Studying the dynamics of the drug processing of pyrazinamide in Mycobacterium tuberculosis

David Requena; Rydberg R. Supo-Escalante; Patricia Sheen; Mirko Zimic

doi:10.1371/journal.pone.0309352

Abstract

Pyrazinamide (PZA) is a key drug in the treatment of Mycobacterium tuberculosis. Although not completely understood yet, the bactericidal mechanism of PZA starts with its diffusion into the cell and subsequent conversion into pyrazinoic acid (POA) after the hydrolysis of ammonia group. This leads to the acidification cycle, which involves: (1) POA extrusion into the extracellular environment, (2) reentry of protonated POA, and (3) release of a proton into the cytoplasm, resulting in acidification of the cytoplasm and accumulation of intracellular POA. To better understand this process, we developed a system of coupled non-linear differential equations, which successfully recapitulates the kinetics of PZA/POA observed in M. tuberculosis. The parametric space was explored, assessing the impact of different PZA and pH concentrations and variations in the kinetic parameters, finding scenarios of PZA susceptibility and resistance. Furthermore, our predictions show that the acidification cycle alone is not enough to result in significant intracellular accumulation of POA in experimental time scales when compared to other neutral pH scenarios. Thus, revealing the need of novel hypotheses and experimental evidence to determine the missing mechanisms that may explain the pH-dependent intracellular accumulation of POA and their subsequent effects.

Citation: Requena D, Supo-Escalante RR, Sheen P, Zimic M (2024) Studying the dynamics of the drug processing of pyrazinamide in Mycobacterium tuberculosis. PLoS ONE 19(8): e0309352. https://doi.org/10.1371/journal.pone.0309352

Editor: Atul Vashist, Bennett University, INDIA

Received: December 31, 2023; Accepted: August 9, 2024; Published: August 29, 2024

Copyright: © 2024 Requena et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: P.S. This study was partially supported the Wellcome Trust Intermediate Fellowship (grant 099805/Z/12/Z). The sponsor did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Tuberculosis (TB) is an infectious bacterial disease caused by Mycobacterium tuberculosis (MTB). According to the 2023 Global Tuberculosis Report by the World Health Organization, the annual global burden of TB comprises more than 10 million cases and 1.3 million deaths. Moreover, multidrug-resistant TB (MDR-TB) accounted for 410, 000 cases in 2022. Thus, TB represents one of the most critical public health challenges worldwide. TB transmission occurs from person to person via droplets expelled from the airways of individuals with active respiratory disease [1]. Following infection, some patients develop granuloma formation, a process in which MTB enters a dormant state and may persist for decades [2]. Nevertheless, infected individuals can develop active disease at any moment, process that is favored under conditions of weakened immunity [3].

Pyrazinamide (PZA) is a crucial drug utilized in the first-line treatment of drug-susceptible TB. Its main advantage lies in its ability to eliminate bacteria in a latent, non-growing state [4], exerting maximum bactericidal effects under acidic conditions [5–7]. Such conditions impact bacteria at sites of active inflammation [8, 9], or those residing within macrophages [10]. PZA is also used in treatment regimens of both MDR-TB [11] and extensively drug-resistant TB (XDR-TB) [12]. Therefore, the emergence of PZA-resistant isolates presents a significant public health concern. Consequently, it is imperative to comprehend its mechanism of action to counteract actual and potential adaptations of MTB leading to PZA resistance.

PZA is a pro-drug that enters the MTB cytoplasm via passive diffusion [13]. Intracellularly, the non-essential enzyme pyrazinamidase (PZase, encoded by the gene pncA) converts PZA into its acidic active form, pyrazinoic acid (POA) [14, 15]. Next, POA is expelled from the cell through an efflux pump system into the extracellular space. Proteins such as Rv0191, Rv3756c, Rv3008, and Rv1667c have been shown to participate in this efflux mechanism [16], among other associated proteins [17, 18]. Extracellularly, acidic conditions promote the protonation of a small fraction of POA, which subsequently reenters the bacteria and releases the proton into the cytoplasm [19]. This cycle of POA expulsion, protonation, and re-entry potentially results in the intracellular accumulation of POA and cytoplasmic acidification [19]. The combined effects of intracellular POA accumulation and cytoplasmic acidification are believed to disrupt the synthesis of vital cellular components and alter the central metabolism, the membrane permeability, causing energy failure and lethal consequences [13, 19, 20]. Nonetheless, although a high cytoplasmic acidification was expected to result from PZA treatment, it was demonstrated that mildly acidic pH conditions (between 5 and 6, known to induce PZA susceptibility) do not cause substantial cytoplasmic acidification [21]. The downstream molecular targets of POA have not been characterized yet, although several potential targets have been proposed, including the enzyme aspartate decarboxylase (PanD) [22], the fatty acid synthase I (FAS-I) [23, 24], and the ribosomal protein S1 (RpsA) [25]. Among them, the enzyme PanD, involved in the synthesis of coenzyme A (CoA) precursors [19, 26], is the candidate with most experimental support.

In vitro, PZA activity has demonstrated a dependence on an acidic extracellular environment with a pH below 6 [12, 27, 28]. Under neutral pH conditions, the sterilizing bacteriostatic effect is not observed [29]. Experiments providing radioactively labeled POA as a drug for MTB strain H37Ra found that an acidic environment effectively promotes POA accumulation [27]. In vivo, this acidic environment is facilitated within the macrophages where MTB resides during the inflammatory process, and may favor the protonation of extracellular POA contributing to recover and accumulate intracellular POA [30]. Additionally, it could aid in decreasing the membrane potential and reducing the bacterial metabolism, thus facilitating PZA activity [13]. Nonetheless, the acidic extracellular environment does not appear to be essential for the bactericidal effect of PZA, as overexpression of pncA [21], coincubation with the efflux pump inhibitors reserpine and valinomycin [31], and most recently constitutive activation of the cell envelope stress response [32] have shown to induce PZA susceptibility independent of environmental pH.

The primary cause of PZA resistance is the loss of PZase activity, resulting from critical mutations in pncA [14, 15, 33]. PZase is considered a non-essential enzyme that does not significantly impact the fitness of the organism [34, 35]. Mutations are dispersed throughout the entire sequence of pncA, although the most disruptive mutations affect the enzymatic active site or the metal binding site [12, 36]. PZase failure is also associated with mutations in its promoter, leading to a low expression level [28, 37]. Numerous efforts have been made to predict PZA resistance associated with PZase dysfunction using computational methods [38, 39]. However, the level of PZase activity is insufficient to fully explain the variability of PZA resistance [28, 40], suggesting alternative resistance mechanisms. It has been discovered that deviations in the POA efflux rate from a critical value towards higher values result in PZA resistance [41–44]. Similarly, intrinsic PZA resistance in Mycobacterium smegmatis is associated with a very active POA efflux mechanism, which has been found to be 900-fold higher than that of MTB strain H37Rv, preventing intracellular POA accumulation [27, 43, 44]. Moreover, resistance could be caused by mutations in downstream targets of POA, hindering proper interaction [19]. This has been observed for rpsA and panD genes in PZA-resistant strains of MTB with a wild-type (wt) pncA gene [22, 25]. However, the exact molecular mechanism by which these potential targets drive PZA susceptibility has not yet been established, but activation of the SigE-dependent cell envelope stress response has been suggested as the main driver of PZA susceptibility [32].

Mathematical modeling of molecular pathways and networks employing systems of non-linear, coupled differential equations serves as a valuable approach in systems biology research. It has been widely used for modeling complex metabolic pathways and concentration dynamics [45–47], drug kinetics and effectiveness [48]. In this study, we developed a mathematical model of the drug processing of PZA in MTB. This model examines the conversion of PZA into intracellular POA, the efflux-influx cycle of POA related to POA accumulation, and the potential cytoplasmic acidification. Overall, this study provides a better understanding of the quantitative impact of alterations in various steps of the processing of PZA, enabling the exploration of alternative modes of PZA resistance and susceptibility, and generating new hypotheses for subsequent investigations.

Materials and methods

Mathematical model of POA accumulation

A mathematical model of the dynamics of PZA processing in MTB was built. The model considers the central elements of the currently accepted drug metabolism (Fig 1A): (1) the entry of extracellular PZA by passive diffusion, (2) the intracellular hydrolysis of PZA to unprotonated POA by the enzyme PZase, (3) the expulsion of intracellular unprotonated POA by an efflux pump system, (4) the loss of some extracellular POA, (5) the protonation of the extracellular unprotonated POA to form protonated POA (HPOA), and its subsequent diffusion into the cytosol, and (6) the protonation and deprotonation between intracellular POA and HPOA. We focused on the concentration of five molecular species, which are our state variables: 1) intracellular pyrazinamide ([PZA]_i), 2) intracellular unprotonated POA ([POA]_i), 3) intracellular protonated POA ([HPOA]_i), 4) intracellular hydrogen ion ([H]_i), and 5) the extracellular total POA ([POA_T]_e). This last variable is defined as the sum of extracellular protonated and unprotonated POA.

Download:

Fig 1. Mathematical model of PZA processing in M. tuberculosis.

(A) Molecular pathway of the mechanism of accumulation and metabolism of pyrazinamide in M. tuberculosis. The arrows indicate the reactions and their directions. The letters next to the arrows represent the kinetic parameters involved in the reaction. The gray boxes indicate the molecular species that are assumed constant in the simulations and treated as fixed parameters. The [POA_T]_e is depicted in their protonated (HPOA) and unprotonated (POA) states but treated as a single molecule in the system of equations. (B) Heatmap of the equilibrium relationships between the ratio [POA_T]_i/[POA_T]_e and model parameters. (C) Heatmap of the equilibrium relationships between the pH_i and model parameters.

https://doi.org/10.1371/journal.pone.0309352.g001

To assemble these variables into equations, we assumed:

The bacterium was assumed to be in a dormant state, neither growing nor replicating. So, the concentration of intracellular species is not reduced by dilution.
PZA is maintained at a steady level, reflecting a sustained-periodic intake during the anti-tuberculosis treatment or in vitro culture conditions, and enters by passive diffusion into the cytoplasm of MTB, represented by: (1)
PZase activity is assumed independent of changes in the pH_i. The intracellular enzymatic conversion of PZA into POA was modeled by Michaelis-Menten kinetics: (2)
Where V_PZAse is the maximum reaction rate, and K_M represents the Michaelis-Menten constant between PZase and PZA.
The POA efflux pump mechanism is assumed to never saturate at POA concentrations attainable inside the cell. Therefore, it operates at first order kinetics proportional to the concentration of intracellular unprotonated POA: (3)
The rate of the reversible protonation reaction between protonated and unprotonated POA was modeled using the law of mass action: (4)
Where q_f, and q_r are the forward and reverse reaction rate constants, and their ratio is K_a, the acid dissociation constant of POA. As protonation and deprotonation are known to be fast scale processes, large values for q_r and q_f were used.
Extracellular POA re-enters the mycobacterium only in its protonated from, justified by the fact that only uncharged molecules can traverse the cell membrane. Passive diffusion was assumed, modeling similarly to Eq (1) (5)
The rate at which protonated POA enters or exits the cell is considered proportional to the concentration difference between compartments.
The loss rate of total extracellular POA surrounding the bacterium is assumed linearly proportional to its concentration. This reflects in vivo conditions, because of the circulation system constantly diluting the local concentration of POA around the lung granuloma, where dormant mycobacteria are located. Also in vitro conditions, where the molecule migrates to areas with less concentration. Therefore: (6)
The external pH is assumed constant, being supported by the homeostatic mechanisms of human hosts that regulate pH. Therefore, by the Henderson–Hasselbalch equation, is also constant. Setting Eq (4) to 0, [POA]_e can be expressed in terms of [HPOA]_e. Then, [POA_T]_e can be expressed as follows: (7)
Rearranging Eq (7): (8)

Thus, a system of first-order non-linear differential equations was developed, modelling the concentration dynamics of each molecular specie.

Non-linear optimization fitting with experimental data of cytoplasm acidification

The values of K_M (1240 μM) and K_a(10^−2.9 M) were retrieved from experimental data [29, 49]. The parameters u, z, V_PZase,Eff_EP, and δ were estimated using an experimental dataset of intracellular pH measurements in MTB after in vitro treatment with PZA [21]. Briefly, this dataset consists of two different experiments with triplicate treatments of 0, 80, 800, and 8000 μM of PZA. The first experiment used pH_e = 5.8 and measured pH_i at 0, 8, 23, 46, 97, and 126 hours. The second experiment used a pH_e = 7 with a 100-fold overexpression of pncA and measured at 0, 7, 23, 47, 71, 95, and 123 hours [21]. To estimate the remaining kinetic parameters, a nonlinear curve-fitting was performed in Matlab R2023b using the lsqcurvefit function (Optimization Toolbox), minimizing the total sum of squares of the difference between experimental data and model predictions: (9)

Where F(p,t) corresponds to the model’s output at time t_i with parameter values provided in the vector p; and y_i corresponds to the experimental measurements at time t_i.

The parameter intervals explored for u, V_PZase, z, Eff_EP, and δ, were [1h⁻¹-10h⁻¹], [10²μMh⁻¹-10⁴μMh⁻¹], [10⁻³h⁻¹-10⁻²h⁻¹], [10³h⁻¹-10⁴h⁻¹], and [10⁻¹h⁻¹-1h⁻¹], respectively. The exploration was performed with a random seed of 2, which was used to choose the starting values from a log-uniform distribution on the provided bounds; performing a maximum of 800 function evaluations per iteration. The optimization problem was run until convergence.

Numerical simulations, parameter sweep and equilibrium analysis

Simulations were carried out in Matlab, starting with an initial concentration of 0 for all the state variables, except for the pH_i which was set to 7.2 to reflect the average value in MTB under normal conditions. To understand the effect of different input values in the dynamics of the system, a bidimensional parameter sweep of [PZA]_e (1 nM up to 1000 M) and pH_e (4 to 8) was performed. The simulations were run for 240 and 10⁷ hours generating heatmaps of pH_i and [POA_T]_i using the results of the bidimensional parameter sweep for both simulation times.

Equilibrium equations were calculated by setting all differential equations to 0 and solving for each state variable in the system. The equilibrium ratio between [POA_T]_i and [POA_T]_e was found to be a function of pH_e and Eff_EP/δ. The equilibrium equations were used to quickly find the stationary concentration of each state variable in all the explored scenarios and to compare them with the simulation results.

Exploration of scenarios of PZA resistance and susceptibility

To explore the system behavior in conditions associated with PZA resistance or susceptibility, parameters V_PZase and Eff_EP were modified to mimic experimentally observed alterations in the PZase enzyme and the POA efflux system, respectively. Resistance scenarios included reduction of PZase activity (a 100-fold reduction of the parameter V_PZase) or increased efflux efficiency (a 100-fold increase of the parameter Eff_EP). Susceptibility scenarios included increase in PZase activity (a 100-fold increase in the value of the parameter V_PZase) or a reduction of efflux efficiency (a 100-fold decrease in the value of the parameter Eff_EP). Simulations and equilibrium analysis was carried out as previously described for the wt scenarios.

Results

Mathematical model of POA accumulation

A mathematical model reflecting the concentration kinetics of the different molecular species involved in the processing of PZA in MTB was developed. It consists of a system of five first-order non-linear, coupled differential equations: (10) (11) (12) (13) (14)

This model simulates the main elements of PZA processing in MTB, including entry and hydrolysis of PZA by the enzyme PZase into POA, efflux-pump-mediated expulsion of POA, which is protonated and reenters as HPOA, and proton transfer between intracellular POA and HPOA (Fig 1A). The state variables and parameters are described in Tables 1 and 2, respectively.

Download:

Table 1. Symbol, description, and units of the molecules studied.

https://doi.org/10.1371/journal.pone.0309352.t001

Download:

Table 2. Model parameters and fitted values.

https://doi.org/10.1371/journal.pone.0309352.t002

Equilibrium analysis of the effect of

When the system reaches the equilibrium, the set of equations can be reduced to: (15) (16) (17) (18) (19)

The positive real root of Eq (15) is the equilibrium value of [PAZ]_i. The flux of PZA to POA in the equilibrium was defined as: (20)

Which leads to a straightforward solution for the rest of equations: (21) (22) (23) (24)

Then, adding Eq (21) and Eq (22), and dividing by Eq (24), gives: (25)

In the right side of Eq (25), the first fraction corresponds to the intracellular unprotonated POA, while the second to the intracellular protonated POA recovered by re-entry through the cell wall. Whereas the first term could be unconstrainedly increased by reducing the efficiency of the efflux pump, the second term has an upper limit of 1 when [H]_e is much larger than K_a. Additionally, the pH_e shows a significant effect on Eq (25) only when Eff_EP/δ is greater than 1 (Fig 1B; S1A Fig). Solving Eq (25) for [POA_T]_i: (26)

There, [POA_T]_i could be increased by raising V, the flux of PZA to POA, which can be achieved by overexpression of PZase or mutations that increase the catalytic activity of PZase. Overall, this equation provides a reasonable approximation which allows to understand the effect of different experimental parameters on [POA_T]_iin MTB.

Parameter sweep of [PZA]_e and pH_e in wt MTB indicates a pH-dependent POA accumulation attainable at equilibrium but not at experimental timescales

To start exploring the dynamics of the model, parameters were fitted using available data on cytoplasmic acidification. The parameter values obtained are displayed in Table 2, and the fitting least square error was 5.99 (Fig 2A and 2B). After that, the effect of different values of [PZA]_e and pH_e on [POA_T]_i at equilibrium were explored, using conditions reflecting MTB strain H37Ra (Fig 3). The highest concentration reached by [POA_T]_i in the explored parameter space is ~200 μM and it is achieved for low extracellular pH and high concentrations of [PZA]_e (Fig 3A). This parameter sweep demonstrated how, in contrast to neutral conditions, acidic conditions give the system the ability to achieve higher [POA_T]_i for similar [PZA]_e values. This becomes more evident for higher values of [PZA]_e, at which the PZase enzyme is expected to be saturated by its substrate. Additionally, the cytoplasmic acidification at equilibrium only depended on pH_e (S2A Fig), and a high acidification is predicted even for neutral environments, with almost all pH_i at equilibrium being lower than 4.

Download:

Fig 2. Parameter estimation using cytoplasmic acidification data.

Experimental and fitted curves of different PZA treatments, under (A) pH_e = 5.8, and (B) pH_e = 7 and a 100-fold expression of the gene pncA.

https://doi.org/10.1371/journal.pone.0309352.g002

Download:

Fig 3. Heatmaps of the predicted [POA_T]_i at equilibrium for the five scenarios studied.

(A) Wild-type scenario. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux. The color scale is the same in all panels.

https://doi.org/10.1371/journal.pone.0309352.g003

Contrarily, after 240 hours of simulation, the pH_i obtained is between 4 and 7 for most of the explored parameter space, showing moderate acidification, thus matching experimental evidence (S3A Fig). Interestingly, the heatmaps of [POA_T]_i indicate that equilibrium is far from being reached, and that [POA_T]_i behaves independently of pH_e (Fig 4A). Similarly, simulation trajectories at 240 hours illustrate that pH_e values representing strong acidic conditions do not help to accumulate more [POA_T]_i than pH_e values in neutral conditions, although slight intracellular acidification is observed (Fig 5A). When simulations run for an extremely long period of time (10⁷ hours), the effect of pH_e becomes visible and the cytoplasmic acidification becomes stronger. However, the time needed to reach this state is experimentally unrealistic (Fig 5A).

Download:

Fig 4. Heatmaps of predicted [POA_T]_i at 240 hours of simulation, for the five studied scenarios.

(A) Wild-type scenario. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux. The color scale is the same in all panels.

https://doi.org/10.1371/journal.pone.0309352.g004

Download:

Fig 5. Concentration trajectories for the five explored scenarios at different time scales.

(A) Wild-type scenario. (B-C) Simulated PZA-susceptibility scenarios: (B) increased PZase activity, and (C) reduction of efflux efficiency. (D-E) Simulated PZA-resistance scenarios: (D) reduction of PZase activity, and (E) increased efflux efficiency. In each column, the first and second plot correspond to the short-term dynamics (240 hours) of [POA_T]_i and pH_i, respectively. Analogously, the third and fourth plot correspond to the long-term dynamics (10⁷ hours).

https://doi.org/10.1371/journal.pone.0309352.g005

Resistance and susceptibility scenarios display different regimes of accumulation of [POA_T]_i and cytoplasmic acidification

The effect of pH_e in increasing [POA_T]_i is present with different strength in all the evaluated susceptibility and resistance scenarios in the equilibrium at saturating [PZA]_e (Fig 3A–3E). In the equilibrium, the susceptibility scenarios (a 100-fold increase in PZase activity or 100-fold reduction of efflux efficiency), show that the values of [POA_T]_i obtained at neutral conditions are similar to those in highly acidic conditions in the wt scenario (Fig 3B, 3C). Additionally, in the scenario of reduced efflux efficiency, [POA_T]_i displayed less dependence on pH_e (Fig 3C; S4C Fig) in the explored range. Conversely, the scenario of increased PZase activity showed a greater dependence on pH_e to reach the maximum [POA_T]_i possible (Fig 3B; S4B Fig), which is predicted to be higher than in the scenario of reduced efflux efficiency (S4B, S4C Fig). Also, in the equilibrium, the resistance scenarios (a 100-fold reduction of PZase activity or a 100-fold increase in efflux efficiency) show different behaviors. For the scenario of reduced PZase activity, even at the most acidic conditions explored (pH = 4), the maximum [POA_T]_i is similar to the values obtained at neutral conditions in wt (Fig 3D; S4D Fig). However, for the scenario of increased efflux efficiency, the same range of values of [POA_T]_i in the wt scenario could be achieved in acidic conditions, but not at neutral conditions (Fig 3E; S4E Fig). Cytoplasmic acidification at equilibrium is never affected by variations in PZase activity and is independent of [PZA]_e (S2B and S2D; S5B and S5D Figs), but is lower when the efflux efficiency is reduced (S2C; S5C Figs), and higher when the efflux efficiency is increased (S2E; S5E Figs).

Noticeably, the effect of pH_e in [POA_T]_i is mostly absent in all the simulations at 240 hours (Fig 4), being barely present in the susceptibility scenario of increased PZase activity (Fig 4B; S6B Fig). In this short simulation time scale, both susceptibility scenarios already have higher [POA_T]_i than in the wt scenario (Fig 4B, 4C; S6B, S6C Fig). The heatmaps of both resistance scenarios are similar, displaying lower values of [POA_T]_i than the wt scenario in the explored parameter range (Fig 4D, 4E; S6D, S6E Fig). Also, at 240 hours, variations in the efflux efficiency result in similar cytoplasmic acidification to the wt scenario (S3C and S3E; S7C and S7E Figs), but null acidification is observed in the scenario of reduced PZase activity, even at acidic conditions (S3D; S7D Figs). Conversely, in the scenario of increased PZase activity, higher acidification than in the wt scenario could be obtained even at neutral pH_e for high values of [PZA]_e (S3B; S7B Figs).

Finally, the [POA_T]_i and pH_i of the susceptibility, resistance, and wt scenarios were compared at different pH_e values (Figs 5; 6). The resistance scenarios have the same dynamics of [POA_T]_i at 240 hours, independently of the pH_e value (Figs 5D, 5E; 6A). At 10⁷ hours, the resistance scenario of reduced PZase activity has much lower [POA_T]_i than the scenario of increased efflux efficiency, but none of them reached equilibrium and are still orders of magnitude below the concentrations reached by the wt (Figs 5D, 5E; 6C). On the other hand, at 240 hours, both susceptibility scenarios have higher [POA_T]_i than the wt, with the scenario of increased PZase activity displaying lower [POA_T]_i than the scenario of reduced efflux efficiency (Figs 5B, 5C; 6A). However, the opposite is seen at 10⁷ hours, although the scenario of reduced efflux efficiency equilibrates faster (Figs 5B, 5C; 6C). As the pH_e is reduced, the wt scenario approaches the susceptibility scenario of reduced efflux efficiency, suggesting that acidic conditions make the wt behave like this scenario (Fig 6C). For cytoplasmic acidification, at 240 hours the wt and scenarios with variations in efflux have similar acidifications levels (Figs 5A and 5C–5E; 6B). Almost null acidification is observed for the scenario of reduced PZase activity, while the scenario of increased PZase activity causes the highest acidification (Figs 5B and 5D; 6B). At 10⁷ hours, acidification is shown to be strong for all simulated scenarios, with the scenarios of reduced PZase activity and reduced efflux efficiency being the most conservative scenarios (Figs 5C, 5D; 6D), and the scenario of increased efflux efficiency leading to the highest acidification (Figs 5E; 6D).

Download:

Fig 6. Simulated trajectories of the five explored scenarios at different values of pH_e.

Short-term dynamics (240 hours) of (A) [POA_T]_i, and (B) pH_i. Long-term dynamics (10⁷ hours) of (C) [POA_T]_i, and (D) pH_i.

https://doi.org/10.1371/journal.pone.0309352.g006

Discussion

In this study, we are presenting a mathematical model reflecting the known elements of the metabolism of PZA in MTB. Model parameters were fitted to experimental data, successfully recapitulating the intracellular acidification observed in wild-type MTB. Using this model, the impact of having an acidic extracellular environment on the dynamics of POA concentration and cytoplasmic acidification was quantitively assessed. Subsequently, the model dynamics was explored under different scenarios of extracellular PZA concentration and extracellular pH, studying conditions leading to PZA resistance. There are two potential roles for the acidic extracellular environment in PZA susceptibility: 1) cytoplasmic acidification, which has been suggested to be non-significant [21], and 2) pH-dependent accumulation of intracellular POA [27]. We investigated these effects in our model, which predicts that at equilibrium, when pH_e is below 7, [POA_T]_i increases with the extracellular acidity. However, when pH_e is 7 or above, it solely depends on [PZA]_e, and the contribution of pH_e becomes negligible. This agrees with previous experimental results. Zhang et al. showed that when POA is provided in the extracellular medium, an acidic pH favors the intracellular accumulation of C14-radiolabelled POA [27]. However, they did not distinguish between the protonated and unprotonated forms of POA. Our model provides further insight, showing that at equilibrium pH_e only contributes to accumulate the protonated form. And, by the Henderson-Hasselbalch equation, this would lead to high intracellular acidification with low pH. Conversely, only slight cytoplasmic acidification has been observed in wt MTB, even at acidic environments for 120 hours [21]. In line with this, neither high cytoplasmic acidification nor dependence between [POA_T]_i and pH_e in the first 10 simulated days were observed. Overall, this suggests that an alternative POA accumulation mechanism without intracellular acidification should be in place. Efflux reduction caused by growth arrest and reduced metabolic activity after culturing M. tuberculosis in a low pH medium is a potential explanation. This would agree with the general observation that susceptibility is mainly achieved under conditions that induce stress and reduce the metabolic activity of M. tuberculosis [32]. If a mechanism for the accumulation of intracellular unprotonated POA (dependent on an acidic environment) exists, then the equilibrium could be achieved within the first 10 days of simulation, without changing the results for all simulated scenarios.

Additionally, scenarios associated with increased PZA susceptibility (mutants with reduced efflux efficiency or increased PZase activity) and with PZA resistance (mutants with increased efflux efficiency or reduced PZase activity) were explored. Under the same conditions, the mutants associated with increased PZA susceptibility achieved higher concentrations of intracellular POA_T than wt MTB, while mutants associated with PZA resistance achieved lower concentrations. The scenarios of PZA resistance explored have been observed in nature: 1) PZA resistance in M. tuberculosis is predominantly associated with a reduction in PZase activity due to missense mutations in the gene pncA and, less commonly, mutations in the promoter [28, 40, 50, 51]. And 2) PZA resistance due to an elevated efflux activity has been observed in M. smegmatis, which exhibits a POA efflux rate approximately 900-fold higher than that of the M. tuberculosis strain H37Rv [43]. Similarly, increased PZA susceptibility has been achieved in vitro by 1) the use of efflux inhibitors like reserpine, valinomycin or piperine [27, 32]; and 2) overexpression of pncA [52]. Additionally, our simulations showed that alterations in efflux efficiency do not influence cytoplasmic acidification in usual experimentation time scales (less than 240 hours). However, higher PZase activity expedites acidification, while lower PZase activity significantly decelerates the process.

Our model, and the conclusions made from it, are limited by the data available and the assumptions made. Steps lacking evidence were not considered, like that the efflux pump is thought to expel only unprotonated POA, but not protonated POA, which is an uncontested assumption in the field. Considering that, intracellularly, the majority of POA exists in an unprotonated state, most POA molecules expelled by the efflux pump would be unprotonated. However, its ability to also expel protonated POA still needs to be resolved. Although our assumptions are sound under the current knowledge, they may differ from reality as new evidence arises or more complexity is considered. However, the main conclusions of our study could be regarded as an upper bound of the system’s ability to increase [POA_T]_i.

Finally, while it is crucial to accurately determine the concentrations of the molecular species involved in the processing of PZA to assess MTB susceptibility, quantifying these molecules remains as a complex task, even in vitro. The mathematical model presented herein serves as a valuable tool to quantitatively study the effect of different alterations in the molecular processing of PZA, or different conditions of pH_e and [PZA]_e in [POA_T]_i. This model proves beneficial not only for quantitively estimating PZA susceptibility and resistance in MTB as response to diverse biological scenarios but also for comprehending the dynamics of the biochemical reactions underlying the antibiotic action of PZA.

Conclusions

We have developed the first mathematical model of the molecular processing of PZA in M. tuberculosis, using a system of ordinary differential equations. This model demonstrated that, although an acidic environment has the potential to promote accumulation of protonated POA and cytoplasmic acidification in MTB, this will not happen at regular experimental time scales. Our findings explain the lack of substantial cytoplasmic acidification after PZA treatment in MTB observed experimentally. Thus, revealing the need of an alternative mechanism to explain the pH-dependent accumulation of intracellular POA. Future research should study the dependence of the efflux efficiency on the extracellular pH. Additionally, the model has proven advantageous for quantitatively assessing the impact of alterations in the molecular processing of PZA, simulating scenarios of susceptibility and resistance. Hence, serving as a tool to generate and simulate novel hypotheses.

Supporting information

S1 Fig. Relationship between model parameters at equilibrium.

(A) The ratio [POA_T]_i/[POA_T]_e at equilibrium as a function of pH_e and the ratio Eff_EP/δ. (B) Values of pH_i at equilibrium, as a function of pH_e and the ratio Eff_EP/δ.

https://doi.org/10.1371/journal.pone.0309352.s001

(TIF)

S2 Fig. Heatmaps of the predicted pH_i at equilibrium for the five studied scenarios.

(A) Wild-type MTB. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux. The color scale is the same in these five panels.

https://doi.org/10.1371/journal.pone.0309352.s002

(TIFF)

S3 Fig. Heatmaps of the predicted pH_i at 240 hours of simulation for the five studied scenarios.

(A) Wild-type MTB. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux. The color scale is the same in these five panels.

https://doi.org/10.1371/journal.pone.0309352.s003

(TIFF)

S4 Fig. Heatmaps of the predicted [POA_T]_i at equilibrium for the susceptibility/resistance studied scenarios.

(A) Wild-type MTB. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux.

https://doi.org/10.1371/journal.pone.0309352.s004

(TIFF)

S5 Fig. Heatmaps of the predicted pH_i at equilibrium for the susceptibility and resistance studied scenarios.

(A) Wild-type MTB. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux.

https://doi.org/10.1371/journal.pone.0309352.s005

(TIFF)

S6 Fig. Heatmaps of the predicted [POA_T]_i at 240 hours of simulation for the susceptibility and resistance studied scenarios.

(A) Wild-type MTB. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux.

https://doi.org/10.1371/journal.pone.0309352.s006

(TIFF)

S7 Fig. Heatmaps of the predicted pH_i at 240 hours of simulation for the susceptibility and resistance studied scenarios.

(A) Wild-type MTB. (B) Increased PZase activity. (C) Decreased efflux. (D) Decreased PZase activity. (E) Increased efflux.

https://doi.org/10.1371/journal.pone.0309352.s007

(TIFF)

References

1. Patterson B, Wood R. Is cough really necessary for TB transmission? Tuberculosis. 2019;117: 31–35. pmid:31378265
- View Article
- PubMed/NCBI
- Google Scholar
2. Gengenbacher M, Kaufmann SHE. Mycobacterium tuberculosis: success through dormancy. FEMS Microbiol Rev. 2012;36: 514–532. pmid:22320122
- View Article
- PubMed/NCBI
- Google Scholar
3. Hopewell PC, Jasmer RM. Overview of clinical tuberculosis. Tuberculosis and the tubercle bacillus. 2004; 13–31.
- View Article
- Google Scholar
4. Hu Y, Coates AR, Mitchison DA. Sterilising action of pyrazinamide in models of dormant and rifampicin-tolerant Mycobacterium tuberculosis. The International Journal of Tuberculosis and Lung Disease. 2006;10: 317–322. pmid:16562713
- View Article
- PubMed/NCBI
- Google Scholar
5. Johnson R, Streicher EM, Louw GE, Warren RM, Van Helden PD, Victor TC. Drug resistance in Mycobacterium tuberculosis. Curr Issues Mol Biol. 2006;8: 97–112. pmid:16878362
- View Article
- PubMed/NCBI
- Google Scholar
6. Mackaness GB. The intracellular activation of pyrazinamide and nicotinamide. American Review of Tuberculosis and Pulmonary Diseases. 1956;74: 718–728. pmid:13372956
- View Article
- PubMed/NCBI
- Google Scholar
7. McDermott W, Tompsett R. Activation of pyrazinamide and nicotinamide in acidic environments in vitro. Am Rev Tuberc. 1954;70: 748–754. pmid:13197751
- View Article
- PubMed/NCBI
- Google Scholar
8. Grosset J. New experimental regimens for preventive therapy of tuberculosis. Bull Int Union Tuberc Lung Dis. 1990;66: 15–16. pmid:1983742
- View Article
- PubMed/NCBI
- Google Scholar
9. Mitchison DA. The action of antituberculosis drugs in short-course chemotherapy. Tubercle. 1985;66: 219–225. pmid:3931319
- View Article
- PubMed/NCBI
- Google Scholar
10. Styblo K, Enarson DA. Epidemiology of tuberculosis. Vol. 24. The Hague: Royal Netherlands Tuberculosis Association. 1984.
11. Raviglione MC, Gupta R, Dye CM, Espinal MA. The burden of drug-resistant tuberculosis and mechanisms for its control. Ann N Y Acad Sci. 2001;953: 88–97. pmid:11795426
- View Article
- PubMed/NCBI
- Google Scholar
12. Zhang Y, Mitchison D. The curious characteristics of pyrazinamide: a review. The international journal of tuberculosis and lung disease. 2003;7: 6–21. pmid:12701830
- View Article
- PubMed/NCBI
- Google Scholar
13. Zhang Y, Shi W, Zhang W, Mitchison D. Mechanisms of pyrazinamide action and resistance. Microbiol Spectr. 2014;2: 2–4. pmid:26104205
- View Article
- PubMed/NCBI
- Google Scholar
14. Corbett EL, Watt CJ, Walker N, Maher D, Williams BG, Raviglione MC, et al. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch Intern Med. 2003;163: 1009–1021. pmid:12742798
- View Article
- PubMed/NCBI
- Google Scholar
15. Wu X, Zhang J, Zhong M. Studies on pncA gene mutations in M. tuberculosis isolates. Zhonghua Jie He He Hu Xi Za Zhi. 2000;23: 40–42. pmid:11778181
- View Article
- PubMed/NCBI
- Google Scholar
16. Zhang Y, Zhang J, Cui P, Zhang Y, Zhang W. Identification of novel efflux proteins Rv0191, Rv3756c, Rv3008, and Rv1667c involved in pyrazinamide resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2017;61: e00940–17. pmid:28584158
- View Article
- PubMed/NCBI
- Google Scholar
17. Liu J, Shi W, Zhang S, Hao X, Maslov DA, Shur K V, et al. Mutations in efflux pump Rv1258c (Tap) cause resistance to pyrazinamide, isoniazid, and streptomycin in M. tuberculosis. Front Microbiol. 2019;10: 216. pmid:30837962
- View Article
- PubMed/NCBI
- Google Scholar
18. Sheen P, Requena D, Gushiken E, Gilman RH, Antiparra R, Lucero B, et al. A multiple genome analysis of Mycobacterium tuberculosis reveals specific novel genes and mutations associated with pyrazinamide resistance. BMC Genomics. 2017;18: 1–11.
- View Article
- Google Scholar
19. Lamont EA, Dillon NA, Baughn AD. The bewildering antitubercular action of pyrazinamide. Microbiology and Molecular Biology Reviews. 2020;84: e00070–19. pmid:32132245
- View Article
- PubMed/NCBI
- Google Scholar
20. Montenegro SH, Gilman RH, Sheen P, Cama R, Caviedes L, Hopper T, et al. Improved detection of Mycobacterium tuberculosis in Peruvian children by use of a heminested IS6110 polymerase chain reaction assay. Clinical Infectious Diseases. 2003;36: 16–23. pmid:12491196
- View Article
- PubMed/NCBI
- Google Scholar
21. Peterson ND, Rosen BC, Dillon NA, Baughn AD. Uncoupling environmental pH and intrabacterial acidification from pyrazinamide susceptibility in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2015;59: 7320–7326. pmid:26369957
- View Article
- PubMed/NCBI
- Google Scholar
22. Zhang S, Chen J, Shi W, Liu W, Zhang W, Zhang Y. Mutations in panD encoding aspartate decarboxylase are associated with pyrazinamide resistance in Mycobacterium tuberculosis. Emerg Microbes Infect. 2013;2: 1–5. pmid:26038471
- View Article
- PubMed/NCBI
- Google Scholar
23. Bhatt A, Molle V, Besra GS, Jacobs WR Jr, Kremer L. The Mycobacterium tuberculosis FAS-II condensing enzymes: their role in mycolic acid biosynthesis, acid-fastness, pathogenesis and in future drug development. Mol Microbiol. 2007;64: 1442–1454. pmid:17555433
- View Article
- PubMed/NCBI
- Google Scholar
24. Zimhony O, Cox JS, Welch JT, Vilcheze C, Jacobs WR. Pyrazinamide inhibits the eukaryotic-like fatty acid synthetase I (FASI) of Mycobacterium tuberculosis. Nat Med. 2000;6: 1043–1047. pmid:10973326
- View Article
- PubMed/NCBI
- Google Scholar
25. Shi W, Zhang X, Jiang X, Yuan H, Lee JS, Barry 3rd CE, et al. Pyrazinamide inhibits trans-translation in Mycobacterium tuberculosis. Science (1979). 2011;333: 1630–1632. pmid:21835980
- View Article
- PubMed/NCBI
- Google Scholar
26. Shi W, Chen J, Feng J, Cui P, Zhang S, Weng X, et al. Aspartate decarboxylase (PanD) as a new target of pyrazinamide in Mycobacterium tuberculosis. Emerg Microbes Infect. 2014;3: 1–8. pmid:26038753
- View Article
- PubMed/NCBI
- Google Scholar
27. Zhang Y, Scorpio A, Nikaido H, Sun Z. Role of acid pH and deficient efflux of pyrazinoic acid in unique susceptibility of Mycobacterium tuberculosis to pyrazinamide. J Bacteriol. 1999;181: 2044–2049. pmid:10094680
- View Article
- PubMed/NCBI
- Google Scholar
28. Sheen P, Ferrer P, Gilman RH, López-Llano J, Fuentes P, Valencia E, et al. Effect of pyrazinamidase activity on pyrazinamide resistance in Mycobacterium tuberculosis. Tuberculosis. 2009;89: 109–113. pmid:19249243
- View Article
- PubMed/NCBI
- Google Scholar
29. Tarshis MS, Weed WA Jr. Lack of significant in vitro sensitivity of Mycobacterium tuberculosis to pyrazinamide on three different solid media. Am Rev Tuberc. 1953;67: 391–395. pmid:13031058
- View Article
- PubMed/NCBI
- Google Scholar
30. Vandal OH, Nathan CF, Ehrt S. Acid resistance in Mycobacterium tuberculosis. J Bacteriol. 2009;191: 4714–4721. pmid:19465648
- View Article
- PubMed/NCBI
- Google Scholar
31. Zhang Y, Permar S, Sun Z. Conditions that may affect the results of susceptibility testing of Mycobacterium tuberculosis to pyrazinamide. J Med Microbiol. 2002;51: 42–49. pmid:11800471
- View Article
- PubMed/NCBI
- Google Scholar
32. Thiede JM, Dillon NA, Howe MD, Aflakpui R, Modlin SJ, Hoffner SE, et al. Pyrazinamide susceptibility is driven by activation of the SigE-dependent cell envelope stress response in mycobacterium tuberculosis. mBio. 2022;13: e00439–21.
- View Article
- Google Scholar
33. Vásquez-Campos L, Asencios-Solis L, Leo-Hurtado E, Quispe-Torres N, Salazar-Lindo E, Bayona J, et al. Drug resistance trends among previously treated tuberculosis patients in a national registry in Peru, 1994–2001. The International Journal of Tuberculosis and Lung Disease. 2004;8: 465–472. pmid:15141740
- View Article
- PubMed/NCBI
- Google Scholar
34. Cheng S-J, Thibert L, Sanchez T, Heifets L, Zhang Y. pncA mutations as a major mechanism of pyrazinamide resistance in Mycobacterium tuberculosis: spread of a monoresistant strain in Quebec, Canada. Antimicrob Agents Chemother. 2000;44: 528–532. pmid:10681313
- View Article
- PubMed/NCBI
- Google Scholar
35. Kent PT. Public health mycobacteriology: a guide for the level III laboratory. US Department of Health and Human Services, Public Health Service, Centers~…; 1985.
- View Article
- Google Scholar
36. Zimic M, Sheen P, Quiliano M, Gutierrez A, Gilman RH. Peruvian and globally reported amino acid substitutions on the Mycobacterium tuberculosis pyrazinamidase suggest a conserved pattern of mutations associated to pyrazinamide resistance. Infection, Genetics and Evolution. 2010;10: 346–349. pmid:19963078
- View Article
- PubMed/NCBI
- Google Scholar
37. Sheen P, Méndez M, Gilman RH, Peña L, Caviedes L, Zimic MJ, et al. Sputum PCR-single-strand conformational polymorphism test for same-day detection of pyrazinamide resistance in tuberculosis patients. J Clin Microbiol. 2009;47: 2937–2943. pmid:19535526
- View Article
- PubMed/NCBI
- Google Scholar
38. Karmakar M, Rodrigues CHM, Horan K, Denholm JT, Ascher DB. Structure guided prediction of Pyrazinamide resistance mutations in pncA. Sci Rep. 2020;10: 1–10.
- View Article
- Google Scholar
39. Supo-Escalante RR, Médico A, Gushiken E, Olivos-Ram\’\irez GE, Quispe Y, Torres Fet al. Prediction of Mycobacterium tuberculosis pyrazinamidase function based on structural stability, physicochemical and geometrical descriptors. PLoS One. 2020;15: e0235643. pmid:32735615
- View Article
- PubMed/NCBI
- Google Scholar
40. Davies AP, Billington OJ, McHugh TD, Mitchison DA, Gillespie SH. Comparison of phenotypic and genotypic methods for pyrazinamide susceptibility testing with Mycobacterium tuberculosis. J Clin Microbiol. 2000;38: 3686–3688. pmid:11015384
- View Article
- PubMed/NCBI
- Google Scholar
41. Endoh T, Yagihashi A, Uehara N, Kobayashi D, Tsuji N, Nakamura M, et al. Pyrazinamide resistance associated with pncA gene mutation in Mycobacterium tuberculosis in Japan. Epidemiol Infect. 2002;128: 337–342. pmid:12002553
- View Article
- PubMed/NCBI
- Google Scholar
42. Hou L, Osei-Hyiaman D, Zhang Z, Wang B, Yang A, Kano K. Molecular characterization of pncA gene mutations in Mycobacterium tuberculosis clinical isolates from China. Epidemiol Infect. 2000;124: 227–232. pmid:10813147
- View Article
- PubMed/NCBI
- Google Scholar
43. Zimic M, Fuentes P, Gilman RH, Gutiérrez AH, Kirwan D, Sheen P. Pyrazinoic acid efflux rate in Mycobacterium tuberculosis is a better proxy of pyrazinamide resistance. Tuberculosis. 2012;92: 84–91. pmid:22004792
- View Article
- PubMed/NCBI
- Google Scholar
44. Zimic M, Loli S, Gilman RH, Gutierrez A, Fuentes P, Cotrina M, et al. A new approach for pyrazinamide susceptibility testing in Mycobacterium tuberculosis. Microbial Drug Resistance. 2012;18: 372–375. pmid:22372927
- View Article
- PubMed/NCBI
- Google Scholar
45. Singh A, Jayaraman A, Hahn J. Modeling regulatory mechanisms in IL-6 signal transduction in hepatocytes. Biotechnol Bioeng. 2006;95: 850–862. pmid:16752369
- View Article
- PubMed/NCBI
- Google Scholar
46. Kiparissides A, Koutinas M, Moss T, Newman J, Pistikopoulos EN, Mantalaris A. Modelling the Delta1/Notch1 pathway: in search of the mediator (s) of neural stem cell differentiation. PLoS One. 2011;6: e14668. pmid:21346804
- View Article
- PubMed/NCBI
- Google Scholar
47. Sedaghat AR, Sherman A, Quon MJ. A mathematical model of metabolic insulin signaling pathways. American Journal of Physiology-Endocrinology and Metabolism. 2002;283: E1084–E1101. pmid:12376338
- View Article
- PubMed/NCBI
- Google Scholar
48. Charusanti P, Hu X, Chen L, Neuhauser D, DiStefano JJ III. A mathematical model of BCR-ABL autophosphorylation, signaling through the CRKL pathway, and gleevec dynamics in chronic myeloid leukemia. Discrete & Continuous Dynamical Systems-B. 2004;4: 99.
- View Article
- Google Scholar
49. Budavari SO’Neil MJ, Heckelman PE. The merck index an encyclopedia of chemical, drugs, and biologicals. The merck index an encyclopedia of chemical, drugs, and biologicals. Merck & Co; 1989.
- View Article
- Google Scholar
50. Lemaitre N, Sougakoff W, Truffot-Pernot C, Jarlier V. Characterization of new mutations in pyrazinamide-resistant strains of Mycobacterium tuberculosis and identification of conserved regions important for the catalytic activity of the pyrazinamidase PncA. Antimicrob Agents Chemother. 1999;43: 1761–1763. pmid:10390238
- View Article
- PubMed/NCBI
- Google Scholar
51. Scorpio A, Lindholm-Levy P, Heifets L, Gilman R, Siddiqi S, Cynamon M, et al. Characterization of pncA mutations in pyrazinamide-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother. 1997;41: 540–543. pmid:9055989
- View Article
- PubMed/NCBI
- Google Scholar
52. Baughn AD, Deng J, Vilcheze C, Riestra A, Welch JT, Jacobs WR Jr, et al. Mutually exclusive genotypes for pyrazinamide and 5-chloropyrazinamide resistance reveal a potential resistance-proofing strategy. Antimicrob Agents Chemother. 2010;54: 5323–5328. pmid:20876380
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Patterson B, Wood R. Is cough really necessary for TB transmission? Tuberculosis. 2019;117: 31–35. pmid:31378265
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Gengenbacher M, Kaufmann SHE. Mycobacterium tuberculosis: success through dormancy. FEMS Microbiol Rev. 2012;36: 514–532. pmid:22320122
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Hopewell PC, Jasmer RM. Overview of clinical tuberculosis. Tuberculosis and the tubercle bacillus. 2004; 13–31.
View Article
Google Scholar

[10] View Article

[11] Google Scholar

[ref4] 4. Hu Y, Coates AR, Mitchison DA. Sterilising action of pyrazinamide in models of dormant and rifampicin-tolerant Mycobacterium tuberculosis. The International Journal of Tuberculosis and Lung Disease. 2006;10: 317–322. pmid:16562713
View Article
PubMed/NCBI
Google Scholar

[13] View Article

[14] PubMed/NCBI

[15] Google Scholar

[ref5] 5. Johnson R, Streicher EM, Louw GE, Warren RM, Van Helden PD, Victor TC. Drug resistance in Mycobacterium tuberculosis. Curr Issues Mol Biol. 2006;8: 97–112. pmid:16878362
View Article
PubMed/NCBI
Google Scholar

[17] View Article

[18] PubMed/NCBI

[19] Google Scholar

[ref6] 6. Mackaness GB. The intracellular activation of pyrazinamide and nicotinamide. American Review of Tuberculosis and Pulmonary Diseases. 1956;74: 718–728. pmid:13372956
View Article
PubMed/NCBI
Google Scholar

[21] View Article

[22] PubMed/NCBI

[23] Google Scholar

[ref7] 7. McDermott W, Tompsett R. Activation of pyrazinamide and nicotinamide in acidic environments in vitro. Am Rev Tuberc. 1954;70: 748–754. pmid:13197751
View Article
PubMed/NCBI
Google Scholar

[25] View Article

[26] PubMed/NCBI

[27] Google Scholar

[ref8] 8. Grosset J. New experimental regimens for preventive therapy of tuberculosis. Bull Int Union Tuberc Lung Dis. 1990;66: 15–16. pmid:1983742
View Article
PubMed/NCBI
Google Scholar

[29] View Article

[30] PubMed/NCBI

[31] Google Scholar

[ref9] 9. Mitchison DA. The action of antituberculosis drugs in short-course chemotherapy. Tubercle. 1985;66: 219–225. pmid:3931319
View Article
PubMed/NCBI
Google Scholar

[33] View Article

[34] PubMed/NCBI

[35] Google Scholar

[ref10] 10. Styblo K, Enarson DA. Epidemiology of tuberculosis. Vol. 24. The Hague: Royal Netherlands Tuberculosis Association. 1984.

[ref11] 11. Raviglione MC, Gupta R, Dye CM, Espinal MA. The burden of drug-resistant tuberculosis and mechanisms for its control. Ann N Y Acad Sci. 2001;953: 88–97. pmid:11795426
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref12] 12. Zhang Y, Mitchison D. The curious characteristics of pyrazinamide: a review. The international journal of tuberculosis and lung disease. 2003;7: 6–21. pmid:12701830
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref13] 13. Zhang Y, Shi W, Zhang W, Mitchison D. Mechanisms of pyrazinamide action and resistance. Microbiol Spectr. 2014;2: 2–4. pmid:26104205
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref14] 14. Corbett EL, Watt CJ, Walker N, Maher D, Williams BG, Raviglione MC, et al. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch Intern Med. 2003;163: 1009–1021. pmid:12742798
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref15] 15. Wu X, Zhang J, Zhong M. Studies on pncA gene mutations in M. tuberculosis isolates. Zhonghua Jie He He Hu Xi Za Zhi. 2000;23: 40–42. pmid:11778181
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref16] 16. Zhang Y, Zhang J, Cui P, Zhang Y, Zhang W. Identification of novel efflux proteins Rv0191, Rv3756c, Rv3008, and Rv1667c involved in pyrazinamide resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2017;61: e00940–17. pmid:28584158
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref17] 17. Liu J, Shi W, Zhang S, Hao X, Maslov DA, Shur K V, et al. Mutations in efflux pump Rv1258c (Tap) cause resistance to pyrazinamide, isoniazid, and streptomycin in M. tuberculosis. Front Microbiol. 2019;10: 216. pmid:30837962
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref18] 18. Sheen P, Requena D, Gushiken E, Gilman RH, Antiparra R, Lucero B, et al. A multiple genome analysis of Mycobacterium tuberculosis reveals specific novel genes and mutations associated with pyrazinamide resistance. BMC Genomics. 2017;18: 1–11.
View Article
Google Scholar

[66] View Article

[67] Google Scholar

[ref19] 19. Lamont EA, Dillon NA, Baughn AD. The bewildering antitubercular action of pyrazinamide. Microbiology and Molecular Biology Reviews. 2020;84: e00070–19. pmid:32132245
View Article
PubMed/NCBI
Google Scholar

[69] View Article

[70] PubMed/NCBI

[71] Google Scholar

[ref20] 20. Montenegro SH, Gilman RH, Sheen P, Cama R, Caviedes L, Hopper T, et al. Improved detection of Mycobacterium tuberculosis in Peruvian children by use of a heminested IS6110 polymerase chain reaction assay. Clinical Infectious Diseases. 2003;36: 16–23. pmid:12491196
View Article
PubMed/NCBI
Google Scholar

[73] View Article

[74] PubMed/NCBI

[75] Google Scholar

[ref21] 21. Peterson ND, Rosen BC, Dillon NA, Baughn AD. Uncoupling environmental pH and intrabacterial acidification from pyrazinamide susceptibility in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2015;59: 7320–7326. pmid:26369957
View Article
PubMed/NCBI
Google Scholar

[77] View Article

[78] PubMed/NCBI

[79] Google Scholar

[ref22] 22. Zhang S, Chen J, Shi W, Liu W, Zhang W, Zhang Y. Mutations in panD encoding aspartate decarboxylase are associated with pyrazinamide resistance in Mycobacterium tuberculosis. Emerg Microbes Infect. 2013;2: 1–5. pmid:26038471
View Article
PubMed/NCBI
Google Scholar

[81] View Article

[82] PubMed/NCBI

[83] Google Scholar

[ref23] 23. Bhatt A, Molle V, Besra GS, Jacobs WR Jr, Kremer L. The Mycobacterium tuberculosis FAS-II condensing enzymes: their role in mycolic acid biosynthesis, acid-fastness, pathogenesis and in future drug development. Mol Microbiol. 2007;64: 1442–1454. pmid:17555433
View Article
PubMed/NCBI
Google Scholar

[85] View Article

[86] PubMed/NCBI

[87] Google Scholar

[ref24] 24. Zimhony O, Cox JS, Welch JT, Vilcheze C, Jacobs WR. Pyrazinamide inhibits the eukaryotic-like fatty acid synthetase I (FASI) of Mycobacterium tuberculosis. Nat Med. 2000;6: 1043–1047. pmid:10973326
View Article
PubMed/NCBI
Google Scholar

[89] View Article

[90] PubMed/NCBI

[91] Google Scholar

[ref25] 25. Shi W, Zhang X, Jiang X, Yuan H, Lee JS, Barry 3rd CE, et al. Pyrazinamide inhibits trans-translation in Mycobacterium tuberculosis. Science (1979). 2011;333: 1630–1632. pmid:21835980
View Article
PubMed/NCBI
Google Scholar

[93] View Article

[94] PubMed/NCBI

[95] Google Scholar

[ref26] 26. Shi W, Chen J, Feng J, Cui P, Zhang S, Weng X, et al. Aspartate decarboxylase (PanD) as a new target of pyrazinamide in Mycobacterium tuberculosis. Emerg Microbes Infect. 2014;3: 1–8. pmid:26038753
View Article
PubMed/NCBI
Google Scholar

[97] View Article

[98] PubMed/NCBI

[99] Google Scholar

[ref27] 27. Zhang Y, Scorpio A, Nikaido H, Sun Z. Role of acid pH and deficient efflux of pyrazinoic acid in unique susceptibility of Mycobacterium tuberculosis to pyrazinamide. J Bacteriol. 1999;181: 2044–2049. pmid:10094680
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref28] 28. Sheen P, Ferrer P, Gilman RH, López-Llano J, Fuentes P, Valencia E, et al. Effect of pyrazinamidase activity on pyrazinamide resistance in Mycobacterium tuberculosis. Tuberculosis. 2009;89: 109–113. pmid:19249243
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref29] 29. Tarshis MS, Weed WA Jr. Lack of significant in vitro sensitivity of Mycobacterium tuberculosis to pyrazinamide on three different solid media. Am Rev Tuberc. 1953;67: 391–395. pmid:13031058
View Article
PubMed/NCBI
Google Scholar

[109] View Article

[110] PubMed/NCBI

[111] Google Scholar

[ref30] 30. Vandal OH, Nathan CF, Ehrt S. Acid resistance in Mycobacterium tuberculosis. J Bacteriol. 2009;191: 4714–4721. pmid:19465648
View Article
PubMed/NCBI
Google Scholar

[113] View Article

[114] PubMed/NCBI

[115] Google Scholar

[ref31] 31. Zhang Y, Permar S, Sun Z. Conditions that may affect the results of susceptibility testing of Mycobacterium tuberculosis to pyrazinamide. J Med Microbiol. 2002;51: 42–49. pmid:11800471
View Article
PubMed/NCBI
Google Scholar

[117] View Article

[118] PubMed/NCBI

[119] Google Scholar

[ref32] 32. Thiede JM, Dillon NA, Howe MD, Aflakpui R, Modlin SJ, Hoffner SE, et al. Pyrazinamide susceptibility is driven by activation of the SigE-dependent cell envelope stress response in mycobacterium tuberculosis. mBio. 2022;13: e00439–21.
View Article
Google Scholar

[121] View Article

[122] Google Scholar

[ref33] 33. Vásquez-Campos L, Asencios-Solis L, Leo-Hurtado E, Quispe-Torres N, Salazar-Lindo E, Bayona J, et al. Drug resistance trends among previously treated tuberculosis patients in a national registry in Peru, 1994–2001. The International Journal of Tuberculosis and Lung Disease. 2004;8: 465–472. pmid:15141740
View Article
PubMed/NCBI
Google Scholar

[124] View Article

[125] PubMed/NCBI

[126] Google Scholar

[ref34] 34. Cheng S-J, Thibert L, Sanchez T, Heifets L, Zhang Y. pncA mutations as a major mechanism of pyrazinamide resistance in Mycobacterium tuberculosis: spread of a monoresistant strain in Quebec, Canada. Antimicrob Agents Chemother. 2000;44: 528–532. pmid:10681313
View Article
PubMed/NCBI
Google Scholar

[128] View Article

[129] PubMed/NCBI

[130] Google Scholar

[ref35] 35. Kent PT. Public health mycobacteriology: a guide for the level III laboratory. US Department of Health and Human Services, Public Health Service, Centers~…; 1985.
View Article
Google Scholar

[132] View Article

[133] Google Scholar

[ref36] 36. Zimic M, Sheen P, Quiliano M, Gutierrez A, Gilman RH. Peruvian and globally reported amino acid substitutions on the Mycobacterium tuberculosis pyrazinamidase suggest a conserved pattern of mutations associated to pyrazinamide resistance. Infection, Genetics and Evolution. 2010;10: 346–349. pmid:19963078
View Article
PubMed/NCBI
Google Scholar

[135] View Article

[136] PubMed/NCBI

[137] Google Scholar

[ref37] 37. Sheen P, Méndez M, Gilman RH, Peña L, Caviedes L, Zimic MJ, et al. Sputum PCR-single-strand conformational polymorphism test for same-day detection of pyrazinamide resistance in tuberculosis patients. J Clin Microbiol. 2009;47: 2937–2943. pmid:19535526
View Article
PubMed/NCBI
Google Scholar

[139] View Article

[140] PubMed/NCBI

[141] Google Scholar

[ref38] 38. Karmakar M, Rodrigues CHM, Horan K, Denholm JT, Ascher DB. Structure guided prediction of Pyrazinamide resistance mutations in pncA. Sci Rep. 2020;10: 1–10.
View Article
Google Scholar

[143] View Article

[144] Google Scholar

[ref39] 39. Supo-Escalante RR, Médico A, Gushiken E, Olivos-Ram\’\irez GE, Quispe Y, Torres Fet al. Prediction of Mycobacterium tuberculosis pyrazinamidase function based on structural stability, physicochemical and geometrical descriptors. PLoS One. 2020;15: e0235643. pmid:32735615
View Article
PubMed/NCBI
Google Scholar

[146] View Article

[147] PubMed/NCBI

[148] Google Scholar

[ref40] 40. Davies AP, Billington OJ, McHugh TD, Mitchison DA, Gillespie SH. Comparison of phenotypic and genotypic methods for pyrazinamide susceptibility testing with Mycobacterium tuberculosis. J Clin Microbiol. 2000;38: 3686–3688. pmid:11015384
View Article
PubMed/NCBI
Google Scholar

[150] View Article

[151] PubMed/NCBI

[152] Google Scholar

[ref41] 41. Endoh T, Yagihashi A, Uehara N, Kobayashi D, Tsuji N, Nakamura M, et al. Pyrazinamide resistance associated with pncA gene mutation in Mycobacterium tuberculosis in Japan. Epidemiol Infect. 2002;128: 337–342. pmid:12002553
View Article
PubMed/NCBI
Google Scholar

[154] View Article

[155] PubMed/NCBI

[156] Google Scholar

[ref42] 42. Hou L, Osei-Hyiaman D, Zhang Z, Wang B, Yang A, Kano K. Molecular characterization of pncA gene mutations in Mycobacterium tuberculosis clinical isolates from China. Epidemiol Infect. 2000;124: 227–232. pmid:10813147
View Article
PubMed/NCBI
Google Scholar

[158] View Article

[159] PubMed/NCBI

[160] Google Scholar

[ref43] 43. Zimic M, Fuentes P, Gilman RH, Gutiérrez AH, Kirwan D, Sheen P. Pyrazinoic acid efflux rate in Mycobacterium tuberculosis is a better proxy of pyrazinamide resistance. Tuberculosis. 2012;92: 84–91. pmid:22004792
View Article
PubMed/NCBI
Google Scholar

[162] View Article

[163] PubMed/NCBI

[164] Google Scholar

[ref44] 44. Zimic M, Loli S, Gilman RH, Gutierrez A, Fuentes P, Cotrina M, et al. A new approach for pyrazinamide susceptibility testing in Mycobacterium tuberculosis. Microbial Drug Resistance. 2012;18: 372–375. pmid:22372927
View Article
PubMed/NCBI
Google Scholar

[166] View Article

[167] PubMed/NCBI

[168] Google Scholar

[ref45] 45. Singh A, Jayaraman A, Hahn J. Modeling regulatory mechanisms in IL-6 signal transduction in hepatocytes. Biotechnol Bioeng. 2006;95: 850–862. pmid:16752369
View Article
PubMed/NCBI
Google Scholar

[170] View Article

[171] PubMed/NCBI

[172] Google Scholar

[ref46] 46. Kiparissides A, Koutinas M, Moss T, Newman J, Pistikopoulos EN, Mantalaris A. Modelling the Delta1/Notch1 pathway: in search of the mediator (s) of neural stem cell differentiation. PLoS One. 2011;6: e14668. pmid:21346804
View Article
PubMed/NCBI
Google Scholar

[174] View Article

[175] PubMed/NCBI

[176] Google Scholar

[ref47] 47. Sedaghat AR, Sherman A, Quon MJ. A mathematical model of metabolic insulin signaling pathways. American Journal of Physiology-Endocrinology and Metabolism. 2002;283: E1084–E1101. pmid:12376338
View Article
PubMed/NCBI
Google Scholar

[178] View Article

[179] PubMed/NCBI

[180] Google Scholar

[ref48] 48. Charusanti P, Hu X, Chen L, Neuhauser D, DiStefano JJ III. A mathematical model of BCR-ABL autophosphorylation, signaling through the CRKL pathway, and gleevec dynamics in chronic myeloid leukemia. Discrete & Continuous Dynamical Systems-B. 2004;4: 99.
View Article
Google Scholar

[182] View Article

[183] Google Scholar

[ref49] 49. Budavari SO’Neil MJ, Heckelman PE. The merck index an encyclopedia of chemical, drugs, and biologicals. The merck index an encyclopedia of chemical, drugs, and biologicals. Merck & Co; 1989.
View Article
Google Scholar

[185] View Article

[186] Google Scholar

[ref50] 50. Lemaitre N, Sougakoff W, Truffot-Pernot C, Jarlier V. Characterization of new mutations in pyrazinamide-resistant strains of Mycobacterium tuberculosis and identification of conserved regions important for the catalytic activity of the pyrazinamidase PncA. Antimicrob Agents Chemother. 1999;43: 1761–1763. pmid:10390238
View Article
PubMed/NCBI
Google Scholar

[188] View Article

[189] PubMed/NCBI

[190] Google Scholar

[ref51] 51. Scorpio A, Lindholm-Levy P, Heifets L, Gilman R, Siddiqi S, Cynamon M, et al. Characterization of pncA mutations in pyrazinamide-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother. 1997;41: 540–543. pmid:9055989
View Article
PubMed/NCBI
Google Scholar

[192] View Article

[193] PubMed/NCBI

[194] Google Scholar

[ref52] 52. Baughn AD, Deng J, Vilcheze C, Riestra A, Welch JT, Jacobs WR Jr, et al. Mutually exclusive genotypes for pyrazinamide and 5-chloropyrazinamide resistance reveal a potential resistance-proofing strategy. Antimicrob Agents Chemother. 2010;54: 5323–5328. pmid:20876380
View Article
PubMed/NCBI
Google Scholar

[196] View Article

[197] PubMed/NCBI

[198] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Mathematical model of POA accumulation

Non-linear optimization fitting with experimental data of cytoplasm acidification

Numerical simulations, parameter sweep and equilibrium analysis

Exploration of scenarios of PZA resistance and susceptibility

Results

Mathematical model of POA accumulation

Equilibrium analysis of the effect of

Parameter sweep of [PZA]e and pHe in wt MTB indicates a pH-dependent POA accumulation attainable at equilibrium but not at experimental timescales

Resistance and susceptibility scenarios display different regimes of accumulation of [POAT]i and cytoplasmic acidification

Discussion

Conclusions

Supporting information

S1 Fig. Relationship between model parameters at equilibrium.

S2 Fig. Heatmaps of the predicted pHi at equilibrium for the five studied scenarios.

S3 Fig. Heatmaps of the predicted pHi at 240 hours of simulation for the five studied scenarios.

S4 Fig. Heatmaps of the predicted [POAT]i at equilibrium for the susceptibility/resistance studied scenarios.

S5 Fig. Heatmaps of the predicted pHi at equilibrium for the susceptibility and resistance studied scenarios.

S6 Fig. Heatmaps of the predicted [POAT]i at 240 hours of simulation for the susceptibility and resistance studied scenarios.

S7 Fig. Heatmaps of the predicted pHi at 240 hours of simulation for the susceptibility and resistance studied scenarios.

References

Parameter sweep of [PZA]_e and pH_e in wt MTB indicates a pH-dependent POA accumulation attainable at equilibrium but not at experimental timescales

Resistance and susceptibility scenarios display different regimes of accumulation of [POA_T]_i and cytoplasmic acidification

S2 Fig. Heatmaps of the predicted pH_i at equilibrium for the five studied scenarios.

S3 Fig. Heatmaps of the predicted pH_i at 240 hours of simulation for the five studied scenarios.

S4 Fig. Heatmaps of the predicted [POA_T]_i at equilibrium for the susceptibility/resistance studied scenarios.

S5 Fig. Heatmaps of the predicted pH_i at equilibrium for the susceptibility and resistance studied scenarios.

S6 Fig. Heatmaps of the predicted [POA_T]_i at 240 hours of simulation for the susceptibility and resistance studied scenarios.

S7 Fig. Heatmaps of the predicted pH_i at 240 hours of simulation for the susceptibility and resistance studied scenarios.