SeeDB2
Publications
1) Meng-Tsen Ke, Yasuhiro Nakai, Satoshi Fujimoto, Rie Takayama, Shuhei Yoshida, Tomoya S. Kitajima, Makoto Sato, and Takeshi Imai. "Super-Resolution Mapping of Neuronal Circuitry with an Index-Optimized Clearing Agent". Cell Reports 14, 2718–2732 (2016)
RIKEN press release (English)
RIKEN press release (Japanese)
JST press release (Japanese)
RIKEN CDB news (Japanese)
2) NEW: Meng-Tsen Ke, Takeshi Imai. "Optical Clearing and Index Matching of Tissue Samples for High-resolution Fluorescence Imaging Using SeeDB2". Bio-protocol (2018)
Step-by-step protocol
Step-by-step protocol (bio-protocol.org) [HTML version] (10-20-18).
Tissue Clearing with SeeDB2 (08-04-15 ver).
Imaging setup (02-06-15 ver).
SeeDB2 Kit commercialized from FUJIFILM Wako Pure Chemical.
See also Supplementary Experimental Procedures of Cell Reports paper for other types of samples, including Drosophila brain, mouse oocytes, and cultured cells on glass-bottomed dishes.
Principles of SeeDB2
SeeDB2 is not just an improved tissue clearing agent with better transparency. SeeDB2 is designed to minimize spherical aberrations, enabling high-resolution imaging with high-NA oil-immersion objective lenses. SeeDB2 is particularly powerful in super-resolution microscopy. SeeDB2 is useful not only for thick brain tissues, but also for thin samples for cell biology and histology; SeeDB2S should work better than conventional mounting media (e.g., ProLong Gold and VECTASHIELD) with high-NA oil-immersion objective lenses.
Previously, 97% TDE has been proposed for this purpose (Staudt et al., 2007), but some of the common fluorescent proteins and dyes were quenched in TDE. SeeDB2 preserves many of the fluorescent proteins and dyes.
TIPs
1. Samples suitable for SeeDB2
1-1. SeeDB2 is useful for high-resolution imaging of all kinds of biological samples. SeeDB2 is useful for mouse brain, liver, muscle, intestine, bone, and whole embryo.
1-2. SeeDB2 is also useful for Drosophila samples including whole brain. SeeDB2 is much better than commercialized mounting media (e.g., Vectashield). Unlike solvent-based mounting media (DPX or TDE), SeeDB2 does not quench fluorescent proteins.
1-3. SeeDB2 is useful for any kinds of cell biology samples, including oocytes and cultured cells. Intracellular organelle, such as mitochondria, microtubules, and actin fibers, is well preserved during clearing process. For thin samples, SeeDB2 can be used similarly to a common glycerol-based mounting media (e.g., ProLong Gold). Alexa dyes, DAPI, and Mitotracker are all stable in SeeDB2.
2. Choice of protocols.
2-1. SeeDB2S is optimized for high-NA oil-immersion objective lenses. Since SeeDB2S is viscous and oil-immersion lenses typically has short WD, we recommend using relatively thin samples (< 500um) for SeeDB2S.
2-2. SeeDB2G is optimized for glycerol-immersion lenses to achieve highest resolution, but water- and multi-immersion objective lenses are also useful with reasonable resolution. SeeDB2G is useful for larger tissues due to lower viscosity. SeeDB2G is similar to earlier methods (ref, and Yang et al., 2014), but is optimized to improve tissue clearing performance and to minimize tissue damages in simple immersion protocols.
2-3. SeeDB (fructose-based) still has advantages for some applications. Since SeeDB does not contain detergents, it is useful for clearing DiI-labelled samples and for correlated light-electron microscopy (CLEM). SeeDB produces some autofluorescence, but this is sometimes useful to identify unlabelled neuronal structures (e.g., neuropils).
2-4. SeeDB2 is optimized for high-resolution imaging using high-NA objective lenses; therefore, CLARITY, 3DISCO, or CUBIC may be a better choice for larger tissues (e.g., whole brain).
3. Sample preparation.
3-1. In the high-resolution imaging with SeeDB2S, the WD (typically 200-300 um) is the limiting factor for depth. Therefore, we recommend using slice samples for large tissues.
3-2. Small samples (e.g., fly brain) are easily lost during clearing because they become transparent. We recommend using a fluorescence stereomicroscope to transfer samples.
3-3. Cultured cells on glass-bottomed dishes can be directly used for clearing.
4. Microscopy tested for SeeDB2S
Confocal (best with high-NA oil-immersion lenses)
STED (Leica Microsystems)
SR-SIM (Carl Zeiss)
SD-OSR (Olympus)
FV-OSR (Olympus)
Airyscan (Carl Zeiss)
HyVolution (Leica Microsystems)
5. Microscopy tested for SeeDB2G
Two-photon (best with multi-immersion lenses)
Confocal (best with high-NA glycerol-immersion lenses)
Light-sheet DLS (Leica, best with a customized mirror devises)
6. Sample storage
SeeDB2 samples become easily moldy. Saponin stock solution should be sterilized by filtration. Sodium azide should be added for longer storage. SeeDB2G samples can be stored in a fridge, but SeeDB2S samples should be stored at room temperature.
7. Antibody staining
Antibody staining should be performed prior to the clearing step.
8. Image analysis and reconstruction
8-1. Airyscan and HyVolution
Using SeeDB2, we can easily obtain large volume of 3D data. Deconvolution processing in Airyscan and HyVolution should be performed using highest-spec workstations (e.g., 192GB RAM for Win7).
8-2. Reconstruction
We are routinely using Neurolucida (MBF) for quantification purpose. We also tested VAST Lite for visualization purpose. Open source software is also available. For example, Vaa3d can handle a huge amount of stitched image data.
Trouble shooting
Additional Notes
1) We recommend preparing Solution 3, 4, and 5 one day prior to usage. To remove bubbles produced during mixing process, the tube should be left for a while before the mounting.
2) Sealing samples in the imaging chambers and glass slides.
Nail polish should not be used to seal the samples. For thick samples, we routinely use silicone rubber sheet. When we seal samples on glass slides, we use an excess amount of SeeDB2, which then form a tight film after water evaporation.
3) Detecting sample surface.
Because refractive index of SeeDB2S is the same as that of coverslip, it is difficult to determine the surface of the sample. Make sure to attach the sample to the coverslip, which can help us estimate the surface depth.
4) Z movements and Clamping.
Because SeeDB2 allows imaging up to the limit of the WD, glass slides are often pushed up by the objective lenses when you go closer to the upper limit, leading to the movement artifacts. We recommend using a clamp or put a weight to avoid the movement of glass slides.
5) Objective lenses for Leica DLS.
SeeDB2G is useful for light-sheet imaging. However, commercialized water-immersion objective lenses and mirror devises from Leica are not suitable for SeeDB2. Please contact Leica for SeeDB2G-optimized mirror devises.
6) Omnipaque 350 is available from several pharmaceutical companies (e.g., Daiichi-Sankyo, GE healthcare, Bayer). Please note that Omnipaque 350 is only supplied for clinical uses (requires Pharmacy licence). You can also prepare SeeDB2G (Omnipaque350) by dissolving iohexol powder (commercialized as Histodenz from Sigma-Aldrich, D2158 or Nycodenz from Axis-shield, #1002424) at 75.5% (w/v) or 56.2% (w/w) in Tris-EDTA solution. For example, mix 10g Histodenz and 7.5ml 10mM Tris-Cl/0.25mM EDTA (pH 7.6) to prepare SeeDB2G. You can also buy ready-to-use SeeDB2 solutions (SeeDB2 Kit) from Fujifilm-Wako Chemical.
7) Silicone rubber sheet (transluent, medical grade) is available from various companies (e.g., Togawa gum, Professional Plastics, CS Hyde). You can cut it to your desired size and shape to make your custom-made imaging chamber. The thiskness should fit your sample to avoid "floating" during imaging. The rubber sheet adheres to coverslips without glue. You can also prepare custom-made imaging chambers using spacers and silicone glue.
NEW 8) We recently noted that Alexa dyes are photobleached more quickly in SeeDB2 than in other mounting media. For this reason, we recommend using 97% TDE (Staudt et al., 2007) for immunostained samples instead of SeeDB2. 97% TDE also has a refractive index of 1.52. It should be noted that fluorescent proteins are best preserved in SeeDB2. In summary, if you want to see fluorescent proteins, SeeDB2 is recommended; if you want to use immunostained samples, 97% TDE is better.
NEW 9) If you need to improve the transparency (especially for myelin-rich axons in the adult), we suggest pre-treatment with ScaleCUBIC-1 (25% (wt/wt) urea (#219–00175, Wako), 25% (wt/wt) N,N,N’,N’-tetrakis(2-hydroxy-propyl)ethylenediamine (#T0781, TCI), and 15% (wt/wt) Triton X-100 (#12967–45, Nakalai-tesque)) for up to 24 hours. Incubation time should be minimum to avoid the loss of fluorescent proteins. After the pre-treatment, samples can be cleared with SeeDB2.
NEW 10) An immersion liquid for some Glycerol objective lenses may have lower refractive index than SeeDB2G. 92% w/w Omnipaque350 + 8% Tris-EDTA matches Type G immersion liquid.
NEW 11) The original protocol may still produce morphological changes, depending on sample types. In that case, we suggest more gradual clearing with intermediate concentration steps with more time.