On the cover!

Interdisciplinary Interdisciplinary The function of the brain depends on the flow of different information between various neurons in various areas of the brain.
Neural information in the brain needs to be transmitted through long-range projection of neural circuits
.

To understand the realization mechanism of brain function, the knowledge of the structure of the long-range connection of neural circuits is very necessary
.

In analyzing the function and structure of neural circuits in normal and pathological states, whole brain cell and subcellular resolution imaging plays an increasingly important role
.

In particular, the whole brain three-dimensional submicron high-contrast imaging of sparsely labeled neurons makes it possible to reconstruct the complete morphology of subsequent single neurons, especially the morphological reconstruction of long-range projection axon terminals.
At the same time, the rich structural knowledge also affects the nerves.
A more precise classification of yuan is of great benefit
.

Existing whole-brain micro-scale imaging methods, such as micro-optical tomography (MOST) and sequential two-photon imaging system (STP), generally take 1 to 2 weeks to complete the mouse whole brain ~0.
3 micron x 0.
3 micron x 1 Micron voxel imaging, while generating 10-20 TB of data, brings great challenges to subsequent storage, processing and analysis
.

In order to solve the above problems faster and more economically, Tsinghua University’s Guo Zengcai group developed two imaging systems, SMART and mLSFM, which achieved high-speed, high-resolution, and high-contrast mouse brain imaging to reconstruct the complete morphology of sparsely labeled neurons.

.

On October 1, the Tsinghua University School of Medicine, Tsinghua-Peking University Joint Center for Life Sciences, and Tsinghua-IDG/McGovern Institute for Brain Science Guo Zengcai's group published an online publication entitled "High-speed and high-resolution whole brain sparse imaging in Cell Reports Methods "Sparse imaging and reconstruction tomography for high-speed high-resolution whole-brain imaging" research paper, the report proposes a SMART (sparse imaging and reconstruction tomography) system that uses a turntable confocal unit for high-resolution rapid imaging
.

At the same time, the research was selected as the cover paper of the October issue of Cell Reports Methods
.

The October issue of Cell Reports Methods was launched on October 25.
▲Long press the picture to identify the QR code to read the original text.
The light path of the SMART system is mainly composed of the turntable confocal unit, sCMOS (scientific complementary-metal-oxide semiconductor) camera and 40x oil mirror ( NA 1.
3) composition, the size of each voxel is 0.
3 × 0.
3 × 1.
0 μm3
.

During the imaging process, the stepper motor controls the stage with the transparent brain sample to move continuously to achieve the acquisition of signals from each area; each time the data of one layer is collected, the vibrating slicer will receive instructions to remove the surface of the sample, and then proceed to the next step.
A cycle continues until the imaging of the whole brain is completed (Figure 1A)
.

Figure 1 The imaging principle of the SMART system.
Since the signal part occupies a very low proportion of the entire transparent brain sample, imaging all areas seems to be time-consuming and laborious but not very effective.
This is also the way the existing whole brain imaging technology improves the imaging speed.
A big obstacle
.

If only the signal part can be imaged, the imaging efficiency will be greatly improved
.

Based on this idea, the researchers added a low-power 16x objective lens for signal detection beside the high-power imaging objective lens
.

When imaging, the surface layer is scanned with a low-power objective lens, and then the picture is semi-automatically judged, and only the area judged to have a signal is imaged with a high-power objective lens, which greatly shortens the imaging time
.

According to estimates, this initiative reduces the area to be imaged, imaging time, and data scale to about 10% of the whole brain imaging (Figure 1 B and C)
.

In order to avoid signal omissions as much as possible, the researchers cleverly used the feature of "signal continuity", that is, all signal parts in the same neuron should be connected to each other, if no signal passes through the side of a cube , Then there is no signal in the cube
.

Based on this idea, the researcher will judge the side of the cube that has been formed, thereby further improving the reliability of this method (Figure 1D I-VII)
.

After calculation, the false negative result obtained by this imaging method is only 0.
13%, which is equivalent to only 9 surfaces with signals passing through more than 7,000 surfaces
.

The majority of these false negative weak signals are caused by the inevitable inhomogeneity of fluorescent signals.
Researchers will subsequently screen the neurons to be reconstructed to avoid signal incompleteness as much as possible
.

Figure 2 SMART system imaging effect display and neuron morphology analysis.
In addition to signal integrity, the sparse labeling strategy of mixed injection of Cre and Cre-dependent GFP virus is supplemented by high-performance signal acquisition equipment to obtain signals with high signal-to-noise ratio.
Whether it is near the cell body, the middle of the axon or the end of the axon, there are good image effects (Figure 2A)
.

In this study, the researchers injected the virus into the primary visual cortex (V1), mPFC, and anterior lateral motor cortex (ALM) and performed whole brain imaging
.

With the help of the existing software Elastix, ANTs, Vaa3d and original Matlab code, the researchers spliced ​​the imaging results and compared them with the standard brain
.

After manually tracing each neuron, a total of 29 complete neurons in different regions were obtained
.

By analyzing this part of neuron data, the researchers found that the distribution of dendrites and axons of neurons at mPFC, V1, and ALM are significantly different in all directions (Figure 2 BC)
.

The longest axon length and total axon length of neurons at mPFC are significantly higher than that at V1 and ALM (Figure 2 DE); the total axon length is linearly related to the number of branches (Figure 2F); in addition, the researchers It is pointed out that PT (pyramidal tract) and IT (intratelencephalic) neurons also have significant differences in dendritic morphology, local axon number and projection sites
.

In addition, in order to further increase the imaging speed to meet the increasing demand for high-throughput single-cell reconstruction, the researchers also updated the 40x objective lens to the 25x objective lens to ensure the image imaging quality while achieving high Fast imaging of flux (Figure 1 BC)
.

Researcher Guo Zengcai from Tsinghua University School of Medicine and Tsinghua-IDG/McGovern Institute for Brain Science is the corresponding author of the SMART imaging system
.

Tsinghua University School of Life Sciences, 2014 Ph.
D.
Chen Han, 2016 Master of School of Medicine Huang Tianyi, and 2018 Ph.
D.
student Yang Yuexin are the first authors; 2020 PhD student Yao Xiao, 2015 PhD student Huo Yan, 2016 PhD student Wang Yu, Class 2019 Doctoral student Wenyu Zhao, laboratory research assistant Ji Runan, and post-doctoral post-doctor Yang Hongjiang from the laboratory have jointly participated in the research on this subject
.

This research was supported by the National Natural Science Foundation of China (31871048) and the Tsinghua-IDG/McGovern Institute for Brain Science Brain+X Fund
.

Guo Zengcai's research group is supported by Tsinghua-Peking University Joint Center for Life Sciences and Tsinghua-IDG/McGovern Institute for Brain Science
.

The author introduces Guo Zengcai, researcher Guo Zengcai, researcher of Tsinghua University School of Medicine, Tsinghua-Peking University Joint Center for Life Sciences, IDG/McGovern Institute of Brain Science PI, PhD supervisor
.

He received a bachelor's degree in engineering mechanics from Tsinghua University in 2002, a bachelor's degree in solid mechanics from Tsinghua University in 2004, a doctorate in applied mathematics from Harvard University in 2010, and a postdoctoral fellow at Howard Seuss Institute Jennings Campus in 2015
.

Guo Zengcai is now focusing on the formation and maintenance of working memory, using advanced research tools in systems neurobiology to study the distribution of memory-related neural activities in the brain and its effects on cognitive behavior
.

A series of research results were published in international authoritative academic journals such as Nature, Nature Methods, Neuron (3 articles), and PNAS as the first or corresponding author
.

The research results of related paper information are published in the journal Cell Reports Methods under Cell Press.
Click "Read Full Text" or scan the QR code below to view the paper
.

▌Paper title: Sparse imaging and reconstruction tomography for high-speed high-resolution whole-brain imaging▌Paper URL: https:// 2▌DOI: https://doi.
org/10.
1016/j.
crmeth.
2021.
100089 Long press the picture to identify the QR code to read the original text.
In 1974, we published the first flagship journal "Cell"
.

Today, CellPress has developed into an international cutting-edge academic publishing house with more than 50 journals in the field of science
.

We firmly believe that the power of science will always benefit mankind
.

CellPress recommends reading the Cell Press journal to add new members.
These areas are coming soon! ▲