DESCRIPTION

" _A N EFFECTIVE NEW DRUG TARGET FOR THE TREATMENT OF

TUBERCULOSIS"

Field of the invention The present invention allows a screening method for identifying novel drugs for the treatment of tuberculosis as well as a diagnostic method for identifying clinical strains that are resistant to these novel drugs. Background art

Tuberculosis (TB) remains the leading cause of mortality due to a single infectious agent, Mycobacterium tuberculosis. This pathogen has latently infected a third of the world population. Currently, TB chemotherapy is made up of a cocktail of first- and second-line drugs. The actual TB therapy takes from six to nine months, bringing to significant toxicity and drug resistance. Drug-resistant M. tuberculosis strains have frequently been encountered from the time anti-TB drugs were introduced. Indeed, mycobacteria are naturally resistant to most of the commonly used antibiotics due to the unusual cell wall. Moreover, genetic changes are considered for acquired drug resistance. Af. tuberculosis strains that are resistant to an increasing number of second-line drugs used to

treat multidrug-resistant tuberculosis (MDR-TB) are becoming a threat to public health worldwide (1 ⁾ . Consequently, there is an urgent need for new drugs to treat tuberculosis. In particular, we require new drugs having novel mechanisms of action that are active against drug-resistant strains. Accordingly, there is also a pressing need to identify new drug targets (2) .

Summary of the invention The present invention is based upon the discovery that the protein Rv3790 from Mycobacterium tuberculosis, whose encoding DNA-sequence can be retrieved at http : //genolist .pasteur . fr/TubercuList (TubercuList World- Wide Web Server; Rv3790 gene coordinates: from 4162306 to 4163718) , is the target for benzothiazinone drugs, a new class of molecules that appear to be very promising in the treatment of tuberculosis ("New thiazinone derivatives, their preparations and their use as antibacterials", pat. appln. No. PCT/EP2006/004942, in the name of Leibniz Institute for Natural Product Research and Infection Biology e.V.Hans-Knδll-Institute (HKI) .

It has been demonstrated that Rv3790 and Rv3791 proteins work in concert and they are involved into arabinogalactan biosynthesis (3) . Arabinogalactan is a fundamental component of mycobacterial cell wall, which covalently binds the outer layer of mycolic acids to

peptidoglycan. In particular, both Rv3790 and Rv3791 are important to perform transformation of decaprenyl-P- ribose to decaprenyl-P-arabinose (3) . The discovery suggests that benzothiazinone drugs interfere with mycobacterial cell wall biosynthesis. It is noteworthy that both rv3790-rv3791 genes were described as essential by Himarl-based transposon mutagenesis in H37Rv strain

(4) .

We found that mutations in the C-ter of the Rv3790 protein as well as its over-expression confer high level of resistance to benzothiazinone derivative drugs in

Mycobacterium tuberculosis, Mycobacterium bovis BCG and

Mycobacterium smegmatis. The rv3790 wild type and mutated genes, as well as rv3791, have been cloned and overexpressed in Escherichia coli.

Rv3790 orthologε sensitive to the benzothiazinone derivatives were found in the following genera: Nocardia, Rhodococcus and Corynebacteriurn.

It may be hypothesized that the excellent results obtained in the laboratory evaluation for benzothiazinone derivatives are correlated with the Rv3790-inhibitory action that has been discovered. This evidence can trigger new research on drugs acting against Rv3790 protein. Rv3790 protein and related homologues and particularly orthologs are thus not only optimal targets for drugs belonging to the class of benzothiazinones, but

also promising target for the discovery of new drugs to be utilized in fighting infections caused by Mycobacterium tuberculosis, Mycobacterium leprae, Nocardia, Rhodococcus and Corynebacterium. Therefore, it is an object of the present invention to provide a method for screening drug candidates for the treatment of tuberculosis by interfering with the arabinogalactan biosynthesis, the said method comprising a step of putting into contact a Mycobacterium tuberculosis cell culture over-expressing a protein that performs the transformation of decaprenyl-P-ribose to decaprenyl-P-arabinose and that can be encoded by rv3790 gene or homologues thereof or any open-reading artificial frame whose gene product is identical or homologue to Rv3790 protein, with a drug candidate and then evaluating the percentage of inhibition caused by the drug candidate against a control (M. tuberculosis transformed with pSODIT-2 vector alone) in an assay test.

Preferably, the assay test is an in vitro assay test of anti-bacterial activity or an enzymatic test.

Another object of the present invention is a cell culture preparation from Mycobacterium tuberculosis mutant strains (having the amino acid cysteine in the 387 position substituted by a serine or a glycine) to be used as a control tool in the method of the present invention.

A further object of the present invention is the preparation of recombinant Rv3790 protein in the wild type and mutated forms for enzymatic activity assay and screening method for identifying novel drugs. The discovery that Rv3790-type proteins encoded by mutated genes are responsible for resistance to drugs interfering with the arabinogalactan biosynthesis, such as benzothiazinone drugs, will have important implications in the diagnosis and treatment of drug- resistant mycobacterial strains.

It is thus another object of the present invention to provide a rapid diagnostic method for drug-resistant mycobacterial strains that involves amplification by Polymerase Chain Reaction (5) of an internal region of rv3790 gene from a Mycobacterium tuberculosis strain isolated from a TB patient, overlapping the 387 codon, and analysing the DNA sequence for the presence of the identified mutations as responsible for resistance to drugs interfering with arabinogalactan biosynthesis, such as benzothiazinones .

In the context of the present invention, the term "drug interfering with the arabinogalactan biosynthesis" means a molecule that prevents the transformation of decaprenyl-P-ribose to decaprenyl-P- arabinose by blocking or silencing a protein that performs the transformation of decaprenyl-P-ribose to

decaprenyl-P-arabinose and that can be encoded by rv ₃ 7 ₉₀ gene or homologues thereof or any open-reading artificial frame whose gene product is identical or homologue to Rv3790 protein. Another object of the present invention are inhibitors of Rv3790 protein obtained by the screening method of the invention, as well as a method of treatment of tuberculosis by using such inhibitors, wherein such inhibitors are preferably selected from benzothiazinones (or other molecules whose metabolism produce benzothiazinones) , monoclonal antibodies raised against Rv3790, antisense RNA sequences or vaccines.

Brief description of the drawings

Figure 1 shows a multi-alignment analysis with Rv3790 orthologs from several genera belonging to actinobacteria;

Figure 2 shows the DNA sequence of rv3790 gene.

Detailed description of the invention

A first object of the present invention is a method for screening drug candidates for the treatment of tuberculosis by interfering with the arabinogalactan biosynthesis, the said method comprising a step of putting into contact a Mycobacterium tuberculosis cell cultures (transformed with pSODIT-2/Rv3790) over- expressing a protein that performs the transformation of decaprenyl-P-ribose to decaprenyl-P-arabinose and

that can be encoded by rv3790 gene or homologues thereof or any open-reading artificial frame whose gene product is identical or homologue to Rv3790 protein, with a drug candidate and then evaluating the percentage of inhibition caused by the drug candidate against a control in an assay test.

Preferably, the said assay test is an in vitro test for antibacterial activity or an enzymatic test.

In the antibacterial test, the evaluation of the percentage of inhibition is preferably performed by determining the Minimum Inhibitory Concentrations

(MICs) . MICs are defined as the lowest concentration of an antimicrobial that will inhibit the visible growth of a microorganism after suitable incubation. Each drug candidate is also tested on a cell culture of M. tuberculosis transformed with pSODIT-2 vector alone.

As a control, isoniazid can be used, this being the reference drug in the treatment of tuberculosis. Alternatively or in addition, a different control experiment can be run, wherein no drug candidate treatment is made, so that a 0% inhibition is given. The % inhibition obtained with the drug candidate with respect to such a control is evaluated.

The said step of putting into contact a Mycobacterium tuberculosis cell culture with a drug candidate can be performed by putting the cell preparation onto one or more solutions of the drug candidate at different concentrations of drug, in a suitable medium.

The drug candidate is usually tested in a suitable form that allows good contact between the substance and the cell culture. Preferably, the drug candidate is utilized in solution or suspension. The drug candidate is used in amounts or concentrations that depend on several factors, among which the molecular weight of the substance, its solubility and so on.

Preferably, two or more concentrations of drugs are used, in order to calculate the MIC (Minimal Inhibitory Concentration) for each candidate.

Preferably, the suitable medium is Middlebrook

7H11 solid medium (Difco supplier) . More preferably, this medium is supplied with oleic acid, bovine serum albumin, dextrose and catalase (OADC) . This medium is typical for mycobacterial strains.

A further object of the present invention is the preparation of recombinant Rv3790 protein in the wild type and mutated forms from Escherichia coli cells for enzymatic activity assay (according to the method

described in reference 3) and a screening method for identifying novel drugs. The enzymatic screening test can be performed by using the recombinant Rv3790 enzyme in a reaction mixture containing FAD, NAD ⁺ , NADPH, and radiolabelled decaprenylphosphoryl ribose (DPR) that through the enzymatic reaction is transformed into Decaprenylphosphoryl-D-Araf (DPA) . This is the control experiment. Another experiment is run, wherein the procedure is repeated in the presence of the drug candidate. The amount of the so formed DPA is detected and the % inhibition of the enzymatic reaction caused by the drug candidate is calculated. The formation of DPA is monitored according to standard procedures well known to the skilled man, such as those described in reference (3) .

The inventive screening method may also comprise an in vivo evaluation of the activity of the drug candidates that pass the in vitro screening step. This in vivo evaluation can be performed as described below, in particular by administering preselected dosage regimens of the drug candidate to mice hematogeneously disseminated with a tuberculosis strain, against a control drug and against a placebo.

The control drug is preferably isoniazid. Preferably, BALB/c line mice are used.

Preferably, the tuberculosis strain is a virulent culture of Mycobacterium tuberculosis H37Rv, more preferably at a dose of 5 x 10 ⁶ CFU (Colony Forming Unit) in saline buffer. The screening dose of the drug candidate can vary in a wide range of doses depending on several factors, such as the MIC calculated from the above in vitro test results or the LD ₅₀ that can be determined for such compound. Generally, the dose of the drug candidate used in the present in vivo screening method will be below its LD ₅₀ .

Preferably, the drugs are introduced orally as suspension in a suitable buffer.

Preferably, the treatment is administered daily 6 times per week.

To determine the efficacy of each treatment regimen, macroscopical changes in parenchymal organs of the mice, growth of mycobacteria from pathologic material on solid media, as well as a bacterioscopical index of organ injury are recorded. A qualitative and quantitative analysis of macroscopical changes in the liver, spleen and lungs is also carried out and an injury index (using a four-score scale) is calculated.

A diagnostic method for drug-resistant mycobacterial strains is also provided as a further

object of the present invention. The diagnostic method involves amplification by Polymerase Chain Reaction (5) of _2TV 3790 gene from a Mycobacterium tuberculosis strain isolated from a TB patient and analysis of DNA sequence for the presence of the identified mutations.

Preferably, the method comprises amplification by Polymerase Chain Reaction (5) of an internal region of rv3790 gene from a Mycobacterium tuberculosis strain isolated from a TB patient, overlapping the 387 codon, and analysis of DNA sequence for the presence of the identified mutations (cysteine versus glycine or cysteine versus serine) as responsible for resistance to drugs interfering with arabinogalactan biosynthesis, such as benzothiazinones . Another object of the present invention are inhibitors of Rv3790 protein obtainable by the screening method of the invention, as well as a method of treatment of tuberculosis by using such inhibitors, wherein such inhibitors are preferably selected from benzothiazinones (or molecules whose metabolism produce benzothiazinones) , monoclonal antibodies raised against to Rv3790, antisense RNA sequences or vaccines.

Drugs interfering with the arabinogalactan biosynthesis The drugs obtainable through the screening method of the invention are preferably selected among

monoclonal antibodies against Rv3790 protein and mRNA antisense sequences for rv3790 gene, such drugs displaying a % inhibition greater than 50% in the in vitro screening test and/or a % inhibition greater than ₅₀ % in the anti-bacterial in vivo test of the invention.

In a preferred embodiment of the present invention, the said drugs are monoclonal antibodies against Rv3790 protein. Such antibodies can be obtained according to conventional procedures as well described in literature (6, 7).

In a further embodiment of the present invention, the said drugs are rv3790 gene antisense sequences.

The present invention provides antisense oligomers having a sequence effective to inhibit or block the expression of the rv3790 encoding gene or mRNA sequence. Antisense technology, which uses specific- oligonucleotides to inhibit expression of target gene products, is developing as a therapeutic modality for human disease. Several selection criteria are available to contribute to the optimization of antisense oligonucleotide antagonists. For example, it is advisable to choose sequences with 50% or more GC content. Preferred sequences span the AUG initiation codon of the target protein, but sites in the coding region and 5 ¹ UTR may perform equally well. Such

sequences are generally about 18-30 nucleotides long. Longer oligomers are often found to inhibit the target to a greater extent, indicating that a preferred length is about 25 mer for the first oligonucleotides chosen as antisense reagents. Typically, three oligonucleotide sequences are chosen with regard to these criteria, and compared for antagonist activity to control oligonucleotide sequences, such as "reverse" oligonucleotides or those in which about every fourth base of the antisense sequence is randomized. Therefore, a preferred sequence for making antisense oligomer sequences to rv3790 is a 25 mer sequence from the rv3790 mRNA sequence: AUG UUG AGC GUG GGA GCU ACC ACU A. The whole sequence of the rv3790 gene is deposited in http: //genolist .pasteur. fr/TubercuList/ under gene name rv3790. The entire rv3790 DNA sequence is reported in Fig. 2. Such preferred sense sequences are used to construct antisense oligonucleotide agents

(and suitable controls) for an in vitro comparison as inhibitors of Rv3790 protein. These in vitro data are predictive of human clinical utility using antisense agents of comparable design.

In a still further embodiment, vaccines constituted by or including the Rv3790 protein sequence can also be provided.

Method of treatment

The present invention further encompasses a method of treatment of a patient affected by tuberculosis, the method including administering to the said patient an antibacterial effective amount of a drug interfering with the arabinogalactan biosynthesis by blocking or silencing a protein that performs the transformation of decaprenyl-P-ribose to decaprenyl-P-arabinose.

Preferably, the said drug is a benzothiazinone molecule or monoclonal antibodies against Rv3790 protein or mRNA antisense sequences for rv3790 gene.

Preferably, the method of treatment further includes administering an antibacterial drug against Mycobacterium tuberculosis having a different mechanism of action, such as isoniazid.

The dose of the said drug interfering with the arabinogalactan biosynthesis that is administered can vary in a wide range of doses, depending on inhibitor's activity, the seriousness of the disease, the conditions of the patient, its weight and age as well as the route of administration, as will be determined by the clinician.

Generally, the drug can be used in doses ranging from 0.001 mg to 50 mg per kg of body weight, from 1 to 4 times a day.

The use of a drug interfering with the arabinogalactan biosynthesis as defined above, for the preparation of a medicament for treating tuberculosis is also claimed. EXPERIMENTAL PART

Determination of the in vitro minimum inhibitory- activity (MIC) of a drug candidate against mycobacteria

A single colony of M. tuberculosis strain is used to inoculate Middlebrook 7H9 enriched with 10% OADC and 0.05% Tween 80 (Difco) . This medium is typical for mycobacterial strains .

The culture is incubated at 37° C until reached exponential growth phase (OD ₆ oonm = 1 -> ~10 ⁸ CFU/ml) . Dilutions of such culture are prepared and used to plate ~10 ⁵ CFU.

MIC EVALUATIONS ON SOLID MEDIA

Middlebrook 7H11 (Difco) agar plates containing 10% OADC are prepared by adding increasing amounts of drugs. Preferably, several concentrations of drugs are used, in order to calculate the MIC for each candidate. Culture dilutions are plated onto media and plates are incubated at 37° C for 21 days. Growth inhibition is visually evaluated.

Determination of the in vivo inhibitory activity of a drug candidates against Mycobacterium tuberculosis in the murine TB model

To determine the chemotherapeutic efficacy, BALB/c line mice with experimental hematogenously disseminated tuberculosis can be used.

The mice are infected with a 2-week virulent culture of Mycobacterium tuberculosis H37Rv by intravenous injection (into tail vein) of the mycobacterial suspension at a dose of 5 x 10 ⁶ CFU

(Colony-Forming-Unit) in 0.5 ml saline. All the experimental animals are divided into groups depending on the treatment regimen used.

Treatment is started the next day after infection.

The drugs are introduced orally as suspension in carboxymethylcellulose/water with a small quantity PEG-

400. Chemotherapy is administered daily 6 times per week.

The animals are then sacrificed. To determine the efficacy of each treatment regimen, macroscopical changes in parenchymal organs of the mice, growth of mycobacteria from pathologic material on solid media, as well as a bacterioscopical index of organ injury are recorded. A qualitative and quantitative analysis of macroscopical changes in the liver, spleen and lungs is also carried out and an injury index (using a fourscore scale) is calculated.

Macroscopical evaluation of the efficacy of each treatment regimen is expressed in the efficacy index, calculated using the following formula:

Efficacy index = 100% - Injury index of the studied group/Injury index of the control group x 100

Microbiological examination includes culture for determination of CFU in parenchymal organs. For this purpose, the right lung and separately the spleen are homogenized with 6% sulfuric acid, centrifuged, washed by water and saline. The yield (about 0.5 mL) is diluted by 1.0 mL of saline and homogenised. This suspension (0.5 mL) of test organs is diluted 100 and 1000 times by saline and is distributed on solid Finn-2 medium. The cultures are incubated at 37 ⁰ C for 1 month and read weekly starting from day 10. After 28 days CFU' s are counted.

Preparation of recombinant Rv3790 In order to obtain protein expression into B. coli, several growth conditions were tested, with or without isopropyl-β-thiogalactoside (IPTG) as inductor, at concentrations ranging between 0.125 mM e 1 πiM. All cultures had been incubated at 37°C for 2 hours, at 27°C for 2 and 4 hours and at 20 ⁰ C for 2 and 20 hours. Cells were harvested by centrifugation and derived protein

extracts were analyzed by SDS-PAGE run. Recombinant proteins were purified according to supplier's instructions (Qiagen) .

_D etermination of Rv3790 as a target for benzothiazionones

In order to identify the molecular target for benzothiazinones, two different scientific/technical approaches have been followed: a) Transformation of Mycobacterium smegmatis cosmid library into λf. smegmatis and selection for resistance to benzothiazinones; b) Conventional mutagenesis: searching for spontaneous mutations by plating mycobacterial cultures onto media containing benzothiazinones. The minimal inhibitory concentration (MIC) of different benzothiazinones, for M. smegmatis mc ² 155, M. bovis BCG and M. tuberculosis H37Rv was determined as described above on Middlebrook 7H11, as solid medium, supplemented with oleic acid, bovine serum albumin, dextrose and catalase (OADC) .

Attention was focused on the most active benzothiazinone: 10526038. MIC determination was carried out also for isoniazid as control compound. MIC values of

10526038 and isoniazid are shown in Table 1.

Table 1. MICs of 10526038 and isoniazid for different mycobacterial species.

Approach a)

A genomic library of M. smegmatis mc ² 155 was constructed by using pYUBlδ cosmid vector and transformed into M. smegmatis mc ² 155; recombinant clones were selected by plating onto different concentrations of benzothiazinones (ranging from 16 ng/ml to 64 ng/ml) . Two cosmids, responsible for benzothiazinone resistance ( ₈ X- MI ^C) , were selected, partially sequenced and compared with the M. smegmatis mc ² 155 genome sequence (www.ti ^g er.or ^g) . These cosmids showed a 22 kb overlapping insert. By subcloning experiments of the common genomic region, a gene responsible for 10526038 resistance was identified. This 1383 bp gene encodes a hypothetical oxidoreductase ⁽ MSMEG_6382) which is highly homologous to R _V ³⁷⁹⁰ of M. tuberculosis

(http : //genolist .pasteur . fr/TubercuList/) . It has been demonstrated that Rv3790 and Rv3791 proteins work in concert and they are involved into arabinogalactan biosynthesis, a fundamental component of mycobacterial cell wall. They both are important to perform transformation of decaprenyl-P-ribose to decaprenyl-P- arabinose (3) .

It is noteworthy that both rv3790-rv3791 genes were described as essential for the life of M. tuberculosis by Himarl-based transposon mutagenesis in H37Rv strain (4) . The rv3790 gene was cloned into pSODIT-2 expression vector and transformed into the following strains: M. smegmatis, M. bovis BCG and M. tuberculosis. As shown in Table 2, the overexpression of Rv3790 confers high level of resistance to 10526038 in all three mycobacterial species.

Table 2. MICs to 10526038 of different mycobacterial species over-expressing Rv3790.

Approach b)

Spontaneous resistant mutants were isolated, respectively, in M. smegmatis, M. bovis BCG and M. tuberculosis.

Results obtained in M. smegmatis mc ² 155 In order to identify the cellular target for benzothiazinones, M. smegmatis mc ² 155 mutants resistant to benzothiazinone 10526038 were isolated by spontaneous conventional mutagenesis by plating ~10 ¹⁰ cells onto solid media containing different concentrations of the compound, ranging from 10 to 30 fold the MIC of the wild type strain. Resistant mutants were isolated at a frequency ranging from 10 ^'7 to 10 ^"9 . The results of this selection are shown in Table 3.

Table 3. M. smegmatis spontaneous mutants resistant to

10526038.

MSMEG_6382 gene and a 200 bp upstream region were sequenced in the 4 M. smegmatis mutants (MN47, MN48 and

MN81, MN84) with a different level of 10526038 resistance

(500/>4000X-MIC) . Such nucleotide region was compared to

wild type sequence (www.tiger.org) . Two point mutations were identified in the nucleotide sequences from M. smegmatis resistant strains. The two mutations are localized into the same codon, but on different nucleotides, according to different level of resistance, as shown in Table 4.

Table 4. Mutations identified in M. smegmatis resistant strains. Mutated nucleotide is underlined.

Results obtained in M. bovis BCG Pasteur In order to identify the cellular target for benzothiazinones, M. bovis BCG high resistant mutants were isolated by plating ~10 ¹⁰ cells onto solid media containing different concentrations of 10526038, ranging from 5 to 40 fold the MIC of the wild type strain. Resistant mutants were obtained at a frequency of 10 ^'8 . The result of this selection is shown in Table 5.

Table 5. Spontaneous M. bovis mutants resistant to

10526038.

Sequencing of the rv3790 homologous gene was performed in the eight M. bovis mutants. A point mutation was identified into the corresponding codon mutated in M. smegma.tis. In all 8 mutants, the second nucleotide was mutated, changing the amino acid from the wild type cysteine to a serine, as shown in Table 6.

Table 6. Mutations identified in M. bovis resistant strains. Mutated nucleotide is underlined.

Results obtained in M. tuberculosis H37Rv To find the target of benzothiazinones in M. tuberculosis (MtJb) , we have isolated some MtJb spontaneous resistant mutants. The MIC value of 10526038 in Mtb is very low, 1.5 ng/ml (versus 0.05 μg/ml of isoniazid) .

The isolation of resistant mutants was performed onto 7Hl1 medium (typical for mycobacteria) with the

addition of 10526038 drug at different concentrations. Four independent Mtb cultures have been used at O.D. ₆ oo=l. Aliquots (250 μl) of the four cultures were plated onto the following concentration of drugs: 1OX MIC, 4OX MIC, 7OX MIC, 150X MIC, and 300X MIC. We obtained Mtb spontaneous resistant mutants at all concentrations of drugs with the exception of 1OX MIC. In Table 7 are shown the isolated mutants, the frequency, and the MIC values.

Table 7. Mtb spontaneous mutants resistant to 10526038.

It is important to underline that the mutants have been isolated only onto high drug concentrations and that

the isolated mutants have two different levels of resistance. Particularly, NTB9 mutant is less resistant and grows very slowly.

We sequenced the rv3790 gene in all the mutants to find the mutation responsible for resistance. Two different missense point mutations mapping in the same codon of rv3790 gene were found (Table 8) .

Table 8. Mutations identified in Mtb Rv3790 resistant strains. Mutated nucleotide is underlined.

The above results point to Rv3790 protein as a target for benzothiazinone drugs. Since two common antibiotic resistance mechanisms include target amplification, or target modification, with the first method we isolated mycobacterial strains resistant to benzothiazinones via amplification mechanism, while with the second one we isolated mycobacterial strains resistant to benzothiazinones through mutations in a specific codon of rv3790 gene. This codon is present in the C-terminal portion of the

Rv ₃₇₉₀ protein; a Cystein in the wild type Rv3790 protein changes into a Glycine in M. tuberculosis and M. smegmatis (see NTB9, MN47 and MN48 mutants) or Serine in M. tuberculosis, M. bovis BCG and M. smegmatis (see NTBl- ₈ , _N TB _lO , BN1-8, MN81, and MN84 mutants) giving different levels of resistance. The highest level of resistance was observed with Serine substitution.

As all mutations are present in the same codon, the cystein could be the active site for the binding with the drug.

Therefore, mutants selected in vitro strongly suggest that the benzothiazinone drugs target the Rv3790 protein, thus affecting arabinogalactan biosynthesis.

Rv3790 orthologues

By BLAST program

(http : //www.ncbi .nlm. nih.gov/sutils/genom table . cgi?organ ism=microb) we searched for the presence of rv3790 orthologs in related microorganisms belonging to actinomycetes family. Mycobacterium avium, Mycobacterium aurum, Mycobacterium leprae, other mycobacterial species, Nocardia spp. , Rhodococcus spp. and Corynebacterium spp. As the rv3790 DNA sequence was not available for some microorganisms, such as Mycobacterium aurum, we cloned and sequenced in our laboratory the corresponding gene. The Rv3790 protein is well conserved in all

microorganisms tested as demonstrated by multi-alignment analysis with ClustalW program (http: //npsa- pbil . ibcp . fr/cgi- bin/npsa automat .pi?page=/NPSA/npsa clustalw.html) . It is noteworthy the conservation of the critical cystein through the tested microorganisms. Indeed, we found a high conservation of Cysteine in these microorganisms, with the exception of Mycobacterium avium, Mycobacterium avium paratuberculosis, and Mycobacterium aurum where Cys is substituited by alanine and serine, respectively (Fig. 1) .

To verify if the absence of cysteine confers resistance to 10526038 benzothiazinone, we performed the MIC evaluations for the following microrganisms : Mycobacterium avium, Mycobacterium aurum, Nocardia spp. , Rhodococcus spp. and CoryneJbacterium spp.

For mycobacteria species the experiment was carried out onto 7H11 medium with different drug concentrations. The MICs were determined from a culture dilution at 0.5 Macfarland.

For the other actinomycetes Nocardia spp., Rhodococcus spp. and CoryneJbacterium sp, the dilutions at 0.5 Macfarland were streacked onto blood-Mueller Hinton agar with the addition of different drug concentrations. A ^s shown in Table 9, 10526038 benzothiazinone is more active against the actinomycetes species. Only M.

avium and M. aurum appear naturally resistant to benzothiazinones; thus confirming our hypothesis that Cys plays a key role in 10526038 benzothiazinone sensitivity.

Table 9. MIC evaluations of 10526038 benzothiazinone in actinomycetes species.

Heterologous expression of Rv3790

As Rv3790 protein work in concert with Rv3791 (3) the following recombinant clones have been constructed: 1) Rv3790 WT enzyme has been produced into Escherichia coli BL21(DE3)pLysS by using pET32 vector.

2) Rv3 ₇ 91 WT enzyme has been produced into Escherichia coli BL21(DE3)pLysS by using pET15 vector.

₃ ) Rv3790 mutated enzyme (Cys->Gly: from M. tuberculosis resistant mutant) has been produced into Escherichia coli BL21 (DE3)pLysS by using pET32 vector.

4) Rv3790 mutated enzyme (Cys->Ser: from M. tuberculosis resistant mutant) has been produced into Escherichia coli BL21 (DE3)pLysS by using pET32 vector. rv3790 and rv3791 genes from M. tuberculosis wild type and resistant strains (NTBl and NTB9) were amplified from M. tuberculosis genome by polymerase chain reaction (PCR) and cloned into the different expression vectors. The recombinant constructs were transformed into E. coli BL21DE3/pLysS cells, by electroporation. In order to obtain protein expression into E. coli, several growth conditions were tested, with or without isopropyl-β- thiogalactoside (IPTG) as inductor, at concentrations ranging between 0.125 mM e 1 mM. All cultures had been incubated at 37 ⁰ C for 2 hours, at 27 ⁰ C for 2 and 4 hours and at 20 ⁰ C for 2 and 20 hours. Cells were harvested by centrifugation and derived protein extracts were analyzed by SDS-PAGE run.

References

1. _S hah NS, Wright A ₁ Bai GH, et al. Worldwide emergence of extensively drug-resistant tuberculosis . Emerg Infect Dis 2007, 13:380-387. 2. Williams KJ ₇ Duncan K. Current strategies for identifying and validating targets for new treatment-shortening drugs for TB. Curr MoI Med 2007, 7:297-307

3. Mikusova K, et al. Decaprenylphosphoryl arabinofuranose, the donor of the D-arabinofuranosyl residues of mycobacterial arabinan, is formed via a two-step epimerization of decaprenylphosphoryl ribose. J Bacteriol 2005, 187:8020-8025.

4. Sassetti CM, et al. Genetic requirements for mycobacterial survival during infection.

Proc Natl Acad Sci USA 2003 100:12989-12994.

5. Mullis K, et al. Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harb Symp Quant Biol 1986, 51:263-73. 6. Andersen P, et al. Recall of long-lived immunityto Mycobacterium tuberculosis infection in mice. J Immunol 1995, 154:3359-3372.

7. Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 1975, 256:495-497.