Abstract:Theoretical and empirical studies have begun to explore the structure and function of the human brain from a system perspective, using network science to abstractly model the brain as a set of nodes (e.g., brain regions) and edges (e.g., structural or functional connections) 1. Within this framework, many studies have identified a small set of highly connected network nodes, referred to as hubs, primarily located in the medial and lateral frontal and parietal cortices. These hubs exhibit higher rates of cerebral blood flow and metabolism activity and mediate long-distance connections between brain modules to allow for efficient communication across remote regions 2-4. The central embedding of the hubs supports their diverse roles across a broad range of cognitive behaviors. For instance, the communication capacities of hubs are significantly associated with individual intellectual and cognitive-control abilities 4, 5 and can predict neuropsychological outcomes after brain injury 6. A hypothesis that naturally follows is that those brain disorders that clinically present with significant cognitive impairments might result from the pathological lesions to the network hubs. In a recent elegant study, Crossley et al. 7 provided compelling evidence for this hypothesis. They first employed diffusion imaging tractography to build healthy human brain structural networks in which nodes represent gray-matter (GM) regions, and edges represent white-matter connections linking these nodes. Highly connected hub regions were identified from the network. Then, GM lesion maps of 26 different brain disorders were generated by carrying out meta-analyses on the structural MRI data of 392 published studies. Two significant findings emerged from this comprehensive analysis. First, a commonality across almost all brain disorders was that the GM lesions were more likely to be anatomically located in hubs of healthy brain networks. Second, distinct subsets of brain hubs were disrupted in different disorders. Take an example, the hubs in the frontal and temporal lobes were specifically associated with higher lesion probability in schizophrenia, whereas the medial temporal and parietal hubs were mainly affected in Alzheimer's disease. For the commonality observation, there are two possible explanations that are not mutually exclusive. One is that the hubs are topologically centralized in terms of global brain communications and play key roles in a variety of cognitive functions, especially for "higher order" cognitive tasks and adaptive behaviors; therefore, lesions in hubs, relative to those in nonhubs, are more likely to lead to clinical symptoms associated with cognitive impairments. It is also possible that brain hubs are biologically costly and therefore are particularly vulnerable to pathogenic processes of brain disorders. That is, it is either that hub lesions tend to be more visible due to its stronger behavioral consequences, or that hubs are just more likely to be damaged, or both. The differences in terms of the specific hubs being involved indicated differences in disease pathogenetic processes: Disorder-specific factors may determine which brain hubs are primary targets and how neurodevelopmental and neurodegenerative processes then propagate along the network structure. Together, Crossley et al. 7 provide an empirical framework for a hub perspective toward understanding the pathophysiological mechanisms of brain disorders, which in turn generates crucial clinical implications about how connectivity patterns of brain hubs could serve as potential biomarkers for early diagnosis and treatment targets for therapeutic intervention (e.g., the optimized locations of brain stimulation). Meanwhile, this study raises several critical questions that warrant further attention. The GM volume or density reduction was used as the signature of brain disorders 7, which might reflect synaptic and neuronal loss. However, there are many other signatures for brain disorders such as synaptic dysfunction, glucose hypometabolism and abnormal axonal projections. These pathological features might exhibit different sensitivities across regions and/or disorders and could also appear prior to GM loss. For instance, white-matter damages or structural dysconnectivities are one of the main pathologies for multiple sclerosis, whereas functional connectivity abnormalities in specific regions may represent an early marker of Alzheimer's disease. Thus, it would be interesting to examine whether the brain hubs are also abnormal while using other hallmarks as brain signatures such as functional and structural connectivity rather than focal GM lesions. Another important concern is that Crossley et al. 7 demonstrated GM lesions in brain disorders primarily located at healthy network hubs. However, whether the hubs in the actual brain networks in patients were targeted remains to be elucidated. Moreover, very little know whether there exist possible compensatory mechanisms that the brain goes through after hub damage. By constructing whole-brain functional connectivity networks derived from resting-state functional MRI data, we reported disrupted connectivity strength of brain hubs (e.g., the medial prefrontal and parietal cortices) in Alzheimer's disease 8. Brain network analyses of autism spectrum disorders and attention-deficit/hyperactivity disorder revealed shared functional disconnectivity in the precuneus hubs and distinct connectivity disruptions in the medial frontal hubs 9. In the future, it would be important to map connectivity patterns of disorder-specific hubs using empirical structural and functional data in patients. Moreover, computational modeling of brain network data 10 that incorporate the hub lesion information and associated disease propagation patterns over network could be a powerful tool to deepen the understanding of the biological mechanisms of brain disorders. Finally, how are the hubs defined in the brain networks? Brain hubs generally reflect the importance of regions in terms of their information communications, which can be defined using different measures (e.g., nodal degree and betweenness centralities and participation coefficient) in structural and functional networks 4. While interrelated, these metrics quantify different aspects of nodal roles in the network: nodal centrality metrics capture global brain communication while nodal participation coefficient captures the integrity ability between different brain modules. Thus, the hub locations are dependent on different definitions. Although Crossley et al. 7 used multiple nodal measures to identify the brain hubs and observed compatible results, it remains open as to which hub measure(s) is the most sensitive, specific and reliable index to quantify the brain network changes for each disorder? Charting the patterns of hub disruption using different nodal measures and imaging modalities would be a potentially productive topic for future studies. This work was supported by the National Key Basic Research Program of China (Grant Nos. 2013CB837300 and 2014CB846100) and the National Science Fund for Distinguished Young Scholars (Grant No. 81225012). The authors declare no conflict of interest.

Abstract:Event Abstract Back to Event Multiple factors shape the neural responses to brain damage: anatomy, topology, etiology and demography Binke Yuan1* and Yanchao Bi2 1 Center for Language and Brain, Shenzhen Institute of Neuroscience, China 2 Beijing Normal University, China Introduction. Focal brain damage may cause widespread disruption of human neural networks (Aerts et al., 2016), and various types of neural responses can (co-) occur, such as diaschisis, transneuronal degeneration, dedifferentiation, compensation and neural reserve (Fornito et al., 2015). Several network-based lesion propagation models have been proposed (Stam et al., 2014; Fornito et al., 2015) but there is no quantitative model-based prediction study to test which factors contribute to the neural functional properties of regions that are not directly within the structural lesion in patients with brain damage. Here we present a group-level general linear model (GLM) to predict the changes of local (e.g., ALFF, Zang et al., 2007) and global (e.g., nodal strength) neural functional properties in regions outside lesion after focal brain lesion, i.e., a functional propagation model for focal structural lesion. Three main types of factors were examined, including the properties of the lesion (volume, functional topological hubness of the lesion epicenter), the properties of the target region (nodal local efficiency, which reflects nodal resilience to random attack; nodal global efficiency, which reflects nodal hubness and stress), and the target region’s relationship to the lesion epicenter (Euclidean distance; topological distance, functional shortest path to the lesion epicenter). Other potential demographic factors and time post-onset were also taken into consideration. Materials and Method. Thirty patients (7 females; age: 42.43 ± 11.62, 22–67 years; education: 12.7 ± 3.45, 6-19 years; lesion onset time: 6.93 ± 15.25 months, 2-87 months; 9 hemorrhagic, 17 ischemic, 4 traumatic brain injury) with brain damage and 36 demographically-matched healthy controls were recruited. Structural MRI data (T1, FLAIR T2) and resting-state fMRI images (28 axial slices, TR = 2000 ms, 120 volumes) were collected on a 1.5 T scanner. Lesion extent was manually drawn and spatially normalized into MNI space (Figure 1a). Resting-state data were preprocessed for ALFF and functional connectivity analyses. The brain lesion was parcellated using a functionally-based atlas (180 nodes covering the whole-brain but the cerebellum, Craddock et al., 2012), and the functional network and node-wise ALFF were calculated on the same atlas. A third metric, the similarity of functional connectivity pattern, which reflects the degree of re-wiring, is defined as the Pearson phi coefficient between a node’s binary whole-brain functional connectivity and a reference node’s binary whole-brain functional connectivity. The reference connectivity patterns were from a normative functional connectome constructed by averaging 144 healthy subjects’ functional networks using the Craddock atlas (Fig. 1b, s = 0.15, Yuan et al., 2017). The normative functional connectome was also used to calculate the nodal local and global efficiency, the functional shortest path to the lesion epicenter, and the hubness of the lesion epicenter. For each subject, the lesion epicenter were those nodes that are damaged. Z-tests were performed to assess the severity of the neural responses of the intact regions. The GLM model. For each metric, the node-wise Z scores were concatenated across patients as the dependent variable and each of the anatomy-related, topology-related, lesion-related and demography-related factors was concatenated across patients as the independent variable (Fig. 1c), resulting in three tests. The GLM model: Yz−scores=β0+ β1 xlocE+β2 xgE + β3 xdistance + β4 xshort−path + β5 xhubness + β6 xonset−time + β7 xetiology + β8 xlesion−volume + β9 xlesion−hemishpere + β10 xage + β11 xsex+ β12xeducation Results. All three prediction models were significant (nodal strength, F (12, 5357) = 25.7, P = 2.46E-39, R2 = 0.037; pattern similarity, F (12, 5357) = 46.19, P = 2.23E-72, R2 = 0.056; ALFF, F (12, 5357) = 26.6, P = 8.82E-41, R2 = 0.038; Table 1). Euclidean distance and nodal local efficiency were significant (uncorrected P < 0.005, Table 1) predictors for all three models. Global efficiency and age were significant predictors for nodal strength and pattern similarity. Lesion volume was a significant predictor for nodal strength and ALFF. The significant predictors of nodal strength also included node hubness, etiology, lesion hemisphere and age. Functional shortest path was a significant predictor for pattern similarity and lesion onset time was a significant predictor for ALFF. Conclusion. Our results showed that multiple factors -- including the properties of the lesion, the properties of intact regions, and the intact regions’ anatomical and topological distance to the lesion epicenter, as well as demographic factors, lesion onset time, etiology, and hemisphere of lesion -- shaped the neural responses after brain damage. Figure 1 Figure 2 References Aerts, H. (2016), ‘Brain networks under attack: robustness properties and the impact of lesions’, Brain, 139:3063-3083. Craddock, R.C. (2012), ‘A whole brain fMRI atlas generated via spatially constrained spectral clustering’, Hum Brain Mapp, 33:1914-1928. Fornito, A. (2015), ‘The connectomics of brain disorders’, Nature Rev. Neurosci, 16: 159-172. Stam, C.J. (2014), ‘Modern network science of neurological disorders’, Nature Rev. Neurosci, 15:683-695. Yuan, B.K. (2017), ‘Brain hubs in lesion models: Predicting functional network topology with lesion patterns in patients’, Scientific Reports, (2017) 7:17908. Zang, Y.F. (2007), ‘Altered baseline brain activity in children with ADHD revealed by resting-state functional MRI’, Brain Dev, 29:83–91. Keywords: Stroke, TBI (Traumatic Brain Injury), Functional Connectivity, ALFF, GLM predictive models Conference: Academy of Aphasia 57th Annual Meeting, Macau, Macao, SAR China, 27 Oct - 29 Oct, 2019. Presentation Type: Poster presentation Topic: Not eligible for student award Citation: Yuan B and Bi Y (2019). Multiple factors shape the neural responses to brain damage: anatomy, topology, etiology and demography. Front. Hum. Neurosci. Conference Abstract: Academy of Aphasia 57th Annual Meeting. doi: 10.3389/conf.fnhum.2019.01.00050 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 30 Apr 2019; Published Online: 09 Oct 2019. * Correspondence: Mx. Binke Yuan, Center for Language and Brain, Shenzhen Institute of Neuroscience, Shenzhen, China, yuanbinke@gmail.com Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Binke Yuan Yanchao Bi Google Binke Yuan Yanchao Bi Google Scholar Binke Yuan Yanchao Bi PubMed Binke Yuan Yanchao Bi Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

Brain networks under attack: robustness properties and the impact of lesions

Simulation of Robustness against Lesions of Cortical Networks

Brain Hubs in Lesion Models: Predicting Functional Network Topology with Lesion Patterns in Patients

Brain networks and their relevance for stroke rehabilitation

Cognitive impairment after focal brain lesions is better predicted by damage to structural than functional network hubs

Characterization of structural and functional network organization after focal prefrontal lesions in humans in proof of principle study

Multivariate and network lesion mapping reveals distinct architectures of domain-specific post-stroke cognitive impairments

Complex brain networks: graph theoretical analysis of structural and functional systems

Brain network topology early after stroke relates to recovery

Functional and structural brain networks in epilepsy: What have we learned?

Joint impact on attention, alertness and inhibition of lesions at a frontal white matter crossroad

Brain Anomaly Networks Uncover Heterogeneous Functional Reorganization Patterns after Stroke

Structural brain networks and functional motor outcome after stroke—a prospective cohort study

Effects of Virtual Lesions on Temporal Dynamics in Cortical Networks Based on Personalized Dynamic Models.

With Great Brain Hub Connectivity Comes Great Vulnerability

Resilience of human brain functional coactivation networks under thresholding

Controllability and Robustness of Functional and Structural Connectomic Networks in Glioma Patients

Disruptions of network connectivity predict impairment in multiple behavioral domains after stroke

Multiple Factors Shape the Neural Responses to Brain Damage: Anatomy, Topology, Etiology and Demography

Assessment of System Dysfunction in the Brain Through MRI-based Connectomics

The hubs of the human connectome are generally implicated in the anatomy of brain disorders