Eric Betzig did not anticipate the experiment to work. Recently, two researchers, Shoh Asano and Ruixuan Gao, sought to use their team’s microscope on brain samples that were expanded four-fold their normal size—puffed up like balloons.
Part of Ed Boyden’s laboratory at the Massachusetts Institute of Technology (MIT), the duo utilizes a chemical method to make tiny specimens bigger so that researchers can view the molecular details more easily.
Although their method, known as expansion microscopy, appeared to work well on thin tissue sections or single cells imaged in traditional light microscopes, Boyden’s group sought to image much larger portions of tissue. They wanted to visualize the entire neural circuits that span millimeters or more. The researchers required a microscope that had excellent resolution and speed and is also comparatively gentle—something that does not damage a sample before they could complete the imaging process.
Hence, they sought Betzig, whose research group at the Howard Hughes Medical Institute’s Janelia Research Campus had applied the lattice light-sheet microscope to obtain a 3D image of the fast subcellular dynamics of delicate living cells. Integrating the two microscopy methods can possibly provide fast and comprehensive images of broad swaths of brain tissue.
“I thought they were full of it,” remembered Betzig. “The idea does sound a bit crude,” Gao stated. “We’re stretching tissues apart.” However, Asano and Gao were invited by Betzig to try out the lattice scope.
“I was going to show them,” laughed Betzig, but instead he was blown away. “I couldn’t believe the quality of the data I was seeing. You could have knocked me over with a feather.”
Currently, Betzig and his Janelia colleagues have partnered with Boyden’s team and imaged the sections of the mouse brain, the whole brain of a fruit fly, and the thickness of the cortex. The researchers’ integrated technique is not only fast but also provides high resolution with the potential to visualize any required protein.
It took only 62.5 hours to image the fly brain in various colors, as opposed to the years it would normally take if an electron microscope had been used, reported Betzig, Boyden, and their coworkers in the Science journal on January 17th, 2018.
“I can see us getting to the point of imaging at least 10 fly brains per day,” stated Betzig, who is presently an HHMI investigator at the University of California, Berkeley. He added that such resolution and speed will allow researchers to pose new questions, like how brains vary between females and males, or how brain circuits differ between the same types of flies.
Boyden’s team dreams of creating a highly comprehensive map of the brain that will allow scientists to simulate it on a computer. “We’ve crossed a threshold in imaging performance,” stated Boyden, who was chosen as an HHMI investigator in 2018. “That’s why we’re so excited. We’re not just scanning incrementally more brain tissue, we’re scanning entire brains.”
Expanding the brain
However, to create comprehensive maps of the brain, its wiring and activity have to be charted—the innumerable number of connections made by each of more than 80 billion neurons in humans. Such kind of maps may help researchers to identify where exactly the brain disease starts, or construct more improved artificial intelligence, or even describe behaviors. “That’s like the holy grail for neuroscience,” Boyden stated.
Decades ago, his team had a concept to find out how everything was arranged: What if the brain could be made bigger—sufficiently big to peer inside? The team infused samples with swellable gels—similar to the stuff in baby diapers—and eventually discovered a method to expand tissues, rendering the interior molecules less crowded and easier to view under a microscope. These molecules lock into a gel scaffold and maintain the same relative positions even after they are expanded.
Yet, it was difficult to image bulky tissue volumes. If a specimen becomes thicker, it becomes harder to illuminate only the potions that need to be visualized. If excess light is illuminated on the samples, it will photobleach them, burning out the fluorescent “bulbs” used by researchers to illuminate cells.
If a sample is expanded four times, it will increase its volume by 64 times, hence imaging speed also becomes crucial, Gao stated. “We needed something that was fast and didn’t have much photobleaching, and we knew there was a fantastic microscope at Janelia.” Gao added.
The advanced lattice light-sheet microscope images an ultrathin sheet of light through a specimen, lighting up only that part in the plane of focus of the microscope. That aids out-of-focus areas to remain dark, preserving the fluorescence of a specimen from being quenched.
When Asano and Gao initially tested their expanded mouse tissues on the lattice light-sheet microscope, they observed a thicket of glowing nubs sticking out from the branches of neurons. Called dendritic spines, the nubs usually resemble mushrooms, featuring bulbous heads on skinny necks that can be difficult to determine. However, Asano informed that the researchers were able to view even “the smallest necks possible,”, and, at the same time, simultaneously imaged the adjacent synaptic proteins.
“It was incredibly impressive,” stated Betzig. The group was convinced that they should further investigate the combined method. “And that’s what we’ve been doing ever since,” he stated.
The brain and beyond
In the last couple of years, Asano and Gao had spent a number of months at Janelia, collaborating with physicists, computer scientists¸ microscopists, and biologists across the campus to capture and study images. “This is like an Avengers-level collaboration,” Gao stated, referring to the group of comic book superheroes.
High-quality fly brain specimens provided by Yoshinori Aso and the FlyLight team were expanded and used by Asano and Gao to obtain some 50,000 cubes of data across individual brains—creating a kind of 3D jigsaw puzzle. Those images needed complex computational stitching to put back the pieces together—a study headed by Igor Pisarev and Stephan Saalfeld. “Stephen and Igor saved our bacon,” Betzig stated. “They dealt with all the horrible little details of image processing and made it work on each multi-terabyte data set.”
As a next step, the paper’s co-first author Srigokul Upadhyayula from Harvard Medical School and Boston Children’s Hospital examined the combined 200 terabytes of data and developed the incredible movies that reveal the intricacies of the brain in bright colors. He and his coauthors analyzed over 1,500 dendritic spines, imaged fatty sheaths insulating mouse nerve cells, emphasized all of the dopaminergic neurons, and calculated all the synapses across the entire brain of the fly.
The nuances of the expansion technique developed by Boyden’s team make it perfectly suitable for the lattice light-sheet microscope; the method creates almost transparent samples. For the microscope, it is nearly like looking through water, instead of a muddy sea of molecular gunk. “The result is that we get crystal clear images at blazingly fast speeds over very large volumes compared to earlier microscopy techniques,” stated Boyden.
Nevertheless, challenges continue to be there. As with any type of super-resolution fluorescence microscopy, it can be difficult to decorate proteins with sufficient fluorescent bulbs to clearly view them at high resolution, Betzig said, Since a number of processing steps are involved in expansion microscopy, artifacts are likely to be introduced. In view of this, “we worked very hard to validate what we’ve done, and others would be well advised to do the same.” he stated.
At present, Gao and the Janelia team are constructing a novel lattice light-sheet microscope, which they have planned to shift to Boyden’s laboratory at MIT. “Our hope is to rapidly make maps of entire nervous systems,” concluded Boyden.
Scientists mapped the location of all synapses – over 40 million – across an adult fruit fly brain. A half million colored balls represent synapses associated with dopaminergic neurons. (Video credit: Gao et al./Science 2019)