Supplementary Discussion
Our analysis focussed on low frequency BOLD signal fluctuations < 0.1 Hz, to reduce contamination by physiological artefacts caused by e.g. vascular pulsation or respiration (Cordes et al., 2001;Auer, 2008). For example, respiration becomes much more regular during deep NREM sleep as compared with wakefulness, which renders it a substantial confound if temporally unfiltered data would be used. In addition, the preprocessing applied in our analysis (residualization with respect to motion, global CSF and WM signal fluctuations) further reduces artefacts induced by physiological noise (Van Dijk et al., 2010) without increasing the occurrence of artificial anti correlations, as observed after correction for global brain signal (Chang and Glover, 2009;Weissenbacher et al., 2009). Another option to consider physiological noise is to model heart rate and respiration directly (Lund et al., 2006), however, no data on respiration were sampled in our study. Restriction to frequencies < 0.1 Hz may shadow contributions of faster frequency bands to the altered hippocampal cortical functional connectivity. EEG slow oscillations (0.5 1.0 Hz) have been shown to play an important role in orchestrating hippocampo neocortical interactions (Diekelmann and Born, 2010;Sirota et al., 2003), and Dang Vu et al. (2008) have shown that these EEG slow oscillations have BOLD signal correlates in the hippocampus. Furthermore, EEG spindle frequencies are in the range of 11 15 Hz, much faster than 0.1 Hz. In our fMRI data, such higher frequencies are suppressed when focussing on signal fluctuations < 0.1 Hz. Our analysis without temporal filtering of the data revealed qualitatively similar results as the one using lowpass filtered data, which may be explained by the fact that not the intrinsic frequency of the specific oscillation is driving the results, but the frequency of the occurrence of the oscillation, which e.g. for sleep spindles is ≤ 0.1 Hz (3 6 spindles per minute). Furthermore, He et al. (2008) showed that throughout the sleep wake cycle, the correlation structure of slow cortical potentials (< 0.5 Hz as well as 1 – 4 Hz) as measured by electrocorticography is similar to the one derived from spontaneous low frequency BOLD signal fluctuations (< 0.1 Hz), showing that fMRI and electrophysiological signals may be linked without sharing the identical frequency range.
1 Supplementary Figures
Supplementary Figure 1: Results of the full factorial design employing factors sleep stage and subregion. F test showing main effects of sleep (A) and subregion (B). Data are
thresholded at pFWE,cluster<0.05 using a collection threshold of p<0.001. Note that the HF has not been masked out in both (A) and (B). MNI coordinates of each slice are given. Color bars depict F values.
2
Supplementary Figure 2: Overlay of cytoarchitectonic probability maps of hippocampal subregions on spatially normalised functional images. Masks for seed extraction of three hippocampal subregions (yellow: dentate gyrus, red: cornu ammonis, green: subiculum) overlaid on the SPM grey matter template (A) and on a study group average of the grey matter compartment (B). The latter image was derived from segmentation of the first (spatially normalised) unequilibrated image of the EPI time series which showed non equilibrium tissue contrast. The spatially normalised images were segmented using the unified segmentation algorithm implemented in SPM5 (Ashburner and Friston, 2005), i.e. the geometry of the resulting grey matter maps reflected exactly the geometry of spatially normalised images used for functional connectivity analysis. Note positioning of the anatomically defined seed regions in the hippocampus proper. Areas intersecting with CSF (i.e. having a probability > 20% of being CSF in the tissue probability maps of SPM) were masked out during individual time course extraction.
3 A B
Supplementary Figure 3: Scalp topography showing the dominance of slow spindles over frontal regions (A) and of fast spindles over centro parietal areas (B). Using the describe algorithm (see Methods), the highest 5% of the root mean square deviation in the frequency band 11 – 13 Hz (at electrode Fz for slow spindles) and 13 – 15 Hz (at electrode Pz for fast spindles), respectively, was calculated in EEG segments of 200 ms. For all recording electrodes, the identified segments of 200 ms each were then Fourier transformed, and topographical maps of the average band power of the segments were created using VisionAnalyzer (version 1.05, Brain Products, Gilching, Germany). 3D mapping of the scalp distribution is shown.
4
Supplementary Figure 4: Comparison of subregional HF functional connectivity contribution within sleep stages. For each sleep stage, strongest contributions of each hippocampal subregion is depicted (yellow: dentate gyrus (DG), red: cornu ammonis (CA), green: subiculum (SUB)). A: wakefulness (W); B: sleep stage 1 (S1); C: sleep stage 2 (S2); D: slow wave sleep (SWS). Data are thresholded at pFWE,cluster<0.05 using a collection threshold of p<0.001 (see Supplementary Table 1). The HF has been masked out to avoid display of autocorrelations. MNI coordinates of each slice are given.
5
Supplementary Figure 5: PPI analysis for all possible interactions spindles x subregion. Subregions are subiculum (SUB, green), cornu ammonis (CA, red) and dentate gyrus (DG, yellow). Data are thresholded at pFWE,cluster<0.05 using a collection threshold of p<0.001. Overlapping activation is identified by mixed color as shown in the upper right corner. MNI coordinates of each slice are given.
6 Supplementary Tables
Supplementary Table 1. Subregional HF functional connectivity per sleep stage.
Peak voxel Brodmann areas, Cluster size Brain Region MNI deep nuclei (voxel) T-value coordinates
Wakefulness
CA
Sub gyral, uncus, fusiform
gyrus, inferior temporal gyrus,
extra nuclear, lentiform
1 R nucleus, claustrum, caudate, 19 21,27,28,34 39 4151 12.30 32 10 20
hippocampus, amygdala,
lateral globus pallidus,
putamen
Uncus, fusiform gyrus, middle
occipital gyrus, culmen, extra
nuclear, lentiform nucleus,
2 L claustrum, caudate, 20,28,34 37 3063 11.91 26 10 20
hippocampus, amygdala,
lateral globus pallidus,
putamen
Inferior/middle temporal 3 L 21, 22 702 5.48 58 24 10 gyrus
Parahippocampal gyrus, 4 L 27 99 5.22 14 38 4 thalamus
5 R putamen 67 4.88 22 4 6
7 6 L Inferior frontal gyrus 47 350 4.85 30 34 12
Medial/lateral
7 L/R Thalamus dorsal nucleus, 196 4.60 12 12 18
midline nucleus
8 L/C/R Anterior cingulate 25 101 4.46 2 8 8
9 R Insula, claustrum 13 132 4.36 40 10 16
10 L/C Posterior cingulate 23,29,30,31 230 4.35 4 52 18
11 R Superior parietal lobule 71 4.31 24 48 58
12 C Midbrain 47 4.31 2 26 18
Anterior cingulate, medial 13 L/R 32 330 4.29 0 58 6 frontal gyrus
Middle/superior temporal 14 R 56 4.25 70 38 8 gyrus
15 L Superior frontal gyrus 74 4.12 14 50 40
16 R Precentral gyrus 64 4.12 56 0 30
Middle/superior temporal 17 R 39 69 4.10 52 60 24 gyrus
18 L Angular gyrus 39 144 3.91 40 66 26
19 R Inferior/middle frontal gyrus 53 3.78 38 36 14
20 R Middle temporal gyrus 21 67 3.76 66 22 10
S1
CA
Uncus, fusiform gyrus, 1 L culmen, hippocampus, 20,28,34 37 2463 12.31 32 14 20 amygdala, caudate tail, lateral
8 globus pallidus, putamen
Sub gyral, uncus, fusiform gyrus, inferior/middle temporal gyrus, culmen, 2 R 20 22,28,34 38 4055 11.36 36 16 18 claustrum, hippocampus, amygdala, caudate tail, lateral globus pallidus
Inferior/middle/superior 3 L 21,38 1120 4.89 54 26 0 temporal gyrus
Inferior/middle temporal 4 R 37,39 294 4.82 58 66 2 gyrus
5 L/C Precuneus, paracentral lobule 5 375 4.81 4 44 62
Middle/superior temporal 6 R 21,22 195 4.66 68 36 0 gyrus
S2
CA
Sub gyral, uncus, fusiform gyrus, inferior occipital gyrus, inferior temporal gyrus, insula, lingual gyrus, middle/superior temporal 13,19,20 22,28, 1 L/C/R 6066 10.62 26 12 16 gyrus, extra nuclear, lentiform 34 38,42 nucleus, claustrum, hippocampus, amygdala, caudate tail, putamen, lateral globus pallidus
Sub gyral, transverse temporal gyrus, uncus, fusiform gyrus, 19 22,28,34 38,41, 2 L inferior occipital gyrus, 5796 10.56 34 12 22 inferior temporal gyrus, 42 insula, lingual gyrus, middle occipital gyrus, middle
9 temporal gyrus, culmen, claustrum, hippocampus, amygdala, caudate tail, lateral globus pallidus, putamen
3 L Lingual gyrus 17 193 5.38 6 92 8
Superior temporal gyrus,
4 L supramarginal gyrus, inferior 13,40 415 4.99 54 46 20
parietal lobule
5 L Inferior frontal gyrus 47 280 4.9 40 28 14
6 R Lingual gyrus 145 4.31 12 82 16
7 R Claustrum, lentiform nucleus 174 4.26 24 16 2
8 L Precentral gyrus 201 4.18 60 0 32
DG
1 R Inferior temporal gyrus 76 4.72 52 24 24
2 R Thalamus, hippocampus 27 165 7.57 22 36 4
3 L Thalamus, hippocampus 27 130 7.10 24 36 0
SUB
Anterior cingulate, 1 L/R 9,10,32 1892 4.69 8 40 8 superior/medial frontal gyrus
2 L Middle/superior frontal gyrus 247 3.91 30 40 20
SW
CA
1 R Sub gyral, uncus, fusiform 20,21,28,34 36,38 2834 11.15 34 18 16
10 gyrus, middle/superior/inferior temporal gyrus, hippocampus, amygdala, caudate tail
Sub gyral, uncus, fusiform gyrus, inferior temporal gyrus, middle/superior temporal 2 L 20 22,28,34 37 3162 10.99 26 10 22 gyrus, hippocampus, amygdala, caudate, lateral globus pallidus
3 L Superior/inferior frontal lobe 38 138 4.80 44 24 18
DG
1 R Thalamus 27 111 5.51 26 32 0
Posterior cingulate, cuneus, 2 L/C/R 18,31 486 5.46 20 68 14 precuneus
SUB
1 R Parahippocampal gyrus 28,35 241 4.78 28 8 32
2 R Superior/medial frontal gyrus 127 4.31 14 38 46
3 L Pre /postcentral gyrus 2 4 279 4.24 34 28 54
4 R Middle/superior frontal gyrus 6 142 3.97 26 8 60
Supplementary Table 1. Clusters resulting from second level random effects analysis
(pFWE,cluster<0.05, collection threshold p<0.001). Regions showing significant stronger functional connectivity with the specified HF subregions as compared to the other two subregions are listed (cornu ammonis, CA; dentate gyrus, DG; subiculum, SUB). Sorting is after T values of the cluster peak voxel. Brodmann areas are identified for clusters covering > 3% of the respective area. Coordinates are given in MNI space.
11
Supplementary Table 2. Activity related to fast (13 15 Hz) and slow (11 13 Hz) sleep spindles.
Cluster Peak voxel Brodmann Brain Region size MNI areas T-value (Voxel) coordinates
Fast spindles Cingulate gyrus, precuneus, pre 1 10, 23 24, /postcentral gyrus, 1 L/C/R 29 32, 12519 6.98 10 20 40 paracentral lobule, R33,L40,42 superior/medial frontal gyrus Transverse/superior 2 L 13,22,29,41,42 935 6.35 18 6 temporal gyrus 32 3 R Thalamus 607 5.44 20 26 4 Transverse/superior 4 R temporal gyrus, 13,22,38,41,42 1128 4.94 34 24 8 insula, claustrum 5 C/R Lingual gyrus/cuneus 17,18 168 4.73 6 88 14 6 L Inferior frontal gyrus 47 112 4.62 12 18 24 Superior temporal 7 L gyrus, inferior frontal 47 138 4.57+ 10 8 54 gyrus Precentral gyrus, 8 R 4,6 448 4.32 50 6 54 middle frontal gyrus
Slow spindles
No suprathreshold clusters
Fast spindles > slow spindles pre /postcentral gyrus, 1 L 2 5 591 4.97 18 42 62 paracentral lobule
Supplementary Table 2. Clusters resulting from second level random effects analysis
(pFWE,cluster < 0.05, collection threshold p<0.001) of fast spindles. No suprathreshold clusters
12 were found for analysis of slow spindles, nor for the negative contrast in both spindle types. Sorting is after T values of the cluster peak voxel. Brodmann areas are identified for clusters covering > 3% of the respective area. Coordinates are given in MNI space. L/C/R denotes + left/central/right clusters. indicates a trend (pFWE,cluster < 0.1).
13 Supplementary Table 3. Activity related to the interaction effect of Spindle × SUB
Peak voxel
Brain Region Brodmann Cluster T- MNI areas size value coordinates (Voxel) 1 L/C/R cingulate gyrus, transverse temporal gyrus, 1 6,13,22 24, 45197 9.84 36 2 2
sub gyral, uncus, fusiform, middle/superior 27 29,
temporal gyrus, 31,32,34,35,
inferior/superior/medial/middle frontal 41 44
gyrus,
pre /para /post central gyrus, ínsula,
parahippocampal gyrus, hippocampus,
amygdala, thalamus, caudate, claustrum,
putamen
2 L superior frontal gyrus 8,9 437 4.96 28 40 32
Supplementary Table 3. Clusters resulting from second level random effects analysis
(pFWE,cluster < 0.05, collection threshold p<0.001). Sorting is after T values of the cluster peak voxel. Brodmann areas are identified for clusters covering > 3% of the respective area. Coordinates are given in MNI space. L/C/R denotes left/central/right clusters
14 Supplementary Table 4. Activity related to the interaction effect of Spindle × DG
Peak voxel
Brain Region Brodmann Cluster T- MNI areas size value coordinates (Voxel)
1 L/R (posterior) cingulate, 1 5, 7, 23, 33459 6.79 24 28 12
(pre )cuneus, sub gyral, 27 31, 35,
transverse/middle/superior temporal gyrus, 40 43
lingual gyrus, pre /para /post central gyrus,
Inferior/superior parietal lobule,
medial/superior frontal gyrus, middle
occipital gyrus, claustrum, thalamus, insula,
parahippocampal gyrus, hippocampus
2 R Middle frontal gyrus 11 156 5.26 34 40 16
3 L Fusiform gyrus, 37 603 5.22 46 56 22
middle temporal gyrus
4 R parahippocampal gyrus, amygdala 34 106 4.75 18 2 18
5 R Superior frontal gyrus 9 156 4.72 32 44 38
6 R Middle occipital gyrus 236 4.69 46 76 12
7 L/R Medial frontal gyrus 158 4.39 2 52 8
Supplementary Table 4. Clusters resulting from second level random effects analysis
(pFWE,cluster < 0.05, collection threshold p<0.001). Sorting is after T values of the cluster peak voxel. Brodmann areas are identified for clusters covering > 3% of the respective area. Coordinates are given in MNI space. L/C/R denotes left/central/right clusters.
15 Supplementary Table 5. Activity related to the interaction effect of Spindle × CA
Peak voxel
Brain Region Brodmann Cluster T- MNI areas size value coordinates (Voxel)
1 R Sub gyral, fusiform gyrus, uncus, 1 4,6,22, 27 11809 7.88 24 12 16
angular gyrus, 29, 34,35,41
transverse/middle/superior temporal 43
gyrus, pre /post central gyrus,
inferior/middle frontal gyrus, inferior
parietal gyrus, insula, lingual gyrus,
parahippocampal gyrus, amygdala,
claustrum
2 L/R Posterior cingulate, fusiform gyrus, L19,23,L27 2102 6.72 32 50 14
parahippocampal gyrus, hippocampus, L28,29,30,L35
lingual gyrus L37
3 L parahippocampal gyrus, hippocampus, 28,34 359 6.65 28 10 24
amygdala
4 R Precuneus 7 231 6.62 16 54 44
5 L/C/R Cingulate gyrus, para /post central L3 L5,6,23,24, 3728 6.37 4 22 54
gyrus, medial frontal gyrus 31,32
6 L Transverse/superior temporal gyrus, 1 4,6,13,22, 3876 5.83 28 24 2
insula, pre /post central gyrus, 41 45,47
inferior/middle frontal gyrus, inferior
parietal lobule
7 L Middle/superior frontal gyrus 9 365 5.08 28 38 36
8 L/R Anterior cingulate 32 254 4.51 0 46 4
9 R Superior frontal gyrus 8 194 3.97 18 36 56
16
Supplementary Table 5. Clusters resulting from second level random effects analysis
(pFWE,cluster < 0.05, collection threshold p<0.001). Sorting is after T values of the cluster peak voxel. Brodmann areas are identified for clusters covering > 3% of the respective area. Coordinates are given in MNI space. L/C/R denotes left/central/right clusters.
17 Supplementary Table 6. Linear mixed models correcting for the variable subject ID
F test Post-hoc test result AAL region df2 F value contrast left precuneus 83.4 9.18** S0>SW** right superior medial frontal gyrus 84.5 7.17** S0>SW** left inferior parietal lobule 79.6 7.61** S0>SW** right inferior parietal lobule 85.4 6.66** S0>SW** left superior temporal gyrus 79.4 12.55** S2>S0/SW** right superior temporal gyrus 81.1 12.32** S2>S0/SW** left inferior orbitofrontal gyrus 86.0 4.94* S2>SW**
Supplementary Table 6. Linear mixed models correcting for the variable subject ID. The primary regions from SPM contrasts in Figure 2 were used for the analysis, in which we extracted the first eigenvariate of the magnitude values from the first level contrasts. F values of the main effect of sleep are provided with the adapted df2 (df1 = 3). Post-hoc test results regarding the contrasts shown in Figure 2 are also provided. **p<0.001, *p<0.003.
Reference List
1. Ashburner J, Friston KJ (2005) Unified segmentation. Neuroimage 26: 839 851.
2. Auer DP (2008) Spontaneous low frequency blood oxygenation level dependent fluctuations and functional connectivity analysis of the 'resting' brain. Magn Reson Imaging 26: 1055 1064.
3. Chang C, Glover GH (2009) Effects of model based physiological noise correction on default mode network anti correlations and correlations. Neuroimage 47: 1448 1459.
4. Cordes D, Haughton VM, Arfanakis K, Carew JD, Turski PA, Moritz CH, Quigley MA, Meyerand ME (2001) Frequencies contributing to functional connectivity in the cerebral cortex in "resting state" data. AJNR Am J Neuroradiol 22: 1326 1333.
5. Diekelmann S, Born J (2010) The memory function of sleep. Nat Rev Neurosci 11: 114 126.
18 6. He BJ, Snyder AZ, Zempel JM, Smyth MD, Raichle ME (2008) Electrophysiological correlates of the brain's intrinsic large scale functional architecture. Proc Natl Acad Sci U S A 105: 16039 16044.
7. Lund TE, Madsen KH, Sidaros K, Luo WL, Nichols TE (2006) Non white noise in fMRI: does modelling have an impact? Neuroimage 29: 54 66.
8. Sirota A, Csicsvari J, Buhl D, Buzsaki G (2003) Communication between neocortex and hippocampus during sleep in rodents. Proc Natl Acad Sci U S A 100: 2065 2069.
9. Van Dijk KR, Hedden T, Venkataraman A, Evans KC, Lazar SW, Buckner RL (2010) Intrinsic functional connectivity as a tool for human connectomics: theory, properties, and optimization. J Neurophysiol 103: 297 321.
10. Weissenbacher A, Kasess C, Gerstl F, Lanzenberger R, Moser E, Windischberger C (2009) Correlations and anticorrelations in resting state functional connectivity MRI: a quantitative comparison of preprocessing strategies. Neuroimage 47: 1408 1416.
19