Lighting Up Neuronal Pathways: The Development of a Novel Transgenic Rat that Identifies Fos-Activated Neurons Using a Red Fluorescent Protein

The immediate early gene c-fos is the most extensively used marker for neuronal activation. A search of the term c-fos in PubMed results in over 18,000 hits. Measuring changes in gene expression is one the simplest ways to determine whether neurons are activated in vivo. The principal advantage of c-fos is that it provides an easily identifiable and fairly sensitive marker to recognize activated neurons in vivo after a given stimuli, such as an injection of a drug or performance of a behavior. Expression of c-fos is sensitive to activation by numerous different stimuli, including growth factors, ion channel activation, neurotransmitter release, and behaviors (for review see Ref. 1). Its expression is turned on through activation of many intracellular signaling pathways, including increases in cAMP, calcium, and activation of the MAPK pathway (for reviews see Refs. 2,3,4). Expression of c-fos gene or protein can be followed temporally and spatially to identify which neurons in a pathway are activated and when. It has been used by many different groups with great success to map out neuronal pathways activated by specific stimuli, such as a behavioral response or an exogenously applied drug (for examples see Refs. 5 and 6) (reviewed in Ref. 4). Colocalization experiments can also be performed relatively easily to determine the neurotransmitter phenotype, receptor, intracellular signaling protein expression or projection site of the activated neurons. Of course, like most things, c-fos expression has its limits. First, precisely because c-fos is turned on by such a wide variety of signaling pathways, c-fos expression alone cannot provide much information about the signaling pathways involved. Second, it provides little information about the degree of activation; for example, immediate early gene expression does not always result in long-term changes in gene expression or correlate with neuronal firing (see Ref. 4 for a review). Third, it does not indicate whether the activation of the neuron is direct or indirect. Fourth, it is possible that a stimulus activates a neuron without turning on c-fos. Last, c-fos expression is useful for measuring only activation, not inhibition, of the neurons of interest. However, other techniques, such as in vivo electrophysiological recordings or imaging, are time consuming, technically demanding, and often cost-prohibitive. Therefore, despite these limitations, c-fos remains the most commonly used method for assessing neuronal activation in vivo due to all the advantages discussed previously.

In this issue, Fujihara and colleagues (7) report the development of a transgenic rat that expresses the fluorescent protein, monomeric red fluorescent protein 1 (mRFP1), under the control of regulatory sequences in the c-fos promoter. RFP is therefore expressed wherever c-fos is activated. Although the methods currently used to detect c-fos are not complicated, they can be time consuming. The development of c-fos-mRFP1 transgenic rats allows the easy measurement of c-fos neuronal activation. The signal appears bright, making immunostaining unnecessary. Therefore, these transgenic rats could save a lot of time, particularly for experiments involving a large number of animals. In this initial study, the authors used the transgenic rats to show that mRFP1 expression is induced in the paraventricular nucleus of the hypothalamus after an osmotic challenge. The expression pattern detected paralleled immunostaining of the native c-fos protein after the same stimulation paradigm. They went on to cross their c-fos-mRFP1 transgenic rats with transgenic rats expressing enhanced green fluorescent protein (GFP) under the control of the arginine vasopressin (AVP) promoter (8). They used these double-transgenic rats to show the osmotic challenge induced mRFP1 florescence in AVP neurons. Their results suggest that this model will be extremely useful to easily detect induction of c-fos (and therefore neuronal activation) in vivo and can be used in combination with other transgenic fluorescent rats.

This new transgenic rat model could also have substantial implications for experiments beyond making detection of c-fos activation easier. Perhaps the greatest potential advantage is for ex vivo preparations. Currently, the standard way to identify specific neurons that are activated by a stimulus in vivo in the rat is to measure changes in gene expression or other markers of neuronal activity, such as phosphorylated cAMP response element-binding protein, phospho-ERK, or phosphorylated signal transducer and activator of transcription. However, to do this, the tissue has to be processed accordingly. What makes the development of these transgenic rats so exciting is that for the first time it may be possible to identify neurons that were activated by a given stimuli in vivo in real time using an in vitro preparation. These activated neurons could then be studied by live imaging, such as with calcium or voltage-sensitive dyes, or electrophysiological techniques. This would combine the advantages of c-fos activation, that it identifies neurons activated in vivo, with the advantages of in vitro recordings, which allow characterization of the electrophysiological properties and signaling pathways within the identified neurons. Currently, in the rat brain slice preparation, electrophysiological recordings can be made only from unlabeled neurons or neurons labeled by neurotransmitter phenotype (e.g. using other transgenic fluorescent reporter rats), projection site (e.g. after retrograde transport after site-specific injections of fluorescent beads), or afferent inputs (by labeling afferent inputs with a fluorescent dye). Additional analysis is possible by post hoc immunostaining or by RT-PCR, but both of these techniques severely restrict the type of recordings that can be made. Furthermore, even strong signals seldom activate all neurons within one group or phenotype, and a subtle behavior, such as whisker barrel stimulation, activates only a very small portion of neurons. Therefore, it is extremely hard to compare changes in electrophysiological properties without knowing whether the neuron you are recording from was one that was activated by the behavior (or drug) or not. These transgenic rats will potentially allow investigators to specifically identify and record from a neuron that was activated by a given stimuli in vivo and compare the properties and responses of that activated neuron to a neighboring neuron that was not activated. These transgenic rats could also be crossed with other transgenic rats that express fluorescent proteins that emit at a different wavelength (such as GFP) under the control of other promoters, as was done by Fujihara and colleagues (7), to allow identification of a specific phenotype of activated neurons.

There are many other applications that one could envision for these transgenic rats. Brain slice or dissociated cultures could be made from these animals that would allow activation of gene transcription to be followed in real time after application of stimuli, such as neurotransmitters or growth factors, to a population of neurons in vitro. This would be particularly useful in slice cultures, such as hippocampal slices, which maintain some of the architecture of the brain region. These techniques could then be combined with transfections targeted to disrupt specific proteins (e.g. dominant-negative or small interference RNA constructs) to determine the cellular signaling pathways involved. These transgenic rats would also be ideal for laser capture microdissection strategies.

Before these exciting new approaches can be taken, the c-fos-mRFP1 transgenic rats remain to be fully characterized, and several questions need to be addressed. First, would methods of euthanasia induce the expression of the mRFP1 protein, thus invalidating any RFP activation you see in the acute slice? This has not proved to be a problem in brain slices prepared from fosGFP transgenic mice (9). Surprisingly, the preparation of the slice alone did not induce fosGFP expression, and these animals have been used successfully to identify neurons in vitro that were activated by stimuli in vivo, e.g. whisker barrel stimulation (9,10,11) and locomotor tasks, such as walking or swimming (12). The time course of activation of the mRFP1 and the half-life of the mRFP1 protein would also need to be established in these transgenic rats. It is probable that there will be a slight temporal delay in the expression of the mRFP1, because maturation of the fluorophore is likely to be required for detection, as has been noted in the fosGFP mice (9). Furthermore, the mRFP1 protein may be more stable than c-fos, and therefore, the marker for activation could be longer lasting, as has also been reported for the fosGFP transgenic mice (13). Both these properties would have implications for the type of applications the c-fos-mRFP1 transgenic rats can be used for. Lastly, it would be interesting to determine the extent of colocalization of the mRFP1 with native c-fos to determine whether the c-fos-mRFP1 transgenic model is more or less sensitive than detection of the native protein or mRNA.

In summary, the report of Fujihara and colleagues (7) of the development of a c-fos-mRFP1 transgenic rat opens up the possibility of a whole array of interesting experiments. Although it remains to be determined whether these transgenic rats have the same useful properties as the fosGFP transgenic mice; the possibility that they could pave the way for innovative new experiments that have not been feasible before is something worth getting excited about.