EZ Clear effectively clears adult mouse organs in 48 hr with three simple steps at room temperature (Figure 1A). Step 1 is to immerse the fixed sample in the lipid removal solution, which consists of 50% (v/v) THF in sterile Milli-Q water. This formula allows THF to dissolve lipid away from the tissue while the organ remains hydrated in an aqueous environment. Following lipid removal, the sample is incubated in sterile Milli-Q water for 4 hr to wash and remove any residual THF from the tissue. The tissue is then rendered transparent by immersing it in aqueous RI matching and imaging solution (EZ View, RI = 1.518) at room temperature for 24 hr. EZ Clear not only renders the whole brain as transparent as 3DISCO and FAST 3D (Figure 1B, Figure 1—figure supplement 1A), but it also has the shortest processing time (48 hr) and simplest procedure (three steps) of current tissue clearing protocols. Because the sample is maintained in an aqueous environment throughout the entire process, the size of processed samples remains constant throughout the protocol (Figure 1C, n=4 for each condition, one-way ANOVA, size change ratio = 1.072 ± 0.062). Conversely, other approaches either significantly shrank (Fast 3D: 0.776±0.025 and 3DISCO: 0.593±0.013) or expanded (X-CLARITY: 1.608±0.049) the brain (Figure 1C and Figure 1—figure supplement 1B). In addition, endogenous transgenic fluorescent reporter activity is preserved throughout the EZ Clear protocol, as a brain from a Thy1-EGFP neuronal reporter line that was processed with EZ Clear retained robust signal within neurons when imaged by lightsheet microscopy across a total imaging depth of 5 mm (Figure 1D–F and Video 1). EZ Clear is also compatible with adult mouse organs other than the brain, with excellent transparency evident following tissue processing (Figure 1G). Multiple adult mouse organs (eye, heart, kidney, testis, and ovary) were collected after vascular perfusion with a far-red fluorescently conjugated Lycopersicon esculentum (tomato) lectin (lectin-DyLight 649) dye that labels the endothelium. Following perfusion, samples were fixed and then processed with EZ Clear. Wholemount imaging of samples by LSFM demonstrated that the lectin-649 signal was robust and well preserved (Figure 1H, maximum intensity projection). These results show that with three simple steps in 48 hr, EZ Clear effectively renders adult mouse organs optically transparent while simultaneously preserving signal from synthetic and endogenous fluorescent reporters.

Figure 1 with 1 supplement see all

Download asset Open asset

Figure 1—source data 1

Quantitative comparison of the size changes of the mouse brains before and after processing with 3DISCO, Fast 3D, X-CLARITY, and EZ Clear.

https://cdn.elifesciences.org/articles/77419/elife-77419-fig1-data1-v2.xlsx

Download elife-77419-fig1-data1-v2.xlsx

Video 1

Download asset

To maintain cleared samples in an aqueous environment after lipid removal and washing, we compared the ability of various sample mounting and imaging buffers to render tissue optically transparent. Given the aqueous nature of EZ Clear, we first tested whether refractive index matching solution (RIMS) (80% (v/v) Nycodenz, RI = 1.46) (McCreedy et al., 2021), or the more affordable alternative sRIMS (80% D-sorbitol, RI = 1.43) (Yang et al., 2014), were compatible with tissues following lipid removal by EZ Clear. Somewhat surprisingly, EZ Clear delipidated samples immersed in either RIMS (Figure 2C) or sRIMS (data not shown) were not rendered transparent. Thus, we pursued novel aqueous solutions with an RI over 1.50. Previous studies showed that urea can hyperhydrate samples and enhance tissue transparency (Hama et al., 2011; Susaki et al., 2014; Zhu et al., 2019). We also noticed that the RIs of aqueous solutions increased linearly as the concentration of urea was increased (Figure 2—source data 1). To test the effect of combining urea and Nycodenz on RI and clearing efficiency under aqueous conditions, we gradually increased the urea concentration in RIMS and measured the RI (Figure 2A). As predicted, the RI of the 80% Nycodenz solution was positively correlated with the concentration of urea. While the solution was saturated with 8 M of urea in 80% Nycodenz, the highest RI (RI = 1.518 ± 0.0003) was obtained with 7 M urea in 80% Nycodenz at room temperature. We then examined sample transparency by comparing brains after lipid removal with EZ Clear, immersed and equilibrated in either PBS (RI = 1.332), RIMS (80% Nycodenz, RI = 1.463 ± 0.0019), or EZ View (80% Nycodenz and 7 M urea, RI = 1.518 ± 0.0003) (Figure 2B–D). While the tissue remained opaque in PBS (Figure 2B), the transparency of the EZ Clear lipid removed brain showed only a minor improvement following immersion and equilibration in RIMS for 24 hr (Figure 2C). However, the EZ Clear lipid removed brain immersed and equilibrated in EZ View solution for 24 hr was comparable to samples cleared using 3DISCO (compare Figure 2D and Figure 1B).

Figure 2

Download asset Open asset

Figure 2—source data 1

Measurements for refractive index changes in response to different concentrations of Nycodenz and urea, quantitative comparison of fluorescence intensity of lectin-649 at different imaging depths between brains equalibrated in EZ View and RIMS, and fluorescence intensity measurement of lectin-649 after storing samples in EZ View for 14, 17, 43, and 70 days.

https://cdn.elifesciences.org/articles/77419/elife-77419-fig2-data1-v2.xlsx

Download elife-77419-fig2-data1-v2.xlsx

Next, we examined how equilibrating samples in different RIMS following EZ Clear lipid removal affected imaging depth. Adult mice perfused with fluorescent lectin-649 were euthanized and the brains were collected and then immersion fixed in 4% PFA overnight at 4°C. After fixation, labeled brains were treated with EZ Clear’s lipid removal and washing steps, and then equilibrated and imaged in either RIMS or EZ View (Figure 2E–J). Imaging by LSFM demonstrated that not only were brains equilibrated in EZ View more transparent than those equilibrated in RIMS (Figure 2C, D), but they also featured deeper imaging depths (over 6 mm) through the dorsal-ventral axis of the brain compared to samples equilibrated in RIMS (Figure 2F, I). Color-coded depth projection stepping at 1 mm intervals showed that while the signal from lectin-649 starts to scatter 1–2 mm deep into RIMS treated tissue (Figure 2G), the signal remains strong and coherent (non-diffuse) at greater depths in the brain equilibrated and imaged in EZ View (Figure 2J). Quantitative comparison of the mean fluorescence intensity at depths of 1, 2, 3, 4, and 5 mm showed that while fluorescence intensity remained consistent in the EZ View equilibrated samples, mean fluorescence intensity was significantly lower after 4 mm in depth in the brains equilibrated and imaged in RIMS compared to those processed and imaged in EZ View (wholemount brains imaged for each condition n=3, two-way ANOVA and multiple comparisons). We also tested the long-term stability of fluorescence signal of samples stored in EZ View. Mouse brains perfused with lectin-649 were cleared with EZ Clear and stored in EZ View and then imaged 14, 17, 43, and 70 days later (Figure 2L). Notably, signal intensity showed no significant change through the imaging depth following prolonged storage in EZ View solution (n=3, two-way ANOVA and multiple comparisons). These results demonstrate that EZ View solution attains a high RI and effectively renders delipidated tissues optically transparent while minimizing light scattering while also maintaining fluorescent signal. Critically, EZ Clear samples remain in an aqueous environment throughout the entire clearing and imaging process, and no corrosive or combustible organic solvents, or potentially damaging immersion oils, are required for RI matching and imaging.

To further assess the clearing efficiency and sample transparency of EZ Clear, we compared samples cleared by two THF-based clearing methods, EZ Clear and Fast 3D, following administration of a synthetic, exogenous fluorescent dye that labels the endothelium. When Evans blue, a non-cell permeable dye, binds to albumin, it undergoes a conformational shift that produces fluorescence in the far-red spectrum (excitation at 620 nm, emission at 680 nm) (Namykin et al., 2019; Saria and Lundberg, 1983; Honeycutt and O’Brien, 2021). A cost affordable alternative to expensive dyes, such as fluorescent tomato lectin, Evans blue can thus be used to robustly label the endothelium following intravenous administration (Robertson et al., 2015). Adult mice perfused with Evans blue were euthanized and the brains were collected and fixed in 4% PFA overnight at 4°C. The brains were then bisected at the midline along the anterior-posterior axis, and the left hemisphere was cleared with Fast 3D, while the right hemisphere was cleared with EZ Clear. Both hemispheres were imaged independently on a Zeiss Lightsheet Z1 from the dorsal to ventral side with an EC Plan-Neofluar 5x/0.16 air detection objective at a resolution of 1.829 μm laterally and 3.675 μm axially (Figure 3A–F and Video 2). The right hemisphere, processed with EZ clear, was equilibrated, and imaged in EZ View solution (RI = 1.518), while left hemisphere, cleared with Fast 3D, was equilibrated, and imaged in Fast 3D imaging solution (RI = 1.512). Although both EZ Clear and Fast 3D effectively rendered the tissue optical transparent (Figure 1B), lightsheet imaging demonstrated that EZ Clear also enabled imaging at depths up to 6.5 mm from dorsal to ventral sides, with focused signal and high contrast (Figure 3A–C), whereas light scattering increased at deeper imaging depths in the hemisphere processed using Fast 3D (Figure 3D–F). We further compared fluorescence intensity and contrast between the brains cleared with EZ Clear and Fast 3D. Mouse brains perfused with lectin-649 were either cleared with EZ Clear or Fast 3D (n=3 brains for each condition) and wholemount imaged through the dorsal-ventral axis by LSFM. A quantitative comparison of the mean fluorescence intensity at depths of 1, 2, 3, 4, or 5 mm into the brain (dorsal to ventral) showed no significant difference in lectin-649 signal intensity between EZ Clear and Fast 3D cleared samples (Figure 3G, n=3, two-way ANOVA and multiple comparisons). However, while the signal contrast remained high and consistent in EZ Clear processed brains, the lectin-649 signal contrast decreased with imaging depth in Fast 3D samples (Figure 3H, n=3). Lectin-649 signal contrast also remained high and consistent when quantitatively measured in other EZ Clear processed adult mouse organs, including the eye, heart, kidney, testis, and ovary (Figure 3I). Thus, in 48 hr and three simple steps, EZ Clear achieves high optical transparency, without sample shrinkage or swelling, it preserves endogenous and synthetic fluorescence, and it is compatible with aqueous-based imaging platforms. Additionally, EZ Clear enables imaging at deeper depths with minimal light scattering compared to other current tissue clearing modalities.

Figure 3

Download asset Open asset

Figure 3—source data 1

Measurements and comparisons of fluorescent intensity and contrast of lectin-649 between EZ Clear and Fast 3D cleared mouse brains and organs.

https://cdn.elifesciences.org/articles/77419/elife-77419-fig3-data1-v2.xlsx

Download elife-77419-fig3-data1-v2.xlsx

Video 2

Download asset

To determine the compatibility of EZ Clear with wholemount immunofluorescent staining, we tested the immunostaining protocol from iDISCO (Renier et al., 2014), as well as a standard immunofluorescent staining procedure, on samples that were delipidated and washed with EZ Clear (Figure 4A). We first stained adult mouse brains with To-Pro 3 to label nuclei and a Cy3-conjugated antibody against αSMA (αSMA-Cy3) to label vascular smooth muscle cell enwrapped blood vessels to determine how efficient the nuclear dye and fluorophore-conjugated antibody can penetrate through the brains after processing with EZ Clear. Both iDISCO and standard immunofluorescent staining protocols enabled effective diffusion and immunolabeling throughout EZ Clear treated samples (Figure 4B–F), as virtual sections from an LSFM imaged wholemount brain showed that To-Pro 3 dye labeled nuclei evenly throughout the entire sample. Similarly, both protocols showed robust antibody penetrance as αSMA-Cy3 staining was evident throughout the wholemount samples (Figure 4C). Quantification of To-Pro 3 signal intensity across the brain sections (in Figure 4B, yellow dashed lines i and ii) also shows the intensity profiles are comparable between iDISCO and standard immunofluorescent protocols, with homogenous staining (Figure 4D). Quantitative comparison of the mean fluorescence intensity of To-Pro 3 (Figure 4E) and αSMA-Cy3 (Figure 4F) at the depths of 1, 2, 3, 4, and 5 mm into the wholemount imaged brains (n=3, dorsal to ventral) also show no significant difference between samples stained with either protocol (Figure 4E and F, n=3, two-way ANOVA and multiple comparisons).

Figure 4

Download asset Open asset

Figure 4—source data 1

Measurements To-Pro 3 penetrance across the lateral axis between iDISCO and standard IHC processed samples.

Quantitative comparison of mean fluorescence intensity of To-Pro 3 and αSMA-Cy3 at different imaging depths (dorsal to ventral) between iDISCO and standard IHC processed samples.

https://cdn.elifesciences.org/articles/77419/elife-77419-fig4-data1-v2.xlsx

Download elife-77419-fig4-data1-v2.xlsx

To further determine whether endogenous fluorescence can be preserved during the wholemount staining procedure, we processed brains from mice containing a Thy1-EGFP pan-neuronal reporter for wholemount immunostaining. After processing the samples with the EZ Clear lipid removal and washing steps, Thy1-EGFP brains were stained with a GFAP antibody to label astrocytes and To-Pro 3 to label nuclei using a standard immunofluorescent staining protocol. Lightsheet imaging showed that not only can the GFAP positive astrocytes be identified throughout the wholemount immunofluorescence-stained brain (Figure 4G and H, Cortex; Figure 4I and J, corpus callosum), robust signals from the Thy1-EGFP positive neurons can be detected throughout the sample. These results demonstrate that EZ Clear processed samples are compatible with wholemount immunofluorescent staining while retaining endogenous fluorescence.

To further determine the downstream uses of samples following EZ Clear treatment, we next tested whether the cleared and wholemount imaged samples could be processed for embedding and cryosectioning. For these studies, we examined brain tumor formation in a native murine model of glioma, as before (Carlson et al., 2021). To induce glioma, the lateral ventricle of E16.5 mouse embryos were injected with a DNA cocktail consisting of three plasmids: (1) a single pX330-variant construct encoding 3xFlag-NLS-Cas9-NLS, along with three human RNA Polymerase III U6 snRNA (RNU6-1) promoter cassettes upstream of guide RNAs targeting the tumor suppressor genes Nf1, Trp53, and Pten, (2) a piggyBac (PB) helper plasmid with the radial glial- and astrocyte-specific promoter, Solute Carrier Family1 Member 3, SLC1A3 (also known as EAAT1, GLAST, Genbank AF448436.1), driving expression of PB transposase, and (3) a PB cargo fluorescent EGFP reporter vector to indelibly label all tumor cells and their descendants (Carlson et al., 2021). Following injection of the DNA cocktail, the embryos were electroporated to allow uptake of the constructs, then the embryos were placed back in the maternal cavity. These in utero electroporated (IUE) animals with tumor suppressor deficient cells were then birthed normally and were collected at postnatal day 104 (P104) for perfusion with lectin-649. Next, the brains were collected, fixed, and processed with EZ Clear and wholemount imaged by LSFM to reveal the distribution of GFP⁺ tumor cells and the lectin-649 labeled vasculature (Figure 5A and B). At a macro level, whole brain imaging showed that GFP⁺ tumor cells can be identified not only as clusters (Figure 5A and B), but also as single cells (Figure 5C and D, Video 3). After imaging, the glioma containing samples were immersed in ×1 PBS overnight to remove the EZ View solution, then sucrose protected and embedded for cryosectioning. P104 glioma brains processed with EZ Clear were then cryosectioned and sections were used for direct immunofluorescence or for hematoxylin and eosin (H&E) staining to visualize tissue architecture. After cryosectioning, tissue sections were mounted and inspected to evaluate whether the fluorescence from the transgenic piggyBac EGFP reporter and synthetic fluorescent lectin dye were preserved. Sections mounted in EZ View solution were highly transparent compared to those which remained in PBS or were mounted in Prolong Glass Antifade medium (Figure 5E–G). Sections mounted in EZ View were compatible with both fluorescence (Figure 5H) and confocal imaging (Figure 5I–J), as signals from EGFP⁺ tumor cells and perfused lectin-649 vessels were well preserved. EZ Clear processed, wholemount imaged, and cryosectioned tissues were also compatible with conventional histology, as evidenced by robust H&E labeling of tissues (Figure 5K–M) comparable to tissues fixed using only 4% PFA (Figure 5—figure supplement 1).

Figure 5 with 1 supplement see all

Download asset Open asset

Video 3

Download asset

Next, we explored whether EZ Clear processed and cryosectioned tissues are compatible with indirect immunofluorescence. Free-floating sections from brains harboring EGFP⁺ glioma cells that were perfused with lectin-649 were processed through the standard immunofluorescent staining protocol (Figure 6A). Sections were stained with antibodies against CD31 (to label endothelial cells), GFAP (astrocytes), smooth muscle α-actin conjugated with Cy3 (αSMA-Cy3) and class III β-tubulin (e.g., Tuj1, to label neurons), and then mounted with EZ View solution. Confocal imaging revealed that in addition to preserving the fluorescence of EGFP⁺ tumor cells and perfused lectin-649 labeled endothelial cells, each of these distinct antigens (GFAP, αSMA, and Tuj1) was robustly detected by indirect immunofluorescence following EZ Clearing and cryosectioning (Figure 6B-D, Video 4). These results demonstrate that EZ Clear processed tissue not only can be interrogated at a macro level by whole organ 3D imaging using LSFM, but can be further used for cryosectioning and investigation at the cellular level via either histological or immunofluorescent staining.

Figure 6

Download asset Open asset

Video 4

Download asset

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All animal research was conducted according to protocols approved by the Institutional Animal Care and Use Committee (IACUC) of Baylor College of Medicine.

All mouse CRISPR-IUE GBM gliomas were generated in the CD-1 IGS mouse background. IUEs were performed as previously described (Carlson et al., 2021). Briefly, a plasmid containing guide RNAs targeting the tumor suppressor genes Nf1, Tp53, and Pten was co-electroporated along with a fluorescent reporter EGFP to label tumor cells. The uterine horns were surgically exposed in a pregnant dam at E16.5 and the embryos were injected with a DNA cocktail containing the following four plasmids: (1) a single pX330-variant (Cong et al., 2013) construct encoding 3xFlag-NLS-Cas9-NLS, along with three human RNA Polymerase III U6 snRNA (RNU6-1) promoter cassettes upstream of validated guide RNA sequences targeting Nf1 (GCAGATGAGCCGCCACATCGA), Trp53 (CCTCGAGCTCCCTCTGAGCC), and Pten (GAGATCGTTAGCAGAAACAAA) (Xue et al., 2014); (2) a piggyBac (PB) helper plasmid with the glial- and astrocyte-specific promoter, Solute Carrier Family1 Member 3, SLC1A3 (also known as EAAT1, GLAST, Genbank AF448436.1), driving expression of PB transposase (pGlast-PBase) (Chen et al., 2015), and a PB cargo fluorescent reporter vector (pbCAG-GFP-T2A-GFP or pbCAG-mRFP1). The PBase helper plasmid promotes stable integration of the cargo fluorescent reporter vector, which indelibly labels all descendant cells, allowing one to visualize tumors over time via fluorescence. Following injection of the glioma-inducing CRISPR-Cas9/PB cocktail (2.0 µg/µL pGLAST-PBase, 1.0 µg/µL all other plasmids) into the lateral ventricle of each embryo, embryos were electroporated six times at 100 ms intervals using BTW Tweezertrodes connected to a pulse generator (BTX 8300) set at 33 V and 55 ms per pulse. Voltage was applied across the entire brain to allow uptake of the constructs. The uterine horns were placed back in the cavity, and these dams developed normally, but their electroporated offspring featured malignancies postnatally, as the tumor suppressor deficient cells expanded. Animals were perfused at P104 with L. esculentum (tomato) lectin fluorescently conjugated with DyLight 649 (Vector Laboratories, Cat. No. DL-1178) before euthanized. A detailed procedure for animal perfusion and lectin labeling is listed below. The brains were then dissected out for further processing.

To perfuse the adult mice, they were deeply anesthetized by CO₂ inhalation and the chest cavity was opened to expose the beating heart. The right atrium was incised, and the mouse was transcardially perfused through the left ventricle with 10 mL of room temperature ×1 PBS, followed by 10 mL of cold 4% PFA. Organs were then dissected from the mouse and then drop fixed in 4% PFA at 4°C for 24 hr with gentle agitation. After fixation, the mouse organs were washed in ×1 PBS three times, 30 min each at room temperature, then stored in ×1 PBS with 0.05% sodium azide at 4°C before proceeding with clearing procedure.

For animals that were labeled with lectin, adult mice were injected in their tail vein with 100 μL of L. esculentum (tomato) lectin fluorescently conjugated with DyLight 649 (lectin-649, Vector Laboratories, Cat. No. DL-1178). Lectin was allowed to circulate for a minimum of 5 min to enable adequate circulation and binding of the lectin to the endothelium. Mice were then deeply anesthetized by CO₂ inhalation and the chest cavity was opened to expose the beating heart. An additional 75 μL of lectin-649 was injected through the left ventricle using a 31-gauge insulin syringe (BD Biosciences, Cat. No. 324911). Lectin was perfused by hand slowly over 1 min and the needle was kept in place for an additional minute after injection to allow the pumping heart to circulate the dye. Afterward, the right atrium was incised, and the animal was transcardially perfused through the left ventricle with an additional 10 mL of room temperature ×1 PBS, followed by 10 mL of cold 4% PFA. Organs were then dissected and drop fixed in 4% PFA at 4°C for 24 hr with gentle agitation. After fixation, all organs were washed in ×1 PBS three times, 30 min each at room temperature, then stored in ×1 PBS with 0.05% sodium azide at 4°C before proceeding with clearing.

For animals that were labeled with Evans blue, adult mice were injected in their tail vein with 100 μL of Evans blue solution (2% (w/v) with 0.9% (w/v) NaCl in sterile Milli-Q water, filter sterilized with a 0.22 µm filter), as in Honeycutt and O’Brien, 2021. The dye was allowed to circulate for a minimum of 5 min. Mice were then deeply anesthetized by CO₂ inhalation and the chest cavity was opened to expose the beating heart. An additional 100 μL of Evans blue solution was injected through the left ventricle using a 31-gauge insulin syringe (BD Biosciences, Cat. No. 324911). The dye was perfused by hand slowly over 1 min and the needle was kept in place for an additional minute after injection to prevent dye leaking back out. The injected Evans blue solution was allowed to circulate until the hindlimbs and tail turned blue, or 5 min was reached. Animals were then euthanized with CO₂ inhalation followed by cervical dislocation. The brain was then dissected from the skull and drop fixed in 4% PFA at 4°C for 24 hr with gentle agitation. After fixation, the brain was washed in ×1 PBS three times, 30 min each at room temperature, then stored in ×1 PBS with 0.05% sodium azide at 4°C before proceeding with clearing procedure.

After perfusion and overnight fixation, samples were placed in individual glass scintillation vials, protected from light with foil, and rocked on an orbital shaker in a vented chemical fume hood, and incubated with 20 mL of the following solutions in sequence for the EZ Clear protocol: (1) Lipid removal: 50% (v/v) THF (with 250 ppm BHT, Millipore-Sigma, 186562) prepared in sterile Milli-Q H₂O for 16 hr. (2) Wash: rinsed with sterile Milli-Q H₂O four times, 1 hr each at room temperature. (3) RI matching: incubated with 5 mL of EZ View sample mounting and imaging solution for 24 hr to render the samples transparent for imaging. The EZ View solution consisted of 80% Nycodenz (Accurate Chemical & Scientific 100334-594), 7 M urea, 0.05% sodium azide prepared in 0.02 M sodium phosphate buffer. To prepare the solution, 52.5 g of urea and 31.25 mg of sodium azide were mixed with 35 mL of 0.02 M sodium phosphate buffer (pH 7.4) in a 250 mL beaker. The solution was stirred and gently heated to 37°C on a hot plate until the urea completely dissolved. Then, 100 g of Nycodenz was slowly stirred in until completely dissolved. The final volume of the solution was adjusted to 125 mL with 0.02 M sodium phosphate buffer. The dissolved solution was filtered through a vacuum filtration system (Nalgene vacuum filtration system filter, pore size 0.2 μm, Z370606) and stored at room temperature. The RI of the EZ View solution was measured on a refractometer (Atago, PAL-RI 3850) and the RI was between 1.512 and 1.518. Cleared samples were protected from light, placed on a shaker, and rocked gently at room temperature for 24 hr to render them transparent. All of the adult mouse organs (brain, eye, heart, lung, liver, kidney, spleen, testis, and ovary) were processed following the same procedure.

3DISCO clearing was performed according to Ertürk et al., 2012. Briefly, after perfusion and fixation, adult mouse brains were incubated in graded series of 20 mL of THF solutions mixed with filtered Milli-Q H₂O at increasing concentrations of 50, 70, 80, and 100% (v/v), 1 hr each, in a scintillation vial at room temperature. The scintillation vial was covered with foil and placed on an orbital shaker within a vented chemical fume hood. Brains were then immersed in fresh 100% THF overnight, followed by a 1 hr incubation in fresh 100% THF. After removing THF completely, samples were then immersed in benzyl ether (DBE, Millipore-Sigma, 108014) overnight to render them transparent.

X-CLARITY clearing was performed according to the manufacturer’s instruction (Logos Biosystems). Briefly, after perfusion and fixation, adult mouse brains were immersed in 5 mL of X-CLARITY Hydrogel Solution with 0.25% (w/v) of polymerization initiator VA-044 (Logos Biosystems, C1310X) at 4°C for 24 hr. The hydrogel infused brain was then thermo-induced crosslinked for 3 hr at 37°C under vacuum (–90 kPa). After the crosslinking reaction, the hydrogel solution was removed and the brain was washed in ×1 PBS three times, 1 hr each, then once overnight at 4°C. The hydrogel infused and crosslinked brain was then cleared in electrophoretic tissue clearing solution (Logos Biosystems, C13001) using the X-CLARITY Tissue Clearing System set at 0.8 A and 37°C for 15 hr (Logos Biosystems). After electrophoresis, the brain was washed in 50 mL of ×1 PBS three times, 1 hr each, then once overnight at room temperature. To render the tissue transparent for imaging, the brains were immersed in sRIMS (70% w/v D-sorbitol in 0.02 M sodium phosphate buffer, pH 7.4) at 4°C until transparent. The RI of the sRIMS was measured on a refractometer (Atago, PAL-RI 3850) and the RI was between 1.42 and 1.43.

Fast 3D clearing was performed according to Kosmidis et al., 2021. Briefly, after perfusion and overnight fixation, samples were placed on an orbital shaker at 4°C, protected from light, and incubated with 20 mL of the following solutions in sequence: (1) 50% (v/v) THF prepared in sterile Milli-Q H₂O with 20 μL of triethylamine (pH 9.0) (Millipore-Sigma, T0886) for 1 hr; (2) 70% THF with 30 μL of triethylamine for 1 hr; and (3) 90% THF with 60 μL of triethylamine overnight. After overnight incubation with 90% THF, samples were then rehydrated with the following solutions: (4) 70% THF with 30 μL of triethylamine for 1 hr followed by (5) 50% THF with 20 μL of triethylamine for 1 hr. Finally, samples were washed with sterile Milli-Q H₂O four times, 10 min each, then washed overnight in sterilized MQ water. To prepare samples for imaging, brains were then incubated with 4 mL of Fast 3D Clear solution. To prepare the Fast 3D Clear solution, 48 g of Histodenz (Millipore-Sigma, D2158), 0.6 g of diatrizoic acid (Millipore-Sigma, D9268), 1.0 g of N-methyl-D-glucamine (Millipore-Sigma, M2004), 10 g of urea, and 0.008 g of sodium azide were mixed in a 100 mL beaker. Twenty mL of sterile Milli-Q H₂O was added to dissolve the powder using a stir bar overnight. The final volume of the solution was approximately 50 mL. The dissolved solution was filtered and stored at room temperature. The RI of the Fast 3D clear solution was measured on a refractometer (Atago, PAL-RI 3850) and the RI was between 1.511 and 1.513. The sample was protected from light, placed on a shaker, and rocked gently at room temperature for 24 hr to render the sample transparent.

A custom-designed sample holder was used for mounting the sample and imaging on a Zeiss Lightsheet Z.1. The sample holder consisted of a 1-inch plastic slotted screw (McMaster-Carr, 94690A724) with a nylon hex nut (McMaster-Carr, 94812A200), a custom-made magnetic nut, and a 1/8” washer (Figure 7A). To mount samples, we first applied a small amount of superglue gel (Loctite Gel Control) to the surface of the screw head, and gently pressed the screw head to the brain stem in order to attach it (Figure 7B). Then, the luer-lock tip of a 5 mL syringe (BD Bioscience, 309646) was removed using a razor blade, and the remaining syringe was used as a casting mold for mounting the brain. The brain was immersed in 3 mL of 1% melted agarose prepared in sterile water, and then the solution and immersed brain were gently aspirated into the sample holder until the agarose solidified completely (Figure 7C). Next, the plunger of the syringe was gently pressed to extend the agarose cylinder containing the brain out onto a 10 cm Petri dish (Figure 7D). Extra solidified agarose at the bottom of the cylinder was trimmed off and then the embedded sample and holder were immersed in 25 mL of EZ View imaging solution inside a 50 mL conical tube and this tube was rocked gently in a vertical position on a horizontal orbital shaker at room temperature overnight to equilibrate the sample.

Figure 7

Download asset Open asset

To image the samples on the Zeiss Lightsheet Z.1, a custom-made, enlarged sample chamber was created, which allows sample diameters up to 1 cm compared to the original Zeiss ×5 chamber which only accommodates samples up to 4 mm (Figure 7E–G). A 1 mL syringe with a recessed magnet was then used as sample probe to attach to the sample holder (Figure 7H, I). To mount the sample on Lightsheet Z.1 for imaging, the 1 mL syringe with a recessed magnet was first loaded onto the system. The sample was placed into the custom-designed imaging chamber by hanging the mounted sample with a PTFE chamber cover with 3 mm opening (Zeiss) by the magnetic nut (Figure 7J). The imaging chamber with the hanging sample was assembled onto the lightsheet chamber (Figure 7K). The sample was then attached to the probe by lifting the sample with the chamber cover and attaching the magnetic nut to the probe (Figure 7L and M). The PTFE chamber cover was removed prior to imaging with a ×5/0.16 air detection lens and ×5/0.1 illumination lens. The data was acquired with a ×0.5 zoom at a resolution of 1.829 μm × 1.829 μm × 3.627 μm (X:Y:Z) in a tiled sequence with 20% overlap between tiled images. Acquired tiles were aligned and stitched together using Arivis Vision4D.

Mouse brains processed with the EZ Clear lipid removal and washing steps were used for standard immunofluorescent staining or iDISCO staining. For standard immunofluorescent staining, EZ Clear processed mouse brains were immersed in blocking buffer (×1 PBS + 0.08% Triton X-100 + 2% donkey serum + 0.05% sodium azide) at room temperature for 2 days, followed by incubation with primary antibodies (GFAP at 1:200 or αSMA-Cy3 at 1:200) diluted in 4 mL of blocking buffer for 4 days at room temperature. Whole brains were washed three times, 2 hr each in ×1 PBS, then once more in ×1 PBS overnight at room temperature. The samples were then incubated with secondary antibodies (1:500) and To-Pro 3 (1:1000) in blocking buffer for 4 days at room temperature with gentle agitation on an orbital shaker. The brains were then washed three times, 2 hr each wash, in ×1 PBS, then once more in ×1 PBS overnight at room temperature. The samples were then incubated in EZ View solution at room temperature for 24 hr to render the sample transparent.

For staining whole brains with the iDISCO protocol (Renier et al., 2014), EZ Clear processed mouse brains were incubated in permeabilization solution (×1 PBS + 0.2% Triton X-100 + 20% DMSO+ 0.3 M glycine + 0.05% sodium azide) at room temperature for 2 days, then incubated in iDISCO’s blocking solution (×1 PBS + 0.2% Triton X-100 +10% DMSO+ 6% donkey serum) at room temperature for 2 days. Samples were then washed in PTwH (×1 PBS + 0.2% Tween-20 with 10 μg/mL heparin) for 1 hr, twice, then incubated with primary antibody (GFAP at 1:200, αSMA-Cy3 at 1:200) in primary antibody incubation solution (PTwH + 5% DMSO+ 3% donkey serum) at room temperature for 4 days. Samples were then washed in PTwH four to five times at room temperature until the following day, then incubated with secondary antibody (1:500) and To-Pro-3 (1:1000) in incubation solution (PTwH + 3% donkey serum) at room temperature for 4 days, then washed in PTwH three times, 2 hr each wash in ×1 PBS, then once more in ×1 PBS overnight at room temperature, then incubated in ×1 PBS with 0.05% sodium azide for another 24 hr at room temperature. The samples were then incubated in EZ View solution at room temperature for 24 hr to render them transparent.

After whole brains were imaged in the EZ View solution, samples were equilibrated in PBS by washing in 50 mL of ×1 PBS four times, 1 hr each, and then once more overnight at room temperature. The brains were then immersed in a three-step sucrose gradient (10%, 20%, and 30%, prepared in ×1 PBS). The samples were then incubated at 4°C for each step until the tissue sank to the bottom of the vial overnight at 4°C. Samples were then embedded in optimal cutting temperature (OCT) medium (Sakura, 4583) and snap-frozen on a bed of crushed dry ice, then stored at –80°C until ready for processing. The frozen block was sectioned on a cryostat (Leica) at 12 μm for H&E staining or at 100 μm for immunofluorescent staining.

Before histological staining, slides containing mounted cryosections were equilibrated at room temperature, followed by rehydration in PBS. To perform H&E staining, slides were then incubated for 2 min in Modified Harris Hematoxylin Solution (Millipore-Sigma, HHS32-1L), rinsed with deionized water five times, then incubated for 5 min in tap water and then dipped 12 times in acidic ethanol (0.25% HCl in 70% ethanol), incubated 1 min in tap water two times, and then incubated for 2 min in deionized water. Slides were then incubated in Eosin Y Phloxine B Solution (EMS, 26051-21) for 30 s and dehydrated through an ethanol graded series (twice in 70%, twice in 80%, twice in 95%, and twice in 99.5%) for 1.5 min at each step. Lastly, slides were washed in Histoclear II (EMS, 64111-04) twice and coverslipped in DPX new (Millipore-Sigma, HX68428779) and stored at room temperature.

One-hundred μm sections were collected in ×1 PBS with 0.05% sodium azide in a 24-well plate (‘free floating’). For staining, sections were then incubated overnight in blocking buffer (×1 PBS + 0.08% Triton X-100 + 2% donkey serum) overnight at 4°C, followed by incubation with primary antibodies (CD31 at 1:200, GFAP at 1:200, beta-III tubulin at 1:200) in blocking buffer overnight at 4°C. Sections were then washed three times, 1 hr each in ×1 PBS, followed by incubation with secondary antibodies (1:500) and Hoechst (Millipore-Sigma, 14533, 10 mg/mL stock at 1:1000 dilution) in blocking buffer overnight at 4°C with gentle agitation on an orbital shaker. Sections stained with αSMA-Cy3 (1:200) were stained together with the Hoechst. The next day, after three, 1 hr washes in ×1 PBS at 4°C, sections were mounted on a charged slide (VWR Micro Slides, 48311-703) with 200 μL of EZ Clear imaging solution and coverslipped with #1 cover glass (VWR Micro Coverglass, 48366-067).

To compare the size changes of brains processed with different clearing protocols, brightfield images were captured using a Zeiss Stemi stereomicroscope and the Labscope Material App at the following stages: after perfusion, lipid removal, and RI matching. For brains that were too large to fit in a single field of view, four tiled images with at least 20% overlap were taken to cover the entire sample and stitched together using Arivis Vision4D. The size of each brain at different stages was quantified using Fiji. The ‘Selection Brush Tool’ under ‘Oval selections’ was used to outline the boundary of the brain from the captured brightfield images to measure the total pixel number within the outlined area.

To measure the fluorescence intensity and contrast at different imaging depths, as well as the preservation of the fluorescence overtime after storing samples in EZ View, Zen (Blue), and Fiji softwares were used to measure the mean (I_mean), maximum (I_max), and minimum (I_min) across the entire image at different imaging depths (z) through each mouse brains and organs imaged on lightsheet. For quantitative analysis of the signal intensity over imaging depth (lectin-649, αSMA-Cy3, and To-Pro 3), each experiment was acquired using the same imaging parameters for all wholemount samples (n=3).

To calculate the contrast of each image acquired through the imaging depth (z), the maximum (I_max) and minimum (I_min) of the perfused Lectin-649 signal of each image was used to calculate the Michelson contrast (Wiebel et al., 2016) and plotted against imaging depth.

Statistics analysis of ANOVA (one-way and two-way), t-test, and multiple comparisons were performed with GraphPad Prism 9 software.

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

We thank Sih-Rong Wu and Dr Huda Zoghbi for providing the Thy1-EGFP mouse brains and Dr Nanbing Li-Villarreal for critical review of the manuscript. We thank Dr. Kristy Red-Horse (Stanford University) for providing the inspiration to adopt the Evans Blue (EB) perfusion methodology by sharing her group's beautiful data in the mouse coronary vasculature at a NAVBO meeting some years ago, and we are indebted to Dr. Lori O'Brien (University of North Carolina, Chappel Hill) for developing the EB method and sharing her detailed protocol with us for EB perfusion. This project was supported by the Optical Imaging and Vital Microscopy (OiVM) Core for all imaging experiments and the Bioengineering Core at Baylor College of Medicine for manufacturing the custom imaging chamber for the Zeiss Lightsheet Z.1. The authors also want to thank Dr Cecilia Ljungberg and the RNA In Situ Hybridization Core at Baylor College of Medicine, which is, in part, supported by a Shared Instrumentation grant from the NIH (1S10OD016167). This work was supported by grants from the National Institutes of Health (5T32GM088129-10 to WDT, R01HL146745 to MED and JDW, R01HD099026 to MED, U42OD026645 to MED, R01HL159159 to JDW), the American Heart Association (22PRE916015 to CFS), the Cancer Prevention Research Institute of Texas (RP200402 to JDW), the Department of Defense (W81XWH-18-1-0350 to JDW) and the Canadian Institutes of Health Research (PJT-155922 to JDW).

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All animal research was conducted according to protocols approved by the Institutional Animal Care and Use Committee (IACUC) of Baylor College of Medicine (AN-4593, AN-8329, and AN-7731).

© 2022, Hsu et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.