Inhabitants-based genomic surveillance
From 2016 to 2021, 16,734 TB instances had been notified in Paraguay, with the vast majority of instances (60%; 10,095/16,734) occurring within the city departments Central and Distrito Capital (which collectively comprise Asunción) and Alto Paraná (Ciudad del Este), the place we performed potential genomic surveillance (Fig. 1a, Fig. S1). In 2021, the TB notification charge was 70 occasions larger in prisons than exterior (3378 instances per 100,000 in prisons/49 instances per 100,000 within the normal inhabitants) (Fig. 1b). Due to this fact, we targeted genomic surveillance within the two largest prisons within the nation, Tacumbú Jail and the Jail of Ciudad del Este, which collectively maintain 36% (4950/13,821) of Paraguay’s incarcerated inhabitants, notification charges are 2000 and 3500 per 100,000 individuals, respectively.
a Map of South America, with Paraguay highlighted, and as an inset. Paraguay’s departments are coloured by inhabitants measurement density and factors point out the 2 largest city facilities in Paraguay, the place we performed targeted genomic surveillance. b Notification charge of TB per 100,000 individuals in prisons (crimson) and within the normal inhabitants (blue) from 2009 to 2020. Supply information are supplied as a Supply Information file.
Of the 7780 TB instances notified in Asunción throughout the research interval, 781 (10%) occurred amongst incarcerated people (Fig. S1). 64% (503/781) of those had been culture-positive and, of those, we sequenced 21% (107/503). 33% (2306/6999) of non-incarcerated people with TB in Asunción had been culture-positive and of those, we sequenced 7% (172/2,306). Of the two,315 TB instances notified in Ciudad del Este throughout the research interval, 422 (18%) occurred amongst incarcerated people (Fig. S1). 64% (269/422) of those had been culture-positive and, of those, we sequenced 20% (55/269). 31% (578/1893) of non-incarcerated people with TB in Ciudad del Este had been culture-positive and of those, we sequenced 27% (158/578) (Fig. S1).
Complete genome sequences (WGS) for a complete of 532 isolates met our protection and high quality standards (Strategies), together with 488 from distinctive TB notifications. Of the samples passing filters, 158 had been from people identified with TB whereas in jail and 330 had been from individuals identified locally. TB isolates had been collected in Asunción (274/488) and in Ciudad del Este (214/488). We excluded 17 isolates with proof of blended an infection with a couple of sub-lineage detected, leading to 471 M. tuberculosis isolates for following analyses.
Genotypic resistance
The bulk, 96% (454/471) of sampled M. tuberculosis, had been drug-sensitive; 3% (15/471) had been immune to a minimum of one drug; and 0.42% (2/471) had been multi-drug resistant, immune to each isoniazid and rifampin. Resistance was not related to sub-lineage (X2 (11) = 7.7, p = 0.74). We recognized three distinctive isoniazid resistance-conferring mutations on the genes fabG1, katG, or each among the many 10 isolates with any isoniazid resistance; the three rifampicin-conferring mutations in rpoB (two on multi-drug resistant isolates) had been distinctive (Fig. 2).
A most probability phylogeny of 471 tuberculosis isolates from Lineage 4 inferred from a a number of sequence alignment of 9966 SNPs and rooted on three sub-lineage 4.8 samples from this research. Department lengths are in items of substitutions per web site. Branches are coloured by sub-lineage. From the within, rings are coloured by antimicrobial resistance; metropolis of sampling; and incarceration standing on the time of TB notification. Different isoniazid consists of isoniazid mono-resistant isolates with out an ahpC promoter mutation. Rifampicin signifies rifampicin mono-resistant isolates. Different resistance consists of isolates with mutations related to resistance to pyrazinamide, streptomycin, or fluoroquinolones and never isoniazid or rifampicin. Supply information are supplied as a Supply Information file.
Secure genomic variety of M. tuberculosis in Paraguay
After excluding blended infections, all M. tuberculosis isolates had been strains from M. tuberculosis lineage 4. A single blended lineage an infection was co-infected with strains from each lineages 1 and 4. Samples predominantly fell into 4 sub-lineages: 4.3.3/LAM (42.5%; 200/471), 4.1.2 /Haarlem (38.2%; 180/471), 4.4.1/S (12.3%; 58/471), and 4.3.4/LAM (3.2%; 15/471) (Fig. 2). The distribution of strains representing these sublineages was secure and didn’t change considerably from a set of 173 M. tuberculosis isolates collected in 200315 (Fig. S1).
Latest growth of M. tuberculosis transmission clusters
We subsequent explored proof of latest M. tuberculosis transmission in Paraguay. As seen in a most probability phylogeny (Fig. 2), sampled M. tuberculosis variety was dominated by a number of extremely associated clones. Seventy-eight % (369/471) of all isolates fell inside 26 genomic clusters (every together with 2 to 159 isolates) outlined by a 12-SNP threshold20, suggesting TB notifications had been typically attributable to latest transmission.
We reconstructed inhabitants measurement dynamics of the three largest genomic clusters—which comprised 56% (264/471) of our pattern—with a Bayesian coalescent inhabitants measurement mannequin. The three largest genomic clusters (together with 159, 91, and 15 samples) elevated in efficient inhabitants measurement by 200, 90, and 40-fold, respectively. Cluster progress was comparatively latest, with cluster most up-to-date widespread ancestors (MRCA) occurring in 1998 (95% HDI: 1994–2001), 1996 (95% HDI: 1991–2000), and 1998 (95% HDI: 1992–2003) respectively, to 2021, when the newest samples had been collected (Fig. 3). All three clusters included isolates from people notified with TB throughout incarceration and people with no incarceration historical past.
a Efficient inhabitants measurement (Ne) estimates for the three largest genomic clusters in our pattern over time. Black strains point out Ne inferred in a Bayesian Skyline Coalescent mannequin and gray shading signifies 95% excessive posterior density estimates. b Median clade credibility timber inferred from the Bayesian Skyline Coalescent mannequin. Department lengths are in years and gray bars point out 95% excessive posterior density estimates of node date. The heatmap to the fitting of the phylogeny signifies affected person incarceration standing on the time of TB notification. Supply information are supplied as a Supply Information file.
We discovered no proof that genomic loci had been related to profitable genomic clusters, which we outlined because the clusters containing greater than 15 M. tuberculosis isolates, utilizing a bacterial GWAS method which controls for clonality and powerful inhabitants construction21. We equally discovered no affiliation between genomic loci and clustered phenotype after we examined membership in a cluster with 10 or extra isolates or 2 or extra isolates.
M. tuberculosis genomic clusters span prisons and the overall inhabitants
In a most probability phylogeny (Fig. 2), M. tuberculosis isolates sampled from incarcerated and non-incarcerated individuals are distributed throughout the tree and didn’t kind distinct clades, indicating a latest shared evolutionary historical past of isolates sampled from prisons and the neighborhood. Nevertheless, sub-lineage was related to incarceration standing (χ2(22) = 52.3, ρ < 0.001), with strains from sub-lineage 4.1.2.1 extra often infecting individuals with a historical past of incarceration (46.1%; 83/180) in comparison with people with no incarceration historical past (33.0%; 96/291; p = 0.006).
Isolates from incarcerated individuals had been extra often clustered (92.6%, 138/149), than these from previously incarcerated (71.0%, 22/31, χ2(1) = 10.1, ρ = 0.001) or by no means incarcerated individuals (71.8%, 209/291; χ2(1) = 24.3, p < 0.001), doubtless reflecting more moderen transmission inside prisons. With a stricter threshold of 5 SNPs, 45.4% (214/471) of all isolates in genomic transmission clusters. With this threshold, isolates from incarcerated people had been once more extra often clustered (58.3%; 87/149) than these from these previously incarcerated (45.2%; 14/31), although not considerably so (χ2(1) = 1.3, ρ = 0.25), and isolates from incarcerated people had been extra often clustered than these from by no means incarcerated people (38.8%; 113/291; χ2(1) = 24.3, ρ < 0.001).
We predicted that if jail and community-associated epidemics had been distinct, isolates from the neighborhood could be most intently associated to and cluster with different isolates from the neighborhood and vice versa. Roughly half (48.0%; 12/25) of genomic clusters, together with individuals with no incarceration historical past additionally included people with a latest historical past of incarceration. The consequence is that 85.2% (178/209) of people with proof of latest transmission and no latest incarceration had been inside transmission clusters, together with people with prior incarceration.
We moreover quantified M. tuberculosis latest transmission with time-scaled haplotype variety, a measure of the centrality of a single tip isolate to all different isolates on the tree22. People who had been incarcerated on the time of TB notification had the next time-scaled haplotype index for a brief epidemic timescale (median: 0.59, IQR: 0.24–0.72) than did previously (median: 0.18, IQR: −0.37–0.71; t(36) = 1.7, p = 0.03) or by no means incarcerated people (median: 0.20, IQR: −0.71–0.66; t(360) = 5.9, p < 0.001) (Fig. S3). This discovering was constant throughout epidemic timescales thought-about (Fig. S3). After adjusting for inhabitants construction, we discovered that incarceration standing was considerably related to time-scaled haplotype variety (one-way ANOVA: F(285) = 85, p < 0.001), proof that the affiliation was unbiased of TB lineage.
Geographic construction regardless of frequent migration throughout M. tuberculosis sub-lineages
We discovered sample of reasonable geographic construction in sampled M. tuberculosis (Fig. 4), with strains from sub-lineage 4.1.2.1 dominant in Asunción (54.1%, 142/262 samples) and strains from sub-lineage 4.3.3 dominant in Ciudad del Este (60.8%, 127/209) (χ2(2) = 72, ρ < 0.001) (Fig. 4). Whereas we noticed geographically distinct patterns of M. tuberculosis variety in Asunción and Ciudad del Este, reconstruction of the ancestral areas for the three most prevalent sub-lineages revealed frequent motion of M. tuberculosis (Fig. 4).
a–c We used discrete ancestral state reconstruction to reconstruct migration between the 2 cities for the three dominant sub-lineages in our pattern; lineages 4.1.2.1, 4.3.3, and 4.4.1.1. Bayesian most clade credibility timber of samples within the three dominant sub-lineages with tip factors coloured by metropolis of sampling and pie charts at nodes indicating the inferred ancestral location. Department lengths are in years and gray bars point out 95% excessive posterior density estimates of node date. A mannequin of uneven charges of motion between the 2 cities was supported for sub-lineages 4.1.2.1 a mannequin of symmetric charges of motion was supported for 4.3.3 and 4.4.1.1. d Map of Paraguay with pie charts indicating the genomic variety sampled in Asunción (at Paraguay’s western border) and Ciudad del Este (jap border). Arrows are coloured by sub-lineage and are weighted by the relative charge of migration between cities. Bi-directional arrows point out equal charges of migration in every course. Supply information are supplied as a Supply Information file.
To check whether or not Asunción and Ciudad del Este served as sources for M. tuberculosis, exporting an infection elsewhere, we in contrast charges of arrival and export of every sub-lineage. Sub-lineage 4.1.2.1 moved extra often Asunción to Ciudad del Este (imply: 75 transitions) in comparison with vice versa (imply: 70 transitions), and a mannequin for uneven charges was supported (χ2(2) = 4.1, ρ = 0.04) (Fig. 4). Each sub-lineages 4.4.1.1 (with the prevalent ahpC mutation) (χ2(2) = 0.16, ρ = 0.69) and 4.3.3 (χ2(2) = 0.56, ρ = 0.46) had related charges of migration to and from Ciudad del Este to Asunción. Regardless of the geographic construction noticed, there was not a enough sign to deduce a probable geographic supply for any of the dominant sub-lineages.
Emergence of a putative resistance-associated ahpC promoter mutation
Eleven % of samples (50/471) shared a mutation in ahpC promoter (G > A, 74 bases upstream of the 5’ begin codon), beforehand thought-about a location for compensatory mutations co-occurring with katG mutations in isoniazid-resistant isolates23,24. Whereas ahpC promoter mutations usually are not included as an unbiased resistance-conferring mutation within the WHO resistance catalogue23, in our assortment, ahpC mutations occurred on in any other case vulnerable genomic background inside sub-lineage 4.4.1.1. The ahpC mutation occurred in a monophyletic clade of 49 samples in sublineage 4.4.1.1 (Fig. 5), which shared a most up-to-date widespread ancestor in 1903 (95% HDI: 1888–1916), doubtless reflecting a single emergence occasion. Among the many basal group of 9 samples and not using a mounted ahpC promoter locus (ahpC−74) mutation, one pattern was polymorphic, with 16% (13/79) of reads representing the ahpC mutation. Among the many samples sharing the ahpC mutation, a single isolate had a co-occurring rifampicin resistance-conferring mutation in rpoB (His445Leu) (Fig. 5).
Time-scaled Bayesian most clade credibility tree for 58 samples in sublineage 4.4.1.1. Department lengths are in years and gray bars point out 95% excessive posterior density estimates of node date. The heatmap to the fitting of the phylogeny signifies affected person incarceration standing on the time of TB notification, ahpC mutation frequency inside a person’s an infection, and the prevalence of a rifampicin resistance-conferring mutation (rpoB His445Leu). Supply information are supplied as a Supply Information file.
We examined whether or not the success of the ahpC mutation within the absence of a katG mutation (i.e. exterior of a compensatory context) we noticed in Paraguay may very well be defined by a rise in related transmissibility. The ahpC mutation was not related to an elevated time-scaled haplotype density (ahpC mutants, median 0.19, IQR: 0.09–0.22; ahpC non-mutants, median: 0.56, IQR: −0.44–0.71, p = 0.93). Additional, people with an incarceration historical past (at present or previously incarcerated) had been no extra more likely to be contaminated with a M. tuberculosis isolate with the ahpC mutation than had been people with no incarceration historical past (χ2(1) = 0.25, ρ = 0.62).
