Inagaki, S., Nakagawa-Tamagawa, N., Huynh, N.Z. et al. Isotonic and minimally invasive optical clearing media for live cell imaging ex vivo and in vivo. Nature Methods (2026). https://doi.org/10.1038/s41592-026-03023-y
[bioRxiv preprint]
[Research Briefing (Nature Methods)] [EurekAlert!] [FocalPlane] [The Transmitter]
Raw image data: https://ssbd.riken.jp/repository/484/
The reagent (SeeDB-Live/ACSF) is commercially available from Nacalai Tesque. Please contact them if you want to try.
Step-by-step protocol (PDF in Google Drive)
Please contact us for any questions (imai.takeshi.457@m.kyushu-u.ac.jp).
For the repeated in vivo imaging with a plastic film window, see also Manita et al. (2022).
Please note that the refractive index of SeeDB-Live is 1.363, which is higher than water (1.33) but lower than silicone (1.41).
It is strongly recommended to use objective lenses with a correction collar (water or multi-immersion) to minimize spherical aberrations. We used XLPLN25XWMP (Evident/Olympus) and LUPLAPO25XS (Evident/Olympus).
It is also recommended to use the immersion of the same index (e.g., 15.6% TDE in H2O, RI 1.363). You cannot make the best use of SeeDB-Live with a water-immersion lens without correction collar, because of the spherical aberrations (RI 1.33 vs 1.363); the transparency will increase with SeeDB-Live, but the brightness and resolution will decrease.
It should be noted that BSA has autofluorescence in a blue-green range (e.g., excitation at 405, 448, and 488 nm), which is most likely derived from endogenous flavin (See Ext Dat Fig. 1n). It is therefore recommended to use confocal or two-photon microscopy, which are more resistant to autofluorescence signals.
When we use wide-field microscopy, we use red dyes, rather than green dyes. For example, in our voltage imaging with Voltron2, we used JF549.
We have been successful with some samples but not with others. Importantly, BSA cannot easily penetrate physical barrier (dura mater, tight junction, etc.). In addition, the high concentration of BSA may limit diffusion of nutrients, oxygen, and cytokines, which may affect normal growth of some types of organoids.
HeLa cell spheroids, intestinal organoids, primary culture of neurons, acute brain slices, mouse cerebral cortex in vivo (after durotomy)
Optic cup induction from ES cells: Growth factors/cytokines in the media may be critical for the long-term induction process. Acute functional assay is possible for ES/iPS-derived organoids.
Long-term culture of organoids: Growth was often affected possibly due to limited diffusion of nutrients and oxygen. We typically limited the SeeDB-Live treatment up to 4 hours/day.
Epithelial tissues: SeeDB-Live cannot easily penetrate tight junction. Therefore, it is recommended to cut the epithelial organoids into pieces.
Myelin-rich brain regions (brainstem, nodose ganglion, etc.): Refractive index of lipid is much higher than SeeDB-Live.
Subcortical regions (hippocampus, striatum, hypothalamus, etc.): Passive diffusion of SeeDB-Live is typically limited to the cerebral cortex. We have not yet achieved 2P imaging of subcortical regions.
Skull should be better cleared with other invasive methods (such as SeeThrough and HOTS).
Q1. Can we use SeeDB-Live for C. elegans, Drosophila (larva, adult), zebrafish, and plants?
Most likely no. BSA cannot penetrate physical bareer of these species. In Drosophila, if you can remove a cuticle and make a cranial window for the brain, you may be able to introduce BSA to a part of the brain, but we have never tried.
Q2. Can we clear heart, liver, and kidney?
Maybe no. Again, BSA probably cannot efficiently penetrate into these organs.
Q3. Can we culture organoids in SeeDB-Live?
Maybe no. We successfully maintained monolayer cultures, such as HeLa cells and primary culture of neurons. However, we observed slower growth of spheroids and organoids in SeeDB-Live culture media. This is possibly due to reduced circulation of oxigen and nutrients. Consistent perfusion of the culture media may improve the growth, but we have never tested.
Q4. Can we clear the myelin-rich brain region?
No. SeeDB-Live is designed to reduce the light scattering happening at the interface of the cytosol and extracellular media. Therefore, SeeDB-Live can reduce the light scattering in the gray matter, but not white matter.
Q5. How can we oxigenize SeeDB-Live/ACSF?
Because SeeDB-Live contains high concentrations of BSA, bubbling will produce a lot of foams. We typically used a glass bottle to equilibrate the SeeDB-Live with oxygen. We just supplied oxygen into the upper space of the bottle and waited for ~1 h until the SeeDB-Live is fully equilibrated with oxygen.
Q6. Can we prepare SeeDB-Live for our favorite culture medium?
We only have experiences for some culture media (e.g., DMEM). See Supplementary Table 1 of our paper. It should be noted that the concentrations of ions (especially divalent cations) are not fully tuned in this recipe.
Q7. I could not purchase the suggested BSA (fraction V, bioWORLD) in our country. Could you suggest alternative products?
BSA fraction V from other vennders may work, but there may be some differences. For example, the amount of residual salts are different among venders. Perhaps, it may be easiest to use the commercialized SeeDB-Live/ACSF from Nacalai Tesque.
Q8. Can I buy the commercialized SeeDB-Live/ACSF (Nacalai Tesque) in my country?
Please check out your local distrubutors in this site.
Q9. Can we use light-sheet microscopy to image organoid samples cleared with SeeDB-Live?
Maybe no. The imaging depth is improved only by two-fold. We recommend using confocal or two-photon microscopy.
Q10. Is the clearing with SeeDB-Live reversible?
Yes. We typically perfuse the in vivo brain for 1 hour and performed imaging for up to 1 hour. Within the next 1-2 hours, SeeDB-Live will be washed out by the endogenous CSF. As for the ex vivo samples, tissues return to the normal appearance (opaque) once the SeeDB-Live is exchaged to the normal media.
Q11. Can we clear a deep part of the brain?
We have not yet succeeded in the whole-brain clearing in vivo. In our published protocol, we can only clear the cerebral cortex. If you are using a GRIN lens, perhaps, you may be able to embed a canula, together with the GRIN lens, to clear the target region.
Q12. Have you ever tried 2-photon imaging of subcortical brain regions, such as hippocampus, in vivo?
No success for now. We are now trying to improve the delivery of SeeDB-Live in vivo.
Q13. Have you ever tried the combination with 3-photon microscopy? Have you tried super-resolution imaging?
Not yet, but we'd be happy to collaborate with you!