(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Optogenetic and chemogenetic approaches reveal differences in neuronal circuits that mediate initiation and maintenance of social interaction [1]

['Karolina Rojek-Sito', 'Laboratory Of Emotions Neurobiology', 'Braincity Centre Of Excellence For Neural Plasticity', 'Brain Disorders', 'Nencki Institute Of Experimental Biology', 'Polish Academy Of Sciences', 'Warsaw', 'Ksenia Meyza', 'Karolina Ziegart-Sadowska', 'Kinga Nazaruk']

Date: 2023-12

Funding: This work was supported by the European Research Council Starting Grant (H 415148) to EK (supporting the work of KRS, KM, KZS, KN, AH, and EK) and the BRAINCITY - Centre of Excellence for Neural Plasticity and Brain Disorders’ project of the Foundation for Polish Science to EK (supporting the work of KRS, AP, and EK). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2023 Rojek-Sito et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Using opsins selectively expressed in the neurons activated by social interaction, projection-specific chemogenetic manipulations, and functional tracing, we demonstrate that the neural circuit composed of neurons in the CeA, VTA, ACC, and OFC is critical for promoting social interactions. In particular, some projections within the circuit affect the initiation of social contact (VTA-ACC, VTA-OFC, OFC-CeA), while others are critical for maintaining it (CeA-VTA, ACC-CeA). Presented results provide novel insights into the social brain and lay a foundation for developing therapeutic approaches targeted at specific aspects of social interaction deficits.

The VTA, OFC, and ACC have previously been associated with facilitating positive social interactions [ 2 – 8 , 10 ]. The role of the CeA in driving social interactions remains largely unexplored. Previous studies have linked different subpopulations of neurons within the CeA to the modulation of rewarding experiences and the pursuit of food rewards [ 12 – 17 ]. Furthermore, it has been also demonstrated that oxytocin signaling in the CeA modulates the discrimination of emotions during social interactions [ 11 , 18 ], which is crucial for responding appropriately to the behavior of social interaction partners. The CeA exhibits strong connections with other structures in the mesocorticolimbic circuitry, and its stimulation recruits the VTA, NAcc, and OFC [ 12 , 19 ]. However, the functional distinction of CeA circuits based on their connectivity to different upstream and downstream brain areas remains unknown.

To explore this problem, we directed our attention to specific brain regions implicated in motivation and approach behaviors, such as the VTA and CeA, as well as regions associated with social cognition and executive functions, such as the ACC and OFC. We sought to investigate the causal role of this neural circuitry, characterized by its long-range functional connectivity, in different aspects of social interaction using a rat model. We analyzed multiple dimensions of social interaction, including initiation (reflecting the ability to initiate social contact), maintenance (reflecting the capacity to appropriately respond when a partner initiates social contact), and blocking (representing attempts to avoid social interaction initiated by the partner). By examining these various aspects, we aimed to gain insights into the specific contributions of this circuitry to different facets of social behavior.

Here, we aimed at dissecting whether neural circuits governing the motivation to initiate social contact and to maintain social interaction overlap or are distinct. The initiation of social contact necessitates the capacity to recognize and interpret social cues and signals from others, a motivation to engage in social interactions, and the willingness to physically approach others. On the other hand, the maintenance of social contact relies on the ability to synchronize behaviors, actions, and emotions with others, the reciprocation of social behaviors between individuals, and sustained engagement.

Social species, including humans, are motivated not only by physiological needs such as food, water, or reproduction but also by the need to interact with other individuals and form relationships. The neural circuits promoting social interactions remain poorly understood. Previous studies implicated the neural circuits within the mesocorticolimbic reward system, including the ventral tegmental area (VTA), nucleus accumbens (NAc), central amygdala (CeA), orbitofrontal cortex (OFC), and anterior cingulate cortex (ACC) in social interaction [ 1 – 11 ].

Finally, we tested the role of the ACC-CeA and OFC-CeA pathways in food motivation using progressive ratio test as described above. We found that inhibition of both projections decreased the number of operant responses for food reward, which contrasts with the effects of manipulating the CeA-VTA-ACC/OFC projections ( S5G Fig ).

To test the role of the ACC-CeA and OFC-CeA projections in social interaction, we inhibited either the ACC-CeA or OFC-CeA projections chemogenetically using DIO-hM4Di and Cav-Cre constructs activated by the DREADD agonist C21. We observed that the vast majority of the terminals were located in the places injected with Cav-CRE, confirming the reliability of the chosen method for the purpose of selective projection manipulation (Figs 4E–4G , S5B , and S5C ). Behavioral analysis showed that the ACC-CeA and OFC-CeA differentially control various aspects of social interaction (Figs 4H–4K , S5D , and S5E ). Specifically, we discovered that inhibition of the OFC-CeA pathway primarily impacts the initiation of social interaction. In contrast, inhibition of the ACC-CeA pathway predominantly affects the maintenance of social interaction when compared to the control group (Figs 4H , 4I , 4K , and S5E ). Notably, when either the OFC-CeA or ACC-CeA pathways are inhibited, we observe an increase in attempts by the treated rats to block social interaction. This is evidenced by contact avoidance and pushing the partner away ( Fig 4J and 4K and S1 and S2 Videos). The increased blocking behavior and decreased maintenance of social interaction in rats with an inhibited OFC-CeA pathway, compared to the partner rats, suggest additional challenges in maintaining social contact in the OFC-CeA group. Inhibition of neither the ACC-CeA nor OFC-CeA changed general locomotor activity ( S5F Fig ).

(A) The schematic of the experiment. Co-housed c-fos-PSD95-Venus rats were injected with anterograde tracer into ACC or OFC, separated for 3 weeks, and then subjected to the social interaction test. (B) Number of the CeA neurons that are activated by the social interaction (left), and a number of neurons that receive projections from the ACC or OFC (right); unpaired t test (t(5) = 6.917, p = 0.001, ACC-CeA: n = 3, OFC-CeA: n = 4. (C) Percent of the CeA neurons activated by the social interaction (green) that receive projections from the ACC (left, red), and OFC (right, blue), respectively. (D) Representative image of the c-fos–driven expression of the GFP in the neurons activated by the social interaction. Middle: anterograde tracer (PHA-L)-labeled projections from the ACC to CeA. Right: overlay of the left and middle images. (E) Schematic of the chemogenetic inhibition of the ACC-CeA, or OFC-CeA projections. (F) The schematic of the experiment. (G) The representative images of the AAV-hSyn-DIO-{hCAR}off-{hM4Di-mCherry}on-W3SL expression in the ACC and OFC. (H, I) Inhibition of the ACC-CeA projection disrupts the maintenance of social contact, while inhibition of the OFC-CeA projection impairs the initiation of social contact. (J) Inhibition of both pathways results in active blocking of social contact. Animals injected with c21 were compared with animals from the control group also injected with c21, as well as conspecific partners (white bars) injected with saline. Initiation: one-way ANOVA (group effect: F(5,28) = 4.941, p = 0.0023) followed by Holm–Sidak post hoc tests; Maintenance: one-way ANOVA (group effect: F(5,28) = 8,960, p < 0.0001) followed by Holm–Sidak post hoc tests; Blocking: one-way ANOVA (group effect: F(5,28) = 21.65, p < 0.0001) followed by Holm–Sidak post hoc tests; Ctrl: NaCl/c21 n = 5/5, ACC-CeA: NaCl/c21 n = 6/6, OFC-CeA: NaCl/c21 n = 6/6. ( K) The change in the initiation (I), maintenance (M), and blocking (B) of social interaction, computed as a percentage of the average results of the control group, during inhibition of the ACC-CeA or OFC-CeA projections yielded the following significant results (comparison to no change level, Wilcoxon signed rank test): I: OFC-CeA: p = 0.0313, M: ACC-CeA: p = 0.0313, B: ACC-CeA: p = 0.0313, B: OFC-CeA: p = 0.0313; ACC-CeA:c21 n = 6, OFC-CeA:c21 n = 6, Ctrl:c21 n = 5. Initiation of social interaction: approaching a partner; Maintenance of social interaction: percentage of the positive reactions to partner’s approaches; Blocking of social interaction: pushing back or avoiding social interaction. All the data are shown as the mean ± SEM, and symbols represent individual data points, * p < 0.05, ** p < 0.01, *** p < 0.001. The data underlying this figure can be found in https://data.mendeley.com/datasets/h49vtpjm8f/3 .

Next, we focused on the projections from the ACC and OFC to the social cells in the CeA. To identify the social cells inputs, we injected the c-fos-PSD95-Venus rats, which express a reporter protein under the control of the c-fos promoter, with an anterograde tracer PHA-L to the ACC or OFC ( Fig 4A ). We induced expression of the reporter protein in the CeA by social interaction and then imaged the activated neurons counting how many of them receive inputs from the ACC or OFC (Figs 4B–4D and S5A ). We confirmed that ACC and OFC neurons project onto the social cells in the CeA. Notably, we discovered that the ACC innervates more of the CeA social cells than the OFC.

Finally, to test the specificity of the involvement of the CeA-VTA-cortical circuits in social interaction, we employed the food motivation test ( S4 Fig ). We tested well-trained animals lever pressing for food under a progressive ratio schedule of reinforcement that reflects their motivation to obtain the reward. We found that inhibition of the CeA-VTA projection, as well as activation of the VTA-ACC/OFC dopaminergic pathways had no effect on motivation for food reward, thus confirming that the observed effects were specific to social interaction.

(A) The representative images of the VTA inputs into the ACC (left) and OFC (right) in Th-Cre rats labeled with AAV-DIO-mCherry. The white lines mark the areas of the ACC and OFC. (B) Representative images of (from the left) the TH-positive cells (anti-Th ab) in the VTA (green), expression of AAV-hSyn-frt-hM3D(Gq):mCherry (red), and the merged image of all the former. ( C) The schematic of the experiment. (D) Schematic of the chemogenetic activation of the VTA-ACC and VTA-OFC projections. (E, F) Activation of the VTA-ACC inputs enhances social contact initiation, while activation of neither the VTA-OFC nor VTA-ACC affects contact maintenance. Instead, activation of both VTA-ACC and VTA-OFC decreases maintenance of social contact by the partners of the treated animals. (G) Activation of both VTA-ACC and VTA-OFC pathways also results in active blocking of social contact by the partner rats. Animals injected with c21 were compared with animals from the control group also injected with c21, as well as conspecific partners (white bars) injected with saline. Initiation: one-way ANOVA (group effect: F(5,32) = 5.362, p = 0.0011) followed by Holm–Sidak post hoc tests; Maintenance: one-way ANOVA (group effect: F(5,32) = 7.1897, p = 0.0001) followed by Holm–Sidak post hoc tests; Blocking: one-way ANOVA (group effect: F(5,32) = 6.295, p = 0.0004) followed by Holm–Sidak post hoc tests; Ctrl: NaCl/c21 n = 6/6, VTA-ACC: NaCl/c21 n = 6/6, VTA-OFC: NaCl/c21 n = 7/7. ( H) The change in the initiation (I), maintenance (M), and blocking (B) of social interaction, computed as a percentage of the average results of the control group, during inhibition of the VTA-ACC or VTA-OFC projections yielded the following significant results (comparison to no change level, one-sample t test): I: VTA-ACC: p = 0.0114 (t = 3,898, df = 5), I: VTA-OFC: p = 0.0253 (t = 2,960, df = 6), B: VTA-OFC: p = 0.0058 (t = 4,190, df = 6). Ctrl:c21 n = 6, VTA-ACC:c21 n = 6, VTA-OFC:c21 n = 7. Initiation of social interaction: approaching a partner; Maintenance of social interaction: percentage of the positive reactions to partner’s approaches; Blocking of social interaction: pushing back or avoiding social interaction. All the data are shown as the mean ± SEM, and symbols represent individual data points, * p < 0.05, ** p < 0.01. The data underlying this figure can be found in https://data.mendeley.com/datasets/h49vtpjm8f/3 .

Firstly, using Th-Cre rats, we confirmed that the VTA sends dopaminergic projections to the ACC and OFC ( Fig 3A ). Then, we injected frt-hM3D(gq) into the VTA and CavFlexFlp into the ACC or OFC of Th-Cre rats, allowing for stimulating of the dopaminergic projections with the DREADD agonist C21 (Figs 3B–3D , S3G , and S3H ). We found that the activation of the VTA-ACC and VTA-OFC had a significant effect on the initiation of social contact but not on its maintenance (Figs 3E–3H and S3I ). Stimulation of the VTA-ACC and VTA-OFC pathways increased active approaches toward the partners. However, simultaneously, rats with activated VTA-OFC projections more frequently blocked their partner’s attempts to interact compared to control animals ( Fig 3G and 3H ). These oversocial and inconsistent behaviors during the activation of the VTA-ACC/OFC led to a decrease in positive responses to contact initiations and an increase in avoidance of social interaction in the partner rats (Figs 3F–3H and S3I ). Activation of neither the VTA-ACC nor VTA-OFC affected general locomotor activity ( S3J Fig ).

Next, we examined the colocalization of CeA inputs to the VTA with GAD67, a marker of GABAergic neurons. The analysis revealed a high degree of colocalization, indicating that the cells in the VTA receiving projections from the CeA are predominantly GABAergic ( Fig 2H and 2I ). Further, activation of the CeA social cells led to a decrease of GABA in the VTA ( S2 Fig ). Together, these results suggest that the CeA output to the VTA may inhibit the tonic inhibition exerted by GABAergic interneurons on dopaminergic neurons. If this hypothesis holds true, the CeA-VTA projection should increase the activity of dopaminergic neurons in the VTA. In our next experiment, we investigated the role of dopaminergic projections from the VTA to the ACC and OFC in social interaction.

(A) Experimental protocol: Each of the co-housed pairs was subjected to inhibition of the CeA-VTA projection in one animal, preceded by the injections of the viral vectors: AAV-DIO-H4 and Cav-CRE. The control (Ctrl) animals were injected with the AAV-DIO-mCherry and Cav-CRE delivering viruses. (B) Assessment of social interaction in each subject (red rat) involved Left: Initiation of social interaction: approaching a partner; Middle: Maintenance of social interaction: percentage of the positive reactions to partner’s approaches; and Right: Blocking of social interaction: pushing back or avoiding social interaction (see Methods for more details). (C) The schematic of the experiment. Co-housed animals were separated for 3 weeks and then subjected to the social interaction test. Prior to the test (30 minutes), one of the rats received an IP injection with the chemogenetic activator (c21), and the other with NaCl. (D–F) Inhibition of the CeA-VTA inputs disturbs the maintenance of social contact and leads to active blocking of social contact, but it does not affect the initiation of social contact, compared to the control group. Animals injected with c21 were compared with animals from the control group also injected with c21, as well as conspecific partners (white bars) injected with saline. It is worth noting the increase in social contact initiation by partners of the CeA-VTA:c21 rats. Initiation: one-way ANOVA (group effect: F(3,16) = 7.364, p = 0.0026) followed by Holm–Sidak post hoc tests; Maintenance: one-way ANOVA (group effect: F(3,16) = 8.848, p = 0.0011) followed by Holm–Sidak post hoc tests; Blocking: one-way ANOVA (group effect: F(3,16) = 6.018, p < 0.0060) followed by Holm–Sidak post hoc tests; Ctrl: NaCl/c21 n = 5/5, CeA-VTA: NaCl/c21 n = 5/5. (G) The change in the initiation (I), maintenance (M), and blocking (B) of social interaction, computed as a percentage of the average results of the control group, during inhibition of the CeA-VTA projection yielded the following significant results (comparison to no change level, one-sample t test): M: CeA-VTA: p = 0.0064 (t = 5.221, df = 4), B: CeA-VTA: p = 0.0484 (t = 5.221, df = 4); CeA-VTA:c21 n = 5, Ctrl:c21 n = 5. ( H) Representative images of (from the left): the GABA-positive cells (anti-GAD67 ab) in the VTA (red), projections from the CeA (blue), nuclei of cells (yellow), and the merged image of all the former. (I) Left: Number of the GABA-positive cells in the VTA receiving projections from the CeA (red) and the number of the Gad67-negative cells receiving projections from the CeA (blue). Right: Majority of the CeA-VTA projections are onto the GABA+ cells; paired t test t(4) = 4.028, p = 0.0158; CeA-VTA: n = 5. All the data are shown as the mean ± SEM, and symbols represent individual data points, * p < 0.05, ** p < 0.01. The data underlying this figure can be found in https://data.mendeley.com/datasets/h49vtpjm8f/3 .

We discovered that inhibiting the CeA-VTA pathway disrupts the maintenance of social interaction, specifically affecting positive reactions to the partner’s approach, such as sniffing or allogrooming, and sustained close physical contact. Inhibiting this pathway leads to an increase in active blocking of social contact. At the same time, rats with inhibited CeA-VTA pathways initiate social interaction, approaching their partners as frequently as animals in the control group (Figs 2D–2G , S3D , and S3E ). Interestingly, the inhibition of the CeA-VTA pathway in one rat also alters the behavior of their partners, who attempt to initiate social contact more frequently than the control animals ( Fig 2D ). Despite the increased initiation of social contacts by the partners, rats with inhibited CeA-VTA pathways do not respond to these attempts and actively avoid social interaction. The manipulation did not affect rats’ general locomotor activity ( S3F Fig ). The methods of tracing the CeA-VTA projection we used could also mark the collaterals of the CeA neurons projecting to the brain structures other than the VTA. Thus, to control for the collateral projections of the CeA neurons, we carefully inspected the brains of the injected animals.

Then, we performed the loss-of-function experiments on the CeA-VTA pathway. Using DIO-hM4Di and Cav-Cre constructs activated by the DREADD agonist c21 (Compound 21), we inhibited the CeA-VTA projection during social interaction (Figs 2A–2C , S3B , and S3C ). We performed 3 types of comparisons: with the control group, with a partner (who was treated with saline), and with the baseline (after a short-time separation). Our primary focus for drawing conclusions was the comparison with the control group and analyzing the changes in partners’ behavior. Furthermore, we measured attempts to block the interaction as they represent the negative aspects of maintenance. The comparison with the baseline was conducted to monitor any potential alterations in sociability that may have arisen during the experimental manipulations. The results of these analyses can be found in S3 and S5 Figs.

First, to investigate whether the CeA-VTA projection is activated by social interaction, we used c-fos-PSD95-Venus rats, which express a reporter protein under the control of the c-fos promoter, and injected them with an anterograde tracer (PHA-L: Phytohemagglutinin-L) into the CeA. Our findings revealed a dense projection from the CeA to the VTA cells activated during social interaction ( S3A Fig ).

The photomanipulation experiments suggested that there is some functional overlap between the social and food cells, but it is not complete. We did not observe anatomical segregation between the social and food cells. Therefore, to investigate the factors that differentiate the social and food cells, we focused on identifying the CeA outputs of these cell populations. To that end, we measured the level of neurotransmitters and their metabolites in various brain areas after manipulating the activity of the social and food cells to identify the brain regions functionally connected with the CeA social but not food cells ( S2 Fig ). Comparing the levels of neurotransmitters in different brain structures after the stimulation of food and social cells, we observed changes in neurotransmitter release in the VTA, ACC, OFC, PFC, and NAc following the activation of social cells but not food cells ( S2 Fig ). We directed our focus towards the CeA-VTA-ACC/OFC pathways. This emphasis stemmed from the tracing experiment we conducted, which demonstrated high activity of the CeA-VTA projection during social interactions (see below). Furthermore, the changes observed in the VTA led us to speculate that the dopaminergic projection could be involved. Accordingly, we observed changes in dopamine levels in the ACC.

Then, using the c-fos-NpHR construct, we expressed halorhodopsin in neurons that were activated by either social interaction or lever pressing for food. We repeated the experiment described above, this time photoinhibiting either the social cells or the food cells. We found that inhibiting both food and social cells reduces motivation for the food reward and decreases cage exploration, suggesting some functional overlap between these populations ( S1D–S1G Fig ). Notably, the behavior during sessions of social interaction or lever pressing for food used for induction of c-fos–dependent constructs did not differ between the groups ( S1H–S1K Fig ). All the procedures were the same in the control groups, except that we did not use laser light during the test. The number of social and food cells activated in the CeA was comparable ( S1J and S1K Fig ).

To further test the role of the CeA social cells and their behavioral specificity, we conducted another experiment to examine their involvement in motivation for food reward. Previous work has shown that optogenetic excitation of the CeA amplifies motivation to pursue food rewards in the progressive ratio lever pressing test [ 15 ]. Thus, we expected that if the social and food cells in the CeA overlap, their activation or inhibition will also affect food motivation. To test this, we induced the activity-dependent expression of opsins (specifically, the c-fos-ChR2 construct described above) in 2 groups of rats using the following methods: (a) social interaction (social cells); and (b) lever pressing for food (referred to as food cells; Figs 1I–1K and S1C ). Then, we tested the effects of their photoactivation on motivation for food in the progressive ratio test. Consistently with the previous reports, we found that photoactivation of the food cells increased motivation for food, as measured by lever pressing. At the same time, activation of the social cells decreased it ( Fig 1K ). As the photoactivation of the social cells promotes social interaction with the partner, this suggests that populations of social and food cells do not overlap completely.

(A) The schematic of the experiment. Cagemate rats were separated for 3 weeks and then subjected to the social interaction (SI group). The control group had the social interaction after a brief, 10-minute separation (Ctrl). (B) The SI rats engage in intensive positive interaction. Number of social contacts and 50 kHz (positive) USVs during the 10-minute social interaction (social contacts: SI, n = 8; Ctrl, n = 8; unpaired t test: t(14) = 5.588, p < 0.0001; USVs: SI, n = 11; Ctrl, n = 8; unpaired t test: t(17) = 3.076, p = 0.0068). (C) Interaction with a partner activates the CeA cells. Representative images of c-Fos expression in the CeA in the SI and Ctrl groups. (D) Quantified c-Fos expression in the CeA (SI, n = 8; Ctrl, n = 8; unpaired t test: t(14) = 11.63, p < 0.0001). (E) The schematic of the photomanipulation experiments—reactivation of social cells during social interaction. (F) Social interaction in the SI group induces strong expression of the optogenetic construct, measured as the YFP fluorescence colocalizing with the endogenous c-Fos (Social cells, n = 3). The baseline expression measured in animals with no exposure to social interaction after 3 weeks of social separation (Ctrl, n = 3). (G) Photoactivation of the CeA cells activated in the SI group increases social contact. Percentage of social contacts during the optogenetic reactivation of the social cells (“light on”, 3-minute period) and without stimulation (“light off”, 3-minute period) normalized to the Ctrl group; paired t test (t(5) = 17, p < 0.001), n = 6. (H) There is no difference in the control group without laser stimulation (“on” 3-minute, “off” 3-minute, n = 7). (I) The schematic of the photomanipulation experiments—reactivation of social cells and food cells during food motivation test. (J) Representative images of the c-fos–dependent expression of the excitatory ChR2-YFP opsin in neurons activated by the social interaction (social cells). (K) Activation of the food cells increases lever pressing, while activation of the social cells decreases lever pressing (compared to baseline); controls show no difference between the baseline and following phases; two-way ANOVA (time × group effect: F(6,52) = 8.052, p < 0.0001), followed by Holm–Sidak post hoc tests. The average number of lever presses per minute during the baseline period (2 minutes), and ON–OFF (3 minutes) laser periods are shown. Ctrl: Social cells: n = 7, Ctrl: Food cells: n = 7, Activation: Social cells: n = 7, Activation: Food cells: n = 9. CeA, central amygdala. All the data are shown as the mean ± SEM, and dots represent individual data points, * p < 0.05, ** p < 0.01, *** p < 0.001. The data underlying this figure can be found in https://data.mendeley.com/datasets/h49vtpjm8f/3 .

To induce a robust drive for social interaction, we separated male rat cagemates [ 20 , 21 ] for 3 weeks. In rats, such procedure induces strong social motivation. Indeed, when the rats got reunited, they engaged in intensive positive interactions without signs of agonistic behaviors. Consistently, the rats produced positive 50-kHz ultrasonic vocalizations (USVs), but they did not produce aversive 22-kHz USVs (Figs 1A, 1B , and S1A ). We found that interaction with a partner activates some CeA neurons as measured by the c-Fos expression (henceforth, we will call them “social cells”; Figs 1C, 1D , and S1B ). To verify the role of the c-Fos–expressing neurons in social interaction, we used the behaviorally driven c-fos–dependent expression of channelrhodopsin [ 18 ]. We injected the AAV-c-fos-ChR2 virus and implanted optic fibers into the CeA of the experimental rats. Next, as previously described, we separated the animals for 3 weeks upon which the animals were reunited, which induced the c-fos–dependent expression of the ChR2 in social cells. To reactivate the neurons recruited by the social interaction, we stimulated them optogenetically on the next day in the presence of the partner rat ( Fig 1E–1H ). In the control group, the rats were treated the same way except for laser being off during the social interaction on the second day. Increase in the activity of social cells caused by photostimulation caused a marked increase in social contact, indicating that the targeted population of the CeA cells is causally involved in social interaction.

Discussion

In this study, we conducted specific neuronal manipulations to investigate the role of the CeA-VTA-ACC/OFC-CeA circuit in social interaction. Successful social interaction relies on 2 fundamental conditions. Firstly, individuals must display the motivation to initiate interaction and an ability to recognize and interpret social cues and signals from others in order to initiate social contact. Secondly, they need the skills and capability to sustain the interaction for a time sufficient to synchronize behaviors, actions, and emotions with others. Consequently, when analyzing the behavioral outcomes of the manipulation of such circuit activity, we comprehensively examined various aspects of social interaction, encompassing both its initiation and maintenance. By doing so, we aimed to gain insights into the distinct neural circuits that underlie the diverse skills required for initiating and maintaining social interaction. Furthermore, we classified avoidance of social interaction and pushing back the partner rat as a third category of social interaction, referred to as “blocking.” This category represents the unwillingness or aversion to engage in social contact and can be considered an attempt to prevent both the initiation of social interaction by others and its maintenance.

Our study provides evidence of the impact of manipulating the OFC-CeA, VTA-ACC, and VTA-OFC projections on the initiation of social contact. In contrast, manipulation of the ACC-CeA and CeA-VTA projections primarily affects responses to partners’ attempts to initiate social interaction. This observation highlights the functional specificity of these circuits in the social domain (Fig 5).

To further investigate the circuits’ involvement in social interaction, we conducted a comparative analysis with food motivation. Our aim was to determine whether these circuits play a general role in motivation or if their function is specific to social interaction. Surprisingly, our findings reveal that manipulating the CeA-VTA-ACC/OFC projections has no discernible impact on food motivation. However, it exerts a specific influence on social interaction. These results underline the specialized nature of these circuits in the social brain.

Social interactions are inherently rewarding experiences that elicit approach behaviors towards other individuals [22,23]. Previous research has implicated several brain structures, including the VTA, OFC, and ACC, which are all part of the reward system, in mediating positive social interactions [2–7,10,17]. However, the specific connectivity between these brain areas and other components of the circuit involved in social interactions remained largely unknown. In our study, we discovered the crucial role of neuronal circuits within the CeA in mediating social interaction. These circuits can be characterized by their inputs from the OFC and ACC, as well as their outputs to the VTA, OFC, and ACC.

The CeA is part of the reward system, and its role in processing food and drug rewards is well established [14]. Previous research has shown that optogenetic activation of the CeA amplifies motivation to pursue food and drug rewards [15,24] or can induce maladaptive attraction to the painful stimuli [12]. Additionally, recent work has implicated the CeA in processing information about the affective states of others [11,18]. Here, we show that photoreactivation of the CeA neurons previously engaged in positive social interaction increases social contact. Additionally, further specification of the CeA social population shows that reactivation of these cells suppresses motivation for pursuing food, in contrast to the reactivation of the CeA cells previously engaged in food reward. Suppression of food motivation during activation of social cells in the CeA suggests that social cells block food cells, promoting social contact over food. Together, these results show that the populations of social and food cells are, at least partially, functionally distinct. In line, we find that stimulation of social but not food cells in the CeA modulates neurotransmitter release in the VTA, OFC, and ACC. We further confirmed the involvement of these projections specifically in social interaction in functional studies.

The amygdala has long been considered a part of the “social brain,” and its abnormality has been linked to deficits in social behavior [25]. However, the structure is heterogeneous, comprising the cortex-like basolateral part and striatum-like part, including the CeA. Previous studies showed that the projections from the basolateral part inhibit social interaction [26,27]. Less is known about which circuits promote social contact. We show that the CeA, which mediates reward motivation, plays such a role. Recently, the medial amygdala–hypothalamus circuit has also been implicated in social reward [28]. As the CeA receives projections from the medial amygdala [29], it is feasible it integrates information from the cortex and medial amygdala in this context.

The VTA, a key source of dopamine in the cortex, plays a crucial role in encoding motivation for social interactions [30,31]. It is a heterogeneous region composed of various cell types, including neurons that release dopamine (DA), GABA, or glutamate [31,32]. The activity of DA neurons is regulated by GABAergic interneurons [31]. Our findings demonstrate that the CeA-VTA projection, which primarily targets GABAergic neurons in the VTA, mediates the maintenance of social interaction, likely through disinhibition of DA neurons in the VTA. However, further studies are necessary to elucidate the interactions between CeA projecting neurons and VTA GABAergic and DA neurons that drive social behavior.

Previous studies have demonstrated that the processing of social stimuli involves dopamine receptors [33] in the ACC and OFC [34]. In particular, recent research suggests that the dopamine system in the OFC may play a crucial role in the development of social anxiety disorder [34]. Furthermore, studies have indicated that the participation of dopamine D2 receptors in the ACC is essential for processing social information during social observational fear learning in animals [33]. Consistent with these findings, the specific stimulation of dopaminergic VTA-ACC or VTA-OFC projections resulted in an increased initiation of social interaction, highlighting the critical role of these projections in modulating social behavior. Interestingly, this heightened sociability persisted despite the partners’ attempts to avoid social contact, indicating potential challenges in recognizing and interpreting social signals from others in rats with overactive VTA-ACC/OFC pathways.

Furthermore, our findings revealed that both the ACC and OFC cortical regions, known to be involved in social decision-making [35–39], project to CeA neurons that are activated during social interaction with a partner. Inhibition of the OFC-CeA projection disrupted the initiation of social interaction, while inhibition of the ACC-CeA pathway disturbed its maintenance. The anterograde transport tracers used to track the ACC-CeA and OFC-CeA projections were injected into the ACC and OFC of separate animals, thereby preventing a direct comparison. However, our findings indicate that the ACC innervates a greater number of social cells in the CeA compared to the OFC. This suggests the possibility that certain cells may receive innervation from the ACC but not the OFC. However, a systematic comparison is necessary to gather evidence regarding whether the same CeA neurons receive projections from both the ACC and OFC.

The specific neuronal circuits responsible for the initiation and maintenance of social interaction had not been previously investigated. Our results provide evidence that distinct neuronal circuitry mediates the different abilities required for initiating and maintaining social contact. Additionally, we demonstrate that manipulating neuronal projections involved in the initiation and/or maintenance of social interaction can impact not only the behavior of the treated animals but also the response of their interaction partners. This can manifest as the discontinuation of interaction and aversion to social contact, potentially indicating abnormalities in the behavior of the treated rats.

Most of the previous work on the CeA circuits focused on the neuronal populations identified by the molecular markers they express. This approach revealed reciprocally connected neuronal circuits that control specific, often opposite behaviors [13]. However, in contrast to the inputs and outputs, the markers usually do not define the function of the neuronal population. Notably, the marker-defined populations are often heterogeneous and can control many behaviors [13]. Thus, our study defined cell populations by their functional connectivity rather than the markers they express.

Understanding the specific circuits involved in initiating and maintaining social interaction can offer valuable insights into neurodevelopmental disorders such as autism spectrum disorder (ASD). The absence of social-seeking tendencies observed in individuals with ASD can be attributed to deficits in social motivation. Research has revealed that infants with a genetic predisposition to ASD exhibit diminished social motivation and interest. This deficiency fundamentally alters how individuals with ASD engage with and perceive the world, resulting in a lack of crucial opportunities to develop social perceptual and social cognitive abilities. Impaired social cognition and skills can thus be considered secondary effects stemming from reduced social motivation, potentially involving distinct neuronal circuits [40–42]. By addressing the core deficits in social motivation, there may be a possibility to enhance the development of social cognition and skills in individuals with ASD.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002343

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.