Fluorescent sensors, be they proteins or small molecules, are extremely useful because they can be used to detect metabolic states and protein interactions in living cells. Fluorescent proteins are particularly useful because they can be produced inside the cell and, using tags, targeted to specific proteins, locations and organelles quite easily. Because of this, a large number of fluorescent proteins have been isolated and engineered, usually using the backbone of the green fluorescent protein. In this week’s Proceedings of the National Academy of Sciences, a group from Columbia University describe a fluorescent protein that can sense the viscosity of its surroundings (1).Kao et al. were working with a protein called Dronpa. Isolated from coral, Dronpa is structurally homologous to GFP, but has unique photoswitching characteristics (2). Irradiation of Dronpa at 488 nm causes fluorescent emission at 518 nm, but the fluorescence is rapidly quenched, and emission ceases.
Dronpa can be restored from this ‘dark state’ by hitting it with light at 405 nm, completely recovering its fluorescence. A pulse of light at 405 nm while also illuminating at 488 nm causes a rapid spike in fluorescence, followed by a slower decay over several milliseconds as the bright state quenches. Kao et al. decided to look further into Dronpa’s photoswitching behavior by examining the kinetics of this process.
When they examined the decay rates, they noticed something interesting: the rate of quenching seemed to depend on the viscosity of the solvent, which they controlled by varying the percentage of glycerol in their measurement buffer. Unfortunately, as you can see below, the distributions overlapped significantly, meaning that this protein would not be a particularly good viscosity sensor.
This prompted Kao et al. to ask if they could do better. The magnitude of the changes in rates that they saw suggested that viscosity-related changes in the photoswitching rate were responsible for the effect. This makes sense, because the structural models suggest that certain translational motions are needed to switch the chromophore from its light state to its dark state. These motions would be subject to drag from the surrounding liquid, which increases with viscosity.
An alternative possibility not discussed by Kao et al. is that the internal fluctuations of the protein are directly coupled to solvent dynamics. This coupling, called ‘solvent slaving’ (3), has long been suggested by Hans Frauenfelder based on his work in myoglobin.
In either case, if protein flexibility is the key to the photoswitching rate, then a more flexible protein might be more sensitive to viscosity. As it turns out, some flexible mutants of Dronpa already exist. Specifically, Dronpa-3 has been engineered to have both a steric clash and a void, and has a reduced quantum yield that is consistent with a more flexible interior.
When Kao et al. repeated their viscosity experiments using Dronpa-3, they found that it had substantially better measurement characteristics. The decay rates still had fairly broad distributions, but were much better separated (see below). Also, consistent with their hypothesis that protein flexibility contributed significantly to the switching rate, the decays were much faster overall (compare the scales). The response is not linear over the entire viscosity range, but it still seems that Dronpa-3 could produce relatively sensitive measurements.
In order to test that idea, the authors expressed Dronpa-3 alone, and as a fusion with a histone protein, in HEK 293T cells. In these experiments, the researchers were able to measure local viscosity in live cells, both during stable phases and mitosis. The results suggested, not surprisingly, that the nucleus is more viscous than the cytoplasm, though the environment seemed to be more heterogeneous once chromatin had condensed for mitosis. All this is more or less as expected, and in some cases cross-validated by other experiments. In the future, Dronpa-3 may be useful for examining solution dynamics during other processes that reshape cells and tissue or depend on molecular diffusion, as well as in calibrating in vitro experiments to better reflect the relevant biological environments.
1. Kao, Y., Zhu, X., & Min, W. (2012). Protein-flexibility mediated coupling between photoswitching kinetics and surrounding viscosity of a photochromic fluorescent protein Proceedings of the National Academy of Sciences, 109 (9), 3220-3225 DOI: 10.1073/pnas.1115311109
2. Ando, R., Mizuno, H., & Miyawaki, A. (2004). Regulated Fast Nucleocytoplasmic Shuttling Observed by Reversible Protein Highlighting Science, 306 (5700), 1370-1373 DOI: 10.1126/science.1102506
3. Fenimore, P.W., Frauenfelder, H., McMahon, B.H., & Parak, F.G. (2002). Slaving: Solvent fluctuations dominate protein dynamics and functions Proceedings of the National Academy of Sciences, 99 (25), 16047-16051 DOI: 10.1073/pnas.212637899