In 2019, the WHO tuberculosis (TB) treatment guidelines were updated to recommend only limited use of streptomycin, in favor of newer agents or amikacin as the preferred aminoglycoside for drug-resistant Mycobacterium tuberculosis. However, the emergence of resistance to newer drugs, such as bedaquiline, has prompted a reanalysis of antitubercular drugs in search of untapped potential. Using 211 clinical isolates of M. tuberculosis from South Africa, we performed phenotypic drug susceptibility testing (DST) to aminoglycosides by both critical concentration and MIC determination in parallel with whole-genome sequencing to identify known genotypic resistance elements.
KEYWORDS: Mycobacterium tuberculosis, aminoglycosides, drug resistance mechanisms, multidrug resistance, tuberculosis, whole-genome sequencing
ABSTRACT
In 2019, the WHO tuberculosis (TB) treatment guidelines were updated to recommend only limited use of streptomycin, in favor of newer agents or amikacin as the preferred aminoglycoside for drug-resistant Mycobacterium tuberculosis. However, the emergence of resistance to newer drugs, such as bedaquiline, has prompted a reanalysis of antitubercular drugs in search of untapped potential. Using 211 clinical isolates of M. tuberculosis from South Africa, we performed phenotypic drug susceptibility testing (DST) to aminoglycosides by both critical concentration and MIC determination in parallel with whole-genome sequencing to identify known genotypic resistance elements. Isolates with low-level streptomycin resistance mediated by gidB were frequently misclassified with respect to streptomycin resistance when using the WHO-recommended critical concentration of 2 μg/ml. We identified 29 M. tuberculosis isolates from South Africa with low-level streptomycin resistance concomitant with high-level amikacin resistance, conferred by gidB and rrs 1400, respectively. Using a large global data set of M. tuberculosis genomes, we observed 95 examples of this corresponding resistance genotype (gidB-rrs 1400), including identification in 81/257 (31.5%) of extensively drug resistant (XDR) isolates. In a phylogenetic analysis, we observed repeated evolution of low-level streptomycin and high-level amikacin resistance in multiple countries. Our findings suggest that current critical concentration methods and the design of molecular diagnostics need to be revisited to provide more accurate assessments of streptomycin resistance for gidB-containing isolates. For patients harboring isolates of M. tuberculosis with high-level amikacin resistance conferred by rrs 1400, and for whom newer agents are not available, treatment with streptomycin may still prove useful, even in the face of low-level resistance conferred by gidB.
INTRODUCTION
Despite recent advances, tuberculosis (TB) remains the number one infectious killer worldwide (1). The ongoing global epidemic of drug-resistant TB and limited effective treatment regimens for drug-resistant Mycobacterium tuberculosis have resulted in significant morbidity and mortality (1). Recognition of the inadequacy of the current antitubercular drug development pipeline, and the emergence of resistance to new drugs—including bedaquiline (2–9), delaminid (3, 4), clofazimine (5, 7), and linezolid (6)—has prompted a reanalysis of the existing arsenal of antitubercular drugs in search of untapped potential. Streptomycin may be one such underutilized drug.
Discovered in 1944, streptomycin, an injectable streptidine aminoglycoside antibiotic, was the first antimicrobial agent with proven activity against M. tuberculosis. In conjunction with isoniazid and para-aminosalicylic acid (PAS), streptomycin formed part of the first multidrug combination chemotherapy for TB, introduced in 1952. Its initial widespread use led to the early emergence of streptomycin resistance, which subsequently limited its clinical utility. Streptomycin remained an integral component of first-line TB therapy until the 1980s, and its empirical use in retreatment TB regimens was recommended until recently (10).
While the majority of the molecular determinants of aminoglycoside resistance are known, commercial diagnostic tests that assay for genotypic streptomycin resistance are lacking. Resistance to streptomycin does not contribute to the definition of extensively drug resistant (XDR) TB, which is defined as multidrug-resistant (MDR) isolates with additional resistance to quinolones and other injectable agents (amikacin, kanamycin) (11). Streptomycin is currently classified as a group C second-line agent for use in longer MDR-TB regimens (10), which are recommended in limited circumstances only. While XDR isolates are frequently cross-resistant to second-line injectable agents, there may be untapped potential for continued use of streptomycin for low-level resistance.
In our large collection of M. tuberculosis isolates from South Africa, we characterized aminoglycoside-resistance phenotypes in conjunction with whole-genome sequencing to identify patterns of aminoglycoside resistance. Subsequently, we used a global data set of over 5,000 M. tuberculosis genomes to assess the occurrence of genotypic low-level streptomycin resistance concomitant with high-level amikacin resistance worldwide.
RESULTS
Using 211 sequenced clinical isolates of M. tuberculosis from South Africa (Table S2 in the supplemental material), we performed critical concentration testing for streptomycin and kanamycin, and observed incomplete cross-resistance between these two aminoglycosides (Table S3). Amikacin critical concentration was not performed due to anticipated near complete cross-resistance with kanamycin (12), which was confirmed by our MIC testing (Fig. S1). Using genomic sequences for these 211 isolates, we sought known drug resistance markers for these aminoglycoside drugs. A total of 140 isolates were found to have genotypic markers of streptomycin resistance, with mutations in rpsL, rrs (non-1400), and gidB, whereas 50 isolates had mutations in rrs 1400, which confers high-level resistance to both amikacin and kanamycin (Table 1). Two isolates contained mutations in the promoter region of eis, which confers resistance to kanamycin, but not to streptomycin or amikacin. Co-occurrence of streptomycin and amikacin/kanamycin resistance genotypes was determined (Table 2), including identification of seven isolates with more than one streptomycin-resistance-determining mutation (Table S4).
TABLE 1.
Distribution of resistance-associated mutations in a South African data seta
| Drug | Gene | Polymorphismb | Number of isolates |
|---|---|---|---|
| Streptomycin | rpsL | K43R | 31 |
| K88R | 8 | ||
| rrs (non-1400) | 513 | 7 | |
| 516 | 6 | ||
| 907 | 1 | ||
| gidB | nt 62, del 1 bp | 1 | |
| nt 103, del 1 bp | 1 | ||
| nt 108, del 1 bp | 3 | ||
| nt 116, del 1 bp | 3 | ||
| nt 282, del 130 bp | 78 | ||
| nt 368, del 2 bp | 1 | ||
| A134E | 3 | ||
| A138V | 2 | ||
| A141E | 2 | ||
| Kanamycin/amikacin | rrs | 1400 | 50 |
| Kanamycin | eis | promoter −14 | 2 |
Of the 211 South African isolates, 140 were found to have genotypic streptomycin resistance with mutations in rpsL, rrs (non-1400), and gidB, as detailed in the table. Fifty strains were found to have genotypic amikacin/kanamycin resistance with mutations in rrs 1400, and two isolates had kanamycin resistance with an eis promoter mutation.
nt, nucleotide; del, deletion; bp, base pairs.
TABLE 2.
Distribution of co-occurring genotypic resistances to streptomycin and amikacin/kanamycin in a South African data seta
| Streptomycin genotype | Amikacin/kanamycin genotype |
|||
|---|---|---|---|---|
| WT | rrs 1400 | eis promoter | Total | |
| WT | 67 | 2 | 2 | 71 |
| rpsL | 33 | 2 | 0 | 35 |
| rrs (non-1400) | 8 | 5 | 0 | 13 |
| gidB | 44 | 41b | 0 | 85 |
| Total | 152 | 50 | 2 | 204 |
Of note, 7 isolates were identified to contain more than one streptomycin resistance mutation, as described in Table S4 in the supplemental material.
The boldface type indicates the number of isolates with co-occurrence of gidB and rrs 1400 mutations that confer low-level streptomycin and high-level amikacin resistance.
In comparing MIC data from Sensititre testing with known aminoglycoside resistance genotypes, we evaluated the relationship between genotypic and phenotypic resistance to streptomycin (Fig. 1A), amikacin (Fig. 1B), and kanamycin (Fig. S1). There was a bell-shaped distribution (Fig. 1A) of isolates containing gidB mutations with low-level streptomycin resistance (median MIC 4 μg/ml; interquartile range [IQR], 2 to 4 μg/ml) (Table 3). By critical concentration testing per the WHO-recommended guidelines, the majority of isolates with gidB mutations (76%, 70/92) were classified as resistant to streptomycin. In contrast, high-level streptomycin resistance was observed in isolates with either rrs (non-1400) or rpsL mutations, with median MIC 32 μg/ml (IQR, 16 to 32 μg/ml) and 32 μg/ml (IQR, 16 to 32 μg/ml), respectively. Three isolates with no identifiable streptomycin mutations were noted to have high MICs to streptomycin (MIC 16 to 32 μg/ml), suggesting that additional streptomycin resistance mutations remain to be discovered, but this could also be due to errors in phenotyping. Nearly every isolate with high-level amikacin and kanamycin resistance contained an rrs 1400 mutation (Fig. 1B, Fig. S2).
FIG 1.
Clinical strains of M. tuberculosis were observed to have a range of susceptibility to aminoglycosides, mediated by resistance genotype. A total of 211 isolates of M. tuberculosis from South Africa underwent MIC determination and aminoglycoside resistance genotyping to identify mutations that confer resistance to streptomycin or amikacin, respectively. (A) Streptomycin MIC testing revealed a bell-shaped curve distribution of gidB strains with low-level streptomycin resistance, whereas strains containing rrs (non-1400) or rpsL mutations had higher resistance. Of note, three isolates containing resistance elements in both gidB and rpsL were included among the rpsL isolates. (B) Amikacin MIC testing revealed high-level resistance among strains containing the rrs 1400 mutation. Kanamycin MIC results mirrored that of amikacin (Fig. S1 in the supplemental material).
TABLE 3.
gidB mutations confer low-level streptomycin resistance, whereas rrs and rpsL mutations confer high-level resistance
| Streptomycin genotype | Median MIC to streptomycin in μg/ml (IQR)a |
|---|---|
| WT | 1 (0.5–2) |
| gidB | 4 (2–4) |
| rrs (non-1400) | 32 (16–32) |
| rpsL | 32 (32–32) |
For each streptomycin resistance genotype, median MIC to streptomycin and interquartile range (IQR) is listed.
When comparing the MIC of each isolate to streptomycin and amikacin, numerous isolates had mismatched phenotypes, indicating that resistance to amikacin did not confer resistance to streptomycin, and vice versa (Fig. 2). In particular, 29 isolates from South Africa exhibited low-level streptomycin resistance (MIC 4 μg/ml or 8 μg/ml) and concomitant high-level amikacin resistance (MIC ≥16 μg/ml) (circled area, Fig. 2). These findings suggest that use of streptomycin instead of amikacin would be the preferred aminoglycoside for treatment of these isolates. The vast majority of isolates with this phenotype (93%, 27/29) contained a gidB resistance genotype, and 100% (29/29) contained an rrs 1400 mutation.
FIG 2.
Significant numbers of M. tuberculosis isolates exhibit concomitant low-level streptomycin resistance and high-level amikacin resistance. Isolates are represented by streptomycin genotype (see key) and plotted as a function of the relative phenotypic resistance to both amikacin and streptomycin. The red dotted circle indicates the 29 isolates with concomitant low-level streptomycin resistance and high-level amikacin resistance.
From the genomic data, we constructed a phylogeny to determine the interrelatedness of isolates with (i) low-level streptomycin resistance and (ii) concomitant low-level streptomycin and high-level amikacin resistance in phenotypic testing (Fig. 3). The 57 isolates with low-level streptomycin resistance were distributed throughout the phylogeny. The majority (25/29, 86%) of South African isolates with low-level streptomycin and high-level amikacin resistance belonged to the Tugela Ferry XDR clone, which was responsible for epidemic XDR in the region in the early 2000s (13). However, there were four isolates outside this cluster, indicating that this phenomenon was not unique to this clone.
FIG 3.
Concomitant low-level streptomycin and high-level amikacin phenotypic resistance in South African M. tuberculosis isolates across the phylogeny. Midpoint rooted maximum-likelihood phylogeny of 211 M. tuberculosis isolates, containing representatives of four of the seven known M. tuberculosis lineages. Phenotypic MDR and XDR are indicated by black and white boxes at the tip of each leaf node. The levels of phenotypic resistance to streptomycin (low, MIC 4 to 8 μg/ml; high, MIC ≥16 μg/ml) and amikacin (low, MIC 4 to 8 μg/ml; high, MIC ≥16 μg/ml) are indicated by box color, per the key. Strains with concomitant low-level streptomycin resistance and high-level amikacin resistance are indicated in red. While the majority of isolates with low-level streptomycin resistance and high-level amikacin resistance pertained to the Tugela Ferry XDR outbreak clone, four examples were observed outside the outbreak clone, indicating that this was not an isolated evolutionary event.
To determine whether the phenomenon of low-level streptomycin and high-level amikacin resistance occurred outside South Africa, we analyzed our large data set of 5,310 M. tuberculosis isolates from 43 countries (14). Within this data set, 257 isolates contained mutations for resistance to all four drugs that define XDR (rifampin, isoniazid, ofloxacin, and amikacin). As phenotypic data were not available for this data set, we used co-occurrence of a gidB resistance mutation and rrs 1400 mutation as a genotypic predictor of this combination of low-level streptomycin resistance and high-level amikacin resistance (Fig. 4). We identified 378 unique isolates with gidB mutations, including 95 isolates with co-occurrence of gidB mutations and rrs 1400 mutation (Table S5). All 95 isolates contained resistance-conferring mutations to both isoniazid and rifampin (MDR genotype) in addition to resistance to either ofloxacin or kanamycin (pre-XDR), and 81/95 of these isolates were XDR. Of the 257 XDR isolates in the 5,310-isolate data set, 81 (31.5%) of the XDR isolates contained this gidB-rrs 1400 combination, indicating frequent occurrence in global XDR-TB. The majority of isolates with the gidB-rrs 1400 pattern were LAM4 and likely members of the Tugela Ferry XDR clade. However, there were nine other spoligotypes with isolates containing this pattern, indicating multiple independent evolutionary events. Beyond South Africa, isolates with this resistance pattern were also identified in Belarus, China, Iran, Portugal, Romania, South Korea, and Sweden, indicating that this phenomenon of streptomycin-low and amikacin-high resistance is of global importance for management of drug-resistant TB.
FIG 4.
Concomitant low-level streptomycin and high-level amikacin genotypic resistance evolved repeatedly in a global data set of M. tuberculosis. Midpoint rooted maximum-likelihood phylogeny of 5,310 M. tuberculosis strains from a global data set containing representatives of all seven known M. tuberculosis lineages. The presence and levels of genotypic resistance to streptomycin (low, gidB; high, rrs [non-1400] and rpsL) and amikacin (high, rrs 1400) are indicated by box color near the leaf nodes. Ninety-five isolates with genotypic mutations predicted to confer both low-level streptomycin resistance and high-level amikacin resistance (gidB-rrs 1400) are indicated in red. Concomitant low-level streptomycin resistance and high-level amikacin resistance occurred across the phylogeny, indicating that this phenomenon is of global relevance for TB control.
DISCUSSION
In both a South African and a global data set, significant numbers of M. tuberculosis isolates contained mutations associated with concomitant low-level streptomycin resistance and high-level amikacin resistance. Current guidelines that recommend only limited use of streptomycin (10) may be unwittingly withholding a potentially lifesaving, inexpensive, and available drug from certain patients with drug-resistant TB. Similarly, current WHO-endorsed laboratory procedures for performing phenotypic DST to streptomycin by critical concentration may obscure the potential utility of streptomycin by not distinguishing between high and low-level resistance.
Given additional newer agents with excellent activity against drug-resistant TB, such as bedaquiline, the updated 2019 WHO guidelines limit use of aminoglycosides (10). Kanamycin is no longer recommended in the treatment of drug-resistant TB patients on longer regimens. Amikacin is now the preferred aminoglycoside, and its use is limited to adults on longer regimens in situations in which DST results confirm susceptibility and for whom high-quality audiometry testing for hearing loss can be performed. Streptomycin use is recommended only when amikacin is not available, and again in situations when DST results confirm susceptibility and in whom safety monitoring can be ensured.
While treatment-related ototoxicity and nephrotoxicity are well established, streptomycin could still hold therapeutic potential for individuals with drug-resistant TB harboring isolates with low-level streptomycin resistance. If an aminoglycoside is being considered for inclusion in a drug-resistant TB regimen, if the rrs 1400 mutation is present, which confers high-level resistance to amikacin, than we recommend selection of streptomycin, even in the face of low-level resistance, such as that conferred by gidB. To our knowledge, clinical outcomes for individuals harboring isolates with low-level streptomycin resistance mediated by gidB and treated with a streptomycin-containing regimen have not been assessed. An expanded role for streptomycin in drug-resistant TB may also increase risk of adverse events related to drug toxicity. Ensuring safety of a streptomycin-based regimen would necessitate implementation of monitoring procedures, including audiometry and measurements of renal function, which constitute an additional burden—especially for resource-limited settings.
In South Africa, due to a clonal outbreak of XDR-TB in Tugela Ferry, a large fraction of circulating XDR-TB isolates contain an 130-bp deletion in gidB that confers low-level streptomycin resistance and an rrs 1400 mutation that confers high-level cross-resistance to amikacin, kanamycin, and capreomycin (15, 16). In a recent long-term cohort study of XDR-TB treatment outcomes in South Africa, only 1% of patients were treated with streptomycin, whereas 98% received capreomycin (17). As treatment outcomes for XDR-TB in South Africa were notoriously abysmal (17), including streptomycin may prove useful for patients in whom new drugs are not available because of resistance or contraindications.
Current WHO-endorsed laboratory procedures for performing phenotypic DST to streptomycin by critical concentration fail to provide key information relevant to streptomycin inclusion in a regimen for drug-resistant TB. The MIC distribution for isolates containing gidB mutations straddles the WHO-recommended critical concentration of 2 μg/ml (Fig. 1A). This modest increase in MIC among isolates containing gidB mutations in comparison to wild-type isolates likely contributes to inconsistencies in testing. Isolates containing gidB mutations are frequently misclassified in terms of their susceptibility to streptomycin on critical concentration testing (as occurred in 24% of isolates in this study). As critical concentration testing is typically performed only at a single concentration, isolates with low-level streptomycin resistance—which may potentially be treated successfully with streptomycin—cannot be distinguished from those with high-level resistance. Similarly, wild-type strains that do not contain genotypes predicted to confer resistance to streptomycin can exhibit low-level streptomycin resistance that is above the critical concentration threshold (as seen in four South African isolates in this study), which may result in withholding a potentially useful drug.
The WHO-recommended critical concentration for streptomycin in M. tuberculosis is based on weak scientific evidence (12). The upper limit of wild-type MIC distribution, termed the epidemiological cutoff value (ECOFF), for streptomycin is 2 μg/ml (18). That this is the same value as the critical concentration in DST reflects the lack of clinical and pharmacokinetic/pharmacodynamic data to inform a more practical selection of a critical concentration. Potential strategies to address this issue include: (i) raising the streptomycin critical concentration; (ii) adding a second streptomycin drug concentration to traditional critical concentration testing (e.g., test at both 2 μg/ml and 8 μg/ml to disambiguate between low-level and high-level streptomycin resistance); (iii) performing additional reflex testing when an isolate is identified by traditional critical concentration DST to be resistant to both streptomycin and kanamycin (e.g., more detailed phenotypic analysis or streptomycin resistance genotype determination); or (iv) forgoing critical concentration testing in all forms and instead expanding genotypic aminoglycoside resistance testing.
Recent efforts to expand the complement of drug resistance mutation panels included on rapid molecular TB diagnostics have not included streptomycin (19). Whole-genome sequencing (WGS) studies of clinical isolates of M. tuberculosis have demonstrated that the majority (92% to 95%) of streptomycin-resistant isolates can be explained by known mutations (20, 21). Thus, omitting streptomycin resistance determinants from rapid drug resistance panels is a missed opportunity to both identify and grade streptomycin resistance relative to amikacin resistance. One potential reason for this exclusion is mutations in gidB can occur anywhere in the gene, where they cause frameshift, nonsense, or deletion mutations. Thus, they are difficult to identify with current SNP-based diagnostics and instead require whole-gene-based strategies, such as high-resolution melt analysis (22) or rapid WGS.
It is important to address several limitations of this study. MIC determination was performed with Sensititre, which is not the gold standard for M. tuberculosis DST. However, prior investigation comparing Sensititre with traditional methods have shown excellent concordance for aminoglycoside testing (23, 24). In addition, phenotyped isolates derived only from South Africa, and the population structure contained clonal XDR isolates from the Tugela Ferry epidemic. However, the phenomenon of genotypic resistance conferring low-level streptomycin and high-level amikacin resistance was also seen outside this clone. Thus, this observation carries implications for M. tuberculosis treatment in other settings.
Our findings suggest that current critical concentration methods for streptomycin resistance determination and the design of molecular diagnostics for resistance may need to be revisited for improved categorization of isolates harboring gidB mutations, which confer low-level streptomycin resistance. In the context of limited therapeutic options for drug-resistant M. tuberculosis, our results show the potential utility of streptomycin, even for isolates observed to have low-level resistance from gidB mutations.
MATERIALS AND METHODS
Clinical isolates.
We selected for inclusion a random subset of 211 clinical isolates of susceptible and drug-resistant M. tuberculosis from South Africa from our larger sequenced strain set (15).
Drug susceptibility testing by critical concentration.
As previously described (15), DST was performed prospectively by critical concentration on Middlebrook 7H11 using the WHO-recommended drug concentrations for streptomycin (2.0 μg/ml) and kanamycin (6.0 μg/ml). Amikacin critical concentration was not performed, as isolates with acquired resistance to amikacin essentially always have resistance to kanamycin (12).
MIC determination.
MIC determination for three aminoglycosides (amikacin, kanamycin, and streptomycin) was performed using MycoTB Sensititre plates (TREK Diagnostic Systems), per the manufacturer’s instructions. The lowest concentration of drug that did not show visible growth was recorded as the MIC to the respective drug.
Whole-genome sequencing and analysis.
Whole-genome sequencing (WGS) and analysis were performed as previously described (15). Genotypic resistance to streptomycin, amikacin, and kanamycin was defined as identification of polymorphisms that are known to be associated with drug resistance, per the refined genotypic resistance definition in Desjardins and Cohen et al. (20) (Table S1). Isolates belonging to the Tugela Ferry XDR clone were identified by phylogenetic clustering with the reference isolates KZN605, collected during the epidemic, as well as the presence of canonical drug-resistance mutations (15). SNP calls from Cohen et al. were used (15). RAxML version 7.3.0 (25) was used to construct a phylogenetic tree from concatenated SNPs, with 1,000 bootstrap replicates.
Data availability. All sequencing data can be found in the Sequence Read Archive NCBI umbrella project identifier PRJNA183624.
Supplementary Material
ACKNOWLEDGMENTS
We acknowledge the support of the Africa Health Research Institute (AHRI) to conduct the research. We thank Nonkqubela Bantubani of the Medical Research Council (MRC) in Durban, in addition to Max O′Donnell and Nesri Padayatchi for contributing clinical isolates of M. tuberculosis, which were used in this study. We thank Koleka P. Mlisana and Nomonde R. Mvelase of the KwaZulu-Natal National Health Laboratory Service.
This work was supported by funding from the National Heart Lung and Blood Institute, National Institutes of Health (K08 HL139994), and a Burroughs Wellcome Fund Career Award for Medical Scientists to K.A.C. K.E.S. is a Wellcome Trust Clinical Ph.D Fellow (203919/Z/16/Z). This project has been supported in part with federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, under contract number HHSN272200900018C and grant number U19AI110818 to the Broad Institute.
None of the authors have a commercial or other association that may pose a conflict of interest.
K.A.C., K.E.S., and A.S.P. conceived of the study and designed the experiments. K.A.C. and V.M. performed the wet-lab experiments. K.A.C., K.E.S., and A.L.M. analyzed the data. K.A.C. and K.E.S. wrote the manuscript. A.M.E. and A.S.P. supervised and coordinated the project. All authors have read the manuscript and confirm that they meet ICMJE criteria for authorship.
Footnotes
Supplemental material is available online only.
REFERENCES
- 1.World Health Organization. 2019. Global tuberculosis report 2019. World Health Organization, Geneva, Switzerland. [Google Scholar]
- 2.Andries K, Villellas C, Coeck N, Thys K, Gevers T, Vranckx L, Lounis N, de Jong BC, Koul A. 2014. Acquired resistance of Mycobacterium tuberculosis to bedaquiline. PLoS One 9:e102135. doi: 10.1371/journal.pone.0102135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bloemberg GV, Keller PM, Stucki D, Trauner A, Borrell S, Latshang T, Coscolla M, Rothe T, Hömke R, Ritter C, Feldmann J, Schulthess B, Gagneux S, Böttger EC. 2015. Acquired resistance to bedaquiline and delamanid in therapy for tuberculosis. N Engl J Med 373:1986–1988. doi: 10.1056/NEJMc1505196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hoffmann H, Kohl TA, Hofmann-Thiel S, Merker M, Beckert P, Jaton K, Nedialkova L, Sahalchyk E, Rothe T, Keller PM, Niemann S. 2016. Delamanid and bedaquiline resistance in Mycobacterium tuberculosis ancestral Beijing genotype causing extensively drug-resistant tuberculosis in a Tibetan Refugee. Am J Respir Crit Care Med 193:337–340. doi: 10.1164/rccm.201502-0372LE. [DOI] [PubMed] [Google Scholar]
- 5.Xu J, Wang B, Hu M, Huo F, Guo S, Jing W, Nuermberger E, Lu Y. 2017. Primary clofazimine and bedaquiline resistance among isolates from patients with multidrug-resistant tuberculosis. Antimicrob Agents Chemother 61:e00239-17. doi: 10.1128/AAC.00239-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zimenkov DV, Nosova EY, Kulagina EV, Antonova OV, Arslanbaeva LR, Isakova AI, Krylova LY, Peretokina IV, Makarova MV, Safonova SG, Borisov SE, Gryadunov DA. 2017. Examination of bedaquiline- and linezolid-resistant Mycobacterium tuberculosis isolates from the Moscow region. J Antimicrob Chemother 72:1901–1906. doi: 10.1093/jac/dkx094. [DOI] [PubMed] [Google Scholar]
- 7.Villellas C, Coeck N, Meehan CJ, Lounis N, de Jong B, Rigouts L, Andries K. 2016. Unexpected high prevalence of resistance-associated Rv0678 variants in MDR-TB patients without documented prior use of clofazimine or bedaquiline. J Antimicrob Chemother 72:684–690. doi: 10.1093/jac/dkw502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Veziris N, Bernard C, Guglielmetti L, Le Du D, Marigot-Outtandy D, Jaspard M, Caumes E, Lerat I, Rioux C, Yazdanpanah Y, Tiotiu A, Lemaitre N, Brossier F, Jarlier V, Robert J, Sougakoff W, Aubry A. 2017. Rapid emergence of Mycobacterium tuberculosis bedaquiline resistance: lessons to avoid repeating past errors. Eur Respir J 49:1601719. doi: 10.1183/13993003.01719-2016. [DOI] [PubMed] [Google Scholar]
- 9.Martinez E, Hennessy D, Jelfs P, Crighton T, Chen S-A, Sintchenko V. 2018. Mutations associated with in vitro resistance to bedaquiline in Mycobacterium tuberculosis isolates in Australia. Tuberculosis (Edinb) 111:31–34. doi: 10.1016/j.tube.2018.04.007. [DOI] [PubMed] [Google Scholar]
- 10.World Health Organization. 2019. WHO consolidated guidelines on drug-resistant tuberculosis. World Health Organization, Geneva, Switzerland. [PubMed] [Google Scholar]
- 11.Jassal M, Bishai WR. 2009. Extensively drug-resistant tuberculosis. Lancet Infect Dis 9:19–30. doi: 10.1016/S1473-3099(08)70260-3. [DOI] [PubMed] [Google Scholar]
- 12.World Health Organization. 2008. Policy guidance on drug-susceptibility testing (DST) of second-line antituberculosis drugs. World Health Organization, Geneva, Switzerland. [PubMed] [Google Scholar]
- 13.Gandhi NR, Moll A, Sturm AW, Pawinski R, Govender T, Lalloo U, Zeller K, Andrews J, Friedland G. 2006. Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa. Lancet 368:1575–1580. doi: 10.1016/S0140-6736(06)69573-1. [DOI] [PubMed] [Google Scholar]
- 14.Manson AL, Cohen KA, Abeel T, Desjardins CA, Armstrong DT, Barry CE, Brand J, Chapman SB, Cho S-N, Gabrielian A, Gomez J, Jodals AM, Joloba M, Jureen P, Lee JS, Malinga L, Maiga M, Nordenberg D, Noroc E, Romancenco E, Salazar A, Ssengooba W, Velayati AA, Winglee K, Zalutskaya A, Via LE, Cassell GH, Dorman SE, Ellner J, Farnia P, Galagan JE, Rosenthal A, Crudu V, Homorodean D, Hsueh P-R, Narayanan S, Pym AS, Skrahina A, Swaminathan S, Van der Walt M, Alland D, Bishai WR, Cohen T, Hoffner S, Birren BW, Earl AM, TBResist Global Genome Consortium . 2017. Genomic analysis of globally diverse Mycobacterium tuberculosis strains provides insights into the emergence and spread of multidrug resistance. Nat Genet 49:395–402. doi: 10.1038/ng.3767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cohen KA, Abeel T, Manson McGuire A, Desjardins CA, Munsamy V, Shea TP, Walker BJ, Bantubani N, Almeida DV, Alvarado L, Chapman SB, Mvelase NR, Duffy EY, Fitzgerald MG, Govender P, Gujja S, Hamilton S, Howarth C, Larimer JD, Maharaj K, Pearson MD, Priest ME, Zeng Q, Padayatchi N, Grosset J, Young SK, Wortman J, Mlisana KP, O'Donnell MR, Birren BW, Bishai WR, Pym AS, Earl AM. 2015. Evolution of extensively drug-resistant tuberculosis over four decades: whole genome sequencing and dating analysis of Mycobacterium tuberculosis isolates from KwaZulu-Natal. PLoS Med 12:e1001880. doi: 10.1371/journal.pmed.1001880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Shah NS, Auld SC, Brust JCM, Mathema B, Ismail N, Moodley P, Mlisana K, Allana S, Campbell A, Mthiyane T, Morris N, Mpangase P, van der Meulen H, Omar SV, Brown TS, Narechania A, Shaskina E, Kapwata T, Kreiswirth B, Gandhi NR. 2017. Transmission of extensively drug-resistant tuberculosis in South Africa. N Engl J Med 376:243–253. doi: 10.1056/NEJMoa1604544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Pietersen E, Ignatius E, Streicher EM, Mastrapa B, Padanilam X, Pooran A, Badri M, Lesosky M, van Helden P, Sirgel F, Warren R, Dheda K. 2014. Long-term outcomes of patients with extensively drug-resistant tuberculosis in South Africa: a cohort study. Lancet 383:1230–1239. doi: 10.1016/S0140-6736(13)62675-6. [DOI] [PubMed] [Google Scholar]
- 18.Juréen P, Ängeby K, Sturegård E, Chryssanthou E, Giske CG, Werngren J, Nordvall M, Johansson A, Kahlmeter G, Hoffner S, Schön T. 2010. Wild-type MIC distributions for aminoglycoside and cyclic polypeptide antibiotics used for treatment of Mycobacterium tuberculosis infections. J Clin Microbiol 48:1853–1858. doi: 10.1128/JCM.00240-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Xie YL, Chakravorty S, Armstrong DT, Hall SL, Via LE, Song T, Yuan X, Mo X, Zhu H, Xu P, Gao Q, Lee M, Lee J, Smith LE, Chen RY, Joh JS, Cho Y, Liu X, Ruan X, Liang L, Dharan N, Cho S-N, Barry CE, Ellner JJ, Dorman SE, Alland D. 2017. Evaluation of a rapid molecular drug-susceptibility test for tuberculosis. N Engl J Med 377:1043–1054. doi: 10.1056/NEJMoa1614915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Desjardins CA, Cohen KA, Munsamy V, Abeel T, Maharaj K, Walker BJ, Shea TP, Almeida DV, Manson AL, Salazar A, Padayatchi N, O'Donnell MR, Mlisana KP, Wortman J, Birren BW, Grosset J, Earl AM, Pym AS. 2016. Genomic and functional analyses of Mycobacterium tuberculosis strains implicate ald in D-cycloserine resistance. Nat Genet 48:544–551. doi: 10.1038/ng.3548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Walker TM, Kohl TA, Omar SV, Hedge J, Del Ojo Elias C, Bradley P, Iqbal Z, Feuerriegel S, Niehaus KE, Wilson DJ, Clifton DA, Kapatai G, Ip CLC, Bowden R, Drobniewski FA, Allix-Béguec C, Gaudin C, Parkhill J, Diel R, Supply P, Crook DW, Smith EG, Walker AS, Ismail N, Niemann S, Peto T. a. 2015. Whole-genome sequencing for prediction of Mycobacterium tuberculosis drug susceptibility and resistance: a retrospective cohort study. Lancet Infect Dis 15:1193–1202. doi: 10.1016/S1473-3099(15)00062-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pholwat S, Liu J, Stroup S, Gratz J, Banu S, Rahman SMM, Ferdous SS, Foongladda S, Boonlert D, Ogarkov O, Zhdanova S, Kibiki G, Heysell S, Houpt E. 2015. Integrated microfluidic card with TaqMan probes and high-resolution melt analysis to detect tuberculosis drug resistance mutations across 10 genes. mBio 6:e02273-14. doi: 10.1128/mBio.02273-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hall L, Jude KP, Clark SL, Dionne K, Merson R, Boyer A, Parrish NM, Wengenack NL. 2012. Evaluation of the Sensititre MycoTB plate for susceptibility testing of the Mycobacterium tuberculosis complex against first- and second-line agents. J Clin Microbiol 50:3732–3734. doi: 10.1128/JCM.02048-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lee JS, Armstrong DT, Ssengooba W, Park J-A, Yu Y, Mumbowa F, Namaganda C, Mboowa G, Nakayita G, Armakovitch S, Chien G, Cho S, Via LE, Barry CE, Ellner JJ, Alland D, Dorman SE, Joloba ML. 2013. The Sensititre MYCOTB MIC plate for testing Mycobacterium tuberculosis susceptibility to first- and second-line drugs. Antimicrob Agents Chemother 58:11-8. doi: 10.1128/AAC.01209-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Stamatakis A. 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22:2688–2690. doi: 10.1093/bioinformatics/btl446. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.