Deficiency of the novel exopolyphosphatase Rv1026/PPX2 leads to metabolic downshift and altered cell wall permeability in Mycobacterium tuberculosis

Abstract

Mycobacterium tuberculosis can persist for decades in the human host. Stringent response pathways involving inorganic polyphosphate [poly(P)], which is synthesized and hydrolyzed by polyphosphate kinase (PPK) and exopolyphosphatase (PPX), respectively, are believed to play a key regulatory role in bacterial persistence. We show here that M. tuberculosis poly(P) accumulation is temporally linked to bacillary growth restriction. We also identify M. tuberculosis Rv1026 as a novel exopolyphosphatase with hydrolytic activity against long-chain poly(P). Using a tetracycline-inducible expression system to knock down expression of Rv1026 (ppx2), we found that M. tuberculosis poly(P) accumulation leads to slowed growth and reduced susceptibility to isoniazid, increased resistance to heat and acid pH, and enhanced intracellular survival during macrophage infection. By transmission electron microscopy, the ppx2 knockdown strain exhibited increased cell wall thickness, which was associated with reduced cell wall permeability to hydrophilic drugs rather than induction of drug efflux pumps or altered biofilm formation relative to the empty vector control. Transcriptomic and metabolomic analysis revealed a metabolic downshift of the ppx2 knockdown characterized by reduced transcription and translation and a downshift of glycerol-3-phosphate levels. In summary, poly(P) plays an important role in M. tuberculosis growth restriction and metabolic downshift and contributes to antibiotic tolerance through altered cell wall permeability.

Importance: The stringent response, involving the regulatory molecules inorganic polyphosphate [poly(P)] and (p)ppGpp, is believed to mediate Mycobacterium tuberculosis persistence. In this study, we identified a novel enzyme (Rv1026, PPX2) responsible for hydrolyzing long-chain poly(P). A genetically engineered M. tuberculosis strain deficient in the ppx2 gene showed increased poly(P) levels, which were associated with early bacterial growth arrest and reduced susceptibility to the first-line drug isoniazid, as well as increased bacterial survival during exposure to stress conditions and within macrophages. Relative to the control strain, the mutant showed increased thickness of the cell wall and reduced drug permeability. Global gene expression and metabolite analysis revealed reduced expression of the transcriptional and translational machinery and a shift in carbon source utilization. In summary, regulation of the poly(P) balance is critical for persister formation in M. tuberculosis.

FIG 1

Recombinant Rv1026 protein exhibits hydrolysis activity against long-chain polyphosphate, which is inhibited by ppGpp. (A) Western blot showing detection of 6×His-Rv1026 (34.6 kDa) using Penta-His antibody. Lanes M, protein markers; E, elution from empty vector control strain after dialysis; P, elution fraction of 6×His-Rv1026 after dialysis. The positions of molecular mass markers (in kilodaltons) are shown to the left of the blot. (B) Exopolyphosphatase activity of recombinant Rv1026. Recombinant protein (5 µg/ml) was incubated with poly(P), specifically sodium hexametaphosphate (P6) (10 µg/ml), 45-mer poly(P) (P45) (10 µg/ml), and 700-mer poly(P) (P700) (50 µM [phosphate monomer]) for 8 h. (C) PPX activity of recombinant Rv1026 as a function of concentration and incubation time. (D) Inhibition of PPX activity of recombinant Rv1026 by ppGpp. Recombinant protein was incubated with P700 (50 µM [phosphate monomer]) for 6 h with (+) or without (−) ppGpp (8 µM). Values that are significantly different (P < 0.05) are indicated by an asterisk. Values are means ± standard deviations (SD) (error bars) from three experiments.

FIG 2

Intrabacillary accumulation of inorganic polyphosphate temporally coincides with M. tuberculosis growth restriction. (A to D) Intrabacillary poly(P) content was measured in wild-type M. tuberculosis during axenic growth in supplemented Middlebrook 7H9 broth (A), nutrient starvation (B), progressive hypoxia (C), and phosphate depletion (D) using a DAPI-based method and normalized to total protein content of extract lysate. Each data point represents the mean of three biological replicates. In the progressive hypoxia model in panel C, the x axis shows the days after change in color of the indicator dye, methylene blue, indicating bacterial entry into nonreplicating persistence stage 2 (mean ± SD). In panels A to D, the poly(P)/total protein ratio is shown on the left-hand x axes, and the optical density at 600 nm (O.D. 600) is shown on the right-hand x axes.

FIG 3

Conditional knockdown of ppx2 leads to M. tuberculosis poly(P) accumulation and growth restriction. (A) Immunoblot confirmation of Rv1026 knockdown strains compared to empty vector control strain. (B) Poly(P) content of ppx2 knockdown and empty vector strains. Values that are significantly different (P < 0.05) are indicated by an asterisk. Values are means ± standard deviations (error bars) from three experiments. (C) Growth curves of ppx2 knockdown and empty vector strains in supplemented Middlebrook 7H9 broth. Values that are significantly different (P < 0.05) compared to the value for the ppx2 knockdown strain are indicated by an asterisk. Values are means ± standard deviations (error bars) from three experiments. (D) Replication rate in ppx2 knockdown strain versus empty vector control strain. The percentage of “replication clock” plasmid-containing bacteria was determined by counting the number of CFU on Middlebrook 7H10 plates containing kanamycin (50 µg/ml) and dividing this number by CFU on nonselective 7H10 plates (*, P < 0.05; n = 3).

FIG 4

Polyphosphate accumulation contributes to M. tuberculosis phenotypic tolerance to isoniazid and stress resistance. (A) Logarithmically growing ppx2 knockdown and empty vector control strains were incubated with isoniazid (10 µg/ml) for 7 days. The bacteria were plated on Middlebrook 7H10 agar with or without isoniazid (INH) (1 µg/ml) to determine the number of drug-sensitive and drug-resistant CFU. Data are the means of three independent samples, and the numbers are the numbers of CFU of isoniazid-resistant (gray bar) or isoniazid-sensitive bacteria (white bar) (*, P < 0.05; n = 3). (B to D) Logarithmically growing cultures of empty vector control and ppx2 knockdown strains were incubated under various stress conditions. The stress conditions were as follows: 40°C for 24 h (B), 0.05% SDS for 4 and 6 h (C), and acidified Middlebrook 7H9 broth (pH 4.5) (D). In panels B and C, the survival ratio is the number of surviving bacteria after challenge divided by the number of bacteria prior to incubation. Values are means ± SD. Values that are significantly different (P < 0.05) compared to the value for the empty vector control strain are indicated by an asterisk.

FIG 5

Polyphosphate accumulation contributes to enhanced M. tuberculosis survival during infection of naive macrophages. (A) Naive macrophages were infected with the empty vector control strain and ppx2 knockdown strain. The numbers of CFU were determined at days 0, 1, 3, 5 and 7 after infection (*, P < 0.05 compared to the empty vector strain; n = 3). (B) Poly(P) accumulation alters cytokine and chemokine release by naive macrophages. Cytokines and chemokines released by naive macrophages were measured after 72 h of infection with empty vector or ppx2 knockdown strain (*, P < 0.05 compared to the empty vector strain; n = 3).

FIG 6

Metabolomics analysis of the ppx2 knockdown strain compared to the empty vector control strain during exponential growth. (A) Hierarchical clustering pathway analysis of ppx2 knockdown and empty vector strains. (B and C) Metabolites altered in the ppx2 knockdown strain compared to the empty vector strain include phosphate compounds (B) and components of the pentose phosphate and glucose utilization pathways (C). Values that are significantly different are indicated as follows: *, P < 0.05; + or †, 0.05 < P < 0.1; ‡, P < 0.005. TCA, tricarboxylic acid.

FIG 7

Polyphosphate accumulation results in decreased ethidium bromide accumulation, increased Nile red staining, and reduced biofilm formation. (A) Mid-log-phase cultures of empty vector control and ppx2 knockdown strains were incubated in PBST with 2 µg/ml ethidium bromide. The values at each time point are normalized to the time zero reading value (mean ± SD; *, P < 0.05; n = 3). RFI, relative fluorescence intensity. (B) Mid-log-phase cultures of each strain were incubated in PBS containing 20 µM Nile red stain (mean ± SD; *, P < 0.05; n = 3). (C) Each strain was incubated in Sauton’s medium lacking detergent for 5 weeks, and biofilms were assessed by crystal violet staining (*, P < 0.05; n = 3).

FIG 8

Polyphosphate accumulation is associated with increased cell wall thickness. (A to D) The empty vector control strain (A and C) and ppx2 knockdown strain (B and D) were evaluated by transmitted electronic microscopy during mid-log-phase growth. The black arrow indicates the cell wall layer. Bars, 100 nm. (E) Dot plot graph of the cell wall thickness in ppx2 knockdown and empty vector strains. The cell wall thickness in the ppx2 knockdown strain and empty vector strain were significantly different (P < 0.001).