Neurons and Astrocytes Interact to Create Day-Night Cycles

Post by Anastasia Sares

The takeaway

Recent work shows how a partnership between neurons and surrounding cells called astrocytes helps to regulate our body’s central clock. This highlights the importance of non-neuronal cells in brain function, and adds a piece to the puzzle of how the brain manages day/night cycles.

What's the science?

The body’s circadian rhythm includes cycles of wake and sleep, hunger and digestion, blood pressure, hormones, and many other daily patterns. The brain region responsible for this is the suprachiasmatic nucleus, which maintains a circadian rhythm even in the absence of any light. But how do all the cells in this nucleus stay synchronized with each other and avoid sending out contradictory signals? This mystery becomes even more puzzling given that the main neurotransmitter in this region, GABA, is inhibitory, which should inhibit activity across the whole network instead of creating the cycling behavior we actually see.

This week in PNAS, Patton and colleagues demonstrated that support cells called astrocytes help to regulate the activity of neurons in this area by “vacuuming up” the GABA floating around outside of cells during the day and letting it accumulate at night.

How did they do it?

The authors obtained the brains of mice and extracted the suprachiasmatic nucleus, slicing it so it was only micrometers thick and mounting these slices on membranes. The slices were kept in a solution that would allow the cells to live and the neurons to keep firing. Each of these slices was then infected with adeno-associated viral vectors (AAV), which introduce genetic material so that the cell itself produces a custom molecule. In this case, the inserted gene encoded for a fluorescent protein that would latch on to GABA molecules. With the fluorescent molecules active, the brain slices would glow when there was GABA present, and go dark when the GABA disappeared. The authors observed that GABA concentrations were low during the day and peaked at night, even though the neurons that should release the GABA were firing more during the day.

The authors then re-analyzed their previously published single-cell RNA-sequencing studies of suprachiasmatic nucleus slices harvested in daytime vs nighttime. Some of the genes being transcribed differently in day and night were involved in GABA transport by astrocytes, which are support cells present in brain tissue. Using the same fluorescent tagging method, they investigated the activity of these GABA transporters, and what happens when they are chemically blocked.

What did they find?

GABA transport proteins in astrocytes were up-regulated during the day, meaning that the astrocytes are likely “installing” them in their membranes and using them to move GABA out of the intercellular space. At night, the opposite is true: there are fewer GABA transport proteins, and thus GABA builds up in the intercellular space. This cycle, in turn, influences how often neurons in the suprachiasmatic nucleus fire, and how often secondary transmitter molecules called neuropeptides are released—these neuropeptides go on to influence circadian behavior.

Inhibiting the activity of GABA transport proteins disrupted the circadian rhythm in the brain slices, and initiating the clock of the astrocytes was able to restore circadian rhythm to “clock-less” neurons in slices genetically engineered to lack certain proteins that would help the circadian clock function. So, although it was previously thought that GABA control of neuronal activity was not important, it is now thought that astrocytes actively remove it during the day instead, and allow it to accumulate at night supporting daily cycles of neuronal activity.

What's the impact?

These findings call attention to the often-forgotten “support” cells that can be found throughout neural tissue, showing that they may in fact be orchestrating important brain functions. It also brings us closer to understanding how our day/night cycles work, how they might be disrupted, and what might be the consequences of that disruption.

Neural Replay as a Proposed Explanation for the Experience of Dreams

Post by Megan McCullough

What is neural replay?

Hippocampal neurons have been observed to spontaneously increase their firing rate during sleep. Recent studies have linked this display of brain activity with prior experiences; neurons that were active during an activity in an awake state are more likely to be reactivated during sleep. This is known as hippocampal neural replay. Neural replay, more broadly known as memory reactivation, occurs when there is a sequence of neuronal activity during rest or sleep that echoes the sequence of activity that occurred in an awake state. The evidence for this phenomenon was first discovered in maze exploration experiments with rodents; brain cells that were active when the rodents were exploring the maze also showed similar activity patterns during sleep. Recent technological advances in neuroimaging and electrical recordings have provided the first evidence for neural replay in humans. Neural replay during NREM has been shown to relay new information to the larger neural network, thus playing a key role in memory consolidation during sleep.  Interestingly, dreams share some features with neural replays, which has led to the idea that neural replays may be one mechanism underlying dreaming.

What is the link between neural replay and dreaming?

One proposed explanation for the purpose of dreams is that they support memory processes like consolidation, the process of transforming short-term memories into long-term ones. Since neural replays have also been shown to support memory consolidation, one hypothesis proposes that dreams are the subjective experience of neural replays that facilitate memory consolidation. Like dreaming, neural replays represent fragments of experiences, can combine multiple memories, and occur in both the hippocampus and cortical regions. Interestingly, neural replay has been shown to occur during sleep onset and NREM stages. These memory reactivations tend to occur for spatial memories, but can also occur for  motor, visual, and social memories.

Neural replay shares some features with the neural correlates of dreaming, but current research shows that memory activation is probably not the main explanation for dreams. Most neural replay events occur in earlier sleep stages, whereas dreams become most vivid in later sleep cycles. The timescales also differ; studies show that dreams occur on a timeline of seconds to minutes and are experienced at "life-like" timescales whereas neural replay occurs in the range of hundreds of milliseconds. These differences suggest that dreaming relies on other mechanisms than neural replay. Because of the number of shared features however, neural replay may relate to dreams in different ways. Dreams that include memories may rely on neural replay to an extent or neural replays could trigger dreaming. But since dreams most vividly occur in the REM stage, don't always include events that the dreamer experienced, and happen at a different timescale than neural replay events, memory activation events alone do not explain the neural basis of dreaming. 

Are there other possible explanations for the basis of dreams?

Beyond memory consolidation, there are other proposed explanations for why we dream, such as improved emotional regulation, future preparation, and the idea that dreams may have evolved to help us adapt to new sets of data. Although there are many hypotheses for why we dream, the neural correlates of dreaming remain unknown. Dreaming is a subjective experience and although new advances in electrical recordings and brain scanning have allowed scientists to monitor brain activity during sleep, the content of dreams is still studied through subjective measures such as dream journaling. More research is needed as we move into the future to further understand the reasons why humans dream, and its neural basis.

References +

Aleman-Zapata et al. Sleep deprivation and hippocampal ripple disruption after one-session learning eliminate memory expression the next day. PNAS (2022). Access the original scientific publication here

Freyja Olasfsdottir et al. The role of hippocampal replay in memory and planning. Current Biology (2018). Access the original scientific publication here

Hoel. The overfitted brain: Dreams evolved to assist generalization. Patterns: Cell Press (2021). Access the original scientific publication here

Mutz et al. Exploring the neural correlates of dream phenomenology and altered states of consciousness during sleep. Neuroscience of Consciousness (2017). Access the original scientific publication here

Picard-Deland et al. Memory reactivations during sleep: A neural basis of dream experiences. Trends in Cognitive Sciences: Cell Press (2023). Access the original scientific publication here

Ruby PM (2020) The Neural Correlates of Dreaming Have Not Been Identified Yet. Commentary on “The Neural Correlates of Dreaming. Nat Neurosci. 2017”. Front. Neurosci. 14:585470. doi: 10.3389/fnins.2020.585470

A New Network for Mapping Movement in the Brain

Post by Christopher Chen 

The takeaway

Traditionally, a region in the motor cortex called the homunculus has described a somatotopic map linked to movement of specific body parts. However, new research suggests that a parallel network (SCAN) in the motor cortex incorporating cognitive aspects of movement also exists. 

What's the science?

For many, learning about the homunculus has become a rite of passage in biology and neuroscience courses. In short, the homunculus is widely-known as the somatotopic map in our brains that controls specific body movements. For example, when we move our pinkie, a specific region (i.e., an effector-specific region) in the homunculus corresponding to pinkie movement becomes activated. Indeed, the father of homunculus theory, Dr. Wilder Penfield, discovered that directly stimulating specific regions in the brain could elicit movement of specific body parts. However, since its characterization nearly 100 years ago, the homunculus theory has been under growing scrutiny, with new theories emerging that suggest body movement may in fact be more complex than Dr. Penfield imagined. 

One of the biggest reasons views on the homunculus theory have changed is that 21st century big data analysis and brain imaging techniques have provided opportunities to more deeply investigate how the brain processes body movement. Specifically, precision-functional mapping (PFM) – which integrates fMRI imaging during resting and active states to generate more detailed images of brain connectivity – has allowed neuroscientists to visualize and analyze brain patterning and activity in unprecedented ways.

In a recent article in Nature, researchers discovered evidence across thousands of human subjects of the existence of a novel somato-cognitive action network (SCAN) that helps inform voluntary body movements in parallel with effector-specific regions in the homunculus, providing a compelling new framework to understand how our brain facilitates movement. 

How did they do it?

This study analyzed public domain fMRI images from thousands of participants from large-scale projects like the Human Connectome Project to inform its conclusions. To find patterns and similarities across such vast amounts of data, researchers used advanced algorithms and big data analysis techniques to generate functional maps of the brain in both resting and active states. Ultimately, the strategy was to apply this repertoire of advanced techniques to determine how the motor cortex communicated with the rest of the brain – including regions linked to cognition, motor planning, and even emotion – during specific body movements. 

On a smaller scale, researchers performed brain imaging studies at the University of Washington to characterize brain connectivity under specific conditions involving specific body parts and movements. They also performed neuroimaging on a wider range of human subjects including infants, to supplement the data generated from the larger neuroimaging studies as well as trace the developmental arc of movement processing in the brain. Researchers also looked at non-human primates (macaques) to determine how evolutionarily conserved the brain’s processing of movement was.

What did they find?

The investigation’s most intriguing finding was the discovery and characterization of regions in the brain termed “inter-effector regions.” Anatomically, these areas are sandwiched between the effector-specific regions (i.e., regions corresponding to a specific body part) in the motor cortex and characterized in the homunculus, but hold a very different function. Rather than correspond to an isolated movement of a specific body part (e.g. lifting your index finger), inter-effector regions are linked to more integrated movements that require coordination of multiple body parts (e.g. reaching for a cup of coffee). Compellingly, researchers found that inter-effector regions shared closer connectivity to a region of the brain called the cingulo-opercular network (CON), a region associated with arousal, error processing, and even pain. Thus, these more cognitive-related inter-effector regions operate in parallel with more motor-related effector-specific regions during movement, collectively making up what the researchers describe as a dual-system model of body movement.   

As for the smaller, more focused fMRI studies, they bolstered as well as extended the study’s core findings. Under controlled conditions, the movements of a range of body parts such as the abdominals, elbows, and eyebrows requiring less specific motor control elicited inter-effector region activity, while more specific body movements elicited activity only in effector-specific regions, highlighting the consistency and specificity of this dual-system network. Furthermore, the authors noted that effector-specific activity appeared to be organized in a ring-like pattern, with distal body parts at the center of the ring and proximal ones at the edges. Finally, researchers found that macaques had similar effector-specific and inter-effector region patterning as humans, and that humans as young as eleven months old expressed the beginnings of this patterning.

What's the impact?

The characterization of the SCAN is a testament to the brain’s incredible connectivity and complexity. With its evidence of a parallel system of effector-specific and inter-effector regions informing body movement, this study highlights how much brain processing occurs during seemingly mundane movements like walking or raising our hand. Naturally, a looming question is whether similar studies can employ big data analysis to generate maps of the brain during more complex, higher-order cognitive tasks. 

Access the original scientific publication here.