Olfactory Disruption Restructures Collective Behavior and Increases Cohesive Group Dynamics

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study examines the contribution of olfactory input to group social organization in mice, addressing a gap in literature that has largely focused on one-to-one social interactions. To address this gap, the authors chemically impaired olfaction in adult mice using two distinct methods: intranasal application of Triton X-100 or intraperitoneal injection of methimazole (MMZ), both of which are well-established approaches to disrupt olfactory function by ablating the olfactory sensory epithelium. After confirming that both methods impaired olfactory function, the authors proceeded to study social dynamics in groups of animals. Four mice of the same sex were introduced into a large open-field arena and allowed to interact over extended periods of time. The behavior of olfactory-impaired animals was video-tracked and compared with control groups treated with vehicles only. Using quantitative analyses and modern behavioral tracking techniques, the authors report an increase in the amount of time animals spent in close proximity, which they describe as a “huddling” behavior. Importantly, the authors further demonstrate that this behavior is not random, but instead reflects the expression of discrete behavioral and locomotor states, including distinct dynamics associated to the “initiators” and “followers”  during the huddling events. Overall, this manuscript presents a novel and conceptually interesting framework for studying group social behavior in mice and is suitable for publication in Biology.  The recommendations below are intended to improve transparency and clarity of interpretation of the findings of the manuscript, and they are just suggested changes

1) The authors could briefly explain and discuss why their analysis focuses exclusively on the first hour of the experiment, despite recordings lasting five hours. As a point of interest, did the animals display similar huddling or aggregation behaviors during the later hours of the recording? The authors could clarify the rationale for selecting only the first hour for analysis and, if no quantitative measures were performed beyond this time window, discuss what they would expect to observe. This point is particularly relevant in the context of social dynamics, which is the topic of this paper. Did the observed huddling behavior remain stable over time, or did it change across hours? Even if formal analyses were not conducted, the authors could include a brief discussion of any preliminary observations and their expectations regarding the temporal evolution of group behavior.

2) Related to the previous point, in the Methods section (“Group social interaction recordings”), the authors state in lines 118-119  that “one day after nasal irrigation or injection, groups of four mice (two males and two females) were placed in a large arena,” and later in the same paragraph mention that behavior was recorded during the dark cycle for five hours starting form 7pm. Please clarify briefly on the methods text whether the recording began immediately after the 4 mice were introduced into the arena, or whether there was a habituation/acclimation period before recording started. Explicitly stating the timing will increase clarity on the behavioral approach especially if other researchers would aim to replicate the interesting findings of this work

3) In Figures 2i and 2j, it appears that the authors plot the calculated minimum pairwise distance among the six dyads for each group. For clarity, please explicitly state in both the plot legend and the figure text that each line corresponds to an individual test groups (the trials). In addition, because many trajectories overlap, it is difficult to follow the time course of the minimum pairwise distance for each group, particularly in the Triton X-100 condition, which includes seven groups. The authors may consider plotting each trajectory in separate panels or alternatively retaining them in a single axis as current version but adding transparency to individual group traces and overlaying a thicker line representing the mean or median to better illustrate the overall trend over time.

4) It would be helpful if the authors clarified on the manuscript, in light of prior studies using large open-fields and/or own controls (if available), whether olfactory-impaired mice are expected to exhibit increased wall or corner occupancy when tested individually without any other mouse around. Such baseline changes in spatial exploration could influence group-level measures, including minimum pairwise distance (Fig. 2), without necessarily reflecting altered social behavior. While this issue is partly addressed by the shuffling analysis in Fig. 3, a clearer acknowledgment of this potential limitation directly in the context of Fig. 2 would improve transparency and facilitate interpretation, without requiring additional experiments.

5) Please double-check the text describing the results in Section 3.2, “Huddling Emerges as a Distinct Pattern of Group Behaviour,” as there appears to be a mismatch between the figure panel references in the text and the corresponding treatment conditions shown in Figure 2.

For example, in lines 245–250 of the current version, the authors state: “In mice administered Triton X-100, snapshots from one example group illustrate animals dispersed early in the session… (Fig. 2a; Video 1).” However, in Figure 2, the Triton X-100 condition appears to be labeled as panel 2b, not 2a.

A similar inconsistency is present for the MMZ condition: the results text refers to Figure 2b, whereas the corresponding MMZ data appear in panel 2d.

Likewise, for control animals, the text refers to panels 2c and 2d, but in the figure these conditions appear to correspond to panels 2a and 2c.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript from Chen et al. address the interesting question of how group dynamics in mice are affected by olfactory dysfunction. They report an interesting novel finding, that mice ‘huddle’ in groups far more frequently and for extended periods, a day after OSN ablation via treatment with either methimazole (MMZ) or Triton-X100. They also include some interesting analysis of behavior leading up to formation of duos. This study is an interesting and important contribution to the field; however, there are some gaps in the analysis and important factors that should be better considered in interpreting the data. These recommended revisions are detailed below:

Major

1. Mouse sex is not considered anywhere in the analysis. Each group of four mice consists of two males and two females, but how sex impacts huddling behavior is not considered. Ideally, the data would be analyzed with sex as a factor (this would seem to only require the sex of each mouse to be known when they are placed in the arena, as each mouse is then tracked individually). If this was a limitation of experimental design, and the sex of the mice in the arena cannot be determined, the authors should at a minimum acknowledge this limitation and discuss the potential role of sex in their data (i.e. could sex have affected likelihood of huddling, initiator vs. follower behavior, etc). 
2. A related point is that the simultaneous formation of two dyads is not considered anywhere in the analysis (see e.g. schematics in Fig. 3m,n). This is probably not an unlikely occurrence given that dyads formed with fairly high probability, although it is notable in the example videos that a dyad plus two individual mice was a common occurrence. This also relates to analysis of sex- perhaps only one sex forms dyads? This seems like an important point that could be better addressed and should be included in the schemes in Fig. 3m,n. 
3. Triads and quads are under-explored in terms of preceding events: the HMM analysis in Fig. 4 that was performed for dyads could be extended to triads and quads. There are many interesting questions here, such as does a single mouse always join a dyad to form a triad, etc. 
4. The differences between MMZ and Triton-X100 in terms of mechanism of action appear to be mis-characterized in the manuscript. While the authors are to be commended for using two different methods for OSN ablation, MMZ and Triton-X100 actually have quite similar effects on the olfactory epithelium, and both cause reversible damage because basal cells are unaffected and can hence repopulate the OE. There are some differences, with Triton-X100 directly damaging OSNs (as well as supporting cells) whereas MMZ is thought to damage supporting cells which results in subsequent OSN cell death. Specific statements that need to be re-written to accurately reflect the literature are: ”Mice received either binasal irrigation with Triton X-100, which temporarily ablates the olfactory epithelium [45, 49], or systemic i.p. injection of MMZ (Fig 1a), an olfactotoxic drug that induces olfactory epithelial injury [50, 51].” (lines 198-200); and “MMZ induces wide- spread olfactory neuron ablation and bulb deafferentation [51], while Triton X-100 transi-ently impairs olfactory signaling [45].” (lines 574-576).
5. While complete ablation of all OSNs after Triton-X100 may be unlikely in all mice given that it relies on intranasal irrigation which is intrinsically variable, system administration of 100 mg/kg MMZ would be expected to ablate virtually all (if not all) OSNs throughout the OE. It was therefore surprising to see that some mice treated with MMZ (and this was also the case for Triton-X100) were able to find buried food two days post-treatment. Based on the literature, all mice would be expected fail this task if OSN ablation is complete (or close to complete). The authors did not discuss this finding, and need to include caveats explaining that OSN ablation was likely incomplete in at least those mice that were able to complete the buried food task. This of course has implications for their group interaction data, and these should be discussed. 

Minor

6. Were experimental mice littermates? How long had they been singly housed? These factors could affect the behavioral outcomes (e.g. huddling might be likely amongst former littermates).

7. Please explain why social behavior was performed one day post-treatment whereas the buried food test was performed two days post-treatment. Using the same time point would have been the obvious choice, especially as different cohorts of mice performed the two tasks. 

8. Typos to correct: line 284 ‘wile’, line 474 ‘laten-state’.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for carefully revising your manuscript in response to my comments. I have on additional concerns.