Dreams and REM Sleep May Be About Memory Consolidation

Rapid Eye Movement (REM) sleep, during which vivid dreaming occurs, may be a period in which memory is consolidated, according to research in Science Advances.

This was shown by a sleep deprivation and recovery animal model created by a team of sleep researchers led by Pierre-Herve Luppi, director of the Center of Neurosciences Research Sleep Lab at the Universite Claude Bernard of Lyon, France.

The research is “very good,” Beth Israel Deaconess Medical Center Department of Neurology Chairman Clif Saper told Bioscience Technology. Saper, who is also a Harvard University neuroscientist, was not involved with the paper. “The Luppi lab does very nice work on REM sleep control using their model of REM sleep deprivation followed by recovery.”

The role of REM

It has long been postulated that REM sleep is involved in certain kinds of memory consolidation. That is, it has been thought that memories are filed away as long or short term during REM sleep—or just discarded—partially accounting for the bizarre nature of dreams.

REM sleep comprises 20 to 25 percent of all sleep, or 90 to 120 minutes of sleep a night. The most vivid dreams tend to occur during the four to five periods a night that people go through REM. Other sleep segments are generally referred to as “slow wave” periods.

Another long-held belief has been that the brain is in a state somewhat like wakefulness during REM. However, after dissecting and analyzing brains of three groups of rats—one getting normal REM sleep, one deprived of it, and one deprived of REM then given it—Luppi has come to new conclusions. Luppi is head of a section on Pathophysiology of Neural Networks Sleep Cycle.

Establish your company as a technology leader. For 50 years, the R&D; 100 Awards, widely recognized as the “Oscars of Invention,” have showcased products of technological significance. Learn more.

The “first” new concept introduced by his paper, Luppi told Bioscience Technology, is the fact that gene expression analysis showed his team that “only selected populations of neurons localized in a small subset of the limbic cortex are activated during REM sleep. In contrast, during waking most cortical structures contain activated neurons. Further, these neurons are not only activated, but also over-express genes involved in synaptic plasticity underlying learning and memory.”

The limbic cortex is a seat of learning and emotion in the brain.

Noted Luppi: “These results are both new and interesting. Indeed, it seems likely that these populations of neurons are involved in emotional learning and memory and that their activation might induce dreams occurring during REM sleep.”

The second new concept introduced by the paper was the fact that retrograde tracing, and neurochemical lesioning, helped Luppi’s team determine that “these cortical neurons are activated by two subcortical structures, the supramammillary nucleus and the claustrum. It was thought before that cortical activation during REM sleep was due to cholinergic neurons in the basal forebrain and the brainstem. We therefore again introduce here a completely new concept.”

The bottom line, Luppi said: “We have now identified the cortical neurons activated during REM sleep as well as the subcortical structures involved. The next step is to study these neurons by unit recordings, and to manipulate their activity by methods such as optogenetics, to demonstrate the exact role of dreaming and REM sleep in emotional learning. If we succeed, it means that we will have discovered the function of REM sleep with the idea that several pathologies, like depression, are due to a dysfunction of these systems.”

Saper told Bioscience Technology that, “During the recovery, they found certain parts of the brain showed activation, as shown by cFos protein. However, it is not entirely clear what those areas were doing for REM sleep. At the end of the results section, the authors lesion the supramammillary nucleus (SUM), one of those structures, and the amount of REM sleep is essentially the same, with only subtle changes in the EEG. This makes it hard to make a strong case that the SUM is really driving REM sleep.”

Still, Saper said, “This is an interesting paper for sleep scientists.”