Fig 1.
Growth of wild-type M. tuberculosis and of the ΔmbtD mutant with different iron sources.
Growth of wild-type Mtb mc26230 (A) and the ΔmbtD::hygr mutant (B) in self-made low-iron 7H9 medium supplemented with: 20 μM ammonium ferric citrate; 10 μM human holo-transferrin; 10 μM human holo-lactoferrin; 5 μM human hemoglobin; 20 μM hemin; 0.2 μM mycobactin (MBT) and 0.2 μM carboxymycobactin (cMBT), respectively. The Mtb strains were grown in self-made low-iron 7H9 medium for 5 days to deplete intracellular iron before growth analysis. The initial OD600 of all the cultures was 0.01. Error bars represent standard deviations from the mean results of biological triplicates.
Fig 2.
Construction of M. tuberculosis transposon libraries.
(A) The Mtb transposon mutagenesis library was generated from the siderophore-deficient mutant ML1600 (ΔmbtD::hygr). The library was cultured on low iron 7H9/ADS agar plates supplemented with (i) 500 ng/mL Fe-cMBT; (ii) 250 ng/mL Fe-MBT (high concentration); (iii) 50 ng/mL Fe-MBT (low concentration); (iv) 20 μM hemin; (v) 5 μM hemoglobin and (vi) 20 μM hemin plus 250 ng/mL Fe-MBT, respectively. The library DNA was isolated from the plates after three weeks. (B) Library building protocol. Genomic DNA pool was digested by HinP1I to generate GC-tails, allowing to the ligation of GC-tailed adapters which compose of short 3’ blocked and long oligonucleotides. The 3’ end of the shorter oligo was blocked (*) by C6-TFA-amino modification so that it could not be extended during amplification. The transposon-chromosome junctions were selectively amplified using primers (S2 Table) that recognize the end of the transposon and the 5’ end of the long adapter oligo. These primers contain all requisite sequences to permit direct multiplex sequencing of amplicons on an Illumina NextSeq 500 platform. (C) Transposon-chromosome junctions from high density mutant libraries were identified by deep sequencing. The six concentric circles in different colors represent the six libraries corresponding to (A). From the outer to the inner, they are (i) cMBT; (ii) high MBT; (iii) low MBT; (iv) hemin; (v) hemoglobin and (vi) hemin plus high MBT, respectively. The number of normalized sequence reads corresponding to each insertion site is represented as one color bars mapped onto the circular chromosome of M. tuberculosis H37Rv. Black contour represents the GC content of the chromosome (G+C content higher or lower than 50% are represented as contours outside or inside the ring, respectively). The genomic deletion regions such as ΔRD1, ΔpanCD and ΔmbtD are indicated in black. The genomic regions such as esx-3, irtA/B, mmpL5 and rv2047c are indicated in red. Nucleotide positions are indicated (million bases). Plots were generated using Circos 0.69–5 [43].
Fig 3.
TnSeq analysis of the M. tuberculosis transposon libraries.
A subset of 165 varying genes were identified by ANOVA. The mean normalized insertion count is calculated for each gene in each of the 6 conditions. Then a log-fold-change is calculated for each condition relative to the mean count across all the conditions. (A) The 165 varying genes were categorized by different physiological functions as annotated in TubercuList [44]. (B) Heatmap with 84 selected genes (See full heatmap in S1 Fig) showing variability in transposon insertion counts across iron-supplementation conditions. The color 'red' means the counts in one condition are lower than the other conditions on average, suggesting a greater requirement for that gene in that condition, and 'blue' means insertion counts are higher than average, suggesting it is less required. The dendrogram shows the hierarchical clustering of the conditions (columns) using complete-linkage clustering. Gene insertion count profiles (rows) are also clustered using the hclust package in R, and gene pathway associations are indicated. (C) Correlation plot between each pair of iron-supplementation conditions. The correlation coefficients between pairs of conditions are calculated between vectors of log-fold-changes for the varying genes. The size of each circle indicates the magnitude of the correlation, and the color indicates the sign, as shown by the scale. (D) Volcano plot of resampling analysis between hemin and high MBT conditions. TnSeq-FC- and false discovery rate-adjusted P values (q values) from the resampling test are plotted for each genetic locus. Loci meeting the significance threshold of a q value of <0.05 are colored. The genes with increased log-fold-change or decreased minus log-fold-change are predicted to have more essentiality in utilizing mycobactin or hemin, respectively. Selected mutants are indicated by name.
Fig 4.
Quantitative classification of the requirement of genes by M. tuberculosis for growth under different iron conditions.
Venn diagram of Mtb genes with ‘essential’ or ‘growth defect’ states from five iron conditions: cMBT (500 ng/mL), high MBT (250 ng/mL), low MBT (50 ng/mL), hemin (Hm) (20 μM) and hemoglobin (Hb) (5 μM). The number of overlapping genes in their respective zones is indicated.
Fig 5.
M. tuberculosis genomic regions involved in siderophore utilization.
The distribution of the transposon insertions in Mtb genomic regions including esx-3 operon (A); rv0455c gene, mmpS5-mmpL5 operon, irtA-irtB operon (B); and pks12-rv2047c operon (C). The iron conditions are marked with different colors. The y-axis (0, 10, 20) represents the counts of the Tn-insertions and x-axis represents the genomic position (bp). Potential TA dinucleotide insertions sites are indicated. Regions containing genes of interest are highlighted by red boxes. Plots were generated using MochiView 1.46 [46].
Table 1.
Genes with differential importance in iron utilization by Mycobacterium tuberculosis.
45 genes selected from the 165 varying genes which were assessed by HMM analysis as either essential (ES), “essential domain” (ESD) or “growth defect” (GD) under different iron conditions. The 21 genes which were classified as essential or “growth defect” under all iron conditions (S6 Table) were excluded from this list. cMBT: carboxymycobactin, MBT: mycobactin at high or low concentrations, hemin (Hm), hemoglobin (Hb) and hemin plus mycobactin (Hm/MBT). NE: non-essential.
Fig 6.
Growth of the M. tuberculosis Δrv2047c mutant.
Growth of Mtb mc26230 (as WT), Δrv2047c (ML2256) and the complementation Δrv2047c mutant (ML2257) in 7H9 Middlebrook medium (A) supplemented with 1 μM carboxymycobactin (B) and 1 μM mycobactin (C), respectively. The initial OD600 of all the cultures is 0.01. Error bars represent standard deviations from the mean results of biological triplicates.
Fig 7.
Growth of the M. tuberculosis mmpL4 and mmpL5 mutants under different iron conditions.
Growth curves of Mtb strains including Mtb mc26230 (as WT), ΔmmpL4 (ML2300), ΔmmpL5 (ML2301), ΔmmpL4/L5 (ML2302), ΔmmpS4/S5 (ML859) and ΔmbtD (ML1600) in low-iron 7H9 medium (A) supplemented with 10 μM hemin (B), 1 μM carboxymycobactin (C) and 1 μM mycobactin (D), respectively. The Mtb strains were grown in self-made low-iron 7H9 medium for 5–7 days to deplete intracellular iron before growth assays. The initial OD600 of all the cultures is 0.01. Error bars represent standard deviations from the mean results of biological triplicates.
Fig 8.
M. tuberculosis genes involved in heme and hemoglobin utilization.
(A) The distribution of the transposon insertions in Mtb genomic regions including rv1085c, ppe37 and smtB-zur operon. The iron conditions are marked with different colors. The y-axis (0, 10, 20) represents the counts of the Tn-insertions and x-axis represents the genomic position (bp). Potential TA dinucleotide insertions sites are indicated. Regions containing genes of interest are highlighted by red boxes. Plots were generated using MochiView. (B-D) Growth assays of wild-type Mtb mc26206, ΔsmtB-zur deletion mutant (ML2277), smtB complementation strain (ML2278), zur complementation strain (ML2279), and smtB-zur complementation strain (ML2280) under low-iron 7H9 medium containing 10 μM 2,2’-dipyridyl (DIP) (B) supplemented with 20 μM hemin (C) and 5 μM hemoglobin (D), respectively. The Mtb strains were grown in self-made low-iron 7H9 medium for 7 days to deplete intracellular iron before growth assays. The initial OD600 of all the cultures is 0.01. Error bars represent standard deviations from the mean results of biological triplicates.