Researchers Investigate Unwanted Color Changes in Fluorescence Microscopy – and Find Ways to Minimize Measurement Errors

A green marker suddenly glows bluish, a signal fades – and the results of an experiment are called into question. Researchers from Leibniz IPHT and Ljubljana have examined why fluorescent dyes change color under the microscope and how measurement errors can be avoided. Their findings help improve the accuracy of microscopic analyses in biology and medicine.

Laser light is a central tool in fluorescence microscopy. By selectively exciting fluorescent dyes, it makes biological structures visible – from individual proteins to complex cellular networks. However, this very illumination can chemically alter the dyes, affecting image accuracy. Intense light exposure can change their structure, causing their signal to fade or shift in the color spectrum.

These light-induced changes are known as photobleaching (fading) and photoconversion (light-induced chemical transformation). Particularly problematic is photoblueing – a color shift toward shorter wavelengths.

“When that happens, it can cause a signal to appear in a different channel or disappear completely – distorting the results,” explains Prof. Dr. Christian Eggeling. The expert in super-resolution microscopy leads the Biophysical Imaging research department at Leibniz IPHT. In 2024, he received an offer from a research institute in Oxford, UK, which Jena University successfully countered.

How to Avoid Color Shifts

A team led by Christian Eggeling from Leibniz IPHT and the University of Ljubljana at the Jožef Stefan Institute has now systematically examined these processes. The researchers tested various fluorescent markers and analyzed under which conditions color shifts occur. Their conclusion: While conventional microscopy methods such as confocal or STED microscopy are relatively robust against these effects, color changes can cause significant measurement errors in certain applications – particularly in techniques that analyze the chemical environment of molecules.

A key finding of their Nature Methods study: Some dyes are particularly susceptible to photoblueing – while others remain much more stable. “The choice of the right marker can make all the difference,” emphasizes Agnes Koerfer, who, together with Dr. Anindita Dasgupta, played a key role in the study. “We were able to show that certain fluorescent dyes minimize the effects and thus lead to more reliable measurements.”

The researchers also found that cleverly adjusting illumination conditions can help. Reducing light intensity or using dyes with an exchangeable labeling mechanism can decrease the risk of color shifts.

Reliable Colors for Cell Biology and Infection Research

The new insights improve the precision of microscopic analyses and have potential implications across many research fields. They not only help optimize existing microscopy techniques but also support the analysis of biological processes – such as in cell biology or infection research.

“For example, when we study how virus particles move within a cell, we need to ensure that our labeling systems are stable,” explains Eggeling. “Our study shows that by carefully selecting dyes and illumination conditions, we can obtain more accurate results.”

This is especially relevant for medical diagnostics and the development of new therapies. “Many biological processes can only be understood through long-term observation of cells,” adds Koerfer. “If signals are distorted by color shifts during this process, it can significantly complicate the interpretation of experiments.”

Toward More Precise Methods

The researchers hope that their results will help ­improve standard methods in fluorescence microscopy. Their insights could support the development of new dyes that are even more stable – and thereby enable even more accurate measurements.

“Microscopy is constantly evolving,” says Eggeling. “Our job is to make sure we’re working with the best methods – so we can capture the molecular world as precisely as possible.”

Original publication: https://doi.org/10.1038/s41592-024-02297-4