At A Glance
Name: Paul Taghert
Position: Professor of Neurobiology, Washington University School of Medicine
Background: Assistant, then Associate Professor of Neurobiology, WU School of Medicine — 1984-2000; Postdoc, Stanford University — 1981-1984; PhD, zoology, University of Washington — 1981
At the Washington University School of Medicine in St. Louis, Paul Taghert and his colleagues in the department of anatomy and neurobiology are investigating circadian rhythms in Drosophila melanogaster. Taghert took some time to talk with Inside Bioassays about the connection between this research and cell-based assays — a preview of a presentation he will give next month at the Assays and Cellular Targets meeting in San Diego.
How did you become involved in your current field of study?
I’m a basic scientist, and I’ve been studying neuropeptide transmitters ever since I can remember. We do our work in insect models — primarily Drosophila, which has its strength in genetics. My lab focuses primarily on circadian rhythms, and neuropeptides figure prominently in the neural circuits underlying circadian rhythms in a variety of animals — mammals and flies. We have been interested in a couple of these neuropeptides, because based on genetic criteria in flies they appear to figure prominently in the circuits underlying circadian rhythms. They’re released by particular circadian pacemaker neurons, and they transmit interesting signals, the absence of which makes the flies generally arrhythmic. To pursue this line of work and to better understand the peptides, it’s important to identify their receptors and define signaling pathways. And it’s with that motivation that we began examining neuropeptide receptors in Drosophila computationally, based on the genomic data. We did this two or three years ago. It was strictly a computational analysis where we looked at predictive genes and categorized them phylogenically. So on that basis we came up with 44 receptor genes predicted in the Drosophila genome, plus or minus one, for neuropeptide receptors — the vast majority of which were orphans. It wasn’t even clear that they were expressed genes at that point, but it turns out that they pretty much all are. So in the interim, we and others have been studying these receptors, and we have now deorphaned a majority, using a variety of assays. And this finally gets to: What’s my participation in this [conference] in San Diego? I’m not a typical participant in this sort of forum. But we use a variety of orthodox signaling assays to try and define the ligands for eight orphan receptors, with some success, but we’ve rapidly plateaued on that. [We have] collaborated with a colleague at Duke, Mark Caron, who has looked at how these receptors turn off and ways of looking at desensitization, and in particular, the beta-arrestin GFP assay.
Mark Caron is part of Norak Biosciences.
Right. That company was formulated to exploit that technique — there’s a method to deorphan. But it’s still largely unproven; well, efficacy in deorphaning has not fully been shown. So in collaboration with them, we used [the assay], and very quickly were able to deorphan a fairly good number of these insect receptors. My presentation in San Diego is simply going to overview that particular usage. Its value to us is that it didn’t require any particular knowledge of the specific signaling pathway, so it seems to be a more [all-purpose] assay.
This is a very drug-discovery oriented conference. What is your understanding as to the relevance of this to people in that field?
Well, that remains to be seen. They invited me, and I’ll be going. It’s predicated on the idea that there remain a large number of orphans, and it will be useful to have a good understanding of what their natural ligands are. In trying to approach the discovery of new drugs, it could be useful — there’s a random approach and a rational approach — and for rational approaches, it would certainly be useful to better understand natural ligand identities.
Likewise, I think the assay certainly has potential even in random drug discovery, given Norak’s ability to quantify and make robotic the use of these — neither of which we did, by the way. Our assays were manually controlled and not quantified, either. So we applied the technique a little bit differently from the patented method, which has its own platform, et cetera.
What assay instrumentation and method do you use in your lab for this?
It was simply done manually — manual drug addition to dishes containing transformed cells, and taking time-lapse images using a confocal microscope. Quantification was by cell numbers, actually, as opposed to fluorescence. So I believe the Norak method is automated, and counts pixels. We did it strictly on a slower time base, because the number of receptors we were screening was smaller, and the number of potential compounds we were screening was much smaller. But then again, we also had a very high success rate. That’s an aside; I don’t know what the moral of that is.
Can you briefly explain the terms “orphaned” and “deorphaned” receptors?
It’s simply a shorthand — a poor shorthand — an attempt to define the natural ligand for a receptor. We fully subscribe to the idea that multiple lines of evidence are best, and the use of this method for us represents what is an entryway to try and deorphan a receptor. So a positive result on this assay is reason to keep going with other traditional methods, such as binding assays, or signaling assays. By deorphaning, [we mean] it’s a long process involving several lines of evidence to implicate a particular ligand for a particular receptor, and we submit that the use of this desensitization assay can be a useful first assay, because it simply doesn’t require a whole lot of information about signaling pathways. So it has virtue in that regard, and people should keep it in mind for that particular reason.
What is the overarching goal of your research on circadian rhythms?
We’re trying to understand behavior, and we’re trying to define how the brain controls particular behaviors, and we’re interested in rhythmic behaviors, like our circadian-controlled ones, to use that as a way to define neurocircuits. And we’re using transmitters as means to define the elements. So finding the transmitters allows us to define which cells are relevant in a circuit — which are signaling and which are responding — and then use those cues to assemble a model of how nerve elements are used operationally to control behavior. So right now we’re just taking the very first steps, and using molecular genetics to identify the pacemaker neurons, or those that contain the clocks. There are special gene elements that cycle every day, and they give sense of time, and then they impart that to other cells — first of all to other pacemakers to help synchronize them so the body receives one sense of time, not several, and then they have to impart that synchronized signal outside the circuit to other neurons that control behavior. So we’ve identified peptides that are released by pacemaker cells. Identifying the receptors simply gives us the means to identify the first set of responders, and we hope to move on from there.
These are primarily GPCR receptors?
Are these typically the primary receptors involved in these neural processes?
There are plenty of receptors, but it happens that neuropeptides are used prominently as signals in these circuits in mammals and flies. These are highly conserved signaling mechanisms. I can’t rule out the involvement of other signaling pathways.
What’s next for your research?
We’ve recently identified the receptor we’ve been chasing for a long time…
Which one is that?
It’s a receptor for a neuropeptide called PDF [pigment dispersing factor]. And having identified it now, we’re interested in its distribution and its signaling turnover. We’d like to know: Is it expressed by clock cells or non-clock cells? Is it a cycling product, or is it static? Does the receptor participate in the gating mechanism — the thing that changes every day to give sense of time? What are the transmitters that are used by the neurons that are receptive, and can we start to form a rudimentary model of the neural circuit on that basis? So we have plenty of neuroanatomy, and pharmacology, and genetics to exploit the identification of the receptor. I’ll also throw in that we’ll use some drugs to block it, in the context of drug discovery and assay development. We’ve certainly tried to identify some antagonists and constitutively active receptor forms, as well.
What would be the usefulness of these?
We restrict our work to flies, but we’re upgrading the guiding notion, and so far it seems to be that the principles that we are trying to establish are common. We can say so by several indicators that indicate evolutionary conservation in the signaling pathways that appear to be useful. For example, these 44 peptide GPCRs are highly related, for the most part, to their mammalian relatives, and appear to derive from common ancestors. The insights that we gain from the fly can be tested in mammalian systems, as well. We can get some answers a little more quickly, because the genetics are a little bit more fleshed out in flies, and the total number of neurons is much smaller, so we have much better resolution. These are the principal motivations to use the fly. So we hope that the things that we learn about neurocircuits using peptide transmitter signaling can be applicable, rather quickly, we hope, to the human condition.