Live self-destructing bacterial adjuvants to enhance induction of immunity

Abstract

Disclosed herein are unique adjuvant compositions comprising an attenuated derivative of a self-destructing bacterial pathogen that undergoes lysis in vivo. In exemplary embodiments, the bacterial pathogen is a Salmonella spp. Also disclosed are methods for enhancing an immune response using the adjuvants disclosed herein.

Claims

1. An adjuvant for the enhancement of vaccine efficacy, the adjuvant comprising an attenuated derivative of a bacterial pathogen that undergoes lysis in vivo, wherein the adjuvant comprises an attenuated Salmonella typhimurium (S. typhimurium) bacterium, comprising (a) one or more mutations facilitating lysis in vivo, comprising (i) P.sub.murA::TT rhaRS P.sub.rhaBAD murA, and/or (ii) a combination of P.sub.asdA::TT araC P.sub.araBAD asdA or P.sub.asdA::TT rhaRS P.sub.rhaBAD asdA, alr, and P.sub.dadB::TT rhaRS P.sub.rhaBAD dadB; and, optionally, a combination of (a) with (b) or (c), wherein (b) is one or more mutations to enhance recruitment of innate immunity comprising (i) pagP::P.sub.lpp lpxE and/or lpxR::P.sub.lpp lpxF, (ii) pagL or pagP or (iii) lpxR, arnT, eptA, waaC, waaG, waaL, wbaP, pmi or rfc or a combination thereof; or (iv) P.sub.stc::P.sub.murA stc and stcABCD or P.sub.saf::P.sub.murA saf and safABCD, or a combination thereof, and wherein (c) is a mutation enhancing safety and effective immunogenicity, comprising sifA, sopF, recA or sopB.

2. The adjuvant of claim 1, wherein the adjuvant comprises a self-destructing attenuated Salmonella typhimurium (S. typhimurium) bacterium, comprising the pagP::P.sub.lpp lpxE mutations.

3. The adjuvant of claim 2, wherein the adjuvant also comprises pagL and lpxR or arnT and eptA mutations.

4. The adjuvant of claim 2, wherein the adjuvant also comprises waaC, waaG, waaL or wbaP mutations.

5. The adjuvant of claim 2, wherein the adjuvant also comprises a sifA mutation.

6. The adjuvant of claim 2, wherein the adjuvant also comprises a recA mutation.

7. The adjuvant of claim 1, wherein the adjuvant also comprises (i) pagP::P.sub.lpp lpxE and/or lpxR::P.sub.lpp lpxF mutations, (ii) pagL and lpxR or arnT and eptA mutations, (iii) waaC, waaG, waaL or wbaP mutations, (iv) a sifA mutation, (v) a recA mutation, or (vi) P.sub.stc::P.sub.murA stc and/or P.sub.saf::P.sub.murA saf mutations.

8. The adjuvant of claim 1 further comprising a relA mutation to dissociate growth leading to lysis being dependent on protein synthesis.

9. The adjuvant of claim 1 further comprising a relA mutation to dissociate growth leading to lysis being dependent on protein synthesis.

10. A method of augmenting induction of protective immunity by a vaccine, the method comprising administering an immune response enhancing amount of the adjuvant of claim 2.

11. A method of augmenting induction of protective immunity by a vaccine, the method comprising administering an immune response enhancing amount of the adjuvant of claim 1.

12. The method of claim 10, wherein the adjuvant is administered by a route of administration comprising oral, intradermal, intravenous, intramuscular, intraocular, intranasal, intrapulmonary, epidermal, subcutaneous, mucosal, or transcutaneous administration.

13. The method of claim 11, wherein the adjuvant is administered by a route of administration comprising oral, intradermal, intravenous, intramuscular, intraocular, intranasal, intrapulmonary, epidermal, subcutaneous, mucosal, or transcutaneous administration.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1. Plasmid maps. (A) Lysis vector pG8R111, pBR ori; pYA4589 p15A ori; and pYA4595 pSC101 ori. (B) Lysis vectors with improved T2SS bla SS pG8R112, pSC101 ori; pG8R113, p15A ori, and pG8R114, pBR ori. (C) Lysis vector with T3SS SopE N-80 pG8R110, p15A ori.

(2) FIG. 2. Induction of TLR4, NOD1 and NOD2 signaling by Salmonella strains (9052 and 12499) on HEK-Blue-mTLR4/mNOD1/mNOD2 cell lines. HEK-Blue cell lines, expressing mTLR4, mNOD1 or mNOD2, were stimulated with 9052 and 12499. TLR/NOD activation was measured by SEAP activity after incubation of HEK-Blue-mTLR4 (A), HEK-Blue-mNOD1 (B) or HEK-Blue-mNOD2 (C) cells with 9052 and 12499. MDP (100 ng/ml) or LPS (100 ng/ml) were used as positive controls. 9052 and 12499 were grown in LB broth with D-alanine+DAP+arabinose, sedimented at room temperature, washed with BSG and suspended in tissue culture medium. MDPmuramyl dipeptide

(3) FIG. 3. Survival of 12517 and 12518 after inoculation into non-permissive growth conditions. X12517 and 12518 were grown in unpurple broth with DAP+D-alanine or unpurple broth with arabinose up to OD 0.9 bacterial cells were harvested and resuspended with PBS to desire concentration. Diluted bacteria were inoculated in un-purple broth for the lysis study. In every hour culture were plated on LB agar plates with DAP+D-alanine or LB agar plates with only arabinose and bacteria were counted

(4) FIG. 4. Swimming motility phenotypes of different S. Typhimurium mutant strains. Bacterial suspension was spotted onto the middle of the supplemented LB plates with 0.3% agar and incubated at 37 C. for 7 h. The diameter of the colonies was measured in centimeters.

(5) FIG. 5. Activation of TLR4 displayed on HEK cells by different S. Typhimurium mutant strains

(6) FIG. 6. Activation of TLR5 displayed on HEK cells by different S. Typhimurium mutant strains. HEK-Blue mTLR5 cells were stimulated with various Salmonella strains at different MOI of 10 or 1 or 0.01 or 0.001. After 24 h incubation, NF-kB-induced SEAP were determined by reading the OD at 650 nm. The response ratio was calculated by dividing the OD at 650 nm for the treated cells by the OD at 650 nm for the untreated cells.

(7) FIG. 7. Activation of Nod1 and Nod2 present in HEK cells by different S. Typhimurium mutant strains

(8) FIG. 8. Enhanced production of antibody production against Ova by co-administration of Family 1 Salmonella adjuvant strain 9052. 5106 CFU of 9052 s.c.; 100 g Ovalbumen s.c.; 50 l Alum s.c.

(9) FIG. 9. Mycobacterium tuberculosis H37Rv titers in spleens of mice vaccinated with Mycobacterium bovis Bacille Calmette-Guerin (BCG) with and without inoculation with ENIIRA strain 12499 or with 12499 alone.

(10) FIG. 10. Mycobacterium tuberculosis H37Rv titers in spleens of mice vaccinated with M. bovis BCG alone or with PIESV 12068 (pYA4891) or with ENIIRA strains 12517 or 12518.

(11) FIG. 11. Mycobacterium tuberculosis H37Rv titers in lungs of mice vaccinated with M. bovis BCG and/or 12068 (pYA4891) with or without inoculation with ENIIRA strain 12518.

(12) FIG. 12. Analysis of splenic lymphocytes from groups of mice used in Experiment 15 showing expansion of total CD3+ cells stimulated with Ag85A with production of INF and the ability of Ag85A to stimulate proliferation using CFSE staining.

(13) FIG. 13. Comparison of the abilities of the AAS strains 12517 and 12518 to activate NF-B synthesis via interactions with TLR4 and TLR5 and with the ability of the wild-type parent 3761 and purified LPS and flagellin.

(14) FIG. 14. Ability of Attenuated Adjuvant Salmonella (AAS) in Experiment 20 to enhance levels of protection of mice against M. tuberculosis H37Rv challenge. Left panel: Titers of Mtb H37Rv in lungs six weeks after challenge. Right panel: Titers of Mtb H37Rv in spleens six weeks after challenge-note that the titers for the last 3 groups in which no colonies were detected by plating undiluted suspensions are based on the titers if 1 colony had been detected. The inocula and routes of administration are denoted at the bottom of the Figure (see text for titers of inocula).

(15) FIG. 15. Antibody titers against the Mtb antigens Ag85A, CFP10 and ESAT-6 in the Groups of mice studied in Experiment 20.

(16) FIG. 16. Western blot analyses of AAS strains to synthesize the truncated FliC180 protein in the presence and absence of synthesizing the FljB phase II flagellin.

(17) FIG. 17. Motility of AAS strains evaluated for flagellin synthesis in FIG. 16 and lack of demonstrated motility in a mutant strain with a (hin-fljBA) phase lock deletion mutation.

(18) FIG. 18. Structure of S. Typhimurium lipid A with indication of the modifications/structures controlled by the indicated S. Typhimurium genes. The bottom panel displays the removal of the 1PO.sub.4 from lipid A caused by the expression of the Francisella tularensis lpxE gene. The expression of the F. tularensis lpxF gene leads to the loss of the 4PO.sub.4 (not shown).

(19) FIG. 19. Determination of endotoxic activities of S. Typhimurium strains dependent on genotype. See text for detailed methods and genotypes of strains.

(20) FIG. 20. Activation of NF-B synthesis by interaction with HEK cells displaying TLR4 in same strains evaluated for endotoxin activities displayed in FIG. 17.

(21) FIG. 21. Activation of NFB synthesis in HEK cells displaying TLR5 by Family A and B SDAAS strains. See text for detailed methods and genotypes of strains evaluated.

(22) FIG. 22. Demonstration that TLR5 activating flagellin is a secreted gene product released into the supernatant fluid and not remaining attached to surface of SDAAS cells. See text for description of methods and genotypes of tested strains.

(23) FIG. 23. Activation of NFB synthesis in HEK cells displaying TLR4 by Family A and B SDAAS strains. See text for detailed methods and genotypes of strains evaluated.

(24) FIG. 24. Demonstration that TLR4 activating modified and unmodified lipid A is present in supernatant of SDAAS strain cultures as well as remaining bound in the outer bacterial cell surface. See text for details of the methods used, the genotype of strains evaluated and an explanation of this unexpected observation.

(25) FIG. 25. Evaluation of Family A and B SDAAS strains for endotoxic activities. See text for description of study and genotypes of strains evaluated.

(26) FIG. 26. Evaluation of SDAAS strains for activation of NFB synthesis in HEK cells internally displaying TLR8 recognizing ssRNA. See text for methods and genotypes of strains evaluated.

(27) FIG. 27. Evaluation of SDAAS strains for activation of NFB synthesis in HEK cells internally displaying TLR9 recognizing CpG sequences in DNA. See text for methods and genotypes of strains evaluated.

DETAILED DISCLOSURE

(28) Successful pathogens have evolved to either infect the host in a stealth mode to be undetectable and/or suppress, modulate or circumvent induction of immunity, synthesize subterfuge antigens that induce immune responses that confer no protective immunity and/or devise means to colonize and persist in the host. Salmonella vaccine vectors have been continuously modified to eliminate these means as they are discovered and characterized. In addition, other modifications to enhance an induction of innate immune response in the absence of excess inflammation have been made with Salmonella vaccine vectors. Additionally, because live bacterial vaccines have the potential to persist in the environment if shed, a method to solve this problem has been devised. This method achieves regulated delayed lysis and thereby ensures that viable vaccine cells do not persist in vivo or survive if shed into the environment.

(29) Based on these attributes and the observation that Salmonella vaccine vector strains with these properties are entirely safe when administered to two-hour old mice or to pregnant mice or to protein-malnourished mice or to immunodeficient SCID mice, Salmonella strains with some of these properties are used herein to design strains with unique attributes to enhance their safety and efficacy when used as adjuvants. These self-destructing attenuated adjuvant Salmonella (SDAAS) strains are designed for use with a diversity of vaccines to augment their abilities to induce protective immunity or have desired abilities to alter physiological functions or combat cancers.

(30) Vaccines for the prevention of multiple infections caused by many pathogens are nonexistent. Currently existing vaccines often only provide partial protective immunity, or, require repeat vaccinations to provide adequate immunity to a subject. Recombinant attenuated Salmonella vaccine (RASV) vectors for synthesizing and delivering protective antigens encoded by genes from pathogens have been studied. In contrast to the RASV vectors previously described, empty vector control strains which do not deliver a protective antigen invariably but surprisingly provide low levels of protective immunity to a challenge pathogen. While this low level often exceed the level of immunity conferred by buffered saline, this level was significantly less than in animals receiving vaccine strains delivering protective antigens. Consequently, this low-level protective immunity against bacterial, viral and parasitic pathogens occurs as a result of vaccinating animal hosts with empty vector control RASVs.

(31) Consequently, the discovery of and improvements provided herein of live self-destructing Salmonella strains for serving as potent adjuvants (Sal-Adj) for enhancing induction of protective immunity by subunit and live ineffectual vaccines is paramount for the production of universally effective adjuvant strains. This discovery and the ensuing improvements enable use of Sal-Adj constructs to induce low-level protective immunity to challenge of unvaccinated animals and humans to various bacterial, viral and parasitic pathogens. Because of the principal means by which these Sal-Adj strains exert their beneficial activities (as determined in the studies conducted), they are also referred to as ENhanced Innate Immunity Response Activators (ENIIRAs) and, more recently, as Self-Destructing Attenuated Adjuvant Salmonella (SDAAS) strains. Optimal doses, routes and times of administration to enhance induction of protective immunity against infection and persistence by pathogens will be elucidated herein. In a similar way, the RASV vector strains now improved with features to diminish their abilities to subvert induction of immunity and increasing their immunogenicity by enhancing stimulation of innate immunity are now termed Protective Immunity Enhanced Salmonella Vaccine (PIESV) vectors.

Definitions

(32) As used herein the specification, a or an may mean one or more, unless clearly indicated otherwise. As used herein in the claims, when used in conjunction with the word comprising, the words a or an may mean one or more than one.

(33) As used herein, codon means, interchangeably, (i) a triplet of ribonucleotides in an mRNA which is translated into an amino acid in a polypeptide or a code for initiation or termination of translation, or (ii) a triplet of deoxyribonucleotides in a gene whose complementary triplet is transcribed into a triplet of ribonucleotides in an mRNA which, in turn, is translated into an amino acid in a polypeptide or a code for initiation or termination or translation. Thus, for example, 5-TCC-3 and 5-UCC-3 are both codons for serine, as the term codon is used herein.

(34) The terms comprise, have, and include are open-ended linking verbs. Any forms or tenses of one or more of these verbs, such as comprises, comprising, has, having, includes, and including, are also open-ended. For example, any method that comprises, has or includes one or more steps is not limited to possessing only those one or more steps and also covers other unlisted steps.

(35) The term consisting essentially of when used in conjunction with adjuvant containing compositions described herein refers to a composition comprising a Sal-Adj or ENIIRA or SDAAS strain and a pharmaceutical carrier without any other immune response enhancing components.

(36) The term agent as used herein refers to either adjuvant and/or vaccine.

(37) As used herein, the term adjuvant refers to an agent that stimulates and/or enhances an immune response in a subject. An adjuvant can stimulate and/or enhance an immune response in the absence of an antigen and/or can stimulate and/or enhance an immune response in the presence of an antigen. Exemplary embodiments of adjuvants disclosed herein are ENIIRA or SDAAS strains.

(38) The term administering or administration of an agent as used herein means providing the agent to a subject using any of the various methods or delivery systems for administering agents or pharmaceutical compositions known to those skilled in the art. Agents described herein may be administered by oral, intradermal, intravenous, intramuscular, intraocular, intranasal, intrapulmonary, epidermal, subcutaneous, mucosal, or transcutaneous administration.

(39) The term co-administration or co-administering as used herein refers to the administration of an active agent before, concurrently, or after the administration of another active agent such that the biological effects of either agents overlap.

(40) The term gene, as used herein, refers to a nucleic acid sequence that encodes and expresses a specific protein. In some embodiments, a gene may include a regulatory sequence of a 5-non-coding sequence and/or a 3-non-coding sequence.

(41) An immune response enhancing amount is that amount of an adjuvant administered sufficient to enhance an immune response of vaccine administration in a subject compared to vaccine administration without adjuvant administration. An immune response enhancing amount can be administered in one or more administrations.

(42) As used herein, the term immunogen refers to an antigen that is recognized as unwanted, undesired, and/or foreign in a subject.

(43) As used herein, the term immune response includes a response by a subject's immune system to a vaccine. Immune responses include both cell-mediated immune responses (responses mediated by antigen-specific T cells and non-specific cells of the immune system) and humoral immune responses (responses mediated by antibodies present in the plasma lymph, tissue fluids and mucosal secretions). The term immune response encompasses both the initial responses to an immunogen as well as memory responses that are a result of acquired immunity.

(44) The term codon-optimized as used herein refers to sequences that have been modified from their natural form to incorporate codons that are highly expressed or preferred by the organism in which they are introduced. Oligonucleotide and/or gene syntheses of pathogen specified protective and putative protective antigens may be codon-optimized to enhance translational efficiency in Salmonella by using codons used most frequently by highly expressed Salmonella genes and with codon selection to make the final sequence within about 2-3% of the average 52% GC content of the Salmonella genome. In some instances, codons are also changed to stabilize mRNA by destroying RNase E cleavage sites to prolong mRNA half-life. (see McDowall K J, Kaberdin V R, Wu S W, Cohen S N, Lin-Chao S. Site-specific RNase E cleavage of oligonucleotides and inhibition by stem-loops. Nature. 1995; 374 (6519): 287-90. Epub 1995 Mar. 16. doi: 10.1038/374287a0. PubMed PMID: 7533896 and Lin-Chao S, Wong T T, McDowall K J, Cohen S N. Effects of nucleotide sequence on the specificity of rne-dependent and RNase E-mediated cleavages of RNA I encoded by the pBR322 plasmid. J Biol Chem. 1994; 269 (14): 10797-803. Epub 1994 Apr. 8. PubMed PMID: 7511607), which are incorporated herein by reference. Provided below are preferred codons (1 letter symbol and codon): F-tcc S-tct Y-tac C-tgc W-tgg I-atc M-atg L-ctg P-ccg H-cac Q-cag R-cgt V-gtt T-acc N-aac K-aaa A-gcg D-gat E-gaa G-ggt
Based on the foregoing list, one or more modifications to the natural sequence may be made to reflect the above codon for the appropriate amino acid.

(45) As used herein, nucleic acid or nucleic acid sequence refers to polynucleotides, such as deoxyribonucleic acid (DNA) or ribonucleic acid (RNA), oligonucleotides, fragments generated by the polymerase chain reaction (PCR), and fragments generated by any of ligation, scission, endonuclease action, and exonuclease action. Nucleic acid molecules can be composed of monomers that are naturally-occurring nucleotides (such as DNA and RNA), or analogs of naturally-occurring nucleotides (e.g., a-enantiomeric forms of naturally-occurring nucleotides), or a combination of both. The term nucleic acid or nucleic acid sequence also pertains to codon optimized nucleic acid sequences as defined herein.

(46) As used herein, the phrase stimulating or enhancing an immune response refers to an increase in an immune response in the subject following administration of a vaccine with an adjuvant of the disclosed embodiments relative to the level of immune response in the subject when a vaccine has been administered without an adjuvant.

(47) As used herein, the term vaccine refers to an immunogen or a composition comprising an immunogen that elicits an endogenous immune response in a subject (e.g., a human or animal). The endogenous immune response may result in, for example, the switching of a Th1 biased immune response to a Th2 biased immune response, the activation or enhancement of T effector cell responses and/or the reduction of T regulatory cell response, the activation of antigen-specific naive lymphocytes that may then give rise to antibody-secreting B cells or antigen-specific effector and memory T cells or both, and/or the direct activation of antibody-secreting B cells.

(48) The term pharmaceutically acceptable carrier as used herein refers to one or more formulation materials suitable for accomplishing or enhancing the successful delivery of the pharmaceutical composition of the Sal-adj or SDAAS strains disclosed herein. As used herein, the term carrier refers to a pharmaceutically acceptable solid or liquid filler, diluent or encapsulating material. A water-containing liquid carrier can contain pharmaceutically acceptable additives such as acidifying agents, alkalizing agents, antimicrobial preservatives, antioxidants, buffering agents, chelating agents, complexing agents, solubilizing agents, humectants, solvents, suspending and/or viscosity-increasing agents, tonicity agents, wetting: agents or other biocompatible materials. A tabulation of ingredients listed by the above 15 categories, may be found in the U.S. Pharmacopeia National Formulary, 1857-1859, (1990). Examples of liquid carriers include, but are not limited to, water, saline, dextrose, glycerol, ethanol and mixtures thereof.

(49) The term protective immunity as used herein refers to induction of an immune response upon administration of a vaccine sufficient to confer protection against a pathogen or to have a therapeutic effect by arresting a disease such as cancer.

(50) Overview and Preliminary Studies

(51) 1. Bacterial Strains for Adjuvant Compositions.

(52) Many S. Typhimurium strains with individual and combinations of deletion and deletion-insertion mutations have been isolated/constructed and all the suicide vectors for these constructions are available to move these mutations into strains to create new Sal-Adj/ENIIRA/SDAAS strains.

(53) Table 1 lists the mutations and their associated phenotypic attributes that were used in these studies. Based on our prior results and the considerations discussed above, we began with three parental strains that exhibit lysis in vivo but with different periods of time needed for lysis and will therefore disperse into tissues away from the inoculation site to different extents. All these strains exhibit complete biological containment features being unable to persist in vivo or survive if released into the environment.

(54) Family A: 9052 alr-3 dadB4 asdA33requires D-alanine (Salmonella has two alanine racemases) and diaminopimelic acid (DAP) that are unique essential constituents of peptidoglycan that provides the rigid layer of the bacterial cell wall. D-alanine and DAP are only synthesized by bacteria and are totally absent in animal tissues.

(55) Family B: 12499 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdrequires presence of arabinose since synthesis of D-alanine and DAP are totally dependent on arabinose-induced synthesis of the dadB-encoded alanine racemase and the asdA-encoded aspartate semialdehyde dehydrogenase. The second alanine racemase that synthesizes D-alanine is absent due to the alr-3 mutation. Arabinose is absent in animal tissues but this strain undergoes several cell divisions in vivo until the dadB- and asdA-encoded enzymes are diluted by cell division so that D-alanine and DAP synthesis are insufficient to maintain peptidoglycan integrity.

(56) Family C: 11730 P.sub.murA25::TT araC P.sub.araBAD murA asdA27::TT araC P.sub.araBAD c2 (wza-wcaM)-8 relA198::araC P.sub.araBAD lacI TTmust be used with a lysis plasmid that also has araC P.sub.araBAD regulation of the murA and asdA genes (see next section) to yield a strain that is totally dependent on arabinose-induced synthesis of the enzymes needed to synthesize DAP and muramic acid (another unique essential constituent of peptidoglycan). Depending on the complementing plasmid copy number used, this strain will disseminate more widely and attain higher titers in animal tissues prior to onset of lysis than strains in Families A and B.

(57) The various mutations that are being used to conduct studies to determine the optimal means to stimulate innate immunity that is not excessively inflammatory are listed in Table 1 along with the mutations present in the Family A, B and C starting strains.

(58) TABLE-US-00001 TABLE 1 Mutations and associated phenotypes in S. Typhimurium adjuvant strains.sup.a Genotype Phenotype A. Deletion and deletion-insertion mutations to facilitate regulated delayed lysis in vivo P.sub.murA::TT makes synthesis of MurA, the first enzyme in the synthesis araC P.sub.araBAD of muramic acid, dependent on arabinose in growth medium and ceases murA synthesis in vivo due to absence of arabinose (1,2). MurA decreases due to cell division in vivo to ultimately cause lysis and death (2). The murA defect is complemented by MurA.sup.+ plasmid vectors (1). mur A mutations are lethal since the product of the gene is phosphorylated that precludes its uptake by Salmonella cells. asdA encodes aspartate semialdehyde dehydrogenase essential for synthesis of diaminopimelic acid (DAP) necessary for peptidoglycan synthesis (3). P.sub.asdA::TT makes synthesis of AsdA dependent on presence of arabinose araC P.sub.araBAD encodes one of two alanine racemases essential for synthesis of D- asdA alr alanine necessary for peptidoglycan synthesis (4). dadB encodes one of two alanine racemases essential for synthesis of D- alanine necessary for peptidoglycan synthesis (4). P.sub.dadB66::TT makes synthesis of DadB dependent on presence of arabinose araC P.sub.araBAD dadB eliminates twenty enzymes needed to synthesize several (wza-wcaM) exopolysaccharides that promote biofilm formation facilitating persistence and synthesis of GDP-fucose required for colanic acid synthesis (5), which protects cells undergoing cell wall-less death from lysing (6). These exopolymers are also immunosuppressive. relA the relA mutation uncouples growth regulation from a dependence on protein synthesis, an important attribute in strains with regulated delayed lysis (7, 8) B. Mutations enabling regulation of genes that might be present on plasmid vectors in conjunction with strains undergoing regulated delayed lysis in vivo P.sub.trc a promoter expressed at high level under both anaerobic and aerobic conditions and repressed by LacI (9, 10) relA::araC the arabinose-dependent synthesis of the LacI repressor is to P.sub.araBAD lacI TT enable a regulated delayed expression of DNA sequences under the control of P.sub.trc (11) asdA::TT the Asd enzyme is essential for the synthesis of DAP required for araC P.sub.araBAD c2 peptidoglycan synthesis (12). The arabinose-dependent synthesis of the C2 repressor enables a regulated delayed expression of DNA sequences under control of C2 repressed promoters (1). The asdA mutation is complemented by Asd.sup.+ plasmids (13). Phage P22 P.sub.R promoter is repressible by arabinose-dependent synthesis of the C2 repressor (14) C. Mutations altering synthesis of LPS components pagP::P.sub.lpp causes regulated delayed in vivo synthesis of the codon-optimized lpxE mutation lpxE gene from Francisella tularensis to cause synthesis of the non- toxic adjuvant form of LPS lipid A monophosphoryl lipid A (MPLA)(15). The pagP mutation also eliminates a means by which Salmonella alters LPS lipid A in vivo to decrease recruitment of innate immunity by interaction with TLR4 (16) pagL eliminates a means by which Salmonella alters LPS components in vivo to decrease recruitment of innate immunity by interaction with TLR4 (16) lpxR eliminates another means by which Salmonella alters LPS components in vivo to decrease recruitment of innate immunity by interaction with TLR4 (16) arnT eliminates a means by which Salmonella alters LPS lipid A in vivo to decrease recruitment of innate immunity by interaction with TLR4 (51) eptA eliminates a means by which Salmonella alters LPS lipid A in vivo to decrease recruitment of innate immunity by interaction with TLR4 (52) waaC eliminates an enzyme needed to synthesize LPS core resulting in a deep-rough phenotype and also renders Salmonella totally attenuated (49) waaG eliminates an enzyme needed to synthesize LPS core resulting in a medium-rough phenotype and also renders Salmonella totally attenuated (49) waaL eliminated the enzyme that joins the LPS O-antigen chain to the LPS core resulting in a moderate-rough phenotype and also renders Salmonella totally attenuated (49) D. Mutations altering synthesis of flagellar components fliC180 a specific mutation in the fliC gene encoding the Phase I flagellin protein (17) that contains the TLR5 receptor and a CD4 T- cell epitope. The mutation prevents synthesis of flagella. fliC2426 deletes fliC gene to eliminate synthesis of Phase I flagellin (17) fliB217 deletes fljB gene to eliminate synthesis of Phase II flagellin (17) (hin-fljBA) deletes the sequence necessary for phase switching of flagellin synthesis and eliminates synthesis of the phase II flagellin and the repressor of the fliC gene E. Mutations altering synthesis of fimbrial components P.sub.stc::P.sub.murA stc causes constitutive synthesis of the in vivo expressed Stc fimbriae that contribute to immunogenicity (18) stcABCD eliminates synthesis of Stc fimbriae (18) P.sub.saf::P.sub.murA saf causes constitutive synthesis of the in vivo expressed Saf fimbriae that contribute to immunogenicity (18) safABCD eliminates synthesis of Saf fimbriae (18) F. Mutations eliminating or diminishing effective immunogenicity sifA enables Salmonella to escape the SCV for lysis in cytosol (19) and eliminates a means of immunosuppression (20) steE G. Mutations leading to degradation of DNA within Salmonella cells recA enhances rate of DNA digestion in Salmonella cells as a consequence of recombination to liberate DNA fragments with CpG sequences and also renders Salmonella totally attenuated (53) .sup.a = deletion; TT = transcription terminator; P = promoter

(59) Table 2 lists the suicide plasmids used to move the mutations including deletion and deletion-insertion mutations listed and described in Table 1 into the Sal-Adj/ENIIRA strains constructed including the Family A, B and C strains listed above and their derivatives described in following sections and Examples as well as in strains yet to be constructed.

(60) TABLE-US-00002 TABLE 2 Suicide vectors for constructing the mutations in Table 1 Genotype Suicide Vector Marker A. Deletion and deletion-insertion mutations to facilitate regulated delayed lysis in vivo P.sub.murA::TT araC P.sub.araBAD murA pYA4686 Cm asdA pYA3736 Cm P.sub.asdA::TT araC P.sub.araBAD asdA pG8R71 Cm alr pYA3667 Cm dadB pYA3668 Cm P.sub.dadB66::TT araC P.sub.araBAD dadB pG8R73 Cm (wza-wcaM) pYA4368 Cm relA pYA3679 Cm B. Mutations enabling regulation of genes that might be present on plasmid vectors in conjunction with strains undergoing regulated delayed lysis in vivo relA::araC P.sub.araBAD lacI TT pYA4064 Cm asdA::TT araC P.sub.araBAD c2 pYA4138 Cm C. Mutations altering synthesis of LPS components pagP::P.sub.lpp lpxE pYA4295 Cm pagL pYA4284 Cm lpxR pYA4287 Cm arnT pYA4286 Cm eptA pYA4283 Cm waaC pYA5473 Cm waaG pYA4896 Cm waaL pYA4900 Cm D. Mutations altering synthesis of flagellar components fliC180 pYA3729 Cm fliC2426 pYA3702 Cm fliB217 pYA3548 Tet (hin-fljBA) pG8R306 Cm E. Mutations altering synthesis of fimbrial components P.sub.stc::P.sub.murA stc pYA5053 Cm stcABCD pYA5007 Tet P.sub.saf::P.sub.murA saf pYA5055 Cm safABCD pYA4586 Tet F. Mutations eliminating or diminishing effective immunogenicity sifA pYA3716 Cm G. Mutations leading to degradation of DNA within Salmonella cells recA pYA4680 Cm .sup.a = deletion; TT = transcription terminator; P = promoter
2. Plasmids for Adjuvant Strains with Regulated Delayed Lysis In Vivo. Family C

(61) Family C Sal-Adj (SDAAS) strains will be used in conjunction with plasmids conferring a regulated delayed lysis in vivo phenotype (FIG. 1). This phenotype is due to the araC P.sub.araBAD-regulated murA and asd genes with GTG start codons to decrease translation efficiency and the P22 P.sub.R (located in opposite orientation to transcription of the araC P.sub.araBAD GTG-murA GTG-asd genes) that is repressed by the C2 repressor encoded in the asdA27::TT araC P.sub.araBAD c2 mutation. MurA, AsdA and C2 are synthesized when Family C strains are grown with arabinose but cease to be synthesized in vivo. Thus, as C2 concentration decreases due to cell division in vivo, P.sub.R-directed anti-sense mRNA synthesis commences to block translation of residual asdA and murA mRNA. Transcription terminators (TT) flank all plasmid domains for controlled lysis, replication and gene expression so that expression in one domain does not affect activities of another domain. The time of onset of in vivo lysis can be controlled, in part, by using plasmids with different copy numbers. These plasmids, especially those with Type 2 and 3 secretion systems (T2SS; T3SS), can be used to synthesize and deliver different effector molecules such as single and double stranded RNA, flagellins, pilins, lipo-proteins or to enable synthesis and delivery of macromolecules such as teichoic acid, lipo-teichoic acid, mannan, etc. to enhance induction of innate immune responses.

(62) 3. Observations with Unexpected Ability of Empty Vector Control RASV Strains to Confer Low-Level Protective Immunity to Pathogens or Decreased Ability of Pathogens to Multiply in Hosts or Reduce Performance.

(63) As stated above, we have observed, in developing RASV/PIESV vectored vaccines, that the empty vector control groups (having a vector plasmid not encoding a protective antigen) invariably had higher survival or performance after challenge than the control groups receiving buffered saline (BS). This was especially true with vaccine vector strains that undergo lysis in various cell compartments in vivo. This implied that we might be recruiting innate immunity via activation of internal Nod and TLR9 receptors by release of peptidoglycan components and DNA intracellularly. We present results from some of these studies in Table 3.

(64) TABLE-US-00003 TABLE 3 Empty vector PIESVs and protective immunity* Percent Survival Pathogen RASV Path challenge BS PIESV-Ag PIESV + Ag Influenza PIESV-Flu Influenza 16 29 90 WSN Av (3) Yersinia PIESV-Yp Y pestis (s.c.) 0 38 83 pestis Av (3) S. PIESV-Sp S. pneumoniae 0 5 61 pneumoniae Av (7) M. PIESV-Mtb M. Empty vector control reduced Mtb colonization tuberculosis tuberculosis more than BS (3 comparisons) C. RASV-Cp C. Empty vector reduced lesions & mortality, perfringens perfringens enhanced feed conversion & weight gain more than BS (4 comparisons) E. tenella PIESV- Eimeria Empty vector enhanced feed conversion & Eimeria weight gain more than BS (2 comparisons) *First 4 studies in inbred mice and last 2 studies in outbred chickens The empty vector PIESV strains used for the results presented in Table 3 are very analogous to the Family C ENIIRA/SDAAS strains although some of these strains did not have the regulated delayed lysis in vivo phenotype.
The empty vector PIESV strains used for the results presented in Table 3 are very analogous to the Family C ENIIRA/SDAAS strains although some of these strains did not have the regulated delayed lysis in vivo phenotype.
4. Interaction of Family A and B Strains with HEK Cells Displaying TLR and Nod Factors.

(65) To evaluate the ability of Salmonella strains with differing genotypes to stimulate innate immune responses, we have used HEK293 cells with a NFB-inducible secreted embryonic alkaline phosphatase (SEAP) reporter gene (InvivoGen) and displaying Nod1, Nod2 and TLR4. We initially determined that all three Family A, B and C strains containing a plasmid encoding GFP were highly invasive into HEK cells. The 9052 and 12499 strains were grown in LB broth to maximize expression of the SPI-1 invasion phenotype, sedimented at room temperature and suspended in tissue culture medium. HEK cells at 210.sup.5 cells/ml were mixed with bacterial cells at a MOI of 10 in a volume of 200 l in 96 well plates. Unattached bacteria were removed by washing in tissue culture medium and then plates were incubated at 37 C. in 5% CO.sub.2 over 24 h with periodic scanning at 650 nm for production of the blue reagent due to secretion of alkaline phosphatase from the HEK cells. FIG. 2 presents the results. 9052 was best at activating Nod1, whereas both strains activated Nod2 equal to or better than muramyl dipeptide (MDP). Both strains activated TLR4 better than LPS at all times measured.

(66) 5. Enhanced NFB in HEK Cells Displaying TLR5.

(67) S. Typhimurium 9026 with fljB217 fliC180 mutations that overproduces a truncated FliC flagellin having the receptor for TLR5 stimulates significantly higher levels of IL6 and TNF production in GALT and MLN cells than the non-flagellated 9028 strain with fljB217 fliC2426 and also stimulates higher levels of NF-kB production in HEK cells displaying the murine TLR5. Expansion of the data in analyzing interaction of Sal-Adj/ENIIRA strains with HEK cells is presented in the Examples.

EXAMPLES

Example 1: Materials and Methods

(68) Bacterial Strains, Media and Bacterial Growth.

(69) All Sal-Adj/ENIIRA/SDAAS strains as well as other strains possessing individual mutations used in strain constructions are derived from the highly virulent S. Typhimurium UK-1 strain 3761 (21). LB broth and agar (22) will be used as complex media for propagation and plating of Salmonella strains. Purple broth (PB) (Difco), which is devoid of arabinose (Ara), was also used since LB contains low levels of arabinose. MacConkey agar with 0.5% lactose (Lac) and 0.1% Ara were used to enumerate bacteria recovered from mice. Mycobacterium tuberculosis H37Rv was propagated in Middlebrook 7H9 broth (Difco) supplemented with 10% albumin, dextrose and sodium chloride (ADS). Middlebrook 7H11 agar, supplemented with 10% ADS was used to determine bacterial CFU titers in the lungs and spleens of immunized mice challenged with M. tuberculosis. Bacterial growth was monitored spectrophotometrically and by plating for colony counts. S. Typhimurium PIESV strain 12068 carrying the plasmid pYA4891 was used in two of the experiments described in Example 5. 12068 has the following genotype: (P.sub.murA25::TT araC P.sub.araBAD murA asdA27::TT araC P.sub.araBAD c2 pmi-2426 (wza-wcaM)-8 relA197::araC P.sub.araBAD lacI TT recF126 sifA26 waaL46 pagL12::TT araC P.sub.araBAD waaL), which has the regulated delayed lysis in vivo phenotype and escapes the SCV to lyse in the cytosol to deliver antigens to the proteosome for class I presentation to induce CD8.sup.+, CD17.sup.+ and NKT-dependent immune responses. The plasmid pYA4891 is designed for use in regulated delayed lysis in vivo PIESVs and is a derivative of pYA4589, described in FIG. 1. Plasmid pYA4891 carries genes encoding three M. tuberculosis antigens: Ag85A, ESAT-6 and CFP-10 (31). Ag85A is one of three highly conserved mycolyl transferases integral to the synthesis of mycolic acids, an essential component of the mycobacterial cell envelope. Antigens 85A, B and C transport mycolyl residues from the mycobacterial cytoplasm to the growing chains of mycolic acids (38). The antigens ESAT-6 and CFP-10 of M. tuberculosis are two secreted proteins that are important virulence factors and are two of the most protective antigens of M. tuberculosis (39-41). M. tuberculosis ESAT-6 and its homologue TB10.4 are components of two subunit vaccines against M. tuberculosis that have been shown to be protective in immunized mice (31,41) and guinea pigs (41) and are currently in Phase IIa human clinical trials (42). In 12068 (pYA4891), Ag85A is exported from 12068 via the Salmonella Type 2 secretion system during in vivo growth; ESAT-6 and CFP-10 are delivered from 12068 by the Salmonella Type 3 secretion system directly into host cell cytoplasm during in vivo growth. All three antigens are delivered into the cytoplasm of host cells upon lysis of 12068, following regulated delayed lysis.

(70) Molecular and Genetic Procedures

(71) Methods for DNA isolation, restriction enzyme digestion, DNA cloning and use of PCR and real-time PCR for construction and verification of bacterial strains and vectors are standard (23). Defined deletion mutations with and without specific insertions have been constructed for all mutations listed in Table 1 using flanking sequences derived from the S. Typhimurium parent 3761. These mutations are introduced using either phage P22HTint (24, 25) transduction of suicide vectors integrated into the deletion mutation followed by selection for sucrose resistance (26) or by conjugational transfer of suicide vectors using standard methods (27, 28) with the suicide vector donor strains 7213 and 7378 (29). Plasmid constructs were evaluated by DNA sequencing and for ability of sugar-regulated sequences to specify synthesis of proteins using gel electrophoresis and western blot analyses (30).

(72) Strain Characterization

(73) Sal-Adj/ENIIRA/SDAAS strains were fully characterized at each step in their construction. Genetic attributes were confirmed by PCR with appropriate primers. Measurement of LPS core and O-antigen were performed after electrophoresis using silver-stained gels (88). This analysis is done after every step in any strain construction to eliminate possible spontaneous variants if they arise. We also validate the complete sensitivity of all Sal-Adj/ENIIRA/SDAAS strains to all antibiotics that might ever be used to treat Salmonella infections. Metabolic attributes are evaluated using API-20E tests. The presence of the recA mutation is determined by inability to undergo recombination using P22 transduction and extreme sensitivity to UV light.

(74) Swimming motility was assessed on LB plates solidified with 0.3% agar and supplemented with appropriate supplements (arabinose or DAP or D-alanine). Strains were grown statically overnight in the appropriate media. The next day, 50 l of this culture was inoculated into 2 ml of the appropriate media and grown with aeration at 37 C. to an optical density at 600 nm (OD600) of 0.8 to 0.85. One milliliter samples of cultures were pelleted at 4,500g and were resuspended in 1 ml BSG. Five microliter samples of bacterial suspension were spotted onto the middle of the plates, which were then incubated at 37 C. for 7 h.

(75) Production and secretion of flagellin was determined by shearing appendages from cells in culture using a Waring blender. After removal of cells by centrifugation, proteins in the supernatant were precipitated with TCA and then hydrolyzed by boiling in SDS buffer prior to electrophoresis. Antiserum against FliC and FljB flagellin were used to detect production of the Phase I and Phase II flagellins, respectively, by western blotting.

(76) Production of fimbrial adhesins is determined using antisera to the Saf and Stc fimbrial subunits by western blot analyses after electrophoresis of precipitated protein fractions obtained after shearing bacterial strains in a blender to remove flagellar and fimbrial appendages (as described above).

(77) Measurement of growth and lysis after inoculation of bacteria into media not permissive for growth was evaluated spectrophotometry and by dilution and plate counting to determine viable cell titers. For these evaluations we used Purple broth since it totally lacks the sugars arabinose, mannose and rhamnose that are present at very low concentrations in LB broth. We also use un-Purple broth that just lacked the pH indicator dye. In some studies, we also used MOPS minimal media with and without supplements DAP, D-alanine, arabinose and/or rhamnose. Bacterial strains were grown statically overnight in the appropriate media supplemented with 0.1% arabinose and/or 50 g DAP and/or 20 g D-alanine. The next day, 50 l of this culture was inoculated into 2 ml of the appropriate media and grown with aeration at 37 C. to an optical density at 600 nm (OD600) of 0.8 to 0.85. Cells were then centrifuged twice and washed and resuspended in un-Purple broth at a cell density of about 510.sup.7 CFU/ml and then grown with rotary aeration at 37 C. Absorbancies at OD.sub.600 nm were monitored continuously and dilutions for plate counts on permissive agar medium made at 30 and/or 60 min intervals.

(78) Cell Culture Methods and Use of HEK293 Cells to Monitor Initiation of Innate Immune Responses

(79) HEK293 cells with the murine TLR1, TLR2, TLR4, TLR4 MD2 CD14, TLR5, TLR5 CD14, TLR6, TLR8, TLR9, Nod1 and Nod2 were used with the NF-B SEAP reporter system to enable read outs at A650 nm. Sal-Adj/ENIIRA/SDAAS strains were grown to maximize their invasiveness and determine bacterial cell attachment to, invasion into and survival in HEK cells and monitor stimulation of NF-B production by HEK cells over a 24 h period. More specifically, Salmonella strains were grown in LB with appropriate supplements at 37 C. to an optical density at 600 nm (OD600) of 0.8 to 0.9. Bacteria were harvested by centrifugation at 4,500g at room temperature and were resuspended in endotoxin free water at the densities required to produce the desired dose ENIIRA/SDAAS strain as multiplicity of infection (MOI 10, 1, 01 and 001) relative to HEK cell density in the appropriate volume. Bacterial samples of 20 l were added per well in a flat-bottom 96-well cell culture plate.

(80) HEK-Blue mTLR4 or HEK-Blue mTLR5 or HEK-Blue mNOD1 or HEK-Blue mNOD2 cells were purchased from InvivoGen, San Diego, CA, USA. The cells were cultured at 37 C. in 5% CO.sub.2 in 25 vented flasks using Eagle's Minimum Essential Medium (EMEM) ATCC 30-2003 containing heat-inactivated fetal bovine serum (FBS), penicillin/streptomycin, and Normocin (InvivoGen). Cells were grown to 50-70% confluence and were resuspended in HEK-Blue detection media at a cell density 2-510.sup.6 viable cells per milliliter. Cell suspensions of 180 l were added to the previous bacterial samples added to the 96-well cell culture plate. Plates were incubated at 37 C. in 5% CO.sub.2 for 24 h. SEAP activity was determined using a spectrophotometer at 650 nm. The response ratios were calculated by dividing the OD at 650 nm for the treated cells by the OD at 650 nm for the untreated cells.

(81) Animal Experimentation to Monitor Safety and Efficacy of Sal-Adj/ENIIRA/SDAAS Constructs Administered by Different Routes to Young Adult Mice

(82) Ten-fold dilutions (10.sup.4-10.sup.8 CFU) of S. Typhimurium 12517 or 12518 strains were administered by intravenous (I.V.) or subcutaneous (S.C.) or intranasal (I.N.) routes to 6-week-old female BALB/c mice by following our standard procedures (49). Briefly, adjuvant strains were grown statically overnight in the appropriate media. The next day, 100 l of each culture was inoculated into 5 ml of the appropriate media and grown with aeration at 37 C. to an optical density at 600 nm (OD600) of 0.8 to 0.85. Cultures were pelleted at 4,500g at room temperature and were resuspended in BSG at the densities required to produce the desired dose in the appropriate volume. A volume of 20 l of bacterial suspension containing the appropriate dose was inoculated by the various routes at day 0. Mice were monitored for death twice a day up to 3 weeks.

(83) Animal Experimentation to Monitor Safety and Efficacy of Sal-Adj/ENIIRA/SDAAS Constructs to Augment Immune Responses to Ovalbumin (Ova)

(84) Imject Alum was purchased from Thermo Scientific (cat #77161) and ova albumin was purchased from Sigma (A2512-1G). Ova albumin was dissolved in BSG with a stock concentration 2 mg/ml. Adjuvant strain 9052 was grown statically overnight in LB supplemented with DAP and D-alanine. The next day, 100 l of this culture was inoculated into 5 ml of the appropriate media and grown with aeration at 37 C. to an optical density at 600 nm (OD600) of 0.8 to 0.9. Bacteria were harvested by centrifugation at 4,500g at room temperature and were resuspended in BSG at the densities required to produce the desired dose in the appropriate volume. Above suspension of 500 l was emulsified with 500 l Imject Alum by repeatedly passing through a syringe needle. Each mouse was inoculated subcutaneously with 100 l of suspension. Six-week-old BALB/c mice were divided into three groups of five mice each named Group-I [Ova (100 g)+9052 (510.sup.6)], Group-II [Ova (100 g)] and Group-III [Ova (100 g)+Imject Alum (50 l)]. Animals were immunized subcutaneously. Blood was sampled from immunized mice for antibody determination by ELISA at 2, 4 and 6 weeks post-inoculations.

(85) Animal Studies to Determine Ability of Sal-Adj/ENIIRA/SDAAS Constructs to Augment Level of Protective Immunity Against M. tuberculosis (Mtb) Challenge of BCG Immunized Mice

(86) C57/BL6 mice of both sexes were given a s.c. dose of 510.sup.6 CFU of M. bovis BCG Pasteur (ATCC 35734) on day 0 without and with varying doses in CFU of different Sal-Adj constructs by s.c. (110.sup.5 CFU), i.n. (110.sup.7 CFU), and i.v. (510.sup.4 CFU) routes. On day 28, sera were collected from all mice. In addition, two mice from each group were euthanized, livers and spleens were removed, processed and used for flow cytometric analyses. On day 35 all mice were infected with aerosolized M. tuberculosis H37Rv (50 to 100 bacteria per mouse) using a Glas-Col Inhalation Instrument (31). Six weeks after challenge, mice were euthanized, lungs and spleens removed aseptically, tissues homogenized, homogenates diluted and samples plated on Middlebrook 7H11+10% ADS agar plates (Difco). Plates were incubated at 37 C. for 3-4 weeks to determine the Mtb CFU in these organs.

(87) Monitoring Immune Responses

(88) Antigen Preparation

(89) Purified Ova was obtained commercially. M. tuberculosis antigens Ag85A, ESAT-6 and CFP-10 were purified as His-tagged proteins from recombinant E. coli. These antigens were used for immunoassays as described below.

(90) ELISA

(91) Serum antibodies were measured in blood collected by submandibular bleeding. We determined the concentrations of IgG and IgA against Ova in g/ml and the concentrations of IgG against Ag85A, ESAT-6 and CFP-10 in g/ml. To distinguish between Th1 and Th2 responses, we determined titers of IgG1 and IgG2a for Ova and titers of IgG1 and IgG2b for Ag85A, ESAT-6 and CFP-10. The 96-well plates were coated with 100 ng Ova or with 0.5 to 1 g of each M. tuberculosis antigen. Free binding sites were blocked with the SEABLOCK Blocking Buffer (Thermo Fisher). Anti-Ova titers or anti-Ag85A, anti-ESAT-6 or anti-CFP-10 titers in the serum (dilution 1:100) were detected with biotinylated goat anti-mouse IgG, IgG1, IgG2a or IgA for Ova and biotinylated goat anti-mouse IgG, IgG1 or IgG2b for Ag85A, ESAT-6 and CFP-10 (Southern Biotechnology) followed by incubation with a streptavidin-alkaline phosphatase conjugate (Southern Biotechnology). Color development (absorbance at 405 nm) with p-nitrophenyl phosphate (Thermo Fisher Scientific) was recorded with an automated ELISA plate reader (EL311SX; Biotek). Unconjugated mouse antibodies (Southern Biotechnology) (IgG, 5 g/ml to 40 ng/ml; IgG1, IgG2a or IgG2b, 1 g/ml to 8 ng/ml; IgA 62.5 ng/ml to 0.46 ng/ml) were serially diluted and coated on a 96-well plate in duplicate. Standard curves were generated by plotting the OD405 values against the representative concentrations of the diluted unconjugated antibody solutions and fitted to a 4-parameter logistic curve (R2>0.98). The absorbance values of experimental samples were fit into the standard curve to interpolate antibody concentrations. All samples were analyzed in triplicate. We also used ELISPOT assays (32) in initial studies to determine whether antigen-specific IgA and IgG secreting peripheral blood lymphocytes are induced 10 to 15 days after vaccination with Ova and adjuvants.

(92) Cellular Immune Responses and Flow Cytometry Analyses

(93) These more detailed evaluations of immune responses induced have been and will be continually used as optimal Sal-Adj/SDAAS constructs are identified and their use evaluated with vaccine constructs and preparations to ultimately be tested.

(94) Flow Cytometry

(95) Flow cytometry was used to quantitate populations of antigen-specific CD4.sup.+ and CD8.sup.+ T cells and antigen-specific cytokine-secreting cells in the lungs and spleens of mice immunized with M. bovis BCG in combination with Sal-Adj/SDAAS constructs. Lymphocytes were isolated from homogenized lungs and spleens of immunized or PBS control mice by centrifugation of the cell lysates through Percoll gradients. After washing the purified lymphocytes with PBS, the cells were simulated with 10 g/ml of purified protein antigens for 24 h. Fc blocking reagent was used to prevent non-specific binding of antibodies to Fc receptors on the lymphocytes. Surface staining using anti-mouse fluorophore-labelled antibodies (Biolegend), followed by fixation of the cells with 4% para-formaldehyde was used to identify subsets of lymphocytes. Gating on both CD4 and CD8 was done on the CD3.sup.+ lymphocyte population to detect the percent of antigen-specific CD4.sup.+ and CD8.sup.+ cells expressing effector KLRG1, PD1 or memory CD62L, CD127 molecules. To detect intracellular cytokines, lymphocytes were stained for surface markers CD3, CD44, CD4 and CD8, followed by treatment with BD Sciences Cytofix/Cytoperm and stained intracellularly with combinations of IFN- and TNF- antibodies. All samples were analyzed in the Department of Infectious Diseases and Immunology Flow Cytometry Core Facility on an 8-color FACSCanto, 18-color FACSFortessa flow cytometer with a SH800Z Cell Sorter. Data analyses and statistical comparisons among the samples from immunized and non-immunized mice was done using the Flow JO software.

(96) Splenomegaly.

(97) To determine protection against challenge with M. tuberculosis H37Rv, the spleens and lungs are weighed individually after they are removed from the euthanized mice. The number of CFUs determined from plating samples of homogenized lungs and spleens are reported as CFU per gram of tissue. When spleens were removed from mice immunized with ENIIRA/SDAAS strains, it was immediately evident by visual observation, that some spleens were significantly enlarged, compared to the spleens of unimmunized mice or mice immunized with M. bovis BCG or mice immunized with PIESV 12068 (pYA4891).

(98) Statistical Analyses

(99) All results were analyzed using the most appropriate statistical test from the SAS program to evaluate the relative significance or lack thereof of results obtained.

Example 2. Construction and Characterization of Sal-Adj/ENIIRA/SDAAS Strains

(100) a. Introduction

(101) Sal-Adj/ENIIRA/SDAAS strains of the starting genotypes for the Family A, B and C strains were initially compared to determine which most enhances induced immune responses to Ova and protective immunity to M. tuberculosis challenge. This was because it is possible that strains from different families will each be more efficacious in one evaluative test than the other. In these initial studies, the Family A and B strains were found to be most efficacious, in all probability since they invade efficiently and undergo lysis more rapidly. In accord with this, Family C strains that underwent regulated lysis in vivo more rapidly, induced higher innate immune responses than did Family C strains that underwent more cell divisions in vivo prior to lysis. These rates and numbers of cell divisions in vivo prior to commencement of lysis is regulated by altering levels of expression of genes for virulence attributes, synthesis of peptidoglycan precursors and repressors by modification of promoter, SD and start-codon sequences and/or by altering the spacing between these elements and/or by alterations in plasmid copy numbers. Based on these results we are continuing to develop improved Family C strains as the PIESV vector strains as vaccine constructs for protective antigen and DNA vaccine delivery to prevent infectious diseases. Since recruitment and induction of innate immunity initially upon vaccination would be most beneficial in augmenting induction of acquired immunity by a vaccine, we have focused on developing the Family A and B ENIIRA/SDAAS strains to serve as superior adjuvants because they invade host cells and lyse quickly to deliver PAMPs/MAMPs to quickly activate induced innate immune responses.

(102) Comparative evaluation of the survival of Family A and B strains after growth in Purple broth with arabinose and inoculation into medium without arabinose indicated rapid lysis and death of the Family A strain 9052 whereas the Family B strain 12499 increased in cell number for several cell divisions before lysis and death commenced (FIG. 3). These results were as expected and suggested that the Family A and B strains might be more efficacious in inducing innate immune responses due to this short-term proliferation in vivo prior to onset of lysis. While this could be beneficial, it was possible that the Family B strains might also be more inflammatory. These considerations thus guided the design of safe efficacious ENIIRA/SDAAS strains to recruit early onset innate immunity. Sal-Adj/ENIIRA/SDAAS strains derived from the Family A and B strains were made with combinations of genetic modifications to enable determining which mutations were beneficial or were marginal in benefit. This then enabled us to further optimize and enhance induced innate immune responses. The phenotypic properties associated with each mutation in the strains constructed are presented in Table 1 and Table 2 lists the suicide vectors used to introduce the various mutations into the strains constructed.

(103) b. Sal-Adj/ENIIRA/SDAAS Strains Constructed

(104) Table 4 lists the strains constructed from the Family A strain 9052. The properties associated with the mutations present are described in following sections.

(105) TABLE-US-00004 TABLE 4 Family A strain genotypes and derivations Chi number Genotype Parent 9052 alr-3 dadB4 asdA33 8901 12503 alr-3 dadB4 asdA33 fliC180 9052 12512 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE 12503 12515 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12512 12517 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12515 lpxR9 12553 alr-3 dadB4 asdA33 fliC180 (hin-fljBA)-219 12503 12554 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12517 lpxR9 (hin-fljBA)-219 12555 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12515 waaC41 12556 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12515 waaG42 12557 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12515 waaL46 12558 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12517 lpxR9 waaC41 12559 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12517 lpxR9 waaG42 12560 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 12517 lpxR9 waaL46

(106) Table 5 lists the strains constructed from the Family 2 strain 12499. The properties associated with the mutations present are described in following sections.

(107) TABLE-US-00005 TABLE 5 Family B strain genotypes and derivations Chi number Genotype Parent 12499 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12498 12504 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 AP.sub.dadB66::TT araC P.sub.BAD dadB 12499 fliC180 12513 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12504 fliC180 pagP81::P.sub.lpp lpxE 12516 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 AP.sub.dadB66::TT araC P.sub.BAD dadB 12513 fliC180 pagP81::P.sub.lpp lpxE pagL7 12518 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12516 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 12542 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 waaC41 12543 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 waaG42 12544 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 waaL46 12545 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12504 fliC180 waaC41 12546 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12504 fliC180 waaG42 12547 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12504 fliC180 (hin fljBA)-219 12548 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin fljBA)-219 12549 P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB 12504 fliC180 waaL46
c. Construction of Strains with Alterations in LPS Structure

(108) S. Typhimurium is a gram-negative bacterium that contains LPS in its outer membrane. One vital component of LPS is lipid A that has a repeating disaccharide with six attached lipid acyl chains. Lipid A constitutes the potent endotoxin that can cause sepsis and death. The level of sensitivity to this endotoxin varies considerably among animal species with cattle, horses, dogs and humans being far more sensitive than chickens and even mice (34). It is also the lipid A that interacts as an agonist with TLR4 to recruit an innate immune response. However, Salmonella has evolved as a successful pathogen so as to infect more successfully by reducing the ability of its lipid A to trigger innate immunity by modifications that reduce the agonist activity of its lipid A. This is accomplished by decorating the lipid acyl chains with small molecules. Since lysis of ENIIRA/SDAAS strains immediately release the LPS with the lipid A endotoxin, we constructed strains 12717 (Family A) and 12518 (Family B) with the pagP81::P.sub.lpp lpxE deletion-insertion mutation so that the strain synthesizes the non-toxic adjuvant mono-phosphoryl lipid A rather than the toxic lipid A while retaining its ability to activate TLR4 via the MD2 (rather than MyD88) pathway (33). Inactivation of the pagP gene is important since its gene product modifies lipid A to reduce its agonist activity. The promoter from the E. coli lipoprotein gene (lpp) is one of the strongest promoters since the lipoprotein synthesized is the most abundant protein in gram negative bacteria. It is used here to drive the constitutive expression of the lpxE gene from Francisella tularensis that eliminates a 4 phosphate decoration of the lipid A disaccharide that is responsible for endotoxicity. This alteration thus precludes excess inflammation due to release of endotoxin in vivo although release of other cell constituents in Family B strains that undergo several cell divisions in vivo prior to onset of lysis might still be too inflammatory at high doses.

(109) We therefore evaluated the relative attenuation/virulence of these two strains in mice by delivery of doses of 10.sup.4, 10.sup.5, 10.sup.6 and 10.sup.7 CFU by the i.v. route, doses of 10.sup.5, 10.sup.6, 10.sup.7 and 10.sup.8 CFU by the i.n, and s.c. routes, and 10.sup.9 CFU by the oral route. All mice survived challenge at all doses by all routes when infected with the Family A strain 12517. However, some mice died when infected with the Family B strain 12518 at doses above 10.sup.6 CFU by the i.v, and i.n. routes and above 10.sup.7 CFU by the s.c. route, while all mice survived oral inoculation. These results indicated that the several cell divisions in vivo of the Family B strains, due to the regulated delayed lysis in vivo attribute would necessitate animal studies using lower doses and ultimately for the need to introduce additional mutations to preclude excess inflammatory responses leading to mortality.

(110) The pagL7 and lpxR9 mutations were also included to eliminate two additional means that Salmonella uses to lessen recruitment of innate immunity by reducing the agonist activity of lipid A (15). The benefits of these two mutations to enhancing TLR4 recruitment is demonstrated below in studies with HEK cells displaying the TLR 4 ligand. The eptA and arnT mutations (Table 1) have also been included in some ENIIRA/SDAAS strains as they also encode enzymes that lessen the agonist activity of lipid A.

(111) Since some of the Family B derived strains administered at high doses by parenteral routes (i.v., s.c.) were too inflammatory, we constructed strains with deletion mutations in the waaC, waaG and waaL genes that eliminate synthesis of the middle LPS core, outer LPS core and LPS O-antigen, respectively, to confer complete attenuation. Since the waaC and waaG mutations reduce synthesis and display of flagella and might possibly reduce secretion of flagellin that is important for recruiting innate immunity via interaction with TLR5, the use of strains with the waaL mutation is preferred. Subsequently, we evaluated SDAAS strains such as 12626 and 12649 with the wbaP45 mutation as a different means to block coupling of the LPS O-antigen to the LPS core.

(112) d. Construction of Strains with Alterations in Flagella and Flagellin Release

(113) A non-flagellated strain 9028 with the fliC2426 and fljB217 mutations and strain 9026 in which the fliC2426 mutation is replaced with the fliC180 mutation to result in synthesis and release of a truncated flagellin that activates TLR5 (see Overview/Preliminary studies) were used in initial evaluations. We initially constructed Family A and B strains with the fliC180 mutation since initial results showed that this mutation enhanced production and release of the FliC flagellin that activates TLR5. In addition, the phase-lock mutation (hin-fljBA)-219 was introduced so that the FliC protein is made by all Sal-Adj/ENIIRA/SDAAS cells. The fljB mutation eliminates the ability of these strains to synthesize the Phase II flagellin and the fljA mutation eliminates synthesis of the repressor that would block transcription of the fliC gene that specifies the Phase I flagellin. A further enhancement in flagellin production was achieved, in one embodiment, by introducing another mutation (cfs) that causes constitutive high-level flagellin synthesis.

(114) As shown in FIG. 4, strains with the fliC180 mutation still display motility if they have a wild-type fljB gene but are unable to exhibit motility when a fljB mutation is present, which is achieved by inclusion of the phase-lock mutation (hin-fljBA)-219 that we introduced into many Family A and B strains. Nevertheless, all strains with the fliC180 mutation produce secreted flagellin as revealed by western blotting but this flagellin is unable to assemble into functional flagella to convey motility to cells.

(115) e. Construction of Strains with Alterations in Display of Fimbriae

(116) S. typhimurium has 12 operons that specify synthesis of fimbriae (18) and most of these are subject to regulation such that they are only synthesized in certain environments. We recently identified four fimbrial operons using the IVET system to identify in vivo expressed genes that are synthesized in the spleen after oral infection whereas the other eight fimbrial operons were not expressed (18). It was also determined that constitutive synthesis of the in vivo-synthesized Saf and Stc fimbriae significantly enhanced induction of protective immunity of a RASV/PIESV strain (18). Strains will therefore be evaluated for interaction with HEK cells and enhancing Ova immune responses lacking the other fimbrial operons expressed in vivo with either the deletion-insertion mutation P.sub.ste53::P.sub.murA stc or P.sub.saf55::P.sub.murA saf to cause constitutive synthesis of Ste or Saf fimbriae.

(117) f. Other Constructions

(118) In the initial comparison of the Family A, B and C parental lysis strains, derivatives were compared with the sifA26 mutation that eliminates a means of immunosuppression and enables S. Typhimurium cells to escape from the Salmonella-containing vesicle (SCV) such that lysis can occur in the cytosol. In RASV/PIESV strains that undergo regulated delayed lysis in vivo, escape from the SCV enables lysis of some vaccine cells to occur in the cytosol for delivery of protective antigens to the proteosome for Class I presentation and enhanced induction of CD8-dependent cellular immunity. Although individual cells of ENIIRA/SDAAS strains with the regulated delayed in vivo lysis attribute, can lyse prior to invasion into host cells or after invasion into cells whether contained in the SCV or after escape from the SCV (due to a sifA mutation) or after escape from a host cell undergoing pyroptosis, such lysis will release peptidoglycan components, RNA and DNA. Since peptidoglycan components activate Nod1 and Nod2, released RNA can activate TLR8 and CpG containing DNA sequences activate TLR9, the location of the release of these components will impact the level of activation of Nod1, Nod2, TLR8 and TLR9 that are located internally in host cells. On average, ENIIRA/SDAAS strains with the sifA26 mutation would therefore be expected to better activate Nod1, Nod2, TLR8 and TLR9.

(119) The recA mutation blocks genetic recombination but about 10 percent of cells at each cell division undergo Rec-less cell death since occasional recombination between replicating chromosomes leads to rapid unregulated degradation of DNA. Consequentially, ENIIRA/SDAAS strains with recA mutations should be more efficient in liberating DNA fragments containing CpG sequences to activate TLR9. An additional benefit is that recA mutants of S. Typhimurium are totally avirulent and do not induce disease symptoms when administered to animal hosts. We have therefore inserted the recA62 mutation into candidate ENIIRA/SDAAS strains.

(120) Some Salmonella pathogenicity island (SPI) 2 genes are effectors that can dampen induction of innate immune responses. The steE gene seems to be in this category such that ENIIRA strains unable to synthesize the steE gene product are investigated for impact on induction of innate immune responses using the HEK cell lines.

(121) g. Characterization of Constructed ENIIRA/SDAAS Strains

(122) All constructed ENIIRA/SDAAS strains are evaluated for the correctness of the introduced mutations and the phenotypes expected are confirmed using all the methods described in Example 1. In this regard, Family A strains such as 12517 commence to lyse as they attempt to grow in any medium lacking DAP and D-alanine such that cultures completely lyse within several hours after transfer from a permissive growth medium to a non-permissive growth condition in the absence of DAP and D-alanine (FIG. 3). Family B strains, however, will undergo several cell divisions after transfer to non-permissive media since the enzymes for synthesis of DAP and D-alanine that cease to be synthesized are diluted by half at each cell division until their concentration is insufficient to enable synthesis of sufficient DAP and D-alanine to enable peptidoglycan synthesis. This behavior is observed for 12518 (FIG. 3) during the first several hours of growth under non-permissive growth conditions. However, for cultures tested under high cell density lysis leads to destruction of the peptidoglycan cell wall layer to release its constituents DAP and D-alanine that are now free to be incorporated into peptidoglycan in surviving cells to cause an increase in viable cell titer. If, however, cultures of 12518 are inoculated into non-permissive growth conditions at densities of 110.sup.5 CFU or lower, death by lysis continues until the culture is devoid of viable cells.

(123) h. Discussion

(124) As each of the Sal-Adj/ENIIRA/SDAAS strains described above is constructed its properties are being fully validated. The abilities of these strains in comparison to parent and control strains to stimulate NF-B production in HEK cells having different TLR and Nod factors were evaluated (Example 3). These results also led to synthesis of strains with combinations of mutations and the conduct of animal studies to determine which combination of mutations yields a Sal-Adj/ENIIRA/SDAAS strain(s) with greatest ability to enhance immune responses to Ova (Example 4) and protection against Mtb (Example 5).

Example 3. Evaluation of Sal-Adj/ENIIRA/SDAAS Strains to Activate Synthesis of NF-kB Production in HEK293 Cells

(125) Strains constructed in the Example 2 research were initially tested and compared for their ability to interact with and stimulate HEK293 cells with TLR1, TLR2, TLR4 (two types), TLR5 (two types), TLR6, TLR9, Nod1 and Nod2 using methods described in Example 1 M&M. Most evaluations have been with Family A strains since they commence to lyse immediately upon placing in non-permissive conditions in the absence of arabinose (see FIG. 3 above) such that lysis would occur during and after invasion of cells. In this regard, Family B strains are able to undergo several cell divisions after placing in non-permissive conditions in the absence of arabinose such that they will not be lysing after adding to HEK cells. This is because Salmonella cell division in infected cells ranges from 8 to more than 20 hours depending on cell type compared to less than an hour for most in vitro growth conditions. These inferences are validated by the results presented in FIG. 2 above in which the Family A strain 9052 is more active in recruiting a response in HEK cells with the internal Nod1 receptor than is the Family B strain 12499 (See FIG. 2B).

(126) Results observed in evaluation of several Family A and Family B strains in assays of NFB recruitment in HEK cells displaying TLR4 are presented in FIG. 5. In these studies, we varied the multiplicity of infection (MOI) of ENIIRA strains to HEK cells from 0.001:1 to 10:1. Based on these and other evaluations, Family B strains such as 12518 were better recruiters than Family A strains such as 12517. In addition, removal of the O-antigen or parts of the LPS core further enhanced recruitment of NFB (compare 12518 versus 12542 and 12543. All ENIIRA/SDAAS strains were more active in TLR4 activation than the wild-type parent 3761.

(127) Results observed in evaluation of several Family A and Family B strains in assays of NFB recruitment in HEK cells displaying TLR5 are presented in FIG. 6. We also include data for control strains in FIG. 6 that evaluate mutant strains derived from the wild-type parent strain 3761. The only strain unable to activate TLR5 is 9028 that is unable to synthesize flagellin or flagella due to the fliC2426 and fljB217 mutations. Thus, all the strains with the fliC180 mutation activate TLR5 even though many are non-motile due to the fljB deletion mutation.

(128) As noted above in reference to the FIG. 2 data, Family A strains are better at activating Nod1 than are Family B strains due to their more rapid lysis after entry into cells. Nevertheless, Family B strains with either waaC41 or waaG42 with or without the pagP81::P.sub.lpp lpxE deletion-insertion mutation are proficient in activating Nod1 and Nod2 (FIG. 7). This unexpected result may be due to the sensitivity of strains lacking the LPS O-antigen and especially parts of the outer LPS core polysaccharide since they are very sensitive to complement-mediated cytotoxicity and other host natural defense barriers such as acidification and production of defensins.

Example 4. Evaluation of Sal-Adj/ENIIRA/SDAAS Strains by s.c., i.n, and Oral Routes with s.c. Administration of Ova Suspended in Buffered Saline or Alum and Monitoring Antibody Responses to Ova

(129) Because of total avirulence, we concentrated on testing the adjuvant activity of the Family A parental strain 9052. The results presented in FIG. 8 reveal very clearly that co-administration of the ENIIRA strain with Ova was far superior to administering Ova with the standard Alum adjuvant in eliciting serum IgG antibodies against Ova. Studies with Family B strains such as with the avirulent derivative 12544 (Table 5) have just commenced with the expectation that they will likely be better than Family A strains in these assays measuring enhanced antibody production to subunit vaccines. In this regard, because of the ability of Family B strains to undergo several rounds of cell division in vivo prior to lysis, we expect them to be effective at lower doses that needed for Family A strains

Example 5. Evaluation of the Ability of ENIIRA/SDAAS Strains to Augment Ability of BCG to Diminish Infection of Mice with M. tuberculosis H37RV (Mtb) and Whether Also Enhances the Ability of a PIESV Construct Delivering Mtb Protective Antigens to Further Protect Against Mtb Infection and Proliferation

(130) Three experiments have been conducted in which ENIIRA/SDAAS strains have been administered in combination with M. bovis BCG, which is currently the only vaccine approved for human use to prevent infections by M. tuberculosis and development of TB. In all of these experiments, BCG was administered s.c. with 510.sup.4 CFU of BCG, either alone or in combination with ENIIRA/SDAAS strains and/or the PIESV 12068 (pYA4891). All immunizations/inoculations were given once. All experiments included a group of mice that were administered 100 l of phosphate-buffered saline PBS on Day 0; these were unimmunized control mice. Thirty-five days after immunization, all mice were challenged with a low-dose aerosol of virulent M. tuberculosis H37Rv, such that each mouse received 50-100 bacteria per lung. On day 28 in Experiments 15 and 20 (FIGS. 10 and 11), blood was collected from all mice to determine antibody titers to the mycobacterial Antigen 85A, which is produced by both BCG and M. tuberculosis. In Experiments 15 and 20, two mice from each group were euthanized on day 28, their lungs and spleens homogenized and processed for flow cytometry, to evaluate T-cell responses to Ag85A and, in Experiment 20, to two M. tuberculosis antigens (ESAT-6 and CFP-10) produced by the PIESV 12068 (pYA4891) construct. Approximately 6 weeks after challenge, mice were euthanized, lungs and spleens were removed, homogenized and plated for CFU determinations.

(131) In Experiment 14 (FIG. 9), the ENIIRA/SDAAS strain was 12499 (Family B) (P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB), which was administered by itself or in combination with BCG by either s.c. administration of 110.sup.5 CFU of 12499 or by i.v. administration of 110.sup.4 CFU of 12499. Only M. tuberculosis CFU determinations were done in this experiment and we found that mice that had been immunized with the combination of BCG+12499 administered i.v. had CFUs of approximately 210.sup.4 in their spleens, compared to titers of 110.sup.5 CFU in the spleens of mice immunized with BCG alone and titers of 310.sup.5 CFU in the spleens of the unimmunized control mice, suggesting that administration of 12499 by the i.v. route enhanced the protection afforded by BCG (FIG. 9). In the lungs of these mice, co-administration of 12499 (i.v.) with BCG yielded CFU titers that were lower than the unimmunized control mice (510.sup.5 for immunized mice compared to 210.sup.6 CFU for unimmunized mice), but were not as low as mice immunized with BCG alone (410.sup.5 CFU).

(132) In Experiment 15 (FIG. 10), co-administration of BCG with two ENIIRA/SDAAS strains (12517 (Family A) (alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9) and 12518 (Family B) (P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9)), both delivered by i.v. (510.sup.4 CFU) or the PIESV 12068 (pYA4891) (P.sub.murA25::TT araC P.sub.araBAD murA asdA27::TT araC P.sub.araBAD c2 (wza-wcaM)-8 pmi-2426 relA197::araC P.sub.araBAD lacI TT recF126 sifA26 waaL46 pagL12::TT araC P.sub.araBAD waaL+M. tuberculosis antigens Ag85A, ESAT-6 and CFP-10)), delivered by i.v. (510.sup.4 CFU) was evaluated. In this experiment, the CFU titers of M. tuberculosis in the lungs of mice immunized with the combination of BCG+12068 (pYA4891) were 5.510.sup.4 CFU, compared to titers of 210.sup.5 CFU in BCG-immunized mice and 210.sup.6 CFU in unimmunized control mice. Mice immunized with BCG+12517 or BCG+12518 had titers that were the same as mice immunized with BCG alone (210.sup.5 CFU). In the spleens of the mice in this experiment (FIG. 10), the titers of M. tuberculosis in mice immunized with the combination of BCG+12518 were 1.510.sup.3 CFU, compared to titers of 410.sup.4 CFU in mice immunized with BCG+12068 (pYA4891) and approximately 610.sup.4 CFU in mice immunized with BCG alone or BCG+12517. The titers in the unimmunized control mice were 2.510.sup.5 CFU. Flow cytometry analyses of the percentage of total T cells and dendritic cells (DCs) secreting Interferon- (IFN-) in response to Ag85A indicated that the highest percentage of IFN--secreting T cells and DCs were from the spleens of mice immunized with BCG+12518, although all immunized mice had higher percentages of IFN--secreting T cells and DCs in their spleens than the unimmunized control mice.

(133) In Experiment 20 (FIG. 11), we repeated the comparisons of mice immunized with BCG+12068 (pYA4891), delivered i.v., and BCG+12518, delivered i.v., compared to mice immunized with BCG alone and to unimmunized mice. We also compared mice immunized with BCG+12068 (pYA4891)+12518, both delivered by i.v. injection of a total of 510.sup.4 CFU and mice immunized with BCG+12068 (pYA4891), delivered by intranasal (i.n.) administration of 110.sup.5 CFU, plus 12518 delivered by i.v. injection (510.sup.4 CFU). (The titers of M. tuberculosis in the spleens of the mice from this experiment are provided in Example 8.) FIG. 11 shows the CFU titers of M. tuberculosis H37Rv in the lungs. Mice vaccinated with BCG alone or BCG+12068 (pYA4891) had titers that were more than one log lower than the CFU titers in the unimmunized mice. Mice immunized with the combination of BCG+12068 (pYA4891)+12518 (both delivered i.v.) had titers that were two logs lower than the titers in the unimmunized mice. Mice that were immunized with the combination of BCG+12518 or BCG+12068 (pYA4891) (delivered i.n.)+12518 (delivered i.v) also had significantly lower CFU titers in their lungs, compared to unimmunized mice. These collective results were most surprising since the highly significant enhancement of the efficacy of BCG in reducing Mtb multiplication has never been previously observed. Thus, the ability of ENIIRA/SDAAS strains to enhance the effectiveness of BCG vaccination could have a profound global benefit in prevention of M. tuberculosis infections. This is highly significant since tuberculosis is now the number one cause of global deaths from any infectious disease. We have determined total IgG antibody responses in the sera of these mice to Ag85A, ESAT-6 and CFP-10. We found that IgG levels to Ag85A were higher in mice immunized with combinations of BCG and an ENIIRA/SDAAS strain or PIESV 12068 (pYA4891), compared to mice immunized with BCG alone, with the highest levels in mice immunized with BCG+12068 (pYA4891)+12518, regardless of whether 12068 (pYA4891) was delivered by i.v. or by i.n. inoculation. We also found that mice immunized with BCG+12068 (pYA4891), delivered i.n., +12518 had the highest levels of total IgG against ESAT-6 and CFP-10. We also conducted flow cytometric analyses on the spleens and lungs of mice in this experiment. In the spleens, mice immunized with the combination of BCG+12068 (pYA4891) delivered i.n.+12518 had the highest percentages of IFN--secreting CD4.sup.+ T cells responding to ESAT-6, compared to mice in each of the other groups. Mice immunized with that combination also had the highest percentages of IFN--secreting CD8.sup.+ T cells responding to Ag85A, ESAT-6 and CFP-10. In the lungs of these mice, immunization with the combination of BCG+12068 (pYA4891) delivered i.n.+12518 induced the highest or close to highest levels of IFN--secreting CD8.sup.+ T cells in response to Ag85A, ESAT-6 or CFP-10 among all of the groups of immunized mice. In both lungs and spleens, mice immunized with the combinations of BCG+12518 strains or BCG+12068 (pYA4891) or BCG+12518+12068 (pYA4891) all induced higher percentages of IFN--secreting CD4.sup.+ and CD8.sup.+ T cells than mice immunized with BCG alone.

(134) The results from the three experiments described above demonstrate that co-administration of BCG with ENIIRA/SDAAS strains plus 12068 (pYA4891) enhances the ability of BCG to protect mice against aerosol challenge with M. tuberculosis and that co-administration enhances both antibody and T-cell responses that is likely to contribute to protection against challenge. If found to be true in humans, the impact will be highly significant.

(135) Splenomegaly: In experiments in which mice were immunized subcutaneously with M. bovis BCG alone or in combination with ENIIRA/SDAAS strains or the PIESV 12068 (pYA4891) or both an ENIIRA strain and 12068 (pYA4891), splenomegaly was observed in mice immunized with BCG and an ENIIRA strain. In Experiment 14, where the ENIIRA strain was 12499 (Family B), 2 out of 9 mice immunized with 12499 (delivered s.c.) alone, 2 out of 10 mice immunized with the combination of BCG+12499 (both delivered s.c) and 7 out of 8 mice immunized with the combination of BCG (delivered s.c.)+12499 (delivered i.v.) had significantly enlarged spleens, compared to the spleens of control mice (PBS administered s.c) and mice immunized with BCG alone, in which none of the mice had enlarged spleens. In Experiment 15, in which mice were immunized with BCG alone (delivered s.c.) or BCG (delivered s.c.) in combination with the PIESV 12068 (pYA4891) or Family A ENIIRA strain 12517 or Family B ENIIRA strain 12518 (each administered i.v), only mice immunized with the combination of BCG+12518 had enlarged spleens. In Experiment 20, mice were immunized with BCG alone (delivered s.c.) and in combination with the PIESV 12068 (pYA4891), delivered i.v., the Family B ENIIRA strain 12518 (delivered i.v.), 12068 (pYA4891) and 12518 (both delivered i.v.) or 12068 (pYA4891), delivered intranasally, and 12518, delivered i.v. All groups of mice immunized with 12518 (which was always administered i.v.) had mice with enlarged spleens. At present, we do not know why some mice in these three experiments developed enlarged spleens.

(136) However, these results are in accord with other observations that indicate that the Family B strain 12518 (P.sub.asdA55::TT araC P.sub.BAD asd alr-3 P.sub.dadB66::TT araC P.sub.BAD dadB fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9) is possibly too inflammatory because it multiplies too many cell divisions prior to lysis. Although this can be addressed by using lower doses, we are now testing the benefits of including the waaL46 (12544), wbaP45 and with or without sifA26 and/or recA62 mutations on ensuring complete attenuation while retaining and even augmenting the beneficial adjuvant activities.

Example 6. Modification of ENIIRA/SDAAS Strains to Display PAMPS/MAMPs that Activate Innate Immune Receptors Displayed by Diverse Bacterial, Viral, Fungal and Parasite Pathogens

(137) Our group has displayed capsular polysaccharides specified by genes from gram-negative and gram-positive bacterial species on the surface of Salmonella strains. We have also expressed lipo-proteins and protein appendages encoded by genes from diverse pathogens. In addition, since the ENIIRA/SDAAS strains are designed to lyse, they can liberate plasmids engineered to display single-stranded and double-stranded RNAs as displayed by and serving as PAMPS for RNA viruses and for RNA released by living pathogenic microbes. It is thus possible to modify ENIIRA strains to induce innate immune responses that could differ and be more appropriate and efficacious in enhancing immunity to be induced by a diversity of vaccines targeting prevention of infection by diverse bacterial, viral, fungal and parasite pathogens.

Example 7. Non-Specific Protection Against Infection by Other Pathogens

(138) It has been found that administration of the Sal-Adj/ENIIRA/SDAAS constructs induces low-level protective immunity to challenge of unvaccinated animals to various bacterial, viral and parasite pathogens as revealed by the data in Table 3 in which the empty vector PIESV strains are representative of Family C ENIIRA strains. These results are impactful in the protection, in some examples, of military personnel and civilians against a biothreat as well as to contend with epidemics. These ENIIRA strains will also have utility in augmenting levels of protective immunity of subunit and killed vaccines and even of attenuated vaccines that do not induce robust protective immunity of long duration, such as BCG (see Example 5). It can also be expected that the level of innate immunity induced by administering ENIIRA strains will be of reasonably long duration. In fact, these strains will probably be effective in inducing memory innate immune responses. The data in Table 3 supports this since the challenge infections to which protection was conferred by vaccination with empty vector PIESV strains were 30 or more days after primary vaccination.

(139) Note: During the course of research to discover and then improve use of live self-destructing Salmonella as adjuvants to enhance recruitment of innate immunity to augment eventual levels of acquired immunity, we simplified terminology from calling these Sal-Adj strains to ENIIRA strains to the simpler and more descriptive Self-Destructing Attenuated Adjuvant Salmonella (SDAAS) strains as used in the following examples.

Example 8. Additional Data Pertaining to Studies Demonstrating Enhancement of Levels of Protection Against Mycobacterium tuberculosis Challenge of Mice Inoculated with Combinations of BCG with SDAAS Strain and a PIESV Delivering Mtb Antigens

(140) Studies on induction of immunity to M. tuberculosis (Mtb) infections are time-consuming due to the very slow growth of Mtb such that colony counts to determine bacterial densities require 4 weeks or so of incubation to obtain data. Thus, data from Experiments 14, 15 and 20 described in Example 5 continued for several months after the PCT application was filed. Analyses of induced immune responses in mice used in these studies also continued. FIG. 12 presents data on immune responses induced in Experiment 15. In this study, spleens were harvested and total CD3+ T cells were analyzed for ability to undergo proliferation in response to the mycobacterial antigen Ag85A using the CFSE dye dilution technology. Compared to the PBS control, lymphocytes from mice stimulated with BCG and the Family B strain 12518 showed slightly more proliferation than the T cells from mice co-inoculated with BCG and the Family A strain 12517. This same trend was also observed in regard to the ability of Ag85A to stimulate production of INF (FIG. 12). In accord with these findings in Experiment 15, we had also evaluated the 12517 and 12518 cultures for their relative abilities to activate synthesis of NF-B by interaction with TLR 4 and TLR5 displayed by HEK cells. Results shown in FIG. 13, demonstrate the clear superiority of the Family B strain 12518 for activation via both TLR4 and TLR5 and also does so in comparison to pure preparations of LPS and flagellin.

(141) At the time of filing the PCT application, there were no colonies on plating undiluted spleen suspensions from mice inoculated with BCG and the Family B SDAAS strain 12518 independent of receiving the PIESV strain 12068 (pYA4891). We thought that maybe there was an inhibition in rate of growth due to the inoculation treatments so we only presented data from the lung suspension cultures (FIG. 11) and continued the incubation of plates for an additional two weeks (6 weeks in total). Still no colonies appeared on the plates with the undiluted spleen suspensions and we thus represent the data from lung cultures in FIG. 11 on an expanded ordinate scale in the left panel of FIG. 14 also displaying the results from the Mtb titers in the spleens of mice subjected to the different treatments in the right panel (using the same ordinate scale). The surprising and unexpected results were that combination of s.c. administration of BCG with i.v. administration of the Family B SDAAS strain 12518 at a dose of 510.sup.4 CFU resulted in apparent inability of the aerosolized Mtb H37Rv to successfully traverse from the lung to systemic tissues such as the spleen or to be precluded from survival if getting to the spleen. Although induction of antibodies against mycobacterial proteins might not contribute to protective immunity against Mtb infections as would cellular immune responses, the co-administration of the SDAAS strain 12518 did enhance induction of antibody responses against Mtb antigens (FIG. 15). It should be emphasized that to the best of our knowledge the results presented in FIG. 14, demonstrating prevention of Mtb to colonize spleens after Mtb aerosol infection of vaccinated mice have never been observed by any investigator for any anti-Mtb vaccine. This offers the hope that co-administration of an SDAAS strain with BCG or some modified BSG or some other anti-Mtb vaccine could result in protective immunity to pulmonary tuberculosis in humans (a feat that BCG immunization alone is incapable of accomplishing). This, of course, will require evaluation in human clinical trials.

Example 9. Further Modifications of Family A and B SDAAS Strains to Investigate Effects of Additional Mutational Alterations to Enhance Safety and Effectiveness of Adjuvant Activities

(142) As described in Example 5, inoculation of the Family B SDAAS strain 12518 at a dose of 510.sup.4 CFU i.v. caused significant splenomegaly that was not observed with oral or s.c inoculation and at lower doses by the i.v. route. As a consequence, we commenced to construct derivatives of the Family A strain 12517 and the Family B strain 12518 with additional mutations to enhance recruitment of innate immunity while reducing potential toxicity upon lysis of the strains in vivo. Many of these SDAAS strains with such modifications were listed in Tables 4 and 5 in Example 2 and additional SDAAS strains were constructed after filing of the PCT application. These newer AAS strains are included in Table 8 with the new mutations and their associated genotypes and phenotypes described in Table 6. Table 7 lists the suicide vectors used to generate the new AAS strains whose properties are presented and discussed in the following Examples.

(143) Table 6. Additional Mutations with Associated Phenotypes Enhancing Adjuvant Activity and/or Safety of AAS Strains

(144) A. Deletion and Deletion-Insertion Mutations to Facilitate Regulated Delayed Lysis In Vivo

(145) P.sub.asdA::TT rhaRS P.sub.rhaBAD asdA and P.sub.asdA::TT P.sub.rhaBAD asdA make synthesis of AsdA dependent on presence of rhamnose. P.sub.dadB::TT rhaRS P.sub.rhaBAD dadB and P.sub.dadB::TT P.sub.rhaBAD dadB make synthesis of DadB dependent on presence of rhamnose.

(146) In one case the repression-activation of the P.sub.rhaBAD promoter can use the encoded E. coli rhaRS gene products or the endogenous Salmonella rhaRS gene products whereas in the case in which the E. coli rhaRS genes were excluded from the inserted regulatory cassette, the P.sub.rhaBAD is subject to control by only the Salmonella rhaRS gene products.

(147) C. Mutations Altering Synthesis of LPS Components

(148) lpxR::P.sub.lpp lpxF mutation causes regulated delayed in vivo synthesis of the codon-optimized lpxF gene from Francisella tularensis to cause synthesis of the non-toxic adjuvant form of LPS lipid A that lacks the 4 phosphate on lipid A. The lpxR mutation eliminates a means by which Salmonella alters LPS lipid A in vivo to decrease recruitment of innate immunity by interaction with TLR4 (16) pmi eliminates gene for phosphomannose isomerase such that LPS O-antigen synthesis is dependent on presence of non-phosphorylated mannose in growth medium (47) rfc eliminates an enzyme that couples units of LPS O-antigen composed of 4 or 5 sugars to result in long-chain LPS O-antigen polymers (48) wbaP eliminates an enzyme needed to couple the first LPS O-antigen subunit onto the LPS core (49)
H. Mutations Enhancing Adjuvant Activity and Safety of Strains sopB decreases induction of inflammatory response for strains delivered on a mucosal surface (43) and increases induction of mucosal immunity (44) sopF enables Salmonella to escape the SCV (45) sseL decreases induction of pyroptosis (46) tlpA decreases induction of pyroptosis (46) purA imposes an obligate requirement for non-phosphorylated adenine that is essential and absent in animal tissues; is thus totally attenuating (50)

(149) The suicide vectors used to introduce the mutations listed in Table 6 into new AAS strains are listed in Table 7.

(150) The new SDAAS strains constructed are listed in Table 8.

(151) TABLE-US-00006 TABLE 7 Suicide vectors for constructing SDAAS strains with mutations listed in Table 6 A. Deletion and deletion-insertion mutations to facilitate regulated delayed lysis in vivo P.sub.dadB66::TT rhaRS P.sub.rhaBAD2 dadB pG8R340 Cm P.sub.dadB66::TT rhaRS P.sub.rhaBAD1 dadB pG8R352 Cm P.sub.dadB66::TT P.sub.rhaBAD1 dadB pG8R353 Cm P.sub.asdA55::TT rhaRS P.sub.rhaBAD1 asdA pG8R354 Cm P.sub.asdA55::TT P.sub.rhaBAD1 asdA pG8R355 Cm C. Mutations altering synthesis of LPS components lpxR:P.sub.lpp lpxF pYA4289 Cm pmi pYA3546 Tet rfc pYA4717 Cm wbaP pYA4899 Cm H. Mutations enhancing adjuvant activity and safety of strains sopB pYA3733 Cm sopF sseL pYA4621 Cm tlpA pYA4620 Cm purA pG8R126 Cm

(152) TABLE-US-00007 TABLE 8 Newly constructed AAS strains Family A strains (genotypes and derivations) 12565 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 12517 recA62 12606 alr-3 dadB4 asdA33 fliC180 (hin-fljBA)-219 pagP8 12553 12608 alr-3 dadB4 asdA33 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 12517 sifA26 12625 alr-3 dadB4 asdA33 fliC180 (hin-fljBA)-219 pagP8 lpxR9 12606 12629 alr-3 dadB4 asdA33 fliC180 (hin-fljBA)-219 pagP8 lpxR9 12625 pagL7 12638 alr-3 dadB4 asdA33 fliC180 (hin-fljB)-219 pagP8 lpxR9 12629 pagL7 eptA4 12640 alr-3 dadB4 asdA33 fliC180 (hin-fljBA)-219 pagP8 lpxR9 12638 pagL7 eptA4 arnT6 12650 alr-3 dadB4 asdA33 fliC180 (hin-fljBA)-219 pagP8 lpxR9 12640 pagL7 epta4 arnT6 sifa26 12661 alr-3 dadB4 asda33 fliC180 (hin-fljBa)-219 pagP8 lpxR9 12650 pagL7 epta4 arnT6 sifa26 recA62 Family B strains (genotypes and derivations) 12564 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12504 fliC180 reca62 12566 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 recA62 12567 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12544 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 waaL46 recA62 12568 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12549 fliC180 waaL46 recA62 12570 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 arnT6 12571 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12548 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 12583 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB .sub.PasdA55::TT araC P.sub.araBAD asdA 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 eptA4 12584 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12548 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 eptA4 12585 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12571 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 12586 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12570 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 arnT6 eptA4 12603 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12547 fliC180 (hin-fljB)-219 pagP8 12604 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12548 fliC180 pagP8 pagL7 lpxR9 (hin-fljB)-219 12605 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12603 fliC180 (hin-fljBA)-219 pagP8 pagL7 12609 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12518 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 sifA26 12610 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12604 fliC180 pagP8 pagL7 lpxR9 (hin-fljBA)-219 sifA26 12611 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12504 fliC180 sifA26 12612 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12585 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 12620 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12612 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 pmi-2426 12621 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12612 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 rfc-112 12623 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12604 fliC180 pagP8 pagL7 lpxR9 (hin-fljBA)-219 eptA4 12626 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12612 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 wbaP45 12639 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA 12623 fliC180 pagP8 pagL7 lpxR9 (hin-fljBA)-219 eptA4 arnT6 12641 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12612 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 waaL46 12648 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadBxx::TT rhaSR P.sub.rhaBAD dadB 12612 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 12649 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadBxx::TT rhaSR P.sub.rhaBAD dadB 12626 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 wbaP45 12668 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12612 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 recA62 12669 P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB 12626 fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 wbaP45 recA62

Example 10. Maximizing Synthesis and Delivery of a FliC Flagellin Subunit to Interact with TLR5 to Activate Induction of Innate Immunity

(153) The FliC180 flagellin subunit was derived by an internal deletion in the fliC gene that eliminated 180 amino acids with retention of the TLR5 binding domain. Also, this FliC180 subunit is secreted by the flagellar type 3 secretion system and since it does not polymerize into flagella, is available for interaction with cell-surface localized TLR5 molecules. It should be emphasized that it is flagellin and not flagella with capability of interacting with TLR5. Thus, SDAAS strains synthesizing and secreting the FliC180 subunit are more active in activating NF-B in HEK cells displaying TLR5 than are wild-type strains synthesizing flagella (see FIG. 6 in Example 3) but are unable to exhibit motility (see FIG. 4). The wild-type motile strains are still capable of interacting with TLR5 however since not all flagellin is incorporated into flagella such that some free flagellin molecules exist to bind to TLR5. S. Typhimurium exhibits phase variation with regard to synthesis and display of flagellar antigens and shifts from synthesis of FliC to synthesis and assembly of FljB. The hin invertible element makes this transition about once in every 10.sup.4 cells in the population. Since we desired to have 100% of the SDAAS cells synthesizing and secreting the FliC180 subunit protein, we constructed the (hin-fljBA)-219 phase-lock deletion mutation that also eliminates synthesis of the FljB flagellin and the FljA protein that is a repressor blocking transcription of the fliC gene in SDAAS strains. FIG. 16 presents western blot data showing the synthesis of the truncated FliC protein and the presence of the full-length FljB protein for strains without the (hin-fljBA)-219 mutation. Since the N- and C-terminal ends of the FliC and FljB flagellins are essentially the same, the antibodies generated against the FliC180 subunit protein react equally well with the FljB and FliC flagellins. Note that the amount of FliC180 protein made is often reduced in cultures of those strains making the FljB flagellin. Also note that the wild-type parent 3761 makes the full-length FliC flagellin. Although not shown, these same strains were analyzed with an antiserum against the middle portion of the full-length FliC flagellin and failed to react with the flagellins produced by all the strains with the fliC180 mutation. Also, all SDAAS strains used for the data presented in FIG. 16 activated NF-B synthesis in HEK cells displaying TLR5.

(154) FIG. 17 presents data on motility of all the strains used to obtain the FIG. 16 data. Thus, all stains with the fliC180 and (hin-fljBA)-219 mutations were non motile whereas the fliC180 mutant strains with the wild-type fljB sequence displayed motility. Collectively, all these results indicate that SDAAS strains with both fliC180 and (hin-fljBA)-219 mutations will be better adjuvants in stimulating innate immunity than strains without these mutations.

Example 11. Modifications in Synthesis of LPS Components Including Modifications of the Lipid A Endotoxin to Enhance Recruitment of Innate Immune Responses and Reduce Toxicity

(155) In Example 3 we noted that waaC and waaG mutations that eliminate the LPS O-antigen and truncate the LPS core enhance the ability of the Family B SDAAS strain 12518 to exhibit enhanced abilities to activate NF-B synthesis in HEK cells displaying TLR4 (FIG. 5). This was a positive result in that such mutants would display decreased virulence by being more readily phagocytized and more sensitive to complement mediated cytotoxicity. We therefore believed that such alterations would be doubly beneficial by reducing toxicity and induction of excessive splenomegaly while enhancing recruitment of innate immunity. However, it was surprisingly discovered that strains like 12542 and 12543 were less able to activate NF-B synthesis in HEK cells displaying TLR5 and also decreased display of motility. We thus, devised other means to lessen toxicity while enhancing stimulation of innate immunity. Nevertheless, this observation that impairment in LPS core synthesis inhibited secretion of flagellin to preclude assembly and function of flagella is apparently a new discovery.

(156) FIG. 18 diagrams the Salmonella lipid A molecule and indicates the gene products that modify the structure. The 1 and 4 phosphate groups contribute to invasiveness and toxicity. A synthetic derivative called MPLA for mono-phosphoryl lipid A is less toxic than lipid A and is used as an adjuvant for many vaccines currently under development. Some years ago, we expressed the Francisella tularensis lpxE gene in S. Typhimurium strains (15) to result is strains that produced mono-phosphoryl lipid A and we introduced the mutation pagP81::P.sub.lpp lpxE into many SDAAS strains including 12518, which nevertheless exhibited unwanted toxicity causing splenomegaly in mice inoculated with this SDAAS strain and BCG (see Examples 5 and 8). We therefore generated strains with various combinations of mutations as well as strains with single mutations pagP8, pagL7, lpxR9, arnT6 and eptA4 since the products of these genes are used by Salmonella to modify lipid A to reduce its ability to interact with TLR4 and thus suppress induction of immunity and/or contribute to its toxicity. We therefore investigated the strains listed in Table 9 for toxicity using the Pierce Endotoxin Quantitation kit.

(157) TABLE-US-00008 TBLE 9 S. Typhimurium mutant strains evaluated for levels of endotoxicity 1. 9429 eptA4 2. 9432 arnT6 3. 9486 arnT eptA 4. 9883 pagL7 pagP8 lpxR9 arnT6 5. 9904 pagL7 pagP8 lpxR9 amT6 eptA4 6. 9430 pagL7 7. 9433 lpxR9 8. 9434 pagP8 9. 3761 wild-type UK-1 parent

(158) For these assays, strains were grown in LB broth overnight at 37 C. with aeration (180 rpm) and then diluted 1:10 into prewarmed LB broth and grown with continued aeration to an OD600 of 1.0 (110.sup.9 CFU/ml). Bacterial cells were harvested by centrifugation, resuspended in endotoxin free water and disrupted by sonication. Cell lysates were diluted, distributed in 96 well plates and the successive reactions conducted according to the instructions provided with the Pierce Kit. Reactions were stopped and the results evaluated in a microplate absorbance reader at 405 nm. The results are presented in FIG. 19. All five of these mutations reduce endotoxicity with the combination lpxR9, arnT6 and eptA4 giving the lowest endotoxin levels. Unexpectedly, inclusion of all 5 mutations in 9904 displayed the lowest endotoxin activity. The surprising result depicted in FIG. 20 was the finding that no combination of these five mutations had an appreciable effect on interacting with TLR4 on HEK cells to activate NF-B synthesis. The Family B SDAAS strain 12585 alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 (Table 8) has these five genes deleted and, in addition, has the P.sub.lpp lpxE inserted into the pagP mutation that should further decrease toxicity.

Example 12. Properties Associated with Constructed SDAAS Strains that should Further Enhance Adjuvant Activities and Safety Attributes

(159) The sifA26 mutation enables Salmonella to escape the Salmonella containing vesicle (SCV) also termed the endosome and this enables all the SDAAS strains with programed abilities to undergo lysis in vivo to escape from the SCV to lyse in the cytosol. This lysis then liberates peptidoglycan constituents and DNA to interact with Nod1, Nod2, TLR8 and TLR9 that are located on internal cell surfaces (and are not available if lysis occurs within the SCV). It would thus be expected that Family A SDAAS strain 12608, 12650 and 12661 (Table 8) and the Family B SDAAS strains 12609, 12610, 12611, 12612, 12620, 12621, 12626, 12641, 12648, 12649, 12668 and 12669 (Table 8) that all possess the sifA26 mutation would exhibit an enhanced ability to activate NF-B synthesis compared to SDAAS strains without this mutation. An added benefit is that S. Typhimurium strains with the sifA26 mutation are more than 10,000 times less virulent when orally administered to mice compared to the wild-type parent. It is also possible that addition of a sopF mutation (Table 6) in place of or in addition to the sifA26 mutation will also be beneficial since the sopF mutation also enables Salmonella to escape from the SCV to lyse in the cytosol.

(160) The recA62 mutation eliminates the ability of Salmonella to carry out genetic recombination and results in increased sensitivity to UV and other stresses. The mutation also renders Salmonella more than 10,000 times less virulent than the wild-type parent after oral inoculation into mice. Importantly, genetic recombination is lethal with approximately 10% of the cells in the population degrading their entire DNA content as a consequence of progeny strand recombination. If this occurs in lysing AAS cells after escape from the SCV, there should be an increased release of CpG sequences targeting the internal TLR9 to activate NF-B synthesis. This recA62 mutation has thus been introduced into the Family A SDAAS strain 12565 and 12661 the Family B SDAAS strains 12566, 12567, 12568, 12668 and 12669 (Table 8) to enable testing of these expectations.

(161) As stated in Example 10 and depicted in FIG. 18, the expression of the F. tularensis lpxF gene can eliminate the 4 phosphate on lipid A to reduce toxicity. As indicated in Table 7, we have generated the lpxR::P.sub.lpp lpxF mutation and have the suicide vector pYA4289 to enable the introduction of this deletion-insertion mutation into a SDAAS strain to conduct tests as described in Example 10. We thus can compare toxicity and adjuvant activities of SDAAS strains with expression of neither, either of both F. tularensis lpxE and lpxF genes.

(162) Although we observed that inability to synthesize the LPS O-antigen and part of the LPS core increased interaction with TLR4 (FIG. 5), mutations such as waaC41 and waaG42 caused an inhibition in FliC180 secretion and thus had an adverse effect on recruiting innate immunity via TLR5. We thus generated derivatives of Family B SDAAS strains such as 12585 (alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB P.sub.asdA55::TT araC P.sub.araBAD asdA fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4) or its derivative 12612 with the sifA26 mutation (Table 8) by addition of the pmi-2426 (12620), rfc-48 (12621), wbaP45 (12626) and waaL46 (12641) mutations (Table 6). These strains are described in Table 8. The wbaP45 and waaL46 mutations both block addition of the LPS O-antigen to the LPS core while the pmi-2426 mutation blocks the synthesis of GDP-mannose such that only the first two sugars of the first O-antigen subunit are attached to the LPS core and the rfc-48 mutation allows addition of the first O-antigen subunit to the core but precludes addition of additional O-antigen subunits. It is our expectation that all of these mutations will enhance innate immunity recruitment via interaction with TLR4 and that one or more of these mutations will also enable maximal secretion of the FliC180 subunit to activate innate immunity via TLR5 binding. Based on comparative studies for interactions with HEK cells displaying TLR4 and TLR5, we selected the wbaP45 deletion strain 12626 and added the recA62 mutation to yield 12669 (Table 8). If desired, further safety can be provided by replacing a mutation for the arabinose-dependent expression of either the asdA or dadB genes with either the rhaRS P.sub.rhaBAD or P.sub.rhaBAD cassette which makes gene expression dependent on rhamnose as a second required sugar for AAS strain viability and survival.

(163) The final outcome of these constructions and evaluations should result in SDAAS strains with maximal abilities to stimulate innate immunity by multiple pathways that will be completely attenuated (avirulent) and non-toxic (non-reactogenic) and thus enhance the immunogenicity of a diversity of subunit and live attenuated or live vectored vaccines to induce protective immunity.

Example 13. Evaluation of Family B Strains with Improvements Based on Inclusion of Mutations to Alter LPS Synthesis, Escape from SCV and Recombination Proficiency and DNA Stability

(164) In evaluating Family B strains with mutations altering LPS structure, we observed that strains 12542 (waaC41) and 12543 (waaG42) interacted with TLR4 on HEK cells at higher levels than their parent strain 12518 (FIG. 5). However, their interaction with TLR5 on HEK cells was sharply reduced. Based on studies on contributions of various mutations in enhancing interactions with TLR5 (FIGS. 16 and 17) and reducing display of endotoxin activities (FIG. 19) without decreasing interaction with TLR4 (FIG. 20), we constructed as series of derivatives of the improved Family B strain 12612 with complete display of the LPS core but in which the amount and structure of the LPS O-sidechain was altered. These strains 12620 (pmi-2426), 12621 (rfc-112), 12626 (wbaP45) and 12641 (waaL46) (Table 8) were then evaluated for interaction with HEK cells displaying TLR4 and TLR5. FIG. 21 presents results from one experiment showing that all these strains activate NFB as well or better than parent strain 12612. (This study also included Family A strains to show that inclusion of mutations such as pagP8, lpxR9 and pagL7 that reduce lipid A toxicity actually enhanced delivery of the FliC180 flagellin to activate synthesis of NFB.) Since FliC180 should be secreted and released into the culture supernatant, we also tested strains with alterations in synthesis of LPS O-antigen for such release. However, since bacteria added to HEK cells continue to be metabolically active and produce flagellin, we conducted studies demonstrating that adding gentamicin at 100 g/ml prevented such metabolic activity without altering TLR5 bearing HEK cells to be activated by flagellin present in culture supernatants containing 100 g gentamicin/ml. FIG. 22 presents data that show that the 12612 derivative strain 12626 with the wbaP45 mutation released the highest amount of the FliC180 flagellin into culture supernatant fluids.

(165) We next evaluated all these strains (including the Family A strains with the pagP8, lpxR9 and pagL7 mutations) for interaction with TLR4 on HEK cells. In this study (FIG. 23), 12621 (rfc-112) and 12626 (wbaP45) were slightly better than the other strains, which was more apparent at earlier times of measurement than at later times. We also conducted a study on the assumption that the TLR4 binding ability to HEK cells would be bound to the surface of bacterial cells and not present in culture supernatants. To our surprise, this was not so as revealed by the data in FIG. 24 that demonstrate as much activity in the culture supernatant fluids as bound to cells. We believe that this is likely due to the continual release of outer membrane vesicles (OMVs) produced continuously during culture growth. Such OMVs that are like liposomes readily attach to host cells and are internalized. This newly discovered attribute in SDAAS strains is likely to be beneficial in enhancing recruitment of innate immune responses.

(166) The Family B and A strains evaluated in the studies generating the data in FIGS. 21 and 23 were then tested for endotoxin assay in the same manner as used to generate the data in FIG. 19 (Example 11). As shown in FIG. 25, the wild-type 3761, the pmi mutant 12620 when grown with mannose to enable LPS O-antigen synthesis) and especially the rfc mutant 12621 (which has a single O-antigen subunit) display the highest levels of endotoxin activities. The other Family A and B strains display the least endotoxin levels and especially 12626, which we selected for further modifications to enhance abilities to recruit innate immune responses and display further attenuation and safety features.

(167) For the studies to investigate activation of NFB synthesis in HEK cells with TLR8 and TLR9, we selected Family A strains that rapidly lyse during infection of cells rather than Family B strains that display regulated delayed lysis features such that they will not lyse during the first 24 h after entry into host cells. As reveled by the results depicted in FIGS. 26 and 27, the Family A strain 12640 and its derivatives 12650 (with the sifA26 mutation) and 12661 (with the sifA26 and recA62 mutations) all are very efficient in stimulating HEK cells with the internal TLR8 and TLR9 pattern receptors that recognize ssRNA and CpG sequences in DNA, respectively. These studies were conducted with MOIs of 10 and 1. We had expected an enhancement of activity dependent on adding the sifA26 mutation to 12640, but this was not observed, presumably since the cells lyse so rapidly and before an appreciable number of still intact cells could escape from the SCV to lyse in the cytosol. A similar inference likely explains the lack of observed enhancement of adding the recA62 mutation in enhancing activation in cells with TLR9. Whether addition of differing concentrations of DAP and D-alanine can enable any distinct differences in reactivity of these strains with HEK cells displaying TLR8 and TLR9 will be investigated. Evaluation of the addition of these two mutations to Family B strains will require animal studies to evaluate enhancement of adjuvant activities in increasing immune responses to antigens such as Ova, attenuated vaccines such as BCG and recombinant PIESV constructs delivering protective antigens from pathogens such as Campylobacter jejuni, Clostridium perfringens, Streptococcus pneumoniae, Clostridium difficile, Eimeria species, Escherichia coli pathovars and virus such as influenza.

Example 14. Selection of Family A and B Strains to Thoroughly Evaluate Adjuvant Activities in Animal Studies

(168) We have selected the Family A strain 12661 (alr-3 dadB4 asdA33 fliC180 (hin-fljBA)-219 pagP8 lpxR9 pagL7 eptA4 arnT6 sifA26 recA62) and the Family B strain 12669 (P.sub.asdA55::TT araC P.sub.araBAD asd alr-3 P.sub.dadB66::TT araC P.sub.araBAD dadB fliC180 pagP81::P.sub.lpp lpxE pagL7 lpxR9 (hin-fljBA)-219 arnT6 eptA4 sifA26 wbaP45 recA62) for the conduct of a diversity of animal studies analogous to those described in Examples 4 and 5 as well as other studies on enhancement of immune responses dependent on genotypes of PIESV vector strains comparing strains not lysing versus strains with rapid versus a much slower rate of in vivo regulated delayed lysis. We are also considering altering the Family B strain 12669 to replace either the P.sub.asdA55::TT araC P.sub.araBAD asd or the P.sub.dadB66::TT araC P.sub.araBAD dadB mutation with a rhamnose regulated rhaRS P.sub.rhaBAD or P.sub.rhaBAD construction, respectively, using the suicide vectors pG8R340, pG8R352, pG8R353, pG8R354 or pG8R355 (Table 7). This will constitute an additional safety feature due to a requirement for two sugars for growth and infectivity not present in animal tissues.

REFERENCES

(169) 1. Kong W, Wanda S Y, Zhang X, Bollen W, Tinge S A, Roland K L, et al. Regulated programmed lysis of recombinant Salmonella in host tissues to release protective antigens and confer biological containment. Proc Natl Acad Sci USA. 2008; 105 (27): 9361-6. 2. Curtiss R, 3rd, Wanda S Y, Gunn B M, Zhang X, Tinge S A, Ananthnarayan V, et al. Salmonella enterica serovar typhimurium strains with regulated delayed attenuation in vivo. Infection and immunity. 2009; 77 (3): 1071-82. 3. Galan J E, Nakayama K, Curtiss R, 3rd. Cloning and characterization of the asd gene of Salmonella typhimurium: use in stable maintenance of recombinant plasmids in Salmonella vaccine strains. Gene. 1990; 94 (1): 29-35. 4. Xin W, Wanda S Y, Zhang X, Santander J, Scarpellini G, Ellis K, et al. The Asd(+)-DadB(+) dual-plasmid system offers a novel means to deliver multiple protective antigens by a recombinant attenuated Salmonella vaccine. Infection and immunity. 2012; 80 (10): 3621-33. 5. Stevenson G, Andrianopoulos K, Hobbs M, Reeves P R. Organization of the Escherichia coli K-12 gene cluster responsible for production of the extracellular polysaccharide colanic acid. J Bacteriol. 1996; 178 (16): 4885-93. 6. Whitfield C. Biosynthesis and assembly of capsular polysaccharides in Escherichia coli. Annual review of biochemistry. 2006; 75:39-68. 7. Pizarro-Cerda J, Tedin K. The bacterial signal molecule, ppGpp, regulates Salmonella virulence gene expression. Mol Microbiol. 2004; 52 (6): 1827-44. 8. Torok I, Kari C. Accumulation of ppGpp in a relA mutant of Escherichia coli during amino acid starvation. The Journal of biological chemistry. 1980; 255 (9): 3838-40. 9. Brosius J, Erfle M, Storella J. Spacing of the 10 and 35 regions in the tac promoter. Effect on its in vivo activity. J Biol Chem. 1985; 260 (6): 3539-41. 10. Amann E, Ochs B, Abel K-J. Tightly regulated tac promoter vectors useful for the expression of unfused and fused proteins in Escherichia coli. Gene. 1988; 69 (2): 301-15. 11. Wang S, Li Y, Scarpellini G, Kong W, Shi H, Baek C H, et al. Salmonella vaccine vectors displaying delayed antigen synthesis in vivo to enhance immunogenicity. Infect Immun. 2010; 78 (9): 3969-80. 12. Black S, Wright N G. Aspartic b-semialdehyde dehydrogenase and aspartic b-semialdehyde. J Biol Chem. 1955; 213 (1): 39-50. 13. Nakayama K, Kelly S M, Curtiss R, III Construction of an Asd.sup.+ expression-cloning Vector: stable maintenance and high level expression of cloned genes in a Salmonella vaccine strain. Nat Biotech. 1988; 6 (6): 693-7. 14. Vander Byl C, Kropinski A M. Sequence of the genome of Salmonella bacteriophage P22. J Bacteriol. 2000; 182 (22): 6472-81. 15. Kong Q, Six D A, Roland K L, Liu Q, Gu L, Reynolds C M, et al. Salmonella synthesizing 1-dephosphorylated [corrected] lipopolysaccharide exhibits low endotoxic activity while retaining its immunogenicity. Journal of immunology (Baltimore, Md: 1950). 2011; 187 (1): 412-23. 16. Kong Q, Six D A, Liu Q, Gu L, Wang S, Alamuri P, et al. Phosphate groups of lipid A are essential for Salmonella enterica serovar Typhimurium virulence and affect innate and adaptive immunity. Infection and immunity. 2012; 80 (9): 3215-24. 17. Liu Q, Liu Q, Yi J, Liang K, Hu B, Zhang X, et al. Outer membrane vesicles from flagellin-deficient Salmonella enterica serovar Typhimurium induce cross-reactive immunity and provide cross-protection against heterologous Salmonella challenge. Scientific reports. 2016; 6:34776. 18. Laniewski P, Baek C H, Roland K L, Curtiss R, 3rd. Analysis of Spleen-Induced Fimbria Production in Recombinant Attenuated Salmonella enterica Serovar Typhimurium Vaccine Strains. mBio. 2017; 8 (4). 19. Beuzon C R, Meresse S, Unsworth K E, Ruiz-Albert J, Garvis S, Waterman S R, et al. Salmonella maintains the integrity of its intracellular vacuole through the action of SifA. The EMBO journal. 2000; 19 (13): 3235-49. 20. Ohlson M B, Huang Z, Alto N M, Blanc M P, Dixon J E, Chai J, et al. Structure and function of Salmonella SifA indicate that its interactions with SKIP, SseJ, and RhoA family GTPases induce endosomal tubulation. Cell host & microbe. 2008; 4 (5): 434-46. 21. Curtiss R, III., Porter S B, Munson M, Tinge S A, Hassan J O, Gentry-Weeks C, et al. Nonrecombinant and recombinant avirulent Salmonella live vaccines for poultry. In: Blankenship L C, Bailey J H S, Cox N A, Stern N J, Meinersmann R J, editors. Colonization control of human bacterial enteropathogens in poultry. New York Academic Press 1991. p. 169-98. 22. Bertani G. Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. J Bacteriol. 1951; 62 (3): 293-300. 23. Sambrook J, Russell D W. Molecular cloning: a laboratory manual. 3rd ed. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press; 2001. 24. Schmieger H. Phage P22-mutants with increased or decreased transduction abilities. Mol Gen Genet. 1972; 119 (1): 75-88. 25. Schmieger H, Backhaus H. Altered cotransduction frequencies exhibited by HT-mutants of Salmonella-phage P22. Mol Gen Genet. 1976; 143 (3): 307-9. 26. Kang H Y, Dozois C M, Tinge S A, Lee T H, Curtiss R, III. Transduction-mediated transfer of unmarked deletion and point mutations through use of counter selectable suicide vectors. J Bacteriol. 2002; 184 (1): 307-12. 27. Edwards R A, Keller L H, Schifferli D M. Improved allelic exchange vectors and their use to analyze 987P fimbria gene expression. Gene. 1998; 207 (2): 149-57. 28. Quandt J, Hynes M F. Versatile suicide vectors which allow direct selection for gene replacement in gram-negative bacteria. Gene. 1993; 127 (1): 15-21. 29. Roland K, Curtiss R, III., Sizemore D. Construction and evaluation of a Dcya Derp Salmonella typhimurium strain expressing avian pathogenic Escherichia coli 078 LPS as a vaccine to prevent airsacculitis in chickens. Avian diseases. 1999; 43 (3): 429-41. 30. Kong Q, Liu Q, Roland K L, Curtiss R, III. Regulated delayed expression of rfaH in an attenuated Salmonella enterica serovar Typhimurium vaccine enhances immunogenicity of outer membrane proteins and a heterologous antigen. Infect Immun. 2009; 77 (12): 5572-82. 31. Juarez-Rodriguez M D, Arteaga-Cortes L T, Kader R, Curtiss R, 3rd, Clark-Curtiss J E. Live attenuated Salmonella vaccines against Mycobacterium tuberculosis with antigen delivery via the type III secretion system. Infection and immunity. 2012; 80 (2): 798-814. 32. Lee F K, Nahmias A J, Lowery S, Nesheim S, Reef S, Thompson S, et al. ELISPOT: a new approach to studying the dynamics of virus-immune system interaction for diagnosis and monitoring of HIV infection. AIDS research and human retroviruses. 1989; 5 (5): 517-23. 33. Park B S, Song D H, Kim H M, Choi B S, Lee H, Lee J O. The structural basis of lipopolysaccharide recognition by the TLR4-MD-2 complex. Nature. 2009; 458 (7242): 1191-5. 34. McCuskey R S, McCuskey P A, Urbaschek R, Urbaschek B. Species differences in Kupffer cells and endotoxin sensitivity. Infection and immunity. 1984; 45 (1): 278-80. 35. Baldridge M T, King K Y, Goodell M A. Inflammatory signals regulate hematopoietic stem cells. Trends in immunology. 2011; 32 (2): 57-65. 36. Essers M A, Offner S, Blanco-Bose W E, Waibler Z, Kalinke U, Duchosal M A, et al. IFNalpha activates dormant haematopoietic stem cells in vivo. Nature. 2009; 458 (7240): 904-8. 37. Takizawa H, Boettcher S, Manz M G. Demand-adapted regulation of early hematopoiesis in infection and inflammation. Blood. 2012; 119 (13): 2991-3002. 38. Belisle J T, Vissa V D, Sievert T, Takayama K, Brennan P J, Besra G S. Role of the major antigen of Mycobacterium tuberculosis in cell wall biogenesis. Science. 1997; 246:1420-1422. 39. Sorensen A L, Nagai S, Houen G, Andersen P, Andersen A B. Purification and characterization of a low molecular mass T-cell antigen secreted by Mycobacterium tuberculosis. Infection and immunity. 1995; 63:1710-1717. 40. Berthet F X, Rasmussen P B, Rosenkrands I, Andersen P, Gicquel B. A Mycobacterium tuberculosis operon encoding ESAT-6 and a novel low-molecular mass culture filtrate protein (CFP-10). Microbiology. 1998; 144:3195-3203. 41. Skjot R L, Oettinger, Rosenkrands I, Ravn P, Brock I, Jacobsen S, Andersen P. Comparative evaluation of low-molecular-mass proteins from Mycobacterium tuberculosis identifies members of the ESAT-6 family as immunodominant T-cell antigens. Infection and immunity. 2000; 68:214-220. 42. Ottenhoff T H, Doherty T M, van Dissel J T, Bang P, Lingnau K, Kromann I, Andersen P. First in humans: a new molecularly defined vaccine shows excellent safety and strong induction of long-lived Mycobacterium tuberculosis-specific Th1-cell like responses. Human Vaccines. 2010:6:1007-1015. 43. Gunn, B. M., S. Y. Wanda, D. Burshell, C. Wang, and R. Curtiss III. 2010. Construction of recombinant attenuated Salmonella enterica serovar Typhimurium vaccine vector strains for safety in newborn and infant mice. Clin. Vaccine Immunol. 17:354-362. PMCID: PMC2837962 44. Li, Y., S. Wang, W. Xin, G. Scarpellini, Z. Shi, B. Gunn, K. L. Roland, and R. Curtiss III. 2008. A sopB deletion mutation enhances the immunogenicity and protective efficacy of a heterologous antigen delivered by live attenuated Salmonella enterica vaccines. Infect. Immun. 76:5238-5246. PMCID: PMC2573344 45. Lau N, Haeberle A L, O'Keeffe B J, Latomanski E A, Celli J, Newton H J, Knodler L A. 2019. SopF, a phosphoinositide binding effector, promotes the stability of the nascent Salmonella-containing vacuole. PLOS Pathog. 2019 Jul. 24; 15 (7): e1007959. doi: 10.1371/journal.ppat.1007959. eCollection 2019 46. Kong, W., M. Brovold, B. A. Koneneman, J. Clark-Curtiss, and R. Curtiss III. 2012. Turning self-destructing Salmonella into a universal DNA vaccine delivery platform. Proc. Natl. Acad. USA 109:19414-19419. PMCID: PMC3511069 47. Curtiss, R. III, X. Zhang, S. Y. Wanda, H. Y. Kang, V. Konjufca, Y. Li, B. Gunn, S. Wang, G. Scarpellini, and I. S. Lee. 2007. Induction of host immune responses using Salmonella-vectored vaccines, p. 297-313. In K. A. Brogden, F. C. Minion, N. Cornick, T. B. Stanton, Q. Zhang, L. K. Nolan, and M. J. Wannemuehler (eds.). Virulence Mechanisms of Bacterial Pathogens, 4.sup.th ed., ASM Press, Washington, D.C. 48. Kong, Q., Q. Liu, A. M. Jansen, and R. Curtiss III. 2010. Regulated delayed expression of rfc enhances the immunogenicity and protective efficacy of a heterologous antigen delivered by live attenuated Salmonella enterica vaccines. Vaccine 28:6094-6103. PMCID: PMC3956130 49. Kong, Q., J. Yang, Q. Liu, P. Alamuri, K. L. Roland, and R. Curtiss III. 2011. Effect of deletion of genes involved in lipopolysaccharide core and O-antigen synthesis on virulence and immunogenicity of Salmonella enterica serovar Typhimurium. Infect. Immun. 79:4227-4239. PMCID: PMC3187260 50. O'Callaghan D, Maskell D, Liew F Y, Easmon C S, Dougan G. 1988. Characterization of aromatic- and purine-dependent Salmonella typhimurium: attention, persistence, and ability to induce protective immunity in BALB/c mice. Infect Immun. 56 (2): 419-23. 51 Trent, M. S., A. A. Ribeiro, S. Lin, R. J. Cotter, and C. R. H. Raetz. 2001. An inner membrane in Salmonella and Escherichia coli that transfers 4-amino-4-deoxy-L-arabinose to lipid A: induction of polymyxin-resistant mutants and role of a novel lipid-linked donor. J. Biol. Chem. 276:43122-43131. 52. Herrera, C. M., J. V. Hankins, and M. S. Trent. 2010. Activation of PmrA inhibits LpxT-dependent phosphorylation of lipid A promoting resistance to antimicrobial peptides. Mol. Microbiol. 76:1444-1460. 53. Buchmeier, N. A., C. J. Lipps, M. Y. So, and F. Heffron. 1993. Recombination-deficient mutants of Salmonella typhimurium are avirulent and sensitive to the oxidative burst of macrophages. Mol. Microbiol. 7:933-936.