Reduction of photo bleaching and long term archiving of chemically cleared GFP-expressing mouse brains

Abstract

Tissue clearing allows microscopy of large specimens as whole mouse brains or embryos. However, lipophilic tissue clearing agents as dibenzyl ether limit storage time of GFP-expressing samples to several days and do not prevent them from photobleaching during microscopy. To preserve GFP fluorescence, we developed a transparent solid resin formulation, which maintains the specimens' transparency and provides a constant signal to noise ratio even after hours of continuous laser irradiation. If required, high-power illumination or long exposure times can be applied with virtually no loss in signal quality and samples can be archived for years.

Figure 1. Fabrication of molds for resin embedding from Silastic E-RTV silicone rubber.

(A) Casting frame made from acrylic glass. (B) Silicon rubber mold. (C) Cured resin block with embedded cleared mouse brain hemisphere.

Figure 2. Quantification of fluorescence preservation during repetitive long-term illumination.

Experiments were performed on three control and three resin-embedded samples. Each sample was exposed for 120 min on 3 consecutive days. Images were recorded any 6 min. On these images, mean pixel intensities were calculated in ROIs either covering areas exclusively containing fluorescence signal or background fluorescence. Size and location of the ROIs, were defined using the first image of each illumination interval. Length of scale bar: 300 µm. (B1–B6) Plots of the relative fluorescence intensity over time. Change of signal and background fluorescence intensities during three successive light exposures (day1 – day3) of 120 min duration each. Photobleaching is much lower in the resin-embedded sample than in the control and limited to the initial illumination phase. During light exposure signal to background ratios (SBRs  =  mean signal intensity/mean background intensity) remain constant for the resin embedded samples (r₁, r₂, r₃), while they exponentially decrease in the controls (c₁, c₂, c₃). The error bars indicate the standard error of the mean (SEM).

Figure 3. GFP signal preservation in resin-embedded mouse brain hemispheres.

UM-images of a selected plane within the hippocampal area in both, a control (A) and a resin-embedded brain hemisphere (B), each recorded at onset (1, 3, 5) and after 120 min of constant high power illumination (2, 4, 6). (A) The control hemisphere shows a pronounced quenching of GFP-fluorescence during the illumination intervals. (B) In the resin-embedded hemisphere photobleaching is markedly reduced. Although a certain amount of photobleaching occurs during the first illumination interval (B2 vs. B1), the fluorescence intensity remains approximately stable during further light exposures. Note the recovery of fluorescence between illuminations. Length of scale bars: 150 µm.

Figure 4. Fluorescence preservation by resin embedding during long-term storage.

No loss of GFP signal quality in a resin-embedded mouse hippocampus can be observed within two months. In contrast, a hippocampus stored in DBE loses its fluorescence within less than two weeks. Length of scale bar: 150 µm.

Figure 5. The refractive index of the cured resin depends on the relative amount of D.E.R.736 in the resin compound mixture (D.E.R.736 + D.E.R.332).

By varying the percentage of D.E.R.736 the refractive index can be adjusted to different clearing reagents as DBE or BABB. The curing agent IPDA should be added 1:5 (vol/vol) to the premixed resin compound mixture.