Seeing Inside Cells

Fluorescent proteins that glow in different colors have revolutionized studies of protein-based processes in cells and whole organisms in recent years. But they’re not perfect, and they don’t do everything. So researchers are trying to develop new tools and approaches for visualizing the locations, interactions, and functions of biomolecules, complexes, and structures in cells.

Fluorescent proteins are relatively large and tend to oligomerize, factors that sometimes cause unwanted effects on the location, activity, and other properties of the proteins being studied. And they can’t be used to image nonprotein biomolecules such as RNA and carbohydrates.

Scientists who are developing new imaging and labeling strategies to address such limitations convened earlier this month at a Division of Biological Chemistry session at the ACS national meeting.

At the symposium, grad student Jingyi Fei of Ruben L. Gonzalez Jr.’s group at Columbia University described visualization experiments that reveal new details about the way the ribosome synthesizes proteins by translating sequence information encoded in messenger RNA.

Gonzalez and coworkers label ribosomal proteins and transfer RNAs (tRNAs) with small-molecule fluorophores. They then monitor those fluorescent groups in real time using single-molecule Förster resonance energy transfer (FRET), a technique that is sensitive to the relative distance and orientation of two fluorophores.

Gonzalez and coworkers showed that the ribosome reversibly fluctuates between two global conformational states, GS1 and GS2. The transitions involve coupled movements of a ribosomal domain (the L1 stalk) and ribosome-bound tRNAs, accompanied by a ratcheting motion between ribosome subunits. They also showed that translation factors play key roles in regulating ribosome dynamics during protein synthesis (Nat. Struct. Mol. Biol. 2009, 16, 861; Proc. Natl. Acad. Sci. USA, DOI: 10.1073/pnas.0908077106).

Whereas Gonzalez’ group labels proteins and RNAs, chemistry professor Carolyn R. Bertozzi and coworkers at the University of California, Berkeley, label carbohydrates. The glycome, the totality of a cell’s glycans, is “a dynamic indicator of a cell’s physiological state,” Bertozzi said. “We could learn something about the state of the cell if we could decode that ­information.”

Glycome change has been found to occur, for example, when embryonic stem cells mature, organs develop, and cells become cancerous. “If we could visualize the glycome in the context of a living organism, perhaps we could study those physiological changes in vivo and even use them to diagnose human disease,” Bertozzi said.

She and her coworkers have developed chemical reactions that make this possible. In the approach, azidosugars are incorporated metabolically into cell-surface glycans. The azides serve as chemical handles for covalent attachment of fluorescent dyes by reaction with cyclooctynes that are introduced into the cells. They have used this strategy to image glycome changes during zebrafish embryonic development.

“Now, we can image glycans in these early embryos to try to get a sense of how the glycome is changing during these very early cell-fate decision-making events,” Bertozzi said. “While we’re focusing on sugars in my laboratory, other groups have already made progress in using the azide as a chemical reporter for lipid groups and even for nucleic acid groups. So hopefully there will be a rich future in chemical biology for this very interesting functional group.”

At Harvard University, meanwhile, professor of chemistry and chemical biology Xiaowei Zhuang and coworkers are exploiting the photophysical properties of fluorophores to image biomolecules at superhigh spatial resolution in studies on the way viruses enter cells. Conventional light microscopy is limited to a resolution of a few hundred nanometers—insufficient to resolve biomolecules. To break this diffraction limit, Zhuang’s group and two others came up with a new approach, which Zhuang calls stochastic optical reconstruction microscopy (STORM).

In STORM, photoswitchable fluorescent probes are used to obtain time-resolved information that is used to improve the resolution of fluorescent imaging data to the range of tens of nanometers. Zhuang and coworkers also recently developed 3-D STORM, which translates flat STORM images into three dimensions, making it possible to better resolve the morphology of cellular structures. Zhuang hopes to continue to improve the resolution of STORM to image smaller biomolecules.

Chemistry professor Alanna Sche­partz of Yale University is also trying to use small-molecule fluorophores to image complex biomolecular events—protein interactions, conformational changes, and posttranslational modifications.

Her group uses bipartite tetracysteine display, in which recombinant proteins are labeled selectively, in vitro or in live cells, with profluorescent biarsenicals, compounds that become fluorescent when they bind to the proteins.

“Bipartite tetracysteine display supports fluorescence-based differentiation between folded and unfolded protein states and can detect protein-protein interactions in live cells,” Schepartz said. She and her coworkers are currently developing a bipartite tetracysteine display-based assay to identify small molecules capable of refolding mutated versions of p53, which is mutationally inactivated in about half of human tumors.

Schepartz and coworkers are also developing nontoxic bis-boronic acid dyes as alternatives to biarsenicals. So far they’ve shown that the dyes can label proteins and glycoproteins in mammalian cells.

“Fluorescence has revolutionized cell biology,” Schepartz said. “Its positive impact on molecular medicine will only continue to deepen and expand with time.”—Stu Borman