The Effects of an APOE Promoter Polymorphism on Human White Matter Connectivity during Non-Demented Aging

Abstract

Polymorphisms of the apolipoprotein E (APOE) promoter rs405509 are related to Alzheimer’s disease (AD). The T/T allele of rs405509 decreases the transcription of the APOE gene and leads to impairments in a specific brain structural network in aged individuals; thus, it is an important risk factor for AD. However, it remains unknown whether rs405509 affects white matter networks during aging. Here, we investigated the effect of the rs405509 genotype (T/T versus G-allele) on age-related brain white matter structural networks via construction of the graph theory-based structural connectome using diffusion MRI data in a large cohort. Network communication efficiency was quantified, along with the network’s betweenness centrality (Bc), global efficiency, local efficiency, and shortest path length. Regarding cognition, TT carriers had significant negative correlations between age and memory performance and between age and executive functions. A network analysis showed that TT carriers had an accelerated age-related loss of Bc and that regional Bc decreased in the left inferior frontal gyrus pars opercularis, the left posterior cingulate cortex, the right inferior occipital gyrus (IOG.R), and the left angular gyrus (ANG.L). Additional brain-behavior relationship analyses showed that polymorphism of rs405509 and age have strong interaction effects on the association of nodal Bc and cognition, mainly in the IOG.R and ANG.L. These results demonstrate that the rs405509 T/T allele of APOE causes an age-related cognitive decline in non-demented elderly people, possibly by modulating brain network communication efficiency, which may be beneficial for understanding the neural mechanisms of rs405509-related cognitive aging and AD pathogenesis.

Keywords

APOE promoter brain connectome cognitive aging diffusion MRI graph theory

INTRODUCTION

Cognitive aging is a complex process influenced by multiple factors including genes. The ɛ4 variant of the apolipoprotein E (APOE) gene has been well established as a risk factor for accelerated cognitive decline and Alzheimer’s disease (AD) [1, 2]. Researchers have systematically documented the role of the ɛ4 variant in cognitive decline [3 –5]. Additionally, accelerated cognitive aging in APOE ɛ4 carriers is accompanied by structural and functional brain changes. Compared with non-carriers, the cognitive normal ɛ4 carriers showed regional brain changes, such as whole brain atrophy [6, 7], reduced metabolism of glucose [8], abnormal functional connectivities in the default mode network (DMN) [9], and decreased white matter (WM) integrity [10 –13]. These patterns of brain structural and functional alterations were similar to that in mild cognitive impairment (MCI) and AD patients [14 –16].

Recently, another gene locus, the rs405509 single-nucleotide polymorphism (SNP) located at the APOE promoter region, was found to be related to cognitive decline and increased risk of AD. As reported by one case-control study [17], homozygous rs405509 T/T significantly increases the risk for developing AD. Interestingly, previous studies also reported the interaction of rs405509 with APOE in terms of AD risk. Its TT genotype was associated with a significantly increased risk of dementia in people with the APOE ɛ3/ɛ3 genotype. Moreover, we recently found that rs405509 TT carriers have an accelerated age-related reduction in the thickness of the left parahippocampal gyrus [18].

In addition to distributed regional abnormalities, researchers have also found an essential role of the rs405509 SNP in the disconnection between brain regions during aging, AD, and MCI [19 –21]. Previously, we found that MCI patients showed both structural and functional connectivity deficits [22, 23]. In AD patients, there is a reduced fidelity of the communication among brain regions [24], eventually causing neuronal death and WM degeneration [25]. APOE ɛ4 carriers exhibited a significant age-related decrease in WM network connectivity [26]. Moreover, disconnection in WM networks has negative cognitive consequences [20]. The ApoE protein functions as the principal cholesterol transporter in the brain and affects diverse cellular processes including development, plasticity, and nerve fiber repair [27]. Considering the role of rs405509 as a promoter that modulates the expression of APOE, the T to G substitution at rs405509 provoked an increase of 169% in promoter activity, rs405509 may lead to abnormal cognitive aging and cognitive impairments possibly via a mechanism similar to APOE, i.e., through the lipid cholesterol-Aβ-WM network pathway. However, no study thus far has examined the potential links between rs405509 and the WM network.

In the present study, we aimed to investigate whether the TT allele of rs405509 is associated with WM matter injury, thus leading to specific cognitive decline in aging. A battery of cognitive tasks was performed to evaluate their cognitive functions. Diffusion tensor imaging (DTI)-based structural network analysis was used to examine the brain WM network. We compared the differences between TT allele carriers and G allele carriers; we also investigated the association between network indices and neuropsychological performance. This study improves our understanding of how rs405509 polymorphisms induce cognitive decline from the perspective of the white matter network.

MATERIALS AND METHODS

Subjects

Participants were selected from the BABRI (Beijing Aging Brain Rejuvenation Initiative) study, an ongoing longitudinal study examining the brain and cognitive decline in elderly community-dwelling individuals aged above 50 years. All enrolled participants were Han Chinese and had received 6 or more years of education. Participants were qualified for our study if they met the following criteria: 1) no history of neurological or psychiatric disorders; 2) clinically non-demented when the MRI scan was performed; 3) a qualified high-resolution T1-weighted MRI image; 4) no visible abnormalities on the MR images, confirmed by an experienced radiologist; and 5) a successful blood sample for genotyping analysis. Specifically, the status “clinically non-demented” was determined by using the following: I) the Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV; II) Petersen’s dementia criteria; III) Clinical Dementia Rating (score = 0); and IV) a score of at least 24 on the Mini-Mental Status Examination (MMSE). Accordingly, 98 non-demented subjects (aged 52∼79 years; 41 males/57 females) were included in the present study. The protocol was approved by the institutional review board (IRB) at the Imaging Center for Brain Research at Beijing Normal University. Written consent was obtained from each subject.

Genotyping

The rs405509 polymorphism (also named Th1/E47cs) is located in the promoter region of the APOE gene. For each subject, we genotyped the rs405509 polymorphism using a TaqMan SNP genotyping assay on a 7900 HT Fast Real-Time PCR system (Applied Biosystems, Foster City, USA). We also genotyped two APOE polymorphisms, rs429358 and rs7412, which collectively form the APOE ɛ2 (with the rs429358-rs7412 haplotype of T-T), ɛ3 (T-C), and ɛ4 alleles (C-C). No other polymorphisms within the APOE gene were tested at this stage. The sample success rates for all three polymorphisms were 100%, and the reproducibility of the genotyping was 100% according to a duplication analysis of at least 10% of the genotypes. According to our sample, the rs405509 polymorphism did not show deviations from Hardy-Weinberg equilibrium (p > 0.5).

All subjects were divided into two groups according to their rs405509 genotype, which included 50 G-allele carriers (including 48 G/T and 2 G/G genotype carriers) and 48 T/T carriers separately. Given the relatively small number of the G/G rs405509 genotype, we combined the G/G and G/T genotypes into a single G-allele carrier group.

Cognitive tests

All participants received a battery of neuropsychological tests that assessed several classical cognitive domains. These tests were the following: 1) Auditory Verbal Learning Test (AVLT) Delayed Recall; 2) Clock Drawing Test; 3) Category Fluency Test; 4) Boston Naming Test (BNT); 5) Symbol Digit Modifying Test (SDMT); 6) Trail Making Test A; 7) Trail Making Test B; 8) Stroop Color and Word Test (SCWT) C; 9) Rey-Osterrieth Complex Figure Test (ROCF); and 10) MMSE. The specific neuropsychological test procedures have been previously described [23].

MRI data acquisition

All participants were scanned with a SIEMENS TRIO 3T scanner in the Imaging Center for Brain Research at Beijing Normal University for high-resolution T1-weighted structural MRI and DTI. T1-weighted, sagittal 3D magnetization prepared rapid gradient echo (MP-RAGE) sequences were acquired and covered the entire brain [176 sagittal slices, repetition time (TR) = 1900 ms, echo time (TE) = 3.44 ms, slice thickness = 1 mm, flip angle = 9°, inversion time = 900 ms, field of view (FOV) = 256×256 mm², acquisition matrix = 256×256]. For each DTI scan, images covering the whole brain were acquired by an echo-planar imaging sequence with the following scan parameters: TR = 9500 ms, TE = 92 ms, 30 diffusion-weighted directions with a b-value of 1000 s/mm², and a single image with a b-value of 0 s/mm², slice thickness = 2 mm, no inter-slice gap, 70 axial slices, acquisition matrix = 128×128, FOV = 256×256 mm², averages = 3.

Preprocessing

All of the image preprocessing and analyses described below were implemented using a pipeline tool (PANDA) for DTI data [28]. First, the skulls were stripped from the T1-weighted image and DTI for each participant. Each diffusion-weighted image was then coregistered to the b0 image using an affine transformation to correct the eddy-current induced distortions and simple head-motion artifacts. The diffusion gradient directions were adjusted accordingly [29].

Brain network construction

The brain network constructions for DTI data are based on a previously reported approach [22, 30] and detailed below. Nodes and edges are the two basic elements of a network. In this study, we defined all network nodes and edges using the procedures described below.

The nodes were defined in native space for each individual using the procedure proposed by Gong and colleagues [31]. Briefly, the skull-stripped T1-weighted image was non-linearly and spatially normalized to the Montreal Neurological Institute (MNI) space using FMRIB’s Linear Image Registration Tool. The individual FA images were coregistered to the individual skull-stripped T1-weighted images. To transform the AAL atlas from MNI space to DTI native space, the inverse transformations achieved in the above two steps were successively applied to the AAL atlas. Using this procedure, we obtained 90 nodes for the WM network. To validate our AAL-90 findings, we also used a high-resolution parcellation network with 1,024 regions of interest to investigate the WM networks, as suggested previously [32] (Supplementary Figure 1). Diffusion tensor tractography was implemented with DTI-studio software (H. Jiang, S. Mori, Johns Hopkins University) using the “fiber assignment by continuous tracking” method [33]. All of the tracts in the dataset were computed by seeding each voxel with an FA that was greater than 0.2. The tractography was terminated if it turned an angle greater than 45 degrees or reached a voxel with an FA of less than 0.2 [33]. For each subject, tens of thousands of streamlines were generated to etch out all of the major WM tracts. For the regional pair-wise connections (referred to as edge) in the network, two regions were considered structurally connected (with an edge) if at least three fibers streamlined with two end-points were located in these two regions [34]. The tractography results were visually inspected by the experts in neuroimaging, and no apparent errors in fiber tracking were found. At the same time, we did not find any significant differences in the tractography quality in all subjects. Specifically, we defined the fiber number (FN) as the weight of each edge. As a result, we constructed the FN-weighted WM network for each participant that was represented by a symmetric 90×90matrix.

Brain network analysis

Graph theoretical measures were used to characterize topological architectures of the DTI-based structural brain networks derived above. To characterize the topological organization of WM structural networks, several basic measures of global topology were examined, including betweenness centrality (Bc), global efficiency (Eg), local efficiency (Eloc), and the shortest path length (Lp) [35].

Eg is a global measure of the parallel information transfer ability in the whole network. It is computed as the average of the inverse of the “harmonic mean” of the characteristic path length: $E_{glob} = \frac{1}{N (N - 1)} \sum_{i \neq j \in G} \frac{1}{L_{ij}}$ where N is the number of nodes in the graph G, and L_ij is the characteristic path length between nodes i and j in graph. Local efficiency quantifies the ability of a network to tolerate faults that correspond to the efficiency of the information flow between the nearest neighbors of any given node.

The Eloc of a network is computed as follows: $E_{loc} = \frac{1}{N} \sum_{i \in G} E_{glob} (G_{i})$ whereG_i is the sub-graph composed of the nearest neighbors of node i and the connections among them. The path length between any pair of nodes (e.g., node i and node j) is defined as the sum of the edge lengths along this path. For weighted networks, the length of each edge was assigned by computing the reciprocal of the edge weight, 1/wij. The shortest path length, Lij, is defined as the length of the path for node i and node j with the shortest length.

The shortest path length of a network is computed as follows: $L_{p} = \frac{1}{N (N - 1)} \sum_{i \neq j \in G} L_{ij}$ where N is the number of nodes in the network. The Lp of a network quantifies the ability for information propagation in parallel.

For regional characteristics, we employed Bc on a nodal level, which was defined as the fraction of all of the shortest paths in the network that pass through a given node [36]. The Bc [36, 57] for each node in a network is defined as the ratio of the number of shortest paths that pass through a specified node to the total number of shortest paths in the network. The Bc of a node i is given as: $N_{bc} (i) = \sum_{i \neq j \neq k \in G} \frac{σ_{jk} (i)}{σ_{jk}}$ where σ_jk is the number of shortest paths between nodes j and k and σ_jk(i) is the number of these shortest paths that pass through node i.

All network analyses were performed using in-house GRETNA software (http://www.nitrc.org/projects/gretna/) and visualized using BrainNet Viewer software (http://www.nitrc.org/projects/bnv/). The detailed definitions of the graph metrics used in the present study are described in the Supplementary Materials.

Statistical analysis

Independent two-sample t-tests were used to assess between-group differences in age and education. The chi-square test was used to compare gender and APOE ɛ4 genotype difference. For neuropsychological measures, an analysis of covariance was used to test between-group differences (adjusted age, gender, education, and APOE ɛ4 genotype). All age-related analyses calculated Pearson’s partial correlation coefficient between age and the neuropsychological assessment and global/regional network metric of interest, separately for T/T and G-allele carriers, after controlling for the effects of gender, education, and APOE ɛ4 genotype.

We further assessed the effect of the rs405509 genotype× age interaction on regional Bc. Specifically, a GLM with “age,” “group,” and “group×age” as predictor variables was used, wherein gender, education, and APOE ɛ4 genotype were included as covariates. For the multiple nodal Bc, we applied the false discovery rate (FDR) procedure to correct the multiple comparisons at a q-value of 0.05.

Rs405509 genotype×age interaction effect on Bc-cognition correlation

For the significant rs405509 genotype×age interaction regions, we applied a GLM to examine the correlation between cognitions and nodal Bc with “group”, “nodal Bc”, “age”, “group×age”, “nodal Bc×group”, “nodal Bc×age”, and “group×age×nodal Bc” as dependent variables. Additionally, gender, education, and APOE ɛ4 genotype were included as covariates.

RESULTS

Cognitive performance

The rs405509 T/T genotype group did not differ significantly from the G-allele carriers in age, education, gender, or APOE ɛ4 genotype. In the T/T group, MMSE, AVLT-delay recall, and Tail Making Test (TMT)-B performances were significantly worse than G-allele carriers. After controlling for the effects of gender, education, and APOE ɛ4 genotype, T/T carriers had a significant age-related reduction in scores on the MMSE, ROCF-delay recall, SDMT, TMT-A, TMT-B, and SCWT-C. A trend of age-related decline in G-allele carriers was also observed in the MMSE, BNT, SDMT, and TMT-A tests (Table 1).

Age effects on global network measures

Including the education, gender, and APOE ɛ4 genotype as covariates, partial correlations were calculated between age and global measures for the T/T and GG/GT groups separately. T/T carriers exhibited significant negative partial correlations with age on the Eg and Eloc and positive correlations on Bc and Lp. For G-allele carriers, only Eg showed a significant age-related reduction (Fig. 1).

Rs405509 genotype×age interaction effect on nodal Bc

After FDR correction for multiple comparisons, there were four regions showing a significant group×age interaction effect, the left inferior frontal gyrus pars opercularis (IFGoper.L), the left posterior cingulate cortex (PCC.L), the right inferior occipital gyrus (IOG.R), and the left angular gyrus (ANG.L) (q < 0.05, FDR-corrected, Fig. 2). In addition, WM networks with AAL-1024 regions were also conducted. In accordance with the AAL-90 results, several frontal and parietal regions showed a significant rs405509 genotype×age interaction on WM nodal Bc (Supplementary Figure 2).

Rs405509 genotype×age interaction effect on Bc-cognition correlation

Further analysis illustrates the three-way group×age×nodal Bc interaction, representing the rs405509 genotype×age interaction effect on the nodal Bc-cognition correlation. There were two tests (AVLT-delay recall and TMT-B) that showed such a significant group×age×nodal Bc interaction. Importantly, the three-way interaction was mainly found in IOG.R and ANG.L. Because there is no simple way to plot a three-way interaction with two continuous variables (i.e., nodal Bc and age), we first divided all subjects into two age sub-groups, a low-age group <62 years of age (48 subjects in total, 52∼61 years) and a high-age group with≥62 years of age (50 subjects in total, 62∼79 years) and then plotted the three-way interaction accordingly (Fig. 3).

DISCUSSION

In the current study, our main findings are as follows: 1) with increasing age, TT carriers of rs405509 exhibited an accelerated cognitive decline in memory and executive function; 2) for DTI-based WM network parameters, TT carriers of rs405509 displayed faster decreases in Bc within several brain regions, including IFGoperc.L, PCC.L, IOG.R, and ANG.L; and 3) further brain-behavior relationship analysis showed that polymorphism of rs405509 and age have strong interaction effects on the association of nodal Bc and cognition, mainly in IOG.R and ANG.L. The interaction results suggested that the TT carriers of the low age group showed a more positive correlation between cognitive performance and nodal Bc than G carriers, and this correlation difference decreased in the high age group.

Although previous studies have documented that the TT allele of the rs405509 polymorphism is associated with a high risk for AD, we, for the first time, evaluated the effects of rs405509 on cognitive aging in a non-demented elderly population. Like the ɛ4 carriers, which is associated with worse cognitive performance in old age [1, 2] but with better cognitive functions in young and middle age [37 –39], the cognitive performances more decreased in old TT carriers. The TT genotype of rs405509 seems to contribute to age-related declines of cognitive functions and brain structure independently of APOE ɛ4.

To find the neural mechanisms for the accelerated age-related cognitive decline in the TT genotype, we collected DTI data and constructed a WM network. Recently, many researchers have adapted graph theory in the study of brain structure alterations in AD and MCI and have made several discoveries. These studies consistently show that both AD and MCI patients have reduced nodal efficiency [34, 40]. In another study, APOE-ɛ4 carriers exhibited an accelerated age-related loss of mean local interconnectivity and regional local interconnectivity decreases in the precuneus, the medial orbitofrontal cortex, and the lateral parietal cortex [20]. The present study, for the first time, found topological alterations of the WM network in the elderly based on another graph theory parameter, Bc. Bc is often used for characterizing the importance of anatomical or functional connections. Nodes with high betweenness play important roles in interacting with other regions and facilitating functional integration and promoting network vulnerability. Largely consistent with those findings applying other network definitions in AD and MCI patients [41, 42], TT carriers of the rs405509 polymorphism also display a reduction of Bc. Previous studies found other network measures like node degree and efficiency representing similar meaning with Bc showing impairments in AD and MCI patients, present findings of rs405509 implied similar mechanism with AD pathology.

In our study, the TT genotype was associated with greater age-related losses of properties of Bc in several brain regions including one frontal region (IFGoperc.L), one occipital region (IOG.R), and two regions in the parietal lobe (PCG.L and ANG.L). According to the theory of cognitive aging, the frontal lobes are regarded as especially vulnerable regions to normal age-changes [43 –45]. The WM networks’ disrupted pattern in TT carriers was consistent with the trajectory of brain aging.

The IFGoperc.L that exhibited a significant negative correlation with age in TT carriers is part of the fronto-striatal circuit [46, 47], which is responsible for executive functions. The circuit is proposed as a hallmark of normal brain aging more than AD [48]. Moreover, some regions (PCG.L and ANG.L) of the DMN also showed a negative correlation between age and local interconnectivity. The DMN is strongly associated with memory performance and is preferentially attacked by amyloid-beta in AD, and abnormal DMN has been observed in non-demented APOE ɛ4 carriers and in AD [49 –51]. Importantly, PCG is an essential hub region with widely distributed connections in the brain network [52]. Researchers have demonstrated that the damage in white matter in the posterior cingulum in MCI patients is associated with memory impairment [53]. Our study supports the notion that white matter lesions of the cingulum represent one of the earliest changes in the development of age-related dementia [54]. The ANG is thought to be associated with complex language functions, which can access both content and episodic memories [55]. Moreover, the angular gyrus is part of the DMN, which is active when the tested subjects are not performing specific cognitive tasks [55]. Our finding suggested that the rs405509 polymorphism may contribute to cognitive impairment by expanding damage along the DMN, mainly at PCG.L and ANG.L.

Moreover, we found significant genotype-age-brain interactions in the AVLT-IOG.R, AVLT-ANG.L, and TMTb-ANG.L. In these three relationships, the TT carriers of the low age group showed a more positive correlation between cognitive performance and brain network efficiency than G carriers, and these correlation differences were diminished or even reversed in the older age group. In other words, this result means that the rs405509 TT effect on cognitive impairment that is mediated through the brain network nodes was weak until the tested subjects reached an “onset age.” As mentioned above, like that of the APOE gene [37], this effect could be considered as antagonistic pleiotropy of the rs405509 polymorphism.

In the present study, we found a gene-WM network relationship in the cognitive aging process that is independent of the established APOE ɛ4 effect. These findings add to the understanding of the potential mechanism of how genes influence cognitive performances through brain structure, especially from a network prospective. In future studies, researchers could make efforts to discover more genes with effects on cognitive aging and build a more explicit gene-cognitive aging map. These studies could be beneficial for the early prevention and preclinical intervention of cognitive-related disorders.

In general, rs405509 TT shares the interesting features of anterograde amnesia, executive function dysfunction and WM network disconnection with APOE ɛ4 and AD. These data support a multiple etiology of age-related WM disconnection, of which the rs405509 polymorphism is one cause.

CONCLUSION

These findings may provide new insights into understanding how the rs405509 polymorphism contributes to cognitive decline and brain structural network and implies how the rs405509 polymorphism poses a high risk for AD.

Footnotes

ACKNOWLEDGMENTS

Financial support for this study was provided by State Key Program of National Natural Science Foundation of China 81430100; by National Natural Science Foundation of China 81173460, 30873458 and 31500922; by Beijing Municipal Science & Technology Commission.Z161100000216135.

Authors’ disclosures available online ().

Appendix

References

Wisdom

, Callahan

, Hawkins

(2011) The effects of apolipoprotein E on non-impaired cognitive functioning: A meta-analysis. Neurobiol Aging 32, 63–74.

Small

, Rosnick

, Fratiglioni

, Backman

(2004) Apolipoprotein E and cognitive performance: A meta-analysis. Psychol Aging 19, 592–600.

Kozauer

, Mielke

, Chan

, Rebok

, Lyketsos

(2008) Apolipoprotein E genotype and lifetime cognitive decline. Int Psychogeriatr 20, 109–123.

Prins

, Scheltens

(2015) White matter hyperintensities, cognitive impairment and dementia: An update. Nat Rev Neurol 11, 157–165.

Grady

, Furey

, Pietrini

, Horwitz

, Rapoport

(2001) Altered brain functional connectivity and impaired short-term memory in Alzheimer’s disease. Brain 124, 739–756.

Shaw

, Lerch

, Pruessner

, Taylor

, Rose

, Greenstein

, Clasen

, Evans

, Rapoport

, Giedd

(2007) Cortical morphology in children and adolescents with different apolipoprotein E gene polymorphisms: An observational study. Lancet Neurol 6, 494–500.

Chen

, Reiman

, Alexander

, Caselli

, Gerkin

, Bandy

, Domb

, Osborne

, Fox

, Crum

, Saunders

, Hardy

(2007) Correlations between apolipoprotein E ɛ4 gene dose and whole brain atrophy rates. Am J Psychiatry 164, 916–921.

Reiman

, Chen

, Alexander

, Caselli

, Bandy

, Osborne

, Saunders

, Hardy

(2005) Correlations between apolipoprotein E epsilon4 gene dose and brain-imaging measurements of regional hypometabolism. Proc Natl Acad Sci U S A 102, 8299–9302.

Machulda

, Jones

, Vemuri

, McDade

, Avula

, Przybelski

, Boeve

, Knopman

, Petersen

, Jack

Jr (2011) Effect of APOE epsilon4 status on intrinsic network connectivity in cognitively normal elderly subjects. Arch Neurol 68, 1131–1136.

10.

Honea

, Vidoni

, Harsha

, Burns

(2009) Impact of APOE on the healthy aging brain: A voxel-based MRI and DTI study. J Alzheimers Dis 18, 553–564.

11.

Ryan

, Walther

, Bendlin

, Lue

, Walker

, Glisky

(2011) Age-related differences in white matter integrity and cognitive function are related to APOE status. Neuroimage 54, 1565–1577.

12.

Gold

, Powell

, Andersen

, Smith

(2010) Alterations in multiple measures of white matter integrity in normal women at high risk for Alzheimer’s disease. Neuroimage 52, 1487–1494.

13.

Smith

, Chebrolu

, Andersen

, Powell

, Lovell

, Xiong

, Gold

(2010) White matter diffusion alterations in normal women at risk of Alzheimer’s disease. Neurobiol Aging 31, 1122–1131.

14.

Jones

, Machulda

, Vemuri

, McDade

, Zeng

, Senjem

, Gunter

, Przybelski

, Avula

, Knopman

, Boeve

, Petersen

, Jack

Jr (2011) Age-related changes in the default mode network are more advanced in Alzheimer disease. Neurology 77, 1524–1531.

15.

Nestor

, Fryer

, Smielewski

, Hodges

(2003) Limbic hypometabolism in Alzheimer’s disease and mild cognitive impairment. Ann Neurol 54, 343–351.

16.

Migliaccio

, Agosta

, Possin

, Rabinovici

, Miller

, Gorno-Tempini

(2012) White matter atrophy in Alzheimer’s disease variants. Alzheimers Dement 8, S78–87 e71-72.

17.

Lambert

, Araria-Goumidi

, Myllykangas

, Ellis

, Wang

, Bullido

, Harris

, Artiga

, Hernandez

, Kwon

, Frigard

, Petersen

, Cumming

, Pasquier

, Sastre

, Tienari

, Frank

, Sulkava

, Morris

, St Clair

, Mann

, Wavrant-DeVrieze

, Ezquerra-Trabalon

, Amouyel

, Hardy

, Haltia

, Valdivieso

, Goate

, Perez-Tur

, Lendon

, Chartier-Harlin

(2002) Contribution of APOE promoter polymorphisms to Alzheimer’s disease risk. Neurology 59, 59–66.

18.

Liang

, Li

, Lv

, Shu

, Chen

, Li

, Zhang

, Hu

, Zhang

(2015) Sex moderates the effects of the Sorl1 gene rs2070045 polymorphism on cognitive impairment and disruption of the cingulum integrity in healthy elderly. Neuropsychopharmacology 40, 1519–1527.

19.

Bondi

, Salmon

, Monsch

, Galasko

, Butters

, Klauber

, Thal

, Saitoh

(1995) Episodic memory changes are associated with the APOE-epsilon 4 allele in nondemented older adults. Neurology 45, 2203–2206.

20.

Brown

, Terashima

, Burggren

, Ercoli

, Miller

, Small

, Bookheimer

(2011) Brain network local interconnectivity loss in aging APOE-4 allele carriers. Proc Natl Acad Sci U S A 108, 20760–20765.

21.

Putcha

, Brickhouse

, O’Keefe

, Sullivan

, Rentz

, Marshall

, Dickerson

, Sperling

(2011) Hippocampal hyperactivation associated with cortical thinning in Alzheimer’s disease signature regions in non-demented elderly adults. J Neurosci 31, 17680–17688.

22.

Shu

, Liang

, Li

, Zhang

, Li

, Wang

, He

, Wang

, Zhang

(2012) Disrupted topological organization in white matter structural networks in amnestic mild cognitive impairment: Relationship to subtype. Radiology 265, 518–527.

23.

Wang

, Li

, Liang

, Zhang

, Li

, Shu

, Wang

, Zhang

(2013) Amnestic mild cognitive impairment: Topological reorganization of the default-mode network. Radiology 268, 501–514.

24.

Greicius

, Srivastava

, Reiss

, Menon

(2004) Default-mode network activity distinguishes Alzheimer’s disease from healthy aging: Evidence from functional MRI. Proc Natl Acad Sci U S A 101, 4637–4642.

25.

Westlye

, Reinvang

, Rootwelt

, Espeseth

(2012) Effects of APOE on brain white matter microstructure in healthy adults. Neurology 79, 1961–1969.

26.

Han

, Bondi

(2008) Revision of the apolipoprotein E compensatory mechanism recruitment hypothesis. Alzheimers Dement 4, 251–254.

27.

Mahley

(1988) Apolipoprotein E: Cholesterol transport protein with expanding role in cell biology. Science 240, 622–630.

28.

Cui

, Zhong

, Xu

, He

, Gong

(2013) PANDA: A pipeline toolbox for analyzing brain diffusion images. Front Hum Neurosci 7, 42.

29.

Leemans

, Jones

(2009) The B-matrix must be rotated when correcting for subject motion in DTI data. Magn Reson Med 61, 1336–1349.

30.

Zhang

, Wang

, Wu

, Kuang

, Huang

, He

, Gong

(2011) Disrupted brain connectivity networks in drug-naive, first-episode major depressive disorder. Biol Psychiatry 70, 334–342.

31.

Gong

, He

, Concha

, Lebel

, Gross

, Evans

, Beaulieu

(2009) Mapping anatomical connectivity patterns of human cerebral cortex using in vivo diffusion tensor imaging tractography. Cereb Cortex 19, 524–536.

32.

Zhang

, Liao

, Chen

, Mantini

, Ding

, Xu

, Wang

, Yuan

, Chen

, Jiao

, Lu

(2011) Altered functional-structural coupling of large-scale brain networks in idiopathic generalized epilepsy. Brain 134, 2912–2928.

33.

Mori

, Crain

, Chacko

, van Zijl

(1999) Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging. Ann Neurol 45, 265–269.

34.

Shu

, Liu

, Li

, Duan

, Wang

, Yu

, Dong

, Ye

, He

(2011) Diffusion tensor tractography reveals disrupted topological efficiency in white matter structural networks in multiple sclerosis. Cereb Cortex 21, 2565–2577.

35.

Rubinov

, Sporns

(2010) Complex network measures of brain connectivity: Uses and interpretations. Neuroimage 52, 1059–1069.

36.

Wang

, Hernandez

, Van Mieghem

(2008) Betweenness centrality in a weighted network. Phys Rev E Stat Nonlin Soft Matter Phys 77, 046105.

37.

Alexander

, Williams

, Gatt

, Dobson-Stone

, Kuan

, Todd

, Schofield

, Cooper

, Gordon

(2007) The contribution of apolipoprotein E alleles on cognitive performance and dynamic neural activity over six decades. Biol Psychol 75, 229–238.

38.

Schultz

, Lyons

, Franz

, Grant

, Boake

, Jacobson

, Xian

, Schellenberg

, Eisen

, Kremen

(2008) Apolipoprotein E genotype and memory in the sixth decade of life. Neurology 70, 1771–1777.

39.

Marchant

, King

, Tabet

, Rusted

(2010) Positive effects of cholinergic stimulation favor young APOE epsilon4 carriers. Neuropsychopharmacology 35, 1090–1096.

40.

Liu

, Yu

, Zhang

, Liu

, Duan

, Alexander-Bloch

, Liu

, Jiang

, Bullmore

(2014) Impaired long distance functional connectivity and weighted network architecture in Alzheimer’s disease. Cereb Cortex 24, 1422–1435.

41.

Yao

, Zhang

, Lin

, Zhou

, Xu

, Jiang

(2010) Abnormal cortical networks in mild cognitive impairment and Alzheimer’s disease. PLoS Comput Biol 6, e1001006.

42.

Tijms

, Wink

, de Haan

, van der Flier

, Stam

, Scheltens

, Barkhof

(2013) Alzheimer’s disease: Connecting findings from graph theoretical studies of brain networks. Neurobiol Aging 34, 2023–2036.

43.

West

(1996) An application of prefrontal cortex function theory to cognitive aging. Psychol Bull 120, 272–292.

44.

Robbins

, James

, Owen

, Sahakian

, Lawrence

, McInnes

, Rabbitt

(1998) A study of performance on tests from the CANTAB battery sensitive to frontal lobe dysfunction in a large sample of normal volunteers: Implications for theories of executive functioning and cognitive aging. Cambridge Neuropsychological Test Automated Battery. J Int Neuropsychol Soc 4, 474–490.

45.

Rabbitt

(2005) Frontal brain changes and cognitive performance in old age. Cortex 41, 238–240.

46.

Jagust

(2013) Vulnerable neural systems and the borderland of brain aging and neurodegeneration. Neuron 77, 219–234.

47.

Fjell

, McEvoy

, Holland

, Dale

, Walhovd

(2014) What is normal in normal aging? Effects of aging, amyloid and Alzheimer’s disease on the cerebral cortex and the hippocampus. Prog Neurobiol 117, 20–40.

48.

Buckner

(2004) Memory and executive function in aging and AD: Multiple factors that cause decline and reserve factors that compensate. Neuron 44, 195–208.

49.

Hafkemeijer

, van der Grond

, Rombouts

(2012) Imaging the default mode network in aging and dementia. Biochim Biophys Acta 1822, 431–441.

50.

Wang

, Roe

, Snyder

, Brier

, Thomas

, Xiong

, Benzinger

, Morris

, Ances

(2012) Alzheimer disease family history impacts resting state functional connectivity. Ann Neurol 72, 571–577.

51.

Petrella

, Sheldon

, Prince

, Calhoun

, Doraiswamy

(2011) Default mode network connectivity in stable vs progressive mild cognitive impairment. Neurology 76, 511–517.

52.

Leech

, Braga

, Sharp

(2012) Echoes of the brain within the posterior cingulate cortex. J Neurosci 32, 215–222.

53.

Choo

, Lee

, Oh

, Lee

, Song

, Youn

, Kim

, Jhoo

, Woo

(2010) Posterior cingulate cortex atrophy and regional cingulum disruption in mild cognitive impairment and Alzheimer’s disease. Neurobiol Aging 31, 772–779.

54.

Delano-Wood

, Stricker

, Sorg

, Nation

, Jak

, Woods

, Libon

, Delis

, Frank

, Bondi

(2012) Posterior cingulum white matter disruption and its associations with verbal memory and stroke risk in mild cognitive impairment. J Alzheimers Dis 29, 589–603.

55.

Seghier

(2013) The angular gyrus: Multiple functions and multiple subdivisions. Neuroscientist 19, 43–61.

56.

Latora

, Marchiori

(2001) Efficient behavior of small-world networks. Phys Rev Lett 87, 198701.

57.

Goh

, Oh

, Kahng

, Kim

(2003) Betweenness centrality correlation in social networks. Phys Rev E Stat Nonlin Soft Matter Phys 67, 017101.