Long Cai is on a mission to develop tools to look at things that have never been seen before. Last year, the up-and-coming California Institute of Technology (Caltech) physical chemist received a New Innovator Award from the National Institutes of Health, and he's putting it to good use by applying genomics and proteomics to single cells using a technique called super-resolution barcoding. "The overarching goal of the project is to turn single cells into microarrays," says Cai.
Using this barcoding strategy, Cai and Lubeck increased the number of molecules that researchers could identify and quantify simultaneously in a cell. Moreover, this method involves direct imaging of samples, so it preserves spatial information about molecules within cells as well as cellular contacts in tissues, allowing scientists to examine complex patterns of signaling interactions among cells in a tissue.
With conventional fluorescence microscopy, it's difficult to resolve individual molecules and therefore transcript levels must be low for barcoding to be useful. By contrast, super-resolution microscopy makes it possible to resolve cellular structures as small as 10–20 nm, which translates into about 108 voxels for a typical yeast cell. Thus, it's possible to resolve all of the genomic messages in a cell by labeling each message with a distinct fluorescent molecular barcode. Super-resolution microscopy uses the sparse activation of subsets of overlapping fluorophores, which is necessary for high-density barcoding when signals from multiple transcripts overlap. This process is similar to colored lights on a Christmas tree that take turns blinking on and off, and it enables researchers to isolate individual spots of light.
"One of the main motivations for going into this was that we realized that for a lot of things that you want to do in single cells, you also have to know what you're looking for in the first place to tag [green fluorescent proteins (GFPs)] to them, and if you want to go beyond that and look at multiple genes at the same time, it's actually quite hard to do with GFP because there aren't that many colors of GFP available," says Cai.
For Cai, the path toward super-resolution barcoding emerged naturally from his past research experiences. His interest in microscopy was apparent even before he started college. During the summer of his senior year of high school, Cai studied diffractive optics in a research lab at the State University of New York at Stony Brook. After studying chemistry and physics as an undergraduate at Harvard, he stayed at the university to pursue a Ph.D. as a student in the lab of single-molecule enzymology researcher Sunney Xie. Using single-molecule spectroscopy, they examined the molecular mechanisms underlying changes in gene expression insingle bacterial cells, publishing high-profile papers in Science and Nature (2–3).
After earning his Ph.D. in 2006, Cai came to Caltech as a Beckman Institute fellow in the lab of Michael Elowitz, an expert in systems and synthetic biology. Together, they studied signaling in single yeast cells using time-lapse microscopy. In 2008, they reported in Nature their discovery that a transcription factor called Crz1 localizes to the nucleus in short bursts whose frequency is controlled by extracellular calcium concentrations (4). This cellular phenomenon could allow transcription factors to coordinate the expression of multiple genes in response to external signals.
"Two years ago, when I started my lab here at Caltech, I started thinking about how to use these single-molecule techniques to do higher-throughput systems biology in single cells," says Cai. Since then, he has focused on developing super-resolution barcoding, which represents the marriage of two diverse approaches: genomics, which is used to simultaneously analyze all genes and proteins in a system but produces an average across many cells, and single-cell biology, which enables the high-precision tracking of gene expression in individual cells in their native environments, but has been limited to the study of only a few genes at a time.
"You can see the natural progression he's taken, from when he was graduate student in Sunney's lab, where he was detecting single-molecule gene expression in bacteria, to when he did a postdoc with Michael Elowitz working in yeast," says Michael Rust, a systems biologist at the University of Chicago who helped to develop the super-resolution imaging technique used by Cai—known as stochastic optical reconstruction microscopy (STORM)—as a graduate student in Xiaowei Zhuang's lab (5). Rust knew Cai when they were both graduate students at Harvard. "He's a very creative scientist, and he's had a string of very impressive work from multiple labs and now from his own lab."
Reaching Full Potential
Eventually, Cai's approach could allow scientists to explore a wider range of questions, not only about the basics of transcriptional patterns in individual cells, but also about how complex patterns of gene expression vary across cells and how this heterogeneity relates to the determination of cell fates during development as well as the origin of disease states such as cancer, bacterial infections, and even autism. In future studies, Cai will use his approach to compare gene expression across generations of yeast cells to try to understand more about the genetic circuits involved in the aging process, and his lab is working on expanding the throughput of the technique to more than 100 genes and applying it to multi-cellular organisms to study developmental processes.
"I think his approach is very exciting," says Sanjay Tyagi, a cell biologist at the University of Medicine and Dentistry of New Jersey who has developed probes to image mRNA single molecules in cells. "As these single-molecule approaches are being used, we are discovering that each cell is very different from the average, and there's a new biology hidden in individual cells that we were completely ignoring. There's this whole field of heterogeneity in gene expression that has come up because it has become possible to do single-molecule imaging."
Ultimately, this work may have important clinical implications. "If you could do this on a single-cell level, and even in a tissue, then you would have potentially a very powerful approach to do disease monitoring and diagnosis. So, I think that the technique has a lot of potential, but there's a lot of technology that has to be developed before this becomes feasible," says Robert Singer, a cell biologist at Albert Einstein College of Medicine, who one decade ago developed a FISH-based barcoding approach for simultaneously detecting multiple mRNAs inside single cells (6). (In contrast to Singer's method, Cai's approach directly barcodes single mRNAs.)
Moving forward, the power of Cai's approach can be more fully harnessed as microscopes become customized for this application and as fluorescent dyes improve. Currently, dyes are limited in the number of photons they emit and the accuracy they provide. "The probes that people have been using since these techniques were created were really ones that people stumbled upon accidentally,” says Rust. “So, I think we have yet to see the first generation of probes that have really been engineered specifically to be used this way. And I think when we get to that point, then we'll really know how well these things can work." Rust is confident that these problems will be worked out in the future. "There's no fundamental physical principle that prevents these dyes from being better," he says.
But the application of this method to mammalian cells may not be so straight-forward. One of Cai's two approaches for barcoding involves physically compressing cells, so some 3D information is lost, and this could pose a greater problem in human cells because of their relatively large size, explains Tyagi. Moreover, Cai's method was implemented in fixed cells, and it would difficult to apply it to live cells, which would allow for real-time monitoring of gene expression. Another hurdle associated with this approach includes the substantial computational requirements for the sophisticated image analysis.
"To do super-resolution microscopy is tremendously labor-intensive at the present time, so it's not really very practical to do at this point. So, what's necessary is some much higher-throughput super-resolution technology," says Singer. But that advance may not be too far off. "I think once in a while there's a breakthrough—like this particular one of Cai's—that allows you to think about leaping to the next level."
- Lubeck, E., and L. Cai. 2012. Single-cell systems biology by super-resolution imaging and combinatorial labeling. Nature Methods9(7):743-8.
- Choi, P. J., L. Cai, K. Frieda, and X. S. Xie. 2008. A stochastic Single-Molecule event triggers phenotype switching of a bacterial cell. Science 322(5900):442-446.
- Cai, L., N. Friedman, and X. S. Xie. 2006. Stochastic protein expression in individual cells at the single molecule level. Nature 440(7082):358-362.
- Cai, L., C. K. Dalal, and M. B. Elowitz. 2008. Frequency-modulated nuclear localization bursts coordinate gene regulation. Nature 455(7212):485-490.
- Rust, M. J., M. Bates and X. Zhuang. 2006. Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM). Nature Methods 3:793-796.
- Levsky, J. M., S. M. Shenoy, R. C. Pezo, and R. H. Singer. 2002. Single-cell gene expression profiling. Science 297(5582):836-40.