Figure 1. a-d, Raster plots of four continuous recording for 400s - 600s of the same cell during successive exposure to test animals of the sex and strain indicated to the left. The colored lines and arrows above and below the raster plots mark the behavioral episodes annotated manually. e-h, Histograms showing the average firing rate during each behavioral episode. Note the cell showed opposite responses to male and female stimuli.
Figure 2. A coronal section of the mouse brain at Bregma level -1.4 mm showing the location of the ventrolateral part of the ventromedial hypothalamus (VMHvl). Image is based on Allan Brain Atlas.
Aggression is an essential social behavior across the animal kingdom including humans. Since this behavior requires no learning, the underlying neural circuit is believed to be hardwired in the brain. Attempts trying to identify the brain substrates mediating aggression go back to 1920's. Early electric stimulation and lesion studies in monkeys, cats, rats and hamsters have identified the medial hypothalamus as critical to aggressive behavior, but the fine anatomic structures remains unclear (Aggression, Violence and Brain, an interesting YouTube movie). Our current and future research is aimed at understanding the aggression circuit in a genetically tractable model system, mice. Genetic studies in mice have identified dozens of genes which cause defects in aggression when disturbed, but the functional studies in mice are largely lacking. Through gain and loss of function manipulations and chronic recording in awake behaving mice (Figure 1), we have identified a small subnucleus in the hypothalamus, the ventrolateral part of the ventromedial hypothalamus (VMHvl, Figure 2), as an essential aggression center in mice. The future study is aimed at expanding our investigation on mouse aggression circuit from the VMHvl in multiple directions.
Identify other relays in the mouse aggression circuit
Other relays in the aggression circuit remain unclear. VMHvl forms an intricate network with over 20 different brain regions. To understand how the sensory information flows and transforms to elicit aggressive behaviors, we will systematically manipulate the connections between the VMHvl and its upstream and downstream targets and observe the behavioral change. We will further exam the cell activities in each potential relays using in vivo chronic recording.
Aggression circuit in females
Social behaviors between males and females differ in their motivation, execution and intensity. Even in human society, the frequency of violence within females is much lower than that in males. While electric stimulation of the similar region in the medial hypothalamus in female rats is able to elicit aggression as in male rats, the VMHvl is known to be sexually dimorphic and indispensable to the female sexual behavior. It is thus interesting to understand whether the aggression circuit in males and females is the same or not. If they indeed differ, is the difference quantitative or qualitative?
Genetic dissection of the aggression circuit
The biggest advantage of studying social behaviors in mice is the opportunity to relate genes to behaviors. Many genes involved in various neurotransmitter neuropeptide and hormonal systems are known to affect aggression but the exact mechanism is unclear. With the knowledge of the important relays of the aggression circuit, we can now exam the effect of the genetic manipulation on the circuitry by in vivo recording in genetically modified animals. Furthermore, combining optogenetics and recording, we could ask whether and how each genetically defined population is involved in aggression. Interestingly, our initial result indicates that aggression and reproduction circuits are topographically intermingled and likely mutually inhibitory (Figure 3). Genetically labeling and manipulating each population will allow us to understand their interaction in further details.
Figure 3. Fos cellular compartment analysis of double fluorescent in situ hybridization (catFISH) analysis shows that fighting and mating induced fos expression in many similar regions, but they recruit largely distinct partially overlapping sets of neurons in the same brain regions