In Vitro Activities of Fourteen Antimicrobial Agents ... - Springer Link

CURRENT MICROBIOLOGY Vol. 33 (1996), pp. 167–175

An International Journal

R Springer-Verlag New York Inc. 1996

In Vitro Activities of Fourteen Antimicrobial Agents Against Drug Susceptible and Resistant Clinical Isolates of Mycobacterium tuberculosis and Comparative Intracellular Activities Against the Virulent H37Rv Strain in Human Macrophages Nalin Rastogi, Vale´rie Labrousse, Khye Seng Goh Unite´ de la Tuberculose et des Mycobacte´ries, Institut Pasteur, Morne Jolivie`re, B.P. 484, 97165-Pointe a` Pitre Cedex, Guadeloupe, French West Indies Received: 23 January 1996 / Accepted: 5 April 1996

Abstract. Minimal inhibitory concentrations (MICs) of 14 first and second-line antituberculous drugs against drug-susceptible and drug-resistant clinical isolates of Mycobacterium tuberculosis (including the multiple drug-resistant or MDR-TB isolates), as well as the type strain H37Rv, were determined radiometrically by the Bactec 460-TB methodols. MICs (µg/ml) of all the fourteen drugs were within an extremely narrow range in case of susceptible strains; isoniazid (0.02–0.04), rifampin (0.2–0.4), ethambutol and streptomycin (0.5–2.0), ethionamide (0.25–0.5), D-cycloserine (25–75), capreomycin (1–2), kanamycin (2–4), amikacin (0.5–1.0), clofazimine (0.1–0.4), ofloxacin (0.5–1.0), ciprofloxacin (0.25–1.0), and sparfloxacin (0.1–0.4). The activity of second-line drugs remained unaltered against MDR-TB isolates resistant to routine first-line drugs. With peak serum level concentrations (Cmax), the intracellular killing of the virulent H37Rv strain was studied in detail in cultured human macrophages. Based on an decreasing order of bactericidal activity, our results showed the following spectrum of intracellular drug action: among the first-line drugs, rifampin . ethionamide 5 isoniazid . ethambutol . streptomycin . D-cycloserine; among second-line drugs, clofazimine 5 amikacin . kanamycin 5 capreomycin; among fluoroquinolones, sparfloxacin . ofloxacin . ciprofloxacin. On the other hand, contrary to atypical mycobacteria, the macrolide drug clarithromycin was inactive against both extracellular and intracellular M. tuberculosis.

A renewed interest in evaluating the activity of a variety of second-line antituberculous drugs, new analogs of existing drugs, and newer drug combinations against Mycobacterium tuberculosis has resulted today because of the recent upsurge in the incidence of tuberculosis (TB) with significant emergence of multidrug-resistant (MDR) cases [1, 18, 23]. A laboratory survey conducted on M. tuberculosis isolates from patients in New York City [4] showed a high proportion of MDR TB isolates: a total of 19%, including 7% resistance from patients without any prior treatment and 30% resistance in patients with prior

Correspondence to: N. Rastogi

treatment. A retrospective nationwide survey showed that 14.2% of all TB isolates in the United States in the year 1991 were resistant to at least one antituberculous drug with nearly 10% being resistant to isoniazid and/or rifampin [1]. On an annualized basis, at least half of the 600 MDR TB isolates were resistant to five first-line drugs (isoniazid, rifampin, ethambutol, streptomycin, and pyrazinamide) with about 90% of patients being coinfected with the human immunodeficiency virus; case fatality rates in this group were as high as 70–90% [1]. The treatment of MDR TB cases remains extremely difficult and requires meticulous laboratory studies to characterize the susceptibility of these isolates to other drugs [7]. However, contrary to the

168 first-line drugs, which can be easily screened in vitro based on previously established ‘‘critical concentrations,’’ such data are unavailable for most of the second-line and/or newer antituberculous drugs. Consequently, there is an urgent need to establish indirect methods, capable of estimating their potential efficacy in vitro [7].

We have been engaged for many years in research aiming to correlate in vitro MICs and MBCs of antituberculous drugs to their intracellular activity in experimentally infected macrophages against both M. tuberculosis and M. avium [14, 15, 17]. We have also been involved in the evaluation of both radiometric and proportion methods in order to establish appropriate methods and critical concentrations of antituberculous drugs [13, 16, 20]. Considering specific parameters for effective therapy of intracellular pathogens like M. tuberculosis [11], in the present investigation we have determined the activity of 14 antituberculous drugs against drug-susceptible and drug-resistant clinical isolates of M. tuberculosis radiometrically, and compared it with intracellular bactericidal activity in cultured human macrophages. Materials and Methods Organisms. Ten strains of M. tuberculosis (the type strain H37Rv and nine clinical isolates) were selected for this study. The strains were kept frozen at 240°C as small aliquots and cultured in the Lo¨wenstein-Jensen medium prior to experiments. Radiometric studies. The MICs were determined radiometrically with the Bactec 460-TB apparatus essentially as reported previously for M. tuberculosis [8, 13, 16, 21, 24, 25]. Briefly, the bacteria were scraped from fresh Lo¨wenstein-Jensen slants, resuspended in 3 ml of diluting fluid, and homogenized with glass beads (2 mm diameter). Letting the suspension stand for a few minutes permitted the bacterial clumps to sediment. The homogeneous supernatant was taken and the turbidity was adjusted to McFarland 1 with diluting fluid. A Bactec 12B vial (Becton-Dickinson, Towson, Maryland) was injected with 0.1 ml of this suspension. This vial was used as a primary inoculum after the growth index (GI) reached 500 as follows: 0.1 ml of bacterial suspension from the preculture vial was injected into drug-containing vials as well as a control vial. A second control vial (the 1:100 control) containing a 100-fold diluted initial bacterial inoculum was also prepared. GI were followed once daily. When the GI of the 1:100 control vial reached 30, the test was read at least one additional day before it was terminated. The results were interpreted as follows: if the difference in the GI values from the previous day (called DGI) in case of drug containing vials was less than the DGI of the 1:100 control, then the bacteria were considered susceptible to the drug concentration tested. The MICs of clarithromycin were determined at two different pHs, i.e., at 6.8 and at 7.4, as reported previously for another macrolide drug roxithromycin against M. tuberculosis [21]. Experiments in a non-weekend schedule were started on a Friday, and the GI readings were performed from Monday onwards, i.e., at day 3, 4, 5, 6, and 7. The bactericidal effect of various drugs in the Bactec system

CURRENT MICROBIOLOGY Vol. 33 (1996)

was also compared between the drug-susceptible and drug-resistant clinical isolates. For these studies, bacterial viability was determined by plating the bacterial suspensions from individual Bactec vials at the beginning and at the end of the experiments on 7H11 agar medium for viable count enumeration, and the results were expressed as mean viable count 6 standard error [13]. Macrophage experiments. Intracellular activity of various firstand second-line drugs against the M. tuberculosis H37Rv strain was evaluated in cultured human monocytes/macrophages essentially as reported previously for mycobacteria [17, 21, 22]. Briefly, bacteria were grown in complete 7H9 broth (supplemented with Middlebrook ADC enrichment, Difco Laboratories, Detroit, Michigan), containing 0.05% (vol/vol) Tween 80 to avoid clumping at 37°C and were harvested in their mid-logarithmic phase at an optical density of 0.15 (measured at 650 nm with a Coleman Junior II spectrophotometer). Cultures of human macrophages (about 105 macrophages/ well) were prepared from adherent peripheral blood monocytes from healthy donors [17, 22] and infected with tubercle bacilli at a ratio of about 10 to 20 bacilli per cell, as reported previously [12, 15]. Any extracellular bacilli were thoroughly washed away with Hanks balanced salt solution after 4 h of phagocytosis at 37°C, and the number of intracellular organisms was determined by lysing the macrophages with 0.25% (wt/vol) sodium dodecyl sulfate (SDS), doing immediate serial dilutions, and plating the lysates on 7H11 agar medium for viable count determinations [14, 15, 17]. After phagocytosis, fresh medium containing desired antimicrobial agents was refed to macrophage-containing wells, and the bacteria were enumerated after lysing the macrophages 3 and 7 days after the drug addition [12, 15]. The results (expressed as mean viable counts 6 standard error) were compared with the growth of bacteria in control, untreated macrophages. A drug was considered bactericidal if it effectively reduced the bacterial viable counts in the test samples compared with the initial inoculum added at the time of drug addition (day 0) by 90% or more. In accordance with our experimental model for determining intracellular action of drugs [12, 14, 15, 17, 22], all the drugs were used at concentrations within their reported peak serum concentration (Cmax) in humans [3, 9, 10, 26]; D-cycloserine, 50 µg/ml; streptomycin, 30 µg/ml; ethambutol, 6 mg/ml; isoniazid, 4 µg/ml; ethionamide, 20 µg/ml; rifampin, 15 mg/ml; clarithromycin, 4 µg/ml; capreomycin and kanamycin, 30 µg/ml each; amikacin, 20 mg/ml; clofazimine, 2.5 mg/ml; ofloxacin and ciprofloxacin, 5 µg/ml each; and sparfloxacin, 1.5 µg/ml. Clarithromycin (Abbott Laboratories, North Chicago, Illinois); clofazimine (Ciba-Geigy, Basel, Switzerland); ofloxacin (Laboratoire Diamant, Puteaux, France), ciprofloxacin (Bayer Pharma, Puteaux, France), and sparfloxacin (Rhone-D.P.C., Antony, France) were kindly provided by their manufacturers, whereas all other drugs were purchased from Sigma Chemical Co. (St. Louis, Missouri).

Results Radiometric MICs. The radiometric tests were performed against a panel of both drug-susceptible and resistant clinical isolates including MDR-TB isolates as well as the type strain H37Rv. Typical radiometric data illustrating the radiorespirometric inhibition by MIC levels of isoniazid, rifampin, ethambutol, streptomycin, ethionamide, and D-cycloserine are shown

169

N. Rastogi et al.: Comparative Activity of Antituberculous Drugs

Fig. 1. Typical radiometric data illustrating the radiorespirometric inhibition by MIC levels of isoniazid, rifampin, ethambutol, streptomycin, ethionamide, and D-cycloserine against a drug-susceptible wild-type clinical isolate of M. tuberculosis (clinical isolate 90-0216; Fig. 1A) and a MDR-TB isolate (clinical isolate 91-0328; Fig. 1B). The growth was monitored radiometrically in 7H12 broth with the Bactec 460-TB apparatus at pH 6.8; INH, isoniazid; RIF, rifampin; EMB, ethambutol; SM, streptomycin; ETH, ethionamide; D-CS, D-cycloserine.

both against a drug-susceptible wild-type clinical isolate of M. tuberculosis (clinical isolate 90-0216; Fig. 1A) and a MDR-TB isolate (clinical isolate 91-0328; Fig. 1B). As shown in Fig. 1A, the radiorespirometry analysis clearly showed that only 0.2 µg/ml of rifampin was able to completely inhibit bacterial metabolism in case of a susceptible strain, whereas even 32 µg/ml of rifampin was without any substantial effect in case of the drug-resistant isolate (Fig. 1B). Similarly, any drug resistance observed on the Middlebrook 7H11 agar medium against other drugs was invariably associated with a lack of activity by use of the Bactec method, as illustrated in Fig. 1; e.g., MICs against drug-susceptible and MDR-TB isolates were, respectively, 0.4 and .10 µg/ml for isoniazid, 1 and 5 µg/ml for ethambutol, 1 and .16 µg/ml for strepto mycin, 0.25 and 4 µg/ml for ethionamide, and 75 and .100 µg/ml for D-cycloserine.

Comparative radiometric MICs of all 14 drugs tested are summarized in Tables 1 and 2. It is interesting to note that the radiometric MICs of all 14 drugs were within an extremely narrow range in case of all the strains, provided they were individually susceptible to the drugs tested (values expressed as µg/ml); isoniazid (0.02–0.04), rifampin (0.2–0.4), ethambutol and streptomycin (0.5–2.0), ethionamide (0.25–0.5), D-cycloserine (25–75), capreomycin (1–2), kanamycin (2–4), amikacin (0.5–1.0), clofazimine (0.1– 0.4), ofloxacin (0.5–1.0), ciprofloxacin (0.25–1.0), and sparfloxacin (0.1–0.4). On the other hand, our data

Table 1. Bactec radiometric MICs (µg/ml) of isoniazid, rifampin, ethambutol, streptomycin, ethionamide, and D-cycloserine against Mycobacterium tuberculosis Drugsa Strain H37Rv 90-0145b 90-0201b 90-0216b 90-0240b 90-0241b 88-0902c 90-0233d 91-0328c 92-0492e

INH

RIF

EMB

SM

ETH

D-CS

0.02 0.04 0.04 0.04 0.04 0.04 2.5 2.5 .10.0 5.0

0.4 0.4 0.2 0.2 0.2 0.4 .32.0 0.2 .32.0 .32.0

0.5 1.0 1.0 1.0 1.0 2.0 20.0 1.0 5.0 2.0

1.0 2.0 1.0 1.0 0.5 2.0 8.0 16.0 .16.0 1.0

0.25 0.25 0.25 0.25 0.50 0.50 2.0 0.25 4.0 0.25

25.0 50.0 50.0 75.0 75.0 75.0 .100.0 75.0 .100.0 50.0

a

INH, isoniazid; RIF, rifampin; EMB, ethambutol; SM, streptomycin; ETH, ethionamide; D-CS, D-cycloserine. b Drug-susceptible clinical isolates. c MDR-TB clinical isolate resistant to INH, RIF, EMB, SM, ETH, and D-CS with the proportional method on Middlebrook 7H11 medium. d Clinical isolate resistant to INH and SM with the 7H11 medium. e MDR-TB clinical isolate resistant to INH and RIF with the 7H11 medium.

showed that the activity of the second-line drugs remained unaltered in case of MDR-TB isolates usually resistant to routine first-line drugs only; e.g., the three quinolone drugs were active against all ten strains, with the order of activity sparfloxacin .

170


Table 2. Bactec radiometric MICs (µg/ml) of second-line anti-tuberculous drugs and the macrolide clarithromycin against Mycobacterium tuberculosis Drugsa CLA Strain H37Rv 90-0145b 90-0201b 90-0216b 90-0240b 90-0241b 88-0902c 90-0233d 91-0328c 92-0492e

CAP

KAN

AMIK

CLOFA

OFL

CIP

SPAR

pH 6.8

pH 7.4

2.0 2.0 1.0 2.0 1.0 2.0 2.0 2.0 2.0 1.0

2.0 4.0 2.0 2.0 2.0 2.0 2.0 4.0 2.0 2.0

0.5 0.5 0.5 0.5 0.5 0.5 1.0 0.5 0.5 1.0

0.1 0.2 0.2 0.4 0.4 0.4 0.2 0.4 0.2 0.4

0.75 0.50 0.50 0.50 0.50 0.75 0.5 0.5 1.0 1.0

0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.25 1.0 0.75

0.2 0.2 0.1 0.2 0.1 0.1 0.1 0.1 0.4 0.2

20.0 10.0 5.0 5.0 10.0 10.0 20.0 5.0 20.0 10.0

8.0 2.0 2.0 2.0 4.0 2.0 4.0 2.0 16.0 4.0

a

CAP, capreomycin; KAN, kanamycin; AMIK, amikacin; CLOFA, clofazimine; OFL, ofloxacin; CIP, ciprofloxacin; SPAR, sparfloxacin; CLA, clarithromycin. b Drug-susceptible clinical isolates. c MDR-TB clinical isolate resistant to INH, RIF, EMB, SM, ETH, and D-CS with the proportional method on Middlebrook 7H11 medium. d Clinical isolate resistant to INH and SM with the 7H11 medium. e MDR-TB clinical isolate resistant to INH and RIF with the 7H11 medium.

ofloxacin . ciprofloxacin. Similarly, there was no difference between CMIs of capreomycin, amikacin, kanamycin, and clofazimine obtained against drug-susceptible or drug-resistant isolates (Table 2).

The MICs of D-cycloserine (range of 25 to 75 µg/ml for susceptible strains) and the macrolide drug clarithromycin (MIC range of 5 to 20 µg/ml) were much higher than those of other first- and second-line drugs. This relative lack of activity of macrolides against M. tuberculosis compared with their high activity against atypical mycobacteria is in agreement with previously published observations [14, 21]. However, initially higher MICs of clarithromycin recorded at the routine pH of 6.8 6 0.2 (pH of the commercialized 7H12 broth) were considerably lowered to an MIC range of 2 to 16 µg/ml, when tested at an alkaline pH of 7.4. Bactericidal activity against drug-susceptible strains. The bactericidal effect of various drugs in the Bactec system was compared between the drugsusceptible and drug-resistant clinical isolates by plating the bacterial suspensions from individual Bactec vials at the beginning and at the end of the experiments on 7H11 agar medium for viable count enumeration. Only selected results (expressed as mean viable count 6 standard error) in case of susceptible and drug-resistant strains are illustrated in Figs. 2 and 3 respectively. As can be seen in the case of six drug-susceptible strains studied (Fig. 2A), about a 1

log growth was observed during 7 days of incubation within the confined atmosphere of the control Bactec vials. Among the first-line drugs, both isoniazid and rifampin resulted in more than 99% (2 log) killing of the bacterial inoculum added at time 0 at a fixed concentration of about 4xMIC, i.e., 0.16 and 1.6 µg/ml respectively, against all strains. Rifampin was uniformly more bactericidal than isoniazid and resulted in a 4 log (99.99%) killing of 2 isolates, 3 log (99.9%) killing of three isolates, and more than 2 log (99.5%) killing of the one remaining isolate (Fig. 2A). Isoniazid was able to kill three isolates by 3 logs (99.9% killing) and the three other isolates between 2 and 3 logs (99 to 99.5% killing). Streptomycin and ethionamide resulted in a 99.9% (3 log) killing in case of two (90-0201 and 90-0240) and one (90-0201) isolates respectively, whereas both ethambutol and D-cycloserine resulted in a 90% (1 log) killing of one isolate each only (Fig. 2A). Consequently, rifampin and isoniazid had the highest bactericidal activity followed by ethionamide and streptomycin, whereas ethambutol and D-cycloserine showed marginal activity only. For other drugs (Fig. 2B), all the three fluoroquinolones were more bactericidal than other second-line drugs. Sparfloxacin used at 0.5 µg/ml was at least equally or more active than ofloxacin used at 2.0 µg/ml and ciprofloxacin at 1.0 µg/ml. Sparfloxacin was uniformly bactericidal compared with other quinolones and resulted in more than 99.9% (3 log)


171

Fig. 2. The comparative bactericidal effect of various drugs in Bactec 7H12 vials against the drug-susceptible strains of M. tuberculosis. Results illustrate mean viable counts 6 standard error in the presence of selected concentrations of various drugs after 7 days of incubation at 37°C, as compared with growth in untreated control vials. Colony-forming units were enumerated by plating the bacterial suspensions from individual Bactec vials at the beginning and at the end of the experiments on 7H11 agar medium. The initial inoculum at time 0 was taken as 1 and is represented by the dotted line on the figure. Abbreviations used (values after drug abbreviations indicate concentration in µg/ml); Control D7, control at day 7; INH, isoniazid; RIF, rifampin; EMB, ethambutol; SM, streptomycin; ETH, ethionamide; D-CS, D-cycloserine; CAP, capreomycin; KAN, kanamycin; AMIK, amikacin; CLOFA, clofazimine; OFL, ofloxacin; CIP, ciprofloxacin; SPAR, sparfloxacin; CLA, clarithromycin.

Fig. 3. Comparative bactericidal effect of second-line drugs in 7H12 vials against three M. tuberculosis isolates with variable drug resistance. The initial inoculum at time 0 was taken as 1 and is represented by the dotted line on the figure. Please refer to legend to Fig. 2 for designations used in figures; values after drug abbreviations indicate concentration in µg/ml.

killing for three isolates, and 99% (2 log) killing for the remaining three isolates. Ciprofloxacin resulted in more than 99.9% and 99% killing for three and one isolates respectively, whereas ofloxacin resulted in 99.9% killing for three isolates (Table 2B). In our hands, some variability in killing by fluoroquinolones

among some isolates was, however, observed; e.g., the H37Rv type strain was more efficiently killed by ofloxacin, whereas clinical isolates 90-0145 and 900241 were more efficiently killed by ciprofloxacin and sparfloxacin (Fig. 2B). For the remaining drugs, amikacin at 2.0 µg/ml was more bactericidal for all

172


Fig. 4. Intracellular activity of 14 first- and second-line antituberculous drugs against the M. tuberculosis H37Rv strain phagocytized in cultured human macrophages. Values represent mean viable counts 6 standard error in control and test wells, 3 and 7 days after drug addition. The dotted line represents the initial inoculum at time 0, i.e., the number of bacilli effectively phagocytized after 4 h of incubation, prior to drug addition. Please refer to legend to Fig. 2 for designations used in figures; values after drug abbreviations indicate concentration in µg/ml and correspond to their reported Cmax in humans.

the strains than capreomycin and kanamycin at 4.0 and 8.0 µg/ml respectively; these drugs resulted in a 2 log (99.9%) killing of five, three, and two isolates respectively. Lastly, both clofazimine and clarithromycin showed only marginal bactericidal activity at 4xMIC to 8xMIC concentrations tested, with none being able to cause a 1 log (90%) reduction of the initial bacterial inoculum added at time 0 (Fig. 2B).

Bactericidal activity against drug-resistant strains. In isolates resistant to first-line drugs, a bactericidal effect was observed at concentrations much higher than in drug-susceptible clinical isolates (results not shown; e.g., in the MDR-TB isolate 88-0902, concentrations as high as 10 µg/ml of isoniazid and 16 µg/ml of ethionamide were needed to cause about 2 log of killing). On the other hand, the bactericidal effect of second-line drugs remained unaltered against this MDR-TB isolate (Fig. 3). Similar results for other isolates with variable drug resistance to first-line drugs (isolate 90-0233 resistant to isoniazid, and streptomycin and isolate 90-0492 resistant to isoniazid and rifampin) were also observed with the full panel of second-line drugs and are illustrated in Fig. 3. Overall, both sparfloxacin and amikacin were the most active drugs resulting in a 99.9% (3 log) of

killing activity for all the isolates. Ciprofloxacin, ofloxacin, and capreomycin showed a 2 log (99%) killing of two of the three drug-resistant isolates and a 1 log (90%) reduction for the third remaining isolate, followed by kanamycin and clofazimine, which were able to result in a 1 log (90%) killing of two and one isolates respectively. On the other hand, clarithromycin did not result in a 1 log killing of any of the isolates tested. Intracellular activity in human macrophages. Results concerning the intracellular activity of all the 14 drugs against the M. tuberculosis H37Rv strain phagocytized in cultured human macrophages are summarized in Fig. 4. In accordance with previously reported investigations concerning intracellular activity of drugs against pathogenic mycobacteria [12, 14, 15, 17, 22], all the drugs were used at their reported Cmax in human. Considering that a 2 log (99%) decrease of viability compared with the initial inoculum in the macrophage model is synonymous high bactericidal activity in vivo [11], rifampin, isoniazid, and ethionamide among first-line drugs, and sparfloxacin and ofloxacin among second-line drugs, were found to be highly bactericidal against phagocytized M. tuberculosis (Fig. 4). Consequently, all these five drugs meet


criteria set for choosing a highly bactericidal molecule to treat tuberculosis patients. However, more than a 3 log (99.9%) killing of the initial inoculum of the virulent H37Rv strain by rifampin was really exceptional (Fig. 4A) and was not paralleled by any other drug.

The next most active drugs relatively close to the 99% reduction limit of the initial inoculum were clofazimine and amikacin (Fig. 4B). Lastly, streptomycin, ethambutol, kanamycin, capreomycin and ciprofloxacin showed more or less similar intracellular activity resulting in more than 1 log (90%) killing activity, whereas D-cycloserine was only moderately bactericidal (about 80% killing of the initial inoculum). Clarithromycin, on the other hand, was the only drug without any bactericidal effect in the intracellular model (Fig. 4B).

Discussion Tuberculosis still results in about 10 million new cases each year with about 3 million deaths. Emergence of AIDS has resulted in profound changes in the epidemiological aspects of the mycobacterial diseases, putting the risk of contracting tuberculosis very high among the HIV1 population. The presence of pulmonary tuberculosis is one of the recommended markers for classification of clinical stage 3 in AIDS, and extrapulmonary tuberculosis is a recommended diagnostic criterion for clinical stage 4 disease. Humans are the main reservoir of the tubercle bacilli in nature. Among about one-third of world’s population estimated to be infected by tubercle bacilli, the majority of the infected individuals carry their infection without ever developing the disease. This infection, however, progresses in disease in about 10 million people each year who become the transmitters of the tubercle bacilli to the noninfected. There exists no possibility until today of stopping this transmission to the noninfected, neither is there any possibility of eliminating the bacilli from as yet healthy but infected individuals. Although the eradication of tuberculosis is not in the realm of immediate possibilities, efficient treatment regimens must be employed to treat the ‘‘transmitters’’ and/or the ‘‘diseased,’’ not only to treat the host but to stop the progression of the disease, an objective that can be achieved only by following efficient chemoprophylaxis and antibiotherapy. As intrinsic resistance observed for M. tuberculosis reflects spontaneous mutations, at least two or more bactericidal drugs are required to avoid the

173

emergence of resistance. The standard regimen to treat tuberculosis today is a 9-month treatment with isoniazid 1 rifampin 1 pyrazinamide with either ethambutol or streptomycin added initially pending drugsusceptibility results. However, prior to the drugsusceptibility screening of tubercle bacilli, which takes about 3–4 weeks, the isolation of bacteria from clinical specimens is needed, which takes another 4–8 weeks with classical methods of laboratory diagnosis. The recent development of a rapid radiometric mycobacterial detection and susceptibility testing system using the Bactec 460TB apparatus now permits a total gain of about 4–6 weeks for isolation, presumptive identification, and drug susceptibility results [11]. The above-mentioned Bactec 460TB method was used in the present investigation to evaluate the activity of 14 drugs against both drug-susceptible and resistant isolates of M. tuberculosis. This investigation showed that the radiometric MICs of all the drugs were within an extremely narrow range in case of drugsusceptible strains, and that the MICs of second-line drugs remained unaltered in case of isolates resistant only to routine first-line drugs. Our results corroborate the conclusion that the Bactec method is as suitable for the regular screening of second-line antituberculous drugs [5, 6] as it has been previously for first-line drugs [8, 16, 24, 25], giving MIC results consistent with the bactericidal effects of the drugs. However, this investigation was performed on only a limited number of strains, and the above observations would now be confirmed by extending this investigation to a larger number of isolates in our routine clinical microbiology setting. One of the unique features of the present investigation was the evaluation of intracellular activity of 14 drugs against M. tuberculosis growing inside human macrophages. Indeed, most of the in vitro antimicrobial testing has been essentially devised keeping in mind extracellular bacteria, and should be modified for M. tuberculosis as intricate bacteria-phagocyte interactions may not only contribute to the pathogenesis of the disease, but may also alter the expected response to therapy [11]. In this context, the following considerations seem important while considering effective therapy of mycobacterial infections [11]: (a) an effective drug must be able to penetrate and eventually concentrate inside M. tuberculosis-infected macrophages, (b) it must be able to cross the lipid-rich domains of the cell wall, and (c) it must eventually act synergistically with the host’s immune system to eliminate the infection. Although the last issue is as yet a field neglected in mycobacteriology, the first two issues are satisfactorily addressed by using M. tubercu-

174


Table 3. Suggested susceptibility criteria for the interpretation of Bactec radiometric MICs (µg/ml) of 14 first- and second-line drugs against Mycobacterium tuberculosis in 7H12 brotha Drug Isoniazid Rifampin Ethambutol Streptomycin Ethionamide D-cycloserinec Capreomycin Kanamycin Amikacin Clofazimine Sparfloxacin Ofloxacin Ciprofloxacin Clarithromycin

Susceptible #0.2b #2.0b #2.5b #2.0b #1.25 #100 #4.0 #4.0 #2.0 #1.0 #1.0 #2.0 #2.0 NAd

Intermediate

Resistant

0.5–1.0 4.0 5.0 4.0

$2.0 $8.0 $10.0 $8.0

2.5 200 8.0 8.0 4.0 2.0 2.0 4.0 4.0 NA

$5.0 $400 $16.0 $16.0 $8.0 $4.0 $4.0 $8.0 $8.0 NA

a Data compiled from this paper and previously published observations [5–8, 11, 13, 16, 19, 24, 25]. b Suggested critical concentrations for so-called SIRE (streptomycin, isoniazid, rifampin, and ethambutol) drugs. The values are from a previous evaluation performed on 100 clinical isolates of M. tuberculosis [16]. c Correlation between Bactec and conventional tests for Dcycloserine may not be reproducible. D-cycloserine is a difficult drug for routine mycobacterial drug-susceptibility testing. d Not applicable, as clarithromycin is not active against M. tuberculosis.

losis-infected macrophages in vitro to mimic their projected intracellular action in a host.

To better correlate with a human host, we used a human macrophage system coupled with intracellular screening of drugs at concentrations within their reported Cmax levels in human [11, 17, 22]. As shown in the Results section above, the intracellular data with all the drugs were in accordance with their extracellular bactericidal activity except for clofazimine, which gave unexpected results. Although this latter drug had radiometric MICs as low as 0.1 for the H37Rv strain, even 8xMIC of this drug (i.e., 0.8 µg/ml) showed a bacteriostatic effect only against the bacterial inoculum. This result was not unexpected as extracellular MBC/MIC ratios for this drug against M. tuberculosis are as high as 32 to 64. On the other hand, cidal activity of 2.5 µg/ml of this drug paralleled that of 20 µg/ml of amikacin in the macrophage model. This discrepancy between extracellular and intracellular activity of clofazimine was in agreement with previously published data, as this drug concentrates manyfold within the phagosomes and phagolysosomes of infected macrophages [11, 15]. In conclusion, based on an decreasing order of bactericidal activity, our

results showed the following spectrum of intracellular drug action: among the first-line drugs, rifampin . ethionamide 5 isoniazid . ethambutol . streptomycin . D-cycloserine; among second-line drugs, clofazimine 5 amikacin . kanamycin 5 capreomycin; among fluoroquinolones, sparfloxacin . ofloxacin . ciprofloxacin. For M. tuberculosis which responds well to chemotherapy, it can be concluded that bactericidal drugs have in common lower MICs, critical concentrations, and intracellular bactericidal concentrations as compared with their higher Cmax levels in human [11]. In this respect, both fluoroquinolones (e.g., sparfloxacin and ofloxacin), aminoglycosides (particularly amikacin), and newer rifamycins (rifabutin and a benzoxazinarifamycin KRM-1648) appear to be good candidates to treat infections by MDR-TB isolates [reviewed in 11]. Another approach to counteract the drug resistance is based on the concomitant use of a drug in association with an inhibitor of the cell wall synthesis [2], or by synthesizing amphipathic derivatives of existing antituberculous drugs; e.g., the fact that amikacin at 2.0 µg/ml was more bactericidal for all the strains than capreomycin and kanamycin at 4.0 and 8.0 µg/ml respectively may also be linked to the fact that it is a semisynthetic amphipathic kanamycin derivative with a butyric acid moiety at the R3 position of kanamycin [11]. Recent reports showing the development of progressive drug resistance while undergoing treatment is of concern, and the development of resistance to drugs that, while related (e.g., newer quinolones and rifamycins), are considered to be more active than the drugs actually used is disturbing [19]. This problem of acquired drug resistance often arises when a patient is initially treated with drugs to which the isolate was already resistant at the time when the treatment was started (awaiting the results of drug susceptibility testing). Thus the Bactec method has its place to overcome not only this avoidable delay to provide rapid susceptibility results, but also in helping to screen both first- and second-line drugs in order to provide a larger choice of tuberculocidal drugs in the present era of emerging MDR-TB cases. On the basis of the present investigation and previously reported observations [5–8, 11, 13, 16, 19, 24, 25], the suggested breakpoints for various first- and second-line drugs are illustrated in Table 3. The breakpoints for second-line drugs should now be tested on a much larger scale, including clinical isolates resistant to second-line drugs. However, not many tuberculosis patients have received the above secondary drugs for treatment yet, and for this reason the suggested three-


point MIC determination during monitoring of patients undergoing chemotherapy compared with their clinical and bacteriologic response would permit validation of breakpoints able to classify strains in near future, as fully susceptible, intermediate, or resistant. ACKNOWLEDGMENTS The mycobacteria project at Guadeloupe was supported through the ‘‘project CORDET’’ of Ministry of Overseas Departments and Territories, French Republic, and the Institut Pasteur Foundation, Paris, France. We thank Becton-Dickinson-France for lending the Bactec 460-TB apparatus.

Literature Cited 1. Bloch AB, Cauthen GM, Onorato IM, Kenneth G, Dansbury G, Kelly GD, Driver CR, Snider DE (1994) Nationwide survey of drug-resistant tuberculosis in the United States. J Am Med Assoc 271:665–671 2. Cavalieri SJ, Biehle JR, Sanders WE Jr. (1995) Synergistic activities of clarithromycin and antituberculous drugs against multidrug-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother. 39:1542–1545 3. Edberg SC, Sabath LS (1980) Determination of antibiotic levels in body fluids: techniques and significance. Bacterial tests in endocarditis and other severe infections. In Lorian V (ed) Antibiotics in laboratory medicine. Baltimore: The Williams & Wilkins Co., pp. 206–264 4. Frieden TR, Sterling T, Pablos-Mendez A, Kilburn JO, Gauthen GM, Dooley SW (1993) The emergence of drug-resistant tuberculosis in New York City. N Engl J Med 328:521–526 5. Heifets LB (1988) Qualitative and quantitative drug-susceptibility tests in mycobacteriology. Am Rev Respir Dis 137:1217– 1222 6. Heifets LB, Lindholm-Levy P (1989) Comparison of bactericidal activities of streptomycin, amikacin, kanamycin, and capreomycin against Mycobacterium avium and M. tuberculosis. Antimicrob Agents Chemother 33:1298–1301 7. Iseman MD (1993) Treatment of multidrug-resistant tuberculosis. N Engl J Med 329:784–791 8. Lee CN, Heifets LB (1987) Determination of minimal inhibitory concentrations of antituberculous drugs by radiometric and conventional methods. Am Rev Respir Dis 136:349–352 9. Levy L (1974) Pharmacological studies with clofazimine (B663) in man. Am J Trop Med Hyg 23:1116–1121 10. McClatchy JK (1980) Antituberculous drugs: mechanisms of action, drug resistance, susceptibility testing, and assays of activity in biological fluids. In Lorian V (ed) Antibiotics in laboratory medicine. Baltimore: The Williams & Wilkins Co., pp. 135–169 11. Rastogi N (1993) Mycobacteria as intracellular pathogens: current notions of pathogenicity, virulence, and drug resistance and their relation to effective therapy, In Raoult D (ed) Antimicrobial agents and intracellular pathogens. Boca Raton: CRC Press, pp. 245–300 12. Rastogi N, Blom-Potar MC (1990) Intracellular bactericidal activity of ciprofloxacin and ofloxacin against Mycobacterium tuberculosis H37Rv multiplying in the J-774 macrophage cell line. Zentralbl Bakteriol 273:195–199

175 13. Rastogi N, Goh KS (1991) In vitro activity of the new difluorinated quinolone sparfloxacin (AT-4140) against Mycobacterium tuberculosis compared with activities of ofloxacin and ciprofloxacin. Antimicrob Agents Chemother 35:1933– 1936 14. Rastogi N, Labrousse V (1991) Extracellular and intracellular activities of clarithromycin used alone and in association with ethambutol and rifampin against Mycobacterium avium complex. Antimicrob Agents Chemother 35:462–470 15. Rastogi N, Potar MC, David HL (1987) Intracellular growth of pathogenic mycobacteria in the continuous murine macrophage cell line J-774: ultrastructure and drug-susceptibility studies. Curr. Microbiol. 16:79–92 16. Rastogi N, Goh KS, David HL (1989) Drug susceptibility testing in tuberculosis: a comparison of the proportion methods using Lowenstein-Jensen, Middlebrook 7H10 and 7H11 agar media and a radiometric method. Res Microbiol 140:405–417 17. Rastogi N, Labrousse V, Goh KS, Carvalho de Sousa JP (1991) Antimycobacterial spectrum of sparfloxacin and its activities alone and in association with other drugs against Mycobacterium avium complex growing extracellularly and intracellularly in murine and human macrophages. Antimicrob Agents Chemother 35:2473–2480 18. Rastogi N, Kochi A, Vareldzis B, Styblo K, Crawford JT, Jarvis WR, McGowan JE Jr, Perrone C, David HL, Zhang Y, Cohn DL, Iseman MD (1993a) 9th Forum in Microbiology, Emergence of multiple-drug-resistant tuberculosis: fundamental and applied research aspects, global issues and current strategies. Res Microbiol 144:103–158 19. Rastogi N, Ross BC, Dwyer B, Goh KS, Clavel-Se´re`s S, Jeantils V, Cruaud P (1993b) Emergence during unsuccessful chemotherapy of multiple drug resistance in a strain of Mycobacterium tuberculosis. Eur J Clin Microbiol Infect Dis 11:901–907 20. Rastogi N, Bauriaud RM, Bourgoin A, Carbonnelle B, Chippaux C, Gevaudan MJ, Goh KS, Moinard D, Ross P (1995a) French multicenter study involving eight test sites for radiometric determination of activities of 10 antimicrobial agents against Mycobacterium avium complex. Antimicrob Agents Chemother 39:638–644 21. Rastogi N, Goh KS, Ruiz P, Casal M (1995b) In vitro activity of roxithromycin against the Mycobacterium tuberculosis complex. Antimicrob Agents Chemother 39:1162–1165 22. Rastogi N, Labrousse V, Bryskier A (1995c) Intracellular activities of roxithromycin used alone and in combination with other drugs against Mycobacterium avium complex in human macrophages. Antimicrob Agents Chemother 39:976–978 23. Riley LW (1993) Drug-resistant tuberculosis. Clin Infect Dis 17 (Suppl 2):S442–S446. 24. Siddiqi SH, Libonati JP, Middlebrook G (1981) Evaluation of a rapid radiometric method for drug susceptibility testing of Mycobacterium tuberculosis. J Clin Microbiol 13:908–912 25. Siddiqi SH, Hawkins JE, Laszlo A (1985) Interlaboratory drug susceptibility testing of Mycobacterium tuberculosis by a radiometric procedure and two conventional methods. J Clin Microbiol 22:919–923 26. Wise R, Griggs D, Andrews JM (1988) Pharmacokinetics of the quinolones in volunteers: a proposed dosing schedule. Rev Infect Dis 10 (Suppl 1):S83–S98