When was the last time you daydreamt? Paying no particular attention to the outside world, engaged in introspection or memory recall, your mental state feels altered. This difference is reflected in global patterns of brain activity – the default mode network (DMN). Identified 20 years ago and the focus of much research activity since, the DMN connects several brain regions through distinct low-frequency oscillations.
“The DMN is also thought to play a key role in a variety of neurological and psychiatric disorders, including Alzheimer’s disease, schizophrenia, depression and autism,” says Tzu-Hao Harry Chao from the University of North Carolina at Chapel Hill’s department of neurology. “Understanding how the DMN functions in health and disease could lead to new treatments and interventions for these conditions.”
Motivated by these goals, Chao and colleagues have combined functional magnetic resonance imaging (fMRI) with a fibre photometry sensor that measures cellular calcium levels to understand how different brain regions come together to establish and disrupt the DMN in rat brains. They report their findings in Science Advances.
When studying large-scale brain connectivity, it is challenging to tap into individual neurons, especially in deep brain regions. To investigate global features, neuroscientists therefore often use a proxy for neuronal activity.
“For example, fMRI detects changes in blood oxygenation/flow to different regions of the brain, which are thought to reflect changes in neuronal activity,” explains Chao, cautioning that “this relationship between blood flow and neuronal activity is not always straightforward, and there can be many sources of noise and variability in fMRI signals.” To complement fMRI data with a direct measure of neuronal activity, the research team developed an fMRI-compatible optical imaging platform that provides multi-site neuronal readout from rat brains.
During signal transmission from one neuron to another, calcium ions enter the cell in response to an action potential, triggering the release of neurotransmitters into the synapse. For the experiments, the team used genetically engineered rats that carry a calcium-sensitive protein. The protein “undergoes a conformational change in response to calcium binding, leading to increased fluorescence intensity that can be used to detect changes in intracellular calcium levels,” says Chao.
The researchers synched up an fMRI machine to a fibre photometry platform that can detect changes in cellular calcium concentration simultaneously in four brain regions. They then scanned the brains of anaesthetized rodents for DMN activity changes, which they aligned to the calcium data.
Three out of the four brain regions observed showed increased neural activity just before the DMN was established, while in the fourth region – the anterior insular cortex – activity was significantly lowered. This is interesting as the anterior insular cortex plays a role in the salience network (SN), an alternative brain connectivity state associated with attention.
In contrast, upon DMN deactivation, activity in the three DMN-associated regions was inhibited, while the anterior insular cortex signal spiked around 8 s before the DMN shut down. Following statistical analysis, these observations reveal that anterior insular cortex activity has a negative causal influence on the other DMN brain regions.
The researchers also derived a model of five latent brain states complete with a cycle of likely transitions between them. Since in some of these latent states the anterior insular cortex correlates with the other regions, while in other states there is an anticorrelation, Chao concludes that “the topology of large-scale brain networks can be very dynamic, and these networks can be somewhat overlapped instead of clearly separated”. The pathway by which the anterior insular cortex induces DMN suppression requires further examination, however, which the team hopes to achieve in future work.
The investigators also studied the brains of awake rats with the calcium-measuring technique. Using an oddball paradigm, where the rats listened to repetitive tones with an occasional odd-one-out, they found a causal network between the studied brain regions, again with the anterior insular cortex having an inhibitory role on other DMN-associated regions.
Experiments on awake rats did not feature fMRI because conventional fMRI acquisitions are very loud, which can cause stress to the animal. “In humans, we can use earplugs plus earmuffs to minimize the acoustic noise to affect human subjects,” Chao explains. “This is practically more difficult for us to mimic in rodents, in part because their skulls are very thin for the acoustic noise to easily penetrate. This being said, we are indeed working on performing fMRI in awake mice with a new silent fMRI technique.”
The team is developing the calcium-sensor approach further by including more channels to enable data acquisition from two subjects at the same time. “This upgrade will enable us to investigate the DMN and SN roles in social interaction using rodent models. We maintain an active collaboration on this topic with Vinod Menon’s lab at Stanford University,” says Chao.
Simultaneous EEG and fMRI measure sleep ‘inertia’
He is confident that their research “paves the way for future translational studies using rodent models to investigate the cellular basis of large-scale, functionally and behaviourally significant brain networks in the healthy brain, and the neuronal mechanisms that lead to network dysfunction in brain disorders”.
“[It] has the potential to transform the landscape of fMRI and the knowledge gained will have widespread implications for the design, analysis and interpretation of human brain fMRI data,” Chao tells Physics World.
