Plasma oxytocin concentrations are lower in depressed vs. healthy
control women and are independent of cortisol
Kaeli W. Yuena, Joseph P. Garnera,b, Dean S. Carsona, Jennifer Kellerc, Anna Lembkea,
Shellie A. Hydea, Heather A. Kennac, Lakshika Tennakoonc, Alan F. Schatzbergc,
Karen J. Parkera,*
aDepartment of Psychiatry and Behavioral Sciences, School of Medicine, Stanford University, 1201 Welch Rd. MSLS Room P104, Stanford,
CA 94305-5485, USA
bDepartment of Comparative Medicine, School of Medicine, Stanford University, 287 Campus Dr., Stanford, CA 94305-5410, USA
cDepartment of Psychiatry and Behavioral Sciences, School of Medicine, Stanford University, 401 Quarry Rd., Stanford, CA 94305-5719, USA
a r t i c l e i n f o
Received 6 August 2013
Received in revised form
5 December 2013
Accepted 17 December 2013
Psychotic major depression
a b s t r a c t
The neuropeptide oxytocin (OT) promotes social behavior and attenuates stress responsivity in mam-
mals. Recent clinical evidence suggests OT concentrations may be dysregulated in major depression. This
study extends previous research by testing whether: 1) OT concentrations vary systematically in
depressive disorders with and without hypercortisolemia, 2) gender differences in OT concentrations are
observed in depressed vs. healthy control participants, and 3) OT concentrations are predictive of clinical
phenotypes. Plasma OT concentrations of psychotic major depressive (PMD; n ¼ 14: 10 female, 4 male),
non-psychotic major depressive (NPMD; n ¼ 17: 12 female, 5 male), and non-depressed, healthy control
(n ¼ 19: 11 female, 8 male) participants were assayed at 2000, 2400, 0400, and 0800 h. Plasma cortisol
concentrations were quantified at 2300 h, and clinical phenotypes were determined. As expected, PMD
participants, compared to NPMD and healthy control participants, showed higher plasma cortisol con-
centrations. Although both depressed groups showed similar OT concentrations, a significant interaction
effect between group and gender was observed. Specifically, depressed females exhibited lower mean OT
concentrations than depressed males. Further, depressed vs. healthy control female participants
exhibited lower mean OT concentrations, whereas depressed vs. healthy control male participants
showed a trend in the opposite direction. OT concentrations were also predictive of desirability, drug
dependence, and compulsivity scores as measured by the Million Clinical Multiaxial Inventory-III. All
findings were independent of cortisol. These data suggest that OT signaling may provide a mechanism by
which to better understand female-biased risk to develop depressive disorders and that plasma OT
concentrations may be a useful biomarker of certain clinical phenotypes.
? 2013 Published by Elsevier Ltd.
Individuals with depressive disorders often exhibit social diffi-
culties, anxietysymptoms, and dysregulation of the hypothalamice
pituitaryeadrenal (HPA) axis (Heuser et al., 1994). A common
feature of HPA axis dysregulation is hypercortisolemia, which is
present in 40e60% of depressed patients (Gold et al., 1986; Parker
et al., 2003). Depressive disorders are also twice as prevalent in
female compared to male patients (Nestler et al., 2002; Young,
The pathophysiology of depressive disorders and enhanced fe-
male risk to develop them are not well understood. It has been
hypothesized that the neuropeptide oxytocin (OT) may play a role
in these phenomena (Frasch et al., 1995). OT is a neuropeptide of
hypothalamic origin with broad ranging central and peripheral
nervous system effects. OT is released during positive social in-
teractions (Knox and Uvnas-Moberg,1998; Young et al., 2008), and
exhibits anxiolytic properties in preclinical animal models (Amico
et al., 2004). OT also exerts a tonic inhibitory influence over the
HPA axis (Neumann et al., 2000). However, stressful social experi-
ences, which are frequent precipitants of depressive disorders,
activate the HPA axis. Often these experiences involve social
isolation or loss, which may decrease OT signaling, thereby
diminishing OT’s ability to inhibit HPA axis drive. Females may be
particularly vulnerable under these stressful circumstances, as they
* Corresponding author. Tel.: þ1 650 736 9863; fax: þ1 (650) 498 7761.
E-mail address: firstname.lastname@example.org (K.J. Parker).
Contents lists available at ScienceDirect
Journal of Psychiatric Research
journal homepage: www.elsevier.com/locate/psychires
0022-3956/$ e see front matter ? 2013 Published by Elsevier Ltd.
Journal of Psychiatric Research 51 (2014) 30e36
respond to social rejection (Stroud et al., 2002) and marital conflict
(Kiecolt-Glaser et al., 1996, 1998) with greater HPA axis responses
compared to men.
Preliminary research has shown that some depressed patients
exhibit lower OT concentrations compared to healthy controls and
that OT concentrations are lower in patients with moderate
compared to mild depression (Frasch et al.,1995; Garcia et al., 2011;
Ozsoy et al., 2009). Several pilot studies have concomitantly
measured cortisol and OT concentrations in major depressed pa-
tients (Garcia et al., 2011; Parker et al., 2010), but no research has
evaluated OT concentrations in patients with psychotic major
depression (PMD), who typically exhibit greater depressive
severity, and who are more likely to exhibit hypercortisolemia than
patients with non-psychotic major depression (NPMD) (Belanoff
et al., 2001).
In addition to the well-described roles of OT biology in human
behavioral functioning, there is emerging evidence that OT con-
centrations are also associated with an array of clinical phenotypes
and personality traits. For example, cerebrospinal fluid (CSF) OT
concentrations are associated with a historyof aggression (Lee et al.,
2009), childhood trauma (Heim et al., 2009), and suicide inten-
tionality (Jokinen et al., 2012). CSF OT concentrations are likewise
negatively correlated with anxiety concentrations in patients with
major depression (Scantamburlo et al., 2007) and plasma OT con-
centrations are lower in female patients with borderline personality
disorder compared to healthy female controls (Bertsch et al., 2013).
Mothers who used cocaine during pregnancy have lower plasma OT
concentrations and greater hostility and depressed mood, as well as
less social support and fewer adaptive strategies for dealing with
stressful life events compared to cocaine naïve mothers (Light et al.,
2004). In contrast, higher plasma OT concentrations are associated
with measures of positive personality traits including increased
extroversion in healthy male and female participants (Andari et al.,
2012) and secure attachment and lower compulsive temperament
in first-time mothers (Strathearn et al., 2009, 2012). Increased
plasma OT concentrations are also associated with secure attach-
ments and lower plasma cortisol and anxiety concentrations in
healthy premenopausal women (Tops et al., 2007).
The present study extends previous research efforts in several
respects. We measured plasma OT and cortisol concentrations in
both male and female PMD, NPMD, and healthycontrol participants
to test whether OT concentrations differ in participants with PMD
vs. NPMD, or in depressed participants with and without hyper-
cortisolism (regardless of depressive subtype). We also sought to
examine whether female and male depressed participants differed
in OT concentrations, and whether cortisol concentrations influ-
enced these differences. Finally, we explored whether OT concen-
trations were associated with depression severity and anxiety
symptomology (Hamilton Depression Rating Scale), as well as
personality and clinical syndromes (Million Clinical Multiaxial
Inventory-III) in depressed participants.
2. Materials and methods
2.1. Study participants
Adult participants were recruited through Stanford University
Medical Center, as well as through online and print advertisements,
as part of a larger research program, which investigates HPA axis
physiology in depressive disorders. The present study was initiated
after the completion of two recruitment rounds of a larger parent
study, and resulted in available samples from 50 participants: 14
PMD (10 female, 4 male), 17 NPMD (12 female, 5 male), and 19
healthy controls (11 female, 8 male). Participants enrolled in the
present study did not overlap with those in our previously
published OT and depression study (Parker et al., 2010). However,
participants’ cortisol data have been reported previously (Keller
et al., 2006; Schatzberg et al., 2013). This research was conducted
in accordance with the Declaration of Helsinki, and the Stanford
University Institutional Review Board approved the study design.
Informed written consent was obtained from all participants after
the study procedures had been fully explained.
All PMD and NPMD participants met DSM-IV-TR criteria for a
current majordepressive episode, with or without psychotic features
required to have a minimum score of 5 on the positive symptom
subscale of the Brief Psychiatric Rating Scale (BPRS) (Overall and
Gorham, 1962). This subscale consists of the following four items:
conceptual disorganization, suspiciousness, hallucinations, and un-
usual thought content. Healthy control participants were required to
have a score <6 on the HDRS, have no psychotic symptoms as
of Axis I disorders as determined by the Structured Clinical Interview
to remain on psychiatric medications (see Table 1) provided the dose
had not been adjusted in the last week. Healthy control participants
were free of psychiatric medications. Additional study exclusion
criteria for all participants were as follows: electroconvulsive shock
therapy or substance abuse problems in the last six months, current
drug use (evaluated by urinary toxicology screening), major medical
illness, history of seizures, major head trauma, abnormal clinical lab-
oratory tests, unstable or untreated hypertension, cardiovascular
disease, endocrine disorders, pregnancy, or lactation.
2.2. Blood collection procedures
Participants were admitted to the Stanford University Hospital
General Clinical Research Center at 1500 h. At 1600 h, an intrave-
nous line, kept patent with saline infusion, was started in one arm
and 4 mL of whole blood was drawn into heparinized vacutainer
tubes once every hour between 1800 and 0900 h. Participants were
required to be supine in bed 15 min prior to each blood sample
collection. No food or sleep restrictions were imposed. Blood
samples were subsequently centrifuged and the plasma fraction
was stored at ?70?C prior to hormone quantification. Blood ali-
quots were available for OT analyses at four time points (2000,
2400, 0400, and 0800 h).
2.3. Plasma hormone quantification
Plasma OTconcentrations were quantified using a commercially
available enzyme immunoassay kit (Enzo Life Sciences, Inc.,
Concomitant psychiatric medications for participants with non-psychotic major
depression (NPMD) and psychotic major depression (PMD).
(N ¼ 14)
(N ¼ 13)
No psychiatric medications
Antidepressant and antipsychotic
Antidepressant and anxiolytic
Antidepressant and mood stabilizer
Antidepressant, antipsychotic, and anxiolytic
Antidepressant, antipsychotic, and mood stabilizer
Antipsychotic and mood stabilizer
Breakdown of psychiatric medications
Any mood stabilizer
K.W. Yuen et al. / Journal of Psychiatric Research 51 (2014) 30e36
Farmingdale, NY). This kit has been validated for use in humans,
and is highly specific and exclusively recognizes OT and not related
peptides (Carter et al., 2007). Per Enzo Life Sciences literature, the
limit of assay sensitivity where the curve is no longer linear is
10 pg/mL. Sample extraction and concentration procedures were
initiated by thawing plasma samples in an ice bath before acidifi-
cation with 0.1% trifluoroacetic acid (TFA). Acidified samples were
then centrifuged at 17,000g for 15 min at 4?C. Phenomenex Strata-
X columns (Phenomenex Inc., Torrance, CA) were activated with
4 mL of HPLC grade methanol followed by 4 mL of molecular
biology grade water. Following column activation, the sample su-
pernatants were applied and drawn through the columns by vac-
uum. Peptides were then eluted by applying 4 mL of pre-elution
buffer (89:10:1 water:acetonitrile:TFA) followed by 4 mL of
elution buffer (80:20 acetonitrile:water). Elutes were evaporated at
room temperature using compressed nitrogen, and reconstituted in
assay buffer prior to quantification. Samples were assayed in
duplicate with a tunable microplate reader for 96-well format ac-
cording to manufacturer’s instructions. A trained technician
without knowledge of experimental conditions performed all as-
says. Intra- and inter-assaycoefficients of variationwerebelow 10%.
Plasma cortisol concentrations were quantified for all partici-
pants using the Access Immunoassay System (Beckman Coulter,
Chaska, MN) as part of the larger parent study. The sensitivity of this
assay is 0.4 ug/dL (11 nmol/L). Cortisol concentrations in our study
were evaluated at2300 h,a timewell-validatedfordetermination of
hypercortisolemia (Boscaro et al., 2001; Raff et al., 1998).
2.4. Phenotypic measurements
Depression severity and anxiety symptomology were deter-
mined using the HDRS total score and the HDRS anxiety subscale,
respectively. The HDRS is a widely used and well-validated, inves-
tigator administered measure of symptom severity and determi-
nation of clinical subtypes for depression (Hamilton, 1960).
Personality characteristics and clinical syndromes were evaluated
with the Million Clinical Multiaxial Inventory (MCMI)-II or -III.
These characteristics correspond to Axis II diagnoses in the DSM-IV.
The MCMI was administered to depressed participants only, as it
was standardized and developed for exclusive use in clinical pop-
ulations (Million et al., 2009). In addition to assessing a wide range
of personality disorders and clinical syndromes, the MCMI includes
three indices that modify a participant’s Base Rate scores: Disclo-
sure, Desirability, Debasement; and, two random response in-
dicators: Validity and Inconsistency.
2.5. Statistical analyses
The plasma OT data were severely skewed, following an expo-
nential distribution. For use as an independent (predictor) variable,
we took the mean of the log of the 2000, 2400, 0400 and 0800 h
measures (which prevents a single data point skewing the mean).
For use as a dependent (predicted) variable, we calculated the
arithmetic mean of the four time points, and then modeled both the
mean and the skewed error distribution directly in a Generalized
Linear Model (GLM; gamma error distribution, reciprocal link
function) implemented using JMP v10 for Windows, and PROC
GENMOD in SAS v9.3 for Windows as needed for detailed post-hoc
tests. This approach is preferable to simple transformation and
modeling in a least-squares framework, particularly because it is
more agnostic and flexible in accommodating the error distribution
of the data (Littell et al., 2002). The 2300 h plasma cortisol data
showed the same skew, and thus were log transformed for use as a
predictor, and treated in the same GLM framework for use as a
We first tested whether Group (PMD, NPMD, and healthy con-
trol) and Gender (female, male), predicted 2300 h plasma cortisol
concentrations. This analysis is essentiallya positive control as PMD
patients are known to show elevated 2300 h cortisol concentra-
tions over NPMD patients (Belanoff et al., 2001). Group was nested
within Depressed, to explicitly test whether Depressed and Non-
depressed participants differed on average; and then whether
NPMD and PMD participants differed from each other; Gender and
Bonferroni-corrected orthogonal planned contrasts werecalculated
for post-hoc-testing. Assumptions of the model were tested
graphically. In addition, this model estimated overdispersion (no
significant evidence for overdispersion was observed).
An identical analysis was used to predict mean plasma OT; and
then also with 2300 h cortisol included as a covariate, yielding
identical results. The simpler model is presented in the Results
section. The post-hoc tests, and model-fit tests discussed above
Finally, the predictive relationships of plasma OT and cortisol
concentrations to measures of 1) depressive severity (HDRS), 2)
anxiety symptoms (HDRS anxiety subscale), and 3) personality
dimensions (MCMI base rate scores) were each tested with the
same multiple linear regression models implemented as a GLM in
JMP v10 for Windows. These analyses excluded healthy controls,
because they either scored low and without variation, or because
no score could be obtained on the instrument. The models were
blocked by Group, Gender, and their interaction, and the log of the
2300 h plasma cortisol, and its interaction with Group. We tested
for an overall effect of the mean of the log of the plasma OT con-
centration, and its interactionwith Group (i.e., whether the effect of
OT differed in PMD and NPMD participants). This model therefore
tested within each blocking factor for consistent effects of OTacross
the blocking factors (e.g., testing whether the effect of OT within
each gender is consistent for both genders). The interaction
explicitly tests whether the effect of OT is different in the two pa-
tient groups. Bonferroni planned contrasts were calculated as
required for post-hoc tests. The assumptions of GLM were tested
graphically, and no transformations were required.
Demographic and clinical characteristics for PMD, NPMD, and
healthy control groups are summarized in Table 2. Because age
differed significantly by group [F2,41 ¼ 3.889; p ¼ 0.028], with
NPMD participants being older than healthy control participants
(p ¼ 0.027; no other groups differed significantly), agewas included
as a statistical covariate in subsequent analyses as appropriate.
Of the 50 participants with available plasma samples, one
participant was omitted for a missing plasma OT value. We also
excluded all female participants prescribed estrogen due to the
known modulatory effects of estrogen on OT (Ochedalski et al.,
2007) and cortisol (White et al., 2006) signaling. These exclusions
yielded a final sample size of 44 participants [13 PMD (9 female, 4
male), 14 NPMD (9 female, 5 male), and 17 healthy controls (9 fe-
male, 8 male)] for the analysis.
As expected, 2300 h cortisol concentrations differed by group.
Specifically the PMD participants showed significantly higher
concentrations than the NPMD participants [Group-nested-within-
depressed Likelihood Ratio (LR) c2¼ 7.76; p ¼ 0.0053; see Table 2],
and further post-hoc tests showed that healthy controls did not
differ from NPMD, but did differ from PMD participants. No other
significant main or interaction effects were observed.
Initial descriptive statistics and subsequent follow up analyses
revealed that PMD and NPMD participants did not differ in plasma
OT concentrations [Group-nested-within-depressed LR c2¼ 0.14;
K.W. Yuen et al. / Journal of Psychiatric Research 51 (2014) 30e36
p ¼ 0.7125]. Instead, the only predictorof plasma OTconcentrations
was a highly significant interaction between depressed group and
gender [Depressed*Gender LR c2¼ 9.54; p ¼ 0.0020]. Males and
females showed opposite patterns between depressed and non-
depressed participants. Depressed females had significantly lower
plasma OT concentrations than healthy control females [Healthy
Control Females vs. Depressed Females: LR c2¼ 6.76; p ¼ 0.0093;
Bonferroni critical alpha correction set at 0.025]. Male participants
generally showed greater OT variability, which rendered some
post-hoc tests non-significant [Healthy Control Males vs.Depressed
Males: LRc2¼ 3.52; p ¼ 0.0605; Bonferroni critical alpha correction
set at 0.025]. Male and female depressed participants significantly
differed [Depressed Males vs. Depressed Females: LR c2¼ 7.33;
p ¼ 0.0068; Bonferroni critical alpha correction set at 0.025], but
their control counterparts did not [Healthy Control Males vs.
Healthy Control Females: LR c2¼ 3.14; p ¼ 0.0762; Bonferroni
critical alpha correction set at 0.025] (see Fig. 1).
For the phenotypic datawe only report significant results for OT,
or its interactions, since the presence of cortisol in the model did
not alter the findings as mentioned above. The MCMI Desirability
Score increased with higher mean log plasma OT [F1,12¼ 5.470;
p ¼ 0.0375; Fig. 2a]. The MCMI Drug Dependence Base Rate Score
decreased with higher mean log plasma OT (F1,12 ¼ 6.428;
p ¼ 0.0262; Fig. 2b). Finally the MCMI Compulsive Base Rate Score,
was predicted by the oxytocin*group interaction (F1,12¼ 9.498;
p ¼ 0.0095; Fig. 2c). For this last result, post-hoc analysis showed
that OT positively predicted compulsivity in the PMD group only
[F1,12¼ 17.19; p ¼ 0.0014; Bonferroni critical alpha correction set at
0.025], but not the NPMD group [F1,12¼ 0.14; p ¼ 0.1361; Bonferroni
critical alpha correction set at 0.025].
This pilot study was designed to investigate whether plasma OT
concentrations differed in the context of hypercortisolemia in PMD,
NPMD, and healthy control participants. We secondarily tested the
relationship between plasma OT concentrations and clinical and
personalitysyndromes in depressed participants. As expected, PMD
participants exhibited significantly higher cortisol concentrations
and NPMD participants did not differ in plasma OT concentrations.
Interestingly, subsequent analyses revealed a significant group by
gender interaction effect, such that depressed females exhibited
lower OTconcentrations compared todepressed males, irrespective
of psychotic status. Further, depressed females, irrespective of psy-
chotic status, exhibited lower OT concentrations compared to
healthy control females. Depressed males, irrespective of psychotic
status, exhibited a strong trend (before controlling for multiple
comparisons), toward higher plasma OT concentrations compared
to healthy control males. These findings were all independent of
cortisol concentrations. A positive relationship was found between
OT concentrations and MCMI Desirability scores for both male and
female depressed participants, irrespective of psychotic status.
Further, an inverse relationship between OT concentrations and
MCMI Drug Dependence Base Rate scores was found for both male
and female depressed participants, irrespective of psychotic status.
Finally, OT concentrations were positively correlated with MCMI
Compulsivity Base Rate scores in PMD, but not NPMD, participants.
The findings that OT concentrations: 1) differed significantly
between female and male depressed participants; and, 2) were
decreased in depressed females but increased (non-significantly) in
depressed males compared to gender-matched healthy controls
were not predicted. One post-hoc explanation is that in females, OT
plays its previously demonstrated role of promoting social bonding
(Lim and Young, 2006; Pedersen et al., 1982; Pedersen and Prange,
1979; Witt et al., 1992), and when females experience social
isolation or loss, as frequently occurs prior to onset of depression
(Bruce and Hoff, 1994), they experience diminished OT release. In
contrast, OT may function as a stress hormone in males, as exoge-
nous OT suppresses cortisol release in response to psychological
stress when paired with social support in males (Heinrichs et al.,
2003). This suggests that males may experience elevated OT
release as a mechanism to reduce stress by promoting social
behavior, increasingsocial support and suppressing cortisol release.
This interpretation may help to explain the disparity in findings
from previous studies examining plasma OT in depression, which
have shown both elevated (Parker et al., 2010) and diminished
(Frasch et al., 1995) plasma OT concentrations in depressed
n i c
o t y
a l P
Fig. 1. Mean plasma oxytocin (OT) concentrations differ by group and sex. Post-hoc
tests revealed that depressed females have lower OT concentrations than healthy
control females, and depressed females have lower OT concentrations than depressed
males. Data show fitted means and standard errors. * indicates a significant finding
after correction for multiple comparisons (p < 0.025). HC, healthy control; NPMD, non-
psychotic major depression; PMD, psychotic major depression.
Participant demographics and clinical characteristics.
(N ¼ 17)
(N ¼ 14)
(N ¼ 13)
p ¼ 0.028
p ¼ 0.817
Education 15.62 (1.62)a
p ¼ 0.638
p < 0.001
p < 0.001
p < 0.001
p < 0.001
HDRS Anxiety 0.47 (1.01)a
p ¼ 0.012
HC, healthy control; NPMD, non-psychotic major depression; PMD, psychotic major
depression; HDRS, Hamilton Depression Rating Scale; BPRS, Brief Psychiatric Rating
Means with the same superscript are not significantly different according to Tukey’s
dOne PMD participant had a missing 11pm plasma cortisol value.
K.W. Yuen et al. / Journal of Psychiatric Research 51 (2014) 30e36
participants. These previous studies consisted of small sample
sizes, and were not powered to examine gender differences in OT
concentrations, raising the possibility that the gender most repre-
sented in each study was perhaps responsible for driving the
Although PMD and NPMD females presented with and without
hypercortisolemia, respectively, they did not differ in OT concen-
trations, with both groups exhibiting equally diminished OT con-
centrations compared to healthy control female participants. There
are two possible explanations for these findings. First, OT biology
may be differentially regulated in females with PMD vs. NPMD.
Specifically, the hypercortisolism that characterizes PMD may
suppress OT release, whereas in NPMD, low circulating OT con-
centrations may be regulated via a cortisol-independent (and as yet
unknown) mechanism. In contrast, it is also possible that dimin-
ished OT concentrations occur via cortisol-independent mecha-
nisms in females for both depressive disorders. The available data
unfortunately do not allow us to differentiate between these two
possibilities, and indicate a clear need for follow up studies to
rigorously test these competing hypotheses.
OT concentrations were significantly correlated with depressed
participants’ concentrations of Desirability scores on the MCMI.
This item is one of the “Modifying Indices” of the MCMI and aims to
measure a participant’s response style to determine whether they
have a propensity to present themselves in a positive light, essen-
tially controlling for the subjective nature of this self-report mea-
sure (Million et al., 2009). Interestingly, it has been hypothesized
that intranasal administration of OT may inflate a person’s desire to
report socially relevant information in a positive manner, thus
altering their responses on self-report measures and interview
questionnaires (Guastella and MacLeod, 2012). Our findings of a
positive correlation between endogenous OT concentrations and
increased MCMI Desirability scores suggest that higher plasma OT
concentrations represent a heightened drive to be viewed posi-
tively byothers, at least for depressed participants. This finding also
aligns with the established evidence that OT is critical in the
expression of prosocial and affiliative behaviors (Young, 2009).
Our finding of an inverse relationship between OT plasma
concentrations and MCMI Drug Dependence scores are supported
by increasing evidence that suggests dysregulation in OT signaling
may play a significant role in the development of substance abuse
problems and associated psychosocial issues, including depression
(see Carson et al., 2013; McGregor and Bowen, 2012; McGregor
et al., 2008 for review). For example, chronic administration of
cocaine to pregnant and post-partum rats results in aberrant
mothering behavior and increased aggression toward a resident
intruder and these behavioral changes are associated with
decreased OT concentrations in the medial preoptic area, ventral
tegmental area, and hippocampus, as well as reduced OT receptor
binding density in the ventral medial hypothalamus and the bed
nucleus of the stria terminalis (Jarrett et al., 2006; Johns et al.,
1997a,b). Importantly, this research is supported by evidence in
humans. As outlined in the Introduction, women who used cocaine
during pregnancy show blunted plasma OT concentrations, expe-
rience increased hostility and depressed mood, and have less social
support and adaptive coping strategies for stressful life events than
do healthy non-cocaine using mothers (Light et al., 2004). In other
relevant preclinical studies repeat cocaine self-administration re-
sults in decreased hippocampal and hypothalamic OT concentra-
tions, chronic exposure to opiates decreases central OT synthesis
(You et al., 2000), and chronic exposure to alcohol is associated
with degeneration of hypothalamic magnocellular OT containing
neurons in both humans and rats (Silva et al., 2002; Sivukhina et al.,
The finding that OT concentrations were positively correlated
with MCMI Compulsivity scores in PMD, but not NPMD, partici-
pants are intriguing. Preclinical studies have shown that both
central and peripheral injections of OT to rodents result in repeti-
tive and compulsive grooming behavior (Sarnyai, 1998). Further,
Leckman et al. (1994) provided evidence for elevated CSF OT
e r o
y t i l i b
a r i s
e r o
Mean Plasma Oxytocin Concentrations (pg/mL)
e r o
v i s l u
Fig. 2. Mean plasma OT concentrations: positively predict MCMI Desirability scores in
depressed participants (Panel A), negatively predict MCMI Drug Dependence Base Rate
scores in depressed participants (Panel B), and positively predict MCMI Compulsivity
Base Rate scores in PMD, but not NPMD, participants (Panel C). To correct for skew,
mean plasma OTconcentration is calculated as the mean of logged titers. To correct for
the controlling variables in the analysis, the regression line is partialed (controlled) for
other variables in the analysis, and calculated at the mean value of those variables.
Similarly, data are shown as residuals from this mean regression line. NPMD, non-
psychotic major depression; PMD, psychotic major depression.
K.W. Yuen et al. / Journal of Psychiatric Research 51 (2014) 30e36
concentrations in patients with obsessiveecompulsive disorder
(OCD) compared to patients with Tourette’s syndrome and healthy
controls. They further showed that CSF OT concentrations were
positively correlated with current severity of OCD symptoms.
However, given the evidence that higher plasma OTconcentrations
and intranasal administration of OT are associated with reduced
positive symptoms in schizophrenia it is difficult to posit why
plasma OT might be related to compulsivity in PMD, but not NPMD,
participants (Pedersen et al., 2011; Rubin et al., 2010).
5. Limitations and conclusions
While producing interesting results, our study has several lim-
itations. First, we did not experimentally control menstrual
cyclicity, which is problematic given the known interaction of fe-
male reproductive hormones and OT on modulating the stress
response (Ochedalski et al., 2007; Taylor et al., 2006). This study
was also limited by the availability of social behavioral data, which
precluded detailed analysis of OT concentrations in relation to so-
cial functioning. In addition, much like other studies in this field
(Frasch et al., 1995; Garcia et al., 2011; Parker et al., 2010) our
sample size was relatively small, particularly for males, so may not
have been statistically powered to find significant relationships
between OT concentrations in both males and females. Research
with more participants, especially males, is necessary to bring
greater clarity to gender effects on OT biology in depression.
Research with more PMD participants is also necessary in order to
determine whether low OT and high cortisol concentrations are
associated. Finally, central and peripheral OTconcentrations can be
independently regulated under certain conditions (Landgraf and
Neumann, 2004; Ludwig and Leng, 2006) thus leaving the possi-
bility that clinical symptoms in our study population reflect dys-
regulation in brain OT systems that are not reflected in plasma
samples. Future studies would benefit from assessing the rela-
tionship between concomitantly collected blood and CSF samples
to more precisely determine the role of OT biology in depressive
In conclusion, these findings further implicate a role for dysre-
gulated OT biology in patients with major depression, and extend
this knowledge for the first time to patients with psychotic major
depression. Evidence from this study indicates that OT signaling
may provide a mechanism by which to better understand female-
biased risk to develop depressive disorders, and suggests that
intranasal OT “replacement” therapy may be efficacious in female
patients with known impairments in OT signaling. Finally, data
from this study suggest that plasma OT concentrations may be a
useful biomarker for the presence of certain clinical and personality
Role of Funding Source
Funding for this study was provided by Stanford University and
The National Institutes of Health, Bethesda, MD (grants MH50604,
MH66537, and RR-00070). The funding sources had no further role
in the collection, analysis and interpretation of data; in the writing
of the report; or in the decision to submit the manuscript for
All authors read and approved the final manuscript. Kaeli Yuen,
Dr. Alan Schatzberg, and Dr. Karen Parker conceptualized the
theoretical framework for the research presented in this paper. Dr.
Jennifer Keller and Dr. Schatzberg oversaw data collection for the
parent study which was conducted by Dr. Keller, Dr. Anna Lembke,
Heather Kenna, and Dr. Lakshika Tennakoon in Dr. Schatzberg’s
laboratory. Shellie Hyde and Ms. Yuen prepared biological samples
and conducted hormone assays in Dr. Parker’s laboratory. Ms. Yuen
and Ms. Hyde also entered experimental data, created the statistical
database, and helped write the methods section of this manuscript.
Dr. Joseph Garner, Dr. Dean Carson, and Dr. Parker conducted the
statistical analyses, Dr. Garner wrote the statistical analysis section
and the experimental results sections, and Dr. Carson created the
figuresand table. Ms. Yuen, Dr. Parker, and Dr. Carsonwrote the first
draft of the manuscript. Dr. Schatzberg, Dr. Parker, and Ms. Yuen
provided financial support for the project.
Conflict of Interest
Ms. Yuen, Dr. Garner, Dr. Carson, Dr. Keller, Dr. Lembke, Ms.
Hyde, Ms. Kenna, Dr. Tennakoon, Dr. Schatzberg, and Dr. Parker
report no biomedical financial interests or potential conflicts of
This research program was supported by grants MH50604,
MH66537, and RR-00070 from the National Institutes of Health,
Bethesda, MD, a Stanford University Undergraduate Advising and
Research Student Major Grant, and Stanford University’s Bio-X
Program. We also thank Dr. Robert Sapolsky for his thoughtful
comments on Kaeli Yuen’s honor’s thesis, from which this research
report is derived.
Amico JA, Mantella RC, Vollmer RR, Li X. Anxiety and stress responses in female
oxytocin deficient mice. Journal of Neuroendocrinology 2004;16:319e24.
Andari E, Schneider FC, Mottolese R, Vindras P, Sirigu A. Oxytocin’s fingerprint in
personality traits and regional brain volume. Cerebral Cortex 2012. http://
Belanoff JK, Kalehzan M, Sund B, Ficek SKF, Schatzberg AF. Cortisol activity and
cognitive changes in psychotic major depression. American Journal of Psychi-
Bertsch K, Schmidinger I, Neumann ID, Herpertz SC. Reduced plasma oxytocin levels
in female patients with borderline personality disorder. Hormones and
Boscaro M, Barzon L, Fallo F, Sonino N. Cushing’s syndrome. Lancet 2001;357:783e
Bruce ML, Hoff RA. Social and physical health risk-factors for 1st-onset major
depressive disorder in a community sample. Social Psychiatry and Psychiatric
Carson DS, Guastella AJ, Taylor ER, McGregor IS. A brief history of oxytocin and its
role in modulating psychostimulant effects. Journal of Psychopharmacology
Carter CS, Pournajafi-Nazarloo H, Kramer KM, Ziegler TE, White-Traut R, Bello D,
et al. Oxytocin: behavioral associations and potential as a salivary biomarker.
Annals of the New York Academy of Sciences 2007;1098:312e22.
First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for
DSM-IV-TR Axis I disorders (Patient Edition). SCID-I/P, 2/2001 revision ed.
Washington, DC: American Psychiatric Press; 1997.
Frasch A, Zetzsche T, Steiger A, Jirikowski GF. Reduction of plasma oxytocin levels in
patients suffering from major depression. Advances in Experimental Medicine
and Biology 1995;395:257e8.
Garcia FD, Coquerel Q, Kiive E, Dechelotte P, Harro J, Fetissov SO. Autoantibodies
reacting with vasopressin and oxytocin in relation to cortisol secretion in mild
and moderate depression. Progress in Neuro-Psychopharmacology & Biological
Gold PW, Loriaux DL, Roy A, Kling MA, Calabrese JR, Kellner CH, et al. Responses to
corticotropin-releasing hormone in the hypercortisolism of depression and
Cushing’s disease. Pathophysiologic and diagnostic implications. The New En-
gland Journal of Medicine 1986;314:1329e35.
Guastella AJ, MacLeod C. A critical review of the influence of oxytocin nasal spray on
social cognition in humans: evidence and future directions. Hormones and
Hamilton M. A rating scale for depression. Journal of Neurololgy, Neurosurgery and
Heim C, Young LJ, Newport DJ, Mletzko T, Miller AH, Nemeroff CB. Lower CSF
oxytocin concentrations in women with a history of childhood abuse. Molecular
K.W. Yuen et al. / Journal of Psychiatric Research 51 (2014) 30e36
Heinrichs M, Baumgartner T, Kirschbaum C, Ehlert U. Social support and oxytocin Download full-text
interact to suppress cortisol and subjective responses to psychosocial stress.
Biological Psychiatry 2003;54:1389e98.
Heuser I, Yassouridis A, Holsboer F. The combined dexamethasone/CRH test: a
refined laboratory test for psychiatric disorders. Journal of Psychiatric Research
Jarrett TM, McMurray MS, Walker CH, Johns JM. Cocaine treatment alters oxytocin
receptor binding but not mRNA production in postpartum rat dams. Neuro-
Johns JM, Lubin DA, Walker CH, Meter KE, Mason GA. Chronic gestational cocaine
treatment decreases oxytocin levels in the medial preoptic area, ventral
tegmental area and hippocampus in Sprague-Dawley rats. Neuropeptides
Johns JM, Noonan LR, Zimmerman LI, Li L, Pedersen CA. Effects of short- and long-
term withdrawal from gestational cocaine treatment on maternal behavior and
aggression in Sprague-Dawley rats. Developmental Neuroscience 1997b;19:
Jokinen J, Chatzittofis A, Hellstrom C, Nordstrom P, Uvnas-Moberg K, Asberg M. Low
Keller J, Flores B, Gomez RG, Solvason HB, Kenna H, Williams GH, et al. Cortisol
circadian rhythm alterations in psychotic major depression. Biological Psychi-
Kiecolt-Glaser JK, Glaser R, Cacioppo JT, Malarkey WB. Marital stress: immunologic,
neuroendocrine, and autonomic correlates. Annals of the New York Academy of
Kiecolt-Glaser JK, Newton T, Cacioppo JT, MacCallum RC, Glaser R, Malarkey WB.
Marital conflict and endocrine function: are men really more physiologically
affected than women? Journal of Consulting and Clinical Psychology 1996;64:
Knox SS, Uvnas-Moberg K. Social isolation and cardiovascular disease: an athero-
sclerotic pathway? Psychoneuroendocrinology 1998;23:877e90.
Landgraf R, Neumann ID. Vasopressin and oxytocin release within the brain: a
dynamic concept of multiple and variable modes of neuropeptide communi-
cation. Frontiers in Neuroendocrinology 2004;25:150e76.
Leckman JF, Goodman WK, North WG, Chappell PB, Price LH, Pauls DL, et al.
Elevated cerebrospinal-fluid levels of oxytocin in obsessive-compulsive disor-
der e comparison with Tourettes-syndrome and healthy controls. Archives of
General Psychiatry 1994;51:782e92.
Lee R, Ferris C, Van de Kar LD, Coccaro EF. Cerebrospinal fluid oxytocin, life history
of aggression, and personality disorder. Psychoneuroendocrinology 2009;34:
Light KC, Grewen KM, Amico JA, Boccia M, Brownley KA, Johns JM. Deficits in
plasma oxytocin responses and increased negative affect, stress, and blood
pressure in mothers with cocaine exposure during pregnancy. Addictive Be-
Lim MM, Young LJ. Neuropeptidergic regulation of affiliative behavior and social
bonding in animals. Hormones and Behavior 2006;50:506e17.
Littell RC, Stroup WW, Freund RJ. SAS for linear models. 4th ed. Cary, NC: SAS
Ludwig M, Leng G. Dendritic peptide release and peptide-dependent behaviours.
Nature Reviews Neuroscience 2006;7:126e36.
McGregor IS, Bowen MT. Breaking the loop: oxytocin as a potential treatment for
drug addiction. Hormones and Behavior 2012;61:331e9.
McGregor IS, Callaghan PD, Hunt GE. From ultrasocial to antisocial: a role for
oxytocin in the acute reinforcing effects and long-term adverse consequences of
drug use? British Journal of Pharmacology 2008;154:358e68.
Million T, Million C, Davis R, Grossman S. MCMI-III manual. Minneapolis, MN:
Pearson Education, Inc.; 2009.
Nestler EJ, Barrot M, DiLeone RJ, Eisch AJ, Gold SJ, Monteggia LM. Neurobiology of
depression. Neuron 2002;34:13e25.
Neumann ID, Kromer SA, Toschi N, Ebner K. Brain oxytocin inhibits the (re)activity
of the hypothalamo-pituitary-adrenal axis in male rats: involvement of hypo-
thalamic and limbic brain regions. Regulatory Peptides 2000;96:31e8.
Ochedalski T, Subburaju S, Wynn PC, Aguilera G. Interaction between oestrogen and
oxytocin on hypothalamic-pituitary-adrenal axis activity. Journal of Neuroen-
Overall JE, Gorham DR. The brief psychiatric rating scale. Psychological Reports
Ozsoy S, Esel E, Kula M. Serum oxytocin levels in patients with depression and the
effects of gender and antidepressant treatment. Psychiatry Research 2009;169:
Parker KJ, Kenna HA, Zeitzer JM, Keller J, Blasey CM, Amico JA, et al. Preliminary
evidence that plasma oxytocin levels are elevated in major depression. Psy-
chiatry Research 2010;178:359e62.
Parker KJ, Schatzberg AF, Lyons DM. Neuroendocrine aspects of hypercortisolism in
major depression. Hormones and Behavior 2003;43:60e6.
Pedersen CA, Prange AJ. Induction of maternal behavior in virgin rats after intra-
cerebroventricular administration of oxytocin. Proceedings of the National
Academy of Sciences of the United States of America 1979;76:6661e5.
Pedersen CA, Ascher JA, Monroe YL, Prange AJ. Oxytocin induces maternal behavior
in virgin female rats. Science 1982;216:648e50.
Pedersen CA, Gibson CM, Rau SW, Salimi K, Smedley KL, Casey RL, et al. Intranasal
oxytocin reduces psychotic symptoms and improves Theory of Mind and social
perception in schizophrenia. Schizophrenia Research 2011;132:50e3.
Raff H, Raff JL, Findling JW. Late-night salivary cortisol as a screening test for
Cushing’s syndrome. The Journal of Clinical Endocrinology and Metabolism
Rubin LH, Carter CS, Drogos L, Pournajafi-Nazarloo H, Sweeney JA, Maki PM. Pe-
ripheral oxytocin is associated with reduced symptom severity in schizo-
phrenia. Schizophrenia Research 2010;124:13e21.
Sarnyai Z. Oxytocin and neuroadaptation to cocaine. Progress in Brain Research
Scantamburlo G, Hansenne M, Fuchs S, Pitchot W, Marechal P, Pequeux C, et al.
Plasma oxytocin levels and anxiety in patients with major depression. Psy-
Schatzberg AF, Keller J, Tennakoon L, Lembke A, Williams G, Kraemer FB, et al. HPA
axis genetic variation, cortisol, and psychosis in major depression. Molecular
Psychiatry 2013 Oct 29. http://dx.doi.org/10.1038/mp.2013.129. [Epub ahead of
print]. PMID: 24166410.
Silva SM, Madeira MD, Ruela C, Paula-Barbosa MM. Prolonged alcohol intake leads
to irreversible loss of vasopressin and oxytocin neurons in the paraventricular
nucleus of the hypothalamus. Brain Research 2002;925:76e88.
Sivukhina EV, Dolzhikov AA, Morozov IE, Jirikowski GF, Grinevich V. Effects of
chronic alcoholic disease on magnocellular and parvocellular hypothalamic
neurons in men. Hormone and Metabolic Research 2006;38:382e90.
Strathearn L, Fonagy P, Amico J, Montague PR. Adult attachment predicts maternal
brain and oxytocin response to infant cues. Neuropsychopharmacology
Strathearn L, Iyengar U, Fonagy P, Kim S. Maternal oxytocin response during
mothereinfant interaction: associations with adult temperament. Hormones
and Behavior 2012;61:429e35.
Stroud LR, Salovey P, Epel ES. Sex differences in stress responses: social rejection
versus achievement stress. Biological Psychiatry 2002;52:318e27.
Taylor SE, Gonzaga GC, Klein LC, Hu PF, Greendale GA, Seeman TE. Relation of
oxytocin to psychologicalstress responses
Tops M, Van Peer JM, Korf J, Wijers AA, Tucker DM. Anxiety, cortisol, and attach-
ment predict plasma oxytocin. Psychophysiology 2007;44:444e9.
White T, Ozel B, Jain JK, Stanczyk FZ. Effects of transdermal and oral contraceptives
on estrogen-sensitive hepatic proteins. Contraception 2006;74:293e6.
Witt DM, Winslow JT, Insel TR. Enhanced social interactions in rats following
chronic, centrally infused oxytocin. Pharmacology Biochemistry and Behavior
You ZD, Li JH, Song CY, Wang CH, Lu CL. Chronic morphine treatment inhibits
oxytocin synthesis in rats. Neuroreport 2000;11:3113e6.
Young EA. Sex differences and the HPA axis: implications for psychiatric disease.
The Journal of Gender Specific Medicine 1998;1:21e7.
Young LJ. Being human: love: neuroscience reveals all. Nature 2009;457:148.
Young KA, Liu Y, Wang Z. The neurobiology of social attachment: a comparative
Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology
K.W. Yuen et al. / Journal of Psychiatric Research 51 (2014) 30e36