1) Richi Sakaguchi, Marcus N. Leiwe, Takeshi Imai. "Bright multicolor labeling of neuronal circuits with fluorescent proteins and chemical tags". eLife (2018)
[SSBD Database] [GitHub] [Addgene] [YouTube (1)] [YouTube (2)] [YouTube (3)] [YouTube (4)]
2) Marcus N. Leiwe, Satoshi Fujimoto, Toshikazu Baba, Daichi Moriyasu, Biswanath Saha, Richi Sakaguchi, Shigenori Inagaki, & Takeshi Imai. "Automated neuronal reconstruction with super-multicolour Tetbow labelling and threshold-based clustering of colour hues". Nature Communications (2024)
[Original microscopy images (SSBD Repository)], [Post-hoc chromatic aberration correction codes (GitHub)], [Linearity correction program for Leica HyD (GitHub)], [ImageJ plugin for Linear Unmixing (GitHub)], [dCrawler (MATLAB and Python) (GitHub)], [QDyeFinder (GitHub)], [SWC exporter for Imaris Filament Tracer (GitHub)] [Addgene] [YouTube]
All plasmids are available from Addgene.
All plasmids are available from Addgene. Addgene will start an AAV packaging service from Addgene plasmids. See the announcemnt from Addgene.
pAAV-SYN1-tTA (neuron-specific, but lower expression)
pAAV-CAG-tTA (higher expression)
pAAV-FLEX-tTA (Cre-dependent, higher expression)
Tetbow is an improved version of Brainbow, allowing for bright multi-color labeling of neurons using fluorescent proteins and chemical tags in cleared brains. Tetbow samples can be best visualized with SeeDB2.
Plasmids (fluorescent proteins, chemical tags, and AAVs) are now available from Addgene.
See also an Addgene blog post on Tetbow.
Protocol (01/26/17 unpublished ver in Japanese)
1) Plasmids are available at Addgene.
2) Prepare plasmids using an endtoxin-free plasmid purification kit (e.g., NucleoBond® Xtra Midi/Maxi EF from MACHEREY-NAGEL). We recommend following plasmid conditions: pCAG-tTA2 (0.1-0.25 ug/uL) and pBS-TRE-XFP-WPRE (0.25 ug/uL each, 3 or more colors). Also label DNA solution with FastGreen for in utero electroporation. Concentrations of these plasmids can be further optimized depending on the conditions and target cells. Note that excessive amont of pCAG-tTA2 leads to reduced expression of XFPs.
3) Perform in utero electroporation as described elsewhere (e.g., Saito, 2006).
4) Deeply anesthetize mice with an overdose nembutal injection. Then intracardially perfuse anamials with 4% PFA in PBS. Postfix the dissected brains in 4% PFA O/N. Prepare brain slices with vibratome (optional).
5) (optional for chemical tags) Label chemical tags with their substrates. Incubate brain slices in 2mL of 2uM substrate solution in 2% saponin and PBS O/N. Wash the slice with PBS (3x 30min).
6) Clear the tissues with SeeDB2 and mount the sample in the imaging chamber as described in SeeDB2 protocol. Fluorescent proteins are very stable in SeeDB2 solution. For tissues labeled with chemical tags, you can chose other clearing agents, such as TDE and 3DISCO.
7) Confocal imaging. To minimize chromatic aberrations, we strongly recommend using Glycerol-immersion lenses or specialized lenses for cleared tissues with a collection collar. Refractive index of SeeDB2G is ~1.46.
1) Prepare four AAVs from Addgene #104109-104112. Each AAV should be prepared separately. See an Addgene blogpost for more details.
2) Inject of AAVs to your preferred brain area.
Optimum ranges of AAV titers: 3×10^7 ~ 4×10^8 gc/mL (pAAV-SynI-tTA)
1×10^10 ~ 1×10^11 gc/mL (pAAV-TRE-XFP-WPRE).
Note: Optimum titer may be different for different cell types. We typically inject 100-1,000 nL of AAV solution. The optimum titer may be different for different injection volumes too.
3) Sacrifice animals at the appropriate time and fix samples.
Note: The timing of the sacrifice is critical. Typically, it takes a few weeks for the optimum expression of XFPs. After that, neurons may start to show morphological abnormalities due to the excessive amount of XFPs expressed. However, the optimum time is different for different cell types.
4) Clear the brain tissues or slices with SeeDB2 protocol.
5) Confocal imaging.
1) Color hue is not enough
Change the amount of TRE-XFP plasmids/viruses. When you cannot find intermediate color hues, increase the amount of TRE-XFPs. When all neurons looks similar, reduce the amount of TRE-XFPs.
2) Brightness is not sufficiently high
Most likely, tTA is too much. Reduce the amount of tTA plasmid/virus (DO NOT increase). Paradoxically, it is important to express a minimum amount of tTA.
3) Clearing is not enough with SeeDB2
When you use thick adult samples, we recommend pre-treatment with ScaleCUBIC-1 before clearing with SeeDB2.
4) Chromatic aberration
Chromatic aberation may occur when the refractive index of the clearing solution is inappropriate. When you use oil immersion lens (RI 1.518), use SeeDB2S. SeeDB2G (RI ~1.46) is recommended for Glycerol-immersion objective lenses. If you use Type-G immersion liquid, use 92% w/w SeeDB2G (Omnipaque350) for clearing.
5) Abnormal morphology
If you use Tetbow AAVs, the labeled neurons may start to show abnormal morphology, and eventually die, due to too much amount of XFPs. It is important to determine the optimum timing of sacrifice (2-4 weeks after infection).
Alternatively, the abnormal morphology may be introduced during clearing process. Abnormal morphology is often found after clearing with CLARITY, CUBIC, BABB etc. To minimize morphological damages, we recommend SeeDB2 with sufficient incubation time (2-3 days in total) and multiple concentration steps (more steps can be added, if necessary).
6) Photo-bleaching of chemical tags
This may occur when samples were cleared with SeeDB2. Some of alexa dyes are prone to photo-bleach in SeeDB2. Alternatively, you could chose TDE for clearing (but fluorescent proteins will be quenched in TDE).
Linear unmixing is powerful for separating multicolor channels. It's very easy and applicable to any types of fluorescent microscopy (including conventional confocal). Here is our ImageJ/Fiji plugin for linear unmixing.
For in-depth analysis of color combinations, see the QDyeFinder page.