Evidence of a Dissociation Pattern in Resting-State Default Mode Network Connectivity in First-Episode, Treatment-Naive Major Depression Patients

Evidence of a Dissociation Pattern in Resting-State Default Mode Network Connectivity in First-Episode, Treatment-Naive Major Depression Patients Xueling Zhu, Xiang Wang, Jin Xiao, Jian Liao, Mingtian Zhong, Wei Wang, and Shuqiao Yao Background: Imaging studies have shown that major depressive disorder (MDD) is associated with altered activity patterns of the default mode network (DMN). However, the neural correlates of the resting-state DMN and MDD-related pathopsychological characteristics, such as depressive rumination and overgeneral autobiographical memory (OGM) phenomena, still remain unclear. Methods: Using independent component analysis, we analyzed resting-state functional magnetic resonance imaging data obtained from 35 first-episode, treatment-naive young adults with MDD and from 35 matched healthy control subjects. Results: Patients with MDD exhibited higher levels of rumination and OGM than did the control subjects. We observed increased functional connectivity in the anterior medial cortex regions (especially the medial prefrontal cortex and anterior cingulate cortex) and decreased functional connectivity in the posterior medial cortex regions (especially the posterior cingulate cortex/precuneus) in MDD patients compared with control subjects. In the depressed group, the increased functional connectivity in the anterior medial cortex correlated positively with rumination score, while the decreased functional connectivity in the posterior medial cortex correlated negatively with OGM score. Conclusions: We report dissociation between anterior and posterior functional connectivity in resting-state DMNs of first-episode, treatment-naive young adults with MDD. Increased functional connectivity in anterior medial regions of the resting-state DMN was associated with rumination, whereas decreased functional connectivity in posterior medial regions was associated with OGM. These results provide new evidence for the importance of the DMN in the pathophysiology of MDD and suggest that abnormal DMN activity may be an MDD trait. Key Words: Autobiological memory, default mode network, independent component analysis, major depressive disorder, restingstate, rumination ver the past 50 years, a large body of psychological research has accumulated with respect to the question of the self (1). More recently, there has been considerable interest in neural correlates of the self (2). Clinically, patients with major depressive disorder (MDD) present with a number of abnormal psychological and psychiatric symptoms characterized by multiple self-abnormalities, such as rumination (3–5), a mode of responding to distress that involves repetitively and passively focusing on negative self-relevant information (6), and overgeneral autobiographical memory (OGM) phenomena (7), the tendency to recall categories of events when asked to provide specific instances from one’s life. Although these clinical phenomena are well-known, the neural networks and the pathophysiological mechanisms underlying them remain unclear. Functional neuroimaging studies performed with healthy subjects have implicated recruitment of several cortical and subcortical midline brain regions, such as the medial prefrontal cortex (MPFC),

O

From the Medical Psychological Institute of the Second Xiangya Hospital (XZ, XW, JX, MZ, SY); the Department of Radiology of the Third Xiangya Hospital (JL, WW), Central South University; and the School of Humanities and Social Sciences (XZ), National University of Defense Technology, Changsha, Hunan, China. Authors XZ and XW contributed equally to this work. Address correspondence to Shuqiao Yao, M.D., Ph.D., Central South University, Medical Psychological Institute of the Second Xiangya Hospital, 139 Renmin (M) Road, Changsha, Hunan 410011, China; E-mail: [email protected]. Received Jan 8, 2011; revised Sep 28, 2011; accepted Oct 30, 2011.

0006-3223/$36.00 doi:10.1016/j.biopsych.2011.10.035

the anterior cingulate cortex (ACC), the posterior cingulate cortex (PCC), the precuneus, and the dorsomedial thalamus, in self-related processing. Moreover, there is strong evidence that these medial regions also exhibit high levels of activity at rest and become suspended or deactivated when specific goal-directed behavior is needed, which means these regions might be crucial components of the so-called default mode network (DMN) in relation to restingstate brain function (8,9). The DMN is a large-scale brain network that encompasses a specific set of brain regions, including MPFC, PCC/precuneus, and medial, lateral, and inferior parietal regions, which indeed demonstrate a consistent pattern of deactivation during the initiation of task-related activity (8 –10). The DMN has recently been shown to be important in self-referential activities, being described as lowfrequency toggling between a task-independent, self-referential, and introspective state and an extrospective state that ensures the individual is alert and attentive to unexpected or novel environmental events (11,12). More specifically, autobiographical memory retrieval, which involves projecting the self into the past, engages a common set of brain regions largely overlapping with the DMN (13,14). In addition, one recent study that directly examined the neural systems underlying rumination showed that the DMN activity was highest in response to a feel strategy condition (designed to facilitate rumination) (15). Considering that both rumination and OGM are risk factors for the onset and course of depression, investigating the relationship between the DMN in MDD and MDDrelated pathopsychological characteristics is of great interest. According to previous studies, there are striking differences between the DMNs of MDD patients and those of healthy control subjects (3,16,17). Studies have shown task-related altered negative blood oxygenation level-dependent responses in regions of the DMN in MDD patients (18,19). For instance, Grimm et al. (20) reported evidence of an association of altered neural activity in a BIOL PSYCHIATRY 2012;71:611– 617 © 2012 Society of Biological Psychiatry

612 BIOL PSYCHIATRY 2012;71:611– 617 subcortical-cortical midline network with increased negative selfattribution during judgment of self-relatedness in subjects with MDD. Additionally, using a seed-based connectivity approach and a self-related short-term memory task, Berman et al. (21) revealed the presence of subgenual-cingulate cortex hyperconnectivity in MDD subjects during off-task periods but not during task engagement; connectivity of this area with the posterior cingulate is highly related to behavioral assays of depressive rumination. However, Berman et al. (21) suggested that studying on-task behavior may mitigate some differences between MDD and healthy control subjects in rumination, which is consistent with the suggestion by Raichle et al. (9) to study the resting state in both health and disease rather than focusing solely on reflexive or on-task performance. To date, few studies have directly examined depression-related changes in resting-state DMN activity, and those that have been done reported inconsistent functional connectivity (FC) patterns. Using independent-components analysis, Greicius et al. (22) reported increased FC in the subgenual cingulate cortex and thalamus in MDD patients but did not observe any decreases in FC. Sheline et al. (23) and Zhou et al. (24) also found that depressed subjects had increased functional connectivities located in the PCC and MPFC, two core components of the DMN. On the other hand, Bluhm et al. (25) described reduced FC between the PCC/precuneus and the bilateral caudate in MDD subjects relative to control subjects but did not report observing any regions with increased FC. Major depressive disorder-related decreased resting-state functional connectivities in major depression have also been reported in two other studies (26,27). Even though the discrepancy between those studies mentioned above could be ascribed, in part, to differences in sample size, medications, and analysis methods, it should be noted that DMN alterations related to MDD may indeed be present in a complex model, not as pure increases or decreases (2,28). A recent metaanalysis of imaging studies on the self suggested that the anterior and posterior cortical midline structures may be particularly influential in negative and positive affective self-referential processing modes (2). Additionally, in an intrinsic connectivity network study, Zhou et al. (29) described a dissociation between the anterior and posterior parts of the DMN. In fact, Greicius et al. (22) have hypothesized that persistent, emotionally laden, self-reflective tendencies like rumination in persons with MDD might generate increased FC in medial prefrontal portions of the network, whereas episodic memory retrieval impairments might lead to reduced FC in the more posterior regions of the network. However, this hypothesis was based only on prior knowledge of the recruited regions and lacks the support of direct behavioral correlates of the resting-state DMN. The first goal of the present study was to investigate the altered FC pattern of DMN underlying MDD psychopathology, using a study cohort of first-episode, treatment-naive young adults with MDD and carefully matched healthy control subjects. To our knowledge, no studies directly examining the relationship between behavior measurements of the self and the resting-state DMN connectivity in MDD have been reported. Therefore, our second goal was to examine the association of the altered resting-state DMN connectivity with rumination and OGM, two MDD-related pathopsychological characteristics related to the self in depressed patients. We hypothesized that MDD and healthy control subjects would show a dissociation connectivity pattern between anterior and posterior parts of resting-state DMN and that altered DMN connectivities in MDD subjects would correlate with indices of rumination and OGM. www.sobp.org/journal

X. Zhu et al.

Methods and Materials Participants Patients with MDD were recruited from the outpatient department at Second Xiangya Hospital of Central South University in Changsha, Hunan, China. Closely matched healthy subjects were recruited through advertisements from several colleges in Changsha. All participants were aware of the purpose of the study before participating in the study and signed an informed consent form approved by the Ethics Committee of the Second Xiangya Hospital. Finally, 37 patients with MDD (21 female patients) and 37 matched healthy control subjects (20 female subjects) were recruited. All subjects were right-handed. The shared exclusion criteria for patients and control subjects were meeting DSM-IV criteria for any other psychiatric disorders other than MDD, any contraindications to magnetic resonance imaging scanning, alcohol or substance abuse history within the past year, history of receiving electroconvulsive therapy in the prior 6 months, chronic neurological disorders, severe or acute medical conditions, a history of loss of consciousness, head trauma, pregnancy, or breast-feeding. Patients with MDD were diagnosed according to the Structured Clinical Interview for DSM-IV (18) by independent assessments of two psychiatrists (X.Z. and S.Y.). All of the patients were experiencing their first episode of depression and had never received psychopharmacological medication. Measures Depression. Depressive severity was assessed using the Center for Epidemiologic Studies Depression Scale (CES-D) (30), a 20item self-report instrument designed to assess depressive symptoms in the general population. In the current study, the CES-D exhibited high internal consistency, with a Cronbach’s value of .94. Rumination. Level of rumination was assessed using the rumination subscale of the Cognitive Emotion Regulation Questionnaire (CERQ) (31), which focuses on the feelings and thoughts that are associated with negative events. The rumination subscale of the CERQ exhibited high degrees of reliability and validity in both the English version (31) and the Chinese version (32). Autobiographical Memory. Autobiographical memory was assessed with an extended version of the Dutch autobiographical memory test (AMT) (33). Briefly, 12 cue words were presented orally in a fixed order, alternating between six positive and six negative words. Given that OGM has been reported as a robust and replicable phenomenon among the clinically depressed, we chose to use AMTOGM as an indicator of depression to correlate with DMN activity. (For more detail about the measures above, please see Supplement 1.) Image Acquisition All magnetic resonance imaging data were acquired using a 1.5-T MRI scanner (Magnetom Symphony; Siemens, Erlangen, Germany) with a standard head coil. During scanning, all the participants were asked to rest with their eyes closed and to try not to think of anything systematically. After scanning, the participants were asked about their statement during scanning. Two patients and two healthy control subjects were excluded according to the criteria mentioned above. Therefore, only the data of 35 participants in each group were analyzed. Functional images were obtained axially using a single-shot, gradient-recalled echo-planar imaging sequence parallel to the line of the anterior-posterior commissure. The acquisition parameters were repetition time ⫽ 2000 msec; echo time ⫽ 40 msec; flip angle ⫽ 90°; field of view ⫽ 24 cm; matrix ⫽ 64 ⫻ 64, 26 slices; slice thickness ⫽ 5.0 mm, without gap; and voxel size ⫽ 5 ⫻ 3.85 ⫻ 3.85

BIOL PSYCHIATRY 2012;71:611– 617 613

X. Zhu et al. mm3. A total of 150 volumes were acquired over 300 sec for each subject. High-resolution T1-weighted images were also acquired with a three-dimensional spoiled gradient-recalled sequence in an axial orientation: repetition time ⫽ 8.5 msec; echo time ⫽ 3.2 msec; flip angle ⫽ 15°; field of view ⫽ 25 cm; matrix ⫽ 256 ⫻ 256; 176 slices; and slice thickness ⫽ 1.0 mm, without gap. Image Preprocessing All imaging preprocessing steps were conducted using statistical parametric mapping (SPM2; Functional Imaging Laboratory, University College London, London, United Kingdom; http://www. fil.ion.ucl.ac.uk/spm). After discarding of the first 10 volumes of each functional time series, slice timing, and realignment of head motion, data from three patients and two healthy subjects were excluded because their translation or rotation exceeded ⫾1.5 mm or ⫾1.5°. The images were then spatially normalized to a standard template (Montreal Neurological Institute, Montreal, Quebec, Canada), resampled to 3-mm cubic voxels, and spatially smoothed by convolution with an isotropic Gaussian kernel (full-width at halfmaximum ⫽ 8 mm) to decrease high spatial frequency noise. Independent Component Analysis and Identification of DMNs Spatial independent component analysis (ICA) was conducted for all 65 participants using the Informix algorithm in the Group ICA of fMRI Toolbox (GIFT) software (Medical Image Analysis Lab, University of New Mexico, Albuquerque, New Mexico; http://icatb.sourceforge. net/). The data obtained from each subject were decomposed into 30 spatially separated components using GIFT software. The number of independent components (ICs) was determined by using the minimum description length criteria (34) and the resulting averaged 30 ICs. Then, the averaged IC was used for each subject for ICA separation. Each data set was processed by means of principal component analysis to reduce its dimensions. Then, a single ICA was performed on each subject, followed by a back reconstruction of single-subject time courses and spatial maps from the raw data. For each IC, the time courses correspond to the waveform of a specific pattern of coherent brain activity, and the intensity of this pattern of brain activity across the voxels is expressed in the associated spatial map (35). To display voxels relevant to a particular IC, the intensity values in each map were converted to z values (22,36). After ICA separation, a template of the DMN was used to select the greatest best-fit component for each subject. The standard DMN template was from a meta-analytic modeling provided by Angela R. Laird, Ph.D. (Research Imaging Institute, University of Texas Health Science Center, San Antonio, Texas) (37) (Supplement 1). Statistical Analysis of the DMN After being extracted from all subjects, the best-fit components representing the DMN were gathered in each group separately for a

random effect analysis using the one-sample t test. Thresholds were set at p ⬍ .05 (false discovery rate [FDR] correction). Subsequently, two-sample t tests were used to compare the best-fit components between two groups (p ⬍ .05 with FDR correction). The group comparisons were restricted (masked) to the voxels within the corresponding DMN. The mask was created by combining the DMN regions of the patients and the control subjects, which were obtained from one-sample t test results (p ⬍ .05 with FDR correction). To obtain a measure of effect size, we calculated Cohen’s d for the mean Z score within the clusters. Correlation Analyses To examine the association of the DMN with the level of rumination and the overgeneral memory, voxel-based correlations were implemented to the best-fit component of each depressed subject and CERQ-rumination and AMT-OGM scores, respectively. To avoid the influence of the depression severity and course of depression, CES-D total score and illness duration were introduced as covariants in the statistical model. Additionally, CERQ-rumination score was controlled when looking at AMT-OGM correlations, and AMTOGM score was controlled when looking at CERQ-rumination correlations. The results were constrained within the mask of the DMN produced in the above step. The statistical results were corrected using AlphaSim program in the Resting-State fMRI Data Analysis Toolkit (REST) software (Forum of resting-state fMRI, http://restfmri.net/forum/index.php) at .05 significance level (combined height threshold p ⬍ .01 and a minimum cluster size ⫽ 33, with the parameters: ⫺fwhm 8 ⫺rmm 5 ⫺pthr .01 ⫺iter 1000 ⫺mask dmn_mask).

Results Participants The demographic and clinical characteristics of the two groups of participants are shown in Table 1. The groups did not differ significantly in terms of age or gender. Compared with healthy control subjects, MDD patients showed higher levels of CES-D (t ⫽ 11.12, p ⬍ .001), CERQ-rumination (t ⫽ 6.24, p ⬍ .001), and AMTOGM (t ⫽ 11.77, p ⬍ .001). Additional correlation analyses showed that there was significantly positive correlation between the CES-D and AMT-OGM (r ⫽ .36, p ⬍ .01) in the present study, whereas no significant correlations were found between the scores of the CES-D and CERQ (r ⫽ .13, p ⬎ .05) and CES-D and CERQ-rumination (r ⫽ .19, p ⬎ .05). DMN Identification The DMNs were extracted from each subject in both the patient and the control groups based on goodness-of-fit measurements. Significant differences in the goodness of fit were found for the DMN (t ⫽ 2.56, p ⬍ .05) between the MDD patients (1.21 ⫾ .10) and control subjects (1.48 ⫾ .09).

Table 1. Demographic and Clinical Characteristics of the MDD and Control Groups Characteristic Age (Mean ⫾ SD Years) Gender (Female/Male) CES-D (Mean ⫾ SD) Rumination (Mean ⫾ SD) AMT-OMG (Mean ⫾ SD) Age at Illness Onset (Mean ⫾ SD Years) Illness Duration (Mean ⫾ SD Months)

MDD

Control

t/␹2

p

Cohen’s d

20.53 ⫾ 1.78 18/14 38.03 ⫾ 6.67 13.91 ⫾ 3.58 10.31 ⫾ 1.33 20.06 ⫾ 2.13 10.53 ⫾ 7.10

20.30 ⫾ 1.63 19/14 17.76 ⫾ 7.94 9.27 ⫾ 2.30 5.39 ⫾ 1.97 NA NA

.54 .02 11.12 6.24 11.77

.59 .90 0 0 0

.13 2.76 1.54 2.93

AMT, autobiographical memory test; CES-D, Center for Epidemiologic Studies Depression Scale; MDD, major depressive disorder; NA, not applicable; OMG, overgeneral autobiographical memory; SD, standard deviation.

www.sobp.org/journal

614 BIOL PSYCHIATRY 2012;71:611– 617

X. Zhu et al. the DMNs of the two groups. Compared with the healthy control subjects, the depressed patients showed increased resting functional connectivity in the dorsal MPFC/ventral ACC (vACC), ventral MPFC (vMPFC), and medial orbital PFC but decreased functional connectivity in the PCC/precuneus, right AG, and left AG/precuneus (Figure 1C, Table 2). In addition, the mean Z score and standard deviation of the MDD and control groups for each cluster and the effect size (Cohen’s d) of each cluster are also reported in Table 2. The Cohen’s d values of the four clusters were conventionally interpreted as constituting a large effect size. Correlation Analyses With the CES-D score, illness duration, and AMT-OGM being controlled as covariants, the CERQ-rumination subscores were correlated with the z values of the voxels within the vMPFC and vACC at .05 significance level (combined height threshold p ⬍ .01 and a minimum cluster size ⫽ 33) (Figure 2) in MDD patients. In addition, the AMT-OGM was inversely correlated with the z values of the voxels within the precuneus and AG regions at .05 significance level (combined height threshold p ⬍ .01 and a minimum cluster size ⫽ 33), with the CES-D score, illness duration, and CERQ-rumination subscore being controlled (Figure 3). The z values of other brain regions with aberrant functional connectivity did not correlate with CERQ-rumination subscore or AMT-OGM.

Discussion Using ICA methodology, our study demonstrated altered resting-state DMNs in first-episode, treatment-naive young adults with MDD. Compared with healthy control subjects, MDD patients showed greater functional connectivity in the MPFC/vACC areas, as well as decreased functional connectivity in the PCC/precuneus and bilateral AG areas. The relatively large effect size (Cohen’s d) of the findings on those clusters lends a quantitative measure to the qualitative impression. Furthermore, the MDD patients’ rumination scores correlated positively with their functional connectivity in the vMPFC and vACC, while the overgeneral memory index values of MDD patients were negatively associated with functional connectivity in the PCC/precuneus and AG areas. Our results demonstrate the presence of aberrant activities of the resting-state DMN in the brains of MDD patients and provide the first empirical evidence for an association between rumination/OGM and abnormal neural activity in the resting-state DMN. In the present study, the MPFC and vACC were found to show increased functional connectivity in patients with MDD. These two brain regions are commonly regarded as key brain regions in MDD, with abnormalities having been observed in blood oxygenation

Figure 1. Axial images showing the group default mode network extracted by independent component analyses in depressed subjects (A) (n ⫽ 32, p ⬍ .05 with false discovery rate [FDR] correction) and in healthy control subjects (B) (n ⫽ 33, p ⬍ .05 with FDR correction). According to the results of two-sample t tests, the major depressive disorder patients showed significantly increased functional connectivity in the medial prefrontal cortex and ventral anterior cingulate cortex (labeled red) but significantly decreased functional connectivity in the posterior cingulate cortex/precuneus and angular gyrus (labeled blue) relative to the healthy control subjects (C) (p ⬍ .05 with FDR correction). The t score bars are shown at right.

Statistical Comparison of DMNs One-sample t tests (p ⬍ .05 with FDR correction) revealed the respective spatial pattern of the DMN in the MDD patients and healthy control subjects (Figure 1A and 1B). Increased functional connectivity was identified in the MPFC, PCC/precuneus, bilateral angular gyrus (AG), inferior temporal cortex, and medial temporal lobes in both groups, which correspond to the typical distribution of the DMN. Furthermore, the two-sample t tests (p ⬍ .05 with FDR correction) showed that there were significant differences between

Table 2. Brain Areas with Significantly Increased and Decreased Activation Between Young Adults with MDD and Control Subjects Mean Z Score Within the Cluster Brain Regions dMPFC/vACC vMPFC mOPFC PCC/Precuneus Right AG Left AG/Precuneus

BA

x, y, z

t

9/24/32 10/11 11/12 30/23/7/31 39 39/19

⫺3 33 11 ⫺9 69 12 ⫺6 56 ⫺15 15 ⫺66 24 45 ⫺72 33 ⫺36 ⫺75 39

4.47 6.17 5.55 ⫺6.97 ⫺7.20 ⫺6.75

Voxels

MDD

Control Subjects

Cohen’s d

1384

1.49 ⫾ .42

.72 ⫾ .38

1.92

777 294 163

.62 ⫾ .41 .67 ⫾ .40 .86 ⫾ .52

1.43 ⫾ .33 1.14 ⫾ .21 1.92 ⫾ .48

⫺2.18 ⫺1.47 ⫺2.11

AG, angular gyrus; BA, Brodmann area; dMPFC, dorsal medial prefrontal cortex; MDD, major depressive disorder; mOPFC, medial orbital prefrontal cortex; PCC, posterior cingulate cortex; vACC, ventral anterior cingulate cortex; vMPFC, ventral medial prefrontal cortex.

www.sobp.org/journal

X. Zhu et al.

Figure 2. Sagittal view of the voxel-wise correlation analysis between z values of the voxels within ventral medial prefrontal cortex (Montreal Neurological Institute coordinates: x ⫽ 6, y ⫽ 50, z ⫽ 12) and ventral anterior cingulate cortex (Montreal Neurological Institute coordinates: x ⫽ 5, y ⫽ 42, z ⫽ 1) and rumination scores in the major depressive disorder patients at .05 significance level (combined height threshold p ⬍ .01 and a minimum cluster size ⫽ 33).

level-dependent functional magnetic resonance imaging (fMRI) activation, as well as in baseline metabolism or perfusion (38 – 40). Major depressive disorder patients have also been reported to have structural differences in the MPFC (18) and vACC (41). Recently, a meta-analysis study using a cross-species translational approach to compare human data with animal data on the functional anatomy and neurochemical modulation of resting-state activity in depression consistently suggested that resting-state hyperactivity in depression occurs in subcortical and cortical midline regions (42). Our study extends these prior findings by providing new evidence for abnormal resting-state neural activity in the MPFC and vACC in MDD patients. It is worth noting that we observed congruently increased functional connectivity in the MPFC and vACC in the resting-state data of MDD subjects. In fact, there has been a convergence of neuroimaging, neuroelectrophysiological, and therapeutic studies showing functional associations between these two brain regions in MDD. In a recent resting-state study using multivariate Granger causality analysis, Hamilton et al. (43) reported that activities in the MPFC and vACC were mutually reinforcing in MDD. In another study combining deep brain stimulation and positron emission tomography, Mayberg et al. (40) showed that reducing activity in the vACC through electrode stimulation of adjacent white-matter tracts also decreased the MPFC activity in depressed subjects (40). In addition, Wu et al. (44) demonstrated greater metabolism in both the vACC and the MPFC of depressed patients who responded therapeutically to sleep deprivation. Recently, both task-related and resting-state functional neuroimaging studies demonstrated that the MPFC and vACC were strongly associated with modulation of emotional behavior and self-referential processing (17,18,41). Our findings of correlations between rumination score and the z values of the voxels in the MPFC and vACC further support the view that the MPFC and vACC are involved in self-abnormalities in MDD. As a crucial component of the cognitive vulnerabilities to depression, rumination is considered lengthy, repetitive thought about a current stressor and negative affect and closely correlated with private self-consciousness, self-reflection, and self-focused attention (31,45,46). Since the DMN appears to mediate internally generated thought processes and is typically inhibited in tasks that require subjects to actively attend to

BIOL PSYCHIATRY 2012;71:611– 617 615 cognitively demanding, external stimuli, it is not surprising that the subjects who obtained higher rumination scores showed greater increases in the functional connectivity of the MPFC and vACC in their DMNs (41). Indeed, prior task-related neurophysiological studies have implicated the MPFC and ACC in the neural mechanisms of rumination (16,47). Our study provides novel evidence for the notion that DMN abnormalities in MDD reflect the ruminative nature of MDD patients and further suggests that the resting-state signal in the MPFC and vACC may be a marker for a ruminative response style in depression. The current data also showed reduced functional connectivity in posterior regions of the DMN (PCC/precuneus and AG) in MDD patients, and less functional connectivity in these areas, especially in precuneus and AG, was associated with higher OGM scores. The PCC and adjacent precuneus, which are centrally located in the DMN, have consistently been reported as having aberrant connectivity in resting-state fMRI scans of individuals with mental disorders (17,28). However, there has been little direct evidence for abnormalities in the PCC/precuneus in resting-state DMN connectivity in MDD. Greicius et al. (22) found altered connectivity in the cuneus/precuneus but not in the PCC, whereas Bluhm et al. (25) found decreased connectivity between the PCC/precuneus and bilateral caudate in resting-state DMN. Our results provide new evidence for reduced functional connectivity within the DMN of MDD patients and further support the notion that the PCC/precuneus and bilateral AG are involved in the pathophysiology of OGM in MDD. A considerable body of evidence for a high level of OGM in depression has accumulated in recent decades (7,48). However, the neural mechanism underlying OGM phenomena has yet to be delineated. Previous task-related neuroimaging studies have indicated that the PCC/precuneus plays an important role in successful retrieval of autobiographical and self-relevant information (14,49). Recently, Conway and Pleydell-Pearce (50) and Conway et al. (51) suggested that autobiographical memory should be viewed as one part of a larger self-memory system with two functions: maintaining adaptive correspondence and ensuring self-coherence. This assertion is consistent with the functional characteristics of restingstate brain and DMN. Last year, a systematic and quantitative metaanalysis using the activation likelihood estimation approach was conducted to investigate the common neural basis of autobiographical memory and default mode. In both domains, correspondence was found within the precuneus and PCC (14). Recent fMRI research results from Spreng and Grady (52) also showed that autobiographical remembering engaged these midline DMN struc-

Figure 3. Sagittal views of the voxel-wise correlation analysis between z values of the voxels within the precuneus (Montreal Neurological Institute coordinates: x ⫽ ⫺5, y ⫽ ⫺61, z ⫽ 25) and angular gyrus (Montreal Neurological Institute coordinates: x ⫽ ⫺42, y ⫽ ⫺72, z ⫽ 39) and autobiographical memory test-overgeneral autobiographical memory scores in the major depressive disorder patients at .05 significance level (combined height threshold p ⬍ .01 and a minimum cluster size ⫽ 33).

www.sobp.org/journal

616 BIOL PSYCHIATRY 2012;71:611– 617 tures to a greater degree than it did other brain areas. To our knowledge, no study, to date, has investigated the relationship between autobiographical memory index score and functional connectivity in resting-state DMN in subjects with MDD. Our findings provide new pathophysiological evidence that the DMN supports the autobiographical memory process and further corroborate the Greicius et al. (22) hypothesis that altered functional connectivity in posterior regions of the DMN are associated with episodic memory impairments in MDD. Further research is needed to distinguish the mechanism underlying the dissociation between anterior and posterior functional connectivity, such as some compensatory mechanism. As one of the group ICA approaches, the reliability or consistency of GIFT software has been addressed by a number of studies. Chen et al. (53) showed the consistency of intrinsic brain networks derived by ICA across five sessions within 16 days. Additionally, the results of Franco et al. (54) suggested high levels of interrater reliability of ICA-derived DMN. Furthermore, GIFT software has been widely used to examine clinical populations, such as patients with depression (22), Alzheimer’s disease (36), schizophrenia (55), epilepsy (56), and social anxiety disorder (57). However, group ICA also has some limitations. First, there is no ideal means for assessing the optimal number of components (58). Second, the results of ICA may vary with different concatenation orders of the individual fMRI data sets (59). Recently, some new algorithms were proposed to cover the shortages of GIFT, such as subject order-independent group ICA (59) and dual regression (58). To further verify the dissociation pattern in resting-state DMN connectivity in MDD patients, we reprocessed the data with the subject order-independent group ICA algorithm and obtained similar results (see Supplement 1 for details). There are several other potential methodological limitations to the interpretation of the data. First, we could not completely avoid the effects of physiological noise during resting fMRI scans, such as cardiac and respiratory pulsation, noise that can alias into the resting-state low-frequency range (.01–.08 Hz) (60). Second, the patients with MDD recruited in this study had relatively mild MDD and no comorbid conditions, so we may be cautious in generalizing the findings of this study to more ill individuals with comorbid diagnoses. In summary, we have demonstrated dissociation between anterior and posterior functional connectivity within resting-state DMNs in first-episode, treatment-naive young adults with MDD. Our findings suggest that increased functional connectivity in anterior medial cortex and decreased functional connectivity in posterior medial regions of the resting-state DMN are associated with rumination and OGM, respectively, as hallmarks of MDD. Our results highlight the important role of the DMN in the pathophysiology of major depression and suggest that abnormal DMN activity may be a trait associated with MDD. This work was supported by the National Natural Science Foundation of China (Grant numbers 30670709 to SY, 30700236 to XW). We are grateful to Dr. Zhiqiang Zhang and Dr. Yuan Zhong from the Department of Medical Imaging in Clinical School of Medical College, Nanjing University, and Wei Liao from the University of Electronic Science and Technology of China for their help on the image analysis and to Dr. Jian Liao for providing expert magnetic resonance imaging support. We thank all the participants and their families. The authors report no biomedical financial interests or potential conflicts of interest. Supplementary material cited in this article is available online. www.sobp.org/journal

X. Zhu et al. 1. Northoff G (2004): Philosophy of the Brain. The Brain Problem. Amsterdam: John Benjamins Publishing. 2. Northoff G, Heinzel A, de Greck M, Bermpohl F, Dobrowolny H, Panksepp J (2006): Self-referential processing in our brain—a meta-analysis of imaging studies on the self. Neuroimage 31:440 – 457. 3. Grimm S, Ernst J, Boesiger P, Schuepbach D, Hell D, Boeker H, Northoff G (2009): Increased self-focus in major depressive disorder is related to neural abnormalities in subcortical-cortical midline structures. Hum Brain Mapp 30:2617–2627. 4. Northoff G (2007): Psychopathology and pathophysiology of the self in depression—neuropsychiatric hypothesis. J Affect Disord 104:1–14. 5. Rimes KA, Watkins E (2005): The effects of self-focused rumination on global negative self-judgements in depression. Behav Res Ther 43:1673– 1681. 6. Nolen-Hoeksema S, Wisco BE, Lyubomirsky S (2008): Rethinking rumination. Perspect Psychol Sci 3:400 – 424. 7. Sumner JA, Griffth JW, Mineka S (2010): Overgeneral autobiographical memory as a predictor of the course of depression: A meta-analysis. Behav Res Ther 48:614 – 625. 8. Greicius MD, Krasnow B, Reiss AL, Menon V (2003): Functional connectivity in the resting brain: A network analysis of the default mode hypothesis. Proc Natl Acad Sci U S A 100:253–258. 9. Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL (2001): A default mode of brain function. Proc Natl Acad Sci U S A 98:676 – 682. 10. Andrews-Hanna JR, Reidler JS, Sepulcre J, Poulin R, Buckner RL (2010): Functional-anatomic fractionation of the brain’s default network. Neuron 65:550 –562. 11. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME (2005): The human brain is intrinsically organised into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 102:9673–9678. 12. Fransson P (2006): How default is the default mode of brain function? Further evidence from intrinsic BOLD signal fluctuations. Neuropsychologia 44:2836 –2845. 13. Buckner RL, Andrews-Hanna JR, Schacter DL (2008): The brain’s default network: Anatomy, function, and relevance to disease. Ann N Y Acad Sci 1124:1–38. 14. Spreng RN, Mar RA, Kim AS (2009): The common neural basis of autobiographical memory, prospection, navigation, theory of mind, and the default mode: A quantitative meta-analysis. J Cogn Neurosci 21:489 – 510. 15. Kross E, Davidson M, Weber J, Ochsner K (2009): Coping with emotions past: The neural bases of regulating affect associated with negative autobiographical memories. Biol Psychiatry 65:361–366. 16. Johnson MK, Nolen-Hoeksema S, Mitchell KJ, Levin Y (2009): Medial cortex activity, self-reflection and depression. Soc Cogn Affect Neurosci 4:313–327. 17. Sheline YI, Barch DM, Price JL, Rundle MM, Vaishnavi SN, Snyder AZ, et al. (2009): The default mode network and self-referential processes in depression. Proc Natl Acad Sci U S A 106:1942–1947. 18. Drevets WC, Price JL, Furey ML (2008): Brain structural and functional abnormalities in mood disorders: Implications for neurocircuitry models of depression. Brain Struct Funct 213:93–118. 19. Gotlib IH, Hamilton JP (2008): Neuroimaging and depression: Current status and unresolved issues. Curr Dir Psychol Sci 17:159 –163. 20. Grimm S, Boesiger P, Beck J, Schuepbach D, Bermpohl F, Walter M, et al. (2009): Altered negative BOLD responses in the default-mode network during emotion processing in depressed subjects. Neuropsychopharmacology 34:932–943. 21. Berman MG, Peltier S, Nee DE, Kross E, Deldin PJ, Jonides J (2011): Depression, rumination and the default network. Soc Cogn Affect Neurosci 6:548 –555. 22. Greicius MD, Flores BH, Menon V, Glover GH, Solvason HB, Kenna H, et al. (2007): Resting-state functional connectivity in major depression: Abnormally increased contributions from subgenual cingulate cortex and thalamus. Biol Psychiatry 62:429 – 437. 23. Sheline YI, Price JL, Yan Z, Mintun MA (2010): Resting-state functional MRI in depression unmasks increased connectivity between networks via the dorsal nexus. Proc Natl Acad Sci U S A 107:11020 –11025. 24. Zhou Y, Yu C, Zheng H, Liu Y, Song M, Qin W, et al. (2010): Increased neural resources recruitment in the intrinsic organization in major depression. J Affect Disord 121:220 –230.

BIOL PSYCHIATRY 2012;71:611– 617 617

X. Zhu et al. 25. Bluhm R, Williamson P, Lanius R, Theberge J, Densmore M, Bartha R, et al. (2009): Resting state default-mode network connectivity in early depression using a seed region-of-interest analysis: decreased connectivity with caudate nucleus. Psychiatry Clin Neurosci 63:754 –761. 26. Anand A, Li Y, Wang Y, Wu J, Gao S, Bukhari L, et al. (2005): Activity and connectivity of brain mood regulating circuit in depression: A functional magnetic resonance study. Biol Psychiatry 57:1079 –1088. 27. Veer IM, Beckmann CF, van Tol MJ, Ferrarini L, Milles J, Veltman DJ, et al. (2010): Whole brain resting-state analysis reveals decreased functional connectivity in major depression. Front Syst Neurosci 4:41. 28. Broyd SJ, Demanuele C, Debener S, Helps SK, James CJ, Sonuga-Barke EJ (2009): Default-mode brain dysfunction in mental disorders: A systematic review. Neurosci Biobehav Rev 33:279 –296. 29. Zhou J, Greicius M, Gennatas E, Growdon ME, Jang JY, Rabinovici GD, et al. (2010): Divergent network connectivity changes in behavioural variant frontotemporal dementia and Alzheimer’s disease. Brain 133:1352– 1367. 30. Radloff LS (1977): The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Meas 1:385– 401. 31. Garnefski N, Kraaij V, Spinhoven P (2001): Negative life events, cognitive emotion regulation and emotional problems. Pers Individ Dif 30:1311– 1327. 32. Zhu X, Auerbach RP, Yao S, Abela JRZ, Xiao J, Tong X (2008): Psychometric properties of the cognitive emotion regulation questionnaire: Chinese version. Cogn Emot 22:288 –307. 33. Williams JM, Broadbent K (1986): Autobiographical memory in suicide attempters. J Abnorm Psychol 95:144 –149. 34. Li YO, Adali T, Calhoun VD (2007): Estimating the number of independent components for functional magnetic resonance imaging data. Hum Brain Mapp 28:1251–1266. 35. Mantini D, Perrucci MG, Del Gratta C, Romani GL, Corbetta M (2007): Electrophysiological signatures of resting state networks in the human brain. Proc Natl Acad Sci U S A 104:13170 –13175. 36. Greicius MD, Srivastava G, Reiss AL, Menon V (2004): Default-mode network activity distinguishes Alzheimer’s disease from healthy aging: Evidence from functional MRI. Proc Natl Acad Sci U S A 101:4637– 4642. 37. Laird AR, Eickhoff SB, Li K, Robin DA, Glahn DC, Fox PT (2009): Investigating the functional heterogeneity of the default mode network using coordinate-based meta-analytic modeling. J Neurosci 29:14496 –14505. 38. Gonul AS, Kula M, Bilgin AG, Tutus A, Oguz A (2004): The regional cerebral blood flow changes in major depressive disorder with and without psychotic features. Prog Neuropsychopharmacol Biol Psychiatry 28:1015–1021. 39. Mayberg HS (1997): Limbic-cortical dysregulation: A proposed model of depression. J Neuropsychiatry Clin Neurosci 9:471– 481. 40. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, et al. (2005): Deep brain stimulation for treatment-resistant depression. Neuron 45:651– 660. 41. Drevets WC, Savitz J, Trimble M (2008): The subgenual anterior cingulate cortex in mood disorders. CNS Spectr 13:663– 681. 42. Alcaro A, Panksepp J, Witczak J, Hayes DJ, Northoff G (2010): Is subcortical-cortical midline activity in depression mediated by glutamate and GABA? A cross-species translational approach. Neurosci Biobehav Rev 34:592– 605. 43. Hamilton JP, Chen G, Thomason ME, Schwartz, Gotlib ME (2011): Investigating neural primacy in major depressive disorder: Multivariate

44.

45.

46. 47.

48.

49. 50.

51. 52.

53.

54.

55.

56.

57.

58.

59.

60.

granger causality analysis of resting-state fMRI time-series data. Mol Psychiatry 16:763–772. Wu J, Buchsbaum MS, Gillin JC, Tang C, Cadwell S, Wiegand M, et al. (1999): Prediction of antidepressant effects of sleep deprivation by metabolic rates in the ventral anterior cingulate and medial prefrontal cortex. Am J Psychiatry 156:1149 –1158. Alloy LB, Abramson LY, Hogan ME, Whitehouse, WG Rose DT, Robinson MS, et al. (2000): The Temple-Wisconsin Cognitive Vulnerability to Depression Project: Lifetime history of axis I psychopathology in individuals at high and low cognitive risk for depression. J Abnorm Psychol 109:403– 418. Nolen-Hoeksema S (1991): Responses to depression and their effects on the duration of depressive episodes. J Abnorm Psychol 100:569 –582. Siegle GJ, Steinhauer SR, Thase ME, Stenger VA, Carter CS (2002): Can’t shake that feeling: Event-related fMRI assessment of sustained amygdala activity in response to emotional information in depressed individuals. Biol Psychiatry 51:693–707. Van Vreeswijk M, deWilde E (2004): Autobiographical memory specificity, psychopathology, depressed mood and the use of the autobiographical memory test: A meta-analysis. Behav Res Ther 42:731–743. Cavanna AE (2007): The precuneus and consciousness. CNS Spectr 12: 545–552. Conway MA, Pleydell-Pearce CW (2000): The construction of autobiographical memories in the self-memory system. Psychol Rev 107:261– 288. Conway MA, Singer JA, Tagini A (2004): The self and autobiographical memory: Correspondence and coherence. Social Cognition 22:491–529. Spreng RN, Grady CL (2010): Patterns of brain activity supporting autobiographical memory, prospection, and theory of mind, and their relationship to the default mode network. J Cogn Neurosci 22:1112–1123. Chen S, Ross T, Zhan W, Myers C, Chuang K, Heishman S, et al. (2008): Group independent component analysis reveals consistent restingstate networks across multiple sessions. Brain Res 1239:141–151. Franco A, Pritchard A, Calhoun V, Mayer A (2009): Interrater and intermethod reliability of default mode network selection. Hum Brain Mapp 30:2293–2303. Jafri MJ, Pearlson G.D., Stevens M, Calhoun VD (2008): A method for functional network connectivity among spatially independent restingstate components in schizophrenia. Neuroimage 39:1666 –1681. Zhang Z, Lu G., Zhong Y, Tan Q, Yang Z, Liao W, et al. (2009): Impaired attention network in temporal lobe epilepsy: A resting fMRI study. Neurosci Lett 458:97–101. Liao W, Chen H, Feng Y, Mantini D, Gentili C, Pan Z, et al. (2010): Selective aberrant functional connectivity of resting state networks in social anxiety disorder. Neuroimage 52:1549 –1558. Zuo XN, Kelly C, Adelstein JS, Klein DF, Castellanos FX, Milham MP (2010): Reliable intrinsic connectivity networks: Test-retest evaluation using ICA and dual regression approach. Neuroimage 49:2163–2177. Zhang H, Zuo XN, Ma SY, Zang YF, Milham MP, Zhu CZ (2010): Subject order-independent group ICA (SOI-GICA) for functional MRI data analysis. Neuroimage 51:1414 –1424. Lowe MJ, Mock BJ, Sorenson JA (1998): Functional connectivity in single and multislice echoplanar imaging using resting-state fluctuations. Neuroimage 7:119 –132.

www.sobp.org/journal