Physical Activity Predicts Population-Level Age-Related Differences in Frontal White Matter

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Physical Activity Predicts Population-Level Age-Related Differences in Frontal White Matter © 2018 the Author(s) Published version Strömmer, Juho; Davis, Simon W.; Henson, Richard N.; Tyler, Lorraine K.; Consortium, Cambridge Centre for Ageing and Neuroscience; Campbell, Karen L. Strömmer, J., Davis, S. W., Henson, R. N., Tyler, L. K., Consortium, C. C. F. A. A. N., & Campbell, K. L. (2020). Physical Activity Predicts Population-Level Age-Related Differences in Frontal White Matter. Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 75(2), 236243. https://doi.org/10.1093/gerona/gly220 2020 © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. 236 Journals of Gerontology: Biological Sciences cite as: J Gerontol A Biol Sci Med Sci, 2020, Vol. 75, No. 2, 236–243 doi:10.1093/gerona/gly220 Advance Access publication October 4, 2018 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Original Article Physical Activity Predicts Population-Level Age-Related Differences in Frontal White Matter Juho M. Strömmer, PhD,1, Simon W. Davis, PhD,2 Richard N. Henson, PhD,3 Lorraine K.Tyler, PhD,4 Cam-CAN,3,4 and Karen L.Campbell, PhD5 1Department of Psychology, University of Jyväskylä, Finland. 2Department of Neurology, Duke University School of Medicine, Durham, North Carolina. 3Medical Research Council Cognition and Brain Sciences Unit and 4Department of Psychology, University of Cambridge, UK. 5Department of Psychology, Brock University, St Catharines, Ontario, Canada. Address correspondence to: Juho M.Strömmer, PhD, Department of Psychology, University of Jyväskylä, PO Box 35, Jyväskylä FI-40014, Finland. E-mail: [email protected] Received: April 30, 2018; Editorial Decision Date: September 17, 2018 Decision Editor: Rozalyn Anderson, PhD Abstract Physical activity has positive effects on brain health and cognitive function throughout the life span. Thus far, few studies have examined the effects of physical activity on white matter microstructure and psychomotor speed within the same, population-based sample (critical if conclusions are to extend to the wider population). Here, using diffusion tensor imaging and a simple reaction time task within a relatively large population-derived sample (N=399; 18–87years) from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN), we demonstrate that physical activity mediates the effect of age on white matter integrity, measured with fractional anisotropy. Higher self-reported daily physical activity was associated with greater preservation of white matter in several frontal tracts, including the genu of corpus callosum, uncinate fasciculus, external capsule, and anterior limb of the internal capsule. We also show that the age-related slowing is mediated by white matter integrity in the genu. Our findings contribute to a growing body of work, suggesting that a physically active lifestyle may protect against agerelated structural disconnection and slowing. Keywords: Brain aging, Exercise, Cognitive decline Aging is associated with profound changes in brain structure, including gray matter atrophy and alterations in the integrity of white matter (WM). Microstructural changes in the intracellular and extracellular components of WM occur throughout the aging brain, but tend to be more pronounced in frontal associative tracts (1,2). These age-related changes are thought to be driven largely by changes in myelin, with axon fibers being relatively unaffected by age (3). Fractional anisotropy (FA), an index of microstructural WM integrity that is sensitive to changes in cerebral myelin levels, as indexed by postmortem histology (4), declines progressively with age in healthy adults, especially in those WM tracts that mature later in life, such as anterior parts of corpus callosum (5). This loss of myelin integrity is considered one of the key mechanisms underlying normal age-related variability in cognitive performance (1,6), often surpassing gray matter volume estimates in the ability to account for age-related cognitive decline (7). Increased WM integrity is often positively correlated with better performance across a number of cognitive domains and, in some circumstances, WM integrity mediates age-related slowing of cognitive processing (8–12). In fact, a substantial portion of the age-related variance in cognitive processing speed has been shown to be attributable to decreases in frontal WM integrity (3,13). The role of WM structures like the genu of corpus callosum in mediating the effect of age on cognitive processing speed has now been replicated many times (2,12), and this relationship appears to be specific to processing speed and executive functioning, rather than other aspects of cognition (eg, language, motor functioning) (10). Thus, maintenance of WM structural connectivity appears to be particularly critical for the prevention of general age-related slowing. However, despite the ubiquity and cognitive relevance of these patterns of change in cerebral WM, the specific mediators explaining these effects—beyond chronological age itself—are unclear. Although several lifestyle factors probably contribute to the maintenance of WM integrity with age, one of the most robust Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020 predictors of WM health appears to be physical activity. High cardiorespiratory fitness and engagement in physical activity have been shown to have protective effects for WM integrity (14–16) and cognitive performance (17–19) in healthy older adults. Evidence from prospective studies also indicates that physical activity considerably reduces the risk of dementia and Alzheimer’s disease (20). Interestingly, Burzynska and colleagues (21) showed that not only engagement in physical activity, but also avoiding sedentary behavior, is important for preserving WM microstructural integrity later in life, possibly via different pathways. Sedentary lifestyle is more likely to be associated with obesity and poor aerobic fitness, and is a leading cause of disease and disability (22), which, in turn, are shown to be associated with lower WM integrity (23). Longitudinal data from aerobic exercise intervention programs in older adults show that the selective increases in fitness associated with aerobic exercise, but not low-intensity control interventions, predict increases in WM integrity in the prefrontal and temporal cerebrum (24) and increases WM volume in the anterior corpus callosum (25). As noted earlier, these brain regions are particularly vulnerable to the detrimental effects of age. Together, these studies emphasize the potential benefits of physical activity in preventing age-related WMloss. Although several studies suggest a link between exercise and differences in WM integrity with age (24,26), it remains to be seen whether this relationship holds within a large, population-based life span sample. Population-based samples are critical if our conclusions are to extend beyond relatively select (and potentially biased) samples of research volunteers to the population in general. Moreover, few studies, if any, have examined the relationship between brain health and participants’ reports of everyday activities and routines (encompassing such activities as cleaning the house and mode of transportation/distance to work), which arguably offer a more ecologically valid counterpoint to standard intervention studies (27). In this study, we examined the relationship between age, self-reported physical activity, WM microstructure, and processing speed within a large, population-based sample from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) (28). Participants (N = 399) completed a physical activity questionnaire (29) and series of cognitive tests, including simple reaction time (RT) task in their homes, before undergoing a series of structural and functional MRI scans, which included diffusion tensor imaging (28). Diffusion tensor imaging was used to estimate FA within 21 major tracts from the John Hopkins University (JHU) White Matter Atlas and related to physical activity and processing speed separately in a series of mediation models. Our first objective was to determine whether physical activity mediates age-related decline in WM within particular tracts and whether these are the tracts that are most susceptible to agerelated decline. To this end, separate mediation models were run for each tract, testing whether the relationship between age and FA was mediated by daily physical activity. Our second objective was to examine whether performance on the simple RT task is associated with WM integrity and whether the age-related decline in this measure is mediated by WM integrity. Only those tracts that showed a significant mediation effect of physical activity in the first model (corrected for multiple comparisons) were included into the second set of models testing the association between age, FA, and RT. Thus, our planned analyses will help to elucidate a possible explanation for age-related declines in WM health and provide evidence for the role of this measure in predicting declines in processingspeed. Methods Participants A healthy, population-based sample of 708 participants (age range 18–88years) was collected as part of the Cam-CAN (for a detailed description of the study, see ref. (28)). The ethical approval for the study was obtained from the Cambridgeshire 2 (now East of England—Cambridge Central) Research Ethics Committee. Participants gave written informed consent. Exclusion criteria included poor vision (below 20/50 on Snellen test (30)), poor hearing (failing to hear 35 dB at 1,000 Hz in either ear), low Mini–Mental Status Examination (24 or lower (31)), self-reported substance abuse (assessed by the Drug Abuse Screening Test [DAST-20 (32)]), poor English knowledge (non-native or nonbilingual English speaker), current psychiatric disorder, or neurological disease. In addition, people with contraindications to MRI or MEG were excluded. Handedness was assessed using Edinburgh Handedness Inventory (33). Of the initial 708, 646 participants had valid T1, T2, and diffusion tensor imaging/diffusion kurtosis imaging data. We also excluded participants who did not complete the RT task (n=75) and those with outlying FA values further than three times interquartile range above or below the age decile mean (n=25; total remaining N=399, 221 females, age range 18–87years). The sample characteristics are described in Table1. Table1. Participant Demographic Information by Age Decile Age Decile 1 2 3 4 5 6 7 N (percentage of the total 399) 28 (7) 72 (18) 70 (17) 59 (15) 67 (17) 60 (15) 43 (11) Age range (y) 18–27 28–37 38–47 48–57 58–67 68–77 78–87 Sex (male/female) 10/18 35/37 33/37 26/33 29/38 25/35 20/23 Highest education University 19 64 54 40 40 29 15 A’ levels 6 4 8 10 15 13 13 GCSE grade 3 4 8 8 9 10 7 None over 16 0 0 0 1 3 8 8 MMSE 29.18 (1.0) 29.49 (1.0) 28.94 (1.2) 29.05 (1.3) 28.93 (1.3) 28.57 (1.5) 28.02 (1.4) Simple RTmean (s) 0.34 (0.04) 0.34 (0.04) 0.35 (0.06) 0.36 (0.06) 0.38 (0.06) 0.40 (0.08) 0.41 (0.07) PAEE 43 (13) 46 (16) 52 (16) 47 (17) 38 (15) 34 (16) 25 (13) Note: MMSE=Mini–Mental Status Examination; PAEE=physical activity energy expenditure (kJ/d/kg); simple RTmean=mean RT on the simple RT task. Values in parentheses are SD. Journals of Gerontology: BIOLOGICAL SCIENCES, 2020, Vol. 75, No. 2 237 Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020 Imaging Preprocessing and Region-wise Analysis The MRI data were collected from a Siemens 3T TIM TRIO (Siemens, Erlangen, Germany). To estimate WM integrity, diffusionweighted images were acquired with a twice-refocused spin echo sequence, with 30 diffusion gradient directions each for b values of 1,000 and 2,000s/mm2, and three images acquired using a b value of 0 (echotime=104ms, repetitiontime=9.1s, voxel size=2×2× 2mm3, field of view=192× 192 mm2, 66 axial slices, GRAPPA acceleration factor=2). All preprocessing was completed using a combination of functions from FSL version 4.1.8 (bet, eddy, dtifit, and TBSS) and custom MATLAB scripts. The diffusion data were preprocessed for eddy currents and subject motion using an affine registration model. After removal of nonbrain tissue, a nonlinear diffusion tensor model was fit to the diffusionweightedimage volumes. Nonlinear fitting of the diffusion tensor provides a more accurate noise modeling than standard linear model fitting and enables various constraints on the diffusion tensor, such as positive definiteness. The tensor’s eigensystem was used to compute the FA at each voxel; FA maps were spatially normalized into a standard stereotactic space using tract-based spatial statistics (34). Images were then smoothed with a 6mm full width at half maximum Gaussian kernel to address possible residual errors and interindividual variability and to ensure the normality requirements of parametric statistics were met and then masked with a binarized version of each participant’s FA map, such that voxels below an FA threshold of 0.35 were not considered for further analysis. Next, the mean FA values over 21 bilaterally symmetrical tract regionofinterests(ROIs) from the JHU White Matter Atlas (http:// cmrm.med.jhmi.edu/) were extracted for subsequent analysis: genu of corpus callosum, body of corpus callosum, splenium of corpus callosum, column and body of fornix, fornix (cres), cerebral peduncle, anterior limb of internal capsule, posterior limb of internal capsule, retrolenticular part of internal capsule, anterior corona radiata, superior corona radiate, posterior corona radiate, posterior thalamic radiation, sagittal stratum, external capsule, cingulate gyrus, hippocampus, superior longitudinal fasciculus, superior fronto-occipital fasciculus, uncinate fasciculus, and tapetum. Physical Activity Questionnaire Information about physical activity energy expenditure (PAEE) was gathered as part of a larger self-completed questionnaire, which asked about education, training, travel, hobbies, and social activities. The questions about physical activity were based on items from the European Prospective Investigation into Cancer Study–Norfolk Physical Activity Questionnaire (EPIC-EPAQ2) (29). The full questionnaire is provided in Supplementary Material. Individual total PAEE per day (kJ/d/kg) was calculated from self-reported activities into metabolic equivalents (35,36), based on the standard definition of 1 MET as 3.5ml O2/min/kg (or 71 J/min/kg) based on the resting metabolic rate (37). In addition, PAEE was divided into subtypes in relation to the nature of the activity to investigate their contribution to total PAEE and age-related differences in it. Work PAEE includes all activities performed at work; Home PAEE includes homeand housework-related activities; Leisure PAEE includes all voluntary leisure activity and exercise; and Commute PAEE includes commuting to work and other travel. Response TimeTask In the simple RT task, participants were seated behind a computer screen and rested their right hand on a response box with four buttons (one for each finger). On the screen, they viewed an image of a hand with blank circles above each finger. Participants were instructed to press with the index finger as quickly as possible whenever the circle above the index finger in the image turned black. On pressing the button, or after maximum 3 seconds, the circle became blank again, and the variable intertrial interval began. The intertrial interval varied pseudorandomly with positively skewed distribution, minimum 1.8 seconds, mean 3.7 seconds, median 3.9 seconds, and maximum 6.8 seconds. The task included 50 trials, and mean RT was calculated for correct trials after applying a 3 SD trim to the data. Statistical Analysis Pearson’s correlation coefficients (partialling out gender and education) were computed to examine the relationship between age and total PAEE. For the PAEE subtypes (which were skewed in their distributions), Spearman’s rank correlation coefficients were computed (partialling out gender and education) to examine age-related changes in the types of activity contributing to totalPAEE. To test whether physical activity helps to predict the effects of age-related WM decline, we ran a series of mediation analyses, in which a third mediator variable fully or partially accounts for the relationship between an independent predictor and dependent outcome variables (38). In each analysis, the independent factor was age, the dependent factor was one of the 21 WM tracts (ie, mean FA within a tract), and the mediator was the amount of physical activity (Figure2D). For those tracks that showed a significant mediation effect, we went on to test the cognitive significance of that effect by examining the relationship between WM in those tracts and age-related slowing (Figure2). To this end, we ran another set of mediation analyses using age as the independent factor, simple RT as the dependent factor and mean FA within each of the previously identified tracts as the mediator (Figure2E). Direct effects of age on FA and RT were also included in these regressions. Statistical significance for mediation analyses is typically signified by a significant attenuation of the relationship (β value) between predictor and outcome variables, denoted here by a 95% confidence interval (CI) for standardized regression coefficient that does not cross zero. All significance tests were two tailed, and false discovery rate (39) at 0.05 was applied to protect against familywise Type Ierror. Results Aging and Physical Activity Total PAEE, controlled for gender and education, showed a gradual decline with age, r=−.37, pfdr < .001 (Figure1A). This is also shown in the results of the mediation models as path a, that is, the direct negative effect of age on total PAEE (Table2). Work-related activity (rho = −.52, pfdr < .001) and commuting-related activity (rho=−.46, pfdr < .001) showed moderate negative correlations with increasing age, but home-related activity showed a very weak correlation (rho=−.099, pfdr=.06) and leisure time activity no correlation (rho=−.09, pfdr=.09) with age (Figure1B). To conclude, leisureand home-related activities seem to remain stable across the life span, whereas workand commuting-related activities decline and probably contribute to the decline in total PAEE. Aging and WM Integrity The direct effect of age on FA was negative in all of the analyzed tracts, except the posterior limb of internal capsule, which showed 238 Journals of Gerontology: BIOLOGICAL SCIENCES, 2020, Vol. 75, No. 2 Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020 a small age-related increase in FA (Table2, path c). The effect of age on FA was relatively large (standardized β’s < −0.5) in the genu and body of the corpus callosum, fornix, anterior corona radiata, posterior thalamic radiation, sagittal stratum, and tapetum (Figure2A and C). Physical Activity and WM Integrity The first mediation analyses tested whether total PAEE mediated the age–FA relationships. Four tracts showed a mediation effect that survived false discovery rate correction: genu of corpus callosum, anterior limb of internal capsule, external capsule, and uncinate fasciculus (Table2, path ab, Figure2A and C). The mediation effects of PAEE on these WM tracts are positive (Table2, path ab), suggesting that higher physical activity is associated with less age-related WM degeneration (see Figure2A). (These effects remain equal when controlling for gender and education, although gender has a direct effect on FA in anterior limb of internal capsule [B=−0.426, SE=0.096, 95% CI = −0.616 to −0.237]. These analyses were also run with body mass index as a covariate to investigate the association between body mass index and PAEE. There were no associations between body mass index and PAEE, although body mass index has a marginal direct effect on external capsule [ß=−0.100, SE=0.051, 95% CI = −0.200 to −0.001].) No mediation effects were found when different PAEE types were used as mediator instead of total PAEE. WM Integrity and Speed of Processing The second mediation analyses tested whether FA (in the tracts related to exercise) mediated the relationship between age and processing speed. As expected, age was associated with slower responding on the simple RT task, B=0.362, CI=0.273 to 0.452, SE=0.047 (Table3, path c). Critically, mean FA in the genu of corpus callosum significantly mediated the effect of age on RT (ab=0.150, CI=0.045 to 0.251, SE=0.050; Table3, path ab, Figure2B), suggesting that preservation of WM in the genu of corpus callosum is associated with less age-related slowing (Figure2B). None of the other tracts showed significant mediation or main effects (Table3, path ab and c). (These effects remain equal when controlling for gender and education, although gender has a direct effect on FA in anterior limb of internal capsule [B=−0.426, SE=0.096, 95% CI=−0.616 to −0.237].) Discussion This study had two major aims. First, we examined whether physical activity mediates the effects of age on WM integrity. In line with previous work, we found higher physical activity to have positive effects that may protect against the damaging effects of age on FA in anterior WM tracts, namely the genu of corpus callosum, uncinate fasciculus, anterior limb of internal capsule, and external capsule. The second aim of this study was to examine whether WM integrity within the tracts that benefit from physical activity mediates agerelated slowing of processing speed. Of the four tracts tested, only the genu of the corpus callosum mediated a significant portion of the variance between age and RT on a simple motortask. This is the first study, to our knowledge, to show a relationship between self-reported everyday activities and FA in a populationderived sample. Although our results rely on a cross-sectional sample, and thus cannot relate physical activity to rates of longitudinal change, these results suggest that those who are more physically active in their day-to-day lives also have more youth-like patterns of WM microstructure. This is consistent with previous studies focusing on healthy older individuals, which have linked higher selfreported physical activity to higher WM volume (40) and lesser WM atrophy (15). Objectively measured cardiorespiratory fitness has also been shown to be associated with FA in the cingulum (23) and large portion of the corpus callosum (14) in older adults. Arecent study with two large samples of older adults demonstrated that WM tracts between prefrontal regions and medial temporal lobe are particularly associated with cardiorespiratory fitness and that these associations mediate spatial working memory performance (41). In our sample, which covers the whole adult age range from 18 to 87years, higher everyday physical activity was associated with less age-related loss of WM in several adjacent anterior tracts. Similarly, a recent study showed that higher cardiorespiratory fitness, assessed with the maximum volume of oxygen uptake (peak VO2), is related to higher FA in several WM tracts in older adults (42). Their study found regional specificity in the sensitivity to cardiorespiratory fitness, including genu of corpus callosum as one of the responsive regions. As with the current results, they showed that not all WM tracts that decline with age are associated with cardiorespiratory fitness. Overall, physical activity declined with increasing age. This appears to be due largely to a decrease in activity related to work and commuting, whereas homeand leisure-related activities remained relatively stable across the age span. These results are in line with a recent review concluding that in childhood, habituation to active lifestyle, like active travel or outdoor play, are important Figure1. (A) The effect of age on total physical activity energy expenditure (PAEE). (B) The effect of age on PAEE subtypes of home-, work-, leisure-, and commuting-related activities. Journals of Gerontology: BIOLOGICAL SCIENCES, 2020, Vol. 75, No. 2 239 Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020 contributors to total daily physical activity, whereas in adulthood, life events have the greatest influence on physical activity behavior (43). In the present data, a drop in work-related activity around Table2. Mediation Models Testing the Mediation of the Relationship Between Age and White Matter Integrity (Fractional Anisotropy) in Genu of Corpus Callosum, Anterior Limb of Internal Capsule, External Capsule, and Uncinate Fasciculus by Physical Activity Energy Expenditure White Matter Tract Path a (Age → PAEE) Path b (PAEE → FA) Path ab (Mediation Effect) Path c′ (Residual Age → FA) Path c (Age → FA) B (95% CI) B SE t p B (95% CI) B SE t p B (95% CI) B SE B (95% CI) B SE t p B (95% CI) B SE t p Genu of corpus callosum −0.382 (−0.474 to −0.291) 0.046 8.24 <.001 0.092 (0.019 to 0.164) 0.037 2.50 .017 −0.035 (−0.066 to −0.006)* 0.015 −0.696 (−0.768 to −0.623) 0.037 18.89 <.001 −0.731 (−0.798 to −0.664) 0.034 21.34 <.001 Anterior limb of internal capsule −0.382 (−0.474 to −0.291) 0.046 8.24 <.001 0.118 (0.014 to 0.221) 0.053 2.23 .033 −0.045 (−0.093 to −0.004)* 0.023 −0.173 (−0.277 to −0.070) 0.053 3.29 .001 −0.218 (−0.315 to −0.122) 0.049 4.46 <.001 External capsule −0.382 (−0.474 to −0.291) 0.046 8.24 <.001 0.102 (0.005 to 0.200) 0.050 2.07 .049 −0.040 (−0.083 to −0.003)* 0.020 −0.365 (−0.463 to −0.268) 0.050 7.82 <.001 −0.404 (−0.495 to −0.314) 0.046 8.81 <.001 Uncinate fasciculus −0.382 (−0.474 to −0.291) 0.046 8.24 <.001 0.141 (0.037 to 0.245) 0.053 2.66 .011 −0.054 (−0.100 to −0.013)* 0.022 −0.138 (−0.242 to −0.034) 0.053 2.61 .012 −0.192 (−0.289 to −0.096) 0.049 3.91 <.001 Notes: B=standardized regression coefficient; CI=confidence interval; FA=fractional anisotropy; PAEE=Physical Activity Energy Expenditure. *Significant mediation effects (for all effects, significance is denoted by a 95% CI that does not cross zero; false discovery rate corrected p value < .05). Figure 2. (A) The relationship between white matter integrity (fractional anisotropy [FA]) and age (steeper line) and age controlled for PAEE (gentlier line) in genu of corpus callosum (GCC), external capsule (EC), anterior limb of internal capsule (ALIC), and uncinated fasciculus (UNC). FA decreases gradually with age within all of the analyzed white matter tracts: GCC: r = –. 731, p < .001; EC: r = –. 404, p < .001; ALIC: r = –. 218, p < .001; UNC: r = –. 192, p < .001. The detrimental effect of age on FA is diminished in all of the analyzed tracts when PAEE is partialled out from age: GCC: r = –. 688, p < .001; EC: r = –. 348, p < .001; ALIC: r = –. 163, p = .001; UNC: r = –. 130, p = .009. The results indicate a positive relationship between higher physical activity and age-related differences in white matter microstructure. (B) The relationship between reaction time and age (steeper line) and age controlled for white matter integrity (FA) in genu of corpus callosum (gentlier line). Reaction times become gradually slower with age: r = .362, p < .001. The effect of age on reaction time is diminished when FA in genu of the corpus callosum is partialled out from age: r = .156, p = .002. The results indicate a positive relationship between white matter integrity in anterior corpus callosum and age-related differences in reaction time performance. (C) White matter tract ROIs from JHU FA atlas. Tracts which survive the first stage of mediation analysis (genu, anterior limb of the internal capsule, and the external capsule) are rendered in (left to right) superior axial, sagittal, and oblique views. (D) Schematic representation of the mediation paths. PAEE mediates the effect of age on FA in genu of corpus callosum. (E) FA in genu of corpus callosum mediates the effect of age on reaction time. Full color version is available within the online issue. 240 Journals of Gerontology: BIOLOGICAL SCIENCES, 2020, Vol. 75, No. 2 Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020 60 years of age coincides with the mean retirement age in our sample. Thus, it may be that people whose everyday activity is highly dependent on the activities associated with work show the greatest drop in the total activity compared with those with an active lifestyle outside of working life. Thus, it seems particularly important to promote physical leisure activities among retired older adults, possibly with the help of societal actions. Age-related slowing of cognitive processing has been proposed to underlie age-related declines within various domains of cognition (44). In the present study, simple RT slowed gradually with increasing age, which is a common finding among various types of agerelated effects on speed of processing (45). Age-related slowing in RT was mediated by FA in the genu of corpus callosum, but not in the other tracts that related to physical activity. These findings are in line with an earlier study suggesting that WM deterioration in the anterior part of the corpus callosum may contribute to general age-related slowing (2), though other studies have also related the splenium of corpus callosum and anterior limb of internal capsule (46) and more global WM structure (12,47) to perceptual-motor speed. A recent study also showed that lower whole brain FA is linked to inefficient brain response to cognitive demands of locomotion (48). We acknowledge that our results do not speak to causality because mediation analyses based on cross-sectional data do not inevitably represent causal relationships between age, physical activity, WM integrity, and RT. Nevertheless, we assume that age, an independent factor in both of the mediation models, cannot be changed by the influence of other factors and, furthermore, that psychomotor speed (RT) is a result of nervous system functioning (WM integrity), rather than the other way round (49). However, the causal interaction between lifestyle factors (eg, physical activity) and brain structure remains unclear: It is well known that environment and behavior, including physical activity, can cause plastic changes in the brain, but at the same time, changes in brain structure and function are known to influence behavior (ie, willingness toward action demanding physical activity). Furthermore, the strength of such inferences, based on self-reported questionnaire data, are necessarily limited. Although the reliability of such questionnaires is high (50), their absolute validity is moderate at best. Thus, observations in large samples such as ours must be validated with more time-intensive vascular measures, such as VO2 uptake and neuroimaging measures of cerebral perfusion. In addition, it is important to note that the Cam-CAN sample represents the population in the United Kingdom, and thus these results may not generalize to a non-Caucasian population. To conclude, we found that self-reported levels of physical activity mediated age-related WM loss in a number of anterior tracts. Although bearing in mind the limitations of cross-sectional data and a mediation-based approach, our findings complement the evidence from previous work suggesting that a physically active lifestyle may have protective benefits against age-related structural disconnection and cognitive decline. The findings of this study further support public health recommendations about the benefits of leading a physically active lifestyle across the life span, including older adults. Supplementary Material Supplementary data is available at The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences online. Funding This work was supported by the Biotechnology and Biological Sciences Research Council (grant number BB/H008217/1). Table3. Mediation Models Testing the Mediation Between Age and Simple Reaction Time, by Correlation to the White Matter Integrity (Fractional Anisotropy) in Genu of Corpus Callosum, Anterior Limb of Internal Capsule, External Capsule, and Uncinate Fasciculus White Matter Tract Path a (Age → FA) Path b (FA → RT) Path ab (Mediation Effect) Path c′ (Residual Age → RT) Path c (Age → RT) B (95% CI) B SE t p B (95% CI) B SE t p B (95% CI) B SE B (95% CI) B SE t p B (95% CI) B SE t p Genu of corpus callosum −0.731 (−0.798 to −0.664) 0.034 21.00 <.001 −0.205 (−0.343 to −0.067) 0.070 2.92 .004 0.150 (0.045 to 0.251)* 0.050 0.213 (0.079 to 0.346) 0.068 3.44 .001 0.362 (0.273 to 0.452) 0.046 7.92 <.001 Anterior limb of internal capsule −0.218 (−0.315 to −0.122) 0.049 4.46 <.001 −0.084 (−0.196 to 0.029) 0.067 1.46 .166 0.018 (−0.006 to 0.051) 0.014 0.344 (0.252 to 0.437) 0.047 7.30 <.001 0.362 (0.273 to 0.452) 0.046 7.92 <.001 External capsule −0.404 (−0.504 to −0.305) 0.051 8.00 <.001 −0.039 (−0.150 to 0.072) 0.053 0.69 .525 0.016 (−0.023 to 0.061) 0.022 0.347 (0.247 to 0.447) 0.051 6.80 <.001 0.362 (0.273 to 0.452) 0.046 7.92 <.001 Uncinate fasciculus −0.192 (−0.290 to −0.095) 0.050 3.87 <.001 −0.013 (−0.101 to 0.075) 0.045 0.30 .768 0.003 (−0.014 to 0.020) 0.009 0.360 (0.269 to 0.451) 0.046 7.75 <.001 0.362 (0.273 to 0.452) 0.046 7.92 <.001 Notes: B=standardized regression coefficient; CI=confidence interval; FA=fractional anisotropy; RT=Reaction Time. *Significant mediation effects (for all effects, significance is denoted by a 95% CI that does not cross zero; false discovery rate corrected p value < .05). Journals of Gerontology: BIOLOGICAL SCIENCES, 2020, Vol. 75, No. 2 241 Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020 Acknowledgments We are grateful to the Cam-CAN respondents and their primary care teams in Cambridge for their participation in this study. We also thank colleagues at the MRC Cognition and Brain Sciences Unit MEG and MRI facilities for their assistance. The Cam-CAN corporate author consists of the project principal personnel: Lorraine K Tyler, Carol Brayne, Edward T.Bullmore, Andrew C.Calder, Rhodri Cusack, Tim Dalgleish, John Duncan, Richard N.Henson, Fiona E. Matthews, William D. Marslen-Wilson, James B. Rowe, Meredith A.Shafto; research associates: Karen Campbell, Teresa Cheung, Simon Davis, Linda Geerligs, Rogier Kievit, Anna McCarrey, Abdur Mustafa, Darren Price, David Samu, Jason R.Taylor, Matthias Treder, Kamen Tsvetanov, Janna van Belle, Nitin Williams; research assistants: Lauren Bates, Tina Emery, Sharon Erzinçlioglu, Andrew Gadie, Sofia Gerbase, Stanimira Georgieva, Claire Hanley, Beth Parkin, David Troy; affiliated personnel: Tibor Auer, Marta Correia, Lu Gao, Emma Green, Rafael Henriques; research interviewers: Jodie Allen, Gillian Amery, Liana Amunts, Anne Barcroft, Amanda Castle, Cheryl Dias, Jonathan Dowrick, Melissa Fair, Hayley Fisher, Anna Goulding, Adarsh Grewal, Geoff Hale, Andrew Hilton, Frances Johnson, Patricia Johnston, Thea Kavanagh-Williamson, Magdalena Kwasniewska, Alison McMinn, Kim Norman, Jessica Penrose, Fiona Roby, Diane Rowland, John Sargeant, Maggie Squire, Beth Stevens, Aldabra Stoddart, Cheryl Stone, Tracy Thompson, Ozlem Yazlik; and administrative staff: Dan Barnes, Marie Dixon, Jaya Hillman, Joanne Mitchell, Laura Villis. The principal personnel, including Lorraine K.Tyler, of the Cam-CAN project designed the study. Simon W.Davis and Juho M.Strömmer analyzed the data. The interpretation of data was performed by Juho M.Strömmer, Karen Campbell, and Simon W.Davis. The manuscript was prepared by Juho M.Strömmer and revised by all the coauthors of this study. All of the authors approve the final version of the manuscript to be published. Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. References 1. Madden DJ, Bennett IJ, Burzynska A, Potter GG, Chen NK, Song AW. Diffusion tensor imaging of cerebral white matter integrity in cognitive aging. Biochim Biophys Acta. 2012;1822:386–400. doi:10.1016/j. bbadis.2011.08.003 2. Salami A, Eriksson J, Nilsson LG, Nyberg L. Age-related white matter microstructural differences partly mediate age-related decline in processing speed but not cognition. Biochim Biophys Acta. 2012;1822:408– 415. doi:10.1016/j.bbadis.2011.09.001 3. Madden DJ, Spaniol J, Costello MC, etal. Cerebral white matter integrity mediates adult age differences in cognitive performance. J Cogn Neurosci. 2009;21:289–302. doi:10.1162/jocn.2009.21047 4. Schmierer K, Wheeler-Kingshott CA, Boulby PA, etal. Diffusion tensor imaging of post mortem multiple sclerosis brain. Neuroimage. 2007;35:467– 477. doi:10.1016/j.neuroimage.2006.12.010 5. Kochunov P, Williamson DE, Lancaster J, etal. Fractional anisotropy of water diffusion in cerebral white matter across the lifespan. Neurobiol Aging. 2012;33:9–20. doi:10.1016/j.neurobiolaging.2010.01.014 6. Madden DJ, Bennett IJ, Song AW. Cerebral white matter integrity and cognitive aging: contributions from diffusion tensor imaging. Neuropsychol Rev. 2009;19:415–435. doi:10.1007/s11065-009-9113-2 7. Westlye LT, Walhovd KB, Dale AM, etal. Life-span changes of the human brain white matter: diffusion tensor imaging (DTI) and volumetry. Cereb Cortex. 2010;20:2055–2068. doi:10.1093/cercor/bhp280 8. Bennett IJ, Madden DJ. Disconnected aging: cerebral white matter integrity and age-related differences in cognition. Neuroscience. 2014;276:187– 205. doi:10.1016/j.neuroscience.2013.11.026 9. Nilsson J, Thomas AJ, O’Brien JT, Gallagher P. White matter and cognitive decline in aging: a focus on processing speed and variability. J Int Neuropsychol Soc. 2014;20:262–267. doi:10.1017/S1355617713001458 10. Gazes Y, Bowman FD, Razlighi QR, O’Shea D, Stern Y, Habeck C. White matter tract covariance patterns predict age-declining cognitive abilities. Neuroimage. 2016;125:53–60. doi:10.1016/j.neuroimage.2015.10.016 11. Price D, Tyler LK, Neto Henriques R, etal.; Cam-CAN. Age-related delay in visual and auditory evoked responses is mediated by whiteand grey-matter differences. Nat Commun. 2017;8:15671. doi:10.1038/ncomms15671 12. Kievit RA, Davis SW, Griffiths J, Correia MM, Cam-Can, Henson RN. A watershed model of individual differences in fluid intelligence. Neuropsychologia. 2016;91:186–198. doi:10.1016/j. neuropsychologia.2016.08.008 13. Kochunov P, Coyle T, Lancaster J, et al. Processing speed is correlated with cerebral health markers in the frontal lobes as quantified by neuroimaging. Neuroimage. 2010;49:1190–1199. doi:10.1016/j. neuroimage.2009.09.052 14. Johnson NF, Kim C, Clasey JL, Bailey A, Gold BT. Cardiorespiratory fitness is positively correlated with cerebral white matter integrity in healthy seniors. Neuroimage. 2012;59:1514–1523. doi:10.1016/j. neuroimage.2011.08.032 15. Gow AJ, Bastin ME, Valde MC, etal. Neuroprotective lifestyles and the aging brain: activity, atrophy, and white matter integrity. Neurology. 2012;79:1802–1808. doi:10.1212/WNL.0b013e3182703fd2 16. Oberlin LE, Verstynen TD, Burzynska AZ, et al. White matter microstructure mediates the relationship between cardiorespiratory fitness and spatial working memory in older adults. Neuroimage. 2016;131:91–101. doi:10.1016/j.neuroimage.2015.09.053 17. Colcombe S, Kramer AF. Fitness effects on the cognitive function of older adults: a meta-analytic study. Psychol Sci. 2003;14:125–130. doi:10.1111/1467-9280.t01-1-01430 18. Sofi F, Valecchi D, Bacci D, etal. Physical activity and risk of cognitive decline: a meta-analysis of prospective studies. J Intern Med. 2011;269:107– 117. doi:10.1111/j.1365-2796.2010.02281.x 19. Sexton CE, Betts JF, Demnitz N, Dawes H, Ebmeier KP, Johansen-Berg H. A systematic review of MRI studies examining the relationship between physical fitness and activity and the white matter of the ageing brain. Neuroimage. 2016;131:81–90. doi:10.1016/j.neuroimage.2015.09.071 20. Hamer M, Chida Y. Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence. Psychol Med. 2009;39:3–11. doi:10.1017/S0033291708003681 21. Burzynska AZ, Chaddock-Heyman L, Voss MW, et al. Physical activity and cardiorespiratory fitness are beneficial for white matter in low-fit older adults. PLoS One. 2014;9:e107413. doi:10.1371/journal.pone.0107413 22. Bauer UE, Briss PA, Goodman RA, Bowman BA. Prevention of chronic disease in the 21st century: elimination of the leading preventable causes of premature death and disability in the USA. Lancet. 2014;384:45–52. doi:10.1016/S0140-6736(14)60648-6 23. Marks BL, Katz LM, Styner M, Smith JK. Aerobic fitness and obesity: relationship to cerebral white matter integrity in the brain of active and sedentary older adults. Br J Sports Med. 2011;45:1208–1215. doi:10.1136/ bjsm.2009.068114 24. Voss MW, Heo S, Prakash RS, etal. The influence of aerobic fitness on cerebral white matter integrity and cognitive function in older adults: results of a one-year exercise intervention. Hum Brain Mapp. 2013;34:2972–2985. doi:10.1002/hbm.22119 25. Colcombe SJ, Erickson KI, Scalf PE, et al. Aerobic exercise training increases brain volume in aging humans. J Gerontol ABiol Sci Med Sci. 2006;61:1166–1170. doi:10.1093/gerona/61.11.1166 26. Marks BL, Madden DJ, Bucur B, etal. Role of aerobic fitness and aging on cerebral white matter integrity. Ann N Y Acad Sci. 2007;1097:171–174. doi:10.1196/annals.1379.022 27. Best JR, Rosano C, Aizenstein HJ, et al.; Health, Aging and Body Composition Study. Long-term changes in time spent walking and subsequent cognitive and structural brain changes in older adults. Neurobiol Aging. 2017;57:153–161. doi:10.1016/j.neurobiolaging.2017.05.023 28. Shafto MA, Tyler LK, Dixon M, etal.; Cam-CAN. The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) study protocol: a cross-sectional, lifespan, multidisciplinary examination of healthy cognitive ageing. BMC Neurol. 2014;14:204. doi:10.1186/s12883-014-0204-1 242 Journals of Gerontology: BIOLOGICAL SCIENCES, 2020, Vol. 75, No. 2 Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020 29. Wareham NJ, Jakes RW, Rennie KL, Mitchell J, Hennings S, Day NE. Validity and repeatability of the EPIC-Norfolk physical activity questionnaire. Int J Epidemiol. 2002;31:168–174. doi:10.1093/ije/31.1.168 30. Snellen H. Probebuchstaben Zur Bestimmung Der Sehschärfe. Utrecht, The Netherlands: Van de Weijer; 1962. 31. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. Apractical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–198. doi:0022-3956(75)90026-6 32. Skinner HA. The drug abuse screening test. Addict Behav. 1982;7:363– 371. doi:10.1016/0306-4603(82)90005-3 33. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia. 1971;9:97–113. doi:10.1016/0028-3932 (71)90067-4 34. Peled S, Friman O, Jolesz F, Westin CF. Geometrically constrained two-tensor model for crossing tracts in DWI. Magn Reson Imaging. 2006;24:1263–1270. doi:10.1016/j.mri.2006.07.009 35. Ainsworth BE, Haskell WL, Herrmann SD, etal. 2011 Compendium of physical activities: a second update of codes and MET values. Med Sci Sports Exerc. 2011;43:1575–1581. doi:10.1249/MSS.0b013e31821ece12 36. Ainsworth BE, Haskell WL, Leon AS, etal. Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc. 1993;25:71–80. doi:10.1249/00005768-199301000-00011 37. Henry CJ. Basal metabolic rate studies in humans: measurement and development of new equations. Public Health Nutr. 2005;8:1133–1152. doi:10.1079/PHN2005801 38. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51:1173–1182. doi:10.1037/0022-3514.51.6.1173 39. Benjamini Y, Hochberg Y. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B. 1995;57:289–300. doi:10.2307/2346101 40. Benedict C, Brooks SJ, Kullberg J, etal. Association between physical activity and brain health in older adults. Neurobiol Aging. 2013;34:83–90. doi:10.1016/j.neurobiolaging.2012.04.013 41. Oberlin LE, Verstynen TD, Burzynska AZ, et al. White matter microstructure mediates the relationship between cardiorespiratory fitness and spatial working memory in older adults. Neuroimage. 2016;131:91–101. doi:10.1016/j.neuroimage.2015.09.053 42. Hayes SM, Salat DH, Forman DE, Sperling RA, Verfaellie M. Cardiorespiratory fitness is associated with white matter integrity in aging. Ann Clin Transl Neurol. 2015;2:688–698. doi:10.1002/acn3.204 43. Condello G, Puggina A, Aleksovska K, et al.; DEDIPAC Consortium. Behavioral determinants of physical activity across the life course: a “Determinants of Diet and Physical Activity” (DEDIPAC) umbrella systematic literature review. Int J Behav Nutr Phys Act. 2017;14:58. doi:10.1186/s12966-017-0510-2 44. Salthouse TA. The processing-speed theory of adult age differences in cognition. Psychol Rev. 1996;103:403–428. doi:10.1037/0033-295X.103.3.403 45. Salthouse TA. Aging and measures of processing speed. Biol Psychol. 2000;54:35–54. doi:10.1016/S0301-0511(00)00052-1 46. Madden DJ, Whiting WL, Huettel SA, White LE, MacFall JR, Provenzale JM. Diffusion tensor imaging of adult age differences in cerebral white matter: relation to response time. Neuroimage. 2004;21:1174–1181. doi:10.1016/j.neuroimage.2003.11.004 47. Johnson MA, Diaz MT, Madden DJ. Global versus tract-specific components of cerebral white matter integrity: relation to adult age and perceptual-motor speed. Brain Struct Funct. 2015;220:2705–2720. doi:10.1007/ s00429-014-0822-9 48. Lucas M, Wagshul ME, Izzetoglu M, Holtzer R. Moderating effect of white matter integrity on brain activation during dual-task walking in older adults. J Gerontol ABiol Sci Med Sci. 2018:1–7. doi:10.1093/ gerona/gly131 49. Gow AJ, Corley J, Starr JM, Deary IJ. Reverse causation in activity-cognitive ability associations: the Lothian Birth Cohort 1936. Psychol Aging. 2012;27:250–255. doi:10.1037/a0024144 50. Helmerhorst HJ, Brage S, Warren J, Besson H, Ekelund U. A systematic review of reliability and objective criterion-related validity of physical activity questionnaires. Int J Behav Nutr Phys Act. 2012;9:103. doi:10.1186/1479-5868-9-103 Journals of Gerontology: BIOLOGICAL SCIENCES, 2020, Vol. 75, No. 2 243 Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/75/2/236/5114653 by University of Jyvaskyla user on 02 February 2020