CardioShift: Resolving cell-to-cell functional heterogeneity in hiPSC-derived cardiomyocytes from high-throughput single-cell imaging
CardioShift is an open-source bioimage informatics framework that resolves hidden cell-to-cell functional heterogeneity in hiPSC-derived cardiomyocytes by properly accounting for nested variance structures in high-throughput single-cell imaging, thereby revealing condition-dependent phenotypes and distinct cellular states that traditional well-average analyses obscure.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
The human heart is not a machine built from identical parts. Even within a single chamber, the muscle cells that pump blood are not carbon copies of one another. Some are larger, some are smaller; some beat with a quick, sharp rhythm, while others move with a slower, more deliberate pace. This natural variation is essential for the heart to function as a whole, allowing it to adapt to the body's changing needs. When scientists try to study heart disease or test new drugs, they often use cells grown in a lab from human stem cells. These cells, which can be coaxed into becoming heart muscle, are a powerful tool because they carry a person's unique genetic code. However, when researchers look at these lab-grown cells, they face a tricky problem. Traditional methods measure the average behavior of thousands of cells mixed together in a single container. This approach is like listening to a crowded room and reporting only the average volume of the noise; it tells you how loud the room is, but it hides the fact that some people are whispering while others are shouting. If a treatment only affects a small group of cells, or if it changes the balance between different types of cells, the average might look completely normal, masking the true effect.
To solve this, a team of researchers developed a new way to look at these heart cells, one that respects their individual differences without losing the big picture. They created a system called CardioShift, which allows them to track the behavior of tens of thousands of individual cells while still treating the group of cells in a single container as the true unit of measurement. This distinction is vital. If a scientist treats every single cell as an independent data point, they risk fooling themselves into thinking they have found a strong effect when they have only seen random noise. By using a statistical approach that acknowledges that cells from the same container are related, the team could separate the natural differences between individual cells from the real effects of a treatment. They tested this method by exposing heart cells to different types of fats, a common stressor for the heart, to see how the cells reacted.
The researchers worked with two specific lines of heart cells. One line came from a patient with a genetic heart condition called hypertrophic cardiomyopathy, caused by a specific change in a gene that controls how the heart muscle contracts. The other line was a genetically corrected version of the same cells, serving as a healthy control. They grew these cells in a 96-well plate, a standard lab tool that holds small amounts of liquid in separate compartments. After letting the cells mature for twenty days, they subjected them to a week of treatment with different fatty acids. Some cells were exposed to palmitate, a fat known to be toxic to heart cells at high levels. Others received oleate, a fat that the body can store safely. A third group received a mix of both, while a control group received no extra fat. The goal was to see if the toxic fat would hurt the cells, if the safe fat would help them, and if the genetic disease made the cells more or less sensitive to these changes.
Using a high-speed camera system, the team recorded the heartbeat and calcium signals of the cells. Calcium is a chemical that acts as a trigger for heart muscle contraction; when calcium levels rise inside a cell, the muscle squeezes. They also used a special dye that glows when it touches fat droplets inside the cells, allowing them to see exactly which cells had stored fat. In total, they analyzed the behavior of over 32,000 calcium signals and more than 10,000 contraction patterns. The results showed that looking at the average of the whole group was not enough. For many measurements, the biggest source of variation was not the treatment or the genetic line, but the natural difference from one cell to the next within the same container. The average value often hid the fact that the cells were splitting into distinct groups with different behaviors.
When the researchers looked at the cells individually, they found that the toxic fat, palmitate, caused significant damage. It made the cells release a protein called troponin-I, which is a sign of heart injury, and it changed how the cells handled calcium. The safe fat, oleate, did not cause this damage. Instead, it caused a large number of cells to store fat droplets. Surprisingly, the cells that stored the most fat did not stop beating or become weak. They maintained a regular rhythm, even though they were carrying a heavy load of lipids. This specific group of fat-storing cells was a distinct subpopulation that would have been invisible if the researchers had only looked at the average of the whole well. In the average, the fat-storing cells were mixed with the non-storing cells, and their unique characteristics were smoothed out.
The team also discovered that the genetic disease changed how the cells responded to the fats. The cells with the genetic mutation were generally more active, contracting with greater force than the healthy cells. When exposed to the mixture of fats, the mutated cells shifted their behavior more dramatically than the healthy ones, moving toward a faster, more energetic state. This suggests that the genetic background of a patient could influence how their heart cells react to metabolic stress. The study confirmed that the genetic mutation made the cells hyperactive, a finding that aligns with previous observations of this specific disease variant.
Perhaps the most important finding was the discovery of a hidden group of cells that accumulated a large amount of fat but kept beating normally. In the healthy control cells, about half of the cells under the oleate treatment belonged to this fat-storing group. In the diseased cells, about a third did. This group was so distinct that it formed its own cluster when the researchers used a computer to group cells with similar traits. Without the ability to see individual cells, this group would have been lost in the noise of the average. The researchers found that the fat-storing cells had a specific profile: they were full of lipids but had a moderate level of contraction, sitting between the very strong, slow-beating cells and the weak, fast-beating cells. This level of detail is crucial for understanding how heart cells survive stress. It shows that the heart does not just fail or succeed as a whole; it contains a mosaic of cells responding in different ways.
The study also highlighted a technical challenge in these experiments. When the researchers tried to measure both the fat content and the contraction speed at the same time, many cells stopped beating properly. This was likely because the combination of dyes and the stress of the experiment made the cells tired. However, the statistical method they developed handled this problem well. Even though many cells dropped out of the analysis, the method remained stable and did not produce false results. This robustness is important because real-world experiments often have missing data or uneven numbers of cells.
The researchers concluded that their new approach, CardioShift, provides a reliable way to map the hidden diversity within heart cell cultures. It allows scientists to see not just the average behavior, but the full range of responses, including rare groups of cells that might be critical for understanding disease or drug effects. By treating the container of cells as the unit of replication while still measuring every individual cell, they avoided a common statistical trap that can lead to false conclusions. This method revealed that the heart cells are far more diverse than previously thought, with different subgroups reacting differently to the same stress. The ability to detect these subtle shifts in cell populations could help researchers develop better models for testing new heart medicines, ensuring that a drug does not just work on the average cell but also protects the most vulnerable ones. The work stands as a proof that looking closer at the individual parts of a system can reveal truths that the whole system hides.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.