Skip to main content



Cognitive deficits are an important aspect of aging, and of a range of psychiatric and neurodegenerative disorders. The growing aging population of Western societies predicts a steady increase in the health and economic burden of mental decline. Hence, the study of the functional circuits and molecular mechanisms of long-term memory (LTM) is a central topic in neuroscience with foreseeable social implications.
One of the intracellular signaling pathways most strongly implicated in LTM is the Erk/MAP kinase–CREB pathway. However, critical gaps of knowledge on the regulation of this pathway remain, particularly as related with the mechanisms that determine its transient activation during learning or recall. Moreover, signal transduction is under tight control of multiple membrane receptors and heterogeneous synaptic inputs. Recent technologies that allow manipulation of neuronal activity in precise synaptic projections or in genetically-defined cell types promise to address these complexities, and are already accelerating our understanding of memory circuits. Therefore, the first goal of this project was to directly test the idea that MAPK tyrosine phosphatases play an important role in LTM by shaping MAPK-CREB signaling in the hippocampus during memory formation. Our second goal was to use optogenetics and designer receptors to dissect the functional neuronal circuits that drive MAPK activation during LTM. Since the beginning of the project, we have addressed the role of MAPK phosphatases, and found strong evidence that they play essential roles in the formation of episodic memories encoded in the hippocampus. Specifically, we found that memory formation involves regulated disinhibition of the MAPK pathway through a cAMP-dependent mechanism that targets tyrosine phosphatase activity. We accomplished these goals by restoring point mutants of select phosphatases on a null background to specifically address the importance of their activation state. Furthermore, by using genetic mouse models in which specific phosphatases are locally or globally ablated in the brain, we found that various tyrosine phosphatases play critical, yet complementary, roles in cognition through regulation of the Erk1/2 pathway. Indeed, different MAPK phosphatases are involved in the induction versus the duration of Erk1/2 activation following memory formation. Our results hence uncovered a missing link between the cAMP and Erk1/2 pathways that is critical for the long-term storage of new memories. These results identify new druggable targets that could potentially restore or enhance cognitive function in diseases where memory is compromised. For the second objective of this project, we have employed activity-dependent reporter mouse lines, cre-recombinase mouse lines, and retrograde viral vectors to map or functionally manipulate neuronal activity in a cell-type or projection-specific manner. First, our experiments have revealed for the first time a neuronal representation of contextual memories with emotional value. Specifically, reactivation of Erk1/2 signaling in CA1 hippocampal neurons is required for contextual memory recall, and it involves the same set of neurons activated during learning. Erk1/2 signaling in CA1 is also necessary for amygdala activation during recall of a fear memory, providing a circuit-level explanation for the observed behavioral effects. Secondly, although the hippocampus receives direct innervation from other brain regions, little is known about the role of extrinsic inputs on memory recall. We found that, besides intrinsic glutamatergic inputs, extrinsic afferents to the hippocampus are necessary for MAPK activation associated with memory retrieval. Moreover, we found that long-range inhibitory projections are essential for memory retrieval itself. Our data indicate that extrinsic inputs provide a source of disinhibition to the hippocampus that is necessary for the full expression of previously learned memories. Together, our results have provided important new insights on the cellular and circuit mechanisms of memory formation and retrieval. Our data provide empirical evidence to support the notion that cognitive deficits associated with neurological disorders may reflect impaired memory access rather than content. This new paradigm suggests that treatments aimed at enhancing memory retrieval could restore or ameliorate memory deficits previously thought irreversible in patients.