Single molecule narrowfield microscopy of protein-DNA binding dynamics in glucose signal transduction of live yeast cells

Single-molecule narrowfield microscopy is a versatile tool to investigate a diverse range of protein dynamics in live cells and has been extensively used in bacteria. Here, we describe how these methods can be extended to larger eukaryotic, yeast cells, which contain sub-cellular compartments. We describe how to obtain single-molecule microscopy data but also how to analyse these data to track and obtain the stoichiometry of molecular complexes diffusing in the cell. We chose glucose mediated signal transduction of live yeast cells as the system to demonstrate these single-molecule techniques as transcriptional regulation is fundamentally a single molecule problem - a single repressor protein binding a single binding site in the genome can dramatically alter behaviour at the whole cell and population level.

💡 Research Summary

This paper presents a comprehensive workflow for applying single‑molecule narrowfield microscopy to study protein‑DNA binding dynamics in the glucose‑signalling pathway of live Saccharomyces cerevisiae cells. While narrowfield microscopy has been widely used in bacteria, the authors extend the technique to the larger, compartmentalised eukaryotic yeast cell, demonstrating that it can resolve the stochastic behaviour of individual transcription regulators in their native nuclear environment.

The experimental setup consists of a high‑NA (1.49) oil‑immersion objective coupled to a 561 nm laser that is focused to a sub‑micron illumination field (≈100 nm radius). By limiting the excitation area, the authors achieve a photon flux sufficient for detecting single fluorophores while keeping phototoxicity low. An EM‑CCD camera records images at 200 frames s⁻¹ (5 ms exposure), providing the temporal resolution needed to capture rapid diffusion and short‑lived binding events.

For biological labeling, the glucose‑responsive repressor Mig1 and the activator Gal4 are fused at their C‑termini to photoconvertible fluorescent proteins (mEos3.2 or mNeonGreen). These tags retain the native function of the transcription factors and enable stochastic photo‑activation, which isolates individual molecules from the dense intracellular background. Yeast cultures are grown under two conditions: high glucose (repressive) and low glucose (inductive). Cells are immobilised in a microfluidic chamber without fixation, preserving physiological dynamics during imaging.

Data processing follows a rigorously defined pipeline. Raw frames are first corrected for background and uneven illumination. Single‑molecule spots are identified by local‑maximum detection and fitted with a 2‑D Gaussian to extract sub‑pixel coordinates and photon counts. Trajectories are assembled across frames using a nearest‑neighbour linking algorithm (u‑track). For each trajectory, mean‑square‑displacement (MSD) analysis yields diffusion coefficients (D). Crucially, the authors apply a hidden‑Markov model (HMM) to classify trajectory segments as “mobile” (free diffusion) or “immobile” (DNA‑bound) and to estimate transition rates between these states, providing quantitative on‑ and off‑rates for protein‑DNA interactions in vivo.

Stoichiometry is inferred from the calibrated fluorescence intensity of each spot. By constructing intensity histograms and fitting them with multiple Gaussian components, the authors resolve complexes containing one, two, three, or more fluorescently labeled proteins. This allows them to determine whether Mig1 binds DNA as a monomer, dimer, or higher‑order oligomer under different glucose conditions.

The biological findings are striking. In high‑glucose media, Mig1 displays rapid nuclear diffusion with brief, infrequent binding events (average dwell time < 0.2 s) and predominantly monomeric stoichiometry. Under glucose limitation, Mig1’s dwell time increases to > 1 s, and the intensity analysis reveals a shift toward dimeric and trimeric complexes, indicating cooperative binding that reinforces transcriptional repression. Conversely, Gal4 shows increased binding frequency and longer residence times in low‑glucose conditions, consistent with activation of GAL genes. The measured on‑ and off‑rates match, within experimental error, kinetic parameters derived from bulk biochemical assays, but provide the added advantage of single‑cell resolution and the ability to correlate binding dynamics with cellular metabolic state.

The authors also discuss technical limitations. High illumination intensities can generate reactive oxygen species, potentially altering cellular physiology; they mitigate this by using pulsed excitation and by limiting the total exposure per cell. Fluorophore blinking introduces ambiguity in spot counting, which is addressed through Bayesian inference of the underlying photon emission rates. Finally, the 2‑D imaging approach cannot fully resolve out‑of‑plane motion; the authors suggest incorporating engineered point‑spread functions or light‑sheet illumination to achieve true 3‑D tracking in future studies.

In summary, this work establishes narrowfield microscopy as a viable, high‑throughput tool for dissecting the stochastic, single‑molecule events that underlie eukaryotic transcription regulation. The detailed methodological guide—including hardware configuration, labeling strategy, image analysis, and statistical modeling—makes the approach readily transferable to other yeast pathways, higher eukaryotes, or even mammalian cell lines. By quantifying diffusion, binding kinetics, and complex stoichiometry in living cells, the study bridges the gap between population‑level biochemical assays and the single‑molecule reality of gene regulation, opening new avenues for investigating how metabolic cues are translated into precise transcriptional outcomes.