Abstract
The dynamics of the bacterial population that comprises the gut microbiota plays key roles in overall mammalian health. However, a detailed understanding of bacterial growth within the gut is limited by the inherent complexity and inaccessibility of the gut environment. Here, we deploy an improved synthetic genetic oscillator to investigate dynamics of bacterial colonization and growth in the mammalian gut under both healthy and disease conditions. The synthetic oscillator, when introduced into both Escherichia coli and Salmonella Typhimurium maintains regular oscillations with a constant period in generations across growth conditions. We determine the phase of oscillation from individual bacteria using image analysis of resultant colonies and thereby infer the number of cell divisions elapsed. In doing so, we demonstrate robust functionality and controllability of the oscillator circuit’s activity during bacterial growth in vitro, in a simulated murine gut microfluidic environment, and in vivo within the mouse gut. We determine different dynamics of bacterial colonization and growth in the gut under normal and inflammatory conditions. Our results show that a precise genetic oscillator can function in a complex environment and reveal single cell behavior under diverse conditions where disease may create otherwise impossible-to-quantify variability in growth across the population.
Introduction: The Gut as a "Black Box"
The mammalian gut is a complex, dynamic environment where bacterial growth and survival are critical to host health. Despite its importance, the gut remains a "black box" for researchers attempting to measure bacterial behavior in real-time. Traditional methods, such as counting Colony Forming Units (CFUs) from fecal samples, provide only a population-wide average of bacterial density. These metrics mask the heterogeneity of individual cells—the "lone wolves" that may grow faster or slower than the mean.
Existing single-cell methods, like fluorescent protein dilution, are limited by signal decay, typically vanishing after approximately ten cell divisions. To truly understand how pathogens invade or how probiotics establish themselves, a method is needed to track cell divisions at the single-cell level over extended periods without losing signal integrity. This study introduces a synthetic biology approach to overcome these hurdles using a genetic "clock."
The Repressilator 2.0 and RINGS Workflow
The researchers utilized a redesigned synthetic gene circuit known as the Repressilator 2.0. This circuit consists of a three-node negative feedback loop (using TetR, cI, and LacI proteins) that oscillates like a pendulum. Unlike clocks that track absolute time, the Repressilator’s "ticks" are linked to bacterial generations (cell divisions). The period of this oscillation remains remarkably stable—approximately 15 generations—regardless of the bacterial species or nutrient conditions.
To translate these molecular oscillations into readable data, the team developed the RINGS (Repressilator-based INference of Growth at Single-cell level) workflow. The process begins by sampling bacteria from mouse feces and plating them on agar. As a single cell grows into a macroscopic colony, it grows radially. Cells at the center stop dividing first, while those at the periphery continue. This growth pattern "freezes" the state of the oscillator in space, creating concentric fluorescent rings. By applying a generative mathematical model to these images, the researchers can back-calculate the exact "phase" of the clock at the moment the bacterium was sampled from the gut.
Experimental Validation
The study validated the Repressilator 2.0 across several models:
- Strains: The circuit was successfully introduced into Escherichia coli (including the probiotic Nissle 1917) and Salmonella Typhimurium.
- Environments: Validation occurred in rich media, "Mother Machine" microfluidic devices using extracted mouse cecum fluid, and living mouse models.
- Stability: The oscillator remained functional and stable in the gut for over 16 days without the need for antibiotic selection pressure, demonstrating its robustness for long-term studies.
Key Insights: Growth in Health and Disease
One of the study's most significant contributions is the ability to "reset" the bacterial clocks in situ. By adding chemical inducers like IPTG to the drinking water of mice, researchers could synchronize the bacterial population's clocks while they were still inside the host. This allows for precise "start-stop" experiments to measure growth rates during specific windows of time.
Inflammation and Heterogeneity
The most striking finding emerged from comparing healthy mice to those with intestinal inflammation (induced via Dextran Sulfate Sodium, or DSS). In a healthy gut, bacterial growth is relatively uniform; the population moves through generations at a similar pace. However, in the inflamed gut, this synchrony collapses. The population splits into subpopulations growing at wildly different rates. This suggests that inflammation creates diverse, localized nutrient niches or physical stressors that affect individual bacteria differently—a phenomenon that population-level CFU counts would fail to detect.
Results and Metrics
- Clock Precision: The oscillator period was measured consistently at 15.3 to 15.6 generations per cycle.
- Dynamic Range: The method captured generation times ranging from ~35 minutes during rapid early colonization to ~10 hours once the native microbiota recovered and increased colonization resistance.
- Single-Cell Variance: Using "angular deviation" as a metric for clock disagreement, the researchers proved that inflammation significantly increases growth rate variability across the bacterial population.
Discussion and Future Directions
While the Repressilator 2.0 and RINGS workflow provide unprecedented insight, the method has limitations. The primary drawback is the sampling lag; because the readout requires growing a colony from a fecal sample, the data is not real-time. Furthermore, the mathematical "unwrapping" of the clock phase assumes that subpopulations do not drift more than 15 generations apart from the mean, which could limit analysis in extremely divergent populations.
Looking forward, this technology paves the way for "growth-linked" synthetic biology. Bacteria could be engineered to trigger therapeutic releases only after reaching a specific number of divisions or when a certain growth rate is detected, allowing for highly localized and context-dependent drug delivery within the mammalian gut.
Summary updated