RESEARCH SUMMARY
Social behaviors are exhibited by a wide range of animal species and are of ubiquitous adaptive value; many social behaviors, such as aggression, pair bonding, and mating, are essential for the health, survival, and reproduction of animals. The control of social behavior is of vital importance in social species such as humans; impairment in social function is a prominent feature of several neuropsychiatric disorders, such as autism and schizophrenia. However, despite its importance, many fundamental questions regarding the neural mechanisms underlying social behavior and its disorders still remain unanswered. One central question is how neural circuits, important computing units in the brain, process and integrate information for the decision and execution of specific social behaviors. To address this question, the Hong lab applies a multidisciplinary approach to investigating social behavior, using techniques including but not limited to:
- Single-cell RNA-seq and bioinformatics to comprehensively identify and characterize diverse neuronal cell types in social behavior circuits and their molecular profiles with single-cell resolution.
- Optogenetics and chemogenetics to manipulate neuronal circuit activities in different neuronal subtypes in different social behaviors, to determine their causal roles, and to map their functional connections.
- Fiber photometry, miniscope imaging, and neural computation to map neural network activities during social behaviors and to uncover the encoding of sensory processing and behavioral decisions in neuronal ensembles.
- Advanced engineering technology and machine learning to make the analysis of social behaviors more efficient, objective, quantitative, and ethologically relevant.
Summary of recent work:
1) Neural circuit mechanisms underlying social behaviors
An important brain region implicated in the control of social behaviors is the medial amygdala (MeA), but little is known about how MeA exerts this control. Using cell-type specific functional manipulations, we identified two neuronal subpopulations in the posterior dorsal subdivision of MeA (MeApd) that separately control social behaviors and repetitive self-grooming (Hong et al, Cell 2014). Interestingly, we found that the GABAergic subpopulations promote social behaviors including attack, mounting, and social grooming, whereas glutamatergic subpopulations evoke repetitive self-grooming, a nonsocial behavior not previously associated with the MeA. Moreover, this glutamatergic subpopulation also inhibits social behaviors, independently of its effect to promote self-grooming. Conversely, the MeApd GABAergic subpopulation suppresses self-grooming, even in a nonsocial context. Together, these data suggest that social vs. repetitive nonsocial behaviors are controlled in an antagonistic manner by inhibitory vs. excitatory amygdala subpopulations. These findings provide a framework for understanding circuit-level mechanisms underlying opponency between innate behaviors, with implications for their perturbation in neuropsychiatric disorders.
An important brain region implicated in the control of social behaviors is the medial amygdala (MeA), but little is known about how MeA exerts this control. Using cell-type specific functional manipulations, we identified two neuronal subpopulations in the posterior dorsal subdivision of MeA (MeApd) that separately control social behaviors and repetitive self-grooming (Hong et al, Cell 2014). Interestingly, we found that the GABAergic subpopulations promote social behaviors including attack, mounting, and social grooming, whereas glutamatergic subpopulations evoke repetitive self-grooming, a nonsocial behavior not previously associated with the MeA. Moreover, this glutamatergic subpopulation also inhibits social behaviors, independently of its effect to promote self-grooming. Conversely, the MeApd GABAergic subpopulation suppresses self-grooming, even in a nonsocial context. Together, these data suggest that social vs. repetitive nonsocial behaviors are controlled in an antagonistic manner by inhibitory vs. excitatory amygdala subpopulations. These findings provide a framework for understanding circuit-level mechanisms underlying opponency between innate behaviors, with implications for their perturbation in neuropsychiatric disorders.
To interrogate neuronal cell type diversity in brain circuits controlling social behavior, we have recently successfully applied scRNA-Seq to uncover the cell type composition in the MeA and have identified multiple transcriptionally distinct neuronal subpopulations (Wu et al, Neuron 2017). Furthermore, we have developed a new scRNA-Seq method, which allows us to systematically map molecularly defined neuronal subpopulations that are activated by specific behaviors and/or experiences. These advances have provided powerful tools for revealing the molecular identify of behaviorally relevant neuronal cell types and abundant molecular markers to facilitate further functional probing of specific neuronal subpopulations in the medial amygdala.
2) Computer vision and machine learning technologies for automated tracking of social behaviors
A lack of automated, quantitative, and accurate assessment of social behaviors in mammalian animal models has limited progress toward understanding mechanisms underlying social interactionsand their disorders such as autism. We developed a new integrated hardware and software system that combines video tracking, depth sensing, and machine learning for automatic detection and quantification of social behaviors involving close and dynamic interactions between two mice in their home cage (Hong et al, PNAS 2015). We designed a hardware setup that integrates traditional video cameras with a depth camera, developed computer vision tools to extract the body “pose” of individual animals in a social context, and used a supervised learning algorithm to classify several well-described social behaviors. We validated the robustness of the automated classifiers in various experimental settings and used them to examine how genetic background, such as that of BTBR mice (a previously reported autism model), influences social behavior. Our integrated approach allows for rapid, automated measurement of social behaviors across diverse experimental designs and also affords the ability to develop new, objective behavioral metrics
A lack of automated, quantitative, and accurate assessment of social behaviors in mammalian animal models has limited progress toward understanding mechanisms underlying social interactionsand their disorders such as autism. We developed a new integrated hardware and software system that combines video tracking, depth sensing, and machine learning for automatic detection and quantification of social behaviors involving close and dynamic interactions between two mice in their home cage (Hong et al, PNAS 2015). We designed a hardware setup that integrates traditional video cameras with a depth camera, developed computer vision tools to extract the body “pose” of individual animals in a social context, and used a supervised learning algorithm to classify several well-described social behaviors. We validated the robustness of the automated classifiers in various experimental settings and used them to examine how genetic background, such as that of BTBR mice (a previously reported autism model), influences social behavior. Our integrated approach allows for rapid, automated measurement of social behaviors across diverse experimental designs and also affords the ability to develop new, objective behavioral metrics
References:
Wu YE, Pan L, Zuo Y, Li X, Hong W^. (2017) Detecting Activated Cell Populations Using Single-Cell RNA-Seq. Neuron. 96, 313-329. [ PDF | Highlighted by Neuron | UCLA News ]
Hong W, Kim DW, Anderson DJ. (2014) Antagonistic control of social versus repetitive self-grooming behaviors by separable amygdala neuronal subsets. Cell. 158, 1348-1361. [ PDF | Previewed by Nature Review Neuroscience | Caltech News ]
Hong W, Kennedy A, Burgos-Artizzu XP, Zelikowsky M, Navonne SG, Perona P, Anderson DJ. (2015) Automated Measurement of Mouse Social Behaviors Using Depth Sensing, Video Tracking, and Machine Learning. Proc. Natl. Acad. Sci. USA. 112, E5351-5360. [ PDF ]
