Endospore Appendages: a novel pilus superfamily from the endospores of pathogenic Bacilli.

Bacillus cereus sensu lato is a group of Gram-positive endospore-forming bacteria with high ecological diversity. Their endospores are decorated with micrometer-long appendages of unknown identity and function. Here, we isolate endospore appendages (Enas) from the food poisoning outbreak strain B. cereus NVH 0075-95 and find proteinaceous fibers of two main morphologies: S- and L-Ena. By using cryoEM and 3D helical reconstruction of S-Enas, we show these to represent a novel class of Gram-positive pili. S-Enas consist of single domain subunits with jellyroll topology that are laterally stacked by ß-sheet augmentation. S-Enas are longitudinally stabilized by disulfide bonding through N-terminal connector peptides that bridge the helical turns. Together, this results in flexible pili that are highly resistant to heat, drought, and chemical damage. Phylogenomic analysis reveals a ubiquitous presence of the ena-gene cluster in the B. cereus group, which include species of clinical, environmental, and food importance. We propose Enas to represent a new class of pili specifically adapted to the harsh conditions encountered by bacterial spores.

The effect of recombination on the evolution of a population of Neisseria meningitidis.

Neisseria meningitidis (the meningococcus) is a major human pathogen with a history of high invasive disease burden, particularly in sub-Saharan Africa. Our current understanding of the evolution of meningococcal genomes is limited by the rarity of large-scale genomic population studies and lack of in-depth investigation of the genomic events associated with routine pathogen transmission. Here, we fill this knowledge gap by a detailed analysis of 2839 meningococcal genomes obtained through a carriage study of over 50,000 samples collected systematically in Burkina Faso, West Africa, before, during, and after the serogroup A vaccine rollout, 2009-2012. Our findings indicate that the meningococcal genome is highly dynamic, with highly recombinant loci and frequent gene sharing across deeply separated lineages in a structured population. Furthermore, our findings illustrate how population structure can correlate with genome flexibility, as some lineages in Burkina Faso are orders of magnitude more recombinant than others. We also examine the effect of selection on the population, in particular how it is correlated with recombination. We find that recombination principally acts to prevent the accumulation of deleterious mutations, although we do also find an example of recombination acting to speed the adaptation of a gene. In general, we show the importance of recombination in the evolution of a geographically expansive population with deep population structure in a short timescale. This has important consequences for our ability to both foresee the outcomes of vaccination programs and, using surveillance data, predict when lineages of the meningococcus are likely to become a public health concern.

Antibiotic Treatment Regimes as a Driver of the Global Population Dynamics of a Major Gonorrhea Lineage.

The Neisseria gonorrhoeae multilocus sequence type (ST) 1901 is among the lineages most commonly associated with treatment failure. Here, we analyze a global collection of ST-1901 genomes to shed light on the emergence and spread of alleles associated with reduced susceptibility to extended-spectrum cephalosporins (ESCs). The genetic diversity of ST-1901 falls into a minor and a major clade, both of which were inferred to have originated in East Asia. The dispersal of the major clade from Asia happened in two separate waves expanding from ∼1987 and 1996, respectively. Both waves first reached North America, and from there spread to Europe and Oceania, with multiple secondary reintroductions to Asia. The ancestor of the second wave acquired the penA 34.001 allele, which significantly reduces susceptibility to ESCs. Our results suggest that the acquisition of this allele granted the second wave a fitness advantage at a time when ESCs became the key drug class used to treat gonorrhea. Following its establishment globally, the lineage has served as a reservoir for the repeated emergence of clones fully resistant to the ESC ceftriaxone, an essential drug for effective treatment of gonorrhea. We infer that the effective population sizes of both clades went into decline as treatment schemes shifted from fluoroquinolones via ESC monotherapy to dual therapy with ceftriaxone and azithromycin in Europe and the United States. Despite the inferred recent population size decline, the short evolutionary path from the penA 34.001 allele to alleles providing full ceftriaxone resistance is a cause of concern.

Evolution of Bordetella pertussis in the acellular vaccine era in Norway, 1996 to 2019.

We described the population structure of Bordetella pertussis (B. pertussis) in Norway from 1996 to 2019 and determined if there were evolutionary shifts and whether these correlated with changes in the childhood immunization program. We selected 180 B. pertussis isolates, 22 from the whole cell vaccine (WCV) era (1996-1997) and 158 from the acellular vaccine (ACV) era (1998-2019). We conducted whole genome sequencing and determined the distribution and frequency of allelic variants and temporal changes of ACV genes. Norwegian B. pertussis isolates were evenly distributed across a phylogenetic tree that included global strains. We identified seven different allelic profiles of ACV genes (A-F), in which profiles A1, A2, and B dominated (89%), all having pertussis toxin (ptxA) allele 1, pertussis toxin promoter (ptxP) allele 3, and pertactin (prn) allele 2 present. Isolates with ptxP1 and prn1 were not detected after 2007, whereas the prn2 allele likely emerged prior to 1972, and ptxP3 before the early 1980s. Allele conversions of ACV genes all occurred prior to the introduction of ACV. Sixteen percent of our isolates showed mutations within the prn gene. ACV and its booster doses (implemented for children in 2007 and adolescents in 2013) might have contributed to evolvement of a more uniform B. pertussis population, with recent circulating strains having ptxA1, ptxP3, and prn2 present, and an increasing number of prn mutations. These strains clearly deviate from ACV strains (ptxA1, ptxP1, prn1), and this could have implications for vaccine efficiency and, therefore, prevention and control of pertussis.

Acquisition of virulence genes by a carrier strain gave rise to the ongoing epidemics of meningococcal disease in West Africa.

In the African meningitis belt, a region of sub-Saharan Africa comprising 22 countries from Senegal in the west to Ethiopia in the east, large epidemics of serogroup A meningococcal meningitis have occurred periodically. After gradual introduction from 2010 of mass vaccination with a monovalent meningococcal A conjugate vaccine, serogroup A epidemics have been eliminated. Starting in 2013, the northwestern part of Nigeria has been affected by yearly outbreaks of meningitis caused by a novel strain of serogroup C Neisseria meningitidis (NmC). In 2015, the strain spread to the neighboring country Niger, where it caused a severe epidemic. Following a relative calm in 2016, the largest ever recorded epidemic of NmC broke out in Nigeria in 2017. Here, we describe the recent evolution of this new outbreak strain and show how the acquisition of capsule genes and virulence factors by a strain previously circulating asymptomatically in the African population led to the emergence of a virulent pathogen. This study illustrates the power of long-read whole-genome sequencing, combined with Illumina sequencing, for high-resolution epidemiological investigations.

A simple stochastic model describing genomic evolution over time of GC content in microbial symbionts.

An organism's genomic base composition is usually summarized by its AT or GC content due to Chargaff's parity laws. Variation in prokaryotic GC content can be substantial between taxa but is generally small within microbial genomes. This variation has been found to correlate with both phylogeny and environmental factors. Since novel single-nucleotide polymorphisms (SNPs) within genomes are at least partially linked to the environment through natural selection, SNP GC content can be considered a compound measure of an organism's environmental influences, lifestyle, phylogeny as well as other more or less random processes. While there are several models describing genomic GC content few, if any, consider AT/GC mutation rates subjected to random perturbations. We present a mathematical model that describes how GC content in microbial genomes evolves over time as a function of the AT â†’ GC and GC â†’ AT mutation rates with Gaussian white noise disturbances. The model, which is suited specifically to non-recombining vertically transmitted prokaryotic symbionts, suggests that small differences in the AT/GC mutation rates can lead to profound differences in outcome due to the ensuing stochastic process. In other words, the model indicates that time to extinction could be a consequence of the mutation rate trajectory on which the symbiont embarked early on in its evolutionary history.

COVID-19 prevalence estimation by random sampling in population - optimal sample pooling under varying assumptions about true prevalence.

BACKGROUND: The number of confirmed COVID-19 cases divided by population size is used as a coarse measurement for the burden of disease in a population. However, this fraction depends heavily on the sampling intensity and the various test criteria used in different jurisdictions, and many sources indicate that a large fraction of cases tend to go undetected. METHODS: Estimates of the true prevalence of COVID-19 in a population can be made by random sampling and pooling of RT-PCR tests. Here I use simulations to explore how experiment sample size and degrees of sample pooling impact precision of prevalence estimates and potential for minimizing the total number of tests required to get individual-level diagnostic results. RESULTS: Sample pooling can greatly reduce the total number of tests required for prevalence estimation. In low-prevalence populations, it is theoretically possible to pool hundreds of samples with only marginal loss of precision. Even when the true prevalence is as high as 10% it can be appropriate to pool up to 15 samples. Sample pooling can be particularly beneficial when the test has imperfect specificity by providing more accurate estimates of the prevalence than an equal number of individual-level tests. CONCLUSION: Sample pooling should be considered in COVID-19 prevalence estimation efforts.

Armed conflict and population displacement as drivers of the evolution and dispersal of Mycobacterium tuberculosis.

The "Beijing" Mycobacterium tuberculosis (Mtb) lineage 2 (L2) is spreading globally and has been associated with accelerated disease progression and increased antibiotic resistance. Here we performed a phylodynamic reconstruction of one of the L2 sublineages, the central Asian clade (CAC), which has recently spread to western Europe. We find that recent historical events have contributed to the evolution and dispersal of the CAC. Our timing estimates indicate that the clade was likely introduced to Afghanistan during the 1979-1989 Soviet-Afghan war and spread further after population displacement in the wake of the American invasion in 2001. We also find that drug resistance mutations accumulated on a massive scale in Mtb isolates from former Soviet republics after the fall of the Soviet Union, a pattern that was not observed in CAC isolates from Afghanistan. Our results underscore the detrimental effects of political instability and population displacement on tuberculosis control and demonstrate the power of phylodynamic methods in exploring bacterial evolution in space and time.

Genotypic and Phenotypic Characterization of the O-Linked Protein Glycosylation System Reveals High Glycan Diversity in Paired Meningococcal Carriage Isolates.

Species within the genus Neisseria display significant glycan diversity associated with the O-linked protein glycosylation (pgl) systems due to phase variation and polymorphic genes and gene content. The aim of this study was to examine in detail the pgl genotype and glycosylation phenotype in meningococcal isolates and the changes occurring during short-term asymptomatic carriage. Paired meningococcal isolates derived from 50 asymptomatic meningococcal carriers, taken about 2 months apart, were analyzed with whole-genome sequencing. The O-linked protein glycosylation genes were characterized in detail using the Genome Comparator tool at the https://pubmlst.org/ database. Immunoblotting with glycan-specific antibodies (Abs) was used to investigate the protein glycosylation phenotype. All major pgl locus polymorphisms identified in Neisseria meningitidis to date were present in our isolate collection, with the variable presence of pglG and pglH, both in combination with either pglB or pglB2 We identified significant changes and diversity in the pgl genotype and/or glycan phenotype in 96% of the paired isolates. There was also a high degree of glycan microheterogeneity, in which different variants of glycan structures were found at a given glycoprotein. The main mechanism responsible for the observed differences was phase-variable expression of the involved glycosyltransferases and the O-acetyltransferase. To our knowledge, this is the first characterization of the pgl genotype and glycosylation phenotype in a larger strain collection. This report thus provides important insight into glycan diversity in N. meningitidis and into the phase variability changes that influence the expressed glycoform repertoire during meningococcal carriage.IMPORTANCE Bacterial meningitis is a serious global health problem, and one of the major causative organisms is Neisseria meningitidis, which is also a common commensal in the upper respiratory tract of healthy humans. In bacteria, numerous loci involved in biosynthesis of surface-exposed antigenic structures that are involved in the interaction between bacteria and host are frequently subjected to homologous recombination and phase variation. These mechanisms are well described in Neisseria, and phase variation provides the ability to change these structures reversibly in response to the environment. Protein glycosylation systems are becoming widely identified in bacteria, and yet little is known about the mechanisms and evolutionary forces influencing glycan composition during carriage and disease.

Modeling of the GC content of the substituted bases in bacterial core genomes.

BACKGROUND: The purpose of the present study was to examine the GC content of substituted bases (sbGC) in the core genomes of 35 bacterial species. Each species, or core genome, constituted genomes from at least 10 strains. We also wanted to explore whether sbGC for each strain was associated with the corresponding species' core genome GC content (cgGC). We present a simple mathematical model that estimates sbGC from cgGC. The model assumes only that the estimated sbGC is a function of cgGC proportional to fixed AT→GC (α) and GC → AT (ß) mutation rates. Non-linear regression was used to estimate parameters α and ß from the empirical data described above. RESULTS: We found that sbGC for each strain showed a non-linear association with the corresponding cgGC with a bias towards higher GC content for most core genomes (66.3% of the strains), assuming as a null-hypothesis that sbGC should be approximately equal to cgGC. The most GC rich core genomes (i.e. approximately %GC > 60), on the other hand, exhibited slightly less GC-biased sbGC than expected. The best fitted regression model indicates that GC → AT mutation rates ß = (1.91 ± 0.13) p < 0.001 are approximately (1.91/0.79) = 2.42 times as high, on average, as AT→GC α = (- 0.79 ± 0.25) p < 0.001 mutation rates. Whether the observed sbGC GC-bias for all but the most GC-rich prokaryotic species is due to selection, compensating for the GC → AT mutation bias, and/or selective neutral processes is currently debated. Residual standard error was found to be σ = 0.076 indicating estimated errors of sbGC to be approximately within ±15.2% GC (95% confidence interval) for the strains of all species in the study. CONCLUSION: Not only did our mathematical model give reasonable estimates of sbGC it also provides further support to previous observations that mutation rates in prokaryotes exhibit a universal GC → AT bias that appears to be remarkably consistent between taxa.

The nucleotide composition of microbial genomes indicates differential patterns of selection on core and accessory genomes.

BACKGROUND: The core genome consists of genes shared by the vast majority of a species and is therefore assumed to have been subjected to substantially stronger purifying selection than the more mobile elements of the genome, also known as the accessory genome. Here we examine intragenic base composition differences in core genomes and corresponding accessory genomes in 36 species, represented by the genomes of 731 bacterial strains, to assess the impact of selective forces on base composition in microbes. We also explore, in turn, how these results compare with findings for whole genome intragenic regions. RESULTS: We found that GC content in coding regions is significantly higher in core genomes than accessory genomes and whole genomes. Likewise, GC content variation within coding regions was significantly lower in core genomes than in accessory genomes and whole genomes. Relative entropy in coding regions, measured as the difference between observed and expected trinucleotide frequencies estimated from mononucleotide frequencies, was significantly higher in the core genomes than in accessory and whole genomes. Relative entropy was positively associated with coding region GC content within the accessory genomes, but not within the corresponding coding regions of core or whole genomes. CONCLUSION: The higher intragenic GC content and relative entropy, as well as the lower GC content variation, observed in the core genomes is most likely associated with selective constraints. It is unclear whether the positive association between GC content and relative entropy in the more mobile accessory genomes constitutes signatures of selection or selective neutral processes.

Whole genome sequencing reveals within-host genetic changes in paired meningococcal carriage isolates from Ethiopia.

BACKGROUND: Meningococcal colonization is a prerequisite for transmission and disease, but the bacterium only very infrequently causes disease while asymptomatic carriage is common. Carriage is highly dynamic, showing a great variety across time and space within and across populations, but also within individuals. The understanding of genetic changes in the meningococcus during carriage, when the bacteria resides in its natural niche, is important for understanding not only the carriage state, but the dynamics of the entire meningococcal population. RESULTS: Paired meningococcal isolates, obtained from 50 asymptomatic carriers about 2 months apart were analyzed with whole genome sequencing (WGS). Phylogenetic analysis revealed that most paired isolates from the same individual were closely related, and the average and median number of allelic differences between paired isolates defined as the same strain was 35. About twice as many differences were seen between isolates from different individuals within the same sequence type (ST). In 8%, different strains were detected at different time points. A difference in ST was observed in 6%, including an individual who was found to carry three different STs over the course of 9 weeks. One individual carried different strains from the same ST. In total, 566 of 1605 cgMLST genes had undergone within-host genetic changes in one or more pairs. The most frequently changed cgMLST gene was relA that was changed in 47% of pairs. Across the whole genome, pilE, differed mostly, in 85% of the pairs. The most frequent mechanisms of genetic difference between paired isolates were phase variation and recombination, including gene conversion. Different STs showed variation with regard to which genes that were most frequently changed, mostly due to absence/presence of phase variation. CONCLUSIONS: This study revealed within-host genetic differences in meningococcal isolates during short-term asymptomatic carriage. The most frequently changed genes were genes belonging to the pilin family, the restriction/modification system, opacity proteins and genes involved in glycosylation. Higher resolution genome-wide sequence typing is necessary to resolve the diversity of isolates and reveals genetic differences not discovered by traditional typing schemes, and would be the preferred choice of technology.

Whole-Genome Characterization of Epidemic Neisseria meningitidis Serogroup C and Resurgence of Serogroup W, Niger, 2015.

In 2015, Niger reported the largest epidemic of Neisseria meningitidis serogroup C (NmC) meningitis in sub-Saharan Africa. The NmC epidemic coincided with serogroup W (NmW) cases during the epidemic season, resulting in a total of 9,367 meningococcal cases through June 2015. To clarify the phylogenetic association, genetic evolution, and antibiotic determinants of the meningococcal strains in Niger, we sequenced the genomes of 102 isolates from this epidemic, comprising 81 NmC and 21 NmW isolates. The genomes of 82 isolates were completed, and all 102 were included in the analysis. All NmC isolates had sequence type 10217, which caused the outbreaks in Nigeria during 2013-2014 and for which a clonal complex has not yet been defined. The NmC isolates from Niger were substantially different from other NmC isolates collected globally. All NmW isolates belonged to clonal complex 11 and were closely related to the isolates causing recent outbreaks in Africa.

CNOGpro: detection and quantification of CNVs in prokaryotic whole-genome sequencing data.

MOTIVATION: The explosion of whole-genome sequencing (WGS) as a tool in the mapping and understanding of genomes has been accompanied by an equally massive report of tools and pipelines for the analysis of DNA copy number variation (CNV). Most currently available tools are designed specifically for human genomes, with comparatively little literature devoted to CNVs in prokaryotic organisms. However, there are several idiosyncrasies in prokaryotic WGS data. This work proposes a step-by-step approach for detection and quantification of copy number variants specifically aimed at prokaryotes. RESULTS: After aligning WGS reads to a reference genome, we count the individual reads in a sliding window and normalize these counts for bias introduced by differences in GC content. We then investigate the coverage in two fundamentally different ways: (i) Employing a Hidden Markov Model and (ii) by repeated sampling with replacement (bootstrapping) on each individual gene. The latter bypasses the complex problem of breakpoint determination. To demonstrate our method, we apply it to real and simulated WGS data and benchmark it against two popular methods for CNV detection. The proposed methodology will in some cases represent a significant jump in accuracy from other current methods. AVAILABILITY AND IMPLEMENTATION: CNOGpro is written entirely in the R programming language and is available from the CRAN repository (http://cran.r-project.org) under the GNU General Public License.

An evolutionary analysis of genome expansion and pathogenicity in Escherichia coli.

BACKGROUND: There are several studies describing loss of genes through reductive evolution in microbes, but how selective forces are associated with genome expansion due to horizontal gene transfer (HGT) has not received similar attention. The aim of this study was therefore to examine how selective pressures influence genome expansion in 53 fully sequenced and assembled Escherichia coli strains. We also explored potential connections between genome expansion and the attainment of virulence factors. This was performed using estimations of several genomic parameters such as AT content, genomic drift (measured using relative entropy), genome size and estimated HGT size, which were subsequently compared to analogous parameters computed from the core genome consisting of 1729 genes common to the 53 E. coli strains. Moreover, we analyzed how selective pressures (quantified using relative entropy and dN/dS), acting on the E. coli core genome, influenced lineage and phylogroup formation. RESULTS: Hierarchical clustering of dS and dN estimations from the E. coli core genome resulted in phylogenetic trees with topologies in agreement with known E. coli taxonomy and phylogroups. High values of dS, compared to dN, indicate that the E. coli core genome has been subjected to substantial purifying selection over time; significantly more than the non-core part of the genome (p<0.001). This is further supported by a linear association between strain-wise dS and dN values (ß = 26.94 ± 0.44, R2~0.98, p<0.001). The non-core part of the genome was also significantly more AT-rich (p<0.001) than the core genome and E. coli genome size correlated with estimated HGT size (p<0.001). In addition, genome size (p<0.001), AT content (p<0.001) as well as estimated HGT size (p<0.005) were all associated with the presence of virulence factors, suggesting that pathogenicity traits in E. coli are largely attained through HGT. No associations were found between selective pressures operating on the E. coli core genome, as estimated using relative entropy, and genome size (p~0.98). CONCLUSIONS: On a larger time frame, genome expansion in E. coli, which is significantly associated with the acquisition of virulence factors, appears to be independent of selective forces operating on the core genome.

Tracing the adaptive evolution of SARS-CoV-2 during vaccine roll-out in Norway.

Vaccination against SARS-CoV-2 has greatly mitigated the impact of the COVID-19 pandemic. However, concerns have been raised about the degree to which vaccination might drive the emergence and selection of immune escape mutations that will hamper the efficacy of the vaccines. In this study, we investigate whether vaccination impacted the micro-scale adaptive evolution of SARS-CoV-2 in the Oslo region of Norway, during the first nine months of 2021, a period in which the population went from near-zero to almost 90 per cent vaccine coverage in the population over 50 years old. Weekly aggregated data stratified by age on vaccine uptake and number of SARS-CoV-2 cases in the area were obtained from the National Immunization Registry and the Norwegian Surveillance System for Communicable Diseases, respectively. A total of 6,438 virus sequences (7.5 per cent of the registered cases) along with metadata were available. We used a causal-driven approach to investigate the relationship between vaccination progress and changes in the frequency of 362 mutations present in at least ten samples, conditioned on the emergence of new lineages, time, and population vaccination coverage. After validating our approach, we identified 21 positive and 12 negative connections between vaccination progress and mutation prevalence, and most of them were outside the Spike protein. We observed a tendency for the mutations that we identified as positively connected with vaccination to decrease as the vaccinated population increased. After modelling the fitness of different competing mutations in a population, we found that our observations could be explained by a clonal interference phenomenon in which high fitness mutations would be outcompeted by the emergence or introduction of other high-fitness mutations.

Transcriptomic and proteomic analysis of the virulence inducing effect of ciprofloxacin on enterohemorrhagic Escherichia coli.

Enterohemorrhagic E. coli (EHEC) is considered to be the most dangerous pathotype of E. coli, as it causes severe conditions such as hemorrhagic colitis (HC) and hemolytic uremic syndrome (HUS). Antibiotic treatment of EHEC infections is generally not recommended since it may promote the production of the Shiga toxin (Stx) and lead to worsened symptoms. This study explores how exposure to the fluoroquinolone ciprofloxacin reorganizes the transcriptome and proteome of EHEC O157:H7 strain EDL933, with special emphasis on virulence-associated factors. As expected, exposure to ciprofloxacin caused an extensive upregulation of SOS-response- and Stx-phage proteins, including Stx. A range of other virulence-associated factors were also upregulated, including many genes encoded by the LEE-pathogenicity island, the enterohemolysin gene (ehxA), as well as several genes and proteins involved in LPS production. However, a large proportion of the genes and proteins (17 and 8%, respectively) whose expression was upregulated upon ciprofloxacin exposure (17 and 8%, respectively) are not functionally assigned. This indicates a knowledge gap in our understanding of mechanisms involved in EHECs response to antibiotic-induced stress. Altogether, the results contribute to better understanding of how exposure to ciprofloxacin influences the virulome of EHEC and generates a knowledge base for further studies on how EHEC responds to antibiotic-induced stress. A deeper understanding on how EHEC responds to antibiotics will facilitate development of novel and safer treatments for EHEC infections.

Scoary2: rapid association of phenotypic multi-omics data with microbial pan-genomes.

Unraveling bacterial gene function drives progress in various areas, such as food production, pharmacology, and ecology. While omics technologies capture high-dimensional phenotypic data, linking them to genomic data is challenging, leaving 40-60% of bacterial genes undescribed. To address this bottleneck, we introduce Scoary2, an ultra-fast microbial genome-wide association studies (mGWAS) software. With its data exploration app and improved performance, Scoary2 is the first tool to enable the study of large phenotypic datasets using mGWAS. As proof of concept, we explore the metabolome of yogurts, each produced with a different Propionibacterium reichii strain and discover two genes affecting carnitine metabolism.

Detection of a historic reservoir of bedaquiline/clofazimine resistance-associated variants in Mycobacterium tuberculosis.

BACKGROUND: Drug resistance in tuberculosis (TB) poses a major ongoing challenge to public health. The recent inclusion of bedaquiline into TB drug regimens has improved treatment outcomes, but this advance is threatened by the emergence of strains of Mycobacterium tuberculosis (Mtb) resistant to bedaquiline. Clinical bedaquiline resistance is most frequently conferred by off-target resistance-associated variants (RAVs) in the mmpR5 gene (Rv0678), the regulator of an efflux pump, which can also confer cross-resistance to clofazimine, another TB drug. METHODS: We compiled a dataset of 3682 Mtb genomes, including 180 carrying variants in mmpR5, and its immediate background (i.e. mmpR5 promoter and adjacent mmpL5 gene), that have been associated to borderline (henceforth intermediate) or confirmed resistance to bedaquiline. We characterised the occurrence of all nonsynonymous mutations in mmpR5 in this dataset and estimated, using time-resolved phylogenetic methods, the age of their emergence. RESULTS: We identified eight cases where RAVs were present in the genomes of strains collected prior to the use of bedaquiline in TB treatment regimes. Phylogenetic reconstruction points to multiple emergence events and circulation of RAVs in mmpR5, some estimated to predate the introduction of bedaquiline. However, epistatic interactions can complicate bedaquiline drug-susceptibility prediction from genetic sequence data. Indeed, in one clade, Ile67fs (a RAV when considered in isolation) was estimated to have emerged prior to the antibiotic era, together with a resistance reverting mmpL5 mutation. CONCLUSIONS: The presence of a pre-existing reservoir of Mtb strains carrying bedaquiline RAVs prior to its clinical use augments the need for rapid drug susceptibility testing and individualised regimen selection to safeguard the use of bedaquiline in TB care and control.