Probing dynamic processes in living cells with high time resolution, spatial precision, and chemical selectivity

Optical microscopy has been indispensable for cell biology. Many cellular processes, however, cannot be studied with existing imaging techniques because of insufficient chemical selectivity, spatial precision, or time resolution. Here, we present several experiments in which we develop and apply advanced optical microscopy to probe dynamic processes in living cells by overcoming the above difficulties.; Specifically, we have used coherent anti-Stokes Raman scattering (CARS) microscopy, a nonlinear, vibrational imaging modality, to study lipogenesis in fat and liver cells, processes associated with obesity and hepatitis, respectively. Lipids are difficult to visualize due to the lack of noninvasive labeling methods, but can be imaged with CARS microscopy in living, unstained cells with high chemical selectivity. With CARS, we monitored the entire fat cell differentiation process and identified a transient period during which lipid droplets (LDs) disappear in the cytoplasm, a previously unrecognized phenomenon. By combining CARS with two-photon fluorescence microscopy, we imaged LDs and hepatitis C (HCV) viral RNA simultaneously in liver cells, and established spatial and temporal correlations between lipogenesis and HCV RNA replication. Furthermore, under carefully optimized illumination conditions, we studied the active transport of LDs, a largely unknown process, in a variety of cell types. We show that the active transport of LDs is regulated, like most other cellular processes, and is correlated with lipid metabolism. These findings provide important information about the roles of lipids in various cellular processes, and demonstrate that CARS microscopy is a powerful tool for probing cellular dynamics.; In parallel, we have developed in vivo particle tracking assays with millisecond to microsecond time resolution and nanometer spatial precision to study the workings of the motor proteins kinesin and dynein that carry cargoes along microtubules. In vitro studies have shown that both motors advance with discrete steps as small as 8 nm in each ATPase enzymatic cycle. Observation of these steps is essential to dissecting their chemomechanical coupling mechanisms. The individual steps, however, are much more difficult to observe in vivo because of the high motor velocities arising from saturating cellular ATP concentration and involvement of multiple motors carrying the same cargo. By using a fast CCD camera to track the movements of vesicles containing endocytosed quantum dots, we observed stepwise movements in both the kinesin and dynein directions with ∼400 mus time resolution and 1.5 nm spatial precision. More importantly, we have achieved 25 mus time resolution and 1.5 nm spatial precision by tracking endocytosed gold nanoparticles (100-200 nm in diameter) with a quadrant photodiode in an objective-type dark field microscope. Individual steps of kinesin and dynein can now be resolved in the entire range of in vivo cargo velocities (0-8 mum/s). We show that while both kinesin and dynein take 8 nm steps in vivo, dynein also takes 12, 16, 20, and 24 nm steps in a living cell. Dynein also has variable step sizes when stepping backward. Interestingly, it takes smaller steps when carrying larger cargoes due to higher drag force, in both the forward and backward directions. These findings suggest that dynein functions like a car's transmission, consistent with a previously proposed 'gear' mechanism of dynein stepping. Lastly, useful information about the coordination among multiple motors can be inferred from our data. Our assays open up exciting new possibilities for studying molecular motors in living cells.