Neurobiology of Disease 167 (2022) 105671 Contents lists available at ScienceDirect Neurobiology of Disease journal homepage: www.elsevier.com/locate/ynbdi Connectomic-genetic signatures in the cerebral small vessel disease Raquel Gutiérrez-Zúñiga a, b, Ibai Diez a, c, Elisenda Bueichekú a, Chan-Mi Kim a, William Orwig a, c, Victor Montal a, d, Blanca Fuentes b, Exuperio Díez-Tejedor b, Maria Gutiérrez Fernández b, Jorge Sepulcre a, c, * a Gordon Center for Medical Imaging, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA b Department of Neurology and Stroke Centre, Neurological Science and Cerebrovascular Research Laboratory, La Paz University Hospital, Neuroscience Area of IdiPAZ Health Research Institute, Universidad Autónoma de Madrid, Madrid, Spain c Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA, USA d Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Biomedical Research Institute Sant Pau, Universitat Autonoma de Barcelona, Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Small vessel disease (SVD) is a disorder that causes vascular lesions in the entire parenchyma of the human brain. Small vessel disease At present, it is not well understood how primary and secondary damage interact to give rise to the complex Grey matter scenario of white matter (WM) and grey matter (GM) lesions. Using novel cross-sectional and longitudinal White matter connectomic approaches, we unveil the bidirectional nature of GM and WM changes, that is, primary cortical Neurodegeneration neurodegeneration that leads to secondary alterations in vascular border zones, and WM lesions that lead to Connectomics Genetics secondary neurodegeneration in cortical projecting areas. We found this GM-WM interaction to be essential for Graph Theory executive cognitive performance. Moreover, we also observed that the interlocked degeneration of GM and WM over time associates with prototypical expression levels of genes potentially linked to SVD. Among these connectomic-genetic intersections, we found that the Androgen Receptor (AR) gene, is a particularly central candidate gene that might confer key vulnerability for brain lesion development in SVD. In conclusion, this study advances in the understanding of the bidirectional relationships between GM and WM lesions, primary and secondary vascular neurodegeneration, and sheds light on the genetic signatures of SVD. 1. Introduction in distributed areas (e.g., spatially distant from the original WM lesion), impacting both GM and WM remotely and impairing the whole-brain Small vessel disease (SVD) is a disorder of the small arterioles and network functionality (Wardlaw et al., 2019). The leading hypothesis capillaries that causes lesions in the brain parenchyma, including white is that the cortical damage is caused by secondary degeneration after the matter (WM) and grey matter (GM) (Ter Telgte et al., 2020; Viswana­ loss of long WM tracts, and some studies have successfully correlated the than et al., 2006; Wardlaw et al., 2019). WM hyperintensities (WMH) of degree of cortical atrophy with the amount of WMH (Habes et al., 2018; presumed vascular origin, lacunar stroke, microbleeds, microinfarcts, Lambert et al., 2016; Rizvi et al., 2018). Interestingly, primary vascular dilated periventricular spaces, or superficial siderosis form the SVD le­ damage within the GM has also been suggested in recent studies of sions’ spectrum. These lesions are related to a high risk of stroke (both cortical microinfarcts (Ter Telgte et al., 2020), particularly relevant in ischemic and hemorrhagic), and the development of a wide variety of genetic forms of SVD, such as CADASIL, cerebral autosomal dominant motor, sensory, and cognitive impairments (Wardlaw et al., 2019). arteriopathy with subcortical infarcts and leukoencephalopathy; that is WMH is one of the most well-characterized features of the SVD neuro­ caused by mutations in the NOTCH3 gene (Viswanathan et al., 2006). imaging manifestations; it is a validated biomarker and it is correlated This proposed bidirectional relationship between GM and WM pathol­ with SVD clinical syndromes in the aging population (Li et al., 2018). It ogy, despite introducing complexity to the study of SVD, also presents is hypothesized that leukoaraiosis could also lead to neurodegeneration exciting new challenges to develop frameworks for elucidating primary * Corresponding author at: 149 13th St, Office 5.209, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA, USA. E-mail address: sepulcre@nmr.mgh.harvard.edu (J. Sepulcre). https://doi.org/10.1016/j.nbd.2022.105671 Received 21 September 2021; Received in revised form 31 January 2022; Accepted 21 February 2022 Available online 26 February 2022 0969-9961/© 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 and secondary damage mechanisms implicated in this pathology. 2015; Sepulcre et al., 2018) Here, we integrate weighted-degree and At present, the etiology of the SVD is not well understood. The most link-level analyses to quantify the centrality influence of specific cere­ reported abnormality in the SVD is a diffuse arteriopathy affecting the bral areas (voxels), as well as to characterize connectivity strength re­ smaller perforating vessels that display histopathological changes in the lationships in the brain’s network architecture of GM and WM vascular-cell integration without consensus about where the primary intersecting neurodegeneration. We also implement a graph theory damage occur (arteriolar smooth muscle cell, endothelium, pericytes strategy to study the longitudinal neurodegenerative changes between and/or blood-brain barrier (BBB); Fig. 1-I) (Craggs et al., 2014). It has WM and GM in SVD. Finally, to improve the biological knowledge of been hypothesized that the perforating vessels arteriopathy causes connectomic findings, we used GM-WM connectomic phenotypes and chronic ischemia in WM due to a reduction in cerebral perfusion and genomic brain-wide gene expression atlas, the Allen Human Brain Atlas cerebral autoregulation with primary affectation of the mural vessel (AHBA; (Shen et al., 2012)), in searching of biological correlates of cells (mostly, smooth muscle cells)(Craggs et al., 2014; Markus, 2008). vascular degeneration vulnerability. In the past, his strategy has Other recent hypothesis suggests that diffuse intrinsic microvascular improved our understanding of individual brain differences (Xin et al., arteriopathy manifests throughout the brain as endothelial dysfunction 2018), brain plasticity (Ortiz-Terán et al., 2017), cortical organization (Wardlaw et al., 2019). The endothelial dysfunction increases BBB (Bueichekú et al., 2020; Diez et al., 2020) and neurogenetic contribu­ permeability and leads to leakage of plasma material into the vessel tions to neurological diseases, including dementia (Sepulcre et al., wall, causing inflammation and loss of structure and integrity of the 2018), among others (Diez and Sepulcre, 2021; Fornito et al., 2019). The WM, both at the microscopic level (only seen with advanced MRI) and overall approach of this study helped us: 1) to bridge the gap between macroscopic level producing leukoaraiosis (Li et al., 2018; Wardlaw the brain phenotypes and the specific genetic factors involved in SVD et al., 2019). The lack of identifiable causal triggers in sporadic SVD and 2) to validate the biological plausibility of the cortical maps for GM (compared to genetically associated forms) – along with multiple sus­ and WM primary or secondary vascular neurodegeneration. ceptibility genes identified in GWAS studies and complex relationships between vascular risk factors in aging populations- make it challenging 2. Methods to disentangle specific vascular vulnerability within the GM or/and the WM of the human brain (e.g. vascular regulation in pial, perforant, 2.1. Participants terminal arteries/capillaries, or other postulated BBB factors)(Pater­ noster et al., 2009; Tan et al., 2017; Tran et al., 2016; Xu et al., 2010). We used two independent MRI datasets to investigate the network In this study, we aimed to analyze the interacting and degenerating relationships between GM neurodegeneration and clinically silent WM patterns of WM and GM in SVD in vivo, using novel MRI graph-theory lesions (flow chart in Fig. 1-II): the Open Access Series of Image Study analytical approaches in cognitively normal participants. Graph-based (OASIS 3; (LaMontage et al., 2019) (Dataset 1) and the stage 2 data of analysis in neuroimaging is generally used to evaluate the complex the Cambridge Center for Aging and Neuroscience (Cam-CAN project; network relationships between brain areas (Diez et al., 2020; Hu et al., (Shafto et al., 2014) (Dataset 2). For our study, we considered the whole Fig. 1. I. The theoretical diagram shows the hypothesis for the SVD damage, with primary affectation of the cerebral blood flow regulation in perforant arteries from cortex to white matter in the vascular border zones. II. Flow chart of the participants included in both datasets. III. Detection of WM lesions in the T1 MRI sequences of both cohorts following by the creation of a WM lesion frequency map (the brightest the colour, the higher the frequency of the lesions with a maximum of 100 for Dataset 1 and 40 for Dataset 2). For each Dataset, the age range is described. IV. WM-GM brain graph. First, GM and WM data were voxel-level extracted, being each voxel a node of the graph. Then, we create correlation matrices between GM and WM lesions based on significant negative correlations (FDR correcteded). 2 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 spectrum of WMH -as a proxy to cerebral SVD. Briefly, the OASIS 3 and normal WM separation in our data was observed at 0.2 partial dataset includes 703 participants between 42 and 95-year-old with volume estimate. Thus, voxel in the WM that reached a higher PVE value normal cognition (controls), that we used in this analysis. Medical his­ than 0.2 were considered as WM lesion. Each resulting WM lesions map tory and cognitive assessment were revised for each participant was registered into the MNI152 standard space and down-sampled to 6 included. Patients with a previous history of stroke or neurological mm. Individual lesion maps were automatically extracted by using a conditions different from SVD were excluded. Within the control sub­ PVE algorithm for WM lesion segmentation in T1-MRIs (see example in jects, we only selected the stable cognitive participant, with a mean Supplementary Fig. 1). Lesion frequency map for all the subjects was centiloid values of amyloid deposition at baseline less than 10 and less computed as the sum of lesions at the voxel-level (Fig. 1-III). than 2 rate of growth in the follow up (LaMontage et al., 2019), to avoid the inclusion of preclinical Alzheimer’s disease. We use 2 subsets of 2.3. Cross-sectional analysis participants: one subset for the cross-sectional study and for the longi­ tudinal study (that includes participants within the cross-sectional 1. Structural connectivity analysis subset with at least one follow-up MRI session with a mean of 4.5 (±2.38) years between baseline and follow-up); and another one for the Whole brain GM and WM lesions maps underwent specific analysis to cognitive analysis, which was non overlapped with the cross-sectional investigate voxel-level network relationships (Fig. 1-IV). First, we subset. The cognitive evaluation included the following tests: Mini­ compute a partial correlation coefficient to create association matrices Mental State Examination (MMSE), logical memory from Weschler using all between (not within) GM and WM voxels’ pairs across the Memory test (WMT) immediate and delayed (units recall and time sample of individuals. The association matrices were controlled by age, elapse), digit span forward and backward (units and longest length), brain size, whole brain GM mean density and WM lesions mean density trail making A and B, one-minute verbal fluency test (animals and using partial correlation. We focused on negative coefficients, as our vegetables), Boston naming test, and Weschler Adult Intelligence Scale investigation targeted the inverse association between GM signal in­ (WAIS)-R Digit Symbol test. Then, Dataset 2 was used for validation/ tensity and WM lesions (the more WM lesions, the less GM volume). replication purposes. Briefly, the Cam-CAN project is a community- Association matrices were corrected using a false discovery rate (FDR) at based prospective study to analyze the influence of age in the struc­ q-level < 0.05 (Benjamini and Hochberg, 1995). An additional analysis ture and function of the brain across the lifespan. It includes 700 par­ corrected by sex and the diagnosis of hypertension were performed in ticipants older than 18-year-old without cognitive decline (MMSE > 24) the Dataset 1 and shown in Supplementary material. Using these asso­ and without history of neurological conditions such stroke, neurode­ ciation matrices, we computed the weighted degree of GM-WM lesions generative diseases, brain tumor, multiple sclerosis, or epilepsy. We relationships by summing all coefficients (Fig. 2-I and 2-II) obtaining used this cohort for the cross-sectional analysis, as follow-up MRIs were two matrices: one for the degree from each GM voxel to WM lesions not available. connections and another one from each WM lesions voxel to GM con­ nections. Additionally, we visualize the connectivity profiles and to­ 2.2. Image acquisition and preprocessing pology distributions of given GM of WM voxels (those with the highest weighted degree) (Fig. 2-III) using a seed-based approach. For the Dataset 1 and Dataset 2, we used the images obtained in 3 T MRIs. High resolution structural sequences, including T1-weighted im­ 2. GM-WM lesions bipartite interactions and modularity ages, were acquired in both protocols. For CAM-Can subjects, a TIM Trio 3 T (Siemens Medical Solutions USA, Inc) with 32 coils was used to An interesting aspect of GM and WM connections, in the context of acquire a Magnetization Prepared Rapid Gradient Echo (MP-RAGE) the appearance of network neurodegeneration and WM lesions, is sequence (TR = 2250 ms; TE = 2.99; FA = 9u and 1 mm isotropic voxel) whether they form close interacting neighborhoods (also known as (Shafto et al., 2014). For OASIS-3, a TIM Trio 3 T (Siemens Medical communities or modules). We know that, for instance, within WM, Solutions USA, Inc) with 20 coils was used to acquire a MP-RAGE voxels tend to cluster together to form specific areas or conglomerates of sequence TR = 2400 ms; TE = 3.1 ms; FA = 8u and 1 mm isotropic WM lesions. However, it is not well-known whether GM and WM form voxel) (LaMontage et al., 2019). tight interacting modules interacting between them. In this analysis, we We used the T1 sequence in both cohorts to perform voxel-based investigate GM-WM modules by searching first for robust bipartite morphometry (VBM) analysis and spatial detection of WM lesions; network motifs among the GM-WM association matrices (Fig. 3-I) using reflecting irreversible axonal damage in the white matter and avoiding a MATLAB toolbox for bipartite network analysis (Flores et al., 2016). the inclusion of potentially transient lesions in T2 and T2-FLAIR se­ First, we identified quadrangle motifs by selecting nodes that connects quences (Al-Janabi et al., 2019) and in agreement with the STandards to other nodes directly (1-Step connectivity), and indirectly in 3-steps. for ReportIng Vascular changes on nEuroimaging (STRIVE) consensus This strategy allowed us to obtain GM-WM modules that showed close (Wardlaw et al., 2013). An optimized VBM protocol (Good et al., 2001) neighborhood relationships, disregarding GM-WM connections without carried out with FSL tools (Douaud et al., 2007; Smith et al., 2004) was graph neighborhood properties. Second, we applied LP-BRIM bipartite used to study GM and WM maps. First, structural images were brain- modularity algorithm to visualize all different communities within that extracted and GM-segmented before being registered to the MNI 152 GM-WM network. Third, a WM degree map was computed counting the standard space using non-linear registration (Andersson et al., 2007). number of GM voxels related to that white matter voxel that were The resulting images were averaged and flipped along the x-axis to assigned to different communities normalized by the total amount of create a left-right symmetric, study-specific GM template. Second, all links of that WM voxel. Finally, we perform a Pearson correlation native GM images were non-linearly registered to this study-specific analysis between each motifs’ cluster (GM and WM together, GM and template and “modulated” to correct for local expansion (or contrac­ WM alone) and the punctuation in the cognitive test described before. tion) due to the non-linear component of the spatial transformation. The The analysis was adjusted by age and the years of education. modulated GM images were then smoothed with an isotropic Gaussian kernel with a sigma of 3 mm. Finally, we use the modulated GM partial 2.4. Longitudinal neuroimaging connectivity analysis volume estimation maps derived from VBM to identify WM lesions. Further details about the process could be found in Supplementary We investigated the temporal and spatial relationship, simulta­ Fig. 1. Briefly, while WM shows partial volume estimates values close to neously, between GM and WM lesions using network science tools, as 0, WM lesions that are visually hypointense in the T1 show higher partial previously described (Sepulcre et al., 2018). We developed a longitu­ volume estimates than normal WM. An optimal threshold for WM lesion dinal weighted-degree analysis (also known as hubs) to investigate the 3 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 Fig. 2. I. GM and WM graph analysis. Here, we calculate the weighted degree as the sum of the correlations for each node of GM (top) and each node of WM (bottom). II. Weighted degree centrality maps for GM and WM of Dataset 1 (A) and Dataset 2 (B): GM regions most influenced by WM lesions are located in orbitofrontal cortex, dorsolateral frontal cortex, insula and anterior cingulate cortex (the highest weighted degree centrality). For WM lesions maps, those regions most connected with GM are located in supratentorial white matter in middle cerebral artery (MCA) and anterior cerebral artery (ACA) watershed territories. III. Seed based analysis: for both cohorts, the WM lesions voxel with highest weighted degree was located in border zone areas, projecting with the highest correlation coefficient in dorsolateral frontal cortex and insula. The voxel of highest weighted degree in GM were located in cingulate cortex for OASIS and supra marginal gyrus for CAM-CAN, both projecting with the highest correlation coefficient widespread in supratentorial WM. temporal relationships between the GM and WM lesions. We used the neurogenetics analysis led to two sets of genes (one on the positive and longitudinal data of Dataset 1 for this analysis. We calculated the GM another one in the negative tail of the histogram; Fig. 5-I) that expressed and WM lesions maps in time 1 and time 2 and computed the association which protein-coding genes might have a biological relevance in the matrices between all voxels of GM signal in time 1 and WM lesions in GM-WM relationships. From the obtained gene set, we used four specific time 2 and GM signal in time 2 and WM lesions in time 1; as well as WM approaches to characterize its biological meaning: 1) we searched for lesions in time 1 and GM in time 2 and WM lesions in time 2 and GM in previously reported genes (a priori genes) related to vascular function­ time 1. The association matrices were controlled by age, brain size, the ality or SVD (bibliographic review); 2) we performed fold enrichment time between the first and the second MRI, whole brain GM mean analyses for Gene Ontology biological processes (Ashburner et al., 2000; density and WM lesions mean density using partial correlations. Links The gene Onthology Consortium, 2021) and Reactome pathways; 3) we surviving a false discovery rate (FDR) at q-level < 0.05 were used. We explored potential single cell gene expression patterns (of pericytes, obtained weighted degree maps to visualize GM and WM hubs in both smooth muscle cells, vascular fibroblast-like cells or microglia) of a relationships, namely, GM1-to-WM2 and WM1-to-GM2 temporal in­ priori genes related with the risk of SVD obtained in the spatial simi­ teractions (Fig. 4-I/II/III). larity analysis using a mouse molecular atlas of brain vasculature (He et al., 2018; Vanlandewijck et al., 2018); and 4) we used Genemania (Mostafavi et al., 2008) gene-gene network interactions from co- 2.5. Neuroimaging-genetics analysis expressions, colocalizations, genetic interactions, pathways, predicted physical interactions and shared proteins domains and Cytoscape (Lopes To gain biological insights about brain connectomic findings, we et al., 2011) software to perform an interactome analysis, including investigated the spatial cerebral similarities between GM1-to-WM2 and centrality assessments (degree and closeness centrality), to confirm WM1-to-GM2 cortical maps and protein-coding genetic profiles using which genes obtained in the spatial similarity analysis (in both left and the AHBA. The AHBA is a unique publicly available database that pro­ right tail of the histogram) play a central role in the GM-WM lesions vides whole-brain, high- resolution genome-wide expression values connectomic-genetic interactions. (Shen et al., 2012). The complete transcriptome dataset consists of 58,692 measurements of gene expression of 20,737 genes in 3702 brain 3. Results samples obtained from 6 individuals (French and Paus, 2015). We used a surface anatomical transformation (French and Paus, 2015) of the AHBA 3.1. Connectivity interactions between grey matter and white matter data based on the 68 cortical regions of the Desikan–Killiany atlas lesions (Desikan et al., 2006), which covers the entire cortex and uses individual vectors of the median cortical expression of protein-coding genes across We found that GM and WM lesions showed significant connectivity 68 cortical regions. To find gene expression profiles putatively associ­ relationships across the human brain (Fig. 2). For example, there are GM ated to vascular damage vulnerability, we performed a spatial similarity regions in orbitofrontal cortex, dorsolateral frontal cortex, insula and analysis between the discovered connectomic phenotypes from the anterior cingulate cortex -in both, Dataset 1 and Dataset 2 cohorts-, longitudinal analysis (GM maps for GM1 to WM2 and WM1 to GM2 and where volumetric signal in GM is heavily associated by the occurrence of AHBA data by means of MATLAB in-house–developed code based on a WM lesions in the supratentorial space (Fig. 2-II; upper row). As for WM linear regression approach (MATLAB R2019b; The MathWorks Inc.) areas, there are leukoaraiotic regions in supratentorial WM, in middle (Bueichekú et al., 2020; Diez and Sepulcre, 2018). Later, we focused our cerebral artery (MCA) and anterior cerebral artery (ACA) watershed investigations on genes that showed high spatial similarity with longi­ territories -in both cohorts- that are negatively related to GM volume tudinal GM-WMH connectomic map (statistically significant R-values across the cortical mantle, meaning that the more lesions in WM, the less were retained using a 95% threshold for the confidence interval (±2 SD) volume of GM (Fig. 2-II; bottom row). Our seed-based analysis in the entire transcriptome distribution, assigning positive R-values to confirmed local regional specificity between WM lesions and GM the GM1-WM2 map and negative R-values to the WM1-GM2 map). The 4 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 Fig. 3. GM-WM bipartite interactions and modularity. I. GM-WM bipartite graphs: we computed the network distances between all bipartite graph nodes (left). Then, we identified quadrangle motifs by selecting nodes that connects to other nodes directly (1-Step connectivity), and indirectly in 3-steps (right). II. Location of modules of GM (blue) and WM (yellow) interconnected. III. Pearson correlation matrix between modules and cognitive test. We found a negative correlation with trail B and module 1 and positive correlations between module 1 and 3 and WAIS-R; and between VEG and module 1. The correlations are shown with 95% confidence interval. Logimem: Logical memory immediate; Digitf: digit spam forward total number of trial correct; Digitflen: Digit spam forward length; Digitb: digit spam backward total number of trial correct; Digitblen: Digit spam backward length; Animals: Category fluency test animals; VEG: Category fluency test vegetables; Trail A: Trail making test part A; Trail B: Trail making test part B; WAIS: WAIS-R Digit Symbol; Memunits: Logical memory delayed, units recall; Memtime: time between logical memory test; Boston: Boston naming test; MMSE: MiniMental State Exam. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) interactions, such as frontal WM and GM, as well as more widespread which prefrontal and frontal lateral GM interacts with supratentorial WM lesions and GM relationships, such as fronto-parietal WM lesions WM; 3) module 3, with GM located in insula and prefrontal cortex in­ and GM volume changes mostly in frontal lobes (Fig. 2-III). For the teracts with WM tracts of prefrontal lobes and 4) module 4, with medial Dataset 1, the voxel with highest weighted degree in WM is located in frontal cortex interacting with tiny regions of WM in frontal lobes. MCA-ACA left border-zone region; and is highly connected with left We observed that GM-WM lesions modules, that is, combined sys­ dorsolateral frontal cortex. The voxel with highest weighted degree in tems integrating GM and WM at the same time, explain the cognitive GM is in the cingulate cortex; and is highly connected with spread voxels performance of the sample, particularly in the WAIS-R, VEG and Trail B of WM, mostly located in MCA-ACA watershed zones. For the Dataset 2 scores (correlations and scatterplots in Fig. 3-III; FDR-corrected) for cohort, the results were similar for the projections from WM to GM modules 1 and 3 (involving the GM-WM connections in frontal and (adding insula as cortex target) but the highest weighted degree region temporal lobes) without significant association with modules 2 and 4 of GM was the supramarginal gyrus, also highly connected with supra­ (involving mostly the connections of supratentorial WM with the pre­ tentorial watershed WM. Finally, we observed the adjustment by the frontal and frontal lateral cortex). We did not find any significant FDR- diagnosis of hypertension did not significantly change the GM and WM corrected correlation with cognitive test when GM or WM modules are lesions weighted degree maps (Supplementary Fig. 1). taken independently. 3.2. Cognition and connectivity neighborhoods of grey matter and white 3.3. Temporal interactions from grey matter to white matter lesions (and matter lesions vice versa) Using a bipartite modularity analysis, we observed 4 main modules Our longitudinal analysis not only corroborated areas of high GM- of highly stable voxels of GM volumetric changes and WM lesions WM lesions connectivity degree found in the cross-sectional study but (Fig. 3-II): 1) module one, in which the GM of medial temporal lobes and also found found that specific regions in the brain display predominant fronto-basal regions interacts with prefrontal WM tracts; 2) module 2 in vulnerability for GM1-to-WM2 or WM1-to-GM2 degeneration (Fig. 4). 5 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 Fig. 4. Longitudinal GM-WM connectivity. I. First, we calculate the weighted degree for GM and WM in time 1 and in time 2. We subtracted the correlations (time 1 minus time 2) and obtained maps for GM and WM. Those regions with positive correlations mean that the causal relationships go from time 1 to time 2 and negative correlations mean the causal relationship go from time 2 to time 1. II. Cortical maps for GM hubs: in orange (positive correlations) are shown the cortical areas that lead to secondary degeneration of the WM and in blue (negative correlation) are shown the cortical areas secondary degenerated due to the presence of WM lesions. III. WM maps for WM hubs: in orange (positive) are shown the WM zones that cause degeneration in the GM, mostly located in the inferior front occipital tract and in blue (negative) are shown the WM areas affected by primary vascular damage from GM, mostly border zone between MCA and ACA. For GM and WM maps, we show the statistically significant weighted-degree t-statistic maps. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) We observed two distinctive progression patterns: 1) one in which GM left column in III). Finally, we observed that the adjustment by the volume changes, in areas such as the dorsolateral frontal cortex, pre and diagnosis of hypertension did not significantly change the GM1-to-WM2 post Rolandic gyrus, supplementary motor area and some cingulate and WM1-to-GM2 maps (Supplementary Fig. 1). areas, that lead to the eventual appearance of secondary WM lesions located in supratentorial WM, in watershed regions of MCA and ACA 3.4. Neurogenetic signatures of grey matter and white matter network (Fig. 4, yellow cortical areas in II, and right column in III), and 2) degeneration another one in which WM lesions, particularly in the inferior fronto- occipital fasciculus, produce subsequent cortical degeneration on areas The biological mechanism that drives GM and WM alterations in SVD such as temporal pole, anterior cingulate cortex, insula, orbitofrontal are likely to be interdependent or causally related. Our cross-sectional cortex and inferior temporal ACA (Fig. 4, blue cortical areas in II, and and longitudinal findings described specific connectomic phenotypes 6 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 Fig. 5. I. Spatial similarity analysis between the discovered connectomic phenotypes (longitudinal GM maps) and AHBA data. In the histogram are represented the 20,737 genes identified in the AHBA (Y-axis) in our cortical map (X-axis). The genes founded beyond +/-2SD were considered significantly spatially associated similar with the connectomic SVD map. II. Fold enrichment analysis (FDR corrected) of of of genes significantly associated with the connectomic SVD map (>2SD) shows the implication of these genes in cell communication, muscle conduction and ion transmembrane transport and electrical conduction). III. Single gene expression from a mouse molecular atlas showed high expression in smooth muscle cells and pericytes of several key genes obtain in approach I. IV and V. Inter­ actomic and spatial similarity analyses of the newly discovered genes associated to the connectomic SVD map showed that AR displ displays a top network centrality (both degree centrality and closeness centrality). that support that view. Consequently, we investigated whether these enrichment significance. The complete lists of genes (for GM1-to-WM2 connectomic phenotypes display biological potential at the SVD mech­ and for WM1-to-GM2) with their description are included in Supple­ anistic level. Thus, we performed a spatial cerebral similarity analysis mentary material. with the AHBA. Interestingly, the GM1-to-WM2 phenotype identified a Next, we used a single-cell atlas atlas to verify whether AR, CNN1, gene set of 683 protein-coding genes, in which several a priori genes ITGA8 and MYLK 2 are specifically expressed in brain vascular cells have been associated to SVD and vascular smooth muscle cells before, (pericytes, smooth muscle cells and vascular fibroblast-like cells). namely AR, CNN1, CSF3, ITGA8 and MYLK2 genes (Fig. 5-I). Moreover, Among these four genes, only CNN1 and AR display high cellular we analyzed the entire set of genes associated to the GM1-to-WM2 specificity for arteriolar smooth muscle cells (Fig. 5-III). Finally, we used phenotype (or GM1-to-WM2 related genes) and we observed signifi­ the entire gene set of 683 genes obtained from the GM1-to-WM2 con­ cant fold enrichments in ionic transmembrane transport, cardiac con­ nectivity map and an interactome analysis to investigate whether any of duction and electrical cell communication functionalities (Fig. 5-I/II), them play central roles using in an independent dataset of gene-gene genetic domains that are known to be related not just to cardiac cells but network interactions. We found that only AR displayed a central inter­ also to vessels cells (such as smooth muscle cells). actomic role in this genetic network (both degree centrality and close­ The WM1-to-GM2 phenotype showed significantly cortical similarity ness centrality; Fig. 5-IV) supporting its its core role in the overall with 476 genes in which we identified IL1-b, IL-6 and TRIBE3 as previous GM-WM lesions connectomic-genetics. Fig. 5-V shows the potentially relevant for SVD pathology. However, the entire set of genes cortical spatial similarity between AR gene (i.e., brain regions linked to associated to this type of connectivity pattern did not reach any fold the expression of this gene) and GM1-to-WM2 connectomic phenotype 7 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 (Fig. 5-V). remotely to focal ischemic areas in the CADASIL brain (Viswanathan et al., 2006). This suggests that secondary degeneration from local le­ 4. Discussion sions either GM or WM can potentially affect distant regions, creating a complex network of interacting cerebral degeneration. Our connectomic The brain pathology of SVD beyond the accumulation of WM lesions findings correspond with these observations. On one hand, we found is poorly understood. Diffuse arteriopathy of the GM and WM perfo­ that primary GM changes can lead to WM lesions in border vascular rating vessels due to dysfunctional cerebral blood flow regulation has zones, meaning that the primary affectation of the pial arteries in the GM been postulated as a key factor in the development of SVD (Promju­ might induce vascular and molecular dysfunctionality in distal capil­ nyakul et al., 2018). The current study aimed to close the gap in un­ laries within vascular-vulnerable zones of the human brain. On the other derstanding the complex relationship of GM and WM pathological hand, we also found that the primary WM lesions in association fascic­ changes from the viewpoint of cerebral connectomics. We hypothesize ulus, such as the fronto-occipital fasciculus, can lead to secondary GM that alterations in GM and WM do not occur in isolation but in modular degeneration in projecting areas at the frontal lobe, with little effect of coordination to arise final SVD phenotypes. Particularly, we speculate hypertension. Thus, our results suggest that there are more pathological that WM lesions and GM neurodegeneration both share the ability to be factors behind the sporadic SVD than the chronic hypoperfusion theory the primary source of vascular-molecular dysfunctions, in which sec­ behind the appearance of WM lesions. In light of the GM-WM bidirec­ ondary degeneration might occur in distant areas of WM or GM, tional causal damage interactions found in this study, we believe that depending on the primary source. This multilayer interdigitation of sporadic SVD is the result not only of primary WM lesions with sec­ primary and secondary damage makes graph theory an ideal framework ondary degeneration of cortical areas, but also the occurrence of pri­ to disentangle them. mary mis-regulation of cerebral blood flow in GM pial arteries that Previous studies have shown that WMH could cause structural con­ infringe damage toward downstream and distal artery territories. nectivity alterations related to SVD. Habes et al. show that normal aging The diversity of underlying pathologies and vascular risk factors and WMH cause different affectation patterns of GM: WMH could pri­ related to SVD make its genetic background a complex scenario which marily affect the orbitofrontal cortex and temporal pole (Habes et al., remains largely unclear (Iadecola, 2013). In this study, the integration of 2016). Liu et al., using a measure of density fiber, found less density fiber neuroimaging connectivity SVD phenotypes and transcriptomic data in frontal, temporal, and parietal regions related to SVD in cognitively from the AHBA has helped us to identify several genetic signatures that normal subjects (Liu et al., 2020). Using novel cross-sectional and lon­ might shed light for future studies. Particularly, we found that AR, gitudinal graph theory metrics, we find converging evidence similar to CNN1, MYLK2 and ITGA8 gene expression levels are highly similar to those structural connectivity descriptions. Critically, we provide new the SVD-related map which primary longitudinal degeneration of in situ data about the bidirectional nature of GM and WM interactions. Our GM led to secondary vascular damage in distant WM. We also observed results show that GM changes can lead to secondary WM degeneration in that the entire genetic profile was highly enriched on genetic function­ vascular border zones areas (irrigated by pial arteries of MCA and ACA), alities that are putatively related to smooth muscle cells and their bio­ whereas WM lesions can lead to secondary GM degeneration in cortical logical processing domains (ion transmembrane transport and electrical projecting areas. Moreover, we found this GM-WM communication is conduction). These findings point to plausible causative roles of cortical organized in close and stable connectivity neighborhoods; a network alterations in the SVD pathological mechanisms. Interestingly, among organization feature that seems critical for an adequate cognitive per­ the 4 candidate genes, AR (gene coding for the androgen receptor pro­ formance. In fact, we observed that the GM-WM connectivity neigh­ tein) stood out as the most central interactomic role in an independent borhoods are related with cognition, particularly with the executive genetic network dataset. AR has been identified previously in GWAS function, which affectation is the main feature of vascular cognitive studies of vascular dementia, with almost 4-times higher risk of vascular decline (O’Brien and Thomas, 2015). Previous studies of structural and dementia when present (Schrijvers et al., 2012). There is an increasing functional connectivity in SVD have shown that the global brain number of studies that demonstrate the role of AR in the vascular network is altered, and this alteration mediates the results in cognitive damage (Son et al., 2010), and vascular risk factors as hypertension scores (Lawrence et al., 2014). Here we go one step further and saw that (Huang et al., 2016). The AR gene is highly expressed in vascular smooth connectivity modules between GM and WM (and not isolated GM or WM muscle cells (Son et al., 2010) and the activity of androgen/AR signaling regions) are key in this SVD-cognitive relationship. were pointed as risk factors of hypertension and stroke (Huang et al., The pathophysiology interpretation of SVD has evolved over the 2016). The results of our study complement the existing evidence, as the years, from the chronic hypoperfusion as the central hypothesis of brain AR gene was significantly expressed in those cortical regions that lead to damage to the endothelium and BBB disruption as critical points of the degeneration in the WM and confirm our hypothesis of primary vascular disease (Wardlaw et al., 2019). Recent studies have shown that vascular damage in GM that might cause WM lesions. Regarding the reverse reactivity is impaired in the cerebral vasculature before developing the pattern -primary vascular damage in WM leading to secondary GM disease (Sam et al., 2016; Shi et al., 2020), supporting the hypothesis of degeneration- we identified several brain damage and inflammatory chronic vascular impairment with global brain hypoperfusion. Some related genes associated to (IL6, IL1b and TRIBE3 genes). This proin­ other studies have pointed out that WMH might be just the consequence flammatory genetic profile did not reach any significant fold enrichment of altered cerebral blood flow somewhere else (Shi et al., 2016). More­ and therefore, they seem less conclusive compare to our findings with over, others have suggested that BBB alterations in WMH could be the the AR centered gene set, however, it deserves future investigations. For primary cause of cerebral damage (Muñoz Maniega et al., 2017; Ward­ instance, IL6 and IL1b are both implicated in the inflammatory cascade, law et al., 2017). However, studies developed in spontaneously hyper­ codifying proinflammatory cytokines and both are well-known ischemic tensive stroke-prone rats models show credible evidence that vascular damage biomarkers; both are related to the outcome of ischemic stroke wall damage with narrowing of the arterioles (Jandke et al., 2018) and (Martínez-Sánchez et al., 2014; Ramiro et al., 2018). Recently, Stas­ an alteration of the vascular reactivity seems to be an early causal zewski et al. found that IL-6 relates to the radiological progression of phenomenon (in which an initial vasodilatation and increase of oxygen small vessel disease (Staszewski et al., 2018)). Moreover, in brain exchange eventually compromises and evolves into vasoconstriction and hypoperfusion models, animals with lower levels of Il1b and Il6 had reduction of oxygen exchange). Interestingly, wall damage activates the lower necrotic cells in the hippocampal cells and better performance in inflammatory response that leads to the BBB break (Mencl et al., 2013). the cognitive test (Ishida et al., 2020). These findings support the idea Moreover, neuropathological studies have found that these primary that IL6 and IL1b genes in cortical areas can explain in part some of the local effects produce changes in distant areas of the brain. For instance, secondary degeneration process occurring in SVD. it has been observed glial apoptosis in cortical-subcortical WM junctions In conclusion, using graph theory-based analysis and connectomic- 8 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 genetic approaches, this study has led us to interpret SVD as a bidirec­ Flores, C.O., Poisot, T., Valverde, S., Weitz, J.S., 2016. BiMat: a MATLAB package to facilitate the analysis of bipartite networks. Methods Ecol. Evol. 7, 127–132. https:// tional degeneration process between GM and WM lesions. This frame­ doi.org/10.1111/2041-210X.12458. work opens new venues to understand the underlying interactomic Fornito, A., Arnatkevičiūtė, A., Fulcher, B.D., 2019. Bridging the gap between mechanism in which the SVD develops as well as to investigate key connectome and transcriptome. Trends Cogn. Sci. 23, 34–50. https://doi.org/ pathophysiology features associated to its molecular background. 10.1016/j.tics.2018.10.005. French, L., Paus, T., 2015. A FreeSurfer view of the cortical transcriptome generated from Importantly, our findings support the central role of AR for the brain the Allen Human Brain Atlas. Front. Neurosci. 9, 1–5. https://doi.org/10.3389/ lesion vulnerability and network breakdown in the context of SVD. fnins.2015.00323. Supplementary data to this article can be found online at https://doi. Good, C.D., Johnsrude, I.S., Ashburner, J., Henson, R.N.A., Friston, K.J., Frackowiak, R.S. J., 2001. A voxel-based morphometric study of ageing in 465 normal adult human org/10.1016/j.nbd.2022.105671. brains. Neuroimage 14, 21–36. https://doi.org/10.1006/nimg.2001.0786. Habes, M., Erus, G., Toledo, J.B., Zhang, T., Bryan, N., Launer, L.J., Rosseel, Y., Authors contributions Janowitz, D., Doshi, J., Van Der Auwera, S., Von Sarnowski, B., Hegenscheid, K., Hosten, N., Homuth, G., Völzke, H., Schminke, U., Hoffmann, W., Grabe, H.J., Davatzikos, C., 2016. White matter hyperintensities and imaging patterns of brain The study protocol was designed and written by the authors (R.G.Z., ageing in the general population. Brain 139, 1164–1179. https://doi.org/10.1093/ I.D., J.S.). R.G.Z. and J.S. obtained grant funding. R.G.Z., J.S., I.D. and E. brain/aww008. Habes, M., Sotiras, A., Erus, G., Toledo, J.B., Janowitz, D., Wolk, D.A., Shou, H., B. analyzed the data and performed statistical analysis. R.G.Z. and J.S. Bryan, N.R., Doshi, J., Völzke, H., Schminke, U., Hoffmann, W., Resnick, S.M., interpreted clinical data. V.M., C.M.K., E.B., I.D., W.O., M.G., B.F. and E. Grabe, H.J., Davatzikos, C., 2018. White matter lesions spatial heterogeneity, links to D.T. provided scientific input during protocol design and interpretation risk factors, cognition, genetics, and atrophy. Neurology 91, E964–E975. https://doi. org/10.1212/WNL.0000000000006116. of the study. The manuscript was written by R.G.Z. and J.S. together He, L., Vanlandewijck, M., Mäe, M.A., Andrae, J., Ando, K., Del Gaudio, F., Nahar, K., with all co-authors, who vouch for the accuracy of the data reported and Lebouvier, T., Laviña, B., Gouveia, L., Sun, Y., Raschperger, E., Segerstolpe, Å., adherence to the protocol. All authors edited and approved the Liu, J., Gustafsson, S., Räsänen, M., Zarb, Y., Mochizuki, N., Keller, A., Lendahl, U., Betsholtz, C., 2018. Single-cell RNA sequencing of mouse brain and lung vascular manuscript. and vessel-associated cell types. Sci. Data 5, 180160. https://doi.org/10.1038/ sdata.2018.160. Hu, C., Cheng, L., Sepulcre, J., Johnson, K.A., Fakhri, G.E., Lu, Y.M., Li, Q., 2015. Declaration of Competing Interest A spectral graph regression model for learning brain connectivity of Alzheimer’s disease. PLoS One 10, 1–24. https://doi.org/10.1371/journal.pone.0128136. The authors declared no potential conflicts of interest with respect to Huang, C.-K., Lee, S.O., Chang, E., Pang, H., Chang, C., 2016. Androgen receptor (AR) in cardiovascular diseases. J. Endocrinol. 229, R1–R16. https://doi.org/10.1530/JOE- the research, authorship, and/or publication of this article. 15-0518. Iadecola, C., 2013. The pathobiology of vascular dementia. Neuron 80, 844–866. https:// Acknowledgment doi.org/10.1016/j.neuron.2013.10.008. Ishida, N., Saito, M., Sato, S., Koepsell, H., Taira, E., Hirose, M., 2020. SGLT1 participates in the development of vascular cognitive impairment in a mouse model of small This research was supported by grants from the National Institutes of vessel disease. Neurosci. Lett. 727, 134929 https://doi.org/10.1016/j. Health (NIH; R01AG061811, and R01AG061445 to J.S.) and from the neulet.2020.134929. Jandke, S., Garz, C., Schwanke, D., Sendtner, M., Heinze, H.J., Carare, R.O., Schreiber, S., Institute of Health Carlos III (CM18/00175 to RGZ). 2018. The association between hypertensive arteriopathy and cerebral amyloid angiopathy in spontaneously hypertensive stroke-prone rats. Brain Pathol. 28, References 844–859. https://doi.org/10.1111/bpa.12629. Lambert, C., Benjamin, P., Zeestraten, E., Lawrence, A.J., Barrick, T.R., Markus, H.S., 2016. Longitudinal patterns of leukoaraiosis and brain atrophy in symptomatic small Al-Janabi, O.M., Bauer, C.E., Goldstein, L.B., Murphy, R.R., Bahrani, A.A., Smith, C.D., vessel disease. Brain 139, 1136–1151. https://doi.org/10.1093/brain/aww009. Wilcock, D.M., Gold, B.T., Jicha, G.A., 2019. White matter hyperintensity regression: LaMontage, P., Benzinger, T., Morris, John C., Keefe, S., Hornbeck, R., Xiong, C., comparison of brain atrophy and cognitive profiles with progression and stable Grant, E., Hassenstab, J., Krista, V., Andrei, G., Raichle, M., Cruchaga, C., Marcus, D., groups. Brain Sci. 9 https://doi.org/10.3390/brainsci9070170. 2019. OASIS-3: Longitudinal Neuroimaging, CLinica, and Cognitive Dataset for Andersson, J.L.R., Jenkinson, M., Smith, S., 2007. Non-linear Registration aka Spatial Normal Aging and Alzheimer Disease. MedXriv. https://doi.org/10.1101/ Normalisation (FMRIB Technical Report TRO7JA2). 2019.12.13.19014902. Ashburner, M., Ball, C.A., Blake, J.A., Botstein, D., Butler, H., Cherry, J.M., Davis, A.P., Lawrence, A.J., Chung, A.W., Morris, R.G., Markus, H.S., Barrick, T.R., 2014. Structural Dolinski, K., Dwight, S.S., Eppig, J.T., Harris, M.A., Hill, D.P., Issel-Tarver, L., network efficiency is associated with cognitive impairment in small-vessel disease. Kasarskis, A., Lewis, S., Matese, J.C., Richardson, J.E., Ringwald, M., Rubin, G.M., Neurology 83, 304–311. https://doi.org/10.1212/WNL.0000000000000612. Sherlock, G., 2000. Gene ontology: tool for the unification of biology. Nat. Genet. 25, Li, Q., Yang, Y., Reis, C., Tao, T., Li, W., Li, X., Zhang, J.H., 2018. Cerebral small vessel 25–29. https://doi.org/10.1038/75556. disease. Cell Transplant. 27, 1711–1722. https://doi.org/10.1177/ Benjamini, Y., Hochberg, Y., 1995. Controlling the false discovery rate: a practical and 0963689718795148. powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57, 289–300. https:// Liu, C., Shi, L., Zhu, Wenhao, Yang, S., Sun, P., Qin, Y., Tang, X., Zhang, S., Yao, Y., doi.org/10.1111/j.2517-6161.1995.tb02031.x. Wang, Z., Zhu, Wenzhen, Wang, D., 2020. Fiber connectivity density in cerebral Bueichekú, E., Aznárez-Sanado, M., Diez, I., d’Oleire Uquillas, F., Ortiz-Terán, L., small-vessel disease patients with mild cognitive impairment and cerebral small- Qureshi, A.Y., Suñol, M., Basaia, S., Ortiz-Terán, E., Pastor, M.A., Sepulcre, J., 2020. vessel disease patients with normal cognition. Front. Neurosci. 14, 1–11. https://doi. Central neurogenetic signatures of the visuomotor integration system. Proc. Natl. org/10.3389/fnins.2020.00083. Acad. Sci. 117, 6836–6843. https://doi.org/10.1073/pnas.1912429117. Lopes, C.T., Franz, M., Kazi, F., Donaldson, S.L., Morris, Q., Bader, G.D., Dopazo, J., Craggs, L.J.L., Yamamoto, Y., Deramecourt, V., Kalaria, R.N., 2014. Microvascular 2011. Cytoscape Web: an interactive web-based network browser. Bioinformatics 27, pathology and morphometrics of sporadic and hereditary small vessel diseases of the 2347–2348. https://doi.org/10.1093/bioinformatics/btq430. brain. Brain Pathol. 24, 495–509. https://doi.org/10.1111/bpa.12177. Markus, H.S., 2008. Genes, endothelial function and cerebral small vessel disease in man. Desikan, R.S., Ségonne, F., Fischl, B., Quinn, B.T., Dickerson, B.C., Blacker, D., Exp. Physiol. 93, 121–127. https://doi.org/10.1113/expphysiol.2007.038752. Buckner, R.L., Dale, A.M., Maguire, R.P., Hyman, B.T., Albert, M.S., Killiany, R.J., Martínez-Sánchez, P., Gutiérrez-Fernández, M., Fuentes, B., Masjuán, J., de Cases, M.A. 2006. An automated labeling system for subdividing the human cerebral cortex on L., Novillo-López, M.E., Díez-Tejedor, E., 2014. Biochemical and inflammatory MRI scans into gyral based regions of interest. Neuroimage 31, 968–980. https://doi. biomarkers in ischemic stroke: translational study between humans and two org/10.1016/j.neuroimage.2006.01.021. experimental rat models. J. Transl. Med. 12, 220. https://doi.org/10.1186/s12967- Diez, I., Sepulcre, J., 2018. Neurogenetic profiles delineate large-scale connectivity 014-0220-3. dynamics of the human brain. Nat. Commun. 9, 1–10. https://doi.org/10.1038/ Mencl, S., Garz, C., Niklass, S., Braun, H., Göb, E., Homola, G., Heinze, H.J., Reymann, K. s41467-018-06346-3. G., Kleinschnitz, C., Schreiber, S., 2013. Early microvascular dysfunction in cerebral Diez, I., Sepulcre, J., 2021. Unveiling the neuroimaging-genetic intersections in the small vessel disease is not detectable on 3.0 Tesla magnetic resonance imaging: a human brain. Curr. Opin. Neurol. 34, 480–487. https://doi.org/10.1097/ longitudinal study in spontaneously hypertensive stroke-prone rats. Exp. Transl. WCO.0000000000000952. Stroke Med. 5, 5–9. https://doi.org/10.1186/2040-7378-5-8. Diez, I., Larson, A.G., Nakhate, V., Dunn, E.C., Fricchione, G.L., Nicholson, T.R., Mostafavi, S., Ray, D., Warde-Farley, D., Grouios, C., Morris, Q., 2008. GeneMANIA: a Sepulcre, J., Perez, D.L., 2020. Early-life trauma endophenotypes and brain real-time multiple association network integration algorithm for predicting gene circuit–gene expression relationships in functional neurological (conversion) function. Genome Biol. 9, S4. https://doi.org/10.1186/gb-2008-9-s1-s4. disorder. Mol. Psychiatry. https://doi.org/10.1038/s41380-020-0665-0. Muñoz Maniega, S., Chappell, F.M., Valdés Hernández, M.C., Armitage, P.A., Makin, S. Douaud, G., Smith, S., Jenkinson, M., Behrens, T., Johansen-Berg, H., Vickers, J., D., Heye, A.K., Thrippleton, M.J., Sakka, E., Shuler, K., Dennis, M.S., Wardlaw, J.M., James, S., Voets, N., Watkins, K., Matthews, P.M., James, A., 2007. Anatomically 2017. Integrity of normal-appearing white matter: influence of age, visible lesion related grey and white matter abnormalities in adolescent-onset schizophrenia. Brain 130, 2375–2386. https://doi.org/10.1093/brain/awm184. 9 R. Gutiérrez-Zúñiga et al. Neurobiology of Disease 167 (2022) 105671 burden and hypertension in patients with small-vessel disease. J. Cereb. Blood Flow 2004. Advances in functional and structural MR image analysis and implementation Metab. 37, 644–656. https://doi.org/10.1177/0271678X16635657. as FSL. Neuroimage 23, S208–S219. https://doi.org/10.1016/j. O’Brien, J.T., Thomas, A., 2015. Vascular dementia. Lancet 386, 1698–1706. https://doi. neuroimage.2004.07.051. org/10.1016/S0140-6736(15)00463-8. Son, B.K., Akishita, M., Iijima, K., Ogawa, S., Maemura, K., Yu, J., Takeyama, K., Kato, S., Ortiz-Terán, L., Diez, I., Ortiz, T., Perez, D.L., Aragón, J.I., Costumero, V., Pascual- Eto, M., Ouchi, Y., 2010. Androgen receptor-dependent transactivation of growth Leone, A., El Fakhri, G., Sepulcre, J., 2017. Brain circuit-gene expression arrest-specific gene 6 mediates inhibitory effects of testosterone on vascular relationships and neuroplasticity of multisensory cortices in blind children. Proc. calcification. J. Biol. Chem. 285, 7537–7544. https://doi.org/10.1074/jbc. Natl. Acad. Sci. U. S. A. 114, 6830–6835. https://doi.org/10.1073/ M109.055087. pnas.1619121114. Staszewski, J., Piusińska-Macoch, R., Brodacki, B., Skrobowska, E., Stepien, A., 2018. IL- Paternoster, L., Chen, W., Sudlow, C.L.M., 2009. Genetic determinants of white matter 6, PF-4, sCD40 L, and homocysteine are associated with the radiological progression hyperintensities on brain scans: a systematic assessment of 19 candidate gene of cerebral small-vessel disease: a 2-year follow-up study. Clin. Interv. Aging 13, polymorphisms in 46 studies in 19 000 subjects. Stroke 40, 2020–2026. https://doi. 1135–1141. https://doi.org/10.2147/CIA.S166773. org/10.1161/STROKEAHA.108.542050. Tan, R., Traylor, M., Rutten-Jacobs, L., Markus, H., 2017. New insights into mechanisms Promjunyakul, N.O., Dodge, H.H., Lahna, D., Boespflug, E.L., Kaye, J.A., Rooney, W.D., of small vessel disease stroke from genetics. Clin. Sci. 131, 515–531. https://doi.org/ Silbert, L.C., 2018. Baseline NAWM structural integrity and CBF predict 10.1042/CS20160825. periventricular WMH expansion over time. Neurology 90, e2107–e2118. https://doi. Ter Telgte, A., Wiegertjes, K., Gesierich, B., Baskaran, B.S., Marques, J.P., Kuijf, H.J., org/10.1212/WNL.0000000000005684. Norris, D.G., Tuladhar, A.M., Duering, M., De Leeuw, F.E., 2020. Temporal dynamics Ramiro, L., Simats, A., García-Berrocoso, T., Montaner, J., 2018. Inflammatory molecules of cortical microinfarcts in cerebral small vessel disease. JAMA Neurol. 77, 643–647. might become both biomarkers and therapeutic targets for stroke management. https://doi.org/10.1001/jamaneurol.2019.5106. Ther. Adv. Neurol. Disord. 11 https://doi.org/10.1177/1756286418789340, The Gene Onthology Consortium, 2021. The gene ontology resource: enriching a GOld 175628641878934. mine. Nucleic Acids Res. 49, D325–D334. https://doi.org/10.1093/nar/gkaa1113. Rizvi, B., Narkhede, A., Last, B.S., Budge, M., Tosto, G., Manly, J.J., Schupf, N., Tran, T., Cotlarciuc, I., Yadav, S., Hasan, N., Bentley, P., Levi, C., Worrall, B.B., Mayeux, R., Brickman, A.M., 2018. The effect of white matter hyperintensities on Meschia, J.F., Rost, N., Sharma, P., 2016. Candidate-gene analysis of white matter cognition is mediated by cortical atrophy. Neurobiol. Aging 64, 25–32. https://doi. hyperintensities on neuroimaging. J. Neurol. Neurosurg. Psychiatry 87, 260–266. org/10.1016/j.neurobiolaging.2017.12.006. https://doi.org/10.1136/jnnp-2014-309685. Sam, K., Crawley, A.P., Conklin, J., Poublanc, J., Sobczyk, O., Mandell, D.M., Vanlandewijck, M., He, L., Mäe, M.A., Andrae, J., Ando, K., Del Gaudio, F., Nahar, K., Venkatraghavan, L., Duffin, J., Fisher, J.A., Black, S.E., Mikulis, D.J., 2016. Lebouvier, T., Laviña, B., Gouveia, L., Sun, Y., Raschperger, E., Räsänen, M., Zarb, Y., Development of white matter hyperintensity is preceded by reduced cerebrovascular Mochizuki, N., Keller, A., Lendahl, U., Betsholtz, C., 2018. A molecular atlas of cell reactivity. Ann. Neurol. 80, 277–285. https://doi.org/10.1002/ana.24712. types and zonation in the brain vasculature. Nature 554, 475–480. https://doi.org/ Schrijvers, E.M.C., Schürmann, B., Koudstaal, P.J., Van Den Bussche, H., Van Duijn, C.M., 10.1038/nature25739. Hentschel, F., Heun, R., Hofman, A., Jessen, F., Kölsch, H., Kornhuber, J., Peters, O., Viswanathan, A., Gray, F., Bousser, M.G., Baudrimont, M., Chabriat, H., 2006. Cortical Rivadeneira, F., Rüther, E., Uitterlinden, A.G., Riedel-Heller, S., Dichgans, M., neuronal apoptosis in CADASIL. Stroke 37, 2690–2695. https://doi.org/10.1161/01. Wiltfang, J., Maier, W., Breteler, M.M.B., Ikram, M.A., 2012. Genome-wide STR.0000245091.28429.6a. association study of vascular dementia. Stroke 43, 315–319. https://doi.org/ Wardlaw, J.M., Smith, E.E., Biessels, G.J., Cordonnier, C., Fazekas, F., Frayne, R., 10.1161/STROKEAHA.111.628768. Lindley, R.I., O’Brien, J.T., Barkhof, F., Benavente, O.R., Black, S.E., Brayne, C., Sepulcre, J., Grothe, M.J., d’Oleire Uquillas, F., Ortiz-Terán, L., Diez, I., Yang, H.-S.S., Breteler, M., Chabriat, H., DeCarli, C., de Leeuw, F.E., Doubal, F., Duering, M., Jacobs, H.I.L.L., Hanseeuw, B.J., Li, Q., El-Fakhri, G., Sperling, R.A., Johnson, K.A., Fox, N.C., Greenberg, S., Hachinski, V., Kilimann, I., Mok, V., van Oostenbrugge, R., 2018. Neurogenetic contributions to amyloid beta and tau spreading in the human Pantoni, L., Speck, O., Stephan, B.C.M., Teipel, S., Viswanathan, A., Werring, D., cortex. Nat. Med. 24, 1910–1918. https://doi.org/10.1038/s41591-018-0206-4. Chen, C., Smith, C., van Buchem, M., Norrving, B., Gorelick, P.B., Dichgans, M., Shafto, M.A., Tyler, L.K., Dixon, M., Taylor, J.R., Rowe, J.B., Cusack, R., Calder, A.J., 2013. Neuroimaging standards for research into small vessel disease and its Marslen-Wilson, W.D., Duncan, J., Dalgleish, T., Henson, R.N., Brayne, C., Cam-CAN, contribution to ageing and neurodegeneration. Lancet Neurol. 12, 822–838. https:// Matthews, F.E., 2014. The Cambridge Centre for Ageing and Neuroscience (Cam- doi.org/10.1016/S1474-4422(13)70124-8. CAN) study protocol: a cross-sectional, lifespan, multidisciplinary examination of Wardlaw, J.M., Makin, S.J., Valdés Hernández, M.C., Armitage, P.A., Heye, A.K., healthy cognitive ageing. BMC Neurol. 14, 1–25. https://doi.org/10.1186/s12883- Chappell, F.M., Muñoz-Maniega, S., Sakka, E., Shuler, K., Dennis, M.S., 014-0204-1. Thrippleton, M.J., 2017. Blood-brain barrier failure as a core mechanism in cerebral Shen, E.H., Overly, C.C., Jones, A.R., 2012. The Allen Human Brain Atlas. Trends small vessel disease and dementia: evidence from a cohort study. Alzheimers Neurosci. 35, 711–714. https://doi.org/10.1016/j.tins.2012.09.005. Dement. 13, 634–643. https://doi.org/10.1016/j.jalz.2016.09.006. Shi, Y., Thrippleton, M.J., Makin, S.D., Marshall, I., Geerlings, M.I., De Craen, A.J.M., Wardlaw, J.M., Smith, C., Dichgans, M., 2019. Small vessel disease: mechanisms and Van Buchem, M.A., Wardlaw, J.M., 2016. Cerebral blood flow in small vessel clinical implications. Lancet Neurol. 18, 684–696. https://doi.org/10.1016/S1474- disease: a systematic review and meta-analysis. J. Cereb. Blood Flow Metab. 36, 4422(19)30079-1. 1653–1667. https://doi.org/10.1177/0271678X16662891. Xin, Q., Ortiz-Terán, L., Diez, I., Perez, D.L., Ginsburg, J., El Fakhri, G., Sepulcre, J., Shi, Y., Thrippleton, M.J., Blair, G.W., Dickie, D.A., Marshall, I., Hamilton, I., Doubal, F. 2018. Sequence alterations of cortical genes linked to individual connectivity of the N., Chappell, F., Wardlaw, J.M., 2020. Small vessel disease is associated with altered human brain. Cereb. Cortex 29, 1–8. https://doi.org/10.1093/cercor/bhy262. cerebrovascular pulsatility but not resting cerebral blood flow. J. Cereb. Blood Flow Xu, H., Stamova, B., Jickling, G., Tian, Y., Zhan, X., Ander, B.P., Liu, D., Turner, R., Metab. 40, 85–99. https://doi.org/10.1177/0271678X18803956. Rosand, J., Goldstein, L.B., Furie, K.L., Verro, P., Johnston, S.C., Sharp, F.R., Smith, S.M., Jenkinson, M., Woolrich, M.W., Beckmann, C.F., Behrens, T.E.J., Johansen- Decarli, C.S., 2010. Distinctive RNA expression profiles in blood associated with Berg, H., Bannister, P.R., De Luca, M., Drobnjak, I., Flitney, D.E., Niazy, R.K., white matter hyperintensities in brain. Stroke 41, 2744–2749. https://doi.org/ Saunders, J., Vickers, J., Zhang, Y., De Stefano, N., Brady, J.M., Matthews, P.M., 10.1161/STROKEAHA.110.591875. 10