Functional connectivity of the amygdala subnuclei in various mood states of bipolar disorder

doi:10.21203/rs.3.rs-2972436/v1

Download PDF

Article

Functional connectivity of the amygdala subnuclei in various mood states of bipolar disorder

https://doi.org/10.21203/rs.3.rs-2972436/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 09 May, 2024

Read the published version in Molecular Psychiatry →

You are reading this latest preprint version

Amygdala functional dysconnectivity lies at the heart of the pathophysiology of bipolar disorder (BD). Recent preclinical studies suggest that the amygdala is a heterogeneous group of nuclei, whose specific connectivity could drive positive or negative emotional valence. We investigated functional connectivity (FC) changes within these circuits emerging from each amygdala’s subdivision in 127 patients with BD in different mood states and 131 healthy controls (HC), who underwent resting-state functional MRI. FC was evaluated between lateral and medial nuclei of amygdalae (lAMY-rh, mAMY-rh, lAMY-lh, mAMY-lh), and key subcortical regions of the emotion processing network: anterior and posterior parts of the hippocampus (aHIP, pHIP), and core and shell parts of the nucleus accumbens (NAc-core and NAc-shell). FC was compared across groups, and subgroups of patients depending on their mood states, using linear mixed models. We also tested correlations between FC and depression (MADRS) and mania (YMRS) scores. We found no difference between the whole sample of BD patients vs. HC but a significant correlation between MADRS and lAMY-rh/aHIP-rh, lAMY-rh/pHIP-rh and lAMY-rh/aHIP-lh FC (r=-0.53, r=-0.36, r=-0.28, respectively, all pFDR<0.05). Subgroup analysis revealed decreased lAMY-rh/aHIP-rh and lAMY-rh/pHIP-rh FC in depressed vs. non-depressed patients, and increased mAMY-lh/NAc-shell-lh FC in manic vs non-manic patients. These results demonstrate that acute mood states in BD concur with FC changes in individual nuclei of the amygdala implicated in distinct emotional valence processing. Overall, our data highlight the importance to consider the amygdala subnuclei separately when studying its FC patterns including patients in distinct homogeneous mood states.

Recent preclinical studies suggested that the amygdala (AMY) is a complex and heterogeneous group of nuclei, whose specific connectivity could drive different emotional valences (1)(2). Numerous electrophysiological studies indeed showed that AMY, and in particular its lateral nucleus (lAMY), has neurons capable of coding for both positive and negative valences (3)(4)(5). While some studies opt for an intermingling of functionally distinct neurons within the same anatomical space, particularly in monkeys (6)(7)(8), others suggest anatomical and functional segregation (9). Differences in the function of amygdala neuronal populations in assigning positive or negative emotional valence could be related to their projection to the nucleus accumbens (NAc), the medial amygdala (mAMY) or the hippocampus (HIP) (3)(10). Specifically, recent studies offered a new perspective on the neurocircuitry of emotion processing, by demonstrating that lAMY neurons projecting to NAc mainly drive preferentially positive valence while those projecting to the ventral HIP and the mAMY preferentially drive negative valence in rodents, despite genetic and functional heterogeneity (3)(7)(11)(12)(13)(14). Based on these results, we proposed a new model of the pathogenesis of bipolar disorder (BD) by considering amygdala as a primary hub for positive and negative valence encoding (15).

BD appears indeed to be a relevant model for studying amygdala connectivity in humans in relation with opposite mood states. Many functional neuroimaging studies have highlighted dysfunctions in neural circuits involving the amygdala in BD (16)(17). Early task-based functional MRI (fMRI) studies demonstrated a predominant pattern of increased AMY activity in response to emotional stimuli (17)(18)(19). More recent studies using resting-state fMRI (rs-fMRI) reinforced these findings and provided evidence of functional connectivity (FC) abnormalities between the amygdala and cortical areas involved in emotional regulation such as the prefrontal cortex (20)(21)(22)(23)(24)(25)(26)(27)(28)(29). However, many of these studies are based on heterogeneous populations, including depressed, manic, or euthymic patients without correlation with the mood states. They also considered the amygdala as a single homogeneous structure, without considering the complexity of the organization of the amygdala and its involvement in the response to different emotional valences, as suggested by recent mouse models (10).

Consistent with the idea that connectivity of amygdala subnuclei has a central role in mood and emotional balance, we sought to investigate functional connectivity (FC) changes within these valence circuits emerging from the two main amygdala’s subnuclei in BD patients, in comparison to healthy participants. We also explored the possibility that these potential FC abnormalities in patients may be correlated to their level of mania or depression. Then, we investigated resting-state FC of the valence circuits in patients with BD in both manic and depressive states. We hypothesized that the lAMY in BD patients with manic symptoms would display increased resting-state FC with the NAc while the lAMY in BD with depressive symptoms would display increased resting-state FC with the ventral HIP.

1. Participants

The participants were included in the psychiatric departments of three European hospital centers: the Creteil University Hospital, France (Center-1), the Grenoble University Hospital, France (Center-2), and the Geneva University Hospital, Switzerland (Center-3). Clinical diagnosis was established by specialized psychiatrists using the DSM-IV-TR criteria and confirmed by the Mini-International Neuropsychiatric Interview (MINI) for the Geneva center or the Structured Clinical Interview (SCID) for the two centers in France, both administered by a trained clinician. Each patient sample was matched on sex and age with healthy controls (HC) from its respective center (Table 1). Control subjects were selected and included in the study after an interview with a psychiatrist, according to the SCID, to exclude any psychiatric disorder.

Table 1

Demographic and clinical characteristics of the samples
	Bipolar disorder (N = 127)			Healthy controls (N = 131)
	N	mean/freq	SD	N	mean/freq	SD
Age	127	35.6	11.6	131	34.2	11.4
Sex (% males)	127	54.3%		131	48.9%
Handedness (% right-handedness)	86	84.9%		83	81.9%
Age of onset	94	22.9	8.0
Illness duration	93	13.4	9.6
BD Type Type 1 Type 2		94 65 28
Depression Score (MADRS)	103	6.3	6.9	30	1.1	1.7
Mania Score (YMRS)	101	2.4	3.6	30	0.4	1.0
Number of depressive episodes	63	5.0	7.9
Number of manic episodes	64	3.3	3.8
On medication Antipsychotics Antidepressants Anticonvulsivants Lithium	101	93.9% 46% 21.7% 44.6% 50.5%	24.1% 50.1% 41.5% 50.0% 50.3%
N = number of available data

The exclusion criteria for all participants consisted of: (i) age outside the range of 18–65; (ii) history of alcohol or drug abuse; (iii) any current or past neurological/medical diseases affecting cognition; (iv) any history of head trauma with loss of consciousness; (v) contraindication for MRI.

For BD patients, additional exclusion criteria were: (i) any current other Axis I psychiatric disorder; (ii) history of electroconvulsive therapy during the previous year.

For HC, additional exclusion criteria included: (i) history of psychiatric illness; (ii) family history of major psychiatric or neurological illness; (iii) medical treatment affecting cerebral activity. All participants were also screened for MRI safety.

In total, 128 BD patients (Center-1: 66, Center-2: 17, Center-3: 45) and 133 HC were included in the study (Center-1: 82, Center-2: 16, Center-3: 35).

The study protocol was approved by the institutional Ethics Committee of each center (i.e., the local University Hospital Ethics Committees in France and the Cantonal Commission on Ethics in Human Research of Geneva) and written informed consent was obtained from all the participants.

2. Clinical assessment

Before MRI scanning, the severity of manic symptoms was evaluated by the Young Mania Rating Scale (YMRS) (30) and the severity of depressive symptoms by the Montgomery-Asberg Depression Rating Scale (MADRS) (31).

A higher YMRS score indicates more severe manic symptoms, and we used a cut-off score ≥ 4 to define mania or residual symptoms of mania (manic BD) (32). Similarly, a higher MADRS score indicates more severe depression, and we used a cut-off score ≥ 7 to define a depressive state (depressed BD) (33)(34). MADRS and YMRS scores were available for 103 and 101 patients with BD, respectively.

3. MRI data acquisition

All participants were instructed to simply rest, let their thoughts wander or not to think about anything specific, and not to fall asleep. A rs-fMRI sequence as well as an anatomical T1-weighted MRI sequence were acquired for each subject and each site with the following parameters:

3.1. Center 1 (Paris)

MRI data were collected at the Neurospin neuroimaging platform (Commissariat à l’Energie Atomique, Saclay, France) using two 3T Siemens MR scanners:

Magnetom TrioTim syngo MR B17, with 12-channel head coils respectively. Functional images were acquired through an echo-planar imaging sequence with the following parameters: TR = 2000 ms, TE = 27 ms, slices = 35, matrix size = 64 x 64 pixels, FA = 81°, FOV = 192 x 192 mm², voxel size = 3 x 3 x 3 mm³, interslice gap = 3 mm, and total 360 volumes with a scan time of 12 minutes. A high-resolution T1-weighted anatomical scan was also acquired (MPRAGE; TR/TI/TE = 2300/900/2.98 ms; FA = 9°; voxel dimensions = 1.0 mm isotropic; 256 x 256 x 192 voxels).
Magnetom Prisma, with 64-channel head coils. Functional images were acquired through an echo-planar imaging sequence with the following parameters: TR = 1430 ms, TE = 20 ms, slices = 62, matrix size = 76 x 76 pixels, FA = 68°, voxel size = 2 x 2 x 2.5 mm³, interslice gap = 2.5 mm, multiband acceleration factor = 2. A high-resolution T1-weighted anatomical scan was also acquired (MPRAGE; TR/TI/TE = 2300/900/2.98 ms; FA = 9°; voxel dimensions = 1.0 mm isotropic; 240 x 256 x 160 voxels).

3.2. Center 2 (Grenoble)

Imaging data were acquired using a whole-body 3T MR scanner (BrukerMedSpecS300- Grenoble MRI facility IRMaGE) with a fMRI acquisition sequence. Functional images were collected through an echo-planar imaging sequence with the following parameters: TR = 2500 ms, TE = 30 ms, slices = 37, matrix size = 72 x 72 pixels, FA = 77°, FOV = 216 x 216 mm², voxel size = 3 x 3 x 3.5 mm³, and total 144 volumes with a scan time of 6 minutes. Finally, a T1-weighted height-resolution three-dimensional anatomical volume was acquired (0.8 mm in plane resolution, 0.8 mm thickness; FOV = 224x256x176; acquisition matrix = 280x320x220 pixels).

3.3. Center 3 (Geneva)

MRI data were obtained from two prospective cohorts, acquired at the University of Geneva on a 3T Siemens Magnetom TrioTim scanner with a 32-channel head coil:

For the first cohort, functional images were collected through an echo-planar imaging sequence with the following parameters: TR = 2100 ms, TE = 30 ms, slices = 36, matrix size = 64x64 pixels, FA = 80°, voxel size = 3.2 x 3.2 x 3.2 mm³, interslice gap = 3.84 mm, and total 250 volumes with a scan time of 8 minutes 45 seconds. A high-resolution whole brain anatomical scan was acquired with a T1-weighted 3D sequence (MPRAGE; TR/TI/TE = 1900/900/2.27 ms; FA = 9°; voxel dimensions = 1.0 mm isotropic; 256 x 256 x 192 voxels).
For the second cohort, whole brain functional images were collected using a BOLD-weighted EPI sequence with the following parameters: TR = 2100 ms, TE = 30 ms, slices = 36, FA = 90°; FOV = 205 mm; matrix size = 64 x 64 pixels, voxel size = 3.2 x 3.2 x 3.2 mm³. 250 volumes were acquired for a total duration of 8 minutes 45 seconds. A T1-weighted volume for anatomical reference was collected with high-resolution three-dimensional T1-weighted MPRAGE sequence (TR/TI/TE = 1900/900/2.32 ms; flip angle = 9°; voxel dimensions = 0.9 mm isotropic; FOV = 230 mm).

4. MRI data processing

We discarded the 5 first frames of functional data, after which the preprocessing of rs-fMRI data was conducted using fMRIPrep (35). Concisely, this standardized pipeline included the following steps: slice-timing correction, motion correction, skull stripping, and estimation of motion parameters and other nuisance signal time series. To remove head motion effects, a Friston 24-parameter model (36) was used to regress out head motion effects from the realigned data (i.e., 6 head motion parameters, 6 head motion parameters one time point before, and the 12 corresponding squared parameters). The physiological noise (cerebrospinal fluid and white matter signals) was also regressed as nuisance covariates. Given the ongoing debates about optimal preprocessing steps in the resting-state functional MRI community (37) (38) and knowing that the global mean signal contains measurable effects of certain pathologies, we decided not to correct for its effect (39). After nuisance covariate regression, the resultant data were linearly detrended and filtered at a range between 0.01 Hz and 0.08 Hz to select low frequency signals. All functional data were spatially normalized to the Montreal Neurological Institute EPI template (MNI 152).

Exclusion criteria after pre-processing for all subjects were: (i) visual artifacts in the medial temporal lobe (n = 1), (ii) signal-to-noise ratio and temporal signal-to-noise ratio less than − 2.5 SD of their cohort (n = 2). We therefore included 127 BD patients and 131 HC in the final analysis.

For each subject, we also excluded volumes with a mean framewise displacement (FD) > 0.5 mm or a mean BOLD data variance from the prior volume (DVARS) > 50% along with the previous and the following volume (40)(41)(42). This represented 0.3% of the total volumes acquired.

5. Functional connectivity analysis

ROI-to-ROI analyses were performed using the AMY, the NAc and the HIP subregions as seeds. Each unilateral subcortical nucleus was divided into two independent areas based on the recently developed functional MRI Atlas of Tian (43). Consequently, for each hemisphere, the AMY was divided into lateral (lAMY) and medial (mAMY) amygdala, the HIP into anterior (aHIP) and posterior (pHIP) hippocampus and the NAc into shell (NAc-shell) and core (NAc-core) subparts. These seed regions were created in the MNI standard space using Nilearn Python library (Fig. 1). The signal time course was extracted from the 12 resulting ROIs (6 by hemisphere) in both patients and controls and the mean signal time course was calculated for each seed region and each subject. Afterwards, we computed temporal correlations between the BOLD signals of four seed of interest (ROIs): right lAMY (MNI coordinates: 28, -2, -22), right mAMY (MNI : 22, -6, -16), left lAMY (MNI : -26, -2, -22), left mAMY (MNI : -20, -6, -16), and eight target ROIs: right aHIP (MNI : 26, -14, -20), right pHIP (MNI : 28, -32, -8), left aHIP (MNI : -24, -14, -20), left pHIP (MNI : -26, -32, -8), right NAc-shell (MNI : 12, 10, -6), right NAc-core (MNI : 14, 18, -2), left NAc-shell (MNI : -10, 10, -6), left NAc-core (MNI : -12, 18, -2). Each ROI was visually inspected for each subject to assure a precise localization within anatomical boundaries.

6. Statistical analysis

6.1. Whole-sample analyses

At the individual subject level, correlations were obtained by extracting the mean BOLD time course from each amygdala subregion i seed (mAMY and lAMY, right and left) and then calculating Pearson’s product moment correlation between that time course and the time course of other regions j of interest (i.e., NAc-shell, NAc-core, pHIP and aHIP, right and left). This matrix of correlation coefficients was then converted to normally distributed Z-scores using Fisher’s r-to-z transformation to allow for second-level analysis. All subsequent statistical analyses were conducted on these transformed data.

FC scores (Z-scores) for each subject were entered into a second-level random-effects analyses to quantitatively compare the amygdala subregions’ FC between the groups (BD vs. HC). We used linear mixed models including the diagnosis as fixed factor (BD vs. HC) as well as Age and Sex as covariates and Site as random effect (R version 4.1.2, lme4 package, Bates et al., 2022) (44). For each pair of ROIs, we fitted the following model:

FC (ROIi-ROIj) = β0 (Intercept) + β1 * Diagnosis + β2 * Age + β3 * Sex + random effect (Site)

We used Benjamini-Hochberg correction to control for multiple comparisons.

6.2. Sub-sample analyses

6.2.1. Correlations with psychiatric symptoms

To evaluate the correlation between the intensity of the mood symptoms and FC in BD patients, FC values (Z-scores) from each pair of ROIs (n = 12) were correlated with the MADRS and the YMRS scores using a non-parametric Pearson’s correlation test (with a significance level threshold of p < 0.05 after FDR correction).

6.2.2. Subgroups analyses

Then, given our initial hypothesis, we confined our analysis and specifically compared AMY/HIP connectivity between depressed (n = 31) and non-depressed (n = 72) BD patients / HC (n = 131) and AMY/NAc connectivity between manic (n = 29) and non-manic (n = 74) BD patients / HC (n = 131). For that purpose, we also used linear mixed models with the subgroup as fixed factor and the Age and Gender as covariates, and Site as random effect, as follows:

FC (ROIi-ROIj) = β0 (Intercept) + β1 * Subgroup + β2 * Age + β3 * Sex + random effect (Site)

Demographic and clinical characteristics from the whole sample are shown in Table 1. Patients and controls did not show differences in terms of age, sex and handedness.

Whole sample analysis

In the whole sample, linear mixed models did not show any significant differences between patients with BD and HC when testing for amygdala lateral and medial nuclei’s functional connections with any of the other target regions (see detailed results in Supplementary Table 1).

Sub-sample analysis

2.1. Correlation between the clinical scores and the amygdala subregions’ connectivity

In the group of patients for whom we had the MADRS score (n = 103), we found a significant negative correlation between MADRS score and lAMY-rh/aHIP-rh, lAMY-rh/pHIP-rh and lAMY-rh/aHIP-lh FC (r = -0.53, p-FDR < 0.001; r = -0.36, p-FDR < 0.01 and r = -0.275, p-FDR < 0.05, respectively) (Table 2; Fig. 2), indicating a decreased FC in patients with higher depressive symptoms. The other correlations tested were not significant (Table 2). No correlation was found between YMRS score and FC of target regions.

Table 2

Correlations between the amygdala, hippocampus and nucleus accumbens subnuclei functional connectivity and clinical scores of depressive (MADRS) and manic (YMRS) symptoms.
Seed	ROI		MADRS N = 103			YMRS N = 101
		r	p_corr	sign	r	p_corr	sign
lAMY-rh	aHIP-rh	-0.533	1.459e-06	*******	-0.044	0.832	NS
	pHIP-rh	-0.364	0.003	**	-0.037	0.832	NS
	NAc-shell-rh	-0.103	0.471	NS	0.073	0.832	NS
	Nac-core-rh	0.014	0.903	NS	0.115	0.832	NS
	aHIP-lh	-0.275	0.030	*	0.029	0.832	NS
	pHIP-lh	-0.105	0.471	NS	0.044	0.832	NS
	NAc-shell-lh	-0.169	0.246	NS	-0.157	0.832	NS
	NAc-core-lh	-0.078	0.551	NS	0.074	0.832	NS
lAMY-lh	aHIP-rh	-0.026	0.930	NS	0.149	0.845	NS
	pHIP-rh	0.002	0.987	NS	0.179	0.845	NS
	NAc-shell-rh	-0.098	0.813	NS	0.034	0.845	NS
	NAc-core-rh	0.074	0.813	NS	0.077	0.845	NS
	aHIP-lh	-0.149	0.745	NS	0.027	0.845	NS
	pHIP-lh	0.073	0.813	NS	0.051	0.845	NS
	NAc-shell-lh	-0.146	0.745	NS	-0.118	0.845	NS
	NAc-core-lh	0.044	0.925	NS	0.066	0.845	NS
mAMY-rh	aHIP-rh	-0.043	0.967	NS	-0.013	0.925	NS
	pHIP-rh	0.005	0.967	NS	0.115	0.765	NS
	NAc-shell-rh	0.018	0.967	NS	0.036	0.908	NS
	NAc-core-rh	0.005	0.967	NS	-0.077	0.770	NS
	aHIP-lh	-0.171	0.956	NS	0.135	0.765	NS
	pHIP-lh	-0.067	0.967	NS	0.132	0.765	NS
	NAc-shell-lh	-0.077	0.967	NS	-0.099	0.765	NS
	NAc-core-lh	-0.037	0.967	NS	-0.261	0.456	NS
mAMY-lh	aHIP-rh	0.032	0.917	NS	0.059	0.672	NS
	pHIP-rh	-0.040	0.917	NS	0.101	0.663	NS
	NAc-shell-rh	0.078	0.917	NS	0.130	0.663	NS
	NAc-core-rh	-0.028	0.917	NS	-0.077	0.663	NS
	aHIP-lh	-0.101	0.917	NS	0.136	0.663	NS
	pHIP-lh	-0.011	0.924	NS	0.079	0.663	NS
	NAc-shell-lh	-0.109	0.917	NS	0.156	0.663	NS
	NAc-core-lh	-0.259	0.140	NS	-0.097	0.663	NS
NS: not significant; * p_corr <0.05 ; p_corr < 0.01; * p_corr < 0.001
Amygdala: Centromedial (mAMY); Laterobasal (lAMY); Hippocampus: Anterior (aHIP); Posterior (pHIP); Nucleus accumbens: Core (NAc-core); Shell (NAc-shell); rh: right hemisphere; lh: left hemisphere

2.2. Subgroups analyses

We observed a significant decrease in FC between lAMY-rh nucleus and aHIP-rh and pHIP-rh nuclei in depressed patients (n = 31) compared to non-depressed patients (n = 72) (t = 7.29 for lAMY-rh/aHIP-rh; t = 4.48 for lAMY-rh/pHIP-rh; both with p-FDR < 0.001) (Table 3). Furthermore, compared with the HC group, depressed patients with BD presented a significant decrease in FC between lAMY-rh and aHIP-lh (t = -2.55, p-FDR < 0.05), between lAMY-rh and aHIP-rh (t = -3.47, p-FDR < 0.01) and between lAMY-rh and pHIP-rh (t = -3.54, p-FDR < 0.01) (see Supplementary Table 2).

Table 3

Results of the subgroup analyses: linear mixed models parameters for functional connectivity differences between depressed and non-depressed, and between manic and non-manic patients with BD
Subgroups	Seed	ROI	β	Std.Err	t-value	P>\|t\|unc	P>\|t\|corr	R²	[0.025	0.975]	sign
Depressed / non-depressed	mAMY-lh	aHIP-rh	-0.0105	0.0459	-0.2282	0.82	1.00E + 00	5E-04	0	0.051	NS
		pHIP-rh	0.0328	0.0433	0.7562	0.4513	1.00E + 00	0.005	0	0.0684	NS
		aHIP-lh	0.071	0.048	1.4792	0.1422	5.69E-01	0.0189	1E-04	0.1037	NS
		pHIP-lh	0.0103	0.0419	0.2458	0.8063	1.00E + 00	5E-04	0	0.0512	NS
	lAMY-lh	aHIP-rh	-0.0028	0.0527	-0.0534	0.9575	1.00E + 00	0	0	0.0489	NS
		pHIP-rh	0.0237	0.0451	0.525	0.6007	1.00E + 00	0.0027	0	0.0601	NS
		aHIP-lh	0.0316	0.042	0.7508	0.4545	1.00E + 00	0.0048	0	0.0676	NS
		pHIP-lh	0.0014	0.048	0.0295	0.9765	1.00E + 00	0	0	0.0488	NS
	lAMY-rh	aHIP-rh	0.3081	0.0423	7.286	0	2.77E-10	0.3445	0.2118	0.4809	*******
		pHIP-rh	0.2167	0.0484	4.4801	0	7.98E-05	0.1522	0.0489	0.2899	*******
		aHIP-lh	0.0675	0.0508	1.3274	0.1873	7.49E-01	0.0169	1E-04	0.0994	NS
		pHIP-lh	0.0402	0.05	0.8056	0.4223	1.00E + 00	0.0063	0	0.0724	NS
	mAMY-rh	aHIP-rh	-0.0028	0.0527	-0.0534	0.9575	1.00E + 00	0	0	0.0489	NS
		pHIP-rh	0.0237	0.0451	0.525	0.6007	1.00E + 00	0.0027	0	0.0601	NS
		aHIP-lh	0.0316	0.042	0.7508	0.4545	1.00E + 00	0.0048	0	0.0676	NS
		pHIP-lh	0.0014	0.048	0.0295	0.9765	1.00E + 00	0	0	0.0488	NS
Manic / non-manic	mAMY-lh	NAc-shell-rh	-0.06	0.0374	-1.6021	0.1122	4.49E-01	0.0248	1E-04	0.1156	NS
		NAc-core-rh	-0.0438	0.0322	-1.3587	0.1774	7.10E-01	0.0182	1E-04	0.1022	NS
		NAc-shell-lh	-0.1247	0.0376	-3.3114	0.0013	5.14E-03	0.097	0.0172	0.2242	**
		NAc-core-lh	-0.0057	0.0354	-0.1603	0.873	1.00E + 00	3E-04	0	0.0499	NS
	lAMY-lh	NAc-shell-rh	0.039	0.0434	0.8975	0.3716	1.00E + 00	0.0079	0	0.0772	NS
		NAc-core-rh	-0.0138	0.0431	-0.3198	0.7498	1.00E + 00	0.001	0	0.0533	NS
		NAc-shell-lh	0.0662	0.045	1.4704	0.1445	5.78E-01	0.021	1E-04	0.108	NS
		NAc-core-lh	-0.0368	0.0366	-1.0051	0.3172	1.00E + 00	0.0099	0	0.0826	NS
	lAMY-rh	NAc-shell-rh	-0.0123	0.053	-0.2314	0.8175	1.00E + 00	5E-04	0	0.0512	NS
		NAc-core-rh	-0.0775	0.0532	-1.4563	0.1484	5.93E-01	0.0206	1E-04	0.1072	NS
		NAc-shell-lh	0.0272	0.0455	0.5981	0.5511	1.00E + 00	0.0035	0	0.0632	NS
		NAc-core-lh	-0.0303	0.0437	-0.6926	0.4901	1.00E + 00	0.0047	0	0.0674	NS
	mAMY-rh	NAc-shell-rh	-0.0509	0.0606	-0.8406	0.4025	1.00E + 00	0.0069	0	0.0744	NS
		NAc-core-rh	0.0034	0.0552	0.0618	0.9508	1.00E + 00	0	0	0.0489	NS
		NAc-shell-lh	-0.0148	0.0469	-0.3163	0.7524	1.00E + 00	0.001	0	0.0532	NS
		NAc-core-lh	0.0723	0.0408	1.7692	0.0798	3.19E-01	0.0301	2E-04	0.1255	NS
NS: not significant; * p_corr <0.05; p_corr < 0.01; * p_corr < 0.001
Amygdala: Centromedial (mAMY); Laterobasal (lAMY); Hippocampus: Anterior (aHIP); Posterior (pHIP); Nucleus accumbens: Core (NAc-core); Shell (NAc-shell); rh: right hemisphere; lh: left hemisphere

We observed a statistically significant increase in FC between the mAMY-lh and NAc-shell-lh in the manic subgroup (n = 28) compared with the non-manic subgroup (n = 73) (t = -3.31, p-FDR < 0.01) (Table 3). Compared with the HC group, increase in FC was not significant in both manic and non-manic subgroups (Supplementary Table 2). Other comparisons were not significant. Results obtained from the subgroups analyses are summarized in Fig. 3. Mean and standard deviation values of the raw FC coefficients obtained for each ROIs pair tested, for both whole sample and subsample analysis, are provided in Supplementary Tables 3 and 4.

This study aimed at examining valence encoding circuits emerging from the amygdala nuclei in BD patients in different mood states. To our knowledge, this is the first study to analyze the connectivity of these nuclei in a systematic way in the BD population. Beyond mood, emotional biases differentiate bipolar states in humans. Mania is associated with positive biases, e.g., emotional stimuli become more rewarding and less aversive, whereas depression is characterized by negative biases. AMY has been implicated in valence coding in both human and animal models (3)(10). Considering studies describing the role of specific circuits on valence assignment in preclinical studies, we hypothesized that in humans, lAMY/HIP FC might also be related to negative valence, and mAMY/NAc FC to positive valence. We tested this idea on a BD population with a range of depressive and manic symptoms. We hypothesized that different states could be associated with changes in FC of these specific circuits. We used a seed-based approach to assess the difference in FC in a sample of 127 patients with BD with either depressive or manic residual mood symptoms and 131 HC. We identified a specific pattern of abnormal FC in depressed patients with BD, mainly characterized by a decrease in FC between the lAMY and the aHIP. In addition, we found abnormal increase in FC between the mAMY and the NAc-shell in manic patients with BD.

Lateral amygdala / anterior hippocampus functional connectivity

Regarding the FC between the AMY and the HIP subnuclei, we found both a decrease in lAMY/HIP FC in depressed patients compared with non-depressed patients and HC, as well as a significant negative correlation between the lAMY/HIP FC and the MADRS score within the whole patients group. Thus, a disturbance in the connectivity between the lAMY and the HIP seems to be related with elevated depressive symptoms.

Connectivity between AMY and HIP has been repeatedly demonstrated by anatomical investigations (45)(46). The AMY/HIP network has previously been identified and described as underlying the regulation of emotional memories (47)(48). Studies in animals have suggested that the enhancing effect of emotional arousal on memory consolidation depends on the influence of the AMY upon the HIP in case of contextual and spatial information (49)(50)(51). Previous task-based fMRI studies in healthy subjects have shown an increased connectivity between the AMY and the HIP during the encoding and retrieval of emotional memories (52)(53)(54)(55). Although very few studies have specifically investigated this network in BD, our results are consistent with previous findings showing decreased AMY/HIP connectivity in patients with BD as compared to HC (56)(57). Li et al. (2015) on the other hand found a significant decrease in AMY/HIP FC in depressed BD patients, but with a fewer number of subjects (n = 10) and without subdividing the amygdala (56). In a small sample of youths with Bipolar I disorder (n = 20), Singh et al. (2015) also found decreased FC between the lAMY and the HIP relative to HC (57). Our results seem to confirm an alteration of the connection between the HIP and the AMY, and more specifically its lateral nucleus, in bipolar depressive patients.

This hypoconnectivity between AMY and HIP has also been described in non-bipolar patients, suffering from major depressive disorder (MDD) (58)(59)(60). A multivariate whole-brain analysis even proposed this reduced AMY/HIP FC as a robust discriminator to distinguish MDD patients from HC (61). These results suggest that decreased lAMY/HIP FC might be related to a depressed state, in both patients with unipolar or bipolar depression. In a study comparing unipolar and bipolar depression, Ambrosi et al. (2017) demonstrated lower FC between the right AMY and left anterior HIP in MDD compared to BD and HC (62) and suggested that this hypoconnectivity may help to distinguish MDD patients from depressed BD patients. However, all participants (BD and MDD patients) were experiencing a major depressive episode at the time of the study, and the correlation between FC and clinical score of depression was not tested. Further studies comparing FC in unipolar vs. bipolar depression, as well as bipolar mania vs. bipolar depression are needed to reinforce the hypothesis that abnormal decreased connectivity between the AMY and the HIP is a direct manifestation of depressive symptoms rather than a specific trait of BD. Our study points to this hypothesis, given the difference in FC between the depressed and non-depressed subgroups of BD patients.

Some results coming from task-based fMRI studies seem to contradict this statement. Indeed, Hamilton et al, in 2008, found an increase in AMY/HIP connectivity during a memory-based fMRI task in depressed patients with MDD (63). Moreover, preclinical studies demonstrated that basolateral amygdala neurons projecting to the hippocampus preferentially respond to negative cues (10). As suggested by Cullen et al., we could assume that the decrease in lAMY/HIP FC observed at rest in both depressed BD patients in the present study and in patients with MDD (58)(59)(60) could be a compensatory mechanism (60). Decrease in the AMY/HIP FC at rest could compensate for the hyperconnectivity of this network during the processing of memories with a negative emotional value. Further prospective studies comparing task-based and resting-state FC between lAMY and HIP in patients suffering from BD are needed to support this hypothesis of a compensatory mechanism at rest.

While the lAMY/HIP network seems involved in negative emotional bias, the absence of significant difference between groups regarding the mAMY/HIP FC can argue for specific connectivity patterns emanating from the different subnuclei of the AMY, that may be related to different mood states. These results appear to confirm in a human population the complex model of the amygdala suggested in preclinical studies (61).

Finally, we did not find a statistically significant difference when comparing lAMY/HIP FC between the whole BD and HC groups. Given the results obtained in our patient subgroups, this could be explained by the heterogeneity of the population, namely their difference in mood, with half of the patients in a non-depressive state (MADRS under 7). The importance of mood status in FC highlights the importance of selecting homogeneous patient groups in future studies on BD patients, and the necessity to compare different mood states. Even more importantly, using a dimensional framework and correlating between the FC indices and the clinical mood scores might be more informative. In addition, longitudinal studies comparing FC at different mood states from the same individual will be crucial to evaluate the causality between both factors.

Medial amygdala-nucleus accumbens functional connectivity

According to our hypotheses, we found an increased FC between the medial region of the AMY and the NAc in manic patients with BD compared to non-manic patients.

The NAc is a reward processing hub and has been hypothesized to function as an interface between limbic cortical structures required for associative conditioning, like the AMY (65). The AMY to NAc projection would have a critical role in the regulation of reward seeking and excitatory input from the AMY to the NAc has been shown to directly modulate reward processing during both reward anticipation and outcome (66)(67).

While the mAMY and the NAc are thought to be central components of the neural system mediating the impact of Pavlovian cues on instrumental responding, the lAMY/NAc FC could modulate cue-triggered motivated behaviors (65)(68).

Only one fMRI study has directly compared the AMY/NAc FC in mood disorders, but it exclusively focused on pediatric patients with BD and reported no difference in comparison to HC (22). The euthymic nature of the patients may explain the lack of difference, and the small number of subjects studied (15 patients) did not allow analysis of the correlation between clinical score and connectivity. In a socially rewarding stimulus (happy or sad face), elevated ventral striatal activity and left amygdala activity have also been observed in patients with BD (18)(69), but without specifically studying the NAc. Here, results support indeed the positive bias associated with manic symptoms, that might lead to increased reward feelings and diminished attention to negative consequences observed in manic episodes.

The small number of studies including manic patients can be explained by the difficulty of performing MRI in these patients. Our results could be confirmed in future studies testing the association between subthreshold clinical scores of mania and FC between the subcortical nuclei.

Limitations

Several limitations can be tackled. First, the spatial resolution of the fMRI is a well-known limitation for the localization of small seed and target regions, particularly the smallest regions like the amygdala subnuclei. The spatial resolution of the MRI scans allows us to individualize small structures of interest, such as the sub-regions of the amygdala, that have an estimated volume of 300 mm³, corresponding to nearly 13 voxels in our study. Almost all previous studies defined ROIs from anatomical or cytoarchitectural atlas, by creating a spherical seed plotted manually or around coordinates defined from the literature. The main issue of these approaches is the variation in the inter-individual's anatomy. Moreover, recent developments in high-resolution MRI have enabled reproducible delineation of functionally distinct and homogeneous subregions and have shown that the functional parcellation does not perfectly overlap with the anatomical or cytoarchitectural divisions (70)(71). Considering this issue, we have defined our ROIs based on a new functional subcortical atlas which accounts for interindividual variations in connectivity (43).

Second, for manic patients, the sample size was relatively small (n = 28) and their YMRS score was low (ranged from 4 to 18; mean = 6.9), corresponding to mostly subthreshold mood symptoms or hypomanic states (32). This probably led to a lack of power and might explain the absence of correlation between abnormal FC patterns and clinical scores in this subgroup. Nevertheless, the significant difference in FC between subgroups is a strong argument for a connectivity anomaly between these regions in case of manic symptoms.

Third, the influence of medication on resting-state FC has not been analyzed in this study as almost all patients were medicated. In addition, most patients were treated with more than one class of drug, making it impossible to analyze the effect of a specific medication type on the FC. The role of drugs remains controversial, and several studies have demonstrated no significant differences in FC between treated and untreated patients (72).

The current study evidences the differential involvement of AMY subnuclei networks connectivity in the opposite mood symptoms of BD. We demonstrated that different mood states are associated with changes in FC along different and specific neural circuits responsible for valence encoding. We propose to extend current pathophysiological models of BD, considering the heterogeneous structure of the amygdala, within which each sub-nucleus has specific connectivity profile according to the mood state. Our results also extend the preclinical literature by revealing – in humans – an altered pattern of FC between the amygdala’s subdivisions and key subcortical regions preferentially encoding for either positive and negative mood states (respectively nucleus accumbens and hippocampus). It may be relevant for future studies on mood disorders to consider the subnuclei of the amygdala separately, to include groups of patients with homogeneous mood states, and to compare these groups of patients with other mood disorders such as MDD, to determine whether these abnormal patterns of connectivity represent of specific biomarkers of BD and how they connect with larger scales networks.

Supplementary information is available at MP’s website.

ACKNOWLEDGEMENTS

The study was supported by research grants from the Grenoble University Hospital, the French University Institute, the Grenoble Cognition Center, and the Health and Society Research Network of the Pierre Mendès-France University (Grenoble). The Grenoble MRI facility IRMaGe/ Neurophysiology was partly funded by the French program “Investissement d’Avenir” run by the “Agence Nationale pour la Recherche”. SK was partially funded by a grant “Société Française de Radiologie – Alain Rahmouni".

The Paris site was funded by a grant “Infrastructure d’avenir en Biologie Santé”—ANR-11-INBS-0006” and the Agence Nationale pour la Recherche (ANR-11-IDEX-0004 Labex BioPsy, ANR-10-COHO-10-01 psyCOH), Fondation pour la Recherche Médicale (Bioinformatique pour la biologie 2014) and the Fondation de l'Avenir (Recherche Médicale Appliquée 2014).

This work was also supported by the Swiss National Center of Competence in Research; “Synapsy: the Synaptic Basis of Mental Diseases” financed by the Swiss National Science Foundation [Grant Number 51NF40-158776], as well as a grant of the Swiss National Science Foundation [Grant Number 32003B_156914].

CONFLICT OF INTEREST

Authors declare there have no competing financial interests in relation to the present study.

Pignatelli M, Beyeler A. Valence coding in amygdala circuits. Curr Opin Behav Sci. 2019 Apr;26:97–106.
O’Neill PK, Gore F, Salzman CD. Basolateral amygdala circuitry in positive and negative valence. Curr Opin Neurobiol. 2018 Apr;49:175–83.
Beyeler A, Namburi P, Glober GF, Simonnet C, Calhoon GG, Conyers GF, et al. Divergent Routing of Positive and Negative Information from the Amygdala during Memory Retrieval. Neuron. 2016 Apr 20;90(2):348–61.
Paton JJ, Belova MA, Morrison SE, Salzman CD. The primate amygdala represents the positive and negative value of visual stimuli during learning. Nature. 2006 Feb 16;439(7078):865–70.
Shabel SJ, Janak PH. Substantial similarity in amygdala neuronal activity during conditioned appetitive and aversive emotional arousal. Proc Natl Acad Sci U S A. 2009 Sep 1;106(35):15031–6.
Gore F, Schwartz EC, Brangers BC, Aladi S, Stujenske JM, Likhtik E, et al. Neural Representations of Unconditioned Stimuli in Basolateral Amygdala Mediate Innate and Learned Responses. Cell. 2015 Jul 2;162(1):134–45.
Namburi P, Beyeler A, Yorozu S, Calhoon GG, Halbert SA, Wichmann R, et al. A circuit mechanism for differentiating positive and negative associations. Nature. 2015 Apr 30;520(7549):675–8.
Zhang W, Schneider DM, Belova MA, Morrison SE, Paton JJ, Salzman CD. Functional circuits and anatomical distribution of response properties in the primate amygdala. J Neurosci Off J Soc Neurosci. 2013 Jan 9;33(2):722–33.
Kim J, Pignatelli M, Xu S, Itohara S, Tonegawa S. Antagonistic negative and positive neurons of the basolateral amygdala. Nat Neurosci. 2016 Dec;19(12):1636–46.
Beyeler A, Chang CJ, Silvestre M, Lévêque C, Namburi P, Wildes CP, et al. Organization of Valence-Encoding and Projection-Defined Neurons in the Basolateral Amygdala. Cell Rep. 2018 Jan 23;22(4):905–18.
Felix-Ortiz AC, Beyeler A, Seo C, Leppla CA, Wildes CP, Tye KM. BLA to vHPC inputs modulate anxiety-related behaviors. Neuron. 2013 Aug 21;79(4):658–64.
Li H, Namburi P, Olson JM, Borio M, Lemieux ME, Beyeler A, et al. Neurotensin orchestrates valence assignment in the amygdala. Nature. 2022 Aug;608(7923):586–92.
Shen CJ, Zheng D, Li KX, Yang JM, Pan HQ, Yu XD, et al. Cannabinoid CB1 receptors in the amygdalar cholecystokinin glutamatergic afferents to nucleus accumbens modulate depressive-like behavior. Nat Med. 2019 Feb;25(2):337–49.
Zhou K, Xu H, Lu S, Jiang S, Hou G, Deng X, et al. Reward and aversion processing by input-defined parallel nucleus accumbens circuits in mice. Nat Commun. 2022 Oct 21;13(1):6244.
Bigot M, Alonso M, Houenou J, Sarrazin S, Dargél AA, Lledo PM, et al. An emotional-response model of bipolar disorders integrating recent findings on amygdala circuits. Neurosci Biobehav Rev. 2020 Nov;118:358–66.
Phillips ML, Ladouceur CD, Drevets WC. A neural model of voluntary and automatic emotion regulation: implications for understanding the pathophysiology and neurodevelopment of bipolar disorder. Mol Psychiatry. 2008 Sep;13(9):829, 833–57.
Strakowski SM, Adler CM, Almeida J, Altshuler LL, Blumberg HP, Chang KD, et al. The functional neuroanatomy of bipolar disorder: a consensus model. Bipolar Disord. 2012 Jun;14(4):313–25.
Lawrence NS, Williams AM, Surguladze S, Giampietro V, Brammer MJ, Andrew C, et al. Subcortical and ventral prefrontal cortical neural responses to facial expressions distinguish patients with bipolar disorder and major depression. Biol Psychiatry. 2004 Mar 15;55(6):578–87.
Blumberg HP, Fredericks C, Wang F, Kalmar JH, Spencer L, Papademetris X, et al. Preliminary evidence for persistent abnormalities in amygdala volumes in adolescents and young adults with bipolar disorder. Bipolar Disord. 2005 Dec;7(6):570–6.
Anand A, Li Y, Wang Y, Lowe MJ, Dzemidzic M. Resting state corticolimbic connectivity abnormalities in unmedicated bipolar disorder and unipolar depression. Psychiatry Res. 2009 Mar 31;171(3):189–98.
Chepenik LG, Raffo M, Hampson M, Lacadie C, Wang F, Jones MM, et al. Functional connectivity between ventral prefrontal cortex and amygdala at low frequency in the resting state in bipolar disorder. Psychiatry Res. 2010 Jun 30;182(3):207–10.
Dickstein DP, Gorrostieta C, Ombao H, Goldberg LD, Brazel AC, Gable CJ, et al. Fronto-Temporal Spontaneous Resting State Functional Connectivity in Pediatric Bipolar Disorder. Biol Psychiatry. 2010 Nov;68(9):839–46.
Anticevic A, Brumbaugh MS, Winkler AM, Lombardo LE, Barrett J, Corlett PR, et al. Global prefrontal and fronto-amygdala dysconnectivity in bipolar I disorder with psychosis history. Biol Psychiatry. 2013 Mar 15;73(6):565–73.
Torrisi S, Moody TD, Vizueta N, Thomason ME, Monti MM, Townsend JD, et al. Differences in resting corticolimbic functional connectivity in bipolar I euthymia. Bipolar Disord. 2013 Mar;15(2):156–66.
Favre P, Baciu M, Pichat C, Bougerol T, Polosan M. fMRI evidence for abnormal resting-state functional connectivity in euthymic bipolar patients. J Affect Disord. 2014 Aug;165:182–9.
Brady RO, Masters GA, Mathew IT, Margolis A, Cohen BM, Öngür D, et al. State dependent cortico-amygdala circuit dysfunction in bipolar disorder. J Affect Disord. 2016 Sep 1;201:79–87.
Altinay M, Karne H, Anand A. Lithium monotherapy associated clinical improvement effects on amygdala-ventromedial prefrontal cortex resting state connectivity in bipolar disorder. J Affect Disord. 2018 Jan 1;225:4–12.
Tang Y, Ma Y, Chen X, Fan X, Jiang X, Zhou Y, et al. Age-specific effects of structural and functional connectivity in prefrontal-amygdala circuitry in women with bipolar disorder. BMC Psychiatry. 2018 Jun 5;18(1):177.
Lee I, Nielsen K, Nawaz U, Hall MH, Öngür D, Keshavan M, et al. Diverse pathophysiological processes converge on network disruption in mania. J Affect Disord. 2019 Feb 1;244:115–23.
Young RC, Biggs JT, Ziegler VE, Meyer DA. A rating scale for mania: reliability, validity and sensitivity. Br J Psychiatry J Ment Sci. 1978 Nov;133:429–35.
Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry J Ment Sci. 1979 Apr;134:382–9.
Weiner L, Ossola P, Causin JB, Desseilles M, Keizer I, Metzger JY, et al. Racing thoughts revisited: A key dimension of activation in bipolar disorder. J Affect Disord. 2019 Aug 1;255:69–76.
Herrmann N, Black SE, Lawrence J, Szekely C, Szalai JP. The Sunnybrook Stroke Study: a prospective study of depressive symptoms and functional outcome. Stroke. 1998 Mar;29(3):618–24.
Duarte-Guerra LS, Gorenstein C, Paiva-Medeiros PF, Santo MA, Lotufo Neto F, Wang YP. Clinical utility of the Montgomery-Åsberg Depression Rating Scale for the detection of depression among bariatric surgery candidates. BMC Psychiatry. 2016 Apr 30;16:119.
Esteban O, Markiewicz CJ, Blair RW, Moodie CA, Isik AI, Erramuzpe A, et al. fMRIPrep: a robust preprocessing pipeline for functional MRI. Nat Methods. 2019 Jan;16(1):111–6.
Friston KJ, Williams S, Howard R, Frackowiak RS, Turner R. Movement-related effects in fMRI time-series. Magn Reson Med. 1996 Mar;35(3):346–55.
Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A. 2005 Jul 5;102(27):9673–8.
Murphy K, Birn RM, Handwerker DA, Jones TB, Bandettini PA. The impact of global signal regression on resting state correlations: are anti-correlated networks introduced? NeuroImage. 2009 Feb 1;44(3):893–905.
Yang GJ, Murray JD, Repovs G, Cole MW, Savic A, Glasser MF, et al. Altered global brain signal in schizophrenia. Proc Natl Acad Sci U S A. 2014 May 20;111(20):7438–43.
Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. NeuroImage. 2012 Feb 1;59(3):2142–54.
Ciric R, Nomi JS, Uddin LQ, Satpute AB. Contextual connectivity: A framework for understanding the intrinsic dynamic architecture of large-scale functional brain networks. Sci Rep. 2017 Jul 26;7(1):6537.
Parkes L, Fulcher B, Yücel M, Fornito A. An evaluation of the efficacy, reliability, and sensitivity of motion correction strategies for resting-state functional MRI. NeuroImage. 2018 May 1;171:415–36.
Tian Y, Margulies DS, Breakspear M, Zalesky A. Topographic organization of the human subcortex unveiled with functional connectivity gradients. Nat Neurosci. 2020 Nov;23(11):1421–32.
Bates D, Mächler M, Bolker B, Walker S (2015). “Fitting Linear Mixed-Effects Models Using lme4.” Journal of Statistical Software, 67(1), 1–48. doi:10.18637/jss.v067.i01
Saunders RC, Rosene DL, Van Hoesen GW. Comparison of the efferents of the amygdala and the hippocampal formation in the rhesus monkey: II. Reciprocal and non-reciprocal connections. J Comp Neurol. 1988 May 8;271(2):185–207.
Nieuwenhuys R, Voogd J, van Huijzen C, editors. Telencephalon: Hippocampus and Related Structures. In: The Human Central Nervous System [Internet]. Berlin, Heidelberg: Springer; 2008 [cited 2022 Nov 27]. p. 361–400. Available from: https://doi.org/10.1007/978-3-540-34686-9_12
Paré D, Collins DR, Pelletier JG. Amygdala oscillations and the consolidation of emotional memories. Trends Cogn Sci. 2002 Jul 1;6(7):306–14.
Ritchey M, Dolcos F, Cabeza R. Role of amygdala connectivity in the persistence of emotional memories over time: an event-related FMRI investigation. Cereb Cortex N Y N 1991. 2008 Nov;18(11):2494–504.
McGaugh JL. Memory consolidation and the amygdala: a systems perspective. Trends Neurosci. 2002 Sep;25(9):456.
Phelps EA. Human emotion and memory: interactions of the amygdala and hippocampal complex. Curr Opin Neurobiol. 2004 Apr;14(2):198–202.
Fastenrath M, Coynel D, Spalek K, Milnik A, Gschwind L, Roozendaal B, et al. Dynamic modulation of amygdala-hippocampal connectivity by emotional arousal. J Neurosci Off J Soc Neurosci. 2014 Oct 15;34(42):13935–47.
Dolcos F, LaBar KS, Cabeza R. Remembering one year later: role of the amygdala and the medial temporal lobe memory system in retrieving emotional memories. Proc Natl Acad Sci U S A. 2005 Feb 15;102(7):2626–31.
Greenberg DL, Rice HJ, Cooper JJ, Cabeza R, Rubin DC, Labar KS. Co-activation of the amygdala, hippocampus and inferior frontal gyrus during autobiographical memory retrieval. Neuropsychologia. 2005;43(5):659–74.
Smith APR, Stephan KE, Rugg MD, Dolan RJ. Task and content modulate amygdala-hippocampal connectivity in emotional retrieval. Neuron. 2006 Feb 16;49(4):631–8.
Zhu Y, Gao H, Tong L, Li Z, Wang L, Zhang C, et al. Emotion Regulation of Hippocampus Using Real-Time fMRI Neurofeedback in Healthy Human. Front Hum Neurosci. 2019;13:242.
Li M, Huang C, Deng W, Ma X, Han Y, Wang Q, et al. Contrasting and convergent patterns of amygdala connectivity in mania and depression: a resting-state study. J Affect Disord. 2015 Mar 1;173:53–8.
Singh MK, Kelley RG, Chang KD, Gotlib IH. Intrinsic Amygdala Functional Connectivity in Youth With Bipolar I Disorder. J Am Acad Child Adolesc Psychiatry. 2015 Sep;54(9):763–70.
Luking KR, Repovs G, Belden AC, Gaffrey MS, Botteron KN, Luby JL, et al. Functional connectivity of the amygdala in early-childhood-onset depression. J Am Acad Child Adolesc Psychiatry. 2011 Oct;50(10):1027-1041.e3.
van der Werff SJA, Pannekoek JN, Veer IM, van Tol MJ, Aleman A, Veltman DJ, et al. Resting-state functional connectivity in adults with childhood emotional maltreatment. Psychol Med. 2013 Sep;43(9):1825–36.
Cullen KR, Westlund MK, Klimes-Dougan B, Mueller BA, Houri A, Eberly LE, et al. Abnormal amygdala resting-state functional connectivity in adolescent depression. JAMA Psychiatry. 2014 Oct;71(10):1138–47.
Janak PH, et Kay MT. From Circuits to Behaviour in the Amygdala. Nature 517, n^o 7534 (janvier 2015): 284-92. https://doi.org/10.1038/nature14188.
Zeng LL, Shen H, Liu L, Wang L, Li B, Fang P, et al. Identifying major depression using whole-brain functional connectivity: a multivariate pattern analysis. Brain J Neurol. 2012 May;135(Pt 5):1498–507.
Ambrosi E, Arciniegas DB, Madan A, Curtis KN, Patriquin MA, Jorge RE, et al. Insula and amygdala resting-state functional connectivity differentiate bipolar from unipolar depression. Acta Psychiatr Scand. 2017 Jul;136(1):129–39.
Hamilton JP, Gotlib IH. Neural substrates of increased memory sensitivity for negative stimuli in major depression. Biol Psychiatry. 2008 Jun 15;63(12):1155–62.
Hall J, Parkinson JA, Connor TM, Dickinson A, Everitt BJ. Involvement of the central nucleus of the amygdala and nucleus accumbens core in mediating Pavlovian influences on instrumental behaviour. Eur J Neurosci. 2001 May;13(10):1984–92.
Wang Y, Liu Z, Cai L, Guo R, Dong Y, Huang YH. A Critical Role of Basolateral Amygdala-to-Nucleus Accumbens Projection in Sleep Regulation of Reward Seeking. Biol Psychiatry. 2020 Jun 1;87(11):954–66.
Janak PH, Tye KM. From circuits to behaviour in the amygdala. Nature. 2015 Jan 15;517(7534):284-92. doi: 10.1038/nature14188. PMID: 25592533; PMCID: PMC4565157.
Heller AS, Cohen AO, Dreyfuss MFW, Casey BJ. Changes in cortico-subcortical and subcortico-subcortical connectivity impact cognitive control to emotional cues across development. Soc Cogn Affect Neurosci. 2016 Dec;11(12):1910–8.
Almeida JRC, Versace A, Hassel S, Kupfer DJ, Phillips ML. Elevated amygdala activity to sad facial expressions: a state marker of bipolar but not unipolar depression. Biol Psychiatry. 2010 Mar 1;67(5):414–21.
Eickhoff SB, Yeo BTT, Genon S. Imaging-based parcellations of the human brain. Nat Rev Neurosci. 2018 Nov;19(11):672–86.
Saygin ZM, Kliemann D, Iglesias JE, van der Kouwe AJW, Boyd E, Reuter M, et al. High-resolution magnetic resonance imaging reveals nuclei of the human amygdala: manual segmentation to automatic atlas. NeuroImage. 2017 Jul 15;155:370–82.
Hafeman DM, Chang KD, Garrett AS, Sanders EM, Phillips ML. Effects of medication on neuroimaging findings in bipolar disorder: an updated review. Bipolar Disord. 2012 Jun;14(4):375–410.
Allen M, Poggiali D, Whitaker K, Marshall TR, Kievit R. Raincloud plots: a multi-platform tool for robust data visualization. Wellcome Open Research, 2019

The authors have declared there is NO conflict of interest to disclose

SupplementaryInformation.docx

Download PDF

Journal Publication

published 09 May, 2024

Read the published version in Molecular Psychiatry →

Editorial decision: revise
21 Sep, 2023
Review #2 received at journal
17 Sep, 2023
Reviewer #2 agreed at journal
06 Sep, 2023
Review #1 received at journal
14 Jun, 2023
Reviewer #1 agreed at journal
31 May, 2023
Reviewers invited by journal
31 May, 2023
Submission checks completed at journal
30 May, 2023
First submitted to journal
23 May, 2023
Editor assigned by journal
23 May, 2023

You are reading this latest preprint version

Functional connectivity of the amygdala subnuclei in various mood states of bipolar disorder

Status:

Journal Publication

Version 1

Abstract

Figures

INTRODUCTION

MATERIALS AND METHODS

3.1. Center 1 (Paris)

3.2. Center 2 (Grenoble)

3.3. Center 3 (Geneva)

6.1. Whole-sample analyses

6.2. Sub-sample analyses

6.2.1. Correlations with psychiatric symptoms

6.2.2. Subgroups analyses

RESULTS

DISCUSSION

Lateral amygdala / anterior hippocampus functional connectivity

Medial amygdala-nucleus accumbens functional connectivity

Limitations

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1