Article activity feed

  1. Evaluation Summary:

    This study implicates excitatory projections from cholecystokinin (CCK) entorhinal cortical neurons to the lateral amygdala in trace fear conditioning in mice. Behavioral, chemogenetic, optogenetic, and electrophysiological work show that these projections are critical for the acquisition of conditioned freezing to a trace conditioned stimulus. The identification of a novel circuit and genetically defined cell type for regulating fear memory formation important. However, whether this pathway is specifically involved in trace fear conditioning is unclear from the present results and further work is needed to address analytic and interpretational concerns.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1, Reviewer #2 and Reviewer #3 agreed to share their names with the authors.)

  2. Reviewer #1 (Public Review):

    In this paper by Feng et al. the authors examine the role of cholecystokinin (CCK) cells in the entorhinal cortex (EC) in fear conditioning. They find that CCK knockout mice are deficient in short and long trace fear conditioning and that this deficit could be rescued by administration of systemic CCKB receptor agonist administration. Using an in-vivo synaptic plasticity assay, they present suggestive evidence that LTP is disrupted in CCK-/- animals. To determine the source of CCK to the LA, they use anatomical tracing techniques to show that the EC contains CCK+ cells which project to the lateral amygdala (LA) and use a DREADD approach to reveal that EC-CCK+ cells are necessary for trace fear conditioning. They then take advantage of a variety of plasticity, shRNA and optogenetic approaches to show that EC-CCK+ cells contribute to plasticity in LA and are necessary for fear conditioning. These results are potentially important as they reveal a role for EC projections to the LA in fear learning and connect this to a specific population of CCK expressing cells. While the findings are compelling, there are issues with the analyses and experimental design (in some cases), validation of the shRNA knockdown technique and some of their interpretations which need to be addressed.

  3. Reviewer #2 (Public Review):

    The authors perform a series of elegant experiments to explore the role of cholecystokinin (CCK) neurons in trace fear conditioning in mice. They show that mice lacking CCK exhibit deficits in trace fear conditioning with both short and long CSs/ISIs--they previously showed these animals also have deficits in delay fear conditioning. Subsequent experiments revealed that CCK-deficient mice showed deficits in LTP-induced potentiation of auditory-evoked potentials in the lateral amygdala (LA), and that systemic activation of CCKBR receptors with CCK4 increases activity CCK in the LA and rescues the deficit in trace fear conditioning. They next used combinatorial tracing methods to reveal a CCK projection from the entorhinal cortex (EC) to the LA in CCK-Cre mice. Chemogenetically silencing LA-projecting CCK neurons in EC impaired trace fear conditioning. Lastly, optogenetic stimulation of CCK-EC axons in LA induced potentiation of auditory-evoked potentials in LA, and this was prevented by RNAi-mediated knockdown of CCK in EC neurons. Optogenetic inhibition of EC->LA CCK neurons also inhibited trace fear conditioning. This is an impressive and thorough set of experiments that reveals a role for CCK-containing EC neurons that project to the LA in trace fear conditioning. However, a shortcoming of the work is that it is not clear whether this projection is involved specifically in trace fear conditioning, or has a more general role in either delay or contextual fear conditioning.

  4. Reviewer #3 (Public Review):

    In the present manuscript, Feng and colleagues used sophisticated techniques to elucidate the role of the neuropeptide cholecystokinin (CCK), and the neurons which produce this peptide, in a model of trace fear memory. First, using global genetic knockout mice, in vivo electrophysiology, and exogenous administration of a CCK receptor agonist, the authors showed that CCK is vital for trace fear memory and associated synaptic plasticity (long-term potentiation; LTP) assessed by changes in auditory evoked potentials (AEPs) in the lateral amygdala (LA). Anatomical tracing revealed diverse inputs to the LA, including those from the entorhinal cortex (EC). Using chemogenetics, the authors showed that the activity of EC neurons, specifically those expressing CCK, is essential to the formation of trace fear memory during conditioning. Further anatomical tracing demonstrated an abundance of CCK-expressing neurons that project from the EC to the LA, and optogenetic excitation of these cells recapitulated AEP-LTP in the LA associated with trace fear conditioning. Next, the authors used viral-genetic techniques to block production of CCK by EC neurons and found that CCK originating in EC neurons is necessary for LTP of the AEP within the LA. Finally, Feng et al. employed optogenetic inhibition of CCK-expressing neurons that project from the EC to the LA during conditioning to demonstrate that these cells are necessary for the formation of trace fear memory. Taken together, this elaborate set of experiments establish an important role of a peptide-signaling circuit in a model of fear memory.

    The results described herein will be useful to behavioral neuroscientists seeking to understand how the brain processes fear.

    This manuscript is well written and the scientific question holds translational relevance, particularly in being able to inform clinical scientists attempting to develop therapeutics targeting peptide signaling to improve symptoms of anxiety disorders. While this study has scientific and practical value, some issues of methodology, interpretation of results, and presentation of data should be addressed by the authors prior to publication.

    1. Statistical and Methodological Concerns:

    1.1) In determining the effects of experimental manipulations on freezing scores, the primary behavioral readout in this study, the authors make inappropriate use of statistical tests. While the authors' comparisons of group averages for freezing are reasonable, the use of t-tests to compare the effects of manipulations across time during trials is inappropriate and would be better suited for repeated measures ANOVAs. This issue can be easily addressed by reanalysis of this set of data.

    1.2) A couple of issues related to the use of viral techniques should be addressed, as well. In using optogenetics to induce LTP, the authors use a particular viral serotype (AAV9) that may lead to anterograde expression of their light-sensitive channel (ChETA) in neurons downstream of their target region. This concern can easily be addressed by additional histology and disclosure of this methodological caveat in the text.

    1.3) The second issue of viral-genetic techniques to be addressed is in the authors' use of shRNA to knockdown CCK expression by EC neurons. The authors failed to show validation of this technique by quantifying the expression of CCK after viral manipulation. This concern can also be easily addressed with additional histology.

    1. Concerns of Interpretation of Results: While the authors elegantly demonstrate a role of CCK-expressing projection neurons originating in the EC and terminating in the LA in their behavioral model, there exists some overextension of interpretation of these results by the authors in the present manuscript. In particular, the authors infer that synaptic release of CCK, per se, by EC neurons in the LA is responsible for the effects observed. However, the authors do not demonstrate that CCK is being released in the LA by neurons originating in the EC. The authors should limit overinterpretation of their results and discuss alternative explanations, such as the possibility of local release of CCK by CCK+ neurons in EC which could be further triggering the release of CCK from local CCK+ neurons in the amygdala.