Neuronal activity engages biochemical signaling pathways that produce alterations in gene expression that lead to long-term adaptive changes in neurons that are fundamental to behavioral plasticity. Since the discovery more than 30 years ago that depolarization can drive transcription of the cfos gene (Greenberg et al., 1986; Morgan and Curran, 1986), a great wealth of knowledge has accumulated on the mechanisms of excitation-transcription coupling. Ca2+ ions entering through NMDA receptors during synaptic transmission and L-type voltage-dependent calcium channels upon subsequent depolarization act as second messengers to engage various signal transduction pathways that activate transcription factors to drive transcription of immediate-early genes (IEGs) (Alberini, 2009; Deisseroth et al., 2003; Greer and Greenberg, 2008; Wheeler et al., 2012).

Traditional IEGs such as fos, arc, and egr1(zif268) have therefore been used as surrogate markers for neuronal activity driven by environmental stimuli, identifying recently active neurons with techniques such as in situ hybridization, immunostaining or transgenic reporter animals. The functional importance of these neurons was demonstrated in recent studies that exploited the activity-dependence of fos to express genetically encoded optogenetic tools or designer drug receptors in neuronal ensembles activated during fear conditioning. Subsequent artificial activation of these neuronal ensembles is sufficient to mimic the fear response (Garner et al., 2012; Liu et al., 2012).

While traditional IEGs have played an invaluable role in identifying activity-dependent circuitry, the expression of these genes may not be entirely tuned to neuronal activity alone. Traditional IEGs are transcribed in response to synaptic activity and Ca2+ entry, but also in response to growth factor signaling, neurotrophins and neuromodulators signaling via cAMP.

Figure 1: Npas4 is a pure marker of neuronal activity.

In contrast, Michael Greenberg’s lab showed that the neuron-specific IEG npas4 is strongly and rapidly induced by neuronal activity, but does not respond to paracrine factors (Lin et al., 2008). This tight tuning for neuronal activity suggested that Npas4 can be a purer marker of neuronal activity. In this background section, we will outline data summarizing properties on npas4 that make it unique among IEGs.

To search for more genes with pure tuning to activity like npas4, we performed RNA-Seq experiments of primary cultured mouse cortical neurons and compared expression driven by synaptic activity (GABAA block with bicuculline) or cAMP (forskolin stimulation).

Figure 2: Genome-wide demonstration of npas4’s unique tuning to activity.

Traditional IEGs were up-regulated by both synaptic activity and cAMP. As expected, npas4 was up-regulated by synaptic activity. In fact, of all ~15,000 transcripts detected, the npas4 gene was the most highly induced activity, yet as predicted by the Greenberg lab, npas4 was not up-regulated by cAMP. Importantly, we did not identify any other genes that displayed similar tuning to neuronal activity.

To examine the induction of npas4 In Vivo, we performed contextual fear conditioning (cFC) experiments and measured gene expression with qPCR and RNA-Seq in both the hippocampus and the retrosplenial cortex (RSC), two brain areas known to be involved in learning and memory (Baumgärtel et al., 2018). In these experiments, animals were not habituated to the fear conditioning box and the response relative to home-cage represents encoding of the context, not fear memory per se. Examination of the time course of fos, arc, and npas4 induction after training showed that npas4 mRNA levels return to baseline within 1 hour in both brain areas, whereas fos and arc expression remained elevated over longer periods of time.

Figure 3: IEG response kinetics in hippocampus & RSC.

This is consistent with data from Yingxi Lin’s lab that showed similar results for these three genes after cFC training (Ramamoorthi et al., 2011). Further, they found that acute npas4 knock-out in neurons prevented the induction of fos, arc, and zif268/egr1, suggesting that Npas4 acts as a master regulator of activity-dependent gene induction.

To demonstrate that IEG induction in retrosplenial cortex is induced by activation of NMDA receptors during cFC, we performed experiments where we exposed mice to the NMDAR antagonist, MK-801, or a vehicle control solution. Importantly, traditional IEGs like arc, fos and egr1 increased expression in response to vehicle injection despite extensive pre-handling, presumably due to the stress of the experience.

Figure 4: Npas4 does not respond to the stress of vehicle injection.

Interestingly, npas4 did not increase upon vehicle injection. All IEGs including increased their expression in response to cFC. This increase in the expression of the traditional IEGs was partially prevented by blocking NMDARs with MK-801, whereas the increase in npas4 was fully dependent upon NMDAR activation. Taken together, these data indicate that npas4 does not respond to pathways activated by vehicle injection. Rather, npas4 responds selectively to cFC and provides a higher-fidelity readout of synaptic activity.

The data above strongly suggest that traditional IEGs respond to signals above and beyond those engaged by neuronal activity per se. Another similar finding came from studies looking at the regulation of IEGs during the circadian cycle. Interestingly, we found that the expression levels of the traditional IEGs fos and arc in the hippocampus had a circadian rhythm, fluctuating in concert over the light-dark cycle, whereas npas4 transcript levels remained relatively constant.

Figure 5: Gene expression over the light-dark cycle.

These data provide further support to the idea that factors other than the acute neuronal activity response to environmental cues control the expression levels of traditional IEGs.

Finally, given the unique regulation of npas4 relative to fos and other traditional IEGs and the fact that c-Fos is very commonly used to mark activity dependent neuronal ensembles thought to be ‘memory traces’ or ‘engrams’, we wanted to determine if npas4 and fos marked the same population of neurons. To do this we performed RNAScope in situ hybridization experiments looking at both transcripts following cFC (Wheeler et al., 2016). In the dentate gyrus region of the hippocampus, we identified a clearly overlapping population of recently activated neurons.

Figure 6: Behavioral training induces cfos and Npas4 in distinct sets of cells.

However, in the CA1 region we saw indications of distinct populations of neurons may be marked with the two different genes. These data are preliminary and cannot be used to make clear conclusions, but they do serve to clearly raise questions about solely relying upon c-Fos to identify recently activated neuronal ensemble and especially the recent trend to equate c-Fos expression to ‘engram cells’.

Alberini, C.M. (2009). Transcription Factors in Long-Term Memory and Synaptic Plasticity. New York 121–145.

Baumgärtel, K., Green, A., Hornberger, D., Lapira, J., Rex, C., Wheeler, D.G., and Peters, M. (2018). PDE4D regulates Spine Plasticity and Memory in the Retrosplenial Cortex. Sci. Rep. 8, 3895.

Deisseroth, K., Mermelstein, P.G., Xia, H., and Tsien, R.W. (2003). Signaling from synapse to nucleus: The logic behind the mechanisms. Curr. Opin. Neurobiol.

Garner, A.R., Rowland, D.C., Hwang, S.Y., Baumgaertel, K., Roth, B.L., Kentros, C., and Mayford, M. (2012). Generation of a synthetic memory trace. Science (80-. ). 335, 1513–1516.

Greenberg, M.E., Ziff, E.B., and Greene, L. a (1986). Stimulation of neuronal acetylcholine receptors induces rapid gene transcription. Science 234, 80–83.

Greer, P.L., and Greenberg, M.E. (2008). From synapse to nucleus: calcium-dependent gene transcription in the control of synapse development and function. Neuron 59, 846–860.

Lin, Y., Bloodgood, B.L., Hauser, J.L., Lapan, A.D., Koon, A.C., Kim, T.-K., Hu, L.S., Malik, A.N., and Greenberg, M.E. (2008). Activity-dependent regulation of inhibitory synapse development by Npas4. Nature 455, 1198–1204.

Liu, X., Ramirez, S., Pang, P.T., Puryear, C.B., Govindarajan, A., Deisseroth, K., and Tonegawa, S. (2012). Optogenetic stimulation of a hippocampal engram activates fear memory recall. Nature 484, 381–385.

Morgan, J.I., and Curran, T. (1986). Role of ion flux in the control of c-fos expression. Nature 322, 552–555.

Ramamoorthi, K., Fropf, R., Belfort, G.M., Fitzmaurice, H.L., McKinney, R.M., Neve, R.L., Otto, T., and Lin, Y. (2011). Npas4 Regulates a Transcriptional Program in CA3 Required for Contextual Memory Formation. Science (80-. ). 334, 1669–1675.

Wheeler, D.G., Groth, R.D., Ma, H., Barrett, C.F., Owen, S.F., Safa, P., and Tsien, R.W. (2012). CaV1 and CaV2 channels engage distinct modes of Ca2+signaling to control CREB-dependent gene expression. Cell 149.

Wheeler, D.G., Beaver, Y., Lapira, J., Ardestani, P., Maruyama, K., Serneo, F., Elow, D., Baumgärtel, K., Scott, R., Shamloo, M. and Peters, M. (2016) Npas4 as a Marker of Recent Neuronal Activity. Annual meeting of the Society for Neuroscience, San Diego.