Impact of drug resistance on fitness of Mycobacterium tuberculosis

FEMS Immunology and Medical Microbiology 42 (2004) 281–290 www.fems-microbiology.org

Impact of drug resistance on fitness of Mycobacterium tuberculosis strains of the W-Beijing genotype Olga S. Toungoussova a,b,*, Dominique A. Caugant a,c, Per Sandven a, Andrey O. Mariandyshev d, Gunnar Bjune b b

a Division of Infectious Disease Control, Norwegian Institute of Public Health, Oslo, Norway Department of General Practice and Community Medicine, Faculty of Medicine, University of Oslo, P.O. Box 1130, Blindern, 0317 Oslo, Norway c Department of Oral Biology, Dental Faculty, University of Oslo, Oslo, Norway d Department of Phthisiopulmonology, Northern State Medical University, Archangel, Russia

Received 20 October 2003; received in revised form 21 April 2004; accepted 29 May 2004 First published online 15 June 2004

Abstract Mycobacterium tuberculosis strains of the W-Beijing genotype became a common cause of tuberculosis during the past years and they are often associated with drug resistance. The biological factors facilitating the selection and wide dissemination of these strains are not known. To determine how acquisition of drug resistance affected growth of strains of the W-Beijing genotype, the growth of 55 M. tuberculosis isolates were studied using the BBL MGIT Mycobacteria Growth Indicator Tube and the BACTEC MGIT 960 System. Susceptible strains of non-Beijing genotypes were found to be the most fit strains. Drug-resistant strains of non-Beijing genotypes were more likely to grow slower than susceptible strains (P ¼ 0:001). Drug-resistant strains of the W-Beijing genotype had two tendencies of growth: some of them showed reduced growth compared to susceptible strains, while others did not show loss of fitness measured as growth. Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords: Mycobacterium tuberculosis; Fitness; W-Beijing family

1. Introduction The incidence of tuberculosis has been rising throughout the world [1] and has been accompanied by increasing anti-tuberculosis drug resistance. Surveillance data obtained from developing countries indicate that nearly 50 million persons worldwide may be infected with drug-resistant Mycobacterium tuberculosis strains [2]. During the past years, the proportion of M. tuberculosis strains having multidrug resistance (MDR), i.e. resistance to at least rifampin and isoniazid [3], has increased dramatically. It is obvious that the evolution and spread of anti-tuberculosis drug resistance is due to improper use of antibiotics. The reversion of this situation *

Corresponding author. Fax: +47-22850672. E-mail addresses: [email protected], [email protected] (O.S. Toungoussova).

and combating tuberculosis caused by drug-resistant strains depend on several measures such as diagnosis, proper use of antibiotics, prescription of adequate treatment regimens, prevention of non-compliance and introduction of new anti-tuberculosis drugs. Many pathogens develop genotypes that have highly effective virulence including more efficient transmission [4]. M. tuberculosis strains belonging to the W-Beijing family is a good example. These isolates exhibit closely related restriction fragment length polymorphism (RFLP) patterns and contain the last 9 of the 43 polymorphic spacer sequences in the chromosomal direct repeat (DR) locus by spoligotyping [5]. Such MDR isolates were identified in the USA in 1990s and designated the W strains [6]. In 1995, M. tuberculosis isolates with similar characteristics were found in the Beijing province, China, and were designated the Beijing genotype [5,7]. These genetically related isolates, variously

0928-8244/$22.00 Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsim.2004.05.012

282

O.S. Toungoussova et al. / FEMS Immunology and Medical Microbiology 42 (2004) 281–290

called W or Beijing, are now referred as the W-Beijing family [6]. Scientific publications indicate that the W-Beijing strains are prevalent in population based studies [8–12]; however most of these studies have bias collections. Strains of the W-Beijing family have been shown to acquire resistance to anti-tuberculosis drugs more frequently than strains of other genotypes [8–12]. Positive selection of the W-Beijing M. tuberculosis strains accompanied by rising resistance to anti-tuberculosis drugs might explain their dissemination worldwide. Mutations leading to drug resistance development may influence the fitness of the microorganism. The definition of fitness includes a microorganism’s ability to survive, reproduce and to be transmitted [13]. Evidence has been provided that drug-resistant mutants of M. tuberculosis have reduced fitness compared to their parent susceptible strains [14–16]. The relative fitness of the W-Beijing M. tuberculosis strains after they acquire resistance to anti-tuberculosis drugs has, however, not been studied. Drug-resistant W-Beijing strains, like ordinary resistant mutants, may loose fitness when they acquire resistance to anti-tuberculosis drugs. If they do, then they should not be competitive towards susceptible M. tuberculosis strains unless antibiotics are present. The reality shows that the W-Beijing strains often constitute the bulk of primary MDR [10–12]. Growth in vitro is one of several fitness indicators. The aim of the present study was to measure and compare growth of drug-susceptible and resistant W-Beijing and non-Beijing M. tuberculosis strains. To determine the impact of drug resistance on fitness, the growth of susceptible and drug-resistant strains belonging to both genotypes was analysed using the BBL MGIT Mycobacteria Growth Indicator Tube and the BACTEC MGIT 960 System. The growth of nine pairs of isolates obtained from the same patients before and after acquisition of resistance to additional drugs was also analysed.

encoding the b-subunit of RNA polymerase (rpoB), leading to rifampin resistance. A cluster of M. tuberculosis strains was defined as two or more strains exhibiting 100% identical IS6110 RFLP patterns [5,18,19]. In addition, nine pairs of M. tuberculosis strains isolated from the same patients with identical IS6110 RFLP patterns, but showing different susceptibility patterns over time were included. Two of the nine pairs were isolated from patients originally from former Yugoslavia (patient 1) and Somalia (patient 2) living in Norway, the remaining seven pairs were obtained from Russian patients living in Archangel, Russia (Table 1). The identification of the isolates was performed using 16S rRNA hybridisation technique (AccuProbe; GenProbe Inc., San Diego, CA, USA) and standard microbiological tests (niacin accumulation test and nitrate reduction test). RFLP analysis of M. tuberculosis DNA was performed according to the internationally standardised methodology [20,21]. Spoligotyping was performed by using a commercially available kit (Isogen Bioscience BV, Maarssen, The Netherlands) according to the instructions supplied by the manufacturer, as previously described [22]. M. tuberculosis strains were defined as belonging to the W-Beijing family if they showed the following characteristics: the strains harboured a high number of IS6110 copies (from 13 to 18) and more than two-thirds of them were present at the same genomic sites (they clustered within 60% similarity); and they had identical spoligotyping results, showing hybridisation only with the last 9 of the 43 possible spacers [5,7,12]. M. tuberculosis strains of the W-Beijing and non-Beijing genotypes included isolates susceptible and resistant to the first line anti-tuberculosis drugs (ethambutol, isoniazid, rifampin and streptomycin). Drug susceptibility testing of all strains was performed using the radiometric broth method (BACTEC, Becton Dickinson Diagnostic Systems, Sparks, MD, USA) [23–25]. Mutations associated with rifampin resistance in the rpoB gene were identified using the InnoLiPA Rif. TB test (Innogenetics N.V., Ghent, Belgium) [26,27].

2. Materials and methods 2.2. Strain cultivation and growth 2.1. M. tuberculosis strains Clinical M. tuberculosis strains belonging to the W-Beijing and non-Beijing genotypes were included in the study. A total of 55 M. tuberculosis strains (9 susceptible of the W-Beijing genotype, 15 resistant of the W-Beijing genotype, 11 susceptible of non-Beijing genotype, and 20 resistant of non-Beijing genotype) were selected from a collection of clinical M. tuberculosis isolates (Table 1), made during 1998–2001 from patients with pulmonary tuberculosis in Archangel, Russia [12,17]. Selection of strains involved clustered strains and strains having different mutations in the gene

The strains were taken from the )70 °C freezer, defrosted, inoculated to Lowenstein–Jensen media and cultivated for three weeks at 37 °C. A suspension of bacilli was then prepared in 4 ml Middlebrook 7H9 broth (BACTEC, Becton Dickinson Diagnostic Systems, Sparks, MD, USA) containing glass beads. The suspension was vortexed for 20 s and then allowed to sediment for 20 min. The supernatant was transferred to another sterile tube and allowed to sediment for another 15 min. The supernatant was transferred to a new sterile tube, adjusted to turbidity comparable to a McFarland No. 0.5 standard (1.5 108 bacterial cells/ml), and then


283

Table 1 Characteristics of the 55 M. tuberculosis strains from Archangel, Russia Strain no.

Resistance toa

Cluster IS6110b

Non-Beijing 418/98 421/98 853/00 5830/00 216/01 736/02 7004/01 722/02 5871/00 350/98 420/98 419/98 5882/00 193/01 396/98 400/98 6964/01 3633/02 398/98 6963/01 7028/01 83/99 5835/00 5849/00 5849/01 6965/01 231/01 219/01 7000/01 7014/01 721/02 368/95 258/96 205/94 352/95 396/98 68/99 446/97 158/98

Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible H H H HS HS HS HS HSE HRS HRS HRSE HRSE HRSE HRSE HRSE HRSE HRSE HRSE HRSE HRSE Susceptible HR H HR HS HRS HS HRS

I I I I Y Y Z Z K K M I K M J J J J M J

W-Beijing 5824/00 3639/02 249/01 6997/01 724/02 729/02 423/98 7012/01 7013/01 7027/01 215/98 5828/00 5868/00 3639/02 423/99 848/00 5866/00 206/01 7002/01 3600/02 360/98 422/99

Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible Susceptible HR HRS HRSE HRSE HRSE HRSE HRSE HRSE HRSE HRSE HRSE HRSE HRSE

P P P P Q Q Q J D

Mutation in rpoB gene

Reproducibility Reproducibility Reproducibility Reproducibility Reproducibility Reproducibility Reproducibility Reproducibility

531 531 531 526 533 533 516 516 516 531 531 531

TCG ! TTG TCG ! TTG TCG ! TTG CAC ! TAC CTG ! CCG CTG ! CCG GAC ! GTC GAC ! GTC GAC ! GTC TCG ! TTG TCG ! TTG TCG ! TTG

531 TCG ! TTG 531 TCG ! TTG 533 CTG ! CCG 531 TCG ! TTG A A B B B B C G G B C A A A A B B B B B C C

Comment

516 516 531 531 531 531 513 513 513 513 531 531 531

GAC ! GGC GAC ! GTC TCG ! TTG TCG ! TTG TCG ! TTG TCG ! TGG CAA ! CTA CAA ! CTA CAA ! CTA CAA ! CTA TCG ! TTG TCG ! TTG TCG ! TTG

Patient Patient Patient Patient Patient Patient Patient Patient

1 1 2 2 3 3 4 4

284


Table 1 (continued) Strain no.

Resistance toa

Cluster IS6110b

Mutation in rpoB gene

7019/01 192/01 503/97 441/98 205/97 160/98 395/98 78/99 346/98 5836/00 443/97 5823/00

HRSE HRSE S HRS S HRSE HS HRS HS HRSE HSE HRSE

G X C C

526 CAC ! TAC 531 TCG ! TTG 516 GAC ! GTC 526 CAC ! CTC

A A B B A A

531 TCG ! TTG 513 CAA ! CTA 531 TCG ! TGG

Comment

Patient Patient Patient Patient Patient Patient Patient Patient Patient Patient

5 5 6 6 7 7 8 8 9 9

a

H, isoniazid; R, rifampin; S, streptomycin; E, ethambutol. b Strains of the same RFLP were assigned to a cluster designed by a capital letter.

diluted 1:500. A volume of 0.5 ml of the dilution (1.5 105 bacterial cells) was added to the BBL MGIT Mycobacteria Growth Indicator Tube (BACTEC, Becton Dickinson Diagnostic Systems, Sparks, MD, USA). The tubes were entered into the BACTEC MGIT 960 System (BACTEC, Becton Dickinson Diagnostic Systems, Sparks, MD, USA) [28], incubated at 37 °C, and monitored for increasing fluorescence. The BACTEC MGIT 960 System performs monitoring for fluorescence (in units) every hour. It does not, however, automatically provide growth curves. The results of strain growth therefore had to be obtained manually by printing out results about every 24th hour during the first 96 h and every 6th hour after the 96th hour. This methodology was tested in a pilot study on four susceptible M. tuberculosis strains (two W-Beijing and two non-Beijing). The reproducibility of the method was demonstrated by testing eight susceptible M. tuberculosis strains of nonBeijing genotypes in three independent experiments starting from cultivation on Lowenstein–Jensen media each time. Standardization of inoculum was achieved by adjusting to turbidity comparable to a McFarland No. 0.5 standard 1 ml of which contains 1.5 108 bacterial cells. The standard provides approximate estimate of number of bacterial cells in a liquid suspension. Measurements of growth for nine pairs of strains obtained from the same patients were performed twice in independent experiments. The remaining strains were tested once. The prepared dilutions were also cultivated on blood agar plates and incubated at 37 °C for 48 h to detect bacterial contamination. None of the dilutions used were contaminated.

t1 ¼ y y1 =y2 y1 , where t ¼ time, and y ¼ growth in units. The lag phase was defined as the time from the start of cultivation to the beginning of detectable growth. In order to test for differences in growth rate, the mean time from the beginning of growth to 200 units of growth was also calculated. SPSS for Windows version 9.0.1 (SPSS, Inc., Chicago, IL) was used for the statistical analysis. Differences between groups were tested by parametric ANOVA test. Difference between two means was used to test for reproducibility of growth of the same strains measured during three independent experiments. A P value of