Optimization of CLARITY for Clearing Whole-Brain and Other Intact Organs

Active clearing of brains with ETC is more efficient than passive clearing. A, Passive clearing is temperature dependent with very little clearing at 21°C occurring after 1 month in SDS. At 37 and 55°C the brain becomes progressively clearer after 1 month in SDS but does not get as clear as with active (ETC) clearing. B, Relative to freshly fixed and uncleared tissue, protein content is well preserved following both active and passive clearing methods, regardless of temperature used. Data are shown relative to brains that were perfused and polymerized with the same hydrogel solution but were not cleared. Olfactory bulb tissue from each clearing condition was homogenized and a BCA assay was performed to estimate protein concentration. There were no significant differences in protein concentration as an effect of active versus passive clearing (p = 0.4) or clearing temperature (p = 0.24).

Clearing conditions for optimal transparency. A, Comparison of ETC temperature and hydrogel composition on tissue transparency and expansion. There were significant main effects of clearing temperature (F_(2,24) = 178.9, p < 0.00001) and hydrogel concentration (F_(3,24) = 16.2, p < 0.00001) on transparency Tissue appears clearest with 37/55°C ETC and 3% acrylamide, 3% formaldehyde, and 0.025% hydrogel. Tissue expansion is less in the 37/55°C condition compared with either 37°C or 55°C alone. B, Both 55 and 37/55°C produce tissue that is more transparent than 37°C, regardless of hydrogel composition (ps = 0.0002 and 0.0009, respectively). Transmission was measured on a light table under standardized illumination conditions. A reference measurement was made without a sample and was set to 100%. The sample was then imaged on the light table and the percentage light intensity through the specimen was recorded as a percentage of the reference value. C, Brains cleared with the combined 37/55°C clearing protocol expand more in lower concentration hydrogel (3% formaldehyde:3% acrylamide:0.025% bis-acrylamide compared with higher concentration hydrogel (4% formaldehyde:4% acrylamide:0.05% bis-acrylamide). The conditions for producing an optimal combination of stable and clear tissue is to polymerize the tissue with hydrogel composed of 4% acrylamide, 4% formaldehyde, and 0.05% bis-acrylamide and then clear the tissue with 5 d of ETC (4 d at 37°C and 1 d at 55°C).

The effect of ETC duration and temperature on imaging depth. A 2mm thick coronal section before (A) and after (B) clearing with out optimized protocol. C, Coronal mouse brain section (2 mm thick) cleared with ETC for 1 d, 3 d, or 5 d at 37°C or for 4 d at 37°C plus an additional day at 55°C. Sections were subsequently stained with propidium iodide and imaged using a confocal microscope under identical excitation conditions without the use of Z-attenuation correction. There is an increase in depth at which a fluorescent signal may be imaged with respect to clearing protocol. The 37/55°C clearing temperature provides deepest imaging penetration.

Imaging depth optimization. Using ArcTRAP-tdTomato mice we visualized the endogenous fluorescence signal in mice brains that underwent 5 d of ETC at 37°C (A) or were cleared for 4 d at 37°C followed by 1 d at 55°C (B). Improved imaging depth is observed using the combined 37/55°C protocol, suggesting that the improvement relates to light rather than dye penetration.

Examining effects of different imaging media (FocusClear vs glycerol). Refractive matching of clarified tissue with either FocusClear or glycerol produces equally transparent tissue, and high-quality images can be collected using either medium. C, D, We imaged the tdTomato signal in the same sections shown in A and B. Excellent image clarity is produced whether the tissue is placed in FocusClear or 80% glycerol.

Examining effects of different microscopes to image clear brains. Although a light sheet or confocal microscope may be ideal, when sufficiently clear, tissue may be imaged using even a very basic epifluorescent microscope. In this case, a Nikon Eclipse 80i with a mercury arc lamp was used to image a 2-mm-thick cleared section. A–F, Single image planes at 100 µm intervals through the z-stack. Bright and sharp cells can be seen through at least 600 µm of the tissue.

Using CLARITY to analyze neuron morphology. A, GFP-labeled cortical pyramidal neuron visualized in a CLARITY-processed tissue section (HSV-GFP viral vector). B–E, The same neuron shown with the Z-depth limited to consecutive 40-µm-thick hypothetical sections. Quantification of dendritic processes based on any individual 40 µm section will produce an incomplete representation of a complex neuron. F, A 40-µm-thick slice of the same neuron centered on the soma is shown. G, A tracing of the entire neuron is shown. H, The neuron is rotated around the y-axis to show how much of the dendritic arbor would be lost using standard section thickness. The inner frame is 40 µm thick. I–L, Example data automatically collected using Imaris from the apical and basal dendrites of the traced neuron. The complete neuron was compared with a single 40-µm-thick portion of the neuron centered on the cell body.

Image resolution in cleared tissue. A, The USAF test target pattern imaged with a 10× air objective on a Nikon light microscope. The lines in element 6-6 and 7-6 are ∼4.4 and 2.2 µm thick, respectively. B, The same test pattern is shown with a cleared brain (∼6 mm thick) over the pattern. The 2.2-µm-thick lines can still be resolved through the tissue. C, Analysis of dendritic spines is possible in cleared tissue using a confocal microscope with 25× oil objective. D, Axonal filopodia in the mossy fiber pathway can be observed in thick tissue sections.

Using CLARITY to clear and image other organs. A, Mouse spleen before and after CLARITY and low- and high-magnification images of CD34 immunohistochemistry on cleared tissue. B, Cleared mouse intestine stained with propidium iodide. C, Cleared mouse kidney labeled for smooth muscle actin. D, Mouse testis stained with propidium iodide. E, Mouse lung immunolabeled for collagen. F, Section of mouse muscle stained with propidium iodide. All fluorescent images were collected using a LaVision Biotec light sheet microscope except for the high-magnification image of the spleen, which was captured using a Zeiss confocal microscope.