• Brain wide engram complex
• Engram
• Episodic memory

Episodic memory formation has been demonstrated to rely on the neuronal computation performed in the hippocampus and the entorhinal cortex. The bidirectional dialogue between these areas allows the neural coding of information about the temporal sequence of events ("when"), the context in which events happened ("where") and the content of the experience itself ("what"). Research from our laboratory previously demonstrated that the lateral part of the entorhinal cortex (LEC) contains neurons with engramlike properties for the recall of episodic memory in the mouse. In particular, using the object-place-context recognition task (OPCRT), it has been shown that the activity of LEC neurons which could be tagged during the learning phase is both necessary and sufficient for subsequent memory expression. However, increasing evidence also indicates that a distributed network in the entire brain should be the substrate for memory storage. This distributed network, known as the "engram complex" (or "memory trace"), has been suggested to be formed by neurons which are connected and undergo cellular and molecular changes through plasticity phenomena. In the present study, to identify a possible network for the encoding and recall of episodic-like memory, we first used the c-Fos immunostaining as a marker of neuronal activity and performed a brain-wide analysis after the execution of the novel object-place-context recognition test (OPCRT). A significant increase in c-Fos positive cell number in the OPCRT group, as compared with the controls, was found in isocortical regions known to be involved in spatial memory and navigation, such as the retrosplenial and posterior parietal cortices, as well as in the primary visual and somatosensory cortices and in the medial prefrontal area (mPFC). It is known that the mPFC is strongly connected with the LEC and hippocampal formation and has been recognized to play a role both in decision making and long term fear memory retrieval. Therefore, we sought to investigate whether activity of c-Fos positive mPFC cells, that were labeled during encoding, is required for subsequent memory expression. We found that the chemogenetic inhibition of mPFC learning tagged cells did not allow memory recall at 12h, a time point at which memory is usually evoked by natural cues. In contrast, activation of excitatory DREADDs on mPFC learning tagged cells was not sufficient for memory expression at 48 h, a time point at which natural cues fail in memory recall. Therefore, the activity in mPFC cells activated during episodic memory encoding is necessary but not sufficient for that memory to be expressed. The present findings validate our methodological approach to investigate the contribution of separate cellular ensembles to the complex engram network that supports episodic-like memory.