Hari Shroff is an early bird. When he was 13 years old, Shroff enrolled at the University of Washington, where he earned a B.S.E. in bioengineering. In 2006, at the age of 24, he completed his Ph.D. in biophysics at the University of California at Berkeley. Since then, Shroff has explored the frontiers of super-resolution microscopy, developing 3-dimensional photoactivated localization microscopy (PALM) techniques alongside the technique’s inventor Eric Betzig, and just last month, President Barack Obama presented Shroff with the Presidential Early Career Award for Scientists and Engineers.
“There’s a big question about how these actually move inside the developing embryo and how they find their targets,” says Shroff. “So what we really need to do is image these things, but the problem is that the microscopes we use either kill the embryos, or they’re too phototoxic or they’re too slow.”
But now, working with neurobiologist Daniel Colón-Ramos from Yale University and computer scientist and developmental biologist Zhirong Bao from Sloan Kettering Memorial Cancer Center, Shroff’s team has developed a microscopy technique that overcomes these limitations; a technique called inverted selective plane illumination microscopy (iSPIM). With this new method, neurobiologists may finally catch their first glimpse of the sequence of neurodevelopment in live C. elegans embryos.
Worm’s Eye View
During the 1970s, South African biologist Sydney Brenner believed that C. elegans was the perfect model for the developmental biology studies (1). His reasons were simple: the organism is relatively simple, consisting of only 959 somatic cells; it’s easy to grow, as self-inseminated hermaphrodites can produce over 300 eggs; and it’s amenable to genetic analysis, with a relatively small, 1 million-base pair genome.
Since then, thousands of biologists have joined the C. elegans community, recognizing additional useful features of the organism for laboratory studies. It is small, about 1 mm in length, so hundreds of C. elegans can fit in a single Petri dish. It only takes about 4 days to generate a generation that lives about 2–3 weeks, making experimental cycles short. Furthermore, C. elegans remains viable after freezing and thawing, making it easy to store in the laboratory. Finally, C. elegans is transparent, making imaging of the organism’s features particularly easy with optical and fluorescent microscopy.
“In C. elegans, one of the things the science community has identified is promoters that can be used to label single cells or single tissues. Using those promoters, they can drive the expression of fluorescent markers and visualize the cell biology and developmental biology with single-cell resolution,” says Colón-Ramos.
For neurobiologists, Brenner proposed that the C. elegans nervous system represents an excellent model for eukaryotes. In C. elegans, the nervous system consists of 302 neurons that are arranged in a structure that rarely varies in the species. Furthermore, the nervous system contains many of the same types of connections—including chemical synapses, neuromuscular junctions, and gap junctions—found in higher organisms.
Earlier this year, a team of researchers from several U.S. institutions developed a method to visualize the chemical synapse network of a C. elegans (2) by combining new electron micrographs with Brenner’s network analysis. Such diagrams can provide insights into how the nervous system responds to specific stimuli. Over the years, C. elegans has provided other significant insights into the nervous system as well.
“And yet, if you want to ask the question, where are these neurons as a function of time, how do they find each other? That’s really the question that hasn’t been answered,” says Shroff.
Getting in Deep
Conventional microscopy techniques—such as epiflourescent, confocal, or multiphoton microscopy—are not quick enough to image C. elegans embryos in four dimensions, longitudinally, over long periods of time. This becomes a major obstacle when you’re studying the development of an organism.
This obstacle stems from a number of factors. First of all, C. elegans embryos are not flat; they are three dimensional and require a technique that can provide clear images at different depths. So illuminating the entire sample at once, as you would in an epiflorescence or confocal microscope, provides a blurred image, because the microscope captures light above and below the focal plane. What you really want is to record the light from one particular plane.
Furthermore, there is a delicate balance between the intensity of the illumination and the embryo’s viability. Embryos are sensitive to light, so shining too much intense light will cause them to die. Obviously, this becomes another problem when attempting to image development over a long period of time.
Finally, most methods are too slow to capture clear images of C. elegans embryos since they twitch and move rapidly during their development. Spinning disc and point scanning confocal microscopy are not rapid enough to capture a clear image of the embryos while they are twitching.
To overcome these obstacles, Shroff and Colón-Ramos looked through the existing microscopy tool box and were particularly intrigued by a technique developed nearly two decades ago and has had a recent renaissance: SPIM.
In 1993, researchers from Shroff’s alma mater, the University of Washington, developed a technique called orthogonal-plane fluorescence optical sectioning (OPFOS), which limited sample illumination to the narrow plane that was being observed (3). An image is taken of each focal depth, and by moving the sheet through the sample and capturing an image per position, researchers can build up a volume of that sample. The instrument that they developed produced focused images of an intact guinea-pig cochlea.
Subsequently in 2004, Ernest Stelzer’s group at the European Molecular Biology Laboratory further developed SPIM from OPFOS to minimize phototoxicity and increase image acquisition speed (4). In their 2004 paper published in Science, Stelzer and colleagues reported the imaging of Drosophila melanogaster embryogenesis in vivo.
“We collect all the signals in the light sheet, and then we reduce the photobleaching and photodamage,” says Yicong Wu, a research fellow in Shroff’s group. “It’s already in focus, so we do not produce any out of focus signals. So, there is reduced photobleaching.”
While SPIM seemed like an excellent technique to image C. elegans neurodevelopment, the technique typically requires mounting the sample in a specialized housing, which has limited the availability of the technique to a few labs. So Shroff and colleagues sought a way of adding SPIM on a commercially available inverted microscope base, thus broadening the applicability of the technique. This had the additional benefit of making it trivial to mount C. elegans embryos.
In a paper published in PNAS (5), Shroff and his team reported the development of the iSPIM technique for high-speed volumetric imaging in developmental and neurobiology studies. iSPIM was developed by using two objectives: excitation and a detection. The mirrors, filters, lenses and a camera capture a focused image with objectives at 90 degree orientation. It was modified from SPIM by replacing the illumination pillar of the inverted microscope and using automated lineaging techniques for neural dynamics. Using the iSPIM technique, the team imaged C. elegans embryos every 2 s for a 14-h period without detectable phototoxicity. “We can image the embryo for the first time in the history of the animal continuously and in vivo from the moment it’s a zygote to until it hatches,” says Colón-Ramos.
But there are still limitations with iSPIM. For example, axial resolution suffers at high-speed image acquisition rates. Axial resolution in the z-axis is less than it is for confocal, and it is the result of using Gaussian beams to make the light sheet, which undergo diffraction at increasing distances. However, Shroff does believe there are ways to overcome these limitations. For instance, Bessel beams could be a solution but may cause out-of-plane illumination.
Commercialization has also been a hot topic for the iSPIM, which could be an add-on module for any typical inverted microscope and have multiple applications. “We did applications in zebrafish, viral infection of cells. You can image faster, so there’s a lot of data, and you could automatically photoanalyze the image,” says Wu.
Currently, the researchers are collaborating on a 4-dimensional movie, on how the nervous system develops in an embryo in three dimensions and across time. They hope this could provide insight into how neurodevelopment occurs in higher organisms.
“Nobody has imaged the development of an intact nervous system continuous and the in vivo, and we think the worm embryo is a good target,” says Shroff. “So the idea would be to make some sort of Google Worm. You can go in from this digital data set you’ve derived from these microscopes, and you can follow a neuron as it finds its partner in a space and in time.”
- Brenner, S. 1974. The genetics of Caenorhabditis elegans. Genetics. 77:71-94.
- Varshney, L.R., B.L. Chen, E. Paniagua, D.H. Hall, and D.B. Chklovskii. 2011. Structural properties of the Caenorhabditis elegans neuronal network. PLoS Comput. Biol. 7:e1001066.
- Voie, A.H., D.H. Burns, and F.A. Spelman. 1993. Orthogonal-plane fluorescence optical sectioning: three-dimensional imaging of macroscopic biological specimens. J. Microsc. 170:229-236.
- Huisken, J., J. Swoger, F. Del Bene, J. Wittbrodt, and E.H. Stelzer. 2004. Optical sectioning deep inside live embryos by selective plane illumination microscopy. Science 305:1007-1009.
- Wu, Y., A. Ghitani, R. Christensen, A. Santella, Z. Du, G. Rondeau, Z. Bao, D. Colón-Ramos, and H. Shroff. 2011. Inverted selective plane illumination microscopy (iSPIM) enables coupled cell identity lineaging and neurodevelopmental imaging in Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 108:17708-17713.