Loss-of-function mutations in ndh do not confer delamanid, ethionamide, isoniazid, or pretomanid resistance in Mycobacterium tuberculosis

Results from clinical strains and knockouts of the H37Rv and CDC1551 laboratory strains demonstrated that ndh (Rv1854c) is not a resistance-conferring gene for isoniazid, ethionamide, delamanid, or pretomanid in Mycobacterium tuberculosis. This difference in the susceptibility to NAD-adduct-forming drugs compared with other mycobacteria may be driven by differences in the absolute intrabacterial NADH concentration.

Evolution and spread of a highly drug resistant strain of Mycobacterium tuberculosis in Papua New Guinea

BACKGROUND

Molecular mechanisms determining the transmission and prevalence of drug resistant tuberculosis (DR-TB) in Papua New Guinea (PNG) are poorly understood. We used genomic and drug susceptibility data to explore the evolutionary history, temporal acquisition of resistance and transmission dynamics of DR-TB across PNG.

METHODS

We performed whole genome sequencing on isolates from Central Public Health Laboratory, PNG, collected 2017-2019. Data analysis was done on a composite dataset that also included 100 genomes previously sequenced from Daru, PNG (2012-2015).

RESULTS

Sampled isolates represented 14 of the 22 PNG provinces, the majority (66/94; 70%) came from the National Capital District (NCD). In the composite dataset, 91% of strains were Beijing 2.2.1.1, identified in 13 provinces. Phylogenetic tree of Beijing strains revealed two clades, Daru dominant clade (A) and NCD dominant clade (B). Multi-drug resistance (MDR) was repeatedly and independently acquired, with the first MDR cases in both clades noted to have emerged in the early 1990s, while fluoroquinolone resistance emerged in 2009 (95% highest posterior density 2000-2016). We identified the presence of a frameshift mutation within Rv0678 (p.Asp47fs) which has been suggested to confer resistance to bedaquiline, despite no known exposure to the drug. Overall genomic clustering was significantly associated with rpoC compensatory and inhA promoter mutations (p < 0.001), with high percentage of most genomic clusters (12/14) identified in NCD, reflecting its role as a potential national amplifier.

CONCLUSIONS

The acquisition and evolution of drug resistance among the major clades of Beijing strain threaten the success of DR-TB treatment in PNG. With continued transmission of this strain in PNG, genotypic drug resistance surveillance using whole genome sequencing is essential for improved public health response to outbreaks. With occurrence of resistance to newer drugs such as bedaquiline, knowledge of full drug resistance profiles will be important for optimal treatment selection.

Computational analysis and prediction of PE_PGRS proteins using machine learning

•

PEPPER is the first machine learning-based predictor for PE_PGRS proteins.

•

PEPPER is based on lightGBM and various sequence and physicochemical features.

•

PEPPER can identify PE_PGRS proteins rapidly and accurately.

•

The webserver of PEPPER and stand-alone tool are publicly available at http://web.unimelb-bioinfortools.cloud.edu.au/PEPPER/.

Mycobacterium tuberculosis genome comprises approximately 10% of two families of poorly characterised genes due to their high GC content and highly repetitive nature. The largest sub-group, the proline-glutamic acid polymorphic guanine-cytosine-rich sequence (PE_PGRS) family, is thought to be involved in host response and disease pathogenicity. Due to their high genetic variability and complexity of analysis, they are typically disregarded for further research in genomic studies. There are currently limited online resources and homology computational tools that can identify and analyse PE_PGRS proteins. In addition, they are computational-intensive and time-consuming, and lack sensitivity. Therefore, computational methods that can rapidly and accurately identify PE_PGRS proteins are valuable to facilitate the functional elucidation of the PE_PGRS family proteins. In this study, we developed the first machine learning-based bioinformatics approach, termed PEPPER, to allow users to identify PE_PGRS proteins rapidly and accurately. PEPPER was built upon a comprehensive evaluation of 13 popular machine learning algorithms with various sequence and physicochemical features. Empirical studies demonstrated that PEPPER achieved significantly better performance than alignment-based approaches, BLASTP and PHMMER, in both prediction accuracy and speed. PEPPER is anticipated to facilitate community-wide efforts to conduct high-throughput identification and analysis of PE_PGRS proteins.

Genomic epidemiology of tuberculosis in eastern Malaysia: insights for strengthening public health responses

Tuberculosis is a leading public health priority in eastern Malaysia. Knowledge of the genomic epidemiology of tuberculosis can help tailor public health interventions. Our aims were to determine tuberculosis genomic epidemiology and characterize resistance mutations in the ethnically diverse city of Kota Kinabalu, Sabah, located at the nexus of Malaysia, Indonesia, Philippines and Brunei. We used an archive of prospectively collected Mycobacterium tuberculosis samples paired with epidemiological data. We collected sputum and demographic data from consecutive consenting outpatients with pulmonary tuberculosis at the largest tuberculosis clinic from 2012 to 2014, and selected samples from tuberculosis inpatients from the tertiary referral centre during 2012-2014 and 2016-2017. Two hundred and eight M. tuberculosis sequences were available for analysis, representing 8 % of cases notified during the study periods. Whole-genome phylogenetic analysis demonstrated that most strains were lineage 1 (195/208, 93.8 %), with the remainder being lineages 2 (8/208, 3.8 %) or 4 (5/208, 2.4 %). Lineages or sub-lineages were not associated with patient ethnicity. The lineage 1 strains were diverse, with sub-lineage 1.2.1 being dominant (192, 98 %). Lineage 1.2.1.3 isolates were geographically most widely distributed. The greatest diversity occurred in a border town sub-district. The time to the most recent common ancestor for the three major lineage 1.2.1 clades was estimated to be the year 1966 (95 % HPD 1948-1976). An association was found between failure of culture conversion by week 8 of treatment and infection with lineage 2 (4/6, 67 %) compared with lineage 1 strains (4/83, 5 %) (P<0.001), supporting evidence of greater virulence of lineage 2 strains. Eleven potential transmission clusters (SNP difference ≤12) were identified; at least five included people living in different sub-districts. Some linked cases spanned the whole 4-year study period. One cluster involved a multidrug-resistant tuberculosis strain matching a drug-susceptible strain from 3 years earlier. Drug resistance mutations were uncommon, but revealed one phenotype-genotype mismatch in a genotypically multidrug-resistant isolate, and rare nonsense mutations within the katG gene in two isolates. Consistent with the regionally mobile population, M. tuberculosis strains in Kota Kinabalu were diverse, although several lineage 1 strains dominated and were locally well established. Transmission clusters - uncommonly identified, likely attributable to incomplete sampling - showed clustering occurring across the community, not confined to households or sub-districts. The findings indicate that public health priorities should include active case finding and early institution of tuberculosis management in mobile populations, while there is a need to upscale effective contact investigation beyond households to include other contacts within social networks.

Cross-Border Movement of Highly Drug-Resistant Mycobacterium tuberculosis from Papua New Guinea to Australia through Torres Strait Protected Zone, 2010-2015

In this retrospective study, we used whole-genome sequencing (WGS) to delineate transmission dynamics, characterize drug-resistance markers, and identify risk factors of transmission among Papua New Guinea residents of the Torres Strait Protected Zone (TSPZ) who had tuberculosis diagnoses during 2010-2015. Of 117 isolates collected, we could acquire WGS data for 100; 79 were Beijing sublineage 2.2.1.1, which was associated with active transmission (odds ratio 6.190, 95% CI 2.221-18.077). Strains were distributed widely throughout the TSPZ. Clustering occurred more often within than between villages (p = 0.0013). Including 4 multidrug-resistant tuberculosis isolates from Australia citizens epidemiologically linked to the TSPZ into the transmission network analysis revealed 2 probable cross-border transmission events. All multidrug-resistant isolates (33/104) belonged to Beijing sublineage 2.2.1.1 and had high-level isoniazid and ethionamide co-resistance; 2 isolates were extensively drug resistant. Including WGS in regional surveillance could improve tuberculosis transmission tracking and control strategies within the TSPZ.

Rapid diagnosis of Capnocytophaga canimorsus septic shock in an immunocompetent individual using real-time Nanopore sequencing: a case report

BACKGROUND

Rapid diagnosis and appropriate treatment is imperative in bacterial sepsis due increasing risk of mortality with every hour without appropriate antibiotic therapy. Atypical infections with fastidious organisms may take more than 4 days to diagnose leading to calls for improved methods for rapidly diagnosing sepsis. Capnocytophaga canimorsus is a slow-growing, fastidious gram-negative bacillus which is a common commensal within the mouths of dogs, but rarely cause infections in humans. C. canimorsus sepsis risk factors include immunosuppression, alcoholism and elderly age. Here we report on the application of emerging nanopore sequencing methods to rapidly diagnose an atypical case of C. canimorsus septic shock.

CASE PRESENTATION

A 62 year-old female patient was admitted to an intensive care unit with septic shock and multi-organ failure six days after a reported dog bite. Blood cultures were unable to detect a pathogen after 3 days despite observed intracellular bacilli on blood smears. Real-time nanopore sequencing was subsequently employed on whole blood to detect Capnocytophaga canimorsus in 19 h. The patient was not immunocompromised and did not have any other known risk factors. Whole-genome sequencing of clinical sample and of the offending dog's oral swabs showed near-identical C. canimorsus genomes. The patient responded to antibiotic treatment and was discharged from hospital 31 days after admission.

CONCLUSIONS

Use of real-time nanopore sequencing reduced the time-to-diagnosis of Capnocytophaga canimorsus in this case from 6.25 days to 19 h. Capnocytophaga canimorsus should be considered in cases of suspected sepsis involving cat or dog contact, irrespective of the patient's known risk factors.

Characterization of pncA mutations in multi-drug and pyrazinamide resistant Mycobacterium tuberculosis isolates cultured from Queensland migrants and Papua New Guinea residents

Outbreak of drug resistant tuberculosis in the Western province, Papua New Guinea is a concern to Queensland, Australia due to migration. We performed pncA mutation analysis and genotyping of multi-drug/pyrazinamide (MDR/PZA) resistant isolates from 18 Queensland (Qld) migrants and 81 Papua New Guinea (PNG) residents, to compare with phenotypic evidence of PZA resistance and to evaluate the genotypes obtained from the two countries. Seven different mutations were seen from Qld isolates of which 2 have not been described previously. A cluster of mutations were found between amino acids L35 and S65. Amongst the PNG isolates, 10 mutations were identified, of which 6 were unique and have not been described previously. Majority of the mutations formed 2 clusters, between amino acids Q10 to A20 and W68 to W119. Mutations identified at nucleotide (nt) position 202 and 307 were found to be the most common types, occurring in 25% and 51% of the PNG isolates respectively. The majority of the mutations were seen in MDR/PZA resistant isolates. These mutations could be utilized for direct screening of PZA resistance from PNG patient samples. Genotypic analysis of the isolates showed strong clustering amongst the PNG isolates as opposed to Qld isolates. A diversity of mutations and genotypes were seen amongst the Qld migrant isolates. Majority of PNG isolates had one genotype with two distinct pncA mutation patterns (T202C and T307G) which highlight on-going transmission. pncA mutation analysis provided a satisfactory alternative to PZA culture DST with high positive predictive value and an improved result turnaround time.

A complete high-quality MinION nanopore assembly of an extensively drug-resistant Mycobacterium tuberculosis Beijing lineage strain identifies novel variation in repetitive PE/PPE gene regions

A better understanding of the genomic changes that facilitate the emergence and spread of drug-resistant Mycobacterium tuberculosis strains is currently required. Here, we report the use of the MinION nanopore sequencer (Oxford Nanopore Technologies) to sequence and assemble an extensively drug-resistant (XDR) isolate, which is part of a modern Beijing sub-lineage strain, prevalent in Western Province, Papua New Guinea. Using 238-fold coverage obtained from a single flow-cell, de novo assembly of nanopore reads resulted into one contiguous assembly with 99.92 % assembly accuracy. Incorporation of complementary short read sequences (Illumina) as part of consensus error correction resulted in a 4 404 064 bp genome with 99.98 % assembly accuracy. This assembly had an average nucleotide identity of 99.7 % relative to the reference genome, H37Rv. We assembled nearly all GC-rich repetitive PE/PPE family genes (166/168) and identified variants within these genes. With an estimated genotypic error rate of 5.3 % from MinION data, we demonstrated identification of variants to include the conventional drug resistance mutations, and those that contribute to the resistance phenotype (efflux pumps/transporter) and virulence. Reference-based alignment of the assembly allowed detection of deletions and insertions. MinION sequencing provided a fully annotated assembly of a transmissible XDR strain from an endemic setting and showed its utility to provide further understanding of genomic processes within Mycobacterium tuberculosis.

Multi-clonal evolution of multi-drug-resistant/extensively drug-resistant Mycobacterium tuberculosis in a high-prevalence setting of Papua New Guinea for over three decades

An outbreak of multi-drug resistant (MDR) tuberculosis (TB) has been reported on Daru Island, Papua New Guinea. Mycobacterium tuberculosis strains driving this outbreak and the temporal accrual of drug resistance mutations have not been described. Whole genome sequencing of 100 of 165 clinical isolates referred from Daru General Hospital to the Supranational reference laboratory, Brisbane, during 2012–2015 revealed that 95 belonged to a single modern Beijing sub-lineage strain. Molecular dating suggested acquisition of streptomycin and isoniazid resistance in the 1960s, with potentially enhanced virulence mediated by an mycP1 mutation. The Beijing sub-lineage strain demonstrated a high degree of co-resistance between isoniazid and ethionamide (80/95; 84.2 %) attributed to an inhA promoter mutation combined with inhA and ndh coding mutations. Multi-drug resistance, observed in 78/95 samples, emerged with the acquisition of a typical rpoB mutation together with a compensatory rpoC mutation in the 1980s. There was independent acquisition of fluoroquinolone and aminoglycoside resistance, and evidence of local transmission of extensively drug resistant (XDR) strains from 2009. These findings underline the importance of whole genome sequencing in informing an effective public health response to MDR/XDR TB.

Use of real time polymerase chain reaction for detection of M. tuberculosis, M. avium and M. kansasii from clinical specimens

Background

The incidence of M. tuberculosis (MTB) and non tuberculous Mycobacterium species (NTMs) like M. avium and M. kansasii has increased due to Human Immunodeficiency Virus (HIV) epidemic. Therefore accurate, rapid and cost effective methods for the identification of these NTMs and MTB are greatly needed for appropriate TB management. Thus in this study we evaluated the performance of Lightcycler® Mycobacterium detection assay to detect MTB, M. avium and M. kansasii in sputum specimens.

Methods

A total of 241 baseline minimally processed sputum specimens from individual adult TB suspected patients were analyzed by Mycobacterium detection assay (Real-time-PCR) on a LightCycler 480® while using liquid culture as a reference standard.

Results

Real time PCR had a sensitivity of 100% (95% CI 96–100) and 100% (CI 19–100) for detection of MTB and M. avium respectively. Additionally the assay had a specificity of 99% (95% CI 96–99) and 95% (95% CI 91–97) for identification of MTB and M. avium respectively. The positive predictive value (PPV) for Real time PCR to identify MTB and M. avium among the specimens was 98% (95% CI 94–99) and 15% (95% CI 2–45) respectively. The kappa statistics for Real time PCR to identify MTB and M. avium was 0.9 (95% CI 0.9–1.0) and 0.3 (95% CI–0.03–0.5) respectively. The median time to detection for Real time PCR assay was 2 hours while overall median time to detection for MGIT-positive cultures was 8 days. The sample unit cost for Real time PCR was $ 12 compared to $ 20 for the reference liquid culture.

Conclusion

The Light cycler® Mycobacterium detection assay rapidly and correctly identified MTB and M avium thus has the potential to be adopted in a clinical setting.