Fig 1.
Hypothesis: A dual role of thymidylate biosynthesis in mycobacterium.
In mycobacteria, dUMP, the precursor for dTTP biosynthesis, is synthesized exclusively by dCTP deamination and coupled dUTP hydrolysis by Dcd:dut. dTTP is used not only for DNA replication by DNA polymerases but also by RmlA that uses this nucleotide for the first step of L-rhamnosyl biosynthesis, a critical step in cell wall biosynthesis. (Abbreviations: linker–L-rhamnosyl linker; Dcd:dut—bifunctional dCTP deaminase:dUTPase; TS–thymidylate synthase; RmlA—D-glucose-1-phosphate thymidylyltransferase; Pol–DNA polymerase).
Table 1.
Strains and plasmids used in the study.
Fig 2.
RmlA overexpression in M. smegmatis and its effect on cell growth.
A) Overexpression was evaluated by measuring mRNA levels of RmlA in WT and overexpressing strains (RmlAi–induced overexpression by the addition of 0.05 μg /ml TC, or RmlAc—constitutive overexpression). RmlA levels in RmlAc and RmlAi strains were 25 and 40 fold higher (p < 0.008) than in the WT M. smegmatis, respectively. B) WT, RmlAi, PknG(-), and PknG(-); RmlAi M. smegmatis strains were grown in liquid culture without any treatment or with 0.1 μg/ml TC. OD (600) was measured every 10 min in a plate reader.
Fig 3.
RmlAi cells are more prone to aggregation.
The indicated strains were grown in liquid culture and streaked onto microscopy slides as described in Methods. RmlA overexpression was induced by the addition of 0.2 μg/ml TC. The control strains do not carry the RmlAi plasmid.
Fig 4.
Effects of RmlA overexpression on the morphology of M. smegmatis.
A) RmlA overexpression in both WT and PknG(-) backgrounds resulted in cell elongation and the appearance of spherical, bubble-like structures breaking off the normal rod shape. B) High-resolution visualization of the morphological changes in the RmlAi; PknG(-) strain upon TC induction. The cell membrane was stained by Bodipy 522/529. PC denotes phase-contrast images.
Fig 5.
Cell elongation upon RmlA overexpression.
The normalized distribution of cell length is shown as bars. Gaussian curve fitting to the data is shown as smooth lines. The number of cells counted in each sample (n) is shown in the legend. RmlA overexpression was induced by 0.1 or 0.2 μg/ml TC (TC 0.1 and TC 0.2 respectively). TC 0 stands for non-induced controls. The highest probability cell length yielded by the Gaussian curves in each case is the following in the order of 0.2 μg/ml TC; 0.1 μg/ml TC and no TC: 2.02±0.03 μm; 2.10±0.02 μm; 2.06±0.08 μm for the WT; 1.98±0.01 μm; 2.22±0.01 μm; 2.20±0.01 μm for the PknG(-); 5.48±0.53 μm; 1.94±0.16 μm; 1.91±0.02 μm for the RmlAi; 3.17±0.20 μm; 1.99±0.01 μm; 2.01±0.02 μm for the PknG(-); RmlAi and 2.32±0.01 μm; 2.10±0.02 μm; 2.01±0.02 μm for the Lxi samples.
Fig 6.
The effect of RmlA overexpression on the cellular dNTP pool.
Concentrations of dNTPs in the extracts were measured according to [30]. The dTTP level in the RmlA overexpressing PknG(-) strain was significantly lower than that in the PknG(-) strain (p < 0.016), although it is unchanged when RmlA was overexpressed in the WT background. The cellular concentration of other dNTPs was not significantly changed in the RmlA overexpressing strains.
Fig 7.
Effects of CIP and EMB treatments on cell viability.
Treatments were performed on exponentially growing WT bacteria in liquid cultures for 8 h using 100 μg/ml EMB and 0.3 μg/ml CIP. CFU was counted on antibiotic-free agar plates. Concentrations were chosen so that growth inhibition was in the range of 20–80%.
Fig 8.
Effects of RmlA overexpression on the expression of different genes.
RadA and LexA were used as indicators for replication stress, while IniA and WhmD indicated cell wall biosynthesis stress. Dcd:dut, ThyA, and ThyX represent the thymidylate biosynthesis pathway, while RmlA-D belongs to the rhamnose biosynthetic pathway. PknG, known to regulate rhamnose biosynthesis, was also measured. The following treatments were used as positive controls: CIP for replication stress and EMB for cell wall biosynthesis stress. Expression levels were normalized to those of the WT strain. Changes in expression levels are shown as a heat map. Blue depicts downregulation, while red is for upregulation. Black star indicates p < 0.05 for the change in expression.
Fig 9.
The cellular localization pattern of fluorescently tagged RmlA.
To investigate the cellular localization of RmlA, we constructed M. smegmatis strains expressing GFP (A) and mOrange-2 (B) tagged RmlA under the control of the endogenous RmlA promoter. A) The localization pattern of the GFP tagged protein was observed using confocal microscopy. Of the 57 cells analyzed, all exhibited a helical fluorescence pattern here. DNA was stained with propidium iodide. B) The localization of mOrange-2 tagged protein was investigated using STED super-resolution imaging. We found that in all 68 cells analyzed, RmlA localized to the cell perimeter in a helix-like pattern, while no helicity was observed in the mOrange controls (mOrange panel in Fig 9B). Cell boundaries are indicated with a dashed line.