Towards a new theory of sleep

Even though slumber consumes about a third of the day for many life forms, we know very little about why it's needed. The need for sleep remains one of the great mysteries of biology.

A leading theory posits that sleep may provide the brain with an opportunity to "rebalance" itself. In this model, waking experiences are associated with powerful processes of learning and development that, over time, result in the saturation of our brains' ability to strengthen connections. Not only would this prevent further learning, but this unbounded increase in connectivity would destabilize the brain, leading to "overexcitation" of neural networks. A leading theory suggests that the core function of sleep is "neuronal homeostasis," the processes whereby neurons self-tune their excitability to restore balanced activity to brain circuits.

Brand new research conducted in the lab of Brandeis neurobiologist Gina Turrigiano suggests this theory isn't true. In a paper published in the March 24th issue of the journal Cell, the Turrigiano lab showed that when the activity of neurons is suppressed in rats, homeostatic rebalancing doesn't occur during sleep; instead, it happened exclusively when animals were awake and active.

This research poses as many questions as it answers. For example, why is homeostasis inhibited during sleep? Turrigiano suggests that homeostatic plasticity may interfere with a sleep-dependent process that strengthens memories. Using behavioral states such as sleep and wake to temporally segregate distinct forms of plasticity may alleviate this interference problem.

Gina Turrigiano is the Joseph J. Levitan Chair in Visual Sciences at Brandeis, and was elected to the National Academy of Sciences in 2013. In 2000, at the age of 37, she won a MacArthur Fellowship, or 'genius' award. Her groundbreaking work has focused on the cellular processes that allow neuronal circuits in the brain to change and adapt. Her lab has played a major role in identifying the key mechanisms underlying homeostatic plasticity, or a neuron's ability to dynamically seek stability despite changes induced by learning or development.

This latest research in Cell, led by postdoctoral fellow Keith Hengen, broke new ground as it explored neuronal homeostasis in the context of freely behaving rats (most research in the past has relied upon cell cultures or anesthetized animals). In this work, rats with occlusion of vision from one eye were observed over nine days during sleep and wake periods. Electrodes inserted in the animals' visual cortex recorded the firings of many individual neurons; these neurons were then followed for nine days, producing a total of six terabytes of data. Algorithms developed in Turrigiano's lab with the help of Brandeis assistant professor Steven Van Hooser enabled the analysis of these enormous and complex datasets.

Turrigiano expects these computational methods to open new avenues of research for her lab, enabling far longer observation of rats' brains and with greater precision.

Some of the key questions arising from the study include:

• What it is about sleep that impedes homeostasis?

• How is it possible for the brain to quickly shift in and out of periods of homeostasis depending on the sleep state?

• What is it about being awake that incites and necessitates the activation of homeostatic regulation?