S

Si

MJa

b

c

d

e

a

ARA

KSMXME

1

(TLownTisrtX

M1

0d

International Journal of Antimicrobial Agents 33 (2009) 479–482

Contents lists available at ScienceDirect

International Journal of Antimicrobial Agents

journa l homepage: ht tp : / /www.e lsev ier .com/ locate / i jant imicag

hort communication

ILA 421, an inhibitor of efflux pumps of cancer cells, enhances the killing ofntracellular extensively drug-resistant tuberculosis (XDR-TB)

arta Martins a,b, Miguel Viveirosa, Jorge Ramosa, Isabel Coutoa,c,oseph Molnard, Martin Boereee, Leonard Amarala,b,∗

Unit of Mycobacteriology, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa (IHMT/UNL), Rua da Junqueira 96, 1349-008 Lisboa, PortugalUPMM, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa (IHMT/UNL), Rua da Junqueira 96, 1349-008 Lisboa, PortugalCentro de Recursos Microbiológicos (CREM), Faculdade de Ciências e Tecnologia, UNL, 2829-516 Caparica, PortugalDepartment of Medical Microbiology and Immunology, Institute of Medical Microbiology, University of Szeged, H-6720, Szeged, Dom ter 10, Szeged, HungaryUniversity Lung Centre Dekkerswald, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

r t i c l e i n f o

rticle history:eceived 31 July 2008ccepted 27 October 2008

eywords:ILA compoundsDR-TBDR-TB

a b s t r a c t

Multidrug-resistant and extensively drug-resistant tuberculosis (MDR/XDR-TB) are problematic to man-age, especially in patients with acquired immune deficiency syndrome (AIDS). There is therefore a direneed for effective anti-MDR/XDR-TB agents. We have previously shown that agents that inhibit the effluxpumps of MDR bacteria and cancer cells also enhance killing of intracellular mycobacteria, possibly byincreasing the availability of K+ and Ca2+ needed for the activation of lysosomal enzymes of the phagolyso-somal unit. In this study, the newly synthesised and recently patented SILA 409 and 421 were testedfor in vitro and ex vivo activity against XDR-TB. The minimum inhibitory concentration (MIC) of SILA

acrophagesfflux pumps

compounds was determined by the BACTEC 460 TB system. The effect of each compound on the killingactivity of human macrophages infected with XDR-TB was determined by exposing the macrophage thathad phagocytosed the bacterium to the compounds and assessing the killing activity by colony-formingunit counting. Amongst the two compounds tested, SILA 421 was shown to have in vitro activity againstXDR-TB (MIC < 3.5 mg/L) and to transform non-killing macrophages into effective killers of phagocytosedbacteria, without any cytotoxic activity. Because SILA 421 revealed good in vitro and ex vivo activities and

activlsevi

is devoid of any cytotoxic© 2008 E

. Introduction

The rates of new cases of multidrug-resistant tuberculosisMDR-TB) and now extensively drug-resistant tuberculosis (XDR-B) continue to increase and this is now a major health concern [1].isbon and Oporto, the two major cities of Portugal, are examplesf where the rates of MDR-TB are exceedingly high [2], althoughith respect to Lisbon there has been a significant reduction inew cases of MDR-TB owing to the implementation of the Faster TBrack Programme and the ability of mycobacteriology laboratoriesn this area to identify within 1 day of receiving a positive acid fast-

tained sputum specimen a patient infected with MDR-TB [3]. Withespect to other cities that have not implemented this programme,he rates of new cases of MDR-TB continue to increase [2]. RegardingDR-TB, retrospective evaluation of MDR-TB strains isolated from

∗ Corresponding author at: Unit of Mycobacteriology, Instituto de Higiene eedicina Tropical, Universidade Nova de Lisboa (IHMT/UNL), Rua da Junqueira 96,

349-008 Lisboa, Portugal. Tel.: +351 21 365 2600; fax: +351 21 363 2105.E-mail address: lamaral@ihmt.unl.pt (L. Amaral).

924-8579/$ – see front matter © 2008 Elsevier B.V. and the International Society of Chemoi:10.1016/j.ijantimicag.2008.10.028

ity, it is a potential candidate as an anti-MDR/XDR-TB drug.er B.V. and the International Society of Chemotherapy. All rights reserved.

patients during the past 8 years clearly shows that a significantnumber are XDR-TB [2]. Undoubtedly, the laboratory plays a majorrole in the control of TB [1,3,4]. However, until TB is completelyeliminated, effective therapy is an absolute requirement for con-trol of the disease. Although antibiotic-susceptible TB infections arenot difficult to manage effectively under conditions that optimisetherapy, inexperience of the physician or non-compliance of thepatient contributes to the selection of antibiotic-resistant strains ofMycobacterium tuberculosis [1,4]. When the patient is infected withMDR-TB, therapy is problematic and cure rates are low [1]. To treatthese MDR/XDR-TB infections, new and effective drugs are neededand, despite the fact that several are at phase II clinical trials, noneare available at this time. We have previously demonstrated thatphenothiazines used for the therapy of psychoses as well as theirderivatives (i) have in vitro activity against a large range of bacteria,in particular M. tuberculosis [5–7], (ii) enhance the killing activity

of infected macrophages [5] and (iii) cure mice infected with M.tuberculosis [8]. However, because thioridazine (TZ) as well as otherneuroleptic phenothiazines cause arrhythmia and even ‘torsade depointes’ (an uncommon variant of ventricular tachycardia), albeit atvery low frequencies, there is resistance to employ these agents for

otherapy. All rights reserved.

4 l of An

ttamcrtaTt

2

2

1apsra((aD(cwSp

2

saEtcCTTR

Fb4paP

80 M. Martins et al. / International Journa

he management of MDR-TB infections. We have previously shownhat agents that enhance the killing of intracellular mycobacteriare potent inhibitors of efflux pumps responsible for phenotypicultidrug resistance of bacteria [5,7]. Because organosilicon (SILA)

ompounds have recently been demonstrated to inhibit multidrugesistance efflux pumps of cancer cells [9] and to reverse the mul-idrug resistance of Escherichia coli, we have investigated the in vitrond ex vivo activities of SILA compounds 409 and 421 against XDR-B as well as their in vitro cytotoxicity against human cells in ordero evaluate their potential as anti-MDR/XDR-TB drugs.

. Materials and methods

.1. Materials

Patented compounds 1,3-dimethyl-1,3-bis(4-fluorophenyl)-,3-bis(3-morpholino-propyl) disiloxan-dihydrochlorid (SILA 409)nd 1,3-dimethyl-1,3-bis(4-fluorophenyl)-1,3-bis{3-[1(4-buthyl-iperazinyl)]-propyl}-disiloxan-tetrahydrochlorid (SILA 421) wereynthesised by Hegyes et al. [10]. SILA compounds 409 and 421 haveeceived patents (Brevet European n◦0099150.6, PCT/DE00/04110)s compounds that reverse multidrug resistance in cancer cellssee Fig. 1 for compound structures). BACTEC 460 vials, isoniazidINH), rifampicin (RIF), streptomycin (STR), ethambutol (EMB)nd pyrazinamide (PZA) were purchased from Becton-Dickinsoniagnostic Instrument Systems (Towson, MD). TZ, amikacin

AMK), kanamycin (KAN), capreomycin (CM), ofloxacin (OFX), d-ycloserine (CS), para-aminosalicylic acid (PAS) and ethionamideere purchased in powder form from Sigma-Aldrich (Madrid,

pain). Stock solutions of TZ and SILA compounds were freshlyrepared in water, protected from light and stored at −20 ◦C.

.2. Bacteria

Mycobacterium tuberculosis strains employed in this study weretrain H37Rv (ATCC 27294), which is susceptible to RIF, INH, STRnd EMB, and a clinical isolate resistant to INH, RIF (MDR-TB), STR,MB and PZA. This latter strain was isolated in February 2005 fromhe bronchoalveolar lavage of a patient, identified as M. tuberculosis

®

omplex by the Accuprobe system (Gen-Probe Inc., San Diego,A) and characterised by the MGIT 960 Antibiotic Susceptibilityesting (AST) SIRE and PZA methods, as previously described [3].his MDR-TB strain was further characterised by the Nationaleference Laboratory for Tuberculosis (INSA Dr Ricardo Jorge,

ig. 1. Structure of SILA 409 and 421. SILA compounds 1,3-dimethyl-1,3-is(4-fluorophenyl)-1,3-bis(3-morpholino-propyl) disiloxan-dihydrochlorid (SILA09) and 1,3-dimethyl-1,3-bis(4-fluorophenyl)-1,3-bis{3-[1(4-buthyl-piperazinyl)]-ropyl}-disiloxan-tetrahydrochlorid (SILA 421) were synthesised by Hegyes etl. [10]. These compounds have received patents (Brevet European n◦0099150.6,CT/DE00/04110).

timicrobial Agents 33 (2009) 479–482

Oporto, Portugal) as resistant to AMK (1.0 mg/L), KAN (5.0 mg/L),CM (1.25 mg/L), OFX (2.0 mg/L), CS (100 mg/L) and PAS (4.0 mg/L),following the World Health Organization recommendations andmethodologies for second-line drug susceptibility testing [1].This second-line drug resistance pattern was also confirmed inour laboratory by the BACTEC 460 TB system [11,12]. This straintherefore satisfies the criteria for classification as an XDR-TB strain[1,4] and is one of the most resistant strains isolated in Portugal(resistant to the five drugs that constitute the ‘first-line of defence’as well as to six of the major ‘second-line of defence’ agents).

2.3. Determination of the in vitro activity of TZ and SILA 409 and421 against an XDR-TB clinical isolate

The minimum inhibitory concentration (MIC) of TZ and SILA409 and 421 was determined individually by the BACTEC 460 TBsystem [11,12]. TZ was used as a positive control for all the exper-iments described. Mycobacterium tuberculosis strains were grownas a primary culture in Middlebrook 7H9 broth in 12B BACTEC 460vials (Becton-Dickinson Diagnostic Instrument Systems) until theyreached an optical density (OD) equal to 1 McFarland standard.From these cultures, aliquots were transferred to new 12B BACTECvials and the OD was adjusted to a 0.5 McFarland scale by visualcomparison. Each vial received 0.1 mL of 12B BACTEC 460 mediumcontaining an adjusted concentration of mycobacteria correspond-ing to ca. 105–106 colony-forming units (CFU). These vials servedas the source for the inoculation of triplicate sets of BACTEC 12Bvials containing varying concentrations of TZ and SILA 421 and 409.Briefly, stock solutions containing varying amounts of each com-pound were separately prepared in 12B BACTEC 460 medium and0.1 mL aliquots of each compound were added to 12B BACTEC 460vials. The final concentration of the compounds in the vials was 0,1.562, 3.125, 6.250, 12.5, 25 and 50 mg/L. A separate 1:100 dilutionof the adjusted inoculum was made and transferred to a BACTEC 12Bvial that contained no drug; this served as the proportional control,as defined by the recommended BACTEC proportional method forM. tuberculosis. Inoculation and growth of each strain into controland experimental BACTEC vials was in accordance with the BACTEC460 TB proportional method procedure [11,12]. MIC determinationswere repeated three times and the values obtained did not dif-fer significantly. These in vitro determinations were also repeatedusing the BACTEC MGIT 960 TB eXiST System [13].

2.4. Determination of the cytotoxicity of TZ and SILA 409 and 421against human lymphocytes

Human peripheral blood (20 mL) was collected with a hep-arinised syringe (300 �L Monoparin, 1000 U; CP PharmaceuticalsLtd., Wrexham, UK) from healthy volunteer donors and purifiedhuman mononuclear cells were obtained by differential cen-trifugation as described previously [5]. For the determination ofcytotoxicity of TZ and SILA compounds, the trypan blue exclusionmethod was conducted as described previously [5]. Concentrationsof TZ and SILA compounds employed for the determination of cyto-toxicity ranged from 0.1 mg/L to 100 mg/L.

2.5. Determination of the ex vivo activity of TZ and SILA 409 and421 against an XDR-TB clinical isolate

Human peripheral blood (20 mL) was collected from healthyvolunteer donors and transferred into tubes containing 50 U of

preservative-free sodium heparin (Monoparin; CP PharmaceuticalsLtd.). Purified mononuclear cells were obtained as described pre-viously [5]. Cells were counted in a Neubauer chamber and theconcentration was adjusted to 1 × 106 cells/mL. From this suspen-sion, aliquots of 1 mL were distributed to the wells of a 24-well

l of Antimicrobial Agents 33 (2009) 479–482 481

ppttwSaTtadolintltftwbetcw7owtd

3

3

mataTse3t

3

hfecot

3

Xcacn

Fig. 2. Ex vivo effect of SILA 421 and 409 in human macrophages infected withextensively drug-resistant tuberculosis (XDR-TB). The average killing activity of

T

M. Martins et al. / International Journa

late (Nalgene Nunc, Thermo Scientific, Rochester, NY) and thelates were incubated at 37 ◦C with 5% CO2 for 3 days. After this,hree washes were performed with RPMI medium to remove allhe non-adhered cells. New medium was added and the platesere incubated at 37 ◦C with 5% CO2 for an additional 2 days.

upernatants were removed, cells were washed with RPMI mediumnd new fresh medium was distributed to the wells of the plate.he contents of one set of triplicate wells were removed by gen-le scraping of the bottom of the wells with a rubber policemannd aliquots of 0.01 mL were used for cell counting. The monocyte-erived macrophages were infected with M. tuberculosis at a ratiof 10:1 (mycobacteria:macrophage), which was previously estab-ished as the optimum ratio for phagocytosis [5]. Cell cultures werencubated at 37 ◦C with 5% CO2 for 1 h. After this period, the super-atants were gently aspirated and washed three times with RPMIo remove all non-phagocytosed bacteria. The supernatants of theast wash were processed for CFU counting to determine whetherhere were any free non-phagocytosed bacteria. One millilitre ofresh medium containing TZ and SILA compounds at a concentra-ion of 0.1 mg/L was added to the macrophage cultures. Controlells received 1.0 mL of drug-free medium. The plates were incu-ated at 37 ◦C for intervals of 0, 1, 2 and 3 days and at the end ofach interval 1.0 mL of 0.01% sodium dodecyl sulphate was addedo each well so that the cells would lyse and release the phago-ytosed bacteria. Then, 0.1 mL aliquots of the lysed supernatantsere serially diluted to 10−4 with phosphate-buffered saline (pH

.4) and an aliquot of 0.1 mL of each serial dilution was platednto 7H11 Middlebrook Mycobacteria agar plates. The agar platesere incubated at 37 ◦C for periods of 3–4 weeks, after which

he CFU were counted and the initial bacterial concentration wasetermined.

. Results

.1. In vitro activity

The in vitro activities of TZ and SILA 409 and 421 were deter-ined against the reference H37Rv and the XDR-TB strains with the

id of the BACTEC 460 TB system, strictly following the manufac-urer’s instructions [11]. Briefly, TZ had a MIC of 15 mg/L and 20 mg/Lgainst the antibiotic-susceptible and XDR-TB strains, respectively.he activity of the SILA compounds against either the antibiotic-usceptible or the XDR-TB strains was identical, with SILA 409xhibiting a MIC of 12.5 mg/L and SILA 421 a much lower MIC of.125 mg/L. The MIC determinations were repeated three times andhe values obtained did not differ significantly.

.2. Cytotoxicity assays

The SILA compounds were evaluated for cytotoxicity againstuman lymphocytes by the trypan blue exclusion method and were

ound not to increase the retention of this dye at concentrations ofach compound as high as 100 mg/L compared with the drug-freeontrols (data not shown). These results are compatible with thosef others who demonstrated that these compounds do not exhibitoxicity in the mouse [14].

.3. Ex vivo activity

The ex vivo activity of SILA 409 and 421 against phagocytosedDR-TB was evaluated at 0.1 mg/L, which is well below the highestoncentration examined for toxicity (100 mg/L). As shown in Fig. 2,t this concentration SILA 421 is able to enhance the killing of intra-ellular XDR-TB by the human macrophage, whereas SILA 409 doesot show any activity.

human monocyte-derived macrophages against XDR-TB when SILA 421 and 409were present in the medium at a concentration of 0.1 mg/L was assessed. Results arethe average number of colony-forming units (CFU) for cultures obtained from threeindependent experiments.

4. Discussion

Whereas therapy of antibiotic-susceptible pulmonary TB infec-tions is readily managed with INH and RIF, infections producedby MDR-TB and XDR-TB strains are highly problematic for therapydespite the use of increasing numbers of agents that make up the‘second-line of defence’ drugs [1]. At the time of writing, no newdrugs are on the horizon and only two clinical trials of new agentsare active with little success in evidence [15]. There is an obviousdire need for effective anti-MDR/XDR-TB compounds and this iswhere our studies have been directed during the past years [5–8].In particular, we have recommended that until effective anti-MDR-B agents are made available, alternative drugs such as TZ and their

derivatives should be considered for the therapy of MDR-TB (fora review, see [7]). The results shown in the present study suggestthat SILA 421 also has the potential to be used as a helper com-pound in the treatment of MDR/XDR-TB infections. In particular,this compound demonstrated significant in vitro activity (with aMIC < 3.5 mg/L) against an XDR-TB strain that was resistant to firstline of defence and six of seven of the most frequently used drugsthat constitute the second line of defence. Although this concen-tration is free of any cytotoxicity for human cells, as shown in thisstudy, what is of far greater significance is that this compound iscapable of enhancing the killing of intracellular XDR-TB by humanmacrophages at concentrations as low as 0.1 mg/L. It is importantto note that whereas hundreds of compounds have in vitro activ-ity against strains of antibiotic-resistant M. tuberculosis, few haveactivity where the macrophage resides in the human lung, namelythe alveolar pulmonary unit. Therefore, because the SILA 421 com-pound is so effective ex vivo against both antibiotic-susceptible andXDR-TB strains, one would expect that a similar concentration ofthe agent in the interstitial fluids of the XDR-TB-infected patientwould also be effective and enhance the killing of XDR-TB at the sitewhere this organism resides, namely the pulmonary macrophage.Hence, barring any unforeseen negative aspects, SILA 421 is a verypromising anti-MDR/XDR-TB agent.

Acknowledgment

The authors thank Dr Andras Varga of the Institute of Molecular

Parasitology, Humboldt University, Berlin, Germany, for donation ofthe SILA compounds.

Funding: This work was supported by EU-FSE/FEDER-POCI/SAU-MMO/59370/2004 and EU-FSE/FEDER-PTDC/BIA-MIC/71280/2006

4 l of An

pPP

R

[

[

[

[

82 M. Martins et al. / International Journa

rovided by the Fundacão para a Ciência e a Tecnologia (FCT) ofortugal. MM was supported by grant SFRH/BD/14319/2003 (FCT,ortugal).

Competing interests: None declared.Ethical approval: Not required.

eferences

[1] World Health Organization. The WHO/IUATLD Global Project on Anti-tuberculosis Drug Resistance Surveillance, 2002-2007. Anti-tuberculosisdrug resistance in the world: report No. 4. Geneva, Switzerland: WHO;2008.

[2] Direccão Geral de Saúde. Programa Nacional de Luta contra a Tuberculose. Pontoda situacão epidemiológica e de desempenho. Programa Nacional de Controloda Tuberculose. Sistema de Vigilância (SVIG-TB). Lisbon, Portugal: DGS; 2006.

[3] Viveiros M, Leandro C, Rodrigues L, Almeida J, Bettencourt R, Couto I, et al.Direct application of the INNO-LiPA Rif.TB line-probe assay for rapid iden-tification of Mycobacterium tuberculosis complex strains and detection ofrifampin resistance in 360 smear-positive respiratory specimens from an areaof high incidence of multidrug-resistant tuberculosis. J Clin Microbiol 2005;43:4880–4.

[4] Raviglione MC, Uplekar MW. WHO’s new Stop TB Strategy. Lancet 2006;367:952–5.

[5] Ordway D, Viveiros M, Leandro C, Bettencourt R, Almeida J, Martins M, etal. Clinical concentrations of thioridazine kill intracellular multidrug-resistantMycobacterium tuberculosis. Antimicrob Agents Chemother 2003;47:917–22.

[

[

timicrobial Agents 33 (2009) 479–482

[6] Martins M, Schelz Z, Martins A, Molnar J, Hajös G, Riedl Z, et al. In vitro and exvivo activity of thioridazine derivatives against Mycobacterium tuberculosis. IntJ Antimicrob Agents 2007;29:338–40.

[7] Amaral L, Martins M, Viveiros M. Phenothiazines as anti-multi-drug resistanttubercular agents. Infect Disord Drug Targets 2007;7:257–65.

[8] Martins M, Viveiros M, Kristiansen JE, Molnar J, Amaral L. The curative activ-ity of thioridazine on mice infected with Mycobacterium tuberculosis. In Vivo2007;21:771–6.

[9] Fusi F, Ferrara A, Zalatnai A, Molnar J, Sgaragli G, Saponara S. Vascular activityof two silicon compounds, ALIS 409 and ALIS 421, novel multidrug-resistancereverting agents in cancer cells. Cancer Chemother Pharmacol 2008;61:443–51.

10] Hegyes P, Molnar J, Mucsi I, Hever A, Szabo D, Kiesig S. Substituted disiloxanes,method for the production thereof and the use thereof for reversal of multidrugresistance (MDR). PCT/DE00/04110; 2000.

11] Siddiqi SH. BACTEC 460 TB SIRE method. Sparks, MD: Becton Dickinson Micro-biology Systems; 1985.

12] Pfyffer GE, Bonato DA, Ebrahimzadeh A, Gross W, Hotaling J, Kornblum J, etal. Multicenter laboratory validation of susceptibility testing of Mycobacteriumtuberculosis against classical second-line and newer antimicrobial drugs byusing the radiometric BACTEC 460 technique and the proportion method withsolid media. J Clin Microbiol 1999;37:3179–86.

13] De Bock A, Moulard P. MGIT 960 eXtended individual Susceptibility Testing ForEpiCenter V5.53 A. TB eXiST for EpiCenter V5. 53A—User’s Manual V1.03. Becton

Dickinson; 2008.

14] Zalatnai A, Molnár J. Effect of SILA-409, a new organosilicon multidrug resis-tance modifier, on human pancreatic cancer xenografts. In Vivo 2006;20:137–40.

15] Zhang Y. Advances in the treatment of tuberculosis. Clin Pharmacol Ther2007;82:595–600.