Non-Dose-Dependent Relationship between Antipredator Behavior and Conspecific Alarm Substance in Zebrafish

Round 1

Reviewer 1 Report

Abstract. Several procedures have been developed to study anxiety-like behaviors in fish, such as the open field test or scototaxis test. I would not be so specific right away in the abstract, rather I would say that there are several tests and among them diving is mentioned (as described in lines 58-60).

Although it is commonly used the term “anxiety” in animals, this term refers to a human pathology. We did not fully comprehend the mechanism of such neural pathology in human, and even less is known in non-human animals. We have similar behavioral responses among specie on stressful condition that mimics human state. I would suggest changing the term “anxiety” to “anxiety-like behavior”.

Lines 12-14 and discussion: the use of these tests is based on a clear anxiety response, so this is optimized by creating a highly stressful situation. It is not necessary to use a substance to trigger the response, as the fish is placed in a new and potentially dangerous environment. The administration of substances is accessory or used for other purposes, for example recognition of chemical cues (i.e., nor test), or triggering a response after habituation with a different setup (i.e., flat tank). Therefore, the investigation of the substance gradient necessary to trigger the response may not be an appealing motivation.

 

Line 86: novel tank diving test

Experimental protocol.

I would have expected to have an additional control condition where subjects were placed in a solution-free baker and then tested. This would have been a control group to use as a baseline. Furthermore, I would have made further samples from a group of fish kept in the experimental tanks, given the effect caused by the double exposure.

I would like the choice of exposing the subjects to two manipulations to be justified. I would have recorded the exploration activity within the new environment and then how this had changed after administration of the alarm substance

Do you have any reference or motivation for choosing 7 minutes of recording?
In the literature, it is commonly measured the first 5-10-20-30minutes, and then an investigation of temporal variations in exposure conducted, or even the first blocks of minutes are considered where the response is expected to be higher.

 

Results

There is no mention of whom the control group is.

Line 153: The authors used a nonparametric test, but report a value of the F distribution (kruskal-wallis uses K). Alternatively, the data can be transformed, the most commonly used is log-transformation or square-root transformation for right-skewed data, and use ANOVA if the assumptions are met. I think it is necessary to review all the statistics anyway.

 

In the response variables “freezing behaviour” and “latency to upper half” there appears to be a linear relationship with dilution. One way to study this relationship could be to use a correlation for each response variable and the dilution ("pearson" method is more robust in case of repeated values as shown in the variable "dilution"); or a generalized linear model in which polynomial contrasts are placed on the "dilution" variable and the linear effect of this variable on the two observed behaviors is studied.

 

In the novel tank diving test literature, stressed fish explore the bottom of the tank less and spend more time on the surface. Conversely, the data shown in this paper suggests otherwise.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper presents a very detailed study on the relationship between alarm substance exposure and  zebrafish behavioural responses. Well executed if I understand correctly what was the number of fish detailed below.

I have some questions regarding the methodology. In particular concerning the total fish number not mentionned. You have used 10 fish to prepare the solution and then 15 naive fish to test each treatment : is it correct ? You are stating that you used 15 fish: is it per treatments ? (6 dilution and one control) = 15*7= 105 fish + 10 = 115 fish in total ?  Please enter the information in manuscript.

In section 2.4, I think you should include the protocol followed for the control fish - it is only explain in legend of Fig. 2.

In the results section : since you have stated that you use male and female fish --> did you test for any significant differences between sexes ? That might be usefull for other research groups.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Li and colleagues followed my suggestions and proceeded to include them in the new version of the manuscript. The result is outstanding and the manuscript has improved chiefly, especially the section concerning data analysis. If thought possible, I would ask the authors to add in the manuscript additional regarding the explorative behavior in fish and the methodologies developed to study such behaviors that authors have already reported in the rebuttal letter (see response 3 to reviewer's #1 comment). From my concern, the reading, and scientific soundness would improve the introduction and the causes that authors were motivated to study.  

Overall, I consider the manuscript as a potentially good contribution to the field and suitable to be published in the journal

Author Response

Thank you for the comments and suggestions. The description of explorative behavior and the relevant methodologies has been added to the introduction.