Background

Bipolar disorder (BD) is a lifelong mood disorder characterized by extreme fluctuations between mania and depression [1]. It affects more than 1% of the world’s population [2] and 0.6% of the Chinese population [3]. However, the pathophysiology of BD is poorly understood. Recent advances in the field of neuroimaging have enhanced our understanding on the neurophysiology of this disease [4]. In particular, the application of resting-state functional magnetic resonance imaging (rs-fMRI), abnormal local spontaneous brain activity, and resting-state functional connectivity (rsFC) have been repeatedly observed in patients with BD [5, 6].

These studies indicate that a brain network, called the default mode network (DMN), is particularly interested in the studies of BD. The DMN consists of the medial prefrontal cortex (MPFC), posterior cingulate cortex (PCC), precuneus [7], inferior temporal and superior frontal cortices, hippocampus, and parahippocampal cortices [8, 9]. When a person is awake, but not actively engaged in a goal-directed task, this network is active with a high degree of rsFC between regions [10]. The DMN is involved in multiple cognitive and affective functions, such as emotional processing, mind wandering, recollection of experiences, self-referential mental activity, social behavior, and episodic memory processes [7], all of which are affected to varying degrees in BD [11]. Broyd and colleagues [12] reviewed the putative mechanisms for DMN-related dysfunction in mental disorders, including BD, from two perspectives. On the one hand, altered rsFC within the DMN indicates that the integrity of DMN is impaired, which may be related to the deficit of working memory and attention, which is often observed in BD. On the other hand, in cognitive tasks that require attention, the activity of certain brain regions increases (referred to as task-positive networks), while the activity of the DMN decreases [13]. This indicates an antagonistic relationship between the two networks [14]. However, in mood disorders, this relationship may be compromised, potentially related to irregular transformation of introspective and extrospective thinking. The disbalance between the DMN and the salience network (SN) (one of the task positive networks) may be associated with emotion, cognitive, and psychomotor symptoms of BD [15].

Empirically, researches have consistently reported that different affective state in BD patients have rsFC abnormalities in the DMN [16,17,18,19,20,21]. For example, one study of drug naive or unmedicated patients with BD-II depression showed decreased rsFC of the PCC with the MPFC and precuneus [19]. Another study found that unmedicated patients with BD during depressive episode exhibited weaker rsFC in the left MPFC and right precuneus compared with healthy controls [22]. Additionally, decreased rsFC between the DMN and the SN was reported in patients with BD, although the sample included patients in different affective states (i.e., manic, depressive, mixed, and euthymic phases) [23]. In these previous studies, researchers often employed independent component analysis (ICA) to investigate the DMN [17, 20,21,22], or selected specific brain regions as seeds for the rsFC analysis of the DMN based on prior hypotheses [15, 19, 23, 24]. While these studies have yielded promising findings, no consistent evidence has emerged regarding the specific brain regions within the DMN that exhibit abnormal spontaneous activity in patients with BD.

In addition to the two methods, other analytical approaches can be applied to the studies of the DMN, such as fractional amplitude of low-frequency fluctuation (fALFF) [25], regional homogeneity (ReHo) [26], and degree centrality (DC) [27]. fALFF measures the magnitude of spontaneous fluctuations in the blood oxygen level-dependent signal (BOLD) at the voxel level, which is considered to reflect spontaneous neural activity [28]. ReHo is defined as the Kendall’s coefficient concordance (KCC) between the time series of voxels and their nearest neighbors, essentially a local FC that reflects the regulation and coordination of local neuronal activity [26]. DC assesses the summed FC of each individual voxel with all voxels in the brain [29], providing an opportunity for an unbiased general search of abnormalities in the entire connectivity matrix of the whole brain functional connectome [30]. With a whole-brain search strategy, these indices can identify abnormalities in the brain. Using these indices, researchers have already found decreased brain activity in the DMN of BD patients [31,32,33]. Therefore, the abnormal spontaneous brain activity in the DMN can be unbiasedly identified using these functional indicators. Finding the abnormal rsFC of the DMN in an unbiased approach is possible by using the regions identified by these functional indicators as the seed region in patients with BD. In a previous study, the researchers have found that the MPFC, an important node of the DMN, showed abnormal brain activity as indicated by ReHo and DC in patients with bipolar mania [34].

Our research hypothesis proposes that BD patients exhibit abnormalities in the DMN, resulting in atypical spontaneous brain activity and altered functional connectivity within distinct brain regions of the DMN. We employed a whole-brain search strategy to identify the regions with specific impairments within the DMN and to examine the rsFC of the DMN in the patients with BD. Specifically, from the point of view of local spontaneous brain activity and functional network, we used multiple brain functional indices including fALFF, ReHo, and DC to unravel abnormal brain spontaneous activities in BD. Subsequently, using the cluster that showed significant between-group differences as a seed, rsFC analyses, which reflect functional interactions of spatially distributed brain regions, were performed to further characterize the abnormal functional connectivity of the seed region in patients with BD.

Methods

Participants

A total of 150 participants were recruited from the community (85 HCs) and the Bei**g Anding Hospital (65 patients with BD); the sample is the same as in our previous publication [35]. In summary, the patients and HC were matched for age, gender, and education. Twenty-three subjects (nine patients with BD and 14 HCs) were excluded from the final analysis due to excessive head motion (details are shown in the “fMRI data preprocessing”). All patients satisfied the diagnostic criteria of the Structured Clinical Interview for DSM-IV (SCID) [36], but in this study, we did not differentiate between BD-I and BD-II. Our inclusion and exclusion criteria were described in our previous study [35]. In this experiment, the Young Mania Rating Scale (YMRS) [37] scores of all patients were less than 7, thereby indicating that all patients were not in a manic mood state. And the Hamilton Depression Rating Scale (HAMD) [38] was used to assess patients’ depressive symptoms. Patients with HAMD scores of ≤ 7 were classified as euthymic patients (EP, n = 30), while those patients with scores above this threshold were considered to be in a depressive state (DP, n = 26).

We focused on patients with BD in a depressive mood state or a euthymic mood state because BD patients spend more than 80% of their time in a non-manic mood state. The current study has been approved by the Institutional Review Board of Brain Image Center, Bei**g Normal University and Bei**g Anding Hospital, Capital Medical University. Written informed consents were obtained from all subjects prior to participation in this study.

fMRI data acquisition

All imaging data were acquired at the Brain Imaging Center of the Bei**g Normal University by using a Siemens TIM Trio 3T scanner (Siemens, Erlangen, Germany). Resting state functional images (240 volumes) were acquired initially, followed by the T1 images. During scanning, subjects were asked to try not to swallow or move their body, especially their heads. During resting state, subjects were also instructed to close their eyes, relax, and remain awake. Resting state functional images were collected using echo-planar imaging (EPI) sequence, as follows: axial scanning, repetition time (TR) = 2000 ms, echo time (TE) = 30 ms, flip angle = 90°, field of view (FOV) = 200 × 200 mm2, matrix size = 64 × 64, slices = 33, interlaced scanning, slice thickness = 3.5 mm, and gap = 0.7 mm. T1 images were collected using T1-weighted sagittal 3D magnetization-prepared rapid gradient echo (MPRAGE) sequence, as follows: TR = 2530 ms, TE = 3.39 ms, flip angle = 7°, FOV = 256 × 256 mm2, matrix size = 192 × 256, slices = 128, thickness = 1.33 mm, and voxel size = 1.33 × 1 × 1 mm3.

fMRI data preprocessing

Resting state fMRI date was preprocessed in the toolbox for the Data Processing and Analysis for (Resting State) Brain Imaging (DPABI v3.1) [39]. The preprocessing steps included removing the first five time points, slice timing, realignment, detrending, removing the head motion effect (using Friston 24-parameter model) [40], regressing out nuisance covariates (the white matter signals, the cerebral spinal fluid signals, and the global mean signals), normalizing the MNI space (using T1 images), resampling (voxel size = 2 × 2 × 2 mm3), smoothing (FWHM = 4 mm, only for the fALFF and rsFC analysis), and band-pass filtering (0.01–0.1 Hz, only for the ReHo, DC, and rsFC analysis). In addition, we used volume-based framewise displacement (FD) [41] to quantify head motion. FD reflects head motion from one volume to the next, and a volume with FD < 0.2 indicates a good time point. Mean FD is calculated by the average of the sum of the absolute values of the differentiated realignment estimates (by backwards differences) at every timepoint [42]. Subjects would be excluded if their mean FD exceeded three standard deviations beyond the mean value of the entire sample or the number of good time points was less than 120.

fMRI data analysis

First, three indices of the fALFF, the ReHo, and the DC were used to unravel abnormal brain activities in the patients with BD. Subsequently, using the overlapped cluster of the significant brain region found in the analysis as seed, rsFC analysis was performed to explore possible brain regions, which showed abnormal rsFC with the overlapped cluster in patients with BD. Finally, two additional strategies were used to reidentify the seed region and then confirmed our rsFC analysis.

fALFF, ReHo, and DC calculation

Using the preprocessed resting state fMRI data, we calculated the fALFF, ReHo, and DC values for each subject in DPABI v3.1. All values were calculated at voxel level. The fALFF value was calculated using the signal strength of the low frequency range (i.e., 0.01–0.1 Hz) to divide the detectable entire frequency range [25]. The ReHo value, also called the Kendall’s coefficient of concordance (KCC), of the time series of a given voxel with its nearest neighbors (27 voxels were considered) was calculated to generate individual ReHo maps [26]. The DC value of a certain voxel was the sum of the Pearson’s correlation coefficients of all possible pairs of voxels (correlation threshold of r0 was set at 0.2) [27, 43]. Lastly, the fALFF, ReHo, and DC maps for each subject were converted into Z-score maps, respectively. For the ReHo and DC maps, smoothing (FWHM = 4 mm) was performed.

Functional connectivity analysis

This analysis was performed in DPABI v3.1. The overlapped brain region, which showed significant group differences in fALFF, ReHo, and DC values between patients with BD and HC, was used as a seed in the following functional connectivity analysis. First, time series of each voxel was extracted within the seed. Second, the mean time series of the seed was calculated by averaging each voxel’s time series. Third, the Pearson correlation coefficients of the mean time series of the seed and all other voxels’ time series were calculated. They were used to construct each subject’s rsFC map. Finally, each subject’s rsFC map was converted into Z-score map.

Statistical analyses

For categorical variables of demographic and clinical data, we used chi-square test for group comparison. For continuous variables, we first used the Shapiro-Wilk test for normality testing. If the variable followed a normal distribution, we used a two sample T-test for group comparison. If it did not follow a normal distribution, the Man-Whitney U test was used to compared the differences between the patients with BD and the HC. The above processes are all calculated in SPSS 20.0.

For each functional index (i.e., the fALFF, the ReHo, the DC, and the rsFC), analyses of the differences between the two groups were conducted in SPM12 (Wellcome Department of Cognitive Neurology, London, UK). In these analyses, two-sample T tests were used with age, gender, education, and mean FD controlled. The significance level was set at the combination of voxel-level uncorrected P < 0.001 and cluster-level Family Wise Error (FWE)-corrected P < 0.05. And in order to further investigate the correlation between imaging indicators and clinical symptoms, we used Pearson correlation analysis to study the correlation between the fALFF, DC, and ReHo of the clusters showing significant inter group differences and the total HAMD score. The significance level was set at 0.05/3 = 0.017. These statistical analyses were limited in a pre-made gray mask. The gray matter mask was generated by averaging the gray matter masks segmented from the T1 images of each subject.

Confirmation analyses of the rsFC

To validate the findings obtained in the rsFC analysis, we used two strategies to confirm the findings. (i) We used the significant cluster obtained in the group analyses of the fALFF and the DC, but not the ReHo (due to the negative finding in a strict statistical threshold) to calculate the overlapped cluster. The overlapped cluster was then used as a seed in the rsFC analysis. (ii). We attempted to map our interested brain regions identified in the analysis of the fALFF, ReHo, and DC to the Brainnetome atlas [S6-A). Then, we used the sum of the brain region Nos. 153 and 154 as a seed in the following rsFC analysis. The rsFC map within each group is shown in Figure S6-B. The whole-brain analysis found a significantly decreased rsFC of the seed with the MPFC (Figure S6-C and Table 3).

Furthermore, in the subgroup analysis, we observed that individuals of the EP group exhibited increased rsFC (indicating decreased negative rsFC) between the precuneus and the left putamen (extended to the left insula), in comparison to HC (Figure S7, Table S1). Subsequently, upon reducing the threshold, we identified that the outcomes for both subgroups were consistent with the comparisons between individuals with BD and HC. Detailed results were presented in the supplementary materials.

Discussion

We used three main functional indices (fALFF, ReHo, and DC) to measure the spontaneous activity from the local activity, local connectivity to network-level functional connectivity. Consistent with our formulated research hypothesis, our findings corroborate the presence of diminished spontaneous activity within the left precuneus, a pivotal node within the DMN, among patients diagnosed with BD. Furthermore, using this overlapped region in the left precuneus as a seed for rsFC, we found the decreased rsFC within the DMN, indicated by the decreased rsFC with the left precuneus and decreased rsFC between the seed precuneus and the MPFC in patients with BD. In addition, we found decreased negative functional connectivity between the left precuneus and the core regions in the SN (i.e., the left putamen/insula), suggesting decreased anticorrelation between the DMN and the SN in patients with BD.

Altered spontaneous activity of the DMN

We used multiple functional indices and found that the patients with BD exhibited decreased fALFF, ReHo, and DC values in the left precuneus (a specific brain region of the DMN) compared with healthy controls. The decrease in the fALFF indicates reduced local BOLD signal fluctuations of the precuneus [25, 45]. The decrease in the ReHo indicates decreased synchronizing ability in relevant voxels within the precuneus [26], and the decrease in DC indicates reduced functional connections of the precuneus with other voxels in the brain [46]. The co-occurrence of abnormalities in fALFF and DC within the precuneus signifies disrupted spontaneous brain activity in this specific brain region. This study revealed a significant correlation (r = 0.671, P < 0.001) between fALFF and DC values in the precuneus across all samples, consistent with findings from Yan’s research that also reported a strong correlation between these two metrics [47]. Our findings are consistent with most previous studies [32], [56] or when they were asked to increase their negative emotions when facing aversive stimuli [57]. One study using a negative emotion task found reduced activation of the left precuneus during negative images condition in the euthymic BD-I patients, with respect to controls [58]. This finding suggested that decreased brain activity of the left precuneus in patients with BD may somehow interfere with their ability to process emotional contents. The results found in the current resting-state fMRI study highlighted the importance of the precuneus, especially its left part, in the pathogenesis of BD. Thus, the patients’ left precuneus was unable to maintain normal signal fluctuation, synchronizing ability, and connectivity with other brain regions.

Decreased rsFC within the DMN

The followed rsFC analysis found hypoconnectivity within the left precuneus in the patients with BD. The results were consistent with previous studies [59, 60]. Wang et al. [44] used FC strength (FCS) to measure the whole-brain FCS patterns, and found that unmedicated patients with BD-II depression showed decreased FCS in the DMN (including the left precuneus). Khadka et al. [60] using ICA found that the rsFC in the posterior DMN consisting mainly of bilateral precuneus was significantly decreased in patients with psychotic BD-I than the HC. Combining these findings with our results, we speculated that the decreased local spontaneous activity of the left precuneus may account for its decreased rsFC. In addition, the precuneus is a functional core of the DMN [61]. It not only shows the highest resting metabolic rate within the network [62], but has widespread connectivity with other brain regions [54]. It suggested that the anomaly of precuneus could have an important effect on the overall functional abnormality of the DMN [54, 61]. The patients with BD also exhibited reduced rsFC between the left precuneus and the MPFC, which is another core node of the DMN [63]. This result was repeated in our confirmation analyses. The current result was, to some extent, consistent with previous report of Wang et al. [64], although two additional studies using smaller sample size (the number of patients with BD is 15 and 30) did not find positive results [24, 65]. In Wang’s study, unmedicated patients with BD under depressive episode were included, and patients showed disrupted intramodular connectivity within the DMN. The reduced rsFC between the MPFC and the left precuneus reflected weaker synchronization of spontaneous neural activities within the DMN, which could reduce the efficiency of the information communication and integration of the network [66]. The precuneus is involved in self-centered mental imagery strategies and episodic memory retrieval [54]. The MPFC is one of the most clearly delineated regions in terms of its functional roles; it serves an important role in emotional experience, social cognition, memory, and decision making [67, 68]. Some researchers suggested that the abnormal rsFC between the anterior (e.g., the MPFC and the superior frontal gyrus) and the posterior (e.g., the PCC/precuneus) components of the DMN sever as the neural basis for the defects of attention and working memory [12]. The patients with BD have been reported to have these cognitive abnormalities [69]. One study on the relationship between the DMN and clinical symptoms of BD found that the decreased rsFC within the posterior DMN in patients with BD depression might be associated with rumination [19], which is a typical symptom for depressed patients. Simply put, rumination is the process in which a ruminant animal continuously regurgitates partially digested food from its stomach for further chewing. From a psychopathological perspective, rumination is defined as a distress response pattern characterized by passive and persistent contemplation of one’s distressing symptoms, their origins, and repercussions, without engaging in active problem-solving to address the underlying causes of the distress [70]. When we engage in rumination, the brain activity of DMN increases [71]. From this aspect, it would be reasonable to speculate that the decreased rsFC between the MPFC and the precuneus may result in the cognitive dysfunction and clinical symptoms for patients with BD. In summary, our findings with all previous studies support that the functional dysconnectivity within the DMN may play a relatively important role in the neural basis of BD.

Decreased anticorrelation between DMN and SN

Decreased negative connectivity was found between the left precuneus and the left putamen/insula, the main nodes of the SN [3, 72], in the patients with BD. This finding reflects the decreased anticorrelation between the DMN and the SN.

In healthy population, resting-state fMRI signals of the DMN are anticorrelated with that of task-positive networks, such as the SN [73,74,75]. The SN generates control signals and causally mediates the “switching” between the DMN and the central executive network [76]. This condition leads to a switch between their functions of introspective and extroversive attention orientation, and then allows individuals to stay vigilant when meeting unexpected environmental events [12]. This anticorrelation between DMN and SN has even been proven more important than the DMN’s own activity [77]. The decreased anticorrelation between these networks may result in the reduction of introspective thinking and attention deficit, which could be the potential interference source of goal-directed behavior [12]. Weaker anticorrelation between the DMN and the SN was also found in many other mental disorders [73, 78]. Our result expanded previous findings of disruption of the DMN-SN functional connectivity in mental disorders [23] by showing that the negative rsFC between the left precuneus and the left putamen/insula was decreased in patients with BD.

Moreover, it is noteworthy that findings obtained from the subgroup analysis were aligned with those obtained from the comparisons between individuals with BD and HC, especially increased rsFC between the precuneus and the left putamen (extended to the left insula). It is suggested that the increased precuneus-left putamen connection may be not unaffected by patients’ mood states, and may be a trait-like feature for BD.

Limitations

Some limitations are as follows. Firstly, although the patients were not in a manic mood state, our sample was still heterogeneous in the subtypes or the different affective status of the illness. Future work should recruit patients in the same mood state (mania, depression, or euthymia) to identify whether the significant results were state-dependent or trait-like characters of BD. Secondly, most patients had taken different types or dosages of drugs. Although no consistent effect of any particular drug was found on BOLD signals, different drugs or dosages may influence BOLD signals in different ways [79], potentially confounding our results. In future studies, unmedicated patients with BD should be recruited or the types and dosages of medication use should be strictly controlled. Thirdly, our study did not examine whether these brain changes could account for any clinical or cognitive symptoms, such as rumination, or whether these brain abnormalities changed over time. Further studies are needed to address these issues.

Conclusions

In summary, using multiple functional indices, we found decreased spontaneous brain activity in DMN, particularly in the left precuneus. We also observed decreased functional connectivity within the DMN and reduced anticorrelation between the DMN and the SN in patients with BD. These findings suggest that the DMN is a key aspect for understanding the neural basis of BD. The altered functional patterns of the DMN may be a potential candidate biomarker for the diagnosis of BD.