INDUCTION OF CTL RESPONSES TO FOREIGN ANTIGENS EXPRESSED IN MYCOBACTERIA

This invention relates to the induction of a T-cell response, in particular a cytotoxic T lymphocyte response. More particularly, this invention relates to the induction of CTL responses to proteins or polypeptideβ expressed by recombinant mycobacteria.

Cell-mediated immunity (or CMI) of infections is thought to be a major line of defense against certain infections, such as viral infections and certain bacterial infections. For example, CMI may be significant in the development of an effective vaccine against human immunodeficiency virus (HIV), or AIDS virus, because HIV vaccines and/or therapies based on the generation of passive transfer of HIV-specific antibody in the absence of eelL-mediated immunity have not yielded consistent protection in primates challenged with the HIV virus. Thus, interest has turned to the induction of cell-mediated responses to various infections, such as for example, HIV infection, and to the identification of proteins or polypeptides that a stimulate a cytotoxic T lymphocyte response, and to methods of administering such proteins or polypeptides.

In accordance with an aspect of the present invention, there is provided a method of inducing a CTL response in an animal comprising administering to the animal mycobacteria transformed with at least one DNA sequence which encodes a protein or peptide or fragment or derivative thereof which includes an epitope which

SUBSTITUTE SHEET

is recognized by cytotoxic T lymphocytes. The mycobacteria are administered in an amount effective to induce a CTL response in an animal.

Proteins or polypeptides for which the at least one D _NA sequence may encode, include, but are not limited to, Mycobacterium leprae antigens; Mycobacterium tuberculosis antigens; Ric ettsia antigens; Chlamydia antigens; Coxiella antigens; malaria sporozoite and merozoite proteins, such as the circumsporozoite protein from Plasmodium berσhei sporozoites; Clostridium antigens; Leishmania antigens; Salmonella antigens; Mycobacterium africanum antigens; Mycobacterium intracellulare antigens; Mycobacterium avium antigens; E.coli antigens; Borrelia antigens; Listeria antigens; Franciscella antigens; Yersinia antigens; Treponema antigens; Schistostoma antigens; Filaria antigens; Pneumococcus antigens; Staphylococcus antigens; Herpe* virus antigens; influenza and parainfluenza virus antigens; measles virus antigens; mumps virus antigens; hepatitis virus antigens; Shi ella antigens; Bordatella antigens; Hemophilus antigens; Streptococcus antigens; polio virus antigens; Rift Valley Fever virus antigens; dengue virus antigens; Human Immunodeficiency Virus (HIV) antigens; and respiratory syncytial virus (RSV) antigens.

In one embodiment, the at least one DNA sequence encodes at least one protein or polypeptide or fragment or derivative thereof which includes an eptiope which is recognized by cytotoxic T lymphocytes induced by an HIV protein or fragment or derivative ^" thereof. The at least one DNA sequence may encode an HIV protein or fragment or derivative thereof. HIV proteins or polypβptidββ which may be encoded by the at least one DNA sequence includes but are not limited to, HIV-I-gp 120; HIV-I-gp 41 _; HIV-I-gp 160; HIV-I-pol; HIV-I-nef; HIV-I-tat; HIV-I-rev; HIV-I-vif _; HIV-I-vpr; HIV-I-vpu; HIV-I-gag; HIV-2-gp 120; HIV-2-gp 160; HIV-2-gp 41; HIV-2-gag; HIV-2-pol; HIV-2-nef;

SUBSTITUTESHEET

HIV-2-tat; HIV-2-rev; HIV-2-vif; HIV-2-vpr; HIV-2-vpu; and HIV-2-vpx.

Mycobacteria which may be transformed with the at least one DNA sequence, which encodes a protein or polypeptide or fragment or derivative thereof which includes an epitope which is recognized by cytotoxic T lymphocytes, include, but are not limited to, Mycobacterium bovis-BCG, M.smeqmatis, M.avium, M.phlei, M.fortiutum, M.lufu, M.paratuberculosis, M.habana, M.scrofalaceum, and M.intracellulare. In a preferred embodiment the mycobacterium is M.bovis-BCG or a mutant thereof.

The at least one DNA sequence may be contained within an expression vector, which is transformed into a mycobacterium, whereby the mycobacterium expresses the protein or polypeptide o fragment or derivative thereof which includes an epitope which i recognized by cytotoxic T lymphocytes.

The expression vector may be, for example, a temperate shuttle phasmid or a bacterial-mycobacterial shuttle plasmid. Each of these vectors may be used to introduce the at least one DNA sequence encoding a protein or polypeptide or fragment or derivative which includes an epitope which is recognized by cytotoxic T lymphocytes, stably into mycobacteria, in which the at least one DNA sequence may be expressed. When a shuttle phasmid, which replicates as a plasmid in bacteria and a phage i mycobactreia, is employed, integration of the phasmid, which includes the at least one DNA sequence encoding a protein or polypeptide, or fragment or derivative thereof, which includes a epitope which is recognized by cytotoxic T lymphocytes, into the mycobactarial chromosome occurs through site-specific integration. The at least one DNA sequence which encodes a protein or polypeptide or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes, is replicated as part of the chromosomal DNA. Whe a bacterial-mycobacterial shuttle plasmid is employed, the at least one DNA sequence which encodes a protein or polypeptide or

SUBSTITUTE SHEET

fragment or derivative thereof, which includes an epitope whic _h is recognized by cytotoxic T lymphocytes, is stably maintaine _d extrachromosomally in a plasmid. Expression of the at least one DNA sequence occurs extrachromosomally (e.g., episomally) . For example, the at least one DNA se ^« quence is cloned into a shuttle plasmid and the plasmid is introduced into a mycobacterium such as those hereinabove described, wherein the plasmid replicates episomally. Examples of such shuttle phasmids and bacterial-mycobacterial shuttle plasmids are further descri _b e _d in Application Serial No. 361,944, filed June 5, 1989, which is hereby incorporated by reference.

In one embodiment the mycobacteria are transformed with an expression vector which comprises at least one DNA sequence encoding a protein or polypeptide which includes an epitope which is recognized by cytotoxic T lymphocytes, and a promoter selected from the class consisting of mycobacterial promoters and mycobacteriophage promoters for controlling expression of the DNA encoding the heterologous protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes.

Mycobacterial and mycobacteriophage promoters which may be employed include, but are not limited to, mycobacterial promoters such as the BCG HSP60 and HSP70 promoters; the mycobactin promoter from M. tuberculosis and BCG; the mycobacterial 14 kda and 12 kda antigen promoters; the mycobacterial α-antigen promoter from M.tuberculosis or BCG; the MBP-70 promoter, the mycobacterial 45 kda antigen promoter from M.tuberculosis or BCG; the supβroxide dismutase promoter; the mycobacterial asd promoter, and mycobacteriophage promoters such as the Bxbl, LI, L5, and TM4 promoters. In one embodiment, the promoter is a mycobacterial heat shock protein promoter such as HSP60 or HSP70.

The promoter sequence may, in one embodiment, be part of an expression cassette which also includes a portion of the gene normally under the control of the promoter. For example, when a

SUB _S TITUTE SHEET

mycobacterial HSP60 or HSP70 promoter is employed, the expressio cassette may, within the scope of the present invention, inclu _d e _, in addition to the promoter, a portion of the gene for the HSP6 ₀ or HSP70 protein. When the expression cassette and the DNA encoding the protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes are expressed, the protein expresse _d by the cassette and the DNA encoding the protein or polypeptide is a fusion protein of a fragment of a mycobacterial protein (eg. _, the HSP60 or HSP70 protein), and the protein or polypeptide or fragment or deriviative thereof which includes an epitope which is recognized by cytotoxic T lymphocytes.

In a preferred embodiment, the transcription initiation site, the ribosomal binding site, and the start codon, which provides for the initiation of the translation of mRNA, are eacfe _. of mycobacterial origin. The stop codon, which stops translation of mRNA, thereby terminating synthesis of the protein or polypeptide or fragment or derivative thereof which includes an epitope which is recognized by cytotoxic T lymphocytes, and the transcription termination site, may be of mycobacterial origin, or of other bacterial origin, or such stop codon and transcription termination site may be those of the DNA encoding the protein or polypeptide which includes an epitope which is recognized by cytotoxic T lymphocytes.

Preferably, the mycobacterial promoter is a BCG promoter, and the mycobacterium is BCG.

In one embodiment, the expression vector may further include DNA which encodes for proteins or polypeptides such as, but not limited to, antigens, anti-tumor agents, enzymes, lymphokines, pharmacologic agents, immunopotentiators, reporter molecules of interest in a diagnostic context, and selectable markers.

Selectable markers which may be encoded include, but are not limited to, the β-galactosidase marker, the kanamycin resistance

_SUB STITUTE SHEET

marker, the chloroamphenicol resistance marker, the neomycin resistance marker, and the hygromycin resistance marker.

In accordance with one embodiment, the vector further includes a mycobacterial origin of replication.

In accordance with another embodiment, the vector may be a plasmid. The plasmid may be a non-shuttle plasmid, or may _b e a shuttle plasmid which further includes a bacterial origin of replication such as an E.coli origin of replication, a Bacillus origin of replication, a Staphylococcus origin of replication, a Streptomyces origin of replication, or a pneumococcal origin of replication. In one embodiment, the shuttle plasmid includes an E. coli origin of replication.

In accordance with yet another embodiment, the vector may further include a multiple cloning site, and the DNA encoding fpr the protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes is inserted in the multiple cloning site.

In addition to the DNA encoding a heterlogous protein or polypeptide, and the mycobacterial promoter for controlling expression of the DNA encoding the protein or polypeptide which includes an epitope which is recognized by cytotoxic T lymphocytes, the expression vector may, in one embodiment, further include a DNA sequence encoding bacteriophage integration into a mycobacterium chromosome. Bacteriophages from which the DNA sequence encoding bacteriophage integration into a mycobacterium chromosome may be derived include, but are not limited to, mycobacteriophages such as but not limited to, the L5, LI, Bxbl, and TM4 mycobacteriophages; the lambda phage of , coli; the toxin phages of Corvnebacteria; phages of Actinomvcetes and Norcardia; the #C31 phage of Streptomvces; and the P22 phage of Salmonella. Preferably, the DNA sequence encodes mycobacteriophage integration into a mycobacterium chromosome. The DNA sequence which encodes bacteriophage integration into a mycobacterium chromosome may include DNA which encodes integrase,

SUBSTITUTE SHEET

which is a protein that provides for integration of the vector into the mycobacterial chromosome. Preferably, the DNA sequence encoding mycobacterial phage integration also includes DNA which encodes an attP site.

The DNA encoding the attP site and the integrase provides for an integration event which is referred to as site-specific integration. DNA containing the attP site and the integrase gene is capable of integrating into a corresponding attB site of a mycobacterium chromosome.

It is to be understood that the exact DNA sequence encoding the attP site may vary among different phages, and that the exact DNA sequence encoding the attB site may vary among different mycobacteria.

Examples of expression vectors which include mycobacterial promoters and mycobacteriophage promoters, for controlling the βt least one DNA sequence encoding a protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes are further described in application Serial No. 642,017, filed January 16, 1991, which is a continuation of Application Serial No. 552,828, filed July 16, 1990, now abandoned. The contents of Application Serial No. 642,017 are hereby incorporated by reference.

In another embodiment, the mycobacteria are transformed with DNA which comprises a first DNA sequence which is a phage DNA portion encoding bacteriophage integration into a mycobacterium chromosome, and DNA including the at least one DNA sequence encoding a protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes.

The term "phage DNA portion", as used herein, means that the DNA sequence is derived from a phage and lacks the DNA which is required for phage replication.

Bacteriophages from which the phage DNA portion may be derived include, but are not limited to, mycobacteriophages, such

SUBSTITUTE SHEET

as but not limited to the those hereinabove described.

Preferably, the phage DNA portion encodes mycobacteriophage integration into a mycobacterium chromosome.

In a preferred embodiment, the first DNA sequence inclu _d es

DNA encoding integrase, which is a protein that provides for integration of the DNA into the mycobacterial chromosome. Most preferably, the first DNA sequence also includes DNA which encodes an AttP site.

The DNA sequence encoding the AttP site and the integrase provides for an integration event which is referred to as site-specific integration. DNA containing the AttP site and the integrase gene is capable of integration into a corresponding AttB site of a mycobacterium chromosome.

It is to be understood that the exact DNA sequence encoding the attP site may vary among different phages, and that the exact DNA sequence encoding the attB site may vary among different mycobacteria.

The integration event results in the formation of two new junction sites called AttL and AttR, each of which contain part of each .of AttP and AttB. The inserted and integrated non-phage DNA which includes the first DNA sequence and the at least one DNA sequence, which encodes a protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes is flanked by the AttL and AttR sites. The insertion and integration of the phage DNA portion results in the formation of a transformed mycobacterium. The DNA may further include DNA which encodes a selectable marker or markers; or other proteins or polypeptides of interest, such as, but not limited to anti-tumor agents, enzymes, lymphokinββ, pharmacologic agents, immunopotentiators, and reporter molecules of interest in a diagnostic context.

Selectable markers which may be encoded include, but are not limited to, the kanamycin resistance marker, the neomycin

SUBSTITUTESHEET

resistance marker, the chloroamphenicol resistance marker, and the hygromycin resistance marker.

The phage DNA portion of the present invention, which includes the first DNA sequence encoding mycobacterium phage integration into a mycobacterium chromosome, and the at least on DNA sequence encoding a protein or polypeptide, or fragment or derivative thereof, which includes an epitope recognized by cytotoxic T lymphocytes, may be constructed through genetic engineering techniques known to those skilled in the art. In a preferred embodiment, the phage DNA portion may be a plasmid including, in addition to the DNA encoding integration and the DNA encoding a protein or polypeptide, or fragment or derivative thereof, which includes an epitope recognized by cytotoxic T lymphocytes, an origin of replication for any of a wide variet of organisms, which includes, but is not limited to, E.coli, Streptomvces species, Bacillus species, Staphylococcus species, _. Shigella species, Salmonella species and various species of pneumococci. Most preferably, the plasmid includes an origin of replication for E.coli.

The phage DNA portion also may include a suitable promoter. Suitable promoters include, but are not limited to, mycobacterial promoters and mycobacteriophage promoters such as those hereinabove described.

The promoter sequence may, in one embodiment, be part of an expression cassette which also includes a portion of the gene normally under the control of the promoter, as hereinabove described. For example, when a mycobacterial HSP60 or HSP70 promoter is employed, the expression cassette may include, in addition to the promoter, a portion of the gene for the HSP60 or HSP70 protein. When the expression cassette and the DNA encodin the protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes are expressed, the protein expressed by the cassette and the DNA encoding the protein or polypeptide is a fusion

SUBSTITUTE SHEET

protein of a fragment of a mycobacterial protein (eg., the H _S P60 or HSP70 protein), and the protein or polypeptide, or fragment o derivaitve thereof, which includes an epitope which is recognize by cytotoxic T lymphocytes.

In a preferred embodiment, the transcription initiation site, the ribosomal binding site, and the start codon, which provides for the initiation of the translation of mRNA, are each of mycobacterial origin. The stop codon, which stops translation of mRNA, thereby terminating synthesis of the protein or polypeptide, or fragment or deriviatve thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes, and the transcription termination site, may be of mycobacterial origin, or of other bacterial origin, or such stop codon and transcription termination site may be those of the DNA encoding the protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes.

Examples of DNA which includes a first DNA sequence which is a phage DNA portion encoding bacteriophage integration into a mycobacterium chromosome, and DNA including the at least one DNA sequence encoding a protein or polypeptide, or fragment or derivative thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes are further described in application Serial No. 553,907, filed July 16, 1990, the contents of which are hereby incorporated by reference.

Mycobacteria which are transformed which DNA which encodes for a protein or polypeptide or fragment(s) or derivative(s) thereof, which includes an epitope which is recognized by cytotoxic T lymphocytes, may be employed in a composition, such as a vaccine, for inducing a CTL response in an aniaml. The vaccine may be administered to a human or non-human animal.

To form such a vaccine, the transformed mycobacteria are administered in conjunction with a suitable pharmaceutical carrier. As representative examples of suitable carriers there

SUBSTITUTE SHEET

may be mentioned: mineral oil, alum, synthetic polymers, etc. Vehicles for vaccines are well known in the art and the selection of a suitable vehicle is deemed to be within the scope of those skilled in the art from the teachings contained herein. The selection of a suitable vehicle is also dependent upon the manner in which the vaccine is to be administered. The vaccine may be in the form of an injectable dose and may be administere _d intramuscularly, intravenously, orally, intradermally, or by subcutaneous administration.

Other means for administering the vaccine or therapeutic agent should be apparent to those skilled in the art from the teachings herein; accordingly, the scope of the invention is not to be limited to a particular delivery form.

When the transformed mycobacteria are employed as a vaccine, such a vaccine has important advantages over other presently available vaccines. Mycobacteria have, as hereinabove indicated, adjuvant properties among the best currently known and, therefore, stimulate a recipient's immune system to respond with great effectiveness. This aspect of the vaccine induces cell-mediated immunity and thus is especially useful in providing immunity against pathogens in cases where cell-mediated immunity appears to be critical for resistance. Also, mycobacteria may stimulate long-term memory or immunity. It thus may be possible to prime long-lasting T cell memory, which stimulates secondary antibody responses neutralizing to the infectious agent. Such priming of T cell memory is useful, for example, against pertussis, malaria, influenza virus, Herpes virus, rabies, Rift Valley fever virus, dengue virus, measles virus, Human Immunodeficiency Virus (HIV), and respiratory syncytial virus.

The invention will now be described with respect to the following examples; however, the scope of the present invention is not to be limited thereby.

Example 1

SUBSTITUTE SHEET

A. Construction of ^p lasmid including mycobacterial promoter expression cassette and lacZ gene.

1. Construction of PYUB125

Plasmid pAL5000, a plasmid which contains an origin of replication of M. fortuitum, and described in Labidi, et al. _, FEMS Microbiol. Lett., Vol. 30, pgs. 221-225 (1985) and in _G ene, Vol. 71, pgs. 315-321 (1988), is subjected to a partial Sau 3A digest, and 5kb fragments are gel purified. A 5kb fragment is then ligated to Bam HI digested pIJ666 (an. E. coli vector containing an E. coli origin of replication and also carries neomycin-kanamycin resistance, as described in Kieser, et al., Gene, Vol. 65, pgs. 83-91 (1988) to form plasmid pYUB12. A schematic of the formation of plasmid pYUB12. A schematic of the formation of plasmid pYUB12 is shown in Figure 1. pYUB12 and pIJ666 were then transformed into M. smeqmatis and BCG. Neomycin-resistant transformants that were only obtained by pYUB12 transformation confirmed that pAL5000 conferred autonomous replication to pI 666 in M. smeqmatis and BCG.

Shotgun mutagenesis by Snapper, et al (1988, hereinabove cited) indicated that no more than half of the pAL5000 plasmid was necessary to support plasmid replication in BCG. This segment presumably carried open reading frames 0RF1 and 0RF2, identified by Rauzier, et al., Gene, Vol. 71, pgs. 315-321 (1988), and also presumably carried a mycobacterial origin of replication. pYUB12 is then digested with Hpal and EcoRV, a 2586 bp carrying this region or segment pAL5000 is removed and ligated to PvuII digested pYUBΘ. Plasmid pYUBβ (a pBR322 derivative) includes an E. coli replicon and a kan (aph) gene. Ligaticn of the 2586 bp pYUB12 fragment to PvuII digested pYUB ^β results in the formation of pYUB53, as depicted in Figure 2. Transformation of pYUB53 confirmed that the EcoRV-Hpal fragment, designated M.rep, was capable of supporting autonomous replication in BCG.

Plasmid pYUB53 was then digested with AatI, EcoRV, and PstI in order to remove the following restriction sites:

_S UB _S TITUTE SHEET

AatI 5707

EcoRI 5783

BamHI 5791

Sail 5797

PstI 5803

Pst 7252

Sail 7258

BamHI 7264

EcoRI 7273

Clal 7298

HindiII 7304; and

EcoRV 7460 Fragment ends are then flushed with T4 DNA polymerase and religated to form plasmid pYUB125, construction of which is show in Figure 3.

2. Elimination of superfluous vector DNA from pYUB125 792 bases of the tet gene, which had been inactivated by prior manipulations, was eliminated by a complete Narl digest, gel purification of the 6407 bp fragment, and ligation/recirculation, transformation of E. coli strain HB101,

R and selection of Kan transformants. The construction of resulting plasmid, pMVlOl, is schematically indicated in Figure 4, and the DNA sequence of pMVlOl, which includes markings of regions which will be deleted, and of mutations, as hereinafter described, is shown in Figure 5.

3. Construction of expression cassette based on BCG HSP60. Among the most abundant proteins in mycobacteria is the

HSP60 heat shock protein (also known as the 65 kda antigen). Because abundance of the HSP60 protein in mycobacteria indicates strong HSP60 gene expression, the sequence controlling HSP60 expression was chosen to control expression of heterologous gene encoding antigens or other proteins in BCG.

The published sequence of the BCG HSP60 gene (Thole, et al, Infect, and Immun. , Vol. 55, pgs. 1466-1475 (June 1987)), and

SUBSTITUTE SHEET

surrounding sequence permitted the construction of a cassette carrying expression control sequences (i.e., promoter an _d translation initiation sequences) by PCR. The BCG HSP61 cassette (Figure 6) contains 375 bases 5' to the BCG HSP60 start codon _, and 15 bases (5 codons) 3' to the start codon. P _C R oligonucleotide primers were then synthesized. Primer Xba-HSP _60, of the following sequence:

CAG ATC TAG ACG GTG ACC ACA ACG CGC C / was synthesized for the 5 ^* end of the cassette, and primer Bam-HSP61, of the following sequence:

CTA σσσ ATC CGC AAT TGT CTT σσc CAT TG Z was synthesized for the 3' end of the cassette. The primers were used to amplify the cassette by PCR from BCG strain Pasteur chromosomal DNA. The addition of the Bam HI site at the 3' end of the cassette adds one codon (Asp) to the first six codons of the HSP60 gene.

Each of pMVlOl and the PCR cassette HSP61 was digested with Nhel and BamHI. The PCR cassette was then inserted between the Nhel and BamHI sites of pMVlOl, then ligated to form plasmid pMV65A (Figure 7).

The E. coli lac 2 gene (Figure 8) was used as a reporter, or marker gene to assay the ability of the HSP61 cassette to express heterologous genes in BCG. A BamHI restriction fragment carrying the lac 2 gene was cloned into the Bam HI site of Bam HI digested pMV65A, resulting in the formation of pMV65A/L2 as indicated schematically in Figure 7. The formation of pMV65A/L2 results in a fusion between the HSP60 and lac 2 genes at the sixth codon of the HSP60 gene and the sixth codon of the lac 2 gene. pMV65A/LZ was then transformed into E. coli. Blue E. coli colonies were selected on x-gal plates for the presence of pMV65A/L2, thus indicating that the HSP60 promoter and translation initiation sequences were also active in E. coli. pMV65A/L2 was then transformed into BCG and plated on Dubos Oleic Agar plates containing x-gal. All BCG colonies resulting

SUBSTITUTE SHEET

from this transformation exhibited blue color, thus indicating that the lac 2 gene product (B-galactosidase) was expressed in BCG. SDS polyacrylamide gel electrophoresis was performed on lysates of the pMV65A/L2 BCG recombinants, revealing that B-galactosidase protein was expressed to levels in excess of 1 ₀ of total BCG protein (as determined by staining with Coomassie brilliant blue). These data indicated that BCG HSP61 expressio cassette was functional in expression vector pMV65A.

Example 2 Cytotoxic T lymphocyte response to E. coli 6-qalactosidase.

E- coli β-galactosidase was expressed in BCG as a six amin acid fusion protein with BCG hsp 60 protein using extrachromosomal plasmid vector pMV65A/L2 utilizing the HSP60 promoter to drive expression. The recombinant BCG was grown tα mid-log phase in Dubos media and concentrated by centrifugation The bacteria were then re-suspended in PBS plus 0.05% Tween 80- and cup sonicated briefly to disperse clumped bacteria. Six we old BALB/c mice were inoculated with a single dose of 2 X 10 ² , 2 X 10 , or 2 X 10 colony forming units (CFU'S determined post-inoculation) by either intradermal (ID), intraperitoneal (IP), or intravenous (IV) injection. At 14 or 19 weeks post-immunization splenocytes were harvested from mice and CTL activity was measured. CTL activity was measured as follows:

Splenocytes (ACK-treated, 5 X 10 /ml) were stimulated in vitro in 10 ml in upright T25 flasks by co-culture for 5 days with mitomycin C-treated cells transfected with the lac 2 gene (C3-4 cells; 5 X 10 ⁶ /ml). A 4 hr. Cr release assay was then performed in triplicate using P815 and P13.1 cells (P815 ceils transfected with the lac Z gene) as targets. Various effector-target ratios were tested using 5,000 targets/well. Specific lysis was calculated as follows: % specific lysis = 100 X [release by effector cells minus spontaneous release/ maximal release minus spontaneous release].

SUBSTITUTE SHEET

At 19 weeks post-immunization, the remaining animals in each group were boosted by intraperitoneal injection at 10 μg of purified lac 2 emulsified in incomplete Freund's adjuvant (IFA). Splenocytes were then harvested from these animals at 23 weeks and CTL activity was again measured. Unimmunized animals or animals immunized with lac 2 emulsified in IFA, or animals immunized with vaccinia virus expressing lac 2 served as controls.

The results of the avove experiments, as determined by % specific lysis of target cells, indicated that a CTL response was induced in mice immunized with BCG transformed with the expression vector pMV65A/LZ.

Example 3 Construction of integrating plasmid including mycobacterial promoter expression cassette and HIV-I-gp 120 gene. 1. Elimination of undesirable restriction sites in aph fkan ). gen .

To facilitate future manipulations, the HindiII and Clal restriction sites in the aph gene in plasmid pMVlOl were mutagenizβd simultaneously by poly eraβe chain reaction (PCR) mutagenesis according to the procedure described in Gene, Vol. 77 pgs. 57-59 (1989). The bases changed in the 'aph gene were at the third position of codons (wobble bases) within each restriction site and the base substitutions made were designed not to change the amino acid sequence of the encoded protein.

Separate PCR reactions of plasmid pMVlOl with primer ClaMut-Kan ^" + HindRMut-Kan and HindFMut-Kan + Barn-Kan were performed at 90°C (1 min.), 50°C (1 min.), and 72°C (1 min.) for 25 cycles. The PCR primers had the following base seqeuences:

ClaMut-Kan

CTT GTA TGG GAA GCC CC -3

HindRMut-Kan

GTG AGA ATG GCA AAA GAT TAT GCA TTT CTT TCC AG f

HindFMut-Kan

SUBSTITUTE SHEET

GTG TGG AAA GAA ATG CAT AAT CTT TTG CCA TTC TCA C _{CG G} -T

Barn-Kan

CGT AGA GGA TCC AGA GGA CG 6

The resulting PCR products were gel purified and mixe _d an _d a single PCR reaction without primers was performed at 94°C (1 min.), 72°C (1 min.) for 10 cycles. Primers ClaMut-Kan and

Barn-Kan were added and PCR was resumed at 94°C (1 min.), 50° _{C (} 1 min.), and 72°C (2 min.) for 20 cylces. The resulting PCR product (Kan. mut) was digested with BamHI and gel purified.

Plasmid pMVlOl was digested with Clal and cohesive ends were filled in by Klenow + dCTP + dGTP. Klenow was heat inactivated and the digest was further digested with BamHI. The 5232 base pair fragment was gel purified and mixed with fragment Kan.mut and ligated. The ligation was transformed into E. coli strain

•a HB101 and Kan colonies were screened for plasmids resistant to

Clal and HindiII digestion. Such plasmids were designated as pMVUO, which is depicted in Figure 4.

2. Elimination of sequences not necessary for plasmid replication in mycobacteria.

Plasmid pMVUO was resected in separate constructions to yield plasmids pMVlll and pMV112. In one construction, pMVUO was digested with Narl and Ball, the ends were filled in, and a 5296 base pair fragment was ligated and recircularized to form pMVlll. In another construct, pMVUO was digested with Ndel and SplI, the ends were filled in, and a 5763 base pair fragment was ligated and recircularized to form pMV112. Schematics of the constructions of pMVlll and pMV112 are shown in Figure 9. These constructions further eliminated superfluous E. coli vector sequences derived from pAL5000 not necessary for mycobacterial replication. Cloning was performed in E. coli. Plasmids pMVlll and pMV112 were tested for the ability to replicate in V _ smegmatis. Because both plasmids replicated in M. sme ^g matis the deletions of each plasmid were combined to construct pMV113. (Figure 9).

SUBSTITUTE SHEET

To construct pMV113, pMVlll was digested with BamHI and EcoRI, and a 1071 bp fragment was isolated. pMV112 was digeste _d with BamHI and EcoRI, and a 3570 bp fragment was isolated, and then ligated to the 1071 bp fragment obtained from pMVlll to form pMV113. These constructions thus defined the region of pAL50 ₀₀ necessary for autonomous replication in mycobacteria as no _l ar _g er than 1910 base pairs.

3.. Mutagenesis of restriction sites in mycobacterial replicon.

To facilitate further manipulations of the mycobacterial replicon, PCR mutagenesis was performed as above to eliminate the Sal I, EcoRI, and Bglll sites located in the open reading frame known as ORFl of pAL5000. PCR mutagenesis was performed at wobble bases within each restriction site and the base substitutions were designed not to change the amino acid sequence of the putative encoded ORFl protein. The restriction sites were eliminated one at a time for testing in mycobacteria. It was possible to eliminate the Sail and EcoRI without altering replication in M. smeqmatis. In one construction PCR mutagenesis was performed at EcoRI1071 of pMV113 with primers Eco Mut - M.rep and Bam-M.rep to form pMV117, which lacks the EcoRI1071 site. Primer Eco Mut - M.rep has the following sequence:

TCC GTG CAA CGA GTG TCC CGG A; 7 and Bam-M.rep has the following sequence:

CAC CCG TCC TGT GGA TCC TCT AC. ° In another construction, PCR mutagenesis was performed at the Sail 13 ^~ 89 site with primer Sal Mut - M.rep and Bam-M.rep to form pMV119, which lacks the Sail 1389 site. Primer Sal Mut- M.rep has the following sequence:

TGG CGA CCG CAG TTA CTC AGG CCT. f pMV117 was then digested with ApaLI and Bglll, and a 3360 bp fragment was isolated. pMV119 was digested with ApaLI and Bglll, and a 1281 bp fragment was isolated and ligated to the 3360 bp fragment isolated from pMV117 to form pMV123. A schematic of the

SUBSTITUTE SHEET

constructions of plasmids pMV117, pMV119, and pMV123 is shown i Figure 10. Elimination of the Bglll site, however, either _b y P mutagenesis or Klenow fill in, eliminated plasmid replication i mycobacteria, thus suggesting that the Bglll site is in proximi to, or within a sequence necessary for mycobacteria plasmid replication.

4. Construction of pMV200 series vectors.

To facilitate manipulations of all the components necessar for plasmid replication in E. coli and mycobacteria, (E. rep. a

M. rep.) and selection of recombinants (Kan ), cassettes of eac component were constructed for simplified assembly in future vectors and to include a multiple cloning site (MCS) containing unique restriction sites and transcription and translation terminators. The cassettes were constructed to allow directiona cloning and assembly into a plasmid where all transcription is unidirectional. e Kan Cassette

R

A DNA cassette containing the aph (Kan ) gene was constructed by PCR with primers Kan5' and Kan3' . An Spel site was added to the 5' end of the PCR primer Kan3' , resulting in th formation of a PCR primer having the following sequence: CTC GAC TAG TGA GGT CTG CCT CGT GAA G. f

Bam HI + Nhel sites were added to the 5' end of the primer Kan5' , resulting in the formation of a PCR primer having the following sequence:

CAG AGG ATC CTT AGC TAG CCA CT GAC GTC GGG G. // PCR was performed at bases 3375 and 4585 of pMV123, and BamHI and Nhel sites were added at base 3159, and an Spel site was added at base 4585. Digestion with BamHI and Spel, followed

D by purification resulted in a 1228/2443 Kan cassette bounded by BamHI and Spel cohesive ends with the direction of transcription for the aph gene proceeding from BamHI to Spe I. E. rep, cassette

_SUBS TITUTE SHEET

A DNA cassette containing the ColEI replicon of pUC19 was constructed by PCR with primers E.rep/Spe and E.rep/Mlu. An Spe site was added to the 5' end of PCR primer E.rep/Spe and an Mlul site was added to the 5' end of PCR primer E.rep./Mlu. The resulting primers had the following se-quences:

E.rep./Spe

CCA CTA GTT CCA CTG AGC GTC AGA CCC /-?

GAC AAC GCG TG CGC TCG GTC GTT CGG CTG. I S

PCR was performed at bases 713 and 1500 of pUC19, and an Mlul site was added to base 713, and a Spel site was added to base 1500. Digestion with Mlul and Spel, followed by purification resulted in an E.rep. cassette bounded by Spel and Mlul cohesive ends with the direction of transcription for RNA I and RNA II replication primers proceeding from Spel to Mlul.

M.rep. cassette

A DNA cassette containing sequences necessary for plasmid replication in mycobacteria was constructed by PCR of pMV123 with primers M.rep/Mlu and M.rep/Bam. An Mlul site was added to the 5' end of PCR primer M.rep/Mlu. A BamHI site was added to the 5' end of PCR primer M.rep/Bam. The resulting PCR primers had the following base sequences:

M.rep.pMlu

M.rep./Bam

CAC CCG TCC TGT GGA TCC TCT AC ^>

PCR was performed at bases 134 and 2082 of pMV123. An Mlul sited was added to base 2082. Digestion with BamHI and Mlul, followed by gal purification resulted in a 1935 base pair DNA cassette bounded by Mlul and BamHI cohesive ends with the direction of transcription for the pAL5000 ORFl and ORF2 genes proceeding from Mlul to Bam HI.

The Kan ^R , E.rep, and M.rep PCR cassettes were then mixed in equimolar concentrations and ligated, and then transformed in E^

SUBSTITUTE SHEET

coli strain HB101 for selection of Kan transformants. Colonies were screened for the presence of plasmids carrying all three cassettes after digestion with BamHI + Mlul + Spel and designate pMV200. An additional restriction site, Ncol, was eliminated from the M.rep cassette by digestion of pMV200 with Ncol, fill i with Klenow, and ligation and recircularization, resulting in th formation of pMV201. A schematic of the formation of pMV200 fro pMV123 and pUC19, and of pMV201 from pMV200, is shown in Figure 11. Plasmids pMV200 and pMV201 were transformed into M. smeqmatis and BCG. Both plasmids yielded Kan ^R transformants, thus indicating their ability to replicate in mycobacteria.

A synthetic multiple cloning sequence (MCS) (Figure 12) was then designed and synthesized to facilitate versatile molecular cloning and manipulations for foreign gene expressions in mycobacteria, and for integration into the mycobacterial chromosome. The synthetic MCS, shown in Figure 12, contains 16 restriction sites unique to pMV201 and includes a region carryin translation stop codons in each of three reading frames, and a transcription terminator derived from E. coli 5S ribosomal RNA (Tl).

To insert the MCS cassette, pMV201 was digested with Nar and Nhel, and the resulting fragment was gel purified. The MCS was digested with HinPI and Nhel and, the resulting fragment was gel purified. The two fragments were then ligated to yield pMV204. A schematic of the construction of pMV204 is shown in Figure 13.

Plasmid pMV204 was then further manipulated to facilitate removal of the M.rep cassette in further constructions. pMV204 was digested with Mlul, and an Mlul - Not I linker was inserted into the Mlul site between the M.rep and the E.rep to generate pMV206. A schematic of the construction of pMV206 from pMV204 i shown in Figure 14, and the DNA sequence of pMV206 is given in Figure 15.

5. Construction of expression cassette based on BCG HSP60.

SUBSTITUTE SHEET

The HSP61 cassette (Figure 6) was constructed as hereina _bov described in Example 1.

Each of pMV206 and the PCR cassette HSP61 was digested with Xbal and BamHI. The PCR cassette was then inserted between the Xbal and BamHI sites of pMV206, then ligated to form plasmi _d pMV261. The construction of this plasmid is shown schematically in Figure 17. The reading frame and the restriction sites of th multiple cloning site of pMV261 is shown in Figure 16.

The E. coli lac 2 gene was used as a reporter, or marker gene to assay the ability of the HSP61 cassette to express heterologous genes in BCG. A BamHI restriction fragment carrying the lac 2 gene was cloned into the Bam HI site of Bam HI digested pMV261, resulting in the formation of pMV261/L2. A schematic of the construction of pMV261/L2 is shown in Figure 18. The formation of pMV261/L2 results in a fusion between the HSP60 anβ lac 2 genes at the sixth codon of the HSP60 gene and the sixth . codon of the lac Z gene. pMV261/L2 was then transformed into E. coli. Blue E. coli colonies were selected on x-gal plates for the presence of pMV261/L2, thus indicating that the HSP60 promoter and translation initiation sequences were also active in E. coli. pMV261/L2 was then transformed into BCG and plated on Dubos Oleic Agar plates containing x-gal. All BCG colonies resulting from this transformation exhibited blue color, thus indicating that the lac 2 gene product (B-galactosidase) was expressed in BCG. SDS polyacrylamide gel electrophoresis was performed on lysates of the pMV261/L2 BCG recobinants, revealing that B-galactoβidaβe protein was expressed to levels in excess of 10% of total BCG protein (as determined by staining with Coomassie brilliant blue). These data indicated that BCG HSP61 expression cassette was functional in expression vector pMV261.

Plasmid pMV261/L2 was then shown to replicate autonomously, an _d express the E. coli B-galactosidase, or lac2 gene, driven by the BCG promoter HSP60, in M. smeqmatis and BCG.

_S UBSTITUTE SHEET

6. Transfer of m ^y cobacteriophage L5 integration sequences to BCG expression vector.

Plasmid pMH9.4, which includes the mycobacteriophage L5 att site, and the L5 integrase gene, was employed in providing the L integration sequences to a BCG expression vector. The construction of pMH9.4, as well as its integration into M. smegmatis and BCG, is described below in sections (i) through (vi). (i) Identification of the DNA sequences of the attachment sites. attB, attL, and attR, of M.smegmatis.

Using standard technologies, a lambda EMBL3 library wa constructed using chromosomal DNA prepared from mc 61 (a strai of M. smeqmatis which includes an M■ smeqmatis chromosome int which has been integrated the genome of mycobacterial phage L-5" and digested with Bam HI. Phage L5 contains DNA havin restriction sites identical to those of phage LI (Snapper, et al. 1988), except that L5 is able to replicate at 42°C and phage LI is incapable of such growth. This library was then probed with 6.7 kb DNA fragment isolated from the L5 genome that had been previously identified as carrying the attP sequence (Snapper, et al 1988). One of the positive clones was plaque purified, DN prepared, and a 1.1 kb Sal I fragment (containing the AttL sequence) sub-cloned into sequencing vector pUC119. The DN sequence of this fragment was determined using a shotgun approac coupled with Sangβr sequencing. By isolating and sequencing the attL junction site and comparing this to the DNA sequence of L5 that was available, a region was determined where the tw sequences aligned but with a specific discontinuity present. Th discontinuity represents one side of a core sequence, which i identical in AttP, attB, and attL. The region containing th recombinational crossover point is shown in Figure 19.

The attL DNA (1.1 kb Sal I fragment) was used as a probe t hybridize to a Southern blot of Bam HI digested mc 6 DNA, whic

SUBSTITUTE SHEET

is a strain of M. sme ^q matis which includes an M. smeqmati chromosome without any phage integration (Jacobs, et al, 198 ₇ hereinabove cited.). A single band of approximately 6.4 kb wa detected corresponding to the attB sequence of M. smeqmati _s This same attL probe was used to screen a cosmid library o _f mc ² (provided by Dr. Bill Jacobs of the Albert Einstein College o Medicine of Yeshiva University), and a number of positive cosmi clones were identified. DNA was prepared from these clones, an a 1.9 kb Sal I fragment (containing the attB site) tha hybridizes to the attL probe was subcloned into pUC119 fo sequencing and further analysis. The DNA sequence containing th core sequence was determined and is shown in Figure 19. The cor sequence, which is identical in attP, attB and attL, has a lengt of 43bp.

2

The mc 61 lambda EMBL3 library was then probed with tΗ

1.9kb Sail fragment containing the attB site. Positive plaque were identified, DNA was prepared, and analyzed by restrictio analysis and Southern blots. Lambda clones were identified that contained a 3.2kb Bam HI fragment containing the putative attR site. The 3.2kb Bam HI fragment was purified and cloned into pUC119 for sequencing and further analysis.

(ii) Determination of attP-integrase region of L5 genome.

Concurrent with the above procedures, a significant portion of the DNA sequence of L5 had been determined and represented in several "contigs" or islands of DNA sequence. Sequences of the 6.7kb Bam HI fragment hereinabove described were determined by (a) analysis of the location of Bam HI sites in the contigs of the DNA of L5, and (b) by determining a short stretch of DNA sequence from around the Bam HI sites of plasmid pJR-1 (Figure 24), which carries the 6.7kb Bam HI fragment of L5.

A segment of DNA sequence was located that represented the 6.7kb Bam HI fragment of phage L5. Studies of other phages have shown that the integrase genes are often located close to the attP site. It was thus determined that the L5 integrase (int)

SUBSTITUTESHEET

gene should lie either within the 6.7kb Bam HI fragment or in DNA sequence on either side of it. The DNA sequence in t regions was then analyzed by translating it into all six possibl reading frames and searching these amino acid sequences fo similarity to the family of integrase related proteins, an through computer-assisted analysis of the DNA sequence. As show in Figure 20, there are shown two domains of reasonably goo conservation among L5 integrase and other integrases, and thre amino acid residues that are absolutely conserved in domain 2 (See Yagil, et al. , J. Mol. Biol.. Vol. 207, pgs. 695-717 (198 ₉₎ and Poyart-Salmeron, et al., J. EMBO. , Vol. 8, pgs. 2425-243 (1989)). A region was identified, and analysis of th corresponding DNA sequence showed a reading frame that coul encode for a protein of approximately 333 amino acids. Thes observations identified the putative int gene.

The location of the int gene was not within the 6.7kb Bam-H fragment; however, it was very close to it with one of the Bam H sites (that defines the 6.7kb Bam HI fragment) less than 100 b upstream of the start of the gene. Analysis of the Bam HI site showed that the int gene lay within a 1.9kb Bam HI fragmen located adjacent to the 6.7kb Bam HI fragment. This 1.9kb Bam H fragment was cloned by purification of the fragment from a Bam H digest of L5 DNA and cloning into pUC 119, to generate pMH (Figure 25) .

From a combination of the above approaches, a schematic o the organization of the attP-int region of L5 was constructe (Figure 26), and the gene sequence of the attP-int region i given in Figure 22.

(iii) Construction of pMH5.

The 6.7kb Bam HI fragment of mycobacteriophage L5, whic contains the attP site, as hereinabove described, was cloned int the Bam HI site of pUC 119 (Figure 23). This was achieved b purifying the 6.7kb Bam HI fragment from a Bam HI digest of L

SUBSTITUTE SHEET

DNA separated by agarose gel electrophoresis and ligating wit Bam HI cut pUC 119. DNA was prepared from candidate recombinan _t s and characterized by restriction enzyme analysis an _d gel electrophoresis. A recombinant was identified that contained the 6.7kb Bam HI fragment of L5 cloned into pUC 119. This plasmid was named pJR-1, as shown in Figure 24.

Analysis of DNA sequence data from a project to sequence L5 showed that a 1.9kb Bam HI fragment adjacent to the 6.7kb Bam HI fragment hereinabove described contained the integrase gene.

A plasmid containing a 1.9kb Bam HI fragment containing the DNA encoding for the integrase cloned into the Bam HI site of pU _C 119 was constructed. The 1.9kb fragment was purified from a Bam HI digest of L5 DNA and cloned into the Bam HI site of pUC 119. Construction of the recombinant was determined by restriction analysis and gel electrophoresis. This plasmid was called pMlfl, the construction of which is shown schematically in Figure 25. pJR-1 was then modified by digestion with EcoRI and SnaBI (both are uni«que cloning sites), between which is a Bam HI site. The EcoRI-SnaBI fragment, including the Bam HI site was excised, and the. plasmid was religated to form plasmid of pMH2, which contains on Bam HI site compared to two Bam HI sites contained in pJR-1. A schematic of the construction of pMH2 is shown in Figure 26.

The 1.9kb Bam HI fragment, which includes the integrase gene, was purified from a Bam HI digest of pMHl and ligated to Bam HI digested pMH2. Recombinants were identified as above and the orientation of the 1.9kb fragment determined. A plasmid called pMH4 was thus constructed (Figure 27) in which the region from the Sna BI site (upstream of attP) through to the Bam HI site ₍ downstream of the integrase gene) was identical to that in

L5. pMH4 was digested with HindiII (unique site) and was ligated to a lkb Hindlll fragment purified from pKD43 (supplied by Keith Derbyshire of the Nigel Gindley Laboratory) that contains the

SU _B STITUTE SHEET

gene determir. ig resistance to kanamycm. .wer identified and characterized as above. This ^* -biasmid i*s -c"a11«et pMH5. A schematic of the construction of pMH5 is shown in Figur 28. fiv) Integration of PMH5 into attB of M. smegmatis.

Plasmids pYUB12 (a gift from Dr. Bill Jacobs, a schematic o the formation of which is shown in Figure 1), pMDOl (Figure 29 ₎ , and pMH5 were electroporated, with four different concentration of plasmid DNA over a 1,000-fold range, into M. smegmatis strai

2 mc 155, a strain which is able to support plasmid replication.

In sections (iv) through (vi), all electroporation procedures o

M. smegmatis, or of BCG, were carried out as follows:

Cultures of organism were grown in Middlebrook 7H9 media, a described by Snapper, et al. (1988), harvested by centrifugation, washed three times with cold 10% glycerol, and resuspended a approximately a 100 x concentration of cells.

1 μl of DNA was added to 100 μl of cells in an ice-col cuvette and pulsed in a Bio-Rad Gene Pulser, and given a singϊ pulse at 1.25 kv at 25 μF. 1 ml of broth was added the cell incubated for 1 hr. at 37°C for expression of the antibiotic- resistant marker. Cells were then concentrated and plated out on Middlebrook or tryptic soy media containing 15 μg/ml kanamycin. Colonies were observed after 3 to 5 days incubation at 37°C.

Each of pYUB12, pMDOl, and pMH5 carries kanamycin resistance. Plasmid pYUB12 carries an origin of DNA replication, while pMDOl lacks a mycobacterial origin of replication. Plasmid pMH5 does not carry a mycobacterial origin of replication, but carries a 2kb region of phage L5 which contains the attP site and the integrase gene (Figure 22). The number of transformants were linear with DNA concentration. Plasmid pYUB12 gives a large number of transformants (2 x 10 5 per μg DNA) in mc2155, while pMH5 gives 6 x 10 ⁴ transformants per μg DNA, and pMDOl gives no transformants.

SUBSTITUTE SHEET

The above experiment was then repeated by electroporati the plasmids pYUB12, pMDOl, and pMH5 into M. smeqmatis strai mc 6, which does not support plasmid replication. _N

2 transformants in mc 6 were obtained from pYUB12 or pMDOl whil ' pMH5 gave approximately 10 kanamycin resistant transformants ι

2 mc 6 per μg of DNA, thus indicating integration of pMH5 into th mc 6 chromosome.

DNA from six independent pMH5 transformants (four in mc ² "■•- 2 and two in mc 6) was prepared. These DNA's (along with DNA fro both mc 155 itself, and mc 155 carrying the plasmid pYUB12) wer digested with a restriction enzyme, and analyzed by Southern blo and hybridization with the M. smeqmatis 1.9kb attB prob hereinabove described. As shown in Figure 30, all si transformants have integrated into the attB site, resulting JL the production of two new DNA fragments with differen mobilities. If pMH5 did not integrate into the attB site, i would be expected that a single band, corresponding to the att

2 site in the mc 155 control, would be obtained.

(v) Construction of pMH9.2 and pMH9.4 pUC119 was digested with Hindlll, and a lkb Hindi11 fragment, containing a kanamycin resistance gene, purified from pKD43, was ligated to the Hindi11 digested pUC119 to form pMH8

(Figure 31). A 2kb Sail fragment (bp 3226-5310), which carries the attP and integrase gene from Sail digested pMH5, was purified and inserted in both orientations relative to the vector backbone of Sail digested pMH8 to form plasmids pMH9.2 and pMH9.4 (Figures

32 and 33).

₂ M. smeqmatis strain mc 155 cells carrying, as a result of electroporation, plasmid pYUB12, pMH9.2 or pMH9.4, or strain mc 6 cells carrying plasmid pMH5, as a result of electroporation as hereinabove described, were grown to saturation in broth with kananmycin. Cultures were then diluted 1:100 into broth without kanamycin and grown to saturation. Two further cycles of dilution and growth were done, corresponding to about 20

SUBSTITUTESHEET

generations of bacterial growth. Cultures were plated out _t single colonies on non-selective plates, and approximately 100 o these colonies were patch plated onto both non-selective an selective plates. The % of colonies that were sensitive t kanamycin, thus corresponding to the percentage of cells whic lost the plasmid, is given below in Table I.

Table I % loss pYUB12 (mc ² 155) 35 pMH5 (mc ² 6) 17

PMH9.2 (mc ² 155) 3

PMH9.4 (mc ² 155) 0

(vi) Transformation of BCG with pMH9.4

The 1.9 kb Sal I fragment, which includes the M. smeqmati attB site as hereinabove described was cloned into pUC119, an the plasmid generated was named pMH-12. (Figure 34).

Gel purified Sal I 1.9kb M. smeqmatis fragment containin attB (isolated from pMH-12) was used to probe a Southern transfe of Bam HI digested mycobacterial DNA's, including BCG substrai Pasteur, shown in Figure 35. This demonstrated that there is on Bam HI fragment of BCG that strongly hybridizes to the M. smeqmatis attB probe and three hybridize weakly. The stronges hybridizing band is the fastest moving band (approximately 1. kb).

The same probe as above was used to probe a BCG cosmi library (provided by Dr. Bill Jacobs) and positive clones wer identified. DNA was prepared from several positive clones an analyzed by restriction analysis and Southern blotting. The 1. kb Bam HI fragment (corresponding to the strongly hybridizin band in the Southern blot was identified, gel purified from th cosmid DNA and cloned into pUC119. The resulting plasmid wa named pMH-15. (Figure 36).

Plasmid pMH-5 and pMH9.4 were electroporated into BC Pasteur. It was observed that pMH9.4 transforms BCG with hig

_S U _B STITUTE SHEET

4 efficiency (approximately 10 transformants/μg DNA), while pMH-5 transforms BCG at low efficiency (1-10 transformants/μg DNA).

DNA was prepared from BCG transformants and analyzed by Bam HI restriction and Southern blot analysis, probing with gel purifie _d

1.9kb Bam HI BCG attB fragment from pMH-15. These data are shown in Figure 37 and show that integration of both pMH5 and pMH9. is specific to the BCG attB site (ie. the strongly cross-hybridizing fragment in BCG). This is illustrated by the loss of the 1.9kb

Bam HI fra«gment from the transformants and the appearance of two new bands representing attL and attR junction fragments. Figure

37 shows just one of the pMH5/BCG transformants, although all of the four that were analyzed show that one of the bands (the largest) is smaller than expected (and different in each of the transformants), indicating that the transformation effiency of pMH-5 is low in BCG. In contrast, the four pMH9.4 transformants are identical to each other (Figure 37) and give attR and attL junction fragments of the predicted sizes.

Plasmid pMH9.4, which includes the mycobacterial phage L5 attP site and the L5 integrase gene, was digested to completion with either Kpnl + PvuII or Xbal + PvuII, and a restriction fragment of 1862 or 1847 base pairs, respectively, each of which contain the attP site and the integrase gene, were purified by agarose gel electrophoresis. Plasmid pMV261/L2 was digested with

Xbal or Dral to generate either a 7569 bp or 7574 bp vector fragment. The 7569 bp fragment was ligated to the 1862 bp fragment derived from pMH9.4 to form pMV460/L2. The 7574 bp fragment was ligated to the 1847 bp fragment derived from pMH9.4 to form pMV460 R L2. Plasmids pMV460 F/L2 and pMV460R/L2 each include a mycobacterial replicon, the L5 attP site, and the L5 integrase gene. A schematic of the formation of plasmids pMV460

F/LZ and pMV460R/LZ is shown in Figure 38. To generate derivatives without the mycobacterial plasmid replicon, plasmids pMV460/LZ and pMV460/LZ were digested with NotI an ^d recircularized by ligation to generate pMV360F/LZ and pMV360R/LZ.

SUB _S TITUTE SHEET

A schematic of the construction of pMV360F/LZ and pMV360R/LZ i shown in Figure 39.

Plasmids pMH9.4, pMV261/LZ, pMV460/LZ, pMV460/LZ, pMV460/LΣ and pMV460/LZ were then transformed into M. smeqmatis and B _CG t test their ability to replicate autonomously or integrate int the M. sme ^q matis or the BCG chromosome. Transformation wit pMH9.4, pMV261/LZ, pMV360F/LZ, and pMV360R/LZ yielded kananmyci resistant transformants of M. smeqmatis and BCG. Transformant of pMV261LZ, pMV360F/LZ, and pMV360R/LZ were shown to express E coli B-galactosidase by SDS-polyacrylamide gel electrophoresi and X-gal assay. Plasmids pMV460F/LZ and pMV460R/LZ failed t yield kanamycin resistant transformants, thus indicating tha chromosomal integration of a plasmid carrying sequences mediatin autonomous replication is lethal to mycobacteria.

7. Construction of pMV307.

Plasmid pMV206 was digested with NotI to remove th mycobacterial replicon. The resulting 2209 bp fragment, whic o includes the aph (Kan ) gene, the E. coli replicon and th multiple cloning site, was ligated and recircularized to for pMV205, the construction of which is schematically depicted i

Figure 14.

PCR with primers Xbal-Att/Int and Nhel-Att/Int was the performed on a Sal I fragment from pMH9.4, which contains th attP site and the L5 integrase gene. The resulting cassette wa then digested with Xbal and Nhel and a 1789 bp fragment was ge purified. pMV205 was then digested with Nhel, and the resultin fragment was ligated to the 1989 bp fragment obtained from pMH9. to form pMV307. A schematic of the construction of pMV307 i shown in Figure 40.

8. Construction of pMV261.HIVl-qp 120.

An Smal-Clal antigen gene fragment, or cassette, wa constructed by PCR, and cloned between the Bam HI and Cla restriction sites of pMV261 to form pMV261/HIVl-gp 120.

_SUB STITUTE SHEET

Plasmid pMV261/HIVl-gp 120 was transformed into BCG, and the presence of the corresponding antigen in BCG was verified by the appearance of immunoreactive protein bands in Western blot analysis of BCG recombinant lysates. 9. Construction of pMV361.HIVl-qp_12_0

The HIVl-gp 120 antigen gene expression cassette, which includes a promoter sequence and an HIVl-gp 120 gene sequence, was excised from the pMV261 derivatives with NotI and a second restriction enzyme site (Pvu II, Eco RI, Sal I, Cla I or Hind III) and cloned into the integrating plasmid pMV307 between the NotI site and a second enzyme site (Pvu II, Eco RI, Sal I, Cla I or Hind III) to form the plasmid pMV361/HIV-Igpl20. The backbone of this plasmid is shown in Figure 41.

Plasmid pMV361/HIV-Igpl20 was transformed into BCG and shown to express the corresponding antigens by Western blot analysis (Figure 42) with the appropriate antigen-specific human sera.

Example 4 Cytotoxic T lymphocyte responses to HIV-l-gp 120

HIV-1 gp 120 was expressed in BCG as a six amino acid fusion protein with BCG hsp 60 protein using vector pMV361/HIV-l-gpl20, using the hsp 60 promoter to control expression.

Two groups of mice were inoculated with 1 X 10 CFU's of recombinant BCG expressing the gp 120 gene from the integrative plasmid pMV361/HIV-l-gpl20. One group received the BCG via intraperitonal injection (100 ul) whereas the other group received the BCG by deposition of the dose (10 ul) rubbed into a tail scratch.

CTL activity was measured at various times after immunization. CTL activity was measured as follows:

Two mice from each group were sacrificed at various times after immunation, and the spleens were removed. Single cell suspensions were made and the red blood cells were lyzed with ammonium chloride. The cells were stimulated in vitro for 5 days with P815 cells- that were pulsed with peptide P18, a fifteen

SU _B STITUTE SHEET

residue synthetic peptide within HIVl-gp 120. A 4-hour chromium (Cr 51) release assay was then carried out using untreated P815 and peptide P18 pulsed P815 cells as targets. Significant

P18-specific CTL activity was observed in the mice immunized by tail scratch 14 weeks after immunization. At 16 weeks, CTL activity was observed in both groups of mice. Upon repeat of this experiement, CTL activity was observed at time points as early as 8 weeks after immunization.

Example 5

Recombinant BCG transformed with pMV361/HIVl-gp 120 were grown to mid-log phase in Dubos media and concentrated by centrifugation. The bacteria were then resuspended in 15% glycerol and frozen using a rate contolled freezing apparatus.

The bacteria were stored at -70°C until use (referred to as

"vaccine"). A second preparation grown in the same way was not frozen and is referred to as a "fresh" preparation. Prior to immunization of animals, the bacteria were resuspended in PBS +

0.05% Tween 80 to the desired concentration and cup sonicated briefly to disperse clumped bacteria. Six week old BALB/c mice were inoculated with a single dose of 5 x 10 4 cfu fresh bacteria

(determined post inoculation) or 1.5 x 10 frozen bacteria

(determined pre-inoculation) by tail scratch (t.s.) injection.

At 8 weeks post-immunization, splenocytes were harvested from animals and CTL activity was measured (described below) .

Splenocytes from unimmunized animals were used as controls in the

CTL assays.

CTL activity was determined as follows:

Splenocytes (ACK-treated, 5 X 10 /ml) were stimulated in vitro in 10 ml in upright T25 flasks by co-culture for 5 days with mitomycin C-treated P815 cells (5 X 10 /ml) that were pulsed with 250 ug/ml of peptide P18 for one hour. A 4 hr. 51Cr release assay was subsequently performed in triplicate using P815 targets with or without pulsing for 1 hour with 250 ug/ml peptide P18.

Various effector-target ratios were tested using 5000

SUBSTITUTE SHEET

targets/well. Specific lysis was calculated as follows: % specific lysis = 100 X [release by effector cells minus spontaneous release/maximal release minus spontaneious release] .

The results are given in Figure 42.

As shown in Figure 43, both groups of mice showed an increased CTL response at 8 weeks after immunization as compared with unimmunized mice.

Example 6

Recombinant BCG transformed with pMV361/HIVl-gp 120. were grown to mid-log phase in Dubos media and concentrated by centrifugation. The bacteria were then resuspended in 15% glycerol and frozen using a rate contolled freezing apparatus.

The bacteria were stored at -70°C until use (referred to as

"vaccine"). A second preparation grown in the same way was not frozen and is referred to as a "fresh" preparation. Prior to immunization of animals, the bacteria were resuspended in PBS +

0.05% Tween 80 to the desired concentration and cup sonicated briefly to disperse clumped bacteria. Six week old BALB/c mice

4 were inoculated with a single dose of 5 x 10 cfu fresh bacteria

(determined post inoculation) or 1.5 x 10 frozen bacteria

(determined pre-inoculation) by tail scratch (t.s.) injection.

At 8 weeks post-immunization, splenocytes were harvested from animals and CTL activity was measured (described below) .

CTL activity was determined as follows:

Lymph node cells (5 x 10 /ml) were stimulated in vitro in 10 ml in upright T25 flasks by co-culture for 5 days with mitomycin

C-treated P815 cells (5 X 10 /ml) that were pulsed with 250 ug/ml of peptide P18 for one hour. A 4 hr Cr release assay was subsequently performed in triplicate using P815 (matched ) or EL4

(mismatched) targets with or without pulsing for 1 hour with 250 ug/ml peptide P18. Various effector target ratios were tested using 5000 targets/well. Specific lysis was calculated as follows: % specific lysis = 100 x [release by effector cells

SUBSTITUTE SHEET

minus spontaneous release/maximal release minus spontaneous release] .

The results of this assay are given in Figure 44.

As shown in Figure 44, a CTL response to HIV-1 gp 120 using lymph node cells was demonstrated following immunization of mice with BCG transformed with pMV361/HIV-l gp 120.

It is to be understood, however, that the scope of the present invention is not to be limited to the specific embodiments described above. The invention may be practiced other than as particularly described and still be within the scope of the accompanying claims.

S _UB STITUTE SHEET