For many years, to study thick tissues like entire organs, biologists have tended to cut thick volumes into thin sections which allowed the study of tissues only in two dimensions. The growing interest in studying large tissues such as the brain in three dimensions has prompted scientists to introduce new methods which involve volumetric imaging. However, one of the limitations in imaging thick or entire organs is that most tissues are not transparent, which is problematic when we want to look into a tissue volume.

The translucency of tissue is caused by light scattering. Any biological tissue is mostly made of water surrounded by lipids and proteins. Each of these components has a different refractive index (RI), which means that their interaction with light passing through the tissue is different. For instance, water has a RI of 1.33, proteins above 1.44, and lipids above 1.45. These RI differences make a tissue translucent which means non-transparent.

All tissue clearing methods, although different in other aspects, invariably aim to normalize the RI of a biological sample by removing, changing or replacing some components.

Since the publication of Atsushi Miyawaki's ScaleA2 in 2011, several clearing methods have been published (also see recent review [3] ). Here, we compare seven main methods currently in use (see Table 1). These seven methods fall into two categories: solvent-based and aqueous-based methods. Additional variations such as FlyClear [4] are not discussed. Small pieces of tissues, such as organoids, can be similarly cleared with much shorter time and simpler formula. For example, Serra D et al used an RI-matched solution of iodixanol, N-methyl-D-glucamine, diatrizoic acid to clear intestinal organoids [5]. So C et al cleared lipid droplets in bovine, ovine, and porcine oocytes, with 4000 U/ml Candida rugose lipase in a mild buffer containing 0.2% sodium taurocholate and protease inhibitor cocktail at room temperature for 20 to 40 min [6]. New protocols have also be devised. For example, Grüneboom A et al used nontoxic ethyl cinnamate to clear bone tissues in a protocol called simpleCLEAR [7] and Gur-Cohen S et al cleared skin tissues in ethyl cinnamate to study lymphatic capillaries [8].

[enlarge]

Figure 1. Steps of the clearing process [1, 2]

Each type of tissue clearing includes four main steps: initial treatment, permeabilization, immunolabeling, and final refractive matching clearing [1] (see Figure 1). The initial treatment (pretreatment) is performed to eliminate pigment molecules which may absorb tissue autofluorescence and affect the results of the clearing. Permeabilization aims to assist the appropriate diffusion of the clearing solution throughout a tissue specimen and is mostly done by Triton X-100 and other detergents or DMSO. Concerning immunolabeling, different fluorescent probes may be applied to the labeling of target proteins. The immunolabeling can be achieved by passive diffusion or by application of electrical fields [9]. Besides, centrifugation may significantly increase antibody diffusion [10]. Tables 2 and 3 list hydrogel- and water-based clearing methods, their advantages, and disadvantages, respectively. More recently, the hydrogel-based CLARITY method has been modified to use passive tissue fixation in 4% paraformaldehyde (PFA) to generate transparent pancreas [11].

The tissue clearing methods based on organic solvents dehydrate the tissue, remove lipids and normalize the RI to a value around 1.55, probably matching the RI of the remaining components [20, 21].

Advantages: high quality and speed in clearing, which is useful when one wants to perform immunostaining on the entire tissue which usually takes several weeks.

Disadvantages: toxic and/or corrosive chemicals are used. Moreover, since all lipids are removed, no lipid staining is possible.

3DISCO uses a series of organic solvents for normalizing the RI of a tissue [22]. This method was developed at the Max-Planck Institute in Munich to clear and image unsectioned mouse brain and spinal cord.

It consists of three steps:

1. Dehydration with growing concentrations of tetrahydrofuran (THF). THF preserves fluorescence better than other dehydrating solutions. Dehydration reduces the sample size;
2. Extraction of lipids with dichloromethane (DCM);
3. Immersion in dibenzyl ether (DBE) for RI normalization which leads to a transparent sample and for imaging.

It works on many tissues including spinal cord [23], lung, spleen, lymph nodes, mammary gland, and tumors.

uDISCO is a variant of the 3DISCO method, and it is used specifically for clearing the whole body of a mouse [24]. This variant uses tert-butanol instead of THF in the first step and also uses a different imaging solution (not DBE) which preserves better the fluorescence. It also enhances the shrinkage of tissue which means that one can observe large samples in one image.

The aqueous-based tissue clearing methods are based on the immersion of the tissue in aqueous solutions that have RIs in the range of 1.44-1.52 [20, 21]. These methods can be further divided into simple immersion (SeeDB, ClearT2), hyperhydration (Scale, ScaleS, CUBIC) and hydrogel embedding (PACT).

Advantages: preservation of the fluorescent protein emission, preservation of lipids and the tissue architecture; technical simplicity.

Disadvantages: slow clearing methods and not effective for big tissues. Thus, only slabs of tissue, organoids or insects can be cleared with these methods.

In this type of methods, the tissue is immersed in an aqueous solution containing a molecule with a high RI, such as sucrose (RI=1.44), fructose (RI=1.50), 2,2’-thiodiethanol (TDE) (RI=1.51) and formamide (RI=1.44) [25]. The methods are not the most efficient method in clearing, but are low-cost and compatible with many fluorescent and protein targeting dyes for staining after the clearing.

It is based on immersion in graded fructose solutions, and it can be used only for small samples.

It is based on immersion in formamide, and it is relatively quick. However, it is considered as one of the worst methods for tissue clearing.

This type of methods allows to both remove lipids and reduce the RI of the sample (1.38-1.48) during the clearing [25]. Lipid removal is achieved using detergents or high concentrations of urea.

Scale was the first method to have been introduced using the hyperhydration technique. It is based on a combination of detergents, such as Triton X-100, to remove lipids and urea which is used to penetrate and thus hydrate hydrophobic regions of proteins with a high RI. This treatment brings the RI down around 1.38.

It is a sorbitol-based clearing method which has replaced the Scale method since it is faster, removes the tissue expansion problem and preserves lipids as well.

It is also a urea-based method but it has an extra step based on sucrose which accelerates the clearing process. It uses a high concentration of Triton to maximize the removal of lipids. The consequence of a high concentration of detergent is an increase in protein loss which might affect the immunostaining after the clearing. Ouadah Y et al cleared thick mouse PFA-fixed lung coronal sections (500 um) with CUBIC [14].

This type of methods can remove lipids with detergents and bring the RI around 1.38-1.45 but is also able to stabilize proteins by cross-linking the proteins to a hydrogel. This method is very efficient in clearing the tissue, but it is also very slow.

The tissue is first embedded in the hydrogel, and then the lipids are removed by a long (weeks) incubation with detergents (8% SDS) or quicker (days) through electrophoresis. The final step is the immersion of the tissue in solutions like FocusClear (a water-soluble clearing agent), RIMs (refractive index matching solution), TDE or glycerol.

Recently, a new method for clearing tissue has been introduced called polyethylene (PEG)-associated solvent system (PEGASOS) [26]. This method efficiently reduces both hard and soft tissue transparent, but still, it can maintain the fluorescence which is usually lost with most of the methods described above.

The PEGASOS method consists of several steps including fixation in 4% PFA, decalcification using EDTA (only for hard tissue like bones), decolorization with quadrol/ammonium, delipidation with tert-butanol gradients, dehydration with tB-PEG and clearing using a solution containing PEG. The PEG component protects the endogenous fluorescence. It is relatively short (7-14 days), inexpensive and easy to perform.

Adipo-Clear method is based on methanol/dichloromethane delipidation of adipose tissues to generate high transparency for high-resolution imaging. Since the delipidation significantly decreases the intensity of endogenously expressed fluorescent proteins, imaging of these proteins can be performed by immunostaining. Adipo-Clear method can be used to study the development of adipose tissue, the structural organization of adipose cells and functions of adipocyte progenitor cells. Adipo-Clear provides simultaneous analysis of various histological structures within the adipose tissue. The method allows to study neuronal projections, vascular network and interactions between immune cells and adipocytes. Thus, Adipo-Clear method guarantees the complete clearance of lipids in adipose tissues, while maintains tissue morphology [27].

According to the original protocol, published Chi et al. [28], the samples were fixed in 4% PFA at 4°C and underwent dehydration in 25%, 50%, 75%, 100%, 100% methanol for 30 min in each solution at RT. The dehydrated samples were washed with 100% dichloromethane and incubated overnight in dibenzyl ether. The delipidation with methanol/dichloromethane is a highly important step, since insufficient treatment would cause dim images. Furthermore, the same group published another study where Adipo-Clear technique was used to generate high-resolution three-dimensional images of the adipose tissue [29]. This method was modified from iDISCO/iDISCO+ and is characterized by complete elimination of the lipid from the tissue while keeping native morphology fully preserved. Adipo-Clear is especially useful for analyzing filamentous structures such as nerves and blood vessels.

This protocol has been adapted by Muller et al. [23] in the study of the microbiota-mediated regulation of gut-extrinsic sympathetic activation via a gut–brain circuit [23]. The authors of the study have reported that the depletion of gut microbiome stimulates the expression of cFos, a neuronal transcription factor. Muller et al. designed the Adipo-Clear method similarly to the protocol published by Chi and co-authors [28]. Moreover, Muller et al. have developed a modified CalAdipoClear method, which was applied for clearing of spinal columns and included tissue decalcification. Similar to the original Adipo-Clear method, after the fixation in 4% PFA, the tissues underwent dehydration in 20/40/60/80/100% methanol followed by dichloromethane and then rehydration in 100/80/60/40/20% methanol. After that, the samples were decalcified in Morse solution consisted of one part 45% formic acid/one part 0.68 mM sodium citrate dihydrate, followed by incubation in primary antibody dilution in PTxwH for 7 days and then in secondary antibody dilution also for seven days. The technique was completed by dehydration in 20/40/60/80/100% methanol followed by dichloromethane and clearing in dibenzyl ether. Since organic solvents can induce the denaturation of proteins, some antibodies may not function well with the delipidation. Hence, selecting an appropriate antibody is a crucial step for this technique.

In addition, Adipo-Clear was combined with light microscopy for visualization of carotid and brachiocephalic arteries to generate three-dimensional vascular architecture [30]. The authors have performed visualization of angiogenesis and the development of a vascular plexus surrounding the carotid artery. The presented technique reconstructed the process of neointima development after vascular injury. Moreover, the study characterized the plaque formation in the arteries of ApoE-deficient mice, which received high-fat diet. Also, it generated three-dimensional imaging of atherosclerotic plaques and was used to evaluate their location, volume and shape.

DJ Jafree et al found that a hybrid solution of benzyl alcohol and benzyl benzoate (BABB) was best suited for clearing renal tissue after evaluating ScaleA2, ScaleS, and ClearT2 [31]. Bossolani GDP et al compared five aqueous-based clearing protocols (SeeDB2, CUBIC, ScaleS, Ce3D, and UbasM) and four organic reagent-based clearing protocols (3DISCO, iDISCO+, uDISCO, and Visikol® ) on segments of small intestine from CX3CR1^GFP/GFP and wild-type mice [32]. They found that three organic-based protocols (except for Visikol) rendered tissue highly transparent; however, all of them caused substantial tissue shrinkage and deformation. Among aqueous-based protocols, only Ce3D achieved full-thickness tissue transparency and displayed excellent GFP retention and preservation of tissue morphology. The authors concluded that Ce3D was the best clearing protocol for mouse intestinal walls [32]. SM Zaytsev et al evaluated four optical clearing conditions: a mixture of polyethylene glycol 400 (PEG-400), propylene glycol and sucrose; a mixture PEG-400 and dimethyl sulfoxide (DMSO), saline solution as control and a "dry" condition with spatially resolved multimodal spectroscopy [33].

Several methods of optical sectioning or microscopy techniques were developed for cleared tissue specimens. In particular, two-photon and confocal microscopes are appropriate for imaging of small areas [2]. Light-sheet fluorescence microscopy (LSFM) is more suitable for large cleared specimens [24, 34]. This method consists of the complete illumination of a single tissue level and acquisition of the emission fluorescence of this narrow level by a microscope camera. Besides, optical projection tomography is applied for cleared samples for imaging with high resolution and may provide both structural and fluorescent data [35, 36].

This step-by-step protocol for optical clearing of biological tissue with ScaleA2 [16] is provided by ScaleA2 inventor - Dr. Atsushi Miyawaki lab.

PDF version.

Stock solution

1. Dissolve 10 g of Triton X-100 (e.g., Nacalai Tesque, Code 355-01 or similar grade) in 80 ml of Milli-Q water by stirring.
2. Add Milli-Q water to make 100 ml and stir until well mixed.
3. Store at 4 ºC.

• 4 M urea
• 0.1% (wt/vol) Triton X-100
• 10% (wt/wt) glycerol

1. Dissolve 240.24 g of urea crystals (e.g., Wako Chemical, Code 217-615 or similar grade) in 800 ml of Milli-Q water by stirring.
2. Add 10 ml of 10% (wt/vol) Triton X-100 solution.
3. Add 100 g of glycerol (e.g., Sigma, Code 191612 or similar grade).
4. Mix well by stirring.
5. Add Milli-Q water to make 1,000 ml and stir until well mixed.
6. Store at room temperature.

• 8 M urea
• 0.1% (wt/vol) Triton X-100

1. Dissolve 480.48 g of urea crystals in 800 ml of Milli-Q water by stirring.
2. Add 10 ml of 10% (wt/vol) Triton X-100 solution.
3. Add Milli-Q water to make 1,000 ml and stir until well mixed.
4. Store at room temperature.

• Allow each solution to stand after preparation for at least 1 day prior to use.
• Sterilization is not required.
• Addition of preservative is not required.

PDF version.

1. Fix a mouse via transcardial perfusion with 4% PFA ( paraformaldehyde) / PBS (w/v) (pH 7.5–8.0) at room temperature (RT).
   Acid fixatives (pH < 7.0) may quench FPs irreversibly. On the other hand, use of alkaline fixatives (pH > 8.0) may results in damage to samples later at Step 8.
2. Remove the brain and postfix it with 4% PFA/PBS (pH 7.5–8.0) at 4 ºC for 8–12 hrs.
   It may be hard to fix the brain of older mice (> 3 weeks old) across the pia mater. For older mice, it is recommended to split the brain into two pieces. For example, unless commissural connections are examined, cut the brain at mid-plane into two cerebral hemispheres. The following procedure assumes that a cerebral hemisphere is processed as a sample to be cleared. Alternatively, if you prefer to keep the entire brain intact, making a few incisions will facilitate fixation throughout the whole brain.
3. After washing with PBS, incubate the sample in 20% sucrose /PBS (w/v) (pH 7.4–7.8) at 4 ºC (or RT) for 1–2 days.
4. Embed the sample in OCT compound (Sakura) and freeze with liquid N2.
5. Thaw the sample in PBS (25 ml/ 0.5 g tissue) at RT for 20 min with gentle shaking.
6. Rinse the sample in PBS (25 ml/ 0.5 g tissue) at RT for 20 min with gentle shaking.
7. Fix the sample again with 4% PFA/PBS (pH 7.5–8.0) for 30 min at RT.
   It is highly recommended that sample fluorescence is checked at every step prior to Scaling (Step 8). It is somewhat common that the fluorescence can easily be lost during the fixation process. For example, incomplete fixation may result in washout of soluble forms of fluorescent proteins (FPs).
8. Transfer the sample into ScaleA2 solution (20 ml/ 0.5 g tissue) in a see-through vial.
   Since ScaleA2 is free of salt, PBS-derived salt remaining in the sample is gradually washed out. In the process of tissue clearing, salts cause white precipitates and thus should be avoided.
   It is important to use a container where you can easily assess sample transparency.
9. Incubate the sample in ScaleA2 at 4 ºC (or RT) for 2–14 days or longer with gentle shaking. Exchange ScaleA2 if necessary.
   Clearing larger or harder (from older animals) samples requires longer incubation times. Check the transparency of the sample intermittently. Scale makes samples soft and fragile, like jelly, so care should be taken not to damage or destroy the sample.
   It may be hard to clear the brain of old mice (> 3 weeks old) across the pia mater. It is thus recommended the brain be split into two pieces or introduced with a few slits (see Step 2).
10. Observe the sample under an upright microscope. Fresh ScaleA2 solution is used as the immersion medium.
    If the sample needs to be stabilized over an extended time period for observation, please see below for “the procedure for immobilizing cleared sample.”
11. Store the cleared sample in fresh ScaleA2 solution at 4 ºC (or RT). Fluorescence is not lost over time during long-term storage. There is no need to add preservatives such as sodium azide.

PDF version. PDF version also has a schematic representation.

0.35% (w/v) agarose dissolved in Milli-Q water (Melt in a microwave oven and cool down to 37–40°C).

1. Transfer a Scaled sample from Scale solution onto a plastic dish measuring 60–100 mm diameter. Air-dry for 10 min at RT. To dry the surface of the sample lightly, wick away excess Scale solution carefully with filter paper.
   Scaled samples are difficult to cut due to their softness. If the sample needs to be trimmed for observation, follow the procedures in Steps 2–6. If not, go to Step 7.
2. Fill the dish with a melted agarose-water solution to embed the sample completely. Avoid introducing air bubbles into the solution.
3. Allow the agarose to get to harden at RT and air-dry its surface.
4. Trim the sample embedded in the agarose gel with a scalpel.
5. Remove the agarose gel from the trimmed sample carefully.
6. Place the (trimmed) Scaled sample onto a plastic dish measuring 60–100 mm diameter for observation using an upright confocal or two-photon excitation microscope. Position the sample with the observation part on top.
7. Pour a melted agarose-water solution gently over the top of the sample. Let the viscous solution run radially down the sides. After hardening, the mountain-shaped agarose gel covers most of the sample deeply, but the summit should be covered with only a thin film of agarose gel. In this setup, the observation part should be accessible to an objective lens.
8. Air-dry the surface of the entire agarose gel for approximately 30 min at RT. Then immobilize the gel edge onto the surface of the plastic dish with fast-drying adhesive (Aron Alpha or Krazy Glue).
9. Pour ScaleA2 solution into the dish until the mountain is submerged. Gently shake the plastic dish using an orbital shaker for 3 hours at RT. The ScaleA2 solution is being diluted with the remaining water. Repeat this step with new ScaleA2 solution.
10. When ScaleA2 is fully substituted as an immersion medium, the sample is ready for observation.

Dr. Atsushi Miyawaki wrote the step-by-step-protocol for ScaleA2. Dr. Elisa Corsiero in February 2019 and Dr. Konstantin Yakimchuk in November 2017 contributed to the rest of this article.