Depressive symptoms in older adults are associated with changes in stress-related markers, functional connectivity and brain volume

doi:10.21203/rs.3.rs-5145799/v1

Download PDF

Research Article

Depressive symptoms in older adults are associated with changes in stress-related markers, functional connectivity and brain volume

https://doi.org/10.21203/rs.3.rs-5145799/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

BACKGROUND: Subclinical depressive symptoms increase the risk of developing Alzheimer’s disease (AD). The neurobiological mechanisms underlying this link may involve stress system dysfunction, notably related to the hippocampus which is particularly sensitive to AD. We aimed to investigate the links between blood stress markers and changes in brain regions involved in the stress response in older adults with or without subclinical depressive symptoms.

METHODS: This cross-sectional study was conducted using baseline data from the Age-Well trial. Cognitively unimpaired (CU) older adults with (DepS; n=73) or without (NoDepS; n=58) subclinical depressive symptoms (defined using the 15-item Geriatric Depression Scale) were included in the analyses. Blood cortisol, epinephrine and norepinephrine were measured; as well as the resting-state functional connectivity (rs-FC) between, and gray matter (GM) volume of, the hypothalamus, hippocampus and hippocampal subfields. Blood stress markers levels and neuroimaging measures were compared between groups; then regression analyses were conducted between these measures.

RESULTS: DepS participants showed higher plasma epinephrine levels, which was associated with greater rs-FC between the CA1 and Subiculum hippocampal subfields and the hypothalamus. Lower GM volume in the CA1 and DG/CA2-3-4 subfields was also found in DepS. No between-group differences were observed for blood cortisol and norepinephrine.

CONCLUSIONS: Our findings show that subclinical depressive symptoms are associated with increased sympatho-adrenomedullary axis activity, together with lower GM volume in a hippocampal subfield (i.e., CA1) particularly sensitive to AD. While causation cannot be inferred, these results suggest that screening and treating subclinical depressive symptoms in CU older adults could reduce AD risk.

TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02977819, Registration Date: 2016-11-25.

Computational Neuroscience

Depressive symptoms

Aging

Cortisol

Catecholamines

Hypothalamus

Hippocampal subfields

Subclinical depressive symptoms are associated with higher risk of cognitive decline in cognitively unimpaired (CU) older adults, with each additional symptom increasing the risk of Alzheimer’s disease (AD) by about 20% [1, 2]. “Subclinical depressive symptoms” here refers to depressive symptoms that do not meet the diagnostic criteria for major depression, such as subsyndromal or subthreshold depression and minor depression, as well as depressive symptoms within a low or subclinical range on depression scales [3, 4]. These symptoms are also associated with increased risk of late-life depression (LLD) [5], which is one of the main later-life potentially modifiable risk factors for AD [6]. However, the neurobiological mechanisms underlying the links between depressive symptoms and AD risk remain unclear. Dysfunctions of the stress system, including the sympatho-adrenomedullary (SAM) and the hypothalamic-pituitary-adrenocortical (HPA) axes, have previously been related to the pathophysiology of both LLD and AD [7–9].

The SAM axis provides the most immediate response to stressor exposure – through neural innervation of end organs – and results in increased circulating epinephrine and norepinepinephrine (catecholamines) [10, 11]. Physiologically, the SAM axis effect is counterbalanced by parasympathetic activation, which exerts the opposite response, in order to restore homeostasis. But in the case of major depressive disorder, a predominance of the SAM axis to the detriment of the parasympathetic axis has been reported [12, 13]. Thus, previous studies found that young adults with major depressive disorder showed higher catecholamines levels compared to healthy controls [14–19]. Higher levels of catecholamines have also been observed in AD patients compared to controls in most studies [8, 9, 20–23] although discrepant results were also reported [24–27]. Moreover, higher levels of catecholamines and/or their metabolites in AD patients have been associated with higher levels of a cerebrospinal fluid (CSF) biomarker of synaptic damage [8], lower hippocampal volume, and higher p-tau levels [28].

HPA axis dysfunction leading to elevated circulating cortisol have also been reported in both LLD and AD [7, 29]. Moreover, high cortisol levels have been associated with increased risk of both major depressive disorder and AD [29, 30]. Interestingly, previous works suggested that elevated cortisol levels could be associated with alterations in the hippocampus that are commonly found in both major depressive disorder and AD [31].

To our knowledge, only a few studies have investigated the links between subclinical depressive symptoms and stress markers in older adults [32, 33], highlighting the need for further exploration of these relationships, particularly in this population.

Overall, the stress system integrates a wide range of brain regions involved in detecting and interpreting potential threats, as well as activating the physiological stress response in the body. Notably, the hypothalamus plays a crucial role in the activation and regulation of the SAM and HPA axes. Indeed, on the one hand, its neurons project through the sympathetic arms to the adrenal medulla to induce epinephrine and norepinephrine release, and on the other hand, through the HPA axis hormonal pathway, the hypothalamus can activate cortisol release into the blood from the adrenal cortex [10, 11]. The hypothalamus is regulated by limbic structures, in particular the hippocampus which sends direct neural projections to the hypothalamus via two of its subfields, namely Cornu Ammonis (CA)1 and Subiculum [10, 11].

In older adults, previous studies assessing the role of brain regions on the links between depressive symptoms and blood stress markers have only focused on cortisol and the hippocampus [31]. None of these studies examined links with catecholamines. Moreover, studying depressive symptoms at a subclinical stage (of depression and AD) may provide a better insight into the mechanisms underlying the link between depressive symptoms and AD risk in older adults. In a previous study, we showed that depressive symptoms in CU older adults were associated with neurodegeneration biomarkers mainly in regions belonging to the frontolimbic network, including the hippocampus [34]. Further specification of the links between regional brain changes and blood stress markers in this population is needed to identify mechanistic targets for future preventive strategies.

This study aimed to provide a better understanding of the changes in blood stress markers associated with the presence of subclinical depressive symptoms in older adults and the relationship with the function and structure of specific brain regions involved in the stress response, namely the hypothalamus and hippocampal subfields. We first assessed the links between the presence of subclinical depressive symptoms in CU older adults and blood stress markers including epinephrine, norepinephrine and cortisol – reflecting the SAM and the HPA activity, respectively. Secondly, we investigated the relationships between blood stress markers found to be altered in individuals with subclinical depressive symptoms, and resting-state functional connectivity (rs-FC) magnetic resonance imaging (MRI) between the hypothalamus and hippocampal subfields, as well as gray matter (GM) volume in these regions. We hypothesized that the presence of subclinical depressive symptoms would be associated with higher levels of blood stress markers as a reflection of brain changes in regions involved in the stress response (i.e., hypothalamus and hippocampus), which may partly overlap with AD-sensitive brain areas.

Participants

We included CU older adults from the baseline visit of the Age-Well randomized controlled trial of the Medit-Ageing European project (Fig. 1), sponsored by the French National Institute of Health and Medical Research (INSERM). The design of the trial has been fully described elsewhere [35, 36]. Participants were recruited from the general population. Main eligibility criteria included being: a native French speaker, aged at least 65 years, retired for at least one year, educated for at least 7 years and showing performance within the normal range for age and educational level on standardized cognitive tests (see Tables 1 and 2 in the Age-Well protocol paper [35] for details). Participants had no evidence of a major neurological or psychiatric disorder, chronic disease or acute unstable illness, history of cerebrovascular disease, and no current or recent medication that may interfere with cognitive functioning (including antidepressants and anxiolytics, see [35] for details). Notably, participants were assessed for psychiatric diagnoses during the inclusion process by a physician who inquired about their psychiatric history and ongoing treatments. This included asking about past psychiatric and depressive episodes, current psychotropic medications (type, dosage, and last use), and consumption of psychoactive substances (drugs, alcohol, psychostimulants). Additionally, information on chronic and occasional medication use (type and dosage) was collected. Current depressive symptoms were evaluated by a psychologist using a semi-structured interview, the Montgomery-Åsberg Depression Rating Scale (MADRS) [37], with a cut-off value of 6 (participants with MADRS > 6, indicating mild to major depressive disorder, were excluded). Of the 135 participants included at the baseline visit, four were excluded from our analyses for medical reasons: one with Cushing’s syndrome, two with dehydroepiandrosterone (DHEA) treatment and one with unavailable catecholamines measurements (total study sample size, n = 131). Eight additional participants were excluded from rs-FC MRI analyses because of data unavailability (Fig. 1). All participants provided written informed consent prior to the examinations, and the Age-Well randomized clinical trial was approved by the ethics committee (Comité de Protection des Personnes Nord-Ouest III, Caen, France; trial registration number: EudraCT: 2016-002441-36; IDRCB: 2016-A01767-44; ClinicalTrials.gov Identifier: NCT02977819; registration date: 2016-11-25). All examinations were performed within a 3-month period.

Assessment of subclinical depressive symptoms and classification of participants

In this study, we were specifically interested in the presence of subclinical depressive symptoms, which was assessed using the 15-item version of the Geriatric Depression Scale (GDS). This self-reported questionnaire ranges from 0 to 15, with a higher score indicating more depressive symptoms [38, 39]. Given that all participants were screened for the lack of depression, GDS scores were rather low showing a floor-effect and a non-normal distribution. Therefore, instead of the severity of the symptoms (which lacked variability), we were interested in comparing participants with or without subclinical depressive symptoms. This strategy revealed to be meaningful in a previous study showing that participants with subclinical depressive symptoms had lower structural and functional integrity in a frontolimbic network [34]. Thus, as in the previous study, we split participants into having subclinical depressive symptoms (DepS; GDS > 0, n = 73) or not (NoDepS; GDS = 0, n = 58). This GDS threshold was selected as, while there is no consensus to date in the current literature to define “subclinical” depressive symptoms [3], previous studies have shown that each additional depressive symptom (including 1-point increase in 15-item GDS) significantly increases the risk of AD in CU older adults [1, 40]. The distribution of depressive symptoms according to GDS items in our sample is shown in Supplementary Fig. 1.

Characterization of the sample

Detailed descriptions of all variables used to characterize the sample and as covariates in the statistical analyses are available in Supplementary Material 1.

Blood stress markers

Fasting blood samples were collected from each participant into heparinized (for plasma) and dry (for serum) BD Vacutainer tubes by intravenous cannulation in the morning, typically between 7:30 and 8:00 AM. Participants were also asked to follow a one-day diet excluding tryptophan-rich food (e.g., dried fruits, avocados, nuts, chocolate, bananas).

Plasma epinephrine and norepinephrine quantification

Epinephrine and norepinephrine were quantified using liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS). Briefly, catecholamines and their deuterated internal standards (NE-d₆ and E-d₆, LGC Standards GmbH, Wesel, Germany) were extracted in basic conditions using activated aluminum oxide. After elution in acid condition and evaporation, the analytes were derivatized by sodium cyanoborohydride and acetaldehyde (Merck, Darmstadt, Germany) to form ethyl derivatives as previously reported [41, 42]. Liquid chromatography was conducted on an UFLC chromatographic system (Shimadzu, Kyoto, Japan) connected to a SCIEX QTRAP® 5500 mass spectrometer (SCIEX, Toronto, Canada). Chromatographic separation was performed using a Synergi™ hydro-RP C18 column (Phenomenex, Torrance, USA) in gradient conditions, and mass spectrometry analysis was conducted using the electrospray ion source (ESI) in positive mode. Limit of quantification was < 0.01 µg/L for epinephrine and norepinephrine (n = 20 participants had epinephrine measures below the limit of quantification). Recoveries were 100 ± 7% and 96 ± 5% in plasma samples spiked with epinephrine and norepinephrine, respectively. In the reproducibility assay, coefficient of variations (CV) were lower than 8.2% for epinephrine or norepinephrine at three levels of concentrations for control samples.

Serum cortisol quantification

Serum cortisol concentration (nmol/L) was measured by chemiluminescent immunoassay on automated analyzer COBAS 8000 (Roche diagnostics, Meylan, France) using ready-made commercial reagent kits (Cortisol II, ROCHE). Limit of quantification was 3 nmol/L (or 1.09 µg/L).

Neuroimaging procedure

MRI data acquisition

All participants were scanned on the same MRI scanner (Philips Achieva; 3.0 T) at the Cyceron Center (Caen, France). T1-weighted structural images were acquired using a 3D fast-field echo sequence. Then, a high-resolution T2-weighted spin echo anatomical acquisition and a non-Echo-Planar Imaging (EPI) T2* volume were obtained. Resting-state functional volumes were obtained using an interleaved 2D T2* SENSE EPI sequence. Further details about MRI data acquisition are provided in the Supplementary Material 2.

Structural MRI data

ROI-based method – Hypothalamus, whole hippocampus and hippocampal subfields

Hypothalamus

The hypothalamus region of interest (ROI) was obtained from the Pauli et al., 2018 [43] atlas which includes the bilateral hypothalamus and mamillary nucleus. This atlas was generated using high-accuracy diffeomorphic registration of T1 and T2 weighted structural images from 168 healthy adults from the Human Connectome Project (HCP) (mean age: 28.9 ± 3.6 years, 84F/84M).

Whole hippocampus and hippocampal subfields

The whole bilateral hippocampus ROI was obtained by manually delineating the hippocampi on coronal slices of a whole brain customized template in MNI space of 50 healthy participants from the Imagerie Multimodale de la maladie d'Alzheimer à un stade Précoce (IMAP) protocol (mean age: 39.9 ± 15.2 years, mean education: 13.3 ± 3.3 years, 31F/19M), as previously described elsewhere [44, 45].

The bilateral hippocampal subfields ROIs were obtained by manually delineating the subfields in native ultra high-resolution proton density weighted scans within an independent group of 50 healthy participants also from IMAP (mean age: 46.3 ± 18.0 years, mean education: 13.7 ± 3.8 years, 31F/19M; see [45, 46] for details). Briefly, three hippocampal regions were delineated: (i) CA1 (ii) the Subiculum; and (iii) CA2-CA3-CA4 and dentate gyrus (DG) pooled together in a unique region. Each high-resolution scan, as well as the manually delineated subfields, was then coregistered to each corresponding T1-weighted image.

T1-MRIs were segmented and spatially normalized to the MNI space using the VBM5 toolbox in the SPM5 software. The normalization parameters were then applied to the coregistred manually delineated subfields. Lastly, the 50 MNI normalized hippocampal subfield regions were averaged. The average image of each hippocampal subregion was then binarized with a threshold of 0.4 in order to obtain one normalized bilateral ROI for each of the three delineated regions (e.g., CA1, Subiculum and DG/CA2-3-4) (see [45, 46] for details).

A mask of total GM (excluding the cerebellum) was applied on each ROI to exclude non-GM voxels. This mask was obtained by combining a group mean T1 GM mask and a mask based on the mean non-EPI-T2* volumes in the MNI space, similar to previously published methods [47].

ROI-based gray matter volume extraction

The GM segments issued from the segmented and spatially normalized T1-weighted images were modulated to correct for non-linear warping so that values in resultant images are expressed as volume corrected for brain size (see Supplementary Material 2 for details). The resultant images were used to extract the mean GM volume within the bilateral hypothalamus, whole bilateral hippocampus and hippocampal subfields ROIs of Age-Well participants.

Additional steps for voxelwise analysis and hippocampal 3D representation

The modulated GM segments described above were smoothed and then used in the voxelwise group comparison (DepS < NoDepS) analysis using an analysis of covariance (ANCOVA) model in SPM12, with age, sex, education and anxiety symptoms as covariates and using the group mean T1 GM mask as an inclusive mask (see Supplementary Material 2 for details). The resultant SPM-T map of lower GM volume in the DepS group was then superimposed onto a 3D template surface representation of the whole bilateral hippocampus. This 3D representation was obtained by converting the whole bilateral hippocampus ROI (described previously) onto 3D mesh using the publicly available “Anatomist/BrainVISA” software (http://www.brainvisa.info/) [44, 45].

Resting-state functional MRI data

Preprocessing of resting-state functional MRI

The mean time course of the resting-state functional MRI was corrected for the mean time courses in the white matter, CSF, the whole brain, their derivatives, and the six movement parameters generated from realignment of head motion, and extracted at the individual level in each ROI (hypothalamus, whole hippocampus, hippocampal subfields) [48, 49]. Then, the extracted time courses were used to compute a ROI-to-ROI (5×5) correlation matrix of Pearson’s correlation coefficients (R values) between all pairs of ROIs. We focused our interest only on the pairs regarding the functional connectivity between the hypothalamus and the whole hippocampus, as well as the hippocampal subfields. We performed one-sample t-tests (mean value set to > 0) on the R values of the correlation matrix within the entire sample, the DepS and the NoDepS groups; all analyses were significant (p < 0.001), indicating that the mean time course of all pairs of ROIs studied were positively correlated. R values of the correlation matrix at individual level were considered as the strength of connection between ROIs and were used as the measure of functional connectivity for the analyses. The detailed acquisition and preprocessing procedures are available in the Supplementary Material 2.

Statistical analyses

Between-group differences for demographic, cognitive, and psychoaffective variables were assessed using Student’s t-tests for continuous variables and χ² tests for categorical variables. Between-group differences in blood stress markers, rs-FC and GM volumes (ROI-based method) were assessed using ANCOVAs corrected for age, sex, education and anxiety symptoms.

Pearson’s partial correlations were performed to assess within group associations (i.e., entire sample, DepS and NoDepS groups) between blood stress markers previously found to be altered in the presence of subclinical depressive symptoms (determined by the previous ANCOVAs) and rs-FC on the one hand, and GM volumes on the other hand – corrected for age, sex, education and anxiety symptoms.

No correction for multiple comparisons was applied because our study was based on well-defined a priori hypotheses, and our analyses were strictly confined to these hypotheses. The significance level was set at p < 0.05 for all statistical analyses.

Covariates

To examine the specific effects of the relationships, we conducted analyses corrected for anxiety (for depressive symptoms), body mass, blood pressure, cardiovascular, hypoglycemic, cholesterol-lowering and anti-inflammatory medications (for blood stress markers), and number of depressive episodes, amyloid positivity and APOEε4 status (for brain measures) due to their close association with depression [50], stress [51], or brain structure and function [52–54].

Participants’ characteristics

Demographic data, cognitive performance and psychoaffective symptoms for each group, as well as between-group differences, are reported in Table 1.

The DepS group had a higher proportion of women and higher anxiety symptoms, negative affect, ruminative brooding, as well as lower psychological defusion compared to the NoDepS group. The two groups did not differ in any other demographic, cognitive or psychoaffective variables.

Table 1

**Participants’ characteristics and between-group comparisons.**
Age-Well Cohort (n = 131)	NoDepS Group (n = 58)	DepS Group (n = 73)	Between-group comparisons
Age-Well Cohort (n = 131)	NoDepS Group (n = 58)	DepS Group (n = 73)	T or χ² value	p-value	Mean difference [95% CI]
Demographic data
Sex, Female, n (%)	28 (48)	51 (70)	5.42	0.02^a
Age, years (Range)	69.9 ± 4.0 (65–84)	69.0 ± 3.7 (65–79)	-1.35	0.18^b	-0.91 [-2.25 to 0.42]
Education, years (Range)	13.1 ± 3.2 (7–22)	13.1 ± 3.1 (7–20)	0.03	0.98^b	0.02 [-1.07 to 1.11]
Health and Medication use
BMI (Range)	25.8 ± 3.6 (19–36)	26.5 ± 4.7 (18–44)	0.96	0.34^b	0.71 [-0.76 to 2.19]
SBP mmHg (Range)	134.1 ± 19.2 (99–179)	135.4 ± 21.1 (88–198)	0.36	0.72^b	1.28 [-5.79 to 8.34]
DBP mmHg (Range)	79.5 ± 10.4 (58–102)	79.9 ± 9.7 (58–107)	0.20	0.84^b	0.36 [-3.13 to 3.84]
Cardiovascular treatments, n (%)	20 (34)	28 (38)	0.08	0.78^a
Hypoglycemic treatments, n (%)	3 (5)	2 (3)	0.07	0.79^a
Cholesterol-lowering treatments, n (%)	11 (19)	12 (16)	0.02	0.88^a
Anti-inflammatory treatments, n (%)	2 (3)	3 (4)	< 0.001	1.00^a
Florbetapir SUVR (Range)	0.96 ± 0.21 (0.73–1.76)	0.98 ± 0.18 (0.72–1.65)	0.47	0.64^b	0.02 [-0.05 to 0.09]
Amyloid positive, n (%)	10 (17)	18 (25)	0.73	0.39^a
APOEε4 Carriers, n (%)	15 (26)	20 (27)	< 0.001	1.00^a
Global cognition
MMSE (Range)	28.9 ± 1.0 (26–30)	29.1 ± 1.1 (26–30)	1.15	0.25^b	0.21 [-0.15 to 0.56]
Psychoaffective variables
GDS (Range)	0.0 ± 0.0 (0–0)	2.2 ± 1.8 (1–11)	NA
STAI-B (Range)	31.6 ± 6.2 (20–51)	36.7 ± 6.8 (24–54)	4.43	< 0.001^b	5.10 [2.82 to 7.37]
History of past depressive episodes, n (%)	7 (12)	15 (21)	1.11	0.29^a
Number of past depressive episodes			4.32	0.12^a
One depressive episode, n (%)	7 (12)	10 (14)
Two depressive episodes, n (%)	0 (0)	5 (7)
PANAS Positive (Range)	33.8 ± 4.8	32.8 ± 4.6	-1.21	0.23^b	-1.01 [-2.65 to 0.64]
PANAS Negative (Range)	11.1 ± 1.7	12.2 ± 3.0	2.65	0.01^b	1.16 [0.30 to 2.03]
RRS Brooding (Range)	7.6 ± 1.9	8.4 ± 2.5	2.00	0.047^b	0.80 [0.01 to 1.58]
DDS Defusion (Range)	36.1 ± 5.8	33.2 ± 5.1	-3.07	0.003^b	-2.92 [-4.79 to -1.04]
ERQ Cognitive reappraisal (Range)	30.1 ± 5.7	29.3 ± 5.8	-0.81	0.42^b	-0.83 [-2.83 to 1.18]
ERQ Expressive suppression (Range)	16.7 ± 5.5	16.4 ± 4.8	-0.35	0.73^b	-0.32 [-2.10 to 1.46]

Data are presented as mean ± standard deviation of participants unless otherwise indicated. Between-group differences were assessed using ^aχ² tests for categorical variables and ^bStudent’s t-tests for continuous variables. Values in bold correspond to significant p-values (p < 0.05). N, Sample size; NoDepS, Group without depressive symptoms; DepS, Group with subclinical depressive symptoms; SUVR, Standard Uptake Value Ratio; MMSE, Mini-Mental Sate Examination; GDS, Geriatric Depression Scale; STAI-B, State-Trait Anxiety Inventory form Y-B; BMI, Body Mass Index; SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; PANAS, Positive and Negative Affect Schedule; RRS, Rumination Response Scale; DDS, Drexel Defusion Scale; ERQ, Emotion Regulation Questionnaire; NA, Not applicable.

Between-group differences in blood stress markers

DepS participants had higher epinephrine levels compared to NoDepS participants (Fig. 2). The difference remained significant without anxiety symptoms as a covariate (F = 5.403 p = 0.022), as well as after additional corrections for body mass index, systolic blood pressure, diastolic blood pressure, cardiovascular, hypoglycemic, cholesterol-lowering and anti-inflammatory treatments, and number of past depressive episodes (Supplementary Table 1). No differences between groups were observed for cortisol and norepinephrine levels.

Between-group differences in hippocampal-hypothalamic functional connectivity

DepS participants showed higher rs-FC between the CA1 subfield and the hypothalamus compared with NoDepS participants (Table 2). The difference remained significant after additional corrections for number of past depressive episodes, amyloid positivity and APOEε4 status (F = 4.314 p = 0.040), but was no longer significant without anxiety symptoms as a covariate (F = 3.801 p = 0.054). No between-group differences were observed between the other hippocampal subfields and the hypothalamus.

Table 2

**Between-group comparisons of hippocampal-hypothalamic resting-state functional connectivity.**
Resting-State Functional Connectivity (R value)	NoDepS Group (n = 53)	DepS Group (n = 70)	Between-group comparisons
Resting-State Functional Connectivity (R value)	NoDepS Group (n = 53)	DepS Group (n = 70)	F value	p-value
Whole Hippocampus – Hypothalamus (Range)	0.07 ± 0.15 (-0.31–0.41)	0.12 ± 0.18 (-0.27–0.53)	2.97	0.09
CA1 – Hypothalamus (Range)	0.04 ± 0.16 (-0.44–0.45)	0.11 ± 0.16 (-0.26–0.52)	4.31	0.04
Subiculum – Hypothalamus (Range)	0.07 ± 0.15 (-0.29–0.40)	0.12 ± 0.18 (-0.30–0.53)	2.51	0.12
DG/CA2-3-4 – Hypothalamus (Range)	0.06 ± 0.16 (-0.28–0.35)	0.11 ± 0.17 (-0.29–0.50)	2.95	0.09

Data are presented as mean ± standard deviation of participants. Between-group differences were assessed using ANCOVAs adjusted for age, sex, education and anxiety symptoms. Values in bold correspond to significant p-values (p < 0.05). N, Sample size; NoDepS, Group without depressive symptoms; DepS, Group with subclinical depressive symptoms; DG, Dentate Gyrus; CA, Cornu Ammonis.

Intra-group associations between epinephrine and functional connectivity

In DepS participants, we found that higher epinephrine was associated with higher rs-FC between the CA1 and Subiculum subfields and the hypothalamus (Fig. 3 and Supplementary Table 2). The same results were found within the entire sample (Supplementary Table 2). These associations remained significant without anxiety symptoms as a covariate within DepS participants (CA1 and hypothalamus, r = 0.329 p = 0.007; Subiculum and hypothalamus, r = 0.305 p = 0.012) and within the entire sample (CA1 and hypothalamus, r = 0.224 p = 0.014; Subiculum and hypothalamus, r = 0.221 p = 0.015), as well as after additional corrections for number of past depressive episodes, amyloid positivity and APOEε4 status (Supplementary Table 3). No significant correlations were found between epinephrine and rs-FC within NoDepS participants.

Between-group differences in hypothalamic and hippocampal gray matter volumes

ROI-based method

Regarding the ROI-based approach analyses, DepS participants showed lower GM volume in the whole hippocampus, more specifically in the CA1 and DG/CA2-3-4 subfields, compared with NoDepS participants (Table 3). GM volume difference remained significant after additional corrections for number of past depressive episodes, amyloid positivity and APOEε4 status in the CA1 (F = 6.296 p = 0.013) and DG/CA2-3-4 (F = 6.206 p = 0.014), but was no longer significant in the whole hippocampus (F = 3.431 p = 0.066). When anxiety symptoms were not included as a covariate, this difference remained significant in the CA1 (F = 3.978 p = 0.048), but not in the DG/CA2-3-4 (F = 3.347 p = 0.070) or in the whole hippocampus (F = 2.293 p = 0.132). No between-group differences were observed for the hypothalamus and the Subiculum.

Table 3

**Between-group comparisons of hypothalamic and hippocampal gray matter volumes.**
Gray Matter Volumes (mm³)	NoDepS Group (n = 58)	DepS Group (n = 73)	Between-group comparisons
Gray Matter Volumes (mm³)	NoDepS Group (n = 58)	DepS Group (n = 73)	F value	p-value
Hypothalamus (Range)	1147.57 ± 99.01 (729.76-1348.39)	1141.60 ± 123.56 (743.69-1327.87)	0.64	0.43
Whole Hippocampus (Range)	6941.96 ± 309.23 (6160.21-7505.97)	6899.08 ± 388.22 (5985.23-7659.28)	4.13	0.04
CA1 (Range)	2074.54 ± 117.33 (1787.21-2339.97)	2041.08 ± 156.96 (1615.14-2359.25)	7.36	0.01
Subiculum (Range)	1408.67 ± 171.72 (977.96-1775.30)	1430.34 ± 153.40 (1058.02-1716.51)	0.77	0.38
DG/CA2-3-4 (Range)	2400.78 ± 98.14 (2156.96-2555.09)	2382.40 ± 125.11 (2040.28-2665.28)	6.29	0.01

VBM-based method – Hippocampal 3D representation

Figure 4A illustrates the hippocampal surface projection of the voxelwise group comparison analysis on GM volume, i.e., lower hippocampal GM volume in DepS participants compared with NoDepS participants. Areas of lowest GM volume were mostly located on the lateral parts of the superior and inferior views, with a left-sided predominance in the anterior part. Based on hippocampal maps of subfields delineation provided in previous publications [44, 45] and schematized in Fig. 4B, regions of lowest GM volume mainly corresponded to the CA1 and DG/CA2-3-4 subfields, while most of the Subiculum was relatively spared. For the sake of comparison, GM atrophy in AD patients and across normal aging are also shown on the same hippocampal views in Fig. 4C (adapted from [55]) – the areas of greatest atrophy mainly corresponded to the CA1 in AD patients, while across normal aging, GM atrophy predominated in the Subiculum.

Intra-group associations between epinephrine and hypothalamic and hippocampal gray matter volumes

We found no significant correlation between epinephrine and hypothalamic and hippocampal GM volumes, in any group (Supplementary Table 2).

The aim of this study was to assess the relationship between the presence of subclinical depressive symptoms in CU older adults and blood stress markers – reflecting SAM and HPA activity, as well as the links with the function and structure of brain regions involved in stress response. We showed that participants with subclinical depressive symptoms had higher levels of plasma epinephrine and greater rs-FC between the CA1 subfield and the hypothalamus, as well as lower GM volume in the whole hippocampus, and more specifically in the CA1 and DG/CA2-3-4 subfields. We also found that higher epinephrine was associated with greater rs-FC between the CA1 and Subiculum subfields and the hypothalamus, but no association was found with hypothalamic or hippocampal (whole and subfields) GM volumes – within participants with subclinical depressive symptoms. Furthermore, we found no significant differences in blood cortisol and norepinephrine between participants with and without subclinical depressive symptoms.

Subclinical depressive symptoms are associated with higher SAM axis activity

We found higher plasma epinephrine in CU older adults with subclinical depressive symptoms compared to those without depressive symptoms. Higher epinephrine levels were consistently reported in previous studies assessing younger patients with major depressive disorder [15, 18], while the only study assessing this link in older adults with preclinical depressive symptoms failed to find such a relationship [32]. Our findings suggest that higher epinephrine is present since the preclinical stage of LLD, reflecting increased SAM axis activity. Consistent with this interpretation, previous studies also found lower heart rate variability (HRV) - which is another marker of increased SAM activity – in both older and young depressive patients [12, 13]. Interestingly, the same evidence for increased SAM activity has been found in AD including both higher epinephrine and lower HRV [8, 23, 56, 57]. SAM dysregulation thus appears to be a common event in both LLD and AD. It is possible that this increase in epinephrine is involved in the links between subclinical depressive symptoms and greater AD risk. Since we did not find an association between higher epinephrine and lower hippocampal GM volume, further investigations, e.g., assessing the links with other AD biomarkers (i.e., amyloid and tau pathologies), are needed to test for other possible mechanisms.

In addition, participants with subclinical depressive symptoms showed higher rs-FC between the CA1 subfield and the hypothalamus, and this increase was associated with greater epinephrine in those participants. The hippocampus is thought to exert an inhibitory influence on the hypothalamus, via projections from the CA1 and Subiculum subfields [10, 11]. One plausible explanation is that this increased CA1-hypothalamus connectivity reflects an attempt to downregulate SAM axis activity and thus reduce the elevated epinephrine in participants with subclinical depressive symptoms (see Fig. 5).

In contrast, we did not find differences in norepinephrine or cortisol in participants with or without subclinical depressive symptoms. Previous findings are discrepant for norepinephrine as, in younger individuals clinical [14–19] or subclinical depressive symptoms [58, 59] were associated with higher norepinephrine, while no relationship was found between subclinical depressive symptoms and norepinephrine in older adults [32]. Regarding cortisol, while several previous studies have shown that clinical levels of depressive symptoms in late life are related to higher cortisol [7], the few studies investigating subclinical depressive symptoms in older adults found no association [32, 33]. It is thus possible that the link with cortisol only exists at the clinical stage of depressive symptoms; otherwise, the lack of link could be due to limitations in the method used to measure cortisol (see below).

Subclinical depressive symptoms are associated with lower gray matter volume in hippocampal subfields sensitive to LLD and AD

Participants with subclinical depressive symptoms showed lower GM volume in the hippocampus (as evidenced in our previous study [34]), more specifically in CA1 and DG/CA2-3-4. These findings are partly consistent with the only two previous studies investigating the links between subclinical depressive symptoms and hippocampal subfield volumes in older adults [60, 61]: one found a link with lower GM volume in the same hippocampal subfields (CA1, DG and CA3-4 [60]), while the other found a link with less age-related GM volume loss in CA1 and Subiculum [61]. Interestingly, DG and CA1 are the most commonly altered hippocampal subfields both in LLD [62–64] and in younger patients with major depressive disorder [65–68]. Moreover, CA1 atrophy is also known to occur specifically in the early stages of AD and is a sensitive predictor of conversion from Mild Cognitive Impairment (MCI) to AD dementia [69, 70]. As illustrated in Fig. 4, that pattern of lower hippocampal GM volume associated with subclinical depressive symptoms is close to that observed in AD but not to that observed across normal, physiological, aging. These findings suggest that these common hippocampal subfields alterations could partly underlie the links between subclinical depressive symptoms and increased risk of LLD and AD.

Strengths and limitations

The main strength of our study was to investigate, within the same study, the relationship between the presence of subclinical depressive symptoms and the main blood stress markers reflecting the SAM and HPA activity in CU older adults, combined with the assessment of the function and structure of brain regions modulating those circulating markers using neuroimaging data. However, the cross-sectional design of our study is a limitation as it prevents us from assessing the links between baseline levels and changes over time in depressive symptoms, blood markers, and brain structure and function. Future longitudinal analyses would allow investigation of this question. Furthermore, it is possible that the lack of link with cortisol reflects the fact that this stress marker was collected from one morning blood sample at the same hour in our study, instead of several times during the day to better capture its overall levels as recommended in several studies [71–73], given that this hormone follows a circadian rhythm. Some of the DepS participants had high epinephrine levels. When these values are excluded from the analysis, the observed between-group difference is no longer significant. However, no medical or technical reasons were reported during the study to justify excluding these measures. Moreover, these values contribute to sufficient variability, which is essential for capturing meaningful differences in our population of interest. In addition, the method used to estimate hippocampal subfields GM volumes is not optimal, segmentations from native ultra high-resolution T2-weighted images should be considered in future studies to improve the accuracy of the measurements. Excluding anxiety symptoms as a covariate reduced the significance of the between-group difference in functional connectivity and gray matter volumes. Future studies investigating the respective relationships between anxiety and depressive symptoms (with or without reciprocal correction) and stress or neuroimaging markers are needed to clarify their specific links. Although our study is based on specific hypotheses, no corrections were made for multiple comparisons and further replication will be needed to address the robustness of these findings. In addition, subclinical depressive symptoms were assessed using a self-report questionnaire; as it is subjective, the measure could be biased by demand characteristics, awareness and introspective ability [74]. Moreover, the threshold we used to define “subclinical” depressive symptoms was not validated; further studies might allow researchers to select the specific threshold for “clinically relevant” subclinical symptoms. Yet, we previously found a link between subclinical depressive symptoms assessed using the same questionnaire (GDS; with the same cut-off value used to split participants) and neurodegeneration biomarkers, both in our sample and in a completely independent sample (ADNI) [34], which suggests that this measure is relatively reliable.

The presence of subclinical depressive symptoms in CU older adults was associated with higher plasma epinephrine, which itself was related to higher rs-FC between the CA1 and Subiculum subfields and the hypothalamus. These associations may indicate a compensatory upregulation of the SAM axis activity. Moreover, subclinical depressive symptoms were also associated with lower GM volume in specific hippocampal subfields (i.e., DG and CA1) which are known to be altered in LLD and AD. Overall, these findings suggest that two distinct mechanisms (i.e., higher SAM axis activity and lower GM volume in sensitive hippocampal subfields) at this preclinical stage may be involved in the increased risk of developing LLD and AD.

AD: Alzheimer’s disease
ANCOVA: Analysis of covariance
CA: Cornu Ammonis
CU: Cognitively unimpaired
CSF: Cerebrospinal fluid
CV: Coefficient of variations
DepS: Group with subclinical depressive symptoms
DG: Dentate gyrus
DHEA: Dehydroepiandrosterone
EPI: Echo-Planar Imaging
ESI: Electrospray ion source
GDS: Geriatric Depression Scale
GM: Gray matter
HCP: Human Connectome Project
HPA: Hypothalamic-pituitary-adrenocortical
IMAP: Imagerie Multimodale de la maladie d'Alzheimer à un stade Précoce
INSERM: National Institute of Health and Medical Research
LC-MS/MS: Liquid chromatography coupled to tandem mass spectrometry
LLD: Late-life depression
MADRS: Montgomery-Åsberg Depression Rating Scale
MRI: Magnetic resonance imaging
NoDepS: Group without depressive symptoms
ROI: Region of interest
rs-FC: Resting-state functional connectivity
SAM: Sympatho-adrenomedullary
STAI-B: State-Trait Anxiety Inventory form Y-B

Availability of data and materials

Data and code are made available on request following a formal data sharing agreement and approval by the consortium and executive committee (https://silversantestudy.eu/2020/09/25/data-sharing). The Material can be mobilized, under the conditions and modalities defined in the Medit-Ageing Charter by any research team belonging to an Academic, for carrying out a scientific research project relating to the scientific theme of mental health and well-being in older people. The Material may also be mobilized by non-academic third parties, under conditions, in particular financial, which will be established by separate agreement between Inserm and by the said third party. Data sharing policies described in the Medit-Ageing charter are in compliance with our ethics approval and guidelines from our funding body.

Acknowledgments

We acknowledge A. Cognet, V. Le Franc, C. Gaubert, G. Le Du, MSc, T. Jorand, M. Botton, MSc, A. Joret Philippe, MSc, S. Egret, MSc, P. Lacheray, MSc, J. Lebahar, PhD, C. Tomadesso, PhD, I. Moulinet, PhD, S. Réhel, PhD, H. Espérou, MD, PhD, E. Frison, MD, PhD, and the Cyceron MRI-PET staff members for their help with recruitment and data acquisition or administrative support. We also acknowledge the Euclid team, the sponsor (INSERM) and all the participants of the study for their contribution.

The Medit-Ageing Research Group:

Florence Allais¹, Claire André², Pierre Champetier^2,3, Fabienne Collette⁴, Sophie Dautricourt², Marion Delarue², Séverine Fauvel², Francesca Felisatti², Victor Ferment², Eglantine Ferrand-Devouge², Eric Frison¹, Julie Gonneaud², Antoine Garnier-Crussard², Anaïs Hamel², Oriane Hébert², Olga M. Klimecki⁵, Pierre Krolak-Salmon⁶, Valérie Lefranc², Antoine Lutz⁷, Valentin Ourry², Anne Quillard², Géraldine Rauchs², Eric Salmon⁴, Anne-Laure Turpin², Patrik Vuilleumier⁸, Miranka Wirth⁹.

¹EUCLID/F-CRIN Clinical Trials Platform, Bordeaux, France

²Normandie Univ, UNICAEN, INSERM, U1237, PhIND "Physiopathology and Imaging of Neurological Disorders", NeuroPresage Team, Institut Blood and Brain @ Caen-Normandie, GIP Cyceron, Caen, France

³Unité 1077 NIMH "Neuropsychologie et Imagerie de la Mémoire Humaine," Institut National de la Santé et de la Recherche Médicale, Normandie Université, Université de Caen, PSL Université, EPHE, CHU de Caen-Normandie, GIP Cyceron, Caen, France.

⁴GIGA–Cyclotron Research Centre, In Vivo Imaging and Psychology and Cognitive Neuroscience Unit, Liège University, Liège, Belgium

⁵Clinical Psychology and Behavioral Neuroscience, Faculty of Psychology, Technische Universität Dresden, 01187, Dresden, Germany

⁶Hospices Civils de Lyon, Lyon, France

⁷Lyon Neuroscience Research Center, Institut National de la Santé et de la Recherche Médicale Unité 1028, Centre National de la Recherche Scientifique Unité Mixte de Recherche 5292, Lyon University, Lyon, France

⁸Swiss Center for Affective Sciences, Department of Medicine, University of Geneva, Geneva, Switzerland

⁹Deutsches Zentrum für Neurodegenerative Erkrankungen, Dresden, Germany

Funding

The Age-Well randomized clinical trial is part of the Medit-Ageing project and is funded through the European Union’s Horizon 2020 Research and Innovation Program (grant 667696), Institut National de la Santé et de la Recherche Médicale, Région Normandie, and Fondation MMA des Entrepreneurs du Futur.

Authors’ Contributions

Dr Poisnel had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: ET, GC, GP. Acquisition, analysis, or interpretation of data: ET, RDF, LC, CP, AC, EK, BL, FM, DR, LC, SH, DV, VDLS, NLM, GC, GP. Drafting of the manuscript: ET, GC, GP. Critical revision of the manuscript for important intellectual content: ET, RDF, LC, CP, AC, EK, BL, FM, DR, LC, SH, DV, VDLS, NLM, GC, GP. Statistical analysis: ET. Obtained funding: ET, NLM, GC, GP. Administrative, technical, or material support: BL, FM, DV, GP. Supervision: GC, GP. Other–Principal investigators: GC, VDLS (MD, principal investigator).

Ethics declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent prior to the examinations, and the Age-Well randomized clinical trial was approved by the ethics committee (Comité de Protection des Personnes Nord-Ouest III, Caen, France; trial registration number: EudraCT: 2016-002441-36; IDRCB: 2016-A01767-44; ClinicalTrials.gov Identifier: NCT02977819; registration date: 2016-11-25).

Consent for publication

Not applicable.

Competing interests

Dr Touron reported grant from French Ministry of Higher Education and Research (PhD grant). Dr Kuhn reported grant from the French Ministry of Higher Education and Research (PhD grant), and support from INSERM (PhD). Dr Marchant, Dr Chételat and Dr Poisnel reported grants from European Union’s Horizon 2020 Research and Innovation Program under grant agreement No 667696 during the conduct of the study. Dr Marchant reported grants from a Senior Fellowship from the Alzheimer’s Society (AS-SF-15b-002). Dr Chételat reported grants, personal fees and nonfinancial support from Institut National de la Santé et de la Recherche Médicale (INSERM); personal fees from Fondation Entrepreneurs MMA, grants and personal fees from Fondation Alzheimer, grants from Région Normandie, grants from Fondation Recherche Alzheimer, grants from Association France Alzheimer, outside the submitted work. No other disclosures were reported.

Wilson RS, Barnes LL, de Leon CFM, Aggarwal NT, Schneider JS, Bach J et al (2002) Depressive symptoms, cognitive decline, and risk of AD in older persons. Neurology 59:364–370
Yin J, John A, Cadar D (2024) Bidirectional Associations of Depressive Symptoms and Cognitive Function Over Time. JAMA Netw Open 7:e2416305
Rodríguez MR, Nuevo R, Chatterji S, Ayuso-Mateos JL (2012) Definitions and factors associated with subthreshold depressive conditions: a systematic review. BMC Psychiatry 12:181
Donovan NJ, Hsu DC, Dagley AS, Schultz AP, Amariglio RE, Mormino EC et al (2015) Depressive Symptoms and Biomarkers of Alzheimer’s Disease in Cognitively Normal Older Adults. J Alzheimers Dis 46:63–73
Jeuring HW, Huisman M, Comijs HC, Stek ML, Beekman ATF (2016) The long-term outcome of subthreshold depression in later life. Psychol Med 46:2855–2865
Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S et al (2020) Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet 396:413–446
Belvederi Murri M, Pariante C, Mondelli V, Masotti M, Atti AR, Mellacqua Z et al (2014) HPA axis and aging in depression: systematic review and meta-analysis. Psychoneuroendocrinology 41:46–62
Henjum K, Watne LO, Godang K, Halaas NB, Eldholm RS, Blennow K et al (2022) Cerebrospinal fluid catecholamines in Alzheimer’s disease patients with and without biological disease. Transl Psychiatry 12:1–9
Fitzgerald PJ (2010) Is elevated norepinephrine an etiological factor in some cases of Alzheimer’s disease? Curr Alzheimer Res 7:506–516
Godoy LD, Rossignoli MT, Delfino-Pereira P, Garcia-Cairasco N, de Lima Umeoka EH A Comprehensive Overview on Stress Neurobiology: Basic Concepts and Clinical Implications. Frontiers in Behavioral Neuroscience [Internet]. 2018 [cited 2022 Apr 6];12. https://www.frontiersin.org/article/10.3389/fnbeh.2018.00127
Ulrich-Lai YM, Herman JP (2009) Neural regulation of endocrine and autonomic stress responses. Nat Rev Neurosci 10:397–409
Kemp AH, Quintana DS, Gray MA, Felmingham KL, Brown K, Gatt JM (2010) Impact of Depression and Antidepressant Treatment on Heart Rate Variability: A Review and Meta-Analysis. Biol Psychiatry 67:1067–1074
Brown L, Karmakar C, Gray R, Jindal R, Lim T, Bryant C (2018) Heart rate variability alterations in late life depression: A meta-analysis. J Affect Disord 235:456–466
Carney RM, Freedland KE, Veith RC (2005) Depression, the autonomic nervous system, and coronary heart disease. Psychosom Med 67(Suppl 1):S29–33
Peacock BN, Scheiderer DJ, Kellermann GH (2017) Biomolecular aspects of depression: A retrospective analysis. Compr Psychiatr 73:168–180
Veith RC, Lewis N, Linares OA, Barnes RF, Raskind MA, Villacres EC et al (1994) Sympathetic nervous system activity in major depression. Basal and desipramine-induced alterations in plasma norepinephrine kinetics. Arch Gen Psychiatry 51:411–422
Lechin F, van der Dijs B, Orozco B, Lechin ME, Báez S, Lechin AE et al (1995) Plasma neurotransmitters, blood pressure, and heart rate during supine-resting, orthostasis, and moderate exercise conditions in major depressed patients. Biol Psychiatry 38:166–173
Koslow SH, Maas JW, Bowden CL, Davis JM, Hanin I, Javaid J (1983) CSF and Urinary Biogenic Amines and Metabolites in Depression and Mania: A Controlled, Univariate Analysis. Arch Gen Psychiatry 40:999–1010
Grossman F, Potter W (1999) Catecholamines in depression: a cumulative study of urinary norepinephrine and its major metabolites in unipolar and bipolar depressed patients versus healthy volunteers at the NIMH. Psychiatry Res
Raskind MA, Peskind ER, Halter JB, Jimerson DC (1984) Norepinephrine and MHPG levels in CSF and plasma in Alzheimer’s disease. Arch Gen Psychiatry 41:343–346
Elrod R, Peskind ER, DiGiacomo L, Brodkin KI, Veith RC, Raskind MA (1997) Effects of Alzheimer’s disease severity on cerebrospinal fluid norepinephrine concentration. Am J Psychiatry 154:25–30
Stefani A, Olivola E, Liguori C, Hainsworth AH, Saviozzi V, Angileri G et al (2015) Catecholamine-Based Treatment in AD Patients: Expectations and Delusions. Front Aging Neurosci 7:67
Peskind ER, Elrod R, Dobie DJ, Pascualy M, Petrie E, Jensen C et al (1998) Cerebrospinal fluid epinephrine in Alzheimer’s disease and normal aging. Neuropsychopharmacology 19:465–471
Pan X, Kaminga AC, Jia P, Wen SW, Acheampong K, Liu A (2020) Catecholamines in Alzheimer’s Disease: A Systematic Review and Meta-Analysis. Front Aging Neurosci 12:184
Umegaki H, Ikari H, Nakahata H, Yoshimura J, Endo H, Yamamoto T et al (2000) Low plasma epinephrine in elderly female subjects of dementia of Alzheimer type. Brain Res 858:67–70
Bemelmans KJ, Noort A, de Rijk R, Middelkoop HAM, van Kempen GMJ, Goekoop JG (2007) Plasma cortisol and norepinephrine in Alzheimer’s disease: opposite relations with recall performance and stage of progression. Acta Neuropsychiatr 19:231–237
Kurup RK, Kurup PA (2003) Hypothalamic digoxin, hemispheric chemical dominance, and Alzheimer’s disease. Int J Neurosci 113:361–381
van Hooren RWE, Verhey FRJ, Ramakers IHGB, Jansen WJ, Jacobs HIL (2021) Elevated norepinephrine metabolism is linked to cortical thickness in the context of Alzheimer’s disease pathology. Neurobiol Aging 102:17–22
Ouanes S, Popp J (2019) High Cortisol and the Risk of Dementia and Alzheimer’s Disease: A Review of the Literature. Front Aging Neurosci [Internet]. [cited 2019 Apr 24];11. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6405479/
Qin D, Rizak J, Feng X, Yang S, Lü L, Pan L et al (2016) Prolonged secretion of cortisol as a possible mechanism underlying stress and depressive behaviour. Sci Rep 6:30187
Geerlings MI, Gerritsen L, Late-Life, Depression (2017) Hippocampal Volumes, and Hypothalamic-Pituitary-Adrenal Axis Regulation: A Systematic Review and Meta-analysis. Biol Psychiatry 82:339–350
Ryff CD, Love GD, Urry HL, Muller D, Rosenkranz MA, Friedman EM et al (2006) Psychological Well-Being and Ill-Being: Do They Have Distinct or Mirrored Biological Correlates? PPS. 75:85–95
Bremmer MA, Deeg DJH, Beekman ATF, Penninx BWJH, Lips P, Hoogendijk WJG (2007) Major Depression in Late Life Is Associated with Both Hypo- and Hypercortisolemia. Biol Psychiatry 62:479–486
Touron E, Moulinet I, Kuhn E, Sherif S, Ourry V, Landeau B et al (2022) Depressive symptoms in cognitively unimpaired older adults are associated with lower structural and functional integrity in a frontolimbic network. Mol Psychiatry. ;1–10
Poisnel G, Arenaza-Urquijo E, Collette F, Klimecki OM, Marchant NL, Wirth M et al (2018) The Age-Well randomized controlled trial of the Medit-Ageing European project: Effect of meditation or foreign language training on brain and mental health in older adults. Alzheimers Dement (N Y) 4:714–723
Chételat G, Lutz A, Klimecki O, Frison E, Asselineau J, Schlosser M et al (2022) Effect of an 18-Month Meditation Training on Regional Brain Volume and Perfusion in Older Adults: The Age-Well Randomized Clinical Trial. JAMA Neurol 79:1165–1174
Montgomery SA, Asberg M (1979) A new depression scale designed to be sensitive to change. Br J Psychiatry 134:382–389
Sheikh JI, Yesavage JA (1986) Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. Clin Gerontologist: J Aging Mental Health 5:165–173
Wancata J, Alexandrowicz R, Marquart B, Weiss M, Friedrich F (2006) The criterion validity of the Geriatric Depression Scale: a systematic review. Acta Psychiatr Scand 114:398–410
Ezzati A, Katz MJ, Derby CA, Zimmerman ME, Lipton RB (2019) Depressive Symptoms Predict Incident Dementia in a Community Sample of Older Adults: Results From the Einstein Aging Study. J Geriatr Psychiatry Neurol 32:97–103
Ellis AG, Zeglinski PT, Coleman KE, Whiting MJ (2017) Dilute, derivatise and shoot: Measurement of urinary free metanephrines and catecholamines as ethyl derivatives by LC-MSMS. Clin Mass Spectrom 4–5:34–41
Ji C, Walton J, Su Y, Tella M (2010) Simultaneous determination of plasma epinephrine and norepinephrine using an integrated strategy of a fully automated protein precipitation technique, reductive ethylation labeling and UPLC-MS/MS. Anal Chim Acta 670:84–91
Pauli WM, Nili AN, Tyszka JM (2018) A high-resolution probabilistic in vivo atlas of human subcortical brain nuclei. Sci Data 5:180063
La Joie R, Fouquet M, Mézenge F, Landeau B, Villain N, Mevel K et al (2010) Differential effect of age on hippocampal subfields assessed using a new high-resolution 3T MR sequence. NeuroImage 53:506–514
Perrotin A, de Flores R, Lamberton F, Poisnel G, La Joie R, de la Sayette V et al (2015) Hippocampal Subfield Volumetry and 3D Surface Mapping in Subjective Cognitive Decline. J Alzheimer’s Disease 48:S141–S150
de Flores R, Mutlu J, Bejanin A, Gonneaud J, Landeau B, Tomadesso C et al (2017) Intrinsic connectivity of hippocampal subfields in normal elderly and mild cognitive impairment patients. Hum Brain Mapp 38:4922–4932
La Joie R, Landeau B, Perrotin A, Bejanin A, Egret S, Pélerin A et al (2014) Intrinsic connectivity identifies the hippocampus as a main crossroad between Alzheimer’s and semantic dementia-targeted networks. Neuron 81:1417–1428
Mutlu J, Landeau B, Gaubert M, de La Sayette V, Desgranges B, Chételat G (2017) Distinct influence of specific versus global connectivity on the different Alzheimer’s disease biomarkers. Brain 140:3317–3328
Xia M, Wang J, He Y (2013) BrainNet Viewer: A Network Visualization Tool for Human Brain Connectomics. PLoS ONE 8:e68910
Braam AW, Copeland JRM, Delespaul PAEG, Beekman ATF, Como A, Dewey M et al (2014) Depression, subthreshold depression and comorbid anxiety symptoms in older Europeans: Results from the EURODEP concerted action. J Affect Disord 155:266–272
Rodrigues SM, LeDoux JE, Sapolsky RM (2009) The influence of stress hormones on fear circuitry. Annu Rev Neurosci 32:289–313
Chételat G, La Joie R, Villain N, Perrotin A, de La Sayette V, Eustache F et al (2013) Amyloid imaging in cognitively normal individuals, at-risk populations and preclinical Alzheimer’s disease. Neuroimage Clin 2:356–365
Chételat G, Fouquet M (2013) Neuroimaging biomarkers for Alzheimer’s disease in asymptomatic APOE4 carriers. Rev Neurol 169:729–736
Ancelin M-L, Carrière I, Artero S, Maller J, Meslin C, Ritchie K et al (2019) Lifetime major depression and grey-matter volume. J Psychiatry Neurosci 44:45–53
Chételat G, Fouquet M, Kalpouzos G, Denghien I, De la Sayette V, Viader F et al (2008) Three-dimensional surface mapping of hippocampal atrophy progression from MCI to AD and over normal aging as assessed using voxel-based morphometry. Neuropsychologia 46:1721–1731
da Silva VP, Ramalho Oliveira BR, Tavares Mello RG, Moraes H, Deslandes AC, Laks J (2018) Heart Rate Variability Indexes in Dementia: A Systematic Review with a Quantitative Analysis. Curr Alzheimer Res 15:80–88
Zulli R, Nicosia F, Borroni B, Agosti C, Prometti P, Donati P et al (2005) QT dispersion and heart rate variability abnormalities in Alzheimer’s disease and in mild cognitive impairment. J Am Geriatr Soc 53:2135–2139
Hughes JW, Watkins L, Blumenthal JA, Kuhn C, Sherwood A (2004) Depression and anxiety symptoms are related to increased 24-hour urinary norepinephrine excretion among healthy middle-aged women. J Psychosom Res 57:353–358
Otte C, Neylan T, Pipkin S, Browner W, Whooley M (2005) Depressive symptoms and 24-hour urinary norepinephrine excretion levels in patients with coronary disease: findings from the Heart and Soul Study. The American journal of psychiatry
Lee J, Ju G, Park H, Chung S, Son J-W, Shin C-J et al (2022) Hippocampal Subfields and White Matter Connectivity in Patients with Subclinical Geriatric Depression. Brain Sci 12:329
Szymkowicz SM, McLaren ME, O’Shea A, Woods AJ, Anton SD, Dotson VM (2017) Depressive Symptoms Modify Age Effects on Hippocampal Subfields in Older Adults. Geriatr Gerontol Int 17:1494–1500
Choi WH, Jung WS, Um YH, Lee CU, Park YH, Lim HK (2017) Cerebral vascular burden on hippocampal subfields in first-onset drug-naïve subjects with late-onset depression. J Affect Disord 208:47–53
Ballmaier M, Narr KL, Toga AW, Elderkin-Thompson V, Thompson PM, Hamilton L et al (2008) Hippocampal Morphology and Distinguishing Late-Onset From Early-Onset Elderly Depression. Am J Psychiatry 165:229–237
Lim HK, Hong SC, Jung WS, Ahn KJ, Won WY, Hahn C et al (2012) Automated hippocampal subfields segmentation in late life depression. J Affect Disord 143:253–256
Huang Y, Coupland NJ, Lebel RM, Carter R, Seres P, Wilman AH et al (2013) Structural changes in hippocampal subfields in major depressive disorder: a high-field magnetic resonance imaging study. Biol Psychiatry 74:62–68
Travis S, Coupland NJ, Silversone PH, Huang Y, Fujiwara E, Carter R et al (2015) Dentate gyrus volume and memory performance in major depressive disorder. J Affect Disord 172:159–164
Roddy DW, Farrell C, Doolin K, Roman E, Tozzi L, Frodl T et al (2019) The Hippocampus in Depression: More Than the Sum of Its Parts? Advanced Hippocampal Substructure Segmentation in Depression. Biol Psychiatry 85:487–497
Han K-M, Kim A, Kang W, Kang Y, Kang J, Won E et al (2019) Hippocampal subfield volumes in major depressive disorder and bipolar disorder. Eur Psychiatry 57:70–77
La Joie R, Perrotin A, de La Sayette V, Egret S, Doeuvre L, Belliard S et al (2013) Hippocampal subfield volumetry in mild cognitive impairment, Alzheimer’s disease and semantic dementia. Neuroimage Clin 3:155–162
de Flores R, La Joie R, Chételat G (2015) Structural imaging of hippocampal subfields in healthy aging and Alzheimer’s disease. Neuroscience 309:29–50
Hellhammer J, Fries E, Schweisthal OW, Schlotz W, Stone AA, Hagemann D (2007) Several daily measurements are necessary to reliably assess the cortisol rise after awakening: state- and trait components. Psychoneuroendocrinology 32:80–86
El-Farhan N, Rees DA, Evans C (2017) Measuring cortisol in serum, urine and saliva – are our assays good enough? Ann Clin Biochem 54:308–322
Chan S, Debono M (2010) Replication of cortisol circadian rhythm: new advances in hydrocortisone replacement therapy. Ther Adv Endocrinol Metab 1:129–138
Althubaiti A (2016) Information bias in health research: definition, pitfalls, and adjustment methods. JMDH 9:211–217

The authors declare potential competing interests as follows: Dr Touron reported grant from French Ministry of Higher Education and Research (PhD grant). Dr Kuhn reported grant from the French Ministry of Higher Education and Research (PhD grant), and support from INSERM (PhD). Dr Marchant, Dr Chételat and Dr Poisnel reported grants from European Union’s Horizon 2020 Research and Innovation Program under grant agreement No 667696 during the conduct of the study. Dr Marchant reported grants from a Senior Fellowship from the Alzheimer’s Society (AS-SF-15b-002). Dr Chételat reported grants, personal fees and nonfinancial support from Institut National de la Santé et de la Recherche Médicale (INSERM); personal fees from Fondation Entrepreneurs MMA, grants and personal fees from Fondation Alzheimer, grants from Région Normandie, grants from Fondation Recherche Alzheimer, grants from Association France Alzheimer, outside the submitted work. No other disclosures were reported.

SubDepSEpinFCHPCHTHAlzResTherSUPPrevisedclean2final.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Depressive symptoms in older adults are associated with changes in stress-related markers, functional connectivity and brain volume

Status:

Version 1

Abstract

Figures

Background

Methods

Participants

Assessment of subclinical depressive symptoms and classification of participants

Characterization of the sample

Blood stress markers

Plasma epinephrine and norepinephrine quantification

Serum cortisol quantification

Neuroimaging procedure

MRI data acquisition

Structural MRI data

ROI-based method – Hypothalamus, whole hippocampus and hippocampal subfields

Hypothalamus

Whole hippocampus and hippocampal subfields

ROI-based gray matter volume extraction

Additional steps for voxelwise analysis and hippocampal 3D representation

Resting-state functional MRI data

Preprocessing of resting-state functional MRI

Statistical analyses

Covariates

Results

Participants’ characteristics

Between-group differences in blood stress markers

Between-group differences in hippocampal-hypothalamic functional connectivity

Intra-group associations between epinephrine and functional connectivity

Between-group differences in hypothalamic and hippocampal gray matter volumes

ROI-based method

VBM-based method – Hippocampal 3D representation

Intra-group associations between epinephrine and hypothalamic and hippocampal gray matter volumes

Discussion

Subclinical depressive symptoms are associated with higher SAM axis activity

Strengths and limitations

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1