Research

Synthetic Cell Biology

Complexity in signaling networks is often derived from co-opting one set of molecules for multiple operations. Understanding how cells achieve such sophisticated processing using a finite set of molecules within a confined space –what we call the “signaling paradox”- is critical to biology and engineering as well as the emerging field of synthetic biology. We have recently developed a series of chemical-molecular tools that allow for inducible, quick-onset and specific perturbation of various signaling molecules. Using this novel technique in conjunction with fluorescence imaging, microfabricated devices, quantitative analysis and computational modeling, we are dissecting intricate signaling networks. In particular, we investigate positive-feedback mechanisms underlying the initiation of neutrophil chemotaxis (known as symmetry breaking), as well as spatio-temporally compartmentalized signaling of Ras and membrane lipids such as phosphoinositides. In parallel, we also try to understand how cell morphology affects biochemical pathways inside cells. Ultimately, we will generate completely orthogonal machinery in cells to achieve existing, as well as novel, cellular functions. Our synthetic, multidisciplinary approach will elucidate the signaling paradox created by nature.


Schematic of AND gate employing rapamycin and gibberellin chemical dimerization systems (Miyamoto et al. Nat Chem Biol, 2012).

In the creation of non-silicon-based computers, various biomolecules have been used to construct Boolean logic gates. However, their ‘computational’ timescale is relatively slow (tens of minutes to hours), as most of these biological logic gates utilize gene expression that requires time-consuming processes such as transcription and translation. For faster processing, we employed chemically inducible protein dimerization systems. To rapidly trigger two chemically inducible signals, we developed an efficient chemical dimerization system using a newly synthesized analog of the plant hormone gibberellin (GA3-AM) and its binding proteins, a system that is completely orthogonal to rapamycin-mediated protein dimerization (Movie #1). With the two chemical inputs (rapamycin and GA3-AM), we achieved AND as well as OR gates, that produced output signals such as fluorescence and membrane ruffling on a timescale of seconds. The use of two orthogonal dimerization systems in the same cell also allows for finer modulation of protein perturbations than is possible with a single dimerizer. We are now working to create more complex logic gates as well as multiplexed logic gates for further complex computations.

Graded activation of Rac with microfluidcs and rapamycin directs cellular polarity (Lin et al. PNAS, 2012).

Polarized cells possess intracellular gradients of active Rho GTPases which serve to properly localize and direct cytoskeletal machinery for optimal motility. The pathways regulating the distribution and activity of the Rho GTPases are still being unraveled and at a fundamental level, it is unknown whether intracellular gradients of the Rho GTPases themselves are sufficient to drive polarity and motility (Movie #2). This is an interesting question, especially because we have previously shown that the neutrophils can polarize with uniform PI3K activation (Movie #3), but not with that of Rac. Thus, we have combined two technologies, chemically-inducible molecular probes and microfluidics, to present cells with a novel perturbation- an induced linear gradient of active, endogenous Rac without receptor actuation. We found that the chemical inducer gradient was sufficient to direct cells towards the chemical source, regardless of their initial direction of polarization or lack thereof (Movie #4). We created a simplified mathematical model of a spatial network of Rho GTPases to model this perturbation and found that the model was fully predictive of the data observed. Consistent with the model prediction, we found that the initial chemotactic response time was primarily dependent upon the steepness of the Rac activity gradient but not the concentration of the active Rac. Our joint microfluidics and chemical activation methodology to direct cells with shallow spatial gradients has great potential utility in further studies of cell polarity and motility. We are now expanding the studies to other Rho GTPases as well as their upstream regulators to systematically map out core components in directed cell migration.

Schematic illustration of two techniques to rapidly manipulate PI(4,5)P2 by chemically inducible dimerization (Ueno et al. Sci Signaling, 2011).

Phosphoinositide membrane lipids regulate diverse cellular functions including proliferation, differentiation, and migration. We have previously developed chemically-inducible dimerization techniques to manipulate phosphoinositides on a timescale of seconds and solved long-standing biological questions regarding their functions. This series of work has greatly impacted the fields of membrane lipids, ion channels and transporters, and the broader field of cell signaling. However, these findings also generated a new question of how this membrane lipid drives multiple cellular functions with high precision. In order to address this, we developed two techniques based on rapamycin-induced protein dimerization to rapidly change plasma membrane PI(4,5)P2. First we increased PI(4,5)P2 by synthesis from PI(4)P using a membrane recruitable 5-kinase, and found that cells form actin comets (Movie #5). We then developed a second technique that increases the amount of available PI(4,5)P2 without consuming PI(4)P. This induced membrane ruffles (Movie #6). These distinct phenotypes were mediated by two signaling events: crosstalk between Rac and RhoA GTPases, and dynamin-mediated vesicular trafficking. Our results indicate that the effect of PI(4,5)P2 on actin reorganization depends on further phosphoinositides such as PI(4)P. Thus, combinatorial regulation may explain the diversity of phosphoinositide functions. Schematic illustration of two techniques to rapidly manipulate PI(4,5)P2 by chemically inducible dimerization. (Ueno et al. Sci Signaling, 2011)

Chemically inducible recruitment of cytoplasmic proteins to various organelles (Komatsu et al. Nat Methods, 2010).

We have previously demonstrated that a series of chemical dimerization probes for Rho GTPases can rapidly induce distinctive morphology changes depending on which Rho GTPase was activated (Movie #7). Using a series of novel chemically-inducible dimerization probes, we generated a system in which proteins were rapidly targeted to individual intracellular organelles such as plasma membrane, endoplasmic reticulum, Golgi, lysosome, and mitochondria (Movie #8-12). We demonstrated that a Ras GTPase can be activated at distinct intracellular locations and that membranes from two organelles can be inducibly tethered. We are also extending this technique to achieve activation at distinct subcellular locations at the cell surface. A newly synthesized photocaged rapamycin derivative induced rapid dimerization of protein dimerization upon UV irradiation. By combining this system with highly spatially confined UV-irradiation, we achieved subcellularly localized activation of Rac, a member of the family of small GTPases (Movie #13). Our technique offers a powerful approach to studies of dynamic intracellular signaling events

Visualization of calcium signals in primary cilia (Su and Phua et al. Nat Methods, 2013).

Primary cilia are short hair-like processes which are found on almost all eukaryotic cells. They are tiny entities - roughly 0.5 microns wide and several microns in length - which only constitute approximately one ten thousandth of a cell’s total volume. The existence of primary cilia has been appreciated for more than 100 years, but it was only recently that scientists begin to explore its function. Unlike normal cilia that move to generate forces, primary cilia are immotile and thought to function as an “antenna” for sensing various environmental factors such as mechanical, chemical, or photothermal stimuli. For example, in the case of renal epithelial cells, primary cilia extend into the apical lumen, putatively functioning as a sensor for liquid flow in tubules. It is increasingly appreciated that abnormal ciliary structure and function results in many disorders such as loss of smell and sight, and kidney failure. These disorders are collectively termed “ciliopathies”. Despite its significant involvement in physiology and pathophysiology, studies of primary cilia have been hampered primarily due to its small size. It is extremely challenging to distinguish signaling events inside primary cilia from those of elsewhere in the same cell. There is virtually no technique to “see” or “touch” dynamic signaling events inside primary cilia. We have previously developed a novel technique that allows for rapid, inducible perturbation of various signaling molecules at specific locations in living cells. We will use intracellular calcium signaling and polycystic kidney disease as a model to elucidate signaling pathways regulating structures and functions of primary cilia. By uniquely combining our perturbation technique with real-time, live-cell fluorescence imaging, we will visualize and manipulate dynamic calcium signaling events specifically in primary cilia. This research will not only provide a powerful technology for probing signaling complexity in primary cilia, but also offer far-reaching implications for the cure of ciliopathies.