To the Editors:
In the Eastern Mediterranean Region (EMRO), there are no accurate estimates of numbers of patients with multidrug-resistant (MDR)/extensively drug-resistant (XDR) tuberculosis (TB). In 2006, we documented the existence and transmission of XDR-TB among patients with MDR-TB 1. In a further study, we isolated more dangerous forms of bacilli, named totally drug-resistant (TDR) 2. This group of strains showed in vitro resistance to all first- and second-line drugs tested (i.e. aminoglycosides, cyclic polypeptides, fluoroquinolones, thioamides, serine analogues and salicylic acid derivatives). TDR-TB patients remained smear- and culture-positive after 18 months’ median treatment with second-line drugs. Even changing the treatment to co-amoxyclav (625 mg per 8 h) or clarithromycin (1,000 mg·day−1), along with a high dose of isoniazid (15 mg·kg−1) led to no improvement 2. In the present study, transmission electron microscopy (TEM) was used to evaluate the differences between such deadly bacilli and MDR or susceptible ones at the ultrastructural level. Furthermore, to evaluate the genotyping patterns, the strains were subjected to spoligotyping and variable number of tandem repeats (VNTR) analysis.
Susceptibility testing against first- and second-line drugs was performed using the proportional method on Löwenstein–Jensen media 3.
For spoligotyping and VNTR analysis, five isolates from each group of strains (i.e., susceptible, MDR- and TDR-TB) were included in this study. The spoligotyping and VNTR analysis were performed on extracted DNA using standard protocols 4, 5.
For TEM analysis, a loop of Löwenstein–Jensen culture media was inoculated into Middlebrook 7H9 broth (Difco, Franklin Lakes, NJ, USA) supplemented with 0.2% glycerol and 10% Middlebrook OADC enrichment (Difco). Cells at an optical density (OD) of 0.6 at 600 nm were used for further experiments. These cells were first centrifuged at 800×g for 5 min, then the supernatant was adjusted to an OD of 580 nm, corresponding to 6.3×107 colony-forming units of Mycobacterium tuberculosis per mL. 10 μL of this supernatant were subjected to acid-fast bacilli and Gram staining. Nontuberculous mycobacterial contamination was check by culturing 100 μL of supernatant on 7H11 agar plates. Niacin production, spoligotyping and IS6110–PCR were performed on isolated colonies from each plate. Thereafter, ultra-thin sections for TEM analysis were prepared from fixed and solidified bacterial cells 6.
The identified spoligotypes were as follows: Haarlem I (n = 3; 20%); Beijing (n = 2; 13%); EAI (n = 4; 26 %); CAS (n = 4; 26%); T (n = 1; 6.6%); and U (n = 1; 6.6%) super families. The VNTR profiles of spoligotype strains were different in each superfamily.
In the TEM analysis, three different cell populations were observed at the exponential phase of TDR tubercle bacilli: 1) ordinary (80–70%); 2) round or oval-shaped (15–20%); and 3) extraordinarily thick cell wall bacilli (21–26 nm) that resembled features of stationary or anaerobic dormant bacilli (5–7%) (fig. 1⇓). These adapted forms were detected in all TDR isolates, irrespective of their superfamilies or genotypes patterns. The adapted forms of TDR tubercle bacilli were not detected in susceptible or MDR strains. Under the TEM, marked differences were observed in the thickness of cell walls: 20.2±1.5 nm and 17.1±1.03 nm for the TDR- and MDR-TB bacilli, respectively and 15.6±1.3 nm for the susceptible isolates (p<0.05).
a) Totally drug-resistant (TDR)-tuberculosis (TB) bacilli with oval or round bodies inside them. Scale bar = 500 nm. b) TDR-TB bacilli with stationary or anaerobic dormant bacilli in dividing stage. The increased size of the cell wall is very clear. Scale bar = 100 nm.
While the problem of XDR-TB cases remains unresolved in much of the world, here we report on a more dangerous form of the disease, which we call TDR-TB 2. In these strains, 30% of the total M. tuberculosis populations transformed into adapted forms, producing either round or oval shaped bacilli (15–20%), or having an extraordinarily thick cell wall (5–7%) (fig. 1⇑). These forms were not found among the susceptible or MDR population. At present, we have not demonstrated the transmission of TB disease by these round or oval-shaped bacilli, but the possibility of transmission cannot be ignored 7. The TDR-TB cells with an extraordinarily thick cell wall (26 nm) resembled stationary or dormant-phase bacilli. Such cells were not found in longitudinal and cross-sectional studies at different levels of MDR and susceptible strains, although it has been reported that these cells (1–3%) are present in normal population of M. tuberculosis 8. Under normal circumstances, stationary cells are viable but non-dividing 8. The TDR stationary bacilli in the process of division may represent the dominant population of TDR tubercle bacilli in the near future.
The next important issue is: what drugs can really penetrate these cells, which have cells walls about 21–26 nm thick? As such the tuberculosis chemotherapy is very different from that of other bacterial infections. Mycobacterium has different populations of bacilli which can survive in an aerobic environment, or low-PH conditions (the conditions necessary for pyrazinamide activity) and microaerophilic conditions 9. Each of anti-TB drugs has a major role in dealing with one of these populations 9. However, there are still persistent bacterial populations that are not killed by any available TB drugs 10. Clinically, we have experienced such cases in our TB wards. These patients did not respond to any standardised treatment and remained smear- and culture-positive after prolonged therapy with second-line drugs. Therefore, the problem of TDR-TB should be taken very seriously and this is an issue requiring urgent attention from the global scientific community.
The genotyping results showed different VNTR profiles among identified superfamilies of M. tuberculosis (Haarlem, Beijing, EAI, CAS, and T). The obtained patterns ruled out recent transmission among the studied cases. TEM analysis showed no ultrastructural differences among identified superfamilies. Therefore, changes at the ultrastructural level of resistant strains may occur progressively and this is always preventable by prompt and reliable laboratory detection of drug-resistant TB cases.
Support statement
This study was sponsored by grant no: 0031/A12/2007 from the Mycobacteriology Research Centre/National Research Institute of Tuberculosis and Lung Disease/World Health Organization.
Statement of interest
None declared.
Acknowledgments
The authors wish to thank the staff of the Microscopy Unit at the Institute of Bioscience, University Putra Malaysia (Serdang, Malaysia). Also we sincerely thank the TB laboratory staff at the Mycobacteriology Research Centre/National Research Institute of Tuberculosis and Lung Disease (Tehran, Iran).
- © ERS Journals Ltd
