Recording from neural circuits is the bread-and-butter of neuroscience. As well as simply recording passively from neurones in vitro or in the live animal, it is important to impose experimental constraints on the system;for example by electrically stimulating a particular neurone and following the responses of others in the circuit, or by applying drugs to the preparation. Such work is informative, but painstaking and prone to artefact.

Recently, transgenic technologies have provided exciting new technologies for neuronal monitoring and monitoring. My own lab was the first to produce animals that carry transgenic calcium reporters, allowing synaptic transmission to be monitored; and recently, Drosophila have been made carrying transgenic potassium channels that act to make neurones less excitable, so shutting them down. Elaine Tan's paper in Neuron adds a valuable new technique that allows mammalian neurones to be selectively and reversibly switched off. To do so, the team has drawn on insect endocrinology.

Previous work from the group had shown in vitro that mammalian neurones transfected with the gene for an insect allatostatin receptor would stop firing action potentials when allatostatin was added. Allatostatin is a peptide hormone that normally acts to inhibit the production of juvenile hormone, a key hormone that regulates the quality of insect moulting. Critically, it is sufficiently different from any vertebrate hormone that one would not anticipate cross-activation of any vertebrate receptor with the insect neuropeptide. Thus, only those cells in which the allatostatin receptor is expressed should respond to allatostatin. The nature of the response is also critical; it appears that in mammalian neurones, the allatostatin receptor activates a potassium channel that hyperpolarizes (makes more negative) the neuronal membrane. Under such circumstances, the neurone becomes most unlikely to fire when neighbouring cells stimulate it: it is effectively`switched off'.

However, studying nerve cells in culture introduces artefacts aplenty. Could this technology be replicated in vivo? Callaway's group put their gene into a modified adenovirus - a cold-like virus that has become popular for its ability to introduce transgenes into a wide variety of mammalian cells. They coupled the gene with a synapsin (neuronal-specific)promoter, ensuring that, although the adenovirus might infect many cell types,it would only make the allatostatin receptor in neurones. The team also incorporated green fluorescent protein into the virus, so that transfected cells would be fluorescently marked.

In essence, the technique worked exactly as predicted. Working first in rat cortex, then in ferret and monkey, the team were able to show that perfusion with allatostatin shut down neuronal signalling, and wash-out of the allatostatin allowed it to start again.

Although these results provide a valuable tool for non-model organisms,they also reveal some of the advantages of working in a suitable model. For example, it is very hard to introduce a transgene into just a few specific cells with adenovirus. However, the team showed that this was indeed possible in transgenic mouse, allowing them to express the allatostatin receptor in specific populations of spinal neurones and selectively inhibit neuronal firing.

Why is this important? The technology has the potential to allow a more quantitative, `systems biology' approach to the study of neuronal circuits,although it will require refinement if it is to be useful beyond mouse. Additionally, it allows in vivo manipulation of neuronal circuits, so protecting the experimenter from the artefacts of cell culture. It is thus an exemplar of the new `integrative physiology'.

Tan, E. M., Yamaguchi, Y., Horwitz, Gosgnach, G. D. S., Lein, E. S., Goulding, M., Albright, T. D. and Callaway, E. M. (
). Selective and quickly reversible inactivation of mammalian neurons in vivo using the Drosophila allatostatin receptor.