Application of fluorescence lifetime imaging microscopy of DNA binding dyes to assess radiation-induced chromatin compaction changes

Abstract

In recent years several approaches have been developed to address the chromatin status and its changes in eukaryotic cells under different conditions-but only few are applicable in living cells. Fluorescence lifetime imaging microscopy (FLIM) is a functional tool that can be used for the inspection of the molecular environment of fluorophores in living cells. Here, we present the use of single organic minor groove DNA binder dyes in FLIM for measuring chromatin changes following modulation of chromatin structure in living cells. Treatment with histone deacetylase inhibitors led to an increased fluorescence lifetime indicating global chromatin decompaction, whereas hyperosmolarity decreased the lifetime of the used dyes, thus reflecting the expected compaction. In addition, we demonstrate that time domain FLIM data based on single photon counting should be optimized using pile-up and counting loss correction, which affect the readout even at moderate average detector count rates in inhomogeneous samples. Using these corrections and utilizing Hoechst 34580 as chromatin compaction probe, we measured a pan nuclear increase in the lifetime following irradiation with X-rays in living NIH/3T3 cells thus providing a method to measure radiation-induced chromatin decompaction.

Keywords: FLIM microcopy; Hoechst 34580; Syto 13; chromatin compaction; histone deacetylation inhibitor (HDACi); irradiation; pile-up.

Figure 1

Modulation of the chromatin density of living NIH/3T3 cells stained with Hoechst 34580 and evaluation by fluorescence lifetime imaging microscopy (FLIM): (a) Intensity (left) and color coded lifetime images (scale ranging from 1225 to 1420 ps) of controls (upper row), upon treatment with histone deacetylase inhibitor, VPA, for 24 h (middle), or 4-fold Phosphate Buffer Saline (PBS, bottom). For 4-fold PBS chromatin compaction becomes obvious also in the intensity image showing the formation of condensed structures. Lookup table (LUT) at left side indicates corrected photon counts (0–1250). The fluorescence lifetime is shown in a continuous pseudo-color scale (right) ranging from 1225 to 1420 ps (b) Normalized frequencies of lifetime distributions from the nuclei of (a). Scale bar, 5 µm. (c) Quantification of relative Hoechst 34580 lifetime changes observed after histone deacetylase inhibitor (VPA) or hyperosmolarity (4-fold PBS). n = 20 for each condition. The lifetime values were normalized to control values. Asterisk (*) shows p < 0.05 (using Student’s t-test) compared to control. Error bars indicate mean ± SD.

Figure 2

Influence of the pixel-wise correction of detector dead-time and pile-up. Confocal FLIM images of NIH/3T3 cell nucleus stained with Hoechst 34580 recorded at an average count rate of 0.9 MHz an 80 MHz laser repetition rate. (a) (a-1) uncorrected intensity image and (a-2) uncorrected lifetime image. (a-3) Intensity image with counting loss correction showing improved contrast and (a-4) lifetime image with the pile-up correction leading to increased values especially at high intensity areas. LUT at left side indicates photon counts for uncorrected (a-1) and corrected (a-3) intensities on same scale (0–900). The fluorescence lifetime is shown in a continuous pseudo-color scale (right) ranging from 1180 to 1520 ps. Scale bar, 5 µm. (b) Quantification of pile-up correction for different laser settings (L6: 2.7 or L8: 10 µW). Lifetime values were normalized to the corrected values of low laser intensities (2.7 µW). Asterisk represents (*) p < 0.05 (using Student’s t-test) compared to the uncorrected values of higher laser intensities. Error bars indicate mean ± SD; n = 15.

Figure 3

Increased Hoechst 34580 lifetime indicating a global chromatin decompaction after irradiation with X-rays. (a) Intensity and lifetime images of the same living NIH/3T3 nucleus before and after 10 Gy X-rays irradiation. Lookup table (LUT) at left side indicates corrected photon counts (0–1000). The fluorescence lifetime is decoded on a continuous pseudo-color scale ranging from 1170 to 1520 ps (right). (b) Normalized frequencies of lifetime values from the nuclei of (a). Quantification of the lifetime indicated a global shift in the lifetime distribution of the Hoechst 34580 after irradiation (c). Mean relative increase of fluorescence lifetime of Hoechst 34580 in NIH/3T3 nuclei upon irradiation with 10 Gy X-rays. Asterisk shows (*) p < 0.05 (using Student’s t-test) compared to pre-irradiated nuclei. Scale bar, 5 µm. Error bars indicate mean ± SD; n = 20.