Mycobacteria functional screening and/or expression vectors

The Mycobacterium genus includes major human pathogens such as

M. leprae

and

M. tuberculosis,

the agents responsible for leprosy and tuberculosis, which remain serious public health problems world-wide.

M. bovis

and

M. tuberculosis,

the causative agents of tuberculosis, are intracellular facultative bacteria. Despite the major health problems linked to these pathogenic organisms, little is known about their exported and/or secreted proteins. In SDS-PAGE analyses of

M. tuberculosis

culture filtrate show at least 30 secreted proteins (1,19,38). Some of them have been characterized, their genes cloned and sequenced (7, 35, 37). Others, although they are immunodominant antigens of major importance for inducing protective immunity (2, 21), have not been completely identified. In addition, it is probable that a great number of exported proteins remain attached to the cell membrane and, consequently, are not present in culture supernatants. It has been shown that proteins located at the outer surface of various pathogenic bacteria, such as the 103 kDa

Yersina pseudotuberculosis

invasin (14) or the 80 kDa

Listeria monocytogenes

internalin (10) play an important role in interactions with the host cells and, consequently, in pathogenicity as in the induction of protective responses. Thus, a membrane-bound protein could be important for

M. tuberculosis

infection as well as for the induction of a protective response against this infection. These proteins could certainly be of interest for the preparation of vaccines.

The BCG (Bacille CalmetteGuérin), an avirulent strain derived from

M. bovis,

has been widely used as vaccine against tuberculosis. It is also a very important vector for the construction of live recombinant vaccines, particularly because of its high immunogenicity. Consequently, the study of the molecular biology of mycobacteria is currently of great interest.

The development of new vaccines against pathogenic mycobacteria, or the improvement of available vaccines required the development of specific tools which make it possible to isolate or obtain immunogenic polypeptide sequences.

The inventors have defined and produced, for this purpose, new vectors allowing the screening of mycobacteria DNA sequences in order to identify, among these sequences, nucleic acids encoding proteins of interest.

Vectors have been defined for evaluating the efficacy of sequences for regulation of expression in mycobacteria.

The invention also relates to new mycobacteria polypeptides which may have been isolated by means of the preceding vectors and capable of entering into the production of compositions for the detection of a mycobacteria infection, or for protection against an infection due to mycobacteria.

The subject of the invention is therefore a recombinant screening and/or cloning and/or expression vector, characterized in that it replicates in mycobacteria, in that it contains

1) a replicon which is functional in mycobacteria;

2) a selectable marker;

3) a reporter cassette comprising

a) a multiple cloning site (polylinker),

b) a transcription terminator which is active in mycobacteria, upstream of the polylinker, and

c) a coding nucleotide sequence derived from a gene encoding a marker for expression and/or export and/or secretion of protein, said nucleotide sequence lacking its initiation codon and its regulatory sequences.

The marker for export and/or secretion is a nucleotide sequence whose expression followed by export and/or secretion depends on regulatory elements which control its expression.

“Sequences or elements for regulation of expression” is understood to mean a promoter sequence for transcription, a sequence comprising the ribosome-binding site (RBS), the sequences responsible for export and/or secretion such as the sequence termed signal sequence.

A first advantageous marker for export and/or expression is a coding sequence derived from the PhoA gene. Where appropriate, it is truncated such that the alkaline phosphatase activity is, nevertheless, capable of being restored when the truncated coding sequence is placed under the control of a promoter and of appropriate regulatory elements.

Other markers for exposure and/or export and/or secretion may be used. There may be mentioned by way of examples a sequence of the gene for β-agarase or for nuclease of a staphylococcus or for β-lactamase of a mycobacterium.

The transcription terminator should be functional in mycobacteria. An advantageous terminator is, in this regard, the T4 coliphage terminator (tT4). Other terminators appropriate for carrying out the invention may be isolated using the technique presented in the examples, for example by means of the vector pJN3.

A vector which is particularly preferred for carrying out the invention is the plasmid pJEM11 deposited at CNCM (Collection Nationale de Cultures de Microorganismes in Paris—France) under the No. I-1375, on Nov. 3, 1993.

For the selection or the identification of mycobacteria nucleic acid sequences encoding products capable of being incorporated into immunogenic or antigenic compositions for the detection of a mycobacteria infection, the vector of the invention will comprise, in one of the polylinker sites, a nucleotide sequence from a mycobacterium in which the presence of regulatory sequences is being sought which are associated with all or part of a gene of interest making it possible, when the vector carrying these sequences (recombinant vector), is intergrated or replicates in a mycobacterium-type cellular host, to obtain the exposure at the level of the cell wall or membrane of the host, and/or export and/or secretion of the product of expression of the abovementioned nucleotide sequence.

The mycobacteria sequence in question may be any sequence for which attempts are made to detect if it contains elements for regulation of expression associated with all or part of a gene of interest and capable of allowing or promoting exposure at the level of the cell membrane of a host in which it might be expressed, and/or export and/or secretion of a product of expression of a given coding sequence and, by way of test, of the marker for export and/or secretion.

Preferably, this sequence is obtained by enzymatic digestion of the genomic DNA or of the DNA complementary to an RNA of a mycobacterium and preferably of a pathogenic mycobacterium.

According to a first embodiment of the invention, the enzymatic digestion of the genomic DNA or of the complementary DNA is carried out using

M. tuberculosis.

Preferably, this DNA is digested with an enzyme such as sau3A.

Other digestive enzymes such as ScaI, ApaI, ScaII, KpnI or alternatively exonucleases or polymerases, may naturally be used, as long as they allow fragments to be obtained whose ends may be inserted into one of the cloning sites of the polylinker of the vector according to the invention.

Where appropriate, digestions with different enzymes will be carried out simultaneously.

Preferred recombinant vectors for carrying out the invention are chosen among the following recombinant vectors deposited at CNCM on Aug. 8, 1994:

pExp53 deposited at CNCM under the No. I-1464

pExp59 deposited at CNCM under the No. I-1465

pExp410 deposited at CNCM under the No. I-1466

pExp421 deposited at CNCM under the No. I-1467.

The vectors of the invention may also be used to determine the presence of sequences of interest, according to what was stated above, in mycobacteria such as

M. africanum, M. bovis, M. avium

or

M. leprae

whose DNA or cDNA will have been treated with determined enzymes.

The subject of the invention is also a process for screening nucleotide sequences derived from mycobacteria, to determine the presence, in these sequences, of regulatory elements controlling the expression, in a cellular host, of nucleic acid sequences containing them, and/or exposure at the surface of the cellular host and/or export and/or secretion of the polypeptide sequences resulting from the expression of the abovementioned nucleotide sequences, characterized in that it comprises the following steps:

a) digestion of mycobacteria DNA sequences with at least one determined enzyme and recovery of the digests obtained,

b) insertion of the digests into a cloning site, compatible with the enzyme of step a), of the polylinker of a vector above,

c) if necessary, amplification of the digest contained in the vector, for example by replication of the latter after insertion of the vector thus modified into a determined cell, for example

E. coli,

d) transformation of cellular hosts by the vector amplified in step c), or in the absence of amplification, by the vector of step b),

e) culture of the transformed cellular hosts in a medium allowing visualization of the marker for export and/or secretion which is contained in the vector,

f) detection of the cellular hosts which are positive for the expression of the marker for exposure and/or export and/or secretion (positive colonies),

g) isolation of the DNA of the positive colonies and insertion of this DNA into a cell which is identical to that of step c),

h) selection of the inserts contained in the vector, which allow clones to be obtained which are positive for the marker for export and/or secretion,

i) isolation and characterization of the fragments of DNA of mycobacteria which are contained in these inserts.

The carrying out of this process allows the construction of DNA libraries containing sequences capable of being exported and/or secreted, when they are produced in recombinant mycobacteria.

Step i) of the process may comprise a step for sequencing the inserts selected.

Preferably, the vector used is the plasmid pJEM11 (CNCM I-1375) and the digestion is carried out by means of the enzyme sau3A.

According to a preferred embodiment of the invention, the screening process is characterized in that the mycobacteria sequences are derived from a pathogenic mycobacteria, for example from

M. tuberculosis, M. bovis, M. avium, M. africanum

or

M. leprae.

The subject of the invention is also the nucleotide sequences of mycobacteria selected after carrying out the process described above.

According to a specific embodiment of the invention, advantageous sequences are for example the mycobacteria DNA fragments contained in the vectors pIPX412 (CNCM I-1463 deposited on Aug. 8, 1994), pExp53, pExp59, pExp410 or pExp421.

When the coding sequence derived from the marker gene for export and/or secretion is a sequence derived from the PhoA gene, the export and/or secretion of the product of the PhoA gene, truncated where appropriate, is obtained only when this sequence is inserted in phase with the sequence placed upstream, which contains the elements controlling the expression and/or export and/or secretion which are derived from a mycobacteria sequence.

The subject of the invention is also recombinant mycobacteria containing a recombinant vector described above. A preferred mycobacterium is a mycobacterium of the

M. smegmatis

type.

M. smegmatis

makes it possible, advantageously, to test the efficiency of mycobacteria sequences for controlling the expression and/or export and/or secretion of a given sequence, for example of a sequence encoding a marker such as alkaline phosphatase.

Another advantageous mycobacterium is a mycobacterium of the

M. bovis

type, for example the BCG strain currently used for vaccination against tuberculosis.

A subject of the invention is, moreover, a recombinant mycobacterium, characterized in that it contains a recombinant vector defined above.

The invention also relates to a nucleotide sequence derived from a gene encoding an exported

M. tuberculosis

protein, characterized in that it is chosen from the following sequences:

a sequence IA corresponding to the chain of nucleotides described in

FIG. 6A

, or a sequence IB corresponding to the chain of nucleotides described in

FIG. 6B

, or hybridizing under stringent conditions with these chains,

a sequence II comprising the chain of nucleotides IA or IB and encoding an

M. tuberculosis

P28 protein having a theoretical molecular weight of about 28 kDa and an observed molecular weight of 36 kDa, determined by denaturing acrylamide gel electrophoresis (SDS-PAGE)

a sequence III contained in the sequence IA or IB and encoding a polypeptide recognized by antibodies directed against the

M. tuberculosis

P28 protein,

a sequence IV comprising the regulatory sequences of the gene comprising the coding sequence IA or IB,

a sequence V corresponding to the chain between nucleotides 1 and 72 of the sequence IA or IB and corresponding to the signal sequence,

a sequence VI corresponding to the chain between nucleotides 62 to 687 of the sequence IA or IB,

a sequence VII corresponding to the chain between nucleotides 688 and 855 of the sequence IA or IB.

Also entering within the framework of the invention is an

M. tuberculosis

polypeptide characterized in that it corresponds to the amino acid chain VIIIA or to the chain VIIIB represented in

FIGS. 6A and 6B

respectively or in that it comprises one of these chains.

A preferred polypeptide is characterized in that it has a theoretical molecular weight of about 28 kDa determined according to the technique described in the examples.

The

M. tuberculosis

p28 protein has been characterized by its capacity to be exported and therefore potentially located across the bacterial plasma membrane or the cell wall. Furthermore, as shown in the sequences presented in

FIG. 6

, some peptide units of the sequence are repeated. For these reasons, the

M. tuberculosis

p28 protein is now most often designated as ERP protein and the gene containing the coding sequence for this protein is called either irsa gene or erp gene.

The theoretical molecular weight of the ERP protein, evaluated at 28 kDa, corresponds to an experimentally observed molecular weight of about 36 kDa (electrophonetic migration on a denaturing polyacrylamide gel (DOS-PAGE)).

Another advantageous polypeptide within the framework of the invention comprises part of the amino acid chain VIII or VIIIB previously described and immunologically reacts with antibodies directed against the

M. tuberculosis

p28 protein.

Preferably, such a polypeptide is, in addition, characterized in that it does not immunologically react with the

M. leprae

p28 protein.

Particularly advantageous amino acid sequences within the framework of the invention are the sequences comprising one of the following chains or corresponding to one of these chains in one or more copies: PGLTS(SEQ ID NO:1), PGLTD(SEQ ID NO:2), PGLTP(SEQ ID NO:3), PALTN(SEQ ID NO:4), PALTS(SEQ ID NO:5), PALGG(SEQ ID NO:6), PTGAT(SEQ ID NO:7), PTGLD(SEQ ID NO:8), PVGLD(SEQ ID NO:9).

Other advantageous sequences are, for example, the signal sequence between the positions of nucleotides 1 and 72 of the sequence of

FIG. 6A

or

6

B or alternatively the sequence between nucleotides 688 and 855 which is capable of behaving like a transmembrane sequence.

These polypeptide sequences may be expressed in the form of recombinant polypeptides. In these recombinant polypeptides, they may be replaced in part especially as regards the sequences of 5 amino acids previously described, by sequences of interest obtained from mycobacteria or other pathogenic organisms, it being possible for this replacement to lead to the inclusion, inside the recombinant polypeptides, of the epitopes or the antigenic determinants of a pathogenic organism or of a protein of interest against which it might be sought to obtain antibodies.

Thus, the polypeptides of the invention, while optionally exhibiting themselves the antigenic or even immunogenic properties, may be used as advantageous carrier molecules for preparing, where appropriate, vaccines having varying properties.

The subject of the invention is also monoclonal antibodies or polyclonal sera directed against a polypeptide as defined above.

As regards monoclonal antibodies, they are preferably directed specifically against a polypeptide of the invention and do not recognize, for example, the

M. leprae

p28 protein.

The subject of the invention is also a composition for the in vitro detection of an

M. tuberculosis

infection, characterized in that it comprises a polypeptide defined above, which is capable of immunologically reacting with antibodies formed in a patient infected with

M. tuberculosis.

Another composition for the in vitro detection of an

M. tuberculosis

infection is characterized by a nucleotide sequence containing at least 9 nucleotides, which is derived from a sequence defined above, or a nucleotide sequence containing at least 9 nucleotides and hybridizing, under stringent conditions, with

M. tuberculosis

DNA and not hybridizing, under the same conditions, with

M. leprae

DNA, this sequence being a DNA or RNA sequence, which is labeled where appropriate.

The subject of the invention is also a prokaryotic or eukaryotic cellular host, characterized in that it is transformed by a nucleotide sequence as described in the preceding pages, under conditions allowing the expression of this sequence and/or its exposure at the level of the membrane of the cellular host and/or its export and/or its secretion from the abovementioned membrane.

Preferably, the cellular hosts are mycobacteria such as

M. smegmatis

or

M. bovis

BCG.

Other cellular hosts are for example

E. coli,

CHO, BHK, Spf9/Baculovirus cells, yeasts such as Saccharomyces cerevisiae, vaccinia virus.

The subject of the invention is also an immunogenic composition comprising a polypeptide as presented above or a cellular host as defined above.

The invention relates, moreover, to a vector for the screening and/or cloning and/or expression of nucleotide sequences which are functional in mycobacteria, and which is derived from a vector described above and characterized in that the coding sequence derived from a gene encoding a marker for export and/or secretion is replaced by a reporter gene or a reporter sequence.

Preferably, the reporter sequence or gene lacks its regulatory sequences, in particular its ribosome binding sequences and/or its sequences which allow the export and/or secretion of the marker produced when the vector is incorporated into a recombinant cellular host.

Preferably, the reporter sequence or gene contains the sequence encoding the lacZ gene or a part of this sequence which is sufficient for the polypeptide to exhibit a β-galactosidase activity.

A preferred vector of the invention is characterized in that it comprises at one of the cloning sites of the polylinker, a chain of nucleotides comprising a promoter and, where appropriate, regulatory sequences, for example for anchorage at the surface, the export or even the secretion of a polypeptide which might be produced under the control of the promoter, for which it is desired to evaluate the capacity to promote or regulate the expression of a reporter nucleotide sequence in mycobacteria.

Preferred vectors are plasmids chosen from the plasmids pJEM12, pJEM13, pJEM14, or pJEM15 as represented in FIG.

12

.

Such a vector may be used to evaluate the value of sequences for regulation of expression or of promoters, for example, the pAN, pblaF*, psul3, pgroES/EL1 sequences.

The invention also comprises a process for determining the activity of a sequence containing at one of the cloning sites of the polylinker a chain of nucleotides comprising a promoter and, where appropriate, regulatory sequences, for example for the exposure, export or even secretion of a polypeptide which might be produced under the control of the promoter in mycobacteria, characterized in that it comprises the steps of:

transforming a mycobacterium strain, for example

M. smegmatis

or

M. tuberculosis,

with a vector described above,

detecting the activity normally associated with the presence of the reporter gene or of the reporter sequence.

Other characteristics and advantages of the invention appear on reading the examples which follow as well as in the figures.

LEGEND TO THE FIGURES

FIG. 1

Construction of pJEM11.

See Materials and Methods. pJEM11 has replication origins (ori) of

E. coli

and mycobacteria. It is therefore a shuttle plasmid. The selectable marker is the kanamycin (Km) resistance gene. The truncated PhoA gene of pPH07 (22) lacks a promoter, a start codon and a signal sequence; thus the expression and export of PhoA depend on the translational fusion with the amino-terminal ends of other proteins. The transcriptional terminator (T) of the omega cassette avoids transcription by “read-through” using plasmid sequences.

FIG. 2

Construction of the plasmids pLA71, pLA72 and pLA73.

The insertion into the BamHI site of pJEM11 of BlaF* fragments (34) of 3 different lengths lead to the expression of fusion proteins with the phoA activity. Colorimetric assays were carried out according to the Brockman and Heppel technique (8), with p-nitrophenyl phosphate as substrate. The protein contents were measured with the aid of the Bio-Rad assay. The arbitrary alkaline phosphatase units (aU) were calculated as described in Materials and Methods.

FIG. 3

Western-blot analyses of PhoA fusion proteins.

Transformed

M. smegmatis

strains were cultured in Beck's medium containing kanamycin (20 μg/ml). Total extracts of sonicated bacteria were solubilized with SDS, resolved by SDS-PAGE and subjected to immunoblotting. The preparation of the rabbit anti-PhoA serum has been previously described (34). PhoA-coupled rabbit antibodies (Promega) and, as substrate, a mixture of X-P and nitro blue tetrazolium (BCIP-NBT, Promega) were used to reveal the PhoA fusions. Column 1: purified bacterial PhoA,

M. smegmatis

transformed by plasmids pJEM11: column 2, pLA71: column 3, pLA72: column 4, pLA73: column 5, pExp410: column 6, pExp53: column 7, pExp59: column 8, pExp421: column 9.

FIG. 4

Nucleotide sequences and deduced amino acid sequences of segments of inserts selected from the plasmids pExp410, pExp53, pExp59 and pExp421.

The

M. smegmatis

clones with the alkaline phosphatase activity were selected on X-P/kanamycin dishes. Their plasmids were amplified in

E. coli

XL-1 B, and the nucleotide sequence of the inserts determined as described in Materials and Methods. A: pExp410 includes part of the 19 kDa lipoprotein. The reading frame is maintained at the junction with phoA (BamHI/Sau3A). B: pExp53 includes part of a gene exhibiting similarities with the 28 kDa

M. leprae

antigen. The divergent amino acids are in bold type. The codon for initiation of translation is GTG. The putative sites of cleavage by signal peptidase are indicated by arrows. C: pExp59 encodes a characteristic signal sequence. A putative ribosome-binding site (RSB) is underlined. The putative site of cleavage by signal peptidase is indicated by an arrow. D: pExp421 encodes conserved amino acid units conserved with proteins of the family of stearoyl-acyl carrier protein (ACP) desaturases. R. comm: R. communis (ricin).

FIG. 5

The gene which is similar to the gene for the 28 kDa

M. leprae

antigen is present in a single copy in the

M. tuberculosis

genome.

The

M. tuberculosis

genomic DNA was extracted according to standard procedures (27), digested with endonucleases PstI, SmaI, BstEII, SphI, BamHI and subjected to migration on a 1% agarose gel. The Southern-blot hybridization was carried out according to standard procedures (27). The 32P-labeled probe was a 180 bp PCR fragment of the pExp53 insert.

FIG. 6

Nucleotide sequence (IA and IB) and amino acid sequence (VIIIA and VIIIB) of the product of the IRSA gene encoding the

M. tuberculosis

P28 protein (two variants are presented). This gene is now designated by the abbreviation “erp” corresponding to the expression “exported repetitive protein”.

FIG. 7

Preliminary nucleotide sequences flanking the

M. tuberculosis

IRSA gene.

FIG. 8

Bacteria genes for the regulation of iron (IRG's).

FIG. 9

Hydrophilicity profile of the

M. leprae

and

M. tuberculosis

P28 PROT2INS.

FIG. 10

A) Alignment of the nucleotide sequences of the gene encoding the

M. tuberculosis

and

M. leprae

p28 proteins.

B) Alignment of the amino acid sequences of the

M. tuberculosis

and

M. leprae

p28 proteins.

FIG. 11

Construction of the plasmids pJN3 and pJN11.

Only the relevant genetic elements and restriction sites are shown. The plasmids pRR3 and pJN1 have been described in the prior art (60) (58). The omega cassette was obtained by digestion of pHP45X with SmaI (59), followed by an agarose gel purification of a 2 kb fragment using the Geneclean kit (Bio 101 Inc.). Standard recombinant DNA techniques were used in accordance with the description given in the state of the art (61). In pJN3 and pJN11, the β lactamase (bla) gene has been interrupted. oriE and oriM designate the replication origins of pUC (

E. coli

) and of pAL5000 (mycobacteria), respectively.

FIG. 12

Structure of the plasmids of the pJEM series.

(A) In the schematic representation of the plasmids, only the relevant genetic elements are indicated. pJEM15 resulted from the cloning, into the ScaI site of pRR3, i) of a fragment obtained by PCR amplification (using OJN1: 5′-AAGCTTCCGATTCGTAGAGCC-3′(SEQ ID NO:10) and OJN2: 5′-GGGCTCGAGCTGCAG TGGATGACCTTTTGA-3′SEQ ID NO:11) as primers; and pJN11 as template) and containing tT4 and the N-terminal end of cII; ii) of the synthetic oligonucleotides corresponding to MCS1; and iii) the HindIII-DraI lacZ′ fragment of pNM480. pJEM12-13-14 were obtained by cloning the PCR-amplified fragment described above, into the ScaI site of pRR3. The synthetic oligonucleotides corresponding to MCS2 were then inserted. Finally, each of the three forms of the pNM480 series were introduced into the HindIII site in MCS2. (B) Nucleotide sequences of the regions between the OJN1 primer and the 8th lacZ′ codon (marked ****). These sequences were checked experimentally. The tT4 region is underlined and the synthetic RBS is in bold type. The amino acid sequence of the N-terminal end of cII is given under the DNA sequence. The HindIII sites are marked by an asterisk because they are not unique. For additional descriptions, see the legend in FIG.

11

.

EXAMPLES

I) Identification of Genes Encoding Exported

M. tuberculosis

Proteins

The results reported here describe the definition, for mycobacteria, of a genetic method of identification of exported proteins. This methodology is based on the translational fusion with bacterial alkaline phosphatase (PhoA). Such fusion proteins must be exported in order to have the PhoA activity (6, 13, 16). A PhoA gene was used after deletion of the promoter region, of the ribosome-binding site and of the entire region encoding the signal sequence whose codon for initiation of translation was used. Thus, the alkaline phosphatase activity is dependent on the translational fusion achieved in the correct reading frame with part of an exported protein. The construction of a phoA plasmid vector for mycobacteria is described first of all since it has been shown that the introduction, into this vector, of the gene for the exported

M. fortuitum

β-lactamase (blaF*) (34) leads to the production, in

M. smegmatis,

of fusion proteins having the PhoA enzymatic activity. A library of sequences for fusion between the

M. tuberculosis

genomic DNA and the phoA gene was then constructed. Twelve independent clones, which exported fusion proteins, were isolated. Among them, it was possible to identify the 19 kDa exported lipoprotein already described in

M. tuberculosis,

a new

M. tuberculosis

sequence exhibiting similarities with the 28 kDa

M. leprae

protein, a protein comprising conserved amino acid residues with stearoylacyl carrier protein (ACP) desaturases, and other new sequences.

Materials and Methods

Bacterial Strains, Plasmids, and Culture Conditions

The bacterial strains and the plasmids used in this study are presented in Table 1. The growth of

E. coli

and

M. smegmatis

strains, the electroporation, the screening on agar containing 20 μg/ml of kanamycin and 20 μg/ml of 5-bromo-4-chloro-3-indolyl phosphate (X-P) were performed as previously described (14)

M. tuberculosis,

an isolate from a patient (strain 103), was cultured on solid Lowënstein-Jensen medium.

Manipulation and Sequencing of DNA

Manipulation of DNA and Southern-blot analyses were carried out with the aid of standard techniques (27). For the determinations of the sequences, the oligonucleotides (5-GGCCCGACGAGTCCCGC-3′(SEQ ID NO:12) and 5′-TTGGGGACCCTAGAGGT-3′(SEQ ID NO:13)) were developed for sequencing across the fusion junctions of the

M. tuberculosis

inserts in pJEM11 (see below). The double-stranded plasmid DNA sequences were determined by the dideoxy chain termination method (28) using the T7 sequencing kit (Pharmacia) according to the manufacturer's instructions, or with the Taq Dyc Deoxy Cycle Terminator sequencing kit (Applied Biosystems), on a GeneAmp 9600 PCR system (Perkin Elmer), and passed over a DNA analysis system—Model 373 (Applied Biosystems).

Analyses of the Databanks

The nucleotide sequences were compared with those of the EMBL and GeneBank databanks using the FASTA algorithm (23) and the derived protein sequences were analyzed to determine a possible similarity with the sequences contained in the databanks for the PIR and SwissProt proteins using the BLAST algorithm (1).

Constructions of the Plasmids

pJEM11: The construction of pJEM11 is summarized in FIG.

1

. Briefly, pJEM2 was constructed using the shuttle plasmid pRR3 of

E. coli

-mycobacteria (26), by insertion of the truncated lacZ fragment of pNM480 (18), a multiple cloning site or polylinker (MCS), and the transcriptional terminator of the omega cassette (24). The N-terminal EcoRV-KpnI fragment of lacZ is replaced with the truncated phoA fragment of pPHO7 (11), without initiation codon or signal sequence to give pJEM10. Finally, a potential initiation codon in the MCS was eliminated in order to give pJEM11.

pLA71, pLA72 and pLA73: Fragments of blaF* (34) of different length, obtained by PCR amplification, were inserted at the Bam H1 site of pJEM11 to give pLA71, pLA72 and pLA73 (FIG.

2

). The oligonucleotides (Genset, Paris) used for the PCR amplification were, upstream, 5′-CGGGATCCTGCTCGGCGGACTCCGGG-3′(SEQ ID NO:14) and, downstream, 5′-CGGGATCCGGTCATCGATCGGTGCCGCGAA-3′(SEQ ID NO:15), 5′-CGGGATCCCGCCGTGCTCGGCCATCTGCAG-3′(SEQ ID NO:16) and 5′-CGGGATCCAGAGTAAGGACGGCAGCACCAG-3′(SEQ ID NO:17), for pLA71, pLA72 and pLA73 respectively. The PCR amplifications were carried out in a DNA Thermal Cycler (Perkin Elmer), using Taq polymerase (Cetus), according to the manufacturer's recommendations.

Construction of the

M. tuberculosis

Genomic Libraries

M. tuberculosis

genomic DNA was extracted according to standard procedures (27). This DNA was partially digested with Sau3A (with 1 U per 2 μg) at 37° C. for 2 min 30 sec. The digestion was stopped by the addition of phenol. This DNA was then run on low-melting point agarose (Gibco, BRL). The fraction containing the fragments having from 400 to 2,000 bp was extracted with agarase (GELase, Epicentre Technologies) and ligated into the compatible Bam HI site of pJEM11 with T4 DNA ligase (Boehringer Mannheim), at 16° C. overnight.

Assay of Alkaline Phosphatase

For the assays of alkaline phosphatase,

M. smegmatis

was cultured in L broth supplemented with 0.05% tylaxopol (Sigma) at 37° for 48 h. The alkaline phosphatase activity was assayed by the Brockman and Heppel method (8), in sonicated extracts as previously described (34), using p-nitrophenyl phosphate as substrate for the reaction. The protein contents were measured with the aid of the Bio-Rad assay (Bio-Rad). The alkaline phosphatase activity is expressed in arbitrary Units (aU)=OD

420

×105×1 g of protein

−1

×min

−1

.

Preparations of Antibodies, SDS-polyacrylamide Gel Electrophoresis and Immunoblottings

The preparation of a rabbit anti-PhoA serum has been previously described (34). Cellular extracts of

M. smegmatis

were prepared by sonication, SDS-PAGE and immunoblotting were performed as previously described (36).

Results

Construction of a Shuttle Plasmid Vector (pJEM11) for the Production of Fusion Proteins with PhoA in

M. smegmatis

pJEM11 has a truncated phoA gene of

E. coli

without initiation codon or any regulatory elements (FIG.

1

). The multiple cloning site allows the insertion of fragments derived from genes encoding putative exported proteins at the same time as their regulatory elements. Thus, fusion proteins were able to be produced, they expressed the activity of alkaline phosphatase when the fusion was exported. pJEM11 is an

E coli

/mycobacteria shuttle plasmid which includes the gene for resistance to the antibiotic kanamycin of tn903 as selectable marker.

Insertion of Genetic Elements Responsible for the Expression and Export of β-lactamase in pJEM11 Lead to the Production of PhoA Fusion Proteins Which are Enzymatically Active in

M. smegmatis

The three plasmids were constructed by insertion of fragments of different length derived from the β-lactamase gene of the overproducing strain

M. fortuitum

D316 (blaF*) (34) at the BamHI site of pJEM11 (FIG.

2

). In pLA71, the 1384 bp fragment includes the promoter, the segment encoding the 32 amino acids of the signal sequence, and the first 5 amino acids of the mature protein (there is no Shine-Dalgarno sequence for ribosomal attachment in the original sequence of blaF*). pLA72 carries a 1550 bp fragment including the elements encoding the signal sequence and the first 61 amino acids of the mature protein. In pLA73, the 2155 bp fragment contains the whole blaF*. These plasmids were used to transform

M. smegmatis

and the transformants were screened for the enzymatically active PhoA fusions by plating on agar media containing kanamycin and X-P. X-P is soluble and is colorless, but after cleavage of the phosphate with alkaline phosphatase, a blue precipitate is produced. Thus, alkaline phosphatase-producing clones could be easily identified by their blue color. The expression of pLA71, 72 and 73 in

M. smegmatis,

leads to blue colonies, whereas colonies with pJEM11 remained white. Western-blot analyses showed the production of phoA fusion proteins with an apparent molecular weight of about 47.5 kDa, 54 kDa and 76 kDa, for pLA71, pLA72 and pLA73 respectively (

FIG. 3

, column 3, 4, 5). These molecular weights are in agreement with the length of the mature protein fused with alkaline phosphatase (apparent MW of 46 kDa,

FIG. 3

, column 1). In pJEM11, there is no expression of PhoA, as expected (

FIG. 3

, column 2). The assay of the alkaline phosphatase activity (see

FIG. 2

) of these bacteria confirms the expression of an enzymatic activity with the 3 pLA constructs. However,

M. smegmatis

with pLA73 expresses an activity which is about twice as high compared with pLA73 and 72. In separate experiments, we have confirmed that the intracellular production of phoA under the control of a mycobacterial promoter, without fusion with an exported protein, was not associated with the expression of the alkaline phosphatase activity. All these results indicate that in this system, the activity of alkaline phosphatase depends on the translational fusion and the actual export of the product. Consequently, pJEM11 is suitable for the genetic identification of the proteins exported by mycobacteria.

Construction in

M. smegmatis

of a Bank of phoA Fusions with

M. tuberculosis

Genomic DNA Fragments

The genomic DNA of a clinical isolate of

M. tuberculosis

was purified and partially digested with Sau3A. The 400/2,000 bp fraction was inserted at the compatible BamHI site of pJEM11. The ligation products were transferred into

E. coli

XL-1 blue by electroporation to obtain an amplification stage. About 2,500 clones containing plasmids with inserts grew on an agar medium containing kanamycin. The plasmids purified from the transformants were combined and transferred by electroporation into

M. smegmatis

MC

2

155. The transformed bacteria were plated on L agar-kanamycin-X-P. About 14,000 clones were obtained. After incubating for 4 days, the first blue, and therefore PhoA

+

, colonies were observed. Each day, the dishes were checked, and new PhoA

+

colonies were isolated. The cloned colonies were lyzed, and their DNA introduced by electroporation into

E. coli

XL-1 blue, for the preparations of plasmids. In all, 12 different inserts allowing the expression of phoA were isolated and sequenced. Three sequences had similarities with known sequences.

Fusion of PhoA with the Gene for the 19 kDa

M. tuberculosis

Lipoprotein

One of the plasmids (pExp410) has an insert corresponding to part of the gene for the 19 kDa protein already known. This gene encodes an exported lipoprotein (5, 31).

FIG. 4A

shows the DNA sequence corresponding to the fusion between this gene and phoA. As expected, the same reading frame is maintained between the two proteins. The expected molecular weight of the fusion protein, according to the sequence, is thought to be close to 57 kDa. However, the true molecular weight observed by Western-blot analysis is identical to the purified PhoA protein (

FIG. 3

, column 1 and 6), which suggests that the fusion protein is cleaved near the PhoA junction.

Fusion with a Sequence Similar to the Gene for the 28 kDa

M. leprae

Protein

The 28 kDa

M. leprae

protein is a major antigen which is very often recognized by the sera from patients suffering from the lepromatous form of leprosy (9). In the

M. tuberculosis

insertion bank prepared, a sequence carried by a recombinant vector (pExp53), exhibiting 77% similarity with the nucleotide sequence of this gene and 68% for the deduced amino acid sequence (FIG.

4

B), was identified. In Western-blot analysis, the molecular weight of the fusion protein is about 52 kDa (

FIG. 3

, column 7), which provides for about 45 amino acids of the mycobacterial protein in the fusion protein, after cleavage of the signal peptide. This is in conformity with the length of the fragment of the

M. tuberculosis

gene fused with phoA (FIG.

4

B).

Southern-blot analyses of the

M. tuberculosis

genomic DNA were carried out. It was shown that a 180 bp fragment of the 2 kb insert of the plasmid pExp53 does not contain any restriction site for the endonucleases PstI, SmaI, BamHI, BstEII and SphI. This fragment was amplified by PCR. The

M. tuberculosis

genomic DNA was digested with the aid of these enzymes, and probed with the 32P-labeled PCR fragment. As can be seen in

FIG. 5

, only one band was observed when the genomic DNA was digested with each of the five enzymes, which suggests that the gene is present in only one copy in the

M. tuberculosis

genome.

Other PhoA Fusions Carrying the Putative Signal Sequences

FIG. 4C

shows the sequence of an insert carried by a recombinant vector (pExp59) fused with phoA. It has a typical signal sequence allowing the export of proteins. The sequence presented is in conformity with the usual rules as established in Gram-negative bacteria (25). It contains two positively charged amino acids (Arg, Asn) after the initiation codon, followed by a hydrophobic peptide, with a Gly, probably corresponding to a loop in the three-dimensional structure of the peptide. A potential site of cleavage by signal peptidase is indicated by an arrow, which gives a fusion protein with a molecular weight close to that of phoA, as shown in

FIG. 3

, column 8, conformably.

PhoA Fusion Proteins with Amino Acid Units Conserved with Stearoyl-acyl Carrier Protein (ACP) Desaturases

The ACP-desaturases are enzymes involved in the pathways for the biosynthesis of fatty acids. In particular, these enzymes are integral membrane proteins (29). Analyses of the plasmid pExp421 of the prepared bank showed two amino acid units conserved with ACP-desaturases, one of 9 amino acids and the second of 14 amino acids (FIG.

4

D). The rest of the sequence did not show any significant similarity with known proteins.

Discussion

More than 30 secreted proteins have been found in BCG or

M. tuberculosis

filtrates in the short term, with a minimum lysis of the bacterium (1, 19, 38). These proteins have been classified according to their molecular weight and their immunological reactivities. Some were characterized more extensively. For example, the secreted proteins of the complex of antigen 85 (antigens 85 A, B and C) are 32 kDa proteins exhibiting serological cross-reactions (7, 35). The antigens 85 A and 85 B exhibit an affinity toward fibronectin and might be involved in the internalization of

M. tuberculosis

in the macrophages. The genes for these immunogenic proteins (7), and for 23 kDa proteins (MPB64) (37) and for 19 kDa proteins (5) have been cloned and sequenced and sequences of signal peptides characteristic of exported proteins have been found. The recombinant proteins produced using these genes are thought to be valuable tools for the serological diagnosis of tuberculosis. Superoxide dismutase (SOD) of 23/28 kDa is abundant in short term culture filtrates, and are thought to be involved in the survival of mycobacteria in the phagolysosome. The gene encoding SOD in

M. tuberculosis

has been cloned and sequenced (39). Advantageously, no characteristic signal peptide sequence has been found. This suggests a specific route for secretion of this enzyme by mycobacteria. Secreted proteins in two narrow molecular weight ranges (6-10 kDa and 26-34 kDa) are major T cell antigens (3) and induce, in mice, T cell immune responses which are protective against a challenge with live mycobacteria of the

M. tuberculosis

complex (4). It has been suggested that the differences in the immune responses observed between live and killed bacteria are due to these exported/secreted proteins (20). These various preliminary results suggest that a better characterization of exported/secreted proteins of pathogenic bacteria of the

M. tuberculosis

complex might be highly useful both for understanding their pathogenicity and for developing new vaccines.

While secreted proteins have been studied by biochemical methods, other genetic methodologies might prove necessary. Using a truncated phoA gene, fusion systems have been developed which allow the attachment of the amino ends of other proteins onto PhoA. This approach is based on the

E. coli

periplasmic bacterial alkaline phosphatase. This enzyme must be located extracytoplasmically to be active. Thus, alkaline phosphatase may be used as subcellular localization probe.

A PhoA methodology has been developed and described here for the identification of proteins exported by mycobacteria. The insertion of blaF* into pJEM11 leads to the production, in

M. smegmatis,

of fusion proteins with alkaline phosphatase activity. Furthermore, PhoA fusions with 3 different fragments of BlaF* were enzymatically active, which suggests that most of the fusions in phase with exported proteins will have a PhoA activity.

A bank of

M. tuberculosis

inserts in pJEM11 has been constructed and expressed in

M. smegmatis.

In this bank, part of the gene encoding the known exported lipoprotein of 19 kDa (pExp410) has been isolated. This

M. tuberculosis

protein is one of the serologically immunodominant antigens found in this bacillus. Analyses of the DNA sequence of the gene encoding this antigen indicate that the hydrophobic NH2-terminal region is a lipoprotein signal peptide (5). Part of this lipoprotein has been fused with the outer surface A protein of

Borrelia burgdorferi

to construct a recombinant BCG vaccine capable of inducing a high immune response (31).

Two other sequences sharing similarities with the exported or membrane proteins have also been identified:

pExp53 was shown to exhibit similarities with the gene for the 28 kDa

M. leprae

antigen. This

M. leprae

antigen has been found by screening a λgt 11 library with serum from patients suffering from the lepromatous form of leprosy. It is a major antigen involved in the humoral immune response to

M. leprae

(9). Advantageously, it has been shown that a peptide of 20 amino acids of this protein exhibits considerable similarity with a peptide of the 19 kDa

M. tuberculosis

antigen, and it is an epitope of T cells exhibiting cross-reactions (12). The DNA sequence of the gene encoding the 28 kDa

M. leprae

antigen suggests that “the abovementioned amino acid sequence of the protein contains a potential signal peptide at its amino-terminal end and two long hydrophobic domains, which suggests that it is screened for localization on the bacterial plasma membrane or the cell wall” (9).

A fusion protein encoded by a plasmid of our bank (pExp421) is thought to share amino acid units with desaturases. The ACP-desaturases are enzymes involved in the pathways of the biosynthesis of fatty acids. In general, these enzymes are integral membrane proteins (39). This result suggests that it is possible to have isolated part of a gene which is important in the metabolism of lipids in

M. tuberculosis,

maybe involved in the lipid cell wall biosynthesis pathway.

Another plasmid (pExp59) with a characteristic putative signal sequence has been found.

In conclusion, the results presented demonstrate that the technology of PhoA for the genetic identification of exported proteins may be successfully adapted for

M. tuberculosis.

Preliminary screenings of an insert bank giving PhoA fusion proteins have revealed sequences exhibiting similarities with known exported proteins.

II) Expression of the P28

M. tuberculosis

Protein

BCG is a live vaccine. It is the only vaccine used to protect against tuberculosis. Its efficacy has proved variable according to the populations vaccinated, ranging from about 80% in Great Britain to 0% in India. It therefore seems essential to search for a more effective vaccine. Moreover, the use of a live vaccine currently poses problems because of the extension of the AIDS epidemic.

Several studies have shown that antigens exported by

Mycobacterium tuberculosis,

the agent for tuberculosis, had a protective effect against a challenge with the virulent strain. The studies reported here consisted in using a genetic method for isolating and studying the

M. tuberculosis

genes encoding exported proteins. We describe here the isolation and characterization of a gene encoding a protein having homologies with the 28 kDa

Mycobacterium leprae

protein already described.

Methodology for the Cloning of Genes Encoding Exported Proteins

The methodology presented in detail in part I is based on the use of translational fusions with the gene encoding the

Escherichia coli

alkaline phosphatase, PhoA. Such fusion proteins have a detectable alkaline phosphatase activity only if they are exported. A plasmid vector carrying a phoA gene lacking its promoter, its ribosomal RNA-binding site and its signal sequence was constructed. Using this vector, a PhoA activity can be observed only after translational fusion in the correct reading frame with an exported protein. The vector, called pJEM11 has a replication origin for

E. coli

and another for mycobacteria. It also has a selectable marker, the kanamycin-resistance gene of the transposon Tn905. A multiple cloning site precedes the truncated phoA gene.

A genomic DNA library obtained from an

M. tuberculosis

strain (Mt103) isolated from a tuberculosis patient was constructed in pJEM11 by inserting DNA fragments derived from a partial hydrolysis by the enzyme Sau3a. The clones selected made it possible to identify a nucleotide fragment of the 28 kDa

M. tuberculosis

gene homologous to the gene encoding the 28 kDa

M. leprae

protein.

In the lepromateous patients, antibodies directed against this 28 kDa protein are observed, suggesting that this protein is an immunodominant antigen. It was hypothesized that in

M. tuberculosis,

the 28 kDa protein possessing homologies with the 28 kDa

M. leprae

protein could also be an immunodominant antigen and that it could serve in the construction of specific immunological tests allowing the detection of the tuberculosis infection or of the tuberculosis disease. It could perhaps be used for the construction of subunit vaccines in different vaccine preparations. Furthermore, it could be useful as vector for the expression of antigens in mycobacteria for the construction of recombinant vaccines.

Cloning and Sequencing of the Gene Encoding a 28 kDa

M. tuberculosis

Protein

Using the insert contained in the plasmid pExp53 as probe, the whole gene encoding the 28 kDa

M. tuberculosis

protein was cloned by colony hybridization of an

M. tuberculosis

DNA library constructed by inserting

M. tuberculosis

DNA fragments of between 2 and 6 kb in size, obtained by total hydrolysis with the enzyme PstI into the vector pBluescript KS-. The

M. tuberculosis

PstI fragment corresponding to the positive clone and comprising a 4.1 kb insert was sequenced.

FIG. 10

shows the nucleotide sequence of the fragment and the similarities with the gene encoding the 28 kDa

M. leprae

protein. The sequence of the 28 kDa

M. tuberculosis

gene is, like that of

M. leprae,

preceded by a sequence possessing similarities with the “iron” boxes found upstream of the genes expressed during an iron deficiency. An iron deficiency situation is encountered during growth in vivo. It is hypothesized that the expression of this gene is induced during the growth, in the macrophages, of the mycobacteria harboring this gene. Furthermore, the 28 kDa

M. tuberculosis

protein possesses, in its central part, two regions containing units of 5 amino acids repeated in tandem, which are absent from the homologous

M. leprae

protein. Analogous repeated structures have been previously identified in major antigens present at the surface of other bacterial or parasitic pathogenic agents such as the M protein (40) of the Streptococcacea and the CS protein of the Plasmodiae (41).

All or part of the 28 kDa

M. tuberculosis

protein, whose gene sequence is presented here, could be a potential protective antigen for the construction of a tuberculosis vaccine. Such an antigen may be obtained by purification from cellular extracts of

M. tuberculosis

or from cellular extracts of genetically recombined heterologous organisms. Furthermore, the 28 kDa

M. tuberculosis

protein, or peptides derived therefrom, could be an antigen capable of being used in ELISA tests for screening tuberculosis patients.

By using the 28 kDa

M. tuberculosis

gene as probe, hybridization under conditions of high stringency was observed only with the genomic DNA of strains belonging to the

M. tuberculosis

complex. Consequently, the sequence corresponding to the 28 kDa

M. tuberculosis

gene is a specific sequence which may be used for tests for detection of the tuberculosis baccilli, using DNA or RNA probes and in vitro methods of gene amplification.

The regulatory region and the 28 kDa

M. tuberculosis

gene may be used as carrier molecules to express heterologous antigens in BCG or any other mycobacterial vector useful for the construction of vaccines.

III) Expression of Mycobacteria Genes; Evaluation of Different Expression Promoters

An important aspect of the results obtained relates to the construction of genetic tools for studying the expression of genes in mycobacteria. Regulatory sequence-probe vectors have been used in the prior art to isolate and analyze regulatory sequences in a large number of bacteria (54). The definition, by the inventors, of such tools specific to mycobacteria facilitates the study of the genetic mechanisms regulating virulence in the pathogenic species, and the isolation of new regulatory sequences which might be useful for developing improved recombinant BCG vaccines.

Initially, the expression of mycobacterial genes was studied in heterologous systems,

Escherichia coli

and

Streptomyces lividans

(46) (51) (60). These analyses suggest that most of the mycobacterial genes are more efficiently expressed in

S. lividans

than in

E. coli.

Subsequently, vectors based on mycobacterial plasmids were constructed which might be used for studies in homologous systems. The vectors pYUB75 and pYUB76 were designed to select gene fusions with a truncated

Escherichia coli

lacZ gene (42). The plasmid pSD7 allows the construction of fusions of operons with a gene for chloramphenicol acetyltransferase (CAT) without promoter (47). By using these vectors a number of mycobacterial regulatory sequences were isolated and evaluated both in

E. coli

and in

Mycobacterium smegmatis.

The inventors have described other constructions of vectors of the pJEM series, which have several advantages: they carry a transcription terminator, suitable multiple cloning sites, and they allow fusions both of operons and of genes with lacZ. lacZ was chosen as reporter gene because the enzyme encoded, β-galactosidase, remains active when heterologous sequences are fused with its amino-terminal end (45) (64). Its activity may be easily measured in vitro, even at very low levels with the aid of fluorescent compounds (48). β-Galactosidase is also highly immunogenic. It induces both humoral and cellular immune responses after presentation to the mouse immune system by recombinant bacteria (44) (56). Thus, β-galactosidase may also be used as reporter of the immunogenicity of a recombinant vaccine. By using pJEM vectors, new regulatory sequences active in BCG could be isolated and the recombinant BCG strains easily tested for their capacity to induce immune responses in mice.

A comparative study of the activities of various promoters in

M. smegmatis

and BCG was also made. The results suggest that the RNA polymerases of

M. smegmatis

and of BCG do not share the same specificity.

The construction of pJEM vectors. Ideally, a plasmid vector promoter-probe should contain five elements:

i) a replicon, ii) a selectable marker, and a reporter cassette containing iii) a transcription terminator followed iv) by multiple cloning sites (MCS) and v) a reporter gene lacking its regulatory sequences.

To construct a promoter cloning vector, mycobacteria, the replicon derived from the plasmid pAL5000 of

Mycobacterium fortuitum,

and the kanamycin resistance gene (aph) of Tn903 (58) were used. These genetic elements are basic components of most plasmids currently used for the transformation of mycobacteria. They appear to confer high stability on transformed clones of

M. smegmatis

and

M. bovis

BCG both in vitro and in vivo (in mice) even in the absence of selection by antibiotics (56). To facilitate the preparation and manipulation of episomal DNA, most of these plasmids also contain an

E. coli

replicon. Thus, we chose the plasmid pRR3, an

E. coli

-mycobacteria shuttle vector which contains these three genetic elements as basic vector (58).

No mycobacterial transcription terminator has yet been characterized. To examine if the T4 coliphage transcription terminator (tT4) was active as termination site for the mycobacteria RNA polymerases, the mega interposon (57) was cloned into the plasmid pJN3, upstream of the sRBS-cII-lacZ element, generating pJN11 (FIG.

11

). The omega fragment is composed of a streptomycin/spectinomycin resistance gene flanked by short inverted repeats containing tT4. The insertion of omega into a DNA fragment leads to termination of the synthesis of RNA in

E. coli

(57). pJN3 was constructed by cloning, into the ScaI site of pRR3, a cassette composed of a truncated lacZ combined with a synthetic RBS (sRBS) and the 5′ end of the lambda phage cII regulatory gene and the pL promoter (FIG.

11

).

M. smegmatis

mc2155 (61) was transformed with pJN3 (pL-sRBS-cII-lacZ) or pJN11 (pL-X-sRBS-cII-lacZ) by electroporation and the transformant clones were identified after growth on LB-Xgal plates. The transformant clones carrying pJN3 gave blue colonies and the transformant clones carrying pJN11 gave white colonies. The β-galactosidase activity in

M. smegmatis

(pJN11) was 50 times as low as that in

M. smegmatis

(pJN3) (Table 2). Thus, tT4 contained in the insert X acts as an efficient transcription terminator in

M. smegmatis.

A DNA fragment containing the tT4 segment followed by the sRBS-cII-lacZ element of pJN11 was synthetized in vitro by amplification by PCR and an MCS (MCS1), containing 6 unique restriction sites, was added. The resulting cassette was then cloned into the ScaI site of pRR3, giving the operon fusion vector pJEM15 (FIG.

12

). The electroporation of

M. smegmatis

MC

2

155 and of BCG with this plasmid led to white colonies on LB-Xgal plates with a very weak β-galactosidase activity (Table 2). On the other hand, in

E. coli,

pJEM15 expressed a higher β-galactosidase activity, and consequently a blue color on LB-Xgal plates. This is probably due to its high copy number. In

E. coli,

pUC vectors are present at a high copy number (greater than 500), whereas in mycobacteria, the replicon-derived plasmids pAL5000 have a copy number of approximately 3 to 10 (50). The testing of DNA fragments for promoter activity, with the aid of pJEM15, by blue-white screening, should thus be carried out directly in mycobacteria.

To obtain vectors allowing fusions of genes with lacZ, we followed a similar strategy. The three forms of truncated lacZ of the pNM480 series (55), which differ from each other in the “placing in translational phase” of a HindIII site located at its 5′ end, were cloned, downstream of tT4 and of an MCS (MCS2) containing 7 unique restriction sites, into the ScaI site of pRR3. The resulting plasmids pJEM12-13-14 (

FIG. 12

) thus allow the cloning of a wide range of restriction fragments in phase with lacZ.

Evaluation of various promoters in

M. smegmatis

and BCG. Operon fusions between the cII-lacZ reporter cassette of pJEM15 and the promoters pAN (56), pblaF* (63), psul3 (52) and pgroES/EL1 (49) were constructed. The activity of these promoters was evaluated in

M. smegmatis

and in

M. bovis

BCG. The first three promoters were isolated from mycobacterial species: pblaF* is a high expression mutant of pblaF, which directs the expression of the

M. fortuitum

β-lactamase gene; pAN is an

M. paratuberculosis

promoter and psul3 a component of a mobile genetic element of

M. fortuitum

Tn610. These promoters were localized on the basis of the mapping of sites of initiation of transcription (pblaF* and pAN) or by deletion analysis (psul3) (62). pgroES/EL1 is a

Streptomyces albus

promoter which regulates the expression of the groES/EL1 operon, and is active both in

M. smegmatis

and BCG (65).

The cloning experiments were carried out directly in

M. smegmatis.

DNA fragments containing each of the promoters were isolated and inserted at MCS1 of pJEM15 disgested with the appropriate restriction enzymes. The resulting ligation mixtures were used to transform

M. smegmatis

mc2155 by electroporation and blue colonies were selected in order to electroduce

E. coli

MC1061 (45) as described above (43). The plasmids were isolated from these

E. coli

clones and analyzed. Those corresponding to the desired constructs pJN29 to pJN32 (table 2) were used for the electroporation of BCG (Pasteur strain).

The β-galactosidase activity was assayed on sonicated extracts of

M. smegmatis

and of BCG (table 2). The activity of the promoters varied considerably both between the promoters in a mycobacterial host and between the hosts for each promoter. The relative strength of these promoters was not the same in

M. smegmatis

and BCG. Although pblaF* was the most powerful promoter both in

M. smegmatis

and in BCG, the situation is different for the other promoters: pAN and pgroES/EL1 were more active than psul3 in BCG, but in

M. smeamatis,

psul3 was more active than pAN or pgroES/El1.

Das Gupta and his colleagues (47) screened

M. smegmatis

and

M. tuberculosis

DNA libraries for the promoter activity in

M. smegmatis.

They reported a promoter frequency 10 to 20 times higher in the

M. smegmatis

DNA. Furthermore, very active promoters were more rare in the

M. tuberculosis

DNA libraries than in those of

M. smegmatis.

These authors suggested that the

M. tuberculosis

promoters may have diverged considerably from those of

M. smegmatis.

The results presented here suggest that the transcriptional machinery of

M. smegmatis

and of

M. bovis

BCG, a species closely related to

M. tuberculosis,

may be different.

In conclusion, the family of vectors constructed facilitates the study of the expression of genes in mycobacteria. A wide range of fragments may be easily cloned in phase with lacZ′ (fusion of genes) or upstream of cII-lacZ (fusion of operons) and evaluated for the promoter activity by blue-white screening of mycobacterial transformants on LB-Xgal plates. The activity of these promoters may also be measured (by assaying the β-galactosidase activity), their sequences determined, and their site for initiation of transcription mapped (by primer extension analysis) using the “universal primer” or related sequences (53) as primer.

IV) Expression of the ERP Protein in Recombinant form in

E. coli

The ERP protein was expressed in recombinant form in

E. coli

and purified by affinity chromatography. Two types of fusions between ERP and peptide fragments having a high affinity for specific chromatographic supports (Amylose, MalE system; chelated Nickel (Ni

2+

) , for the Histidine system) were carried out. They are:

ERP lacking its signal sequence fused at the C-ter with the maltose-binding protein (MalE) of

E. coli

(MalE-ERP);

ERP lacking its signal sequence (ERP(His)

6

ss) or in its entirety (ERP(His)

6

), and possessing 6 C-ter Histidine amino acids.

After purification, analysis of these three fusion proteins by SDS-PAGE electrophoresis indicates that the ERP polypeptide possesses a relative molecular weight (MW) of 36 kDa. There is a major difference between the MW calculated from the sequence (28 kDa) and the MW observed experimentally (36 kDa). This delay in the electrophoretic migration could be due to the high content of Proline residues, or from post translational modifications.

REFERENCES

1. Altschul, S. F. et al., 1990, J. Mol. Biol., 215: 403-410.

2. Andersen, P. et al., 1991, Infect. Immun. 59: 1905-1910.

3. Andersen, P. et al., 1991, Infect. Immun. 59: 1558-1563.

4. Andersen, P. et al., 1994, Immun. 62: 2536-2544.

5. Ashbridge, K. R. et al., 1989, Nucl. Acid. Res. 17: 1249.

6. Boquet, P. et al., 1987, J. Bacteriol. 169: 1663-1669.

7. Borremans, M. et al., 1989, Infect. Immun. 57: 3123-3130.

8. Brockman, R. W. et al., 1968, Biochemistry 7: 2554-2561.

9. Cherayil, B. et al., 1988, J. Immunol. 12: 4370-4375.

10. Gaillard, J.-L. et al., 1991, Cell 65: 1127-1141.

11. Gutierrez, C. et al., 1989, Nucl. Acids. Res. 17: 3999.

12. Harris, D. P. et al., 1991, J. Immunol. 147: 2706-2712.

13. Hoffman, C. S. et al., 1985, Proc. Natl. Acad. Sci. USA 82: 5107-5111.

14. Isberg, R. R. et al., 1987, Cell 50: 769-778.

15. Knapp, S. et al., 1988, J. Bact. 170: 5059-5066.

16. Manoil, C. et al., 1990, J. Bacteriol. 172: 515-518.

17. Miller, V. L. et al., 1987, cell. 48: 271-279.

18. Minton, N. P., 1984, Gene. 31: 269-273.

19. Nagal, S. et al., 1991, Infect. Immun. 59: 372-382.

20. Orme, I. M., 1988, Infect. Immun. 56: 3310-3312.

21. Orme, I. M. et al., 1993, J. Infect. Disea. 167: 1481-1497.

22. Pearce, B. J. et al., 1993, Mol. Microbiol. 9: 1037-1050.

23. Pearson, W. R. et al., 1988, Proc. Natl. Acad. Sci. USA. 85: 2444-2448.

24. Prentki, P. et al., 1984, Gene. 29: 303-313.

25. Pugsley, A. P., 1993, Microbiol. Rev. 57: 50-108.

26. Ranes, L. G. et al., 1990, J. Bacteriol. 172: 2793-2797.

27. Sambrook, J. et al., 1989. Molecular Cloning: a Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

28. Sanger, F. et al., 1977, Proc. Natl. Acad. Sci. USA 74: 5463-5467.

29. Shanklin, J. et al., 1991, Proc. Natl. Acad. Sci. USA 88: 2510-2514.

30. Snapper, S. B. et al., 1990, Mol. Microbiol. 11: 1911-1919.

31. Stover, K. C. et al., 1993, J. Exp. Med. 178: 197-209.

32. Taylor, R. K. et al., 1987, Proc. Natl. Acad. Sci. USA 84: 2833-2837.

33. Taylor, R. K. et al., 1989, J. Bact. 171: 1870-1878.

34. Timm, J. et al., 1994, Mol. Microbiol. 12: 491-504.

35. Wiker, H. G. et al., 1992, Microbiol. Rev. 56: 648-661.

36. Winter, N. et al., 1991, Gene. 109: 47-54.

37. Yamaguchi, R. et al., 1989, Infect. Immun. 57: 283-288.

38. Young, D. B. et al., 1992, Mol. Microbiol. 6: 133-145.

39. Zhang, Y. et al., 1991, Mol. Microbiol. 5: 381-391.

40. Hollingstead S. et al., 1986, J. Biol. Chem. 262: 1677-1686.

41. Zavala, F. et al., J. Exp. Med. 157: 194-1957.

42. Barletta, R. G. et al., 1992, J. Gen. Microbiol. 138: 23-30.

43. Baulard, A. et al., 1992, Nucleic Acids Res. 20: 4105.

44. Brown, A. et al., 1987, J. Infect. Dis. 155: 86-92.

45. Casabadan, M. J. et al., 1980, J. Bacteriol. 143: 971-980.

46. Clark-Curtiss, J. E. et al., 1985, J. Bacteriol. 161: 1093-1102.

47. Das Gupta, S. K. et al., 1993, J. Bacteriol. 175: 5186-5192.

48. Garcia-del-Portillo, F. et al., 1992, Mol. Microbiol. 6: 3289-3297.

49. Guglielmi, G. et al., 1993, Basic and Applied Genetics. Americain Society for Microbiology, Washington, D.C.

50. Hatfull, G. H. et al., 1993. Genetic transformation of mycobacteria. TIM 1: 310-314.

51. Kieser, T. et al., 1986, J. Bacteriol. 168: 72-80.

52. Martin, C. et al., 1990, Nature 345: 739-743.

53. Messing, J., 1983, New M13 vectors for cloning, p.20-78. In R. Wu, L. Grossman and K. Moldave (eds.), Methods in Enzymology. Academic Press, New York.

54. Miller, J. H., 1991, Bacterial Genetic Systems, In J. N. Abelson and M. I. Simon (eds.), Methods in Enzymology, Academic Press, San Diego.

55. Minton, N. P., 1984, Gene 31: 269-273.

56. Murray, A. et al., 1992, Mol. Microbiol. 6: 3331-3342.

57. Prentki, P. et al., 1984, Gene 29: 303-313.

58. Ranes, M. G. et al., 1990, J. Bacteriol. 172: 2793-2797.

59. Sambrook, J. et al., 1989, Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

60. Sirakova, T. D. et al., 1989, FEMS Microbiol. Lett. 59: 153-156.

61. Snapper, S. B. et al., 19[illegible]0, Mol. Microbial. 4: 1911-1919.

62. Timm, J. et al. Unpublished data.

TABLE 1

Strain/Plasmid

Relevant characteristics

Reference

E. coli

supE44 hsdR17 recA1 gyrA46 thi relA1

27

XL1-Blue

lac

−

F′

M. smegmatis

High-transformant mutant of

M. smegmatis

30

mc

2

155

ATCC607

pRR3

E. coli

-mycobacteria shuttle vector

26

pPHO7

pUC derivative carrying a truncated phoA

11

gene

pNM480

pUC derivative carrying a truncated lacZ

18

gene

pJEM11

E. coli

-mycobacteria shuttle vector carrying

this work

a truncated phoA gene

pLA71

pJEM11 in which has been cloned a

34,

1,364 bp fragment from blaF*

this work

pLA72

pJEM11 in which has been cloned a

34,

1,550 bp fragment from blaF*

this work

pLA73

pJEM11 in which has been cloned the

34,

complete blaF*

this work

pExp410

pJEM11 in which has been cloned part of the

this work

M. tuberculosis

19 kDa antigen gene

pExp53

pJEM11 in which has been cloned part of a

this work

M. tuberculosis

gene similar to the

M. leprae

28 kDa antigen gene

pExp59

pJEM11 in which has been cloned the signal

this work

sequence of a

M. tuberculosis

unidentified gene

pExp421

pJEM11 in which has been cloned a

this work

M. tuberculosis

gene encoding a protein with

amino acids motives similar to desaturases

Mycobacteria functional screening and/or expression vectors

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

Priority Claims (1)

PCT Information

Non-Patent Literature Citations (9)

Entry
Das Gupta, S. et al., “Cloning and Assessment of Mycobacterial Promoters By Using a Plasmid Shuttle Vector,” J. Bacteriology, vol. 175, No. 16, pp. 5186-5192 (1993).
Timm, J. et al., “Transcription and Expression Analysis, Using IacZ and phoA Gene Fusions, of Mycobacterium Fortuitum beta-lactamase Genes Cloned From a Natural Isolate . . . ,” Molecular Microbiology, vol. 12, No. 3, pp. 491-504 (1994).
Nagai, S. et al., “Isolation and Partial Characterization of Major Protein Antigens In the Culture Fluid of Mycobacterium Tuberculosis,” Infection and Immunity, vol. 59, No. 1, pp. 372-382 (1991).
Stover, K. et al., “Protective Immunity Elicited by Recombinant Bacille Calmette-Guerin (BCG) Expressing . . . ,” J. Experimental Medicine, vol. 178, pp. 197-209 (1993).
Boquet, P. et al., “Use of TnphoA to Detect Genes to Exported Proteins in Escherichia Coli: Identification . . .,” J. Bacteriology, vol. 169, pp. 1663-1669 (1987).
Andersen, P. et al., “Identification of Immunodominant Antigens During Infection with Mycobacteria Tuberculosis,” Scandinavian J. of Immunology, vol. 36, pp. 823-831 (1992).
Timm, J. et al., “Escherichia Coli-Mycobacteria Shuttle Vectors for Operon and Gene Fusions to lacZ: the pJEM Series,” J. Bacteriology, vol. 176, pp. 6749-6753 (1994).
Bigi, F. et al., “Characterization of a Novel Mycobacterium Bovis Secreted Antigen Containing PGLTS Repeats,” Infection and Immunity, vol. 63, pp. 2581-2586 (1995).
Lim, E. et al., “Identification of Mycobacterium Tuberculosis DNA Sequences Encoding Exported Proteins . . .,” vol. 177, pp. 59-65 (1994).