The role of the hippocampal CA2 subfield and CA2 Input-Timing-Dependent Plasticity (ITDP) for social fear and social fear specificity

Animals are capable of successfully navigating through and adapting to a complex and continuously changing environment. One of the most fundamental capabilities necessary for such a successful navigation is the ability to learn, i.e. to persistently alter a behavior, for instance to avoid aversive and pursue rewarding stimuli. Dysfunctions related to learning are in turn associated with a wide range of cognitive disorders (Bliss et al., 2014). It has been proposed that one cellular substrate for learning and memory storage and hence for persistent changes to behavior is the modification of synaptic efficacy. Correlative and synapse-specific mechanisms underlying synaptic plasticity are commonly termed 'Hebbian' (Hebb 1949; Abraham et al. 2019). While the Hebbian postulate of 'firing together wiring together' has been tremendously fruitful, the link between synaptic plasticity and behavior has remained elusive. Furthermore, the Hebbian postulate has been challenged by experimental results indicating other forms of plasticity at play (Földiak, 1990; Dudman et al., 2007; Basu et al., 2016; Bittner et al., 2016). One form of non-Hebbian synaptic plasticity which promises new insights into the mechanisms underlying learning and memory formation is the input-timing-dependent plasticity (ITDP; Dudman et al. 2007). ITDP is a heterosynaptic form of synaptic plasticity, where repeated 20ms-interval pairings of long-range entorhinal cortex (EC) projections and CA3 Schaffer collaterals (SC) were initially shown to potentiate the SC to hippocampal CA1 synapse. Strikingly, the long-range EC projections only evoke subthreshold changes in the postsynaptic cell. This has given rise to the idea that these distal synaptic inputs serve as an instructive signal which gate synaptic plasticity (Dudman et al. 2007). Accordingly, this mechanism could serve as a way to gate plasticity for reward-based or unpredicted signals in the animal’s environment, thus bridging the gap between synaptic plasticity and behavior. Recent work has demonstrated that ITDP is also at play in the hippocampal subfield CA2 (Leroy et al., 2017) and hence potentially serves as a general plasticity rule. Targeted blockade of specific dorsal CA2 ion channels within a mouse model of the human 22q11.2 microdeletion (the strongest known genetic risk factor for schizophrenia) as developed in the Siegelbaum lab has been demonstrated to rescue social memory (Donegan et al., 2020). Dorsal CA2 has been shown to be essential for social memory (Hitti and Siegelbaum; 2014; Donegan et al., 2020; Oliva et al., 2020), the encoding of contextual information in general and social contexts in particular (Donegan et al. 2020) and social aggression (Leroy et al., 2018). A natural role for CA2 would accordingly be its involvement in social defensive behaviors. Such a role of dorsal CA2 is, however, yet unknown. As CA1 ITDP has been shown to be relevant for fear specificity (Basu et al., 2016), CA2 ITDP might in turn underlie social fear specificity. Within the scope of this project, we plan to assess the role of dorsal CA2 for social defensive behaviors and the role of CA2 ITDP for social fear specificity within a social fear conditioning paradigm. First, we ask how social memory as encoded by dorsal CA2 is essential for social fear learning and retrieval. During the fear conditioning paradigm and the relevant control conditions we will employ in vivo calcium imaging to collect neural population activity data from cells in dorsal CA2 in the respective Cre mouse lines. To link cellular mechanisms of synaptic plasticity to behavior, we then aim at assessing the relevance of CA2 ITDP with optogenetic inhibition of long-range EC projections to CA2 during relevant behaviors. Here we assess in particular the role of CA2 ITDP for social fear specificity. Finally, we will implement computational models which will assist us in understanding the link between CA2 ITDP and memory specificity and allow us to make experimentally testable predictions.