American Journal of Primatology 74:528542 (2012)

RESEARCH ARTICLE
Modeling Depression in Adult Female Cynomolgus Monkeys
(Macaca fascicularis)
STEPHANIE L. WILLARD1 AND CAROL A. SHIVELY2
1
Integrative Neuroscience Graduate Program, Wake Forest School of Medicine, Winston-Salem, North Carolina
2
Department of Pathology, Section on Comparative Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina

Depressive disorders are prevalent, costly, and poorly understood. Male rodents in stress paradigms
are most commonly used as animal models, despite the two-fold increased prevalence of depression
in women and sex differences in response to stress. Although these models have provided
valuable insights, new models are needed to move the eld forward. Social stress-associated
behavioral depression in adult female cynomolgus macaques closely resembles human depression in
physiological, neurobiological, and behavioral characteristics, including reduced body mass,
hypothalamicpituitaryadrenal axis perturbations, autonomic dysfunction, increased cardiovascular
disease risk, reduced hippocampal volume, altered serotonergic function, decreased activity levels,
and increased mortality. In addition, behaviorally depressed monkeys also have low ovarian
steroid concentrations, even though they continue to have menstrual cycles. Although this type of
ovarian dysfunction has not been reported in depressed women and is difcult to identify, it may be the
key to understanding the high prevalence of depression in women. Depressive behavior in female
cynomolgus monkeys is naturally occurring and not induced by experimental manipulation. Different
social environmental challenges, including isolation vs. subordination, may elicit the depression-like
response in some animals and not others. Similarly, social subordination is stressful and depressive
behavior is more common in socially subordinate monkeys. Yet, not all subordinates exhibit behavioral
depression, suggesting individual differences in sensitivity to specic environmental stressors and
enhanced risk of behavioral depression in some individuals. The behavior and neurobiology of
subordinates is distinctly different than that of behaviorally depressed monkeys, which affords the
opportunity to differentiate between stressed and depressed states. Thus, behaviorally depressed
monkeys exhibit numerous physiological, neurobiological, and behavioral characteristics same as those
of depressed human beings. The nonhuman primate model represents a new animal model of
depression with great promise for furthering our understanding of this prevalent and debilitating
disease and identifying novel therapeutic targets. Am. J. Primatol. 74:528542, 2012. r 2011 Wiley
Periodicals, Inc.

Key words: depression; female; nonhuman primate; animal models; stress; ovarian function;
cardiovascular disease

INTRODUCTION
Depression is the leading cause of disability and
the fourth leading cause of disease burden worldwide, with a lifetime prevalence rate of 16.6%
[Kessler et al., 2005; World Health Organization,
2001]. Given that depression begins early in life and
results in substantial role impairment in more than
half the depressed patients, depression has severe
economic ramications [Kessler et al., 2005; Wang
et al., 2003; World Health Organization, 2001].
Depression is also highly comorbid with numerous
medical and psychiatric illnesses, worsens comorbid
disease outcomes, and is associated with increased
mortality risk [Cuijpers & Schoevers, 2004; Jackson
et al., 2004; Krishnan, 2003]. With less than half the

r 2011 Wiley Periodicals, Inc.

depressed patients responding to standard pharmacotherapies, depression remains a signicant treatment challenge [Nemeroff, 2007].
Contract grant sponsor: NIH; Contract grant numbers: RO1
HL39789; RO1 HL87103; RO1 MH56881; R21 MH086731;
Contract grant sponsor: John D. and Catherine T. MacArthur
Foundation.
Correspondence to: Carol A. Shively, Department of Pathology,

Section on Comparative Medicine, Wake Forest School of

Medicine, Medical Center Boulevard, Winston-Salem, NC
27157-1040. E-mail: cshively@wakehealth.edu
Received 15 April 2011; revised 13 September 2011; revision
accepted 23 September 2011
DOI 10.1002/ajp.21013
Published online 10 November 2011 in Wiley Online Library
(wileyonlinelibrary.com).

2 / Willard and Shively

Despite the worldwide prevalence and burden of

depression, progress toward understanding the various mechanisms involved in its pathophysiology has
been slow compared with other prevalent disease
states owing largely to the challenge of assessing the
living brain. Noninvasive neuroimaging technology
has provided some insight into neural mechanisms,
but currently does not allow an investigation of
mechanisms at the cellular and molecular level.
Moreover, depression is a whole-body disease that
is heterogeneous in manifestation, subjectively diagnosed, and often accompanied by alterations in most
of the major physiological systems. Thus, our ability
to uncover the mechanisms of depression depends
largely on the development of appropriate animal
models of depression that reect the complexity of
this disease.
ANIMAL MODELS OF DEPRESSION
Challenge of Modeling a Complex Disease
Animal models are evaluated based on their
construct, face, and predictive validities [McKinney,
1984; Willner, 1984]. As such, an ideal animal model
is created through the same etiologic processes as the
human disease being modeled (construct validity),
has phenotype similarity to the human disease (face
validity), and predicts human responses to interventions (predictive validity). Developing appropriate
animal models of depression has proven difcult,
given the vast heterogeneity in symptomology,
causes and course of illness, and treatment of human
depression. There are multiple subtypes within the
depression syndrome, and the symptom diagnosis
pool is almost twice that of the number of symptoms
that must be present for a particular diagnosis
[Rush, 2007]. In addition, many of the symptoms
may be present in one direction or the other,
including increased or decreased appetite, weight,
sleep, and motor activity [American Psychiatric
Association, 2000], resulting in patients with a wide
variety of symptom proles. Reproducing such
complex depression symptomology in animal models
is highly challenging, as is assessing the validity of
such models when diagnosis is subjective and so little
is known about the pathophysiology of depression.
There are several excellent reviews in the recent
literature that detail the validity and utility of
animal models of depression [i.e., Duman, 2010;
Krishnan & Nestler, 2010; Nestler & Hyman, 2010].
Thus, animal models will be discussed only briey
below.
Traditional Models
Historically, animal models of depression have
been mostly male rodents in stress paradigms that
capture behavioral and physiological responses to
stress. These models are used for a number of

Am. J. Primatol.

Adult Female Monkey Model of Depression / 529

reasons: (1) it is well-established that stress often

precipitates human depression [Kessler, 1997]; (2)
the depression-like behavior produced by these stress
paradigms is generally reversed with antidepressants; and (3) rodents are inexpensive, have relatively short life spans, and thus provide a fast
screening tool for novel pharmacotherapies [Duman,
2010; Krishnan & Nestler, 2010; Nestler & Hyman,
2010].
Some rodent models are routinely used as
pharmacologic screening tools, such as the learned
helplessness, forced swim, and tail suspension tests.
These models employ acute stress to produce
immobility or helplessness phenotypes, both considered forms of behavioral despair. Although much has
been gleaned from these models in the way of
understanding monoamine mechanisms of antidepressant action, the ability of these models to
uncover novel mechanisms in depression may
be thwarted by their selectivity for monoamines
[Berton & Nestler, 2006; Cryan et al., 2002].
However, less than half the depressed patients
respond to the standard pharmacotherapies that were
developed using these rodent models [Nemeroff,
2007]. In addition, some of these models use relatively
acute stress to invoke an immediate antidepressant
response [Berton & Nestler, 2006], whereas chronic
treatment is necessary to produce a response in
humans. Unlike the forced swim and tail suspension
tests, the learned helplessness model produces a
syndrome of physiological endpoints that actually
look similar to human depression, such as weight loss,
decreased
activity
and
motivation,
and
increased corticosteroids, and the time course to
antidepressant response is longer [Maier & Watkins,
2005]. The degree to which acute stress-induced
behavioral despair in rodents promotes understanding of the complex etiology and heterogenous symptomology of human depression remains an open
question.
Given that chronic stress has long been associated with human depression [Kessler, 1997; Liu &
Alloy, 2010], more recent animal model development
has utilized chronic stress paradigms. For example,
chronic mild or unpredictable stress uses physical
stressors and/or social stressors to produce anhedonia, a hallmark feature of depression, which can be
reversed by chronic, but not acute, antidepressant
treatment [Willner, 2005]. Alternatively, social
defeat models use social conict and the resulting
social subordination as a means of inducing a
depression-like syndrome that may include anhedonia, reduced social interaction, metabolic abnormalities, and suppression of hippocampal neurogenesis,
and these effects can last weeks to months after
the removal of the stressor [Duman, 2010; Van
Bokhoven et al., 2011]. The anhedonia and social
withdrawal in these defeat models are reversed by
chronic, but not acute, antidepressant treatment

Am. J. Primatol.

530 / Willard and Shively

[Krishnan et al., 2007]. Moreover, not all animals

exposed to social defeat stress will exhibit the
depression-like syndrome, which allows for the study
of why some populations are resilient to stress,
whereas others are at risk, as is observed in humans
[Charney, 2004; Krishnan et al., 2007]. Thus, these
chronic stress models hold the potential to provide
greater mechanistic insight than traditional drugscreening models, given their ability to model more
naturalistic stress-induced depression-like behavior.
Importance of Sex Differences
Traditionally, most rodent stress models use
only males, although (1) depression is twice as
common in women, (2) response to antidepressant
therapy may be gender specic, and (3) human sex
differences are reported in the function and morphology of mood-related brain regions [Gorman,
2006; Maller et al., 2007; Marcus et al., 2005]. The
higher incidence of female depression is not lost on
rodent researchers; many are limited by the low
levels of aggression characteristic of female rodents
[Palanza et al., 2001]. Indeed, some groups have
recently attempted to apply stress models of depression to female rodents, with varying degrees of success
[Dalla et al., 2010; Schmidt et al., 2010; Ter Horst
et al., 2009]. Emerging evidence indicates that rodent
neurobiological and behavioral responses are sex
specic in some of the commonly used paradigms,
including the learned helplessness, chronic mild
stress, social defeat, and the forced swim test [Dalla
et al., 2010; Ter Horst et al., 2009; Trainor et al.,
2011]. This evidence suggests that some of these
paradigms may be less appropriate for female rodents,
research on male animals in these models may not
generalize to women, and other animal models are
needed to elucidate mechanisms of female depression.
Invaluable insights into the neurobiology of
depression have been provided by rodent stress
models; however, the limitations to validity and
utility of these models for human depression are
well recognized [Nestler & Hyman, 2010]. New
models that capture more and different aspects of
the complex etiology and heterogenous symptomology of human depression are needed. More in-depth
investigations of mechanisms underlying the complex pathophysiology of depression would benet
substantially from the development of sex-specic
animal models that closely resemble human physiology, neurobiology, and behavior.
Attributes of Nonhuman Primates That May
Be Valuable to Depression Research
Nonhuman primates offer the unique opportunity to translate between basic and clinical science.
Macaque species, in particular, have proven advantageous for modeling human disease [Shively &
Clarkson, 2009]. Macaques are small enough to

Am. J. Primatol.

Adult Female Monkey Model of Depression / 3

house and handle easily, yet large enough to

noninvasively image effectively. They have 95%
overall genetic coding sequence identity to humans
[Magness et al., 2005] and are more similar to
humans in multiple physiological systems than are
rodents. Importantly for sex-specic disease processes, female macaques have a true menstrual cycle
like women in which the endometrium is shed,
unlike rodents which absorb the endometrium
during the estrous cycle. In macaques, the length
and associated hormone uctuations of the menstrual
cycle are likewise similar to women, thus providing
an ideal model for investigating the relationship
between depression and ovarian steroids. Finally, the
macaque brain is far more similar to that of the
human in terms of cortex elaboration, nuclear
organization, connectivity patterns, and overall
evolutionary conservation of functional areas [Amaral
& Lavenex, 2007; Hofman, 1989; Preuss, 1995;
Wise, 2008]. Rodent models of depression have
focused on the hippocampus as it is stress responsive.
However, cortical areas, particularly the prefrontal
cortex, anterior cingulate, and subgenual cingulate
cortex seem to be critically involved in human
depression. These are well represented in the macaque brain, but not present or only rudimentarily
dened in the rodent brain [Preuss, 1995; Wise,
2008]. This neuroanatomical difference limits the
translation between the neurobiology of stress models
of depression in rodents and the neurobiology of
depression in primates.
Recently, Perera et al. reported the results of a
small study of six adult female bonnet macaque
(Macaca radiata) behavioral responses to social
isolation stress [Perera et al., 2011]. Briey, socially
housed macaques were separated from their pen
mates and socially isolated 2 days/week for 15 weeks;
three of the monkeys were simultaneously treated
with the selective serotonin reuptake inhibitor
(SSRI) uoxetine. The three untreated monkeys
seemed to gradually increase anhedonic behaviors
(dened as a behavioral composite of collapsed
postures, inactivity, and blank stares) and decrease
social status scores (calculated as total subordinate
behaviors subtracted from total dominant behaviors). Fluoxetine seemed to ameliorate the behavioral effects of the social isolation stress in the three
treated monkeys. Further study will reveal the
utility of this potential model.
Notably, all the animal models previously discussed induce depressive-like behavior by stressing
the animal. Thus, the effects of stress are not
separable from the depression-like behavior.
PRIMATE MODEL OF SPONTANEOUSLY
OCCURRING DEPRESSIVE BEHAVIOR
We have developed an adult female nonhuman
primate model of depression using cynomolgus

Am. J. Primatol.

4 / Willard and Shively

macaques (Macaca fascicularis) that were wildcaught as adults from a purpose-bred free-ranging
island colony in Indonesia. In order to approximate
typical dietary constituents consumed in Western
societies, the monkeys are fed a diet containing
moderate amounts of fat and cholesterol. The
monkeys are housed under a 12/12 light/dark cycle
in small social groups of 35 monkeys, and live in
rooms (810 m3) with visual access to the outside
that are enriched with perches, barrels, and manipulanda, including mirrors and toys.
When placed in social groups, adult female
cynomolgus monkeys quickly organize themselves
into linear social status hierarchies that are stable
over long periods of time [Shively & Kaplan, 1991].
Social status is evaluated by recording the outcomes
of aggressive interactions between cage mates, and
has been described in detail previously [Shively et al.,
1997, 2002]. Briey, the animal to which all other
cage mates send submissions is considered rst
ranking. The animal to which all other cage mates,
except the rst-ranking animal, send submissions is
considered second ranking and so forth. On average,
the rst- and second-ranking animals are considered
dominant, whereas the third- and fourth-ranking
animals are considered subordinate. Compared with
their dominant counterparts, subordinate females
receive more aggression, are groomed less, are more
vigilant, and spend more time alone [Shively, 1998;
Shively et al., 1997]. Subordinates are also hypercortisolemic, insensitive to glucocorticoid negative
feedback in a dexamethasone suppression test
(DST), respond to a standardized stressor with heart
rates that recover more slowly than dominants, and
have poor ovarian function [Kaplan et al., 1986,
2010; Shively, 1998; Shively et al., 1997]. As such,
socially subordinate female cynomolgus monkeys are
stressed relative to dominant monkeys. Although
subordinate animals are more susceptible than
dominants to various disease pathologies, this review
focuses on behavioral depression. It is important to
note here that, unlike rodent stress paradigms,
stress is not used as an experimental manipulation
to produce depressive-like behavior in this monkey
model. Social subordination stress is not an experimental manipulation, but instead a normal facet of
macaque life that exists in both captivity and the
wild. The relationship between social subordination
stress and depression is described in more detail
below.
Denition and Incidence of Behavioral
Depression
In the late 1980s, we began observing and
documenting a behavior displayed by some monkeys
that was similar to a behavioral phenotype exhibited
by maternally separated infant macaques [Harlow &
Suomi, 1974]. As depicted in Figure 1, this

Am. J. Primatol.

Adult Female Monkey Model of Depression / 531

Fig. 1. A nondepressed monkey (A) compared with a monkey

displaying the depressed posture (B). The nondepressed monkey
responds to the photographer, a potential threat, by being alert
and attentive, whereas the monkey displaying depressive
behavior is inattentive and in a slumped posture, with open
eyes in a downward gaze.

behavioral depression is operationally dened as a

slumped or collapsed body posture, head at or below
shoulders, in which an animals eyes are open, yet
the animal lacks interest or responsivity to environmental events [Shively et al., 1997, 2005; Suomi
et al., 1975]. The lack of interest and responsivity to
the environment may reect anhedonia, similar to
that observed in the Perera model mentioned above.
Behavioral depression in captive cynomolgus macaques seems to be spontaneous and not induced by a
specic experimental manipulation. Details of behavior collection technique and depressive classication have been described in detail previously [Shively
et al., 1997, 2005]. Briey, we have observed and
recorded the percent time spent in depressive
behavior during weekly observations in 78 adult
female monkeys over two experiments, with interobserver reliability of at least 92% across studies. All
animals exhibit behavioral depression at different
rates, ranging from never to very often. Much like
humans, the incidence of depressive behavior varies
among individuals. Thus, the behavior can be plotted
as a distribution, and that distribution drives the
manner in which behaviorally depressed and
nondepressed monkeys are characterized. In the
rst experiment (Experiment I), the distribution
tended to be bimodal (monkeys either did or did not
engage in this behavior) [Shively et al., 1997].
Therefore, those that were ever observed in the
depressed posture were classied as behaviorally
depressed and those that never exhibited behavioral depression were characterized as nondepressed.
In the second experiment (Experiment II), percent
time spent in the depressed posture was more
broadly distributed and was divided into two groups
at the mean [Shively et al., 2005]. Monkeys that fell
above the mean were designated as behaviorally
depressed and those that fell below the mean
were designated nondepressed. The incidence of

Am. J. Primatol.

532 / Willard and Shively

behavioral depression was similar between the two

experiments: 16 of 42 monkeys (38%) in Experiment
I [Shively et al., 1997] and 15 of 36 monkeys (42%) in
Experiment II [Shively et al., 2005] displayed
depressive behavior.
Relationship Between Depressive Behavior
and Social Status
Following a 3-month quarantine, the 42 animals
in Experiment I lived in their rst social groups for 8
weeks and their second social groups for 26 months
(Fig. 2). All social groups were counterbalanced for
body weight before their formation. Social status was
determined in their rst social groups, and then the
constituency of the social groups was changed.
Subordinates (the two bottom-ranking monkeys)
were housed with subordinates and dominants (the
two top-ranking monkeys) were housed with dominants. Linear social status hierarchies reformed
resulting in four groups: subordinates that had a
history of subordination; subordinates that became
dominant; dominants that became subordinate; and
dominants with a history of social dominance.
Current subordinates were more likely to exhibit
depression than current dominants (14/23 5 61% vs.
2/19 5 10%). However, it was subordinates with a
history of social subordination that were most
affected: 83% (10/12) of these monkeys exhibited
depressive behavior (Fig. 3). Thus, low social
status may increase the risk of depressive behavior
in subordinate female monkeys. Furthermore,

Fig. 2. Experimental design of Experiments I and II. Following

a 3-month quarantine in single cages, monkeys were placed in
their rst social groups (n 5 35/group) and linear social status
hierarchies were formed (initial social status). During social
group reorganization, group constituencies were changed such
that dominants were housed with dominants and subordinates
housed with subordinates. Liner social status hierarchies
reformed (current social status) in the second social groups.
Half the monkeys that were dominant in the rst social group
became subordinate in the second social group and half the
monkeys that were subordinate in the rst social group became
dominant in the second social group. The other half of the
dominant and subordinate monkeys maintained their initial
social status in the second social group [Shively et al., 1997,
2005].

Am. J. Primatol.

Adult Female Monkey Model of Depression / 5

subordinates that are most likely to be subordinate,

no matter their social environment, seem to be at
enhanced risk, suggesting that some inherent characteristic of these individuals also contributes to
their depression risk.
The increased risk of depressive behavior in
subordinates is not surprising, given the observation
that subordinate animals are stressed and that
stressful events often precipitate depression in
humans [Kessler, 1997]. Rates of depression are also
inversely related to socioeconomic status in
depressed humans [Adler & Rehkopf, 2008; Lorant
et al., 2003]. For persons of lower socioeconomic
status, this relationship may be mediated by inadequate personal resources, such as coping style, selfesteem, and locus of control, as well as the presence
of ongoing life stressors, poor social support, low
education level, and insufcient access to health care
[Adler & Rehkopf, 2008; Lorant et al., 2003]. These
observations suggest that characterization of individual differences in personality variables might
explain some of the variance in depressive behavior
in the nonhuman primate model. Although depressive behavior is more common in socially housed
subordinate females, many subordinates do not
display depressive behavior and some dominants
do. Like humans and other species, there seems to be
individual variation in resilience within the monkey
model. Taken together, these results suggest that
there is indeed a relationship between social subordination stress and depression. Yet, they are not
the same constructs, as socially subordinate and
behaviorally depressed monkeys are behaviorally
and neurobiologically distinguishable [Shively &
Willard, 2011; Shively et al., 2005].

Fig. 3. A history of social subordination increases risk of

depression. Social group reorganization resulted in four groups:
current subordinates that had a history of social subordination,
current subordinates with a history of social dominance,
current dominants with a history of social subordination, and
current dominants with a history of social dominance. Current
subordinates were more likely to exhibit behavioral depression
than current dominants (14/23 5 61% vs. 2/19 5 10%). However,
it was subordinates with a history of social subordination that
were most affected: 83% (10/12) of these monkeys exhibited
depressive behavior [Shively et al., 1997].

Am. J. Primatol.

6 / Willard and Shively

Housing Considerations
Experiment II (Fig. 2) included 36 monkeys and
was divided into three phases. During Phase 0, the
animals were housed in single cages for 12 months,
in Phase 1 the monkeys were housed in their rst
social groups for 12 months, and in Phase 2 the social
groups were systematically reorganized similar to
Experiment I, so that half of the previous subordinates became dominant and half the previous
dominants became subordinate. The monkeys lived
in these social groups for 22 months. Although
depression did occur in single cages, it did not
predict depression in later social groups, suggesting
that monkeys responding to social isolation with
depression are different from those that become
behaviorally depressed in social situations. This may
also indicate that the behavioral and physiological
antecedents of depression are not the same for
socially isolated and socially housed individuals,
and that housing condition should be carefully
considered when studying depressive behavior in
primates.
Predictors of Depression
In Experiment I, heart rates were recorded
while the animals were in single cages, just before
being placed into social groups. The monkeys that
later exhibited depression in social groups had
higher overnight heart rates while in single cages
compared with those that never exhibited depressive
behavior [Shively et al., 2002]. In Experiment II, the
monkeys lived in single cages for a year throughout
Phase 0. During this time, monkeys that subsequently became behaviorally depressed in Phase 1
social groups had decreased circulating insulin-like
growth factor-1 concentrations, lower activity levels,
decreased cortisol secretion in a corticotropin-releasing hormone (CRH) challenge test, and higher total

Adult Female Monkey Model of Depression / 533

plasma cholesterol (TPC) and the ratio of TPC:high

density lipoprotein (HDL) cholesterol concentrations. None of these characteristics were associated
with depressive behavior during Phase 0 [Shively
et al., 2005]. As such, monkeys that become
behaviorally depressed can be distinguished physiologically from those that do not become behaviorally
depressed.
CHARACTERISTICS OF BEHAVIORALLY
DEPRESSED FEMALE MONKEYS
Body Composition
Diagnostic criteria for human depression subtypes reect the heterogeneity of this prevalent
disorder by including such criteria as an increase or
a decrease in appetite, weight, and physical activity.
It seems that the relationship between body mass
and depression in humans is U-shaped, with the
leanest and heaviest patients being the most
depressed [de Wit et al., 2009]. Behaviorally
depressed monkeys have 17% lower body weight
and 20% lower body mass indices than nondepressed
female cynomolgus monkeys [Shively et al., 2005]. By
comparison, body weight is unrelated to social status
in female cynomolgus macaques [Kaplan et al., 2002,
2010].
Ovarian Function (Fig. 4A)
As mentioned previously, depressive disorders
occur in women at nearly twice the rate of men,
particularly during the reproductive years [Gorman,
2006; Kessler, 2003]. Depression is associated with
changes in reproductive system function [Spinelli,
2005], which suggest an important role for ovarian
steroids in this disease. Primary ovarian insufciency, characterized by 4 or more months of
amenorrhea, is associated with an increased lifetime
risk of major depression [Schmidt et al., 2011].

Fig. 4. Characteristics of behaviorally depressed monkeys. Compared with nondepressed monkeys, peak luteal phase progesterone levels
are lower in behaviorally depressed monkeys (A) suggesting impaired ovarian function; behaviorally depressed monkey have chronically
elevated heart rates around the clock (B) signifying autonomic dysfunction; and the ratio of the sum of circulating o-6 to o-3 fatty acids
is higher in behaviorally depressed monkeys (C) indicating a greater imbalance of fatty acids in the circulation. Po0.05 [Chilton et al.,
2011; Shively et al., 1997, 2002, 2005].

Am. J. Primatol.

Am. J. Primatol.

534 / Willard and Shively

However, the nature of the relationship between

depression and lack of ovarian cyclicity remains
to be elucidated. In addition, women are more likely
than men to respond to stress with depression
[Maciejewski et al., 2001], perhaps owing in part to
the ovarian steroid modulation of stress responsivity
in brain systems that mediate mood [Goldstein et al.,
2010; Shansky, 2009; Ter Horst et al., 2009].
Female cynomolgus monkeys have menstrual
cycles that are similar in length and hormonal
uctuations to those of women, and thus provide
an ideal model in which to study the relationship
between depression and ovarian steroids. Peak
progesterone values during the luteal phase of the
menstrual cycle reveal the quality of the menstrual
cycle, with high values indicating the occurrence of
ovulation and low values indicating impaired or no
ovulation. Socially subordinate females have low
peak progesterone concentrations [Kaplan et al.,
2010], which may be a factor in their increased risk
of depression. Low peak progesterone concentrations
are also characteristic of behaviorally depressed
monkeys, indicating impaired ovarian function in
these animals [Shively et al., 2002, 2005]. Cynomolgus monkeys with low luteal-phase progesterone
levels also produce less estrogen during the follicular
phase of the menstrual cycle [Adams et al., 1985],
suggesting that behaviorally depressed monkeys
with low peak progesterone levels are likely estrogen
decient. During a cycle in which peak progesterone
is low, menses may still occur. Thus, depressed
women may exhibit normal patterns of menstrual
cyclicity, but still be decient in ovarian steroids,
increasing their risk of comorbidities, such as
coronary heart disease (CHD) and osteoporosis.
Ovarian steroids have been shown to affect
antidepressant efcacy, which is evidenced by a
number of studies reporting age-related gender
differences in response to various pharmacotherapies [Bigos et al., 2009]. Women respond differently
than men to tricyclic antidepressants (TCAs) and
SSRIs [Kornstein et al., 2000; Young et al., 2009].
Similarly, premenopausal women respond differently than postmenopausal women to both TCAs
and SSRIs [Pae et al., 2009; Thase et al., 2005].
In postmenopausal depressed women, administration of hormone replacement therapy in addition to
an SSRI markedly improves antidepressant efcacy
to that of premenopausal women [Cohen et al., 2005;
Schneider et al., 1997; Thase et al., 2005]. These
results suggest an important role for ovarian function
in antidepressant efcacy, which is not surprising
because estrogen has been shown to modulate
serotonergic
neurotransmission
in
female
macaques [Bethea et al., 2002]. Given the ability to
track menses and determine ovarian steroid
concentrations over long periods of time, as well
as the observation that behaviorally depressed
monkeys have reduced serotonin-1A receptor

Am. J. Primatol.

Adult Female Monkey Model of Depression / 7

(5-HT1A) binding potential like depressed women

(described below) [Shively et al., 2006], the female
monkey model of depression would be a useful model
system for studying mechanisms of subclinical
ovarian dysfunction in depression and antidepressant
efcacy.
HypothalamicPituitaryAdrenal Axis
Function (Fig. 5)
The hypothalamicpituitaryadrenal (HPA) axis
is the central neuroendocrine stress response system. Briey, the HPA axis (Fig. 5A) is activated by
the secretion of CRH from the hypothalamus in
response to stress. This leads to the secretion of
adrenocorticotropin (ACTH) from the pituitary,
which in turn triggers the secretion of glucocorticoids from the adrenal cortex. Glucocorticoids then
activate glucocorticoid receptors in target tissues,
located throughout the periphery and the brain to
regulate key stress-related functions, including
immunity and neural plasticity. Glucocorticoids also
activate glucocorticoid receptors along the HPA axis,
resulting in feedback inhibition of subsequent CRH

Fig. 5. Hypothalamicpituitaryadrenal (HPA) axis. (A) The

core central nervous system stress response system consists of
the hypothalamus, pituitary, and adrenal cortex. In response to
stress, the hypothalamus secretes corticotropin-releasing hormone (CRH), which leads to the secretion of adrenocorticotropin
(ACTH) from the pituitary, and this in turn triggers the
secretion of cortisol from the adrenal cortex (black arrows).
Cortisol activates glucocorticoid receptors in neural and peripheral tissues to regulate key stress-related functions. Cortisol also
activates glucocorticoid receptors along the HPA axis (gray
arrows), resulting in feedback inhibition of subsequent CRH and
ACTH release, which reduces subsequent cortisol secretion.
CRH release is centrally mediated by the amygdala and the
hippocampus. Although activation of the amygdala promotes the
stress response by stimulating CRH release (white arrow),
activation of glucocorticoid receptors in the hippocampus as
part of the HPA negative feedback loop (gray arrows) results in
feedback inhibition of CRH and suppression of subsequent
cortisol production. (B) Behaviorally depressed monkeys secrete
less cortisol than nondepressed monkeys in a CRH challenge
test, as indicated by the average cortisol area under the curve
(AUC). (C) Behaviorally depressed monkeys are insensitive to
glucocorticoid negative feedback, with reduced suppresion of
cortisol in response to a dexamethasone suppression test (DST)
compared with nondepressed monkeys. Pr0.01 [Shively et al.,
1997, 2005].

Am. J. Primatol.

8 / Willard and Shively

and ACTH release. CRH release is centrally

mediated by the amygdala and the hippocampus,
two neural structures whose function and morphology are implicated in the pathophysiology of mood
and anxiety disorders. Although activation of the
amygdala via thalamic and cortical inputs promotes
continuation of the stress response by stimulating
CRH release, activation of glucocorticoid receptors in
the hippocampus as part of the HPA negative
feedback loop results in feedback inhibition of
CRH. Over time, chronic exposure to stress may
result in HPA dysfunction at various levels along the
axis, particularly in regions densely populated with
glucocorticoid receptors [Sapolsky et al., 1992].
These dysfunctions are well-established in a number
of stress-related disease states, including depression
[Pariante & Lightman, 2008]. HPA axis dysfunction
is not surprising in depression, given that stress
often precipitates depression [Kessler, 1997]. Indeed,
stressed socially subordinate females exhibit perturbations in HPA axis function [Kaplan et al., 2010;
Shively et al., 1997; Shively, 1998]. Various manifestations of HPA dysfunction are reported in depressed
humans, the degree to which seems to depend upon
depression subtype and severity [Pariante & Lightman,
2008].
Similar to depressed humans, HPA axis function
is altered in behaviorally depressed monkeys. Unlike
their nondepressed counterparts, behaviorally
depressed monkeys are insensitive to glucocorticoid
negative feedback, as determined in a DST, with a
greater cortisol response to dexamethasone as well as
an inability to suppress cortisol in response to
dexamethasone suppression (Fig. 5C) [Shively
et al., 1997, 2002]. Given that the HPA negative
feedback loop is centrally mediated by the hippocampus, these results indicate hippocampal dysfunction. Insensitivity to glucocorticoid negative feedback
promotes sustained high levels of cortisol, and
chronically elevated glucocorticoids in brain areas
rich in glucocorticoid receptors, such as the hippocampus, can lead to signicant remodeling and even
atrophy in those regions [McEwen, 2005; Sapolsky,
2000]. There are no differences between behaviorally
depressed and nondepressed monkeys in cortisol
levels in response to an ACTH challenge test [Shively
et al., 1997, 2005], indicating that adrenal HPA
function is normal in behaviorally depressed monkeys. However, behaviorally depressed monkeys
have lower cortisol responses to a CRH challenge
test than nondepressed monkeys (Fig. 5B), and
tended to have lower ACTH levels in response to
the CRH test, though this latter difference did not
reach signicance [Shively et al., 2005]. These results
suggest that behaviorally depressed monkeys may
have reduced sensitivity to CRH and are hypocortisolemic in response to CRH challenge. Similarly, in
depressed humans, exogenous CRH administration
leads to a blunted ACTH response, which has been

Am. J. Primatol.

Adult Female Monkey Model of Depression / 535

shown to be a function of endogenous CRH hypersecretion and the resulting desensitization of pituitary
CRH receptors [Gold et al., 1986; Heim et al., 2001;
Holsboer et al., 1986]. Whether these alterations in
HPA axis function predispose one to depression or
are the consequence of a depressed state remains to
be determined.
Autonomic Function (Fig. 4B)
On average, behaviorally depressed animals
maintain higher heart rates throughout the day
(measured over 24 hr) than nondepressed animals
[Shively et al., 2002, 2005], suggesting that autonomic function is perturbed in behaviorally
depressed monkeys. This may be indicative of
impairments in autonomic balance and is consistent
with the observation that behaviorally depressed
monkeys are less capable of suppressing HPA
responses to stress, as evidenced by the insensitivity
to glucocorticoid negative feedback. Increased heart
rate is also characteristic of depressed humans
[Carney et al., 2005; Lahmeyer & Bellur, 1987; Nabi
et al., 2010] and is associated with markedly
increased CHD risk compared with healthy controls
[Carney et al., 2002; Musselman et al., 2008; Rudisch
& Nemeroff, 2003].
Cardiovascular Disease Risk
Increased heart rate is also associated with
increased coronary artery atherosclerosis (CAA) in
monkeys [Kaplan et al., 1987]. Indeed, depression is
closely and positively correlated with CAA in the
female monkey model, and monkeys have nearly four
times the plaque area of their nondepressed counterparts [Shively et al., 2008, 2009]. Moreover, behaviorally depressed female monkeys are dyslipidemic,
with higher TPC and lower HDLC than nondepressed monkeys [Shively et al., 2005]. Although the
literature is mixed regarding the direction of
relationship between cholesterol and depression in
humans, some studies suggest that low HDLC may
be characteristic of long-term depression [Lehto
et al., 2008], and perhaps more prevalent in
depressed women than men [Chen et al., 2001;
Zhang et al., 2005]. Altered lipid metabolism has
been purported to be a mechanism at play in the
relationship between depression and CHD [Grippo &
Johnson, 2002]. Behaviorally depressed female monkeys also have altered circulating serum fatty acid
proles, with a higher ratio of o-6 to o-3 fatty acids
(Fig. 4C) and higher o-6 fatty acid levels [Chilton
et al., 2011]. An imbalance among o-3 and o-6 fatty
acids in humans is likewise associated with depression [Freeman et al., 2006; Wolfe et al., 2009] as well
as CHD risk [Iso et al., 2006]. Taken together,
these observations provide further evidence
supporting the validity of using this monkey model
for studies aimed at revealing the mechanisms

Am. J. Primatol.

536 / Willard and Shively

underlying the relationship between depression,

autonomic function, and cardiovascular disease.
Neurobiological Characteristics (Fig. 6)
Compared with rodents, the macaque brain is
far more similar to that of the human in terms of
nuclear organization, connectivity patterns, and
overall evolutionary conservation of functional areas
[Amaral & Lavenex, 2007; Hofman, 1989; Preuss,
1995; Wise, 2008]. As such, this monkey model of
depression is advantageous for furthering our understanding of the neurobiology of human depression.
We have observed that behaviorally depressed
monkeys exhibit similar neurobiological perturbations as those observed in depressed humans. Using
positron emission tomography, we found reduced
5-HT1A receptor binding potential in behaviorally
depressed compared with nondepressed animals
[Shively et al., 2006]. This reduction was observed
across 11 areas bilaterally, including the raphe
nucleus, amygdala, hippocampus, and anterior cingulate cortex, and seemed to be linearly related to
depression severity (Fig. 6A). Likewise, decreased
5-HT1A binding potential has been reported in
multiple brain areas in both currently depressed
and remitted humans, indicating that this characteristic reects a trait marker of depression [Bhagwagar
et al., 2004; Drevets et al., 1999; Sargent et al., 2000].
Reduced 5-HT1A binding potential was inversely
correlated with telemetered heart rate as well as
atherosclerotic extent in behaviorally depressed
monkeys [Shively et al., 2009], which provides
evidence to support a relationship between serotonergic decits and autonomic dysfunction in depression.

Fig. 6. Neurobiological characteristics of behaviorally depressed

monkeys. (A) Serotonin (5-HT) 1A receptor binding potential is
reduced in behaviorally depressed compared with nondepressed
monkeys. In this group of 18 monkeys from Experiment II, the
distribution of depressive behavior was divided into tertiles, such
that one group of monkeys never displayed behavioral depression (nondepressed, n 5 6), one group displayed behavioral
depression in 2574% of the months (mid-depressed, n 5 6),
and one group displayed behavioral depression in 75% or more of
the months (high-depressed, n 5 5). Distribution volume ratio 5 (ROI % injected dose/c.c. tissue)/(cerebellum % injected
dose/c.c. tissue). (B) Behaviorally depressed monkeys have
smaller anterior hippocampi than their nondepressed counterparts. Po0.05 [Shively et al., 2006; Willard et al., 2009].

Am. J. Primatol.

Adult Female Monkey Model of Depression / 9

As discussed previously, the hippocampus is

involved in regulating negative feedback in the
HPA axis and is implicated in the pathophysiology
of depression. Our observation of glucocorticoid
negative feedback insensitivity in behaviorally
depressed monkeys, as described above, provides
evidence that hippocampal function is altered in
depression. The hippocampus is prone to changes
induced by environmental stressors, given its plastic
nature and high density of glucocorticoid receptors
[Sapolsky, 2000]. Indeed, we have observed reduced
hippocampal volume both post mortem and in vivo in
behaviorally depressed adult female cynomolgus
monkeys [Willard et al., 2009, 2011]. The hippocampus is functionally heterogeneous, as the anterior
hippocampus plays a larger role in mood- and
emotion-related functioning than does the posterior
hippocampus, which seems to function more in
memory [Bannerman et al., 2004]. Most human
volumetric studies do not detail separate investigations of the anterior and posterior hippocampus or
control for hormone levels in female participants,
even though volume changes in the anterior hippocampus are associated with menstrual cycle phase in
women [Protopopescu et al., 2008]. We initially
reported a reduction in anterior hippocampal volume
measured post mortem in behaviorally depressed
compared with nondepressed adult female monkeys
that were controlled for menstrual cycle phase and
with no prior antidepressant exposure (Fig. 6B)
[Willard et al., 2009]. To conrm these ndings in
vivo and examine whether a lack of ovarian steroids
affects the relationship between depression and
hippocampal volume, we then measured whole,
anterior, and posterior hippocampal volumes in
antidepressant-naive, surgically postmenopausal
animals using MRI. Global reductions in hippocampal volume were observed, further supporting an
association between hippocampal morphometric
alterations and depressive behavior in adult female
cynomolgus macaques [Willard et al., 2011]. We also
observed reduced bilateral whole hippocampal
volume in behaviorally depressed monkeys, which
is consistent with a recent meta-analysis conrming
signicant bilateral reductions in depressed humans
[Koolschijn et al., 2009]. Given that we observed
posterior hippocampal reductions in the ovariectomized animals, but not in our previous post mortem
study of intact females, it is possible that endogenous
female hormones play a protective role against
hippocampal size reduction in depression, particularly in the posterior hippocampus, and may be
relatively more involved in memory processing. The
observation of smaller hippocampal volume in
behaviorally depressed monkeys is compelling,
because many of the characteristics that differentially affect hippocampal volume in human studies
are controlled in this primate model. Furthermore,
these ndings suggest that the neurobiological

Am. J. Primatol.

10 / Willard and Shively

substrates of depression are similar between humans

and nonhuman primates.

Behavioral Characteristics
Given the extensive differences in neural structure and function between behaviorally depressed
and nondepressed monkeys, perhaps it is not
surprising that behavioral differences extend well
beyond the occurrence of depressive behavior. Behaviorally depressed monkeys spend more time in body
contact with conspecics and less time alone than
their nondepressed counterparts [Shively et al.,
2005]. In addition, behaviorally depressed animals
are far less active than nondepressed animals, as
indicated by a signicant reduction in percent time
locomoting [Shively et al., 2005] and an inverse
correlation between depression and percent time
locomoting [Shively et al., 2008]. The relationship
between body composition, activity levels, and time
spent in body contact is intriguing. Although human
depression is often characterized by social withdrawal, the majority of depressive behavior in these
animals was recorded while they were in physical
contact with others. On the surface this might
appear contradictory; however, physical contact
may provide valuable resources for behaviorally
depressed monkeys. As mentioned above, behaviorally depressed monkeys are relatively lean and less
active than nondepressed monkeys, and thus may
benet from the warmth generated by close physical
proximity to others. In addition, behaviorally
depressed monkeys are, by denition, relatively
unresponsive to environmental stimuli, and as a
consequence might use the startle and orienting
behavior of another monkey close by to warn of
impending environmental threats. In this way,
physical contact may be an adaptive, purposeful
behavior for behaviorally depressed animals.

Mortality
The distribution of percent time spent behaviorally depressed was divided into quartiles (n 5 9 per
quartile) and mortality over the course of Experiment II being examined [Shively et al., 2005]. Zero to
two monkeys died before the end of the study in each
of the three lowest quartiles of behavioral depression. However, 5/9 or 56% of the animals died out of
the most behaviorally depressed quartile. Although
complete gross and histological pathology evaluations were done, in some cases although physiological
abnormalities were identied, causes of death were
not clear. Increased mortality (not including suicide)
is also associated with major depression in humans
[Kouzis et al., 1995; Michaud et al., 2001].

Am. J. Primatol.

Adult Female Monkey Model of Depression / 537

SYNTHESIS AND FUTURE DIRECTIONS

Depression is a prevalent, costly, and poorly
understood disease. Advancement in understanding
and treating depression requires the development of
new animal models. The adult female monkey model
of depression has many characteristics in common
with the physiology, neurobiology, and behavior of
human depression. Like depressed human beings,
behaviorally depressed female cynomolgus monkeys
have reduced body mass, HPA axis disturbances,
autonomic dysfunction, increased cardiovascular
disease risk, reduced hippocampal volume, altered
serotonergic function, decreased activity levels, and
increased mortality (Table I). They also have low
ovarian steroid concentrations, even though they
continue to have menstrual cycles. Although this
type of ovarian dysfunction has not been reported in
depressed women and is difcult to identify, it may
be key to understanding the high prevalence of
depression in women.
Different social environmental challenges, such
as isolation vs. subordination, may elicit the depressive response in some animals and not others,
suggesting individual differences in sensitivity to
specic environmental stressors. Depressive behavior is more common in socially subordinate monkeys, but the behavior and neurobiology of
subordinates is different than that of behaviorally
depressed monkeys, suggesting that stress and
depression are related yet separate constructs
[Shively & Willard, 2011]. Owing to the similarities
in central nervous system structure and function
between cynomolgus monkeys and humans, the
ability to differentiate between stressed and
depressed populations, and that the depressive
behavior is naturally occurring, the nonhuman
primate model of depression represents a new
TABLE I. Characteristics of Depressed Adult Female
Cynomolgus Monkeysa

 Low body mass indices

 Impaired ovarian function
 Normal cortisol to ACTH
 Poor suppression in DST
 Hypocortisolemic to CRH
 High heart rate
 Exacerbated CAA
 Dyslipidemia
 High ratio of o-6 to o-3 fatty acids
 Decreased 5-HT1A binding potential
 Smaller hippocampus
 More time in body contact
 Less time alone
 Less active
 Increased mortality
a

Compared with nondepressed adult female cynomolgus monkeys. ACTH,

adrenocorticotropin hormone; DST, dexamethasone suppressions test;
CRH, corticotropin-releasing hormone; CAA, coronary artery atherosclerosis; 5-HT1A, serotonin-1A receptor.

Am. J. Primatol.

538 / Willard and Shively

animal model of depression with great promise for

moving the eld forward.
The known characteristics of behaviorally
depressed monkeys point to other features of
depression that have not yet been investigated in
this model. For example, behaviorally depressed
monkeys have high overnight heart rates, HPA axis
dysfunction, and reduced ovarian function, all
indicators of perturbed endogenous rhythms. Thus,
these animals may also have disturbed sleep patterns
and changes in other circadian rhythms similar to
depressed humans. Furthermore, low body weight,
body mass, and activity levels indicate disruptions in
energy homeostasis, which could be accompanied by
abnormal eating behavior, a diagnostic criterion in
human depression. HPA axis dysfunction, increased
CAA, and altered circulating fatty acid levels in
behaviorally depressed monkeys suggest chronic
inammation, which could be determined in studies
of neural and peripheral levels of inammatory
cytokines. Given the neurobiological similarities
reported in behaviorally depressed monkeys compared with humans, it is likely that in-depth
investigations of the cellular and molecular mechanisms of hippocampal volume reduction in the post
mortem monkey brain would provide substantial
insight into the pathophysiology of human depression. Likewise, macaques have an elaborate cingulate
cortex, more like the human than the rodent brain.
Based on imaging studies, the cingulate cortex
appears to be important in human depression, and
thus the monkey model of depression may be used to
further understand the molecular and cellular
processes characteristic of depression in this region,
leading to new therapeutic targets.
It is reasonable to consider whether monkey
depression is similar to a particular type (or types) of
human depression. In humans, a number of depressive disorders and their subtypes are diagnosed,
based on the presentation of a wide variety of
symptoms [American Psychiatric Association,
2000]. The symptom diagnosis pool is almost twice
that of the number of symptoms that must be
present for a diagnosis, and many of the symptoms
may be present in one direction or the other, such as
increased or decreased appetite, weight, sleep, and
motor activity. However, given this heterogeneity in
symptomology of human depression, what we know
of depression in monkeys to date suggests some
similarity to major depression with melancholic
features. Depressed patients with melancholic features have major depression and anhedonia, and may
have sleep disturbances, weight loss, and psychomotor retardation or agitation. One of the most
consistent clinical features of melancholic depressed
patients is a particular pattern of HPA dysfunction
[Carroll et al., 1981; Gold & Chrousos, 2002].
Melancholic depressives have an insensitivity to
glucocorticoid negative feedback and hypersecrete

Am. J. Primatol.

Adult Female Monkey Model of Depression / 11

CRH, resulting in a blunted ACTH response to

exogenous CRH [Gold & Chrousos, 2002]. This
pattern of HPA dysfunction is similar to that of
behaviorally depressed female monkeys. Given that
behaviorally depressed monkeys also have low body
weight, reduced activity, and are generally unresponsive to environmental stimuli, the female monkey model of depression may approximate depression
with melancholic features. Moreover, slumped body
posture and downcast gaze like that observed in
behaviorally depressed monkeys are common manifestations of psychomotor retardation in melancholic
depressed humans [Sobin et al., 1998]. Further
studies of sleep, anxiety, eating behaviors, and other
features of melancholic depression are warranted in
the monkey model.
Another approach to understanding the heterogeneity in depression symptomology is to consider
that depressed persons experience a range of life
events, and that these varied experiences might
differentially contribute to specic depression characteristics. HPA axis function is thought to change
over time and vary among individuals with regard to
number and duration of depressive episodes, as well
as exposure to acute vs. prolonged stress [Gold &
Chrousos, 1985; Parker et al., 2003]. For example,
adverse early life experiences are associated with
heightened vulnerability to stress and increased risk
of mood disorders, effects that are mediated by HPA
axis function [Heim et al., 2001; McCauley et al.,
1997]. Women with a history of child abuse without
adult depression have increased ACTH response to
CRH [Heim et al., 2001]. Alternatively, depressed
women with a history of child abuse have reduced
cortisol levels and lower ACTH in response to a CRH
challenge, indicating pituitary insensitivity to CRH
[Heim et al., 2001]. These results suggest that
adverse early life events contribute to a greater
HPA sensitivity to stress and increased vulnerability
for depression in adulthood. The pattern exhibited
by depressed women with a history of abuse is
similar to that displayed by behaviorally depressed
monkeys, although the reduced ACTH response to
CRH challenge did not reach signicance in behaviorally depressed monkeys. It is interesting to note
that social subordination stress exposure begins
early in life for subordinate monkeys, because
cynomolgus macaques assume the social status of
their mothers at birth. Given that depressive
behavior is more common in subordinate females,
perhaps early life adversity results in an HPA axis
sensitivity that predisposes subordinates to
increased vulnerability for depressive behavior.
In conclusion, this nonhuman primate model of
depression distinguishes between stressed and
depressed populations, is sensitive to social environmental characteristics, and includes individuals
at enhanced risk of depression. Unlike rodent
models, the stress of social subordination is not an

Am. J. Primatol.

12 / Willard and Shively

experimental manipulation, but instead a normal

facet of macaque life that exists in both captivity and
the wild. As such, the etiology of behavioral depression in this model may be more relevant to the
etiology of human depression than the experimentally manipulated stress of rodent models. Moreover,
behaviorally depressed monkeys exhibit numerous
multisystem physiological, neurobiological, and
behavioral characteristics like that of depressed
humans, many of which cannot be evaluated in
rodent models. Although the female monkey model
seems to have good face and construct validity, we
are currently investigating the models predictive
validity through studies of commonly prescribed
antidepressant treatments on depressive behavior.
Though the female monkey model has been under
study for 20 years, there are a number of characteristics of the model yet to be explored that may enrich
our understanding of human depression and lead to
new avenues for therapeutic intervention.
ACKNOWLEDGMENTS
This work was supported by RO1 HL39789, RO1
HL87103; RO1 MH56881, R21 MH086731, and the
John D. and Catherine T. MacArthur Foundation.
All procedures involving monkeys were conducted in
accordance with state and federal laws, standards of
the Department of Health and Human Services and
Institutional Animal Care and Use Committee
guidelines, and the American Society of Primatologists Principles for the Ethical Treatment of Nonhuman Primates.
REFERENCES
Adams MR, Kaplan JR, Clarkson TB, Koritnik DR. 1985.
Ovariectomy, social status, and atherosclerosis in cynomolgus monkeys. Arteriosclerosis 5:192200.
Adler NE, Rehkopf DH. 2008. U.S. disparities in health:
descriptions, causes, and mechanisms. Annual Review of
Public Health 29:235252.
Amaral D, Lavenex P. 2007. Hippocampal neuroanatomy.
In: Anderson P, Morris R, Amaral D, Bliss T, OKeefe J, editors.
The hippocampus book. New York: Oxford UP. p 37114.
American Psychiatric Association. 2000. Diagnostic and
statistical manual of mental disorders, fourth edition, text
revision. Washington DC: APA.
Bannerman DM, Rawlins JN, McHugh SB, Deacon RM, Yee
BK, Bast T, Zhang WN, Pothuizen HH, Feldon J. 2004.
Regional dissociations within the hippocampusmemory
and anxiety. Neuroscience and Biobehavioral Reviews
28:273283.
Berton O, Nestler EJ. 2006. New approaches to antidepressant
drug discovery: beyond monoamines. Nature Reviews
Neuroscience 7:137151.
Bethea CL, Lu NZ, Gundlah C, Streicher JM. 2002. Diverse
actions of ovarian steroids in the serotonin neural system.
Frontiers in Neuroendocrinology 23:41100.
Bhagwagar Z, Rabiner EA, Sargent PA, Grasby PM,
Cowen PJ. 2004. Persistent reduction in brain serotonin1A
receptor binding in recovered depressed men measured by
positron emission tomography with [11C]WAY-100635.
Molecular Psychiatry 9:386392.

Am. J. Primatol.

Adult Female Monkey Model of Depression / 539

Bigos KL, Pollock BG, Stankevich BA, Bies RR. 2009. Sex
differences in the pharmacokinetics and pharmacodynamics
of antidepressants: an updated review. Gender Medicine
6:522543.
Carney RM, Freedland KE, Miller GE, Jaffe AS. 2002.
Depression as a risk factor for cardiac mortality and
morbidity: a review of potential mechanisms. Journal of
Psychosomatic Research 53:897902.
Carney RM, Freedland KE, Veith RC. 2005. Depression, the
autonomic nervous system, and coronary heart disease.
Psychosomatic Medicine 67:S29S33.
Carroll BJ, Feinberg M, Greden JF, Tarika J, Albala AA,
Haskett RF, James NM, Kronfol Z, Lohr N, Steiner M, de
Vigne JP, Young E. 1981. A specic laboratory test for the
diagnosis of melancholia. Archives of General Psychiatry
38:1522.
Charney DS. 2004. Psychobiological mechanisms of resilience
and vulnerability: implications for successful adaptation to
extreme stress. American Journal of Psychiatry 161:195216.
Chen CC, Lu FH, Wu JS, Chang CJ. 2001. Correlation between
serum lipid concentrations and psychological distress.
Psychiatry Research 102:153162.
Chilton FH, Lee TC, Willard SL, Ivester P, Sergeant S,
Register TC, Shively CA. 2011. Depression and altered
serum lipids in cynomolgus monkeys consuming a Western
diet. Physiology and Behavior 2011 Jan 20 [Epub ahead of
print]. PMID: 21256145 [PubMed as supplied by publisher] DOI:10.1016/j.physbeh.2011.01.013
Cohen LS, Soares CN, Joffe H. 2005. Diagnosis and management of mood disorders during the menopausal transition.
American Journal of Medicine 118:9397.
Cryan JF, Markou A, Lucki I. 2002. Assessing antidepressant
activity in rodents: recent developments and future needs.
Trends in Pharmacological Science 23:238245.
Cuijpers P, Schoevers RA. 2004. Increased mortality in
depressive disorders: a review. Current Psychiatry Reports
6:430437.
Dalla C, Pitychoutis PM, Kokras N, Papadopoulou-Daifoti Z.
2010. Sex differences in animal models of depression and
antidepressant response. Basic and Clinical Pharmacology
and Toxicology 106:226233.
de Wit LM, van Straten A, van Herten M, Penninx BW,
Cuijpers P. 2009. Depression and body mass index, a
u-shaped association. BMC Public Health 13:9:14. PMCID:
PMC2631467 DOI: 10.1186/1471-2458-9-14.
Drevets WC, Frank E, Price JC, Kupfer DJ, Holt D, Greer PJ,
Huang Y, Gautier C, Mathis C. 1999. PET imaging of
serotonin 1A receptor binding in depression. Biological
Psychiatry 46:13751387.
Duman CH. 2010. Models of depression. Vitamins and
Hormones 82:121. DOI:10.1016/S0083-6729(10)82001-1
Freeman MP, Hibbeln JR, Wisner KL, Davis JM, Mischoulon
D, Peet M, Keck Jr PE, Marangell LB, Richardson AJ, Lake
J, Stoll AL. 2006. Omega-3 fatty acids: evidence basis for
treatment and future research in psychiatry. Journal of
Clinical Psychiatry 67:19541967.
Gold PW, Chrousos GP. 1985. Clinical studies with corticotropin releasing factor: implications for the diagnosis and
pathophysiology of depression, Cushings disease, and adrenal
insufciency. Psychoneuroendocrinology 10:401419.
Gold PW, Chrousos GP. 2002. Organization of the stress
system and its dysregulation in melancholic and atypical
depression: high vs low CRH/NE states. Molecular Psychiatry 7:254275.
Gold PW, Loriaux DL, Roy A, Kling MA, Calabrese JR,
Kellner CH, Nieman LK, Post RM, Pickar D, Gallucci W,
Avgerinos P, Paul S, Oldeld EH, Cutler Jr GB,
Chrousos GP. 1986. Responses to corticotropin-releasing
hormone in the hypercortisolism of depression and Cushings disease. New England Journal of Medicine
314:13291335.

Am. J. Primatol.

540 / Willard and Shively

Goldstein JM, Jerram M, Abbs B, Whiteld-Gabrieli S,

Makris N. 2010. Sex differences in stress response circuitry
activation dependent on female hormonal cycle. Journal of
Neuroscience 30:431438.
Gorman JM. 2006. Gender differences in depression and
response to psychotropic medication. Gender Medicine
3:93109.
Grippo AJ, Johnson AK. 2002. Biological mechanisms in the
relationship between depression and heart disease. Neuroscience and Biobehavioral Reviews 26:941962.
Harlow HF, Suomi SJ. 1974. Induced depression in monkeys.
Behavioral Biology 12:273296.
Heim C, Newport DJ, Bonsall R, Miller AH, Nemeroff CB.
2001. Altered pituitary-adrenal axis responses to provocative challenge tests in adult survivors of childhood abuse.
American Journal of Psychiatry 158:575581.
Hofman MA. 1989. On the evolution and geometry of the brain
in mammals. Progress in Neurobiology 32:137158.
Holsboer F, Gerken A, Von Bardeleben U, Grimm W, Beyer H,
Mu
ller OA, Stalla GK. 1986. Human corticotropin releasing
hormone (CRH) in patients with depression, alcoholism and
panic disorder. Biological Psychiatry 21:601611.
Iso H, Kobayashi M, Ishihara J, Sasaki S, Okada K, Kita Y,
Kokubo Y, Tsugane S, JPHC Study Group. 2006. Intake of
sh and n3 fatty acids and risk of coronary heart disease
among Japanese: the Japan Public Health Center-Based
(JPHC) Study Cohort I. Circulation 113:195202.
Jackson JL, DeZee K, Berbano E. 2004. Can treating
depression improve disease outcomes? Annals of Internal
Medicine 140:10541056.
Kaplan JR, Adams MR, Koritnik DR, Rose JC, Manuck SB.
1986. Adrenal responsiveness and social status in intact and
ovariectomized Macaca fascicularis. American Journal of
Primatology 11:181193.
Kaplan JR, Manuck SB, Clarkson TB. 1987. The inuence of
heart rate on coronary artery atherosclerosis. Journal of
Cardiovascular Pharmacology 10:S100S103.
Kaplan JR, Manuck SB, Fontenot MB, Mann JJ. 2002. Central
nervous system monoamine correlates of social dominance
in cynomolgus monkeys (Macaca fascicularis). Neuropsychopharmacology 26:431443.
Kaplan JR, Chen H, Appt SE, Lees CJ, Franke AA, Berga SL,
Wilson ME, Manuck SB, Clarkson TB. 2010. Impairment of
ovarian function and associated health-related abnormalities are attributable to low social status in premenopausal
monkeys and not mitigated by a high-isoavone soy diet.
Human Reproduction 25:30833094.
Kessler RC. 1997. The effects of stressful life events on
depression. Annual Review of Psychology 48:191214.
Kessler RC. 2003. Epidemiology of women and depression.
Journal of Affective Disorders 74:513.
Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR,
Walters EE. 2005. Lifetime prevalence and age-of-onset
distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry
62:593602.
Koolschijn PC, van Haren NE, Lensvelt-Mulders GJ, Hulshoff
Pol HE, Kahn RS. 2009. Brain volume abnormalities in
major depressive disorder: a meta-analysis of magnetic
resonance imaging studies. Human Brain Mapping
30:37193735.
Kornstein SG, Schatzberg AF, Thase ME, Yonkers KA,
McCullough JP, Keitner GI, Gelenberg AJ, Davis SM,
Harrison WM, Keller MB. 2000. Gender differences in
treatment response to sertraline versus imipramine in
chronic depression. American Journal of Psychiatry
157:14451452.
Kouzis A, Eaton WW, Leaf PJ. 1995. Psychopathology and
mortality in the general population. Social Psychiatry and
Psychiatric Epidemiology 30:165170.

Am. J. Primatol.

Adult Female Monkey Model of Depression / 13

Krishnan KR. 2003. Comorbidity and depression treatment.

Biological Psychiatry 53:701706.
Krishnan V, Nestler EJ. 2010. Linking molecules to mood: new
insight into the biology of depression. American Journal of
Psychiatry 167:13051320.
Krishnan V, Han MH, Graham DL, Berton O, Renthal W,
Russo SJ, Laplant Q, Graham A, Lutter M, Lagace DC,
Ghose S, Reister R, Tannous P, Green TA, Neve RL,
Chakravarty S, Kumar A, Eisch AJ, Self DW, Lee FS,
Tamminga CA, Cooper DC, Gershenfeld HK, Nestler EJ.
2007. Molecular adaptations underlying susceptibility and
resistance to social defeat in brain reward regions. Cell
131:391404.
Lahmeyer HW, Bellur SN. 1987. Cardiac regulation and
depression. Journal of Psychiatric Research 21:16.
Lehto SM, Hintikka J, Niskanen L, Tolmunen T, Koivumaaki H. 2008. Low HDL
Honkanen H, Honkalampi K, Viinama
cholesterol associates with major depression in a sample with a
7-year history of depressive symptoms. Progress in Neuropsychopharmacology and Biological Psychiatry 32:15571561.
Liu RT, Alloy LB. 2010. Stress generation in depression:
a systematic review of the empirical literature and recommendations for future study. Clinical Psychology Review
30:582593.
Lorant V, Delie`ge D, Eaton W, Robert A, Philippot P, Ansseau M.
2003. Socioeconomic inequalities in depression: a metaanalysis. American Journal of Epidemiology 157:98112.
Maciejewski PK, Prigerson HG, Mazure CM. 2001. Sex
differences in event-related risk for major depression.
Psychological Medicine 31:593604.
Magness CL, Fellin PC, Thomas MJ, Korth MJ, Agy MB,
Proll SC, Fitzgibbon M, Scherer CA, Miner DG, Katze MG,
Iadonato SP. 2005. Analysis of the Macaca mulatta
transcriptome and the sequence divergence between Macaca
and human. Genome Biol 6:R60.
Maier SF, Watkins LR. 2005. Stressor controllability and
learned helplessness: the roles of the dorsal raphe nucleus,
serotonin, and corticotropin-releasing factor. Neuroscience
and Biobehavioral Reviews 29:829841.
Maller JJ, Daskalakis ZJ, Fitzgerald PB. 2007. Hippocampal
volumetrics in depression: the importance of the posterior
tail. Hippocampus 17:10231027.
Marcus SM, Young EA, Kerber KB, Kornstein S, Farabaugh
AH, Mitchell J, Wisniewski SR, Balasubramani GK, Trivedi
MH, Rush AJ. 2005. Gender differences in depression:
ndings from the STARD study. Journal of Affective
Disorders 87:141150.
McCauley J, Kern DE, Kolodner K, Dill L, Schroeder AF,
DeChant HK, Ryden J, Derogatis LR, Bass EB. 1997.
Clinical characteristics of women with a history of childhood
abuse: unhealed wounds. Journal of the American Medical
Association 277:13621368.
McEwen BS. 2005. Glucocorticoids, depression, and mood
disorders: structural remodeling in the brain. Metabolism
54:2023.
McKinney WT. 1984. Animal models of depression: an overview. Psychiatric Developments 2:7796.
Michaud CM, Murray CJ, Bloom BR. 2001. Burden of
diseaseimplications for future research. Journal of the
American Medical Association 285:535539.
Musselman DL, Cowles M, McDonald W, Nemeroff CB. 2008.
Effects of mood and anxiety disorders on the cardiovascular
system: implications for treatment. In: Fuster V,
Alexander RW, ORourke RA, Roberts R, King SB,
Wellens HJJ, editors. Hursts the heart, 12th edition. New
York: McGraw Hill. p 21692187.
ki M, Empana JP, Sabia S, Britton A,
Nabi H, Kivima
Marmot MG, Shipley MJ, Singh-Manoux A. 2010. Combined
effects of depressive symptoms and resting heart rate on
mortality: the Whitehall II prospective cohort study.
Journal of Clinical Psychiatry 2010 Dec 14 [Epub ahead of

Am. J. Primatol.

14 / Willard and Shively

print]. PMID: 21208592 [PubMed as supplied by publisher]

DOI:10.4088/JCP.09m05901blu
Nemeroff CB. 2007. The burden of severe depression: a review
of diagnostic challenges and treatment alternatives. Journal
of Psychiatric Research 41:189206.
Nestler EJ, Hyman SE. 2010. Animal models of neuropsychiatric disorders. Nature Neuroscience 13:11611169.
Pae CU, Mandelli L, Kim TS, Han C, Masand PS, Marks DM,
Patkar AA, Steffens DC, De Ronchi D, Serretti A. 2009.
Effectiveness of antidepressant treatments in pre-menopausal versus post-menopausal women: a pilot study on
differential effects of sex hormones on antidepressant
effects. Biomedicine and Pharmacotherapy 63:228235.
Palanza P, Gioiosa L, Parmigiani S. 2001. Social stress in mice:
gender differences and effects of estrous cycle and social
dominance. Physiology and Behavior 73:411420.
Pariante CM, Lightman SL. 2008. The HPA axis in major
depression: classical theories and new developments.
Trends in Neurosciences 31:464468.
Parker KJ, Schatzberg AF, Lyons DM. 2003. Neuroendocrine
aspects of hypercortisolism in major depression. Hormones
and Behavior 43:6066.
Perera TD, Dwork AJ, Keegan KA, Thirumangalakudi L,
Lipira CM, Joyce N, Lange C, Higley JD, Rosoklija G,
Hen R, Sackeim HA, Coplan JD. 2011. Necessity of
hippocampal neurogenesis for the therapeutic action of
antidepressants in adult nonhuman primates. PLoS ONE
6:e17600.
Preuss TM. 1995. Do rats have a prefrontal cortex? The
RoseWoolseyAkert program reconsidered. Journal of
Cognitive Neuroscience 7:124.
Protopopescu X, Butler T, Pan H, Root J, Altemus M,
Polanecsky M, McEwen B, Silbersweig D, Stern E. 2008.
Hippocampal structural changes across the menstrual cycle.
Hippocampus 18:985988.
Rudisch B, Nemeroff CB. 2003. Epidemiology of comorbid
coronary artery disease and depression. Biological Psychiatry 54:227240.
Rush AJ. 2007. The varied clinical presentations of major
depressive disorder. Journal of Clinical Psychiatry 68:410.
Sapolsky RM. 2000. Glucocorticoids and hippocampal atrophy
in neuropsychiatric disorders. Archives of General Psychiatry 57:925935.
Sapolsky RM, Krey LC, McEwen BS. 1992. The neuroendocrinology of stress and aging: the glucocorticoid cascade
hypothesis. Endocrine Reviews 7:284301.
Sargent PA, Kjaer KH, Bench CJ, Rabiner EA, Messa C,
Meyer J, Gunn RN, Grasby PM, Cowen PJ. 2000. Brain
serotonin1A receptor binding measured by positron emission tomography with [11C]WAY-100635: effects of depression and antidepressant treatment. Archives of General
Psychiatry 57:174180.
Schmidt MV, Scharf SH, Liebl C, Harbich D, Mayer B,
Holsboer F, Muller MB. 2010. A novel chronic social stress
paradigm in female mice. Hormones and Behavior
57:415420.
Schmidt PJ, Luff JA, Haq NA, Vanderhoof VH, Koziol DE,
Calis KA, Rubinow DR, Nelson LM. 2011. Depression in
women with spontaneous 46, XX primary ovarian insufciency. Journal of Clinical Endocrinology and Metabolism
96:E278E287.
Schneider LS, Small GW, Hamilton SH, Bystritsky A, Nemeroff CB, Meyers BS. 1997. Estrogen replacement and
response to uoxetine in a multicenter geriatric depression
trial. Fluoxetine Collaborative Study Group. American
Journal of Geriatric Psychiatry 5:97106.
Shansky RM. 2009. Estrogen, stress and the brain:
progress toward unraveling gender discrepancies in major
depressive disorder. Expert Review of Neurotherapeutics
9:967973.

Am. J. Primatol.

Adult Female Monkey Model of Depression / 541

Shively CA. 1998. Social subordination stress, behavior, and

central monoaminergic function in female cynomolgus
monkeys. Biological Psychiatry 44:882891.
Shively CA, Clarkson TB. 2009. The unique value of primate
models in translational research. American Journal of
Primatology 71:715721.
Shively CA, Kaplan JR. 1991. Stability of social status
rankings of female cynomolgus monkeys, of varying reproductive condition, in different social groups. American
Journal of Primatology 23:239245.
Shively CA, Willard SL. 2011. Behavioral and neurobiological
characteristics of social stress versus depression in nonhuman primates. Experimental Neurology Sep 24. [Epub
ahead of print].
Shively CA, Laber-Laird K, Anton RF. 1997. Behavior and
physiology of social stress and depression in female
cynomolgus monkeys. Biological Psychiatry 41:871882.
Shively CA, Williams JK, Laber-Laird K, Anton RF. 2002.
Depression and coronary artery atherosclerosis and reactivity in female cynomolgus monkeys. Psychosomatic Medicine
64:699706.
Shively CA, Register TC, Friedman DP, Morgan TM, Thompson J, Lanier T. 2005. Social stress-associated depression in
adult female cynomolgus monkeys (Macaca fascicularis).
Biological Psychology 69:6784.
Shively CA, Friedman DP, Gage HD, Bounds MC, BrownProctor C, Blair JB, Henderson JA, Smith MA, Buchheimer
N. 2006. Behavioral depression and positron emission
tomography-determined serotonin 1A receptor binding
potential in cynomolgus monkeys. Archives of General
Psychiatry 63:396403.
Shively CA, Register TC, Adams MR, Golden DL, Willard SL,
Clarkson TB. 2008. Depressive behavior and coronary
artery atherogenesis in adult female cynomolgus monkeys.
Psychosomatic Medicine 70:637645.
Shively CA, Musselman DL, Willard SL. 2009. Stress, depression, and coronary artery disease: modeling comorbidity in
female primates. Neuroscience and Biobehavioral Reviews
33:133144.
Sobin C, Mayer L, Endicott J. 1998. The motor agitation and
retardation scale: a scale for the assessment of motor
abnormalities in depressed patients. Journal of Neuropsychiatry and Clinical Neurosciences 10:8592.
Spinelli MG. 2005. Neuroendocrine effects on mood. Reviews
in Endocrine and Metabolic Disorders 6:109115.
Suomi SJ, Eisele CD, Grady SA, Harlow HF. 1975. Depressive
behavior in adult monkeys following separation from family
environment. Journal of Abnormal Psychology 84:576578.
Ter Horst GJ, Wichmann R, Gerrits M, Westenbroek C, Lin Y.
2009. Sex differences in stress responses: focus on ovarian
hormones. Physiology and Behavior 97:239249.
Thase ME, Entsuah R, Cantillon M, Kornstein SG. 2005.
Relative antidepressant efcacy of venlafaxine and SSRIs:
sex-age interactions. Journal of Womens Health
14:609616.
Trainor BC, Pride MC, Villalon Landeros R, Knoblauch NW,
Takahashi EY, Silva AL, Crean KK. 2011. Sex differences in
social interaction behavior following social defeat stress in
the monogamous california mouse (Peromyscus californicus).
PLoS ONE 6:e17405.
Van Bokhoven P, Oomen CA, Hoogendijk WJ, Smit AB,
Lucassen PJ, Spijker S. 2011. Reduction in hippocampal
neurogenesis after social defeat is long-lasting and responsive to late antidepressant treatment. European Journal of
Neuroscience 33:18331840.
Wang PS, Beck A, Berglund P, Leutzinger JA, Pronk N,
Richling D, Schenk TW, Simon G, Stang P, Ustun TB,
Kessler RC. 2003. Chronic medical conditions and work
performance in the health and work performance questionnaire calibration surveys. Journal of Occupational and
Environmental Medicine 45:13031311.

Am. J. Primatol.

542 / Willard and Shively

Willard SL, Friedman DP, Henkel CK, Shively CA. 2009.

Anterior hippocampal volume is reduced in behaviorally
depressed female cynomolgus macaques. Psychoneuroendocrinology 34:14691475.
Willard SL, Daunais JB, Cline JM, Shively CA. 2011.
Hippocampal volume in postmenopausal cynomolgus macaques with behavioral depression. Menopause 18:582586.
Willner P. 1984. The validity of animal models of depression.
Psychopharmacology (Berl) 83:116.
Willner P. 2005. Chronic mild stress (CMS) revisited:
consistency and behavioural-neurobiological concordance
in the effects of CMS. Neuropsychobiology 52:90110.
Wise SP. 2008. Forward frontal elds: phylogeny and fundamental function. Trends in Neurosciences 31:599608.
Wolfe AR, Ogbonna EM, Lim S, Li Y, Zhang J. 2009. Dietary
linoleic and oleic fatty acids in relation to severe depressed

Am. J. Primatol.

Adult Female Monkey Model of Depression / 15

mood: 10 years follow-up of a national cohort. Progress in

Neuropsychopharmacology and Biological Psychiatry
33:972977.
World Health Organization. 2001. The world health report
2001. Mental health: new understanding, new hope.
Geneva: World Health Organization.
Young EA, Kornstein SG, Marcus SM, Harvey AT, Warden D,
Wisniewski SR, Balasubramani GK, Fava M, Trivedi MH,
John Rush A. 2009. Sex differences in response to
citalopram: a STARD report. Journal of Psychiatric
Research 43:503511.
Zhang J, McKeown RE, Hussey JR, Thompson SJ, Woods JR,
Ainsworth BE. 2005. Low HDL cholesterol is associated with
suicide attempt among young healthy women: the Third
National Health and Nutrition Examination Survey.
Journal of Affective Disorders 89:2533.

Am. J. Primatol.