At A Glance
Name: Jin Zhang
Position: Assistant Professor, pharmacology and molecular sciences; neuroscience, Johns Hopkins University School of Medicine
Background: Postdoc, pharmacology, University of California, San Diego — 2000-2003; PhD, chemistry, University of Chicago — 2000
Having conducted her postdoctoral fellowship in the lab of fluorescent protein pioneer Roger Tsien, Johns Hopkins researcher Jin Zhang knows a thing or two about using fluorescent indicators to probe living cells. She is the corresponding author of a PNAS paper published today [PNAS, 2004 Nov 23; 101(47): 16513-16518] describing a new FRET-based assay for monitoring cAMP signaling in living cells. Last week, she discussed her work with Inside Bioassays.
Tell me a little about how you developed your interest in live-cell imaging and more specifically, studying cell signaling pathways.
If we’re talking about the general field, I was trained as a biological chemist, and I have always been interested in signaling pathways, even in my graduate studies. One of the very fundamental questions is: How does the cell achieve specificity in signal transduction? Signal transduction, of course, is the mechanism by which cellular information is processed. If you think about it, there are so many signal inputs and signal outputs, so how does one set of inputs lead to a specific output under a specific condition? So this could be called a long-standing question. The way I like to think about it is although we know there are so many cellular components involved, and there’s a lot of cellular information being processed, it’s very likely that just a limited number of fundamental strategies are used for signal transduction.
One of the probable strategies is to achieve specificity by temporal coordination and spatial compartmentalization of signaling components. And because I was trained as a biological chemist, I realized the importance of tool making. My graduate work and my postdoctoral work was with Roger Tsien, where we were making molecular tools to study biological questions. So I would like to develop tools to help us to understand this specificity issue — the spatial and temporal regulation of signaling components. That’s the major theme surrounding my still short career.
Can you encapsulate what the importance is of being able to monitor cAMP in real time and real space?
Cyclic AMP, of course, is the classic second messenger — I think the second messenger concept actually started with it. So many cellular responses through cAMP are mediated by PKA, which is cAMP-dependent protein kinase. This molecule, along with cAMP, is involved in numerous signaling cascades, and this PKA molecule can phosphorylate a huge number of substrates. So the question again is the specificity issue: How does the molecule know when and where to phosphorylate which molecule, which downstream substrate? It has become clear that subcellular compartmentalization is a major theme to achieve specificity, and people have speculated or suggested micro-domains that exist to control the specificity — meaning different pools of cAMP generated at different sub-cellular sites would be important to achieve cAMP actions. In order to study the temporal and spatial dynamics of this, we have to do it in intact cells. Because if you destroy the cells or use artificial systems, then you lose all of the context and all the information about spatio-temporal regulation.
Prior to this reporter assay, what has been the standard way to monitor this in live cells?
In live cells, we can start with the really traditional assay method that has been used for many years, which is the radioassay or immunoassay. In live cells, in the early 90’s in really pioneering work from Roger Tsien’s lab, they developed a fluorescent reporter for imaging cAMP in living cells. This involves generating the PKA, chemically labeling that with fluorophores, and then micro-injecting that into living cells. They used that methodology to reveal quite interesting cAMP dynamics in neurons, et cetera.
More recently, fluorescent protein versions have been developed in collaboration between an Italian lab and Tsien’s group. And another useful method is based on a cAMP-dependent channel — you can look at the channel activity, and then translate that back to the cAMP dynamics. So we kind of stand on the giant’s shoulder — we’re making improvements on those existing methods and developing them into more robust assays. Especially in the early 90’s, with Roger’s work, we realized that when people start to use these assays, like any assay, there are limitations. For example, all the PKA-based approaches are not as easy to use as we would have hoped. The PKA itself is a tetramer, so all the PKA-based probes are bimolecular probes, and you need to label them with different fluorophores, and of course you have to match the expression level of those two very carefully to generate a functional reporter. And even then, these fluorophore-labeled subunits will sometimes pair up with an endogenous partner and give you a mixed tetramer, so that’s not really as robust as we would have hoped.
When I started my own lab just about a year ago, we were basically interested in a lot of aspects of cAMP and PKA signaling, so I had noticed that there’s a relatively newer cAMP receptor called Epac, and it’s been suggested that it undergoes a conformational change upon cAMP binding. So we had the idea of sandwiching that between CFP and YFP, two fluorescent proteins that are known to undergo FRET. Actually, that was developed by then, a rotation student, Lisa DiPilato, at the beginning of her second year, in collaboration with Xiadong Cheng.
Does this differ from traditional FRET assays?
FRET itself is very useful because you can detect protein-protein interactions or conformational changes, and that kind of spatial resolution is very important. FRET occurs within the range of 100 angstroms, and typically around the 50 to 70 angstrom range, and that’s the range of molecular interactions. This type of spatial resolution is below any type of light microscopy, and that’s the spatial resolution we need.
And with FRET imaging, we’ll be able to do ratiometric imaging — that is, you monitor both the donor and acceptor emissions, and then calculate the ratio, which allows you to cancel out a lot of the variations; for example, the cell size, and thickness, and light source variations.
What type of instrumentation do you use currently to monitor these assays?
For this we routinely use an epifluorescence microscope. People have used two-photon microscopy, and we’re also trying to see if we can get consistent results in plate reader formats.
Do you think this is something that can be adapted to higher-throughput methods?
That’s definitely one direction we want to push forward. One potential difficulty for the whole family of reporters is the dynamic range. Right now this one gives about a 30-percent change in the emission ratio, and we’re testing whether this can be consistently detected in a higher-throughput format. Very likely, to get a robust assay in [such a] format, there needs to be more improvement and adaptation, which is something we’re doing.
Improvement of the assay, or instrumentation?
Improvement of the reporter itself, by protein engineering. That’s something we’re good at. For the machine part, we’re not so good at it. That’s probably for the commercial side.
Are you working with anyone on that?
There are a lot of people here — the biomedical engineering department here is also very strong, so hopefully we can find some collaborators. But mostly we want to further develop this for basic research applications.
Do you have an opinion as to whether this will be applicable to drug discovery?
Yes, I think it will have potential applications along that line, because GPCRs are one of the classic molecules that are important drug targets, and GPCR coupled to Gs activates adenylyl cyclase, and then increases cAMP; and on the other hand, Gi activation will decrease cAMP, so cAMP is the second messenger that directly reports that. I know I’ve recently read about real-time cAMP assays that are being developed for drug discovery, so that’s why we think there may be some potential interest. And our assay is a fluorescent assay, and because of all the high-throughput instruments that are available for fluorescence-based assays, this may be attractive.
You had mentioned that there was some interest in commercializing this assay. Where does that currently stand?
So far we’ve just filed provisional patents, and a couple of friends from the commercial side have expressed interest, but it’s still in the very early stages.
What’s next for this research? Is it going to primarily focus on improving the reporter construct?
Yes, and as I touched on before, we’re interested in many aspects of cAMP/PKA signaling, defining the signaling micro-domains. And also we’re interested in, for example, looking at the involvement of this pathway in integrin-mediated cell migration, and also how hormones regulate this pathway as an example for cross-talk between different signal transduction pathways. We’re also using this method to look at other second messengers — protein phosphorylation events — so sometime in the near future we’d like to integrate some aspects of disease-related research, because at Johns Hopkins Medical School, cancer biology is very strong. We hope to not only have an understanding of the signaling network and dynamics, but also to contribute to the understanding to some of the disregulation in some diseases like cancer.