3D "super-resolution" images show tiny structures in living mouse brains
Researchers have adapted an advanced microscopy technique to take “super-resolution” 3D images inside the brains of living mice. The method is so precise the team was able to image the tiny twigs on the branches of neurons, and could watch how they changed over the course of a few days.
In tests, the team was able to snap three-dimensional images at very high resolution relatively deep in the animal’s tissue – 76 micrometers deep in a living mouse brain, and 164 micrometers deep in a tissue sample. The method was able to image some incredibly tiny cellular structures, called dendric spines. If a neuron is a tree, the dendrites are the branches that reach out to others, and dendric spines are the twigs protruding from those.
“Our microscope is the first instrument in the world to achieve 3D STED super-resolution deep inside a living animal,” says Joerg Bewersdorf, lead researcher on the study. “Such advances in deep-tissue imaging technology will allow researchers to directly visualize subcellular structures and dynamics in their native tissue environment."
The new technique builds on a technology called stimulated emission depletion (STED) microscopy, which was developed in the 1990s and won scientist Stefan Hell the Nobel Prize in Chemistry in 2014. This method allowed optical microscopes to bypass a physical size limit in imaging, by making nano-scale objects fluoresce.
STED microscopy is usually done with molecules that glow when excited by a laser. A target sample is dyed with these molecules, then two lasers are swept over them – the first excites all of the molecules, while the second causes the larger molecules to discharge their energy and stop glowing. The end result is that only the smallest molecules are still fluorescing, allowing for an image with extremely fine detail. In recent years, scientists even managed to adapt the technique to produce three-dimensional images.
The problem is, 3D STED imaging has so far only really worked in thin samples. That’s because the laser light has a hard time getting through too much tissue to reach the molecules and excite them. So for the new study, the researchers overcame this limitation by combining 3D STED imaging with another technique called two-photon excitation (2PE).
“2PE enables imaging deeper in tissue by using near-infrared wavelengths rather than visible light,” says Mary Grace Velasco, first author of the study. “Infrared light is less susceptible to scattering and, therefore, is better able to penetrate deep into the tissue.”
The end result is a technology the team calls 3D-2PE-STED imaging. To improve the image even further, the team used adaptive optics as well, which corrects aberrations or distortions of the light shape that can be produced when imaging through tissue.
“During imaging, the adaptive element modifies the light wavefront in the exact opposite way that the tissue in the specimen does,” says Velasco. “The aberrations from the adaptive element, therefore, cancel out the aberrations from the tissue, creating ideal imaging conditions that allow the STED super-resolution capabilities to be recovered in all three dimensions.”
The technology performed admirably in tests. The first experiments were conducted in cultured cells, where the 3D-2PE-STED imaging was able to reveal details 10 times smaller than using 2PE alone.
In tests on living mice, the researchers were able to zoom right in on the dendric spines, revealing their 3D structure in great detail. They were even able to show natural differences in the structure when imaging the same region three days later.
“Dendritic spines are so small that without super-resolution it is difficult to visualize their exact 3D shape, let alone any changes to this shape over time,” says Velasco. “3D-2PE-STED now provides the means to observe these changes and to do so not only in the superficial layers of the brain, but also deeper inside, where more of the interesting connections happen.”
While the team says they didn’t see any signs of damage caused by the technique in either the structure of the neurons or the behavior of the mice, they say that further study will be required to ensure its safety.
It also could use some tweaking before it can be used to image human tissue – in this version the fluorescent molecules need to be directly “painted” onto the target cells, which could cancel out the otherwise non-invasive nature of the technology. Future advances, the team says, could involve injectable dyes.
The research was published in the journal Optica. The 3D image produced by the method can be seen in the video below.
Source: The Optical Society