Tractography of the external capsule and cognition: A diffusion MRI study of cholinergic fibers

Experimental Gerontology 130 (2020) 110792

Contents lists available at ScienceDirect

Experimental Gerontology journal homepage: www.elsevier.com/locate/expgero

Tractography of the external capsule and cognition: A diffusion MRI study of cholinergic fibers

T

Geneviève Nolze-Charrona,b, Raphaël Dufort-Rouleaua, Jean-Christophe Houdec, Matthieu Dumontc, Christian-Alexandre Castellanoa,d, Stephen Cunnanea,d, Dominique Lorraind, ⁎ Tamàs Fülöpa,d, Maxime Descoteauxc, Christian Boctia,d,e, a

Faculty of Medicine and Health Sciences, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, Quebec J1H 5N4, Canada Division of Neurology, Department of Medicine, Hôpital de Rouyn-Noranda – CISSS de l'Abitibi-Témiscamingue, 4, 9e Rue, Rouyn-Noranda, Quebec J9X 2B2, Canada c Sherbrooke Connectivity Imaging Lab (SCIL), Computer Science Department, Université de Sherbrooke, 2500 Boulevard de l'Université, Sherbrooke, Quebec J1K 0A5, Canada d Research Centre on Aging, CIUSSS de l'Estrie-CHUS, 1036 rue Belvédère Sud, Sherbrooke, Quebec J1H 4C4, Canada e Division of Neurology, Department of Medicine, CIUSSS de l'Estrie-Centre Hospitalier Universitaire de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, Quebec J1H 5N4, Canada b

ARTICLE INFO

ABSTRACT

Section Editor: Thomas Foster

Introduction: White matter changes (WMC) in the cholinergic tracts contribute to executive dysfunction in the context of cognitive aging. WMC in the external capsule have been associated with executive dysfunction. The objectives of this study were to: 1) Characterize the lateral cholinergic tracts (LCT) and the superior longitudinal fasciculus (SLF). 2) Evaluate the association between diffusion measures within those tracts and cognitive performance. Methods: Neuropsychological testing and high angular resolution diffusion imaging (HARDI) of 34 healthy elderly participants was done, followed by anatomically constrained probabilistic tractography reconstruction robust to crossing fibers. The external capsule was manually segmented on a mean T1 image then merged with an atlas, allowing extraction of the LCT. Diffusion tensor imaging (DTI) and HARDI-based measures were obtained. Results: Correlations between diffusion measures in the LCT and the time of completion of Stroop (left LCT radial and medial diffusivity), the Symbol Search score (right LCT apparent fiber density) and the motor part of Trail-B (left LCT axial and radial diffusivity) were observed. Correlations were also found with diffusion measures in the SLF. WMC burden was low, and no correlation was found with diffusion measures or cognitive performance. Discussion: DTI and HARDI, with isolation of strategic white matter tracts for cognitive functions, represent complimentary tools to better understand the complex process of brain aging.

Keywords: Cholinergic tract Superior longitudinal fasciculus Tractography HARDI Cognition

1. Introduction Alzheimer's disease and vascular dementia are the most common

etiologies of cognitive decline in the aging population. White matter changes of the brain have been associated with both these conditions (Brun and Englund, 1986; Iadecola, 2013). Nearly ubiquitous cerebral

Abbreviations: AD, axial diffusivity; AFD, Fiber Orientation Distribution Function Apparent Fiber Density; BET, Brain Extraction Tool; dMRI, diffusion Magnetic Resonance Imaging; DTI, diffusion tensor imaging; FA, fractional anisotropy; GFA, generalized FA; GMWMI, gray matter/white matter interface; HARDI, High Angular Resolution Diffusion Imaging; LCT, lateral cholinergic tracts; MCI, mild cognitive impairment; MD, medial diffusivity; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; PFT, Particle Filter Tractography; RD, radial diffusivity; ROI, region of interest; SLF, superior longitudinal fasciculus; TMT-B, Trail-Making-test part B; WM, white matter; WMH, white matter hyperintensities ⁎ Corresponding author at: Division of Neurology, Faculty of Medicine and Health Sciences, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, Quebec J1H 5N4, Canada. E-mail addresses: [email protected] (G. Nolze-Charron), [email protected] (R. Dufort-Rouleau), [email protected] (J.-C. Houde), [email protected] (M. Dumont), [email protected] (C.-A. Castellano), [email protected] (S. Cunnane), [email protected] (D. Lorrain), [email protected] (T. Fülöp), [email protected] (M. Descoteaux), [email protected] (C. Bocti). https://doi.org/10.1016/j.exger.2019.110792 Received 14 June 2019; Received in revised form 23 October 2019; Accepted 21 November 2019 Available online 25 November 2019 0531-5565/ © 2019 Elsevier Inc. All rights reserved.

Experimental Gerontology 130 (2020) 110792

G. Nolze-Charron, et al.

white matter changes (WMC) mostly reflect damage to small blood vessels (Longstreth et al., 1998) and have been correlated to slower information processing speed (Prins et al., 2005), impaired memory (Libon et al., 2008), cognitive decline (Inzitari et al., 2007; Boyle et al., 2016), loss of functional autonomy, and increased risk of death (Debette and Markus, 2010). Cognitive changes associated with normal aging could be related to these WMC and alterations in white matter integrity (Rizio and Diaz, 2016; Head, 2004; Sevigny-Dupont, 2019 ). Thus, a better understanding of the associations between white matter structure and cognition during aging is important. Diffusion Magnetic Resonance Imaging (dMRI) allows for in vivo reconstruction of fiber tract pathways within the brain via the estimation of the principal diffusion orientations of water molecules (Basser et al., 1994). The direction of diffusion of water molecules in white matter is governed by the physical constraints created by the parallel orientation and structural properties of axons, including their myelin sheath, membranes and walls (Concha, 2014). DMRI tractography algorithms permit the reconstruction of such fiber bundles by integrating the diffusion directions of adjacent voxels (Mori and van Zijl, 2002). Diffusion measures such as fractional anisotropy (FA), and axial (AD), mean (MD) and radial diffusivity (RD) give an indication as to underlying directionality and microstructure of white matter bundles. Possible alterations in their structure typically decrease FA and increase diffusivity measures (Salat, 2014). An inherent problem of diffusion modeling is the management of crossing fibers. This is a fairly important problem, as up to 90% of brain voxels contain crossing fibers (Jeurissen et al., 2012). This problem can however be partly overcome by using High Angular Resolution Diffusion Imaging (HARDI), which captures more directions within every voxel (Descoteaux, 2015). HARDI thus permits the measurement of novel parameters such as generalized FA (GFA) and Fiber Orientation Distribution Function Apparent Fiber Density (fODF-AFD or AFD) (Raffelt et al., 2012). HARDI models also allow tractography algorithms to better recover complex white matter bundles and better deal with regions of partial volume effects (Girard et al., 2014). The relative potential contribution of DTI and HARDI varies according to the structural complexity of the white matter tracts under study (Caiazzo et al., 2016). Tractography can also help establish whether there is an association between the average measures of white matter fiber bundles and cognition. For instance, reduced fiber length was associated with decreased speed of information processing and executive function in healthy aging subjects (Behrman-Lay et al., 2015). In MCI (Zhuang et al., 2010; Wang et al., 2012) and Alzheimer's disease (Bosch et al., 2012; Fieremans et al., 2013; Lee et al., 2015), changes in FA and diffusion measures were found in multiple white matter bundles (corpus callosum, inferior fronto-occipital fasciculus, superior and inferior longitudinal fasciculus, among others.) The most important fiber bundles for cognition and the most useful measures to estimate white matter diffusion are not yet clearly identified. It is also of importance to note that DTI and HARDI are complementary investigational tools given their relative strengths for different anatomical tracts of varying spatial complexity (Behrens et al., 2007; Caiazzo et al., 2016). Acetylcholine is a neurotransmitter that plays an important role in memory and attention (Hasselmo, 2006; Klinkenberg et al., 2011). Studies have reported a decrease in cholinergic markers of the cortex in Alzheimer's disease patients (McGeer et al., 1984). The trajectory of the cholinergic pathways originating from the basal nucleus of Meynert in the substantia innominata was described by dissection and immunohistochemistry (Mesulam and Geula, 1988; Selden et al., 1998). This white matter bundle divides into the medial and lateral cholinergic tracts (LCT), the latter coursing through the external capsule. Cholinergic fibers seem more densely packed in the inferior portion. Cognitive decline is associated with a loss of cholinergic neurons of the substantia innominata (Cullen and Halliday, 1998; Geula et al., 2008; Gao et al., 2013) and a loss of thickness of this same region on MRI

imaging (Hanyu et al., 2007). Furthermore, white matter hyperintensities in the presumed cholinergic bundles graded by a visual scale on T2 MRI have been associated with lower cognitive performance; these effects could be due to the co-localization of other critical tracts that are part of large-scale neuronal networks that subserve cognitive functions (Bocti et al., 2005; Behl et al., 2007). The superior longitudinal fasciculus (SLF) is a bidirectional association fiber tract, connecting anterior (frontal, opercular) to posterior (temporal, parietal, occipital) parts of each hemisphere, thereby also linking the expressive to the receptive areas of language. Most of its fibers overlie the superior part of the external capsule (Schmahmann and Pandya, 2006), coursing through the corona radiata and thus potentially crossing the projecting cholinergic fibers. Its most important function is in language (verbal repetition, speech fluency, phonological and complex syntactic processing in speech production and comprehension) (Dick et al., 2014), but it also seems to play a role in global cognition, attention, and executive functions (Schmahmann et al., 2008). Cholinergic bundles have seldom been characterized using tractography, and most studies attempting to do so have either used only the external capsule as a global region of interest (ROI) (Sun et al., 2014), or have isolated the medial cholinergic tract (Hong and Jang, 2010; Hong et al., 2012). One previous study isolated both the medial and LCT in a group of non-demented subjects with vascular cognitive impairment and compared them to healthy elderly subjects; lower FA in the cholinergic bundles was reported in the vascular cognitive impairment group, as well as associations with neuropsychological testing performance (Liu et al., 2017). The objectives of this study were to: 1) Use HARDI tractography to characterize white matter bundles from the external capsule, comprising presumed lateral cholinergic fibers and also the superior longitudinal fasciculus within the overlying corona radiata. 2) Evaluate the association between the obtained diffusion measures (FA, AD, MD, RD, GFA, AFD) within these bundles and neuropsychologic testing results in a sample of cognitively healthy older adults. 2. Material and methods 2.1. Study population Participant aged between 65 and 85 years old were recruited from the community based on having a normal MMSE (Mini-Mental State Examination; Folstein et al., 1975) score, of at least 26/30. Participants who smoked, had an excessive consumption of alcohol, had a history of uncontrolled cerebrovascular risk factors, neurological disease, psychiatric illness, or took psychoactive medication were excluded from the study. Baseline data were collected by an initial interview. Every participant gave their informed written consent. This study was approved by the ethics committee of the Centre Hospitalier de l'Université de Sherbrooke-CHUS and the Research Centre on Aging. 2.2. Neuropsychological evaluation All subjects underwent a detailed neuropsychological battery including, in addition to the MMSE, the Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005) as a measure of general cognition. Visual analysis and working memory were assessed with the TrailMaking-test-B (TMT-B; Partington and Leiter, 1949). Information-processing speed was assessed with Coding and Symbol search from WAISIV (Wechsler, 2008), and visuospatial memory span with WMS-III (Wechsler, 1997). Executive functions, inhibition and cognitive flexibility were assessed with Stroop - D-KEFS Color-Word Interference Test (Delis et al., 2001). 2

Experimental Gerontology 130 (2020) 110792

G. Nolze-Charron, et al.

2.3. Imaging data acquisition MR images were acquired using a 1.5-Tesla Siemens scanner (Sonata, Siemens Medical Solutions, Erlangen, Germany). For every participant, three-dimensional T1-weighted MR images were acquired. Parameters for the gradient echo sequence were as follows: repetition time/echo time - 16.00/4.68 ms, 20° flip angle, 1 mm3 isotropic reconstructed voxel size, 256 × 240 × 192 mm3 field of view, matrix size of 256 × 256 × 164, number of averages of 1, acquisition time of 9.14 min. A set of 20 axial fluid attenuated inversion recovery (FLAIR) images were acquired, with parameters: repetition time/echo time 8500/91 ms, 2400 ms inversion time, echo train length of 17, matrix size of 256 × 192, for a 230 × 172.5 mm2 field of view, slice thickness of 6 mm, spacing between slices of 1.2 mm, number of acquisitions of 1, acquisition time of 3.09 min. FLAIR images were used for assessment of white matter hyperintensities on visual scales: 1 - general WMH (using the Age-Related White Matter Changes – ARWMC scale; Wahlund et al., 2001) and 2 - WMH in presumed cholinergic tracts (using the Cholinergic Pathways Hyperintensities - CHIPS scale; Bocti et al., 2005). Sixty-four diffusion-weighted images were acquired along 64 uniformly distributed directions using a b-value 1000 s/mm2 and a single b = 0 s/ mm2 image using the single-shot echo-planar imaging (EPI) sequence (128 × 128 matrix, 2 mm isotropic resolution, TR/TE 11000/98 ms and GRAPPA factor 2).

Fig. 1. Manual segmentation of the external capsules and peri-insular white matter using the MI-Brain Software.

Freesurfer atlas (Reuter et al., 2012). Streamlines passing through the external capsule were filtered using the White Matter Query Language (Wassermann et al., 2013) as fibers passing through irrelevant regions were excluded (see Appendix A). The resulting presumed lateral cholinergic pathways are presented in Fig. 2. Another white matter tract that overlies the ROI and plays a putative role in cognitive functions, the SLF, was also extracted using the Freesurfer atlas and the White Matter Query Language (Fig. 3). A population-specific template was created using ANTs Multivariate Template Construction due to the major structural changes appearing with age. T1w images and the fractional anisotropy (FA) maps from the pre-supplementation acquisitions (n = 23) were used as the input. The resulting average T1w template is seen in the background of the leftright group-average SLFs and LCTs in the template space (Figs. 2, 3). To produce these, each track-file (SLF right, SLF left, LCT right and LCT left) of each subject was warped non-linearly to the new template space and group-averaged.

2.4. DWI processing and tractography Diffusion images were denoised using a non-local means denoising technique (Coupé et al., 2008). Eddy currents and motion were then corrected using the FSL eddy command (Andersson and Sotiropoulos, 2016). The corrected diffusion images were then upsampled to a 1 mm3 resolution. The Brain Extraction Tool (BET; Smith, 2002) from FSL was used on the b = 0 image to extract the brain mask and to restrict further processing to relevant zones of the brain. Diffusion tensors and their derived measures (FA, AD, MD, RD) were computed using the Diffusion Imaging in PYthon (DIPY) software library (Garyfallidis et al., 2014). The fiber ODF model was computed using Spherical Constrained Deconvolution (Tournier et al., 2007; Descoteaux et al., 2009) with maximal spherical harmonics order 8, as implemented in DIPY. The GFA and AFD were also computed using DIPY. BET was also run on the T1-weighted image to extract a mask of the brain. The T1-weighted image was registered to the upsampled DWI using Advanced Normalization Tools affine and non-linear registration (Avants et al., 2011). Probabilistic tissues maps were extracted from the T1-weighted image using FMRIB's Automated Segmentation Tool (Zhang et al., 2001) and were then used as seeding and constraint masks for the Particle Filter Tractography (PFT; Girard et al., 2014) algorithm. Tractography used the probabilistic version of the PFT algorithm, using 10 seeds per voxel of the white matter (WM) and gray matter/white matter interface (GMWMI). Other tracking parameters: minimum/maximum length: 10/ 200 mm, no propagation of 0,2 mm, minimum radius of curvature of 0.575877 mm, and amplitude threshold of fODF of 0,1. This generated streamlines files in the order of 5,000,000 streamlines for each subject. Streamlines were compressed using the Fiber Compression algorithm (Presseau et al., 2015), using an error threshold of 0.2 mm.

2.6. Statistical analysis AFD, AD, RD, MD and FA were obtained on cholinergic bundles isolated by the process described above, and on superior longitudinal fasciculus (SLF) bundles. Descriptive statistics (mean, standard deviation) were used to describe all variables. Demographic, DTI (FA, AD, RD, MD) and HARDI (GFA, AFD) measures were correlated with results of neuropsychological testing using Pearson's simple correlations after normality was confirmed. The alpha was 0,05. Analyses were performed using SPSS Statistics 20. 3. Results 3.1. Demographics Patient characteristics for the 34 healthy elderly subjects included in this study are presented in Table 1. MMSE was near maximal (29 on average) and MoCA was at 27. Semi-quantitative scale scores were relatively low for both general and cholinergic tracts white matter hyperintensities, indicating low cerebral microvascular pathology burden in this group. There was no correlation between these WMH visual scores and the DTI/HARDI measures within the isolated bundles. Neuropsychological testing results are shown in Table 2. Variability in results was present as shown by large standard deviations despite the relative homogeneity in age, years of education and health status of

2.5. Cholinergic streamlines isolation To isolate the presumed cholinergic white matter tracts, regions of interest, in this instance the external capsule and peri-insular white matter, were manually segmented using the MI-Brain software (Rheault et al., 2016; Rheault et al., 2017), on a mean T1-weigthed image (MNI 152 Cornell model; Fig. 1). Registration was accomplished on the segmented T1 using FLIRT linear registration (Jenkinson and Smith, 2001; Jenkinson et al., 2002), and then ANTS non-linear registration (Avants et al., 2009). Finally, both segmented zones were included in the 3

Experimental Gerontology 130 (2020) 110792

G. Nolze-Charron, et al.

Fig. 2. Tracked lateral cholinergic pathways.

participants, but remained within the norms for a cognitively normal elderly population.

Table 1 Patient characteristics.

3.2. Cholinergic pathways and cognitive testing The lateral cholinergic bundles were successfully isolated using probabilistic tractography with the HARDI technique (see Fig. 2). The SLF were also isolated (Fig. 3). No significant correlation was found between cognitive performance and the WM hyperintensities scales (ARWMC and CHIPS). Correlations between diffusion parameters within the LCT and neuropsychological test results were tested. A significant positive association between RD and MD of the left LCT and the time of completion of Stroop (inhibition/switching component) was observed,

Variable

Mean

Standard deviation

Age (years) Years of education (years) White matter volume (mm3) General white matter lesions (ARWMC, score; range 0–30) Cholinergic pathways white matter lesions (CHIPS, score; range 0–100) MMSE MOCA

72 14 988,150 3

5 4 119,801 3

14

13

29.5 27

0.8 2

Fig. 3. Tracked superior longitudinal fasciculi. 4

Experimental Gerontology 130 (2020) 110792

G. Nolze-Charron, et al.

moment the mainstay of symptomatic treatment for Alzheimer's disease (Hampel et al., 2018). The optimal way of measuring the correlation between cognition and white matter tracts in vivo is yet to be defined. Concerning the lateral cholinergic tracts, previous reports have shown loss of integrity at autopsy (Tomimoto et al., 2005) or have used semiquantitative measures using visual landmarks for white matter lesions in vivo (Bocti et al., 2005). The use of tractography, and furthermore of HARDI, might isolate the presumed cholinergic bundles. The resulting diffusion measures might also be more sensitive to detect microstructural changes, as opposed to visible microvascular damage only, especially as no significant correlation with cognitive performance was found using a semi-quantitative scale of white matter lesions in the cholinergic tracts. In the left LCT and right SLF, correlations were found between higher time of completion, and thus lower performance, and higher RD and MD, thereby reflecting perpendicular mobility of water molecules perpendicular to the presumed fiber direction. This could be explained by multiple processes including but not limited to myelin loss (for example less coherently arranged axons and increased membrane permeability.) These values are also subjected to partial volume effects and the crossing of fibers. AFD, a diffusivity measure of relative intra-axonal volume taken up by fibers aligned in one direction, can be given to a fiber even when a voxel contains multiple orientations, thus accounting for the fiber crossing problem (Raffelt et al., 2012). Surprisingly, in our study, lower AFD was associated with better performance in the Stroop test. This counterintuitive result suggests that there is more work needed to better understand the neurobiological underpinnings of dMRI results. In general, white matter damage and subcortical dementia is most typically associated with executive dysfunction in the cognitive profile (Jokinen et al., 2006). Several correlations between HARDI measures and neuropsychological test results were statistically significant in this cohort of healthy older adults. Lost directionality in the right and left LCT and right SLF was associated with lower performance in executive function, information-processing speed and attention tests, results consistent with previous studies showing an age-related change in diffusivity in large ROI of white matter (Head, 2004). Changes in white matter with aging have been reported specifically within the external capsule (Bohnen et al., 2009) and associated with executive dysfunction. Our findings are also in line with the results of a more recent study (Liu et al., 2017), which found in participants with vascular cognitive impairment but without dementia that reduced FA within both the medial and lateral cholinergic tracts explained in great part decreases in executive function testing results and in lesser part decreased memory and global cognition performances. Taken together, the data from these studies suggest that diffusion measures in the cholinergic tracts are a correlate of executive dysfunction even in non-demented elderly individuals. Our study also supports that information processing speed is in association of apparent diminished integrity in the LCT. We show that DTI and HARDI parameters within the white matter tracts of interest studied here, represent complementary tools to better understand cognition and aging. Historically, the SLF has mostly been linked to language, and was thought to link the Broca and Wernicke areas, especially with its arcuate component. However, some call into question this connection (Bernal and Altman, 2010), and few cases of conduction aphasia have been associated with an actual lesion in these bundles, and even agenesis of the SLF might not lead to this defect. It has more recently been linked to speech fluency, phonological and complex syntactic processing in speech production and comprehension. Thus, its functions have been more examined and questioned in recent years. Furthermore, some parts of the SLF, especially the SLF2, is thought to have important functions in the maintenance of attention and engagement in the environment (Schmahmann et al., 2008), which might explain the correlations found with the Stroop, Symbol Search and motor part of Trail Making test B in the current study. Such changes were also found in

Table 2 Neuropsychological testing results. Test

Mean

Standard deviation

WAIS-IV coding (scaled score) WMS-III visuospatial memory span (scaled score) WAIS-IV symbol search (scaled score) Trail making test B (seconds) Stroop (seconds)

11 11 10 114 75

2 3 3 48 16

Table 3 Associations between DTI and diffusion measures within isolated WM tracts and neuropsychological test results. Stroop Correlation

Symbol search

TMT-B (motor)

p

Correlation

p

Correlation

p

Right LCT FA −0.134 AD 0.206 RD 0.262 MD 0.261 GFA 0.039 AFD 0.256

0.45 0.24 0.14 0.14 0.83 0.14

−0.017 0.008 −0.010 −0.003 −0.060 −0.362⁎

0.92 0.96 0.96 0.99 0.73 0.04

−0.147 0.285 0.321 0.333 −0.126 −0.048

0.41 0.10 0.06 0.06 0.48 0.79

Left LCT FA −0.306 AD 0.240 RD 0.371 MD 0.348 GFA −0.168 AFD 0.287

0.08 0.17 0.03 0.04 0.34 0.10

0.234 0.026 −0.154 −0.096 0.126 −0.286

0.18 0.88 0.38 0.59 0.48 0.10

−0.165 0.432 0.412 0.450 −0.121 −0.048

0.35 0.01 0.02 0.01 0.50 0.79

Right SLF FA −0.239 AD 0.270 RD 0.407 MD 0.407 GFA −0.139 AFD 0.275

0.17 0.12 0.02 0.02 0.43 0.12

0.046 −0.007 −0.041 −0.033 0.008 −0.372

0.80 0.97 0.82 0.86 0.97 0.03

0.280 0.450 0.115 0.280 0.357 −0.082

0.11 0.01 0.52 0.11 0.04 0.65

Left SLF FA −0.158 AD 0.231 RD 0.299 MD 0.301 GFA −0.064 AFD 0.198

0.37 0.19 0.09 0.08 0.72 0.26

−0.107 0.028 0.060 0.053 −0.130 −0.257

0.55 0.88 0.74 0.77 0.46 0.14

0.138 0.334 0.160 0.242 0.278 −0.170

0.44 0.05 0.37 0.17 0.11 0.34

⁎

Data in bold are statistically significant results.

reflecting a link between higher perpendicular diffusivity in theses tracts and lesser performance on executive functions testing. A similar correlation was identified between the right SLF and Stroop performance. Surprisingly, the AFD of the right LCT and SLF showed a negative correlation with the Symbol Search score, reflecting a possible correlation between a decrease in this parameter and better information processing speed. Finally, there was a positive association between AD, RD and MD of the left LCT and AD and GFA of the right SLF and the motor part of the TMT-B (linked to motor speed). These results are shown in Table 3. 4. Discussion This study successfully isolated white matter tracts that contain the presumed lateral cholinergic tracts within the external capsule using dMRI tractography. Indeed, the obtained bundles using the external capsule as a ROI followed the pathways previously described pathologically by Selden et al. (1998). The role of acetylcholine on cognition is important, whether in the pathophysiology of AD or in many cases of vascular cognitive impairment. Anticholinergic medications have negative effects on cognition, and cholinesterase inhibitors are for the 5

Experimental Gerontology 130 (2020) 110792

G. Nolze-Charron, et al.

recent studies, where right- or left-sided SLF diffusivity measures were also related to attention skills in samples of different ages (Frye et al., 2010; Rizio and Diaz, 2016; Lunven and Bartolomeo, 2017). To further complete the study of the external capsule, our group tried to extract the inferior fronto-occipital fasciculus. Unfortunately, the resulting tracts did not appear to be reliable, and had very inconsistent spatial coverage. These unreliable results were not analyzed. In order to more globally study the external capsule, a further study could also extract the uncinate and inferior longitudinal bundles. Also, the LCT obtained by our method yielded a large number of streamlines. This might be reflective of fibers from other tracts being contained in our LCT. However, the cholinergic tracts have diffuse projections and a large number of streamlines might be expected. Another limit of this study was the relatively small sample size, with a commensurate reduction in the power to detect associations between imaging parameters and cognitive functions. This might explain the lack of significant correlations between cognitive performance and FA measures. This was an exploratory study, so no correction for multiple comparisons was applied in order to identify targets for potential further studies. Also, the medial cholinergic tracts were not isolated in this study. Biomarkers for dementia were not obtained; these could account for some differences in cognitive performance among participants. Finally, this study was done in a mostly homogenous sample with little variation in health and cognitive status. Hence, future applications for the use of this particular technique might include clinical groups and longitudinal data, and eventually with the use of potential therapies.

not in wm_posteriorcingulate.left not in wm_postcentral.left not in wm_parsopercularis.left not in wm_parstriangularis.left not in ctx_rostralmiddlefrontal.left SLF was extracted using the Free Surfer Atlas SLF1.left |= 6010 SLF1.right |= 6020 SLF3.left |= 6030 SLF3.right |= 6040 SLF2.left |= 6070 SLF2.right |= 6080 SLF1_waypoint.left |= 5050 SLF1_Start.left |= 5051 SLF1_End.left |= 5052 SLF1_waypoint.right |= 5053 SLF1_Start.right |= 5054 SLF1_End.right |= 5055 SLF2_waypoint.left |= 5056 SLF2_Start.left |= 5057 SLF2_End.left |= 5058 SLF2_waypoint.right |= 5059 SLF2_Start.right |= 5060 SLF2_End.right |= 5061 SLF3_waypoint.left |= 5062 SLF3_Start.left |= 5063 SLF3_End.left |= 5064 SLF3_waypoint.right |= 5065 SLF3_Start.right |= 5066 SLF3_End.right |= 5067

5. Conclusions This study successfully isolated the LCT in vivo using tractography in an older healthy individuals group. HARDI, with isolation of key tracts for cognitive functions and better analysis of crossing fibers, represents an innovative tool to better understand the complex process of brain aging and cognitive decline.

References Andersson, J.L.R., Sotiropoulos, S.N., 2016. An integrated approach to correction for offresonance effects and subject movement in diffusion MR imaging. Neuroimage 125, 1063–1078. Avants, B., Tustison, N., Song, G., 2009. Advanced Normalization Tools, ANTS 1.0. Sourceforge. Avants, B.B., Tustison, N.J., Song, G., Cook, P.A., Klein, A., Gee, J.C., 2011. A reproducible evaluation of ANTs similarity metric performance in brain image registration. Neuroimage 54 (3), 2033–2044. Basser, P.J., Mattiello, J., LeBihan, D., 1994. MR diffusion tensor spectroscopy and imaging. Biophys. J. 66 (1), 259–267. Behl, P., Bocti, C., Swartz, R.H., Gao, F., Sahlas, D.J., Lanctot, K.L., Streiner, D.L., Black, S.E., 2007. Strategic subcortical hyperintensities in cholinergic pathways and executive function decline in treated Alzheimer patients. Arch. Neurol. 64, 266–272. Behrens, T.E., Berg, H.J., Jbabdi, S., Rushworth, M.F., Woolrich, M.W., 2007. Probabilistic diffusion tractography with multiple fibre orientations: what can we gain? Neuroimage 34 (1), 144–155. Behrman-Lay, A.M., Usher, C., Conturo, T.E., Correia, S., Laidlaw, D.H., Lane, E.M., Bolzenius, J., Heaps, J.M., Salminen, L.E., Baker, L.M., Cabeen, R., Akbudak, E., Luo, X., Yan, P., Paul, R.H., 2015. Fiber bundle length and cognition: a length-based tractography MRI study. Brain Imaging Behav. 9, 765–775. Bernal, B., Altman, N., 2010. The connectivity of the superior longitudinal fasciculus: a tractography DTI study. J. Magn. Reson. Imaging 28 (2), 217–225. Bocti, C., Swartz, R.H., Gao, F.Q., Sahlas, D.J., Behl, P., Black, S.E., 2005. A new visual rating scale to assess strategic white matter hyperintensities within cholinergic pathways in dementia. Stroke 36, 2126–2131. Bohnen, N.I., Bogan, C.W., Müller, M.L., 2009. Frontal and periventricular brain white matter lesions and cortical deafferentation of cholinergic and other neuromodulatory axonal projections. Eur. J. Neurol. 1 (1), 33–50. Bosch, B., Arenaza-Urquijo, E.M., Rami, L., Sala-Llonch, R., Junqué, C., Solé-Padullés, C., Peña-Gómez, C., Bargalló, N., Molinuevo, J.L., Bartrés-Faz, D., 2012. Multiple DTI index analysis in normal aging, amnestic MCI and AD: relationship with neuropsychological performance. Neurobiol. Aging 33, 61–74. Boyle, P.A., Yu, L., Fleischman, D.A., Leurgans, S., Yang, J., Wilson, R.S., Schneider, J.A., Arvanitakis, Z., Arfanakis, K., Bennett, D.A., 2016. White matter hyperintensities, incident mild cognitive impairment, and cognitive decline in old age. Ann. Clin. Transl. Neurol. 3 (10), 791–800. Brun, A., Englund, E., 1986. A white matter disorder in dementia of the Alzheimer type: a pathoanatomical study. Ann. Neurol. 9, 253–262. Caiazzo, G., Trojsi, F., Cirillo, M., Tedeschi, G., Esposito, F., 2016. Q-ball imaging models: comparison between high and low angular resolution diffusion-weighted MRI protocols for investigation of brain white matter integrity. Neuroradiology 58 (2), 209–215.

Funding This research was supported by a grant from the Research Centre in Aging of the CIUSSS de l'Estrie-Centre Hospitalier Universitaire de Sherbrooke. Declaration of competing interest Maxime Descoteaux is co-founder of Imeka Solutions Inc. Appendix A. White matter query language Right LCT FINAL_capsd_filt_18 = capsextd3 not in brain_stem not in cc_full not in hemisphere.left not in insulad3 not in ctx_parsorbitalis.right not in wm_medialorbitofrontal.right not in wm_middletemporal.right not in ctx_supramarginal.right not in wm_superiortemporal.right not in wm_posteriorcingulate.right not in wm_postcentral.right not in wm_parsopercularis.right not in wm_parstriangularis.right not in ctx_rostralmiddlefrontal.right Left LCT FINAL_capsg_filt_18 = capsextg3 not in brain_stem not in cc_full not in hemisphere.right not in insulag3 not in ctx_parsorbitalis.left not in wm_medialorbitofrontal.left not in wm_middletemporal.left not in ctx_supramarginal.left not in wm_superiortemporal.left 6

Experimental Gerontology 130 (2020) 110792

G. Nolze-Charron, et al. Concha, L., 2014. A macroscopic view of microstructure: using diffusion-weighted images to infer damage, repair, and plasticity of white matter. J. Neurosci. 276, 14–28. Coupé, P., Yger, P., Prima, S., Hellier, P., Kervrann, C., Barillot, C., 2008. An optimized blockwise nonlocal means denoising filter for 3-D magnetic resonance images. IEEE Trans. Med. Imaging 27 (4), 425–441. Cullen, K.M., Halliday, G.M., 1998. Neurofibrillary degeneration and cell loss in the nucleus basalis in comparison to cortical Alzheimer pathology. Neurobiol. Aging 19, 297–306. Debette, S., Markus, H.S., 2010. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. BMJ 341, c3666. Delis, D.C., Kaplan, E., Kramer, J.H., 2001. Delis-Kaplan Executive Function System: Technical Manual. Harcourt Assessment Company, San Antonio. Descoteaux, M., 2015. High Angular Resolution Diffusion Imaging (HARDI). Wiley Encyclopedia of Electrical and Electronics Engineering, pp. 1–25. Descoteaux, M., Deriche, R., Knosche, T.R., Anwander, A., 2009. Deterministic and probabilistic tractography based on complex fibre orientation distributions. IEEE Trans. Med. Imaging 28 (2), 269–286. Dick, A.S., Bernal, B., Tremblay, P., 2014. The language connectome: new pathways, new concepts. Neuroscientist 20 (5), 453–467. Fieremans, E., Benitez, A., Jensen, J.H., Falangola, M.F., Tabesh, A., Deardorff, R.L., Spampinato, M.V.S., Babb, J.S., Novikov, D.S., Ferris, S.H., Helpern, J.A., 2013. Novel white matter tract integrity metrics sensitive to Alzheimer Disease progression. AJNR Am. J. Neuroradiol. 34, 2105–2112. Folstein, M.F., Folstein, S.E., McHugh, P.R., 1975. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J. Psychiatr. Res. 12, 189–198. Frye, R.E., Hasan, K., Malmberg, B., Desouza, L., Swank, P., Smith, K., Landry, S., 2010. Superior longitudinal fasciculus and cognitive dysfunction in adolescents born preterm and at term. Dev. Med. Child. Neurol. 52 (8), 760–766. Gao, F.Q., Pettersen, J.A., Bocti, C., Nestor, S.M., Kiss, A., Black, S.E., 2013. Is encroachment of the carotid termination into the substantia innominata associated with its atrophy and cognition in Alzheimer’s disease? Neurobiol. Aging 34 (7), 1807–1814. https://doi.org/10.1016/j.neurobiolaging.2013.01.009. Garyfallidis, E., Brett, M., Amirbekian, B., Rokem, A., van der Walt, S., Descoteaux, M., Nimmo-Smith, I., 2014. Dipy, a library for the analysis of diffusion MRI data. Front. Neuroinform. 8 (8). Geula, C., Nagykery, N., Nicholas, A., Wu, C.K., 2008. Cholinergic neuronal and axonal abnormalities are present early in aging and in Alzheimer disease. J. Neuropathol. Exp. Neurol. 67, 309–318. Girard, G., Whittingstall, K., Deriche, R., Descoteaux, M., 2014. Towards quantitative connectivity analysis: reducing tractography biases. Neuroimage 98, 266–278. Hampel, H., Mesulam, M.M., Cuello, A.C., Farlow, M.R., Giacobini, E., Grossberg, G.T., Khachaturian, A.S., Vergallo, A., Cavedo, E., Snyder, P.J., Khachaturian, Z.S., for the Cholinergic System Working Group, 2018. The cholinergic system in the pathophysiology and treatment of Alzheimer’s disease. Brain 141, 1917–1933. Hanyu, H., Shimizu, S., Tanaka, Y., Hirao, K., Iwamoto, T., Abe, K., 2007. MR features of the substantia innominata and therapeutic implications in dementias. Neurobiol. Aging 28, 548–554. Hasselmo, M.E., 2006. The role of acetylcholine in learning and memory. Curr. Opin. Neurobiol. 16 (6), 710–715. Head, D., 2004. Differential vulnerability of anterior white matter in nondemented aging with minimal acceleration in dementia of the Alzheimer type: evidence from diffusion tensor imaging. Cereb. Cortex 14 (4), 410–423. Hong, J.H., Jang, S.H., 2010. Neural pathway from nucleus basalis of Meynert passing through the cingulum in the human brain. Brain Res. 1346, 190–194. Hong, J.H., Jang, S.H., Kim, O.L., Kim, S.H., Ahn, S.H., Byun, W.M., Hong, C.P., Lee, D.H., 2012. Neuronal loss in the medial cholinergic pathway from the nucleus basalis of meynert in patients with traumatic axonal injury: a preliminary diffusion tensor imaging study. J. Head Trauma Rehabil. 27 (3), 172–176. Iadecola, C., 2013. The pathobiology of vascular dementia. Neuron 80 (4), 1–46. Inzitari, D., Simoni, M., Pracucci, G., Poggesi, A., Basile, A.M., Chabriat, H., Erkinjuntti, T., Fazekas, F., Ferro, J.M., Hennerici, M., Langhorne, P., O’Brien, J., Barkhof, F., Visser, M.C., Wahlund, L.O., Waldemar, G., Wallin, A., Pantoni, L., LADIS Study Group, 2007. Risk of rapid global functional decline in elderly patients with severe cerebral age-related white matter changes: the LADIS study. Arch. Intern. Med. 167 (1), 81–88. Jenkinson, M., Smith, S.M., 2001. A global optimisation method for robust affine registration of brain images. Med. Image Anal. 5 (2), 143–156. Jenkinson, M., Bannister, P.R., Brady, J.M., Smith, S.M., 2002. Improved optimisation for the robust and accurate linear registration and motion correction of brain images. NeuroImage 17 (2), 825–841. Jeurissen, B., Leemans, A., Tournier, J.D., Jones, D.K., Sijbers, J., 2012. Investigating the prevalence of complex fiber configurations in white matter tissue with diffusion magnetic resonance imaging. Hum. Brain Mapp. 11, 2747–2766. Jokinen, H., Kalska, H., Mantyla, R., Pohjasvaara, T., Ylikoski, R., Hietanen, M., Salonen, O., Kaste, M., Erkinjuntti, T., 2006. Cognitive profile of subcortical ischaemic vascular disease. J. Neurol. Neurosurg. Psychiatry 77, 28–33. Klinkenberg, I., Sambeth, A., Blokland, A., 2011. Acetylcholine and attention. Behav. Brain Res. 221 (2), 430–442. Lee, S.H., Coutu, J.P., Wilkens, P., Yendiki, A., Rosas, H.D., Salat, D.H., for the Alzheimer’s disease Neuroimaging Initiative (ADNI), 2015. Tract-based analysis of white matter degeneration in Alzheimer’s disease. Neuroscience 301, 79–89. Libon, D.J., Price, C.C., Giovannetti, T., Swenson, R., Bettcher, B.M., Heilman, K.M., Pennisi, A., 2008. Linking MRI Hyperintensities with patterns of neuropsychological impairment: evidence for a threshold effect. Stroke 39 (3), 806–813.

Liu, Q., Zhu, Z., Teipel, S.J., Yang, J., Xing, Y., Tang, Y., Jia, J., 2017. White matter damage in the cholinergic system contributes to cognitive impairment in subcortical vascular cognitive impairment, no dementia. Front Aging Neurosci. 9 (47), 1–9. Longstreth, W.T., Bernick, C., Manolio, T.A., Bryan, N., Jungreis, C.A., Price, T.R., 1998. Lacunar infarcts defined by magnetic resonance imaging of 3660 elderly people: the Cardiovascular Health Study. Arch. Neurol. 55 (9), 1217–1225. Lunven, M., Bartolomeo, P., 2017. Attention and spatial cognition: neural and anatomical substrates of visual neglect. Ann. Phys. Rehabil. Med. 60 (3), 124–129. McGeer, P.L., McGeer, E.G., Suzuki, J., Dolman, C.E., Nagai, T., 1984. Aging, Alzheimer’s disease, and the cholinergic system of the basal forebrain. Neurology 34, 741–745. Mesulam, M.M., Geula, C., 1988. Nucleus basalis (Ch4) and cortical cholinergic innervation in the human brain: observations based on the distribution of acetylcholinesterase and choline acetyltransferase. J. Comp. Neurol. 275, 216–240. Mori, S., van Zijl, P.C.M., 2002. Fiber tracking: principles and strategies – a technical review. NMR Biomed. 15, 468–480. Nasreddine, Z.S., Phillips, N.A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., Cummings, J.L., Chertkow, H., 2005. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J. Am. Geriatr. Soc. 53 (4), 695–699. Partington, J.E., Leiter, R.G., 1949. Partington’s pathway test. Psychol. Serv. Center Bull. 1, 9–20. Presseau, C., Jodoin, P.M., Houde, J.C., Descoteaux, M., 2015. A new compression format for fiber tracking datasets. Neuroimage 109, 73–83. Prins, N.D., van Dijk, E.J., den Heijer, T., Vermeer, S.E., Jolles, J., Koudstaal, P.J., Hofman, A., Breteler, M.M., 2005. Cerebral small-vessel disease and decline in information processing speed, executive function and memory. Brain 128 (9), 2034–2041. Raffelt, D., Tournier, J.D., Rose, S., Ridgway, G.R., Henderson, R., Crozier, S., Salvado, O., Connelly, A., 2012. Apparent fibre density: a novel measure for the analysis of diffusion-weighted magnetic resonance images. Neuroimage 59 (4), 3976–3994. Reuter, M., Schmansky, N.J., Rosas, H.D., Fischl, B., 2012. Within-subject template estimation for unbiased longitudinal image analysis. Neuroimage 61 (4), 1402–1418. Rheault, F., Houde, J.C., Goyette, N., Morency, F., Descoteaux, M., 2016. MI-Brain, a software to handle tractograms and perform interactive virtual dissection. In: Lisbon: Proceeding of: Breaking the Barriers of Diffusion MRI (ISMRM workshop). Rheault, F., Houde, J.C., Descoteaux, M., 2017. Visualization, interaction and tractometry: dealing with millions of streamlines from diffusion MRI tractography. Front. Neuroinform. 11 (42). Rizio, A., Diaz, T., 2016. Language, aging, and cognition: frontal aslant tract and superior longitudinal fasciculus contribute to working memory performance in older adults. Aging Clin. Exp. Res. 27, 689–693. Salat, D.H., 2014. Imaging small vessel-associated white matter changes in aging. Neuroscience 276, 174–186. Schmahmann, J.D., Pandya, D., 2006. Fiber Pathways of the Brain. Oxford University Press, Oxford. Schmahmann, J.D., Smith, E.E., Eichler, F.S., Filley, C.M., 2008. Cerebral white matter: neuroanatomy, clinical neurology, and neurobehavioral correlates. Ann. N. Y. Acad. Sci. 1142, 266–309. Selden, N.R., Gitelman, D.R., Salamon-Murayama, N., Parrish, T.B., Mesulam, M.M., 1998. Trajectories of cholinergic pathways within the cerebral hemispheres of the human brain. Brain 121 (12), 2249–2257. Sevigny-Dupont, P., et al., 2019. Amyloid Burden and White Matter Hyperintensities Mediate Age-Related Cognitive Differences. Neurobiology of aging In press. Smith, S.M., 2002. Fast robust automated brain extraction. Hum. Brain Mapp. 17 (3), 143–155. Sun, X., Salat, D., Upchurch, K., Deason, R., Kowall, N., Budson, A., 2014. Destruction of white matter integrity in patients with mild cognitive impairment and Alzheimer disease. J. Investig. Med. 62 (7), 927–933. Tomimoto, H., Ohtani, R., Shibata, M., Nakamura, N., 2005. Loss of cholinergic pathways in vascular dementia of the Binswanger type. Dement. Geriatr. Cogn. Disord. 19, 282–288. Tournier, J.D., Calamante, F., Connelly, A., 2007. Robust determination of the fibre orientation distribution in diffusion MRI: non-negativity constrained super-resolved spherical deconvolution. Neuroimage 35 (4), 1459–1472. Wahlund, L.O., Barkhof, F., Fazekas, F., Bronge, L., Augustin, M., Sjögren, M., Wallin, A., Ader, H., Leys, D., Pantoni, L., Pasquier, F., Erkinjuntti, T., Scheltens, P., 2001. A new rating scale for age related white matter changes applicable to MRI and CT. Stroke 32, 1318–1322. Wang, Y., West, J.D., Flashman, L.A., Wishart, H.A., Santulli, R.B., Rabin, L.A., Pare, N., Arfanakis, K., Saykin, A.J., 2012. Selective changes in white matter integrity in MCI and older adults with cognitive complaints. Biochim. Biophys. Acta 1822 (3), 423–430. Wassermann, D., Makris, N., Rathi, Y., Shenton, M., Kikinis, R., Kubicki, M., Westin, C.F., 2013. On describing human white matter anatomy: the white matter query language. Med. Image Comput. Comput. Assist. Interv. 16 (1), 647–654. Wechsler, D., 1997. Wechsler Memory Scale - Third Edition. The Psychological Corporation, San Antonio. Wechsler, D., 2008. Wechsler Adult Intelligence Scale, fourth edition. The Psychological Corporation, San Antonio (WAIS-IV). Zhang, Y., Brady, M., Smith, S., 2001. Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm. IEEE Trans. Med. Imag. 20 (1), 45–57. Zhuang, L., Wen, W., Zhu, W., Trollor, J., Kochan, N., Crawford, J., Reppermund, S., Brodaty, H., Sachdev, P., 2010. White matter integrity in mild cognitive impairment: a tract-based spatial statistics study. NeuroImage 53, 16–25.

7