Polyphosphate Deficiency in Mycobacterium tuberculosis Is Associated with Enhanced Drug Susceptibility and Impaired Growth in Guinea Pigs

Abstract

Inorganic polyphosphate (polyP), a linear polymer of hundreds of phosphate residues linked by ATP-like phosphoanhydride bonds, is found in all organisms and performs a wide variety of functions. This study shows that polyP accumulation occurs in Mycobacterium tuberculosis upon exposure to various stress conditions. M. tuberculosis possesses a single homolog of ppk-1, and we have disrupted ppk-1 in the M. tuberculosis genome by allelic replacement. The mutant strain exhibited negligible levels of intracellular polyP, decreased expression of sigF and phoP, and reduced growth in the stationary phase and displayed a survival defect in response to nitrosative stress and in THP-1 macrophages compared to the wild-type strain. We report that reduction in polyP levels is associated with increased susceptibility of M. tuberculosis to certain TB drugs and impairs its ability to cause disease in guinea pigs. These results suggest that polyP contributes to persistence of M. tuberculosis in vitro and plays an important role in the physiology of bacteria residing within guinea pigs.

INTRODUCTION

Mycobacterium tuberculosis leads to approximately 8 million new cases of active tuberculosis (TB) and 2 million deaths each year (1). In addition to these, one-third of the world population is estimated to be latently infected with M. tuberculosis. These individuals act as reservoirs, develop disease by reactivation of latent bacilli, and can infect people in their surroundings. Efforts to combat TB are further aggravated by the emergence of drug-resistant strains of M. tuberculosis and HIV-TB coinfections (2). The emergence and spread of drug-resistant TB have not been prevented by current TB therapy. The low efficiency of TB therapy has been attributed to multiple factors, such as noncompliance of patients, high bacterial loads, reduced metabolic rates of the organism, and the emergence of persisters or drug-tolerant bacteria due to their adaptability to the local microenvironment in the host lungs (3–6). A number of pathways, including energy production, stringent response, trans-translation pathways, proteasomal protein degradation, toxin-antitoxin modules, efflux pumps, or switching to utilization of cholesterol and fatty acids as a carbon source have been implicated as contributing to M. tuberculosis persistence (7–14).

M. tuberculosis encounters a variety of insults, such as low oxygen and nutritional stress in the human lungs, and as a mechanism to adapt to these changing environmental conditions, the bacterium enters into a nonreplicating or slowly replicating state characterized by antibiotic tolerance. This dormant or persistent bacillus ultimately causes latent infection, which has the potential to reactivate into active disease. The stringent response is one of the regulatory mechanisms by which bacteria adapt to poor nutrient conditions through the production of various alarmones, such as guanosine pentaphosphate [(p)ppGpp]. In several bacteria, levels of (p)ppGpp are maintained by the synthetic activity of RelA and SpoT and the hydrolytic activity of SpoT (15). (p)ppGpp regulates numerous cellular processes, such as replication, transcription, and translation and the cell's virulence and persistence (16–19). The ability to fine-tune the levels of (p)ppGpp has been shown to enhance the survival of M. tuberculosis under nutrient-starved conditions in mice and guinea pigs (8, 11, 20). Another important player in adaptation of bacteria to various stress conditions is polyphosphate (polyP). In bacteria, some of the enzymes involved in polyP metabolism are polyphosphate kinase 1 (PPK-1), which catalyzes the reversible transfer of the terminal (γ) phosphate of ATP to form polyP, and polyphosphatase (PPX), which processively hydrolyzes the terminal residues of polyP to liberate P_i (21–23). Polyphosphate kinase 2 (PPK-2) is another enzyme involved in polyP metabolism that drives synthesis of GTP and ATP using polyP as a phosphate donor (24–26).

PolyP serves as reservoir of energy and phosphate and a chelator of metal ions and plays an important role in imparting microbial resistance to different stress conditions (27–36). PolyP levels have been shown to regulate expression of Escherichia coli rpoS, which codes for the sigma factor responsible for adaptation of bacteria to the stationary phase (37). It also enables the bacteria to adapt to nutritional downshift by stimulating protein degradation by Lon protease, thereby providing the amino acids needed to respond to starvation (38). PolyP-deficient strains of various organisms have been shown to be defective for survival in the stationary phase and are more sensitive to either heat, acidic pH, UV radiation, or oxidative stress (28, 30, 38–42). PolyP-deficient strains have also been shown to be defective in quorum sensing, biofilm formation, motility, growth, surface attachment, and virulence (31, 43–46). Deletion of ppk1 in the genome of Shigella and Salmonella results in reduced invasiveness of epithelial cells, loss of viability, and increased sensitivity to polymyxin B sulfate (28).

The genome of M. tuberculosis codes for enzymes involved in both polyP synthesis (PPK-1 [Rv2984]) and its utilization (Rv3232c, Rv0496, and Rv1026) (47). A ppk-1 mutant strain of Mycobacterium smegmatis has been shown to be compromised in its ability to survive under oxidative stress, surface stress, and anaerobic conditions (48). In another study, it has been shown that downregulation of ppk-1 in M. tuberculosis using the IPTG (isopropyl-β-d-thiogalactopyranoside) antisense-inducible system results in a bacteriostatic effect for nearly 15 days, followed by a rapid bactericidal effect, thereby suggesting that ppk-1 is essential for M. tuberculosis growth in vitro and a potential drug target to combat tuberculosis (49). Rv3232c (ppk-2) has also been shown to be widely conserved in a number of human pathogens, regulates nucleotide diphosphate kinase activity, and is important for intracellular survival in mycobacteria (26). Rv0496 (ppx) has also been shown to possess polyphosphatase activity, showing a distinct preference for short-chain polyP as the substrate (50). In the same study, it was also reported that Rv1026 lacks exopolyphosphatase activity but possesses a modest ATPase/ADPase activity (50). A ppx mutant strain of M. tuberculosis displayed a significant growth defect in axenic cultures as well as a survival defect in activated macrophages and guinea pigs (51).

In the present study, we show that accumulation of polyP occurs in late-log-phase and stationary-phase cultures of mycobacteria in vitro. We report accumulation of polyP upon exposure to various stress conditions as well as to various TB drugs. We found that polyP accumulation is dependent on (p)ppGpp levels in M. tuberculosis. In contrast to earlier studies, we report that PPK-1 is nonessential for M. tuberculosis growth in vitro and that the mutant strain displayed a growth defect at later stages of growth and under nitrosative stress. We also show that polyP deficiency is associated with increased susceptibility of M. tuberculosis to various TB drugs as well as impairing its ability to survive in macrophages and guinea pigs.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

The strains and plasmids used in this study are described in Table 1. The Escherichia coli XL-1 Blue and HB-101 strains were used for cloning purposes. E. coli strain BL21(λDE3, plysS) was used for protein expression and purification studies. Culturing of E. coli and mycobacterial strains was carried out in Luria-Bertani (LB) and Middlebrook (MB) medium, respectively, as per standard protocols. Optical densities at 600 nm (OD₆₀₀) of 0.2, 0.5, 2.0, 2.5, and 3.0 were considered early log phase, mid-log phase, late log phase, stationary phase, and late stationary phase, respectively. All chemicals, drugs, and antibiotics used in this study were procured from Sigma-Aldrich (St. Louis, MO). PA-824 was a kind gift received from Clifton E. Barry (NIAID, NIH). The concentrations of drugs used in the study for RNA extraction and persistence experiments were 4 μg/ml, 0.4 μg/ml, or 0.04 μg/ml rifampin (Rif), 10 μg/ml isoniazid (Inh), 10 μg/ml levofloxacin (Levo), 10 μg/ml gentamicin (Gm), and 10 μg/ml ethambutol (Eth). For polyP isolation, the concentrations of drugs used in the study were either 10×, 1×, or 0.1× the MIC₉₉ values.

Table 1.

List of strains and plasmids used in this study

Strain or plasmid	Description	Source or reference
Strains
H37Rv	Laboratory strain (ATCC 27294) of M. tuberculosis	ATCC
Δppk-1 mutant	ppk-1 mutant strain of M. tuberculosis	This work
Δppk-1 attB::ppk-1 mutant	ppk-1::hyg complemented with ppk-1 at attB site	This work
ΔrelA mutant	relA mutant strain of M. tuberculosis	Kind gift from Clifton E. Barry (8)
Plasmids
pYUB854	Cloning vector, Hygr	52
pYUB-Δppk1	pYUB854 vector carrying upstream and downstream regions of ppk-1, hygromycin cassette	This work
phAE87	Phagemid DNA, Ampr	52
phAE87-Δppk1	Phagemid DNA carrying upstream and downstream regions of ppk-1, hygromycin cassette	This work
pMV306K	E. coli mycobacterium shuttle vector, Kanr	53
pMV306K-ppk1	pMV306K carrying ppk-1 along with its upstream regions.	This work
pET28b	E. coli T7-based expression system	Novagen
pET-28b-ppk1	pET28b carrying ppk-1	This work
pTE-mcs-1	E. coli-Mycobacterium shuttle vector, Hygr, Atc inducible in M. tuberculosis	Kind gift from Sabine Ehrt (70)
pTE-ppk1	pTE-mcs-1 carrying ppk-1	This work

Cloning, expression, purification, and characterization of PPK-1 enzyme.

The ppk-1 gene of M. tuberculosis was PCR amplified and cloned into pET28b, resulting in pET28b-ppk1. The recombinant protein was purified by affinity chromatography using Ni-nitrilotriacetic acid (NTA) resin (Qiagen GmbH; Hilden, Germany). Purified protein was dialyzed, concentrated using a 30-kDa-cutoff Amicon-ultra filter (Millipore), and used for raising antibodies in rabbits (Enzene Biosciences, Hyderabad) and polyP quantification assays. The PPK-1 reverse enzymatic reaction was performed in the presence of 50 mM Tris (pH 7.4), 40 mM (NH₄)₂SO₄, 4 μM MgCl₂, 5 μM ADP, and 1 μM His₆-tagged PPK-1 enzyme. For PPK-1 kinetic experiments, the initial velocities of ATP generation were plotted against various concentrations of either polyP₄₅, polyP₁₇, or polyP₃ and analyzed using nonlinear regression to the Michaelis-Menten (MM) equation. The kinetic constants V_max, K_m, and k_cat/K_m for any given reaction were determined from the plotted data.

PolyP quantification using PPK-1 in the reverse direction.

To study cell density-dependent accumulation of polyP, mycobacteria were grown in MB 7H9 medium and harvested at different growth stages. For oxidative stress, mid-log-phase cultures were exposed to 5 mM H₂O₂ for 6 or 20 h. For acidic stress, bacteria were harvested, washed twice, and resuspended in acidic medium (pH 5.2) for 6 or 20 h. For nitrosative stress, mid-log-phase bacilli were harvested, washed with acidic medium, and exposed to 5 mM NaNO₂ for 6 or 20 h. For nutritional stress, mid-log-phase cultures were harvested, washed twice with 1× Tris-buffered saline–Tween 80 (TBST), and resuspended in TBST for 3 days. For polyP isolation, bacteria were harvested, washed twice with 1× phosphate-buffered saline (PBS), and resuspended in 500 μl of GITC lysis buffer (4 M guanidine isothiocyanate, 50 mM Tris [pH 8.0]). Cells were lysed either by bead beating or by sonication, and polyP was extracted using Glassmilk (Geneclean kit; MP Biomedicals) as per standard protocols (39). PolyP levels were assayed using the PPK-1 reverse reaction as described above. ATP levels in the enzyme reaction were measured by mixing 50 μl of the enzyme reaction with an equal volume of BacTiter-Glo reagent, and luminescence was measured using a 96-well plate reader (Promega Corporation, Madison, WI). The amount of ATP generated in the enzymatic reaction was normalized to the amount of protein in the samples; polyP levels are depicted as pmol/min/mg.

ATP measurement assays.

Intracellular ATP levels were determined using the BacTiter-Glo microbial cell viability kit (Promega Corporation). Mid-log-phase cultures were exposed to various stress conditions as described above. At 6 and 20 h postexposure, aliquots of bacterial cultures were collected, immediately inactivated at 90°C for 30 min, and stored at −80°C before use. Intracellular ATP levels were measured by mixing 25 μl of cell suspension with an equal volume of the BacTiter-Glo reagent and measuring the luminescence using a 96-well plate reader. The ATP levels were normalized to the OD₆₀₀ of the samples.

Extraction of RNA and qRT-PCR assay.

For quantitative reverse transcription (qRT)-PCR analysis, M. tuberculosis was grown up to the mid-log phase and subsequently exposed to various drugs and oxidative stress for 6 h. For nitrosative stress, bacilli were harvested, resuspended in acidic medium, and exposed to 5 mM NaNO₂ for 6 h. For nutritional stress, mid-log-phase bacilli were harvested and washed and resuspended in 1× TBST for 3 days. The bacterial pellets were resuspended in 1.0 ml TRIzol (Invitrogen Corporation, Carlsbad, CA); mRNA was extracted as per standard protocols and cleaned using RNAeasy columns (Qiagen GmbH, Hilden, Germany). Extracted mRNA (1 μg) was treated with DNase I using the Ambion DNA-free kit (Invitrogen Corporation, Carlsbad, CA). First-strand cDNA synthesis was carried out using random primers and Superscript III reverse transcriptase (Invitrogen Corporation, Carlsbad, CA). Real-time quantitative PCR was performed on the 7500 Fast real-time PCR system (Applied Biosystems, Austin, TX). mRNA levels of ppk-1 (Rv2984), ppk-2 (Rv3232c), relA (Rv2583c), and ppx (Rv0496) were quantified and normalized to transcript levels of the housekeeping gene sigA using gene-specific TaqMan probes (Invitrogen Corporation, Carlsbad, CA). Transcript levels of various sigma factors and response regulators in the wild-type and ppk-1 mutant strains were quantified using gene-specific primers and SYBR green mix (Invitrogen Corporation, Carlsbad, CA). The data obtained were normalized to the transcript levels of sigA. The sequence of primers used for cDNA amplification is listed in Table S1 in the supplemental material.

Generation of ppk-1 mutant and complemented strains of M. tuberculosis.

A mutant strain of M. tuberculosis lacking the polyP synthetic activity associated with PPK-1 was generated by homologous recombination using temperature-sensitive mycobacteriophages (52). To generate the mutant strain, ∼800-bp upstream and downstream flanking regions of ppk-1 were PCR amplified and cloned into pYUB854 flanking the hygromycin cassette. For generation of the complemented strain, ppk-1 was PCR amplified along with its upstream region (∼200 bp) and cloned into pMV306K (53). The recombinant plasmid, pMV306K-ppk1, was electroporated into the ppk-1 mutant strain resulting in the ppk-1 complemented strain.

In vitro susceptibility of the wild-type, ppk-1 mutant, or ppk-1 complemented strain to various stress conditions.

The wild-type, ppk-1 mutant, and ppk-1 complemented strains were grown to the early log stage. To measure susceptibility to oxidative stress, cultures were diluted and incubated with 5 mM H₂O₂ for 24 h at 37°C. To test susceptibility to nitrosative stress, cultures were harvested and incubated at pH 5.5 with or without 5 mM NaNO₂ for either 3 days or 7 days at 37°C. For bacterial enumeration, 10-fold serial dilutions were prepared in MB 7H9 medium and 100 μl was plated in duplicates on MB 7H11 plates.

EM analysis of various strains of M. tuberculosis.

The wild-type, ppk-1 mutant, and ppk-1 complemented strains were grown in MB 7H9 medium until the late log phase. Cultures were harvested by centrifugation, and pellets were washed with 0.1 M phosphate buffer and fixed overnight in 4% paraformaldehyde–2.5% glutaraldehyde in 100 mM sodium cacodylate buffer (pH 7.4). Samples were fixed in 1% osmium tetraoxide, dehydrated in acetone, and embedded in araldite CY212. For electron microscopy (EM) analysis, thin sections were cut and mounted onto 300-mesh copper grids. Sections were stained with alcoholic uranyl acetate and alkaline lead citrate, washed gently with distilled water, and observed under a Morgagni 268D transmission electron microscope (Fei Company).

Infection of THP-1 human macrophages with various strains of M. tuberculosis.

THP-1 monocytes were cultured in RPMI medium containing 10% fetal bovine serum (FBS) (Invitrogen Corporation, Carlsbad, CA). THP-1 monocytes were differentiated into macrophages by addition of 15 nM phorbol 12-myristate 13-acetate (PMA). Cells were seeded in a 12-well plate for infection at a cell density of 2 × 10⁵ per well and infected with either the wild-type, ppk-1 mutant, or ppk-1 complemented strain (in triplicates) at a multiplicity of infection (MOI) of 1:5 macrophage to bacteria for 4 h. Following infection, extracellular bacteria were removed by addition of amikacin (200 μg/ml) for 2 h. Subsequently, macrophages were washed twice with antibiotic-free RPMI medium and overlaid with RPMI medium supplemented with 10% fetal bovine serum (FBS). At the designated time points (days 0, 2, 4, and 6 postinfection), macrophages were lysed with 1 ml PBS–0.1% Triton X-100, and bacterial enumeration was performed as described above.

In vitro persistence experiments.

Various strains were grown to early log phase and diluted to an OD₆₀₀ of 0.1 in MB 7H9 medium containing drugs. To investigate the role of polyP levels in formation of persisters at later stages of growth, various strains were grown to the late log phase and subsequently exposed to various drugs. After drug exposure, bacilli were harvested, washed twice with antibiotic-free medium, and resuspended in 1 ml of MB 7H9 medium, and bacterial enumeration was performed as described above.

Determination of MIC₉₉ values.

Drugs were dissolved in dimethyl sulfoxide (DMSO) to make 50 mM stock solutions. Drugs (100-μl solutions) were added to the first row of the 96-well plate at a final concentration of 100 μM. Two-fold serial dilutions were made, and 11 dilutions of each drug (from 50 μM to 0.04875 μM) were tested for antimycobacterial activity. Row 12 of the 96-well plates was used as the no-drug control. For MIC₉₉ determination, various strains were grown to the early log phase and diluted 1:1,000 in MB 7H9 medium before being aliquoted at 50 μl into each well of a 96-well plate. The plates were incubated at 37°C for either 7 days for MIC₉₉ determination, using the alamarBlue method as per standard protocols, or 14 days where growth inhibition was measured macroscopically using an inverted plate reader. MIC₉₉ value determination using the alamarBlue method was done two independent times and by the inverted plate reader method three independent times.

Influence of intracellular polyP levels on the growth of M. tuberculosis in guinea pigs.

Pathogen-free outbred female guinea pigs (Hartley strain) in the weight range 200 to 300 g were procured from the Disease-Free Small Animal House Facility, Chaudhary Charan Singh Haryana Agricultural University, Hisar, India. The animals were maintained at the International Centre of Genetic Engineering and Biotechnology Tuberculosis Aerosol Challenge Facility (TACF) (biosafety level 3 [BSL3]) and routinely cared for in accordance with the guidelines provided by the Committee for the Purpose of Control and Supervision on Experiments on Animals (CPCSEA; Government of India). The animal experiments were approved by the Animal Ethics Committee of the International Centre for genetic engineering and biotechnology, New Delhi, India (approval no. ICGEB/AH/2012/1/TACF-THSTI-2).

Guinea pigs were infected by the aerosol route using a Madison aerosol chamber. Animals were sacrificed (n = 6) at 28 days and 70 days postinfection for determination of splenic and lung bacterial loads and histopathological analysis. For histological studies, lung tissues were fixed in 4% neutral buffered formalin, embedded in paraffin wax, sectioned, and stained with hematoxylin and eosin (HE). The tissue samples were coded and evaluation of lung sections for granulomatous organization was performed by a pathologist having no prior knowledge of the experimental groups. The whole section of lung was examined, and the number of granulomas was counted. Each granuloma was graded by the following criteria. (i) Granulomas with necrosis were given a score of 5, (ii) granulomas with centrally placed epithelioid cell clusters but no necrosis were given a score of 2.5, and (iii) granulomas with fibrous connective tissue surrounding lymphocytes and epithelioid cells were given a score of 1 (54). The total granuloma score was obtained by multiplying the number of granulomas of each type by the score and then adding them up to obtain a total granuloma score for each sample.

Statistical analysis.

Graph Pad Prism 5 software (version 5.01; GraphPad Software, Inc., CA) was used for the generation of graphs. Differences were considered statistically significant with a P value of <0.05.

RESULTS

Biochemical characterization of M. tuberculosis PPK-1 enzyme.

M. tuberculosis PPK-1 was expressed and purified as an NH₂-terminus 84-kDa His-tagged fusion protein using Ni-NTA chromatography (see Fig. S1A in the supplemental material). To understand the effect of the length of phosphate residues in polyP on the kinetic parameters of the reverse reaction of M. tuberculosis PPK-1, we determined the steady-state kinetic parameters K_m, V_max, and K_cat/K_m for PPK-1 using various substrate concentrations of polyP₄₅, polyP₁₇, or polyP₃. Purified enzyme displayed a substrate preference for long-chain polyP compared to short-chain polyP (see Fig. S1B). The apparent K_m for ADP was 3 μM (see Fig. S1C).

Intracellular polyP quantification under various growth and stress conditions.

In our in vitro polyP quantification assays, 3- to 4-fold-higher accumulation of polyP was observed in mid-log- and late-log-phase cultures compared to early-log-phase cultures of M. smegmatis (Fig. 1A). Approximately 18- to 20-fold-higher accumulation of polyP was observed in stationary-phase cultures compared to early-log-phase cultures (Fig. 1A; P < 0.05). A similar pattern of polyP accumulation in the stationary phase was also observed in M. tuberculosis (Fig. 1B; P < 0.01). However, a decline in polyP levels was observed in the late stationary phase as reported earlier in M. tuberculosis and Bacillus cereus (32, 51). Since mycobacteria encounter a number of adverse stress conditions in the host, such as nitrosative, oxidative, nutritional, and drug-induced stress, polyP levels were quantified from mycobacteria exposed to various stress conditions for either 6 or 20 h (Fig. 1C). The accumulation of polyP was higher in bacteria exposed to various stresses for 20 h than that in bacteria exposed for 6 h (Fig. 1C). Exposure of mycobacteria to acidic, oxidative, nitrosative, or nutritional stress for 20 h led to 19-fold, 17-fold, 40-fold, or 15-fold greater accumulation of polyP, respectively, compared to that in untreated cells (Fig. 1C; P < 0.05). Exposure of mycobacteria to various drugs with different modes of action, such as Rif (transcription inhibitor), Gm (translational inhibitor), or Levo (replication inhibitor), at either 10×, 1×, or 0.1× MIC₉₉ values also led to 10- to 20-fold-higher accumulation of polyP compared to that in untreated samples (Fig. 1D; P < 0.05).

Fig 1.

(A and B) Intracellular polyP quantification at different stages of growth. PolyP was quantified from M. smegmatis (A) and M. tuberculosis (B) growing in MB 7H9 medium during different growth stages as described in Materials and Methods. Values were normalized to the amount of protein in the sample and are expressed as pmol/min/mg. The data shown are mean values ± standard errors (SE) obtained from three independent experiments. *, P < 0.05; **, P < 0.01. (C and D) PolyP accumulation in cells exposed to various stress conditions or various drugs. Mid-log-phase growing cultures were exposed to various stress conditions (C) or drugs (D). PolyP levels in bacteria exposed to various conditions and untreated samples were quantified and normalized to the amount of protein. The data depicted are the ratio of polyP levels in treated samples with respect to (w.r.t.) those in untreated samples and are mean values ± SE of three independent experiments. *, P < 0.05.

Next, we sought to understand whether these elevated levels of polyP could be attributed to higher mRNA levels of ppk-1 (Fig. 2A). By qRT-PCR analysis, we observed that ppk-1 was upregulated by 2-fold in the late log phase and by 3-fold upon exposure of M. tuberculosis to nitrosative stress. Despite accumulation of polyP under oxidative and nutritional stress, we did not observe any upregulation of ppk-1 at the mRNA level. The transcript level of ppk-1 was also observed to be upregulated by 4- to 6-fold when M. tuberculosis was exposed to a different concentration of Rif (Fig. 2A). We did not observe any upregulation of ppk-1 in bacilli exposed to Levo, Gm, Inh, or Eth (Fig. 2A). To gain further insight into the mechanism for accumulation of intracellular polyP, we measured ATP levels in M. tuberculosis exposed to various stress conditions or drugs for either 6 h or 20 h. In our experiments, the maximum reduction in ATP levels was observed when M. tuberculosis was exposed to nitrosative stress for 20 h, a condition in which polyP accumulation was highest compared to the level in untreated bacilli (Fig. 2B). As shown in Fig. 2B, we also observed that ATP levels were increased by approximately 2-fold when M. tuberculosis was exposed to nutritional stress for 6 or 20 h compared to that in untreated cells. No significant difference was observed when M. tuberculosis was exposed to oxidative stress or Rif or Gm for 6 h. However, approximately 2-fold downregulation of ATP levels was observed when M. tuberculosis was exposed to oxidative stress, Rif for 20 h, and Levo for 6 h and 20 h compared to the effect in untreated cells (Fig. 2B).

Fig 2.

(A) qRT-PCR analysis of ppk-1 under various stress conditions. mRNA was isolated from M. tuberculosis cultures grown to the mid-log phase and exposed to various stress conditions. qRT-PCR was performed, and the relative expression of ppk-1 was calculated after normalization to transcript levels of the housekeeping gene sigA. A standard curve ranging from 0.15 pg to 1.5 ng genomic DNA was included in the PCR for calculation of transcript numbers. The data depicted represent the mean values ± SE obtained from three independent experiments. (B) Intracellular ATP levels in bacteria exposed to various stress conditions or drugs. Mid-log-phase growing cultures were exposed to various stress conditions or drugs. ATP levels were measured by using the Promega Bactiter-Glo microbial cell viability kit. ATP levels were normalized to the OD₆₀₀ of the samples. The data depicted are the ratio of ATP levels in treated samples to those in untreated samples and are mean values ± SE of two independent experiments.

With disruption of ppk-1 in the M. tuberculosis genome, the ppk-1 mutant displays a marginal growth defect in the stationary phase.

Earlier studies suggest PPK-1 is nonessential in M. smegmatis, but it might be essential for M. tuberculosis growth in vitro (48, 49). However, we were able to generate the ppk-1 mutant strain of M. tuberculosis using temperature-sensitive mycobacteriophages, in which the PPK-1 open reading frame (ORF) has been replaced by the hygromycin cassette (Fig. 3A). Southern blot analysis showed the presence of a 1.8-kb band in the parental strain, whereas a 4.0-kb band was observed in the mutant strain (Fig. 3B). Western blot analysis using polyclonal sera revealed the absence of an ∼84.0-kDa PPK-1 protein in the mutant strain (see Fig. S2A in the supplemental material). Integration of the wild-type copy of ppk-1 in the complemented strain was confirmed by Southern blotting and immunoblot analysis (Fig. 3B; see Fig. S2A). However, in Western blots, a band corresponding to 72 kDa was also observed in all lanes, which might be due to a nonspecific interaction between a bacterial protein and polyclonal sera. In our polyP quantification assay, levels of polyP were nondetectable in the ppk-1 mutant strain, and these levels were restored in the complemented strain (Fig. 3C).

Fig 3.

(A) Schematic representation of the ppk-1 locus in the wild-type (WT) and ppk-1 mutant (MT) strains of M. tuberculosis. Details of the ppk-1 locus in the genomes of WT and MT strain of M. tuberculosis are shown. The loci gpdA2 and Rv2983 are genes upstream of ppk-1, and mutT1 is a gene downstream of ppk-1. In the MT strain, the region containing ppk-1 has been replaced by the hygromycin cassette allelic exchange substrate (AES) using the temperature-sensitive mycobacteriophage phAE87. (B) Southern blot analysis of the WT, MT, and complemented (CT) strains of M. tuberculosis. Genomic DNA was isolated from various strains of M. tuberculosis using the cetyltrimethylammonium bromide (CTAB) method. Five micrograms of genomic DNA was digested with XhoI, resolved on 1.2% agarose gel, transferred to Hybond N membrane using the Turboblotter kit, and probed with the digoxigenin (DIG)-labeled probe as shown in panel A. (C) Quantification of polyP levels in WT, MT, and CT strains of M. tuberculosis. Various strains were grown to the late log phase, and intracellular polyP levels were estimated and normalized to the amount of protein; the polyP levels and are depicted as pmol/min/mg. The data shown are means ± SE obtained from three different experiments. *, P < 0.05.

Inability to accumulate polyP has also been shown to be associated with impaired bacterial survival in the stationary phase and in response to oxidative, osmotic, and thermal stress (30, 37, 39, 42, 55). We further evaluated whether polyP accumulation enables M. tuberculosis to survive better under stress conditions. We did not observe any difference in the growth of the ppk-1 mutant strain versus the wild-type strain until the late log stage of growth (Fig. 4A and B). However, the mutant strain displayed a marginal growth defect during the late log and stationary phases of growth in MB 7H9 medium (Fig. 4A and B). This growth defect was found to be statistically significant and was restored in the complemented strain (P < 0.05). We did not observe any significant alterations in the colony morphologies of the wild-type and ppk-1 mutant strains (see Fig. S2B in the supplemental material). To evaluate whether polyP levels contribute to the cell wall architecture of M. tuberculosis, the cell envelopes of various strains were analyzed by transmission electron microscopy (TEM). TEM analysis revealed that the inner membrane, outer membrane, and periplasmic space were visible in all strains, thereby suggesting that polyP does not contribute to maintaining the cell wall structure of M. tuberculosis (Fig. 4C).

Fig 4.

(A and B) Influence of ppk-1 deletion on M. tuberculosis growth in vitro. The growth phenotype of WT, MT, or CT strains was monitored either by measuring OD₆₀₀ (A) or by CFU determination (B). *, P < 0.05. (C) Transmission electron microscopic analysis of sections of WT, MT, or CT strains of M. tuberculosis. The electron microscopic pictures of various strains shown are typical for the appearance of bacteria in different fields. The scales for corresponding images are identical.

PolyP deficiency increases the sensitivity of M. tuberculosis to nitrosative stress and attenuates its growth in macrophages.

We compared the survival of the wild-type, ppk-1 mutant, and complemented strains under either acidic, nitrosative, or oxidative stress. The mutant strain survived at levels comparable to the parental strain upon exposure to acidic medium (Fig. 5A). However, upon exposure to nitrosative stress for 7 days, 6-fold more killing of the mutant strain was observed compared to that of the wild-type strain. This growth defect was restored in the complemented strain (Fig. 5A). However, no significant differences were observed in the survival of the wild-type and mutant strains after 3 days of exposure to nitrosative stress (data not shown). We did not observe any difference in survival rates of the wild-type and mutant strains upon exposure to oxidative stress for 24 h (Fig. 5A). These results suggest that polyP accumulation in M. tuberculosis contributes to its survival under nitrosative stress but is not essential for its survival in acidic and oxidative stress. In order to evaluate whether depletion in polyP levels is associated with impaired survival of M. tuberculosis within human macrophages, THP-1 macrophages were infected with the wild-type, ppk-1 mutant, or complemented strain. Both the wild-type and complemented strains displayed similar growth kinetics in macrophages; however, the mutant strain exhibited an intracellular growth defect by 5-fold compared to the wild-type and complemented strains in THP-1 macrophages after day 2 postinfection (Fig. 5B). This growth defect was further magnified at later time points, and a 12-fold difference in the CFU between the wild-type and ppk-1 mutant strains was observed at day 6 postinfection (Fig. 5B; P < 0.01). These observations imply that PPK-1 is required for the ability of M. tuberculosis to survive in human macrophages.

Fig 5.

(A) Susceptibility of various strains to acidic, oxidative, and nitrosative stress. Various strains were grown until an OD₆₀₀ of 0.2 and subsequently exposed to either acidic or nitrosative stress for 7 days or oxidative stress for 24 h. Bacterial enumeration was performed as described in Materials and Methods. The data depicted are means ± SE for triplicate samples and are representative of three independent experiments. (B) Influence of polyP levels on M. tuberculosis growth in macrophages. THP-1 macrophages were infected with WT, MT, or CT strains as described in Materials and Methods. At designated time points, triplicate wells of infected cells were washed and then lysed in PBS containing 0.1% Triton X-100, and bacterial counts were enumerated. The data depicted are means ± SE from three independent experiments.

PolyP regulates the sigF and phoP response regulator in M. tuberculosis.

It has been reported earlier that polyP acts as a phosphate donor for MprB-mediated phosphorylation of MprA, facilitating transcription of the sigE gene and thereby leading to enhancement of relA in mycobacteria (48). In Helicobacter pylori, polyP has been shown to interact with the principal sigma factor (33). To evaluate whether polyP regulates the transcript levels of sigma factors or response regulators in M. tuberculosis, qRT-PCR was performed. We observed that sigF transcript levels were marginally downregulated in the ppk-1 mutant strain compared to the level in the wild-type strain (Fig. 6A). We also observed upregulation of sigF in M. tuberculosis overexpressing ppk-1 (data not shown). However, contrary to earlier published studies, we did not observe any differences in the transcript levels of the sigE and mprA response regulators of the mprAB two-component system (TCS) between the wild-type and ppk-1 mutant strains (48) (Fig. 6A and B). We also observed that phoP transcript levels were reduced in the ppk-1 mutant strain compared to the wild-type strain (Fig. 6B).

Fig 6.

Effect of polyP depletion on the transcript levels of sigma factors and response regulators of TCS in M. tuberculosis. mRNA was extracted from exponentially growing cultures of WT and MT strains of M. tuberculosis. qRT-PCR for sigma factors (A) and response regulators (B) was performed as described in Materials and Methods. qRT-PCR analysis was done for gene expression in the MT strain, and the data are shown as the change in the cycle threshold relative to that in the WT strain.

(p)ppGpp levels regulate polyP levels in M. tuberculosis.

In concordance with our earlier results, where we did not observe any difference in sigE transcript levels between the wild-type and ppk-1 mutant strains, we observed that the transcript levels of relA were also similar in these two strains (Fig. 7A). We next evaluated whether polyP regulates the transcript levels of other enzymes involved in polyP metabolism. The transcript levels of both ppk-2 and ppx were similar in both the wild-type and ppk-1 mutant strains (Fig. 7A). In E. coli, it has been reported that (p)ppGpp accumulation leads to polyP accumulation, which has been attributed to inhibition of PPX by (p)ppGpp (55). Therefore, to evaluate whether polyP accumulation in M. tuberculosis is (p)ppGpp dependent, we compared polyP levels in mid-log-phase and late-log-phase cultures of the wild-type and relA mutant strains. Less accumulation of polyP was observed in the relA mutant strain than in the parental strain in both the mid-log and late log stages of growth (Fig. 7B). Next, we compared the transcript levels of ppk-1 in both the wild-type strain and relA mutant strain and observed that ppk-1 mRNA levels were reduced in the relA mutant strain in the late log phase, which might be responsible for lower polyP levels in the relA mutant strain (Fig. 7C).

Fig 7.

(A) Effect of polyP levels on the transcript levels of relA, ppk-2, and ppx in M. tuberculosis. mRNA was extracted from early-log-phase, mid-log-phase, and late-log-phase growing cultures of WT and MT strains. qRT-PCR analysis for the transcript levels of relA, ppk-2, and ppx was done, and the data are shown as the change in the cycle threshold relative to that in the WT strain from three independent experiments. (B) Intracellular polyP quantification in WT and relA mutant strains of M. tuberculosis. WT and relA mutant strains of M. tuberculosis were grown to the mid-log and late log phases. PolyP was isolated and quantified as described in Materials and Methods. PolyP levels are represented as pmol/min/mg and are mean ± SE values obtained from three independent experiments. (C) Effect of (p)ppGpp levels on transcript levels of ppk-1 in M. tuberculosis. mRNA was isolated from mid-log-phase and late-log-phase cultures of the wild-type and relA mutant strains. qRT-PCR analysis for the transcript levels of ppk-1 was done with the relA mutant strain, and the data are shown as the change in cycle threshold relative to that in the WT strain from three independent experiments.

PolyP deficiency is associated with increased drug susceptibility of M. tuberculosis.

We observed a 10- to 20-fold-higher polyP accumulation in mycobacteria when exposed to various TB drugs. To evaluate whether polyP accumulation contributes to drug-induced tolerance in vitro, we measured the number of persisters in early-log-phase cultures of the wild-type, ppk-1 mutant, and ppk-1 complemented strain in the presence of drugs with different mechanisms of action. PolyP deficiency significantly reduced the number of persisters by ∼9-fold, 5-fold, or 4-fold upon exposure to either the cell wall inhibitor (Inh; P < 0.01), replication inhibitor (Levo; P < 0.01), or transcription inhibitor (Rif; P < 0.05), respectively (Fig. 8A). This defect in formation of persisters was fully restored in the complemented strain. We also observed that polyP deficiency does not affect persister cell formation in the presence of Gm, Eth, and PA-824 (data not shown). Since polyP accumulation was observed at later stages of growth in vitro, we further investigated the effect of polyP deficiency on persister formation in the late log phase of M. tuberculosis in vitro. At later stages of growth, polyP deficiency enhances the susceptibility of M. tuberculosis by approximately 25-fold and 19-fold upon exposure to either Inh or Levo, respectively, for 14 days (Fig. 8B; P < 0.01). In our MIC₉₉ determination experiments, no difference was observed in the susceptibilities of strains to Inh, Levo, Eth, amoxicillin plus potassium clavulanic acid, and cefradine plus potassium clavulanic acid (Table 2). These results suggest that polyP levels control drug tolerance by modulating formation of persisters. Since polyP levels were lower in the relA mutant strain than those in the parental strain, we have also evaluated the ability of relA mutant strain to persist in the presence of Inh, Levo, or Rif. Disruption of relA in the genome of M. tuberculosis reduced the number of persisters by 8-fold in the presence of Levo, but no difference was observed in persister cell formation between the wild-type and mutant strains upon exposure to either Inh or Rif (Fig. 8C). These results suggest that polyP levels in the relA mutant strain were sufficient for the relA mutant strain to persist as well as the wild-type strain in the presence of Inh or Rif. MIC₉₉ determination experiments showed that the relA mutant strain was as susceptible as the wild-type strain to the various drugs used in this study (Table 2).

Fig 8.

Influence of deletion of ppk-1 on drug tolerance in vitro. Various strains were grown to an OD₆₀₀ of 0.2 (A and C) or 2.0 (B). These cultures were diluted and exposed to Rif (4 μg/ml), Inh (10 μg/ml), Levo (10 μg/ml), or Gm (10 μg/ml) for 7 or 14 days at 37°C. The percentage of survival was measured as number of CFU/ml in the culture after incubation with the drug relative to the CFU of the culture before the addition of drug. The results shown are representative data from three independent experiments.*, P < 0.05; **, P < 0.01.

Table 2.

MIC₉₉ values of various drugs against the wild-type, ppk-1 mutant and ppk-1 complemented, and relA mutant strains of M. tuberculosis

Drug	MIC99 (μM) of drug fora:
	WT	MT1	CT	MT2
Isoniazid	0.39	0.195	0.195	0.195
Levofloxacin	0.78	0.78	0.78	0.78
Ethambutol	3.125	3.125	1.56	1.56
Amoxicillin + clavulanate	>50	>50	>50	>50
Cefradine + clavulanate	6.25	6.25	6.25	6.25

^aWT, wild type; MT1, ppk-1 mutant; CT, ppk-1 complemented; and MT2, relA mutant strain.

PolyP levels are required for survival of M. tuberculosis in guinea pigs.

Guinea pigs were infected with the wild-type, ppk-1 mutant, or ppk-1 complemented strain via the aerosol route. Aerosol infection in guinea pigs led to implantation of ∼50 to 100 bacteria in lungs at day 1 postinfection. Gross evaluation of lungs and spleens infected with the wild-type and complemented strains revealed discrete tubercles distributed throughout the tissues, whereas markedly reduced inflammation was observed in lungs and spleens of animals infected with the mutant strain (Fig. 9A). However, no significant differences were observed in the lung weights of ppk-1 mutant-infected guinea pigs (3.75 ± 0.42 g) compared to the lung weights of wild-type-infected guinea pigs (4.22 ± 0.58 g) at day 70 postinfection. At 4 weeks postinfection, we observed a 10-fold significant reduction in lung bacillary loads in animals infected with the mutant strain in comparison to bacillary loads in animals infected with the wild-type strain (Fig. 9B; P < 0.05). This difference in the growth rates of the wild-type and ppk-1 mutant strains was more evident at later stages of growth. The lung bacterial loads decreased slightly to 4.95 log₁₀ CFU in animals infected with the wild-type strain at 10 weeks postinfection; however, in animals infected with the mutant strain, lung CFU numbers decreased dramatically (Fig. 9B). Approximately 2.2 log₁₀ CFU fewer bacteria were observed in lungs of animals infected with the ppk-1 mutant strain in comparison to bacterial loads in lungs of animals infected with the wild-type strain. We also observed that CFU counts in spleen paralleled those in lungs. The counts were 4.6 and 3.8 log₁₀ CFU in animals infected with the wild-type or ppk-1 mutant strain, respectively, at 4 weeks postinfection. The trend of lower CFU in animals infected with the mutant strain persisted at 10 weeks postinfection; the mean counts were 4.3 and 2.5 log₁₀ CFU in guinea pigs infected with the wild-type or ppk-1 mutant strain, respectively. The bacterial loads were restored in animals infected with the complemented strain (Fig. 9B). This reduction in bacterial counts was found to be statistically significant, thereby suggesting that polyP levels play an important role in pathogenesis of M. tuberculosis in guinea pigs (Fig. 9B; P < 0.01). To further investigate the role of polyP levels in disease progression of M. tuberculosis, we also performed histopathological analysis. As shown in Fig. 9C, tissue damage was almost similar in animals infected with either the wild-type, mutant, or complemented strain at 4 weeks postinfection. However, at 10 weeks postinfection, lung sections of animals infected with the wild-type (granuloma score, 19.5) and complemented strains (granuloma score, 12.5) revealed significant lung involvement exhibiting either large granulomas or multiple small granulomas. In contrast, histological analysis of lung sections from animals infected with the mutant strain (granuloma score, 1.5) displayed significantly fewer granulomas, with overall reduced inflammation and significant restoration of the lung architecture (Fig. 9C and D).

Fig 9.

Influence of disruption of ppk-1 on growth of M. tuberculosis in guinea pigs. (A) Gross pathology of lungs and spleens of animals infected with various strains via the aerosol route at 10 weeks postinfection. (B) Bacterial loads in lungs of guinea pigs infected with various strains were determined. The data depicted are means ± SE for each group (n = 3 for time zero and n = 6 for the 4- and 10-week time points). *, P < 0.05; **, P < 0.01. (C) The total granuloma score in hematoxylin-and-eosin (HE)-stained lung sections (n = 5) of animals infected with various strains was determined as described in Materials and Methods. The data depicted are means ± SE for each group. (D) The images show HE-stained (4× magnification) lung sections from guinea pigs infected with WT, MT, or CT strains after 4 weeks and 10 weeks postinfection. The data are from individual sections representative of five infected animals.

DISCUSSION

Numerous studies have shown that polyP plays a critical role in bacterial metabolism, and polyP has been described as a “molecule of multi-purpose” (56). The presence of polyP in M. tuberculosis has been documented for more than 5 decades (57). The principle bacterial enzyme responsible for synthesis of polyP, PPK-1, is widely conserved in bacteria and serves a number of essential roles. M. tuberculosis PPK-1 forms dimers, undergoes autophosphorylation on His₄₉₁ and His₅₁₀ residues, and transfers the γ-phosphate of ATP to generate polyP chains with a length of 200 to 800 residues (48). In the present study, we developed a nonradioactive luciferase-based assay system for intracellular quantification of polyP in mycobacteria. The kinetic constants (k_cat/K_m) for the PPK-1 reverse reaction with polyP₁₇ and polyP₄₅ were 3.1 μM⁻¹ min⁻¹ and 2 μM⁻¹ min⁻¹, respectively. However, the kinetic constant for PPK-1 enzyme for polyP₃ was 0.05 μM⁻¹ min⁻¹, thereby suggesting that PPK-1 displays a substrate preference for long-chain polyP (polyP₄₅ and polyP₁₇) rather than short-chain polyP (polyP₃).

We have also quantified polyP levels in various growth stages of mycobacteria in vitro and observed maximum accumulation of polyP in the stationary phase in both fast-growing and slow-growing mycobacteria. A similar pattern of polyP accumulation has also been reported in E. coli, Acinetobacter johnsonii, and Pseudomonas aeruginosa, thereby suggesting that polyP enables the bacteria to adapt to various conditions (39). In the present study, polyP accumulation was also observed upon exposure to acidic, oxidative, nitrosative, or nutritional stress. Accumulation of polyP was also observed upon exposure to various antitubercular drugs with different modes of action. This observed accumulation of polyP under various physiological conditions could be either due to (i) activation of PPK-1 enzyme, (ii) the expression levels of PPX enzyme, (iii) intracellular conditions kinetically favoring the forward reaction of PPK-1 enzyme, or (iv) a combination of all three factors. Interestingly, in our RT-PCR analysis, ppk-1 was observed to be upregulated by ∼5-fold during inhibition of transcription by Rif, ∼3-fold under nitrosative stress, and ∼2-fold in the late log phase of growth. Similar upregulation of transcript levels of ppk-1 in M. tuberculosis had also been reported earlier in the late log or stationary phase of growth (48). We did not observe upregulation of ppk-1 under other stress conditions, where accumulation of polyP was observed. Under most of these conditions, intracellular ATP levels were reduced compared to those of untreated cells. These observations suggest that under these conditions, the forward reaction of PPK-1 enzyme is favored and polyP might be the preferred storage form of energy. Apart from these, there are other factors also that might regulate polyP levels in M. tuberculosis. One such factor is phoY2, which downregulates pstSCAB, the phosphate transport system of M. tuberculosis. It is possible that under various stress conditions, phoY2 is downregulated, thereby leading to upregulation of pstSCAB operon. This leads to increase in uptake of P_i, which then might be responsible for polyP accumulation under some of these stress conditions (58).

The observed accumulation in levels of intracellular polyP suggests that PPK-1 might be essential for M. tuberculosis to cope with these stresses. To address this question, we constructed a ppk-1 mutant strain of M. tuberculosis, thereby implicating that PPK-1 is nonessential in vitro. The ppk-1 mutant strain displayed a slight growth defect compared to the wild-type strain in MB 7H9 medium. This observed slight growth defect of the ppk-1 mutant strain is contrary to the findings of earlier studies, in which downregulation of ppk-1 using the IPTG-inducible system was shown to be bactericidal (49). The ppk-1 mutant strain was compromised in its ability to survive under nitrosative stress compared to the wild-type and complemented strains. Since polyP levels have been shown to regulate activity of Lon protease in E. coli, it is possible that polyP regulates the activity of the mycobacterial PrcBA proteasome system, thereby promoting degradation of misfolded proteins and providing amino acids for synthesis of new proteins critical for the tubercle bacilli to persist in the host (9, 38, 41, 59). To evaluate the importance of polyP levels in survival of M. tuberculosis inside the macrophage, the growth kinetics of the ppk-1 mutant strain were compared to those of the wild-type and complemented strains. Disruption of ppk-1 impaired the survival of M. tuberculosis by ∼12-fold after 6 days postinfection in THP-1 macrophages. These observations are similar to earlier reports that downregulation of ppk-1 by an antisense approach impairs multiplication of M. tuberculosis in macrophages (48).

A large number of studies have shown that alternative sigma factors and TCS contribute to the ability of M. tuberculosis to survive under various stress conditions and persist in the host. Numerous studies have also shown that polyP levels regulate expression of sigma factors and response regulators of TCS (33, 37, 48); hence, we focused our attention on studying regulation of these factors by polyP levels in M. tuberculosis. Contrary to previously published reports, we did not observe any difference in the transcript levels of sigE between the wild-type and ppk-1 mutant strains of M. tuberculosis. However, we observed marginal downregulation of sigF in the ppk-1 mutant strain. SigF, a key player in mycobacterial adaptation to the stationary phase and various stress conditions, has been reported to be upregulated in the stationary phase, in persisters, and during starvation (60–63). In addition to sigF, we also observed downregulation in the transcript levels of the phoP response regulator in the ppk-1 mutant strain. It is possible that both polyP and ATP could serve as phosphate donors for PhoR histidine sensor kinase, thereby leading to regulation of the phoPR TCS, which has been shown to be involved in complex lipid biosynthesis and virulence of M. tuberculosis (64).

The stringent response pathways have also been shown to contribute to the ability of M. tuberculosis to survive under various stress conditions and to persist in mice and guinea pigs (8, 20, 48). In M. tuberculosis, it has been demonstrated that polyP regulates (p)ppGpp levels in a sigE-dependent manner (48). Elevated intrabacillary polyP levels in the ppx mutant of M. tuberculosis have been shown to downregulate ppk-1 and upregulate Rv1026 (51). Therefore, we next compared the transcript levels of enzymes involved in stringent response and polyP metabolism in both the wild-type and ppk-1 mutant strains. We did not observe any difference in the transcript levels of these enzymes. However, in our study, we observed less accumulation of polyP in the relA mutant strain compared to the parental strain of M. tuberculosis at both the mid-log and late-log stages of growth, thereby suggesting that (p)ppGpp regulates polyP levels in M. tuberculosis. The reduced intracellular polyP levels in the relA mutant strain were partly due to lower transcript levels of ppk-1 in the relA mutant strain. Recently, it has also been shown that M. tuberculosis ppx is inhibited by (p)ppGpp in vitro, which might also explain elevated polyP levels in the parental strain (50).

It has been hypothesized that metabolically less active bacteria are more tolerant to drugs in vivo (5). In such a metabolic state, macromolecular synthesis slows down dramatically, and the bacteria adopt an antibiotic-tolerant state that can be maintained indefinitely (65, 66). The formation of these persisters has been attributed to various stress-responsive pathways of bacteria. Since polyP accumulation was observed upon exposure to various drugs, we hypothesized that polyP accumulation would induce a drug-tolerant state in M. tuberculosis. We observed a biphasic pattern of killing in our in vitro persistence experiment. However, we observed less bactericidal activity of Inh compared to Rif, which might be associated with tolerance of metabolically less active persisters against Inh (67–69). We also observed that polyP deficiency is associated with increased susceptibility of M. tuberculosis to both Inh and Levo, therefore, suggesting that accumulation of polyP in M. tuberculosis induces a metabolic state that enables the bacteria to persist better in the presence of drugs. Another piece of evidence for the role of polyP in M. tuberculosis pathogenesis emerges from our guinea pig experiments. We show that polyP deficiency impairs the ability of M. tuberculosis to cause disease in guinea pigs, as measured by bacillary loads in lungs and spleens and pathological damage to the tissues. It has also been reported that elevated intracellular levels of polyP impair survival of M. tuberculosis under hypoxia conditions in macrophages and guinea pigs (51). These findings suggest that any deregulation in polyP levels contributes to impaired survival of M. tuberculosis in the host.

In conclusion, this study demonstrates that polyP accumulates in M. tuberculosis under various stress conditions and polyP deficiency is associated with increased susceptibility of M. tuberculosis to antitubercular drugs and impaired M. tuberculosis growth in guinea pigs. This study, along with some other recent findings, reveals an interesting association between polyP levels and M. tuberculosis persistence (51, 58). Deregulation of intrabacillary polyP levels might lead to altered ATP levels, NAD⁺/NADH ratios, and redox status of the bacilli, which might be the link between polyP levels and M. tuberculosis persistence. Future experiments are being carried out to address these questions and characterize the role of polyphosphatases in physiology and persistence of M. tuberculosis. Although polyP is present in all cells, enzymes homologous to polyP metabolism enzymes are absent in eukaryotes, suggesting that these enzymes might be attractive therapeutic targets for the design of small molecules that can target these persister bacterial populations. In addition, PPK-1 might perform the essential role of synthesizing ATP from stored polyP in the nonreplicating phase (dormant, latent phase); therefore, molecules inhibiting this enzyme might be useful for targeting these dormant bacilli.