Last Thursday, an article published in the internationally renowned journal Nature piqued my interest. This paper, titled “Activating Positive Memory Engrams Suppresses Depression-like Behavior,” was written by Ramirez and colleagues and was published in the June 18th issue. I was initially drawn to this paper because my current research project is focused on understanding the hippocampus, a region of the brain important for learning and memory, and understanding the mechanisms by which antidepressants work in animal models. Judging by the title alone, it seemed to me like the research group was able to reduce depression symptoms through some sort of memory manipulation, which could have important ramifications on future depression and antidepressant research.
I decided to take this opportunity to explore the paper in more depth and share my thoughts with the SPEaC blog as a way of communicating new and exciting science to a broader audience. In general I think that it is important for scientists to discuss scientific findings and their caveats publicly in order to disseminate information, but I also think that the public deserves access to the realities of the scientific process instead of having to depend on promotional press releases and flashy headlines to educate themselves. I hope that posts like this one can become a recurring feature on our blog.
What research question drove the creation of this study?
Depression is a mental illness characterized by emotional, cognitive, and behavioral symptoms that disrupt a person’s ability to function normally. (For more about the symptoms of depression, please see the National Institute of Mental Health’s website.) It is a major cause of disability worldwide, yet scientists still know little about the brain changes involved in developing depression or in recovering from it after treatment. The authors state in the abstract of the paper that stress is a major risk factor for depression and that the hippocampus may be involved in modulating behavioral responses to stress. For this reason, they wanted to test whether manipulating the hippocampus could alter the development of depression-like behaviors after exposure to a stressful experience.
What did the scientists actually do?
To determine whether the hippocampus modulates depression-like behaviors, the authors chose to study mice. You may be wondering how scientists measure depression-like behaviors in mice. Although we can never diagnose a mouse as depressed in the same way a person may be depressed (since we cannot talk with a mouse about its thoughts, feelings, and motivations), we can measure a mouse’s behavior, like their tendency to pursue rewards or to give up during a stressful event. For example, if a mouse drinks less sugar water in the Sucrose Preference Test (in which a mouse is given a choice of regular water or sugar water to drink), then this is thought to mimic depressed people’s tendency to lose interest in otherwise pleasurable experiences. If a mouse struggles less during a Tail Suspension Test or a Forced Swim Test (in which a mouse is held by the tail or put into a container of water and observed), then scientists think this correlates to the hopelessness often felt by people with depression. Although it sounds like a stretch, these behaviors have been well-validated and respond to the same antidepressants that work in depressed people.
The authors of the current paper began their experiment by growing a light-sensitive channel in the brains of mice so that individual brain cells called neurons could be experimentally controlled (through a technique called optogenetics). They did this by putting a drug (doxycycline) into mouse food and feeding it to mice that have a particular set of genes that use the drug as a signal to stop making the channel. For the major experiment in this paper, the authors stopped feeding the mice the drug before exposing them to some sort of experience: a positive experience (being allowed to socialize with mice of the opposite sex), a neutral experience (being in a new enclosure), or a negative experience (being temporarily restrained and unable to move). This caused the light-sensitive channel to be made by the small subset of neurons that was particularly active during the experience. After the experience was over, the mice were given the drug again to prevent neurons from producing any more of the channel. The aim of this technique was to express the light-sensitive channel only in the individual neurons that were activated during the experience.
To introduce a depression-like state in the mice prior to final testing, the authors then exposed some of the mice to a chronic immobilization stress. This stress paradigm involves restraining the mouse (like in the “negative experience” described above) on ten separate occasions and is known to produce depression-like behaviors. The authors then used fiber optics to deliver light pulses to the hippocampus of each mouse’s brain while simultaneously testing for these behaviors. The light pulses opened the light-sensitive channel and caused the neurons to fire action potentials, which are the electrical signals by which neurons communicate. Only the neurons that were active during the positive, neutral, or negative experiences have the channel, so the authors presumed that delivering light to the mouse’s brain essentially reactivated this previous memory. (As an aside: the physical basis of memory has been referred to as an engram, in case you were wondering what this word meant in the paper’s title.) The authors observed the mice during the Sucrose Preference Test and the Tail Suspension Test while these memories were re-activated.
What were the results of these experiments?
The most exciting result obtained from the study was that the mice with reactivated positive memories after the chronic immobilization stress performed similarly to non-stressed (i.e. non-depressed) mice on their final behavior tests. The mice with neutral or negative reactivated memories displayed the same amount of depression-like behaviors as stressed mice that did not have any memories reactivated. The authors concluded that reactivation of positive memories reduces depression-like symptoms in mice.
Depression, however, is a chronic condition, and this experiment only activated the memories once: during the single behavioral test performed on the mouse. This does not speak at all to whether positive memory activation can be used as a long-term treatment for depression-like states. To answer that question, the authors used a new set of mice to reactivate memories twice a day for five days with light pulses after the chronic immobilization stress was completed. This time, the authors waited 24 hours before testing the mice for depression-like behaviors (rather than activating the memory during the behavioral test itself). In this scenario, five days of light exposure, but not a shorter treatment, reduced depression-like behaviors in stressed mice. Although this is still not proof of a long-lasting effect, it does suggest that chronically reactivating a positive memory may produce more than fleeting positive effects for the mice.
Interestingly, the authors also tried a more “natural” exposure to positive experiences. Instead of using light to activate neurons twice a day for five days, they let male mice socialize with female mice twice a day for five days. This did not reduce depression-like symptoms, suggesting that there is something very particular about light-induced activation of neurons that allows it to provide more benefit than actual positive experiences.
I find this result fascinating. Is the mouse reliving (like a hallucination) the prior experience, or is it just a pleasant memory? Is the effect due to some sort of confusion in the brain because positive and negative experiences are happening simultaneously (rather than “thinking about a positive experience” per se)? Does the effect depend more on the novelty of the positive experience rather than the positive experience itself, since the neurons being activated by light are those that were “tagged” with the channel after the first exposure to the positive experience? Is all of this some side effect of the experimental design that we haven’t figure out yet? The authors hypothesize that this difference between “natural” and light-induced positive experiences may be due to activation of different downstream brain circuits, which is a mechanism that could underlie many of the hypotheses I just proposed. There are many unanswered questions left in the wake of this paper that will hopefully be explored in further research.
Now, what does this mean for depression treatment in people?
Because I don’t want to get bogged down in the technical aspects of the study, I will not go further into any criticisms (positive or negative) of the experimental design in this particular paper. Instead, I would rather focus on how we take data like these (assuming the experiments are valid) and interpret them in a way that is socially responsible while still retaining the excitement that a paper like this can bring. (As a disclaimer, please remember that I am not a physician and cannot render advice about any particular individual’s medical treatment.)
Firstly, it is important to note that this entire study was performed in mice, and we don’t know how humans would respond to this treatment. Performing a similar experiment in people is difficult for a number of reasons. Expressing a light-sensitive channel and implanting fiber optics in people is fraught with health concerns, although gene therapy techniques and implants are currently being developed and could, one day, be safe enough for this situation. Also, exposing people to a novel positive experience before they develop depression, so that a nice memory can be “tagged” without any confounding current disease, is currently impossible. Even if neuronal tagging were feasible, we do not know with certainty who will or will not develop depression, and it will never be ethical to purposefully induce depression in people. We also don’t know whether persistent reactivation of positive memories will have negative side effects, for example by constantly distracting the person with false memories and preventing them from functioning normally. We cannot say, therefore, that this study gives us direct insight into relieving the symptoms of human disease.
Secondly, we need to be careful about extrapolating the current study to situations that people can more easily relate to, which could lead to inappropriate simplification. Although a positive memory was activated in the current study, it may be a completely different experience for the subject than simply thinking about a pleasant past experience. Look again at all the questions I asked in the previous section about how we do not know what this manipulation actually did to the mouse. Also, keep in mind that exposure to positive experiences in the “natural” experimental condition did not improve behavior. For this reason, it is unhelpful to people suffering from depression for anyone to conclude that one simply needs to “think happy thoughts” or “submit to the power of positive thinking” to cure depression. Dismissive attitudes like this only serve to make the person with depression feel unsupported, and if thinking about happy memories were such an easy cure then there would be no depression in the first place. We need to take the affected individuals and their treatment seriously and accept the fact that any cure stemming from this work will likely be much more complicated than we currently realize. That being said, disrupting negative thinking and controlling thought patterns is already known to be an important component of cognitive behavioral therapies for mental illness, so the results of this study do fit (however cautiously) into current knowledge about depression and its treatments.
Lastly, despite all the caveats we should still appreciate how this study gave us important cellular-level insights into a behavior with clear ties to human disease. This paper did not describe a feasible cure for depression, but it made great strides in describing how a few cells in a single brain region control a cascade of events that leads to an incredibly complicated behavior. This and other studies like it open the door to looking at how different cells in the brain control cognition and behavior. For example, the optogenetics technique is already being used to parse out how individual populations of neurons (amongst the billions of neurons in the brain) control specific brain circuits. Through research like this, we have learned more about how the brain functions, which in turn has taught us more about ourselves and our fellow organisms. Also, who knows? Maybe this study just gave another scientist an idea about the next experiment he or she could do that will, in turn, inspire clinical trials and expand our knowledge about one of the most fascinating (in my opinion) organs ever to have evolved.
Editor: Mary Topalovski