Imagine trying to read the small print on a sign while someone shakes it in front of your face. That, in a nutshell, is the challenge of measuring the electrical activity of living heart cells. Every heartbeat begins with an electrical spike, the action potential, which triggers a rush of calcium inside the cell, which in turn makes the cell contract. Scientists call this chain of events excitation–contraction coupling, and understanding it is key to understanding heart disease, still the world’s leading cause of death.
Today, much of this research is done on heart muscle cells grown from human induced pluripotent stem cells (hiPSC-CMs). These lab-grown cells beat spontaneously in a dish, can be made from individual patients, and are increasingly used to test how new drugs affect the heart. But to learn anything from them, we need to watch their voltage and calcium while they do what heart cells do best: beat.
The problem with a moving target
The most common way to image cardiac electrical activity is to stain the cells with fluorescent dyes that get brighter or dimmer as the voltage or calcium level changes. It works well, until the cells move. When a cell contracts, it shifts and squashes under the microscope, and the brightness recorded in each pixel changes for reasons that have nothing to do with electricity. These motion artefacts distort the shape and timing of the very signals we want to measure.
However, there is a type of measurement that ignores brightness altogether: fluorescence lifetime imaging, which measures how long a fluorescent molecule stays “lit up” after a pulse of laser light. The problem is speed. A heart cell’s action potential rises in just 5 to 10 milliseconds, so capturing it requires exceedingly fast imaging speeds. The issue: commercial lifetime imaging systems top out at about 30 frames per second on small images, and drop to roughly one frame per second for large images.
How Does It Work?
In our new paper, we introduce a technique we call fluorescence-lifetime optical electrophysiology, or FLOE. The key feature we rely on here is that a fluorescence lifetime is an intrinsic property of a dye molecule and its immediate surroundings. It doesn’t depend on how many molecules are in a pixel or how brightly they are illuminated. So when a contracting cell slides a little to one side, the brightness in a pixel may jump, but the lifetime barely changes.
To do this in a way that overcomes the speed limitations of traditional lifetime systems, we used a quarter-megapixel single-photon avalanche diode (SPAD) camera developed by our colleagues at EPFL. This camera allows us to detect individual photons at each pixel and “gated” these detections. For every frame, the camera records two images at once: the total fluorescence, and the fluorescence arriving in a short time window after each pulse. The ratio between the two tells us how quickly the fluorescence is fading, and therefore its lifetime. Because every single frame carries its own lifetime information, we can record at almost 200 frames per second without averaging over many heartbeats, and we get a conventional intensity image for free.
What did we find?
To put FLOE to the test, we recorded action potentials from beating monolayers of heart cells. With conventional intensity imaging, the shape of the action potential was severely distorted by the motion, a sign good motivation to use a different modality. With lifetime imaging, the action potentials from beating and non-beating cells were statistically indistinguishable, showing that FLOE can measure the true electrical signal, regardless of any motion in the sample.
An extra layer of information
Beyond removing the motion artefacts, because lifetime can be calibrated, we can also go beyond relative changes and estimate actual physiological values. This lets us produce spatial maps of estimated membrane potential or calcium levels across entire monolayers, revealing that neighbouring groups of cells can behave quite differently within a single dish.
Why does it matter?
Watching heart cells behave naturally, without drugs to hold them still, opens up some exciting possibilities:
-
Safer drug testing:
New medicines are routinely screened for unwanted effects on the heart. Measuring cells in their natural, beating state removes a confounding drug from the experiment and gives a truer picture of how a compound acts.
-
Understanding heart disease:
Patient-derived heart cells can carry the genetic signature of inherited conditions. Single-beat, spatially resolved measurements could help reveal the regional differences and beat-to-beat irregularities that underlie arrhythmias.
-
Beyond the heart:
The same approach could be applied wherever fast electrical or chemical signals matter, from mapping voltage in networks of neurons to monitoring forces in cells with tension-sensitive probes.
This work is a collaboration between our group in the School of Physics and Astronomy at the University of Glasgow (Euan Millar, Vytautas Zickus, Giedrė Astrauskaitė, Caroline Müllenbroich and Daniele Faccio), the School of Cardiovascular and Metabolic Health at the University of Glasgow (Eline Huethorst and Godfrey Smith), the Center for Physical Sciences and Technology in Vilnius, Lithuania, and the Advanced Quantum Architecture Laboratory at EPFL, Switzerland (Jiuxuan Zhao, Gregor Taylor, Claudio Bruschini and Edoardo Charbon).
By letting the heart keep beating while we watch, we hope FLOE will give researchers a clearer, more honest view of how our most important muscle really works.
The full paper is open access and can be read here: Fluorescence-lifetime optical electrophysiology in contracting cardiomyocytes, PNAS 123 (23), e2533881123 (2026).