The brain consumes a great deal of a sugar called glucose, which is delivered to the brain through blood vessels. Active regions of the brain need more glucose, and so the brain has a metabolic network that controls when and where glucose is metabolized. Yet precisely how this metabolic network changes during brain development is not yet understood.

Choi et al. have now monitored the patterns of glucose metabolism in the brains of awake rats as they matured from 'childhood' to early adulthood. The experiments involved injecting the rats with radioactive glucose, and then using a technique called positron emission tomography (commonly known as 'PET scan') to monitor the metabolism of these radioactive sugar molecules in the animals’ brains.

Choi et al. showed that the patterns of glucose consumption in the brain shift drastically as the rats mature. Importantly, the findings showed that these shifts in glucose metabolism seem to support the activity of long distance connections that develop as the brain matures. The findings also showed that the increased long distance connections were energy efficient. The results suggest that these metabolic changes are likely a way of maintaining high-energy efficiency that is crucial for the brain to perform normally.

Finally, in addition to revealing the changes involved in normal brain development, these findings may have implications in neurological and psychiatric disorders in which the brain fails to achieve efficient metabolic networks as it matures.

Recent neuroimaging studies unraveled the developmental changes of the adolescent brains (Blakemore, 2012). Among those neuroimaging techniques, positron emission tomography (PET) provides quantitative information regarding regional metabolism or synaptic neurochemicals with high sensitivity. As ¹⁸F-Fluorodeoxyglucose (FDG) is taken up in the brain proportional to cerebral energy consumption and neuronal activity, FDG PET studies have been used to investigate regional energy metabolism during physiologic and pathologic processes. Previous brain developmental studies using FDG PET in humans assumed that the regional increase of metabolism was accompanied by regional increase of neuronal activity during brain maturation (Chugani et al., 1987; Phelps and Mazziotta, 1985). Despite these previous data, the developmental changes are not understood in terms of dynamic organization of metabolic networks between brain regions.

Since a series of regional brain activity is being coherently organized changing throughout time, sets of specific brain regions are activated or deactivated during resting or various cognitive tasks. These specific regional activity patterns, so-called functional brain networks, are identified by neuroimaging studies (Deco et al., 2011). Among several imaging approaches, resting state functional magnetic resonance image (fMRI)-based connectivity analyses revealed maturation of the functional networks during childhood and adolescence (Fair et al., 2008; Fair et al., 2007; Supekar et al., 2009). FDG PET can also reveal functional metabolic network and its maturation, exploiting the coupling between neuronal activity and metabolism. While the fMRI-based functional connectivity measures correlation of fast temporal fluctuations, metabolic connectivity measured by FDG PET reflects a cumulative energy consumption in several minutes and provides relatively stable information regarding steady resting state (Choi et al., 2014; Di et al., 2012; Lee et al., 2008; Lee et al., 2012; Toussaint et al., 2012; Yakushev et al., 2013). Furthermore, FDG PET measures metabolic activity and connection patterns during awaken states even in animals as in humans, because FDG is taken up in the brain mainly during the period after injection and before the imaging under anesthesia.

To organize highly connected functional brain networks, which can be measured by fMRI or FDG PET, the brain needs efficient energy metabolism (Bullmore and Sporns, 2012). As the high energetic costs are required for brain wiring to construct and maintain hub regions with dense connections, hub regions will show high glucose metabolism (Lord et al., 2013). Brain regions were found to show different regional energy efficiency so that, for instance, subcortical regions consumed relatively lower metabolic energy per connection (Tomasi et al., 2013). Therefore, we can say that metabolism of brain regions could be determined both by the regional complexity of functional networks and their energy efficiency for wiring. Since the developmental changes in functional networks reconfigure the metabolic demands, energy efficiency of the brain networks are going to be changed during maturation.

We primarily aimed to find the maturation-related changes in both regional metabolic activity and metabolic connectivity in adolescent period with a longitudinal study in rats. As the animals allow repeated imaging studies, we could study longitudinal maturation of regional metabolism and connectivity in the same animals. As the small animals also have default mode network (DMN) (Lu et al., 2012) defined as brain regions activating during resting state and deactivating during attention-demanding tasks (Raichle et al., 2001), specific network components were extracted in the rat brain from minutes-scale metabolic covariance patterns and independent component analysis (ICA). We hypothesized the metabolic networks are dynamically organized during maturation to achieve connection efficiency. We investigated maturation of large-scale metabolic connectivity and analyzed their energy efficiency for wiring in rat brains.

Voxelwise comparisons were firstly performed among rats aged 5, 10, and 15 weeks to disclose temporal changes of regional metabolic activity. The comparison using statistical parametric mapping revealed that the metabolism increased in 10-week-old age in bilateral frontal cortices and anterior aspect of striatum compared with 5-week-old age and decreased in bilateral cerebellar cortices, thalamus, parieto-occipital cortices, and retrosplenial cortices (Figure 1A) (representative FDG PET images for each age are shown in Figure 1—figure supplement 1). Again, metabolism increased in 15-week-old age compared with 10-week-old age, in the clusters of both hippocampi and decreased in the small clusters of right striatum and left frontal cortex (Figure 1B, Figure1—figure supplement 2).

Figure 1 with 2 supplements see all

Download asset Open asset

Figure 2 with 2 supplements see all

Download asset Open asset

Interregional metabolic correlation on FDG PET was supposed to reflect interregional covariance patterns of neuronal activities. An ICA was performed to yield regional contributing components of metabolic networks. A total of 13 metabolic independent components (ICs) were identified in rats similar to components found in previous reports in humans on FDG PET (Di et al., 2012; Toussaint et al., 2012; Yakushev et al., 2013). All ICs were displayed in Figure 2—figure supplement 1, and a threshold z > 1.5 was applied for the voxels for display purposes. Four particular components, IC1, IC5, IC8, and IC9, were selected (Figure 2), which anatomically corresponded to previously alleged limbic/anterior DMN (IC1), posterior DMN (IC5), motor (IC8), and somatosensory network (IC9), respectively (Lu et al., 2012). IC1 included the clusters of dorsal hippocampi and medial prefrontal cortex. IC5 included the retrosplenial cortex, known as a hub of posterior DMN in rats (Lu et al., 2012). It is anatomically adjacent to the posterior cingulate and precuneus in human as cores of DMN (Raichle et al., 2001). These were used further to identify maturing patterns of metabolic networks consisting of IC-derived volume-of-interests (VOIs). Eight VOIs (3 for IC1, 1 for IC5, 2 for IC8 and IC9, respectively) were selected taking anatomical structures into consideration (Figure 2—figure supplement 2) (stereotaxic coordinates and abbreviations of VOIs are summarized in Table 1).

Interregional correlations were calculated between paired VOIs to yield 28 pairs (28 edges upon 8 nodes). This VOI-based metabolic interregional correlation was assumed to yield functionally coherent covariance patterns among spatial components to represent metabolic connectivity. Metabolic connectivity obtained this way was used to investigate the changes of metabolic networks during maturation. Figure 3A shows interregional correlation representing connection strength of pairs of VOIs in rats aged 5, 10, and 15 weeks. Significantly different connections in interregional correlation were examined using nonparametric permutation tests (Figure 3—figure supplement 1) using pseudorandom relabeling 5-week/10-week-old rat images or 10-week/15-week-old ones, followed by FDR correction for multiple comparisons (Kim et al., 2015). Connectivity matrices (Figure 3A) and p-values for the difference in connection strength (Figure 3—figure supplement 2) between paired groups suggested that the connection between pairs of retrosplenial, medial prefrontal, and sensorimotor cortices was strengthened from 5 to 10 weeks, while connectivity involving limbic regions did not change. According to aging, pairs of anterior-posterior connections were significantly strengthened when comparing 10-week-old rats with 5-week-old rats (Figure 3B). The increase became prominent when we compared 15-week-old rats from 5-week-old rats. Connectivity was not significantly weakened between areas during maturation (Figure 3—figure supplement 2).

Figure 3 with 2 supplements see all

Download asset Open asset

We further investigated energy efficiency to configure the metabolic connectivity between spatial components at each period of maturation. The energy efficiency was defined as the ratio of metabolic connection strength, a sum of weights of links connected to each VOI, to normalized FDG uptake of each VOI. Significantly different metabolic energy efficiency was also examined using the permutation test (Figure 3—figure supplement 1). In 5-week-old rats, energy efficiency was relatively lower in midline structures, that is, medial prefrontal and retrosplenial cortices than sensorimotor cortices. In 10- and 15-week-old age compared with 5-week-old age, energy efficiency of the midline structures significantly increased almost reaching those of the sensorimotor cortices (Figure 4).

Figure 4

Download asset Open asset

Figure 4—source data 1

Energy efficiency of each brain region.

https://doi.org/10.7554/eLife.11571.014

Download elife-11571-fig4-data1-v2.xlsx

In the present study, large-scale brain network based on glucose metabolic usage measured by PET was analyzed to evaluate the maturation of the metabolic brain connectivity in rats. We identified the specific pattern of changes in regional metabolism and metabolic connectivity during brain development. In adolescent (10 weeks) and early-adult (15 weeks) period, metabolism increased in the bilateral frontal lobes compared to childhood (5 weeks), whereas metabolism decreased in the posterior cerebrum, thalamus, and cerebellum. Metabolic network components extracted using ICA were default-mode, limbic, motor, and somatosensory ones, which have been regarded as important functional components also in humans. Further connectivity analyses using VOIs put on these components showed connection strength increased particularly in the anterior-posterior connections during maturation. This connection strength increase was accompanied by increased energy efficiency in the midline structures including medial prefrontal and retrosplenial cortices. Our work suggested large-scale functional networks matured to increase anterior-posterior long-distance connections and achieved energetically efficient wiring in the midline structures during maturation from childhood via adolescence to early adulthood.

Voxelwise comparison revealed metabolism increase in bilateral frontal cortices during maturation, prominent during 510 weeks of ages, which complied with previous perfusion studies. A study using ^99mTc-HMPAO SPECT showed perfusion increase in the neocortex at juvenile to young adult periods in mice, particularly in frontal lobes relative to the cerebellum and subcortical structures (Apostolova et al., 2012). In humans, perfusion was lower in the neonates’ neocortex while higher in the basal ganglia and cerebellum (Fockele et al., 1990). Additionally, positive correlation was found between age and frontal lobe perfusion (Kuji et al., 1999). FDG PET studies in human showed significantly higher metabolism mostly in anterior cingulate and thalamus before 25-year-old (Van Bogaert et al., 1998) and linear metabolism increase from the age of 115 in prefrontal/orbitofrontal cortices (Kang et al., 2004). As a first study in terms of longitudinal age-related brain metabolism in rats, our findings of age-related frontal metabolism increase corresponded to human PET findings as well as perfusion studies in rodents.

Analysis of metabolic network are based on the sequential coupling of neuronal activity, brain metabolism, and blood flow, which more closely represent the neuronal networks than blood flow networks. Because metabolic activity measured by FDG uptake reflects the cumulative energy consumption in steady states, while BOLD signal from fMRI reflects fast temporal fluctuation of physiologic factors such as blood flow, blood oxygenation, and cerebral metabolic rate of oxygen, the metabolic network provides distinct information from the functional network on fMRI as well as from the structural network disclosed on diffusion tensor imaging (Di et al., 2012; Wehrl et al., 2013). Furthermore, network construction using neuronal activity-coupled metabolism found on FDG PET could take regional metabolic activities into account to disclose energy efficiency rather than only concentrate on the changes of fluctuating neural activities without considering absolute amount of regional perfusion or metabolism (Hyder et al., 2011; Smith et al., 2002).

Using metabolic network analyses, we could identify networks of rat brains similar to humans using ICA. In human brains, ICA for metabolic networks revealed DMN successfully (Toussaint et al., 2012; Yakushev et al., 2013). We also found homologous patterns in rats to those of humans. According to the studies comparing the networks acquired from fMRI and PET, each DMN from these modalities shared quite the similar anatomical regions, although distinctive patterns were found on these two modalities (Di et al., 2012; Wehrl et al., 2013). Of note, our PET-derived IC5 shared the regions of fMRI-derived DMN including retrosplenial cortices and small clusters of bilateral posterior hippocampi. The IC1 shared the regions of anterior part of DMN derived from fMRI, medial frontal cortices (Lu et al., 2012). Considering that DMN-related regions are deactivated during attention-demanding and working memory tasks (Raichle et al., 2001), DMN may be constructed during brain maturation in adolescence to learn attention-demanding tasks. Previous human studies on the fMRI-derived functional networks revealed connectivity increase in DMN regions during maturation, particularly between the anterior and posterior midline structures (Fair et al., 2008), which might be closely associated with the developmental process of working memory (Satterthwaite et al., 2013). Moreover, fMRI-derived DMN showed two distinctive clusters, temporal-prefrontal and parietal subsystems, which considerably corresponded to our IC1 and IC5 (Lu et al., 2012). According to our results, the metabolic connections of these two clusters of DMN developed during brain maturation in rats. In spite of homologous network patterns of rat brain, metabolic networks derived from FDG PET were partly different from fMRI-derived networks. A discrepancy in DMN between fMRI and PET was also reported in human (Di et al., 2012). In PET-derived networks in the rat brain, anterior and posterior part of DMN was separately extracted (IC1 and IC5). Furthermore, cingulate cortex was rarely included in IC1 or IC5, which was different from previous fMRI-derived DMN (Lu et al., 2012; Sierakowiak et al., 2015). Nevertheless, a hub region of posterior part of rat DMN was retrosplenial cortex and that of anterior part was medial prefrontal cortex in fMRI study, which corresponded to PET-derived IC5 and IC1, respectively.

During connectivity maturation, metabolic connections between component VOIs strengthened, particularly in anterior-posterior long-distance ones. Along with increased long-distance connections, the retrosplenial and medial prefrontal cortices showed efficient energy consumption. This maturation pattern may represent enhancement of efficiency of information flow between network components, which has been suggested in human brain maturation (Fair et al., 2007; Stevens et al., 2009). The strengthened connections between long-distance network components are likely to support dynamically organizing brain networks to achieve efficient connections according to cognitive demands (Bullmore and Sporns, 2012). The functional interactions of large-scale brain networks are dynamically evolving, which could be associated with cognitive, behavioral developments during maturation and pathophysiology of brain disorders (Deco et al., 2011). In this investigation using metabolic connectivity analysis, we could localize the components, that is, medial prefrontal and retrosplenial areas, showing strengthened connections and at the same time enhanced connection efficiencies.

Cognitive developments demand highly connected functional brain networks, particularly in hub regions (Micheloyannis et al., 2009; van den Heuvel and Sporns, 2013). They could accompany high-energy consumption in spite of limited supply of glucose for brain (Smith et al., 2002; Tomasi et al., 2013). Our results of increased energy efficiency in retrosplenial and medial prefrontal cortex imply long-distance connectivity maturation coincide with the redistribution of energy consumption. High-energy efficiency in hubs was previously reported in adult human brain (Tomasi et al., 2013), which was recapitulated in our study in rats, and we also found that their high efficiency developed during brain maturation. Comprehensively, large-scale networks are formed during adolescent period accompanied by these networks’ efficient energy consumption for long-distance connectivity. As deficits in these relationships are found in several brain disorders (Broyd et al., 2009), metabolic network analyses might be used to find the relevant functional connectivity abnormalities. Specifically, the hub regions are well known as culprit components for several developmental and degenerative brain disorders (Fornito et al., 2015). Our results could explain the underlying biological background of the vulnerability that hub regions might be easily influenced by developmental problems in reconfiguration of metabolic connectivity efficiency.

To find the disease-related abnormalities differentiated from normal developmental changes in neurological or psychiatric disorders, we need to characterize first the normal developmental patterns of metabolic networks. Although metabolic networks have advantages to reveal steady-state connectivity in longer term scale than resting fMRI, it is difficult to perform longitudinally repeated PET in the growing normal children. And thus, our findings in rats of maturational changes of metabolic connectivity can be referred to as surrogates, considering the homology between humans and rats in network components disclosed by ICA (Lu et al., 2012; Wehrl et al., 2013).

Although rat brains have similar features of metabolic network to human brains, there are several technical issues of interest. Unlike human studies, rats should be anesthetized during either FDG injection or image acquisition. Since anesthesia could affect brain metabolism, rats were awake after the injection until imaging to minimize the anesthesia effects. The images acquired from small animal PET had the voxel size of 0.3875 mm and physical spatial resolution of PET scanner is more than 1 mm due to positron range and acollinearity. Thus, we were afraid that it might prevent the delineation of network components during ICA. However, despite this limited spatial resolution, the brain regions with specific covariance patterns were identified and further metabolic connectivity analyses were finally feasible to provide characteristic pattern of brain maturation.

1. 1. Apostolova I
   2. Wunder A
   3. Dirnagl U
   4. Michel R
   5. Stemmer N
   6. Lukas M
   7. Derlin T
   8. Gregor-Mamoudou B
   9. Goldschmidt J
   10. Brenner W
   11. Buchert R

   (2012) Brain perfusion SPECT in the mouse: normal pattern according to gender and age

   NeuroImage 63:1807–1817.

   https://doi.org/10.1016/j.neuroimage.2012.08.038

   • Google Scholar
2. 1. Blakemore S-J

   (2012) Imaging brain development: the adolescent brain

   NeuroImage 61:397–406.

   https://doi.org/10.1016/j.neuroimage.2011.11.080

   • Google Scholar
3. 1. Broyd SJ
   2. Demanuele C
   3. Debener S
   4. Helps SK
   5. James CJ
   6. Sonuga-Barke EJS

   (2009) Default-mode brain dysfunction in mental disorders: a systematic review

   Neuroscience & Biobehavioral Reviews 33:279–296.

   https://doi.org/10.1016/j.neubiorev.2008.09.002

   • Google Scholar
4. 1. Bullmore E
   2. Sporns O

   (2012) The economy of brain network organization

   Nature Reviews Neuroscience 13:336–349.

   https://doi.org/10.1038/nrn3214

   • Google Scholar
5. 1. Choi H
   2. Kim YK
   3. Kang H
   4. Lee H
   5. Im H-J
   6. Hwang DW
   7. Kim EE
   8. Chung J-K
   9. Lee DS

   (2014) Abnormal metabolic connectivity in the pilocarpine-induced epilepsy rat model: a multiscale network analysis based on persistent homology

   NeuroImage 99:226–236.

   https://doi.org/10.1016/j.neuroimage.2014.05.039

   • Google Scholar
6. 1. Chugani HT
   2. Phelps ME
   3. Mazziotta JC

   (1987) Positron emission tomography study of human brain functional development

   Annals of Neurology 22:487–497.

   https://doi.org/10.1002/ana.410220408

   • Google Scholar
7. 1. Deco G
   2. Jirsa VK
   3. McIntosh AR

   (2011) Emerging concepts for the dynamical organization of resting-state activity in the brain

   Nature Reviews Neuroscience 12:43–56.

   https://doi.org/10.1038/nrn2961

   • Google Scholar
8. 1. Di X
   2. Biswal, and Alzheimer's Disease Neu BB

   (2012) Metabolic brain covariant networks as revealed by FDG-PET with reference to resting-state fMRI networks

   Brain Connectivity 2:275–283.

   https://doi.org/10.1089/brain.2012.0086

   • Google Scholar
9. 1. Fair DA
   2. Dosenbach NUF
   3. Church JA
   4. Cohen AL
   5. Brahmbhatt S
   6. Miezin FM
   7. Barch DM
   8. Raichle ME
   9. Petersen SE
   10. Schlaggar BL

   (2007) Development of distinct control networks through segregation and integration

   Proceedings of the National Academy of Sciences of the United States of America 104:13507–13512.

   https://doi.org/10.1073/pnas.0705843104

   • Google Scholar
10. 1. Fair DA
    2. Cohen AL
    3. Dosenbach NUF
    4. Church JA
    5. Miezin FM
    6. Barch DM
    7. Raichle ME
    8. Petersen SE
    9. Schlaggar BL

    (2008) The maturing architecture of the brain's default network

    Proceedings of the National Academy of Sciences of the United States of America 105:4028–4032.

    https://doi.org/10.1073/pnas.0800376105

    • Google Scholar
11. 1. Fockele DS
    2. Baumann RJ
    3. Shih Wei-Jen
    4. Ryo Uyun

    (1990) Tc-99m HMPAO SPECT of the brain in the neonate

    Clinical Nuclear Medicine 15:175–177.

    https://doi.org/10.1097/00003072-199003000-00008

    • Google Scholar
12. 1. Fornito A
    2. Zalesky A
    3. Breakspear M

    (2015) The connectomics of brain disorders

    Nature Reviews Neuroscience 16:159–172.

    https://doi.org/10.1038/nrn3901

    • Google Scholar
13. 1. Hyder F
    2. Herman P
    3. Sanganahalli BG
    4. Coman D
    5. Blumenfeld H
    6. Rothman DL

    (2011) Role of ongoing, intrinsic activity of neuronal populations for quantitative neuroimaging of functional magnetic resonance imaging–based networks

    Brain Connectivity 1:185–193.

    https://doi.org/10.1089/brain.2011.0032

    • Google Scholar
14. 1. Kaiser M

    (2011) A tutorial in connectome analysis: topological and spatial features of brain networks

    NeuroImage 57:892–907.

    https://doi.org/10.1016/j.neuroimage.2011.05.025

    • Google Scholar
15. 1. Kang E
    2. Lee DS
    3. Kang H
    4. Lee JS
    5. Oh SH
    6. Lee MC
    7. Kim CS

    (2004) Age-associated changes of cerebral glucose metabolic activity in both male and female deaf children: parametric analysis using objective volume of interest and voxel-based mapping

    NeuroImage 22:1543–1553.

    https://doi.org/10.1016/j.neuroimage.2004.04.010

    • Google Scholar
16. 1. Kim H
    2. Hahm J
    3. Lee H
    4. Kang E
    5. Kang H
    6. Lee DS

    (2015) Brain networks engaged in audiovisual integration during speech perception revealed by persistent homology-based network filtration

    Brain Connectivity 5:245–258.

    https://doi.org/10.1089/brain.2013.0218

    • Google Scholar
17. 1. Kuji I
    2. Sumiya H
    3. Niida Y
    4. Takizawa N
    5. Ikeda E
    6. Tsuji S
    7. Tonami N

    (1999)

    Age-related changes in the cerebral distribution of 99mTc-ECD from infancy to adulthood

    Journal of Nuclear Medicine 40:1818–1823.

    • Google Scholar
18. 1. Lee DS
    2. Kang H
    3. Kim H
    4. Park H
    5. Oh JS
    6. Lee JS
    7. Lee MC

    (2008) Metabolic connectivity by interregional correlation analysis using statistical parametric mapping (sPM) and FDG brain PET; methodological development and patterns of metabolic connectivity in adults

    European Journal of Nuclear Medicine and Molecular Imaging 35:1681–1691.

    https://doi.org/10.1007/s00259-008-0808-z

    • Google Scholar
19. 1. Lee H
    2. Kang H
    3. Chung MK
    4. Kim BN
    5. Lee DS

    (2012) Persistent brain network homology from the perspective of dendrogram

    IEEE Transactions on Medical Imaging 31:2267–2277.

    https://doi.org/10.1109/TMI.2012.2219590

    • Google Scholar
20. 1. Li Y-O
    2. Adalı T
    3. Calhoun VD

    (2007) Estimating the number of independent components for functional magnetic resonance imaging data

    Human Brain Mapping 28:1251–1266.

    https://doi.org/10.1002/hbm.20359

    • Google Scholar
21. 1. Lord L-D
    2. Expert P
    3. Huckins JF
    4. Turkheimer FE

    (2013) Cerebral energy metabolism and the brain’s functional network architecture: an integrative review

    Journal of Cerebral Blood Flow & Metabolism 33:1347–1354.

    https://doi.org/10.1038/jcbfm.2013.94

    • Google Scholar
22. 1. Lu H
    2. Zou Q
    3. Gu H
    4. Raichle ME
    5. Stein EA
    6. Yang Y

    (2012) Rat brains also have a default mode network

    Proceedings of the National Academy of Sciences of the United States of America 109:3979–3984.

    https://doi.org/10.1073/pnas.1200506109

    • Google Scholar
23. 1. Micheloyannis S
    2. Vourkas M
    3. Tsirka V
    4. Karakonstantaki E
    5. Kanatsouli K
    6. Stam CJ

    (2009) The influence of ageing on complex brain networks: a graph theoretical analysis

    Human Brain Mapping 30:200–208.

    https://doi.org/10.1002/hbm.20492

    • Google Scholar
24. 1. Phelps M
    2. Mazziotta J

    (1985) Positron emission tomography: human brain function and biochemistry

    Science 228:799–809.

    https://doi.org/10.1126/science.2860723

    • Google Scholar
25. 1. Raichle ME
    2. MacLeod AM
    3. Snyder AZ
    4. Powers WJ
    5. Gusnard DA
    6. Shulman GL

    (2001) A default mode of brain function

    Proceedings of the National Academy of Sciences of the United States of America 98:676–682.

    https://doi.org/10.1073/pnas.98.2.676

    • Google Scholar
26. 1. Satterthwaite TD
    2. Wolf DH
    3. Erus G
    4. Ruparel K
    5. Elliott MA
    6. Gennatas ED
    7. Hopson R
    8. Jackson C
    9. Prabhakaran K
    10. Bilker WB
    11. Calkins ME
    12. Loughead J
    13. Smith A
    14. Roalf DR
    15. Hakonarson H
    16. Verma R
    17. Davatzikos C
    18. Gur RC
    19. Gur RE

    (2013) Functional maturation of the executive system during adolescence

    Journal of Neuroscience 33:16249–16261.

    https://doi.org/10.1523/JNEUROSCI.2345-13.2013

    • Google Scholar
27. 1. Schiffer WK
    2. Mirrione MM
    3. Biegon A
    4. Alexoff DL
    5. Patel V
    6. Dewey SL

    (2006) Serial microPET measures of the metabolic reaction to a microdialysis probe implant

    Journal of Neuroscience Methods 155:272–284.

    https://doi.org/10.1016/j.jneumeth.2006.01.027

    • Google Scholar
28. 1. Schiffer WK
    2. Mirrione MM
    3. Dewey SL

    (2007)

    Optimizing experimental protocols for quantitative behavioral imaging with 18F-FDG in rodents

    Journal of Nuclear Medicine 48:277–287.

    • Google Scholar
29. 1. Sierakowiak A
    2. Monnot C
    3. Aski SN
    4. Uppman M
    5. Li T-Q
    6. Damberg P
    7. Brené S
    8. Gozzi A

    (2015) Default mode network, motor network, dorsal and ventral basal ganglia networks in the rat brain: comparison to human networks using resting state-fMRI

    PLOS ONE 10:e0120345.

    https://doi.org/10.1371/journal.pone.0120345

    • Google Scholar
30. 1. Smith AJ
    2. Blumenfeld H
    3. Behar KL
    4. Rothman DL
    5. Shulman RG
    6. Hyder F

    (2002) Cerebral energetics and spiking frequency: the neurophysiological basis of fMRI

    Proceedings of the National Academy of Sciences of the United States of America 99:10765–10770.

    https://doi.org/10.1073/pnas.132272199

    • Google Scholar
31. 1. Stevens MC
    2. Pearlson GD
    3. Calhoun VD

    (2009) Changes in the interaction of resting-state neural networks from adolescence to adulthood

    Human Brain Mapping 30:2356–2366.

    https://doi.org/10.1002/hbm.20673

    • Google Scholar
32. 1. Supekar K
    2. Musen M
    3. Menon V
    4. Friston KJ

    (2009) Development of large-scale functional brain networks in children

    PLoS Biology 7:e1000157.

    https://doi.org/10.1371/journal.pbio.1000157

    • Google Scholar
33. 1. Toga AW
    2. Santori EM
    3. Hazani R
    4. Ambach K

    (1995) A 3D digital map of rat brain

    Brain Research Bulletin 38:77–85.

    https://doi.org/10.1016/0361-9230(95)00074-O

    • Google Scholar
34. 1. Tomasi D
    2. Wang G-J
    3. Volkow ND

    (2013) Energetic cost of brain functional connectivity

    Proceedings of the National Academy of Sciences of the United States of America 110:13642–13647.

    https://doi.org/10.1073/pnas.1303346110

    • Google Scholar
35. 1. Van Bogaert P
    2. Wikler D
    3. Damhaut P
    4. Szliwowski HB
    5. Goldman S

    (1998) Regional changes in glucose metabolism during brain development from the age of 6 years

    NeuroImage 8:62–68.

    https://doi.org/10.1006/nimg.1998.0346

    • Google Scholar
36. 1. van den Heuvel MP
    2. Sporns O

    (2013) Network hubs in the human brain

    Trends in Cognitive Sciences 17:683–696.

    https://doi.org/10.1016/j.tics.2013.09.012

    • Google Scholar
37. 1. Toussaint P-J
    2. Perlbarg V
    3. Bellec P
    4. Desarnaud S
    5. Lacomblez L
    6. Doyon J
    7. Habert M-O
    8. Benali H

    (2012) Resting state FDG-PET functional connectivity as an early biomarker of alzheimer's disease using conjoint univariate and independent component analyses

    NeuroImage 63:936–946.

    https://doi.org/10.1016/j.neuroimage.2012.03.091

    • Google Scholar
38. 1. Wehrl HF
    2. Hossain M
    3. Lankes K
    4. Liu C-C
    5. Bezrukov I
    6. Martirosian P
    7. Schick F
    8. Reischl G
    9. Pichler BJ

    (2013) Simultaneous PET-MRI reveals brain function in activated and resting state on metabolic, hemodynamic and multiple temporal scales

    Nature Medicine 19:1184–1189.

    https://doi.org/10.1038/nm.3290

    • Google Scholar
39. 1. Xu L
    2. Groth KM
    3. Pearlson G
    4. Schretlen DJ
    5. Calhoun VD

    (2009) Source-based morphometry: the use of independent component analysis to identify gray matter differences with application to schizophrenia

    Human Brain Mapping 30:711–724.

    https://doi.org/10.1002/hbm.20540

    • Google Scholar
40. 1. Yakushev I
    2. Chételat G
    3. Fischer FU
    4. Landeau B
    5. Bastin C
    6. Scheurich A
    7. Perrotin A
    8. Bahri MA
    9. Drzezga A
    10. Eustache F
    11. Schreckenberger M
    12. Fellgiebel A
    13. Salmon E

    (2013) Metabolic and structural connectivity within the default mode network relates to working memory performance in young healthy adults

    NeuroImage 79:184–190.

    https://doi.org/10.1016/j.neuroimage.2013.04.069

    • Google Scholar

Author details

1. Hongyoon Choi

   1. Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea
   2. Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology, Seoul National University, Seoul, Republic of Korea

   Contribution

   HC, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

2. Yoori Choi

   Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

   Contribution

   YC, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

3. Kyu Wan Kim

   Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

   Contribution

   KWK, Acquisition of data

   Competing interests

   The authors declare that no competing interests exist.

4. Hyejin Kang

   Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

   Contribution

   HK, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

5. Do Won Hwang

   1. Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea
   2. Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology, Seoul National University, Seoul, Republic of Korea

   Contribution

   DWH, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

6. E Edmund Kim

   1. Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea
   2. Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology, Seoul National University, Seoul, Republic of Korea

   Contribution

   EEK, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

7. June-Key Chung

   Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

   Contribution

   JKC, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

8. Dong Soo Lee

   1. Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea
   2. Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology, Seoul National University, Seoul, Republic of Korea

   Contribution

   DSL, Conception and design, Drafting or revising the article

   For correspondence

   dsl@snu.ac.kr

   Competing interests

   The authors declare that no competing interests exist.

• 1,849

  views

• 378

  downloads

• 26

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.