Experience, circuit dynamics, and forebrain recruitment in larval zebrafish prey capture

Acceptance summary:

In this manuscript, the authors looked at prey capture behaviors (eye convergence and tail flick) of zebrafish larvae in a virtual reality environment while at the same time they observed the neuronal activity of the brain. Importantly, they focused on the effect of previous experience on the prey capture performance and its neuronal correlates. They compared fish fed live paramecia (experienced) and fish fed fish flakes (naïve), the former of which showed higher response to live paramecia. As a result, authors found the correlate of learning-dependent improvement of prey capture in the brain, and propose a new idea of forebrain involvement in the reflex-like behavior in larval zebrafish. All three major concerns by the reviewers were addressed by additional experiments and textual revisions, supporting and strengthening authors' conclusions.

Decision letter after peer review:

Thank you for submitting your article "Experience, circuit dynamics and forebrain recruitment in larval zebrafish prey capture" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and K VijayRaghavan as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

As the editors have judged that your manuscript is of interest, but as described below that additional experiments are required before it is published, we would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). First, because many researchers have temporarily lost access to the labs, we will give authors as much time as they need to submit revised manuscripts. We are also offering, if you choose, to post the manuscript to bioRxiv (if it is not already there) along with this decision letter and a formal designation that the manuscript is "in revision at eLife". Please let us know if you would like to pursue this option. (If your work is more suitable for medRxiv, you will need to post the preprint yourself, as the mechanisms for us to do so are still in development.)

Summary:

In this manuscript, the authors looked at prey capture behaviors (eye convergence and tail flick) of zebrafish larvae in a virtual reality environment while at the same time they observed the neuronal activity of the brain in a fixed focal plane. Importantly, they focused on the effect of previous experience on the prey capture performance and its neuronal correlates. They compared fish fed live paramecia (experienced) and fish fed fish flakes (naïve).

They first looked at the behavior and found that eye convergence of experienced fish in response to prey was higher than in naïve fish (Figure 1). They then monitored the brain and found that the activity of the visual areas, pretectum and tectum, are not very different between experienced and naïve fish. There was also no difference between neuronal activities at eye convergence events that spontaneously occurred without prey and those at prey-evoked eye convergence (Figure 2, Figure 2—figure supplement 1). To see whether encoding of visual stimuli is different, they performed regularized regression analysis on the neuronal map in the pretectum, tectal neuropil and tectal PVN, and found no difference between experienced and naïve fish (Figure 3). To look at information flow in and out of visual areas, they performed Granger causality analyses. Granger causality between tectum and PVN was increased by presentation of prey, but again, prior experience made no difference (Figure 4). On the other hand, the authors did observe a difference in information flow to the telencephalon from visual areas. Interestingly, looking at individual fish, eye convergence performance was correlated with Granger causality from pretectum and other visual areas, and from visual areas to the telencephalon. This raised the possibility that the forebrain may be involved in increased performance resulting from experience (Figure 5). Triggered by this observation, the authors looked at activity of each area at the time of pretectal activity peaks, and found that responses of the telencephalon and habenular region were different between experienced and naïve fish (Figure 6). To confirm the relevance of brain activity measurements in the habenulae, they performed a functional test to kill habenular neurons using a toxin, and indeed saw decreased success at prey capture (Figure 7).

Essential revisions:

The writing and figures are generally very good, and most of the small-scale interpretations are sound. However, it is the consensus of the reviewers that the following concerns need to be addressed before the study is publishable.

1) There is a simple and appealing (maybe preferable) alternative explanation for the results: flake-fed larvae are more satiated than paramecium-fed larvae are. We acknowledge the great lengths that the authors have gone to in demonstrating that larvae in the two groups are similar in size, move similarly, and have had access to saturating amounts of food. This does not mean that they are equally hungry (flakes are easier to catch and may be more nutritious, as this is what they're designed for). The behavioral effect, an increased initiation of pursuit, appears to be more a function of motivation than hunting proficiency. The authors have demonstrated that the visual coding does not depend on experience, nor is there evidence of improved efficacy (say, improved motor coordination) once an initiation occurs. These seem like more motivated animals. Greater hunger is the simplest explanation, and the authors capably outline in their Discussion the mechanisms by which hunger could mediate this stronger motivation to prey.

We do not think that the authors' model, "experience leads to greater visually-evoked activity in the forebrain", is supported until the effects of hunger and experience are uncoupled. Unfed fish, which would undoubtedly be hungry and naïve, would provide the answer. If they initiate more and show other effects reported in the manuscript, then it is likely that hunger is responsible. If they act more like flake-fed larvae (or are even farther in this direction), then the authors' model has much stronger support. While it would be less than ideal, and would leave caveats, a clear behavioral effect (even in the absence of brain imaging) could be used to provide this support.

2) The habenular ablation experiments support that the habenulae are necessary for normal initiation and capture. This is only circumstantially related to the manuscript's other findings. The solution is to do the ablations of flake-fed larvae, where the model would predict no effect.

3) Subsection “Fluorescence Analysis”. The description of quantifying habenular neuronal loss is confusing. The authors state that "NTR+ fish were fixed in 4% formaldehyde overnight and then washed and mounted in low-melting agarose. Z-stacks of both habenulae were taken on a Zeiss LSM 880 upright laser scanning confocal at equal laser intensity, making sure no pixels were oversaturated. The total volume of expressing cells was quantified.…." However, it is well known that fixation in paraformaldehyde quenches the fluorescence signal. Please clarify how the authors were able to use this signal (mCherry?) to quantify neuronal loss.