Protease inhibitor:protease sensitive expression system and method improving the therapeutic activity and specificity of proteins and phage and phagemids delivered by bacteria

Abstract

A genetically engineered live bacterium which is adapted to selectively replicate in and colonize a selected tissue type within the mammal, and concurrently produce within the selected tissue type at least one protease-sensitive cytotoxic molecule which is degradable by proteases within the selected tissue type, and at least one protease inhibitor peptide to inhibit the proteases within the selected tissue type from proteolytically degrading the protease sensitive cytotoxic molecule. The combination results in higher concentrations of the cytotoxic molecule local to the colonization, while permitting protease degradation of the cytotoxic molecule further away from the colonization.

Claims

1. A chimeric furin inhibitor peptide, comprising: a furin inhibitor peptide sequence, linked to and terminated with a terminal peptide comprising a bacterial signal sequence; and an intervening protease cleavage site between the furin inhibitor peptide sequence and the terminal peptide.

2. The chimeric furin inhibitor peptide according to claim 1, wherein the furin inhibitor peptide sequence is TABLE-US-00015 (SEQ ID NO: 2) TKKVAKRPRAKRAA, (SEQ ID NO: 3) TKKVAKRPRAKRDL, (SEQ ID NO: 4) GKRPRAKRA, (SEQ ID NO: 5) CKRPRAKRDL, (SEQ ID NO: 6) CVAKRPRAKRDL, or (SEQ ID NO: 7) CKKVAKRPRAKRDL.

3. The chimeric furin inhibitor peptide according to claim 1, wherein the terminal peptide is a complete C-terminal bacterial signal sequence.

4. The chimeric furin inhibitor peptide according to claim 1, wherein the terminal peptide is an hlyA C-terminal bacterial signal sequence.

5. The chimeric furin inhibitor peptide according to claim 1, wherein the terminal peptide is an N-terminal bacterial signal sequence.

6. The chimeric furin inhibitor peptide according to claim 1, wherein the terminal peptide is configured to interact with a non-lytic bacterial peptide secretion system of a live bacterium to export the chimeric furin inhibitor peptide by the bacterial secretion system.

7. The chimeric furin inhibitor peptide according to claim 1, wherein the intervening protease cleavage site comprises a furin cleavage site.

8. The chimeric furin inhibitor peptide according to claim 7, wherein the furin cleavage site comprises RXRAKRJ (SEQ ID NO: 54).

9. A chimeric furin inhibitor peptide, comprising: a furin inhibitor peptide sequence comprising a furin inhibitor site; a terminal peptide sequence comprising a bacterial secretion peptide sequence adapted to interact with a bacterial secretion system for export of the chimeric furin inhibitor peptide; and an intervening protease cleavage site between the furin inhibitor peptide sequence and the terminal peptide sequence.

10. The chimeric furin inhibitor peptide according to claim 9, wherein the furin inhibitor peptide sequence is TABLE-US-00016 (SEQ ID NO: 2) TKKVAKRPRAKRAA, (SEQ ID NO: 3) TKKVAKRPRAKRDL, (SEQ ID NO: 4) GKRPRAKRA, (SEQ ID NO: 5) CKRPRAKRDL, (SEQ ID NO: 6)  CVAKRPRAKRDL, or (SEQ ID NO: 7) CKKVAKRPRAKRDL.

11. The chimeric furin inhibitor peptide according to claim 9, wherein the terminal peptide sequence is configured to interact with a non-lytic bacterial peptide secretion system of a live bacterium to export the chimeric furin inhibitor peptide by the bacterial secretion system.

12. The chimeric furin inhibitor peptide according to claim 9, wherein the intervening protease cleavage site comprises a furin cleavage site.

13. A chimeric peptide, comprising a furin inhibitor peptide sequence, linked through a linker and an intervening protease cleavage site to a terminal peptide sequence comprising a terminal bacterial signal sequence.

14. The chimeric peptide according to claim 13, wherein the terminal peptide sequence is a complete C-terminal bacterial signal sequence HlyA.

15. The chimeric peptide according to claim 13, wherein the terminal peptide sequence is configured to interact with a non-lytic bacterial peptide secretion system of a live bacterium to export the chimeric furin inhibitor peptide by the bacterial secretion system.

16. The chimeric peptide according to claim 13, wherein the furin inhibitor peptide sequence comprises: TABLE-US-00017 (SEQ ID NO: 4) GKRPRAKRA, followed by a furin cleavage signal RXRAKR ⬇ DL (SEQ ID NO: 57), followed by the C-terminal signal sequence of hlyA TABLE-US-00018 (SEQ ID NO: 44) STYGSQDYLNPLINEISKUSAAGNLDVKEERSAASLLQLSGNASDES YGRNSITLTASA.

Description

5 BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIGS. 1A and 1B show a comparison of tumor-protease activated toxin with tumor protease inhibitor (FIG. 1A) and protease sensitive toxin expression system (FIG. 1B).

(2) FIGS. 2A-2C show secreted protease inhibitors.

(3) FIGS. 3A to 3F show chimeric colicins.

(4) FIGS. 4A to 4D show lytic peptide chimeras.

(5) FIGS. 5A to 5D show protease activated lytic peptide chimera prodrugs.

(6) FIGS. 6A to 6D show cytolethal distending toxin subunit B (cldtB) chimeras.

(7) FIGS. 7A to 7D show repeat in toxin (RTX) family members and hybrid operons.

(8) FIG. 8 shows a non-conjugative bacterium with and without the F′ factor.

(9) FIG. 9 shows segregation of required colicin toxin and immunity factors.

(10) FIG. 10 shows a non-conjugative bacterium capable of releasing phage/phagemids carrying expression constructs for DNA and RNA therapeutics.

6 DETAILED DESCRIPTION OF THE INVENTION

(11) The present invention provides, according to various embodiments, improved live attenuated therapeutic bacterial strains that express one or more therapeutic molecules together with one or more protease inhibitor polypeptides that inhibit local proteases that could deactivate the therapeutic molecules. In particular, one aspect of the invention relates to live attenuated tumor-targeted bacterial strains that may include Salmonella vectoring novel chimeric anti-tumor toxins to an individual to elicit a therapeutic response against cancer. The types of cancer may generally include solid tumors, leukemia, lymphoma and multiple myeloma. In addition, certain of the therapeutic molecules have co-transmission requirements that are genetically unlinked to the therapeutic molecule(s), limiting certain forms of genetic exchange. Another aspect of the invention relates to live attenuated tumor-targeted bacterial strains that may include Salmonella vectoring filamentous phage that encode anti-tumor DNA and RNA molecules to an individual to elicit a therapeutic response against cancer including cancer stem cells. The filamentous phage may also be targeted to tumor matrix cells, and immune cells. Another aspect of the invention relates to reducing or eliminating the bacteria's ability to undergo conjugation, further limiting incoming and outgoing exchange of genetic material.

(12) For reasons of clarity, the detailed description is divided into the following subsections: protease sensitivity; protease inhibitors; targeting ligands; chimeric bacterial toxins; co-expression of protease inhibitors with bacterial toxins, segregation of required colicin cofactors; limiting bacterial conjugation; phage/phagemid producing gram-negative bacteria encoding therapeutic DNA and RNA molecules.

(13) 6.1 Protease Sensitivity.

(14) The therapeutic proteins of the invention are sensitive to proteases (in contrast pro-aerolysin or urokinase chimeric toxins that are activated by proteases). Protease digestion sites may be added to the therapeutic agent to enhance protease sensitivity. Preferred proteases for conferring greater sensitivity are those that are under-expressed in tumors and over-expressed in normal tissues. Other proteases for which sensitivity sites may be added include tissue plasminogen activator, activated protein C, factor Xa, granzyme (A, B, M), cathepsins, thrombin, plasmin, urokinase, matrix metalloproteases, prostate specific antigen (PSA) and kallikrein 2.

(15) 6.2.1 Protease Inhibitors

(16) Protease inhibitors of the invention are preferably based on known polypeptide inhibitors. The inhibitors include both synthetic peptides and naturally occurring, endogenous peptides.

(17) To result in the desired activity, the peptides should be secreted outside of the bacteria. Accordingly, the peptides are modified by fusing them to secretion signals. The secretion signals may be either N-terminal (derived from ompA, ompF, M13pIII, cldt) or C-terminal (last 60 amino acids of the E. coli HlyA hemolysin, together with the required HlyBD supplied in trans and endogenous tolC) as shown in FIGS. 2A-2C. The N-terminal signal sequences are well known and characterized by the presence of a protease cleavage site for an endogenous bacterial protease. Thus, N-terminal signal sequences provide free protease inhibitors, free from the signal sequence. The C-terminal signal sequence may be further engineered to have a protease cleavage site in between the protease inhibitory peptide and the signal sequence. The cleavage site may be for the same protease that the peptide inactivates. Thus, the protease activates its own inhibitor. The protease cleavage site may also be for a protease other than for the protease inhibitor, thus deactivating another protease. Proteases upregulated within tumors for which protease cleavage sites may be engineered include: tissue plasminogen activator, activated protein C, factor Xa, granzyme (A, B, M), cathepsin, thrombin, plasmin, urokinase, matrix metalloproteases, prostate specific antigen (PSA) and kallikrein 2.

(18) Suitable protease inhibitors, include, but are not limited to, those listed below.

(19) Inhibitors of Kallikrein 2:

(20) TABLE-US-00002 SEQ ID NO: 9 SRFKVWWAAG SEQ ID NO: 10 AARRPFPAPS SEQ ID NO: 11 PARRPFPVTA

(21) Tissue Protease Inhibitor

(22) TABLE-US-00003 SEQ ID NO: 12 DSLGREAKCYNELNGCTKIYDPVCGTDGNTYPNECVLCFENRKRQTSILI QKSGPC (serine protease inhibitor, Kazal type 1, mature)

(23) Furin Inhibitors:

(24) TABLE-US-00004 SEQ ID NO: 1 PAAATVTKKVAKSPKKAKAAKPKKAAKSAAKAVKPK SEQ ID NO: 2 TKKVAKRPRAKRAA SEQ ID NO: 3 TKKVAKRPRAKRDL SEQ ID NO: 4 GKRPRAKRA SEQ ID NO: 5 CKRPRAKRDL SEQ ID NO: 6 CVAKRPRAKRDL SEQ ID NO: 7 CKKVAKRPRAKRDL SEQ ID NO: 8 RRRRRR [L6R] (hexa-L-arginine)

(25) Other suitable protease inhibitors are described in Rawlings et al., 2010, MEROPS: The Peptidase Database, Nucleic Acids Res. 2010 (Database issue):D227-33, the entirety of which ius expressly incorporated herein by reference. Suitable protease inhibitors also encompass functional fragments, respective homologs, and respective analogs, of the sequences described in Rawlings et al., and also other known peptide protease inhibitors including those described in Brinkmann et al, 1991 Eur J. Biochem 202: 95-99; Dunn et al., 1983 Biochem J 209: 355-362; Feng et al., (WO 2004/076484) Peptide Inhibitors Of Thrombin As Potent Anticoagulants); and Markowska et al., 2008, Effect of tripeptides on the amidolytic activities of urokinase, thrombin, plasmin and trypsin. Int. J. Peptide Research and Therapeutics 14: 215-218, each of which is expressly incorporated herein by reference.

(26) Targeting Ligands

(27) Targeting ligands are used to both confer specificity to chimeric proteins or phages, but also to direct internalization. The ligands of various aspects of the present invention are peptides that can be expressed as fusions with other bacterially-expressed proteins. The peptides may be further modified, as for gastrin and bombesin, in being amidated by a peptidylglycine-alpha-amidating monooxygenase or C-terminal amidating enzyme, which is co-expressed in the bacteria that use these peptides using standard molecular genetic techniques.

(28) TABLE-US-00005 TABLE 2 Examples of targeting peptides Peptide sequence or ligand name Receptor or Target Reference TGF-alpha EGFR SYAVALSCQCALCRR Rivero-Muller et al., CG-beta Molecular and Cellular SEQ ID NO: 13 Endocrinology 2007: 17- 25 Morbeck et al., 1993 AVALSCQCALCRR Jia et al., Journal of CG-beta (ala truncation) Pharmacy and SEQ ID NO: 14 Pharmacology 2008; 60: 1441-1448 Leuteinizing hormone-releasing LHRH receptor hormone (LHRH, also known as GnRH) pyroGlu-His-Trp-Ser-Tyr-Gly-Leu- Arg-Pro-Gly-NH.sub.2 pyro-EHWSYGLRPG SEQ ID NO: 15 IL2 IL2R Frankel et al. 2000, Clinical Cancer Research 6: 326-334. Tf TfR Frankel et al. 2000, Clinical Cancer Research 6: 326-334. IL4 IL4R Frankel et al. 2000, Clinical Cancer Research 6: 326-334. GM-CSF GM-CSFR Frankel et al. 2000, Clinical Cancer Research 6: 326-334. CD-19 Bombesin Gastrin releasing Dyba M., Tarasova N. I., peptide receptor Michejda C. J. Small molecule toxins targeting tumor receptors. Curr. Pharm. Des., 2004, 10(19), 2311-2334. Gastrin releasing peptide Gastrin releasing peptide receptor somatostatin octapeptide RC-121 (D-Phe-Cys-Tyr-D-Trp-Lys-Val- Cys-Thr-NH2 SEQ ID NO: 16 somatostatin Vasoactive intestinal peptide (VIP Neurotensin) Parathyroid hormone-related Parathyroid hormone protein PTHrP N-terminal 36 receptor G-protein residues, also has nuclear coupled receptor targeting KLAKLAKKLALKLA Proapoptotic peptide SEQ ID NO: 17 Endoglin (CD105) KCNK9 Mesothelin EGFR Mucin Heat stable enterotoxin (ST) Guanylyl cyclase C NSSNYCCELCCNPACTGCY SEQ ID NO: 18 Mature peptide VLSFSPFAQDAKPVESSK Heat stable enterotoxin EKITLESKKCNIAKKSNK unprocessed SDPESMNSSNYCCELCC NPACTGCY SEQ ID NO: 19 CM-CSF AML Alfa(V)Beta(3) integrin STEAP-1 (six transmembrane antigen of the prostate) CDCRGDCFC RGD 4C: active peptide Line et al. 46 (9): 1552. SEQ ID NO: 20 targeting the vß.sub.3 (2005) Journal of Nuclear integrin) Medicine LGPQGPPHLVADPSKKQGP bind to the gastrin WLEEEEEAYGWMDF receptor, also known in SEQ ID NO: 54 the art as the (gastrin-34) or big gastrin cholecystokinin B (CCKB) receptor MGWMDF SEQ ID NO: 21 N-terminal truncation of gastrin VPLPAGGGTVLTKM Gastrin releasing YPRGNHWAVGHLM peptide SEQ ID NO: 22 CAYHLRRC AML Nishimra et al., 2008. J SEQ ID NO: 23 Biol Chem 283: 11752- 11762 CAY (cys-ala-tyr) Lymph node homing Nishimra et al., 2008. J SEQ ID NO: 24 Biol Chem 283: 11752- 11762 RLRR (arg-le-arg-arg) Cell penetrating Nishimra et al., 2008. J SEQ ID NO: 25 Biol Chem 283: 11752- 11762 VRPMPLQ Colonic dysplasia Hsiung et al, Nature SEQ ID NO: 26 Medicine 14: 454-458 HVGGSSV 2622 Radiation-Induced International Journal of SEQ ID NO: 27 Expression of Tax- Radiation Oncology Interacting Protein 1 Biology Physics, Volume (TIP-1) in Tumor 66, Issue 3, Pages S555- Vasculature S556 Binds irradiated tumors H. Wang, A. Fu, Z. Han, i.e., ones responding to D. Hallahan therapy CGFECVRQCPERC Lung vasculature - Mori 2004 Current SEQ ID NO: 28 MOSE Pharmaceutical Design Binds membrane 10: 2335-2343 dipeptidase (MDP) SMSIARL MURINE PROSTATE Mori 2004 Current SEQ ID NO: 29 VASCULATURE Pharmaceutical Design 10: 2335-2343 VSFLEYR MURINE PROSTATE Mori 2004 Current SEQ ID NO: 30 VASCULATURE Pharmaceutical Design 10: 2335-2343 Fragment 3 of the high mobility group (HMG)N2 CKDEPQRRSARLSAKPAPP KPEPKPKKAPAKK SEQ ID NO: 31 H-VEPNCDIHVMW VEGF BINDING (WO/2006/116545) EWECFERL-NH2 PEPTIDE Spatial Control Of Signal SEQ ID NO: 32 Transduction RLLDTNRPLLPY L-PEPTIDE Let al., 2004. Cancer SEQ ID NO: 33 Nasopharyngeal Phage Research 64: 8002-8008. derived - caused internalization of phage RGDLATL truncated Alfa(v) beta (6) integrin Shunzi et al. (Kathyll C RGDLATLRQLAQEDGVVGVR Brown SEQ ID NO: 34

(29) 6.3 Small Lytic Peptides

(30) Small lytic peptides (less than 50 amino acids) are used to construct chimeric proteins for more than one purpose. The chimeric proteins containing lytic peptides may be directly cytotoxic for the cancer cells, and/or other cells of the tumor including the tumor matrix cells and immune cells which may diminish the effects of the bacteria by eliminating them. In order to be cytotoxic, they must be secreted (FIGS. 4A to 4D and 5A to 5D) and may be provided with cell specificity by the addition of a targeting ligand. Furthermore, the lytic peptides are useful in chimeric proteins for affecting release from the endosome. Small lytic peptides have been used in the experimental treatment of cancer. However, it is evident that most, if not all, of the commonly used antitumor small lytic peptides have strong antibacterial activity, and thus are not compatible with delivery by a bacterium (see Table 1 of Leschner and Hansel, 2004 Current Pharmaceutical Design 10: 2299-2310, expressly incorporated herein by reference). Small lytic peptides useful in the invention are those derived from Staphylococcus aureus, S. epidermidis and related species, including the phenol-soluble modulin (PSM) peptides and delta-lysin (Wang et al., 2007 Nature Medicin 13: 1510-1514, expressly incorporated herein by reference). The selection of the lytic peptide depends upon the primary purpose of the construct, which may be used in combination with other constructs providing other anticancer features. That is, the therapies provided in accordance with aspects of the present invention need not be provided in isolation, and the bacteria may be engineered to provide additional therapies or advantageous attributes. Constructs designed to be directly cytotoxic to cells employ the more cytotoxic peptides, particularly PSM-alpha-3. Constructs which are designed to use the lytic peptide to affect escape from the endosome use the peptides with the lower level of cytotoxicity, such as PSM-alpha-1, PSM-alpha-2 or delta-lysin.

(31) TABLE-US-00006 TABLE 3 Membrane lytic peptides useful in the invention Peptide and source Peptide Sequence Processed MAQDIISTISDLVKWIIDTVNKFTKK « short » active SEQ ID NO: 35 delta lysin S. aureus Delta lysin MMAADIISTI GDLVKWIIDTVNKFKK processed SEQ ID NO: 36 S. epidermitidis Delta lysin from MAQDIISTISDLVKWIIDTVNKFTKK CA-MRSA SEQ ID NO: 35 PSM-alpha-1 MGIIAGIIKVIKSLIEQFTGK SEQ ID NO: 37 PSM-alpha-2 MGIIAGIIKFIKGLIEKFTGK SEQ ID NO: 38 PSM-alpha-3 MEFVAKLFKFFKDLLGKFLGNN SEQ ID NO: 39 PSM-alpha-4 MAIVGTIIKIIKAIIDIFAK SEQ ID NO: 40 PSM-beta-1 MEGLFNAIKDTVTAAINNDGAKLGTSI VSIVENGVGLLGKLFGF SEQ ID NO: 41 PSM-beta-2 MTGLAEAIANTVQAAQQHDSVKLGTSI VDIVANGVGLLGKLFGF SEQ ID NO: 42

(32) 6.4 Chimeric Bacterial Toxins

(33) Chimeric toxins are used to adapt secreted bacterial proteins to provide therapeutic molecules that are effective in treating tumor cells, tumor stem cells as well as immune infiltrating cells. Targeting to a particular cell type uses the appropriate ligand from the Table 2 above or from other known sources.

(34) 6.4.1 Chimeric Colicins.

(35) Colicins lack tumor cell targeting. In the present invention, the colicin targeting and translocation domains are replaced with an M13pIII-derived signal sequence and truncated membrane anchor together with a targeting ligand. A lytic peptide may also be added. Examples of the unique organization for chimeric colE3, colE7 and col-la are shown in FIGS. 3A to 3F.

(36) 6.4.2 Chimeric Cytolethal Distending Toxin.

(37) Cytolethal distending toxin (cldt) is a three-component toxin of E. coli, Citrobacter, Helicobacter and other genera. Cldt is an endonuclease toxin and has a nuclear localization signal on the B subunit. Chimeric toxins are provided that utilize fusion to apoptin, a canary virus protein that has a tumor-specific nuclear localization signal, a normal cell nuclear export signal (FIGS. 6A to 6D). The cytolethal distending toxin B and chimeric cltdB may be expressed as a polycistronic construct consisting of cldtABC. The cytolethal distending toxin B and chimeric cltdB may be expressed as a polycistronic construct consisting containing the typhoid pertussis-like toxin (plt) AB genes.

(38) 6.4.3 RTX Toxins and Hybrid Operons.

(39) E coli HlyA(s) operon hlyCABD (+TolC), Actinobacillus actinomycetemcomitans leukotoxin ltxCABD, and a hybrid CABD operon are shown in FIGS. 7A to 7D. The ltxA may be generated as a chimera wherein it contains the C-terminal 60 amino acids of the E. coli HlyA. The ltx genes and chimeras may be expressed together with prtF and/or cyaE.

(40) 6.4.4 Saporin and Ricin Chimeras.

(41) Saporin and ricin can be replaced for the active portion of the colicin chimeras (FIGS. 3A to 3F). It can also be generated as a targeting peptide, saporin, HlyA C-terminus.

(42) 6.4.5 Cytotoxic Necrotic Factor (Cnf) and Bordetella Dermonecrotic Factor (Dnf) Chimeras.

(43) Cnf and dnf can be expressed as chimeras, where the N-terminal binding domain (amino acids 53 to 190 of cnf) is replaced with a tumor cell binding ligand, such as TGF-alpha.

(44) 6.4.6 Shiga Toxin (ST) and Shiga-Like Toxin (SLT) Chimeras.

(45) ST and SLT chimeras are generated wherein the GB3-binding domain is replaced with a tumor cell binding ligand, such as TGF-alpha.

(46) 6.4.7 Subtilase Toxin Chimeras.

(47) Subtilase chimeras are generated by replacing the binding domain with a tumor cell binding ligand, such as TGF-alpha.

(48) 6.5 Limiting Bacterial Conjugation.

(49) The fertility inhibition complex (finO and finP), are cloned onto the chromosome using standard genetic techniques such that strains either with or without the pilus resistant to mating with F′ bacteria (FIG. 8). Other known inhibitory factors may also be used.

(50) The F′ pilus factors in a Salmonella strain needed for phage to be able to infect the cell are provided by the F′ plasmid using standard mating techniques from an F′ E coli. The F′ factor provides other functions such as traD and the mating stabilization which are deleted using standard techniques.

(51) 6.6 Co-Expression of Protease Inhibitors with Bacterial Toxins and Determination of Synergy

(52) Each of the bacterial toxins listed herein may be improved in its therapeutic activity by co-expression with a protease inhibitor. Inhibitors are expressed as secreted proteins as described above. The effect of the protease inhibitor on in vitro cytotoxicity is determined using standard cell culture techniques and cytotoxicity assays such as MTT known to those skilled in the arts. The contribution of the protein cytotoxin and protease inhibitors is determined individually and in combination. Synergy may be determined using the median effect analysis (Chou and Talaly 1981 Eur. J. Biochem. 115: 207-216) or other standard methods. The assay may be further modified to include addition of a specific protease. The assay may also be used to determine synergy, additivity or antagonism of two or more bacterial cytotoxins. The assay may also be used to determine synergy, additivity or antagonism a bacterial cytotoxin together with a conventional small molecule cytotoxin (e.g., Cisplatin, doxorubicin, irinotecan, Paclitaxel or vincristine), targeted therapeutic (e.g., imatinib, irissa, cetuximab), proteosome inhibitor (bortezomib), mTOR inhibitor. In vivo studies may also be performed with antiangiogenic inhibitors such as Avastin, combretastatin, or thalidomide. In vivo studies with reticuloendothelial system (RES) blocker such as chlodronate which have the potential to improve the circulation time of the bacteria, vascular permeability inducing agents such as bradykinin, hyperthermia or carbogen which have the potential to improve the permeability of the tumor enhancing entry of the bacteria, or aldose reductase inhibitors.

(53) 6.7 Segregation of Required Colicin Toxin Cofactors.

(54) The chimeric colicin toxins have active colicin components that require their respective immunity proteins, which are usually genetically linked. By unlinking the two genes and separating them on the chromosome, a single fragment or phage transduction is highly unlikely to contain both elements. Without both elements, the toxin portion cannot be carried and will kill most bacteria. Any additional genes such as other chimeric therapeutic molecules genetically linked to the colicin will also be inhibited from being transferred to other bacteria (FIG. 9)

(55) 6.8 Phage/Phagemid Producing Gram-Negative Bacteria Encoding Therapeutic DNA and RNA Molecules (FIG. 10).

(56) The F′ pilus containing bacterium (FIG. 8) with deletions relating to conjugation and is expressing a protease inhibitor (PI) that is secreted into the medium are first infected with a helper phage, such as M13K07 which is able to use the pilus for entry. The helper phage may be further modified to lack an antibiotic resistance maker such as the kanamycin marker. Next, a phagemid (hybrid plasmid:phage which has the F′ origin such as one derived from pEFGP-N1) containing a pIII fusion with a targeting peptide, and optionally, a lytic peptide fusion to pVIII, and one or more therapeutic genes which could be a DNA encoding a functional p53 protein, or a gene encoding small interfering RNA molecules (siRNA) or microRNA (miRNA) molecules or other RNA interfering (RNAi) molecules or constructs that mediate RNA interference for an oncogene such as KRAS is transfected into the bacterial cell. The phagemid may also encode the T7 polymerase, and the effector gene such as one encoding the siRNA and/or miRNA and/or RNAi construct may be driven by the T7 promoter. The phage may also contain self-complementary sequences that induce the formation of double-stranded filamentous phage. Pieto and Sanchez 2007 Biochmica et Biophysica Acta 1770:1081-1084 regarding self-complementary sequences that induce the formation of double-stranded filamentous phage), expressly herein incorporated by reference. Now, the phagemid, in the presence of the helper phage, is replicated as single stranded DNA and packaged into a filamentous phagemid that is secreted outside of the bacterium. Because the phagemid contains pIII fusions with a targeting ligand, such as TGF-alpha, the phage are able to bind to the target cell, enter, release their DNA which then is transcribed into the respective therapeutic molecules and results in an antitumor effect. When administered to a patient with a tumor for which the appropriate receptor has been selected, the bacterium carrying the phagemids results in a therapeutic effect. The effect may be further enhanced by co-administration of camptothecin as described by Burg et al. See, Burg et al., “Enhanced Phagemid Particle Gene Transfer in Camptothecin-treated Carcinoma Cells”, Cancer Research 62: 977-981 (2002), expressly incorporated herein by reference.

7 FIGURE LEGENDS

(57) FIGS. 1A and 1B show a comparison of tumor-protease activated toxin with tumor protease inhibitor and protease sensitive toxin expression. FIG. 1A. Intravenously injected tumor protease activated toxin remains active if it diffuses out of the tumor. FIG. 1B. Intratumoral bacteria co-expressing a protease inhibitor and a protease sensitive toxin achieve high intratumoral activity and degradation following diffusion out of the tumor. The co-expression system results in high intratumoral activity, achieving a therapeutic benefit with low toxicity.

(58) FIGS. 2A-2C shows secreted protease inhibitors. A) An N-terminal signal sequence from a cytolethal distending toxin gene followed by a protease inhibitor (PI). B) A protease inhibitor followed by the hlyA C-terminal signal sequence. C) A protease inhibitor followed by the hlyA C-terminal signal sequence with a protease cleavage site (downward arrow).

(59) FIGS. 3A to 3F show chimeric colicins. FIG. 3A shows an M13 pIII signal sequence with amino acids 1 to 18 followed by a targeting peptide (TGF-alpha), a membrane anchor truncated M13 pIII amino acids 19 to 372 and the C-terminus of ColE3 (ribonuclease). The colicin is secreted, the signal sequence cleaved and the targeting peptide targets the EGFR-expressing cancer cell. FIG. 3B shows a lytic peptide is added between the signal sequence and the targeting peptide. Following cleavage of the signal sequence, the targeting peptide localizes to the EFGF-expressing cancer cell and the lytic peptide assists in its release from the endosome. FIG. 3C shows a ColE7 (DNase) chimera. FIG. 3D shows a ColE7 chimera with a lytic peptide. FIG. 3E shows a Col-la (membrane channel forming peptide) chimera. FIG. 3F shows a Col-la chimera with a lytic peptide.

(60) FIGS. 4A to 4D show lytic peptide chimeras. FIG. 4A shows a lytic peptide followed by the hlyA signal sequence. FIG. 4B shows a lytic peptide, targeting peptide (TGF-alpha), hlyA signal peptide chimera. FIG. 4C shows the M13 pIII signal sequence followed by a lytic peptide, the membrane anchor truncated M13 pIII amino acids 19 to 372 and a targeting peptide (TGF-alpha). FIG. 4D shows the M13 pIII signal sequence followed by a lytic peptide and a targeting peptide (TGF-alpha).

(61) FIGS. 5A to 5D show protease activated lytic peptide chimera prodrugs. FIG. 5A shows a blocking peptide followed by a tumor protease cleavage site, a lytic peptide followed by the hlyA signal sequence. The bracket underneath shows the active portion of the chimera following proteolytic cleavage. FIG. 5B shows a blocking peptide followed by a tumor protease cleavage site, a lytic peptide, targeting peptide (TGF-alpha) followed by a second tumor protease cleavage site and the hlyA signal peptide. FIG. 5C shows the M13 pIII signal sequence followed by a blocking peptide with a tumor protease cleavage site, a lytic peptide, the membrane anchor truncated M13 pIII amino acids 19 to 372 and a targeting peptide (TGF-alpha). FIG. 5D shows the M13 pIII signal sequence followed by a blocking peptide with a tumor protease cleavage site, a lytic peptide, a targeting peptide (TGF-alpha) with a tumor protease cleavage site and the membrane anchor truncated M13 pIII amino acids 19 to 372.

(62) FIGS. 6A to 6D show cytolethal distending toxin subunit B (cldtB) chimeras. It is understood that full functionality requires cltdA and cltdC. FIG. 6A shows CldtB followed by apoptin 1 to 121. FIG. 6B shows CldtB followed by apoptin 33 to 121. FIG. 6C shows CldtB followed by apoptin 33-46. FIG. 6D shows CldtB followed by apoptin 81-121.

(63) FIGS. 7A to 7D show repeat in toxin (RTX) family members and hybrid operons. FIG. 7A shows HlyCABD from E. coli. FIG. 7B shows LtxCABD from Actinobacillus. FIG. 7C shows a hybrid CABD of E coli (HlyBD) and Actinobacillus (HlyCA). FIG. 7D shows a hybid ltxCA with E. coli BD where the ltxA contains the C-terminal 60 amino acids of HlyA.

(64) FIG. 8 shows a non-conjugative bacterium with and without the F′ factor. The bacterial chromosome contains a secreted protease inhibitor construct (PI) that results in a secreted protease inhibitor. The chromosome also contains the FinO and FinP genes in order to inhibit conjugation. When present, the F′ factor containing the pilus genes with deletions relating to conjugation in traD and the mating stabilization (MS) results in a pilus expressed by the bacterium.

(65) FIG. 9 shows segregation of colicin toxin and required immunity factor(s). The bacterial chromosome has a colicin immunity protein integrated into a neutral site (e.g., attenuating mutation or IS200 element). The colicin, or colicin hybrid is not linked to the immunity protein, but is distal to it. Other therapeutic molecules may be in the same proximity, such as in a polycistronic organization. Based on this organization, a random DNA fragment, or a portion of the genome packaged by a transducing phage, could not contain the immunity protein. If such a fragment were transferred to another bacterium, expression of the colicin without the immunity protein would kill the bacterium.

(66) FIG. 10 shows: (A) A1. The bacterial chromosome contains a secreted protease inhibitor construct (PI) that results in a secreted protease inhibitor. A2. The F′ factor containing the pilus genes with deletions relating to conjugation in traD and the mating stabilization results in a pilus expressed by the bacterium. The FinO and FinP genes are inserted onto the F′ in order to further inhibit conjugation. A3. A helper phage such as M13K07 provides phage functions for replication and packaging. A4. A phagemid (hybrid plasmid:phage which has the F′ origin) containing a pIII fusion with a targeting peptide, and optionally, a lytic peptide fusion to pVIII, and one or more therapeutic genes which could be a DNA encoding a functional p53 protein, or a gene encoding small interfering RNA or microRNA molecules (siRNA or miRNA) that mediate RNA interference for an oncogene such as KRAS has been transfected into the bacterial cell (B). Then, the phagemid, in the presence of the helper phage, is replicated as single stranded DNA and packaged into a filamentous phagemid that is secreted outside of the bacterium. Because the phagemid contains pIII fusions with a targeting ligand, such as TGF-alpha, the phage are able to bind to the target cells (C), enter, release their DNA which then is transcribed into the respective therapeutic molecules and results in an antitumor effect. When administered to a patient with a tumor for which the appropriate receptor has been selected, the bacterium carrying the phagemids results in a therapeutic effect.

EXAMPLES

(67) In order to more fully illustrate the invention, the following examples are provided.

8.1 Example 1: Secreted Protease Inhibitors

(68) Secreted protease inhibitors are generated using standard molecular genetic techniques and expressed in bacteria using methods known to those skilled in the arts, operably linking a promoter, ribosomal binding site and initiating methionine if not provided by the first portion of the construct. The construct may either be a plasmid or a chromosomal integration vector, for which many different integration sites exist, including but not limited to any of the attenuation mutations or any of the IS200 elements. The constructs may also be polycistronic, having multiple genes and/or gene products separated by ribosomal binding sites. The different forms of the protease inhibitor constructs are shown in FIGS. 2A-2C. The constructs used have three basic forms: 1) An N-terminal signal sequence, such as that from M13pIII MKKLLFAIPLVVPFYSHS SEQ ID NO: 42, followed by a protease inhibitor such as the furin inhibitor GKRPRAKRA SEQ ID NO: 4; 2) a protease inhibitor such as the furin inhibitor GKRPRAKRA SEQ ID NO: 4 followed by the C-terminal signal sequence of hlyA STYGSQDYLNPLINEISKIISAAGNLDVKEERSAASLLQLSGNASDFSYGRNSITLTASA SEQ ID NO: 43, or 3) a protease inhibitor such as the furin inhibitor GKRPRAKRA SEQ ID NO:4, followed by a furin cleavage signal RXRAKR ⬇ DL SEQ ID NO: 57 followed by the C-terminal signal sequence of hlyA

(69) TABLE-US-00007 SEQ ID NO: 43 STYGSQDYLNPLINEISKIISAAGNLDVKEERSAASLLQLSGNASDFSYG RNSITLTASA

8.2 Example 2: A Targeted Colicin E3 (colE3) Chimera

(70) First, the colicin colE3 immunity protein is synthesized as an expression cassette and cloned into a chromosomal localization vector for an integration site distal to the that of the chimeric effector gene vector described below, e.g., an IS200 deletion vector at location. The amino acid sequence of the immunity protein is given as:

(71) TABLE-US-00008 SEQ ID NO: 56 MGLKLDLTWFDKSTEDFKGEEYSKDFGDDGSVMESLGVPFKDNVNNGCFD VIAEWVPLLQPYFNHQIDISDNEYFVSFDYRDGDW

(72) The sequence is reverse translated using codons optimal for Salmonella. The entire chimeric effector protein and expression cassette components are synthesized using standard DNA synthesis techniques at a contract DNA synthesis facility and integrated into the chromosome (Donnenberg and Kaper, 1991, Low et al., 2003, each of which is expressly incorporated herein by reference). The recipient stain can be any tumor-targeted gram-negative bacterium.

(73) This example follows the chimeric pattern shown in FIG. 3A. This chimera is targeted to cancer cells over-expressing EGFR via a TGF-alpha ligand. The chimera consists of the M13 filamentous phage pIII protein 18 amino acid signal sequence, followed by the natural alanine and a 3-glycine spacer. The spacer is followed by the mature 50 amino acid peptide for TGF-alpha, the remaining pIII protein truncated after amino acid 372 of pIII, followed by the enzymatically active (ribonuclease) C-terminus of colicin E3, followed by a stop signal. The complete amino acid sequence is:

(74) TABLE-US-00009 SEQ ID NO: 46 MKKLLFAIPLVVPFYSHSAGGGVVSHFNDCPDSHTQFCFHGTCRFLVQED KPACVCHSGYVGARCEHADLLAAETVESCLAKSHTENSFTNVWKDDKTLD RYANYEGCLWNATGVVVCTGDETQCYGTWVPIGLAIPENEGGGSEGGGSE GGGSEGGGTKPPEYGDTPIPGYTYINPLDGTYPPGTEQNPANPNPSLEES QPLNTFMFQNNRFRNRQGALTVYTGTVTQGTDPVKTYYQYTPVSSKAMYD AYWNGKFRDCAFHSGFNEDLFVCEYQGQSSDLPQPPVNAGGGSGGGSGGG SEGGGSEGGGSEGGGSEGGGSGGGSGSGDFDYEKMANANKGAMTENADEN ALQSDAKGKLDSVATDYGAAIDGFIGDVSGLANGNGATGDFAGSNSQMAQ VGDGDNSPLMNNFRQYLPSLPQSVECRFAHDPMAGGHRMWQMAGLKAQRA QTDVNNKQAAFDAAAKEKSDADAALSSAMESRKKKEDKKRSAENNLNDEK NKPRKGFKDYGHDYHPAPKTENIKGLGDLKPGIPKTPKQNGGGKRKRWTG DKGRKIYEWDSQHGELEGYRASDGQHLGSFDPKTGNQLKGPDPKRNIKKY L*

(75) The entire chimeric effector protein and expression cassette components are synthesized using standard DNA synthesis techniques, for example, at a contract DNA synthesis facility, and cloned into a chromosomal localization vector, e.g., an IS200 deletion vector, and integrated into the chromosome (Donnenberg and Kaper, 1991, Low et al., 2003, each of which is expressly incorporated herein by reference).

8.3 Example 3: A Targeted Colicin Chimera Containing a Lytic Peptide Resulting in Endosomal Release and/or Increased Anti-Cancer Cell Cytotoxicity

(76) The lytic peptide PSM-alpha-3 is inserted between the pIII signal sequence and the TGF-alpha (FIG. 3B). The complete sequence of the construct is as follows:

(77) TABLE-US-00010 SEQ ID NO: 47 MKKLLFAIPLVVPFYSHSAMEFVAKLFKFFKDLLGKFLGNN VVSHFNDCPDSHTQFCFHGTCRFLVQEDKPACVCHSGYVGARCEHADLLA AETVESCLAKSHTENSFTNVWKDDKTLDRYANYEGCLWNATGVVVCTGDE TQCYGTWVPIGLAIPENEGGGSEGGGSEGGGSEGGGTKPPEYGDTPIPGY TYINPLDGTYPPGTEQNPANPNPSLEESQPLNTFMFQNNRFRNRQGALTV YTGTVTQGTDPVKTYYQYTPVSSKAMYDAYWNGKFRDCAFHSGFNEDLFV CEYQGQSSDLPQPPVNAGGGSGGGSGGGSEGGGSEGGGSEGGGSEGGGSG GGSGSGDFDYEKMANANKGAMTENADENALQSDAKGKLDSVATDYGAAID GFIGDVSGLANGNGATGDFAGSNSQMAQVGDGDNSPLMNNFRQYLPSLPQ SVECRFAHDPMAGGHRMWQMAGLKAQRAQTDVNNKQAAFDAAAKEKSDAD AALSSAMESRKKKEDKKRSAENNLNDEKNKPRKGFKDYGHDYHPAPKTEN IKGLGDLKPGIPKTPKQNGGGKRKRWTGDKGRKIYEWDSQHGELEGYRAS DGQHLGSFDPKTGNQLKGPDPKRNIKKYL

8.4 Example 4: A Chimeric Colicin E7

(78) As for the other colicin E3 constructs, the colicin colE7 immunity protein is synthesized as an expression cassette and cloned into a chromosomal localization vector for an integration site distal to the that of the chimeric effector gene vector described below, e.g., an IS200 deletion vector at location.

(79) The genetic construct of the first colicin E7 chimera follows the same pattern as shown in FIG. 3A, except that the ColE3 C-terminus is replaced with the colE7 (a DNase) C-terminus comprising amino acids 444 to 576 (FIG. 3C).

(80) The genetic construct of the second colicin E7 chimera follows the same pattern as shown in FIG. 3C, except that the lysis peptide is inserted between the M13pIII signal sequence and the targeting peptide (TGF-alpha) (FIG. 3D).

8.5 Example 5: A Chimeric Colicin Ia

(81) As for the other colicin E3 constructs, the colicin Ia immunity protein is synthesized as an expression cassette and cloned into a chromosomal localization vector for an integration site distal to the that of the chimeric effector gene vector described below, e.g., an IS200 deletion vector at location.

(82) The genetic construct of the first colicin Ia chimera follows the same pattern as shown in FIG. 3A, except that the ColE3 C-terminus is replaced with the Ia (pore forming) C-terminus comprising amino acids 450 to 626 (FIG. 3 E).

(83) The genetic construct of the second colicin Ia chimera follows the same pattern as shown in FIG. 3B, except that the lysis peptide is inserted between the M13pIII signal sequence and the targeting peptide (TGF-alpha) (FIG. 3F).

8.6 Example 6: Expression of a C-Terminal Amidating Enzyme Required to Post-Translationally Modify Gastrin and Bombesin Targeting Peptides

(84) A C-terminal amidating enzyme composition known form serum or plasma which comprises a C-terminal amidating enzyme capable of amidating a C-terminal glycine which amidates the carboxy-terminus of the C-terminal glycine of a peptide terminating in Gly-Gly. The enzyme participating in such amidation is called peptidylglycine-α-amidating monooxygenase (C-terminal amidating enzyme) (EC.1.14.17.3) (Bradbury et al, Nature, 298, 686, 1982: Glembotski et al, J. Biol, Chem., 259, 6385, 1984, expressly incorporated herein by reference), is considered to catalyze the following reaction:

(85) —CHCONHCH.sub.2 COOH.fwdarw.—CHCONH.sub.2+glyoxylic acid is produced by the recombinant.

8.7 Example 7: Expression of Antitumor Lytic Peptides

(86) Examples of antitumor lytic peptides are shown in FIGS. 4A to 4D. It is understood that those peptides utilizing the hlyA signal sequence requires hlyBD in trans together with a functional tolC. The lytic peptide constructs consist of (FIG. 4A) lytic peptide joined to the HlyA signal sequence, (FIG. 4B) lytic peptide, targeting peptide, signals sequence, (FIG. 4C) M13 pIII signal sequence, lytic peptide, M13 pIII amino acids 19 to 372, targeting peptide, (FIG. 4D) M13 signal sequence, lytic peptide, targeting peptide, M13 pIII amino acids 19 to 372.

8.8 Example 8: Expression of Antitumor Lytic Peptide Prodrugs

(87) Examples of antitumor lytic peptide prodrugs are shown in FIGS. 5A to 5D. It is understood that those peptides utilizing the hlyA signal sequence requires hlyBD in trans together with a functional tolC. The lytic peptide prodrug constructs consist of (FIG. 5A) a neutral (e.g., beta sheet) blocking peptide of 50 amino acids, a protease cleavage site shown by downward arrow (for a protease not being blocked by a protease inhibitor), a lytic peptide, and the hlyA signal sequence, which may contain the same protease cleavage site shown by a downward arrow, (FIG. 5B) a neutral (e.g., beta sheet) blocking peptide of 50 amino acids, a lytic peptide, a targeting peptide (e.g., TGF-alpha), a protease cleavage site shown by downward arrow (for a protease not being blocked by a protease inhibitor), and the hlyA signal sequence, which may contain the same protease cleavage site shown by a downward arrow, (FIG. 5C) the M13 pIII signal sequence, a blocking peptide, a protease cleavage sequence, a lytic peptide, M13 pIII amino acids 19 to 372, and a targeting peptide (e.g., TGF-α), and (FIG. 5D) the M13 pIII signal sequence, a blocking peptide, a protease cleavage sequence, a lytic peptide, a targeting peptide (e.g., TGF-alpha), and M13 pIII amino acids 19 to 372.

8.9 Example 9: Cytolethal Distending Toxin cltdB Fusion with Apoptin (FIGS. 6A to 6D)

(88) A cytolethal distending toxin subunit B with tumor-specific nuclear localization and normal cell nuclear export is generated by a fusion with apoptin containing a five-glycine linker in between (FIG. 6A). The complete sequence of the construct is as follows:

(89) TABLE-US-00011 SEQ ID NO: 48 MKKYIISLIVFLSFYAQADLTDFRVATWNLQGASATTESKWNINVRQLIS GENAVDILAVQEAGSPPSTAVDTGTLIPSPGIPVRELIWNLSTNSRPQQV YIYFSAVDALGGRVNLALVSNRRADEVFVLSPVRQGGRPLLGIRIGNDAF FTAHAIAMRNNDAPALVEEVYNFFRDSRDPVHQALNWMILGDFNREPADL EMNLTVPVRRASEIISPAAATQTSQRTLDYAVAGNSVAFRPSPLQAGIVY GARRTQISSDHFPVGVSRRGGGGGMNALQEDTPPGPSTVFRPPTSSRPLE TPHCREIRIGIAGITITLSLCGCANARAPTLRSATADNSESTGFKNVPDL RTDQPKPPSKKRSCDPSEYRVSELKESLITTTPSRPRTAKRRIRL

8.10 Example 10: Cytolethal Distending Toxin cltdB Fusion with a Truncated Apoptin

(90) A cytolethal distending toxin subunit B with tumor-specific nuclear localization and normal cell nuclear export is generated by a fusion with a truncated apoptin amino acids 33 to 121 containing a five-glycine linker in between (FIG. 6B). The complete sequence of the construct is as follows:

(91) TABLE-US-00012 SEQ ID NO: 49 MKKYIISLIVFLSFYAQADLTDFRVATWNLQGASATTESKWNINVRQLIS GENAVDILAVQEAGSPPSTAVDTGTLIPSPGIPVRELIWNLSTNSRPQQV YIYFSAVDALGGRVNLALVSNRRADEVFVLSPVRQGGRPLLGIRIGNDAF FTAHAIAMRNNDAPALVEEVYNFFRDSRDPVHQALNWMILGDFNREPADL EMNLTVPVRRASEIISPAAATQTSQRTLDYAVAGNSVAFRPSPLQAGIVY GARRTQISSDHFPVGVSRRGGGGGITPHCREIRIGIAGITITLSLCGCAN ARAPTLRSATADNSESTGFKNVPDLRTDQPKPPSKKRSCDPSEYRVSELK ESLITTTPSRPRTAKRRIRL

8.11 Example 11: Cytolethal Distending Toxin cltdB Fusion with a Truncated Apoptin

(92) A cytolethal distending toxin subunit B with tumor-specific nuclear retention signal is generated by a fusion with a truncated apoptin amino acids 33 to 46 containing a five-glycine linker in between (FIG. 6C). The complete sequence of the construct is as follows:

(93) TABLE-US-00013 SEQ ID NO: 57 MKKYIISLIVFLSFYAQADLTDFRVATWNLQGASATTESKWNINVRQLIS GENAVDILAVQEAGSPPSTAVDTGTLIPSPGIPVRELIWNLSTNSRPQQV YIYFSAVDALGGRVNLALVSNRRADEVFVLSPVRQGGRPLLGIRIGNDAF FTAHAIAMRNNDAPALVEEVYNFFRDSRDPVHQALNWMILGDFNREPADL EMNLTVPVRRASEIISPAAATQTSQRTLDYAVAGNSVAFRPSPLQAGIVY GARRTQISSDHFPVGVSRRGGGGGIRIGIAGITITLSL

8.12 Example 12: Cytolethal Distending Toxin cltdB Fusion with a Truncated Apoptin

(94) A cytolethal distending toxin subunit B with a normal cell nuclear export signal is generated by a fusion with a truncated apoptin amino acids 81 to 121 containing a five-glycine linker in between (FIG. 6D). The complete sequence of the construct is as follows:

(95) TABLE-US-00014 SEQ ID NO: 50 MKKYIISLIVFLSFYAQADLTDFRVATWNLQGASATTESKWNINVRQLIS GENAVDILAVQEAGSPPSTAVDTGTLIPSPGIPVRELIWNLSTNSRPQQV YIYFSAVDALGGRVNLALVSNRRADEVFVLSPVRQGGRPLLGIRIGNDAF FTAHAIAMRNNDAPALVEEVYNFFRDSRDPVHQALNWMILGDFNREPADL EMNLTVPVRRASEIISPAAATQTSQRTLDYAVAGNSVAFRPSPLQAGIVY GARRTQISSDHFPVGVSRRGGGGGTDQPKPPSKKRSCDPSEYRVSELKES LITTTPSRPRTAKRRIRL

8.13 Example 13: Exchange of the Variable Loop in cldtB to Enhance Activity

(96) The amino acid sequence FRDSRDPVHQAL SEQ ID NO:52 which is associated with dimerization and inactivation can be exchanged for the loop NSSSSPPERRVY SEQ ID NO:54 from Haemophilus which is associated with stabile retention of cytotoxicity.

8.14 Example 14: Expression of Repeat in Toxin (RTX) Family Members

(97) RTX family members, including E. coli hemolysin operon hlyCABD and Actinobacillus actinomycetemcomitans leucotoxin ltxCABD are expressed in coordination with protease inhibitors as shown in FIGS. 7A to 7D. E coli hemolysin operon hlyCABD is expressed as a non-chimera (FIG. 7A). Actinobacillus actinomycetemcomitans leucotoxin ltxCABD operon is expressed as either a non-hybrid (FIG. 7B) or as a hybrid (FIG. 7C). It is understood that a functional tolC gene is required in the gram-negative bacterial strain for functional expression of each of these operons.

(98) It will be understood that the foregoing is only illustrative of the principles of the invention, and that various modifications can be made by those skilled in the art without departing from the scope and spirit of the invention.