Identification of bacterial determinants of tuberculosis infection and treatment outcomes: a phenogenomic analysis of clinical strains

Abstract

Background: Bacterial diversity could contribute to the diversity of tuberculosis infection and treatment outcomes observed clinically, but the biological basis of this association is poorly understood. The aim of this study was to identify associations between phenogenomic variation in Mycobacterium tuberculosis and tuberculosis clinical features.

Methods: We developed a high-throughput platform to define phenotype-genotype relationships in M tuberculosis clinical isolates, which we tested on a set of 158 drug-sensitive M tuberculosis strains sampled from a large tuberculosis clinical study in Ho Chi Minh City, Viet Nam. We tagged the strains with unique genetic barcodes in multiplicate, allowing us to pool the strains for in-vitro competitive fitness assays across 16 host-relevant antibiotic and metabolic conditions. Relative fitness was quantified by deep sequencing, enumerating output barcode read counts relative to input normalised values. We performed a genome-wide association study to identify phylogenetically linked and monogenic mutations associated with the in-vitro fitness phenotypes. These genetic determinants were further associated with relevant clinical outcomes (cavitary disease and treatment failure) by calculating odds ratios (ORs) with binomial logistic regressions. We also assessed the population-level transmission of strains associated with cavitary disease and treatment failure using terminal branch length analysis of the phylogenetic data.

Findings: M tuberculosis clinical strains had diverse growth characteristics in host-like metabolic and drug conditions. These fitness phenotypes were highly heritable, and we identified monogenic and phylogenetically linked variants associated with the fitness phenotypes. These data enabled us to define two genetic features that were associated with clinical outcomes. First, mutations in Rv1339, a phosphodiesterase, which were associated with slow growth in glycerol, were further associated with treatment failure (OR 5·34, 95% CI 1·21-23·58, p=0·027). Second, we identified a phenotypically distinct slow-growing subclade of lineage 1 strains (L1.1.1.1) that was associated with cavitary disease (OR 2·49, 1·11-5·59, p=0·027) and treatment failure (OR 4·76, 1·53-14·78, p=0·0069), and which had shorter terminal branch lengths on the phylogenetic tree, suggesting increased transmission.

Interpretation: Slow growth under various antibiotic and metabolic conditions served as in-vitro intermediate phenotypes underlying the association between M tuberculosis monogenic and phylogenetically linked mutations and outcomes such as cavitary disease, treatment failure, and transmission potential. These data suggest that M tuberculosis growth regulation is an adaptive advantage for bacterial success in human populations, at least in some circumstances. These data further suggest markers for the underlying bacterial processes that contribute to these clinical outcomes.

Funding: National Health and Medical Research Council/A∗STAR, National Institutes of Allergy and Infectious Diseases, National Institute of Child Health and Human Development, and the Wellcome Trust Fellowship in Public Health and Tropical Medicine.

Figure 1:. Mycobacterium tuberculosis clinical isolates genetically barcoded for pooled competition experiments to examine relative metabolic and antibiotic fitness phenotypes

(A) Approximate maximum-likelihood tree of the 158 M tuberculosis strains selected for the study and the reference strain, M tuberculosis Erdman. The scale indicates the number of mutations per site; the tree is rooted at the midpoint. (B) Dot plot of relative fitness values for all 159 strains per indicated condition and timepoint. All carbon sources were added to concentrations of 0·1–0·2% w/v. Relative fitness values were normalised to strain abundance in input pool inoculum, so all day 0 values are set to zero. Kruskal-Wallis comparison of the distribution of relative fitness values across conditions for each timepoint is shown. (C) Ward’s linkage clustering of the stress conditions based on strain day 6 relative fitness values, with agglomerative coefficient=0·76. Boxes denote four different k-mean clusters. DMSO=dimethylsulfoxide.

Figure 2:. Lineage patterns of antibiotic and metabolic relative fitness phenotypes of Mycobacterium tuberculosis clinical isolates

Dot plot of strain relative fitness values for the indicated conditions, grouped by lineage. Horizontal lines denote medians. Kruskal-Wallis p values adjusted by the Dunn’s multiple comparisons test indicated.

Figure 3:. Identification of genetic determinants associated with intermediate phenotypes and poor tuberculosis treatment outcomes with a GWAS

(A)Heat map of strain day 6 relative fitness values ordered according to the phylogeny shown in figure 1. Colour squares mark strains carrying mutations in the indicated genes returned as significant hits from the GWAS after multiple test correction; reported mutations have an allele frequency of >2%. Conditions are ordered according to the hierarchal clustering from figure 1C.(B) Percentage and fraction of strains associated with the indicated clinical outcomes, grouped by genotype for the indicated genes. Coloured dots indicate the associated intermediate phenotype as determined by the GWAS. Data from input-normalised relative fitness values are shown in (A) and (B). GWAS=genome-wide association study. OR=odds ratio.

Figure 4:. Link between phenogenotypes of Mycobacterium tuberculosis associated with poor tuberculosis outcomes and in-vivo fitness and transmission

(A) ATP concentrations in recombinant Mycobacterium smegmatis strains expressing the indicated M tuberculosis Rv1339 genotype. The solid line denotes the mean and the p values denote the results of an unpaired t test. Each point represents the mean of technical replicates for an independent experiment. (B) Review of published genome-wide LOF screens using the M tuberculosis reference strain H37Rv.^,, Fold change refers to the representation of Rv1339 LOF mutants in the input inoculum compared with the abundance in the indicated condition. q values indicate p values adjusted for multiple testing correction. (C) Violin plots of terminal branch lengths determined by SNPs for different clades of L1. p values from Kruskal-Wallistest and Dunn’s multiple comparison test in comparison to L1.1.1.1 are indicated. Black line denotes the median. (D) Phylogenetic tree of 952 clinical isolates representing the global diversity of L1.1.1.1 and L1.1.1 M tuberculosis clinical strains. LOF=loss-of-function. SNP=single-nucleotide polymorphism. WT=wild type.