PreprintPDF Available

Disaggregating between-and within-person relations for depression and burnout symptoms with cumulative hair cortisol and endocannabinoid concentrations: A multilevel analysis of four annual assessment waves

Authors:
Andreas Walther at University of Zurich
Andreas Walther
Clemens Kirschbaum at Technische Universität Dresden
Clemens Kirschbaum
Susanne Wehrli at University of Zurich
Susanne Wehrli
Nicole Rothe at Technische Universität Dresden
Nicole Rothe
Show all 7 authorsHide
Preprints and early-stage research may not have been peer reviewed yet.
Download file PDF
Read file
Download citation

Abstract and Figures

Background: The endocannabinoid system (ECS) is increasingly being recognized as key regulatory system coupled with the glucocorticoid system implicated in the pathophysiology of stress-related mental disorders. However, prior studies examining the ECS in depression or burnout have been inconclusive, of small sample size or of cross-sectional nature limiting interpretations of causal inferences or time-dependent effects. Methods: In a prospective community based cohort study including 128 individuals (women: 108), depression (PHQ-9) and burnout symptoms (MBI-GS) as well as hair cortisol and endocannabinoids were measured annually over four years. Cortisol, arachidonylethanolamide (AEA) and 2-arachidonyl-sn-glycerol/1-arachidonyl-sn-glycerol (2-AG/1-AG) were extracted from 3 cm hair segments reflecting cumulative concentrations of the last 3 months prior hair sampling. Results: Cross sectional group comparisons at baseline revealed reduced cortisol and AEA levels in the group with a positive major depressive disorder screening compared to individuals with low depression symptomatology (both p < .05). Cortisol was also reduced in the group with a positive burnout screening compared to individuals with a negative screening at baseline (p < .05). Longitudinal multilevel analysis, showed for the total sample that a within-person increase in burnout symptoms was associated with a decrease in cortisol levels over time (p < .05). In the male subsample, a between-person increase in AEA levels across time was associated with a decrease in depression symptoms and vice versa. However, a within-person increase of AEA levels across time was associated with an increase in depression symptoms and vice versa (both p < .05). Further, a within-person increase in burnout symptoms lead to a significant increase in AEA levels in the male subsample (p < .05). Conclusions: While cross-sectional analyses suggest higher depression or burnout symptomatology to be associated with reduced cortisol and AEA release, longitudinal analyses disaggregating between- and within-person effects reveal a complex pattern. A within-person increase in burnout symptoms precedes a decrease in cortisol secretion suggesting an exhaustion of the HPA axis. Between-person analysis show only for men a negative association between depression symptoms and AEA levels over time. However, within-person effects, indicate an increase in depression symptoms to preced and to follow an increase in AEA levels, suggesting a dynamic counterregulatory mechanism between the EC and depression in men differing on the between- and within-person levels. These longitudinal associations further elucidate the time and sex dependent relation between depression, burnout, glucocorticoid and endocannabinoid secretion.
Figures - uploaded by Andreas Walther
Author content
All figure content in this area was uploaded by Andreas Walther
Content may be subject to copyright.
Content uploaded by Andreas Walther
Author content
All content in this area was uploaded by Andreas Walther on Aug 31, 2021
Content may be subject to copyright.
Disaggregating between- and within-person relations for depression and
burnout symptoms with cumulative hair cortisol and endocannabinoid
concentrations: A multilevel analysis of four annual assessment waves
Walther, A.1,2, Kirschbaum, C.2, Wehrli, S.1,2, Rothe, N.2, Wekenborg, M.2, Penz, M.3, Gao,
W.2,†
1Clinical Psychology and Psychotherapy, University of Zurich, Zurich, Switzerland
2Biopsychology, TU Dresden, Dresden, Germany
3Psychology, University of Linz, Linz, Austria
†Corresponding author
Article type: Original Research
Running title: Longitudinal cortisol, AEA, depression and burnout relations
Key words: depression; burnout, endocannabinoids; cortisol, longitudinal; multilevel analysis
Word count (Text): 5757
Word count (Abstract): 388
Tables: 7
Figures: 5
Correspondence to: Dr. Wei Gao, Zellescher Weg 19, 1069, Dresden; Biopsychology, TU
Dresden, Dresden Germany
Wei.gao@tu-dresden.de
Longitudinal cortisol, AEA, depression and burnout relations
2
Abstract
Background: The endocannabinoid system (ECS) is increasingly being recognized as key
regulatory system coupled with the glucocorticoid system implicated in the pathophysiology
of stress-related mental disorders. However, prior studies examining the ECS in depression or
burnout have been inconclusive, of small sample size or of cross-sectional nature limiting
interpretations of causal inferences or time-dependent effects.
Methods: In a prospective community based cohort study including 128 individuals (women:
108), depression (PHQ-9) and burnout symptoms (MBI-GS) as well as hair cortisol and
endocannabinoids were measured annually over four years. Cortisol,
arachidonylethanolamide (AEA) and 2-arachidonyl-sn-glycerol/1-arachidonyl-sn-glycerol (2-
AG/1-AG) were extracted from 3 cm hair segments reflecting cumulative concentrations of
the last 3 months prior hair sampling.
Results: Cross sectional group comparisons at baseline revealed reduced cortisol and AEA
levels in the group with a positive major depressive disorder screening compared to
individuals with low depression symptomatology (both p < .05). Cortisol was also reduced in
the group with a positive burnout screening compared to individuals with a negative screening
at baseline (p < .05). Longitudinal multilevel analysis, showed for the total sample that a
within-person increase in burnout symptoms was associated with a decrease in cortisol levels
over time (p < .05). In the male subsample, a between-person increase in AEA levels across
time was associated with a decrease in depression symptoms and vice versa. However, a
within-person increase of AEA levels across time was associated with an increase in depression
symptoms and vice versa (both p < .05). Further, a within-person increase in burnout
symptoms lead to a significant increase in AEA levels in the male subsample (p < .05).
Conclusions: While cross-sectional analyses suggest higher depression or burnout
symptomatology to be associated with reduced cortisol and AEA release, longitudinal analyses
disaggregating between- and within-person effects reveal a complex pattern. A within-person
increase in burnout symptoms precedes a decrease in cortisol secretion suggesting an
exhaustion of the HPA axis. Between-person analysis show only for men a negative association
between depression symptoms and AEA levels over time. However, within-person effects,
indicate an increase in depression symptoms to preced and to follow an increase in AEA levels,
suggesting a dynamic counterregulatory mechanism between the EC and depression in men
differing on the between- and within-person levels. These longitudinal associations further
Longitudinal cortisol, AEA, depression and burnout relations
3
elucidate the time and sex dependent relation between depression, burnout, glucocorticoid
and endocannabinoid secretion.
Introduction
Depressive disorders are considered stress-related mental disorders characterized by its core
symptoms of depressive mood and anhedonia (Smoller, 2015). Similarly, also the burnout
syndrome is considered a stress-related psychological state identified by emotional
exhaustion, reduced professional efficacy, and cynicism caused by an excess of work-related
stress (Schaufeli et al., 2009). Although, the underlying pathophysiology of depressive
disorders and burnout are insufficiently understood, a dysregulation of the hypothalamus-
pituitary-adrenal (HPA) axis has been identified in both conditions in many independent
investigations (Otte et al., 2016; Rothe et al., 2020).
However, the endocannabinoid system (ECS) is increasingly being recognized for its
stress-responsive and -regulatory properties functionally interacting with the HPA axis
(Balsevich et al., 2017). That is, glucocorticoids such as cortisol – the main effector of the HPA
axis – recruit endocannabinoid signaling and thereby the ECS modulates many of the
neurobiological effects of glucocorticoids (Balsevich et al., 2017). The HPA axis together with
the ECS are therefore regarded as two interlocking key regulatory systems implicated in the
pathophysiology of depressive disorders (Hillard, 2017) and burnout (Gao et al., 2020). This
perspective is substantiated by the fact that both, glucocorticoid receptors (GRs) as well as
cannabinoid receptors (CBR) are expressed in all of the main brain structures involved in
depressive disorders and burnout such as the hypothalamus, hippocampus, habenula, cortex,
limbic system and amygdala (Biselli et al., 2019; Gururajan et al., 2019).
Although it seemed widely accepted that major depressive disorder (MDD) is
associated with increased HPA activity and increased basal cortisol concentrations (Otte et al.,
2016; Stetler and Miller, 2011), this is no longer unambiguous given reduced cortisol stress
reactivity in MDD (Miller and Kirschbaum, 2019; Rothe et al., 2020) and the integration of new
specimen such as hair cortisol showing mixed associations (Behnke et al., 2021; Gerritsen et
al., 2019; Rothe et al., 2021, 2020; Stalder et al., 2017a; Steudte-schmiedgen et al., 2017).
With regard to the ECS, on the one hand, rodent studies using cannabinoid type 1 receptor
(CB1R) knock-out models, pharmacological blockade or stimulation indicate increased
signaling of the CB1R in the central nervous system to exert antidepressant and anxiolytic
Longitudinal cortisol, AEA, depression and burnout relations
4
effects (Aso et al., 2008; Bambico et al., 2007; Gobbi et al., 2005; Haller et al., 2001; Kathuria
et al., 2003), while on the other hand, there is a report of impaired CB1R signaling by
pharmacological blockade being associated with antidepressant effects but increased HPA
activation (Steiner et al., 2008). By contrast, chronic unpredictable stress significantly
increasing HPA activity was shown to reduce the endocannabinoid N-
arachidonylethanolamine (anandamide [AEA]) in the central nervous system (Hill et al., 2008),
while concomitantly decreasing CB1R binding site density in the hippocampus and
hypothalamus, but increasing it in the prefrontal cortex. Examining acute and chronic
corticosterone administration further substantiated that the CB1R is under negative regulation
by glucocorticoids (Hill et al., 2008a). However, examining specific brain regions such as the
lateral habenula, increased levels of 2-arachidonoylglycerol (2-AG) are reported after chronic
unpredictable or social defeat stress, while site specific CB1R blockade reduced anxiety
behavior and increased basal corticosterone (Berger et al., 2018). Compelling evidence for an
antidepressive effect of CB1R activation in the amygdalaCCK-nucleus accumbensD2 circuit was
recently provided (Shen et al., 2019), suggesting functionality of the CB1R and the availability
of activating compounds such as AEA and 2-AG to be crucial for preventing the establishment
of depressive disorders.
Human studies including patients suffering from major depressive disorder (Hill et al.,
2009; Hill et al., 2008b), or post-traumatic stress disorder (PTSD) show on the one hand
reduced serum AEA and 2-AG levels (Hill et al., 2013; Neumeister et al., 2013), but on the other
hand no associations or increased levels of AEA and 2-AG are also reported for these disorders
or related conditions (Giuffrida et al., 2004; Harfmann et al., 2020; Hauer et al., 2013; Schaefer
et al., 2014). However, experimental research showed that under stress, the enzyme fatty acid
amide hydrolase (FAAH), which downregulates AEA and to a lesser extent 2-AG, is activated
in the amygdala (Hermanson et al., 2013). Inhibition of FAAH and thereby an increase in AEA
emerges as protective treatment against the anxiogenic effects of stress in healthy individuals
(Mayo et al., 2020). Another study showed that chronic stress (520 days of isolation and
confinement) reduced 2-AG, but not AEA levels in healthy men (Yi et al., 2016). A recent study
further highlights that inflammation-induced depression is not associated with circulating AEA
or 2-AG levels suggesting alterations in endocannabinoid levels in depression to be unrelated
to inflammatory-dependent depression states (Zajkowska et al., 2020).
Longitudinal cortisol, AEA, depression and burnout relations
5
While serum endocannabinoid levels or levels extracted from the central nervous
system vary with respect to the acute situation or time of the day, concentrations obtained
from hair – suggesting to be long-term integrated concentrations over several months – are
not subject to many of these confounding variables. A study revealed reduced hair AEA, but
not hair 2-AG, levels in patients with borderline personality disorder compared to healthy
controls, while a negative correlation between depressive symptoms and hair AEA levels was
observed (Wingenfeld et al., 2018). Another study further reported negative correlations
between hair endocannabinoid levels of N-acyl-ethanolamides palmitoylethanolamide (PEA),
oleoylethanolamide (OEA), stearoylethanolamide (SEA) and PTSD symptom severity (Wilker
et al., 2016). However, another report examining a sample of male refugee minors in a cross-
sectional design did not identify consistent associations between psychological
symptomatology and hair endocannbinoids except for a positive correlation between 2-AG/1-
AG and depressive symptoms (Croissant et al., 2020). Related to this line of research is a study
suggesting higher hair 1-AG and lower SEA levels in women with childhood maltreatment
(Koenig et al., 2018). In a recent analysis, we further replicated significant negative cross-
sectional relations between hair AEA levels and anxiety and burnout, and on trend-level with
depression symptoms in a sample of 207 individuals of the general population (Gao et al.,
2020). However, a recent analysis comparing 21 depressed women with 27 nondepressed
controls identified no differences between groups for hair endocannabinoids (Behnke et al.,
2021).
Taken together, based on rodent studies and cross-sectional human studies a state of
endocannabinoid deficiency is suggested to represent a stress-susceptible endophenotype
predisposing to depression or other stress-related mental conditions (Giacobbe et al., 2021;
Gorzalka and Hill, 2011; Hill et al., 2018). However, although there is compelling evidence that
endocannabinoid levels and CB1 signaling are reduced in depression or other stress-related
mental conditions, longitudinal human studies dissecting between- and within-person effects
by examining alterations in long-term integrated endocannabinoid and glucocorticoid levels
in parallel with regard to changes in depression or burnout symptoms are lacking.
Methods and Materials
Sample characteristics and study procedure
Longitudinal cortisol, AEA, depression and burnout relations
6
Participants were members of the Dresden Burnout Study, a prospective cohort study, which
tracks changes in psychological symptoms and corresponding changes in different biological
systems related to mental health approved by the local ethics committee and conducted in
accordance with the Declaration of Helsinki (Penz et al., 2018a). Recruitment of participants
was achieved via a local population registry sending invitation letters of the study to
inhabitants of the city of Dresden meeting age inclusion criteria selected by random sampling.
Furthermore, public media platforms, newspaper announcements and radio broadcasts
advertised the study and invited to participation.
Participants had to register for the study via the Dresden Burnout Study - Online
Platform and were subsequently selected and invited for biomarker testing when living in the
nearer area of Dresden. One week prior to their examination participants were invited via
email to complete an online psychometric test battery including depression and anxiety
measures. At the day of the biological examination, participants arrived overnight fasting and
responded to a standardized set of questions providing information on the current
constitution. Subsequently, several biological measures were obtained including blood serum
and hair sampling as described in more detail in the official study protocol (Penz et al., 2018a).
Should the participants not have completed the online questionnaire battery by the date of
examination, they would complete the questionnaire after the biological examination on site.
For the purpose of this longitudinal examination, only participants who participated in at least
three out of four examinations of the annual biomarker survey during the period between
2015 and 2018 were included.
Psychometric evaluation by self-report questionnaires
Depressive symptoms were measured using the Patient Health Questionnaire-9 (PHQ-9;
Kroenke et al., 2001). The PHQ-9 comprises 9 items reflecting the DSM-IV criteria for a major
depressive disorder (e.g. feeling down, depressed, or hopeless). Assessing how often the
respondent has been bothered by specific symptoms over the past two weeks using a four-
point Likert scale of 0 to 3 points (0 - not at all, 3 - nearly every day), the PHQ-9 ranges between
0 and 27. Large validation studies examining PHQ-9 cut-off scores (e.g. ≥10) compared to
structured clinical interviews confirm good detection performance of depressive disorders
indicating individuals with scores of 10 or higher to have high depressive burden and to be
Longitudinal cortisol, AEA, depression and burnout relations
7
consistently identified as having MDD by clinical diagnostic interviews (Levis et al., 2020). A
Cronbach’s alpha of .88 emerged for this sample suggesting a good internal consistency.
Burnout symptomatology was measured using the Maslach Burnout Inventory-
General Survey (MBI-GS; Büssing and Glaser, 1999; Schutte et al., 2000), which is considered
the gold standard for assessing burnout by self-report. The MBI-GS comprises of 16 items
rated on a seven-point Likert scale (0 - never, 6 - daily) examining symptoms experienced in
relation to one’s work. Thre subscales representing the burnout dimensions emotional
exhaustion, professional inefficacy, and cynicism can be derived, while the MBI-GS weighted
sum score ranges between 0 and 6. As cumulative cut off value indicating a clinically relevant
form of burnout symptomatology, a summed value of 3.5 or higher was suggested (Kalimo et
al., 2003). A Cronbach’s alpha of .92 emerged for the total score for this sample suggesting a
good internal consistency.
Endocannabinoid and glucocorticoid quantification
For the quantification of endocannabinoids and glucocorticoids, three hair strands (each with
a minimum of 20 mg of hair) were cut as close as possible to the scalp from the posterior
vertex. Hair samples were processed and analyzed as described in our published protocols for
glucocorticoids (Gao et al., 2013) and endocannabinoids (Gao et al., 2020). Sample pre-
processing and subsequent biochemical analysis were conducted at the Dresden LabService
GmbH (Tatzberg 47, 01307, Dresden, Germany). Samples were cut into 3cm segments from
the scalp-near site representing the integrated hormone concentration over the last three
months (e.g. average hair growth rate 1cm per month; Wennig 2000) and weighted into 7.5
mg of whole, nonpulverized hair and subsequently washed with 2.5 ml isopropanol following
the protocol by Gao et al. (2020, 2013). Biochemical analysis was conducted using liquid
chromatography coupled with tandem mass spectrometry (LC-MS/MS) described in detail
elsewhere (Gao et al., 2016). To avoid batch effects, all samples of the 128 participants were
quantified in a single run after the study was completed end of 2018. The period between hair
sampling and biochemical analysis varies therefore for the samples between 0 to 4 years,
depending on the study visit of the participants and the obtainment of the sample. This
variable was further included as covariate in the analysis. However, storage duration effects
have so far not been reported, while batch effects pose a well-recognized issue in biochemical
research across the globe. Recently, strong test-retest associaitons of hair endocannabinoid
Longitudinal cortisol, AEA, depression and burnout relations
8
levels were reported for the used method with intraclass correlation coefficients between
0.80 and 0.92 were reported (Gao et al., 2021). Intra- and inter-assay coefficient of variation
for endocannabinoids are below 14% suggesting acceptable quantification performance (CV <
15 %) (Caruso et al., 2008), and for cortisol and cortisone are below 8.8 % and 8.5 %,
respectively. The lower limit of detection for endocannabinoids is below (or equal to) 0.3
pg/mg−1 for AEA, 2-AG/1-AG, and 6 pg/mg−1 for PEA, OEA, SEA, while for cortisol and
cortisone it is 0.09 pg/mg and 0.07 pg/mg.
Statistical analysis
Cortisol and endocannabinoid data was log-transformed in order to approach a normal
distribution. Level of significance was set at p < .05. For calaculation of means, standard
deviations (SD) and cross-sectional comparisions, the first timepoint was used. Two-tailed t-
tests were conducted in order to perform cross-sectional comparisons between men and
women with regard to general demographics, endocannabinoids and clinical parameters
(depression and burnout).
Two-tailed t-tests were also used in order to conduct cross-sectional comparisions with
regard to severity of psychological disorders, cortisol and endocannabinoids. Based on the
PHQ-9 screening measure a cut off of ≥10 was used in order to create a major depression
syndrome subsample (Levis et al., 2020). For the MBI-GS scale a cut-off of ≥3.5 was used in
order to generate a subsample consisting of individuals suffering from burnout (Kalimo et al.,
2003). Due to the small portion of men in the present sample, we did not conduct subsample
analyses based on sex with regard clinical parameters.
Based on the nested structure of the present data, multilevel modelling was used
(Raudenbush and Bryk, 2002). The data had a two-level structure, meaning that years were
nested within individuals. Multilevel regression analyses were used to examine whether
cortisol and endocannbinoid levels predicted clinical parameters and vice versa. The models
were estimated with the lme4 package (Bates et al., 2015) in RStudio (version 3.3.3, (R Core
Team, 2017)). Models were fitted with maximum likelihood estimation. Model comparisions
were calculated with the anova function of the car package (Fox et al., 2007) and revealed
that a random-intercept model provided the best fit for the present dataset. 500 simulations
were used in order to compute parametric bootstrap intervals with the confint.merMod
Longitudinal cortisol, AEA, depression and burnout relations
9
function from the lme4 package (Bates et al., 2015). The lme4 package does not provide p-
values due to estimation of degrees of freedom and denominator degrees due to
disagreements in the literature (Bates et al., 2015). The Satterrthwaite’s degree of freedom
approach (Satterthwaite, 1941) and the Kenward-Roger approximation (Kenward and Roger,
1997) represent methods which produce the most acceptable Type I error rate, even for small
sample size (Luke, 2017). We applied the Satterthwaite approximation as it can be applied to
models fitted with a maximum likelihood approach, whereas the Kenward-Rogers
approximation only applies to REML models (Luke, 2017).
All predictors were person-mean centered to capture within-person variations of
endocannbinoids and clinical parameters. Additionally, to control for between-person effects,
endocannabinoids, depression and burnout were grand-mean centered and included into the
respective models (Hoffman and Stawski, 2009). Furthermore, in our analyses we controlled
for age, sex, BMI, medication intake, and alcohol consumption (Best et al., 2021; Stalder et al.,
2017a, 2010). We conducted subsample analyses for both sexes.
Results
Descriptive Statistics
Descriptive statistics separated by sex are summarized in Table 1. The total sample consisted
of 128 participants, with a mean age of 44.96 years, 84.38% of the overall sample was female.
In total 30 participants (23.4%) were identified as suffering from a major depressive syndrome
based on the cut-off point of the PHQ-9 ≥ 10, while 19 participants (14.8%) reached the cut-
off point for a burnout syndrome of the MBI ≥ 3.5. A mean BMI of 25.8 was observed and
15.6% self-identified as regular tobacco smokers and 19.5% reported to drink alcohol once in
a week or more often. A total of 32.8% reported to suffer from a chronic disease and the same
percentage reported to use medication due to their condition, while 14.8% reported to use
multiple medications. We used unpaired two-sided t-tests and t (χ2) tests to assess sex
differences. Men and women did not significantly differ with regard to age, BMI, chronic
disease status, alcohol consumption, smoking status, medication intake, mental disorder
status, burnout, or major depressive syndrome. As shown in the Supplementary Figures 1 and
2, there were no clear sex differences in depression symptoms and burnout symtpoms,
cortisol, AEA, 2-AG/1-AG, OEA, PEA and SEA levels. Baseline distribution of the parameters
separated by sex are provided in Supplementary Figures 3 (for pg/mg values) and 4 (for log
Longitudinal cortisol, AEA, depression and burnout relations
10
values). Intraclass correlation coefficients (ICCs) for depressive symptoms from T1 to T4 were
.76 for the overall sample, .72 for women and .87 for men. ICCs concerning burnout symptoms
were .66 for the overall sample, .62 for women and .85 for men, respectively. ICCs concerning
cortisol and endocannabinoids levels were .25 - .39 for the overall sample, .20 - .42 for women
and .24 - .53 for men.
Cross-sectional baseline comparisons of endocannbinoid- and cortisol levels by
clinical features
As presented in Figure 1, significant differences emerged between the major depressive
syndrome group (based on the PHQ-9 screening measure cut off ≥10) and participants with
low depressive symptoms concerning cortisol (t = -2.554, p < .05) and AEA (t = -2.053, p < .05).
Individuals in the major depressive syndrome group exhibited significantly lower cortisol (M =
.550, SD = .339) and AEA levels (M = -.483, SD = .287) whereas the group with low depressive
symptomatology had significantly higher cortisol (M = .804, SD = .370) and AEA levels (M = -
.225, SD = .452). No differences were found for 2-AG/1-AG, OEA, PEA and SEA (all p > .05).
As shown in Figure 2, individuals with burnout (based on the MBI-GS screening
measure cut off ≥3.5) significantly differed from invidivuals with low burnout symptoms
concerning cortisol levels (t = -3.160, p < .01). Participants with burnout had significantly lower
cortisol levels (M = .458, SD = .289) compared to individuals who had low burnout symptoms
(M = .798, SD = .367). No significant differences were found between the two groups with
respect to AEA, 2-AG/1-AG, OEA, PEA and SEA (all p > .05). Figures 3 and 4 present longitudinal
change patterns for the six different analytes separated by syndrome group at baseline
indicating no consistent differences between the groups across time (data not shown).
Therefore, multilevel regression analysis disaggregating between- and within person
associaitons shall be described in more detail in the following.
Multilevel regression
Depressive symptoms (PHQ-9), endocannabinoides and cortisol
In the male subample, a within-person increase of AEA significantly predicted an increase in
symptoms of depression across time (b = 27.504, p < .05, 95% CI [10.324; 44.460]) (see table
2). A within-person increase of PHQ-9 scores did significantly predict an increase in AEA levels
(b = .061, p < .001, 95% CI [.020; .103]) (see table 3). The direction of the slope in question was
reversed when looking at between person differences of the male subsample, an increase in
Longitudinal cortisol, AEA, depression and burnout relations
11
AEA significantly predicted a decrease in PHQ-9 levels (b = -24.389, p < .05, 95% CI [-41.138; -
7.150]). This was also the case in the male subsample for the relation with PHQ-9 levels as
predictor and AEA as outcome showing that a between-person increase in PHQ-9 levels led to
a significant decrease in AEA across the four timepoints (b = -.020, p < .05, 95% CI [-.035; -
.005]). No other significant result was found in regard to the within-person and between-
person relation between PHQ-9 scores and AEA levels in the overall sample or the female
subsample (see table 2 and 3).
Either a between-person nor a within-person increase in cortisol, 2-AG/1-AG, OEA, PEA
or SEA did signficiantly predict depression symptomathology and depression
symptomathology did not predict cortisol, 2-AG/1-AG, OEA, PEA or SEA levels in the overall-
and the two subsamples (all p > .05).
Burnout symptoms (MBI), endocannabinoides and cortisol
In the male subsample, a within person increase in MBI scores, lead to a significant increase
in AEA levels (b = .212, p < .05, 95% [CI .025; .429]) over time (see table 5), however, a within
person increase in AEA levels did not significantly predict MBI scores (see table 4). This relation
seemed reversed when considering between-person differences for the direction of the
association. However, an increase in MBI scores did not significantly predict a decrease of AEA
levels (b = -.056, p = .283, 95% CI [-.158; .043]). No other significant findings were found
regarding the within-person and between-person relation between MBI scores and AEA levels
in the overall sample or the female subsample (see table 4 and 5).
In the overall sample a within-person increase in MBI scores lead to a significant
decrease in cortisol levels (b = -.094, p < .05, 95% CI [-.184; -.006]) across time (see table 7),
while a within person increase in cortisol did not predict MBI scores across time (see table 6).
The aforementioned findings exhibited the same direction of effects when looking at
between-person differences, but not identifying a significant effect for an increase in MBI
scores predicting a decrease in cortisol levels (b = -.001, p = .981, 95% CI [-.048; .049]). No
significant association was found regarding the within-person and between-person relation
between MBI scores and cortisol levels in the female or the male subsample alone (see table
6 and 7).
Longitudinal cortisol, AEA, depression and burnout relations
12
No between-person or a within-person increases in 2-AG/1-AG, OEA, PEA or SEA did
significantly predict MBI scores and MBI scores did not predict 2-AG/1-AG, OEA, PEA or SEA
levels in the overall- and the male or female subsamples (all p > .05).
Exploratory Analyses
Cortisol did not significantly moderate the between and within person relation between AEA,
2-AG/1-AG , OEA, PEA or SEA with regard to PHQ-9 scores and vice versa. AEA, 2-AG/1-
AG, OEA, PEA and SEA did not moderate the relation between the predictor cortisol and the
criterion PHQ-9 and vice versa.
Cortisol significantly moderated the within person relation between MBI as predictor
and 2-AG/1-AG as criterion in the overall sample (b = .184, p < .05, 95% CI [.015; .346]).
Additionally, cortisol significantly moderated the relation between MBI as predictor and 2-
AG/1-AG as criterion in the male subample (b = -8.594, p < .05, 95% CI [-16.141; -.583]) but
not in the female subsample (b = .171, p = .065, 95% CI [-.006; .334]). Cortisol did not
significantly moderate the within person relation between 2-AG/1-AG as predictor and MBI as
criterion in the overall sample (b = .041, p = .968, 95% CI [-2.055; . 2.006]) or the female (b =
.036, p = .976, 95% CI [-2.408; 2.359]). However, cortisol did significantly moderate the within
person relation between 2-AG/1-AG as predictor and MBI as criterion in the male subsample
(b = -8.594, p < .05, 95% CI [-8.104; 2.302]).
2-AG/1-AG significantly moderated the relation between the predictor MBI and the
criterion cortisol in the overall sample (b = .850, p < .01, 95% CI [.226; 1.487]) and the female
subsample (b = .952, p < .01, 95% CI [.260; 1.614]) but not in the male subsample
(b = .026, p = .984, 95% CI [-2.537; 2.440]). This was not the case the other way around,
meaning that when cortisol was the predictor and MBI was the outcome variable the relation
was not moderated by 2-AG/1-AG in the overall sample (b = 1.194, p = .865, 95% CI [-1.670;
3.852]) and the female subsample (b = 2.139, p = .158, 95% CI [-.772; 5.177]). However, the
aforementioned relation was significant for the male subample (b = -8.277 p < .05, 95% CI [-
16.401; -.327]).
Cortisol did not significantly moderate the within person relation between AEA, OEA,
PEA and SEA with regard to MBI scores and vice versa. AEA, OEA, PEA and SEA did not
moderate the relation between the predictor cortisol and the criterion MBI and vice versa.
Longitudinal cortisol, AEA, depression and burnout relations
13
Cortisol did not significantly moderate the between person relation between AEA, 2-
AG/1-AG , OEA, PEA and SEA with regard to MBI scores and vice versa. AEA, 2-AG/1-AG, OEA,
PEA and SEA did not moderate the relation between the predictor cortisol and the criterion
MBI and vice versa.
Discussion
Summary
The present study is worldwide the first to examine cross-sectional and longitudinal relations
on the between- and within-person level for depression and burnout symptomatology and
hair cortisol and endocannabinoids. Cross-sectional comparisons reveal for depression and
burnout groups reduced cortisol levels, while only for the depression group also reduced AEA
levels were observed. Longitudinal multilevel analyses suggest that a within-person increase
in burnout, but not depression, symptoms preceds a decrease in cortisol levels in the overall
sample. In the female subsample no longitudinal associatons were identified. However, in the
male subsample, on the between-person level an increase in depression symptoms preceded
and followed a decrease in AEA levels, while on the within-person level an increase in
depression symptoms preceded and followed an increase in AEA levels. In men, also a within-
person increase in burnout symptomatology preceded an increase in AEA levels. Finally,
moderation analyses revealed that cortisol moderates the relation between burnout
symptoms and 2-AG/1-AG over time and that 2-AG/1-AG moderates the relation between
burnout symptoms and cortisol over time.
Integration of findings
The finding of reduced hair cortisol levels in the depression group as compared to the low
symptom group adds to the emerging picture of generally lower hair cortisol levels in
depressive disorders (Rothe et al., 2020). For example, in a study of 28 participants (4 males)
the MDD group exhibited significantly lower hair cortisol levels than controls (Steudte-
schmiedgen et al., 2017). Similarly, in two only female samples, reduced hair cortisol levels
were identified in the MDD cases as compared to healthy controls (Behnke et al., 2021;
Pochigaeva et al., 2017). However, there ist conflicting literature showing increased hair
cortisol levels in depressed individuals compared to controls (Dettenborn et al., 2012; Wei et
al., 2015) or no differences (Gerritsen et al., 2019). A meta-analysis on hair cortisol did not
identify any association between hair cortisol and mood disorders or depressive
Longitudinal cortisol, AEA, depression and burnout relations
14
symptomatology (Stalder et al., 2017b). Therefore, our cross-sectional data add to the mixed
findings on hair cortisol and depression suggesting either additional factors to influence this
association or time-dependent effects of the disease course in regard to its pathophysiology.
The fact that the subjects studied were most likely at very different points in the disease
course at the measurement time points may have introduced too much variance in terms of
the highly adaptive glucocoritcoid secretion. This is also reflected by the fact, that there
emerged no significant association in longitudinal analyses between depression symptoms
and cortisol levels.
In addition, a positive burnout screening at baseline was also associated with reduced
cortisol levels compared to controls, and this finding was corroborated in the longitudinal
analysis revealing a within-person increase in burnout symptoms to be associated with a
decrease in cortisol levels across time over the entire sample. This now confirms previous
findings on the within-person level, while previously a prospective association between
increased work-related stress measured by the effort-reward imbalance scale and reduced
hair cortisol levels at two-year follow-up on the group level was highlighted (Penz et al., 2019).
Notably, two previous reports using single time point measures of hair cortisol suggested
burnout to be associated with higher hair cortisol levels (Penz et al., 2018b; Wendsche et al.,
2020), highlighting the importance of repeated measures analyses to map the dynamics
between changes in burnout symptoms and changes in hair cortisol. It is further relevant to
highlight that although within-person associations corroborated the initially identified cross-
sectional negative association between burnout symptoms and hair cortisol levels on the
within-person level, this was not identified in the between-person analysis. Large differences
in the intercepts between persons may have obscured between-person associations with
regard to the centered grand mean, whereas differences becomes more visible in the person
mean centered differences (Curran and Bauer, 2011).
Based on the recognition of the functional interaction between the HPA axis and the
ECS, endocannabinoids such as AEA and 2-AG/1-AG have been discussed on the one hand as
potential biomarkers and on the other hand as factors that in concert with the HPA axis system
may cause depression or stress-related syndromes such as burnout or as a consequence of
the disorder be dysregulated (Behnke et al., 2021; Gao et al., 2020; Giacobbe et al., 2021; Hill
et al., 2018). However, except for the present study, it has never been investigated wether
the between- and within-person changes in depression or burnout symptoms are related to
Longitudinal cortisol, AEA, depression and burnout relations
15
to endocannabinoid and cortisol levels using a baseline assessment and three annual follow-
up examinations. Corroborating the at baseline observed difference in AEA levels with
reduced levels in the group with a positive depression screening as compared to the group
with low depression symptoms, only in the male subsample this pattern was confirmed in the
between-person examination. We found that in men, in general an increase in depression
symptoms led to a decrease in AEA levels over time and vice versa. This findings are in line
with the majority of previous reports showing lower AEA levels to be associated with
depression or stress-related mental disorders (Hill et al., 2013, 2009; Hill et al., 2008b;
Neumeister et al., 2013). This further suggests that in men, AEA levels may be reduced as a
consequence of a depressive disorder and that a pharmacological supplementation of AEA or
FAAH inhibition may achieve symptom alleviation as previously reported for social anxiety
disorders or in principle for PTSD (Mayo et al., 2020; Schmidt et al., 2021). Yet, with respect
to women, confounding factors such as sex hormone fluctuations might obscure this relation
(Balsevich et al., 2017; Fiacco et al., 2019). Introducing more complexity with regard to the
relation between the ECS and depression, recent findings suggest that a history of trauma
influences basal endocannabinoid levels and CB1R availablility depending on the
developmental stage (Nia et al., 2019).
Focusing on hair endocannabinoids, it can be said that small sample sizes and the cross-
sectional nature of previous studies limit the interpretability of existing reports suggesting a
negative trend between AEA levels and depressive symptoms in 31 women (Wingenfeld et al.,
2018). Similarly, in a sample of 207 individuals of the general population, a significant negative
cross-sectional correlation between hair AEA levels and anxiety and burnout symptoms was
identified, while also a negative trend with depression symptoms emerged (Gao et al., 2020).
Furthermore, a negative correlation between PEA, OEA, SEA and PTSD symptom severity in 76
rebel war survivors was reported (Wilker et al., 2016), while Croissant et al. (2020) report for
93 male refugee minors a positive correlation between 2-AG/1-AG with depression symptoms,
and Behnke et al. (2021) identify no association between hair endocannabinoids and
depression status in women. Thus, the present results overall support and extend pre-clinical
and clinical findings on AEA from blood serum and hair (Behnke et al., 2021; Gao et al., 2020;
Hill et al., 2008; Hill et al., 2008b; Shen et al., 2019), and highlight hair AEA as potential
biomarker for depression and burnout (Gao et al., 2020).
Longitudinal cortisol, AEA, depression and burnout relations
16
Therefore, it was unexpected to the authors that in the male subsample, within-person
increases in depression symptoms significantly predicted within-person increases in AEA
levels over time and vice versa, while the between-person relation was negative (see Figure
5). On the one side, this could indicate a compensatory mechanism raising AEA levels during
acute periods of increased depression symptomatology, which is then followed by a general
reduction of AEA secretion over time being catched by the conducted multilevel analysis
(Curran and Bauer, 2011). More specifically, the positive within-person effect reflects that, on
average, men tend to exhibit higher AEA levels in periods when their depression
symptomatology is elevated relative to their typical stable level (Berger et al., 2018); this might
be attributable to the energy-providing effect of increased AEA recuruitment and CB1R
signaling when depressed men need to surmount current challenges (Balsevich et al., 2017).
In contrast, the negative between-person effect reflects that, on average, men who are more
depressed tend to exhibit lower AEA levels over time. This might be attributable to an
exhaustion of the ECS after depressive periods of excess need of energy as similarly suggested
for the HPA axis (Rothe et al., 2020; Steudte-Schmiedgen et al., 2017). These results therefore
reflect a time-sensitive pattern in which the ECS system supports countermechanims against
depression establishment. However, if the increase in depression symptoms is not stopped,
the ECS may eventually exhaust and release blunted AEA levels as is typical in MDD. The
finding of a within-person effect in the male subsample revealing increased burnout
symptoms leading to increased AEA levels over time, lends further support for the hypothesis
of an identified counterregulatory mechanism of the ECS. To further test this hypothesis,
however, more measurements with shorter intervals, a longer study period and an
aggregation of the stress period would be needed to catch this point of AEA level reversal as
a consequence of ongoing symptom burden.
The fact that in the female subsample no significant associations were identified points
to additional confounding variables such as sex hormone levels or age potentially interacting
with the ECS (Fiacco et al., 2018; Walther et al., 2019; Walther et al., 2019; Walther et al.,
2020). However, conducting exploratory analyses controlling for birth control measures did
not reveal further associations. Another route of explanation might be the low cross-sectional
variance in depression and burnout symptomatology in the female subsample as well as less
change over time reducing the probability to identify statistical significance. This highlights
the need for larger samples to identify between- and within-person associaitons for
Longitudinal cortisol, AEA, depression and burnout relations
17
depression and burnout symptoms in relation to hair glucocorticoids and endocannabinoids
in women.
Finally, our results suggest that with regard to burnout symptoms, hair 2-AG/1-AG and
cortisol are significant moderators of each other, which also corroborated the previous finding
that changes in 2-AG levels following stress seem to be mediated by glucocorticoids (Nia et
al., 2019). However, systematic interaction effects between these two parameters should be
further examined in future research.
Limitations
Imitations of the present study must be considered when interpreting the results. First, no
clinical diagnostic interview or medical examination was performed at baseline. This leaves
the possibility of large heterogeneity in the sample with respect to underlying diagnoses. Thus,
although standardized screening instruments for depression or burnout were used, other
underlying conditions causing the depression or burnout symptoms cannot be excluded.
Possible somatic diseases could also have an impact on cortisol or endocannabinoid levels.
Furthermore, even though the Dresden Burnout Study comprises a large overall sample, the
number of subjects who participated in at least three of the four measurement time points
between 2015 and 2018 is relatively small for multilevel analyses with 128 subjects. Although
there is worldwide no other sample like this, longitudinal subgroup analyses are only
moderately powered with sample sizes ranging from 20 to 108 subjects. This also limits the
ability to test a broad set of covariates simultaneously. Finally, it is important to note the small
number of male participants, although it is in this subgroup that consistent longitudinal
associations between AEA level and depressive symptoms were identified.
Conclusion
Depressive disorders and burnout are complex conditions requiring in-depth analysis of time-
dependent symptom and biomarker changes. The present study represents a first attempt to
map cross-sectional and longitudinal between- and within-person relations between
depression and burnout symptoms and hair cortisol and endocannabinoid levels. For the
overall sample a within-person increase in burnout symptoms predicted a decrease in cortisol
levels over time highlighting the HPA axis as a stress-physiological system that appears to
enter exhaustion with reduced glucocorticoid secretion after a one-year stress period. Only
Longitudinal cortisol, AEA, depression and burnout relations
18
after this stressful period is terminated recovery of HPA axis activity seems possible. With
regard to endocannabinoids, reduced AEA levels seem to precede and follow increased
depression symptoms in men, while during acute depression phases increased AEA levels
seem to be present indicative of a counterregulatory mechanisms. This further highlights AEA
supplementation and FAAH inhibition as promising adjunct treatment especially in men with
depression symptoms, while more research is needed in female samples examining potential
confounding factors.
Funding
This work was supported by an intramural research grant issued by the TU Dresden (Support
the Best programme, part of the TUD Institutional Strategy, Excellence Initiative, awarded to
CK). The research was additionally supported by a research grant from the Deutsche
Forschungsgemeinschaft awarded to WG (GA 2772/2-1)
Declaration of interest
All authors declare no conflict of interest.
Author contributions
Conceptualization: CK, WG, AW, MP, MW; Recruitment and testing: NR, MP, MW, AW;
Biochemical analyses: WG; Data analysis and interpretation: AW, SW, WG; Writing of first
draft: AW, SW, WG; Editing of first draft: CK, AW, WG, NR, MP, MW.
Acknowledgements
First, we would like to thank all participants for study participation. Second, we would like to
thank the entire Dresden Burnout Study team for the support in conducting the study.
Longitudinal cortisol, AEA, depression and burnout relations
19
References
Aso, E., Ozaita, A., Valdizán, E.M., Ledent, C., Pazos, Á., Maldonado, R., Valverde, O., 2008.
BDNF impairment in the hippocampus is related to enhanced despair behavior in CB1
knockout mice. J. Neurochem. 105, 565–572.
Balsevich, G., Petrie, G.N., Hill, M.N., 2017. Frontiers in Neuroendocrinology
Endocannabinoids : E ff ectors of glucocorticoid signaling 47, 86–108.
doi:10.1016/j.yfrne.2017.07.005
Bambico, F.R., Katz, N., Debonnel, G., Gobbi, G., 2007. Cannabinoids Elicit Antidepressant-
Like Behavior and Activate Serotonergic Neurons through the Medial Prefrontal Cortex
27, 11700–11711. doi:10.1523/JNEUROSCI.1636-07.2007
Bates, D., Maechler, M., Bolker, B., 2015. Walker., S. Fitting linear mixed-effects models
using lme4. J Stat Softw 67, 1–48.
Behnke, A., Gumpp, A.M., Krumbholz, A., Bach, A.M., Schelling, G., Kolassa, I.-T., Rojas, R.,
2021. Hair-based biomarkers in women with major depressive disorder:
Glucocorticoids, endocannabinoids, N-acylethanolamines, and testosterone. Compr.
Psychoneuroendocrinology 7, 100068. doi:10.1016/j.cpnec.2021.100068
Berger, A.L., Henricks, A.M., Lugo, J.M., Wright, H.R., Warrick, C.R., Sticht, M.A., Morena, M.,
Bonilla, I., Laredo, S.A., Craft, R.M., Parsons, L.H., Grandes, P.R., Hillard, C.J., Hill, M.N.,
Mclaughlin, R.J., 2018. Archival Report The Lateral Habenula Directs Coping Styles
Under Conditions of Stress via Recruitment of the Endocannabinoid System. Biol.
Psychiatry 84, 611–623. doi:10.1016/j.biopsych.2018.04.018
Best, L.M., Wardell, J.D., Tyndale, R.F., McPhee, M.D., Le Foll, B., Kish, S.J., Boileau, I.,
Hendershot, C.S., 2021. Association of the Fatty Acid Amide Hydrolase C385A
Polymorphism With Alcohol Use Severity and Coping Motives in Heavy-Drinking Youth.
Alcohol. Clin. Exp. Res. 45, 507–517. doi:10.1111/acer.14552
Biselli, T., Lange, S., Sablottny, L., Steffen, J., Walther, A., 2019. Optogenetic and
chemogenetic insights into the neurocircuitry of depression- like behaviour: A
systematic review. Eur. J. Neurosci. accepted, 327–331.
Büssing, A., Glaser, J., 1999. Deutsche Fassung des Maslach burnout inventory–General
survey (MBI-GS-D). Munich München, Tech. Univ. Lehrstuhl für Psychol.
Caruso, D., Scurati, S., Maschi, O., De Angelis, L., Roglio, I., Giatti, S., Garcia-Segura, L.M.,
Melcangi, R.C., 2008. Evaluation of neuroactive steroid levels by liquid
chromatography–tandem mass spectrometry in central and peripheral nervous system:
effect of diabetes. Neurochem. Int. 52, 560–568.
Croissant, M., Glaesmer, H., Klucken, T., Kirschbaum, C., Gao, W., Stalder, T., Sierau, S., 2020.
Endocannabinoid concentrations in hair and mental health of unaccompanied refugee
minors. Psychoneuroendocrinology 104683.
Curran, P.J., Bauer, D.J., 2011. The disaggregation of within-person and between-person
effects in longitudinal models of change. Annu. Rev. Psychol. 62, 583–619.
doi:10.1146/annurev.psych.093008.100356
Dettenborn, L., Muhtz, C., Skoluda, N., Stalder, T., Steudte, S., Hinkelmann, K.I.M.,
Kirschbaum, C., Otte, C., 2012. Introducing a novel method to assess cumulative steroid
concentrations : Increased hair cortisol concentrations over 6 months in medicated
patients with depression. Stress 15, 348–353. doi:10.3109/10253890.2011.619239
Fiacco, S., Walther, A., Ehlert, U., 2019. Steroid secretion in healthy aging.
Psychoneuroendocrinology 105, 64–78. doi:10.1016/j.psyneuen.2018.09.035
Fiacco, S., Walther, A., Ehlert, U., 2018. Steroid secretion in healthy aging.
Psychoneuroendocrinology. doi:10.1016/j.psyneuen.2018.09.035
Longitudinal cortisol, AEA, depression and burnout relations
20
Fox, J., Friendly, G.G., Graves, S., Heiberger, R., Monette, G., Nilsson, H., Ripley, B., Weisberg,
S., Fox, M.J., Suggests, M., 2007. The car package. R Found. Stat. Comput.
Gao, W., Kirschbaum, C., Grass, J., Stalder, T., 2016. LC – MS based analysis of endogenous
steroid hormones in human hair. J. Steroid Biochem. Mol. Biol. 162, 92–99.
doi:10.1016/j.jsbmb.2015.12.022
Gao, W., Schmidt, K., Enge, S., Kirschbaum, C., 2021. Intra-individual stability of hair
endocannabinoid and N-acylethanolamine concentrations. Psychoneuroendocrinology
105395. doi:10.1016/j.psyneuen.2021.105395
Gao, W., Stalder, T., Foley, P., Rauh, M., Deng, H., Kirschbaum, C., 2013. Quantitative analysis
of steroid hormones in human hair using a column-switching LC – APCI – MS / MS assay.
J. Chromatogr. B 928, 1–8. doi:10.1016/j.jchromb.2013.03.008
Gao, W., Walther, A., Wekenborg, M., Penz, M., Kirschbaum, C., 2020. Determination of
endocannabinoids and N-acylethanolamines in human hair with LC-MS/MS and their
relation to symptoms of depression, burnout, and anxiety. Talanta 217, 121006.
doi:10.1016/j.talanta.2020.121006
Gerritsen, L., Staufenbiel, S.M., Penninx, B.W.J.H., van Hemert, A.M., Noppe, G., de Rijke,
Y.B., van Rossum, E.F.C., 2019. Long-term glucocorticoid levels measured in hair in
patients with depressive and anxiety disorders. Psychoneuroendocrinology 101, 246–
252.
Giacobbe, J., Marrocu, A., Di Benedetto, M.G., Pariante, C.M., Borsini, A., 2021. A systematic,
integrative review of the effects of the endocannabinoid system on inflammation and
neurogenesis in animal models of affective disorders. Brain. Behav. Immun. 93, 353–
367. doi:10.1016/j.bbi.2020.12.024
Giuffrida, A., Leweke, F.M., Gerth, C.W., Schreiber, D., Koethe, D., Faulhaber, J.,
Klosterkötter, J., Piomelli, D., 2004. Cerebrospinal anandamide levels are elevated in
acute schizophrenia and are inversely correlated with psychotic symptoms.
Neuropsychopharmacology 29, 2108.
Gobbi, G., Bambico, F.R., Mangieri, R., Bortolato, M., Campolongo, P., Solinas, M., Cassano,
T., Morgese, M.G., Debonnel, G., Duranti, A., 2005. Antidepressant-like activity and
modulation of brain monoaminergic transmission by blockade of anandamide
hydrolysis. Proc. Natl. Acad. Sci. 102, 18620–18625.
Gorzalka, B.B., Hill, M.N., 2011. Putative role of endocannabinoid signaling in the etiology of
depression and actions of antidepressants. Prog. Neuropsychopharmacol. Biol.
Psychiatry 35, 1575–1585. doi:10.1016/j.pnpbp.2010.11.021
Gururajan, A., Reif, A., Cryan, J.F., Slattery, D.A., 2019. The future of rodent models in
depression research. Nat. Rev. Neurosci. 20, 686–701. doi:10.1038/s41583-019-0221-6
Haller, J., Varga, B., Ledent, C., Freund, T.F., 2001. CB1 cannabinoid receptors mediate
anxiolytic effects : convergent genetic and pharmacological evidence with CB1-specific
agents 299–304. doi:10.1097/01.fbp.0000135704.56422.40
Harfmann, E.J., McAuliffe, T.L., Larson, E.R., Claesges, S.A., Sauber, G., Hillard, C.J., Goveas,
J.S., 2020. Circulating endocannabinoid concentrations in grieving adults.
Psychoneuroendocrinology 120, 104801. doi:10.1016/j.psyneuen.2020.104801
Hauer, D., Schelling, G., Gola, H., Campolongo, P., Morath, J., Roozendaal, B., Hamuni, G.,
Karabatsiakis, A., Atsak, P., Vogeser, M., 2013. Plasma concentrations of
endocannabinoids and related primary fatty acid amides in patients with post-traumatic
stress disorder. PLoS One 8, e62741.
Hermanson, D.J., Hartley, N.D., Gamble-George, J., Brown, N., Shonesy, B.C., Kingsley, P.J.,
Colbran, R.J., Reese, J., Marnett, L.J., Patel, S., 2013. Substrate-selective COX-2
Longitudinal cortisol, AEA, depression and burnout relations
21
inhibition decreases anxiety via endocannabinoid activation. Nat. Neurosci. 16, 1291–
1298. doi:10.1038/nn.3480
Hill, M.N., Bierer, L.M., Makotkine, I., Golier, J.A., Galea, S., McEwen, B.S., Hillard, C.J.,
Yehuda, R., 2013. Reductions in circulating endocannabinoid levels in individuals with
post-traumatic stress disorder following exposure to the World Trade Center attacks.
Psychoneuroendocrinology 38, 2952–2961.
Hill, M.N., Campolongo, P., Yehuda, R., Patel, S., 2018. Integrating endocannabinoid signaling
and cannabinoids into the biology and treatment of posttraumatic stress disorder.
Neuropsychopharmacology 43, 80.
Hill, Matthew N, Carrier, E.J., Ho, W.V., Shi, L., Patel, S., Gorzalka, B.B., Hillard, C.J., 2008a.
Prolonged glucocorticoid treatment decreases cannabinoid CB1 receptor density in the
hippocampus. Hippocampus 18, 221–226.
Hill, M. N., Carrier, E.J., McLaughlin, R.J., Morrish, A.C., Meier, S.E., Hillard, C.J., Gorzalka,
B.B., 2008. Regional alterations in the endocannabinoid system in an animal model of
depression: effects of concurrent antidepressant treatment. J. Neurochem. 106, 2322–
2336. doi:10.1111/j.1471-4159.2008.05567.x
Hill, M.N., Miller, G.E., Carrier, E.J., Gorzalka, B.B., Hillard, C.J., 2009. Circulating
endocannabinoids and N-acyl ethanolamines are differentially regulated in major
depression and following exposure to social stress. Psychoneuroendocrinology 34,
1257–1262.
Hill, Matthew N, Miller, G.E., Ho, W.-S., Gorzalka, B.B., Hillard, C.J., 2008b. Serum
endocannabinoid content is altered in females with depressive disorders: a preliminary
report. Pharmacopsychiatry 41, 48–53.
Hillard, C.J., 2017. Circulating Endocannabinoids : From Whence Do They Come and Where
are They Going ? 43, 155–172. doi:10.1038/npp.2017.130
Hoffman, L., Stawski, R.S., 2009. Persons as contexts: Evaluating between-person and within-
person effects in longitudinal analysis. Res. Hum. Dev. 6, 97–120.
Kalimo, R., Pahkin, K., Mutanen, P., Topipinen-tanner, S., 2003. Staying well or burning out at
work : Work characteristics and personal resources as long- term predictors. Work
Stress 17, 109–122. doi:10.1080/0267837031000149919
Kathuria, S., Gaetani, S., Fegley, D., Valiño, F., Duranti, A., Tontini, A., Mor, M., Tarzia, G., La
Rana, G., Calignano, A., 2003. Modulation of anxiety through blockade of anandamide
hydrolysis. Nat. Med. 9, 76.
Kenward, M.G., Roger, J.H., 1997. Small sample inference for fixed effects from restricted
maximum likelihood. Biometrics 983–997.
Koenig, A.M., Gao, W., Umlauft, M., Schury, K., Reister, F., Kirschbaum, C., Karabatsiakis, A.,
Kolassa, I.T., 2018. Altered hair endocannabinoid levels in mothers with childhood
maltreatment and their newborns. Biol. Psychol. 135, 93–101.
doi:10.1016/j.biopsycho.2018.03.006
Kroenke, K., Spitzer, R.L., Williams, J.B.W., 2001. The phq-9. J. Gen. Intern. Med. 16, 606–
613.
Levis, B., Sun, Y., He, C., Wu, Y., Krishnan, A., Bhandari, P.M., Neupane, D., Imran, M.,
Brehaut, E., Negeri, Z., Fischer, F.H., Benedetti, A., Thombs, B.D., 2020. Accuracy of the
PHQ-2 Alone and in Combination with the PHQ-9 for Screening to Detect Major
Depression: Systematic Review and Meta-analysis. JAMA - J. Am. Med. Assoc. 323,
2290–2300. doi:10.1001/jama.2020.6504
Luke, S.G., 2017. Evaluating significance in linear mixed-effects models in R. Behav. Res.
Methods 49, 1494–1502.
Longitudinal cortisol, AEA, depression and burnout relations
22
Mayo, L.M., Asratian, A., Lindé, J., Morena, M., Haataja, R., Hammar, V., Augier, G., Hill,
M.N., Heilig, M., 2020. Elevated Anandamide, Enhanced Recall of Fear Extinction, and
Attenuated Stress Responses Following Inhibition of Fatty Acid Amide Hydrolase: A
Randomized, Controlled Experimental Medicine Trial. Biol. Psychiatry 87, 538–547.
doi:10.1016/j.biopsych.2019.07.034
Miller, R., Kirschbaum, C., 2019. Cultures under stress: A cross-national meta-analysis of
cortisol responses to the Trier Social Stress Test and their association with anxiety-
related value orientations and internalizing mental disorders.
Psychoneuroendocrinology 105, 147–154. doi:10.1016/j.psyneuen.2018.12.236
Neumeister, A., Normandin, M.D., Pietrzak, R.H., Piomelli, D., Zheng, M.-Q., Gujarro-Anton,
A., Potenza, M.N., Bailey, C.R., Lin, S.F., Najafzadeh, S., 2013. Elevated brain
cannabinoid CB 1 receptor availability in post-traumatic stress disorder: a positron
emission tomography study. Mol. Psychiatry 18, 1034.
Nia, A.B., Bender, R., Harpaz-Rotem, I., 2019. Endocannabinoid system alterations in
posttraumatic stress disorder: A review of developmental and accumulative effects of
trauma. Chronic Stress 3, 1–17. doi:10.1177/2470547019864096
Otte, C., Gold, S.M., Penninx, B.W., Pariante, C.M., Etkin, A., Fava, M., Mohr, D.C.,
Schatzberg, A.F., 2016. Major depressive disorder. Nat. Publ. Gr. 2, 1–21.
doi:10.1038/nrdp.2016.65
Penz, M., Kanthak, M.K., Pieper, L., Beesdo-baum, K., Walther, A., Miller, R., Stalder, T.,
Kirschbaum, C., 2018a. The Dresden Burnout Study: Study protocol of a prospective
cohort study for the bio-psychological investigation of burnout Marlene Penz. Int. J.
Methods Psychiatr. Res. 1–11. doi:10.1002/mpr.1613
Penz, M., Siegrist, J., Wekenborg, M.K., Rothe, N., Walther, A., Kirschbaum, C., 2019. Effort-
reward imbalance at work is associated with hair cortisol concentrations: Prospective
evidence from the Dresden Burnout Study. Psychoneuroendocrinology 109, 104399.
doi:10.1016/j.psyneuen.2019.104399
Penz, M., Stalder, T., Miller, R., Ludwig, V.M., Kanthak, M.K., Kirschbaum, C., 2018b. Hair
cortisol as a biological marker for burnout symptomatology. Psychoneuroendocrinology
87, 218–221.
Pochigaeva, K., Druzhkova, T., Yakovlev, A., Onufriev, M., Grishkina, M., Chepelev, A.,
Guekht, A., Gulyaeva, N., 2017. Hair cortisol as a marker of hypothalamic-pituitary-
adrenal Axis activity in female patients with major depressive disorder. Metab. Brain
Dis. 32, 577–583.
R Core Team, 2017. A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna.
Raudenbush, S.W., Bryk, A.S., 2002. Hierarchical linear models: Applications and data
analysis methods. sage.
Rothe, N., Ste, J., Penz, M., Kirschbaum, C., Walther, A., 2020. Examination of peripheral
basal and reactive cortisol levels in major depressive disorder and the burnout
syndrome : A systematic review. Neurosci. Biobehav. Rev. 113, 1–39.
doi:10.1016/j.neubiorev.2020.02.024
Rothe, N., Vogel, S., Schmelzer, K., Kirschbaum, C., Penz, M., Wekenborg, M.K., Gao, W.,
Walther, A., 2021. The moderating effect of cortisol and dehydroepiandrosterone on
the relation between sleep and depression or burnout. Compr.
Psychoneuroendocrinology 7, 100051. doi:10.1016/j.cpnec.2021.100051
Satterthwaite, F.E., 1941. Synthesis of variance. Psychometrika 6, 309–316.
Schaefer, C., Enning, F., Mueller, J.K., Bumb, J.M., Rohleder, C., Odorfer, T.M., Klosterkötter,
Longitudinal cortisol, AEA, depression and burnout relations
23
J., Hellmich, M., Koethe, D., Schmahl, C., 2014. Fatty acid ethanolamide levels are
altered in borderline personality and complex posttraumatic stress disorders. Eur. Arch.
Psychiatry Clin. Neurosci. 264, 459–463.
Schaufeli, W.B., Leiter, M.P., Maslach, C., 2009. Burnout : 35 years of research and practice.
Career Dev. Int. 14, 204–220. doi:10.1108/13620430910966406
Schmidt, M.E., Liebowitz, M.R., Stein, M.B., Grunfeld, J., Van Hove, I., Simmons, W.K., Van
Der Ark, P., Palmer, J.A., Saad, Z.S., Pemberton, D.J., Van Nueten, L., Drevets, W.C.,
2021. The effects of inhibition of fatty acid amide hydrolase (FAAH) by JNJ-42165279 in
social anxiety disorder: a double-blind, randomized, placebo-controlled proof-of-
concept study. Neuropsychopharmacology 46, 1004–1010. doi:10.1038/s41386-020-
00888-1
Schutte, N., Toppinen, S., Kalimo, R., Schaufeli, W.B., 2000. The factorial validity of the
Maslach Burnout Inventory-General Survey (MBI-GS) across occupational groups and
nations. J. Occup. Organ. Psychol. 73, 53–66. doi:10.1348/096317900166877
Shen, C.J., Zheng, D., Li, K.X., Yang, J.M., Pan, H.Q., Yu, X.D., Fu, J.Y., Zhu, Y., Sun, Q.X., Tang,
M.Y., Zhang, Y., Sun, P., Xie, Y., Duan, S., Hu, H., Li, X.M., 2019. Cannabinoid CB1
receptors in the amygdalar cholecystokinin glutamatergic afferents to nucleus
accumbens modulate depressive-like behavior. Nat. Med. 25, 337–349.
doi:10.1038/s41591-018-0299-9
Smoller, J.W., 2015. The Genetics of Stress-Related Disorders : PTSD , Depression , and
Anxiety Disorders 41, 297–319. doi:10.1038/npp.2015.266
Stalder, T., Kirschbaum, C., Heinze, K., Steudte, S., Foley, P., Tietze, A., Dettenborn, L., 2010.
Use of hair cortisol analysis to detect hypercortisolism during active drinking phases in
alcohol-dependent individuals. Biol. Psychol. 85, 357–360.
doi:10.1016/j.biopsycho.2010.08.005
Stalder, T., Steudte-schmiedgen, S., Alexander, N., Klucken, T., Vater, A., Wichmann, S.,
Kirschbaum, C., Miller, R., 2017a. Stress-related and basic determinants of hair cortisol
in humans: A meta-analysis 77, 261–274.
Stalder, T., Steudte-schmiedgen, S., Alexander, N., Klucken, T., Vater, A., Wichmann, S.,
Kirschbaum, C., Miller, R., 2017b. Stress-related and basic determinants of hair cortisol
in humans: A meta-analysis. Psychoneuroendocrinology 77, 261–274.
Steiner, M.A., Marsicano, G., Nestler, E.J., Holsboer, F., Lutz, B., Wotjak, C.T., 2008.
Antidepressant-like behavioral effects of impaired cannabinoid receptor type 1
signaling coincide with exaggerated corticosterone secretion in mice.
Psychoneuroendocrinology 33, 54–67. doi:10.1016/j.psyneuen.2007.09.008
Stetler, C., Miller, G.E., 2011. Depression and hypothalamic-pituitary-adrenal activation: a
quantitative summary of four decades of research. Psychosom. Med. 73, 114–126.
doi:10.1097/PSY.0b013e31820ad12b
Steudte-schmiedgen, S., Wichmann, S., Stalder, T., Hilbert, K., Muehlhan, M., Lueken, U.,
Beesdo-baum, K., 2017. Hair cortisol concentrations and cortisol stress reactivity in
generalized anxiety disorder , major depression and their comorbidity. J. Psychiatr. Res.
84, 184–190. doi:10.1016/j.jpsychires.2016.09.024
Steudte-Schmiedgen, S., Wichmann, S., Stalder, T., Hilbert, K., Muehlhan, M., Lueken, U.,
Beesdo-Baum, K., 2017. Hair cortisol concentrations and cortisol stress reactivity in
generalized anxiety disorder, major depression and their comorbidity. J. Psychiatr. Res.
84, 184–190. doi:10.1016/j.jpsychires.2016.09.024
Walther, A., Tsao, C., Pande, R., Kirschbaum, C., Field, E., Berkman, L., 2019. Do
dehydroepiandrosterone, progesterone, and testosterone influence women’s
Longitudinal cortisol, AEA, depression and burnout relations
24
depression and anxiety levels? Evidence from hair-based hormonal measures of 2105
rural Indian women. Psychoneuroendocrinology 109, 104382.
doi:10.1016/j.psyneuen.2019.104382
Walther, Andreas, Wasielewska, J.M., Leiter, O., 2019. The antidepressant effect of
testosterone: An effect of neuroplasticity? Neurol. Psychiatry Brain Res. 32, 104–110.
doi:10.1016/j.npbr.2019.05.004
Walther, A., Wehrli, S., Kische, H., Penz, M., Wekenborg, M., Gao, W., Rothe, N., Beesdo-
Baum, K., Kirschbaum, C., 2020. Depressive symptoms are not associated with long-
term integrated testosterone concentrations in hair. World J. Biol. Psychiatry 0, 1–13.
doi:10.1080/15622975.2020.1795253
Wei, J., Sun, G., Zhao, L., Yang, X., Liu, X., Lin, D., Li, T., Ma, X., 2015. Analysis of hair cortisol
level in first-episodic and recurrent female patients with depression compared to
healthy controls. J. Affect. Disord. 175, 299–302.
Wendsche, J., Ihle, A., Wegge, J., Penz, M.S., Kirschbaum, C., Kliegel, M., 2020. Prospective
associations between burnout symptomatology and hair cortisol. Int. Arch. Occup.
Environ. Health 93, 779–788. doi:10.1007/s00420-020-01528-3
Wennig, R., 2000. Potential problems with the interpretation of hair analysis results.
Forensic Sci. Int. 107, 5–12.
Wilker, S., Pfeiffer, A., Elbert, T., Ovuga, E., Karabatsiakis, A., Krumbholz, A., Thieme, D.,
Schelling, G., Kolassa, I., 2016. Psychoneuroendocrinology Endocannabinoid
concentrations in hair are associated with PTSD symptom severity.
Psychoneuroendocrinology 67, 198–206. doi:10.1016/j.psyneuen.2016.02.010
Wingenfeld, K., Dettenborn, L., Kirschbaum, C., Gao, W., Otte, C., Roepke, S., Wingenfeld, K.,
Dettenborn, L., Kirschbaum, C., Gao, W., 2018. Reduced levels of the endocannabinoid
arachidonylethanolamide ( AEA ) in hair in patients with borderline personality disorder
– a pilot study 3890. doi:10.1080/10253890.2018.1451837
Yi, B., Nichiporuk, I., Nicolas, M., Schneider, S., Feuerecker, M., Vassilieva, G., Thieme, D.,
Schelling, G., Choukèr, A., 2016. Reductions in circulating endocannabinoid 2-
arachidonoylglycerol levels in healthy human subjects exposed to chronic stressors.
Prog. Neuro-Psychopharmacology Biol. Psychiatry 67, 92–97.
doi:10.1016/j.pnpbp.2016.01.004
Zajkowska, Z., Borsini, A., Nikkheslat, N., Russell, A., Romano, G.F., Tomassi, S., Hepgul, N.,
Forton, D., Agarwal, K., Hotopf, M., 2020. Differential effect of Interferon-alpha
treatment on AEA and 2-AG levels. Brain. Behav. Immun.
Longitudinal cortisol, AEA, depression and burnout relations
25
Table 1. Sample characteristics (N=128)
Total sample
Women
Men
t (χ2)
p
n = 128
n = 108
n = 20
Age, M (SD)
44.96 (12.04)
44.40 (12.13)
48 (11.35)
BMI, M (SD)
25.79 (4.81)
25.47 (4.71)
27.55 (5.07)
Tobacco smokers, n (%)
20 (15.63)
18 (16.67)
2 (10)
0.584
.747
Weekly alcohol consumers, n
(%)
25 (19.531)
18 (16.67)
7 (35)
5.493
.905
Medication intake, n (%)
42 (32.81)
38 (35.19)
4 (20)
1.765
.184
Antidepressants intake, n (%)
11 (8.59)
9 (8.33)
2 (10)
.060
.807
Thyroid supplementation, n (%)
26 (20.31)
25 (23.15)
1 (5)
3.434
.064
Birth control, n (%)
17 (13.28)
17 (15.74)
0 (0)
Multiple medication, n (%)
19 (14.84)
16 (14.81)
3 (15)
0.000
.983
Chronic disease, n (%)
42 (32.81)
38 (35.19)
4 (20)
1.765
.184
Psychological disorder, n (%)
22 (17.19)
19 (17.59)
3 (15)
.080
.778
Depressive Symptoms (PHQ-9),
M (SD)
8.35 (5.31)
8.51 (4.89)
7.74 (6.83)
-.567
.572
Positive screening for current
major depressive syndrome
(MDD) PHQ-9 ≥ 10, n (%)
30 (23.44)
25 (23.15)
5 (25)
.389
.534
Burnout Symptoms (MBI-GS),
M (SD)
1.90 (1.16)
1.87 (1.17)
2.02 (1.09)
.530
.597
Positive screening for burnout
MBI-GS  ≥ 3.5, n (%)
19 (14.84)
18 (16.67)
1 (5)
1.817
.178
Cortisol (pg/mg), M (SD)
13.54 (46.63)
14.97 (51.15)
6.67 (6.92)
-.509
.613
Cortisol (log), M (SD)
.72 (.38)
.72 (.38)
.67 (.36)
-.391
.700
Anandamide (AEA; pg/mg), M
(SD)
.75 (.75)
.69 (0.70)
.98 (.97)
.964
.341
Anandamide (AEA; log), M (SD)
-.32 (.41)
-.34 (.40)
-.25 (.51)
.557
.580
Arachidonoyl glycerol (AG-1 &
AG-2; pg/mg), M (SD)
33.21 (13.49)
32.88 (13.97)
34.85 (11.31)
.417
.678
Arachidonoyl glycerol (AG-1 &
AG-2; log), M (SD)
1.48 (.19)
1.48 (.20)
1.52 (.13)
.725
.471
Oleoylethanolamide (OEA;
pg/mg), M (SD)
1629.28
(1447.88)
1611.72
(1412.86)
1717.07
(1692.48)
.208
.836
Oleoylethanolamide (OEA; log),
M (SD)
3.05 (.39)
3.05 (.38)
3.03 (.47)
-.143
.887
Longitudinal cortisol, AEA, depression and burnout relations
26
Palmitoylethanolamide (PEA;
pg/mg), M (SD)
1927.83
(1424.82)
1988.80
(1499.26)
1622.98
(972.95)
-.738
.463
Palmitoylethanolamide (PEA;
log), M (SD)
3.17 (.31)
3.18 (.32)
3.12 (.31)
-.565
.575
Stearoylethanolamine (SEA;
pg/mg), M (SD)
381.55 (303.02)
386.66 (320.42)
355.96
(205.15)
-.290
.773
Stearoylethanolamine (SEA;
log), M (SD)
2.47 (.31)
2.47 (.32)
2.48 (.28)
.078
.938
Note: N: sample size, M: mean, SD: standard deviation, BMI: body mass index; PHQ-9: Patient Health
Questionnaire 9, MBI-GS: Maslach Burnout Inventory-General Survey.
Hair-based biomarkers in women with major depressive disorder: Glucocorticoids, endocannabinoids, N-acylethanolamines, and testosterone
Article
Full-text available
  • Jun 2021
Background Stress-related alterations in the regulation of several endocrine systems, including the hypothalamus-pituitary-adrenal (HPA) and -gonadal (HPG) axes and the endocannabinoid system are proposed to be involved in the etiology of major depressive disorder (MDD). Therefore, this study examines whether altered concentrations of glucocorticoids, testosterone, endocannabinoids, and related N-acylethanolamines accumulated in hair are present in MDD. Methods Female participants (range: 19–59, Mdn = 30.5 years) were recruited, including n = 21 with a current MDD episode and n = 27 nondepressed controls without any current mental disorder. Weight-standardized samples of 3 cm hair segments were analyzed which equals to three months of retrospectively assessed biomarkers in hair. Concentrations of cortisol, cortisone, testosterone, endocannabinoids (i.e., anandamide [AEA], 2-arachidonylglycerol [2-AG]), and selected N-acylethanolamines (i.e., stearoylethanolamide [SEA], oleoylethanolamide [OEA], palmitoylethanolamide [PEA]) were measured using mass spectrometry. Results Female MDD patients exhibited lower cortisol and cortisone levels in hair than nondepressed controls, whereas the hair concentrations of endocannabinoids, N-acylethanolamines, and testosterone did not differ between the groups. Conclusions Our results add to the heterogeneous body of findings on alterations in hair-stored glucocorticoids and endocannabinoids in MDD. As in previous studies, there was no evidence of altered testosterone concentrations in the hair of MDD patients. Larger and longitudinal studies are needed to comprehensively explore the overall picture of endocrine alterations in MDD.
View
The moderating effect of cortisol and dehydroepiandrosterone on the relation between sleep and depression or burnout
Article
Full-text available
  • Mar 2021
For poor sleep quality (SQ) as well as major depressive disorder (MDD) and burnout, a dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis has been identified. Although poor SQ is often reported as an early symptom of MDD or burnout, it is not clear whether HPA axis-related hormones can influence the association between SQ and MDD or burnout. This manuscript addresses this question by examining HPA axis-related hormones as potential moderators influencing the association between SQ and MDD or burnout. In the fourth annual examination wave of the Dresden Burnout Study, we measured general SQ (including sleep duration and efficiency), depressive and burnout symptoms, and obtained hair samples for quantification of long-term integrated steroid concentrations (cortisol [hC], cortisone [hCn], dehydroepiandrosterone [hDHEA]) from 462 participants (67 % female). Data on SQ, depressive and burnout symptoms were available from 342 participants from the preceding examination wave (average time span between examinations 13.2 months). Cross-sectional analyses showed that the negative association between sleep duration and depressive symptoms was buffered by higher levels of hC, and hCn, whereas the negative association between sleep duration and burnout symptoms was buffered by higher levels of hDHEA. The negative association between sleep efficiency and burnout symptoms was intensified by higher levels of hC and hC/hCn ratio and the negative association between general SQ and burnout symptoms was intensified by higher levels of hC/hCn ratio. Longitudinal analyses revealed an intensifying effect of a higher hC/hCn ratio for the negative association between sleep duration and burnout symptoms. Our results suggest opposed moderation effects of hair glucocorticoids on the association between SQ and depressive or burnout symptoms. This points toward opposed glucocorticoid receptor functioning in depression and burnout. To fully elucidate the negative consequences of poor SQ on MDD and burnout, the complex underlying mechanisms of action including HPA axis-related hormones need to be investigated in MDD and burnout separately.
View
Association of the Fatty Acid Amide Hydrolase C385A Polymorphism With Alcohol Use Severity and Coping Motives in Heavy‐Drinking Youth
Article
Full-text available
  • Feb 2021
  • ALCOHOL CLIN EXP RES
Background Reduced function of fatty acid amide hydrolase, the catabolic enzyme for the endocannabinoid anandamide, can be inherited through a functional genetic polymorphism (FAAH rs324420, C385A, P129T). The minor (A) allele has been associated with reduced FAAH enzyme activity and increased risk for substance use disorders in adults. Whether this inherited difference in endocannabinoid metabolism relates to alcohol use disorder etiology and patterns of alcohol use in youth is unknown. Methods To examine this question, heavy‐drinking youth (n = 302; mean age = 19.74 ± 1.18) were genotyped for FAAH C385A. All subjects completed a comprehensive interview assessing alcohol use patterns including the Timeline Follow‐back Method, Alcohol Use Disorders Identification Test (AUDIT), and Drinking Motives Questionnaire. Analyses of Covariance (ANCOVAs) were conducted to assess differences in drinking patterns and drinking motives between genotype groups, and mediation analyses investigated whether drinking motives accounted for indirect associations of genotype with alcohol use severity. Results Youth with the FAAH minor allele (AC or AA genotype) reported significantly more drinking days (p = 0.045), significantly more frequent heavy episodic drinking (p = 0.003), and significantly higher alcohol‐related problems and consumption patterns (AUDIT score p = 0.045, AUDIT‐C score p = 0.02). Mediation analyses showed that the association of FAAH C385A with drinking outcomes was mediated by coping motives. Conclusions These findings extend previous studies by suggesting that reduced endocannabinoid metabolism may be related to heavier use of alcohol in youth, prior to the onset of chronic drinking problems. Furthermore, differences in negative reinforcement‐related drinking could account in part for this association.
View
A systematic, integrative review of the effects of the endocannabinoid system on inflammation and neurogenesis in animal models of affective disorders
Article
Full-text available
  • Dec 2020
The endocannabinoid (eCB) system is considered relevant in the pathophysiology of affective disorders, and a potential therapeutic target, as its hypoactivity is considered an important risk factor of depression. However, the biological mechanisms whereby the eCB system affects mood remain elusive. Through a systematic review, thirty-seven articles were obtained from the PubMed/Medline, Web of Science, Embase, PsychInfo, and CINAHL databases, investigating the role of the eCB system on the immune system and neurogenesis, as well as resulting behavioural effects in rodent models of affective disorders. Overall, activation of the eCB system appears to decrease depressive-like behaviour and to be anti-inflammatory, while promoting neuro- and synaptogenesis in various models. Activation of cannabinoid receptors (CBRs) is shown to be crucial in improving depressive-like and anxiety-like behaviour, although cannabidiol administration suggests a role of additional mechanisms. CB1R signalling, as well as fatty acid amide hydrolase (FAAH) inhibition, are associated with decreased pro-inflammatory cytokines. Moreover, activation of CBRs is required for neurogenesis, which is also upregulated by FAAH inhibitors. This review is the first to assess the association between the eCB system, immune system and neurogenesis, alongside behavioural outcomes, across rodent models of affective disorders. We confirm the therapeutic potential of eCB system activation in depression and anxiety, highlighting immunoregulation as an important mechanism whereby dysfunctional behaviour and neurogenesis can be improved.
View
The effects of inhibition of fatty acid amide hydrolase (FAAH) by JNJ-42165279 in social anxiety disorder: a double-blind, randomized, placebo-controlled proof-of-concept study
Article
Full-text available
  • Oct 2020
JNJ-42165279 is a selective inhibitor of fatty acid amide hydrolase (FAAH), the enzyme responsible for the degradation of fatty acid amides (FAA) including anandamide (AEA), palmitoylethanolamide (PEA), and N-oleoylethanolamide (OEA). We assessed the efficacy, safety, tolerability, pharmacokinetics, and pharmacodynamics of treatment with JNJ-42165279 in subjects with social anxiety disorder (SAD). This was a multicenter, double-blind, placebo-controlled study randomizing subjects to 12 weeks of treatment with either JNJ-42165279 (25 mg daily) or placebo (PBO). The primary endpoint was the change in the Liebowitz Social Anxiety Scale (LSAS) total score from baseline to end of study. Secondary endpoints included the Hamilton Anxiety Scale (HAM-A), Hamilton Depression Rating Scale (HDRS17), and the Clinical Global Impression-Improvement (CGI-I). Samples were collected for plasma concentration of AEA, PEA, OEA, and JNJ-42165279. A total of 149 subjects were enrolled with a mean baseline LSAS total score of 102.6 (SD 16.84). The mean change from baseline (SD) in LSAS total score at week 12 was numerically greater for JNJ-42165279: −29.4 (27.47) compared to PBO: −22.4 (23.57) but not significant. The percentage of subjects with ≥30% improvement from baseline in the LSAS total score was significantly higher for JNJ-42165279 (42.4%) compared to PBO (23.6%) (p value = 0.04). The percentage of subjects with a CGI-I score of much or very much improved was also significantly higher for JNJ-42165279 (44.1%) than for PBO (23.6%) (p value = 0.02). The drug was well tolerated. JNJ-42165279 appears to elicit an anxiolytic effect in subjects with SAD although trough concentrations with 25 mg once daily appeared to be insufficient to completely inhibit FAAH activity which may have led to suboptimal efficacy. ClinicalTrials.gov Identifier: NCT02432703.
View
Differential effect of Interferon-alpha treatment on AEA and 2-AG levels
Article
Full-text available
  • Aug 2020
The endocannabinoid (eCB) system is one of the key players in immunoregulation, and reduced activity of the eCB system has been linked with depressive-like behaviours in animal studies and depression in clinical samples. There is a well-established link between immune activation and depression, such as following the administration of the pro-inflammatory cytokine, interferon-α (IFN-α), used to treat hepatitis C viral (HCV) infection. However, the role of peripheral endocannabinoids (eCBs), anandamide (AEA) and 2-arachidonoylglycerol (2-AG), following immunotherapy with IFN-α and in IFN-α -induced depression, have not been examined yet. In this study, we investigated whether circulating AEA and 2-AG were modified by treatment with IFN-α and whether they were involved in the development of IFN-α-induced depression. We also explored whether circulating eCBs were associated with peripheral cytokines during and after IFN-α treatment. We measured serum concentrations of AEA and 2-AG using High Performance Liquid Chromatography with Tandem Mass Spectrometry, and serum concentrations of cytokines using Meso Scale Discovery electrochemiluminescence V-PLEX assay, in 70 patients with HCV infection and 41 healthy subjects. We assessed HCV patients at baseline, IFN-α-treatment weeks (TW) 4 and 24, end of treatment (END) and at six months follow-up (FU). We assessed depression using M.I.N.I. International Neuropsychiatric Interview. We found a different pattern of change in peripheral AEA and 2-AG during and after IFN-α treatment. Whilst 2-AG increased earlier in immunotherapy (TW4), remained elevated throughout treatment, and reduced at six months follow-up (FU), AEA increased later in treatment (TW24) and remained elevated six months post-treatment. We also found that baseline levels of AEA were lower in HCV patients compared with healthy controls, whereas there were no differences in 2-AG levels. Interestingly, AEA, but not 2-AG, was significantly, negatively correlated with interleukin (IL)-2 and IL-17a at six months follow-up. We did not find any difference in both eCBs between patients with and without IFN-α-induced depression, at any time point. Our findings suggest that AEA and 2-AG are involved in different stages of immunoregulation following IFN-α treatment, where AEA might be involved in chronic inflammation. Lack of association between peripheral eCBs and IFN-α-induced depression suggests that different biological mechanisms may underpin inflammation-induced depression compared with classic “psychiatric” depression, or that any changes in the eCB system in depression may not be captured by peripheral AEA and 2-AG.
View
Intra-individual stability of hair endocannabinoid and N-acylethanolamine concentrations
Article
  • Aug 2021
  • PSYCHONEUROENDOCRINO
Hair analysis of endocannabinoids and N-acylethanolamines presents a promising methodological advancement for the retrospective assessment of long-term cumulative endocannabinoids and N-acylethanolamines secretion over extended periods of time. A main assumption of this method application that hair endocannabinoid and N-acylethanolamine concentrations show intra-individual stability has not been confirmed yet. Thus, in the current study hair endocannabinoid and N-acylethanolamine levels were measured over a period of two and a half years with six months between each hair sample collection in 100 female participants. We found strong test-retest associations of hair endocannabinoid and N-acylethanolamine levels with intraclass correlation coefficients between 0.80 and 0.92. Furthermore, no correlations between perceived stress and hair endocannabinoids or N-acylethanolamines was observed. The current findings support the notion that endocannabinoids and N-acylethanolamines in hair are rather trait biomarkers that are stable over a considerable period of time rather than rapidly changing state markers.
View
Circulating endocannabinoid concentrations in grieving adults
Article
  • Jul 2020
  • PSYCHONEUROENDOCRINO
Bereavement is one of the most intense, distressing, and traumatic events an elderly person will experience. The symptom responses to bereavement vary, particularly during the first year. However, the neurobiology underlying the symptom variance in grief is poorly understood. The endocannabinoid signaling (ECS) system is stress-responsive; mounting evidence implicates abnormalities in the central ECS system could contribute to psychopathology. The current study aimed to investigate the hypothesis that the ECS is abnormal in grief, using circulating eCB concentrations as a biomarker of central ECS. A predominantly older sample of grief participants, within 13 months following the death of a loved one, and healthy comparison (HC) participants were studied. Associations of circulating eCBs with symptom variance in grievers were also examined. A total of 61 (grief: n = 44; HC: n = 17) adults completed cross-sectional clinical assessments and a fasting blood draw. Assessments included the Inventory of Complicated Grief scale; the 17-item Hamilton Depression Rating Scale; and the Hamilton Anxiety scale. Serum eCB concentrations (i.e., N-arachidonoylethanolamine [AEA] and 2-arachidonoylglycerol [2-AG]) were quantified using isotope dilution, liquid chromatography-mass spectrometry. Relative to HC participants, grievers had significantly elevated serum AEA but similar 2-AG concentrations. In grievers, serum AEA concentrations were positively associated with depressive and anxiety symptoms, but only in those with low grief symptoms. These novel findings indicate that elevated circulating eCB concentrations are found following bereavement. The eCB signaling response varies based on the degree of grief severity. Circulating eCB measures may have the potential to serve as biomarkers of prolonged grief disorder.
View
Depressive symptoms are not associated with long-term integrated testosterone concentrations in hair
Article
  • Jul 2020
  • WORLD J BIOL PSYCHIA
Objectives: The association between depressive symptomatology and endogenous testosterone levels is inconclusive. Large inter- and intra-individual testosterone differences suggest point measurements from saliva or serum to be inadequate to map basal testosterone concentrations highlighting the potential for long-term integrated testosterone levels from hair. Methods: Using data from a prospective cohort study, a total of 578 participants (74% female) provided complete data on depressive symptomatology, clinical features, and hair samples for quantification of testosterone concentrations at baseline. Available data of three annual follow-up examinations was used for longitudinal analyses. Results: Correlation analysis showed in both, men and women, hair testosterone across all four time points not to be significantly related to depressive symptoms. Examined clinical features were not associated with testosterone levels, except for having a current diagnosis of a psychological disorder, which was associated with reduced testosterone levels in men, but not in women. Acceptable model fit for an autoregressive cross-lagged panel analysis emerged only for the female subsample suggesting inverse cross-relations for the prediction of testosterone by depressive symptomatology and vice versa. Conclusions: Findings from this study add to the literature by showing no association between long-term integrated testosterone in hair and depressive symptomatology in men and women.
View
Accuracy of the PHQ-2 Alone and in Combination With the PHQ-9 for Screening to Detect Major Depression: Systematic Review and Meta-analysis
Article
  • Jun 2020
Importance The Patient Health Questionnaire depression module (PHQ-9) is a 9-item self-administered instrument used for detecting depression and assessing severity of depression. The Patient Health Questionnaire–2 (PHQ-2) consists of the first 2 items of the PHQ-9 (which assess the frequency of depressed mood and anhedonia) and can be used as a first step to identify patients for evaluation with the full PHQ-9. Objective To estimate PHQ-2 accuracy alone and combined with the PHQ-9 for detecting major depression. Data Sources MEDLINE, MEDLINE In-Process & Other Non-Indexed Citations, PsycINFO, and Web of Science (January 2000-May 2018). Study Selection Eligible data sets compared PHQ-2 scores with major depression diagnoses from a validated diagnostic interview. Data Extraction and Synthesis Individual participant data were synthesized with bivariate random-effects meta-analysis to estimate pooled sensitivity and specificity of the PHQ-2 alone among studies using semistructured, fully structured, or Mini International Neuropsychiatric Interview (MINI) diagnostic interviews separately and in combination with the PHQ-9 vs the PHQ-9 alone for studies that used semistructured interviews. The PHQ-2 score ranges from 0 to 6, and the PHQ-9 score ranges from 0 to 27. Results Individual participant data were obtained from 100 of 136 eligible studies (44 318 participants; 4572 with major depression [10%]; mean [SD] age, 49 [17] years; 59% female). Among studies that used semistructured interviews, PHQ-2 sensitivity and specificity (95% CI) were 0.91 (0.88-0.94) and 0.67 (0.64-0.71) for cutoff scores of 2 or greater and 0.72 (0.67-0.77) and 0.85 (0.83-0.87) for cutoff scores of 3 or greater. Sensitivity was significantly greater for semistructured vs fully structured interviews. Specificity was not significantly different across the types of interviews. The area under the receiver operating characteristic curve was 0.88 (0.86-0.89) for semistructured interviews, 0.82 (0.81-0.84) for fully structured interviews, and 0.87 (0.85-0.88) for the MINI. There were no significant subgroup differences. For semistructured interviews, sensitivity for PHQ-2 scores of 2 or greater followed by PHQ-9 scores of 10 or greater (0.82 [0.76-0.86]) was not significantly different than PHQ-9 scores of 10 or greater alone (0.86 [0.80-0.90]); specificity for the combination was significantly but minimally higher (0.87 [0.84-0.89] vs 0.85 [0.82-0.87]). The area under the curve was 0.90 (0.89-0.91). The combination was estimated to reduce the number of participants needing to complete the full PHQ-9 by 57% (56%-58%). Conclusions and Relevance In an individual participant data meta-analysis of studies that compared PHQ scores with major depression diagnoses, the combination of PHQ-2 (with cutoff ≥2) followed by PHQ-9 (with cutoff ≥10) had similar sensitivity but higher specificity compared with PHQ-9 cutoff scores of 10 or greater alone. Further research is needed to understand the clinical and research value of this combined approach to screening.
View