上海科技大学知识管理系统

Introduction:

Understanding the brain's dynamic processes, particularly traveling waves in resting-state fMRI (rsfMRI) [1], is the key to unraveling its complex functionalities. These waves, which are tied to electrophysiological measures in animal models [2], have been increasingly recognized for their role across various brain activity scales. These directionally constrained waves are found to propagate along a spatial axis representing cortical hierarchical organization [3]. Ongoing arousal fluctuations are found to be associated with global waves of spontaneous brain activity in both rodents and humans, by using physiological arousal indicators [4-6]. However, it is yet to be understood how traveling waves propagate under different sleep/wakefulness states in these species. A comprehensive examination of the relationship between cortical hierarchical organization and propagation patterns of traveling waves under different sleep/wakefulness states is not only critical for illuminating the underlying mechanism of dynamic brain connectivity, but also essential for providing insights into the evolutionary aspects of functional architecture of the brain across species.

Methods:

Two open-source datasets (Dataset 1: OpenNeuro ds003768; Dataset 2: Mouse sleep fMRI with simultaneous ECoG) were employed in this study [6-8]. Dataset 1 included simultaneous EEG-rsfMRI data from human sleep, and Dataset 2 included simultaneous ECoG-rsfMRI data from mouse sleep. The synchronized electrophysiological data were used to classify sleep/wakefulness stages. To extract propagation patterns of spontaneous brain activity, each subject's global mean signal was segmented into chunks based on peak global activity, and in each chunk, every voxel's rsfMRI signal peak had either an advance or a delay relative to the global signal peak. This was used to form a specific vector containing time delay information. The vectors of the same sleep/wakefulness stage were grouped together to be merged into a matrix [3]. Singular value decomposition was applied to each matrix, extracting the principal propagation delay profile for each sleep/wakefulness stage. To obtain cortical hierarchical organization for each sleep/wakefulness stage, functional gradients were calculated using the average resting-state functional connectivity matrix of that stage based on the diffusion embedding algorithm [9]. Spatial correlations between propagation delay profiles and functional gradients were assessed to determine their relationships.

Results:

Fig. 1a shows human propagation delay profile of each stage was spatially similar to the principal gradient of that stage, indicating the propagations of spontaneous brain activity at different stages followed cortical hierarchical organization. Fig. 1b shows similar relationships in the mouse brain, but in the awake state, the propagation delay profile was similar to the secondary functional gradient instead, indicating the difference between these two species. Fig. 2 shows the comparison of propagation delay profiles between these two species. Specifically, under the awake state, the propagation in the human brain occurred between default mode regions and primary sensory areas (Fig. 2a). In the mouse brain, it was between the anterior cingulate and primary sensory cortex plus the amygdala (Fig. 2d). Under NREM, the propagation in the human brain involved the visual cortex and limbic system at one end, and primary sensory areas at the other end (Fig. 2b and c). In the mouse brain, one end of the propagation still included the primary sensory cortex, but the amygdala was no longer involved (Fig. 2e).

Conclusions:

This study reveals the relationship between cortical hierarchy and propagation patterns of spontaneous brain activity across wakefulness and NREM states in both human and mouse brains. The propagation features conserved between the human and mouse brains open avenues for further research into the evolutionary aspects of dynamic brain connectivity across species.

1. Majeed, W. (2011), ‘Spatiotemporal dynamics of low frequency BOLD fluctuations in rats and humans’, Neuroimage 54:1140-1150.
2. Thompson, G.J. (2014), ‘Quasi-periodic patterns (QPP): large-scale dynamics in resting state fMRI that correlate with local infraslow electrical activity’, Neuroimage 84:1018-1031.
3. Gu, Y. (2021), ‘Brain Activity Fluctuations Propagate as Waves Traversing the Cortical Hierarchy’, Cerebral Cortex 31:3986-4005.
4. Liu, X. (2021), ‘Single-neuron firing cascades underlie global spontaneous brain events’, Proceedings of the National Academy of Sciences of the United States of America 118.
5. Raut, R.V. (2021), ‘Global waves synchronize the brain's functional systems with fluctuating arousal’, Science Advances 7.
6. Gu, Y. (2022), ‘An orderly sequence of autonomic and neural events at transient arousal changes’, Neuroimage 264:119720.
7. Gu, Y. (2023), ‘Simultaneous EEG and functional MRI data during rest and sleep from humans’, Data Brief 48:109059.
8. Yu, Y. (2023), ‘Sleep fMRI with simultaneous electrophysiology at 9.4 T in male mice’, Nature Communications 14:1651.
9. Margulies, D.S. (2016), ‘Situating the default-mode network along a principal gradient of macroscale cortical organization’, Proceedings of the National Academy of Sciences of the United States of America 113:12574-12579.