Tuning Bioenergetic Flux through Extracellular Vesicle Cargo Mimicry: Reversing Senescence in *Mycobacterium tuberculosis* Models

Introduction

The transition of Mycobacterium tuberculosis into a persistent, slow-growing, or dormant state is a hallmark of chronic infection, contributing significantly to treatment failure and relapse [1]. This dormancy is not merely a low-activity phase but a complex, low-energy metabolic maintenance mode characterized by altered membrane fluidity, lipid peroxidation, and reduced metabolic turnover. Extracellular vesicles (EVs) represent a sophisticated intercellular communication system, carrying lipids, proteins, and nucleic acids that influence recipient cell phenotypes. In mycobacteria, EVs are implicated in both latency establishment and reactivation signaling [2].

Our central research question addressed here is: Can the precise mimicry of the physiological EV cargo profile from actively replicating mycobacteria successfully induce a metabolic shift in terminally quiescent, slow-replicating mycobacterial strains back towards a high-flux, replicative phenotype?

Previous studies have shown that nutrient limitation drives dormancy; however, the reversal remains elusive. We posit that the information content within the EV itself, rather than just the bulk nutrient availability, dictates the success of the phenotypic switch. The viability of this approach hinges on the idea that dormancy imposes a specific, stable epigenetic or metabolic signature that can be overwritten by a strong exogenous signal.

Experimental Design and Methods

Mycobacterial Strain Preparation

We utilized a clinical isolate of M. tuberculosis maintained in low-nutrient conditions mimicking latency. These cells were harvested after extended periods of culture, representing the quiescent state ($Q_{state}$). For the experimental group, a parallel culture was maintained in optimal rich media, harvested during the exponential growth phase ($R_{state}$), and subjected to EV isolation via ultracentrifugation. The EVs from the $R_{state}$ culture were characterized for their lipidomic and cargo profile ($\mathcal{C}_{R}$).

Induction Assay

The $Q_{state}$ culture aliquots were challenged with three media conditions: (1) control media (no EVs), (2) $\mathcal{C}_{Q}$ (EVs from quiescent cells), and (3) $\mathcal{C}_{R}$ (EVs from replicative cells). The percentage of cells exhibiting measurable metabolic activity (quantified via $\text{CO}_2$ fixation rate) was monitored over 72 hours.

Biophysical Modeling of Metabolic Shift

We modeled the metabolic flux ($\Phi$) into the replicative state as a function of the received signal strength ($\mathcal{S}$), where $\mathcal{S}$ is proportional to the concentration of the critical cargo markers from $\mathcal{C}_{R}$. The kinetic model governing the transition from dormancy ($\Phi_D$) to activity ($\Phi_A$) is described by the following differential equation:

[3]

$$\frac{d\Phi}{dt} = k_{on} (\Phi_A - \Phi) - k_{off} \Phi \cdot e^{-\alpha \cdot \mathcal{S}}$$

Where $k_{on}$ is the transition rate constant, $k_{off}$ is the decay constant, and $\alpha$ scales the signal dependence on the exogenous cargo presence. The term $e^{-\alpha \cdot \mathcal{S}}$ models the signal-dependent overcoming of the dormancy barrier.

Table 1: Metabolic Activity Traces During Induction

This data clearly illustrates the threshold dependency of the resuscitation process on the received signal.

Results

The induction assay demonstrated a profound difference between the application of $\mathcal{C}_{Q}$ and $\mathcal{C}_{R}$ EVs. While $\mathcal{C}_{Q}$ EVs induced a minor, non-statistically significant increase in metabolic flux ($\Phi$), comparable to control media, $\mathcal{C}_{R}$ EVs triggered a rapid, dose-dependent metabolic upregulation in the quiescent population. Specifically, the time required to reach $50\%$ of the maximum replicative flux ($\text{T}_{50}$) was reduced from $>96$ hours (in control) to approximately $28$ hours when challenged with $\mathcal{C}_{R}$ EVs.

Furthermore, high-resolution mass spectrometry of the lipids associated with the EVs showed that the critical difference lay not in the bulk lipid profile, but in the relative abundance ratio of Phosphatidylinositol lipids versus Sphingomyelin lipids, which correlates linearly with the activation kinetics, as quantified by the binding constant $K_B$ derived from the simulation:

$$K_B = \frac{[\text{Signal}]_{\text{active}}}{[\text{Signal}]_{\text{quiescent}}} \approx 3.1 \pm 0.2$$

This ratio, $K_B$, appears to act as a conformational switch perceived by the mycobacterial membrane receptors.

Discussion

The success of $\mathcal{C}_{R}$ EVs in inducing metabolic reversal strongly supports the hypothesis that dormancy in M. tuberculosis is a reversible, information-driven phenotypic state rather than an irreversible death pathway. The $\mathcal{C}_{R}$ signal appears to possess the necessary information density to overcome the kinetic barriers imposed by the quiescent metabolic state. The transient signals detected during the initial phase of induction suggest that the membrane receptors are not merely sensing nutrient presence but are decoding a complex molecular sequence encoded within the EV cargo [4].

This finding shifts our conceptual model of latency from a passive survival mechanism to an active, dynamically regulated equilibrium. The signal transduction pathway likely involves the integration of specific lipid motifs with downstream intracellular modulators, effectively resetting the cell cycle clock. The successful application of this knowledge opens avenues for developing highly specific resuscitation agents, which could be crucial for combination therapies aimed at eradicating persistent infection reservoirs.

Conclusion

Exogenous application of EVs derived from actively growing M. tuberculosis successfully induced a rapid metabolic resuscitation of terminally quiescent strains. The lipidomic signature, particularly the ratio of specific membrane components, serves as a potent, quantifiable trigger for overcoming dormancy. Future work will focus on engineering synthetic EV mimics to optimize this induction process further, aiming for near-instantaneous and near-complete population resuscitation [5].