close

Вход

Забыли?

вход по аккаунту

?

j.neuroimage.2018.08.036

код для вставкиСкачать
Accepted Manuscript
Functional coherence of striatal resting-state networks is modulated by striatal iron
content
Alireza Salami, Bárbara Avelar-Pereira, Benjamín Garzón, Rouslan Sitnikov, Grégoria
Kalpouzos
PII:
S1053-8119(18)30734-1
DOI:
10.1016/j.neuroimage.2018.08.036
Reference:
YNIMG 15197
To appear in:
NeuroImage
Received Date: 6 April 2018
Revised Date:
14 August 2018
Accepted Date: 16 August 2018
Please cite this article as: Salami, A., Avelar-Pereira, Bá., Garzón, Benjamí., Sitnikov, R., Kalpouzos,
Gré., Functional coherence of striatal resting-state networks is modulated by striatal iron content,
NeuroImage (2018), doi: 10.1016/j.neuroimage.2018.08.036.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
1
Functional coherence of striatal resting-state networks is modulated
2
by striatal iron content
3
Alireza Salami*1,2,3, Bárbara Avelar-Pereira*1,2, Benjamín Garzón1, Rouslan
5
Sitnikov4, Grégoria Kalpouzos1
6
1
7
Sweden
8
2
Umeå Center for Functional Brain Imaging, Umeå University, Umeå, Sweden
9
3
Department of Integrative Medical Biology, Wallenberg Centre for Molecular
RI
PT
4
M
AN
U
SC
Aging Research Center, Karolinska Institutet and Stockholm University, Stockholm,
10
Medicine, Umeå University, Umeå, Sweden
11
4
12
*These authors have equally contributed to this work
MRI Research Center, Karolinska University Hospital, Stockholm, Sweden
14
15
TE
D
13
Running head: Resting-state networks and iron content
Correspondence: Alireza Salami, Aging Research Center, Department of
17
Neurobiology, Care Sciences and Society, Karolinska Institutet, Tomtebodavägen
18
18A, 171 65 Solna, Sweden, e-mail: Alireza.Salami@ki.se
20
21
22
23
24
25
AC
C
19
EP
16
ACCEPTED MANUSCRIPT
Abstract
27
Resting-state spontaneous fluctuations have revealed individual differences in the
28
functional architecture of brain networks. Previous research indicates that the striatal
29
network shows alterations in neurological conditions but also in normal aging.
30
However, the neurobiological mechanisms underlying individual differences in
31
striatal resting-state networks (RSNs) have been less explored. One candidate that
32
may account for individual differences in striatal spontaneous activity is the level of
33
local iron accumulation. Excessive iron in the striatum has been linked to a loss of
34
structural integrity and reduced brain activity during task performance in aging. Using
35
independent component analysis in a sample of 42 younger and older adults, we
36
examined whether higher striatal iron content, quantified using relaxometry, underlies
37
individual differences in spontaneous fluctuations of RSNs in general, and of the
38
striatum in particular. Higher striatal iron content was linked to lower spontaneous
39
coherence within both caudate and putamen RSNs regardless of age. No such links
40
were observed for other RSNs. Moreover, the number of connections between the
41
putamen and other RSNs was negatively associated with iron content, suggesting that
42
iron modulated the degree of cross-talk between the striatum and cerebral cortex.
43
Importantly, these associations were primarily driven by the older group. Finally, a
44
positive association was found between coherence in the putamen and motor
45
performance, suggesting that this spontaneous activity is behaviorally meaningful. A
46
follow-up mediation analysis also indicated that functional connectivity may mediate
47
the link between striatal iron and motor performance. Our preliminary findings
48
suggest that striatal iron potentially accounts for individual differences in spontaneous
49
striatal fluctuations, and might be used as a locus of intervention.
51
52
53
54
55
56
SC
M
AN
U
TE
D
EP
AC
C
50
RI
PT
26
Keywords: aging, iron, functional connectivity, resting-state fMRI, striatum
ACCEPTED MANUSCRIPT
57
1. Introduction
Resting-state spontaneous fluctuations have shown to be a powerful tool to
59
map the functional architecture of the brain (Biswal et al. 2010; Fox and Greicius
60
2010; Buckner et al. 2013). These intrinsic neuronal signals are organized in a
61
hierarchy of networks that form large-scale functional circuits and integrate resources
62
needed to perform complex tasks (Allen et al. 2011; Power et al. 2011; Yeo et al.
63
2011; van den Heuvel and Sporns 2013). Several studies have shown that resting-state
64
networks (RSNs) are altered in aging (Chan et al. 2014; Geerligs et al. 2014; Ng et al.
65
2016; Salami et al. 2016; Avelar-Pereira et al. 2017), and have been linked to age-
66
related differences in a variety of cognitive measures, including episodic memory
67
(Salami et al. 2014; Fjell et al. 2015), executive functions (Andrews-Hanna et al.
68
2007), self-reference processes (Grady et al. 2016), and processing speed (Ng et al.
69
2016).
M
AN
U
SC
RI
PT
58
Despite the fact that resting-state connectivity has been extensively studied in
71
the context of aging (Grady, 2016; Campbell and Schacter 2017; Geerligs and
72
Tsvetanov 2017), the neurobiological mechanisms underlying individual differences
73
in RSNs have been less explored. A number of hypotheses have been put forward in
74
an attempt to explain age-related alterations in functional networks and their
75
concomitant cognitive decline [for a review see Ferreira and Busatto, 2013]. More
76
specifically, metabolic (cerebral blood flow), structural (gray- and white-matter
77
integrity), and molecular (dopamine and amyloid deposition) characteristics of the
78
brain have been shown to influence RSNs, in both healthy and pathological aging
79
(van den Heuvel et al. 2009; Zhang and Raichle 2010; Liang et al. 2013; Betzel et al.
80
2014; Salami et al. 2014; Fjell et al. 2015; Nyberg et al. 2016; Salami et al. 2016).
81
Another candidate that may account for individual differences in RSNs is iron
82
accumulation in the striatum, which has been previously linked to age-related
83
cognitive and motor decline (Daugherty and Raz, 2015).
EP
AC
C
84
TE
D
70
There are essentially two types of iron in the brain: heme iron, found in the
85
blood, and intracellular non-heme iron, which binds to the storage protein of iron,
86
ferritin. Non-heme iron is crucially involved in numerous biological processes, such
87
as mitochondrial ATP production (adenosine triphosphate, a major substrate of
88
cellular energy metabolism), neurotransmitter synthesis, synaptic plasticity, and
89
myelination (Hare et al. 2013; Daugherty and Raz 2015). Although almost absent at
90
birth, brain iron concentration rapidly increases during childhood (Hallgren &
ACCEPTED MANUSCRIPT
Sourander, 1958) and has been linked to beneficial effects in cognitive performance
92
(Hect et al., 2018). After a relative stabilization over adulthood, brain iron levels tend
93
to further increase in older age. A higher permeability of the blood-brain barrier
94
(BBB) with aging may lead to leakage of iron, which could contribute to elevated iron
95
content in older age (Mesquita et al., 2012; Marques et al., 2013). When iron
96
concentration passes a certain threshold, its accumulation becomes deleterious for
97
brain cells by inducing oxidative stress via the Fenton reaction (Winterbourn, 1995;
98
Zecca et al., 2004). In normal aging, accumulation of iron in the striatum triggers
99
striatal atrophy over time (Daugherty et al. 2015; Daugherty and Raz 2016). Beyond
100
this relationship between iron and volume, we have recently shown a link between
101
striatal iron and brain activity during a motor imagery task, where older individuals
102
with more striatal iron had reduced activity in frontostriatal regions (Kalpouzos et al.
103
2017). We hypothesized that iron content could directly impact brain activity via
104
astrocytic dysfunction: the astrocytes, cells where iron concentration increases with
105
aging, are involved in the neurovascular coupling on which the blood oxygen level-
106
dependent (BOLD) signal measured with functional MRI (fMRI) relies (Connor et al.
107
1990; Koehler et al., 2009; Kowianski et al., 2013; Hillman, 2014; Ward et al. 2014).
108
Spontaneous fluctuations consume a substantial portion of the brain’s metabolism via
109
cerebral blood flow (Raichle and Mintun, 2006) and, as such, it is reasonable to
110
hypothesize that individual differences in brain iron might be linked to differences in
111
intrinsic functional connectivity.
TE
D
M
AN
U
SC
RI
PT
91
We aimed to extend the relationship between iron and task-related brain
113
activity to system-level intrinsic connectivity at rest. Iron content was quantified
114
using relaxometry R2*, a reliable marker of iron in ferritin (Langkammer et al.,
115
2010). This technique is highly correlated with measurements of iron taken from
116
postmortem brain tissue (Langkammer et al., 2010; Deistung et al., 2013) and is used
117
to study age-related differences in brain iron in vivo (Cherubini et al., 2009; Haacke et
118
al., 2010). Since no previous study has investigated the link between iron and resting-
119
state connectivity, we first explored possible relationships between striatal iron and
120
different RSNs identified by independent component analysis [ICA; (Calhoun et al.,
121
2001)]. Alterations in striatal connectivity have been demonstrated in disorders
122
affecting the motor system, such as Parkinson’s disease (Helmich et al., 2009), but
123
also in normal aging (Ystad et al. 2011). Thus, we predicted that higher striatal iron
124
content would be related to lower local functional coherence within the striatum
AC
C
EP
112
ACCEPTED MANUSCRIPT
which, in turn, would be related to less efficient motor processing. Moreover, given
126
that the striatum is structurally connected to virtually all cortical areas of the brain
127
(Postuma and Dagher 2006), the degree of striatal-cortical connectivity could also be
128
modulated by striatal iron. Such possible associations within the striatum and/or
129
between striatal and cortical regions would potentially be stronger in older individuals
130
(Kalpouzos et al., 2017).
RI
PT
125
131
2. Materials and Methods
The study was approved by the Regional Ethical Review Board in Stockholm.
133
134
All participants signed an informed consent form prior to data collection.
SC
132
135
136
2.1. Participants
Forty-two (25 younger and 17 older) adults were included in this study (Table
138
1). All participants were right-handed, did not report any previous or current
139
neurological or psychiatric diseases, and none was taking psychoactive medication.
140
The educational level was slightly lower in the older group than in the younger group
141
(p < 0.05). The two groups did not differ in vocabulary (p = .78) (Dureman, 1960).
142
All participants had a score ≥ 23 at the Montreal Cognitive Assessment (MoCA)
143
(Nasreddine et al. 2005), which corresponds to a score of 28 on the Mini-Mental State
144
Examination (MMSE) (Roalf et al., 2013). Younger subjects had higher MoCA scores
145
than older subjects (p = 0.03).
TE
D
M
AN
U
137
Table 1. Participants’ characteristics
Younger group
Older group
N (Women)
25 (13)
17 (8)
Age (Mean ± SD)
36.2 ± 4.4
70.1 ± 3.1 a
Age range
26 - 42
65 - 77
Educationb (max = 3)
2.8 ± 0.4
2.4 ± 0.9 a
MoCA (max = 30)
28 ± 1.6
26.8 ± 1.9 a
Vocabulary (max = 30)
24.4 ± 2.8
24.7 ± 4.6
AC
C
147
EP
146
148
a
Age-group differences significant at p < 0.05
149
b
Education level was assessed according to the highest degree obtained (1 = lower
150
school certificate; 2 = high school; 3 = university).
ACCEPTED MANUSCRIPT
151
152
2.2. MRI acquisition
Participants were scanned on a Discovery MR750 3T scanner (General
154
Electric, Milwaukee, WI, USA), using an 8-channel phased array receiving coil at the
155
MR Research Center of Karolinska Hospital Solna, Stockholm. T1-weighted 3D IR-
156
SPGR images were obtained with the following parameters: voxel size = 0.94 × 0.94
157
× 1 mm3, repetition time [TR] = 7.908 ms, echo time [TE] = 3.06 ms, inversion time
158
[TI] = 450 ms, field of view [FOV] = 24 cm, no gap, 176 axial slices, and flip angle
159
12°. Additionally, a 3D multi-echo gradient-recalled echo sequence (meGRE) was
160
acquired with the following parameters: voxel size = 0.94 × 0.94 × 1 mm3, TR =
161
37.52 ms, FOV = 24 cm, no gap, 146 axial slices, and flip angle 20°. The first TE was
162
3.74 ms, followed by seven additional ones with a 3.75 ms interval between
163
consecutive echoes. Functional images were acquired with a Gradient Echo EPI
164
sequence: TR = 2200 ms, TE = 30 ms, FOV= 22 cm, acquisition matrix 72x72 and
165
slice thickness 3 mm, in-plane resolution 3 x 3 mm2, total accelerated (R=2) EPI
166
readout duration = 16.4 ms, 47 axial slices acquired in an interleaved bottom/up order,
167
180 volumes, and flip angle 70°. To avoid signals arising from progressive saturation,
168
the first 5 scans were discarded from this sequence. The functional data were acquired
169
during resting-state (6 min and 30 seconds). Subjects were instructed to stare at a
170
fixation cross and think freely.
TE
D
M
AN
U
SC
RI
PT
153
171
2.3. Preprocessing of the MRI data
EP
172
173
2.3.1. Relaxometry and extraction from regions of interest: Transverse
175
relaxation rates R2* were estimated by fitting a monoexponential model to the square
176
of the signal at each voxel of the meGRE sequence (Deistung et al. 2013):
177
AC
C
174
S2 = S02 · exp (-2 · TE · R2*),
178
where S is the measured signal magnitude, TE is the echo time, and S0 is the signal
179
amplitude corresponding to TE=0. These analyses were performed with in-house
180
software written in Python (www.python.org).
Average regional R2* values were extracted from the segmented subcortical
181
182
nuclei
performed
with
the
Freesurfer
5.3
image
analysis
suite
183
(http://surfer.nmr.mgh.harvard.edu/) (Fischl et al., 2002), after removing 15% of the
184
most extreme values to avoid influence of high signal from neighboring vessels and
ACCEPTED MANUSCRIPT
185
obtain more robust estimates (for details of the procedure see Kalpouzos et al., 2017).
186
Striatal iron content as reported in the present study is the average of the left and right
187
caudate and putamen.
R2* relaxometry is sensitive to myelin and, thus, may overestimate iron
189
concentration values. An alternative phase-based method, which overcomes this bias
190
and provides more sensitive and specific estimates of iron is quantitative
191
susceptibility mapping (QSM; Daugherty & Raz, 2015). QSM estimations were
192
carried out for all but 3 participants, for whom phase data were corrupted. Given the
193
already modest sample size, our main findings are based on R2* estimates, whereas
194
results using QSM are presented as complementary information. Data processing for
195
QSM estimates have already been described in detail in a previous study (Garzón et
196
al., 2017).
SC
RI
PT
188
M
AN
U
197
2.3.2 Resting-state fMRI: Functional images were preprocessed using the Data
199
Processing Assistant for Resting-State fMRI (DPARSFA) (Chao-Gan and Yu-Zeng,
200
2010), which is based on the Statistical Parametric Mapping Software (SPM;
201
Wellcome Department of Imaging Science, Functional Imaging Laboratory,
202
University College London). The functional images were first corrected for
203
differences in slice-time acquisition within each volume and motion corrected. This
204
was followed by a within-subject rigid registration in order to align functional and
205
structural images. Next, motion was regressed out using the Friston 24-parameters
206
model (Yan et al., 2013). Finally, the realigned nuisance-corrected images were non-
207
linearly normalized to a sample specific template created with Diffeormorphic
208
Anatomical Registration using Exponentiated Lie Algebra (DARTEL), affine aligned
209
to the Montreal Neurological Institute (MNI) template, and smoothed using a 6 mm
210
full-width at half-maximum (FWHM) Gaussian filter.
212
EP
AC
C
211
TE
D
198
2.4. Group ICA and ICA-driven measures
213
Independent Component Analysis (ICA) was applied to the resting-state
214
images using the group ICA fMRI toolbox (GIFT v2.0a; Calhoun et al., 2001; Allen
215
et al., 2011). ICA is a multivariate data-driven approach that identifies independent
216
spatial maps (SMs) and respective time courses (TCs) from the fMRI signal. First, the
217
time-series in each voxel were intensity-normalized in order to improve accuracy and
218
test-retest reliability (Allen et al., 2011). The resulting volumes were then
ACCEPTED MANUSCRIPT
concatenated across time. The optimal number of independent sources was estimated
220
by minimum description length criteria (MDL), resulting in a total of 30 independent
221
components (ICs) (Calhoun et al., 2001). A two-step data reduction was performed
222
with Principal Component Analysis (PCA): a subject-specific data reduction step was
223
carried out in order to decrease computational complexity while preserving most of
224
the information in the data, followed by a group-level data reduction step according to
225
the estimated number of ICs. After data reduction, ICA was performed using the
226
Infomax algorithm to optimally extract the 30 identified ICs (Allen et al., 2011).
227
Finally, a back-reconstruction was carried out, and SMs and respective TCs were
228
computed for each subject using GICA3 (Erhardt et al., 2011).
SC
RI
PT
219
After visually inspecting all 30 ICs and comparing their topology to those of
230
previous studies, 20 were considered to represent relevant RSNs (Supplementary A).
231
These networks exhibited spatial overlap with RSNs identified in previous studies
232
(Biswal et al., 2010; Allen et al., 2011), showed peak activation in the gray-matter,
233
and had little to no overlap with ICs known to reflect vascular, ventricles, motion, and
234
susceptibility artifacts. Four ICA-driven measures, reflecting different aspects of
235
within- or between-network connectivity, were computed for each subject. Within-
236
network functional connectivity was quantified by computing a measure of coherence
237
and a voxel-wise measure of local connectivity strength.
TE
D
M
AN
U
229
First, power spectra were estimated on the detrended and despiked subject-
239
specific TCs using the multi-taper approach as implemented in Chronux
240
(http://chronux.org; number of tapers = 5; time-bandwidth = 3; 129 frequency bins).
241
Power spectra reflect coherence of functional connectivity within a network with 129
242
frequency bins [0-0.22 Hz]. To provide a more aggregated subject-specific measure,
243
we also computed the average power spectra across low-frequency bands (0.009-0.1),
244
which have been previously suggested to contain the most relevant signal in
245
spontaneous activity (van Dijk et al., 2010).
AC
C
246
EP
238
Second, the intensities in each component SM were used to compute a voxel-
247
wise connectivity strength measure (Calhoun et al., 2001). This was done using a
248
back-reconstruction with GICA3, an improved version of dual regression (Erhardt et
249
al., 2011). Thus, both coherence and intensity are measures of within-network
250
connectivity, with the former being based on each component’s TCs, and the latter
251
using each component’s SM.
ACCEPTED MANUSCRIPT
To determine the between-network connectivity of striatal components to
253
other RSNs, we used a graph theoretical approach. The ICA-derived RSNs (20
254
components) were taken as separate nodes of a graph, where functional connectivity
255
between ICs (Fisher’s z-transformed Pearson correlation coefficients) represents
256
edges of the weighted graph. Degree centrality and strength were computed for each
257
striatal component and each participant separately using functions implemented in the
258
brain connectivity toolbox [(Rubinov and Sporns, 2010), www.brain-connectivity-
259
toolbox.net]. The measure of degree centrality was defined as the number of ICs
260
positively connected to each striatal network. Strength was defined as the sum of
261
positive weights from edges connected to each striatal network and represents how
262
strongly connected the network of interest (i.e., striatum) is to all other networks
263
identified with ICA.
SC
RI
PT
252
265
M
AN
U
264
2.5. Motor performance: Purdue Pegboard task
The Purdue Pegboard task (Tiffin and Asher, 1948) is divided into four sub-
267
sections and measures both unilateral and bilateral fine manual dexterity. It includes a
268
board with two parallel vertical rows, each with 25 peg-holes, and compartments with
269
three types of objects at the top (pins, collars, and washers). In the first test, one has to
270
place the maximum number of pins within 30 seconds using the dominant hand; in the
271
second and third test, the task is the same but one has to use the other hand and both
272
hands at the same time, respectively. The last section requires the subject to use
273
alternate hands to make assemblies of 4 objects (pin – washer – collar – washer)
274
within 60 seconds. This yields four separate scores: 1 and 2) number of pins inserted,
275
3) number of pairs of pins, and 4) number of assemblies. The task was chosen given
276
the high test-retest reliability reported in the literature (Desrosiers et al., 1995;
277
Buddenberg et al., 2000), and the fact that different subtests may reflect different
278
levels of motor and cognitive speed, and coordination. Although the task is most
279
commonly known for simple motor skills, performing it with the left non-dominant
280
hand may be more attentionally demanding (Spreen and Strauss, 1998; Strenge et al.,
281
2002; Cowan et al., 2009). Moreover, while studies have shown that the Purdue
282
Pegboard Test with the dominant hand relies primarily on finger dexterity (Fleishman
283
and Ellison, 1962), the assembly subtest seems to additionally measure arm-hand
284
control and attention (Spreen and Strauss, 1998; Strenge et al., 2002).
285
AC
C
EP
TE
D
266
ACCEPTED MANUSCRIPT
286
2.6. Statistical analyses
287
2.6.1. Age differences in striatal iron content: In order to track possible
289
outliers, the Outlier Labeling Rule was applied (Tukey, 1977). This was done by
290
computing the upper and lower limits using the following formula: Q3 + (2.2 * (Q3 –
291
Q1)) and Q1 – (2.2 * (Q3-Q1)), where Q1 is the lower quartile (25th percentile of the
292
data), Q3 is the upper quartile (75th percentile), and 2.2 is the g-value (Hoaglin and
293
Iglewicz, 1987). After this, we used a two-sample t-test to determine whether there
294
was an age difference in striatal iron content.
SC
295
RI
PT
288
2.6.2. Resting-state networks and iron content
297
2.6.2.1. Striatal within-network connectivity. We applied a multivariate model
298
selection strategy to identify possible covariates and reduce the number of subsequent
299
univariate tests, thus decreasing the risk of spurious findings. This method is similar
300
to a standard analysis of variance (ANOVA) followed by pairwise comparisons, but
301
instead uses a multivariate analysis of covariance (MANCOVA), testing if each of the
302
predictors explains variability (Allen et al., 2011). Our design matrix included age,
303
iron, as well as an age x iron interaction as covariates of interest. In addition, sex was
304
included as a covariate of no interest. We included connectivity strength and power
305
spectra in each network as dependent variables.
TE
D
M
AN
U
296
Based on the results of the multivariate analysis, we restricted the analyses to
307
power spectra and independent components corresponding to the caudate and
308
putamen. These analyses were followed by partial correlations between striatal iron
309
and power spectra for both networks, while controlling for age and sex. This was
310
done for the entire sample, followed by within-group analyses.
AC
C
311
EP
306
2.6.2.2. Between-network connectivity. As the striatum is well-connected to
312
most cortical regions, we had also hypothesized that iron could influence striatal-
313
cortical functional connectivity. As such, we investigated whether striatal iron
314
modulated connectivity between the caudate and putamen networks to the rest of the
315
brain. This was done by performing partial correlations between striatal iron content
316
and the two between-network connectivity measures (degree centrality and strength)
317
for the caudate and putamen. All analyses were carried out for the whole group and
318
within age groups, and were controlled for age and sex. When necessary, a
319
comparison between correlations was conducted using the Steiger’s z-test.
ACCEPTED MANUSCRIPT
320
2.6.2.2. Striatal resting-state networks, iron content, and motor performance:
322
Given the well-established link between putamen and motor function (Mattay et al.,
323
2002; Leh et al., 2007; Di Martino et al., 2008), we tested for correlations between
324
coherence in the putamen RSN and the Purdue Pegboard motor task scores. We
325
expected that the putamen would correlate with all subtests but more strongly with the
326
dominant (right) hand given that it reflects the most pure motor dexterity measure.
327
Finally, we also tested for possible correlations between the caudate RSN and motor
328
performance. All analyses were controlled for age and sex in the entire sample and
329
within age groups.
330
3. Results
332
3.1. Age differences in striatal iron content
M
AN
U
331
SC
RI
PT
321
The Outlier Labeling Rule (see Materials and Methods) showed that none of
334
the subjects should be excluded from the analyses based on iron content. One subject
335
collided with the upper-limit, but was not an outlier in the connectivity measures and
336
was also the oldest subject in the sample, and thus likely to have the highest iron
337
content. The two-sample t-test indicated that iron content (R2* rate) in the striatum
338
was significantly higher in the older than in the younger group (t = -3.807, p = 0.001).
339
340
TE
D
333
3.2. Mapping resting-state networks
The ICA estimated a total of 30 ICs, 20 of which were considered to reflect
342
RSNs. In line with previous reports (Biswal et al. 2010; Allen et al. 2011; Salami et
343
al. 2014; Salami et al. 2016; Avelar-Pereira et al. 2017), these RSNs encompassed the
344
auditory cortex, insula, fronto-parietal regions, cerebellum as well as several visual,
345
sensory-motor, default mode, and attentional networks (Supplementary A). Critically,
346
two striatal components, including the caudate and putamen networks were part of the
347
observed patterns. More specifically, the putamen network encompassed the bilateral
348
putamen, whereas the caudate network included bilateral caudate regions, the anterior
349
cingulate and inferior frontal cortex (Figure 1). Two different ICA-driven measures
350
were computed for the two networks of interest. They consisted of 1) a component
351
average power spectrum, reflecting the level of coherence within a network and 2) the
352
intensities in each component SM, indicating local connectivity strength for every
353
voxel within the network.
AC
C
EP
341
ACCEPTED MANUSCRIPT
354
[Figure 1]
355
356
357
Figure 1. Maps of the caudate and anterior cingulate cortex (in blue) and putamen (in
358
red) networks.
360
3.3. Intrinsic resting-state connectivity and iron content
RI
PT
359
We applied a multivariate model selection strategy to determine the effect of
362
iron on RSNs SMs connectivity strength and coherence. This was done while
363
accounting for variance associated with age and sex. Results from the multivariate
364
analysis are displayed in Figure 2. As expected, we found a significant effect of age
365
and sex on the SMs of several RSNs. Similarly, age and sex also had a significant
366
effect on the level of coherence in some of the networks. There was no effect of
367
striatal iron (R2* rate) on SMs, but a significant association between R2* rate and
368
coherence was found for two components: the caudate and putamen networks. The
369
interaction between iron and age was not significant. There were also no age-group
370
differences in coherence for either the caudate (t (40) = 1.67, p = 0.1) or putamen (t
371
(40) = -0.24, p = 0.8). Follow-up univariate analyses revealed a decrease in low
372
frequency power with increasing iron for both the caudate and putamen (Figure 2A).
373
The effect was primarily found between 0.01 and 0.10 Hz, which is consistent with
374
the frequency range associated with spontaneous fluctuations of RSNs. This suggests
375
that higher iron content was associated with lower functional connectivity as
376
measured using time-course power spectra (Cordes et al., 2001; Sun et al., 2004).
EP
TE
D
M
AN
U
SC
361
Further univariate analyses with average coherence across the RS frequency
378
bins exhibited a significant negative association between striatal iron R2* rate and
379
coherence of the caudate (r = - 0.41, p = 0.008; False Discovery Rate (FDR)
380
corrected: p = 0.040) and putamen (r = -0.32, p = 0.047; FDR corrected: p = 0.078)
381
RSNs across the whole sample (Figure 2B). Within-group analyses revealed a
382
significant association between striatal iron and coherence of the caudate RSN in the
383
older group (r = -0.53, p = 0.04), but not in the younger group (r = -0.24, p = 0.27);
384
however these two correlations were not significantly different (p = 0.32). There were
385
no significant associations for the putamen RSN within age groups (younger: r = -
386
0.11, p = 0.60; older: r = - 0.32, p = 0.24).
AC
C
377
ACCEPTED MANUSCRIPT
We also ran these analyses using QSM, despite having a smaller sample size
388
available and, consequently, reduced statistical power. In line with the literature
389
(Deistung et al., 2013), striatal R2* and QSM were highly correlated in our sample (r
390
= 0.762, p < 0.001). The results using QSM pointed in the same direction as those
391
using R2*. Specifically, the association between QSM and coherence in the caudate (r
392
= -0.398, p = 0.015) remained significant, whereas the association to the putamen (r =
393
-0.301, p = 0.07) was at trend level. Finally, the correlation between QSM and degree
394
centrality in the putamen (r = -0.482, p = 0.002) was also significant.
RI
PT
387
Taken together, higher levels of striatal iron were associated with lower intra-
396
striatal intrinsic connectivity across the whole sample, especially of the caudate
397
network. This association was more pronounced in the older group, most notably for
398
the caudate network.
SC
395
M
AN
U
399
[Figure 2]
400
401
Figure 2. Relation between iron content and resting-state networks: (A1) Effect of
403
iron on SMs connectivity strength and TCs power spectra across various frequency
404
bins for different ICs; (A2) The significant effect of iron on the caudate and putamen
405
RSNs was found primarily within the frequency range associated with spontaneous
406
fluctuations; (B) Negative correlation between striatal iron content (R2* rate) and
407
(B1) coherence of the caudate RSN across the whole sample, (B2) coherence of the
408
putamen RSN across the whole sample.
411
EP
410
3.4. Between-network connectivity and iron content
AC
C
409
TE
D
402
In the previous analyses, we showed that striatal iron affects intra-network
412
coherence in striatal circuits. However, it remains unknown whether iron also impacts
413
connectivity between the striatum and other RSNs. As such, we investigated whether
414
striatal iron modulated the degree centrality and/or strength of the striatal
415
components. The results showed that striatal iron was negatively associated with
416
degree centrality of the putamen across the two groups (r = - 0.42, p = 0.007; FDR
417
corrected: p = 0.02; Figure 3). Within age-group analyses revealed that this negative
418
association was present in the older (r = - 0.55, p = 0.03) but not in the younger (r = -
419
0.3, p = 0.12) group. The correlation between iron in the striatum and strength of the
420
putamen network did not reach conventional significance (r = -0.22, p=0.18). No
ACCEPTED MANUSCRIPT
421
significant association was found between striatal iron with either degree (r = 0.04,
422
p=0.82) or strength (r = 0.03, p = 0.86) of the caudate RSN across the whole group.
Altogether, these findings suggest that higher striatal iron content was
424
associated with a decrease in functional cross-talk between the putamen and other
425
RSNs across both age groups. Moreover, this association was more pronounced in the
426
older group.
RI
PT
423
427
[Figure 3]
428
429
Figure 3. Negative correlation between striatal iron content (R2* rate) and degree
431
centrality of the putamen RSN across the whole sample.
SC
430
432
3.5. Striatal resting-state networks, iron content, and motor performance
M
AN
U
433
Based on well-established links from previous literature, we hypothesized that
435
the caudate and putamen networks would tax executive and motor processes
436
(Middleton and Strick, 2000; Gerardin et al., 2004; Monchi et al., 2006; Postuma and
437
Dagher 2006; Di Martino et al. 2008; Choi et al., 2012; Pauli et al., 2016; Haber,
438
2016). To investigate this, we tested for associations between the caudate and
439
putamen and the Purdue Pegboard task. No significant correlations were found
440
between caudate resting-state measures and any of the tasks (p < 0.05). We found a
441
significant positive association between coherence of the putamen RSN and task
442
performance with the dominant (right) hand (r = 0.45, p = 0.04) across both age
443
groups (Figure 4). However, there were no associations with either the left hand or the
444
assembly task (p > 0.05). Correlations between striatal iron and the motor task at a
445
whole-group level did not survive after controlling for age and sex. However, more
446
exploratory analyses showed that, after performing a median split and looking only at
447
individuals with high iron content in the caudate and putamen compared to others in
448
the same age group, there was a significant association between striatal iron and
449
motor performance with the dominant hand (r = -0.539, p = 0.047) across the sample.
AC
C
EP
TE
D
434
450
451
[Figure 4]
452
453
Figure 4. Positive correlation between coherence of the putamen and task
454
performance with the dominant (right) across age groups.
ACCEPTED MANUSCRIPT
455
456
3.6. Control analyses
Although there was no relationship between striatal R2* rate and striatal
458
volume in our sample (r = 0.2, p = 0.2), prior work on the relationship between striatal
459
iron and cognitive and brain measures indicates that these may be confounded by
460
striatal volume. As such, we performed control analyses adjusting for striatal volume,
461
which revealed very similar results to the ones reported in previous sections
462
[Supplementary B]. A similar analysis was carried out with white-matter
463
hyperintensities (WMH) as a covariate of no interest. This analysis also yielded very
464
similar results to the ones reported above [details in Supplementary C]. Given that
465
decreases in signal-to-noise ratio (SNR) are associated with increases in R2* values,
466
we have taken SNR in the components of interest as a covariate [Supplementary D].
467
Importantly, we also tested for an association between SNR and R2* in the caudate (r
468
= -0.1, p = 0.537) and putamen (r = 0.0021, p = 0.897) but these were not significant.
469
Finally, we were also interested to know whether the relationship between iron R2*
470
and motor performance was mediated by functional connectivity. For this, we have
471
performed a tentative mediation analysis with iron as the predictor variable,
472
functional connectivity as the mediator, and motor performance as the outcome
473
variable [Supplementary E]. We found that functional connectivity mediated the
474
relation between iron and motor performance.
475
4. Discussion
EP
476
TE
D
M
AN
U
SC
RI
PT
457
Our main aim was to investigate the relationship between striatal iron content
478
and inter-individual differences in resting-state functional connectivity, and whether
479
age affected this association. We have demonstrated for the first time that striatal iron
480
was associated with functional connectivity in striatal resting-state networks. More
481
specifically, higher iron content correlated with lower coherence within the caudate
482
and putamen networks. Our results also indicated that iron modulated connectivity of
483
striatal networks to the rest of the brain. In general, these associations seemed to be
484
more pronounced in the older group, however this should be interpreted with caution
485
given that there were no group differences in degree of correlation between iron and
486
functional connectivity. Finally, we showed that functional connectivity features
487
affected by iron are behaviorally meaningful, so that the level of coherence within the
488
putamen correlated with performance in a motor task. Overall, the present work
AC
C
477
ACCEPTED MANUSCRIPT
makes a contribution to the literature on iron accumulation and its relation to
490
functional brain integrity and cognition. Currently, little is known about the factors
491
that modulate brain iron levels, including those that are modifiable (e.g., life style; for
492
a review see Kalpouzos, accepted for publication). Investigating these factors may
493
help preventing or slowing down age-related iron accumulation and, consequently,
494
preserving brain integrity.
RI
PT
489
Higher levels of iron in the striatum were associated with lower resting-state
496
coherence within the caudate and putamen networks. This association was more
497
pronounced in the older group, especially for the caudate, suggesting that increased
498
iron accumulation has a greater impact on striatal functional architecture in older age.
499
As no study has previously investigated associations between brain iron and
500
functional connectivity, the nature of the relationships we describe in this study and
501
their underlying mechanisms are still poorly understood. Previous work suggests that
502
iron interferes with the BOLD signal (Kalpouzos et al., 2017), which is itself related
503
to brain metabolism. Moreover, neurophysiological studies have provided support for
504
a tight coupling between coherent spontaneous activity and brain metabolism (Liang
505
et al., 2013; Jann et al., 2015; Salami et al., 2016; Bernier et al., 2017). This
506
relationship might be reduced with advancing age and thus, it is possible that iron
507
impacts spontaneous activity via alterations in brain metabolism. Finally, iron
508
increases in the striatum might be threshold-dependent and become detrimental in old
509
age only, after iron accumulation reached a specified level. Nonetheless, longitudinal
510
studies with other neuroimaging measures (e.g., perfusion and spectroscopy) are
511
needed to shed light on the exact mechanisms underlying the association between iron
512
and resting-state connectivity.
M
AN
U
TE
D
EP
AC
C
513
SC
495
The lack of association between iron and SMs indicates that the relationship
514
between iron and functional connectivity is perhaps more sensitive to oscillations in
515
activity than to local strength within a network, that is, the shape of a network (Allen
516
et al., 2012). This finding may suggest that the effect of iron on coherence is perhaps
517
more lasting than its effect on SMs. This hypothesis finds support from a study that
518
showed that alcohol and substance use were associated with widespread differences in
519
resting-state power spectra, with no impact on SMs (Thijssen et al., 2017). Moreover,
520
a number of previous studies have shown that the Amplitude of Low Frequency
521
Fluctuation (ALFF), which corresponds to the square root of the power spectra, is a
522
more sensitive measure compared to its corresponding conventional functional
ACCEPTED MANUSCRIPT
connectivity (Zuo et al., 2010; Wang et al., 2012; Tadayonnejad, 2015). The reason
524
why, in our study, power spectrum seems more sensitive to iron is unclear. SM
525
intensities correspond to the shape of a given network, so it is plausible that a
526
substantial effect of iron would be necessary to elicit changes in this measure. Since
527
the effect of iron in functional connectivity is perhaps less pronounced, it might first
528
impact coherence in regional time-series, which, in turn, could progressively
529
deteriorate networks’ shape.
RI
PT
523
The second aim of our study was to investigate whether iron in the striatum
531
modulated the relationship between striatal resting-state networks to the rest of the
532
brain. We were specifically interested in whether the influence of iron was restricted
533
to connectivity within the striatum, or if it was also related to connectivity with other
534
brain regions. We measured both strength and degree centrality, which represent 1)
535
how much the rest of the brain communicates with the caudate and putamen and 2)
536
how many other networks communicate with the caudate and putamen. Degree
537
centrality is usually measured in graph theory when defining hubs, so that a region
538
that has many possible connections with other regions will be a potential hub
539
(Bullmore and Sporns, 2009; Zuo et al., 2011; Kaboodvand et al., 2018). Given that
540
the striatum is virtually connected to most of the brain, it is relevant in this context to
541
not only investigate if the level of connectivity between striatal networks to other
542
brain networks is influenced by striatal iron, but to also measure the number of
543
networks possibly affected. The correlation between iron and strength of the putamen
544
was negative but did not reach statistical significance. However, our results showed
545
that iron was negatively associated with degree centrality of the putamen across the
546
sample, and that this effect was mostly driven by the older group. This indicates that
547
higher levels of iron in the striatum were associated with less communication between
548
the putamen and the rest of the brain, and implies that striatal iron accumulation might
549
have a broad influence on striatal dynamics. If the effects of iron are indeed threshold-
550
dependent, it is understandable that these disruptions are more pronounced for the
551
putamen (Schenck and Zimmerman, 2004; Pfefferbaum et al., 2009; Ghadery et al.,
552
2015). Post-hoc analyses showed that subjects have less iron in the caudate compared
553
to the putamen, even after controlling for corresponding brain volume in these
554
regions. This suggests that, although coherence within both the putamen and caudate
555
networks might be influenced by iron accumulation, spillover effects to other brain
556
regions may occur for the putamen, whereas connectivity between the caudate and the
AC
C
EP
TE
D
M
AN
U
SC
530
ACCEPTED MANUSCRIPT
557
rest of the brain could remain relatively intact. Consistent with this argument, a
558
previous resting-state study demonstrated that the degree of connectivity of the
559
putamen (but not the caudate) with both task-positive (e.g., motor) and task-negative
560
(e.g., DMN) networks decreased with advancing age (Manza et al., 2015).
Given a well established link between motor functions and the striatum in
562
general, and putamen in particular (Mattay et al., 2002; Leh et al., 2007; Di Martino et
563
al. 2008), we predicted that an iron overload would be associated to lower coherence
564
within the striatum and, consequently, to worse motor performance. The results
565
showed a positive correlation between coherence within the putamen network and
566
motor performance in the Purdue Pegboard task using the dominant hand. This is in
567
line with studies that have shown a link between connectivity in the striatum and
568
motor performance (Wu et al., 2010; Wu et al., 2015a; Wu et al., 2015b; Manza et al.,
569
2016). A follow-up mediation analysis suggested that coherence in the putamen might
570
mediate the association between iron and motor performance. This is in line with a
571
previous study where functional connectivity was also found to mediate the link
572
between cognition and a different molecular measure, dopamine (Nyberg et al., 2016).
573
No associations were found between the caudate network and task performance.
574
Although the association between coherence in the putamen and performance should
575
be interpreted with caution due to a lack of control for multiple comparisons, it is
576
important to note that the Purdue Pegboard task is primarily motor-dependent and
577
reflects both motor and cognitive speed as well as coordination. The assembly test is
578
perhaps more attentionally demanding but its focus is still on fine motor skills and
579
action planning. Thus, it might not be ideal for the study of executive functions. In the
580
future, other tasks tapping executive functions should be used. Previous work has also
581
shown that higher levels of iron in the basal ganglia were negatively associated to
582
both executive and psychomotor functioning (Ghadery et al., 2015). Although we
583
failed to see any direct associations between iron and task performance, post-hoc
584
analyses showed that, for individuals with high iron content in the caudate and
585
putamen, there was a negative association between iron and task performance. There
586
is evidence suggesting that manual dexterity may be more sensitive to iron content
587
than executive functions (Li et al., 2015), which could explain why we only found an
588
association when the task was performed with the dominant hand since that would
589
represent the closest measure to optimal manual dexterity. There is also evidence
590
suggesting that the relation between iron and cognitive functioning is more evident
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
561
ACCEPTED MANUSCRIPT
(or perhaps stronger) in the putamen or pallidum, whereas relationships between iron
592
in other brain regions, including the caudate, and cognition show less consistent
593
patterns (Ghadery et al., 2015). It is also plausible that iron accumulation in the
594
caudate did not reach an extent that would damage functional connectivity and,
595
consequently, lead to cognitive impairments (Li et al., 2015). Future studies should
596
investigate these links using different measures of motor and executive functions and
597
larger sample sizes.
RI
PT
591
There are some limitations to our work. First, although R2* has been widely
599
validated as an in vivo measure of iron in the brain (Langkammer et al., 2010), it is
600
also susceptible to myelin which might have biased our findings. Results using QSM
601
were similar to those reported with R2*, although there was one association that did
602
not reach statistical significance. This may be due to reduced statistical power given
603
that QSM could not be computed for the whole sample. Second, current estimates of
604
iron cannot differentiate between non-heme and heme iron even though we are only
605
interested in the former. It is reasonable to assume that R2* mostly detects ferritin and
606
thus measures iron content. However, the contribution of circulating blood in living
607
tissues remains a concern. Given that we only found associations with striatal
608
networks (not with other networks where cerebral blood flow [CBF] is also a
609
confounding factor), it seems unlikely that blood perfusion is driving the main
610
findings present in this study but we cannot rule out its influence. Future studies,
611
where CBF is also collected, should use this measure as a potential covariate. Another
612
limitation of our study is its small sample size and cross-sectional nature, which does
613
not allow us to investigate changes in the relationship between iron content and
614
functional connectivity. Moreover, cross-sectional estimates of age-related changes
615
may deviate from those found in longitudinal studies (Nyberg et al., 2010).
M
AN
U
TE
D
EP
AC
C
616
SC
598
In summary, iron in the striatum affects functional coherence in striatal
617
networks as well as connectivity with distant brain areas. Moreover, these effects are
618
more pronounced in older adults. We have also shown that functional coherence in
619
the putamen network is related to motor dysfunction. Nonetheless, in this study, we
620
cannot establish causal relations between iron content and functional connectivity or
621
motor performance. Our results shed some light into the functional consequences of
622
iron accumulation in normal aging, but future longitudinal studies with a larger
623
sample size and an extensive cognitive test battery are needed to understand possible
ACCEPTED MANUSCRIPT
624
effects of brain iron overload on age-related within- and between-network disruptions
625
associated with motor and cognitive decline.
626
Acknowledgements
628
This work was supported by grants from Loo and Hans Ostermans Foundation for
629
Medical Research (G.K.), Karolinska Institutet’s Research Foundation (G.K.),
630
Foundation for Geriatric Diseases at Karolinska Institutet (G.K.), Gun and Bertil
631
Stohne’s Foundation (G.K.), Stiftelsen för Gamla Tjänarinnor (G.K.), the Swedish
632
Research Council (grant number 421-2014-940, G.K.), and the Karolinska Institutet
633
doctoral (KID) grant. We want to thank Carmel Heiland for assistance with data
634
collection and the participants for their contribution.
TE
D
EP
AC
C
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
M
AN
U
635
SC
RI
PT
627
ACCEPTED MANUSCRIPT
References
RI
PT
Allen, E. A., Erhardt, E. B., Damaraju, E., Gruner, W., Segall, J. M., Silva, R. F., ... &
Michael, A. M., 2011. A baseline for the multivariate comparison of resting-state
networks. Front Syst Neurosci, 5.
SC
Allen, E. A., Erhardt, E. B., Wei, Y., Eichele, T., & Calhoun, V. D., 2012. Capturing
inter-subject variability with group independent component analysis of fMRI data: a
simulation study. NeuroImage, 59(4), 4141-4159.
M
AN
U
Andrews-Hanna, J. R., Snyder, A. Z., Vincent, J. L., Lustig, C., Head, D., Raichle, M.
E., & Buckner, R. L., 2007. Disruption of large-scale brain systems in advanced
aging. Neuron, 56(5), 924-935.
Avelar-Pereira, B., Bäckman, L., Wåhlin, A., Nyberg, L., & Salami, A., 2017. Agerelated differences in dynamic interactions among default mode, frontoparietal
control, and dorsal attention networks during resting-state and interference
resolution. Front Aging Neurosci, 9.
TE
D
Bernier, M., Croteau, E., Castellano, C. A., Cunnane, S. C., & Whittingstall, K., 2017.
Spatial distribution of resting-state BOLD regional homogeneity as a predictor of
brain glucose uptake: a study in healthy aging. NeuroImage, 150, 14-22.
EP
Betzel, R. F., Byrge, L., He, Y., Goñi, J., Zuo, X. N., & Sporns, O., 2014. Changes in
structural and functional connectivity among resting-state networks across the human
lifespan. NeuroImage, 102, 345-357.
Biswal, B. B., Mennes, M., Zuo, X. N., Gohel, S., Kelly, C., Smith, S. M., ... &
Milham, M. P., 2010. Toward discovery science of human brain function. Proc Natl
Acad Sci U S A, 107(10), 4734-4739.
AC
C
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
Buckner, R. L., Krienen, F. M., & Yeo, B. T., 2013. Opportunities and limitations of
intrinsic functional connectivity MRI. Nat Neurosci, 16(7), 832-837.
Buddenberg, L. A., & Davis, C., 2000. Test–retest reliability of the Purdue Pegboard
Test. Am J Occup Ther, 54(5), 555-558.
Bullmore, E., & Sporns, O., 2009. Complex brain networks: graph theoretical analysis
of structural and functional systems. Nature Rev Neurosci, 10(3), 186-198.
ACCEPTED MANUSCRIPT
Calhoun, V. D., Adali, T., Pearlson, G. D., & Pekar, J. J., 2001. A method for making
group inferences from functional MRI data using independent component
analysis. Hum Brain Mapp, 14(3), 140-151.
Campbell, K. L., & Schacter, D. L., 2017. Ageing and the resting state: is cognition
obsolete?. Lang Cogn Neurosci, 32(6), 661-668.
RI
PT
Chan, M. Y., Park, D. C., Savalia, N. K., Petersen, S. E., & Wig, G. S., 2014.
Decreased segregation of brain systems across the healthy adult lifespan. Proc Natl
Acad Sci U S A, 111(46), E4997-E5006.
SC
Chao-Gan, Y., & Yu-Feng, Z., 2010. DPARSF: a MATLAB toolbox for “pipeline”
data analysis of resting-state fMRI. Front Syst Neurosci, 4.
M
AN
U
Cherubini, A., Péran, P., Caltagirone, C., Sabatini, U., & Spalletta, G., 2009. Aging of
subcortical nuclei: microstructural, mineralization and atrophy modifications
measured in vivo using MRI. NeuroImage, 48(1), 29-36.
Choi, E. Y., Yeo, B. T., & Buckner, R. L., 2012. The organization of the human
striatum estimated by intrinsic functional connectivity. J Neurophysiol, 108(8), 22422263.
TE
D
Connor, J. R., Menzies, S. L., St Martin, S. M., & Mufson, E. J., 1990. Cellular
distribution of transferrin, ferritin, and iron in normal and aged human brains. J
Neurosci Res, 27(4), 595-611.
EP
Cordes, D., Haughton, V. M., Arfanakis, K., Carew, J. D., Turski, P. A., Moritz, C.
H., ... & Meyerand, M. E., 2001. Frequencies contributing to functional connectivity
in the cerebral cortex in “resting-state” data. Am J Neuroradiol, 22(7), 1326-1333.
Cowan, D. M., Zheng, W., Zou, Y., Shi, X., Chen, J., Rosenthal, F. S., & Fan, Q.,
2009. Manganese exposure among smelting workers: relationship between blood
manganese–iron ratio and early onset neurobehavioral alterations. Neurotoxicology,
30(6), 1214-1222.
AC
C
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
Daugherty, A. M., Haacke, E. M., & Raz, N., 2015. Striatal iron content predicts its
shrinkage and changes in verbal working memory after two years in healthy adults. J
Neurosci, 35(17), 6731-6743.
Daugherty, A. M., & Raz, N., 2015. Appraising the role of iron in brain aging and
cognition: promises and limitations of MRI methods. Neuropsychol Rev, 25(3), 272287.
ACCEPTED MANUSCRIPT
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
Daugherty, A. M., & Raz, N., 2016. Accumulation of iron in the putamen predicts its
shrinkage
in
healthy
older
adults:
a
multi-occasion
longitudinal
study. NeuroImage, 128, 11-20.
770
D., 2011. Comparison of multi‐subject ICA methods for analysis of fMRI data. Hum
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
Brain Mapp, 32(12), 2075-2095.
RI
PT
Deistung, A., Schäfer, A., Schweser, F., Biedermann, U., Turner, R., & Reichenbach,
J. R., 2013. Toward in vivo histology: a comparison of quantitative susceptibility
mapping (QSM) with magnitude-, phase-, and R2*-imaging at ultra-high magnetic
field strength. NeuroImage, 65, 299-314.
Desrosiers, J., Hebert, R., Bravo, G., & Dutil, E., 1995. The Purdue Pegboard Test:
normative data for people aged 60 and over. Disabil Rehabil, 17(5), 217-224.
M
AN
U
SC
Di Martino, A., Scheres, A., Margulies, D. S., Kelly, A. M. C., Uddin, L. Q., Shehzad,
Z., ... & Milham, M. P., 2008. Functional connectivity of human striatum: a resting
state FMRI study. Cereb Cortex, 18(12), 2735-2747.
Dureman, I., 1960. SRB:1. Stockholm, Sweden: Psykologiförlaget.
Erhardt, E. B., Rachakonda, S., Bedrick, E. J., Allen, E. A., Adali, T., & Calhoun, V.
TE
D
Ferreira, L. K., & Busatto, G. F., 2013. Resting-state functional connectivity in
normal brain aging. Neurosci Biobehav Rev, 37(3), 384-400.
EP
Fischl, B., Salat, D. H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C., ... &
Dale, A. M., 2002. Whole brain segmentation: automated labeling of neuroanatomical
structures in the human brain. Neuron, 33(3), 341-355.
AC
C
Fjell, A. M., Sneve, M. H., Storsve, A. B., Grydeland, H., Yendiki, A., & Walhovd,
K. B., 2015. Brain events underlying episodic memory changes in aging: a
longitudinal investigation of structural and functional connectivity. Cereb
Cortex, 26(3), 1272-1286.
Fleishman, E. A., & Ellison, G. D., 1962. A factor analysis of fine manipulative
tests. J Appl Psychol, 46(2), 96.
Fox, M. D., & Greicius, M., 2010. Clinical applications of resting state functional
connectivity. Front Syst Neurosci, 4.
Garzón, B., Sitnikov, R., Bäckman, L., & Kalpouzos, G., 2017). Automated
segmentation of midbrain structures with high iron content. NeuroImage.
ACCEPTED MANUSCRIPT
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
Geerligs, L., Renken, R. J., Saliasi, E., Maurits, N. M., & Lorist, M. M., 2014. A
brain-wide study of age-related changes in functional connectivity. Cereb
Cortex, 25(7), 1987-1999.
822
Hallgren, B., & Sourander, P., 1958. The effect of age on the non‐haemin iron in the
823
824
825
826
827
828
829
830
831
832
833
834
835
836
human brain. J Neurochem, 3(1), 41-51.
RI
PT
Geerligs, L., & Tsvetanov, K. A., 2017. The use of resting state data in an integrative
approach to studying neurocognitive ageing–commentary on Campbell and Schacter
(2016). Lang Cogn Neurosci, 32(6), 684-691.
Gerardin, E., Pochon, J. B., Poline, J. B., Tremblay, L., Van de Moortele, P. F., Levy,
R., ... & Lehéricy, S., 2004. Distinct striatal regions support movement selection,
preparation and execution. Neuroreport, 15(15), 2327-2331.
M
AN
U
SC
Ghadery, C., Pirpamer, L., Hofer, E., Langkammer, C., Petrovic, K., Loitfelder, M., ...
& Schmidt, R., 2015. R2* mapping for brain iron: associations with cognition in
normal aging. Neurobiol Aging, 36(2), 925-932.
Grady, C., Sarraf, S., Saverino, C., & Campbell, K., 2016. Age differences in the
functional interactions among the default, frontoparietal control, and dorsal attention
networks. Neurobiol Aging, 41, 159-172.
Grady, C., 2016. Cognition is never obsolete: a comment on Campbell and Schacter,
2016. Lang Cogn Neurosci, 32, 681-683.
AC
C
EP
TE
D
Haacke, E. M., Miao, Y., Liu, M., Habib, C. A., Katkuri, Y., Liu, T., ... & Wu, J.,
2010. Correlation of putative iron content as represented by changes in R2* and phase
with age in deep gray matter of healthy adults. J Magn Reson Imaging, 32(3), 561576.
Haber, S. N., 2016. Corticostriatal circuitry. Dialogues Clin Neurosci, 18(1), 7-21.
Hare, D., Ayton, S., Bush, A., & Lei, P., 2013. A delicate balance: iron metabolism
and diseases of the brain. Front Aging Neurosci, 5.
Hect J. L., Daugherty A. M., Hermez K. M., Thomason, M. E., 2018. Developmental
variation in regional brain iron and its relation to cognitive functions in childhood.
Dev Cogn Neurosci, 34: 18-26.
Helmich, R. C., Derikx, L. C., Bakker, M., Scheeringa, R., Bloem, B. R., & Toni, I.,
2009. Spatial remapping of cortico-striatal connectivity in Parkinson's disease. Cereb
Cortex, 20(5), 1175-1186.
ACCEPTED MANUSCRIPT
Hillman, E. M., 2014. Coupling mechanism and significance of the BOLD signal: a
status report. Annu Rev Neurosci, 37, 161-181.
Hoaglin, D. C., & Iglewicz, B., 1987. Fine-tuning some resistant rules for outlier
labeling. J Am Stat Assoc, 82(400), 1147-1149.
RI
PT
Jann, K., Gee, D. G., Kilroy, E., Schwab, S., Smith, R. X., Cannon, T. D., & Wang,
D. J., 2015. Functional connectivity in BOLD and CBF data: similarity and reliability
of resting brain networks. NeuroImage, 106, 111-122.
SC
Kaboodvand N, Bäckman L, Nyberg L, Salami A., 2018. The retrosplenial cortex: A
memory gateway between the cortical default mode network and the medial temporal
lobe. Hum Brain Mapp.
M
AN
U
Kalpouzos G, Brain iron accumulation, and motor and cognitive decline in normal
aging. Revue de Neuropsychologie.
Kalpouzos, G., Garzón, B., Sitnikov, R., Heiland, C., Salami, A., Persson, J., &
Bäckman, L., 2017. Higher Striatal Iron Concentration is Linked to Frontostriatal
Underactivation and Poorer Memory in Normal Aging. Cereb Cortex, 27(6), 34273436.
TE
D
Koehler, R. C., Roman, R. J., & Harder, D. R., 2009 (in press). Astrocytes and the
regulation of cerebral blood flow. Trends Neurosci, 32(3), 160-169.
EP
Kowiański, P., Lietzau, G., Steliga, A., Waśkow, M., & Moryś, J., 2013. The
astrocytic contribution to neurovascular coupling–still more questions than
answers?. J Neurosci Res, 75(3), 171-183.
Langkammer, C., Krebs, N., Goessler, W., Scheurer, E., Ebner, F., Yen, K., ... &
Ropele, S., 2010. Quantitative MR imaging of brain iron: a postmortem validation
study. Radiology, 257(2), 455-462.
AC
C
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
Leh, S. E., Ptito, A., Chakravarty, M. M., & Strafella, A. P., 2007. Fronto-striatal
connections in the human brain: a probabilistic diffusion tractography
study. Neurosci. Lett., 419(2), 113-118.
Li, W., Langkammer, C., Chou, Y. H., Petrovic, K., Schmidt, R., Song, A. W., ... &
Liu, C., 2015. Association between increased magnetic susceptibility of deep gray
matter nuclei and decreased motor function in healthy adults. NeuroImage, 105, 4552.
ACCEPTED MANUSCRIPT
880
881
882
883
884
885
886
887
Liang, X., Zou, Q., He, Y., & Yang, Y., 2013. Coupling of functional connectivity
and regional cerebral blood flow reveals a physiological basis for network hubs of the
human brain. Proc Natl Acad Sci U S A, 110(5), 1929-1934.
888
Manza, P., Zhang, S., Li, C. S. R., & Leung, H. C., 2016. Resting‐state functional
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
connectivity of the striatum in early‐stage Parkinson's disease: Cognitive decline and
motor symptomatology. Hum Brain Mapp, 37(2), 648-662.
RI
PT
Manza, P., Zhang, S., Hu, S., Chao, H. H., Leung, H. C., & Chiang-shan, R. L., 2015.
The effects of age on resting state functional connectivity of the basal ganglia from
young to middle adulthood. NeuroImage, 107, 311-322.
SC
Marques, F., Sousa, J. C., Sousa, N., & Palha, J. A., 2013). Blood–brain-barriers in
aging and in Alzheimer’s disease. Mol Neurodegener, 8(1), 38.
M
AN
U
Mattay, V. S., Fera, F., Tessitore, A., Hariri, A. R., Das, S., Callicott, J. H., &
Weinberger, D. R., 2002. Neurophysiological correlates of age-related changes in
human motor function. Neurology, 58(4), 630-635.
TE
D
Mesquita, S. D., Ferreira, A. C., Sousa, J. C., Santos, N. C., Correia-Neves, M.,
Sousa, N., ... & Marques, F., 2012. Modulation of iron metabolism in aging and in
Alzheimer's disease: relevance of the choroid plexus. Front Cell Neurosci, 6, 25.
Middleton, F. A., & Strick, P. L., 2000. Basal ganglia output and cognition: evidence
from anatomical, behavioral, and clinical studies. Brain Cogn, 42(2), 183-200.
EP
Monchi, O., Petrides, M., Strafella, A. P., Worsley, K. J., & Doyon, J., 2006.
Functional role of the basal ganglia in the planning and execution of actions. Ann
Neurol, 59(2), 257-264.
AC
C
Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V.,
Collin, I., ... & Chertkow, H., 2005. The Montreal Cognitive Assessment, MoCA: a
brief screening tool for mild cognitive impairment. J Am Geriatr Soc, 53(4), 695-699.
Ng, K. K., Lo, J. C., Lim, J. K., Chee, M. W., & Zhou, J., 2016. Reduced functional
segregation between the default mode network and the executive control network in
healthy older adults: A longitudinal study. NeuroImage, 133, 321-330.
Nyberg, L., Karalija, N., Salami, A., Andersson, M., Wåhlin, A., Kaboovand, N., ... &
Bäckman, L., 2016. Dopamine D2 receptor availability is linked to hippocampal–
caudate functional connectivity and episodic memory. Proc Natl Acad Sci U S A,
201606309.
ACCEPTED MANUSCRIPT
Nyberg, L., Salami, A., Andersson, M., Eriksson, J., Kalpouzos, G., Kauppi, K., ... &
Nilsson, L. G. 2010. Longitudinal evidence for diminished frontal cortex function in
aging. Proc Natl Acad Sci U S A, 201012651.
927
Pauli, W. M., O’Reilly, R. C., Yarkoni, T., & Wager, T. D., 2016. Regional
928
specialization within the human striatum for diverse psychological functions. Proc
929
Natl Acad Sci U S A, 113(7), 1907-1912.
SC
Pfefferbaum, A., Adalsteinsson, E., Rohlfing, T., & Sullivan, E. V., 2009. MRI
estimates of brain iron concentration in normal aging: comparison of field-dependent
(FDRI) and phase (SWI) methods. NeuroImage, 47(2), 493-500.
M
AN
U
Postuma, R. B., & Dagher, A., 2006. Basal ganglia functional connectivity based on a
meta-analysis of 126 positron emission tomography and functional magnetic
resonance imaging publications. Cereb Cortex, 16(10), 1508-1521.
Power, J. D., Cohen, A. L., Nelson, S. M., Wig, G. S., Barnes, K. A., Church, J. A., ...
& Petersen, S. E., 2011. Functional network organization of the human
brain. Neuron, 72(4), 665-678.
TE
D
Raichle, M. E., & Mintun, M. A., 2006. Brain work and brain imaging. Annu. Rev.
Neurosci., 29, 449-476.
EP
Roalf, D. R., Moberg, P. J., Xie, S. X., Wolk, D. A., Moelter, S. T., & Arnold, S. E.,
2013. Comparative accuracies of two common screening instruments for
classification of Alzheimer's disease, mild cognitive impairment, and healthy
aging. Alzheimers Dement, 9(5), 529-537.
Rubinov, M., & Sporns, O., 2010. Complex network measures of brain connectivity:
uses and interpretations. NeuroImage, 52(3), 1059-1069.
AC
C
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
RI
PT
923
924
925
926
Salami, A., Pudas, S., & Nyberg, L., 2014. Elevated hippocampal resting-state
connectivity underlies deficient neurocognitive function in aging. Proc Natl Acad Sci
U S A, 111(49), 17654-17659.
Salami, A., Wåhlin, A., Kaboodvand, N., Lundquist, A., & Nyberg, L., 2016.
Longitudinal evidence for dissociation of anterior and posterior MTL resting-state
connectivity in aging: links to perfusion and memory. Cereb Cortex, 26(10), 39533963.
963
Schenck, J. F., & Zimmerman, E. A., 2004. High‐field magnetic resonance imaging of
964
brain iron: birth of a biomarker?. NMR in Biomedicine, 17(7), 433-445.
ACCEPTED MANUSCRIPT
RI
PT
Sun, F. T., Miller, L. M., & D'esposito, M., 2004. Measuring interregional functional
connectivity using coherence and partial coherence analyses of fMRI
data. NeuroImage, 21(2), 647-658.
M
AN
U
SC
Tadayonnejad, R., Yang, S., Kumar, A., & Ajilore, O., 2015. Clinical, cognitive, and
functional connectivity correlations of resting-state intrinsic brain activity alterations
in unmedicated depression. J Affect Disord, 172, 241-250.
Thijssen, S., Rashid, B., Gopal, S., Nyalakanti, P., Calhoun, V. D., & Kiehl, K. A.,
2017. Regular cannabis and alcohol use is associated with resting-state time course
power spectra in incarcerated adolescents. Drug Alcohol Depend, 178, 492-500.
Tiffin, J., & Asher, E. J., 1948. The Purdue Pegboard: norms and studies of reliability
and validity. J Appl Psychol, 32(3), 234.
Tukey, J. W., 1977. Exploratory data analysis. Reading (MA): Addison-Wesley
TE
D
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
Strenge, H., Niederberger, U., & Seelhorst, U., 2002. Correlation between tests of
attention and performance on grooved and Purdue pegboards in normal
subjects. Percept Mot Skills, 95(2), 507-514.
van den Heuvel, M. P., Mandl, R. C., Kahn, R. S., Pol, H., & Hilleke, E., 2009.
Functionally linked resting‐state networks reflect the underlying structural
connectivity architecture of the human brain. Hum Brain Mapp, 30(10), 3127-3141.
EP
991
Spreen, O., & Strauss, E., 1998. A Compendium of Neuropsychological Tests. Oxford
University Press, New York.
van den Heuvel, M. P., & Sporns, O., 2013. Network hubs in the human brain. Trends
Cogn. Sci, 17(12), 683-696.
AC
C
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
van Dijk, K. R., Hedden, T., Venkataraman, A., Evans, K. C., Lazar, S. W., &
Buckner, R. L., 2010. Intrinsic functional connectivity as a tool for human
connectomics: theory, properties, and optimization. J Neurophysiol, 103(1), 297-321.
Wang, L., Dai, W., Su, Y., Wang, G., Tan, Y., Jin, Z., ... & Si, T., 2012. Amplitude of
low-frequency oscillations in first-episode, treatment-naive patients with major
depressive disorder: a resting-state functional MRI study. PloS One, 7(10), e48658.
Ward, R. J., Zucca, F. A., Duyn, J. H., Crichton, R. R., & Zecca, L., 2014. The role of
iron in brain ageing and neurodegenerative disorders. Lancet Neurol, 13(10), 10451060.
ACCEPTED MANUSCRIPT
Winterbourn, C. C., 1995. Toxicity of iron and hydrogen peroxide: the Fenton
reaction. Toxicol Lett, 82, 969-974.
Wu, T., Chan, P., & Hallett, M., 2010. Effective connectivity of neural networks in
automatic movements in Parkinson's disease. NeuroImage, 49(3), 2581-2587.
RI
PT
Wu, T., Hallett, M., & Chan, P., 2015a. Motor automaticity in Parkinson's
disease. Neurobiol Dis, 82, 226-234.
SC
Wu, T., Zhang, J., Hallett, M., Feng, T., Hou, Y., & Chan, P., 2015b. Neural
correlates underlying micrographia in Parkinson’s disease. Brain, 139(1), 144-160.
M
AN
U
Yan, C. G., Cheung, B., Kelly, C., Colcombe, S., Craddock, R. C., Di Martino, A., ...
& Milham, M. P., 2013. A comprehensive assessment of regional variation in the
impact of head micromovements on functional connectomics. NeuroImage, 76, 183201.
Yeo, B. T., Krienen, F. M., Sepulcre, J., Sabuncu, M. R., Lashkari, D., Hollinshead,
M., ... & Buckner, R. L., 2011. The organization of the human cerebral cortex
estimated by intrinsic functional connectivity. J Neurophysiol, 106(3), 1125-1165.
TE
D
Ystad, M., Hodneland, E., Adolfsdottir, S., Haász, J., Lundervold, A. J., Eichele, T.,
& Lundervold, A., 2011. Cortico-striatal connectivity and cognition in normal aging:
a combined DTI and resting state fMRI study. NeuroImage, 55(1), 24-31.
EP
Zecca, L., Youdim, M. B., Riederer, P., Connor, J. R., & Crichton, R. R., 2004. Iron,
brain ageing and neurodegenerative disorders. Nature Rev Neurosci, 5(11), 863.
Zhang, D., & Raichle, M. E., 2010. Disease and the brain's dark energy. Nat Rev
Neurol, 6(1), 15-28.
AC
C
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
Zuo, X. N., Di Martino, A., Kelly, C., Shehzad, Z. E., Gee, D. G., Klein, D. F., ... &
Milham, M. P., 2010. The oscillating brain: complex and reliable. NeuroImage, 49(2),
1432-1445.
Zuo, X. N., Ehmke, R., Mennes, M., Imperati, D., Castellanos, F. X., Sporns, O., &
Milham, M. P., 2011. Network centrality in the human functional connectome. Cereb
Cortex, 22(8), 1862-1875.
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
Документ
Категория
Без категории
Просмотров
1
Размер файла
2 861 Кб
Теги
2018, 036, neuroimage
1/--страниц
Пожаловаться на содержимое документа