Solid samples as Fluorescent Lifetime Imaging (FLIM) standards

Session

Poster Session

Authors

Mariano Gonzalez Pisfil (1), Sonja Rottmeier (1), Brigitte Bergner (1), Andreas W. Thomae (1), Steffen Dietzel (1)

Affiliations

1. Core Facility Bioimaging and Cardiovascular Physiology, Biomedical Center, Ludwig-Maximilians-Universität München, Großhaderner Str. 9, 82152 Planegg-Martinsried

Keywords

Fluorescence Lifetime Imaging Microscopy, FLIM

Abstract text

FLIM microscopy as an add-on to confocal microscopes used to be tedious and slow, with a separate computer and software user interface. Therefore the application was limited to specialized labs and specialized research questions. In recent years, however, several companies implemented technical developments in time correlated single photon counting (TCSPC) that allow much faster recordings of microscopic lifetime images with speeds comparable to normal confocal imaging, ideally with software that is fully integrated in the confocal control software. One major supplier of confocal microscopes now builds at least some lifetime hardware and analysis tools into all shipped systems. These developments make it likely that FLIM will reach a much broader distribution in life science research in the near future, for example to use several fluorochromes that can be distinguished by lifetime in the same color channel.

Checking for correct lifetime measurements on a given FLIM system is typically done with a solution of a fluorescent dye with well-known lifetime. These solutions have a limited shelf life and thus have to be made repeatedly. They also must be prepared carefully with regard to dye concentration and pH. With a more widespread usage of FLIM in life science research it can be expected that not all groups who have access to FLIM will want to prepare a fresh dye solution for calibration measurements. Therefore solid fluorescent samples with defined fluorescence lifetimes would be helpful. Ideally with a simple, mono-exponential lifetime decay.

With this goal in mind, we tested three kinds of samples: fluorescent plastic slides from Chroma, Argolight calibration slides, and the SF series from Starna. The latter consist of a polymer doped with organic fluorescent compounds in a microscope slide format.

In particular the Starna slides hold great promise for the outlined task and for comparing the lifetimes of different systems. With mono-exponential decay and a lifetime close to 4 ns for the green fluorescing SFG slide and 5.4 for the red fluorescing SFR slide their lifetime is in a range that every FLIM system should be capable to resolve. We performed series of measurements with detection in varying wavelength ranges and with various excitation wavelengths and found the measured lifetimes to be within a span of 0.1 ns under those conditions. For comparison, we studied the lifetime of ATTO488 solutions which are often used for calibration. Here we found a lifetime change of about 0.05 ns when changing the pH from pH 8 to pH 3 and thus in a comparable range. We conclude that with a defined excitation wavelength and a reproducible detection window the tested Starna SF slides provide a good calibration sample for FLIM systems. Even with slightly different detection windows, for example with filters from different producers for the same dye, the lifetimes between different systems will be close enough to be comparable.

Chroma slides showed a span of up to a quarter nanosecond in the lifetimes of various detection windows along the wavelength range and also multi-exponential decay. In addition the manufacturer warns explicitly that properties may vary from one batch to another. Still, thanks to their excellent fluorescent yield any specific slide may be used to track performance of a given system over time or to compare two systems in the same lab when always the same excitation wavelength and detection windows are used.

The Argolight slide contains fluorescent patterns and is not fluorescent throughout. It thus was difficult to obtain a sufficient amount of photons for exact lifetime determinations. Clearly, the lifetime was multi-exponential. .