Stimulating neurons in the brain's hippocampus that are normally activated by pleasurable experiences protects mice from the depressive consequences of stress. See Letter p.335
Our ability to construct a personal narrative and to establish a sense of identity depends on our recollection of past events, the most significant of which are inevitably associated with emotions. Understanding how emotion-laden memories affect behaviour forms the bedrock of psychotherapy treatments, but the biology of this process is poorly understood. On page 335 of this issue, Ramirez et al.1 shed light on the subject, demonstrating that depressive-like stress responses in mice can be acutely suppressed through artificial reactivation of a small population of neurons that had previously been activated by a positive experience.
Understanding how memories are stored in the brain has been the subject of intensive research2, much of which has focused on the hippocampus — an elongated structure in the brain's temporal lobe that is essential for forming long-term memories and regulating responses to stress. The hippocampus contributes to the generation of cellular representations of experiences through the activation of ensembles of neurons. In mice, ensembles in a hippocampal subregion called the dentate gyrus can be genetically tagged with light-sensitive molecules while being activated by an experience3. Subsequent experimental reactivation of the ensembles using pulses of light causes the animals to behave in a way that suggests they are recalling the original experience4. As a result, these neuronal ensembles are thought to constitute the physical substrate of memory, called an engram.
Ramirez et al. used a previously described system4 to selectively express light-sensitive molecules in a subset of mouse dentate-gyrus cells while the cells were being activated by experiences, tagging them for optical reactivation at a later time. The authors tagged these cells in male mice under three conditions: a rewarding experience, in which the mice were placed in a cage with a female; a neutral experience, in which they were placed in an empty cage; or a negative experience, in which they were immobilized, leading to stress. The animals were subsequently exposed to stress for 10 consecutive days, which induced increased anxiety and depression-related behaviours such as passivity and decreased preference for enjoyable activities (anhedonia).
The investigators found that the depression-like behaviours seen in animals that had been subjected to this stressing could be reversed by optically reactivating dentate-gyrus neurons tagged during the positive experience, whereas activation of neutral- or negative-tagged neurons did not have this effect (Fig. 1).
Repeated exposure to positive imagery has been used in attempts to battle the symptoms of human depression5. Could repeated activation of a positive engram provide enduring protection from the depression-like consequences of stress in mice? Ramirez and colleagues found that when they reactivated the positive engram every day for five days in previously stressed animals, passivity and anhedonia were reversed on day six. Importantly, animals exposed to the positive experience itself for five days after the repeated stress did not show this enduring protection. Moreover, the decreased production of new neurons that is routinely observed following stress was reversed by repeated engram stimulation, but not by the repeated positive experience.
Ramirez and colleagues' provocative results raise several questions relevant to stress-related disorders in humans. Both anxiety and depression are associated with repeated exposure to stress, and the two disorders often occur together. However, acutely activating the dentate-gyrus engram in mice primarily reversed depression-associated behaviours. Might an engram that can affect anxiety be found elsewhere in the hippocampus or other regions of the brain?
Because a single reactivation of the positive engram did not relieve anxiety, the authors did not investigate the effects of repeated engram stimulation on this trait. But although fast-acting treatments for depression and anxiety differ, chronic treatments for the two conditions overlap. Could repeated activation of the positive engram, unlike the single intervention, prevent anxiety-associated manifestations of stress? Finally, it will be interesting to investigate the other side of the coin — whether reactivation of traumatic memories is sufficient to cause stress-related disease. A first step to exploring this avenue in mice would be to test whether reactivating an ensemble of dentate-gyrus neurons tagged by negative experiences could cause anxiety or depression-like behaviours.
Why is reactivating an engram of a positive experience a more effective stress-reliever than the experience itself? We are left wondering whether the repeated recall of positive memories will result in resilience to adversity. It is intriguing to speculate that nostalgia serves a similar stress-reducing purpose in humans4,6. Perhaps an experience itself results simply in memory storage, whereas its recollection activates a neuronal network associated with reward, thereby changing behaviour.
If, as has been suggested7, stress results in a cognitive bias against rewarding associations, then reactivation of a 'positive' engram laid down in an animal or human that was previously stressed would not combat the consequences of later stress — a hypothesis that could easily be tested in the authors' animals. In support of this suggestion, Ramirez et al. found that inhibiting a component of the reward neuronal circuitry while activating a positive engram interfered with the antidepressant properties of reactivating that engram. Consistent with this, a study8 of people with depression suggests that, unlike healthy individuals, they cannot improve their mood by recalling positive memories. So the ability to remember your way to happiness may be a normal regulatory process that is disrupted in depression.
Ramirez, S. et al. Nature 522, 335–339 (2015).
Squire, L. R. Neurobiol. Learn. Mem. 82, 171–177 (2004).
Ramirez, S., Tonegawa, S. & Liu, X. Front. Behav. Neurosci. 7, 226 (2014).
Liu, X. et al. Nature 484, 381–385 (2012).
Blackwell, S. E. et al. Clin. Psychol. Sci. 3, 91–111 (2015).
Cheung, W.-Y. et al. Pers. Soc. Psychol. Bull. 39, 1484–1496 (2013).
Harding, E. J., Paul, E. S. & Mendl, M. Nature 427, 312 (2004).
Joorman, J., Siemer, M. & Gotlib, I. H. J. Abnorm. Psychol. 116, 484–490 (2007).