Microfluidics and single-cell microscopy to study stochastic processes in bacteria

Abstract

Bacteria have molecules present in low and fluctuating numbers that randomize cell behaviors. Understanding these stochastic processes and their impact on cells has, until recently, been limited by the lack of single-cell measurement methods. Here, we review recent developments in microfluidics that enable following individual cells over long periods of time under precisely controlled conditions, and counting individual fluorescent molecules in many cells. We showcase discoveries that were made possible using these devices in various aspects of microbiology, such as antibiotic tolerance/persistence, cell-size control, cell-fate determination, DNA damage response, and synthetic biology.

Figure 1

The importance of single-cell and time-lapse measurements. a) Two dose-response mechanisms with identical average behaviors (blue lines), but different single-cell responses (dots, colored by density). On the left, cells progressively respond to the dose following the population average; on the right, cells are either completely on or off, and the fraction of the active population increases with the dose. b) Two stochastic processes can have the same probability distribution, but fluctuate on different timescales. The numbers of molecules are shown fluctuating over time (main panel), where the blue trace fluctuates more quickly than the red trace. Although they have identical probability distributions (inset, top), the two stochastic processes can be distinguished by their autocorrelation functions (inset, bottom).

Figure 2

Microfluidic devices for single-cell analyses. a) Diagram of the first device to support growth of bacteria in linear colonies with continuous time-lapse imaging. Cells grow and divide in strait grooves etched in PDMS (top panel, top-down view). Trenches are enclosed by a permeable membrane and fed diffusely from flowing media above (bottom panel, cross-sectional view). b) Top-down depiction of monolayer growth device. Bacteria are confined to a two-dimensional monolayer, fed by flowing media on both sides of the colony. c) Top-down schematic of the mother machine device. Cells are confined to dead-end linear trenches and fed by flowing media. Red dotted line represents segmentation of old-pole mother cells; yellow dotted line marks a single lineage with daughter cells sharing the same mother. d) Top-down cartoon of the single-cell chemostat device. Bacteria are geometrically constrained to a single dimension, but no old pole accumulates due to the open-ended structure. Flowing media feeds the trapped bacteria and washes away daughter cells. e) Cross-sectional view of the MACS platform. A soft ceiling is expanded by controlled pressure to trap single bacteria from a liquid culture. Squeezing the cells aids single-molecule detection (inset). f) Single field-of-view from MACS. Bacteria with a constitutive segmentation reporter (red) are captured and imaged with MACS. A rare transient phenotype is observed in yellow. g) Kymograph of mother machine cells with fluorescent protein expression reporting oscillations of a synthetic gene circuit [33]. Images, taken at regular time points, of a single lineage (as marked by the yellow-dotted box in panel c) are concatenated, with time represented on the horizontal axis. h) Single-cell trace from the mother machine. A reporter of the synthetic oscillator (from panel g) is shown in blue; cell area, capturing growth and division, of the same cell is plotted in red.