Escribir un comentario

Social isolation is a long-lasting stressor that can negatively shape social behavior, including promoting aggression. How these persistent internal states are managed within the brain has been a major focus in neuroscience research. Peptidergic signaling mechanisms, such as Tachykinin, regulate both social behaviors and the effects of social isolation across species. Yet the exact way in which this peptidergic signaling functions with other local circuit mechanisms to coordinate isolation-dependent changes in aggression remains unknown. 

Employing a mouse model of chronic isolation across sexes, Gatlin et al. revealed that Tachykinin-2 expressing (Tac2+) GABAergic neurons in the medial prefrontal cortex (mPFC) regulate both pre-consummatory (social investigation) and consummatory (attack) phases of aggression. A combination of single nucleus RNA sequencing, spatial transcriptomics, and circuit manipulations further characterized these neurons and revealed that the neuropeptide encoded by Tac2 (Neurokinin B) coordinates investigative behaviors, while GABA promotes attack behaviors through releasing inhibition. In addition to implicating Tac2+ cells in layers 1-3 of the cortex and performing a commendable amount of data collection across sexes, the authors clearly describe the context and motivation for their experiments. This work is important for our understanding of how slower peptidergic signaling coordinates with fast-acting neurotransmitters to modulate social state progression and the strength of evidence supports their conclusions. However, there are limitations in the definition of the neuronal population involved and behaviors as well as a need for more consistency in the presentation of data across sexes. One major caveat of this peer review that should be noted is the lack of access to the supplemental data in the current version of the preprint.

Major points:

1. Throughout the text, the authors claim that Neurokinin B (NkB) and GABA are co-released from “the same population of cells.” Their work on characterizing the Tac2-expressing cells and eliminating other major peptides demonstrates that Tac2+ neurons with cell bodies in layers 1-3 and projections in layer 5 appear to be involved. However, there is a disconnect between the cells characterized in Figure 3 and the manipulations performed in Figure 5 and 6. Based on the images shown in panels 5B and 6B, the Tac2+ and GABAnergic populations manipulated appear more distributed than refined to the exact same subset. To this reviewer’s knowledge, there is a current lack of precise tools needed in mouse models to identify if co-transmission is occurring, with a specific neuropeptide and neurotransmitter being released by one cell type at the same time. Such technical limitations makes the claim that co-release by exact same neuronal population accounts for changes in behavior difficult to prove since alternative hypotheses, such as multiple similar populations or different vesicle release timing, cannot be eliminated. The author should clarify these limitations and adjust their terminology throughout the text to account for this. Further, the authors should also consider changing the title to account for these limitations. One possibility is “Opposing signals from cortical neurons modulate distinct phases of aggression.” Additional quantification of the cell types effected in each layer of the mPFC in their Tac2knockdown and vGAT mutagenesis experiments in Figures 5 and 6 would also assist with improving the strength of these claims.

2. The authors define investigative behaviors prior to the first attack bout as aggressive escalation, which is first mentioned in the last paragraph of the results section entitled “Prolonged social isolation induces a state of aggression comprised of attack behavior and social behaviors escalating to attack.” However, it is unclear if these do constitute “escalatory aggression” or an increase in investigation. To clarify this terminology and claim, it would be important to quantify the behaviors that the control animals exhibit in response to such social investigative behaviors. For example, if these investigative behaviors constitute aggressive escalation, the control animals might move away from the socially isolated mice at a higher frequency.

3. One of the strengths of the work is that the authors performed their experiments across sexes and thereby identified consistent changes in aggressive behaviors following isolation. In the previous version of the manuscript, the authors separated out the majority of this data. The current version clearly distinguishes male and female animals in some figures, but in others it is either difficult to see or clearly not stated. Therefore, it remains unclear how consistent the behavior and role of Tac2+ neurons is across sexes in this current version. Below are points that would assist in clarifying the consistency across sexes:

a.    For Figure 1G, the authors should state if both sexes are used for this transition probability and if there is variability across sexes as seen in the first version of the manuscript. It would be helpful to note if any variability found across sexes is similar to that found across individuals.

b.    For Figures 1B, 1K-M, 2H-L, 4C-F, 5C-F, it is very difficult to distinguish between the light and dark colors that are showing the males and females. This makes it difficult to note if there are separate effects across sexes. The authors should choose a different color scheme (i.e. those from Color Universal Design to assist color-blind individuals) or shapes to distinguish males and females.

c.     For Figures 2A – E and 2H – K, the authors do not distinguish between male and female samples. Additionally, the authors do not state the sex of the animals used in Figure 3 in the figure legend. It is therefore unclear if there are different distributions of Tac2+ cells in the mPFC layers across sexes as well as differences in persistent activity in these neurons. The authors should detail the sexes used in the figure legend or figure and if any variability exists.

Minor points:

1. The authors switch from measuring total duration, transition probability, and proportion of time in Figure 1 to only measuring total duration in Figures 5 and 6. It is unclear why total duration is selected and why the authors do not continue to analyze transition probability and proportion of time. Since the transition probability helped to identify the escalatory phase of aggression, it would be important to analyze if there are changes in transition probability even though the total duration does not differ in Figure 6.

2. The authors investigated the gene encoding the Tac2-specific Neurokinin 3 receptor, Tacr3. During this investigation, they state that the Tacr3+ neurons were also activated in socially isolated animals during the resident-intruder assay (Figure S2L – O). It is unclear if receptor activation mediates the “releasing the breaks” on aggression. To clarify this point, it would be important for the authors to state if the number of Tacr3+ cells scaled with attack behavior.

3. The graphics in Figures 4G, 5G, and 6G are helpful but difficult to understand in their current form. One suggestion would be to redraw the diagram as a group of neurons with vesicles releasing either GABA or NkB.

4. To improve reproducibility, the authors should provide more detail on both the exact parameters for analysis of their single nucleus RNAseq experiments and code for calculating transition probability instead referring to references in the methods section.

5.  Detailed below are two minor points not central to the author’s claims but may be of further interest.

a.    In the “Social isolation induces an aggressive state comprised of mutually exclusive actions” section of the discussion, the authors state “…these two classes of behavior tile the trajectory of an aggressive encounter, consistent with the ideas that once animals enter the attack phase, other behaviors are shut down.” As the authors have shown disruption of GABA inhibits entering into attack, it would be interesting for the authors to examine the other behaviors that arose in the absence of attack.

b.    In the same section of the discussion, the authors suggest face interaction is “controlled by a distinct signaling system operating parallel to Tac2 and sensitive to isolation.” Following from this, it would be interesting for the authors to elaborate more on if they think that the body and anogenital investigation is due to other sensory information that these mPFC neurons receive.  

Competing interests

The authors declare that they have no competing interests.