WO2005021790A2 - Use of gene sequences specific for mycobacterium tuberculosis and their related proteins for diagnosis and prevention of tubercular infection - Google Patents

Abstract

The present invention relates to the use of gene sequences specific Mycobacterium tuberculosis and their corresponding proteins for the diagnosis and the prevention of the tubercular infection.

Description

USE OF GENE SEQUENCES SPECIFIC FOR MYCOBACTERIUM TUBERCULOSIS AND THEIR RELATED PROTEINS FOR DIAGNOSIS AND PREVENTION OF TUBERCULAR INFECTION.

The present invention relates to the use of gene sequences specific for Mycobacterium tuberculosis and their related proteins for diagnosis and prevention of tubercular infection. More in particular, the invention refers to the use of gene sequences specific for Mycobacterium tuberculosis and their related proteins for the specific diagnosis of tubercular infection able to remove the problem of false positives, for example, in vaccinated subjects.

Further, the invention relates to the use of the above mentioned sequences and their related proteins, as well as for the diagnosis, for the prevention of tubercular infection, and for the preparation of diagnostic kits and vaccines. Estimations of the World Health Organization show that one third of the world population is infected with Mycobacterium tuberculosis (MTB), with 8 million new cases of tuberculosis and about 3 million deaths per year (World Health Organisation, 1996). The initial interaction between the microorganism and the host dramatically influences the outcome of the infection, either proceeding towards the pathogen containment or estabilishment of the tubercular pathology. Nowadays it is opportune to dwell upon the examination of the

"dialogue" between the pathogen and the host, instead of the "monologue" of each of them (Kaufmann et al., 2002). On the basis of the last observations, the research addressed the M. tuberculosis studies towards an in-depth knowledge of the gene expression of the bacterium in response to different stimuli, like heat shock, oxidative stress and phagocytosis by resting or activated macrophages (Stewart et al., 2002, Dubnau et al., 2002, Russell et al ,

2002, Fenhalls et al., 2002). These studies have showed that M. tuberculosis adopts a fine tuning of its transcriptome according to the environment in which it replicates (Betts et al., 2002, Alland et al., 1998,

Graham et al., 1999, Wilson et al., 1999, Chan et al., 2002, Rosenkrands et al., 2002). The laboratory diagnosis of the tubercular pathology in its early stage is very important to guarantee the therapeutic intervention quickness. The current diagnostic kits, based on cultural or biochemical methods, do not often guarantee the medical report to be issued in acceptable time, indeed M. tuberculosis colonies are visible only 10-24 days after the spread on cultural media (Konemann et al., 1995). The bacterium grows slowly in vitro and the method has the disadvantages to give false positives and to be expensive. The tuberculine skin test is the method currently in use which avails itself of a diagnostic assay which is very faster and cheaper. Recently, molecular biology techniques were also developed, such as PCR analysis of restriction fragments to detect the presence of specific MTB genes, or hybridization using DNA probes, which are techniques both requiring specie specific sequences for the identification of the microorganism and for the diagnosis of tubercular disease. For example the use of insertion sequence such as IS6110 can show the disadvantage to create false positives, since such a sequence was found also in non-tubercular mycobacteria (Kent et al., 1995). The use of the sequence 16SrRNA shows the same problem of specie specificity, as it is a genus specific sequence, common to all the mycobacteria (Moussa et al., 2000). As far as prophylaxis of the disease is concerned, the sole existing vaccine against tuberculosis used in clinical practice consists of M. bovis Bacillus Calmette-Guerin (BCG) (ATCC 27291). The BCG strain is an avirulent bacterium which has been used worldwide for the antitubercular vaccination for about 75 years. During tuberculosis testing the identification of latent infected individual with M. tuberculosis is very important. As previously disclosed the currently in use diagnostic assay is the aforesaid tuberculine skin test (Mantoux), which shows some limits due to the low specificity. In fact, the mycobacterium protein derivate (PPD) by which the test works, shows a cross-reactivity with both M. bovis Bacillus Calmette-Guerin (BCG) used for the antitubercular vaccination and several environmental mycobacteria (Lalvani et al., 2001). Thus, if both individuals who show immunological memory induced by BCG vaccination and individuals who recently get to contact with the virulent strain H37Rv (ATCC 27294) underwent the routine skin test with the protein extract (PPD), they could react positive on the test in an indistinguishable way. Therefore, a positive reaction to the skin test is not able to distinguish between M. tuberculosis infection and BCG vaccination or the individual exposure to environmental mycobacteria (MMWR Morb Mortal Wkly Rep., 1997). In the light of the above, it is evident the need to dispose of new biotechnological instruments to be used in the recent biomolecular techniques both for the diagnosis and for the prophylaxis of M. tuberculosis infections, which allow to overcome the limits of the adopted techniques so far. During the course of previous studies the authors of the present invention have identified a gene group which was significantly transcribed by M. tuberculosis H37Rv in infected human macrophages. Such in vivo expression is the result of a choice that M. tuberculosis makes, according to the replication environment in which it is situated (Mariani et al., 2000; Cappelli et al., 2001 ). The authors of the present invention have now found M. tuberculosis specific gene sequence groups that are absent in the genome of the attenuated strain BCG and that can be advantageously used for the preparation of diagnostic kits for tubercular disease without the limits of the known methodologies. In the course of the present study the gene expression patterns, particularly the transcriptomes, of the two M. tuberculosis strains, H37Rv and CMT97, were analyzed and compared respectively in synthetic medium, pure human macrophages and human macrophages in co- culture with autologous peripheral blood lymphocytes. Among the group of genes significantly transcribed in human macrophages infected with M. tuberculosis H37Rv, it has been found that nine of such genes are not present in BCG genome and belong to the deletion regions of the chromosome which distinguish BCG from H37Rv (Behr et al., 1999). The list of M. tuberculosis genes expressed in human macrophages and belonging to the corresponding deleted DNA regions in M. bovis BCG is shown in Table 1.

The above listed genes can be used for a more accurate identification of asymptomatic individuals, who have been recently exposed to M. tuberculosis, in such a way as to allow the improvement of tuberculosis control. Genes countermarked with an asterisk are not present in M. bovis BCG Aventis Pasteur. In fact, the advantage in the use of the gene sequences in diagnostic kits according to the present invention consists in the fact that it is possible to distinguish between immunological memory and new exogenous infection, thus restricting the possibility to obtain false positives because of the cross-reactivity between BCG and PPD or vice versa false negatives because of the poor reactivity. On these assumptions, the individual positive result of a test which uses the gene sequences according to the present invention provides a certain indication of a contingent infection with a mycobacterial specie belonging to the tubercular complex (M. tuberculosis, M. bovis, M. africanum). Obviously, also the proteins related to the above mentioned gene sequences can be employed for the same diagnostic purpose. The gene sequences and their related proteins according to the present invention can be also advantageously used for the prevention of M. tuberculosis infection, for example, for the preparation of DNA vaccines specific for the tubercular complex. Therefore it is an object of the present invention the use of oligonucleotidic sequences, and their encoded proteins, or their portions, expressed by M. tuberculosis in human macrophages, characterized in that belong to a deletion region of M. bovis BCG, for the preparation of means for the diagnosis and the prevention of tubercular disease. The oligonucleotidic sequences used according to the present invention, that can be deoxyribonucleotidic or ribonucleotidic sequences and their complementary sequences, may be selected from the group consisting of the genes Rv1767, Rv1772, Rv1966, Rv3119, Rv1511 , Rv3425, Rv1255c, Rv1575, Rv1576c, Rv1584c, Rv0222, Rv2350c,

Rv1768 (http://genolist.pasteur.fr/TubercuList/). As above said, such genes expressed by M. tuberculosis in human macrophage belong to the deletion regions RD3, RD4, RD5, RD6, RD7, RD10, RD11 , RD14, RD15 (http://molepi.stanford.edu/bcg) of M. bovis BCG. The oligonucleotidic sequences according to the present invention may be used as whole sequences or portions thereof which will encode for whole proteins or portions thereof. Said oligonucleotidic sequences, encoded proteins thereby, or their portions can be advantageously used as means for the diagnosis of tubercular disease and in particular as probes, antisense oligonucleotides, primers, antigens, antibodies. The techniques used for the diagnosis may be PCR, in situ hybridization, ELISA, T cell proliferation assay, ELISPOT and Immunoblotting. In particular, the use of recombinant proteins could be useful for the screening of serum antibodies against the specific MTB proteins or to stimulate human T lymphocytes and to analyze their cytokines production. The use as DNA sequences could be advantageous for the design of primer for the molecular diagnosis of mycobacterial DNA by PCR or as probes for the in situ hybridization assay. As regards the use of the oligonucleotidic sequences, encoded proteins thereby, or their portions, as means for the prevention of the tubercular disease, said means can be inserted into a DNA vector for the preparation of a vaccine. It is another object of the present invention a diagnostic kit for the detection of M. tuberculosis infection in a biological sample which comprises at least one oligonucleotidic sequence or an encoded protein thereby, or their portions, according to the present invention. The detection of the tubercular infection carried out by said kits can be done by PCR techniques, in situ hybridization, ELISA, T cell proliferation assay, ELISPOT and Immunoblotting and the biological sample to be analyzed can be saliva, sputum, urine, broncho alveolar lavage and blood in the case of disseminateTB. It is a further object of the present invention a vaccine for the prevention of M. tuberculosis infection comprising at least one oligonucleotidic sequence or a protein encoded thereby, or their portions, as previously disclosed, together with one or more pharmaceutically acceptable excipients and/or adjuvants. Preferably the vaccine is a DNA vaccine specific for the tubercular infection (M. tuberculosis, M. bovis, M. africanum). Further object of the present invention are oligonucleotidic probes comprising at least one of the oligonucleotidic sequences, or their portions, for the diagnosis of the tubercular infection through detection of M. tuberculosis by hybridization. Subject of the present invention are also primer comprising at least one of the oligonucleotidic sequences, or their portions, for the diagnosis of tubercular disease through the detection of M. tuberculosis in biological samples by PCR amplification of the genie sequences from mycobacterium. Also in this case the biological sample can be blood, saliva, sputum, urine, broncho alveolar lavage. According to another aspect, the present invention provides an antigen comprising at least one oligonucleotidic sequence or protein encoded thereby, or their portions, for the diagnosis of tubercular disease through the detection of the presence of human serum antibodies against said sequences or their related proteins by ELISA. Finally the object of the invention can be represented by monoclonal antibodies able to recognize and to selectively bind at least one protein, or its portion, corresponding to the oligonucleotidic sequences for the detection of M. tuberculosis in biological samples. For reasons of clearness the abbreviations used through the text are explained as follows: MTB: M. tuberculosis; MΦ: human macrophages; PBL: Peripheral Blood Lymphocytes; PBMC: Peripheral Blood Mononuclear Cells; FCS: foetal calf serum; MOI: Multiplicity of Infection; BP: Binding Proteins; ROI: Reactive Oxygen Intermediates; RNI: Reactive Nitrogen Intermediates; ORFs: Open Reading Frame; BCG: M. Bovis Bacillus Calmette Guerin; CDs: Coding Sequences; CFU: Colony Forming Unit; ISs: Insertion Sequences. The present invention will be now described with reference to exemplifying, but not limiting, preferred embodiments thereof, with a particular reference to the figures of the enclosed drawings, wherein: Figure 1 shows the classification of patterns of expression of MTB genes, selected and analyzed in the different environments of mycobacterial replication; Figure 2 shows the analyses of the pattern of MTB genes expression in the different environments analyzed; in panel A it is shown the grouping and numbering of the positive genes according to their expression, respectively in, all environments (housekeeping genes), synthetic medium and human phagocytes (all the cultural conditions); in panel B is shown the number of genes whose function is still unknown and the relative percentage of unknown protein in each gene group; in panel C is shown the splitting of the genes expressed always in human phagocytes in the three pattern of expression: such as the genes expressed in human macrophages, whether or not or in co-colture with autologous PBLs (both pure MΦ and PBMCs), genes expressed in human macrophages only when they are in co-colture with autologous PBL (PBMCs), gene expressed only when MTB replicates inside pure human macrophages (pure MΦ); in panel D is shown the number of genes whose function is still unknown in each "in vivo" sub-group and the relative percentage of unknown proteins in the gene group; Figure 3 shows the functional categories of MTB gene expression; in each of the expression group the genes were analyzed according to their function in MTB physiology; a part from unknown proteins, PE_PGRS and other categories, the listed categories refer to metabolism of coenzymes, lipids, nucleotides, aminoacids and carbohydrates, information storage and processing (transcription, translation and replication) and cellular processes (cell envelope, signal transduction, post-translational modifications, cell motility and inorganic ion transport); Figure 4 shows the pattern of gene expression of two MTB strains in human macrophages; Figure 5 shows the bi-dimensional histogram of the transcriptomes comparison between the two MTB strains in human macrophages; the expression in pure human macrophages of H37Rv strain was plotted against CMT97 strain expression; the 45 degree line represents the genes that are equally expressed by the two strains; it has been used a logarithmic scale to plot the signal-to-noise ratio and to compare the different arrays; Figure 6 shows the bi-dimensional histogram of the transcriptomes comparison between the two MTB strains in Sauton's medium; gene expression in synthetic medium of H37Rv strain was plotted against CMT97 strain expression in the same medium; the 45 degree line represents the genes that are equally expressed by the two strains; it has been used a logarithmic scale to plot the signal-to-noise ratio and to compare the different arrays; Figure 7 shows the bi-dimensional histogram of the CMT97 strain transcriptomes comparison between MΦ and PBMC intracellular growth; CMT97 gene expression in pure MΦ was plotted against and compared to the expression of the same strain in PBMCs; the 45 degree line represents the genes that are equally expressed by the two strains; it has been used a logarithmic scale to plot the signal-to-noise ratio and to compare the different arrays; Figure 8 shows the CFU number determination of the two MTB strains in human macrophages; the colony number of CMT97 and H37Rv was obtained by plating serial dilutions of triplicate samples of cell lysates from pure MΦ and PBMCs after 7 days of infection with each of the two strains; Figure 9 shows the histogram exhibiting the frequency of distribution of gene's activities depending on the expression intensity; Figure 10 shows the posterior distributions of the positively (lower panel) and negatively (upper panel) expressed genes populations.

EXAMPLE 1 : Studies on the relative amount of genes expressed in the different environmental conditions, functional categories and incidence of unknown proteins. MATERIALS AND METHODS M. tuberculosis strains and cultural conditions The M. tuberculosis strains used herein are the laboratory H37Rv strain and the clinical CMT97 strain (isolated at the Monaldi

Hospital, Naples from a patient sputum) (Mariani et al., 2001), which were transferred, every two months in Sauton medium (Sauton et al., 1912;

Wills et al., 1962; Landi et al., 1984) allowing them to grow on the medium surface. Every two months, in order to infect MΦs, mycobacterial layers were harvested, spun down and resuspended in sterile PBS. To get an homogeneous resuspension the MTBs were sonicated in a water bath sonicator (UST, 50W, 20 Khz), regulated at a maximal power that is 50W, in sterile glass tubes. The samples were aliquoted and stored at -80°C. Before infection one MTB aliquot was grown on 7H10 plates to titre the corresponding CFU/ml ratio after freezing. For each infection experiment was employed the same frozen master batch. Human cells cultures and infection "Buffy coats" were collected from healthy donors. The blood samples were diluted 1:1 with PBS and mononuclear cells were separated on Ficoll gradient (Eurobio, Paris). The cells wre harvested, washed twice, plated at a concentration of 2 X 10⁶ /ml in 75 ml culture flasks. The cells were cultivated in glutamine enriched RPMI 1640 medium supplemented with gentamicin and 10% of foetal calf serum (FCS). These cultures were incubated at 37°C in a 5% C0₂ - 95% air atmosphere. Purification of monocvtes Human macrophage (MΦ) were isolated by magnetic depletion of non-monocytes cells (Monocytes Isolation Kit, Miltenyi Biotec, Bergish Gladbach) using a cocktail of CD3, CD7, CD19, CD45RA, CD56 and anti- IgE antibodies. The percentage of differentiated macrophages was checked at the FACscan with monoclonal antibodies specific for CD14. These analysis showed a degree of purity higher than 99%. 5 x 10⁶ of purified macrophages were plated and cultivated in 10 ml culture flasks; the infection was performed after 7 days at a multiplicity of infection (MOI) of 10:1 keeping MΦ and MTB in 5 ml of medium for 3 hours. After incubation, the extracellular bacteria were washed out with warm PBS. The infected MΦ cells were grown for other 7 days, without adding growth factors and by replacing the culture medium every three-four days. PBMCs: PBMC cells samples (MΦ plus autologous PBLs) were prepared according to the following method: after Ficoll gradient separation, the cells were harvested, washed twice and plated at a concentration of 2x10⁶ in 75 ml culture flasks. After one hour of adherence the cells were washed twice and detached with cold PBS through gentle scraping and 30x10⁶ PBMCs were plated in 10 ml culture flasks. The infection was performed on the seventh day at a multiplicity of infection (MOI) of 10:1 keeping MTB and MΦs in 5 ml of medium for 3 hours. Total RNA extraction The total RNA extraction from infected human macrophages was carried out by 4 M Guanidium iso-thiocyanate in a single-step GTC method (Chomczynski and Sacchi, 1987). The solutions were prepared with diethyl pyrocarbonate (DPC) treated distilled water, while the extraction was performed in a separate RNase-free room. In order to achieve the maximum yield and a good quality of MTB total RNA, without using CsCI gradient, French press or silica beads to lyse mycobacteria, we slightly modified the GTC protocol. The infected macrophages MΦ, lysed with 4M GTC, were sonicated three times, 10 minutes each, in a water bath sonicator (UST, 50W, 20Khz). Total RNA was then extracted with warm water-equilibrated phenol, the extract being warmed at 65 °C for 5 minutes, then spun down and re-extracted with clorophorm/isoamilic alcohol 24:1 (CCIA) for 10 minutes at 65 °C. Finally the supernatant was precipitated with 2,5 volumes of ethanol and stored at -80 °C overnight. Macroarrav experiments The whole genome profile of MTB was analyzed by macroarray technique. The whole genome of MTB spotted on a nylon membrane

(Panorama M. tuberculosis Gene Array , SIGMA Genosys, USA) was tested by hybridization with a α-dCT³³P-labeled cDNA, corresponding to different experimental conditions. For mycobacterial RNA extracted from culture medium 1 μg of total RNA was employed, as recommended by the manufacturer's instructions. For the experiments conducted with infected human macrophages, 10 μg of total RNA were employed, taking into account that almost 90% of the RNA comes from human cells. The experiments in each environment, with the two MTB strains, were repeated three times. The cDNA was reverse transcribed according to the

Panorama™ Kit instructions, and hybridized overnight with the nylon membrane containing 3,875

ORFs. Each membrane was then washed and exposed in a Storage Phosphor Screen

(Storage Phosphor Screen, Amersham Bioscience, UK) and the signal was detected in a detector (Storm Phosphor Imager, Molecular Dynamics). Macroarrav analysis The analysis of the macroarray results was done by a dedicated software Array Vision 7.0 (Imaging Research Inc., Canada), which was employed to analyse the M. tuberculosis gene expression profile on each membrane, and to compare the different experimental conditions. It was accepted every spot whose intensity was above four-fold the background of each membrane. The determination of CFU number was carried out allowing mycobacteria to grow for 7 days in 10⁶ human macrophages at the MOI of 10:1. Then the cells were washed twice with sterile PBS and incubated 30 minutes in 0,5 ml of lysis buffer (0.1 % saponin in PBS), at 37°C. The dilutions 10^"1, 10^'2, 10^"3, 10^"4, 10^'5, 10^"6 and 10^"7 of each lysate (in 0.01% Tween 80 in PBS) were plated as 50 μl of lisate on 7H10 Middlebrook medium. CFU were checked after 21 days of culture in 5% C0₂ incubator. The RT-PCR of selected genes was performed according to the following protocol. Total RNA (0,5-1 μg) was digested with 2 units of

DNase I (Gibco-BRL) for 30 minutes at 37°C; the enzyme was then inactivated at 75°C for 5 minutes, without further phenol extraction. In the same tube, samples were then reverse transcribed to the corresponding cDNAs with 200 units of MMuLV Reverse Transcriptase (RT); 5 μg/μl p(dN6) random primers (Pharmacia) were used, for one hour at 45°C, in a total volume of 20 μl. The retrotranscriptase was then inactivated at 95°C for 5 minutes. For each sample, one aliquot of DNasel digested RNA, without RT, was used as a negative control for PCR amplification. For each PCR reaction, 1 μl of cDNA was used. The internal control of mycobacterial cDNA was amplified by PCR with a pair of primers belonging to the MT10Sa gene, which encodes for a small stable RNA (Mariani et al., 2000). Identification of immunodominant peptides inside the selected MTB genes One of the main objectives, after having identified genes of M. tuberculosis that might be relevant in vivo, is to identify immunodominant peptide sequences encoded by the selected M. tuberculosis genes, by analysing their binding ability to MHC molecules. MHC genomic, structural and functional studies have shown that it is possible to identify protein sequences, i.e., T-cell epitopes, capable of binding several MHC molecules. In this context, it was attempted to identify protein sequences playing a role in the infection process which may be used to produce an anti-TB vaccine. To this end, the isolated genes were analyzed by means of dedicated software (SYFTPHEI, BIMAS, EPI-predict, ProPred, T-Epitope) and of algorithms developed for MHC binding studies. For the purpose of this study, a panel of allelic MHC variants was used that cover >95% of the Caucasoid population, >91 % of Northern African population, >93% of the African population, >95% of Asian population, >97% of the Amerindian population for class I genes and >91% of the Caucasoid population, >90% of the Northern African population , >92% of the African population >90% of the Asian population, >95% of the Amerindian population for class II genes (data XII International HLA Workshop). IFN-v ELISPOT assay to test the immunogenicitv of the selected peptides The ELISpot (Enzyme Linked Immuno-Spot) assay is powerful method used for the detection of individual cells secreting a protein or a cytokine of interest. Based on the traditional and commonplace ELISA technique, ELISpot protocols use many common reagents and steps associated with the original microwell assay. ELISpot assays have the added benefit of permitting the ex-vivo identification of cells actively secreting cytokines in the one-in-one million cell range. An antibody that recognizes the antigen of interest was used to coat a PVDF or nitrocellulose bottomed 96 well plate. Appropriately stimulated cells were added to each well, and the bound antibodies in each well will capture the cells secreting the antigen of interest. After several washing steps, a biotinylated antibody was added and then enzyme linked to the streptavidin. The detection of positive cells was achieved via a chromogenic substrate that produces a stained precipitate on the membrane. This stained end product can then be used to count individual cells under a microscope or an ELISpot plate reader. The high specificity and sensitivity of the ELISpot assay makes this an ideal procedure if compared to the traditional quantitative methodologies. Cytokine ELISPOT assay has recently emerged as the most sensitive and robust technique for direct ex vivo monitoring of antigen-specific CD4 or CD8 T cells. By measuring the frequencies and analyzing cytokine profiles of individual cells, it has been established the clonal size and effector class of the specific T cell pool. Statistical analysis Among the 4675 different spots were selected 970 genes with t scores ≥ ±2:5, where the t score refers to the standardized difference among the sample means of gene expression in synthetic medium and in human macrophages. The resulting 970 t scores were represented in figure 9. The histogram shape suggests the presence of different groups of genes in terms of expression intensity. The natural statistical way to model this observed heterogeneity is the use of a mixture model with three components, whose importance in microarrays experiment is well documented (Mc Lachlan et al., 2002; Garrett et al., 2003). In practice, it was assumed that each score might arise from one out of three different normal distributions, namely , f₂, and h, with parameters (μ_1t σι), (μ₂, σ₂), (μ3, σ₃) respectively. Also p-i, p₂, and p₃ represent the weights of the three sub-populations among the genes. This amounts to say that each single gene score follows a mixture model with density: f(zj) = i fι(zι\μι, σι) + p₂f₂(z\μ_2l σ₂) + Psfstølμs, σ₃)

In the above formula fi represents the score distribution for those gene which were more expressed in synthetic medium, f₂ represents the score distribution for those genes which did not show significant differences of expression intensity when measured either in synthetic medium or in human macrophages, and is the score distribution for those genes which were more expressed in human macrophages. A Bayesian analysis of the above mixture model has been performed (Robert, 2001 ). The figure 10 shows the posterior distributions of the means μσi and μσ₃ corresponding to the positively (lower panel) and negatively (upper panel) expressed genes populations. As the indicator of expressivity of a gene it has been calculated, for each gene, the probability of belonging to the three different above described populations, and a gene has been declared negatively (positively) expressed when the probability of being in the first (third) group was estimated greater than 0,90. The estimated percentage of non expressed genes among the 970 under analysis was equal to 64,2%. Computations were done using R with specific microarrays oriented package SMA: see the website http://lib.stat.cmu.edU/R/CRAN. Code is available at the website http://3w.eco.uniroma1.it /utenti/liseo/micro.r. Note that the Bayesian approach substantially differs from a more common p-value calculation, since it explicitly provides, for each single gene, the probability that the gene is expressed or not. The Bayesian approach in biostatistics and epidemiology is better in comparison to P evaluation, according to Goodman (1999) and as far as multiple test is concerned, according to Garret and Parmigiani (2003) and to Scott and Berger (2004). RESULTS Genes were classified on the basis of their relative expression amounts in the different environments as: a) constitutively expressed genes; b) genes expressed only when MTB grows on a synthetic medium; c) genes expressed only when MTB replicates in pure human macrophages; d) genes expressed in human macrophages whether or not in co-colture with autologous peripheral blood mononuclear cells (PBMCs); e) genes expressed in human macrophages only in co-colture with autologous PBL (PBMC), as shown in figure 1. Note that the gene expression of the two MTB strains reveals an imbalance. In fact, as is shown in figure 2, panel A, the major number of genes (60,4%), whose expression is statistically validated, is transcripted by the two mycobacterial strains in human phagocytes. A lower number of genes is expressed constitutively (13%), while an intermediate number is transcripted in synthetic medium (26,6%). The number of unknown proteins in each group as shown in figure 2, panel B, is in an average of 25%, ranging from 18,4% for the synthetic medium expressed genes, to 29,7% for the genes expressed in a constitutive fashion. After splitting the in vivo data (genes expressed in human macrophages alone or in co-culture with PBLs) in the three specific patterns of gene expression, as shown in figure 2 panel C, it has been showed that the major number of in vivo genes were expressed in pure human macrophages (51 ,8 % of in vivo expressed genes), followed by the gene expressed both in pure macrophages and macrophages in co-colture with PBLs (37,7% of in vivo expressed genes), while the genes expressed only in macrophages in co-culture with autologous PBLs, represent the minority (13,5% of in vivo expressed genes). The average value of unknown proteins in those three sub-groups is 31 %, as shown in panel D of figure 2. Further the different genes expressed in each of the groups were analyzed and splitted in functional categories as to the MTB metabolism. The analyses of these categories, as shown in Figure 3, disclosed the presence of a fine tuning modulation of the chosen functions and of short distances among the gene groups. In the case of genes expressed in PBMCs many gene functional categories are not represented, and this might easily be explained by the low number of genes transcribed by MTB in an hostile environment for its replication. In PBMCs group of peculiar genes groups there is another peculiarity: 21 ,7% of them belongs to the MTB 600 kb chromosomal region, close to the origin of replication, that lacks insertion sequences. Given that these mobile sequences can be responsible for insertions in ORFs, the areas in which there are no insertion sequences is likely to contain genes vital for MTB survival (Gordon et al., 1999). The percentage of genes belonging to this "protected" region is 27% in constitutively expressed genes, 17% in genes expressed in vitro, 14,7% in genes expressed in pure macrophages and 9,2% in genes expressed in macrophages both without or in co-culture with PBLs. The last observation is that many gene clusters contemporarily expressed were absent, contrarily to the evidences of other studies (Fisher et al., 2002).

The first group of genes identified whose expression is exclusive in a synthetic medium is composed of 37 genes. The selection of 30 MTB expressed genes in synthetic medium is shown in table 2; there are not proteins with unknown function.

Table 2 GENE FUNCTION Rv 0137c msrA Defence from RNl (only by CMT97) Rv 0418 Probable aminopeptidase Rv 0956 purU Formil-transferase Rv 1221 sigE Transcription Rv 1866 Energy production and conversion Rv 1909c furA Iron metabolism Rv 2373c dnaJ2 Chaperones Rv 2428 ahpC Defence from ROI Rv 2626c Hypoxic response Rv 3219 IΛV/7/^'B1 Regulatory protein Rv 3286c sigF Transcription Rv0232 Transcriptional regulator protein Rv0334 rmlA Cell envelope biogenesis Rv0382c umpA Prob. phosphoribosyltransferase Rv0525 Carbohydrate transport and metabolism Rv0684 fusA Translation, ribosomal structure Rv0709 rpmC Translation, ribosomal structure Rv0709 rpmC Translation, ribosomal structure Rv0912 Possible transmembrane protein Rv0920 Probable transposase for IS1554 Rv1042 Probable IS-like 2 transposase Rv1106 Probable cholesterol dehydrogenase Rv1799 IppT Probable lipoprotein Rv2224c Probable protease Rv2245 kasA INH exposure Rv2597 Possible membrane protein Rv2639c Prob. integral membrane protein (only by CMT97) Rv3078 hab Nitrobenzene degradation Rv3460c rpsM Translation, ribosomal structure Rv3475 Possible IS6110 transposase

In this selection of highly expressed genes we found two alternative sigma factors, sigE (Rv1221 ) and sigf (Rv3286c), the first of which was reported to be slightly expressed in exponentially growing mycobacteria (Manganelli et al., 1999). The MTB mutant strain, lacking sigE, was shown to be more sensitive than the wild type strain to heat shock, SDS and oxidative stress.

The results obtained are in contrast with two previous observations which concern respectively the sigE specific mRNA expression from macrophage-grown bacteria (Graham et al., 1999), and the growth of a sigE mutant strain of MTB in human non activated macrophages (Manganelli et al., 2001). The discrepancies are only apparent and are due to the different experimental conditions concerning the time, the cells and the bacterial strains that were employed during sigE mRNA detection and that do not render comparable the obtained results. Another gene whose exclusive expression in synthetic medium contrasts with other studies results kasA (Rv2245), known to be upregulated in response to isoniazid exposure. Fenhalls found the presence of kasA mRNA in human lung granulomas (Fenhalls et al., 2002), and this observation suggests to be very cautious in conferring general relevance to the results of the experiments performed in synthetic medium.

The authors of the present invention have identified in Sauton's medium the genes involved in the oxidative stress, such as msrA (Rv0137c), ahpC (Rv2428) and furA (Rv1909c). The msrA protein is involved in the response to oxidative damage and reactive NO intermediates (St John et al., 2001); the second protein ahpC is involved in the resistance to peroxynitrite (Master et al., 2002); the third protein furA is a ferric up-take regulator, whose transcription is induced upon oxidative stress (Milano et al., 2001). Finally, it has been found a relevant expression of the three mycobacterial transposases, such as: IS1554 (Rv0920), IS6110 (Rv3465) and IS-like2 (Rv1042) (Mariani et al. 1993), a family of mobile elements of MTB which are able to duplicate, cut themselves from a certain locus in the chromosome and re-integrate their sequence in another locus. Their expression and the effect of their mobility could be the reason of the genomic plasticity of MTB.

The second group of genes comprise the genes expressed both in MΦ and in PBMCs. The percentage of hypothetical expressed proteins in this group is 30,5 %, thus suggesting that mycobacterial survival strategy in human cells needs quite a good number of proteins whose functions are still to be identified. A selection of the 44 MTB genes expressed only in human phagocytes (pure MΦ and MΦ + PBL) is shown in table 3; there are 11 proteins whose function is unknown (25%).

Table 3 GENE FUNCTION Rv0098 Unknown Rv0125 pepA Post-translational modification, turnover, chaperones Rv0182c sigG Transcription Rv0353 hspR Heat shock response

Rv0467 aceA Latency

Rv0682 rpsl Translation, ribosomal structure/ function

Rv0738 Unknown, prob. membrane protein

Rv0755c PPE

Rv0877 Unknown

Rv0921 Possible resolvase for IS 1535

Rv1003 Unknown

Rv1009 rpfB Resuscitating factor

Rv1020 mfd DNA replication, recombination and repair

Rv1047 Probable transposase

Rv1124 ephC Probable epoxide hydratase

Rv1343c IprD Probable conserved lipoprotein

Rv1382 Possible transport or membrane protein

Rv1481 Probable membrane protein

Rv1503c Cell envelope, outer membrane

Rv1546 Unknown

Rv1610 Possible conserved membrane protein

Rv1612 trpB Aminoacid transport and metabolism

Rv1745c Lipid metabolism

Rv1767 Unknown (RD14 Deletion Region BCG)

Rv1893 Unknown

Rv1966 mce3 mce operon (RD15 Deletion region BCG)

Rv2108 PPE

Rv2289 cdh Similar to cdp-dyacilglycerol pyrophosphatase

Rv2536 Probable transmembrane protein

Rv2633c Unknown

Rv2666 Partial IS 1081 transposase

Rv2681 Translation, ribosomal structure/function

Rv2796c IppV Probable lipoprotein

Rv2810c C- domain of transposase for IS 1555

Rv2955c Unknown

Rv3050c Probable transcription regulatory protein

Rv3086 adhϋ Probabile alcohol dehydrogenase

Rv3110 moaB Prob. involved in molybdopterin synthesis

Rv3119 moaE Coenzyme metabolism (RD5 Deletion Region BCG) Rv3363c Unknown Rv3396c guaA Transcriptional regulator Rv3571 hmp Flavohemoglobin Rv3629c Probable integral membrane protein Rv3702c Unknown

Some genes belonging to this group are noteworthy, like sigG (Rv0182c), hmp (Rv3571), guaA (Rv3396c), mce3 (Rv1966), aceA (Rv0467) and rpfB (Rv1009). The first one sigG is an alternative sigma transcription factor. Interestingly, it was shown that its expression is downregulated in mycobacteria exposed to various stresses (heat shock, exposure to SDS, low aeration) (Manganelli et al., 1999). On the contrary, its strong expression in human macrophages, either resting or activated by autologous PBLs, might suggest that it plays a key role in directing the transcription of new genes in response to the intracellular environment stimuli. Moreover, a recent study showed that a Mycobacterium bovis strain exhibiting a deletion at the sigG gene was less virulent than the wild type strain (Collins et al., 2002). The second selected gene, hmp, encoding for a flavohemoglobin which has been described as a protein involved in mycobacterial response to oxygen limitation and NO derivates or radicals stress (Hu et al., 1999), and therefore its relevant expression in human macrophages after the exposure of MTB to oxygen and nitrogen reactive species is consistent with this function. The third gene, guaA, encoding for the guanosine monophosphate (GMP) synthetase, is regulated by a novel transcriptional regulator found in MTB, shown to be involved in the determination of the growth rate also in stringent conditions (Kamalakannan et al., 2002). Therefore, the expression of the downstream gene guaA might be the evidence of the activity of its promoter in the course of in vivo replication of MTB. The gene mceZ belongs to the MTB mce3 gene family, firstly shown to be involved in the genes capacity of entry into eukaryotic cells (Arruda et al., 1993) and to code for proteins exposed on the external face of the cytoplasmic membrane (Tekaia et al., 1999). This gene has been shown to be normally repressed by an upstream regulator (Rv1963), and a deletion at this level lead to an increase of the mce3 expression in a macrophage-like cell lines, like J774, compared to the in vitro culture (Santangelo et al., 2002). So the fact that the authors of the present invention show that mce3 is transcribed even in the presence of its repressor, might confirm its role in the entry phase of MTB in human cells. Interestingly, the A7?ce3 gene DNA sequence is not present in BCG genome, since it belongs to the RD15 deletion region of the avirulent strain of mycobacterium, in which, is present the upstream regulator Rv1963c (Behr et al., 1999). The fifth gene aceA, codes for isocitrate lyase. In a previous study it was found to be expressed by MTB in response to phagocytosis by human macrophages (Graham et al., 1999). Its presence was shown to be necessary to MTB persistence and virulence in immune competent mice (McKinney et al., 2000). The authors findings about the transcription of the gene transcribed by MTB in human macrophages is coherent with the results of the previous studies relating to its expression features. The last gene rpfB is a particular "resuscitation promoting factor", identified previously in Micrococcus luteus (Mukamolova et al., 1998). This gene codes for a factor which promotes the resuscitation and growth of dormant, non-growing cells, and therefore it was defined as a bacterial cytokine. Its expression in human macrophages, either alone or in co-culture with autologous PBLs, is particularly relevant to awake cells, considering the dormant cellular as a MTB strategy to escape the macrophages defence mechanisms. The last observation regards two other expressed genes, moaE

(Rv3119) and Rv1767, both of them belonging to two different deletion regions of BCG genome, respectively RD5 and RD14. Their expression in human macrophages might confirm that BCG becomes a live attenuated strain also because it loses important virulence genes, instead present in MTB. Genes whose expression was exclusive only in pure human macrophage culture were also identified. This group is composed of 88 genes. A selection of the 39 MTB genes only in pure macrophages is shown in table 4; 7 proteins whose functions have not been identified yet (17,9%) were identified. Table 4

GENE FUNCTION Rv0099 ac DIO Lipid metabolism

Rv0159c PE

Rv0222 Unknown (Deletion Region RD14 BCG)

Rv0256c PPE

Rv0492c Probable oxidoreductase

Rv0510 hemC Coenzyme metabolism

Rv0567 Possible methyltransferase (only by CMT97)

Rv0703 rpl\N Translation, ribosomal structure/function

Rv0705 rpsS Translation, ribosomal structure/function

Rv0722 rpmD Translation, ribosomal structure/function

Rv0818 Probable transcriptional regulator (only by CMT97)

Rv0885 Unknown

Rv0965c Unknown (repressed by acidic treatment)

Rv0973c accA2 Lipid metabolism (only by CMT97)

Rv1274 IprB N-term signal seq., lipoprotein

Rv1359 Probable transcriptional regulator

Rv1391 dfp Coenzyme metabolism

Rv1460 Probable transcriptional regulator

Rv1511 gmdA GS gene (RD6 Deletion Region BCG)

Rv1584c Prob. PhiRVI phage prot. (RD3 Deletion Region BCG)

Rv1718 Unknown

Rv1772 Unknown (RD14 Deletion Region BCG)

Rv1811 mgtC MgCI₂ transport, P type ATPase C

Rv1869c Probable reductase

Rv2098c PE_PGRS

Rv2278 Probable IS6110 transposase

Rv2350c pclB Intermediate metabolism (RD7 Deletion Region BCG)

Rv2517c IppS Unknown

Rv2545 Unknown (repressed by acidic treatment)

Rv2893 Energy production/conversion (repressed by acidic treatment)

Rv2989 Transcription

Rv3004 Low MW antigen 6 (only in CMT97)

Rv3016 IpqA Probable lipoprotein Rv3194c Possible secreted protein Rv3239c Probable transmembrane protein Rv3246c mtrA Signal transduction Rv3250c rubB Energy production/ conversion (repressed by acidic treatment) Rv3425 PPE (RD11 Deletion region BCG) Rv3478 PPE Rv3507 PE_PGRS Rv3508 PE_PGRS Rv3571 Energy production and conversion Rv3643 Unknown (repressed by acidic treatment)

Some of the selected genes display particular features: for example in the case of gmdA (Rv1511 ) which is homologous to M.avium gsbA, a gene located in a low GC content island, with about a G+C content 9% lower than the mean G+C values of this specie. Such a region displays a number of features in common with the pathogenicity island, and represents the first of such element identified in mycobacteria (Tizard et al., 1998). Furthermore, gmdA gene belongs to the RD6 deletion region, that is absent in the attenuated strain BCG. Another interesting gene strongly expressed in pure macrophages culture, belonging to a deletion region absent in BCG (RD11) is the PPE gene (Rv3425), belonging to the PPE family protein. During the selection process of the most expressed genes other two genes encoding for the PPE proteins, such as, Rv0256c and Rv3478, were found. In the same group shown in table 4, 4 out of 39 strongly expressed genes belong to another family of related proteins, the PE/PE_PGRS (Ramakrishnan et al., 2000), that are Rv2098c, Rv3507, Rv3508, and Rv0159c. These proteins are rich in prolin and glycin, that might represent a source of antigenic diversity for MTB clinical strains, due to their inner presence of a variable number of repeat sequences (VNTR), that can be reduced or increased by MTB during its replication. Furthermore, there are three genes involved in translation processes, rplW (Rv0703), rpsS (Rv0705) and rpmD, (Rv0722), and two genes involved in coenzyme metabolism dfp (Rv1391) and hemC (Rv0510). The genes involved in lipid metabolism are three: facfDIO (Rv0099), accA2 (Rv0973c), and IpqA (Rv3016), and only one gene is necessary for signal transduction, mfrA (Rv3246c). Finally, also two reductases were found (Rv1869c and Rv0492c) and four genes dedicated to the regulation of transcription (Rv2989, Rv1359, Rv0818 and Rv1460), findings that confirm how accurate the MTB tuning of gene expression could be. The authors showed also a relevant expression of a probable 1S6110 transposase (Rv2278). The last group of genes (Rv3250c, Rv0965c, Rv2545, Rv3643 and Rv2893) was recently found to have a remarkable repressed expression by acidic treatment on MTB synthetic cultural medium. On the contrary, the authors of the present invention found that such a gene group is strongly expressed during pure macrophage infection, and this finding might suggest that those intracellular mycobacteria were not facing an acidic environment: This could be due either to the inhibition of the phagosome-lysosome fusion and therefore to an intra-phagosomal environment not acidic, or to the capacity of MTB to escape the acidic phago-lysosome towards the cytosol. Both of these mechanisms are well known mycobacterial tools to avoid the destruction by activated macrophages (Malik et al., 2003; McDonough et al., 1993), and therefore the expression of those genes represent a good indicator of the intracellular pH value that mycobacteria are facing in the experimental conditions. Subsequently genes expressed only in macrophages co- cultivated with autologous PBLs were studied and this group of genes was carefully being esteemed, given that their expression by MTB in the human macrophages activated by the co-culture with autologous PBLs, could play a central role in mycobacterial defence mechanisms from host phagocyte response. Particularly, these genes are expressed almost exclusively by H37Rv strain, because CMT97 strain transcribes many genes in MΦ in co-culture with PBLs, but not exclusively (they are expressed also in pure MΦ culture). Viceversa, H37Rv strain gene expression decreases dramatically in PBL activated macrophages, but the few genes that this strain transcribes were not found elsewhere. The group is composed of 25 genes; the selection of 24 of these MTB genes expressed only in macrophages co-cultivated with PBLs, is shown in table 5; 9 proteins were identified whose function is still unknown (37,5%). Table 5 GENE FUNCTION Rv 0108c Unknown Rv 0149 Energy production and conversion Rv 0201 Unknown Rv 0356 Unknown Rv 0991c Unknown, serine-rich in C-term Rv 1096 Possible glycosyl hydrolase Rv 1114 Unknown Rv 1130 Unknown Rv 1255c Transcription (RD10 Deletion Region of BCG) Rv 1517 Hypotetical protein, similar to M.leprae Rv 1600 hisC Aminoacid transport and metabolism Rv 1613 trpA Aminoacid transport and metabolism Rv 1713 Probable GTP-BP Rv 1751 Possible hyroxylase Rv 2091 c Repetitive ORF, possible trans-membrane Rv 2163c pbpB Probable penicillin-binding-protein Rv 2244 acpM Lipid metabolism Rv 2269c Unknown Rv 2489c Unknown Rv 3042c serB2 Probable phosphoserine phosphatase Rv 3082c virS Transcription Rv 3140 fadE23 Lipid metabolism Rv 3724 Probable cutinase precursor Rv 3911 s/fifM Transcription

Among the 15 genes already characterised 3 of them deserve a particular note: virS (Rv3082c), pbpB (Rv2163c) and s/gM (Rv3911). The first one, virS, exhibits homology with several virulence-regulating and controlling proteins of other pathogenic bacteria (Gupta and Tyagi, 1993), hence suggesting the possible involvement of the gene in the regulation of pathogenesis of MTB (Gupta et al., 1999). The second gene, pbpB, has been described as a possible penicillin-binding-protein (Cole et al., 1998), whose expression in the activated macrophages could work as a protection factor of mycobacteria from the effect of the antibiotic. The expression of such a gene is not the consequence of drug exposure of MTB. The third gene, sigM, belongs to the family of alternative sigma factors, that provide mycobacteria with a means of regulating gene expression in response to changing environmental stimuli. This gene was low expressed in exponential growing MTB cultures, while a slight increase in response to heat shock can be seen and a significant decrease following SDS exposure, in conditions of low aeration and in the stationary phase of growth of mycobacteria (nota of Manganelli, R. Mol. Microbiol., 1999). Therefore, the gene expression in activated human macrophages suggests a specific role of this sigma factor in directing the transcription of new genes by MTB. The last observation concerns Rv1255, a gene involved in transcription, which belongs to the R10 Deletion Region of BCG. Finally, genes that are constantly expressed and in a substantial way were studied: the selection of 24 genes among which only two code for protein of unknown function (8,3%), is shown in table 6.

Table 6 GENE FUNCTION Rv 0001 dnaA DNA replication Rv 0172 mcel D Mce-family protein Rv 0176 mce-associated transmembrane protein Rv 0187 O-metyl-transferase Rv 0350 dnaK Heat shock response Rv 0352 oV?aJ Chaperon Rv 0394c Secreted protein Rv 0491 regX3 Signal transduction Rv 0667 rpoB Transcription Rv 0710 rpsQ Translation ribosomal structure Rv 0747 PE_PGRS Rv 0829 DNA replication, recomb. and repair Rv 1265 Macrophage infection Rv 1434 Unknown Rv 1519 Unknown Rv 1661 pks7 Energy production and conversion Rv 1868 Cell envelop biogenesis Rv 2374c hrcA Heat shock response Rv 2620c Probable transmembrane protein Rv 2711 ideR Transcriptional regulator Rv 2929 Similar to C-term urease accessory prot. Rv 2994 Membrane protein, similar to FQ pump Rv 3608c folP Coenzyme metabolism Rv 3841 fba Hypoxic response

In this group of 24 highly expressed genes many of the functional categories identified in the present study are represented: two genes responsible for the heat shock response (dnaK, Rv0350, and hrcA, Rv2374c), two genes for DNA replication (Rv 0829, and dnaA, Rv0001), the well known transcriptional regulator iron-dependent ideR (Rv2711) (Rodriguez et al., 2002), the beta sub-unit of RNA polymerase beta rpoB (Rv0667), a gene involved in hypoxic state response, bfrB (Rv3841), one gene encoding for a protein belonging to the PE_PGRS family (Rv0747), a gene involved in signal transduction, regX3 (Rv0491) (Ewann et al., 2002), and a membrane protein, Rv2994, similar to fluoroquinolones (FQ) efflux pumps. It is quite reasonable that genes that were transcribed by MTB in a constitutive fashion span over the general bacterial metabolism categories, and in fact there is also a gene for energy production (pks7, Rv1661), another for coenzyme metabolism (folP, Rv3608c), a gene encoding for a chaperone protein (dnaJ, Rv0352), and finally another involved in translation processes (rpsQ, Rv0710). Surprisingly, the authors of the present invention found Rv1265, a gene that seems to be induced by macrophage infection (Hobson et al., 2002), highly expressed in all the environments. As above mentioned, 27% of the genes constantly expressed belong to the 600 kb MTB chromosomal region, close to the origin of replication, in which insertion sequences are absent, suggesting that DNA insertions in this areas could be detrimental for the bacterium survival. The comparison of MTB strains H37Rv versus CMT97 in global gene expression profile patterns and of their replication conditions are shown in relation to the gene expression pattern in figure 4. It's easy to note that CMT97 strain of MTB expresses roughly the same number of genes when it replicates in macrophages alone and/or in macrophages in co-culture with PBLs and this might suggest a better adaptation of this clinical isolate strain to the hostile environment of the activated macrophages, as previously shown (Cappelli et al. 2001). On the contrary, H37Rv diminishes considerably the gene transcription in macrophages in co-culture with PBLs, compared to macrophages as such, but the few genes that it transcribes in PBL- activated macrophages are expressed only by H37Rv. Consequently, when the two "in vivo" transcriptome were plotted one against the other it has been found a good number of genes whose expression was much higher in one strain rather than other one, as shown in the diagram of figure 5. The two transcriptome were more similar when the two strains replicated in synthetic medium, as shown in the diagram of figure 6. Then, the authors have selected a group of highly expressed genes by the two strains, according to their differential gene expression in human macrophages, as shown in table 7, panels A and B. In panel A the group of genes expressed by CMT97 strain in human macrophages in an amount ranging from 10 to 100-fold more than H37Rv strain, is shown. In panel B, genes expressed by the H37Rv strain in human macrophages in an amount ranging from 10 to 100-fold more than CMT97 strain, are shown.

Rv3899c Unknown

Rv0106 Unknown

Rv0858c Probable aminotransferase

Rv2493 Unknown

Rv2730 Unknown

Rv2240c Unknown

Rv2828c Unknown

Rv1695 Catalyses of the phosphorylation of NAD

Rv1134 Unknown

Rv2980c rpsB Probable 30S ribosomal protein S2

It's clearly evident that in the group of genes expressed by H37Rv in human macrophages more than in CMT97 strain, as shown in table 7 panel A, there are several unknown proteins, more than in the group expressed by CMT97 (panel B). In the same group of expressed genes there are also two probable transposases (Rv 2961 and Rv1047), and two probable phiRvl phage protein (Rv1575 and Rv1576c), coming from the expression of a chromosomal region (the biotin operon) in which the phiRvl mycobacterial prophage had integrated (Cole et al, 1998). Interestingly, both genes belong to the RD3 Deletion Region of BCG. Quite differently, in the group of genes expressed in human macrophages more by CMT97 strain rather than H37Rv strain, there are less unknown proteins, two PPE genes, and many genes involved in mycobacterial metabolism. Particularly leuC gene (Rv2988c), involved in leucine biosynthesis, is expressed by the CMT97 clinical strain in human MΦ, not only more than H37Rv strain, but also much more than by CMT97 strain itself in MΦ co-cultivated with PBLs, as shown in table 8, panels A and B. In panel A is reported a selection of the genes that CMT97 strain expresses in MΦ + PBLs much more than in isolated MΦ. In panel B is reported the selection of genes that CMT97 strain expresses much more than MΦ + PBLs. Table 8 PANEL A GENE FUNCTION Rv1434 Unknown Rv1767 Unknown (RD14 Deletion Region of BCG) Rv1768 PE_PGRS (RD14 Deletion Region of BCG) Rv2289 cdh Phospholipid biosynthesis Rv2376c Secretion protein, prob. involved in the immune response, low MW antigen Rv2487c PE_PGRS Rv2545 Unknown Rv2768c PPE Rv2920c amt Membrane transport of ammonium Rv2933 ppsC Phenopthiocerol synthesis type I polyketide synthase Rv3056 dinP Possible DNA damage inducible protein Rv3086 adhϋ Prob.able zinc-type alcohol dehydrogenase Rv3129 Unknown

PANEL B GENE FUNCTION Rv0125 pepA Probable serine protease Rv0701 rplC Probable ribosomal protein Rv0703 rpl\N Probable 50S ribosomal protein L23 Rv0877 Unknown Rv0942 Unknown Rv0973c accA2 Lipid metabolism Rv2988c leuC Leucine biosynthesis Rv3571 Possible oxidoreductase, electron transfer

In order to compare the two MTB strains gene expression in PBL-activated MΦ, the authors analyzed the genes expressed by CMT97 strain, 10 to 100-fold more in PBMCs than in pure MΦ and viceversa, as shown in figure 7 and the results are shown in table 8 (the same comparison for H37Rv strains was not possible given the few genes that this strain expresses in PBMCs). The selected ORFs of CMT97 in PBMCs were transcribed as single-gene mRNA, whereas those coming from infection of pure MΦ were co-transcribed as multigene transcripts. In the table 8 panel A are reported genes that CMT97 strain transcribes much more in PMBCs than in MΦ and among them there are Rv1767 and Rv1768 genes. The first is still of unknown function, while the second belongs to the PE_PGRS family, and both of them belong to the RD14 deletion region of BCG. In the table 8, panel B are shown genes that CMT97 strain transcribes more in MΦ than in PBMCs, and among them leuC is an important one (Rv2988c), a gene involved in leucine biosynthesis, that is probably transcripted under the control of the seπX3-regX3 two- component regulatory system, most likely required for virulence (Parish et al., 2003). Another gene preferentially expressed in MΦ by CMT97 is rp/C (Rv0701 ), a ribosomal protein, most likely regulated by the transcriptional repressor hspR, responsible for the control of the expression of the heat shock proteins of MTB (Stewart et al., 2002). As far as the mycobacterial replication in the two populations of infected cells is concerned, the CFU enumeration shows that both strains are better kept under control by PBMCs than by MΦ, with a CFU reduction in PBMCs slightly less than 1 log. In addition, the CMT97 strain CFU in macrophages are 2,5-fold more than the CFU of the H37Rv strain. In conclusion, the data shown confirm at the whole MTB genome level, the previous results concerning the dependence of the MTB gene expression on the environment in which it finds itself. Moreover, the present study confirm that the CMT97 clinical strain reacts to human macrophage activation enhancing its gene expression and reducing itself replication. In fact in PBL-activated MΦ this strain continues to transcribe many genes, thus reducing its intracellular replication roughly of one sixth, as shown in figure 8. Whereas the H37Rv strain decreases its gene expression dramatically. Some important features are worth noting: i) more than 60% of expressed genes were transcripted by the two MTB strains only when they replicate in human phagocytes; ii) there are genes belonging to the DNA regions deleted in the genome of the attenuated strain BCG, that are strongly expressed by MTB only in human macrophages and not in synthetic medium or in a housekeeping fashion. We do not attribute the virulence feature to these genes, but most likely their strong expression only in human cells, put them into a group of genes that MTB employs to survive in vivo; iii) an average of 25% of the genes expressed by the two MTB strains have not known function yet; iv) the relevant expression of ISs takes place only in synthetic medium, and this might depend on the absence of selective pressure, that MTB encounters in human macrophages, given the importance that DNA insertion might have for the intracellular survival of the bacillus . The comparison of global gene expression profiles between the two MTB strains revealed a better adaptation of the clinical strain to human macrophages, demonstrated by the higher number of transcripts in vivo from the mycobacterial metabolism genes, which were transcribed highly in human macrophages than in H37Rv comparison strain. Another evidence is given by the lower CFU number that the CMT97 strain reaches in PBL-activated macrophages, compared with the CFU number reached by the same strain in macrophages. If we mean for adaptation a process by which the pathogen survives in the macrophage without hurting too much the host cell, the CMT97 strain and human macrophages are better adapted one each other. On each of the MTB genes relevantly up-regulated, and belonging to some RD regions of BCG compared to MTB, was selected an average of 6 epitopes, that were subsequently subjected to data bank NCBI BLAST analysis to discard to the most conserved sequences among bacteria. Different combinations of the selected peptides were employed in IFN-γ ELISPOT assay to ascertain their immunogenicity for TB patients and healthy contacts. The study is still in course, and the selected peptides are being compared to some peptides belonging to the region RD1 , absent in each BCG strain, that are well known to be strong inducers of T CD4+ cells IFN- v production. From this analyses the more immunogenic peptides for human being will appear, to be selected both for the preparation of diagnostic kit and for the preparation of an anti- tubercular vaccine In conclusion, these data obtained during the study will be important both to add new insights into the knowledge of MTB mechanisms and behaviour in the course of infection, to the aim of identifying the important factors able to trigger enhanced resistance of mycobacteria, and to provide for new diagnostic or protective tools for the preparation of diagnostic kits and/or vaccines specific for the tubercular disease. BIBLIOGRAPHY

Alland D., Kramnik I., Weisbrod T.R., Otsubo L, Cerny R., Miller L.P., Proc. Natl. Aca. Sci., 1998; 95:13227-13232. - Arruda S., Bomfim G., Knights R, Huima-Byron T., and Riley L.W., Science, 1993; 261 :1454-1457. Behr M.A., Wilson M.A., Gill W.P., Salamon H., Schoolnik G.K., Rane S., Science, 1999; 284: 1520-1523. Betts JC, Lukey P.T., Robb L.C., McAdam R.A., and Duncan K. , Mol. Microbiol.,2002; 43: 717-731. Camus J.C., Pryor M.J., Medigue C, and Cole S., Microbiology, 2002; 148: 2967-2973. Cappelli G., P., Volpe, A., Sanduzzi, A., Sacchi, V., Colizzi, and F. Mariani., Infect. Immun., 2001 ; 69: 7262-70. - Chan K., Knaak T., Satkamp L., Humbert O., Falkow S., and Ramakrishan L, Proc. Natl. Aca. Sci., 2002; 99: 3920-3925. Chomczynski, P., and N. Sacchi., Anal. Biochem., 1987; 162:156- 159. Cole ST., Brosch R., Parkhill J., Gamier T., Churcher C, Harris D. , et al. , Nature, 1998; 393: 537-544. Collins D.M., Wilson T., Campbell S., Buddie B.M., Wards B.J., Hotter G., De Lisle G.W., Microbiology, 2002; 148: 3019-3027. Dubnau E., Fontan P., Manganelli R., Soares-Appel S., and Smith I. Infect. Immun., 2002; 70: 2787-2795. - Ewann F., Jackson M., Pethe K., Cooper A., Mielcarek N., Ensergueix D., Gicquel B., Locht C, and P. Supply., Infect. Immun., 2002; 70: 2256-2263. Fenhalls G., Stevens L., Moses L., Bezuidenhout J., Betts J.C., Helden P.V. et al., Infect. Immun., 2002; 70: 6330-6338. - Fisher M.A., Plikaytis B.B., and Shinnick T.M., J. Bacteriol., 2002; 184: 4025-4032. Garrett E.S: and Parmigiani G., The analysis of gene expression data: methods and software, 2003. Gillespie, SH., Int. J Antimic. Agents, 2001 ; 17: 171-176. - Goodman S. Annals of internal medicine, 1999a; 130, 995-1004. Goodman S. Annals of internal medicine, 1999b; 130, 1005-1014. Gordon S.V. Heym B., Parkhill J., Barrel B., and Cole ST., Microbiology , 1999; 145: 881-892. Graham JE., and J.E. Clark-Curtiss. Proc. Natl. Aca. Sci., 1999; 96: 11554-11559. Gupta S, Jain S., and Tyagi A.K. FEMS Microbiol. Letters, 1999; 172: 137-143. - Gupta S., and Tyagi A.K., Gene, 1993; 126: 157-158. Hobson R.J., McBride A.J., Kempsell K.E., and J.W. Dale, Microbiology, 2002; 148: 1571-1579. Hu Y., Butcher P.D., Mangan J.A., Rajandream M.A., and A.R. Coates, J. Bacteriol., 1999; 181: 3486-3493. - Kamalakannan V., Ramachandran G., Narayanan S., Vasan S.K., and P.R. Narayanan, FEMS Microbiol. Lett, 2002; 209: 261-266. Kaufmann SHE, Sher A., and R. Ahmed, Washington, D.C.: ASM Press, pp.1-2, 2002. Kent L, T.D. McHuge, O. Billington, J.W. Dale, and S.H. Gillespie, J Clin Microbiol, 1995; 33:2290-2293. Landi S., The Mycobacteria, Part A, 1984; 15: 505-535; Lalvani A, et al., The lancet, 2001 ; 357:2017-2021. Malik Z.A., Thompson C.R., Hashimi S., Porter B., Iyer S.S., and Kusher D.J., J. Immunol., 2003; 170: 2811-2815. - Manganelli R., Dubnau E., Tyagi S., Kramer F.R., and Smith I., Mol. Microbiol., 1999; 31 : 715-724. Manganelli, R., Voskuil M.I., Schoolnik G.K., and Smith I., Mol. Microbiol., 2001; 41 : 423-437. Mariani F., Bocchino M.L., Cappelli G., Persichini T., Colizzi V., Bonanno E., Ponticiello A., and S. Sanduzzi, Monaldi Arch, for Chest Diseases , 2001 ; 56: 30-32. Mariani F. Cappelli G. Riccardi G. and V. Colizzi, Gene, 2000; 253: 281-291. Mariani F., Piccolella E., Colizzi V., Rappuoli R., and Gross, R., Journal of General Microbiology, 1993; 139, 1767-1772. Master SS., Springer B., Sander P., Boettger E.C., Deretic V., and Timmins G.S., Microbiology, 2002; 148: 3139-3144. McDonough K.A, Kress Y., and Bloom B. R., Infect Immun., 1993 ; 61 : 2763-2773. - McKinney J.D., et al., Nature, 2000; 406: 735-738. McLachlan G., Bean R.W. and Peel D:, Bioinformatics, 2002; 18, 3: 413-422. Milano A., Forti F., Sala C, Riccardi G., and Ghisotti D., J Bact., 2001 ; 183: 6801-6806. MMWR Morb Mortal Wkly Rep., 1997; 46: 846-851. Moussa OM., Eraky I., El-Far MA., Osman HG., e Ghoneim MA. J. Urology, 2000; 164: 584-588. Mukamolova G.V., Kaprelyants A.S., Young D.I., Young M., and Kell D.B., Proc. Natl. Aca. Sci., 1998; 95: 8916-8921. Parish T, Smith D.A., Roberts G., Betts J., and Stoker N.G., Microbiology, 2003; 149: 1423-1435. - Ramakrishnan L., Federspiel N.A., and S. Falkow, Science, 2000; 288: 1436-1439. Robert C, The Bayesian Choice, Springer Verlag, 2001. Rodriguez G. M., Voskuil M.I., Gold B., Schoolnik G.K., and I. Smith, Infect. Immun., 2002; 70: 3371-3381. - Rosenkrands I., Slayden R.A., Crawford J., Aagaard C, Barry III C.E., and Andersen P., J Bacteriol., 2002; 184: 3485-3491. Russell J., Hobson RJ., McBride A.J., Kempsell K.E., and Dale J.W., Microbiology, 2002; 148: 1571-1579. Santangelo M.P., Goldstein J., Alito A., Gioffre A., Caimi K., Zabal O. , et al. , Microbiology, 2002; 148: 2997-3006. Sauton B., Acad. Sci., 1912; 166:1860-1865. Scott and Berger J.O. An exploration of aspects of Bayesian Multiple Testing, Technical Report, Department of Statistics, Duke University, 2004. - St John G., Brot N., Erdjument-Bromage H., Tempst P., Weissbach H., and Nathan C, Proc. Natl. Aca. Sci., 2001 ; USA. 98: 9901-9906. Stewart G.R., Wernisch L, Stabler R., Mangan J.A., Hinds J., Laing KG. et al., Microbiology, 2002; 148: 3129-3138. - Tekaia F., Gordon S.V., Gamier T., Brosch R., Barrel B.G., and Cole ST., Tubercle and Lung Disease, 1999; 79: 329-342. Tizard M., Bull T., Millar D., Doran T., Martin H., Sumar N., Ford J., and J. Hermon-Taylor, Microbiology, 1998; 144: 3413-3423. World Heath Organisation. Report on the Tuberculosis epidemic, Geneva, 1996. Wills H.S., M.M., Diagnostic and experimental methods in tuberculosis, 1962; pag. 126. Wilson M., DeRisi J., Kristensen H.H., Imboden P., Rane S., Brown P.O. etal., Proc. Natl. Aca. Sci., 1999; 96: 12833-12838.

Claims

CLAIMS 1. Use of oligonucleotidic sequences, encoded proteins thereby or their portions expressed by M. tuberculosis in human macrophages, characterized in that belong to a deletion region of M. bovis BCG, for the preparation for the preparation of means for the diagnosis and the prevention of tubercular disease. 2. Use according to claim 1 , wherein said oligonucleotidic sequences are deoxiribonucleotidic or ribonucleotidic sequences and their complementary sequences. 3. Use according to any one of the claims 1 or 2, wherein said oligonucleotidic sequences are selected from the group consisting of the genes Rv1767, Rv1772, Rv1966, Rv3119, Rv1511 , Rv3425, Rv1255c, Rv1575, Rv1576c, Rv1584c, Rv0222, Rv2350c, Rv1768. 4. Use to any one of the claims 1 to 3, wherein the deletion regions of M. bovis BCG are selected from the group consisting of RD3, RD4, RD5, RD6, RD7, RD10, RD11 , RD14, RD15. 5. Use according to any one of the claims 1 to 4, wherein said means for the diagnosis of tubercular disease are selected from the group consisting of probes, antisense oligonucleotides, primers, antigens, antibodies. 6. Use according to claim 5, wherein the diagnosis is carried out by PCR, in situ hybridization, ELISA, T cell proliferation assay, ELISPOT, Immunoblotting. 7. Use according to any one of the claims 1 to 4, wherein said means for the prevention of tubercular disease consists in at least an antigen for the preparation of a vaccine. 8. Use according to claim 7, wherein said vaccine is a DNA vaccine specific for the tubercular complex. 9. Diagnostic kit for the detection of M. tuberculosis infection in a biological sample, which comprises at least one oligonucleotidic sequence or an encoded protein thereby, or their portions, as defined in claims 1 to

4. 10. Diagnostic kit according to claim 9, wherein said detection is carried out by PCR techniques, in situ hybridization, ELISA, T cell proliferation assay, ELISPOT, Immunoblotting.

11. Diagnostic kit according to any one of the claims 9 and 10, wherein the biological sample is selected from the group consisting in blood, saliva, sputum, urine, broncho alveolar lavage. 12. Vaccine for the prevention of M. tuberculosis infection comprising at least one oligonucleotidic sequence or an encoded protein thereby, or their portions, as defined in claims 1 to 4, together with one or more pharmaceutically acceptable excipients and/or adiuvants. 13. Vaccine according to claim 12, wherein said vaccine is selected from the group consisting in a DNA vaccine and in a vaccine based on at least one protein. 14. Vaccine according to any one of the claims 12 and 13, wherein said vaccine is a DNA vaccine specific for the tubercular complex. 15. Oligonucleotidic probes comprising at least one of the gene sequences, or their portions, as defined in claims 1 to 4, for the diagnosis of the tubercular disease through detection of M. tuberculosis by hybridization. 16. Primers comprising at least one of the oligonucleotidic sequences, or their portions, as defined in claims 1 to 4, for the diagnosis of the tubercular disease through M. tuberculosis detection in biological samples by PCR amplification of the gene sequences from mycobacterium. 17. Primers according to claim 16, wherein the biological sample is selected from the group consisting in blood, saliva, sputum, urine, broncho alveolar lavage. 18. Antigen comprising at least one oligonucleotidic sequence, or correlated protein thereby, or their portions, as defined according to any one of the claims 1 to 4, for the diagnosis of the tubercular infection through the detection of the antibodies presence in human serum against said sequences or correlated proteins by ELISA or Immunoblotting. 19. Monoclonal antibodies able to recognise and bind in a selective manner at least one protein, or its portions, correlated to the nucleotidic sequences as defined according to any one of the claims 1 to 4, for the detection of M. tuberculos s in biological samples.