Trying to keep tabs on how an organism grows from a single fertilized egg into an embryo, cell by cell, is hard work. So hard in fact, that no one’s quite figured out how to do it.
Digital fruit fly embryo, reconstructed from live imaging data recorded with a SiMView light-sheet microscope. Each colored circle in the image shows one of the embryo’s cells, and the corresponding tail indicates that cell’s movement over a short time interval at around 3 hours post-fertilization
[Credit: Kristin Branson, Fernando Amat, Bill Lemon and Philipp Keller (HHMI/Janelia Research Campus)]
The problem, as researchers have lamented, is that there’s just too much happening—all at the same time—for the human eye to parse through all the data, even with the aid of the most powerful microscopes.
But now, scientists at the Howard Hughes Medical Institute (HHMI) have devised a high-tech shortcut: a new computational program that measures in real-time the three-dimensional development of each individual cell in a developing fetus.
This program stands to revolutionize how scientists understand the microscopic cellular ‘universe.’ As lead author, HHMI Group Leader Dr. Philipp Keller, explained in a news release:
“We wanted to reconstruct the elemental building plan of animals, tracking each cell from very early development until the late stages, so that we know everything that has happened in terms of cell movement and cell division.”
This technique, which is described in the latest issue of the journal Nature Methods, was built upon Keller’s 2012 development of a something called SiMView, a one-of-a-kind microscope that can capture precise 3D images of cells over a period of hours or even days.
But this was only the first step. Since the development of SiMView, Keller has been working on improving the system so that it could be used more broadly and over the course an organism’s development as an embryo. Specifically, Keller had sought to use this technique to look at how specific parts of the body develop—cell by cell. As Keller elaborated:
“In particular, we wanted to understand how the nervous system forms. Ultimately, we could like to collect the developmental history of every cell in the nervous system and link that information to the cell’s final function.”
In collecting and analyzing these vast datasets, researchers would then be well-poised to understand underlying molecular mechanisms of nervous system diseases.
Keller and his team have been looking for ways to both capture and analyze the vast amount of data hidden within each cell as it grows, matures and divides, with limited success—even the SiMView system was only active at a much smaller scale than what the team desired. One of the main issues is that as the cells in the embryo grow and divide, they become densely packed. They also shift around constantly, making tracking incredibly difficult to view.
The solution, Keller said, was to simplify the data. First, they clustered groups of 3D pixels called ‘voxels’ together into larger units, called ‘supervoxels.’ Next, they programmed the software to recognize the nuclei of each cell within the supervoxels. Then, using high-speed microscopy, they could capture images in a very quick sequence—so quick that individual cells wouldn’t be able to move out of the frame.
In this way, they are able to gather about 95% of all available data, a far higher number than that achieved by traditional methods. For the remaining 5%, the team employed even more complex algorithms to sort through the data. The end result, Keller says, is a wealth of knowledge that reveals more than many ever thought possible. According to Keller:
“You know the path, you know where it is at a certain time point. You know it divided from a certain point, you know the daughter cells, you know what mother cell it came from.”
In the early tests, the team studied the cellular ‘lineages’ of 295 early-stage nerve cells, called neuroblasts. Interestingly, they were not only able to trace these lineages in their entirety, but they could also predict their behavior later in their lifespan based on how they behaved early on.
The software, which is free and readily available to interested researchers, can be applied to a wide variety of data types—including different organisms and different microscopes.
This development stands to potentially become highly valuable to the stem cell research community. Increasingly, stem cell scientists are finding that in order to drive stem cells towards a desired adult tissue efficiently and completely, they need to try to recreate the stem cells’ natural environment. This should make it easier to build the right cellular “Neighborhood,” and help foster the transition from basic research into effective therapies.