105 cells into C57B/6 mice subcutaneously on Day 0. Aliquots of 0.2 ml bacterial suspension (approximately 1 x 107 CFU/ml) were injected intraperitoneally into mice on Day 14. On Day 16, the tumors and livers were harvested and homogenized and bacterial counts determined by plating serial dilutions. Results of the relative distribution are shown in Table 20(H).
TABLE 20(H) TUMOR SPECIFIC ACCUMULATION OF
CLONES YS721 and YS721 1 IN MICE

Strain No. pathogens No. pathogens Ratio:
/g Liver /g Tumor tumor/liver YS721 9.8 x 106 4.7 x 1010 4.8 x 103 4.3 x 105 3.2 x 1010 7.3 x 104 1.1 x 106 1.4x 109 1.3x 103 1.6x106 1.0x1012 6.2x105 3.0 x 104 2.3 x 109 7.7 x 104 YS721 1 1.4 x 104 2.6 x 1010 1.9 x 106 1.9 x 105 2.7 x 106 1.4 x 103 2.3 x 105 6.0 x 10" 2.6 x 106 1.Ox 106 5.Ox 10" 5.Ox 105 Extremely high levels of bacteria were localized to these tumors, as well as others indicating that the auxotrophic mutations retain tumor specific accumulation of bacteria for a spectrum of tumor models.

21. EXAMPLE: TREATMENT OF B16F10 MELANOMA
METASTATIC TUMORS
Metastases constitute one of the major problems for treatment of solid tumors. While larger tumors can be detected and removed surgically, smaller metastases constitute the untreated reservoir which is frequently the cause of death. Therefore, an effective cancer therapeutic should be effective against metastatic tumors.
An experimental metastasis model of B16F10 was set up by injecting 1x105 CELLS into C57B/6 mice via the lateral tail vein on Day 0. Aliquots of 0.2 ml YS7211/p5-3 and YS7212/p5-3 (YS7211 and YS7212 each carrying the HSV
thymidine kinase expression plasmid) bacterial suspensions (approximately 1 x 107 CFU/ml) were. injected intraperitoneally into mice on Day 5. Ganciclovir, at 0.1 ML aliquots dissolved in PBS at 22 mg/ml (final dose: 100 mg/kg), was injected into mice intraperitoneally on Day 7. Tumor progression was monitored by periodic sacrifice and examination of the lungs. At day 28, all the animals were sacrificed and the normal and tumor-bearing lungs weighted.
Figure 19 clearly demonstrates that animals inoculated with YS7212 carrying the HSV thymidine kinase gene (YS7212/p5-3) and further treated with GCV show reductions in the number and extent of B16F10 lung metastases.
22. EXAMPLE: DIAGNOSIS OF TISSUE BIOPSIES FOR
MELANOMA USING SALMONELLA TYPHIMURIUM
Diagnosis of melanoma according to the methods of the present invention can be performed using, for example, Salmonella typhimurium as follows: A portion of a biopsied specimen suspected of melanoma is minced with scissors in tris-buffered saline (TBS) and then incubated in Ca"/Mg"- -free saline containing trypsin, collagenase, and EDTA (Sigma Chemicals) for 60 minutes at 37 C to dissociate the tissue into individual cells. The cells are then rinsed free of the dissociation enzymes by centrifugation. The cells are resuspended in 1 ml DMEM/10% FBS and added to a 24 well Corning tissue culture chamber containing cover slips in the wells. The cells are then incubated in a gassed (5% C02/95%
air) humidified incubator for 3 hours at 37 C to allow for attachment to the cover slip.
After attachment of the biopsied cells is achieved, an attenuated, super-infective, melanoma-specific strain of Salmonella typhimurium (106-107 c. f .u. /ml) is added. The bacteria and biopsied cells are incubated together at 37 C
for 15 minutes to allow for infection of melanoma cells by the S. typhimurium, and the cells are then rinsed with TBS to remove non-infecting bacteria. The cells are then permeablized with 0.01% saponin in 3% bovine serum albumin for 5 min, stained for DNA for 10 minutes with 2.5 mg/ml 4'-6 Diamidino-2-phenyhndole (DAPI) and saponin (.01%) in TBS, washed with TBS, mounted in Mowiol (Calbiochem) containing 1, 4-Diazabicyclo (2,2,2) octane (DABCO, Kodak) and observed by phase and fluorescence microscopy. The presence of DAPI-stain in the cytoplasm of the biopsied cells would indicate that they were melanoma cells, i.e., cells that were infected by the melanoma-specific S. typhimurium are melanoma cells rather than melanocytes.

23. EXAMPLE: MELANOMA TUMOR TARGETING BY LISTERIA
MONOCYTOGENES
This example demonstrates that Listeria monocytogenes targets to and proliferates in tumor cells when administered to melanoma bearing animals.
C57B/6 mice were inoculated s.c. in the left flank with 5 x 105 B16F10 melanoma cells. When the tumors reached about 1-2 g (16 days post implantation of tumor cells) the animals were inoculated i.p. with 7 x 105 cfu of Listeria monocytogenes wild type strain 43251. Prior to inoculation into mice, the Listeria culture was grown overnight in LB
media to an OD600 of 0.25. At the times indicated animals were sacrificed and the tumors and livers were removed, homogenized and quantitated for bacterial numbers by plating serial dilutions onto L.B. plates.
Tumors were analyzed at 24, 48, and 96 hours post-inoculation of Listeria monocytogenes. Results are shown in Table 20(I).
TABLE 20(I) Time L. monocytogenes/g tissue:
Post-inoc. Tumor Liver Tumor: Liver 24 hrs 1.5 1.4 x 103 8.0 6.1 x 104 1:5 48 hrs 6.3 8.5 x 102 1.3 1.0 x 105 1:210 96 hrs 5.2 6.8 x 105 5.7 9.0 x 105 1:1 The results represent the average SD of triplicate determinations.

As shown in Table 20(I) it was found that the levels of bacteria within the tumors rose about 100 fold during this time period, indicating that wild type Listeria monocytogenes can target tumors and proliferate within them.
Listeria monocytogenes strain 43251 was virulent in the C57B/6 mice, causing death about 5 days pot-inoculation i.p.
of 7 x 105 cfu.

24. EXAMPLE: LEISRMWXA AMAZONENSIS SHOWS TUMOR CELL
SPECIFICITY

24.1. LEISEMANIA AMAZONENSIS SPECIFICALLY
ATTACHES TO HUMAN MELANOMA CELLS IN VITRO
Leishmania amazonensis trypomastigotes are regarded as being highly biospecific, in that they are unable to infect virtually any cell types other than macrophages.
Since human melanomas are known to express some macrophage-like traits it was determined whether Leishmania amazonensis would be able to enter into human melanoma cells in culture.
Leishmania amazonensis promastigotes were grown in Schneider's Drosophila media (GIBCO BRL) containing 15a heat-inactivated fetal calf serum at 24 C until the parasites were in late log phase (usually 3 to 4 days). Animal cells used in the L. amazonensis infection assays were a mouse melanoma cell line which forms non-metastatic tumors when injected into C57B6 mice (B16/F1), two human metastatic melanoma cell lines (M2 and M2-A7, and as a negative control human foreskin fibroblasts, HFF. These different cell types were grown on glass coverslips in 24 well plates or on plastic Lab-Tec (Nunc) slides in MEM culture medium with 10% fetal calf serum, for HFF cells; Ham's F10 medium with 10% horse serum for B16/F1 cells; and DMEM with 10% fetal calf serum buffered with 10 mM HEPES, for M2 and M2-A7 cells.
The Leishmania parasites were pre-incubated for about one hour with 5% normal human serum and the cultured cells were infected with 0.5 to 5.0x106 parasites/ml for about two hours at 32 C. After incubation the cells were washed twice with phosphate buffered saline (PBS) and fixed with 3%
paraformaldehyde for about 30 minutes at about 4 C. An anti-Leishmanial antibody was incubated with the fixed cells at a normal working dilution (1: 100,000) in PBS with 3% bovine serum albumen (BSA) for about one hour. After washing, a fluorescent-conjugated anti-mouse antibody (Boehringer Mannheim) was incubated with the cells at a normal working dilution (1:500) for about one hour and then washed from the cells. The cells were then permeabilized with 0.02% Saponin (Sigma, a detergent used to remove lipids, thereby allowing penetration by antibodies) in Tris buffered saline (TBS) for 10 min and stained for DNA with 5.0 mg/ml DAPI stain (Sigma) in TBS with 0.02% Saponin. The cells were washed with TBS
and mounted on glass slides using Mowiol (CalBiochem) with DABCO (Kodak, a compound that sustains fluorescent emissions). The presence of internalized parasites was determined by failure to react with an anti-Leishmania monoclonal antibody in the absence of host cell membrane permeabilization.
Observations made immediately after addition of the live parasites using an inverted phase microscope showed that among these cell lines, motile parasites were immediately adherent only when they encountered the metastatic melanoma cells, suggesting that the metastatic cells may possess an appropriate receptor for Leishmania. These results are shown in Figure 13.
To determine whether or not the Leishmania parasites were internalized, M2 human melanoma cells were grown, infected with Leishmania and fixed as described above with 3% paraformaldehyde at 4 C for about 30 minutes without permeabilization, washed and immunostained with a monoclonal antibody directed toward a Leishmania surface protein, followed by a rhodamine-conjugated anti-mouse antibody.
After washing in Tris-buffered saline (TBS), the cells were permeablized with 0.01% saponin in 3% BSA for 5 minutes, and stained for DNA 10 minutes with 2.5 mg/ml 4'-6 Diamidino-2-phenylindole (DAPI) and 0.01% saponin in TBS, washed with TBS, mounted in Mowiol (Calbiochem) containing 1, 4-Diazabicyclo (2,2,2) octane (DABCO, Kodak) and observed by phase and fluorescence microscopy. This procedure detects all parasites and distinguishes between those which are internalized (inaccessible to antibody staining in non-permeabilized cells), and those which are attached but not internalized.
Parasites were internalized by M2 cells (data not shown). Internalization was estimated to occur in 3% of the melanoma cells. These findings demonstrate that a) live Leishmania parasites were able to enter the melanoma cells, and b) possibly only a sub-population of the melanoma cells were involved in the process.

24.2. LYSOSOMAL FUSION FOLLOWS INTERNALIZATION
OF LEISHMANIA BY MELANOMA
In the normal course of invasion of macrophages by Leishmania, lysosomes fuse with the phagosome. To determine whether or not this also occurs when Leishmania invade melanoma cells, parasites were co-localized with a lysosomal glycoprotein (lgp) marker. The cells were grown, infected and fixed as described above except the cells were immunostained with a monoclonal antibody directed against a human lysosomal glycoprotein, LAMP-1, followed by a rhodamine-conjugated antimouse antibody. After washing in TBS, the cells were stained for DNA 10 minutes with 2.5 mg/ml DAPI, washed with TBS, mounted in Mowiol containing DABCO and observed by phase and fluorescence microscopy. Parasites co-localizing with LAMP-1 are shown in Figures 14A-C. Co-localization corroborates the internalization of the parasite and demonstrates that the process of lysosomal fusion occurs when Leishmania is internalized into the melanoma cells. In summary, Leishmania amazonensis in its wild type state shows invasion ability for human melanoma cells that has been heretofore unreported.

25. EXAMPLE: DIAGNOSIS OF MELANOMA IN HUMAN TISSUE
BIOPSIES USING LEISEMANIA AMAZONENSIS
Diagnosis of melanoma according to the methods of the present invention can be performed using, for example, Leishmaina Amazonensis as follows: A portion of a biopsied specimen suspected of melanoma is minced with scissors in tris-buffered saline (TBS) and then incubated in Ca+`/Mg"-free saline containing trypsin, collagenase, and EDTA at 37 C for 60 minutes to dissociate the tissue into individual cells.
The cells are then rinsed free of the dissociation enzymes by centrifugation. The cells are resuspended in 1 ml DMEM/10%
FBS and added to a 24 well Corning tissue culture chamber containing cover slips in the wells. The cells are then incubated in a gassed, 5% CO2, humidified incubator at 37 C
for about three hours to allow for attachment to the cover slip.
After attachment of the biopsied cells is achieved, a melanoma-specific strain of Leishmania amazonensis promastigotes which has been isolated according to the methods of the present invention is added. The parasites and biopsied cells are incubated together at 37 C for about two hours to allow for infection of melanoma cells by the Leishmania amazonensis and the cells are then rinsed with TBS
to remove non-infecting parasites. The cells are then permeablized with 0.010i saponin in 3% bovine serum albumin for five minutes, stained for DNA with 2.5 mg/ml 4'-6 Diamiclino-2-phenylindole (DAPI) and saponin (.01%) in TBS
for 10 minutes, washed with TBS, mounted in Mowiol (Calbiochem) containing 1,4-Diazabicyclo (2,2,2) octane (DABCO, Kodak) and observed by phase and fluorescence microscopy. The presence of DAPI-stain in the cytoplasm of the biopsied cells would indicate that they were melanoma cells.
26. EXAMPLE: DIAGNOSIS OF HUMAN TISSUE BIOPSIES
FOR MELANOMA USING MYCOBACTERIUM AVIUM
Mycobacterium avium were found associated with the melanoma cells but not with the normal melanocytes. This discriminatory ability for melanoma cells demonstrates the ability of Mycobacterium avium as a vector in the diagnosis and treatment of melanoma. Diagnosis of melanoma using Mycobacterium avium is as follows: A portion of a biopsied specimen suspected of melanoma is minced with scissors in tris-buffered saline (TBS) and then incubated in Ca"/Mg"-free saline containing trypsin, collagenase, and EDTA at 37 C for 60 minutes to dissociate the tissue into individual cells.
The cells are then rinsed free of the dissociation enzymes by centrifugation. The cells are resuspended in 1 ml DMEM/10%
FBS and plated onto 12mm glass cover slips in 24 well plates with 1x105 cells per well. The cells are then incubated in a gassed 5% C02, humidified incubator 37 C for 3 hours to allow for attachment of the cells to the cover slip.
After attachment of the biopsied cells is achieved, a melanoma-specific strain of Mycobacterium avium which has been isolated by the methods of the present invention is added. The bacteria and biopsied cells are incubated together at 37 C for 15 minutes for infection of melanoma cells by the Mycobacterium avium. The cells are then rinsed with TBS to remove non-infecting bacteria. The cells are then permeablized with 0.01% saponin in 3% bovine serum albumin for five minutes, stained for DNA with 2.5 mg/ml 4'-6 Diamidino-2-phenylindole (DAPI) and saponin (.01%) in TBS for minutes, washed with TBS, mounted in MowiolTM (Calbiochem) containing 1,4-Diazabicyclo (2,2,2) octane (DABCO' ', Kodak) and observed by phase and fluorescence microscopy. The 5 presence of DAPI-stain in the cytoplasm of the biopsied cells would indicate that they were melanoma cells.

27. DEPOSIT OF MICROORGANISMS
The following microorganisms were deposited with 10 the American Type Culture Collection (ATCC), Rockville, MD on June 1, 1995 and have been assigned the indicated Accession numbers:
Microorganism ATCC Accession No.
Clone #70 55686 Clone #71 55685 Clone #72 55680 Clone #72-5-3-2 97179 Population #72p p-1 55684 Population #72p p-2 55683 Population #14028P P-1 55682 Population 014028p p-2 55681 The following microorganisms were deposited with the American Type Culture Collection (ATCC), Rockville, MD on May 29, 1996, and have been assigned the indicated Accession numbers:
Microorganism ATCC Accession No.
Clone YS721 55788 Clone YS7211 55787 Clone YS7212 55789 Clone YS7213 55786 The following plasmids were deposited with the American Type Culture Collection (ATCC), Rockville, MD on May 29, 1996, and have been assigned the indicated Accession numbers:
Microorganism ATCC Accession No.
pTK-Sec3 97592 pCD-Secl 97593 pSP-SAD4-5 97591 The invention claimed and described herein is not to be limited in scope by the specific embodiments herein disclosed since these embodiments are intended as illustrations of several aspects of the invention. Indeed, various modifications of the invention in addition to those shown and described herein will become apparent to those skilled in the art from the foregoing description. Such modifications are also intended to fall within the scope of the appended claims.
SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: Pawelek, John M.
Bermudes, David Low, Kenneth Brooks (ii) TITLE OF INVENTION: VECTORS FOR THE DIAGNOSIS AND TREATMENT
OF SOLID TUMORS INCLUDING MELANOMA

(iii) NUMBER OF SEQUENCES: 10 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: Pennie & Edmonds (B) STREET: 1155 Avenue of the Americas (C) CITY: New York (D) STATE: New York (E) COUNTRY: U.S.A.
(F) ZIP: 10036-2711 (v) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
is (D) SOFTWARE: Patentln Release #1.0, Version #1.30 (vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER: To Be Assigned (B) FILING DATE: On Even Date Herewith (C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:
(A) NAME: Baldwin, Geraldine F.
2 0 (B) REGISTRATION NUMBER: 31,232 (C) REFERENCE/DOCKET NUMBER: 8002-036 (ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (212) 790-9090 (B) TELEFAX: (212) 869-9741/8864 (C) TELEX: 66141 PENNIE

(2) INFORMATION FOR SEQ ID NO:1:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 27 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:

(2) INFORMATION FOR SEQ ID NO:2:

3 S (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 28 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 31 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

(2) INFORMATION FOR SEQ ID NO:4:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 30 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 32 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

(2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:
3 5 (A) LENGTH: 31 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:
Tyr Thr Ser Gly Tyr Ala His Arg Ser (2) INFORMATION FOR SEQ ID NO:8:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 6 amino acids (B) TYPE: amino acid (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:8:
Ser Gly Tyr Arg Ile Pro (2) INFORMATION FOR SEQ ID NO:9:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:

(2) INFORMATION FOR SEQ ID NO:10:

(i) SEQUENCE CHARACTERISTICS:
3 5 (A) LENGTH: 30 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:10:
International Application No: PCT/ US96 ,10250 MICROORGANISMS
Optional Sheet in connection vdth the microorganism referred to on page 189-140. Ones 135 of the description A. IDENTIFICATION OF DEPOSIT' Further deposits are Identified on an additional sheet Name of depositary instuution' American Tyype Culture Collection Address of depositary Institution (Including postal code and country) ' 12301 Paridawn Drive Rockville, MD 20852 US

Date of deposit' ,tune 1, 19.95 Accession Number - 55 886 B. ADDITIONAL INDICATIONS . pave blank If lax rpptiabte). TWa iafarautioe is ooadaaed as a aeparste aaad ed sheet C. DESIGNATED STATES FOR WHICH INDICATIONS ARE MADE' se-w D. SEPARATE FURNISHING OF INDICATIONS ' (eave Nut if not applicable) The Indications listed below we[ be submitted to the International Bureau later' (Specify the general nature of the Indications e.g..
'Accession Number of Deposit') E. 0 This sheet was received with the International application when filed (to be checked by the receiving Office) (Authorized Officer) 0 The date of receipt (from the applicant) by the internationalBureau -was (Authorized 'Officer) Form 1 34 (January International Application No: PCT/ US96/10250 Form PCT/RO/134 (cont.) American Type Culture Collection 12301 Parkiawn Drive Rockville, MD 20852 us Accession No. Date of Deposit 55685 June 1, 1995 65680 June 1. 1995 97179. June 1, 1995 65684 June 1, 1995 55683 June 1, 1995 55681 June 1, 1995 55682 June 1, 1995 55788 May 29, 1996 55787 May 29, 1996 55789 May 29, 1996 55786 May 29, 1996 97592 May 29, 1996 May 29, 1996 97591 May 29, 1996

Claims (11)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:
1. An isolated population of attenuated strain of Salmonella typhimurium, wherein the strain is selected from the group consisting of Salmonella typhimurium strain #70, having ATCC
Accession No. 55686;
Salmonella typhimurium strain #71, having ATCC
Accession No. 55685;
Salmonella typhimurium strain #72, having ATCC
Accession No. 55680;
Salmonella typhimurium strain #72 5-3-2, having ATCC
Accession No. 97179;
Salmonella typhimurium strain #14028 pop-1, having ATCC
Accession No. 55682;
Salmonella typhimurium strain #14028 pop-2, having ATCC
Accession No. 55681;
Salmonella typhimurium strain #72 pop-1, having ATCC
Accession No. 55684;
Salmonella typhimurium strain #72 pop-2, having ATCC
Accession No. 55683;
Salmonella typhimurium strain YS721, having ATCC
Accession No. 55788;
Salmonella typhimurium strain YS7211, having ATCC
Accession No. 55787;
Salmonella typhimurium strain YS7212, having ATCC
Accession No. 55789; and Salmonella typhimurium strain YS7213, having ATCC
Accession No. 55786.
2. The attenuated strain of Salmonella typhimurium of claim 1, further comprising a plasmid selected from the group consisting of a plasmid encoding the Herpes Simplex Virus thymidine kinase gene designated pTK-Sec3 and having ATCC
Accession No. 97592;
a plasmid encoding the E. coli cytosine deaminase gene designated pCD-Sec1 and having ATCC Accession No.
97593; and a plasmid encoding the human p450 oxidoreductase gene designated pSP-SAD4-5 and having ATCC Accession No.
97591.
3. The attenuated strain of Salmonella typhimurium of claim 1, wherein the Salmonella bacteria induce TNF-.alpha.
expression to a lesser degree compared to non-attenuated Salmonella bacteria.
4. Use of the attenuated strain of Salmonella typhimurium of claim 1 for treating solid tumor cancer.
5. Use of the attenuated strain of Salmonella typhimurium of claim 1 for the preparation of a medicament for treating solid tumor cancer.
6. The use of claim 4 or 5, wherein the solid tumor cancer is melanoma cancer.
7. The use of claim 4 or 5, wherein the solid tumor cancer is colon carcinoma.
8. The use of claim 4 or 5, wherein the solid tumor cancer is selected from the group consisting of lung cancer, liver cancer, kidney cancer, prostate cancer and breast cancer.
9. The use of claim 4 or 5, wherein the attenuated strain expresses a suicide gene.
10. The use of claim 9, wherein the suicide gene is encoded by a plasmid selected from the group consisting of pTK-Sec3 (ATCC Accession No. 97592), pCD-Sec1 (ATCC Accession No. 97593) and pSP-SAD4-5 (ATCC Accession No. 97591).
11. The use of claim 9, wherein the suicide gene is selected from the group consisting of p450 oxidoreductase, HSV
thymidine kinase, E. coli cytosine deaminase, carboxypeptidase G2, .beta.-glucuronidase, penicillin-V-amidase, penicillin-G-amidase, .beta.-lactamase, .beta.-glucosidase, nitroreductase and carboxypeptidase A.
CA2224075A 1995-06-07 1996-06-05 Vectors for the diagnosis and treatment of solid tumors including melanoma Expired - Lifetime CA2224075C (en)

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
US48642295A 1995-06-07 1995-06-07
US08/658,034 US6190657B1 (en) 1995-06-07 1996-06-04 Vectors for the diagnosis and treatment of solid tumors including melanoma
US08/658,034 1996-06-04
US486,422 1996-06-04
PCT/US1996/010250 WO1996040238A1 (en) 1995-06-07 1996-06-05 Vectors for the diagnosis and treatment of solid tumors including melanoma

Publications (2)

Publication Number Publication Date
CA2224075A1 CA2224075A1 (en) 1996-12-19
CA2224075C true CA2224075C (en) 2011-10-04

Family

ID=27048679

Family Applications (1)

Application Number Title Priority Date Filing Date
CA2224075A Expired - Lifetime CA2224075C (en) 1995-06-07 1996-06-05 Vectors for the diagnosis and treatment of solid tumors including melanoma

Country Status (15)

Country Link
US (3) US6190657B1 (en)
EP (1) EP0833660B1 (en)
JP (1) JP3482213B2 (en)
KR (1) KR100435932B1 (en)
CN (1) CN1138563C (en)
AT (1) ATE348634T1 (en)
AU (1) AU719446B2 (en)
BR (1) BR9609016A (en)
CA (1) CA2224075C (en)
DE (1) DE69636777D1 (en)
HK (1) HK1017253A1 (en)
IL (1) IL122407A (en)
MX (1) MX9709439A (en)
NZ (1) NZ312341A (en)
WO (1) WO1996040238A1 (en)

Families Citing this family (98)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6020144A (en) * 1996-09-12 2000-02-01 Symbiontics, Inc. Sustained delivery device comprising a Leishmania protozoa and methods of making and using the same
WO1998044788A2 (en) * 1997-04-09 1998-10-15 Chang Lung Ji Animal model for evaluation of vaccines
US6730512B2 (en) 1997-04-09 2004-05-04 Amdl, Inc. Combination immunogene therapy
US6080849A (en) * 1997-09-10 2000-06-27 Vion Pharmaceuticals, Inc. Genetically modified tumor-targeted bacteria with reduced virulence
EP1012232B1 (en) 1997-09-10 2009-10-28 Vion Pharmaceuticals, Inc. Genetically modified tumor-targeted bacteria with reduced virulence
DE19754938B4 (en) * 1997-12-11 2006-04-20 Christoph von Dr. Eichel-Streiber TGC method for induction of targeted, somatic transgenicity
CN1420783A (en) * 1999-10-04 2003-05-28 维昂药品公司 Compositions and methods for tumor-targeted delivery of effector molecules
AU7856100A (en) * 1999-10-04 2001-05-10 Vion Pharmaceuticals, Inc. Methods for treating solid tumors with irradiation and bacteria
AU7993600A (en) * 1999-10-04 2001-05-10 Sloan-Kettering Institute For Cancer Research Non-invasive tumor imaging by tumor-targeted bacteria
US6962696B1 (en) * 1999-10-04 2005-11-08 Vion Pharmaceuticals Inc. Compositions and methods for tumor-targeted delivery of effector molecules
CA2342040C (en) * 2000-09-21 2012-07-10 Kyowa Hakko Kogyo Co., Ltd. Anaerobic bacterium as a drug for cancer gene therapy
DE60141773D1 (en) * 2001-04-20 2010-05-20 Inst Systems Biology GREAT SIMILAR RECEPTOR 5-LIGANDS AND USE PROCESS
US8703146B2 (en) * 2001-04-20 2014-04-22 Institute For Systems Biology Toll-like receptor 5 ligands and methods of use
EP1281767A3 (en) 2001-07-31 2003-05-28 Aladar A. Szalay Light emitting microorganisms and cells for diagnosis and therapy of tumors
PT1567641E (en) * 2001-08-24 2012-08-21 Uvic Industry Partnerships Inc Proaerolysin containing protease activation sequences and methods of use for treatment of prostate cancer
US7037718B2 (en) 2001-10-26 2006-05-02 Cornell Research Foundation, Inc. Mutant purine nucleoside phosphorylase proteins and cellular delivery thereof
US7488598B2 (en) 2001-10-26 2009-02-10 Cornell Center For Technology Enterprise And Commercialization Mutant purine nucleoside phosphorylase proteins and cellular delivery thereof
US7015027B1 (en) 2001-11-20 2006-03-21 Mds (Canada) Inc. Radiation therapy by accumulation of therapeutic radionuclides in tumor-targeting bacteria
US20030228261A1 (en) * 2002-06-05 2003-12-11 Aladar Szalay Light emitting microorganisms and cells for diagnosis and therapy of diseases associated with wounded or inflamed tissue
EP1369491A1 (en) * 2002-06-05 2003-12-10 Aladar A. Szalay Light emitting microorganisms and cells for diagnosis and therapy of diseases associated with wounded or inflamed tissue
US20050026866A1 (en) * 2002-08-02 2005-02-03 Pawelek John M. Agents and methods for treatment of disease by oligosaccharide targeting agents
WO2004062597A2 (en) * 2003-01-09 2004-07-29 The Trustees Of The University Of Pennsylvania Compositions, methods and kits for enhancing the immunogenicity of a bacterial vaccine vector
US20050008618A1 (en) * 2003-02-27 2005-01-13 Howard Kaufman Composition for delivering an agent to a target cell and uses thereof
EP1610824A2 (en) * 2003-03-03 2006-01-04 University Of Victoria Innovation And Development Corporation Modified aerolysin linked to a lung cancer binding agent and methods of use for treating lung cancer
DK1622648T3 (en) * 2003-03-24 2010-09-06 Scripps Research Inst DNA vaccines against tumor growth and methods for their use
AU2004289953B2 (en) * 2003-06-18 2008-09-25 Genelux Corporation Modified recombinant vaccina viruses and other microorganisms, uses thereof
CN1838942A (en) * 2003-07-11 2006-09-27 普罗医药公司 Compositions and methods for hydrophobic drug delivery
US20050053664A1 (en) * 2003-09-08 2005-03-10 Eliezer Zomer Co-administration of a polysaccharide with a chemotherapeutic agent for the treatment of cancer
US7390646B2 (en) 2003-09-17 2008-06-24 The Regents Of The University Of California Bacterial vectors and methods of use thereof
ES2642565T3 (en) * 2003-11-28 2017-11-16 National Research Council Of Canada Anticarcinoma antibodies and their uses
US20070078109A1 (en) * 2004-02-13 2007-04-05 David Platt Compositions and methods used to treat acne and candida
US8221739B2 (en) 2004-04-29 2012-07-17 Botanic Oil Innovations, Inc. Method of cancer treatment
CN101022826A (en) * 2004-06-07 2007-08-22 哈罗尔·戴维·贡 Bacterial compositions for the treatment of cancer
CA2572756C (en) * 2004-06-29 2016-01-26 Anticancer, Inc. Cancer selective auxotrophs
EP1773128A2 (en) * 2004-08-02 2007-04-18 Pro-Pharmaceuticals, Inc. Compositions and methods for the enhancement of chemotherapy with microbial cytotoxins
KR101243662B1 (en) 2004-11-24 2013-03-14 가부시키가이샤 아네로파마·사이엔스 Novel shuttle vector
US7785872B2 (en) * 2004-12-08 2010-08-31 Simpson Biotech Co., Ltd. Nucleic acids for enhancing gene expression and use thereof
KR101302088B1 (en) * 2005-04-08 2013-08-30 가부시키가이샤 아네로파마·사이엔스 5-fluorouracil-resistant bacteria and method for production thereof
ATE502954T1 (en) 2005-06-14 2011-04-15 Protox Therapeutics Inc METHOD FOR TREATING OR PREVENTING BENICAENT PROSTATE HYPERPLASIA USING MODIFIED PORE-FORMING PROTEINS
WO2007079351A2 (en) * 2005-12-15 2007-07-12 Aeras Global Tb Vaccine Foundation Novel prime-boost combinations of attenuated mycobacterium
TW200819540A (en) 2006-07-11 2008-05-01 Genelux Corp Methods and compositions for detection of microorganisms and cells and treatment of diseases and disorders
US20080124355A1 (en) 2006-09-22 2008-05-29 David Gordon Bermudes Live bacterial vaccines for viral infection prophylaxis or treatment
EP2426142A3 (en) 2006-10-16 2012-06-13 Genelux Corporation Modified vaccinia virus strains for use in a diagnostic and therapeutic method
EP2150616A4 (en) 2007-05-10 2011-07-27 Univ Arizona Regulated synthesis of antigen and/or regulated attentuation to enhance vaccine immunogenics and/or safety
EP2152883A4 (en) * 2007-05-10 2010-10-27 Univ Arizona Recombinant bacteria comprising vectors for expression of nucleic acid sequences encoding antigens
JP5213075B2 (en) * 2007-06-15 2013-06-19 ジェネラックス・コーポレイション Microorganisms for tumor imaging and / or treatment
EP2202297B1 (en) * 2007-07-18 2014-05-14 Genelux Corporation Use of a chemotherapeutic agent in the preparation of a medicament for treating or ameliorating an adverse side effect associated with oncolytic viral therapy
CN101112398A (en) * 2007-08-28 2008-01-30 罗舒仓 Application of typhoid fever and paratyphoid fever salmonella in the aspect of fast anti-tumor
US9040059B2 (en) 2007-10-05 2015-05-26 The Arizona Board Of Regents For And On Behalf Of Arizona State University Recombinant bacterium capable of eliciting a protective immune response against C. perfringens
US20090136917A1 (en) * 2007-10-25 2009-05-28 Szalay Aladar A Systems and methods for viral therapy
US20110191868A1 (en) 2008-04-10 2011-08-04 Massachusetts Institute Of Technology Methods for identification and use of agents targeting cancer stem cells
EP2300023A2 (en) * 2008-05-16 2011-03-30 Genelux Corporation Microorganisms for preventing and treating neoplasms accompanying cellular therapy
US8647642B2 (en) 2008-09-18 2014-02-11 Aviex Technologies, Llc Live bacterial vaccines resistant to carbon dioxide (CO2), acidic PH and/or osmolarity for viral infection prophylaxis or treatment
WO2010045620A1 (en) 2008-10-17 2010-04-22 The Arizona Board Of Regents For And On Behalf Of Arizona State University Recombinant bacterium capable of eliciting an immune response against streptococcus pneumoniae
US8241623B1 (en) 2009-02-09 2012-08-14 David Bermudes Protease sensitivity expression system
US9163219B2 (en) * 2009-04-14 2015-10-20 Arizona Board Of Regents On Behalf Of Arizona State University Single expression vector for generation of a virus with a segmented genome
WO2010135563A1 (en) 2009-05-22 2010-11-25 The Arizona Board Of Regents For And On Behalf Of Arizona State University Recombinant bacterium and methods of antigen and nucleic acid delivery
US9339018B2 (en) 2009-05-27 2016-05-17 Industry Foundation Of Chonnam National Univeristy Selective infarcted-tissue-targeting bacteria and use thereof
US9045742B2 (en) * 2009-05-29 2015-06-02 The Arizona Board Of Regents For And On Behalf Of Arizona State University Recombinant Edwardsiella bacterium
US9062297B2 (en) 2010-01-13 2015-06-23 The Arizona Board Of Regents For And On Behalf Of Arizona State University Yersinia pestis vaccine
WO2011091291A1 (en) 2010-01-22 2011-07-28 The Arizona Board Of Regents For And On Behalf Of Arizona State University BACTERIUM COMPRISING A REGULATED rfaH NUCLEIC ACID
US9597379B1 (en) 2010-02-09 2017-03-21 David Gordon Bermudes Protease inhibitor combination with therapeutic proteins including antibodies
US8524220B1 (en) 2010-02-09 2013-09-03 David Gordon Bermudes Protease inhibitor: protease sensitivity expression system composition and methods improving the therapeutic activity and specificity of proteins delivered by bacteria
US8771669B1 (en) 2010-02-09 2014-07-08 David Gordon Bermudes Immunization and/or treatment of parasites and infectious agents by live bacteria
US9598697B2 (en) 2010-05-28 2017-03-21 The Arizona Board Of Regents For And On Behalf Of Arizona State University Recombinant bacterium to decrease tumor growth
WO2012008860A2 (en) 2010-07-16 2012-01-19 Auckland Uniservices Limited Bacterial nitroreductase enzymes and methods relating thereto
JP6415977B2 (en) 2011-04-15 2018-10-31 ジェネラックス・コーポレイションGenelux Corporation Cloned strain of attenuated vaccinia virus and method of using the same
US9956028B2 (en) * 2011-05-09 2018-05-01 Innovolink, Llc Apparatus and method for heating biological targets
US9901387B2 (en) 2011-05-09 2018-02-27 Innovolink, Llc Apparatus and method for heating adipose cells
US9901742B2 (en) 2011-05-09 2018-02-27 Innovolink, Llc Apparatus and method for heating a treatment region with an alternating electric field
US9675814B2 (en) 2011-05-09 2017-06-13 Innovolink, Llc Apparatus and method for obtaining a substantially constant current across a treatment region
US8859256B2 (en) 2011-10-05 2014-10-14 Genelux Corporation Method for detecting replication or colonization of a biological therapeutic
US9303264B2 (en) 2012-05-18 2016-04-05 The Arizona Board Of Regents For And On Behalf Of Arizona State University Photosynthetic microorganisms expressing thermostable lipase
DK2941258T3 (en) 2013-01-02 2019-12-16 Decoy Biosystems Inc COMPOSITIONS AND PROCEDURES FOR TREATING CANCER USING BACTERIA
US9593339B1 (en) 2013-02-14 2017-03-14 David Gordon Bermudes Bacteria carrying bacteriophage and protease inhibitors for the treatment of disorders and methods of treatment
US9580718B2 (en) 2013-06-17 2017-02-28 Arizona Board Of Regents On Behalf Of Arizona State University Attenuated live bacteria with increased acid resistance and methods of use thereof
WO2015076663A1 (en) * 2013-11-22 2015-05-28 Universiti Putra Malaysia (Upm) Salmonella enterica serovar agona for anti-tumor therapy
US9737592B1 (en) 2014-02-14 2017-08-22 David Gordon Bermudes Topical and orally administered protease inhibitors and bacterial vectors for the treatment of disorders and methods of treatment
US9616114B1 (en) 2014-09-18 2017-04-11 David Gordon Bermudes Modified bacteria having improved pharmacokinetics and tumor colonization enhancing antitumor activity
EP3250215A4 (en) 2015-01-29 2018-08-08 Oxyrase, Inc. Methods for inhibiting tumor growth
CN105983103A (en) * 2015-03-17 2016-10-05 广州华津医药科技有限公司 Application of genetically engineered bacterium VNP-20009-M in preparation of medicines for preventing and treating cancer metastasis
US10676723B2 (en) 2015-05-11 2020-06-09 David Gordon Bermudes Chimeric protein toxins for expression by therapeutic bacteria
CA3003157A1 (en) * 2015-10-30 2017-05-04 The Jackson Laboratory Compositions and methods relating to tumor analysis
US11185555B2 (en) 2016-04-11 2021-11-30 Noah James Harrison Method to kill pathogenic microbes in a patient
US11129906B1 (en) 2016-12-07 2021-09-28 David Gordon Bermudes Chimeric protein toxins for expression by therapeutic bacteria
US11180535B1 (en) 2016-12-07 2021-11-23 David Gordon Bermudes Saccharide binding, tumor penetration, and cytotoxic antitumor chimeric peptides from therapeutic bacteria
EP3550976A4 (en) 2016-12-07 2020-06-10 MD Biosciences, Inc. Methods of synergistic treatment of cancer
US20180275027A1 (en) * 2017-03-22 2018-09-27 Slmp, Llc Cell yield for synthetic tissue controls and synthetic tissue microarray controls
JP7097438B2 (en) 2017-07-11 2022-07-07 アクティム・セラピューティクス・インコーポレイテッド Genetically engineered immunostimulatory bacterial strains and their use
MX2021000165A (en) 2018-07-11 2021-05-28 Actym Therapeutics Inc Engineered immunostimulatory bacterial strains and uses thereof.
US11242528B2 (en) 2018-08-28 2022-02-08 Actym Therapeutics, Inc. Engineered immunostimulatory bacterial strains and uses thereof
EP3699263A1 (en) * 2019-02-22 2020-08-26 GlaxoSmithKline Biologicals S.A. Fermentation process
AU2020229875A1 (en) 2019-02-27 2021-09-02 Actym Therapeutics, Inc. Immunostimulatory bacteria engineered to colonize tumors, tumor-resident immune cells, and the tumor microenvironment
US11471497B1 (en) 2019-03-13 2022-10-18 David Gordon Bermudes Copper chelation therapeutics
KR20220113943A (en) 2019-11-12 2022-08-17 액팀 테라퓨틱스, 인코퍼레이티드 Immunostimulatory bacterial delivery platforms and uses thereof for delivery of therapeutic products
US10973908B1 (en) 2020-05-14 2021-04-13 David Gordon Bermudes Expression of SARS-CoV-2 spike protein receptor binding domain in attenuated salmonella as a vaccine
CA3191433A1 (en) 2020-08-12 2022-02-17 Actym Therapeutics, Inc. Immunostimulatory bacteria-based vaccines, therapeutics, and rna delivery platforms
WO2023086796A2 (en) 2021-11-09 2023-05-19 Actym Therapeutics, Inc. Immunostimulatory bacteria for converting macrophages into a phenotype amenable to treatment, and companion diagnostic for identifying subjects for treatment

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4436727A (en) 1982-05-26 1984-03-13 Ribi Immunochem Research, Inc. Refined detoxified endotoxin product
EP0142641B1 (en) 1983-09-26 1991-01-16 Udo Dr. Ehrenfeld Means and product for the diagnosis and therapy of tumours and for the treatment of weaknesses of the cellular and humoral immune system
US5830702A (en) 1990-10-31 1998-11-03 The Trustees Of The University Of Pennsylvania Live, recombinant listeria monocytogenes and production of cytotoxic T-cell response
US5695983A (en) 1990-12-18 1997-12-09 The General Hospital Corporation Salmonella vaccines
ATE190660T1 (en) 1990-12-18 2000-04-15 Gen Hospital Corp IMPROVED VACCINES
HU219535B (en) 1991-03-05 2001-05-28 Wellcome Foundation Ltd. Vaccines comprising attenuated bacteria expressing recombinant proteins
IL101409A0 (en) 1992-03-29 1992-11-15 Era Masis Ltd Method for the early diagnosis of cancer
IL101410A0 (en) 1992-03-29 1992-11-15 Era Masis Ltd Formulation for the treatment of cancer
US5631236A (en) 1993-08-26 1997-05-20 Baylor College Of Medicine Gene therapy for solid tumors, using a DNA sequence encoding HSV-Tk or VZV-Tk
IT1270123B (en) 1994-10-05 1997-04-28 Dompe Spa PHARMACEUTICAL COMPOSITIONS CONTAINING ENGINEERED MICROORGANISMS AND THEIR USE FOR THERAPY
US6051237A (en) 1994-11-08 2000-04-18 The Trustees Of The University Of Pennsylvania Specific immunotherapy of cancer using a live recombinant bacterial vaccine vector

Also Published As

Publication number Publication date
ATE348634T1 (en) 2007-01-15
CA2224075A1 (en) 1996-12-19
EP0833660A4 (en) 2003-04-09
EP0833660B1 (en) 2006-12-20
MX9709439A (en) 1998-02-28
EP0833660A1 (en) 1998-04-08
AU6385196A (en) 1996-12-30
DE69636777D1 (en) 2007-02-01
BR9609016A (en) 1999-07-06
WO1996040238A1 (en) 1996-12-19
HK1017253A1 (en) 1999-11-19
NZ312341A (en) 1999-05-28
IL122407A (en) 2007-09-20
KR19990022686A (en) 1999-03-25
US6190657B1 (en) 2001-02-20
US6685935B1 (en) 2004-02-03
CN1192694A (en) 1998-09-09
KR100435932B1 (en) 2004-08-25
CN1138563C (en) 2004-02-18
JP3482213B2 (en) 2003-12-22
US20050036987A1 (en) 2005-02-17
JP2000514400A (en) 2000-10-31
AU719446B2 (en) 2000-05-11
IL122407A0 (en) 1998-06-15

Similar Documents

Publication Publication Date Title
CA2224075C (en) Vectors for the diagnosis and treatment of solid tumors including melanoma
US10828356B1 (en) Modified bacteria having improved pharmacokinetics and tumor colonization enhancing antitumor activity
Nallar et al. Bacteria and genetically modified bacteria as cancer therapeutics: Current advances and challenges
Rathman et al. The unique trafficking pattern of Salmonella typhimurium-containing phagosomes in murine macrophages is independent of the mechanism of bacterial entry
Pawelek et al. Tumor-targeted Salmonella as a novel anticancer vector
Smith et al. The two distinct phospholipases C of Listeria monocytogenes have overlapping roles in escape from a vacuole and cell-to-cell spread
Stritzker et al. Enterobacterial tumor colonization in mice depends on bacterial metabolism and macrophages but is independent of chemotaxis and motility
KR100242798B1 (en) Stable pura vectors and uses therefor
KR20010015577A (en) Genetically modified tumor-targeted bacteria with reduced virulence
Bergman et al. CD8+ T cells restrict Yersinia pseudotuberculosis infection: bypass of anti-phagocytosis by targeting antigen-presenting cells
Mackie et al. Bacterial cancer therapy in autochthonous colorectal cancer affects tumor growth and metabolic landscape
US8075879B2 (en) Brucella melitensis mutants and methods
Liang et al. Attenuated Salmonella Typhimurium with truncated LPS and outer membrane-displayed RGD peptide for cancer therapy
US20100303861A1 (en) Live Attenuated Vaccine Strain for Prevention of Tularemia
US6562348B2 (en) Recombinant M. tuberculosis auxotrophic for leucine and vaccines using same
US20230346851A1 (en) Salmonella strain for treating cancer and use thereof
WO2022006748A1 (en) Genetically engineered live bacteria and methods of constructing the same
EP4332228A1 (en) Attenuated salmonella gallinarum strain and use thereof
Liang et al. Recombinant Listeria ivanovii strain expressing listeriolysin O in place of ivanolysin O might be a potential antigen carrier for vaccine construction
Shao Effects of cyclic adenosine monophosphate independent cyclic adenosine monophosphate receptor protein in recombinant attenuated Salmonella vaccines

Legal Events

Date Code Title Description
EEER Examination request
MKEX Expiry

Effective date: 20160606

MKEX Expiry

Effective date: 20160606