CA2319380A1

CA2319380A1 - Nucleic acid fragments and polypeptide fragments derived from m. tuberculosis

Info

Publication number: CA2319380A1
Application number: CA002319380A
Authority: CA
Inventors: Peter Andersen; Rikke Skjot
Original assignee: Individual
Current assignee: Statens Serum Institut SSI
Priority date: 1997-11-10
Filing date: 1998-10-08
Publication date: 1999-05-20
Also published as: EP1029053A1; AU750173B2; AU9433898A; NZ504951A; WO1999024577A1

Abstract

The present invention is based on the identification and characterization of a number of M. tuberculosis derived novel proteins and protein fragments (SEQ ID NOs: 175, 177, 179, 181, 183, and 185). The invention is directed to the polypeptides and immunologically active fragments thereof, the genes encoding them, immunological compositions such as vaccines and skin test reagents containing the polypeptides. Another part of the invention is based on the surprising discovery that CFP7A induces a high protective immune response.

Description

NUCLEIC ACID FRAGMENTS AND POLYPEPTIDE FRAGMENTS DERIVED FROM
M. TUBERCULOSIS
FIELD OF THE INVENTION
The present invention relates to a number of immunologically active,-novel polypeptide fragments derived from the Mycobacterium tuberculosis, vaccines and other immuno-logic compositions containing the fragments as immunogenic components, and methods of production and use of the polypeptides. The invention also relates to novel nucleic acid fragments derived from M. tuberculosis which are useful in the prepara-tion of the polypeptide fragments of the invention or in the diagnosis of infection with M. tuberculosis .
BACKGROUND OF THE INVENTION
Human tuberculosis (hereinafter designated "TB") caused by Mycobacterium tubercu-losis is a severe global health problem responsible for approximately 3 million deaths annually, according to the WHO. The worldwide incidence of new TB cases has been progressively falling for the last decade but the recent years has markedly changed this trend due to the advent of AIDS and the appearance of multidrug resistant strains of M. tuberculosis.
The only vaccine presently available for clinical use is BCG, a vaccine which efficacy remains a matter of controversy. BCG generally induces a high level of acquired resis-tance in animal models of TB, but several human trials in developing countries have failed to demonstrate significant protection. Notably, BCG is not approved by the FDA
for use in the United States.
This makes the development of a new and improved vaccine against TB an urgent matter which has been given a very high priority by the WHO. Many attempts to de-fine protective mycobacterial substances have been made, and from 1950 to 7 several investigators reported an increased resistance after experimental vaccination.
However, the demonstration of a specific long-term protective immune response with the potency of BCG has not yet been achieved by administration of soluble proteins or cell wall fragments, although progress is currently being made by relying on polypep-tides derived from short term-culture filtrate, cf. the discussion below.
Immunity to M. tuberculosis is characterized by three basic features; i) Living bacilli ' efficiently induces a protective immune response in contrast to killed preparations; ii) Specifically sensitized T lymphocytes mediate this protection; iii) ThE most important mediator molecule seems to be interferon gamma (INF-y).
Short term-culture filtrate (ST-CF) is a complex mixture of proteins released from M.
tuberculosis during the first few days of growth in a liquid medium (Andersen et al., 1991 ). Culture filtrates has been suggested to hold protective antigens recognized by the host in the first phase of TB infection (Andersen et al. 1991, Orme et al.
1993).
Recent data from several laboratories have demonstrated that experimental subunit vaccines based on culture filtrate antigens can provide high levels of acquired resis-tance to TB (Pal and Horwitz, 1992; Roberts et al., 1995; Andersen, 1994;
Lindblad et al., 1997). Culture filtrates are, however, complex protein mixtures and until now very limited information has been available on the molecules responsible for this pro-tective immune response. In this regard, only two culture filtrate antigens have been described as involved in protective immunity, the low mass antigen ESAT-6 (Andersen et al., 1995 and EP-A-0 706 571 ) and the 31 kDa molecule Ag85B fEP-0 432 203).
There is therefore a need for the identification of further antigens involved in the in-duction of protective immunity against TB in order to eventually produce an effective subunit vaccine.
OBJECT OF THE INVENTION
It is an object of the invention to provide novel antigens which are effective as com-ponents in a subunit vaccine against TB or which are useful as components in diag-nostic compositions for the detection of infection with mycobacteria, especially viru-lence-associated mycobacteria. The novel antigens may also be important drug tar-gets.

SUMMARY OF THE INVENTION
The present invention is i.a. based on the identification and characterization of a num-ber of previously uncharacterized culture filtrate antigens from M.
tuberculosis. In animal models of TB, T cells mediating immunity are focused predominantly to anti-gens in the regions 6-12 and 17-30 kDa of ST-CF. In the present invention 6 antigens in the low molecular weight region (ORF7-1, ORF7-2, ORF11-1, ORF11-2, ORF11-3, ORF1 1-4? have been identified.
Furthermore immunological and biological data on several important antigens are presented.
The encoding genes for 8 antigens have been determined. The panel hold antigens with potential for vaccine purposes as well as for diagnostic purposes, since the antigens are all secreted by metabolizing mycobacteria.
The following table lists the antigens of the invention by the names used herein as well as by reference to relevant SEQ ID NOs of N-terminal sequences, full amino acid sequences and sequences of DNA encoding the antigens:

Antigen N-terminal sequenceNucleotide sequenceAmino acid sequence SEQ ID NO: SEQ ID NO: SEQ ID NO:

CFP7B 168 i46 147 CFP8B 74 150 151 ' CFP25 21 i3 14 CFP50 86 61 . 62 RDi-ORF9B 69 70 It is well-known in the art that T-cell epitopes are responsible for the elicitation of the acquired immunity against TB, whereas B-cell epitopes are without any significant in-fluence on acquired immunity and recognition of mycobacteria in vivo. Since such T-5 cell epitopes are linear and are known to have a minimum length of 6 amino acid resi-dues, the present invention is especially concerned with the identification and utifisa-Lion of such T-cell epitopes.
Hence, in its broadest aspect the invention relates to a substantially pure polypeptide fragment which a) comprises an amino acid sequence selected from the sequences shown in SEQ
ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, and 185 b) comprises a subsequence of the polypeptide fragment defined in a) which has a length of at least 6 amino acid residues, said subsequence being immunologically equivalent to the polypeptide defined in a) with respect to the ability of evoking a pro-tective immune response against infections with mycobacteria belonging to the tuber-culosis complex or with respect to the ability of eliciting a diagnostically significant immune response indicating previous or ongoing sensitization with antigens derived from mycobacteria belonging to the tuberculosis complex, or c) comprises an amino acid sequence having a sequence identity with the poly-peptide defined in a) or the subsequence defined in b) of at least 70% and at the same time being immunologically equivalent to the polypeptide defined in a) with respect to the ability of evoking a protective immune response against infections with mycobacteria belonging to the tuberculosis complex or with respect to the ability of eliciting a diagnostically significant immune response indicating previous or ongoing ' sensitization with antigens derived from mycobacteria belonging to the tuberculosis complex, with the proviso that i) the polypeptide fragment is in essentially pure form when consisting of the amino acid sequence 1-96 of SEQ ID NO: 2 or when consisting of the amino acid sequence 87-108 of SEQ ID NO: 4 fused to ~3-galactosidase, ii) the degree of sequence identity in c) is at least 95% when the polypeptide com-prises a homologue of a polypeptide which has the amino acid sequence SEQ ID
NO:
12 or a subsequence thereof as defined in b), and iii) the polypeptide fragment contains a threonine residue corresponding to position 213 in SEQ ID NO: 42 when comprising an amino acid sequence of at least 6 amino acids in SEQ ID NO: 42.
Other parts of the invention pertains to the DNA fragments encoding a polypeptide with the above definition as well as to DNA fragments useful for determining the pre sence of DNA encoding such polypeptides.
DETAILED DISCLOSURE OF THE INVENTION
In the present specification and claims, the term "polypeptide fragment"
denotes both short peptides with a length of at least two amino acid residues and at most 10 amino acid residues, oligopeptides (1 1-100 amino acid residues), and longer peptides (the usual interpretation of "polypeptide", i.e. more than 100 amino acid residues in length) as well as proteins (the functional entity comprising at least one peptide, oiigopeptide, or polypeptide which may be chemically modified by being glycosylated, by being lipi-dated, or by comprising prosthetic groups). The definition of polypeptides also com-prises native forms of peptides/proteins in mycobacteria as well as recombinant pro-teins or peptides in any type of expression vectors transforming any kind of host, and also chemically synthesized peptides.
In the present context the term "substantially pure polypeptide fragment"
means a polypeptide preparation which contains at most 5% by weight of other polypeptide material with which it is natively associated (lower percentages of other polypeptide material are preferred, e.g. at most 4%, at most 3%, at most 2%, at most 1 %, and at most %z %). It is preferred that the substantially pure polypeptide is at least 96% pure, i.e. that the polypeptide constitutes at least 96% by weight of total polypeptide mate-rial present in the preparation, and higher percentages are preferred, such as at least 97%, at least 98%, at least 99%, at least 99,25%, at least 99,5%, and at least 99,75%. It is especially preferred that the polypeptide fragment is in "essentially pure form", i.e. that the polypeptide fragment is essentially free of any other antigen with which it is natively associated, i.e. free of any other antigen from bacteria belonging to the tuberculosis complex. This can be accomplished by preparing the poiypeptide fragment by means of recombinant methods in a non-mycobacterial host cell as will be described in detail below, or by synthesizing the polypeptide fragment by the well-known methods of solid or liquid phase peptide synthesis, e.g. by the method de-scribed by Merrifield or variations thereof.
The term "subsequence" when used in connection with a polypeptide of the invention having a SEQ ID NO selected from 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, and 185 denotes any continuous stretch of at least 6 amino acid residues taken from the M. tuberculosis derived polypeptides in SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, and 185 and being immunological equivalent thereto with respect to the ability of conferring increased resistance to infections with bacteria be-longing to the tuberculosis complex. Thus, included is also a polypeptide from diffe-rent sources, such as other bacteria or even from eukaryotic cells.
When referring to an "irnmunologically equivalent" polypeptide is herein meant that the polypeptide, when formulated in a vaccine or a diagnostic agent (i.e.
together with a pharmaceutically acceptable carrier or vehicle and optionally an adjuvant), will 1) confer, upon administration /either atone or as an immunologically active con-stituent together with other antigens), an acquired increased specific resistance in a mouse and/or in a guinea pig and/or in a primate such as a human being against infec-tions with bacteria belonging to the tuberculosis complex which is at least 20% of the acquired increased resistance conferred by Mycobacterium bouts BCG and also at least 20% of the acquired increased resistance conferred by the parent polypeptide com-prising SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, or 185 (said parent polypeptide having substantially the same relative location and pattern in a 2DE
gel prepared as the 2DE gel shown in Fig. 6, cf. the examples), the acquired increased resistance being assessed by the observed reduction in mycobacterial counts from spleen, lung or other organ homogenates isolated from the mouse or guinea pig re-ceiving a challenge infection with a virulent strain of M. tuberculosis, or, in a primate such as a human being, being assessed by determining the protection against develop-ment of clinical tuberculosis in a vaccinated group versus that observed in a control group receiving a placebo or BCG (preferably the increased resistance is higher and corresponds to at least 50% of the protective immune response elicited by M.
bouts BCG, such as at least 60%, or even more preferred to at least 80% of the protective immune response elicited by M. bouts BCG, such as at least 90%; in some cases it is expected that the increased resistance will supersede that conferred by M.
bouts BCG, and hence it is preferred that the resistance will be at least 100%, such as at least 1 10% of said increased resistance); and/or II) elicit a diagnostically significant immune response in a mammal indicating pre-vious or ongoing sensitization with antigens derived from mycobacteria belonging to the tuberculosis complex; this diagnostically significant immune response can be in the form of a delayed type hypersensitivity reaction which can e.g. be determined by a skin test, or can be in the form of IFN-y release determined e.g. by an IFN-y assay as described in detail below. A diagnostically significant response in a skin test setup will be a reaction which gives rise to a skin reaction which is at least 5 mm in diameter and which is at least 65% (preferably at least 75% such as at the least 85%) of the skin reaction (assessed as the skin reaction diameter) elicited by the parent polypep- -tide comprising SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, or 185.

The ability of the polypeptide fragment to confer increased immunity may thus be as-sessed by measuring in an experimental animal, e.g. a mouse or a guinea pig, the re-duction in mycobacterial counts from the spleen, lung or other organ homogenates isolated from the experimental animal which have received a challenge infection with a virulent strain of mycobacteria belonging to the tuberculosis complex after previously having been immunized with the polypeptide, as compared to the mycobacterial counts in a control group of experimental animals infected with the same virulent strain, which experimental animals have not previously been immunized against tuber-culosis. The comparison of the mycobacterial counts may also be carried out with my-cobacterial counts from a group of experimental animals receiving a challenge infec-tion with the same virulent strain after having been immunized with Mycobacterium bovis BCG.
The mycobacterial counts in homogenates from the experimental animals immunized with a polypeptide fragment according to the present invention must at the most be 5 times the counts in the mice or guinea pigs immunized with Mycobacterium bovis BCG, such as at the most 3 times the counts, and preferably at the most 2 times the counts.
A more relevant assessment of the ability of the polypeptide fragment of the invention to confer increased resistance is to compare the incidence of clinical tuberculosis in two groups of individuals (e.g. humans or other primates) where one group receives a vaccine as described herein which contains an antigen of the invention and the other group receives either a placebo or an other known TB vaccine le.g. BCG). In such a setup, the antigen of the invention should give rise to a protective immunity which is significantly higher than the one provided by the administration of the placebo (as de-termined by statistical methods known to the skilled artisan).
In the context of the present application, the term "wide genetically" should be under-stood in a meaning of at least two strains. That is, if a polypeptide is recognised by at least two different strains, it is considered to have a wide genetically recognition.
A subunit vaccine component is defined as a reagent which stimulates protective im-munity in an animal model of infection with an organism of the M. tuberculosis com-plex, when given prior to infection and which also generates a significant immune re-sponses in human volunteers.
The "tuberculosis-complex" has its usual meaning, i.e. the complex of mycobacteria 5 causing TB which are Mycobacterium tuberculosis, Mycobacterium bovis, Mycobacte rium bovis BCG, and Mycobacterium africanum. -In the present context the term "metabolizing mycobacteria" means live mycobacteria that are multiplying logarithmically and releasing polypeptides into the culture medium 10 wherein they are cultured.
The term "sequence identity" indicates a quantitative measure of the degree of ho-mology between two amino acid sequences or between two nucleotide sequences of equal length, of if not of equal length aligned to best possible fit: The sequence identity can be calculated as ("~' ""')I°" 1, wherein Nd;, is the total number of non-h., l identical residues in the two sequences when aligned and wherein N,e~ is the number of residues in one of the sequences. Hence, the DNA sequence AGTCAGTC will have a sequence identity of 75 % with the sequence AATCAATC (Nd;, = 2 and N,e, =
8). A
gap is counted as non-identity of the specific residue(s), i.e. the DNA
sequence AGTGTC will have a sequence identity of 75% with the DNA sequence AGTCAGTC
(Nd;, = 2 and N,e, = 8). Sequence identity can alternatively be calculated by the BLASTP
program ((Pearson W.R and D.J. Lipman (1988) PNAS USA 85:2444-2448) in the EMBL database (www.ncbi.nlm.govlcgi-bin/BLAST). Generally, the default settings with respect to e.g. "scoring matrix" and "gap penalty" will be used for alignment.
The sequence identity is used here to illustrate the degree of identity between the amino acid sequence of a given polypeptide and the amino acid sequence shown in SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, or 185. The amino acid sequence to be compared with the amino acid sequence shown in SEQ ID NO:
2, 4, 6, 8, 10, 12, 14, 7 6, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, WO 99124577 PCT/DK98/flfl438 any one of 168-171, 175, 177, 179, 181, 183, or 185 may be deduced from a DNA
sequence, e.g. obtained by hybridization as defined below, or may be obtained by conventional amino acid sequencing methods. The sequence identity is preferably de-termined on the amino acid sequence of a mature polypeptide, i.e. without taking any leader sequence into consideration.
As appears from the above disclosure, polypeptides which are not identical to the polypeptides having SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62; 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 7 83, or 185 are embraced by the present invention. The invention allows for minor varia-tions which do not have an adverse effect on immunogenicity compared to the parent sequences and which may give interesting and useful novel binding properties or bio-logical functions and immunogenicities etc.
Each polypeptide fragment may thus be characterized by specific amino acid and nu-cleic acid sequences. It will be understood that such sequences include analogues and variants produced by recombinant methods wherein such nucleic acid and poiypeptide sequences have been modified by substitution, insertion, addition and/or deletion of one or more nucleotides in said nucleic acid sequences to cause the substitution, in-section, addition or deletion of one or more amino acid residues in the recombinant polypeptide. When the term DNA is used in the following, it should be understood that for the number of purposes where DNA can be substituted with RNA, the term DNA
should be read to include RNA embodiments which will be apparent for the man skilled in the art. For the purposes of hybridization, PNA or LNA may be used instead of DNA. PNA has been shown to exhibit a very dynamic hybridization profile (PNA is described in Nielsen P E et al., 1991, Science 254: 1497-1500?. LNA (Locked Nucleic Acids) is a recently introduced oligonucleotide analogue containing-bicyclo nucleoside monomers (Koshkin et al., 1998, 54, 3607-3630;Nielsen, N.K. et al.
J.Am.Chem.Soc 1998, 120, 5458-5463).
In both immunodiagnostics and vaccine preparation, it is often possible and practical to prepare antigens from segments of a known immunogenic protein or poiypeptide.

Certain epitopic regions may be used to produce responses similar to those produced by the entire antigenic polypeptide. Potential antigenic or immunogenic regions may be identified by any of a number of approaches, e.g., Jameson-Wolf or Kyte-Doolittle an-tigenicity analyses or Hopp and Woods (1981 ) hydrophobicity analysis (see, e.g., Ja-meson and Wolf, 1988; Kyte and Doolittle, 1982; or U.S. Patent No. 4,554,101).
Hy-drophobicity analysis assigns average hydrophilicity values to each amino acid residue from these values average hydrophilicities can be calculated and regions of greatest hydrophilicity determined. Using one or more of these methods, regions of predicted antigenicity may be derived from the amino acid sequence assigned to the polypep-tides of the invention.
Alternatively, in order to identify relevant T-cell epitopes which are recognized during an immune response, it is also possible to use a "brute force" method: Since T-cell epitopes are linear, deletion mutants of poiypeptides having SEQ !D NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, or 185 will, if constructed systematically, re-veal what regions of the polypeptides are essential in immune recognition, e.g. by subjecting these deletion mutants to the IFN-y assay described herein. Another method utilises overlapping oligomers (preferably synthetic having a length of e.g.
20 amino acid residues) derived from polypeptides having SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, or 185. Some of these will give a positive response in the IFN-Y assay whereas others will not.
In a preferred embodiment of the invention, the polypeptide fragment of the invention comprises an epitope for a T-helper cell.
Although the minimum length of a T-cell epitope has been shown to be at least amino acids, it is normal that such epitopes are constituted of longer stretches of amino acids. Hence it is preferred that the polypeptide fragment of the invention has a length of at least 7 amino acid residues, such as at least 8, at least 9, at least 10, at feast 12, at least 14, at least 16, at least 18, at least 20, at least 22, at least 24, and at least 30 amino acid residues.
As will appear from the examples, a number of the polypeptides of the invention are natively translation products which include a leader sequence (or other short peptide sequences), whereas the product which can be isolated from short-term culture fil-trates from bacteria belonging to the tuberculosis complex are free of these sequen-ces. Although it may in some applications be advantageous to produce these polypep-tides recombinantly and in this connection facilitate export of the poiypeptides from the host cell by including information encoding the leader sequence in the gene for the polypeptide, it is more often preferred to either substitute the leader sequence with one which has been shown to be superior in the host system for effecting export, or to totally omit the leader sequence (e.g. when producing the polypeptide by peptide synthesis. Hence, a preferred embodiment of the invention is a polypeptide which is free from amino acid residues -30 to -1 in SEQ ID NO: 6 and/or -32 to -1 in NO: 10 and/or -8 to -1 in SEQ ID NO: 12 and/or -32 to -1 in SEQ ID NO: 14 and/or -33 to -1 in SEQ ID NO: 42 andlor -38 to -1 in SEQ ID NO: 52 andlor -33 to -1 in SEQ
ID NO: 56 and/or -56 to -1 in SEQ ID NO: 58 andlor -28 to -1 in SEQ ID NO:
151.
In another preferred embodiment, the polypeptide fragment of the invention is free from any signal sequence; this is especially interesting when the polypeptide fragment is produced synthetically but even when the polypeptide fragments are produced re-combinantly it is normally acceptable that they are not exported by the host cell to the periplasm or the extracellular space; the polypeptide fragments can be recovered by traditional methods (cf. the discussion below) from the cytoplasm after disruption of the host cells, and if there is need for refolding of the polypeptide fragments, general refolding schemes can be employed, cf. e.g. the disclosure in WO 94/18227 where such a general applicable refolding method is described. -A suitable assay for the potential utility of a given polypeptide fragment derived from SEQ ID NO: 2, 4, 6, 8, 1 O, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, or 185 is to assess the ability of the polypeptide fragment to effect IFN-y release from primed memory T-lymphocytes. Poiypeptide fragments which have this capability are according to the invention especially interesting embodiments of the invention: It is contemplated that polypeptide fragments which stimulate T lymphocyte immune response shortly after the onset of the infection are important in the control of the mycobacteria causing the infection before the mycobacteria have succeeded in multiplying up to the number of bacteria that would have resulted in fulminant infection. _' ' It is presently contemplated that when this application refers to IFN-y release as a measure of immunogenicity, other cytokines could be relevant, such as IL-12, TNF-a., IL-4, IL-5, IL-10, IL-6, TGF-~3. Usually one or more cytokines will be measured utilising for example the PCR technique or ELISA. It will be appreciated by the person skilled in the art, that a significant increase or decrease in any~of these cytokines will be indica-tive of an immunological effective polypeptide or polypeptide fragment.
Thus, an important embodiment of the invention is a polypeptide fragment defined above which 1 ) induces a release of IFN-y from primed memory T-lymphocytes withdrawn from a mouse within 2 weeks of primary infection or within 4 days after the mouse has been re-challenge infected with mycobacteria belonging to the tuberculosis complex, the induction performed by the addition of the polypeptide to a suspension comprising about 200,000 spleen cells per ml, the addition of the polypeptide resulting in a con-centration of 1-4 ~g potypeptide per mi suspension, the release of IFN-y being assess-able by determination of IFN-'y in supernatant harvested 2 days after the addition of the polypeptide to the suspension, and/or 2) induces a release of 1FN-y of at least 1,500 pglml above background level from about 1,000,000 human PBMC (peripheral blood mononuclear cells) per ml isolated -from TB patients in the first phase of infection, or from healthy BCG
vaccinated do-nors, or from healthy contacts to TB patients, the induction being performed by the -addition of the polypeptide to a suspension comprising the about 1,000,000 PBMC
per ml, the addition of the polypeptide resulting in a concentration of 1-4 ~g polypep-tide per ml suspension, the release of IFN-y being assessable by determination of IFN-y WO 99lZ4577 PCT/DK98/00438 in supernatant harvested 2 days after the addition of the polypeptide to the suspen-sion; and/or 3) induces an IFN-y release from bovine PBMC derived from animals previously 5 sensitized with mycobacteria belonging to the tuberculosis complex, said release being at least two times the release observed from bovine PBMC derived from animals not previously sensitized with mycobacteria belonging to the tuberculosis complex.
Preferably, in alternatives 1 and 2, the release effected by the polypeptide fragment 10 gives rise to at least 1,500 pg/ml IFN-y in the supernatant but higher concentrations are preferred, e.g. at least 2,000 pglml and even at least 3,000 pg/ml IFN-y in the su-pernatant. The IFN-y release from bovine PBMC can e.g. be measured as the optical density (OD) index over background in a standard cytokine ELISA and should thus be at least two, but higher numbers such as at least 3, 5, 8, and 10 are preferred.
The poiypeptide fragments of the invention preferably comprises an amino acid se-quence of at least 6 amino acid residues in length which has a higher sequence iden-tity than 70 percent with SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, any one of 17-23, 42, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, any one of 72-86, 88, 90, 92, 94, 141, 143, 145, 147, 149, 151, 153, any one of 168-171, 175, 177, 179, 181, 183, or 185. A preferred minimum percentage of sequence identity is at least 80%, such as at least 85%, at least 90%, at feast 91 °~, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, and at least 99.5%.
As mentioned above, it will normally be interesting to omit the leader sequences from the polypeptide fragments of the invention. However, by producing fusion polypep-tides, superior characteristics of the polypeptide fragments of the invention can be achieved. For instance, fusion partners which facilitate export of the polypeptide when 3O produced recombinantiy, fusion partners which facilitate purification of the polypep-tide. and fusion partners which enhance the immunogenicity of the potypeptide frag-ment of the invention are all interesting possibilities. Therefore, the invention also per-tains to a fusion polypeptide comprising at least one polypeptide fragment defined above and at least one fusion partner. The fusion partner can, in order to enhance irn-munogenicity, e.g. be selected from the group consisting of another polypeptide frag-ment as defined above (so as to allow for multiple expression of relevant epitopes), and an other polypeptide derived from a bacterium belonging to the tuberculosis com-plex, such as ESAT-6, CFP7, CFP10, CFP17, CFP21, CFP25, CFP29, MPB59, MPT59, MPB64, and MPT64 or at least one T-cell epitope of any of these antigens.
Other immunogenicity enhancing polypeptides which could serve as fusion partners are T-cell epitopes (e.g. derived from the polypeptides ESAT-6, MPB64, MPT64, or MPB59) or other immunogenic epitopes enhancing the immunogenicity of the target gene product, e.g. lymphokines such as INF-y, IL-2 and IL-12. In order to facilitate expression and/or purification the fusion partner can e.g. be a bacterial fimbrial protein, e.g. the pilus components pilin and papA; protein A; the ZZ-peptide (ZZ-fusions are marketed by Pharmacia in Sweden); the maltose binding protein;
gluthatione S-transferase; ~i-galactosidase; or poly-histidine.
Other interesting fusion partners are polypeptides which are lipidated and thereby ef-fect that the immunogenic polypeptide is presented in a suitable manner to the im-mune system. This effect is e.g. known from vaccines based on the Borrelia burgdor-feri OspA polypeptide, wherein the lipidated membrane anchor in the polypeptide con-fers a self-adjuvating effect to the polypeptide (which is natively lipidated) when iso-lated from cells producing it. In contrast, the OspA pofypeptide is relatively silent im-munologically when prepared without the lipidation anchor.
As evidenced in Example 6A, the fusion polypeptide consisting of MPT59 fused di rectly N-terminally to ESAT-6 enhances the immunogenicity of ESAT-6 beyond what would be expected from the immunogenicities of MPT59 and ESAT-6 alone. The pre-cise reason for this surprising finding is not yet known, but it is expected that either the presence of both antigens lead to a synergistic effect with respect to immuno-genicity or the presence of a sequence N-terminally to the ESAT-6 sequence protects this immune dominant protein from loss of important epitopes known to be present in the N-terminus. A third, alternative, possibility is that the presence of a sequence C-terminally to the MPT59 sequence enhances the immunoiogic properties of this anti-gen.

Hence, one part of the invention pertains to a fusion polypeptide fragment which comprises a first amino acid sequence including at least one stretch of amino acids constituting a T-cell epitope derived from the M. tuberculosis protein ESAT-6 or MPT59, and a second amino acid sequence including at least one T-cell epitope de-rived from a M. tuberculosis protein different from ESAT-6 (if the first stretch of amino acids are derived from ESAT-6) or MPT59 (it the first stretch o~ amino acids are derived from MPT59) and/or including a stretch of amino acids which protects the first amino acid sequence from in vivo degradation or post-translational processing.
The first amino acid sequence maybe situated N- or C-terminally to the second amino acid sequence, but in line with the above considerations regarding protection of the ESAT-6 N-terminus it is preferred that the first amino acid sequence is C-terminal to the second when the first amino acid sequence is derived from ESAT-6.
Although only the effect of fusion between MPT59 and ESAT6 has been investigated at present, it is believed that ESAT6 and MPT59 or epitopes derived therefrom could be advantageously be fused to other fusion partners having substantially the same ef-fect on overall immunogenicity of the fusion construct. Hence, it is preferred that such a fusion polypeptide fragment according of the invention is one, wherein the at least one T-cell epitope included in the second amino acid sequence is derived from a M.
tuberculosis polypeptide (the "parent" polypeptide) selected from the group consisting of a polypeptide fragment according to the present invention and described in detail above and in the examples, or the amino acid sequence could be derived from any one of the M. tuberculosis proteins DnaK, GroEL, urease, glutamine synthetase, the proline rich complex, L-alanine dehydrogenase, phosphate binding protein, Ag 85 complex, HBHA (heparin binding hemagglutinin), MPT51, MPT64, superoxide dismutase, 19 kDa lipoprotein, a-crystallin, GroES, MPT59 (when the first amino acid sequence is derived from ESAT-6), and ESAT-6 (when the first amino acid sequence is derived from MPT59). It is preferred that the first and second T-cell epitopes each have a se-quence identity of at least 70% with the natively occurring sequence in the proteins from which they are derived and it is even further preferred that the first and/or second amino acid sequence has a sequence identity of at least 70% with the protein from which they are derived. A most preferred embodiment of this fusion polypeptide is one wherein the first amino acid sequence is the amino acid sequence of ESAT-6 or MPT59 and/or the second amino acid sequence is the full-length amino acid sequence of the possible "parent" polypeptides listed above.
In the mast preferred embodiment, the fusion polypeptide fragment comprises fused to MPT59 (advantageously, ESAT-6 is fused to the C-terminus of MPT59) and in one special embodiment, there are no linkers introduced between the two amino acid sequences constituting the two parent polypeptide fragments.
Another part of the invention pertains to a nucleic acid fragment in isolated form which 1 ) comprises a nucleic acid sequence which encodes a polypeptide or fusion poly-peptide as defined above, or comprises a nucleic acid sequence complementary there-to, and/or 2) has a length of at least 10 nucleotides and hybridizes readily under stringent hybridization conditions (as defined in the art, i.e. 5-10°C under the melting point Tm, cf. Sambrook et al, 1989, pages 1 1.45-11.49) with a nucleic acid fragment which has a nucleotide sequence selected from SEQ ID NO: 1 or a sequence complementary thereto, SEQ ID NO: 3 or a sequence complementary thereto, SEQ ID NO: 5 or a sequence complementary thereto, SEQ ID NO: 7 or a sequence complementary thereto, SEQ ID NO: 9 or a sequence complementary thereto, SEQ ID NO: 11 or a sequence complementary thereto, SEQ ID NO: 13 or a sequence complementary thereto, SEQ ID NO: 15 or a sequence complementary thereto, SEQ ID NO: 41 or a sequence complementary thereto, _ SEQ ID NO: 47 or a sequence complementary thereto, ' SEQ ID NO: 49 or a sequence complementary thereto, SEQ ID NO: 51 or a sequence complementary thereto, ' SEQ ID NO: 53 or a sequence complementary thereto, SEQ ID NO: 55 or a sequence complementary thereto, SEQ ID NO: 57 or a sequence complementary thereto, SEQ ID NO: 59 or a sequence complementary thereto, SEQ ID NO: 61 or a sequence complementary thereto, SEQ ID NO: 63 or a sequence complementary thereto, SEQ ID NO: 65 or a sequence complementary thereto, SEQ ID NO: 67 or a sequence complementary thereto, SEQ ID NO: 69 or a sequence complementary thereto, SEQ ID NO: 71 or a sequence complementary thereto, SEQ ID NO: 87 or a sequence complementary thereto, SEQ ID NO: 89 or a sequence complementary thereto, SEQ ID NO: 91 or a sequence complementary thereto, SEQ ID NO: 93 or a sequence complementary thereto, SEQ ID NO: 140 or a sequence complementary thereto, SEQ ID NO: 142 or a sequence complementary thereto, SEQ ID NO: 144 or a sequence complementary thereto, SEQ ID NO: 146 or a sequence complementary thereto, SEQ ID NO: 148 or a sequence complementary thereto, SEQ ID NO: 150 or a sequence complementary thereto, SEQ ID NO: 152 or a sequence complementary thereto, SEQ ID NO: 174 or a sequence complementary thereto, SEQ ID NO: 176 or a sequence complementary thereto, SEQ ID NO: 178 or a sequence complementary thereto, SEQ ID NO: 180 or a sequence complementary thereto, SEQ ID NO: 182 or a sequence complementary thereto, and SEQ ID NO: 184 or a sequence complementary thereto with the proviso that when the nucleic acid fragment comprises a subsequence of SEQ ID NO: 41, then the nucleic acid fragment contains an A corresponding to posi-tion 781 in SEQ ID NO: 41 and when the nucleic acid fragment comprises a subse-quence of a nucleotide sequence exactly complementary to SEQ ID NO: 41, then the nucleic acid fragment comprises a T corresponding to position 781 in SEQ ID
NO: 41.
It is preferred that the nucleic acid fragment is a DNA fragment.

To provide certainty of the advantages in accordance with the invention, the preferred nucleic acid sequence when employed for hybridization studies or assays includes se-quences that are complementary to at least a 10 to 40, or so, nucleotide stretch of the selected sequence. A size of at least 10 nucleotides in length heaps to ensure that 5 the fragment will be of sufficient length to form a duplex molecule that is both stable and selective. Molecules having complementary sequences over stretches greater than 10 bases in length are generally preferred, though, in order to increase stability and selectivity of the hybrid, and thereby improve the quality and degree of specific hybrid molecules obtained.
Hence, the term "subsequence" when used in connection with the nucleic acid frag-ments of the invention is intended to indicate a continuous stretch of at feast 10 nu-cleotides exhibits the above hybridization pattern. Normally this will require a minimum sequence identity of at least 70% with a subsequence of the hybridization partner having SEQ ID NO: 1, 3, 5, 7, 9, 1 1, 12, 15, 21, 41, 47, 49, 51, 53, 55, 57, 59, 61, 63, 65, 67, 69, 71, 87, 89, 91, 93, 140, 142, 144, 146, 148, 150, 152, 174, 176, 178, 180, 182, or 184. tt is preferred that the nucleic acid fragment is longer than 10 nucleotides, such as at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at feast 45, at least 50, at least 55, at least 60, at least 65, at least 70, and at least 80 nucleotides long, and the sequence identity should preferable also be higher than 70%, such as at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 94%, at least 96%, and at least 98%. It is most preferred that the sequence identity is 100%. Such fragments may be readily prepared by, for example, directly synthesizing the fragment by chemical means, by application of nucleic acid reproduction technology, such as the PCR technology of U.S. Patent 4,603,102, or by introducing selected sequences into recombinant vectors for recombinant production.
it is well known that the same amino acid may be encoded by various codons, the co don usage being related, inter alia, to the preference of the organisms in question ex pressing the nucleotide sequence. Thus, at least one nucleotide or codon of a nucleic acid fragment of the invention may be exchanged by others which, when expressed, result in a polypeptide identical or substantially identical to the polypeptide encoded by the nucleic acid fragment in question. The invention thus allows for variations in the sequence such as substitution, insertion (including introns?, addition, deletion and rearrangement of one or more nucleotides, which variations do not have any substantial effect on the polypeptide encoded by the nucleic acid fragment or a subsequence thereof. The term "substitution" is intended to mean the replacement of one or more nucleotides in the full nucleotide sequence with one or more different nucleotides, "addition" is understood to mean the addition of one or more nucleotides at either end of the full nucleotide sequence, "insertion" is intended to mean the introduction of one or more nucleotides within the full nucleotide sequence, "deletion"
is intended to indicate that one or more nucleotides have been deleted from the full nucleotide sequence whether at either end of the sequence or at any suitable point within it, and "rearrangement" is intended to mean that two or more nucleotide residues have been exchanged with each other.
The nucleotide sequence to be modified may be of cDNA or genomic origin as discus-sed above, but may also be of synthetic origin. Furthermore, the sequence may be of mixed cDNA and genomic, mixed cDNA and synthetic or genomic and synthetic origin as discussed above. The sequence may have been modified, e.g. by site-directed mu-tagenesis, to result in the desired nucleic acid fragment encoding the desired polypep-tide. The following discussion focused on modifications of nucleic acid encoding the polypeptide should be understood to encompass also such possibilities, as well as the possibility of building up the nucleic acid by ligation of two or more DNA
fragments to obtain the desired nucleic acid fragment, and combinations of the above-mentioned principles.
The nucleotide sequence may be modified using any suitable technique which results in the production of a nucleic acid fragment encoding a polypeptide of the invention.
The modification of the nucleotide sequence encoding the amino acid sequence of the polypeptide of the invention should be one which does not impair the immunological function of the resulting polypeptide.
A preferred method of preparing variants of the antigens disclosed herein is site-di-rected mutagenesis. This technique is useful in the preparation of individual peptides, or biologically functional equivalent proteins or peptides, derived from the antigen se-quences, through specific mutagenesis of the underlying nucleic acid. The technique further provides a ready ability to prepare and test sequence variants, for example, incorporating one or more of the foregoing considerations, by introducing one or more nucleotide sequence changes into the nucleic acid. Site-specific mutagenesis allows the production of mutants through the use of specific oligonucleotide sequences which encode the nucleotide sequence of the desired mutation, as well as a sufficient number of adjacent nucleotides, to provide a primer sequence of su#icient size and sequence complexity to form a stable duplex on both sides of the deletion junction being traversed. Typically, a primer of about 17 to 25 nucleotides in length is pre-ferred, with about 5 to 10 residues on both sides of the junction of the sequence be-ing altered.
In general, the technique of site-specific mutagenesis is well known in the art as ex-emplified by publications (Adelman et al., 1983). As will be appreciated, the technique typically employs a phage vector which exists in both a single stranded and double stranded form. Typical vectors useful in site-directed mutagenesis include vectors such as the M 13 phage (Messing et al., 1981 ). These phage are readily commercially available and their use is generally well known to those skilled in the art.
fn general, site-directed mutagenesis in accordance herewith is performed by first ob-taining a single-stranded vector which includes within its sequence a nucleic acid se-quence which encodes the polypeptides of the invention. An oligonucleotide primer bearing the desired mutated sequence is prepared, generally synthetically, for example by the method of Crea et al. (1978). This primer is then annealed with the single-stranded vector, and subjected to DNA polymerizing enzymes such as E. coli polymer-ase I Klenow fragment, in order to complete the synthesis of the mutation-bearing strand. Thus, a heteroduplex is formed wherein one strand encodes the original non-mutated sequence and the second strand bears the desired mutation. This hetero-duplex vector is then used to transform appropriate cells, such as E. coli cells, and clones are selected which include recombinant vectors bearing the mutated sequence arrangement.
The preparation of sequence variants of the selected nucleic acid fragments of the in-vention using site-directed mutagenesis is provided as a means of producing poten-tially useful species of the genes and is not meant to be limiting as there are other ways in which sequence variants of the nucleic acid fragments of the invention may be obtained. For example, recombinant vectors encoding the desired genes may be treated with mutagenic agents to obtain sequence variants (see, e.g., a method de-scribed by Eichenlaub, 1979y for the mutagenesis of plasmid DNA using hydroxyl-s amine.
The invention also relates to a replicable expression vector which comprises a nucleic acid fragment defined above, especially a vector which comprises a nucleic acid frag-ment encoding a polypeptide fragment of the invention.
The vector may be any vector which may conveniently be subjected to recombinant DNA procedures, and the choice of vector will often depend on the host cell into which it is to be introduced. Thus, the vector may be an autonomously replicating vector, i.e. a vector which exists as an extrachromosomal entity, the replication of which is independent of chromosomal replication; examples of such a vector are a plasmid, phage, cosmid, mini-chromosome or virus. Alternatively, the vector may be one which, when introduced in a host cell, is integrated in the host cell genome and replicated together with the chromosomes) into which it has been integrated.
Expression vectors may be constructed to include any of the DNA segments disclosed herein. Such DNA might encode an antigenic protein specific for virulent strains of mycobacteria or even hybridization probes for detecting .mycobacteria nucleic acids in samples. Longer or shorter DNA segments could be used, depending on the antigenic protein desired. Epitopic regions of the proteins expressed or encoded by the disclosed DNA could be included as relatively short segments of DNA. A wide variety of expres-sion vectors is possible including, for example, DNA segments encoding reporter gene products useful for identification of heterologous gene products andlor resistance genes such as antibiotic resistance genes which may be useful in identifying trans-formed cells.
The vector of the invention may be used to transform cells so as to allow propagation of the nucleic acid fragments of the invention or so as to allow expression of the.poly-peptide fragments of the invention. Hence, the invention also pertains to a transform-ed cell harbouring at least one such vector according to the invention, said cell being one which does not natively harbour the vector and/or the nucleic acid fragment of the invention contained therein. Such a transformed cell (which is also a part of the invention) may be any suitable bacterial host cell or any other type of cell such as a unicellular eukaryotic organism, a fungus or yeast, or a cell derived from a multiceilular organism, e.g. an animal or a plant. It is especially in cases where glycosylation is de-sired that a mammalian cell is used, although glycosylation of proteins is a rare event in prokaryotes. Normally, however, a prokaryotic cell is preferred such as a bacterium belonging to the genera Mycobacterium, Salmonella, Pseudomonas, Bacillus and Es-chericia. It is preferred that the transformed cell is an E. coli, B.
subtilis, or M. bovis BCG cell, and it is especially preferred that the transformed cell expresses a polypep-tide according of the invention. The latter opens for the possibility to produce the polypeptide of the invention by simply recovering it from the culture containing the transformed cell. In the most preferred embodiment of this part of the invention the transformed cell is Mycobacterium bovis BCG strain: Danish 1331, which is the My-cobacterium bovis strain Copenhagen from the Copenhagen BCG Laboratory, Statens Seruminstitut, Denmark.
The nucleic acid fragments of the invention allow for the recombinant production of the polypeptides fragments of the invention. However, also isolation from the natural source is a way of providing the polypeptide fragments as is peptide synthesis.
Therefore, the invention also pertains to a method for the preparation of a polypeptide fragment of the invention, said method comprising inserting a nucleic acid fragment as defined above into a vector which is able to replicate in a host cell, introducing the resulting recombinant vector into the host cell ltransformed cells may be selected us-ing various techniques, including screening by differential hybridization, identification of fused reporter gene products, resistance markers, anti-antigen antibodies and the like), culturing the host cell in a culture medium under conditions sufficient to effect expression of the polypeptide (of course the cell may be cultivated under conditions appropriate to the circumstances, and if DNA is desired, replication conditions are used), and recovering the polypeptide from the host cell or culture medium; or isolating the polypeptide from a short-term culture filtrate as defined in claim 1; or isolating the poiypeptide from whole mycobacteria of the tuberculosis complex or from lysates or fractions thereof, e.g. cell wall containing fractions, or synthesizing the polypeptide by solid or liquid phase peptide synthesis.

The medium used to grow the transformed cells may be any conventional medium suitable for the purpose. A suitable vector may be any of the vectors described above, and an appropriate host cell may be any of the cell types listed above. The methods employed to construct the vector and effect introduction thereof into the host cell 10 may be any methods known for such purposes within the field of recombinant DNA. In the following a more detailed description of the possibilities will be given:
In general, of course, prokaryotes are preferred for the initial cloning of nucleic se-quences of the invention and constructing the vectors useful in the invention.
For ex-1 5 ample, in addition to the particular strains mentioned in the more specific disclosure below, one may mention by way of example, strains such as E. coli K12 strain (ATCC No. 31446), E. coli B, and E. coli X 1776 (ATCC No. 31537). These examples are, of course, intended to be illustrative rather than limiting.
20 Prokaryotes are also preferred for expression. The aforementioned strains, as well as E. coli W31 10 (F-, lambda-, prototrophic, ATCC No. 2733251, bacilli such as Bacillus subtilis, or other enterobacteriaceae such as Salmonella typhimurium or Serratia mar-cesans, and various Pseudomonas species may be used. Especially interesting are rapid-growing mycobacteria, e.g. M. smegmatis, as these bacteria have a high degree 25 of resemblance with mycobacteria of the tuberculosis complex and therefore stand a good chance of reducing the need of performing post-translational modifications of the expression product.
In general, plasmid vectors containing replicon and control sequences which are de-rived from species compatible with the host cell are used in connection with these hosts. The vector ordinarily carries a replication site, as well as marking sequences which are capable of providing phenotypic selection in transformed cells. For example, E. coli is typically transformed using pBR322, a plasmid derived from an E.
coli species (see, e.g., Bolivar et al., 1977, Gene 2: 95). The pBR322 plasmid contains genes for ampiciliin and tetracycline resistance and thus provides easy means for identifying transformed cells. The pBR plasmid, or other microbial plasmid or phage must also contain, or be modified to contain, promoters which can be used by the microorgan-ism for expression.
Those promoters most commonly used in recombinant DNA construction include the B-lactamase (penicillinase) and lactose promoter systems (Chang et al., 1978;
Itakura et al., 1977; Goeddef et al., 1979) and a tryptophan (trp) promoter system (Goeddel et al., 1979; EPO Appl. Publ. No. 0036776). While these are the most commonly used, other microbial promoters have been discovered and utilized, and details con-cerning their nucleotide sequences have been published, enabling a skilled worker to ligate them functionally with plasmid vectors (Siebwenlist et al., 1980).
Certain genes from prokaryotes may be expressed efficiently in E. coli from their own promoter se-quences, precluding the need for addition of another promoter by artificial means.
After the recombinant preparation of the polypeptide according to the invention, the isolation of the polypeptide may for instance be carried out by affinity chromatography (or other conventional biochemical procedures based on chromatography), using a monoclonal antibody which substantially specifically binds the polypeptide according to the invention. Another possibility is to employ the simultaneous electroelution tech-pique described by Andersen et a!. in J. lmmunol. Methods 7 6'! : 29-39.
According to the invention the post-translationai modifications involves lipidation, gly-cosylation, cleavage, or elongation of the polypeptide.
In certain aspects, the DNA sequence information provided by this invention allows for the preparation of relatively short DNA for RNA or PNA) sequences having the ability to specifically hybridize to mycobacterial gene sequences. In these aspects, nucleic acid probes of an appropriate length are prepared based on a consideration of the rele-want sequence. The ability of such nucleic acid probes to specifically hybridize to the mycobacterial gene sequences lend them particular utility in a variety of embodiments.
Most importantly, the probes can be used in a variety of diagnostic assays for detect-ing the presence of pathogenic organisms in a given sample. However, either uses are envisioned, including the use of the sequence information for the preparation of mu-tant species primers, or primers for use in preparing other genetic constructs.
Apart from their use as starting points for the synthesis of polypeptides of the inven tion and for hybridization probes (useful for direct hybridization assays or as primers in e.g. PCR or other molecular amplification methods) the nucleic acid fragments of the invention may be used for effecting in vivo expression of antigens, i.e. the nucleic acid fragments may be used in so-called DNA vaccines. Recent research have revealed that a DNA fragment cloned in a vector which is non- replicative in eukaryotic cells may be introduced into an animal (including a human being) by e.g. intramuscular injection or percutaneous administration (the so-called "gene gun" approach). The DNA is taken up by e.g. muscle cells and the gene of interest is expressed by a promoter which is functioning in eukaryotes, e.g. a viral promoter, and the gene product thereafter stimulates the immune system. These newly discovered methods are reviewed in UI-mer et al., 1993, which hereby is included by reference.
Hence, the invention also relates to a vaccine comprising a nucleic acid fragment ac-cording to the invention, the vaccine effecting in vivo expression of antigen by an ani-mal, including a human being, to whom the vaccine has been administered, the amount of expressed antigen being effective to confer substantially increased resis-tance to infections with mycobacteria of the tuberculosis complex in an animal, inclu-ding a human being.
The efficacy of such a "DNA vaccine" can possibly be enhanced by administering the gene encoding the expression product together with a DNA fragment encoding a poly-peptide which has the capability of modulating an immune response. For instance, a gene encoding lymphokine precursors or lymphokines (e.g. IFN-y, IL-2, or IL-12) could be administered together with the gene encoding the immunogerlic protein, either by administering two separate DNA fragments or by administering both DNA
fragments included in the same vector. It also is a possibility to administer DNA
fragments com-prising a multitude of nucleotide sequences which each encode relevant epitopes of the polypeptides disclosed herein so as to effect a continuous sensitization of the im-mune system with a broad spectrum of these epitopes.

As explained above, the polypeptide fragments of the invention are excellent candi-dates for vaccine constituents or for constituents in an immune diagnostic agent due to their extracellular presence in culture media containing metabolizing virulent myco-bacteria belonging to the tuberculosis complex, or because of their high homologies with such extracellular antigens, or because of their absence in M. bovis BCG.
Thus, another part of the invention pertains to an immunologic composition comprising a polypeptide or fusion polypeptide according to the invention. In order to ensure optimum performance of such an immunologic composition it is preferred that it comprises an immunologically and pharmaceutically acceptable carrier, vehicle or adjuvant.
Suitable carriers are selected from the group consisting of a polymer to which the polypeptide(s) is/are bound by hydrophobic non-covalent interaction, such as a plastic, e.g. polystyrene, or a polymer to which the polypeptide(s) is/are covalently bound, such as a polysaccharide, or a polypeptide, e.g. bovine serum albumin, ovalbumin or keyhole limpet haemocyanin. Suitable vehicles are selected from the group consisting of a difuent and a suspending agent. The adjuvant is preferably selected from the group consisting of dimethyldioctadecylammonium bromide (DDA), Quil A, poly I:C, Freund's incomplete adjuvant, IFN-y, IL-2, IL-12, monophosphoryl lipid A
(MPL), and muramyi dipeptide (MDP).
A preferred immunologic composition according to the present invention comprising at least two different polypeptide fragments, each different polypeptide fragment being a poiypeptide or a fusion polypeptide defined above. It is preferred that the immunologic composition comprises between 3-20 different polypeptide fragments or fusion poly-peptides.
Such an immunologic composition may preferably be in the form of a vaccine or in the form of a skin test reagent.
In line with the above, the invention therefore also pertain to a method for producing an immunologic composition according to the invention, the method comprising pre-paring, synthesizing or isolating a polypeptide according to the invention, and solubi-lining or dispersing the potypeptide in a medium for a vaccine, and optionally adding other M. tuberculosis antigens and/or a carrier, vehicle and/or adjuvant substance.
Preparation of vaccines which contain peptide sequences as active ingredients is gen-erally well understood in the art, as exemplified by U.S. Patents 4,608,251;
4,601,903; 4,599,231; 4,599,230; 4,596,792; and 4,578,770, all incorporated herein by reference. Typically, such vaccines are prepared as injectables either as liquid solutions or suspensions; solid forms suitable for solution in, or suspension in, liquid prior to injection may also be prepared. The preparation may also be emulsified.
The active immunogenic ingredient is often mixed with excipients which are pharma-ceutically acceptable and compatible with the active ingredient. Suitable excipients are, for example, water, saline, dextrose, glycerol, ethanol, or the like, and combina-tions thereof. In addition, if desired, the vaccine may contain minor amounts of auxil-iary substances such as wetting or emulsifying agents, pH buffering agents, or adju-vants which enhance the effectiveness of the vaccines.
The vaccines are conventionally administered parenterally, by injection, for example, either subcutaneously or intramuscularly. Additional formulations which are suitable for other modes of administration include suppositories and, in some cases, oral for-mutations. For suppositories, traditional binders and carriers may include, for example, polyalkalene glycols or triglycerides; such suppositories may be formed from mixtures containing the active ingredient in the range of 0.5% to 1.0%, preferably 1-2%. Oral formulations include such normally employed excipients as, for example, pharmaceu-tical grades of mannitol, lactose, starch, magnesium stearate, sodium saccharine, cel-lulose, magnesium carbonate, and the like. These compositions take the form of solu-tions, suspensions, tablets, pills, capsules, sustained release formulations or powders and contain 10-95% of active ingredient, preferably 25-70%.
The proteins may be formulated into the vaccine as neutral or salt forms.
Pharmaceu-tically acceptable salts include acid addition salts (formed with the free amino groups of the peptide) and which are formed with inorganic acids such as, for example, hy-drochloric or phosphoric acids, or such organic acids as acetic oxalic, tartaric, man-delic, and the like. Salts formed with the free carboxyl groups may also be derived from inorganic bases such as, for example, sodium, potassium, ammonium, calcium, or ferric hydroxides, and such organic bases as isopropylamine, trimethytamine, 2-eth-ylamino ethanol, histidine, procaine, and the like.
The vaccines are administered in a manner compatible with the dosage formulation, 5 and in such amount as will be therapeutically effective and immunogenic. The quantity to be administered depends on the subject to be treated, including, a g., the capacity of the individual's immune system to mount an immune response, and the degree of protection desired. Suitable dosage ranges are of the order of several hundred micro-grams active ingredient per vaccination with a preferred range from about 0.1 ug to 10 1000 gg, such as in the range from about 1 ~g to 300 ug, and especially in the range from about 10 ug to 50 pg. Suitable regimens for initial administration and booster shots are also variable but are typified by an initial administration followed by subse-quent inoculations or other administrations.

15 The manner of application may be varied widely. Any of the conventional methods far administration of a vaccine are applicable. These are believed to include oral applica-tion on a solid physiologically acceptable base or in a physiologically acceptable dis-persion, parenterally, by injection or the like. The dosage of the vaccine will depend on the route of administration and will vary according to the age of the person to be vac-20 cinated and, to a lesser degree, the size of the person to be vaccinated.
Some of the polypeptides of the vaccine are sufficiently immunogenic in a vaccine, but for some of the others the immune response will be enhanced if the vaccine fur-ther comprises an adjuvant substance.
Various methods of achieving adjuvant effect for the vaccine include use of agents such as aluminum hydroxide or phosphate lalum), commonly used as 0.05 to 0.1 per-cent solution in phosphate buffered saline, admixture with synthetic polymers of sug-ars (Carbopol) used as 0.25 percent solution, aggregation of the protein in the vaccine by heat treatment with temperatures ranging between 70° to 101 °C for 30 second to 2 minute periods respectively. Aggregation by reactivating with pepsin treated (Fab) antibodies to albumin, mixture with bacterial cells such as C. parvum or endotoxins or lipopolysaccharide components of gram-negative bacteria, emulsion in physiologically acceptable oil vehicles such as mannide mono-oleate (Aracel A) or emulsion with 20 percent solution of a perfluorocarbon (Fluosol-DA) used as a block substitute may also be employed. According to the invention DDA (dimethyldioctadecylammonium bro-mide) is an interesting candidate for an adjuvant, but also Freund's complete and in-complete adjuvants as well as QuilA and RIBI are interesting possibilities.
Further pos-y sibilities are monophosphoryl lipid A (MPL), and muramyl dipeptide (MDP).
Another highly interesting (and thus, preferred) possibility of achieving adjuvant effect is to employ the technique described in Gosselin et al., 1992 (which is hereby incorpo-rated by reference herein). In brief, the presentation of a relevant antigen such as an antigen of the present invention can be enhanced by conjugating the antigen to anti-bodies (or antigen binding antibody fragments) against the Fcy receptors on mono-cyteslmacrophages. Especially conjugates between antigen and anti-FcyRl have been demonstrated to enhance immunogenicity for the purposes of vaccination.
Other possibilities involve the use of immune modulating substances such as lympho-kines (e.g. IFN-y, IL-2 and IL-12) or synthetic IFN-y inducers such as poly I:C in combi-nation with the above-mentioned adjuvants. As discussed in example 3, it is contem-plated that such mixtures of antigen and adjuvant will lead to superior vaccine formu-lations.
In many instances, it will be necessary to have multiple administrations of the vaccine, usually not exceeding six vaccinations, more usually not exceeding four vaccinations and preferably one or more, usually at least about three vaccinations. The vaccinations will normally be at from two to twelve week intervals, more usually from three to five week intervals. Periodic boosters at intervals of 1-5 years, usually three years, will be desirable to maintain the desired levels of protective immunity. The course of the im-munization may be followed by in vitro proliferation assays of PBL (peripheral blood lymphocytes) co-cultured with ESAT-6 or ST-CF, and especially by measuring the levels of IFN-y released form the primed lymphocytes. The assays may be performed using conventional labels, such as radionuclides, enzymes, fluorescers, and the like.
These techniques are well known and may be found in a wide variety of patents, such as U.S. Patent Nos. 3,791,932; 4,174,384 and 3,949,064, as illustrative of these types of assays.

Due to genetic variation, different individuals may react with immune responses of varying strength to the same polypeptide. Therefore, the vaccine according to the in-vention may comprise several different polypeptides in order to increase the immune response. The vaccine may comprise two or more polypeptides, where all of the poly-peptides are as defined above, or some but not all of the peptides may be derived from a bacterium belonging to the M. tuberculosis complex. In the latter example the polypeptides not necessarily fulfilling the criteria set forth above for polypeptides may either act due to their own immunogenicity or merely act as adjuvants.
Examples of such interesting polypeptides are MPB64, MPT64, and MPB59, but any other sub-stance which can be isolated from mycobacteria are possible candidates.
The vaccine may comprise 3-20 different polypeptides, such as 3-10 different poly-peptides.
One reason for admixing the polypeptides of the invention with an adjuvant is to ef-fectively activate a cellular immune response. However, this effect can also be achieved in other ways, for instance by expressing the effective antigen in a vaccine in a non-pathogenic microorganism. A well-known example of such a microorganism is Mycobacterium bouts BCG.
Therefore, another important aspect of the present invention is an improvement of the living BCG vaccine presently available, which is a vaccine for immunizing an animal, including a human being, against TB caused by mycobacteria belonging to the tuber-culosis-complex, comprising as the effective component a microorganism, wherein one or more copies of a DNA sequence encoding a polypeptide as defined above has been incorporated into the genome of the microorganism in a manner allowing the microorganism to express and secrete the poiypeptide.
In the present context the term "genome" refers to the chromosome of the microor- -ganisms as well as extrachromosomally DNA or RNA, such as plasmids. It is, how-ever, preferred that the DNA sequence of the present invention has been introduced -into the chromosome of the non-pathogenic microorganism, since this will prevent loss of the genetic material introduced.

It is preferred that the non-pathogenic microorganism is a bacterium, e.g.
selected from the group consisting of the genera Mycobacterium, Salmonella, Pseudomonas and Eschericia. It is especially preferred that the non-pathogenic microorganism is My-cobacterium bouts BCG, such as Mycobacterium bouts BCG strain: Danish 1331.
The incorporation of one or more copies of a nucleotide sequence encoding the poly-peptide according to the invention in a mycobacterium from a M. bouts BCG
strain will enhance the immunogenic effect of the BCG strain. The incorporation of more than one copy of a nucleotide sequence of the invention is contemplated to enhance the immune response even more, and consequently an aspect of the invention is a vaccine wherein at least 2 copies of a DNA sequence encoding a polypeptide is incorporated in the genome of the microorganism, such as at least 5 copies. The copies of DNA
se-quences may either be identical encoding identical polypeptides or be variants of the same DNA sequence encoding identical or homologues of a polypeptide, or in another embodiment be different DNA sequences encoding different polypeptides where at least one of the polypeptides is according to the present invention.
The living vaccine of the invention can be prepared by cultivating a transformed non-pathogenic cell according to the invention, and transferring these cells to a medium for a vaccine, and optionally adding a carrier, vehicle and/or adjuvant substance.
The invention also relates to a method of diagnosing TB caused by Mycobacterium tuberculosis, Mycobacterium afiicanum or Mycobacterium bouts in an animal, includ-ing a human being, comprising intradermaNy injecting, in the animal, a polypeptide ac-cording to the invention or a skin test reagent described above, a positive skin re-sponse at the location of injection being indicative of the animal having TB, and a negative skin response at the location of injection being indicative of the animal not having TB. A positive response is a skin reaction having a diameter of at least 5 mm, but larger reactions are preferred, such as at least 1 cm, 1.5 cm, and at least 2 cm in diameter. The composition used as the skin test reagent can be prepared in the same manner as described for the vaccines above.
In line with the disclosure above pertaining to vaccine preparation and use, the inven-lion also pertains to a method for immunising an animal, including a human being, against TB caused by mycobacteria belonging to the tuberculosis complex, comprising administering to the animal the polypeptide of the invention, or a vaccine composition of the invention as described above, or a living vaccine described above.
Preferred routes of administration are the parenteral (such as intravenous and intraarterially), in-s traperitoneal, intramuscular, subcutaneous, intradermal, oral, buccal, sublingual, nasal, rectal or transdermal route.
The protein ESAT-6 which is present in short-term culture filtrates from mycobacteria as well as the esat-6 gene in the mycobacterial genome has been demonstrated to have a very limited distribution in other mycobacterial strains that M.
tuberculosis, e.g.
esat-6 is absent fn both BCG and the majority of mycobacterial species isolated from the environment, such as M. avium and M. terrae. It is believed that this is also the case for at least one of the antigens of the present invention and their genes and therefore, the diagnostic embodiments of the invention are especially well-suited for performing the diagnosis of on-going or previous infection with virulent mycobacterial strains of the tuberculosis complex, and it is contemplated that it will be possible to distinguish between 1 ) subjects (animal or human) which have been previously vacci-nated with e.g. BCG vaccines or subjected to antigens from non-virulent mycobacteria and 2) subjects which have or have had active infection with virulent mycobacteria.
A number of possible diagnostic assays and methods can be envisaged:
When diagnosis of previous or ongoing infection with virulent mycobacteria is the aim, a blood sample comprising mononuclear cells (i.a. T-lymphocytes) from a patient could be contacted with a sample of one or more polypeptides of the invention. This con-tacting can be performed in vitro and a positive reaction could e.g, be proliferation of the T-cells or release cytokines such as y-interferon into the extracellular phase (e.g.
into a culture supernatant); a suitable in vivo test would be a skin test as described above. It is also conceivable to contact a serum sample from a subject to contact with a polypeptide of the invention, the demonstration of a binding between antibodies in the serum sample and the polypeptide being indicative of previous or ongoing infec-Lion.

The invention therefore also relates to an in vitro method for diagnosing ongoing or previous sensitization in an animal or a human being with bacteria belonging to the tuberculosis complex, the method comprising providing a blood sample from the ani-mal or human being, and contacting the sample from the animal with the polypeptide 5 of the invention, a significant release into the extracelluiar phase of at least one cyto-kine by mononuclear cells in the blood sample being indicative of the- animal being sensitized. By the term "significant release" is herein meant that the release of the cytokine is significantly higher than the cytokine release from a blood sample derived from a non-tuberculous subject (e.g. a subject which does not react in a traditional 10 skin test for TB). Normally, a significant release is at least two times the release ob-served from such a sample.
Alternatively, a sample of a possibly infected organ may be contacted with an anti-body raised against a polypeptide of the invention. The demonstration of the reaction 15 by means of methods well-known in the art between the sample and the antibody will be indicative of ongoing infection. It is of course also a possibility to demonstrate the presence of anti-mycobacterial antibodies in serum by contacting a serum sample from a subject with at least one of the polypeptide fragments of the invention and using well-known methods for visualizing the reaction between the antibody and antigen.
Also a method of determining the presence of mycobacterial nucleic acids in an ani-mal, including a human being, or in a sample, comprising administering a nucleic acid fragment of the invention to the animal or incubating the sample with the nucleic acid fragment of the invention or a nucleic acid fragment complementary thereto, and de-tecting the presence of hybridized nucleic acids resulting from the incubation (by using the hybridization assays which are well-known in the art), is also included in the inven-tion. Such a method of diagnosing TB might involve the use of a composition compri-sing at least a part of a nucleotide sequence as defined above and-detecting the pre-sence of nucleotide sequences in a sample from the animal or human being to be tested which hybridize with the nucleic acid fragment (or a complementary fragment) by the use of PCR technique.
The fact that certain of the disclosed antigens are not present in M. bouts BCG but are present in virulent mycobacteria point them out as interesting drug targets;
the anti-gens may constitute receptor molecules or toxins which facilitate the infection by the mycobacterium, and if such functionalities are blocked the infectivity of the mycobac-terium will be diminshed.
To determine particularly suitable drug targets among the antigens of the invention, the gene encoding at least one of the polypeptides of the invention arid the necessary control sequences can be introduced into avirulent strains of mycobacteria (e.g. BCG) so as to determine which of the polypeptides are critical for virulence. Once particular proteins are identified as critical for/contributory to virulence, anti-mycobacterial agents can be designed rationally to inhibit expression of the critical genes or to attack the critical gene products. For instance, antibodies or fragments thereof (such as Fab and (Fab')2 fragments can be prepared against such critical polypeptides by methods known in the art and thereafter used as prophylactic or therapeutic agents.
Alterna-tively, small molecules can be screened for their ability to selectively inhibit expression of the critical gene products, e.g. using recombinant expression systems which in-clude the gene's endogenous promoter, or for their ability to directly interfere with the action of the target. These small molecules are then used as therapeutics or as pro-phylactic agents to inhibit mycobacterial virulence.
Alternatively, anti-mycobacterial agents which render a virulent mycobacterium aviru-lent can be operably linked to expression control sequences and used to transform a virulent mycobacterium. Such anti-mycobacterial agents inhibit the replication of a specified mycobacterium upon transcription or translation of the agent in the myco-bacterium. Such a "newly avirulent" mycobacterium would constitute a superb alter-native to the above described modified BCG for vaccine purposes since it would be immunologically very similar to a virulent mycobacterium compared to e.g. BCG.
Finally, a monoclonal or polyclonal antibody, which is specifically reacting with a poly-peptide of the invention in an immuno assay, or a specific binding fragment of said antibody, is also a part of the invention. The production of such polyclonal antibodies requires that a suitable animal be immunized with the polypeptide and that these anti- -bodies are subsequently isolated, suitably by immune affinity chromatography.
The production of monoclonals can be effected by methods well-known in the art, since the present invention provides for adequate amounts of antigen for both immunization and screening of positive hybridomas.
LEGENDS TO THE FIGURES
Fig. 1: Long term memory immune mice are very efficiently protected towards an in-fection with M. tuberculosis. Mice were given a challenge of M_ tuberculosis and spleens were isolated at different time points. Spleen lymphocytes were stimulated in vitro with ST-CF and the release of IFN-y investigated (panel A). The counts of CFU in the spleens of the two groups of mice are indicated in panel B. The memory immune mice control infection within the first week and produce large quantities of IFN-y in response to antigens in ST-CF.
Fig. 2: T cells involved in protective immunity are predominantly directed to molecules from 6-12 and 17-38 kDa. Splenic T cells were isolated four days after the challenge with M. tuberculosis and stimulated in vitro with narrow molecular mass fractions of ST-CF. The release of IFN-'y was investigated Fig. 3: Nucleotide sequence (SEQ ID NO: 1 ) of cfp7. The deduced amino acid se-quence (SEQ ID NO: 2) of CFP7 is given in conventional one-letter code below the nu-cleotide sequence. The putative ribosome-binding site is written in underlined italics as are the putative -10 and -35 regions. Nucleotides written in bold are those encoding CFP7.
Fig. 4. Nucleotide sequence (SEQ ID NO: 3) of cfp9. The deduced amino acid se-quence (SEQ ID NO: 4) of CFP9 is given in conventional one-letter code below the nu-cleotide sequence. The putative ribosome-binding site Shine Delgarno sequence is written in underlined italics as are the putative -10 and -35 regions.
Nucleotides in bold writing are those encoding CFP9. The nucleotide sequence obtained from the lambda 226 phage is double underlined.
Fig. 5: Nucleotide sequence of mpt5l. The deduced amino acid sequence of MPT51 is given in a one-letter code below the nucleotide sequence. The signal is indicated in italics. The putative potential ribosome-binding site is underlined. The nucleotide differ-ence and amino acid difference compared to the nucleotide sequence of MPB51 (Ohara et al., 1995) are underlined at position 780. The nucleotides given in italics are not present in M. tuberculosis H37Rv.
Fig. 6: the position of the purified antigens in the 2DE system have been determined and mapped in a reference gel. The newly purified antigens are encircled and the posi-tion of well-known proteins are also indicated.

Identification of single culture filtrate antigens involved in p~oteciive immunity A group of efficiently protected mice was generated by infecting 8-12 weeks old fe-male C57B1/6j mice with 5 x 10° M. tuberculosis i.v. After 30 days of infection the mice were subjected to 60 days of antibiotic treatment with isoniazid and were then left for 200-240 days to ensure the establishment of resting long-term memory immu-nity. Such memory immune mice are very efficiently protected against a secondary infection (Fig. 1 ). Long lasting immunity in this model is mediated by a population of highly reactive CD4 cells recruited to the site of infection and triggered to produce large amounts of IFN-y in response to ST-CF (Fig. 1 ) /Andersen et al. 1995).
We have used this model to identify single antigens recognized by protective T
cells.
Memory immune mice were reinfected with 1 x 106 M. tuberculosis i.v. and splenic lymphocytes were harvested at day 4-6 of reinfection, a time point where this popula-tion is highly reactive to ST-CF. The antigens recognized by these T cells were mapped by the multi-elution technique IAndersen and Heron, 1993). This technique divides complex protein mixtures separated in SDS-PAGE into narrow fractions in a physiological buffer. These fractions were used to stimulate spleen lymphocytes in vi-tro and the release of IFN-y was monitored (Fig. 2). Long-term memory immune mice did not recognize these fractions before TB infection, but sptenic lymphocytes ob-tained during the recall of protective immunity recognized a range of culture filtrate antigens and peak production of IFN-y was found in response to proteins of apparent molecular weight 6-12 and 17-30 kDa (Fig. 2). It is therefore concluded that culture filtrate antigens within these regions are the major targets recognized by memory ef-fector T-cells triggered to release IFN-y during the first phase of a protective immune response.

Cloning of genes expressing low mass culture filtrate antigens In example 1 it was demonstrated that antigens in the low molecular mass fraction are recognized strongly by cells isolated from memory immune mice. Monoclonal anti-bodies (mAbs) to these antigens were therefore generated by immunizing with the low mass fraction in RIBI adjuvant (first and second immunization) followed by two injec-lions with the fractions in aluminium hydroxide. Fusion and cloning of the reactive cell lines were done according to standard procedures (Kohler and Milstein i 975).
The procedure resulted in the provision of two mAbs: ST-3 directed to a 9 kDa culture fil-trate antigen (CFP9) and PV-2 directed to a 7 kDa antigen (CFP7), when the molecular weight is estimated from migration of the antigens in an SDS-PAGE.
In order to identify the antigens binding to the Mab's, the following experiments were carried out:
The recombinant ~.gt1 1 M. tuberculosis DNA library constructed by R. Young (Young, R.A. et al. 1985? and obtained through the World Health Organization IMMTUB
pro-gramme (WH0.0032.wibr) was screened for phages expressing gene products which would bind the monoclonal antibodies ST-3 and PV-2.
Approximately 1 x 105 pfu of the gene library (containing approximately 25 %
recom-binant phages) were plated on Eschericia colt Y 1090 (DIacU 169,-proA+, Dlon, araD139, supF, trpC22::tn10 (pMC9) ATCC#37197) in soft agar and incubated for 2,5 hours at 42°C.
The plates were overlaid with sheets of nitrocellulose saturated with isopropyl-~3-D-thiogalactopyranoside and incubation was continued for 2,5 hours at 37°C. The nitro-cellulose was removed and incubated with samples of the monoclonal antibodies in PBS with Tween 20 added to a final concentration of 0.05%. Bound monoclonal anti-bodies were visualized by horseradish peroxidase-conjugated rabbit anti-mouse immu-noglobulins (P260, Dako, Glostrup, DK) and a staining reaction involving 5,5',3,3'-tetramethylbenzidine and HZOz.

Positive plaques were recfoned and the phages originating from a single plaque were used to lysogenize E. coli Y1089 (DIacU169, proA+, Dlon, araD139, strA, hf1150 [chr::tn 10] [pMC9] ATCC nr. 37196). The resultant lysogenic strains were used to propagate phage particles for DNA extraction. These lysogenic E. coli strains have 10 been named:
AA226 (expressing ST-3 reactive polypeptide CFP9) which has been deposited 28 June 1993 with the collection of Deutsche Sammlung von Mikroorganismen and Zell-kulturen GmbH (DSM) under the accession number DSM 8377 and in accordance with 15 the provisions of the Budapest Treaty, and AA242 (expressing PV-2 reactive polypeptide CFP7) which has been deposited 28 June 1993 with the collection of Deutsche Sammiung von Mikroorganismen and ZeN-kulturen GmbH (DSM) under the accession number DSM 8379 and in accordance with 20 the provisions of the Budapest Treaty.
These two lysogenic E. coli strains are disclosed in WO 95101441 as are the myco-bacterial polypeptide products expressed thereby. However, no information concerning the amino acid sequences of these polypeptides or their genetic origin are given, and 25 therefore only the direct expression products of AA226 and AA242 are made available to the public.
The st-3 binding protein is expressed as a protein fused to G3-galactosidase, whereas the pv-2 binding protein appears to be expressed in an unfused version.
Seauencing of the nucleotide se4uence encoding the PV-2 and ST-3 binding protein In order to obtain the nucleotide sequence of the gene encoding the pv-2 binding pro-tein, the approximately 3 kb M. tuberculosis derived EcoRl - EcoRl fragment from AA242 was subcloned in the EcoRl site in the pBluescriptSK + (Stratagene) and used to transform E. coli XL-1 Blue (Stratagene).
Similarly, to obtain the nucleotide sequence of the gene encoding the st-3 binding pro-tein, the approximately 5 kb M. tuberculosis derived EcoRl - EcoRl fragment from AA226 was subcloned in the EcoRl site in the pBluescriptSK + (Stratagene) and used to transform E. coli XL-1 Blue (Stratagene).
The complete DNA sequence of both genes were obtained by the dideoxy chain termi-nation method adapted for supercoiled DNA by use of the Sequenase DNA
sequencing kit version 1.0 (United States Biochemical Corp., Cleveland, OH) and by cycle se-quencing using the Dye Terminator system in combination with an automated gel reader (model 373A; Applied Biosystems) according to the instructions provided. The sequences DNA are shown in SEQ ID NO: 1 (CFP7) and in SEQ ID NO: 3 (CFP9) as well as in Figs. 3 and 4, respectively. Both strands of the DNA were sequenced.

An open reading frame (ORF) encoding a sequence of 96 amino acid residues was identified from an ATG start codon at position 91-93 extending to a TAG stop codon at position 379-381. The deduced amino acid sequence is shown in SEQ ID NO: 2 (and in Fig. 3 where conventional one-letter amino acid codes are used).
CFP7 appear to be expressed in E. coli as an unfused version. The nucleotide se-quence at position 78-84 is expected to be the Shine Delgarno sequence and the se-quences from position 47-50 and 14-19 are expected to be the -10 and -35 regions, respectively:

The protein recognised by ST-3 was produced as a G3-galactosidase fusion protein, when expressed from the AA226 lambda phage. The fusion protein had an approx.
size of 1 16 - 1 l7kDa (Mw for (3-galactosidase 116.25 kDa) which may suggest that only part of the CFP9 gene was included in the lambda clone (AA226).

Based on the 90 by nucleotide sequence obtained on the insert from lambda phage AA226, a search of homology to the nucleotide sequence of the M. tuberculosis genome was performed in the Sanger database (Sanger Mycobacterium tuberculosis database):
http:/lwww.sanger.ac.uk/pathogenslTB-blast-server.html;
Williams, 1996). 100% identity to the cloned sequence was found on the MTCY48 cosmid. An open reading frame (ORF) encoding a sequence of 109 amino acid resi-dues was identified from a GTG start codon at position 141 - 143 extending to a TGA
stop codon at position 465 - 467. The deduced amino acid sequence is shown in Fig.
4 using conventional one letter code.
The nucleotide sequence at position 123 - 130 is expected to be the Shine Delgarno sequence and the sequences from position 73 - 78 and 4 - 9 are expected to be the -10 and -35 region respectively (Fig. 4). The ORF overlapping with the 5'-end of the sequence of AA229 is shown in Fig. 4 by double underlining.
Subcloning CFP7 and CFP9 in expression vectors The two ORFs encoding CFP7 and CFP9 were PCR cloned into the pMST24 (Theisen et al., 1995) expression vector pRVN01 or the pQE-32 (QiAGEN) expression vector pRVN02, respectively.
The PCR amplification was carried out in a thermal reactor (Rapid cycler, Idaho Tech-nology, Idaho) by mixing 10 ng plasmid DNA with the mastermix (0.5 NM of each oli-gonucleotide primer, 0.25 uM BSA (Stratagene), low salt buffer (20 mM Tris-HCI, pH
8.8, 10 mM KCI, 10 mM (NH4)ZSO4, 2 mM MgSO, and 0,1 % Triton X-100) (Strata- ~
gene), 0.25 mM of each deoxynucleoside triphosphate and 0.5 U Taq Plus Long DNA
polymerase (Stratagene)). Final volume was 10 NI (all concentrations given are con-centrations in the final volume). ~Predenaturation was carried out at 94°C for 30 s. 30 cycles of the following was performed; Denaturation at 94°C for 30 s, annealing at 55°C for 30 s and elongation at 72°C for 1 min.

The oligonucleotide primers were synthesised automatically on a DNA
synthesizer (Applied Biosystems, Forster City, Ca, ABI-391, PCR-mode?, deblocked, and purified by ethanol precipitation.
The cfp7 oligonucleotides (TABLE 1 ) were synthesised on the basis ef the nucleotide sequence from the CFP7 sequence (Fig. 31. The oligonucleotides were engineered to include an Smal restriction enzyme site at the 5' end and a BamHl restriction enzyme site at the 3' end for directed subcloning.
The cfp9 oligonucleotides (TABLE 1 ) were synthesized partly on the basis of the nu-cleotide sequence from the sequence of the AA229 clone and partly from the identical sequence found in the Sanger database cosmid MTCY48 (Fig. 4). The oligonucleotides were engineered to include a Smal restriction enzyme site at the 5' end and a Hindlll restriction enzyme site at the 3' end for directed subcloning.

By the use of PCR a Smal site was engineered immediately 5' of the first codon of the ORF of 291 bp, encoding the cfp7 gene, so that only the coding region would be ex-pressed, and a BamHl site was incorporated right after the stop codon at the 3' end.
The 291 by PCR fragment was cleaved by Smal and BamHl, purified from an agarose gel and subcloned into the Smal - BamHl sites of the pMST24 expression vector.
Vec-tor DNA containing the gene fusion was used to transform the E. coli XL1-Blue (pRVN01 ).

By the use of PCR a Smal site was engineered immediately 5' of the first codon of an ORF of 327 bp, encoding the cfp9 gene, so that only the coding region would be ex-pressed, and a Hindlll site was incorporated after the stop codon at the 3' end. The 327 by PCR fragment was cleaved by Smal and Hindlll, purified from an agarose gel, and subcloned into the Smai - Hindlll sites of the pQE-32 (QIAGEN) expression vector.

Vector DNA containing the gene fusion was used to transform the E. coli XL1-Blue (pRVN02).
Purification of recombinant CFP7 and CFP9 The ORFs were fused N-terminally to the (His)6-tag (cf. EP-A-0 282- 242).
Recombi-nant antigen was prepared as fotlows: Briefly, a single colony of E. coli harbouring either the pRVN01 or the pRVN02 plasmid, was inoculated into Luria-Bertani broth containing 100 Ng/ml ampicillin and 12.5 ug/ml tetracycline and grown at 37°C to OD6~°m = 0.5. IPTG (isopropyl-[3-D-thiogalactoside) was then added to a final concen-tration of 2 mM (expression was regulated either by the strong IPTG inducible Pta~ or the T5 promoter) and growth was continued for further 2 hours. The cells were har-vested by centrifugation at 4,200 x g at 4°C for 8 min. The pelleted bacteria were stored overnight at -20°C. The pellet was resuspended in BC 40/100 buffer (20 mM
Tris-HCI pH 7.9, 20% glycerol, 100 mM KCI, 40 mM lmidazole) and cells were broken by sonication (5 times for 30 s with intervals of 30 s) at 4°C.
followed by cenirifuga-tion at 12,000 x g for 30 min at 4°C, the supernatant (crude extract) was used for purification of the recombinant antigens.
The two Histidine fusion proteins (His-rCFP7 and His-rCFP9) were purified from the crude extract by affinity chromatography on a Ni2r-NTA column from QIAGEN with a volume of 100 ml. His-rCFP7 and His-rCFP9 binds to Ni2+. After extensive washes of the column in BC 401100 buffer, the fusion protein was eluted with a BC

buffer containing 100 mM imidazole, 20 mM Tris pH 7.9, 20% glycerol and 1 M
KCI.
subsequently, the purified products were dialysed extensively against 10 mM
Tris pH
8Ø His-rCFP7 and His-rCFP9 were then separated from contaminants by fast protein liquid chromatography (FPLC) over an anion-exchange column (Mono Q, Pharmacia, Sweden). in 10 mM Tris pH 8.0 with a linear gradient of NaCI from 0 to 1 M.
Aliquots of the fractions were analyzed by 10%-20% gradient sodium dodecyl sulphate poly-acrylamide gel electrophoresis (SDS-PAGE). Fractions containing purified either purified His-rCFP7 or His-rCFP9 were pooled.

TABLE 1. Sequence of the cfp7 and cfp9 oligonucleotidesa.
Orientation and Sequences (5' -~ 3') Position° (nucleo-oligonucleotide tide) Sense pvR3 GCAACACCCGGGATGTCGCAAATCATG 91-105 (SEQ ID NO: 43) (SEQ ID NO: 1 ) stR2 GTAACACCCGGGGTGGCCGCCGACCCG 141-155 (SEQ ID NO: 44) (SEQ ID NO: 3) Antisense pvF4 CTACTAAGCTTGGATCCCTAGCCG- 381-362 CCCCATTTGGCGG (SEQ ID NO: 1 ) (SEQ ID NO: 45) stF2 CTACTAAGCTTCCATGGTCAGGTC- 467 - 447 TTTTCGATGCTTAC (SEQ ID NO: 3) (SEQ 1D NO: 46) a The cfp7 oligonucfeotides were based on the nucleotide sequence shown in Fig. 3 (SEQ ID NO: 1 ). The cfp9 oligonucleotides were based on the nucleotide sequence 5 shown in Fig. 4 (SEQ ID NO: 3).
Nucleotides underlined are not contained in the nucleotide sequence of cfp7 and cfp9.
The positions referred to are of the non-underlined part of the primers and corre-spond to the nucleotide sequence shown in Fig. 3 and Fig. 4, respectively.

Identification of antigens which are not expressed in BCG strains.
In an effort to control the treat of TB, attenuated bacillus Calmette-Guerin (BCG) has 15 been used as a live attenuated vaccine. BCG is an attenuated derivative of a virulent Mycobacterium bovis. The original BCG from the Pasteur Institute in Paris, France was developed from 1908 to 1921 by 231 passages in liquid culture and has never been shown to revert to virulence in animals, indicating that the attenuating mutations) in BCG are stable deletions and/or multiple mutations which do not readily revert. While 20 physiological differences between BCG and M. tuberculosis and M, bovis has been noted, the attenuating mutations which arose during serial passage of the original BCG
strain has been unknown until recently. The first mutations described are the loss of the gene encoding MPB64 in some BCG strains (Li et al., 1993, Oettinger and Ander-sen, 1994) and the gene encoding ESAT-6 in all BCG strain tested (Harboe et al., 1996), later 3 large deletions in BCG have been identified (Mahairas et al., 1996). The region named RD 1 includes the gene encoding ESAT-6 and an otheF (RD2) the gene encoding MPT64. Both antigens have been shown to have diagnostic potential and ESAT-6 has been shown to have properties as a vaccine candidate (cf.
PCT/DK94100273 and PCT/DK100270). In order to find new M. tuberculosis specific diagnostic antigens as well as antigens for a new vaccine against TB, the RD 1 region (17.499 bp) of M. tuberculosis H37Rv has been analyzed for Open Reading Frames (ORF). ORFs with a minimum length of 96 by have been predicted using the algorithm described by Borodovsky and Mclninch (1993), in total 27 ORFs have been predicted, of these have possible diagnostic and/or vaccine potential, as they are deleted 15 from all known BCG strains. The predicted ORFs include ESAT-6 (RD1-ORF7) and CFP10 (RD1-ORF6) described previously (Srarensen et al., 1995), as a positive control for the ability of the algorithm. In the present is described the potential of 7 of the predicted antigens for diagnosis of TB as well as potential as candidates for a new vaccine against TB.
Seven open reading frames (ORF) from the 17,499kb RD1 region (Accession no.
034848) with possible diagnostic and vaccine potential. have been identified and cloned.
Identification of the ORF's rdl-orf2, rd 1-orf3 rd 1-orf4 rd 1-orf5 rd 1-orf8 rd 1-orf9a, and rdl-orf9b.
The nucleotide sequence of rd 1-orf2 from M. tuberculosis H37Rv is set forth in SEQ
ID NO: 71. The deduced amino acid sequence of RD1-ORF2 is set forth in SEQ ID
NO:
72.
The nucleotide sequence of rd 1-orf3 from M. tuberculosis H37Rv is set forth in SEQ
iD NO: 87. The deduced amino acid sequence of RD1-ORF2 is set forth in SEQ 1D
NO:
88.

The nucleotide sequence of rd 1-orf4 from M. tuberculosis H37Rv is set forth in SEQ
ID NO: 89. The deduced amino acid sequence of RD1-ORF2 is set forth in SEQ ID
NO:
90.
The nucleotide sequence of rd 1-orf5 from M. tuberculosis H37Rv is set forth in SEQ
ID NO: 91. The deduced amino acid sequence of RD1-ORF2 is set forth in SEQ ID
NO:
92.
The nucleotide sequence of rd 1-orf8 from M. tuberculosis H37Rv is set forth in SEQ
ID NO: 67. The deduced amino acid sequence of RD1-ORF2 is set forth in SEQ ID
NO:
68.
The nucleotide sequence of rd 1-orf9a from M. tuberculosis H37Rv is set forth in SEQ
ID NO: 93. The deduced amino acid sequence of RD1-ORF2 is set forth in SEQ ID
NO:
94.
The nucleotide sequence of rd 1-orf9b from M. tuberculosis H37Rv is set forth in SEQ
ID NO: 69. The deduced amino acid sequence of RD1-ORF2 is set forth in SEQ ID
NO:
70.
The DNA sequence rd 1-orf2 (SEQ ID NO: 71 ) contained an open reading frame start-ing with an ATG codon at position 889 - 891 and ending with a termination codon (TAA) at position 2662 - 2664 (position numbers referring to the location in RD 1 ). The deduced amino acid sequence (SEQ ID NO: 72) contains 591 residues corresponding to a molecular weight of 64,525.
The DNA sequence rdl-orf3 (SEQ 1D NO: 87) contained an open reading frame start-ing with an ATG codon at position 2807 - 2809 and ending with a termination codon (TAA) at position 3101 - 3103 (position numbers referring to the location in RD1 ). The deduced amino acid sequence (SEQ ID NO: 88) contains 98 residues corresponding to a molecular weight of 9,799.

The DNA sequence rdl-orf4 (SEQ ID NO: 89) contained an open reading frame start-ing with a GTG codon at position 4014 - 4012 and ending with a termination codon (TAG) at position 3597 - 3595 lposition numbers referring to the location in RD 1 ). The deduced amino acid sequence (SEQ ID NO: 90) contains 139 residues corresponding -to a molecular weight of 14,210.
The DNA sequence rdl-orf5 (SEQ ID NO: 91) contained an open reading frame start-ing with a GTG codon at position 3128 - 3130 and ending with a termination codon (TGA) at position 4241 - 4243 (position numbers referring to the location in RD1 ). The deduced amino acid sequence (SEQ ID N0: 92) contains 371 residues corresponding to a molecular weight of 37,647.
The DNA sequence rdl-orf8 (SEQ ID NO: 67) contained an open reading frame start-ing with a GTG codon at positron 5502 - 5500 and ending with a termination codon ~ (TAG) at position 5084 - 5082 (position numbers referring to the location in RD1 ), and the deduced amino acid sequence (SEQ ID NO: 68) contains 139 residues with a mo-lecular weight of 1 1,737.
The DNA sequence rdl-orf9a (SEQ ID NO: 93) contained an open reading frame starting with a GTG codon at position 6146 - 6148 and ending with a termination co-don (TAA) at position 7070 - 7072 (position numbers referring to the location in RD1 ).
The deduced amino acid sequence (SEQ 1D NO: 94) contains 308 residues corre-sponding to a molecular weight of 33,453.
The DNA sequence rdl-orf9b (SEQ ID NO: 69) contained an open reading frame starting with an ATG codon at position 5072 - 5074 and ending with a termination codon (TAA) at position 7070 - 7072 (position numbers referring to the location in RD1). The deduced amino acid sequence (SEQ ID NO: 70) contains 666 residues cor-responding to a molecular weight of 70,650.

Cloning of the ORF's rd1-orf2, rdl-orf3. rdl-orf4, rdl-orf5. rd1-orf8, rdl-orf9a, and rd 1-orf96.
The O R F's rd 1-orf2, rd 1-orf3, rd 1-orf4, rd 1-orf5, rd 1-orf8, rd 1-orf9a and rd 1-orf9b were PCR cloned in the pMST24 (Theisen et al., 1995) (rdl-orf3) or the pQE32 (QIAGEN) (rd l -orf2, rd l-orf4, rd l-orf5, rd l-orf8, rd 1-orf9a and rd l orf96) expression vector. Preparation of oligonucleotides and PCR amplification of the rd1-orf encoding genes, was carried out as described in example 2. Chromosomal DNA from M.
tuber culosis H37Rv was used as template in the PCR reactions. Oligonucleotides were syn-thesized on the basis of the nucleotide sequence from the RD1 region (Accession no.
U34848). The oligonucleotide primers were engineered to include an restriction en-zyme site at the 5' end and at the 3' end by which a lafer subcloning was possible.
Primers are listed in TABLE 2.
rdl-orf2. A BamHl site was engineered immediately 5' of the first codon of rdl-orf2, and a Hindlll site was incorporated right after the stop codon at the 3' end.
The gene rdl-orf2 was subcloned in pQE32, giving pT096.
rd 1-orf3. A Smal site was engineered immediately 5' of the first codon of rd 1-orf3, and a Ncol site was incorporated right after the stop codon at the 3' end. The gene rd1-orf3 was subcloned in pMST24, giving pT087.
rd 1-orf4. A BamHl site was engineered immediately 5' of the first codon of rd 1-orf4, and a Hindlll site was incorporated right after the stop codon at the 3' end.
The gene rd 1-orf4 was subcloned in pQE32, giving pT089.
rd1-orf5. A BamHl site was engineered immediately 5' of the first codon of rdl-orf5, and a Hindllt site was incorporated right after the stop codon at the 3' end.
The gene rd 1-orf5 was subcloned in pQE32, giving pT088.
_ rdl-orf8. A BamHl site was engineered immediately 5' of the first codon of rd1-orf8, and a Ncol site was incorporated right after the stop codon at the 3' end. The gene rd1-orf8 was subcloned in pMST24, giving pT098.

rdl-orf9a. A BamHl site was engineered immediately 5' of the first codon of rd1-orf9a, and a Hindlll site was incorporated right after the stop codon at the 3' end. The gene rd 1-orf9a was subcloned in pQE32, giving pT091.
5 rd1-orf96. A Scal site was engineered immediately 5' of the first codon of rdl-orf9b, and a Hind llt site was incorporated right after the stop codon at the 3' end.
The gene rdl-orf96 was subcloned in pQE32, giving pT090.
The PCR fragments were digested with the suitable restriction enzymes, purified from 10 an agarose gel and cloned into either pMST24 or pQE-32. The seven constructs were used to transform the E. coli XL1-Blue. Endpoints of the gene fusions were determined by the dideoxy chain termination method. Both strands of the DNA were sequenced.
Purification of recombinant RD1-ORF2, RD1-ORF3, RD1-ORF4, RD1-ORF5, RD1-ORFB, 15 RD1-ORF9a and RD1-ORF9b.
The rRD1-ORFs were fused N-terminally to the (His)6 -tag. Recombinant antigen was prepared as described in example 2 (with the exception that pT091 was expressed at 30°C and not at 37°C?, using a single colony of E. coli harbouring either the pT087, 20 pT088, pT089, pT090, pT091, pT096 or pT098 for inoculation. Purification of re-combinant antigen by Niz+ affinity chromatography was also carried out as described in example 2. Fractions containing purified His-rRD1-ORF2, His-rRD1-ORF3 His-rRD1-ORF4, His-rRD1-ORFS, His-rRD1-ORFB, His-rRD1-ORF9a or His-rRD1-ORF9b were pooled. The His-rRD1-ORF's were extensively dialysed against 10 mM Tris/HCI, pH
25 8.5, 3 M urea followed by an additional purification step performed on an anion ex-change column (Mono Q) using fast protein liquid chromatography (FPLC) (Pharmacia, Uppsala, Sweden). The purification was carried out in 10 mM Tris/HCI, pH 8.5, urea and protein was eluted by a linear gradient of NaCI from 0 to 1 M.
Fractions con-taining the His-rRD1-ORF's were pooled and subsequently dialysed extensively against 30 25 mM Hepes, pH 8.0 before use.

Table 2. Sequence of the rd l-orf's oiigonucteotidesa.
Orientation and Sequences (5'--> 3') Position (nt) oligo-nucleotide Sense RD1-ORF2f CTGGGGATCCGCATGACTGCTGAACCG -886 - 903 RD1-ORF3f CTTCCCGGGATGGAAAAAATGTCAC 2807 - 2822 RD1-ORF4f GTAGGATCCTAGGAGACATCAGCGGC 4028 - 4015 RD1-ORFSf CTGGGGATCCGCGTGATCACCAT- 3028 - 3045 GCTGTGG

RD 1-ORF8f CTCGGATCCTGTGGGTGCAGGTCCGGC 5502 - 5479 GATGGGC

RD1-ORF9af GTGATGTGAGCTCAGGTGAAGAA- 6144 - 6160 GGTGAAG

RD1-ORF9bf GTGATGTGAGCTCCTATGGCGGCCGAC- 5072 - 5089 TACGAC

Antisense RD1-ORF2r TGCAAGCTTTTAACCGGCGCTTGGGGGT 2664 - 2644 GC

RD1-ORF3r GATGCCATGGTTAGGCGAAGACGC- 3103 - 3086 CGGC

RD1-ORF4r CGATCTAAGCTTGGCAATGGAGGTCTA 3582 - 3597 RD1-ORFSr TGCAAGCTTTCACCAGTCGTCCT- 4243 - 4223 CTTCGTC

RD1-ORF8r CTCCCATGGCTACGACAAGCTCTTC- 5083 - 5105 CGGCCGC

RD1-ORF9a/br CGATCTAAGCTTTCAACGACGTCCAGCC 7073 - 7056 a The oligonucteotides were constructed from the Accession number U34484 nucleo-tide sequence (Mahairas et al., 1996). Nucleotides (nt) underlined are not contained in the nucleotide sequence of RD1-ORF's. The positions correspond to the nucleotide se-quence of Accession number U34484.

The nucleotide sequences of rd 1-orf2, rd 1-orf3, rd 1-orf4, rd 1-orf5, rd 1-orf8, rd 1-orf9a, and rdl-orf96 from M. tuberculosis H37Rv are set forth in SEQ ID NO:
71, 87, 89, 91, 67, 93, and 69, respectively. The deduced amino acid sequences of rd 1-orf2, rd 1-orf3, rd 1-orf4 rd 1-orf5, rd 1-orf8, rd 1-orf9a, and rd 1-orf96 are set forth in SEQ ID
NO: 72, 88, 90, 92, 68, 94, and 70, respectively.

Cloning of the genes expressing 17-30 kDa antigens from ST CF
Isolation of CFP17, CFP20, CFP21, CFP22, CFP25, and CFP28 ST-CF was precipitated with ammonium sulphate at 80% saturation. The precipitated proteins were removed by centrifugation and after resuspension washed with 8 M
urea. CHAPS and glycerol were added to a final concentration of 0.5% (w/v) and 5%
(v/v) respectively and the protein solution was applied to a Rotofor isoelectrical Cell (BioRad). The Rotofor Cell had been equilibrated with an 8 M urea buffer containing 0.5% (w/v) CHAPS, 5% (v/v) glycerol, 3% (v/v) Biolyt 3/5 and 1 % (v/v) Biolyt (BioRad). Isoelectric focusing was performed in a pH gradient from 3-6. The fractions were analyzed on silver-stained 10-20% SDS-PAGE. Fractions with similar band pat-terns were pooled and washed three times with PBS on a Centriprep concentrator (Amicon) with a 3 kDa cut off membrane to a final volume of i-3 ml. An equal volume of SDS containing sample buffer was added and the protein solution boiled for 5 min before further separation on a Prep Cell (BioRad) in a matrix of 16%
polyacrylamide under an electrical gradient. Fractions containing pure proteins with an molecular mass from 17-30 kDa were collected.
Isolation of CFP29 Anti-CFP29, reacting with CFP29 was generated by immunization of BALB/c mice with crushed gel pieces in RIBI adjuvant (first and second immunization) or aluminium hydroxide (third immunization and boosting) with two week intervals. SDS-PAGE
gel pieces containing 2-5 Ng of CFP29 were used for each immunization. Mice were boosted with antigen 3 days before removal of the spleen. Generation of a monoclonal WO 99!24577 PCT/DK98100438 cell line producing antibodies against CFP29 was obtained essentially as described by Kohler and Milstein (1975). Screening of supernatants from growing clones was car-ried out by immunoblotting of nitrocellulose strips containing ST-CF separated by SDS-PAGE. Each strip contained approximately 50 Ng of ST-CF. The antibody class of anti-s CFP29 was identified as IgM by the mouse monoclonal antibody isotyping kit, (Amersham) according to the manufacturer's instructions. -CFP29 was purified by the following method: ST-CF was concentrated 10 fold by ul-trafiltration, and ammonium sulphate precipitation in the 45 to 55% saturation range was performed. The pellet was redissolved in 50 mM sodium phosphate, 1.5 M am-monium sulphate, pH 8.5, and subjected to thiophilic adsorption chromatography (Po-rath et al., 1985) on an Affi-T gel column (Kem-En-Tec). Protein was eluted by a linear 1.5 to 0 M gradient of ammonium sulphate and fractions collected in the range 0.44 to 0.31 M ammonium sulphate were identified as CFP29 containing fractions in West-ern blot experiments with mAb Anti-CFP29. These fractions were pooled and anion exchange chromatography was performed on a Mono Q HR 5/5 column connected to an FPLC system (Pharmacia). The column was equilibrated with 10 mM Tris-HCI, pH
8.5 and the elution was performed with a linear gradient from 0 to 500 mM
NaCI.
From 400 to 500 mM sodium chloride, rather pure CFP29 was eluted. As a final puri-fication step the Mono Q fractions containing CFP29 were loaded on a 12.5% SDS-PAGE gel and pure CFP29 was obtained by the multi-elution technique (Andersen and Heron, 1993).
N-terminal seguencing and amino acid analysis CFP17, CFP20, CFP21, CFP22, CFP25, and CFP28 were washed with water on a Centricon concentrator (Amicon) with cutoff at 10 kDa and then applied to a ProSpin concentrator (Applied Biosystems) where the proteins were collected on a PVDF
mem-brane. The membrane was washed 5 times with 20% methanol before sequencing on a Procise sequencer (Applied Biosystems).
CFP29 containing fractions were blotted to PVDF membrane after tricine SDS-PAGE
(Ploug et al., 19891. The i-elevant bands were excised and subjected to amino acid analysis (Barkholt and Jensen, 1989) and N-terminal sequence analysis on a Procise sequencer (Applied Biosystems).
The following N-terminal sequences were obtained:
ForCFPI7:A/SELDAPAQAGTEXAV (SEQIDN0:17) ForCFP20:AQITLRGNAINTVGE (SEQIDN0:18) ForCFP2I:DPXSDIAVVFARGTH (SEQIDN0:19) ForCFP22:TNSPLATATATLHTN (SEQIDN0:20) ForCFP25:AXPDAEV VFARGRFE (SEQIDN0:21) For CFP28: X I/V Q K S L E L I V/T V/F T A D/Q E (SEQ ID NO: 22) ForCFP29:MN NLYRDLAPVTEA AWAEI (SEQIDN0:23) "X" denotes an amino acid which could not be determined by the sequencing method used, whereas a "/" between two amino acids denotes that the sequencing method could not determine which of the two amino acids is the one actually present.
Cloning the gene encoding CFP29 The N-terminal sequence of CFP29 was used for a homology search in the EMBL da-tabase using the TFASTA program of the Genetics Computer Group sequence analysis software package. The search identified a protein, Linocin M 18, from Brevibacterium linens that shares 74% identity with the 19 N-terminal amino acids of CFP29.
Based on this identity between the N-terminal sequence of CFP29 and the sequence of the Linocin M 18 protein from Brevibacterium linens, a set of degenerated primers were constructed for PCR cloning of the M. tuberculosis gene encoding CFP29.
PCR
reactions were containing 10 ng of M. tuberculosis chromosomal DNA in 1 x low salt Taq + buffer from Stratagene supplemented with 250 ~M of each of the four nucleo- -tides (Boehringer Mannheim), 0,5 mg/ml BSA (IgG technologyl. 1 % DMSO (Merck), pmoles of each primer and 0.5 unit Tag+ DNA polymerase (Stratagene) in 10 ~I
reac- ' tion volume. Reactions were initially heated to 94°C for 25 sec. and run for 30 cycles of the program; 94°C for 15 sec., 55°C for 15 sec. and 72°C for 90 sec, using ther-mocycler equipment from Idaho Technology.

An approx. 300 by fragment was obtained using primers with the sequences:
1: 5'-CCCGGCTCGAGAACCTSTACCGCGACCTSGCSCC (SEQ ID NO: 24) 5 2:5'-GGGCCGGATCCGASGCSGCGTCCTTSACSGGYTGCCA
(SEQ-ID NO: 25) -where S = GlC and Y = T/C
The fragment was excised from a 1 % agarose gel, purified by Spin-X spine columns 10 (Costar), cloned into pBluescript SK II + - T vector (Stratagene) and finally sequenced with the Sequenase kit from United States Biochemical.
The first 150 by of this sequence was used for a homology search using the Blast program of the Sanger Mycobacterium tuberculosis database:
(http//www.sanger.ac.uk/projects/M-tuberculosis/blast_server).
This program identified a Mycobacterium tuberculosis sequence on cosmid cy444 in the database that is nearly 100% identical to the 150 by sequence of the CFP29 pro-tein. The sequence is contained within a 795 by open reading frame of which the 5' end translates into a sequence that is 100% identical to the N-terminally sequenced 19 amino acids of the purified CFP29 protein.
Finally, the 795 by open reading frame was PCR cloned under the same PCR condi-tions as described above using the primers:
3: 5'-GGAAGCCCCATATGAACAATCTCTACCG (SEQ ID NO: 26) 4: 5'-CGCGCTCAGCCCTTAGTGACTGAGCGCGACCG (SEQ ID NO: 27) The resulting DNA fragments were purified from agarose gels as described above se-quenced with primer 3 and 4 in addition to the following primers:
5: 5'-GGACGTTCAAGCGACACATCGCCG-3' (SEQ ID NO: 115) 6: 5'-CAGCACGAACGCGCCGTCGATGGC-3' (SEQ ID NO: 1 16) Three independent cloned were sequenced. All three clones were in 100%
agreement with the sequence on cosmid cy444.
All other DNA manipulations were done according to Maniatis et al. E7 989).
All enzymes other than Taq polymerase were from New England Biolabs.
Homology searches in the Sanger database For CFP17, CFP20, CFP21, CFP22, CFP25, and CFP28 the N-terminal amino acid se-quence from each of the proteins were used for a homology search using the blast program of the Sanger Mycobacterium tuberculosis database:
http://www.sanger.ac.uk/pathogenslTB-blast-server.html.
For CFP29 the first 150 by of the DNA sequence was used for the search.
Further-more, the EMBL database was searched for proteins with homology to CFP29.
Thereby, the following information were obtained:

Of the 14 determined amino acids in CFP17 a 93% identical sequence was found with MTCY 1 A 1 1.16c. The difference between the two sequences is in the first amino acid:
It is an A or an S in the N-terminal determined sequenced and a S in MTCY1A1 1.
From the N-terminal sequencing it was not possible to determine amino acid number 13.
Within the open reading frame the translated protein is 162 amino acids long.
The N-terminal of the protein purified from culture filtrate starts at amino acid 31 in agree-meat with the presence of a signal sequence that has been cleaved off. This gives a length of the mature protein of 132 amino acids, which corresponds to a theoretical molecular mass of 13833 Da and a theoretical pl of 4.4. The observed mass in SDS-PAGE is 17 kDa.

A sequence 100% identical to the 15 determined amino acids of CFP20 was found on the translated cosmid cscy09F9. A stop codon is found at amino acid 166 from the amino acid M at position 1. This gives a predicted length of 165 amino acids, which corresponds to a theoretical molecular mass of 16897 Da and a pl of 4.2. The ob-served molecular weight in a SDS-PAGE is 20 kDa.
Searching the GenEMBL database using the TFASTA algorithm (Pearson and Lipman, 1988) revealed a number of proteins with homology to the predicted 164 amino acids tong translated protein.
The highest homology, 51.5% identity in a 163 amino acid overlap, was found to a Haemophilus influenza Rd toxR reg. (HIH10751 ).

A sequence 100% identical to the 14 determined amino acids of CFP21 was found at MTCY39. From the N-terminal sequencing it was not possible to determine amino acid number 3; this amino acid is a C in MTCY39. The amino acid C can not be detected on a Sequencer which is probably the explanation of this difference.
Within the open reading frame the translated protein is 217 amino acids long.
The N-terminally determined sequence from the protein purified from culture filtrate starts at amino acid 33 in agreement with the presence of a signal sequence that has been cleaved off. This gives a length of the mature protein of 185 amino acids, which cor-responds to a theoretical molecular weigh at 18657 Da, and a theoretical pl at 4,6.
The observed weight in a SDS-PAGE is 21 kDa.
In a 193 amino acids overlap the protein has 32,6% identity to a cutinase precursor with a length of 209 amino acids (CUTI ALTBR P41744).

A comparison of the 14 N-terminal determined amino acids with the translated region (RD2) deleted in M. bovis BCG revealed a 100% identical sequence (mb3484) (Mahai-ras et al. (1996)).

A sequence 100% identical to the 15 determined amino acids of CFP22 was found at MTCY10H4. Within the open reading frame the translated protein is 182 amino acids long. The N-terminal sequence of the protein purified from culture filtrate starts at amino acid 8 and therefore the length of the protein occurring in M.
tuberculosis cul-ture filtrate is 175 amino acids. This gives a theoretical molecular weigh at 18517 Da and a pl at 6.8. The observed weight in a SDS-PAGE is 22 kDa.
In an 182 amino acids overlap the translated protein has 90,1 % identity with E235739; a peptidyl-prolyl cis-trans isomerase.

A sequence 93% identical to the 15 determined amino acids was found on the cosmid MTCY339.08c. The one amino acid that differs between the two sequences is a C
in MTCY339.08c and a X from the N-terminal sequence data. On a Sequencer a C can not be detected which is a probable explanation for this difference.
The N-terminally determined sequence from the protein purified from culture filtrate begins at amino acid 33 in agreement with the presence of a signal sequence that has been cleaved off. This gives a length of the mature protein of 187 amino acids, which corresponds to a theoretical molecular weigh at 19665 Da, and a theoretical pl at 4.9.
The observed weight in a SDS-PAGE is 25 kDa.
In a 217 amino acids overlap the protein has 42.9% identity to CFP21 (MTCY39.35).

No homology was found when using the 10 determined amino acid residues 2-8, 1 1, 12, and 14 of SEQ ID NO: 22 in the database search.

Sanger database searching: A sequence nearly 100% identical to the 150 by sequen-ce of the CFP29 protein was found on cosmid cy444. The sequence is contained within a 795 by open reading frame of which the 5' end translates into a sequence that is 100% identical to the N-terminally sequenced 19 amino acids of the purified CFP29 protein. The open reading frame encodes a 265 amino acid protein.
The amino acid analysis performed on the purified protein further confirmed the iden-tity of CFP29 with the protein encoded in open reading frame on cosmid 444.
EMBL database searching: The open reading frame encodes a 265 amino acid protein that is 58% identical and 74% similar to the Linocin M18 protein (61 %
identity on DNA levely. This is a 28.6 kDa protein with bacteriocin activity (Valdes-Stauber and Scherer, 1994; Valdes-Stauber and Scherer, 1996). The two proteins have the same length (except far 1 amino acid) and share the same theoretical physicochemical pro-perties. We therefore suggest that CFP29 is a mycobacterial homolog to the Brevibac-terium linens Linocin M 18 protein.
The amino acid sequences of the purified antigens as picked from the Sanger database are shown in the following list. The amino acids determined by N-terminal sequencing are marked with bold.
CFP17 (SEQ ID NO: 6):

CFP20 (SEQ ID NO: 8):

CFP21 (SEQ ID NO: 10):

CFP22 (SEQ ID NO: 12):
1 MADCDSVTNS PLATATATLH TNRGDIKiAL FGNHAPKTVA NFVGLAQGTK

CFP25 (SEQ ID NO: 14):

CFP29 (SEQ ID NO: 16):

For all six proteins the molecular weights predicted from the sequences are in agree-ment with the molecular weights observed on SDS-PAGE.
Cloning of the genes encoding CFP17, CFP20 CFP21 CFP22 and CFP25 The genes encoding CFP17, CFP20, CFP21, CFP22 and CFP25 were all cloned into the expression vector pMCT6, by PCR amplification with gene specific primers, for recombinant expression in E. coli of the proteins.
PCR reactions contained 10 ng of M. tuberculosis chromosomal DNA in 1 x low salt Taq + buffer from Stratagene supplemented with 250 mM of each of the four nucleo-tides (Boehringer Mannheim), 0,5 mg/ml BSA (IgG technology), 1 % DMSO (Merck), pmoles of each primer and 0.5 unit Tag+ DNA polymerise (Stratagene) in 10 pl reac-Lion volume. Reactions were initially heated to 94°C for 25 sec. and run for 30 cycles according to the following program; 94°C for 10 sec., 55°C for 10 sec. and 72°C for 90 sec, using thermocycler equipment from Idaho Technology.
The DNA fragments were subsequently run on 1 % agarose gels, the bands were ex-cised and purified by Spin-X spin columns (Costar) and cloned into pBluescript SK 11 +
- T vector (Stratagene). Plasmid DNA was thereafter prepared from clones harbouring the desired fragments, digested with suitable restriction enzymes and subcloned into the expression vector pMCT6 in frame with 8 histidine residues which are added to the N-terminal of the expressed proteins. The resulting clones were hereafter sequen-ced by use of the dideoxy chain termination method adapted for supercoiled DNA
us-ing the Sequenase DNA sequencing kit version 1.0 (United States Biochemical Corp., USA) and by cycle sequencing using the Dye Terminator system in combination with an automated gel reader (model 373A; Applied Biosystems) according to the instruc-tions provided. Both strands of the DNA were sequenced.

s2 For cloning of the individual antigens, the following gene specific primers were used:
CFP17: Primers used for cloning of cfp17:
OPBR-51: ACAGATCTGTGACGGACATGAACCCG (SEQ ID NO: 1 17) OPBR-52: TTTTCCATGGTCACGGGCCCCCGGTACT (SEQ ID NO: 118) OPBR-51 and OPBR-52 create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.
CFP20: Primers used for cloning of cfp20:
OPBR-53: ACAGATCTGTGCCCATGGCACAGATA (SEQ ID NO: 119) OPBR-54: TTTAAGCTTCTAGGCGCCCAGCGCGGC (SEQ ID NO: 120) OPBR-53 and OPBR-54 create Bglll and HinDlll sites, respectively, used for the cloning in pMCT6.
CFP21: Primers used for cloning of cfp21:
OPBR-55: ACAGATCTGCGCATGCGGATCCGTGT (SEQ ID NO: 121 ) OPBR-56: TTTTCCATGGTCATCCGGCGTGATCGAG (SEQ ID NO: 122) OPBR-55 and OPBR-56 create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.
CFP22: Primers used for cloning of cfp22:
OPBR-57. ACAGATCTGTAATGGCAGACTGTGAT (SEQ ID NO: 123) OPBR-58: TTTTCCATGGTCAGGAGATGGTGATCGA (SEQ ID NO: 124) -OPBR-57 and OPBR-58 create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.

WO 99/245?? PCTlDK98/00438 CFP25: Primers used for cloning of cfp25:
OPBR-59: ACAGATCTGCCGGCTACCCCGGTGCC (SEQ iD NO: 125) OPBR-60: TTTTCCATGGCTATTGCAGCTTTCCGGC (SEQ ID NO: 126) OPBR-59 and OPBR-60 create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.
Expression/purification of recombinant CFP17, CFP20, CFP21, CFP22 and CFP25 pro-teins.
Expression and metal affinity purification of recombinant proteins was undertaken es-sentially as described by the manufacturers. For each protein, 1 I LB-media containing 100 Ng/ml ampicillin, was inoculated with 10 ml of an overnight culture of XL1-Blue cells harbouring recombinant pMCT6 plasmids. Cultures were shaken at 37 °C until they reached a density of ODsoo = 0.4 - 0.6. IPTG was hereafter added to a final con-centration of 1 mM and the cultures were further incubated 4 - 16 hours. Cells were harvested, resuspended in 1 X sonication buffer + 8 M urea and sonicated 5 X

sec. with 30 sec. pausing between the pulses.
After centrifugation, the lysate was applied to a column containing 25 ml of resus-pended Talon resin (Clontech, Palo Alto, USA). The column was washed and eluted as described by the manufacturers.
After elution, all fractions (1.5 ml each) were subjected to analysis by SDS-PAGE us-ing the Mighty Small (Hoefer Scientific Instruments, USA) system and the protein con-centrations were estimated at 280 nm. Fractions containing recombinant protein were pooled and dialysed against 3 M urea in 10 mM Tris-HCI, pH 8.5. The dialysed protein was further purified by FPLC (Pharmacia, Sweden) using a 6 ml Resource-Q
column, eluted with a linear 0-1 M gradient of NaCI. Fractions were analyzed by SDS-PAGE
and protein concentrations were estimated at ODZ8o. Fractions containing protein were pooled and dialysed against 25 mM Hepes buffer, pH 8.5.

Finally the protein concentration and the LPS content were determined by the BCA
(Pierce, Holland) and LAL (Endosafe, Charleston, USA) tests, respectively.

Identification of CFP7A, CFPBA, CFPBB, CFP 16, CFP 19, CFP 19B, CFP22A, CFP23A, CFP23B, CFP25A, CFP27, CFP30A, CWP32 and CFP50.
Identification of CFP16 and CFP19B.
ST-CF was precipitated with ammonium sulphate at 80% saturation. The precipitated proteins were removed by centrifugation and after resuspension washed with 8 M
urea. CHAPS and glycerol were added to a final concentration of 0.5 % (w/v) and 5 (v/v) respectively and the protein solution was applied to a Rotofor isoelectrical Cell (BioRad). The Rotofor Cell had been equilibrated with a 8M urea buffer containing 0.5 (w/v) CHAPS, 5% (v/v) glycerol, 3% (v/v) Biolyt 3/5 and 1 % (vlv) Biolyt 4/6 (Bio-Rad). Isoelectric focusing was performed in a pH gradient from 3-6. The fractions were analyzed on silver-stained 10-20% SDS-PAGE. Fractions with similar band pat-terns were pooled and washed three times with PBS on a Centriprep concentrator (Amicon) with a 3 kDa cut off membrane to a final volume of 1-3 mi. An equal volume of SDS containing sample buffer was added and the protein solution boiled for 5 min before further separation on a Prep Cell (BioRad) in a matrix of 16%
polyacrylamide under an electrical gradient. Fractions containing well separated bands in SDS-PAGE
were selected for N-terminal sequencing after transfer to PVDF membrane.
Isolation of CFP8A, CFPBB, CFP19, CFP23A, and CFP23B.
ST-CF was precipitated with ammonium sulphate at 80% saturation and redissolved in PBS, pH 7.4, and dialysed 3 times against 25mM Piperazin-HCI, pH 5.5, and subjected to chromatofocusing on a matrix of PBE 94 (Pharmacia) in a column connected to an -FPLC system (Pharmacia). The column was equilibrated with 25 mM Piperazin-HCI, pH
5.5, and the elution was performed with 10% PB74-HCI, pH 4.0 (Pharmacia). Frac-tions with similar band patterns were pooled and washed three times with PBS
on a Centriprep concentrator (Amicon) with a 3 kDa cut oft membrane to a final volume of 1-3 ml and separated on a Prepcell as described above.
Identification of CFP22A

ST-CF was concentrated approximately 10 fold by ultrafiltration and-proteins were precipitated at 80 % saturation, redissolved in PBS, pH 7.4, and dialysed 3 times against PBS, pH 7.4. 5.1 ml of the dialysed ST-CF was treated with RNase (0.2 mg/ml, QUIAGEN) and DNase (0.2 mg/ml, Boehringer Mannheim) for 6 h and placed 10 on top of 6.4 ml of 48 % (wlv) sucrose in PBS, pH 7.4, in Sorvall tubes (Ultracrimp 03987, DuPont Medical Products) and ultracentrifuged for 20 h at 257,300 x gmax.
10°C. The pellet was redissolved in 200 NI of 25 mM Tris-192 mM
glycine, 0.1 SDS, pH 8.3.
15 Identification of CFP7A, CFP25A, CFP27, CFP30A and CFP50 For CFP27, CFP30A and CFP50 ST-CF was concentrated approximately 10 fold by ultrafiltration and ammonium sulphate precipitation in the 45 to 55 %
saturation range was performed. Proteins were redissolved in 50 mM sodium phosphate, 1.5 M ammo-20 nium sulphate, pH 8.5, and subjected to thiophilic adsorption chromatography on an Affi-T gel column (Kem-En-Tec). Proteins were eluted by a 1.5 to 0 M
decreasing gra-dient of ammonium sulphate. Fractions with simitar band patterns in SDS-PAGE
were pooled and anion exchange chromatography was performed on a Mono Q HR 5/5 col-umn connected to an FPLC system (Pharmacia). The column was equilibrated with 25 mM Tris-HCI, pH 8.5, and the elution was performed with a gradient of NaCI
from 0 to 1 M. Fractions containing well separated bands in SDS-PAGE were selected.
CFP7A and CFP25A were obtained as described above except far the following modi-fication: ST-CF was concentrated approximately 10 fold by ultrafiltration and proteins 30 were precipitated at 80 % saturation, redissolved in PBS, pH 7.4, and dialysed 3 times against PBS, pH 7.4. Ammonium sulphate was added to a concentration of 1.5 M, and ST-CF proteins were loaded on an Affi T-gel column. Elution from the Affi T-gel column and anion exchange were performed as described above.

Isolation of CWP32 Heat treated H37Rv was subfractionated into subcellular fractions as described in Se-rensen et al 1995. The Cell wall fraction was resuspended in 8 M urea, 0.2 %
(w/v) N-octyl (3-p gfucopyranoside (Sigma) and 5 % (v/v) glycerol and the protein solution was applied to a Rotofor isoelectrical Cell (BioRad) which was equilibrated with the same buffer. Isoetectric focusing was performed in a pH gradient from 3-6. The frac-tions were analyzed by SDS-PAGE and fractions containing well separated bands were polled and subjected to N-terminal sequencing after transfer to PVDF membrane.
N-terminal seguencing Fractions containing CFP7A, CFPBA, CFPBB, CFP16, CFP19, CFP19B, CFP22A, CFP23A, CFP23B, CFP27, CFP30A, CWP32, and CFP50A were blotted to PVDF
membrane after Tricine SDS-PAGE IPloug et al, 1989). The relevant bands were ex-cised and subjected to N-terminal amino acid sequence analysis on a Procise 494 se-quencer (Applied Biosystems): The fraction containing CFP25A was blotted to PVDF
membrane after 2-DE PAGE (isoelectric focusing in the first dimension and Tricin SDS-PAGE in the second dimension). The relevant spot was excised and sequenced as de-scribed above.
The following N-terminal sequences were obtained:
CFP7A: AEDVRAEIVA SVLEVVVNEG DQIDKGDVVV
LLESMYMEIP

VLAEAAGTVS (SEQ ID NO: 81 ) CFPBA: DPVDDAFtAKLNTAG (SEQ ID NO: 73) CFPBB: DPVDAtINLDNYGX (SEQ ID NO: 74) CFP16: AKLSTDELLDAFKEM (~EQ ID NO: 79) CFP19: TTSPDPYAALPKLPS (SEQ ID NO: 82) -CFP19B: DPAXAPDVPTAAQLT (SEQ ID NO: 80) CFP22A: TEYEGPKTKF HALMQ (SEQ tD NO: 83) CFP23A: VIQ/AGMVT/GHIHXVAG (SEQ ID NO: 76) CFP23B: AEMKXFKNAIVQEtD (SEQ tD NO: 75) CFP25A: AIEVSVLRVF TDSDG (SEQ ID NO: 78) CWP32: TNIVVLIKQVPDTWS (SEQ ID NO: 77) CFP27: TTIVALKYPG GVVMA (SEQ ID NO: 84) CFP30A: SFPYFISPEX AMRE (SEO ID NO: 85) CFP50: THYDVVVLGA GPGGY (SEQ ID NO: 86) N-terminal homology searching in the Sanger database and identification of the corre-sponding genes.
The N-terminal amino acid sequence from each of the proteins was used for a homo-logy search using the blast program of the Sanger Mycobacterium tuberculosis data-base:
http://www.Banger.ac.uk/projects/m-tuberculosis/TB-blast-server.
For CFP23B, CFP23A, and CFP19B no similarities were found in the Sanger database.
This could be due to the fact that only approximately 70% of the M.
tuberculosis ge-nome had been sequenced when the searches were performed. The genes encoding these proteins could be contained in the remaining 30% of the genome for which no sequence data is yet available.
For CFP7A, CFPBA, CFPBB, CFP16, CFP19, CFP19B , CFP22A, CFP25A, CFP27, CFP30A, CWP32, and CFP50, the following information was obtained:
CFP7A: Of the 50 determined amino acids in CFP7A a 98% identical sequence was found in cosmid csCY07D1 (contig 256):
Score = 226 (100.4 bits), Expect = 1.4e-24, P = 1.4e-24 Identities = 49150 (98%), Positives = 49/50 (98%), Frame = -i Query: 1 AEDVRAEIVASVLEVVVNEGDQIDKGDVVVLLESMYMEIPVLAEAAGTVS

Sbjct: 257679 AEDVRAEIVASVLEVVVNEGDOIDKGDVVVLLESMKMEIPVLAEAAGTVS

(SEQ ID NOs: 127, 128, and 129) The identity is found within an open reading frame of 71 amino acids length corre-sponding to a theoretical MW of CFP7A of 7305.9 Da and a pl of 3.762. The ob-served molecular weight in an SDS-PAGE gel is 7 kDa.
CFPBA: A sequence 80% identical to the 15 N-terminal amino acids-was found on coniig TB-1884. The N-terminally determined sequence from the protein purified from culture filtrate starts at amino acid 32. This gives a length of the mature protein of 98 amino acids corresponding to a theoretical MW of 9700 Da and a pl of 3.72 This is in good agreement with the observed MW on SDS-PAGE at approximately 8 kDa. The full length protein has a theoretical MW of 12989 Da and a pl of 4.38.
CFPBB: A sequence 71 % identical to the 14 N-terminal amino acids was found on contig TB_653. However, careful re-evaluation of the original N-terminal sequence data confirmed the identification of the protein. The N-terminally determined sequence from the protein purified from culture filtrate starts at amino acid 29. This gives a length of the mature protein of 82 amino acids corresponding to a theoretical MW of 8337 Da and a pl of 4.23. This is in good agreement with the observed MW on SDS-PAGE at approximately 8 kDa. Analysis of the amino acid sequence predicts the pres-ence of a signal peptide which has been cleaved of the mature protein found in culture filtrate.
CFP16: The 15 as N-terminal sequence was found to be 100% identical to a sequence found on cosmid MTCY20H1.
The identity is found within an open reading frame of 130 amino acids length corre-sponding to a theoretical MW of CFP16 of 13440.4 Da and a pl of 4.59. The ob-served molecular weight in an SDS-PAGE gel is 16 kDa.
CFP19: The 15 as N-terminal sequence was found to be 100% identical to a sequence found on cosmid MTCY270.

The identity is found within an open reading frame of 176 amino acids length corre-sponding to a theoretical MW of CFP19 of 18633.9 Da and a pl of 5.41 . The ob-served molecular weight in an SDS-PAGE gel is 19 kDa.
CFP22A: The 15 as N-terminal sequence was found to be 100% identical to a se-quence found on cosmid MTCY 1 A6.
The identity is found within an open reading frame of 181 amino acids length corre-sponding to a theoretical MW of CFP22A of 20441.9 Da and a pl of 4.73. The ob-served molecular weight in an SDS-PAGE gel is 22 kDa.
CFP25A: The 15 as N-terminal sequence was found to be 100% identical to a se-quence found on contig 255.
The identity is found within an open reading frame of 228 amino acids length corre-sponding to a theoretical MW of CFP25A of 24574.3 Da and a pl of 4.95. The ob-served molecular weight in an SDS-PAGE gel is 25 kDa.
CFP27: The 15 as N-terminal sequence was found to be 100% identical to a sequence found on cosmid MTCY261.
The identity is found within an open reading frame of 291 amino acids length.
The N-terminally determined sequence from the protein purified from culture filtrate starts at amino acid 58. This gives a length of the mature protein of 233 amino acids, which corresponds to a theoretical molecular weigh at 24422.4 Da, and a theoretical pl at 4.64. The observed weight in an SDS-PAGE gel is 27 kDa.
CFP30A: Of the 13 determined amino acids in CFP30A, a 100% identical sequence was found on cosmid MTCY261.
The identity is found within an open reading frame of 248 amino acids length corre-sponding to a theoretical MW of CFP30A of 26881.0 Da and a pl of 5.41. The ob-served molecular weight in an SDS-PAGE gel is 30 kDa.

WO 99/24577 PCTlDK98/00438 CWP32: The 15 amino acid N-terminal sequence was found to be 100% identical to a sequence found on contig 281. The identity was found within an open reading frame of 266 amino acids length, corresponding to a theoretical MW of CWP32 of 28083 Da and a pl of 4.563. The observed molecular weight in an SDS-PAGE gel is 32 kDa.

CFP50: The 15 as N-terminal sequence was found to be 100% idEntical to a sequence found in MTV038.06. The identity is found within an open reading frame of 464 amino acids length corresponding to a theoretical MW of CFP50 of 49244 Da and a pl of 5.66. The observed molecular weight in an SDS-PAGE gel is 50 kDa.
Use of homology searchinca in the EMBL database for identification of CFP19A
and CFP23.
Homology searching in the EMBL database (using the GCG package of the Biobase, Arhus-DK) with the amino acid sequences of two earlier identified highly immunoreac-tive ST-CF proteins, using the TFASTA algorithm, revealed that these proteins tCFP21 and CFP25, EXAMPLE 3) belong to a family of fungal cutinase homologs. Among the most homologous sequences Were also two Mycobacterium tuberculosis sequences found on cosmid MTCY13E12. The first, MTCY13E12.04 has 46% and 50% identity to CFP25 and CFP21 respectively. The second, MTCY13E12.05, has also 46% and 50% identity to CFP25 and CFP21. The two proteins share 62.5% as identity in a 184 residues overlap_ On the basis of the high homology to the strong T-cell antigens CFP21 and CFP25, respectively, it is believed that CFP19A and CFP23 are possible new T-cell antigens.
The first reading frame encodes a 254 amino acid protein of which the first 26 as constitute a putative leader peptide that strongly indicates an extracellular location of the protein. The mature protein is thus 228 as in length corresponding to a theoretical MW of 23149.0 Da and a Pi of 5.80. The protein is named CFP23.
The second reading frame encodes an 231 as protein of which the first 44 as consti-tute a putative leader peptide that strongly indicates an extracellular location of the protein. The mature protein is thus 187 as in length corresponding to a theoretical MW of 19020.3 Da and a Pi of 7.03. The protein is named CFP19A.

The presence of putative leader peptides in both proteins (and thereby their presence in the ST-CF) is confirmed by theoretical sequence analysis using the signalP
program at the Expasy molecular Biology server (http:/lexpasy.hcuge.chlwwwltools.html). -Searching for homologies to CFP7A, CFP16, CFP19, CFP19A, CFP19B, CFP22A, CFP23, CFP25A, CFP27, CFP30A, CWP32 and CFP50 in the EMBL database.
The amino acid sequences derived from the translated genes of the individual antigens were used for homology searching in the EMBL and Genbank databases using the TFASTA algorithm, in order to find homologous proteins and to address eventual func-tional roles of the antigens.
CFP7A: CFP7A has 44% identity and 70% similarity to hypothetical Methanococcus jannaschii protein (M. jannaschii from base 1 162199-1 175341 ), as well as 43% and 38% identity and 68 and 64% similarity to the C-terminal part of B.
stearothermophi-lus pyruvate carboxylase and Streptococcus mutans biotin carboxyl carrier protein.
CFP7A contains a consensus sequence EAMKM for a biotin binding site motif which in this case was slightly modified (ESMKM in amino acid residues 34 to 38). By incuba-tion with alkaline phosphatase conjugated streptavidin after SDS-PAGE and transfer to nitrocellulose it was demonstrated that native CFP7A was biotinylated.
CFP16: RAIL gene, 130 aa. Identical to the M. bovis 50s ribosomal protein (acc. No P37381 ).
CFP19: CFP19 has 47% identity and 55% similarity to E.coli pectinesterase homolog (ybhC gene) in a 150 as overlap.
CFP19A: CFP19A has between 38% and 45% identity to several cutinases from dif-ferent fungal sp.

In addition CFP19A has 46% identity and 61 % similarity to CFP25 as well as 50%
identity and 64% similarity to CFP21 (both proteins are earlier isolated from the ST-CF).
CFP19B: No apparent homology CFP22A: No apparent homology CFP23: CFP23 has between 38% and 46% identity to several cutinases from different fungal sp.
In addition CFP23 has 46% identity and 61 % similarity to CFP25 as well as 50%
identity and 63% similarity to CFP21 (both proteins are earlier isolated from the ST-CF).
CFP25A: CFP25A has 95% identity in a 241 as overlap to a putative M.
tuberculosis thymidylate synthase (450 as accession No p28176).
CFP27: CFP27 has 81 % identity to a hypothetical M. leprae protein and 64%
identity and 78% similarity to Rhodococcus sp. proteasome beta-type subunit 2 (prcB(2) gene).
CFP30A: CFP30A has 67% identity to Rhodococcus proteasome alfa-type 1 subunit.
CWP32: The CWP32 N-terminal sequence is 100% identical to the Mycobacterium leprae sequence MLCB637.03.
CFP50: The CFP50 N-terminal sequence is 100% identical to a putative lipoamide de-hydrogenase from M. leprae (Accession 415183) Cloning of the genes encoding CFP7A, CFPBA, CFPBB, CFP16 CFP19 CFP19A, CFP22A, CFP23, CFP25A, CFP27, CFP30A, CWP32, and CFP50.
The genes encoding CFP7A, CFPBA, CFP8B, CFP16, CFP19, CFP19A, CFP22A, CFP23, CFP25A, CFP27, CFP30A, CWP32 and CFP50 were all cloned into the ex-pression vector pMCT6, by PCR amplification with gene specific primers, for recombi-nant expression in E. coli of the proteins.
PCR reactions contained 10 ng of M. tuberculosis chromosomal DNA in 1 X low salt Taq+ buffer from Stratagene supplemented with 250 mM of each of the four nucleo-tides (Boehringer Mannheim), 0,5 mg/ml BSA (IgG technology), 1 % DMSO (Merck), pmoles of each primer and 0.5 unit Tag + DNA poiymerase (Stratagene) in 10 ml reac-tion volume. Reactions were initially heated to 94°C for 25 sec. and run for 30 cycles of the program; 94°C for 10 sec., 55°C for 10 sec. and 72°C for 90 sec, using ther-mocycler equipment from Idaho Technology.
The DNA fragments were subsequently run on 1 % agarose gels, the bands were ex-cised and purified by Spin-X spin columns (Costar) and cloned into pBluescript SK II+
T vector (Stratagene). Plasmid DNA was hereafter prepared from clones harbouring the desired fragments, digested with suitable restriction enzymes and subcloned into the expression vector pMCT6 in frame with 8 histidines which are added to the N-terminal of the expressed proteins. The resulting clones were hereafter sequenced by use of the dideoxy chain termination method adapted for supercoiled DNA using the Sequenase DNA sequencing kit version 1.0 (United States Biochemical Corp., USA) and by cycle sequencing using the Oye Terminator system in combination with an automated gel reader (model 373A; Applied Biosystems) according to the instructions provided. Both strands of the DNA were sequenced.
For cloning of the individual antigens, the following gene specific primers were used:
CFP7A: Primers used for cloning of cfp7A:
OPBR-79: AAGAGTAGATCTATGATGGCCGAGGATGTTCGCCz. (SEQ ID NO:
95) OPBR-80: CGGCGACGACGGATCCTACCGCGTCGG (SEQ ID NO: 96) OPBR-79 and OPBR-80 create Bglll and BamHl sites, respectively, used for the cloning in pMCT6.

CFPBA: Primers used for cloning of cfpBA:
CFPBA-F: CTGAGATCTATGAACCTACGGCGCC (SEa ID NO: 154) CFPBA-R: CTCCCATGGTACCCTAGGACCCGGGCAGCCCCGGC (SEQ ID NO:
155) CFPBA-F and CFPBA-R create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.
CFPBB: Primers used for cloning of cfp88:
CFPBB-F: CTGAGATCTATGAGGCTGTCGTTGACCGC . (SEQ ID NO: 156) CFPBB-R: CTCCCCGGGCTTAATAGTTGTTGCAGGAGC (SEQ ID NO. 157) CFPBB-F and CFPBB-R create Bglll and Smal sites, respectively, used for the cloning in pMCT6.
CFP16: Primers used for cloning of cfpl6:
OPBR-104: CCGGGAGATCTATGGCAAAGCTCTCCACCGACG
(SEQ ID NOs: 1 1 1 and 130) OPBR-105: CGCTGGGCAGAGCTACTTGACGGTGACGGTGG
(SEQ ID NOs: 1 12 and 131 ) OPBR-104 and OPBR-105 create Bglll and Ncol sites, respectively, used for the clon-ing in pMCT6.
CFP19: Primers used for cloning of cfpl9:
OPBR-96: GAGGAAGATCTATGACAACTTCACCCGACCCG
(SEQ ID NO: 107) OPBR-97. CATGAAGCCATGGCCCGCAGGCTGCATG
(SEQ ID NO: 108?

WO 99!24577 PCT/DK98/00438 OPBR-96 and OPBR-97 create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.
CFP19A: Primers used for cloning of cfp19A:

OPBR-88: CCCCCCAGATCTGCACCACCGGCATCGGCGGGC -(SEQ ID NO: 99) OPBR-89. GCGGCGGATCCGTTGCTTAGCCGG (SEQ ID NO: 100) 10 OPBR-88 and OPBR-89 create Bglll and BamHl sites, respectively, used for the cloning in pMCT6.
CFP22A: Primers used for cloning of cfp22A:
15 OPBR-90: CCGGCTGAGATCTATGACAGAATACGAAGGGC
(SEQ ID NO: 101 ) OPBR-91: CCCCGCCAGGGAACTAGAGGCGGC (SEQ ID NO: 102) OPBR-90 and OPBR-91 create Bglll and Ncol sites, respectively, used for the cloning 20 in pMCT6.
CFP23: Primers used for cloning of cfp23:
OPBR-86: CCTTGGGAGATCTTTGGACCCCGGTTGC
25 (SEQ ID NO: 97) OPBR-87: GACGAGATCTTATGGGCTTACTGAC (SEQ 1D NO: 98) OPBR-86 and OPBR-87 both create a Bglll site used for the cloning in pMCT6.
30 CFP25A: Primers used for cloning of cfp25A:
OPBR-106: GGCCCAGATCTATGGCCATTGAGGTTTCGGTGTTGC
(SEQ ID NO: 113) OPBR-107: CGCCGTGTTGCATGGCAGCGCTGAGC (SEQ ID NO: 1 14) OPBR-106 and OPBR-107 create BgRI and Ncol sites, respectively, used for the clon-ing in pMCT6.
CFP27: Primers used for cloning of cfp27:
OPBR-92: CTGCCGAGATCTACCACCATTGTCGCGCTGAAATACCC
(SEQ ID NO: 103?
OPBR-93: CGCCATGGCCTTACGCGCCAACTCG (SEQ ID NO: 104) OPBR-92 and OPBR-93 create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.
CFP30A: Primers used for cloning of cfp30A:
OPBR-94: GGCGGAGATCTGTGAGTTTTCCGTATTTCATC
ISEQ ID NO: 105) OPBR-95: CGCGTCGAGCCATGGTTAGGCGCAG (SEQ ID NO: 106) OPBR-94 and OPBR-95 create Bglll and Ncol sites, respectively, used for the cloning in pMCT6.
CWP32: Primers used for cloning of cwp32:
CWP32-F: GCTTAGATCTATGATTTTCTGGGCAACCAGGTA
(SEQ ID NO: 158) CWP32-R: GCTTCCATGGGCGAGGCACAGGCGTGGGAA (SEQ 1D NO: 1591 CWP32-F and CWP32-R create Bg/11 and Ncol sites, respectively, used for the cloning in pMCT6.
CFP50: Primers used for cloning of cfp50:

OPBR-100: GGCCGAGATCTGTGACCCACTATGACGTCGTCG
(SEQ ID NO: 109) OPBR-101: GGCGCCCATGGTCAGAAATTGATCATGTGGCCAA
(SEQ ID NO: 110) OPBR-100 and OPBR-101 create Bglll and Ncol sites, respectively, used for the clon-ing in pMCT6.
Expression/purification of recombinant CFP7A, CFPBA, CFP8B CFP16 CFP19 CFP19A. CFP22A, CFP23, CFP25A, CFP27, CFP30A, CWP32 and CFP50 proteins Expression and metal affinity purification of recombinant proteins was undertaken es-sentially as described by the manufacturers. For each protein, 1 I LB-media containing 100 ug/ml ampicillin, was inoculated with 10 ml of an overnight culture of XL1-Blue cells harbouring recombinant pMCT6 plasmids. Cultures were shaken at 37°C until they reached a density of OD6~ = 0.4 - 0.6. IPTG was hereafter added to a final con-centration of 1 mM and the cultures were further incubated 4-16 hours. Cells were harvested, resuspended in 1 X sonication buffer + 8 M urea and sonicated 5 X
30 sec.
with 30 sec. pausing between the pulses.
After centrifugation, the iysate was applied to a column containing 25 ml of resus-pended Talon resin (Clontech, Palo Alto, USA). The column was washed and eluted as described by the manufacturers.
After elution, all fractions (1.5 ml each) were subjected to analysis by SDS-PAGE us-ing the Mighty Small (Hoefer Scientific Instruments, USA) system and the protein con-centrations were estimated at 280 nm. Fractions containing recombinant protein were pooled and dialysed against 3 M urea in 10 mM Tris-HCI, pH 8.5. The dialysed protein was further purified by FPLC (Pharmacia, Sweden) using a 6 ml Resource-Q
column, eluted with a linear O-1 M gradient of NaCI. Fractions were analyzed by SDS-PAGE
and protein concentrations were estimated at ODZeo. Fractions containing protein were pooled and dialysed against 25 mM Hepes buffer, pH 8.5.

Finally the protein concentration and the LPS content were determined by the BCA
(Pierce, Holland) and LAL (Endosafe, Charleston, USA) tests, respectively.

Identification of CFP7B, CFP10A, CFPI 1 and CFP30B.
Isolation of CFP7B
ST-CF was precipitated with ammonium sulphate at 80% saturation and redissolved in PBS, pH 7.4, and dialyzed 3 times against 25 mM Piperazin-HCI, pH 5.5, and subject-ed to cromatofocusing on a matrix of PBE 94 (Pharmacia) in a column connected to an FPLC system (Pharmacia). The column was equilibrated with 25 mM Piperazin-HCI, pH
5.5, and the elution was performed with 10% PB74-HCI, pH 4.0 (Pharmacia). Frac-tions with similar band patterns were pooled and washed three times with PBS
on a Centriprep concentrator (Amicon) with a 3 kDa cut off membrane to a final volume of '!-3 ml. An equal volume of SDS containing sample buffer was added and the protein solution boiled for 5 min before further separation on a MultiEluter (BioRad) in a matrix of 10-20 % polyacrylamid (Andersen,P. & Heron,l., 1993). The fraction containing a well separated band below 10 kDa was selected for N-terminal sequencing after trans-fer to a PVDF membrane.
Isolation of CFP1 1 ST-CF was precipitated with ammonium sulphate at 80% saturation. The precipitated proteins were removed by centrifugation and after resuspension washed with 8 M
urea. CHAPS and glycerol were added to a final concentration of 0.5 % (wlv) and 5%
(v/v) respectively and the protein solution was applied to a Rotofor isoelectrical Csll (BioRadl. The Rotofor Cell had been equilibrated w'tth an 8M urea buffer containing 0.5 % (w/v) CHAPS, 5% (viv) glycerol, 3% (v!v) Biolyt 3/5 and 1 % (vlv) Biolyt (BioRad). Isoelectric focusing was performed in a pH gradient from 3-6. The fractions were analyzed on silver-stained 10-20% SDS-PAGE. The fractions in the pH
gradient 5.5 to 6 were pooled and washed three times with PBS on a Centriprep concentrator (Amicon) with a 3 kDa cut off membrane to a final volume of 1 ml. 300 mg of the protein preparation was separated on a 10-20% Tricine SDS-PAGE (Ploug et al 1989) and transferred to a PVDF membrane and Coomassie stained. The lowest band occur-ring on the membrane was excised and submitted for N-terminal sequencing.
Isolation of CFP10A and CFP30B
ST-CF was concentrated approximately 10-fold by ultrafiltration and ammonium sul-phate precipitation at 80 % saturation. Proteins were redissolved in 50 mM
sodium phosphate, 1.5 M ammonium sulphate, pH 8.5, and subjected to thiophilic adsorption chromatography on an Affi-T gel column (Kem-En-Tec). Proteins were eluted by a 1.5 to 0 M decreasing gradient of ammonium sulphate. Fractions with similar band pat-terns in SDS-PAGE were pooled and anion exchange chromatography was performed on a Mono Q HR 5/5 column connected to an FPLC system (Pharmacia). The column was equilibrated with 10 mM Tris-HCI, pH 8.5, and the elution was performed with a gradient of NaCI from 0 to 1 M. Fractions containing well separated bands in SDS-PAGE were selected.
Fractions containing CFP10A and CFP30B were blotted to PVDF membrane after 2-DE
PAGE (Ploug et al, 1989). The relevant spots were excised and subjected to N-termi-nal amino acid sequence analysis.
N-terminal seauencing N-terminal amino acid sequence analysis was performed on a Procise 494 sequencer (applied Biosystems).
The following N-terminal sequences were obtained:
CFP7B: PQGTVKWFNAEKGFG (SEQ ID NO: 168) CFP10A: NVTVSIPTILRPXXX (SEQ ID NO: 169) CFP1 1: TRFMTDPHAMRDMAG (SEQ lD NO: 170) CFP30B: PKRSEYRQGTPNWVD (SEQ ID NO: 171 ) "X" denotes an amino acid which could not be determined by the sequencing method used.
N-terminal homolo4y searching in the Sanger database and identification of the corre- ' 5 s~ondinc,~genes.
The N-terminal amino acid sequence from each of the proteins was used for a homo-logy search using the blast program of the Sanger Mycobacterium tuberculosis ge-nome database:
http//www.sanger.ac.uklprojectslm-tuberculosis/TB-blast-server.
For CFP1 1 a sequence 100% identical to 15 N-terminal amino acids was found on contig TB_1314. The identity was found within an open reading frame of 98 amino acids length corresponding to a theoretical MW of 10977 Da and a pl of 5.14.
Amino acid number one can also be an Ala (insted of a Thr) as this sequence was also obtained (results not shown), and a 100% identical sequence to this N-terminal is found on contig TB_671 and on locus MTC1364.09.
For CFP7B a sequence 100% identical to 15 N-terminal amino acids was found on contig TB-2044 and on locus MTY15C10.04 with EMBL accession number: z95436.
The identity was found within an open reading frame of 67 amino acids length corre-sponding to a theoretical MW of 7240 Da and a pl of 5.18.
For CFP10A a sequence 100% identical to 12 N-terminal amino acids was found on contig TB 752 and on locus CY130.20 with EMBL accession number: Q10646 and 273902. The identity was found within an open reading frame of 93 amino acids length corresponding to a theoretical MW of 9557 Da and a pl of 4.78. -For CFP30B a sequence 100% identical to 15 N-terminal amino acids was found on .
contig TB 335. The identity was found within an open reading frame of 261 amino acids length corresponding to a theoretical MW of 27345 Da and a pl of 4.24.

The amino acid sequences of the purified antigens as picked from the Sanger database are shown in the following list.
CFP7B (SEQ ID NO: 147) CFP10A (SEQ ID NO: 141 ) CFP1 1 protein sequence (SEQ ID NO: 143) CFP30B (SEQ ID NO: 145) 101 MMPAFDiGDA GRMSFITDPT GAAVGLWQAN RHIGATLVNE TGTLIWNELL

20i VPNHWHVYFA VDDADATAAK AAAAGGQV1A EPADIPSVGR FAVLSDPQGA

Cloning of the 4enes encoding CFP7B, CFP10A, CFP11. and CFP30B.
PCR reactions contained 10 ng of M. tuberculosis chromosomal DNA in 1 X low salt Taq + buffer from Stratagene supplemented with 250 mM of each of the four nucleo-tides (Boehringer Mannheim!. 0,5 mg/ml BSA (IgG technology), 1 % DMSO (Merck), pmoies of each primer and 0.5 unit Tag + DNA polymerase (Stratagene) in 10 ml reac-tion volume. Reactions were initially heated to 94°C for 25 sec. and run for 30 cycles of the program; 94°C for 10 sec., 55°C for 10 sec. and 72°C for 90 sec., using ther-mocycler equipment from Idaho Technology.
The DNA fragments were subsequently run on 1 % agarose geis, the bands were ex-cised and purified by Spin-X spin columns (Costar) and cloned into pBluscript SK ll + -T vector (Stratagene). Plasmid DNA was hereafter prepared from clones harbouring the desired fragments, digested with suitable restriction enzymes and subcloned into the expression vector pMCT6 in frame with 8 histidines which are added to the N-terminal of the expressed proteins. The resulting clones were hereafter sequenced by use of the dideoxy chain termination method adapted for supercoiled DNA using the Sequenase DNA sequencing kit version 1.0 (United States Biochemical Corp., USA) and by cycle sequencing using the Dye Terminator system in combination with an automated gel reader (model 373A; Applied Biosystems) according to the instructions provided. Both strands of the DNA were sequenced.
For cloning of the individual antigens, the following gene specific primers were used:
CFP7B: Primers used for cloning of cfp7B:
CFP7B-F: CTGAGATCTAGAATGCCACAGGGAACTGTG (SEQ iD NO: 160) CFP7B-R: TCTCCCGGGGGTAACTCAGAGCGAGCGGAC (SEQ ID NO: 161 ) CFP7B-F and CFP7B-R create Bglfl and Smal sites, respectively, used for the cloning in pMCT6.
CFP10A: Primers used for cloning of cfplOA:
CFP10A-F: CTGAGATCTATGAACGTCACCGTATCC (SEQ ID NO: 162) CFP10A-R: TCTCCCGGGGCTCACCCACCGGCCACG ISEQ ID NO: 163) CFP10A -F and CFP10A -R create Bglll and Smal sites, respectively, used for the cloning in pMCT6.
CFP1 1: Primers used for cloning of cfpll CFP11-F: CTGAGATCTATGGCAACACGTTTTATGACG (SEQ ID NO: 164) CFP11-R: CTCCCCGGGTTAGCTGCTGAGGATCTGCTH (SEQ ID NO: 165) CFP1 1-F and CFP1 1-R create Bglll and Smal sites, respectively, used for the cloning in pMCT6.
CFP30B: Primers used for cloning of cfp30B:
CFP30B-F: CTGAAGATCTATGCCCAAGAGAAGCGAATAC (SEQ ID NO: 166) CFP30B -R: CGGCAGCTGCTAGCATTCTCCGAATCTGCCG (SEQ ID NO: 167) CFP30B-F and CFP30B-R create BgAI and Pvull sites, respectively, used for the cloning in pMCT6.
Expression/purification of recombinant CFP7B. CFP10A, CFP11 and CFP30B
protein.
Expression and metal affinity purification of recombinant protein was undertaken es-sentially as described by the manufacturers. 1 I LB-media containing 100 Nglml ampi-cillin, was inoculated with 10 ml of an overnight culture of XL1-Blue cells harbouring recombinant pMCT6 plasmid. The culture was shaken at 37 °C until it reached a den-sity of OD6~ = 0.5. IPTG was hereafter added to a final concentration of i mM
and the culture was further incubated 4 hours. Cells were harvested, resuspended in 1 X
sonication buffer + 8 M urea and sonicated 5 X 30 sec. with 30 sec. pausing be-tween the pulses.
After centrifugation, the lysate was applied to a column containing 25 ml of resus-pended Talon resin (Clontech, Palo Alto, USA). The column was washed and eluted as described by the manufacturers.
After elution, all fractions (1.5 ml each) were subjected to analysis by SDS-PAGE us-ing the Mighty Small (Hoefer Scientific Instruments, USA) system and the protein con-centrations were estimated at 280 nm. Fractions containing recombinant protein were pooled and dialysed against 3 M urea in 10 mM Tris-HCI, pH 8.5. The dialysed protein was further purified by FPLC (Pharmacia, Sweden) using a 6 ml Resource-Q
column, eluted with a linear 0-1 M gradient of NaCI. Fractions were analysed by SDS-PAGE
and protein concentrations were estimated at ODZeo. Fractions containing protein were pooled and dialysed against 25 mM Hepes buffer, pH 8.5. ' Finally the protein concentration and the LPS content was determined by the BCA
(Pierce, Holland) and LAL (Endosafe, Charleston, USA) tests, respectively.

Using homology searching for identification of ORF11-1, ORF1 1-2, ORF11-3 and ORFI 1-4.
A search of the Mycobacterium tuberculosis Sanger sequence database with the amino acid sequences of CFP1 1, a previously identified ST-CF protein, identified 4 new very homologous proteins. All 4 proteins were at least 96% homologous to CFP1 1.
On the basis of the strong homology to CFP1 1, it is belived that ORF1 1-1, ORF11-2, ORF11-3 and ORF1 1-4 are potential new T-cell antigens.
The first open reading frame, MTCY10G2.1 1, homologous to CFP11, encodes a pro-tein of 98 amino acids corresponding to a theoretical molecular mass of 10994Da and a pl of 5.14. The protein was named ORF11-1.
The second open reading frame, MTC1364.09, homologous to CFP11, encodes a pro-tein of 98 amino acids corresponding to a theoretical molecular mass of 10964Da and a pl of 5.14. The protein was named ORF11-2.
The third open reading frame, MTV049.14, has an in frame stop codon. Because of the very conserved DNA sequence in this position amongst the 4 open reading frames it is however suggested that this is due to a sequence mistake.

The "T" in position 175 of the DNA sequence is therefor suggested to be a "C"
as in the four other ORF's. The Q in position 59 in the amino acid sequence would have been a "stop" if the T in position 175 in the DNA sequence had not been substituted.
The open reading frame encodes a protein of 98 amino acids corresponding to a theo-5 retical molecular mass of 10994Da and a pl of 5.14. The protein was named 3.
The fourth open reading frame, MTCY15C10.32, homologous to CFP1 1, encodes a protein of 98 amino acids corresponding to a theoretical molecular mass of 1 1024Da 10 and a pl of 5.14. The protein was named ORF1 1-4.
Using homology searching for identification of ORF7-1 and ORF7-2.
A search of the Mycobacterium tuberculosis Sanger sequence database with the 15 amino acid sequences of a previously identified immunoreactive ST-CF
protein, CFP7, identified 2 new very homologous proteins. The protein ORF7-1 (MTV012.33) was 84% identical to CFP7, with a primary structure of the same size as CFP7, and the protein ORF7-2 (MTV012.31? was 68% identical to CFP7 in a 69 amino acid overlap.
On the basis of the strong homology to the potent human T-cell antigen CFP7, 20 and ORF7-2 are belived to be potential new T-cell antigens.
The first open reading frame homologous to CFP7, encodes a protein of 96 amino ac-ids corresponding to a theoretical molecular mass of 10313Da and a pl of 4.186. The protein was named ORF7-1.
The second open reading frame homologous to CFP7, encodes a protein of 120 amino acids corresponding to a theoretical molecular mass of 1 2923.00 Da and a pl of 7.889. The protein was named ORF7-2.
Clonin4 of the homoloctous orf7-1 and orf7-2.
Since ORF7-l and ORF7-2 are nearly identical to CFP7 it was nessesary to use the flanking DNA regions in the cloning procedure, to ensure the cloning of the correct ORF. Two PCR reactions were carried out with two different primer sets. PCR
reaction WO 99/24577 PCTlDK98/00438 1 was carried out using M. tuberculosis chromosomal DNA and a primerset corre-sponding to the flanking DNA. PCR reaction 2 was carried out directly on the first PCR
product using ORF specific primers which introduced restriction sites for use in the later cloning procedure. ' The sequences of the primers used are given below;
Orf7-1:
Primers used for the initial PCR reaction (1) using M. tuberculosis chromosomal DNA
as template;
Sence: MTV012.33-R1: 5'- GGAATGAAAAGGGGTTtGTG -~3' (SEQ ID NO: 186) Antisence:MTV012.33-F1: 5'- GACCACGCCCGCGCCGTGTG - 3'(SEQ ID N0:187) Primers used for the second round of PCR (2) using PCR product 1 as template;
Sence: MTV012.33-R2: 5' - GCAACACCCGGGATGTCGCAGATTATG - 3' (SEQ ID NO: 188) (introduces a Smal upstream of the orf7-1 start codon) Antisence:MTV012.33-F2: 5' - CTAAGCTTGGATCCCTAGCCGCCCCACTTG - 3' ((SEQ ID NO: 189) (introduces a BamHl downstream of the orf7-1 stop codon).
Orf7-2:
Primers used for the initial PCR reaction (1 ) using M. tuberculosis chromosomal DNA
as template;
Sence: MTV012.31-R1: 5'- GAATATTTGAAAGGGATTCGTG - 3' (SEQ 1D NO: 190) .
Antisence:MTV012.31-F1: 5'- CTACTAAGCTTGGATCCTTAGTCTCCGGCG - 3' (SEQ ID NO: 191 ) .
(introduces a BamHl downstream of the orf7-2 stop codon) Primers used for the second round of PCR (2) using PCR product 1 as template;

Sence: MTV012.31-R2: 5' -GCAACACCCGGGGTGTCGCAGAGTATG- 3' (SEQ ID NO: 192) (introduces a Smal upstream of the orf7-2 start codon) Antisence:MTV012.31-F1: 5'- CTACTAAGCTTGGATCCTTAGTCTCCGGCG - 3' (SEQ ID NO: 193) (introduces a BamHl downstream of the orf7-2 stop codon) The genes encoding ORF7-1 and ORF7-2 were cloned into the expression vector pMST24, by PCR amplification with gene specific primers, for recombinant expression in E. coli of the proteins.
The first PCR reactions contained either 10 ng of M. tuberculosis chromosomal DNA
(PCR reaction 1 ) or 1 Ong PCR product 1 (PCR reaction 2) in 1 x low salt Taq + buffer from Stratagene supplemented with 250 mM of each of the four nucleotides (Boehrin-ger Mannheim), 0,5 mg/ml BSA (IgG technology), 1 % DMSO (Merck), 5 pmoles of each primer and 0.5 unit Tag + DNA polymerase (Stratagene) in 10 ml reaction volume. Reactions were initially heated to 94°C for 25 sec. and run for 30 cycles of the program; 94°C for 10 sec., 55°C for 10 sec. and 72°C
for 90 sec, using thermo-cycler equipment from Idaho Technology.
The DNA fragments were subsequently run on 1 % agarose gels, the bands were ex-cised and purified by Spin-X spin columns (Costar) and cloned into pBluscript SK II + -T vector (Stratagene). Plasmid DNA was hereafter prepared from clones harbouring the desired fragments, digested with suitable restriction enzymes and subcloned into the expression vector pMST24 in frame with 6 histidines which are added to the N-terminal of the expressed proteins. The resulting clones were hereafter sequenced by use of the dideoxy chain termination method adapted for supercoiled DNA using the Sequenase DNA sequencing kit version 1.0 (United States Biochemical Corp., USA) and by cycle sequencing using the Dye Terminator system in combination with an automated gel reader (model 373A; Applied Biosystems) according to the instructions provided. Both strands of the DNA were sequenced.
Expression/purification of recombinant ORF7-1 and ORF7-2 protein.
Expression and metal affinity purification of recombinant protein was undertaken es-sentially as described by the manufacturers. 1 I LB-media containing 100 Ng/ml ampi-cillin, was inoculated with 10 ml of an overnight culture of XL1-Blue cells harbouring recombinant pMCT6 plasmid. The culture was shaken at 37 °C until it reached a den-sity of OD600 = 0.5. IPTG was hereafter added to a final concentration of 1 mM
and the culture was further incubated 2 hours. Cells were harvested, resuspended in 1 X
sonication buffer + 8 M urea and sonicated 5 X 30 sec. with 30 sec. pausing be-tween the pulses.
After centrifugation, the lysate was applied to a column containing 25 ml of resus-pended Talon resin (Clontech, Palo Alto, USA). The column was washed and eluted as described by the manufacturers.
After elution, all fractions (1.5 ml each) were subjected to analysis by SDS-PAGE us-ing the Mighty Small (Hoefer Scientific Instruments, USA) system. Fractions contain-ing recombinant protein were pooled and dialysed against 3 M urea in 10 mM
Tris-HCI, pH 8.5. The dialysed protein was further purified by FPLC fPharmacia, Sweden) using a 6 ml Resource-Q column, eluted with a linear 0-1 M gradient of NaCI.
Frac-tions were analysed by SDS-PAGE. Fractions containing protein were pooted and dia-lysed against 25 mM Hepes buffer, pH 8.5.
Finally the protein concentration and the LPS content was determined by the BCA
(Pierce, Holland) and LAL (Endosafe, Charleston, USA) tests, respectively.

Cloning of the gene expressing CFP26 (MPT51 J
Svnthesis and desi4n of probes Oligonucleotide primers were synthesized automatically on a DNA synthesizer (Applied Biosystems, Forster City, Ca, ABI-391, PCR-mode) deblocked and purified by ethanol precipitation.
Three oligonucleotides were synthesized (TABLE 3) on the basis of the nucleotide se-quence from mpb51 described by Ohara et al. (1995). The oligonucleotides were en-gineered to include an EcoRl restriction enzyme site at the 5' end and at the 3' end by which a later subcloning was possibte.

Additional four oligonucleotides were synthesized on the basis of the nucleotide se-quence from MPT51 (Fig. 5 and SEQ ID NO: 41 ). The four combinations of the primers were used for the PCR studies.
DNA cloning and PCR technology ' Standard procedures were used for the preparation and handling of DNA
(Sambrook et al., 1989). The gene mpt51 was cloned from M. tuberculosis H37Rv chromosomal DNA by the use of the polymerase chain reactions (PCR) technology as described pre-viously (Oettinger and Andersen, 1994). The PCR product was cloned in the pBluescriptSK + (Stratagene).
Cloning of mpt51 The gene, the signal sequence and the Shine Delgarno region of MPT51 was cloned by use of the PCR technology as two fragments of 952 by and 815 by in pBluescript SK + , designated pT052 and pT053.
DNA Seguencing The nucleotide sequence of the cloned 952 by M. tuberculosis H37Rv PCR
fragment, pT052, containing the Shine Dalgarno sequence, the signal peptide sequence and the structural gene of MPT51, and the nucleotide sequence of the cloned 815 by PCR
fragment containing the structural gene of MPT51, pT053, were determined by the dideoxy chain termination method adapted for supercoiled DNA by use of the Seque-nase DNA sequencing kit version 1.0 (United States Biochemical Corp., Cleveland, OH) and by cycle sequencing using the Dye Terminator system in combination with an automated gel reader (model 373A; Applied Biosystems) according to the instructions provided. Both strands of the DNA were sequenced.
The nucleotide sequences of pT052 and pT053 and the deduced amino acid sequence are shown in Figure 5. The DNA sequence contained an open reading frame starting with a ATG colon at position 45 - 47 and ending with a termination colon (TAA) at position 942 - 944. The nucleotide sequence of the first 33 codons was expected to encode the signal sequence. On the basis of the known N-terminal amino acid sequence (Ala - Pro - Tyr - Giu - Asn) of the purified MPT51 (Nagai et al., 1991 ) and the features of the signal peptide, it is presumed that the signal peptidase 5 recognition sequence (Ala-X-Ala) (von Heijne, 1984) is located in front of the N-terminal region of the mature protein at position 144. Therefore, a structural gene encoding MPT51, mpt5l, derived from M. tuberculosis H37Rv was found to be located at position 144 - 945 of the sequence shown in Fig. 5. The nucleotide sequence of mpt51 differed with one nucleotide compared to the nucleotide sequence 10 of MPB51 described by Ohara et al. (1995) (Fig. 5). In mpt51 at position 780 was found a substitution of a guanine to an adenine. From the deduced amino acid sequence this change occurs at a first position of the codon giving a amino acid change from alanine to threonine. Thus it is concluded, that mpt51 consists of 801 by and that the deduced amino acid sequence contains 266 residues with a molecular 15 weight of 27,842, and MPT51 show 99,8% identity to MPB51.
Subcloninq of mpt51 An EcoRl site was engineered immediately 5' of the first codon of mpt51 so that only 20 the coding region of the gene encoding MPT51 would be expressed, and an EcoRl site was incorporated right after the stop codon at the 3' end.
DNA of the recombinant plasmid pT053 was cleaved at the EcoRl sites. The 815 by fragment was purified from an agarose gel and subcloned into the EcoRl site of the 25 pMAL-cRl expression vector (New England Biolabs), pT054. Vector DNA
containing the gene fusion was used to transform the E. coli XL1-Blue by the standard proce-dunes for DNA manipulation.
The endpoints of the gene fusion were determined by the dideoxy chain termination 30 method as described under section DNA sequencing. Both strands of the DNA
were sequenced.

Preearation and purification of rMPT51 Recombinant antigen was prepared in accordance with instructions provided by New England Biolabs. Briefly, single colonies of E. colt harbouring the pT054 plasmid were inoculated into Luria-Bertani broth containing 50 Ng/mf ampiciliin and 12.5 ~rg/ml tetra-cycline and grown at 37°C to 2 x 10$ cells/ml. Isopropyl-~i-D-thiogalaCtoside (IPTG) was then added to a final concentration of 0.3 mM and growth was continued for fur-ther 2 hours. The pelleted bacteria were stored overnight at -20°C in new column buffer (20 mM Tris/HCI, pH 7.4, 200 mM NaCI, 1 mM EDTA, 1 mM dithiothreitol (DTT))and thawed at 4 °C followed by incubation with 1 mg/ml lysozyme on ice for 30 min and sonication (20 times for 10 sec with intervals of.20.sec). Atter centrifugation at 9,000 x g for 30 min at 4°C, the maltose binding protein -MPT51 fusion protein (MBP-rMPT51 ) was purified from the crude extract by affinity chromatography on amyfose resin column. MBP-rMPT51 binds to amylose. After extensive washes of the column, the fusion protein was eluted with 10 mM maltose. Aliquots of the fractions were analyzed on 10% SDS-PAGE. Fractions containing the fusion protein of interest were pooled and was dialysed extensively against physiological saline.
Protein concentration was determined by the BCA method supplied by Pierce (Pierce Chemical Company, Rockford, IL).

TABLE 3.
Sequence of the mpt51 oligonucleotidesa.
Orientation Sequences (5'-a 3') Positionb and oli- (nucleotide) gonucleotidea Sense (SEQ ID NO: 28) (SEQ ID NO: 41) -(SEQ ID NO: 29) . (SEQ ID NO: 41) -(SEQ ID NO: 30) (SEQ ID NO: 41) -(SEQ ID NO: 31) (SEQ ID NO: 41) Antisense -(SEQ ID NO: 32) (SEQ ID NO:
41) -(SEQ ID NO: 33) (SEQ ID NO:
41) MPT51-6 GTCCAGCAGATACAC ~ 242 - 228 (SEQ ID NO: 34) (SEQ ID NO:
41) a The oligonucieotides MPT51-1 and MPT51-2 were constructed from the MPB51 nu-cleotide sequence (Ohara et al., 1995). The other oligonucleotides constructions were based on the nucleotide sequence obtained from mpt51 reported in this work.
Nucleo-tides (nt) underlined are not contained in the nucleotide sequence of MPB/T51.
The positions referred to are of the non-underlined parts of the primers and corre- -spond to the nucleotide sequence shown in SEQ 1D NO: 41.

Cloning of mpt51 in the expression vector pMST24 A PCR fragment was produced from pT052 using the primer combination MPT51-F
and MPT51-R (TABLE 4). A BamHl site was engineered immediately 5' of the first co-don of mpt51 so that only the coding region of the gene encoding MPT51 would be expressed, and an Ncol site was incorporated right after the stop codon at the 3' end.
The PCR product was cleaved at the BamHl and the Ncol site. The 81 1 by fragment was purified from an agarose gel and subcloned into the BamHl and the Ncol site of the pMST24 expression vector, pT086. Vector DNA containing the gene fusion was used to transform the E. colt XL1-Blue by the standard procedures for DNA
manipula-Lion.
The nucleotide sequence of complete gene fusion was determined by the dideoxy chain termination method as described under section DNA sequencing. Both strands of the DNA were sequenced.
Preparation and purification of rMPT51.
Recombinant antigen was prepared from single colonies of E. colt harbouring the pT086 pfasmid inoculated into Luria-Bertani broth containing 50 Ng/ml ampicillin and 12.5 Ng/ml tetracycline and grown at 37°C to 2 x 1 O$ cellslml.
Isopropyl-(3-D-thioga-lactoside (IPTG) was then added to a final concentration of 1 mM and growth was continued for further 2 hours. The pelleted bacteria were resuspended in BC

buffer (100 mM KCI, 20 mM Imidazole, 20 mM Tris/HCI, pH 7.9, 20 % glycerol).
Cells were broken by sonication (20 times for 10 sec with intervals of 20 sect'.
After cen-trifugation at 9,000 x g for 30 min. at 4°C the insoluble matter was resuspended in BC 100/20 buffer with 8 M urea followed by sonication and centrifugation as above.
The 6 x His tag-MPTSi fusion protein (His-rMPT51 ) was purified by affinity chroma-tography on Ni-NTA resin column (Qiagen, Hilden, Germany). His-rMPT51 binds to Ni-NTA. After extensive washes of the column, the fusion protein was eluted with BC
100/40 buffer (100 mM KCI, 40 mM Imidazole, 20 mM Tris/HCI, pH 7.9, 20 % gly-cerol) with 8 M urea and BC 1000/40 buffer (1000 mM KCI, 40 mM Imidazole, 20 mM Tris/HCI, pH 7.9, 20 % glycerol) with 8 M urea. His-rMPT51 was extensive dia-lysed against 10 mM TrisiHCl, pH 8.5, 3 M urea followed by purification using fast protein liquid chromatography (FPLC) (Pharmacia, Uppsala, Sweden), over an anion exchange column (Mono Q) using 10 mM TrisiHCl, pH 8.5, 3 M urea with a 0 - 1 M
NaCI linear gradient. Fractions containing rMPT51 were pooled and subsequently dia-fysed extensively against 25 mM Hepes, pH 8.0 before use.
Protein concentration was determined by the BCA method supplied by Pierce (Pierce Chemical Company, Rockford, IL).
The iipopoiysaccharide (LPS) content was determined by the limulus amoebocyte ly-sate test (LAL) to be less than 0.004 nglug rMPT5l; and this concentration had no influence on cellular activity.
TABLE 4. Sequence of the mpt51 oligonucleotides.
Orientation and Sequences (5' --~ 3') Position oligonucleotide (nt) Sense Antisense Cloning of the ESA T6-MPT59 and the MPT59-ESA T6 hybrides.
Background for ESAT-MPT59 and MPT59-ESAT6 fusion -Several studies have demonstrated that ESAT-6 is a an immunogen which is relatively difficult to adjuvate in order to obtain consistent results when immunizing therewith.
To detect an in vitro recognition of ESAT-6 after immunization with the antigen is very difficult compared to the strong recognition of the antigen that has been found during the recall of memory immunity to M. tuberculosis. ESAT-6 has been found in ST-CF in a truncated version were amino acids 1-15 have been deleted. The deletion includes the main T-cell epitopes recognized by C57BLi6j mice (Brandt et al., 1996).
This result indicates that ESAT-6 either is N-terminally processed or proteolytically degraded in STCF. In order to optimize ESAT-6 as an immunogen, a gene fusion between ESAT-5 and another major T cell antigen MPT59 has been constructed. Two different con-struct have been made: MPT59-ESAT-6 (SEQ ID NO: 172) and ESAT-6-MPT59 iSEQ
ID NO: 173). In the first hybrid ESAT-6 is N-terminally protected by MPT59 and in the latter it is expected that the fusion of two dominant T-cell antigens can have a syner-gistic effect.
The genes encoding the ESAT6-MPT59 and the MPT59-ESAT6 hybrides were cloned into the expression vector pMCT6, by PCR amplification with gene specific primers, for recombinant expression in E. coli of the hybrid proteins.
Construction of the hybrid MPT59-ESAT6.
The cloning was carried out in three steps. First the genes encoding the two cornpo-nests of the hybrid, ESAT6 and MPT59, were PCR amplified using the following primer constructions:
ESAT6:
OPBR-4: GGCGCCGGCAAGCTTGCCATGACAGAGCAGCAGTGG
(SEQ ID NO: 132) OPBR-28: CGAACTCGCCGGATCCCGTGTTTCGC (SEQ ID NO: 133) OPBR-4 and OPBR-28 create HinDlll and BamHl sites, respectively.
MPT59:
OPBR-48: GGCAACCGCGAGATCTTTCTCGCGGCCGGGGC (SEQ tD NO: 134) OPBR-3: GGCAAGCTTGCCGGCGCCTAACGAACT (SEQ ID NO: 135) OPBR-48 and OPBR-3 create Bglll and HinDlll, respectively. Additionally OPBR-3 de-letes the stop codon of MPT59.
PCR reactions contained 10 ng of M. tuberculosis chromosomal DNA in 1 x low salt Taq+ buffer from Stratagene supplemented with 250 mM of each of the four nucleo-tides (Boehringer Mannheim), 0,5 mg/ml BSA (IgG technology), 1 % DMSO (Merck), pmoles of each primer and 0.5 unit Tag + DNA polymerase (Stratagene) in 10 u!
reac-tion volume. Reactions were initially heated to 94°C for 25 sec. and run for 30 cycles of the program; 94°C for 10 sec., 55°C for 10 sec. and 72°C for 90 sec, using ther-mocycler equipment from Idaho Technology.
The DNA fragments were subsequently run on 1 % agarose gels, the bands were ex-cised and purified by Spin-X spin columns (Costar). The two PCR fragments were di-gested with HinDlll and ligated. A PCR amplification of the iigated PCR
fragments en-coding MPT59-ESAT6 was carried out using the primers OPBR-48 and OPBR-28. PCR
reaction was initially heated to 94°C for 25 sec. and run for 30 cycles of the pro-gram; 94°C for 30 sec., 55°C for 30 sec. and 72°C for 90 sec. The resulting PCR
fragment was digested with Bglil and BamHl and cloned into the expression vector pMCT6 in frame with 8 histidines which are added to the N-terminal of the expressed protein hybrid. The resulting clones were hereafter sequenced by use of the dideoxy chain termination method adapted for supercoiled DNA using the Sequenase DNA
se-quencing kit version 1.0 (United States Biochemical Corp., USA) and by cycle se-quencing using the Dye Terminator system in combination with an automated gel reader (model 373A; Applied Biosystems) according to the instructions provided. Both strands of the DNA were sequenced.
Construction of the hybrid ESAT6-MPT59.
Construction of the hybrid ESAT6-MPT59 was carried out as described for the hybrid MPT59-ESAT6. The primers used for the construction and cloning were:
ESAT6:
OPBR-75: GGACCCAGATCTATGACAGAGCAGCAGTGG (SEQ ID NO: 136) WO 99124577 PCTlDK98/00438 OPBR-76: CCGGCAGCCCCGGCCGGGAGAAAAGCTTTGCGAACATCCCAGTGACG
(SEQ ID NO: 137) OPBR-75 and OPBR-76 create Bglll and HinDlll sites, respectively. Additionally OPBR-76 deletes the stop colon of ESAT6.
MPT59:
OPBR-77: GTTCGCAAAGCTTTTCTCCCGGCCGGGGCTGCCGGTCGAGTACC
(SEQ 1D NO: 138) OPBR-18: CCTTCGGTGGATCCCGTCAG (SEQ iD NO: 139) OPBR-77 and OPBR-18 create HinDlll and BamHl sites, respectively.
Expression/purification of MPT59-ESAT6 and ESAT6-MPT59 hybrid proteins.
Expression and metal affinity purification of recombinant proteins was undertaken es-sentially as described by the manufacturers. For each protein, 1 I LB-media containing 100 ~uglmt ampicillin, was inoculated with 10 ml of an overnight culture of XL1-Blue cells harbouring recombinant pMCT6 plasmids. Cultures were shaken at 37 °C until they reached a density of ODsoo = 0.4 - 0.6. IPTG was hereafter added to a final con-centration of 1 mM and the cultures were further incubated 4 - 16 hours. Cells were harvested, resuspended in 1 X sonication buffer + 8 M urea and sonicated 5 X

sec. with 30 sec. pausing between the pulses.
After centrifugation, the lysate was applied to a column containing 25 ml of resus-pended Talon resin (Clontech, Palo Alto, USA). The column was washed and eluted as described by the manufacturers.
After elution, all fractions (1.5 ml each) were subjected to analysis by SDS-PAGE us-ing the Mighty Small (Hoefer Scientific Instruments, USA) system and the protein con-centrations were estimated at 280 nm. Fractions containing recombinant protein were pooled and dialysed against 3 M urea in 10 mM Tris-HCI, pH 8.5. The dialysed protein was further purified by FPLC (Pharmacia, Sweden) using a 6 ml Resource-Q
column, eluted with a linear O-1 M gradient of NaCI. Fractions were analyzed by SDS-PAGE
and protein concentrations were estimated at oD28o. Fractions containing protein were pooled and dialysed against 25 mM Hepes buffer, pH 8.5.
Finally the protein concentration and the LPS content were determined by the BCA
(Pierce, Holland) and LAL (Endosafe, Charleston, USA) tests, respectively.
The biological activity of the MPT59-ESAT6 fusion protein is described in Example 6A.

Mapping of the purified antigens in a 2DE system.
In order to characterize the purified antigens they were mapped in a 2-dimensional electrophoresis (2DE) reference system. This consists of a silver stained gel containing ST-CF proteins separated by isoelectrical focusing followed by a separation according to size in a polyacrylamide gel electrophoresis. The 2DE was performed according to Hochstrasser et al. (1988). 85 ug of ST-CF was applied to the isoelectrical focusing tubes where BioRad ampholytes BioLyt 4-6 (2 parts) and BioLyt 5-7 (3 parts) were in-ciuded. The first dimension was performed in acrylamide/piperazin diacrylamide tube gels in the presence of urea, the detergent CHAPS and the reducing agent DTT
at 400 V for 18 hours and 800 V for 2 hours. The second dimension 10-20% SDS-PAGE was performed at 100 V for 18 hours and silver stained. The identification of CFP7, CFP7A, CFP7B, CFPBA, CFPBB, CFP9, CFP11, CFP16, CFP17, CFP19, CFP20, CFP21, CFP22, CFP25, CFP27, CFP28, CFP29, CFP30A, CFP50, and MPT51 in the 2DE reference gel were done by comparing the spot pattern of the purified antigen with ST-CF with and without the purified antigen. By the assistance of an analytical 2DE software system (Phoretix International, UK) the spots have f?een identified in Fig.
6. The position of MPT51 and CFP29 were confirmed by a Western blot of the 2DE
gel using the Mab's anti-CFP29 and HBT 4.

Biological activity of the purified antigens.
IFN-y induction in the mouse model of TB infection The recognition of the purified antigens in the mouse model of memory immunity to TB (described in example 1 ) was investigated. The results shown in TABLE 5 are re-presentative for three experiments.
A very high IFN-y response was induced by two of the antigens CFP17 and CFP21 at almost the same high level as ST-CF.

IFN-y release from splenic memory effector cells from C57BL16J mice isolated after reinfection with M. tuberculosis after stimulation with native antigens.
Antigena IFN-y (pg/ml)b CFP7 NDd CFP22 + CFP25~ 5342 CFP26 (MPT51) ND

The data is derived from a representative experiment out of three.
a ST-CF was tested in a concentration of 5 pg/ml and the individual antigens in a con-centration of 2 wg/ml.

° Four days after rechallenge a pool of cells from three mice were tested. The results are expressed as mean of duplicate values and the difference between duplicate cul-tures are < 15% of mean. The IFN-y release of cultures incubated without antigen was 390 pg/ml.
' A pool of CFP22 and CFP25 was tested.
d ND, not determined. -Skin test reaction in TB infected guinea pigs The skin test activity of the purified proteins was tested in M. tuberculosis infected guinea pigs.
1 group of guinea pigs was infected via an ear vein with 1 x 10' CFU of M.
tubercu-losis H37Rv in 0,2 ml PBS. After 4 weeks skin tests were performed and 24 hours after injection erythema diameter was measured.
As seen in TABLES 6 and 6a all of the antigens induced a significant Delayed Type Hypersensitivity (DTH) reaction.

DTH erythema diameter in guinea pigs infected with 1 x 10° CFU of M.
tuberculosis, after stimulation with native antigens.
Antigens Skin reaction (mm)b Control 2.00 PPD~ 15.40 (0.53) CFP 7 NDe CFP17 11.25 (0.84) CFP20 8.88 (0..13) CFP21 12.44 (0.79) CFP22 + CFP25d 9.19 (3.10) CFP26 (MPT51) ND

CFP28 2.90 (1.28) CFP29 6.63 (0.88) The values presented are the mean of erythema diameter of four animals and the SEM's are indicated in the brackets. For PPD and CFP29 the values are mean of ery-thema diameter of ten animals.
a The antigens were tested in a concentration of 0,1 ~g~ except for CFP29 which was tested in a concentration of 0,8 fig.
b The skin reactions are measured in mm erythema 24 h after intradermat injection.
' i 0 TU of PPD was used.
A pool of CFP22 and CFP25 was tested.
a ND, not determined.
Together these analyses indicate that most of the antigens identified were highly bio-logically active and recognized during TB infection in different animal models.

TABLE 6a DTH erythema diameter of recombinant antigens in outbred guinea pigs infected with 1 x 10° CFU of M. Tuberculosis.
Antigens Skin reaction (mm}b Control 2.9 (0.3) PPD 14.5 (1.0) CFP 7a 13.6 (1.4) CFP 17 6.8 (1.9) CFP 20 6,.4 . (1.4) CFP 21 5.3 (0.7) CFP 25 10.8 (0.8) CFP 29 7.4 (2.2) MPT 51 4 . 9 ( 1 . 1 ) The values presented are the mean of erythema diameter of four animals and the SEM's are indicated in the brackets. For Control, PPD, and CFP 20 the values are mean of erythema diameter of eight animals.
a The antigens were tested in a concentration of 1,0 ug.
The skin test reactions are measured in mm erythema 24 h after intradermal infec-tion.
10 TU of PPD was used.
Table 6B.
DTH erythema diameter in guinea pigs i.v. infected with 1 x i 04 CFU M.
tuberculo-sis, after stimulation with 10 g antigen.
Antigen Mean (mm) SEM
PBS 3,25 0,48 PPD (2TU) 10,88 1 nCFP7B 7,0 0,46 nCFPl9 6,5 0,74 MPT59-ESAT6 14,75 I,5 The values presented are the mean of erythema diameter of four animals.
The results in Table 6B indicates biological activity of nCFP7B, nCFP19 and ESAT-6. MPT59-ESAT-6 resulting in a DTH response at the level of PPD.
Biological activity of the purified recombinant antigens.
Interferon-Y induction in the mouse model of TB infection.
Primary infections. 8 to 12 weeks old female C57BU6jtH-2b), CBAIJ(H-2k), DBA.2iH-2d) and A.SWiH-25) mice (Bomholtegaard, Ry) were given intravenous infections via the lateral tail vein with an inoculum of 5 x 104 M. tuberculosis suspended in PBS in a vol. of 0.1 ml. 14 days postinfection the animals were sacrificed and spleen cells were isolated and tested for the recognition of recombinant antigen.
As seen in TABLE 7 the recombinant antigens rCFP7A, rCFP17, rCFP21, rCFP25, and rCFP29 were all recognized in at least two strains of mice at a level comparable to ST-CF. rMPT51 and rCFP7 were only recognized in one or two strains respectively, at a level corresponding to no more than 1 /3 of the response detected after ST-CF
stimula-tion. Neither of the antigens rCFP20 and rCFP22 were recognized by any of the four mouse strains.
As shown in TABLE 7A, the recombinant antigens rCFP27, RD1-ORF2, MPT59-ESAT6, rCFPIOA, rCFPl9, and rCFP25A were all recognized in at least two strains of mice at a level comparable to ST-CF, whereas ESAT6-MPT59, rCFP23, and rCFP30B
only were recognized in one strain at this level. rCFP30A , RD1-ORFS, rCFP16 gave rise to an IFN-y release in two mice strains at a level corresponding to 2/3 of the re-sponse after stimulation with ST-CF. RD1-ORF3 was recognized in two strains at a level of 1/3 of ST-CF.
The native CFP7B was recognized in two strains at a level of 1/3 of the response seen after stimulation with ST-CF.

WO 99lZ4577 PCTIDK98/00438 Memory responses. 8-12 weeks old female C57BL/6j(H-2b) mice (Bomholtegaard, Ry) were given intravenous infections via the lateral tail vein with an inoculum of 5 x 10°
M. tuberculosis suspended in PBS in a vol. of 0.1 ml. After 1 month of infection the mice were treated with isoniazid (Merck and Co., Rahway, NJ) and rifabutin (Farma-talia Carlo Erba, Milano, Italy) in the drinking water, for two months. The mice were rested for 4-6 months before being used in experiments. For the study of the recall of memory immunity, animals were infected with an inoculum of 1 x 106 bacteria i.v. and sacrificed at day 4 postinfection. Spleen cells were isolated and tested for the recogni-tion of recombinant antigen.
As seen from TABLE 8, IFN-y release after stimulation with rCFP17, rCFP21 and rCFP25 was at the same level as seen from spleen cells stimulated with ST-CF.
Stimulation with rCFP7, rCFP7A and rCFP29 all resulted in an IFN-y no higher than 1 /3 of the response seen with ST-CF. rCFP22 was not recognized by IFN-y producing cells. None of the antigens stimulated iFN-y release in naive mice.
Additionally non of the antigens were toxic to the cell cultures.
As shown in TABLE 8A, IFN-y release after stimulation with RD1-ORF2, MPT59-ESAT6, ESAT6-MPT59, and rCFP19 was at the same level as seen from spleen cells stimulated with ST-CF. Stimulation with rCFPIOA and rCFP30A gave rise to an IFN-y release of 213 of the response after stimulation with ST-CF, whereas rCFP27, ORFS, rCFP23, rCFP25A and rCFP30B all resulted in an IFN-Y release no higher than 1/3 of the response seen with ST-CF. RD1-ORF3 and rCFPl6 were not recognized by IFN-y producing memory cells.

WO 99124577 PC'T1DK98100438 TABLE 7. T cell responses in primary TB infection.
Name c57BL/6J (H2b) DBA. 2 (H2d) CBA/J (H2k) A. SW (H2S
rCFP7 + + -rCFP7A +++ +++ +++ +

rCFPl7 +++ + +++ -t rCFP20 - - - -rCFP21 +++ +++ +++ +

rCFP22 - - - -rCFP25 +++ ++ +++ +

rCFP29 +++ +++ ++_ . ++

rMPT51 + - - -Mouse IFN-y release 14 days after primary infectiowrvith M. tuberculosis.
-:no response; + : 1 I3 of ST-CF; + + : 2/3 of ST-CF; + + + : level of ST-CF.

TABLE 7A. T cell responses in primary TB infection.
Name C57B1 6j (H2 DBA. 2 (H2) CBA/J (H2") A. SW (H2 ) ) rCFP27 ++ ++ +++ +++ ' rCFP30A - + ++ ++

RDl-ORF2 +++ +++ +++ _ ++

RDl-ORF3 - - + +

RD1-ORFS + + ++ ++

MPT59-ESAT6 +++ +++ +++ ++

ESAT6-MPT59 +++ - + -rCFPIOA +++ n.d. +++ n.d.

rCFPl6 ++ n.d. -~+ n.d.

rCFPl9 +++ n.d. +++ n.d.

rCFP23 ++ n.d. +++ n.d.

rCFP25A +++ n.d. +++ n.d.

rCFP30B + n.d. +++ n.d.

CFP7B(native) + n.d. + n.d.

Mouse IFN-y release 14 days after primary infection with M. tuberculosis.
no response; + : 113 of ST-CF; + + : 213 of ST-CF; + + + : level of ST-CF.
n.d. = not determined.

TABLE 8. T cell responses in memory immune animals.
Name Memory response rCFP7 +

rCFP7A ++

rCFPl7 +++

rCFP21 +++

rCFP22 -rCFP29 +

rCFP25 +++

rMPT51 +

Mouse IFN-y release during recall of memory immunity to M. tuberculosis.
-:no response; + : 1 /3 of ST-CF; + + : 2/3 of ST-CF; + + + : level of ST-CF.
TABLE 8A. T cell responses in memory immune animals.
Name Memory response rCFP27 +

rCFP30A ++

RD1-ORF2 +++

RD1-ORF5 +

MPT59-ESAT6 +++

ESAT6-MPT59 +++

rCFPIOA ++

rCFPl6 -rCFPl9 +++

rCFP23 +

rCFP25A +

rCFP30B +

Mouse IFN-y release during recall of memory immunity to M. tuberculosis.
-: no response; + : 1 i3 of ST-CF; + + : 2/3 of ST-CF; + + + : level of ST-CF.

WO 99/24577 PC'T/DK98/00438 Interferon-y induction in human TB patients and BCG vaccinated people.
Human donors: PBMC were obtained from healthy BCG vaccinated donors with no v known exposure to patients with TB and from patients with culture or microscopy ' proven infection with Mycobacterium tuberculosis. Blood samples were drawn from the TB patients 1-4 months after diagnosis.
Lymphocyte preparations and cell culture: PBMC were freshly isolated by gradient cen-trifugation of heparinized blood on Lymphoprep (Nycomed, Oslo, Norway). The cells were resuspended in complete medium: RPMI 1640 (Gibco, Grand Island, N.Y.) sup-plemented with 40 ~rglml streptomycin, 40 U/ml penicillin, and 0.04 mM/ml glutamine, (all from Gibco Laboratories, Paisley, Scotland) and 10% normal human ABO
serum (NHS) from the local blood bank. The number and the viability of the cells were de-termined by trypan blue staining. Cultures were established with 2,5 x 105 PBMC in 200 ~ul in microtitre plates (Nunc, Roskilde, Denmark) and stimulated with no antigen, ST-CF, PPD (2.5Ng1m1); rCFP7, rCFP7A, rCFP17, rCFP20, rCFP21, rCFP22, rCFP25, rCFP26, rCFP29, in a final concentration of 5 Ng/ml. Phytohaemagglutinin, 1 Ng/ml (PHA, Difco laboratories, Detroit, MI. was used as a positive control.
Supernatants for the detection of cytokines were harvested after 5 days of culture, pooled and stored at -80°C until use.
Cytokine analysis: Interferon-y (IFN-y ) was measured with a standard ELISA
tech-nique using a commercially available pair of mAb's from Endogen and used according to the instructions for use. Recombinant lFN-y (Gibco laboratories) was used as a standard. The detection level for the assay was 50 pg/ml. The variation between the duplicate wells did not exceed 10 % of the mean. Responses of 9 individual donors are shown in TABLE 9.
A seen in TABLE 9 high levels of IFN-y release are obtained after stimulation with sev-eral of the recombinant antigens. rCFP7a and rCFPl7 gives rise to responses compa-rable to STCF in almost all donors. rCFP7 seems to be most strongly recognized by BCG vaccinated healthy donors. rCFP21, rCFP25, rCFP26, and rCFP29 gives rise to a mixed picture with intermediate responses in each group, whereas low responses are obtained by rCFP20 and rCFP22.

As is seen from Table 9A RD1-ORF2 and RD1-ORES give rise to IFN-y responses close to the level of ST-CF. Between 60% and 90% of the donors show high IFN-y respon-ses (> 1000 pglml). rCFP30A gives rise to a mixed response with 40-50% high re-sponders, whilst low responses are obtained with RD1-ORF3.
As seen from Table 9B MPT59-ESAT6 and ESAT6-MPT59 both give rise to IFN-y re-sponses at the level of ST-CF and 67-89 % show high responses ( > 1000 pg/ml).

Y
C

U N lIla0 r-i N dltf1V~M

C4 lD01N 00 ~1CO~ a0U1 Cc. lflO O N .-I fs-rU1.-1N D1 U aoN ~ ~ ao U ~rW n a~

M

Y
N lD N COO ,-) C far tolDr d' (~t~r N M

Cxa d'N O O r ~ O M v-iM

'+- U omnooN o~ U ~

Q

1~ ~ M

N r .-i~ r1 N O N a1 CL NM M ~ o farO COl0O

ti. 00~ InM ~-1 CLVIO O O

U '-1.-iN r1 N U N M N ri C N N

N N

W a~ W t' U Cxr Mr-il0 . [t,00l0M C

U rIn~ '-r~ U ~ ~ ~r C

v z ,~ r U tL N O~V' 00 N '-1LI1 r c0- (3r NV'01d' O C1rM M r In pI IL tf1'-ir-iOO O Lia'-1M O 01 d U r'tlpM N M U ,-~

U

m .C O O

>-C N N

L N ~ r CIrV' 01 Ci-1a001(h Cz.va0r-IrIr-1 U u~N a~r-+~ U N ~-1~ a~

al ~C 01 In~ ~ ~ a~r ~ .--rN r amn ~ .-a O r o cn C~. ar o~~ ~r.-ao W ~ o m M

'"' X Cia rN ~DN U1 CLO t~M lD

v U mn oom ,~ U v<<rM

v-N

N
- r I~ V'1, r-i V'l001 CO f-iO lI)M M

U C ~ LL O t>7M r1 C~rtf1r I11 0 g Cs. a~o ,-tav o CL1N M t~r U vDr-i.-1,-IaDU7 U d~~ N o1 . _.-a C
cp 'n ~ r o r VWD u1t')N

~ tarN aod~II1 O

~ ~ ~T'u1l~O d~
U

U M oDrl .-i U co~--1rico M

p = ~ G4 l0I~01r N C~alIlM LIlrl ~ r~ U m~ o~M o ~ U cro~t~N

c O Li [-i 01O N tf1t'~1~ E-~r-1M M r O Cl) 1'~10000M ri~ V1~DM I~r-1 U

r L o 01 Cr1O ~O O
~Dr-1v--In-I

rO O O N
Q

nr rW o .-~a ~ Aro N ~o a. ~om ,~~r ,-~ A,Lmo r ~

~ , E

W o o~ov o _ M u~o cca > ~ ~ ~

. ~ ~~ m o ~ , ~ ~ ~
-I ' o ~ x InN r-i.-1a0~ 61N ~ N

"'a ~ Cl~ 01.-1r-iN r-i Clra0W -IN

to U ~ rn bl U x ~ ~ ~

o O a0O1O
N O N

-ri -Sir01~ ~f'f'1 . 1J

tLf C~r .Ln ~ ~

Q C O O al I-'(13 U T.y -1N M V' U1E-1 \Dr 40dl ~

CSI IOC~TITI ITf- C~L11-'~

Table 9A_ Results from the stimulation of human blood cells from 10 healthy BCG
vaccinated or non vaccinated ST-CF responsive healthy donors and 10 Tb patients with recombinant antigen are shown. ST-CF, PPD and PHA are included for comparison. Results are given in pg IFN-y/ml and negative values below 300 pg/ml are shown as " < ". nd = not done.
Controls, Healthy BCG vaccinated, or non vaccinated ST-CF positive Donor no ag PHA PPD STCF RDl-ORF2 RD1-ORF3 RDl-ORFS rCFP30A

< nd 3500 4200 1250 < 690 nd 11 < nd 5890 4040 5650 880 9030 nd 12 < nd 6480 3330 2310 < 3320 nd 13 < nd 7440 4570 920 < 1230 nd 14 < 8310 nd 2990 1870 < 4880 <

< 10820 nd 4160 5690 < 810 3380 16 < 8710 nd 5690 1630 < 5600 <

17 < 7020 4480 5340 2030 nd 670 <

18 < 8370 6250 4780 3850 nd 370 1730 19 < 8520 1600 310 5110 nd 2330 1800 10 Tb patients, 1-4 month after diagnosis Donor no ag PHA PPD STCF RD1-ORF2 RD1-ORF3 RD1-ORF5 rCFP30A

- -c nd 10670 12 2020 c 9670 nd 21 < nd 3010 1420 850 < 350 nd 22 < nd 8450 7850 430 < 1950 nd 23 < 10060 nd 3730 < < 350 <

24 < 10830 nd 6180 2090 < 320 730 < 9000 nd 3200 4760 < 4960 2820 26 < 10740 nd 7650 4710 < 1170 2280 27 < 7550 6430 6220 2030 nd 3390 3069 28 < 8090 5790 4850 1100 nd 2095 550 29 < 7790 4800 4260 2800 nd 1210 420 Table 9B.
Results from the stimulation of human blood cells from 9 Healthy BCG
vaccinated, or -non vaccinated ST-CF positive and 8 Tb patients with recombinant MPT59-ESAT6 and ESAT6-MPT59 are shown. ST-CF, PPD and PHA are included for comparison. Results are given in pg IFN-ylml and negative values below 300 pg/ml are shown as " <
". nd = not done. -Controls, Healthy BCG vaccinated, or non vaccinated ST-CF positive.
Donor no ag PHA PPD STCF MPT59- ESAT6-1 < 9560 6770 3970 2030 <

2 < 12490 6600 8070 5660 5800 4 < 21030 4100 3540 < <

5 < 18750 14200 13030 8540 <

11 < nd 5890 4040 4930 8870 12 < nd 6470 3330 2070 6450 14 < 8310 nd 2990 10270 11030 < 10830 nd 4160 3880 4540 16 < 8710 nd 5690 2240 5820 10 Tb patients, 1-4 month after diagnosis Donor no ag PHA PPD STCF MPT59- ESAT6-6 < 8970 5100 6150 4150 4120 7 < 12410 6280 3390 5050 2040 8 < 11920 7670 7370 800 1350 9 < 22130 16420 17210 13660 5630 23 < 10070 nd 3730 1740 2390 24 < 10820 nd 6180 1270 1570 < 9010 nd 3200 3680 5340 26 < 10740 nd 7650 2070 620 Four groups of 6-8 weeks old, female C578116J mice (Bomholtegard, Denmark) were immunized subcutaneously at the base of the tail with vaccines of the following com-positions:
Group 1: 10 Ng ESAT-fi/DDA (250,ug) Group 2: 10 ~g MPT59lDDA (250Ng) Group 3: 10 Ng MPT59-ESAT-6 /DDA (250 Ng) Group 4: Adjuvant control group: DDA (250 fig) in NaCI
The animals were injected with a volume of 0.2 ml. Two weeks after the first injection and 3 weeks after the second injection the mice were boosted a little further up the back.
One week after the last immunization the mice were bled and the blood cells were isolated. The immune response induced was monitored by release of IFN-y into the cul-ture supernatants when stimulated in vitro with relevant antigens (see the following table).
Immunogen For restimulation Ag vitro . in 10 pg/dose no antigen ST-CF ESAT-6 MPT59 Hybrid: 127 127 7453 581 15133 861 16363 a' Blood cells were isolated 1 week after the last immunization and the release of IFN-y (pglml) after 72h of antigen stimulation (5 ~rg/ml) was measured.
The values shown are mean of triplicates performed on cells pooled from three mice ~ SEM
°' - not determined The experiment demonstrates that immunization with the hybrid stimulates T
cells which recognize ESAT-6 and MPT59 stronger than after single antigen immunization.
Especially the recognition of ESAT-6 was enhanced by immunization with the ESAT-6 hybrid. IFN-y release in control mice immunized with DDA never exceeded 1000 pg/mf.

The recombinant antigens were tested individually as subunit vaccines in mice.
Eleven groups of 6-8 weeks old, female C57B116j mice (Bomholtegard, Denmark) were immu-nized subcutaneously at the base of the tail with vaccines of the following composi-tion:
Group 1: 10 Ng CFP7 Group 10 ~g CFP17 2:

Group 3: 10 Ng CFP21 Group 4: 10 Ng CFP22 Group 5: l0lrg CFP25 Group 6: 10 Ng CFP29 Group 10 Ng MPT51 7:

Group 8: 50 dug ST-CF

Group 9: Adjuvant control group Group 10: BCG 2,5 x 1051m1, 0,2 ml Group 1 Control group: Untreated 1:

All the subunit vaccines were given with DDA as adjuvant. The animals were vacci-nated with a volume of 0.2 ml. Two weeks after the first injection and three weeks after the second injection group 1-9 were boosted a little further up the back. One week after the last injection the mice were bled and the blood cells were isolated. The immune response induced was monitored by release of IFN-y into the culture super-natant when stimulated in vitro with the homologous protein.
fi weeks after the last immunization the mice were aerosol challenged with 5 x viable Mycobacterium tuberculosislml. After 6 weeks of infection the mice were killed and the number of viable bacteria in lung and spleen of infected mice was determined by plating serial 3-fold dilutions of organ homogenates on 7H11 plates.
Colonies were counted after 2-3 weeks of incubation. The protective efficacy is expressed as the dif-ference between 1 og,o values of the geometric mean of counts obtained from five mice of the relevant group and the geometric mean of counts obtained from five mouse of the relevant control group.
The results from the experiments are presented in the following table.
Immunogenicity and protective efficacy in mice, of ST-CF and 7 subunit vaccines Subunit Vaccine Immunogenicity Protective efficacy ST-CF +++ +++

CFP7 ++ -CFP17 +++ +++

CFP21 +++ ++

CFP25 +++ +++

CFP29 +++ +++

MPT51 +++ ++

+ + + Strong immunogen / high protection (level of BCG) + + Medium immunogen / medium protection - No recognition / no protection In conclusion, we have identified a number of proteins inducing high levels of protec-tion. Three of these CFP17, CFP25 and CFP29 giving rise to similar levels of protec-tion as ST-CF and BCG while two proteins CFP21 and MPT51 induces protections around 2/3 the level of BCG and ST-CF. Two of the proteins CFP7 and CFP22 did not induce protection in the mouse model.
As is described for rCFP7, rCFP17, rCFP21, rCFP22, rCFP25, rCFP29 and rMPT51 the two antigens rCFP7A and rCFP30A were tested individually as subunit vaccines in mice. C57BI/6j mice were immunized as described for rCFP7, rCFP17, rCFP21, rCFP22, rCFP25, rCFP29 and rMPT51 using either 10~g rCFP7A or 10~g rCFP30A.
Controls were the same as in the experiment including rCFP7, rCFP17, rCFP21, rCFP22, rCFP25, rCFP29 and rMPT5l.

Immunogenicity and protective efficacy in mice, of ST-CF and 2 subunit vaccines.
Subunit vaccine Immunogenicity Protective efficacy ST-CF +++ +++
rCFP7A +++ +++
rCFP30A +++ -+ + + Strong immunogen/high protection (level of BCG) + + Medium immunogen/medium protection - No recognition/no protection In conclusion we have identified two strong immunogens of~which one, rCFP7A, in-duces protection at the level of ST-CF.

Species distribution of cfp7, cfp9, mpt5l, rd 1-orf2, rd 1-orf3, rd 1-orf4, rd 1-orf5, rd 1-orf8, rd 1-orf9a and rd 1-orf9b as well as of cfp7a, cfp7b, cfp lOa, cfp 17, cfp20, cfp2l, cfp22, cfp22a, cfp23, cfp25 and cfp25a.
Presence of cfp7 cfp9 mpt51 rd 1-orf2 rd 1-orf3 rd 1-orf4 rd 1-orf5 rd 1-orf8, rd 1-orf9a and rdl-orf96 in different mycobacterial species.
In order to determine the distribution of the cfp7, cfp9, mpt51, rdl-orf2, rdl-orf3, rd 1-orf4, rd 1-orf5, rd 1-orf8, rd 1-orf9a and rd 1-orf9b genes in species belonging to the M. tuberculosis-complex and in other mycobacteria PCR and/or Southern blotting was used. The bacterial strains used are listed in TABLE 10. Genomic DNA was prepared from mycobacterial cells as described previously (Andersen et al. 1992). "
, PCR analyses were used in order to determine the distribution of the cfp7, cfp9 and mpt51 gene in species belonging to the tuberculosis-complex and in other mycobacte- , ria. The bacterial strains used are listed in TABLE 10. PCR was performed on genomic DNA prepared from mycobacterial cells as described previously (Andersen et al., 1992).

The oligonucleotide primers used were synthesised automatically on a DNA
synthe-sizer (Applied Biosystems, Forster City, Ca, ABI-391, PCR-mode), deblocked, and puri-' fied by ethanol precipitation. The primers used for the analyses are shown in TABLE
11.
The PCR amplification was carried out in a thermal reactor (Rapid cycler, Idaho Tech-nology, Idaho) by mixing 20 ng chromosomal with the mastermix (contained 0.5 NM
of each oligonucleotide primer, 0.25 NM BSA (Stratagene), low salt buffer (20 mM
Tris-HCI, pH8.8 , 10 mM KCI, 10 mM (NHQ)ZS04, 2 mM MgS04 and 0,1 % Triton X-100) (Stratagene), 0.25 mM of each deoxynucleoside triphosphate and 0.5 U Taq Plus Long DNA polymerase (Stratagene)). Final volume was 10 girl (all concentrations given are concentrations in the final volume). Predenaturation was carried out at 94°C for 30 s. 30 cycles of the following was performed: Denaturation at 94°C
for 30 s, annealing at 55°C for 30 s and elongation at 72°C for 1 min.
The following primer combinations were used (the length of the amplified products are given in parentheses):
mpt5l: MPT51-3 and MPT51-2 (820 bp), MPT51-3 and MPT51-6 (108 bp), MPT51-5 and MPT51-4 (415 bp), MPT51-7 and MPT51-4 (325 bp).
cfp7: pVF1 and PVR1 (274 bp), pVF1 and PVR2 (197 bp), pVF3 and PVR1 (302 bp), pVF3 and PVR2 (125 bp).
cfp9: stR3 and stF1 (351 bp).

WO 99rZ4577 PCTIDK98/00438 TABLE 10_ Mycobacterial strains used in this Example.
Species and strains) Source 1. M. tuberculosis H37Rv ATCC°
(ATCC 27294) .
2_ H37Ra ATCC _ (ATCC 25177}
3, Erdman Obtained from A. Lazlo, Ottawa, Canada 4. M. bovis BCG substrain:Danish SSIb 5. Chinese SSI' 6_ Canadian SSI' 7, Glaxo SSI' g_ Russia SSI' 9. Pasteur SSI' 10. Japan WHOe 11. M. bovis MNC 27 SSI' 12. M. africanum Isolated from a Danish patient 13. M. leprae (armadillo-derived} Obtained from J. M. Colston, London, UK
14. M. avium (ATCC 15769} ATCC
15. M. kansasii (ATCC 12478) ATCC
16. M. marinum (ATCC 927) ATCC
17. M. scrofulaceum (ATCC 19275) ATCC
18. M. interce11u1are (ATCC 15985} ATCC
19. M. fortuitum (ATCC 6841} ATCC

20. M. xenopi Isolated from a Danish patient 21. M. flavescens Isolated from a Danish patient 22. M. szulgai Isolated from a Danish patient 23. M. terrae SSI' 24. E. cola SSI°

25. S.aureus SSId a American Type Culture Collection, USA.
b Statens Serum institut, Copenhagen, Denmark.
' Our collection Department of Mycobacteriology, Statens Serum Institut, Copenha-gen, Denmark.
Department of Clinical Microbiology, Statens Serum Institut, Denmark.
a WHO International Laboratory for Biological Standards, Statens Serum Institut, Co-penhagen, Denmark.

TABLE 1 1.
Sequence of the mpt5l , cfp7 and cfp9 oligonucleotides.
Orientation and Sequences (5'--~3')a Position°
oligonucleotide (nucleotides) Sense 1 (SEQ ID NO: 28) (SEQ ID NO:
41 ) 3 (SEQ ID NO: 29) (SEQ ID NO:
41 ) (SEQ ID NO: 30) (SEQ ID NO:
41 ) 7 (SEQ ID NO: 31 ) (SEQ ID NO:
41 ) pvR1 GTACGAGAATTCATGTCGCAAATCATG ' 91 - 105 (SEQ ID NO: 35) (SEQ ID NO:
1) pvR2 GTACGAGAATTCGAGCTTGGGGTGCCG 168 - 181 (SEQ lD NO: 36) (SEQ ID NO:
1 ) stR3 CGATTCCAAGCTTGTGGCCGCCGACCCG 141 - 155 (SEO ID NO: 37) (SEQ ID NO:
3) Antisense 2 (SEQ ID NO: 32) (SEQ ID NO:
41) 4 (SEQ ID NO: 33) (SEQ ID NO:
41 ) 6 (SEQ ID NO: 34) (SEQ ID NO:
41 ) pvF1 CGTTAGGGATCCTCATCGCCATGGTGTTGG 340 - 323 (SEQ ID NO: 38) ~ (SEQ ID NO:
1 ) pvF3 CGTTAGGGATCCGGTTCCACTGTGCC 268 - 255 (SEQ ID NO: 39) (SEQ ID NO:
1 ) stF1 CGTTAGGGATCCTCAGGTCTTTTCGATG 467 - 452 (SEQ ID NO: 40) (SEQ ID NO:
3) a Nucleotides underlined are not contained in the nucleotide sequences of mpt51, cfp7, and cfp9.
b The positions referred to are of the non-underlined parts of the primers and corre-spond to the nucleotide sequence shown in SEQ ID NOs: 41, 1, and 3 for mpt51, cfp7, and cfp9, respectively.
The Southern blotting was carried out as described previously (Oettinger and Ander-sen, 1994) with the following modifications: 2 erg of genomic DNA was digested with Pvull, electrophoresed in an 0.8% agarose gel, and transferred onto a nylon membrane WO 99124577 PCTlDK98100438 (Hybond N-plus; Amersham International plc, Little Chalfont, United Kingdom) with a vacuum transfer device (Milliblot, TM-v; Millipore Corp., Bedford, MA). The cfp7, cfp9, m pt 51, rd 1-orf2, rd 1-orf3, rd 1-orf4, rd 1-orf5, rd 1-orf8, rd 1-orf9a and rd 1-orf9b gene fragments were amplified by PCR from the plasmids pRVN01, pRVN02, pT052, pT087, pT088, pT089, pT090, pT091, pT096 or pT098 by using the primers shown in TABLE 1 1 and TABLE 2 (in Example 2a). The probes were labelled non-radioactively with an enhanced chemiluminescence kit (ECL; Amersham International plc, Little Chalfont, United Kingdom). Hybridization and detection was performed according to the instructions provided by the manufacturer. The results are summarized in TABLES
12 and 13.

TABLE 12. Interspecies analysis of the cfp7, cfp9 and mpt51 genes by PCR
and/or Southern blotting and of MPT51 protein by Western blotting.
,PCR ,Southern ,Western blot ' ' ;blot Species and strain , cfp7cfp mpt51 cfp9 mpt5 , MPT51 , cfp7 ' 9 ' 1 1. M. tub. H37Rv ~+ + + ,+ + + ~+

2. M. tub. H37Ra + + + N.D. N.D. + +

3. M. tub. Erdmann+ + + + + + +

4. M. bovis + + + + +

S. M. bovis BCG :+ + + .+ + + :+

Danish 331 : ' 6. M. bovis BCG ;+ + N.D. ;+ + + ;N.D.

Japan ' ' , , 7. M. bovis BCG + + N.D. + +' N.D. N.D.

Chinese 8. M. bovis BCG + + N.D. + + N.D. N.D.

Canadian 9. M. bovis BCG ;+ + N.D. ;+ + N.D. ;N.D.

Glaxo ; ; ;

10. M. bovis BCG + + N.D. + + N.D. N.D.

Russia 11. M. bovis BCG + + N.D. + + N.D_ N.D.

Pasteur 12. M. africanum .+ + + .+ + + :+

13. M. leprae . - - . - -I4. M. avium ;+ + - ;+ + + , 15. M. kansasii ;+ - - ;+ + + ;-16 . marinum - ( - + + + -M. +
) 17. M. scrofulaceum- - - - - - -18. M. intercellul-+ (+) - + ~ + + -are 19. M. fortui tum : - - : - -20. M. flavescens ;+ (+} - ;+ + + ;N.D.

2I. M. xenopi ;- - - ;N.D_ N.D. + ;-22. M. szulgai (+) (+} - - + - -23. M. terrae - - N.D. N.D. N.D. N.D. N.D.

+, positive reaction; -, no reaction, N.D. not determined.
cfp7, cfp9 and mpt51 were found in the M. tuberculosis complex_including BCG
and the environmental mycobacteria; M. avium, M. kansasii, M. marinum, M.
intracellular and M. flavescens. cfp9 was additionally found in M. szulgai and mpt51 in M.
xenopi.
Furthermore the presence of native MPT51 in culture filtrates from different mycobac-terial strains was investigated with western blots developed with Mab HBT4.

There is a strong band at around 26 kDa in M, tuberculosis H37Rv, Ra, Erdman, M.
bovis ANS, M. bovis BCG substrain Danish 1331 and M. africanum. No band was seen in the region in any other tested mycobacteriai strains.
TA8LE 13a. Interspecies analysis of the rd 1-orf2, rd 1-orf3, rd 1-orf4, rd 1-orf5, rd 1-orf8, rdl-orf9a and rdl-orf9b genes by Southern blotting. -Species and rd1- rd1- rd1- rd1- rd1- rd1- rd1-strain orf2 orf3 orfg orf5 orf8 orf9a orf9b 1. M. tub. + + + + + + +

H37Rv 2. M. bovis + + + + N.D. + +

3. M. bovis + - - - N.D. - -BCG

Danish . .

4. M. bovis + - - - N.D. - -BCG Japan 5. M. aviu.m - - - - N.D. - -6. M. - _ _ _ N.D. - _ kansasii 7. M. marinum+ - + - N.D. - -8. M. scrofu-+ - - - N.D. - -laceum 9. M. - - - - N.D. - -intercellu-lare 10. M. - - - - N.D. - -fortui tum 11. M. xenopi- - - - N.D. - -12_ M. + - - - . N.D. - -szulgai +, positive reaction; -, no reaction, N.D. not determined.
Positive results for rd 1-orf2, rd 1-orf3, rd 1-orf4, rd 1-orf5, rd 1-orf8, rd 1-orf9a and rd 1-orf96 were only obtained when using genomic DNA from M. tuberculosis and M. bo-vis, and not from M. bovis BCG or other mycobacteria analyzed except rd 1-orf4 which also was found in M. marinum.
Presence of cfA7a cfv7b cfp lOa cfp 17 cfp20 cfp21 cfp22 cfp22a cfp23 cfp25 and cfp25a in different mycobacterial species.
Southern blotting was carried out as described for rd l-orf2, rd l-orf3, rd 1-orf4, rd l-orf5, rd 1-orf8, rd l-orf9a and rd 1-orf9b. The cfp7a, cfp7b, cfp lOa, cfp 17, cfp20, cfp2l, cfp22, cfp22a, cfp23, cfp25 and cfp25a gene fragments were amplified by PCR from the recombinant pMCT6 plasmids encoding the individual genes. The primers used (same as the primers used for cloning) are described in example 3, 3A
and 3B. The results are summarized in Table 13b.
TABLE 13b. Interspecies analysis of the cfp7a, cfp7b, cfp 10a, cfp 17, cfp20, cfp2l, cfp22, cfp22a, cfp23, cfp25, and cfp25a genes by Southern blotting.
Species and cfp7acfp7 cfp- cfp cfp cfp2cfp2 cfp-cfp cfp cfp-strain b l0a 17 20 1 2 22a 23 25 25a 1. M. tub. + + + + + + + + + + +

H37Rv 2. M. bovis + + + + + + + + + + +

3. M. bovis + + + + + N.D.+ + + + +

BCG

Danish 1331 4. M. bovis + + + + + + + + + + +

BCG Japan 5. M. avium + N.D. - + - + + + + + -6. M. kansasii- N.D. + - - - + - + - -7. M. marinum+ + - + + + + + + + +

8. M. scrofu-- - + - + + - + + + -laceum 9. M. intercel-+ + - + - + + - + + -lulare 10. M. fortui-- N.D. - - - - - - + - -tum 11. M. xenopi+ + + + + + + + + + +

12. M. szulgai+ + - + + + + + + + +
-+, positive reaction; -, no reaction, N.D. not determined.
LIST OF REFERENCES
Andersen, P. and Heron, I, 1993, J. Immunot. Methods 161: 29-39.

WO 99!24577 PCT/DK98/00438 Andersen, A. B. et al., 1992, Infect. Immun. 60: 2317-2323.
Andersen P., 1994, Infect. Immun. 62: 2536-44.
Andersen P. et al., 1995, J. Immunol. 154: 3359-72 Barkholt, V. and Jensen, A. L., 1989, Anal. Biochem. 177: 318-322.
Borodovsky, M., and J. Mclninch. 1993, Computers Chem. 17: 123-133.
van Dyke M. W. et al., 1992. Gene pp. 99-104.
Gosselin et al., 1992, J. Immunol. 149: 3477-3481.
Harboe, M. et al., 1996, Infect. Immun. 64: 16-22.
von Heijne, G., 1984, J. Mol. Biol. 173: 243-251.
Hochstrasser, D.F. et al., 1988, AnaI.Biochem. 173: 424-435 Kohler, G. and Milstein, C., 1975, Nature 256: 495-497.
Li, H. et al., 1993, Infect. Immun. 61: 1730-1734.
Lindblad E.B. et al., 1997, Infect. Immun. 65: 623-629.
Mahairas, G. G. et al., 1996, J. Bacteriol 178: 1274-1282.
Maniatis T. et al., 1989, "Molecular cloning: a laboratory manual", 2nd ed..
Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Nagai, S. et al., 1991, infect. Immun. 59: 372-382.
Oettinger, T. and Andersen, A. B., 1994, Infect. Immun. 62: 2058-2064.

WO 99!14577 PGT/DK98/00438 Ohara, N. et al., 1995, Scand. J. imn,unol. 41: 233-442.
Pal P. G. and Horwitz M. A., 1992, Infect. lmmun. 60: 4781-92.
Pearson, W. R. and Liprnan D. J., 1988. Proc. Natl. Acad. Sci. USA -85: 2444-2448.
Ploug, M. et al. , 1989, Anal. Biochem. 181: 33-39.
Porath, J. et al., 1985, FEES Lett. 185: 306-310.
Roberts, A.D. et aL, 1995, Immunol. 85: 502-508.
Sorensen, A.L. et aL, 1995, Infect. Immun. 63: 1710-1717.
Theisen, M. et al., 1995, Clinical and Diagnostic Laboratory Immunology, 2: 30-34.
Valdes-Stauber, N. and Scherer, S., 1994, Appl. Environ. Microbiol. 60: 3809-3814.
Valdes-Stauber. N. and Scherer, S., 1996, Appl. Environ. Microbiol. 62: 1283-1286.
Williams, N., 1996, Science 272: 27.
Young, R. A. et al., 1985, Proc. Natf. Acad. Sci. USA 82: 2583-2587.

Claims

1. A substantially pure polypeptide fragment which a) comprises an amino acid sequence selected from the sequences shown in SEQ
ID NO: 175, 177, 179, 181, 183, and 185, b) comprises a subsequence of the polypeptide fragment defined in a) which has a length of at least 6 amino acid residues, said subsequence being immunologically equivalent to the polypeptide defined in a) with respect to the ability of evoking a protective immune response against infections with mycobacteria belonging to the tuberculosis complex or with respect to the ability of eliciting a diagnostically significant immune response indicating previous or ongoing sensitization with antigens derived from mycobacteria belonging to the tuberculosis complex, or c) comprises an amino acid sequence having a sequence identity with the polypeptide defined in a) or the subsequence defined in b) of at least 70% and at the same time being immunologically equivalent to the polypeptide defined in a) with respect to the ability of evoking a protective immune response against infections with mycobacteria belonging to the tuberculosis complex or with respect to the ability of eliciting a diagnostically significant immune response indicating previous or ongoing sensitization with antigens derived from mycobacteria belonging to the tuberculosis complex.

2. The polypeptide fragment according to claim 1 in essentially pure form.

3. The polypeptide fragment according to claim 1 or 2, which comprises an epitope for a T-helper cell.

4. The polypeptide fragment according to any of the preceding claims, which has a length of at least 7 amino acid residues, such as at least 8, at least 9, at least 10, at least 12, at least 14, at least 16, at least 18, at least 20, at least 22, at least 24, and at least 30 amino acid residues.

5. The polypeptide fragment according to any of the preceding claims which is free from any signal sequence.

6. The polypeptide fragment according to any of the preceding claims which 1) induces a release of IFN-.gamma. from primed memory T-lymphocytes withdrawn from a mouse within 2 weeks of primary infection or within 4 days after the mouse has been re-challenge infected with mycobacteria belonging to the tuberculosis complex, the induction performed by the addition of the polypeptide to a suspension comprising about 200.000 spleen cells per ml, the addition of the polypeptide resulting in a concentration of 1-4 µg polypeptide per ml suspension, the release of IFN-.gamma. being assessable by determination of IFN-.gamma. in supernatant harvested 2 days after the addition of the polypeptide to the suspension, and/or 2) induces a release of IFN-.gamma. of at least 300 pg above background level from about 1000,000 human PBMC (peripheral blood mononuclear cells) per ml isolated from TB patients in the first phase of infection, or from healthy BCG
vaccinated donors, or from healthy contacts to TB patients, the induction being performed by the addition of the polypeptide to a suspension comprising the about 1,000,000 PBMC
per ml, the addition of the polypeptide resulting in a concentration of 1-4 µg polypeptide per ml suspension, the release of IFN-.gamma. being assessable by determination of IFN-.gamma.
in supernatant harvested 2 days after the addition of the polypeptide to the suspension;
and/or 3) induces an IFN-.gamma. release from bovine PBMC derived from animals previously sensitized with mycobacteria belonging to the tuberculosis complex, said release being at least two times the release observed from bovine PBMC derived from animals not previously sensitized with mycobacteria belonging to the tubercutosis complex.

7. A polypeptide fragment according to any of the preceding claims, wherein the sequence identity in c) is at least 80%, such as at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, and at least 99.5%.

8. A fusion polypeptide comprising at least one polypeptide fragment according to any of the preceding claims and at least one fusion partner.

9. A fusion polypeptide according to claim 8, wherein the fusion partner is selected from the group consisting of a polypeptide fragment as defined in any of claims 1-8, and an other polypeptide fragment derived from a bacterium belonging to the tuber-culosis complex, such as ESAT-6 or at least one T-cell epitope thereof, MPB64 or at least one T-cell epitope thereof, MPT64 or at least one T-cell epitope thereof, and MPB59 or at least one T-cell epitope thereof.

10. A fusion polypeptide fragment which comprises 1) a first amino acid sequence including at least one stretch of amino acids constituting a T-cell epitope derived from the M. tuberculosis protein ESAT-6, and a second amino acid sequence including at least one T-cell epitope derived from a M.tuberculosis protein different from ESAT-6 and/or including a stretch of amino acids which protects the first amino acid sequence from in vivo degradation or post-translational processing; or 2) a first amino acid sequence including at least one stretch of amino acids constituting a T-cell epitope derived from the M. tuberculosis protein MPT59, and a second amino acid sequence including at least one T-cell epitope derived from a M.tuberculosis protein different from MPT59 and/or including a stretch of amino acids which protects the first amino acid sequence from in vivo degradation or post-translational processing.

11. A fusion polypeptide fragment according to claim 10, wherein the first amino acid sequence is situated C-terminally to the second amino acid sequence.

12. A fusion polypeptide fragment according to claim 10, wherein the first amino acid sequence is situated N-terminally to the second amino acid sequence.

13. A fusion polypeptide fragment according to any of claims 10-12, wherein the at least one T-cell epitope included in the second amino acid sequence is derived from a M. tuberculosis polypeptide selected from the group consisting of a polypeptide fragment according to any of claims 1-55, DnaK, GroEL, urease, glutamine synthetase, the proline rich complex, L-alanine dehydrogenase, phosphate binding protein, Ag 85 complex, HBHA (heparin binding hemagglutinin), MPT51, MPTB4, superoxide dismutase, 19 kDa lipoprotein, .alpha.-crystallin, GroES, MPT59 when the first T-cell epitope is derived from ESAT-6, and ESAT-6 when the first T-cell epitope is derived from MPT59.

14. A fusion polypeptide tragment according to any of claims 10-13, wherein the first and second T-cell epitopes each have a sequence identity of at least 70% with the natively occurring sequence in the proteins from which they are derived.

15. A fusion polypeptide according to any of claims 10-14, wherein the first and/or second amino acid sequence have a sequence identity of at least 70% with the protein from which they are derived.

16. A fusion polypeptide fragment according to any of claims 10-15, wherein the first amino acid sequence is the amino acid sequence of ESAT-6 or of MPT59 and/or the second amino acid sequence is the amino acid sequence of. a M. tuberculosis polypeptide selected from the group consisting of a polypeptide fragment according to any of claims 1-7, DnaK, GroEL, urease, glutamine synthetase, the proline rich complex, L-alanine dehydrogenase, phosphate binding protein, Ag 85 complex, HBHA
(heparin binding hemagglutinin), MPT51, MPT64, superoxide dismutase, 19 kDa lipoprotein, .alpha.-crystallin, GroES, ESAT-6 when the first amino acid sequence is that of MPT59, and MPT59 when the first amino acid sequence is that of ESAT-6.

17. A fusion polypeptide fragment according to any of claims 10-16, wherein no linkers are introduced between the two amino acid sequences.

18. A polypeptide according to any of the preceding claims which is lipidated so as to allow a self-adjuvating effect of the polypeptide.

19. A substantially pure polypeptide according to any of claims 1-18 for use as a pharmaceutical.

20. The use of a substantially pure polypeptide according to any of claims 1-19 in the preparation of a pharmaceutical composition for the diagnosis of tuberculosis caused by Mycobacterium tuberculosis, Mycobacterium africanum or Mycobacterium bouts.

21. The use of a substantially pure polypeptide according to any of claims 1-19 in the preparation of a pharmaceutical composition for the vaccination against tuberculosis caused by Mycobacterium tuberculosis, Mycobacterium africanum or Mycobacterium bovis.

22. A nucleic acid fragment in isolated form which 1) comprises a nucleic acid sequence which encodes a polypeptide as defined in any of claims 1-18, or comprises a nucleic acid sequence complementary thereto, 2) has a length of at least 10 nucleotides and hybridizes readily under stringent hybridization conditions with a nucleic acid fragment which has a nucleotide sequence selected from SEQ ID NO: 174 or a sequence complementary thereto, SEQ ID NO: 176 or a sequence complementary thereto, SEQ ID NO: 178 or a sequence complementary thereto, SEQ ID NO: 180 or a sequence complementary thereto, SEQ ID NO: 182 or a sequence complementary thereto, and SEQ ID NO: 184 or a sequence complementary thereto,

23. A nucleic acid fragment according to claim 22, which is a DNA fragment.

24. A nucleic acid fragment according to claims 22 or 23 for use as a pharmaceutical.

25. The use of a nucleic acid fragment according to any of claims 22-24 in the preparation of a pharmaceutical composition for the vaccination against tuberculosis caused by Mycobacterium tuberculosis, Mycobacterium africanum or Mycobacterium bovis.

26. The use of a nucleic acid fragment according to any of claims 22-24 in the preparation of a pharmaceutical composition for the diagnosis of tuberculosis caused by Mycobacterium tuberculosis, Mycobacterium africanum or Mycobacterium bovis.

27. A vaccine comprising a nucleic acid fragment according to any of claims 22-24, the vaccine effecting in vivo expression of antigen by an animal, including a human being, to whom the vaccine has been administered, the amount of expressed antigen being effective to confer substantially increased resistance to infections with myco-bacteria of the tuberculosis complex in an animal, including a human being.

28. An immunologic composition comprising a polypeptide according to any of claims 1-19.

29. An immunologic composition according to claim 28, which further comprises an immunologically and pharmaceutically acceptable carrier, vehicle or adjuvant.

30. An immunologic composition according to claim 29, wherein the carrier is selected from the group consisting of a polymer to which the polypeptide(s) is/are bound by hydrophobic non-covalent interaction, such as a plastic, e.g, polystyrene, a polymer to which the polypeptide(s) is/are covalently bound, such as a polysaccharide, and a polypeptide, e.g. bovine serum albumin, ovalbumin or keyhole limpet hemocyanin; the vehicle is selected from the group consisting of a diluent and a suspending agent; and the adjuvant is selected from the group consisting of dimethyfdioctadecylammonium bromide (DDA), Quil A, poly I:C, Freund's incomplete adjuvant, IFN-.gamma., IL-2, IL-12, monophosphoryl lipid A (MPL), and muramyl dipeptide (MDP).

31. An immunologic composition according to any of claims 28-30, comprising at least two different polypeptide fragments, each different polypeptide fragment being a polypeptide according to any of claims 1-19.

32. An immunologic composition according to claim 31, comprising 3-20 different polypeptide fragments, each different polypeptide fragment being according to any of claims 1-19.

33. An immunologic composition according to any of claims 28-32, which is in the form of a vaccine.

34. An immunologic composition according to any of claims 28-32, which is in the form of a skin test reagent.

35. A vaccine for immunizing an animal, including a human being, against tuberculosis caused by mycobacteria belonging to the tuberculosis complex, comprising as the effective component a non-pathogenic microorganism, wherein at least one copy of a DNA fragment comprising a DNA sequence encoding a polypeptide according to any of claims 1-19 has been incorporated into the genome of the microorganism in a mar ner allowing the microorganism to express and optionally secrete the polypeptide.

36. A vaccine according to claim 35, wherein the microorganism is a bacterium.

37. A vaccine according to claim 36, wherein the bacterium is selected from the group consisting of the genera Mycobacterium, Salmonella, Pseudomonas and Eschericia.

38. A vaccine according to claim 37, wherein the microorganism is Mycobacterium bovis BCG, such as Mycobacterium bovis BCG strain: Danish 1331.

39. A vaccine according to any of claims 35-38, wherein at least 2 copies of a DNA
fragment encoding a polypeptide according to any of claims 1-12 are incorporated into the genome of the microorganism.

40. A vaccine according to claim 39, wherein the number of copies is at least 5.

41. A replicable expression vector which comprises a nucleic acid fragment according to any of claims 22-24.

42. A vector according to claim 41, which is selected from the group consisting of a virus, a bacteriophage, a plasmid, a cosmid, and a microchromosome.

43. A transformed cell harbouring at least one vector according to claim 41 or 42.

44. A transformed cell according to claim 43, which is a bacterium belonging to the tuberculosis complex, such as a M. tuberculosis bovis BCG cell.

45. A transformed cell according to claim 43 or 44, which expresses-a polypeptide according to any of claims 1-19.

46. A method for producing a polypeptide according to any of claims 1-19, comprising inserting a nucleic acid fragment according to any of claims 15-17 into a vector which is able to replicate in a host cell, introducing the resulting recombinant vector into the host cell, culturing the host cell in a culture medium under conditions sufficient to effect expression of the polypeptide, and recovering the polypeptide from the host cell or culture medium; or isolating the polypeptide from a short-term culture filtrate; or isolating the polypeptide from whole mycobacteria of the tuberculosis complex or from lysates or fractions thereof, e.g. cell wall containing fractions; or synthesizing the polypeptide by solid or liquid phase peptide synthesis.

47. A method for producing an immunologic composition according to any of claims 28-34 comprising preparing, synthesizing or isolating a polypeptide according to any of claims 1-19, and solubilizing or dispersing the polypeptide in a medium for a vaccine, and optionally adding other M. tuberculosis antigens and/or a carrier, vehicle and/or adjuvant substance, or cultivating a cell according to any of claims 41-45, and transferring the cells to a medium for a vaccine, and optionally adding a carrier, vehicle and/or adjuvant substance.

48. A method of diagnosing tuberculosis caused by Mycobacterium tuberculosis, Mycobacterium africanum or Mycobacterium bouts in an animal, including a human being, comprising intradermally injecting, in the animal, a polypeptide according to any of claims 1-19 or an immunologic composition according to any of claims 28-34, a positive skin response at the location of injection being indicative of the animal having tuberculosis, and a negative skin response at the location of injection being indicative of the animal not having tuberculosis.

49. A method for immunising an animal, including a human being, against tuberculosis caused by mycobacteria belonging to the tuberculosis complex, comprising administering to the animal the polypeptide according to any of claims 1-19, the immunologic composition according to any of claims 28-34, or the vaccine according to any of claims 35-40.

50. A method according to claim 49, wherein the polypeptide, immunologic composition, or vaccine is administered by the parenteral (such as intravenous and intraarterially), intraperitoneal, intramuscular, subcutaneous, intradermal, oral, buccal, sublingual, nasal, rectal or transdermal route.

51. A method for diagnosing ongoing or previous sensitization in an animal or a human being with bacteria belonging to the tuberculosis complex, the method comprising providing a blood sample from the animal or human being, and contacting the sample from the animal with the polypeptide according to any of claims 1-19, a significant release into the extracellular phase of at least one cytokine by mononuclear cells in the blood sample being indicative of the animal being sensitized.

52. A composition for diagnosing tuberculosis in an animal, including a human being, comprising a polypeptide according to any of claims 1-19, or a nucleic acid fragment according to any of claims 22-24, optionally in combination with a means for detection.

53. A monoclonal or polyclonal antibody, which is specifically reacting with a polypeptide according to any of claims 1-19 in an immuno assay, or a specific binding fragment of said antibody.

54. Use of CFP7A or CFP30A, or a T-cell epitope thereof, for the induction of a strong immune response in a mammal including a human being.

55. Use of CFP7A, or a T-cell epitope thereof, for the induction of a high protective immune response in a mammal including a human being.

56. Use of CFP7B, CFP19, or MPT59-ESAT6, or a T-cell epitope thereof, for the diagnosis of tuberculosis in a mammal including a human being by performing a DTH type skin test.

57. Use of CFP27, CFP30A, RD1-ORF2, RD1-ORF3, RD1-ORF5, MPT59-ESAT6, ESAT6-MPT59, CFP10A, CFP16, CFP19, CFP23, CFP25A, CFP30B, CFP7B, or a T-cell epitope thereof, for the preparation of an immunological composition with a wide genetically recognition.

58. Use of CFP27, CFP30A, RD1-ORF2, RD1-ORF5, MPT59-ESAT6, ESAT6-MPT59, CFP10A, CFP19, CFP23, CFP25A, CFP30B, or a T-cell epitope thereof, for the preparation of a vaccine such as a subunit vaccine.