Physiology & Behavior 103 (2011) 540–546

Contents lists available at ScienceDirect

Physiology & Behavior

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p h b

Altered aortic vascular reactivity in the unpredictable chronic mild stress model of
depression in mice
UCMS causes relaxation impairment to ACh
Elsa Isingrini a,⁎, Alexandre Surget a, Catherine Belzung a, Jean-Louis Freslon b, Jefferson Frisbee c,d,
James O'Donnell e, Vincent Camus a,f, Alexandre d'Audiffret d,g
a
INSERM U930 and, Université François Rabelais de Tours, 37200 Tours, France
b
EA4433, Université François Rabelais, 37200 Tours, France
c
Department of Physiology and Pharmacology, West Virginia University, School of Medicine, Morgantown, WV 26506, USA
d
Center for Interdisciplinary Research in Cardiovascular Sciences, West Virginia University, School of Medicine, Morgantown, WV 26506, USA
e
Department of Behavioral Medicine & Psychiatry, West Virginia University, School of Medicine, Morgantown, WV 26506, USA
f
Clinique psychiatrique universitaire, CHRU de Tours, Tours, France
g
Department of Surgery, Division of Vascular and Endovascular Surgery, West Virginia University, School of Medicine, Morgantown, WV 26506, USA

a r t i c l e i n f o a b s t r a c t

Article history: Major depression is an independent risk factor for the development of cardiovascular disease. This impact of
Received 2 September 2010 depression on vascular function seems to be mediated by the endothelial dysfunction, defined as an
Received in revised form 20 March 2011 impairment of endothelium-dependent vasorelaxation, which represents a reliable predictor of atheroscle-
Accepted 4 April 2011
rosis and has been regularly found to be associated with depression. This study aimed at investigating aortic
vascular reactivity in mice submitted to the unpredictable chronic mild stress (UCMS) procedure, a reliable
Keywords:
Depression
model of depression. The results confirm the effectiveness of the UCMS procedure to induce neuroendocrine,
Cardiovascular disease physical and behavioral depression-like alterations as well as a significant decrease of acetylcholine-induced
Endothelial dysfunction vasorelaxation without any effect on phenylephrine-induced vasoconstriction. In this study, we reveal an
Vasorelaxation altered vascular reactivity in an animal model of depression, demonstrating an endothelial dysfunction
Rodent model reminiscent to the one found in depressed patients.
Chronic stress © 2011 Elsevier Inc. All rights reserved.

1. Introduction vasorelaxant properties and exerts protective effects on the vascula-

In developed countries, major depression is the second leading
cause of death and disability after coronary heart disease [1].
Interestingly, these two pathologies have a high co-morbidity [2,3].
Several hypotheses have been proposed to explain depression-
induced cardiovascular disease including neurohormonal activation,
autonomic dysfunction, immuno-inflammatory processes, platelet
activation and endothelial dysfunction [4,5].
The vascular endothelium plays a key role in maintaining vascular
homeostasis and consequently regulates several physiological func-
tions, including vascular tone, smooth muscle cell proliferation,
inflammation, platelet aggregation, thrombosis, fibrinolysis and
oxidation. Nevertheless, the term endothelial dysfunction is limited
to describe the impairment of endothelium-dependant vasorelaxation
caused by a loss of nitric oxide (NO) bioavailability. While NO has
⁎ Corresponding author at: INSERM U930 – Team 4, Affective disorders, Université
François Rabelais de Tours, UFR Sciences et Techniques, Parc grandmont (Bat O), 37200
Tours, Cedex 01, France. Tel.: + 33 247 47 36 69 96.
E-mail address: Isingrini.elsa@voila.fr (E. Isingrini).
ture, endothelial dysfunction leads to a modification in the vascular
homeostasis which contributes to atherosclerosis formation [6],
specifically in large conductive arteries such as aorta where NO is
the most important vasorelaxant mediator.
The functional and structural assessment of endothelial dysfunc-
tion is based on studies demonstrating that normal endothelium
mediates the vasodilatation effect of acetylcholine (ACh). When the
endothelium is removed or altered, either ACh exerts a vasoconstric-
tive effect by a direct action on smooth muscles [7] or the
endothelium releases contractile factors instead of relaxing factors.
Consequently, the vasomotor response to various pharmacological or
physical stimuli can be considered as surrogate markers of NO
bioavailability. Using high resolution ultrasound equipment for
assessing the vascular remodeling of the blood vessel, several studies
demonstrated endothelial function altered in depressed patients [8–
10]. Moreover, the level of biological surrogate markers of endothelial
dysfunction in patients with depression frequently shows increased
concentrations of the soluble chemokine monocyte chemotactic
protein-1, E-selectin and ICAM-Is [9,11], as well as decreased plasma
levels of NO metabolites and endothelial nitric oxide synthase (eNOS)

0031-9384/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.physbeh.2011.04.002
 E. Isingrini et al. / Physiology & Behavior 103 (2011) 540–546
activity [12]. This endothelial dysfunction is particularly interesting in
the context of the main mechanistic hypothesis of the relationship
between depression and the increased risk of cardiovascular disease.
The unpredictable chronic mild stress (UCMS) animal model is
valid, reliable and sensitive [13] for studying depressive disorders in
rodents. UCMS involves subjecting mice to a period of mild socio-
environmental stressors. This procedure replicates several depres-
sion-related behavioral and physiological impairments which are
reversed by chronic, but not acute, antidepressant treatment [14].
The aim of the present study was to investigate the effect of UCMS,
intended to induce depression-like behaviors, on the vaso-reactive
properties of the thoracic aorta, including the vasoconstriction to
phenylephrine (L-Phe) and the vasodilatation to acetylcholine (ACh).
We hypothesize that UCMS, as observed in depression, can alter
endothelium vascular reactivity by impairing acetylcholine-induced
vasodilatation.
2. Methods
2.1. Animals
Seven week-old male BALB/cJ mice were housed in groups of 4 at their
arrival and allowed to become used to their new environment for 2 weeks.
The animals had free access to water and food in standard laboratory
conditions (12 h day/night cycle, light on at 08:00; T=22±2 °C). All
protocols were approved by the West Virginia University IACUC.
2.2. General procedure
The mice were randomized into the control group (n = 18), kept in
standard laboratory conditions or the UCMS group (n = 18) subjected
to 8 weeks of random daily socio-environmental stressors. A 7–
9 week of UCMS protocol is commonly used in order to demonstrate
that UCMS-induced behavioral disturbance can be reverse by
antidepressant treatment [14–16]. In the present study, 8-week
UCMS procedure without any antidepressant treatment was used. The
control mice were housed in groups of 4 in standard laboratory cages
(42 cm × 28 cm × 18 cm) while mice in the UCMS group were
singularly housed in individual cages (8 cm × 13.5 cm × 8.1 cm). The
coat state and body weight of each animal were evaluated every
Monday in a blind manner. The total score of the coat state was
computed by attributing a score of 0 (clean coat) or 1 (dirty coat) to
eight different body parts (head, neck, dorsal coat, ventral coat, tail,
forelimb, hind-limb, and genital region) [14,17]. At the seventh week,
after performance in the reward test (anhedonic test), 8 animals per
group were sacrificed for a plasma corticosterone radioimmunoassay.
During the last week, after the splash test, the novelty suppression of
feeding (NSF) test and locomotor activity test, the other half of the
animals (UCMS, n = 10 and control n = 10) were euthanized for the
vascular reactivity study.
2.3. Unpredictable chronic mild stress
The UCMS regimen used is a variation of the procedure previously
described [18–21] and adapted to mice. Several times a day during the
8 weeks of the UCMS procedure, the mice were exposed to various
socio-environmental mild stressors (between two and four stressors
per day). The environmental stressors were: altered bedding (change
or removal of sawdust, damp sawdust, substitution of sawdust with
2 cm of 21 °C water), cage tilting (45°) and altered length and time of
light/dark cycle (inversion of the light/dark cycle and lights on for a
short time during the dark phase). The social stressors were: isolation
(control mice were housed in groups of 4 while UCMS mice were
singularly housed) and social stress (mice were placed in the empty
cage of another male). The presentation of the different stressors was
randomized each week to maximize their unpredictability (Table 1).
541
For ethical reasons, the stress procedure did not involve food and
water deprivation.
2.4. Behavioral testing
All behavioral testing was performed during the light phase of the
light/dark cycle and were recorded using video equipment.
2.4.1. Reward test
The reward test is a model of anhedonia, used to assess UCMS-
induced effects on the motivation to obtain a reward. It was
demonstrated that exposure to repeated unpredictable stress pre-
vents the acquisition of an appetitive behavior induced and
maintained by a highly palatable food [22]. The reward test requires
a device containing three aligned compartments with the same
dimension (20 × 20 × 20 cm). Only the colors of the walls and the floor
are different: white for the first one, gray for the second one and black
for the third one. The three compartments are linked by two openings
controlled by the experimenters. The device is illuminated with 200-
lux white light. Four and half weeks before the first session, a small
portion (2 g ± 1) of a chocolate cookie (Pepito, Lu, France) was placed
in the home cage every two days during 2.5 weeks in order to
familiarize the mice with the palatable stimulus. The last two weeks
before the first session were cookie-free. One hour before testing, food
was removed in order to avoid inter-individual differences in
the drive for feeding (hunger). At the time of testing, a small amount
(2 g ± 1) of chocolate cookie (or of regular food in a supplemental
control experiment) was positioned at the center of the black room.
The white room was the compartment of the departure where the
mouse is placed with its head facing the opposite side to the opening.
Four sessions of testing are performed within 9 days, each session
being separated from the previous one by 3 days. Each session lasted
5 min; the door of the first opening was closed after the transition of
the mouse within a maximum time of 2 min (at 2 min, the mouse was
gently guided to the second room when required). The cookie
consumption (number of bites) was recorded within the 5 min test
period. The validation of this test was demonstrated by the stronger
drive to chew a chocolate cookie than to chew a regular food pellet as
a robust increase of the chewing frequency with the chocolate cookie
when compared with the regular food [18,23]. This test allowed
measuring the locomotion and exploratory behavior (number of
passages through the second door over the session) and anhedonia
(number of bites over the session).
2.4.2. Splash test
The splash test, performed under a red light (15 W), consists of
squirting two sprays of an atomizer containing a 10% sucrose solution
on the dorsal coat of a mouse in its home cage. Because of its viscosity,
the sucrose solution dirties the mouse's fur and animals initiate
grooming behavior. After applying sucrose solution, the grooming
latency and frequency were recorded for a period of 5 min as an index
of self-care and motivational behavior. The splash test, pharmacolog-
ically validated, demonstrates that UCMS decreases grooming
behavior [17,19,23,24], which can be interpreted as a loss of
motivational behavior considered to parallel some symptoms of
depression such as apathic behavior [25].
2.4.3. NSF test
The NSF test is a modified version of the test used by Santarelli et al.
[19]. This test induces a motivational conflict between eating behavior
and the fear of novelty. Latency to initiation of sniffing and eating a
food pellet is used as an index of depressive-like behavior since
chronic antidepressant treatment decreases this measure [14,17]. This
test was conducted in an open-field (30 × 30 × 30 cm) with a sawdust
covered floor and under a red light (230 V, 15 W). Twelve hours
before testing, the mice were food restricted. At the start of testing, a
542 E. Isingrini et al. / Physiology & Behavior 103 (2011) 540–546
Table 1
The unpredictable chronic mild stress (UCMS) protocol during the first week of the experiment.
Week Friday Saturday Sunday
1 Coat state, 4 light/dark Social stress (9 h)
weighing (10 h) succession every
social stress 30 min (9 h30–11 h30)
(12 h) without sawdust
(9 h–16 h)
Without sawdust Dark (4 h–6 h) Social stress (16 h)
(15 h–17 h) 4 light/dark succession
every 30 min
(23 h30–1 h30,
4 h–6 h)
single food pellet was placed on a white paper (2 × 2 cm) at the center
of the apparatus. Mice were placed in a corner of the open field. The
sniffing and eating latency were recorded during the 5-min testing
period. Immediately after, each animal was transferred to an
individual cage for 3 min before being replaced in his home cage
and the amount of food consumed was measured in order to assess
change in appetite as a confounding factor.
2.4.4. Measurement of locomotor activity
The measurement of locomotor activity was performed using an
open-field chamber. The floor of the open-field chamber
(50 × 50 × 25 cm), made of white Plexiglas, was divided into 16
identical squares. The naive mouse was placed at one corner of the
chamber and allowed to explore for 5 min. We recorded the number
of line crossings (with all four paws crossing the line) they made.
2.5. Corticosterone radioimmunoassay
The animals were killed by CO2 asphyxiation and then decapitated
(between 4.00 pm and 6.30 pm). The trunk blood was collected and
spun at 1200 ×g for 12 min to separate the serum, which was stored at
−20 °C until corticosterone radioimmunoassay. Serums were analyzed
for total corticosterone levels using a 125I-labeled corticosterone double-
antibody radioimmunoassay kit (MP Biomedicals, NY) according to the
manufacturer's protocol. To avoid inter-assay variability, all the samples
were run in a single assay (the intra-assay variability was 5.9%). Assay
limit of detection was 7.7 ng/ml. Because of technical problems, 5 mice
were excluded of the analysis (3 in the UCMS group and 2 in the control
group).
2.6. Vascular reactivity study
Mice were anesthetized with sodium pentobarbital (60 mg/kg, i.p.)
and after thoracotomy, the thoracic aorta was removed. Segments
(length 2–3 mm) were immediately placed into ice-cold physiological
salt solution (PSS, which contained in mM: 120 NaCl, 25 NAHCO3, 11.1
glucose, 4.76 KCl, 1.18 MgSO4, 1.18 KH2PO4, 2.5 CaCl2), cleaned of all
visible connective tissue and mounted in organ bath myographs
(myobath II-8 channel, WPI; Transbridge Force Analyzer, WPI) for
measurement of isometric tension. Equilibration of the rings for at least
1 h under a 2-g tension measured, in PSS maintained at 37 °C and
gassed with 5% CO2/95% O2, was performed. Contractile reactivity of
each preparation was checked using high potassium (K+, 80 mM)
depolarizing solution. Contractile responses were normalized by
dividing the force developed by the rings in g by their weight in mg.
An average weight was used to calculate a normalization coefficient.
Each response (g) was multiplied by this coefficient. Aortic rings that
did not contract up to 0.01 g/mg of tissue were discarded. Moreover,
Monday Tuesday Wednesday Thursday
Social stress (12 h) Cages tilt at 45° Bath (15 min) (10 h) Damp sawdust
(9 h–10 h30) social stress (13 h) (9 h30–11 h)
Damp sawdust (15 h–18 h) + Social stress Light/dark succession Cages tilt at 45°
social stress (16 h30) (15 h) light every 15 min (14 h–15 h30)
light/dark succession (15 h30–17 h) (14 h–17 h) light/dark succession
(15 min–22 h–23 h15) without sawdust every 15 min
(30 min–9 h30–11 h30) (17 h–8 h) (23 h–2 h)
rings were pre-conditioned by treatment with 10− 7 M L-Phe for 5 min,
at which time 10− 5 M ACh was added to the bath to assess endothelial
integrity. Any ring that failed to demonstrate both a brisk constrictor
response to L-Phe and viable endothelial function were discarded. Two
animals per group which did not respond to these criteria were
excluded of both the vascular and the behavioral analysis. A preliminary
experiment we conducted in male BALB/c mice indicates that the
response to 10− 5 M ACh ranges from 40 to 60% of relaxation in intact
endothelium rings and between 0 and 5% of relaxation in endothelium
rubbed rings.
In the present experiment, a cumulative curve response to L-Phe
(10− 9 M to 10− 5 M) was generated. Then, after precontraction with
10− 7 M of L-Phe for 10 min (time necessary to reach a plateau), a
cumulative response to ACh (10− 9 M to 10− 5 M) was generated and
expressed as tension change from the baseline L-Phe precontraction.
2.7. Statistical analysis
The mechanical response following challenge with L-Phe or ACh
were analyzed by the sigmoidal fit function using Origin Scientific
software. The logistical equation for curve fitting was: y = A2 + [(A1 −
A2) / (1 + (x / x0)p] were y represents the isometric tension, A1 and A2
represent the lower and upper bound respectively of the change in
tone with the agonist concentration (x), x0 represents the agonist
concentration (x) where the response y is halfway between the bound
(EC50), and p is power. The use of this logistic equation is appropriate
for the analysis of sigmoidal concentration–effect curve, as it
simultaneously provides estimates of the curve maximum response
and the dose at which the dependant variable reaches 50% of
maximum (EC50).
The results are presented as mean ± SEM. One-way ANOVA with
environment (control vs UCMS) as a main factor was used except
when there was a repeated measure. One-way repeated measures
ANOVA, with environment (control vs UCMS) as a main factor,
followed by a Fisher post-hoc analysis were performed to study 1) the
evolution of the coat state and the body weight (with the 8-week of
experiment as repeated factor), 2) the frequency to chew the cookie in
the reward test (with the 4-session as repeated factor) and 3) the
dose-dependent response to L-Phe and ACh (with dose of L-Phe or
ACh as repeated factor). A value of p b 0.05 was considered statistically
significant.
3. Results
3.1. Coat status and body weight
ANOVA revealed a significant main effect of UCMS (F1, 34 = 28.5,
p b 0.001) and weeks (F7, 238 = 43.5, p b 0.001) and a significant
 E. Isingrini et al. / Physiology & Behavior 103 (2011) 540–546 543

Fig. 1. Results (mean ± SEM) of the coat status (A) and the body weight (B) of mice submitted to 8 weeks of unpredictable chronic mild stress (UCMS, n = 18) and control mice
(n = 18). *p b 0.05: comparison between control and UCMS groups.

interaction between the factor week and UCMS (F7, 238 = 5.6,
p b 0.001) on the coat state. The UCMS protocol resulted in a
deterioration of the coat state, with the difference between the
control and UCMS groups reaching significance from week 4 and this
persisted until the end of the procedure (Fig. 1A, p b 0.05). In addition,
ANOVA revealed a significant effect of UCMS (F1, 34 = 11.3, p b 0.01)
and week (F7, 238 = 185.8, p b 0.001) on body weight and a significant
interaction between the factors week and UCMS (F7, 238 = 11.4,
p b 0.001). An overall increase in body weight with time is observed in
each group and UCMS induced a significant increase of the body
weight only in week 5 (Fig. 1B, p b 0.05).
3.2. Behavioral tests
In the reward test, there is no statistical difference between the
control and UCMS groups on the number of passages through the
second door (UCMS effect: F1, 14 = 3.9, p N 0.05, ns and session effect:
F1, 14 = 1.43, p N 0.05, ns) indicating no effect of UCMS on locomotion
and exploratory behavior (data not shown). The one-way repeated
measure ANOVA on the number of chews of the cookie revealed a
significant effect of UCMS (F1, 14 = 5.6, p b 0.05), of session (F3, 42 = 8.2,
p = 0.001) and a significant interaction between the factors UCMS and
session (F3, 42 = 8.3, p = 0.001). In the control group, the number of

Fig. 2. Results (mean ± SEM) of the frequency to chew the chocolate cookie over the sessions in the reward test (A), the latency to sniff and eat the food pellet in the novelty-
suppressed feeding (NSF) test (B), the latency and frequency of grooming behavior following a 10% sucrose solution spray in the splash test (C), and number of square crossings in
the open field (D) in mice submitted to 8 weeks of unpredictable chronic mild stress (UCMS) and control mice (in each group n = 10 except for the reward test n = 8). *p b 0.05:
comparison between the control and UCMS groups. #p b 0.05: significant difference with the first session in the reward test.
544 E. Isingrini et al. / Physiology & Behavior 103 (2011) 540–546
chews significantly increased over the different sessions, an effect
attenuated in the UCMS group. Moreover, UCMS significantly
decreased the number of chews of the cookie during the last two
sessions compared to control group, indicating that UCMS induced a
decrease of cookie consumption (Fig. 2A).
The results of the NSF test (Fig. 2B) reveal that mice subjected to
the UCMS protocol exhibit an increase in sniffing (F1, 18 = 11.6,
p b 0.01) and eating latency (F1, 18 = 12.9, p b 0.01) compared to the
control group. Food consumption after the test was not influenced by
the UCMS procedure (F1, 18 = 0.09, p = 0.8, ns), indicating that effects
on appetite do not account for the behavioral changes observed in this
test (data not shown).
The data presented in Fig. 2C demonstrates that mice subjected to
the UCMS protocol neglected coat grooming when compared to the
control group. This is illustrated by increased latency (F1, 18 = 8.7,
p b 0.01) and decreased frequency (F1, 18 = 11.4, p b 0.01) of grooming
behavior.
Finally, the results of the open-field test (Fig. 2D) show
increased locomotor activity in UCMS mice compared to the control
mice (F1, 18 =14.4, pb 0.01).
By inducing a decrease of cookie consumption in the reward test,
disturbance of grooming behavior and an increase eating latency in
the NSF test, these data indicate that the UCMS procedure was
effective in inducing a depression-like behavioral state, similar to
previously published reports [17,21].
3.3. Corticosterone radioimmunoassay
From blood taken between 4.00 and 6.30 PM, serum corticosterone
level was assessed by radioimmunoassay. Results indicate that UCMS
induced a significant increase of the total corticosterone level in
comparison to the control mice (mean corticosterone, control: 82.3 ±
15.1 pg/ml vs UCMS: 156.5 ±19.1 pg/ml; F1, 9 = 9.5, p b 0.05).
3.4. Vascular reactivity study
Concentration–effect curves to L-Phe obtained in intact aortic
segments were not significantly different between control and UCMS
groups (Fig. 3A; UCMS effect: F1, 15 = 0.10, p = 0.7, ns; dose effect:
F8, 120 = 13.8, p b 0.001). The maximal responses to L-Phe (control:
0.055 ± 0.010 g/mg vs UCMS: 0.050 ± 0.020 g/mg; F1, 15 = 0.5, p = 0.8,
ns), the EC50 (control: 1.0.10− 6 ± 0.5.10− 6 M vs UCMS: 1.5.10− 6 ±
0.1.10− 6 M; F1, 15 = 0.15, p = 0,7, ns) and the EC20 (control: 4.3.10− 7 ±
2.1.10− 7 M vs UCMS 3.7.10− 7 ± 1.1.10− 7 M; F1, 15 = 0.07, p = 0.8, ns)
did not differ between groups.
The maximum response to the second pre-contraction with 10− 7
L-Phe did not differ either between both groups (Control: 0.07 ±
Fig. 3. Results (mean ± SEM) of the contractile response to increasing concentrations of phenylephrine (L-Phe, 10− 9 to 3.10− 5 mol/L) (A) and the relaxant response to increasing
concentrations of acetylcholine (ACh, 10− 9 to 3.10− 5 mol/L) (B) in mice submitted to 8 weeks of unpredictable chronic mild stress (UCMS, n = 10) and control mice (n = 10).
*p b 0.05: comparison between the control and UCMS groups.
0.013 g/mg of tissue vs UCMS: 0.04 ± 0.008 g/mg of tissue; F1, 18 = 3.9,
p = 0.07, ns).
With regard to the concentration-dependent relaxation to ACh, the
one-way repeated measures ANOVA revealed a significant effect of UCMS
(F1, 18 = 6.3, pb 0.05) and of ACh concentration (F8, 144 =3.9, p b 0.001)
and a significant interaction between the two factors (F8, 144 =5.9,
pb 0.001). The acetylcholine-induced relaxation was blunted in the
UCMS group as shown by a significant difference in the decrease of the
L-Phe-induced contraction between control and UCMS groups for the
higher concentrations of 3.10− 6 M and 10− 5 M of ACh (Fig. 3B; p b 0.05).
The maximum response to ACh was significantly lower in the UCMS
compared to the control group (UCMS: 0.005±0.005 g/mg vs control:
0.040 ± 0.009 g/mg; F1, 18 = 11.3, p b 0.05) with relaxations which
represented only 9.6% of the initial contraction in the UCMS group
compared to 54.5% in the control group (F1, 18 =6.4, pb 0.05).
4. Discussion
In this study we investigated the relationship between depression-
like state and vascular reactivity in mice submitted to the UCMS
protocol. The present observations show that UCMS led to both
depressive-like behavior and altered vascular reactivity illustrated by
a blunted vasorelaxation response to an acetylcholine challenge.
These results are in accordance with previous study demonstrating
that UCMS attenuates the endothelium-dependent dilatation but had
no effect on constriction in response to phenylephrine [26].
The validity of the UCMS procedure as an animal model of
depression has been demonstrated in previously published reports
[13,14,17,19,27–30]. Indeed, the unpredictable psycho-social stress
mimics the etiology of human depression and recruits equivalent
neuroendocrine systems [13]. Moreover, the UCMS paradigm elicits
behavioral and physiological alterations analogous to symptoms of
depression [19,27] which are reversed by chronic antidepressant
treatment [14,17]. In our study, the effectiveness of the UCMS protocol
was confirmed by the increase in corticosterone plasma levels.
Moreover, mice subjected to UCMS developed anhedonia (Reward
test), a core symptom of human depression. At the same time, UCMS
induced a progressive deterioration of the coat state and reduced
grooming behavior (Splash test) indicating a loss of motivational and
self-care behavior and an exaggerated emotional reactivity (NSF test).
Mice also exhibited weight gain and increased locomotor activity;
symptoms which are analogous to human depression where an
increase or decrease in appetite and inhibition of activity or agitation
is commonly observed. Changes in body weight due to UCMS were
reported in a protocol-dependent manner [17,31,32]. As this change
occurred only at one time point over the 8 week period with no
differences at the end of the procedure, we consider it as nonrelevant.
Through multiple behavioral and neuroendocrine readouts, UCMS
 E. Isingrini et al. / Physiology & Behavior 103 (2011) 540–546
elicits a depressive-like state demonstrating UCMS as a valid model to
mimic the altered vascular reactivity observed in depressed subject.
Endothelial function is most commonly measured as the vasomotor
response to pharmacological or physical stimuli. In our study, the UCMS
procedure lead to an impairment of the aortic vasodilatation induced by
acetylcholine, suggestive of an alteration of the endothelium-dependent
relaxation in mice subjected to the UCMS procedure. Indeed, Furchgott
and Zawadzki[7] demonstrate that the relaxation of vascular smooth
muscle cells in response to acetylcholine is dependent on the integrity of
the endothelium. This possibility is strengthened by a study of d'Audiffret
and colleagues who also found an impairment of the endothelium-
dependent relaxation under UCMS conditions without any alteration
of the endothelium-independent relaxation in response to sodium
nitropusside [26]. Moreover, as NO is the most important factor implicated
in the endothelium-dependent relaxation of the smooth muscle cells in
aorta, situations in which the relaxation to acetylcholine is altered
supposed an impairment of the NO bioavailability [6,33]. Consequently, by
impairing vasorelaxation, UCMS may have a direct impact on the
endothelium-dependent relaxation induced by the NO release. The
prostacyclin (PGI2) as well as the endothelium-derived hyperpolarizing
factor (EDHF), which are also two endothelium mediators with
vasorelaxant properties could be implicated in UCMS-induced vasor-
elaxation impairment. Accordingly, assessing the response to acetylcho-
line in the presence of inhibitors of eNOS such as the L-NAME, of PGI2
synthesis may provide an explanation about the specific role of NO, PGI2
and EDHF in UCMS-induced impairment of the aortic relaxation. In
addition, demonstrating that UCMS-altered relaxation is an effect
dependent on the endothelium and on the release of NO would
demonstrate the similarity between endothelial dysfunction in the
UCMS model of depression and that observed in depressed subjects [8–
10,12].
At the cellular level, the decrease of the endothelium-dependent
relaxation induced by NO can be explained either by an increase of the
NO-degradation induced by oxidative stress activation or by a
decrease of the NO-production induced by a low eNOS availability.
In the UCMS model, both increasing of the oxidative stress pathway
and decreasing of eNOS production has been demonstrated [34].
However, studies also demonstrated an increased eNOS activity
induced by UCMS [35]. An hypothesis can be advanced in order to
explain this apparent controversial effect: the increase eNOS activity
can serve as a compensatory mechanism in order to hide the NO
degradation by the oxidative stress pathway and stabilize the NO-
dependent vasodilatation.
The glucocorticoid hypothesis of depression [36] could serve as a
potential pathophysiological mechanism involved in these links.
Glucocorticoid-induced cardiovascular disease is mediated by endo-
thelial alterations, and NO is a strong candidate mediator [37]. Indeed,
excessive glucocorticoids, as found in our study, may increase
oxidative stress eliciting the overproduction of reactive oxygen
species, thereby reducing NO availability and causing impairment of
the relaxing activity of vascular smooth muscle cells [38]. Thus, the
increased corticosterone levels observed in mice subjected to UCMS
might be related to the hyposensitivity of thoracic aorta to ACh. In
support of the implication of the glucocorticoid pathway in the
relationship between endothelial dysfunction and depression, Broadley
et al. [39] found that metyrapone, a glucocorticoid synthesis inhibitor,
improves the endothelial dysfunction observed in depressed subjects.
However, other mechanisms, such as a change in sympathetic drive,
already demonstrated in a rat model of chronic mild stress [40], but not
in mice [26], may also contribute to this dysfunction by increasing blood
pressure. It is well known that sympathetic stimulation results in sharp
increases in blood pressure and heart rate, which may damage vascular
endothelium and impair endothelium-dependent vasomotion [41].
In conclusion, this study represents a step forward in elucidating
the mechanism responsible for the association between depression
and vascular diseases. While further investigations are required, they
545
suggest that a decrease in endothelial NO availability could be a
common factor in the main mechanistic hypotheses for this
association.
Acknowledgment
This study was supported by the medical research foundation
(FRM) grant to E.I.
References
[1] Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause
1990–2020: Global Burden of Disease Study. Lancet 1997;349(9064):1498–504.
[2] Lett HS, Blumenthal JA, Babyak MA, Sherwood A, Strauman T, Robins C, et al.
Depression as a risk factor for coronary artery disease: evidence, mechanisms, and
treatment. Psychosom Med 2004;66(3):305–15.
[3] Plante GE. Depression and cardiovascular disease: a reciprocal relationship.
Metabolism 2005;54(5 Suppl 1):45–8.
[4] Grippo AJ. Mechanisms underlying altered mood and cardiovascular dysfunction:
the value of neurobiological and behavioral research with animal models.
Neurosci Biobehav Rev 2009;33(2):171–80.
[5] Isingrini E, Desmidt T, Belzung C, Camus V. Endothelial dysfunction: a potential
therapeutic target for geriatric depression and brain amyloid deposition in
Alzheimer's disease? Curr Opin Investig Drugs 2009;10(1):46–55.
[6] Verma S, Anderson TJ. Fundamentals of endothelial function for the clinical
cardiologist. Circulation 2002;105(5):546–9.
[7] Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation
of arterial smooth muscle by acetylcholine. Nature 1980;288(5789):373–6.
[8] Broadley AJ, Korszun A, Jones CJ, Frenneaux MP. Arterial endothelial function is
impaired in treated depression. Heart 2002;88(5):521–3.
[9] Rajagopalan S, Brook R, Rubenfire M, Pitt E, Young E, Pitt B. Abnormal brachial
artery flow-mediated vasodilation in young adults with major depression. Am J
Cardiol 2001;88(2):196–8 [A7].
[10] Rybakowski JK, Wykretowicz A, Heymann-Szlachcinska A, Wysocki H. Impair-
ment of endothelial function in unipolar and bipolar depression. Biol Psychiatry
2006;60(8):889–91.
[11] Dimopoulos N, Piperi C, Salonicioti A, Mitsonis C, Liappas I, Lea RW, et al. Elevation
of plasma concentration of adhesion molecules in late-life depression. Int J Geriatr
Psychiatry 2006;21(10):965–71.
[12] Chrapko WE, Jurasz P, Radomski MW, Lara N, Archer SL, Le Melledo JM. Decreased
platelet nitric oxide synthase activity and plasma nitric oxide metabolites in major
depressive disorder. Biol Psychiatry 2004;56(2):129–34.
[13] Willner P. Validity, reliability and utility of the chronic mild stress model of
depression: a 10-year review and evaluation. Psychopharmacology (Berl)
1997;134(4):319–29.
[14] Ibarguen-Vargas Y, Surget A, Touma C, Palme R, Belzung C. Multifaceted strain-
specific effects in a mouse model of depression and of antidepressant reversal.
Psychoneuroendocrinology 2008;33(10):1357–68.
[15] Surget A, Wang Y, Leman S, Ibarguen-Vargas Y, Edgar N, Griebel G, et al.
Corticolimbic transcriptome changes are state-dependent and region-specific in a
rodent model of depression and of antidepressant reversal. Neuropsychopharma-
cology 2009;34(6):1363–80.
[16] Piato AL, Detanico BC, Jesus JF, Lhullier FL, Nunes DS, Elisabetsky E. Effects of
Marapuama in the chronic mild stress model: further indication of antidepressant
properties. J Ethnopharmacol 2008;118(2):300–4.
[17] Surget A, Saxe M, Leman S, Ibarguen-Vargas Y, Chalon S, Griebel G, et al. Drug-
dependent requirement of hippocampal neurogenesis in a model of depression
and of antidepressant reversal. Biol Psychiatry 2008;64(4):293–301.
[18] Isingrini E, Camus V, Le Guisquet AM, Pingaud M, Devers S, Belzung C. Association
between repeated unpredictable chronic mild stress (UCMS) procedures with a
high fat diet: a model of fluoxetine resistance in mice. PLoS One 2010;5(4):
e10404.
[19] Santarelli L, Saxe M, Gross C, Surget A, Battaglia F, Dulawa S, et al. Requirement of
hippocampal neurogenesis for the behavioral effects of antidepressants. Science
2003;301(5634):805–9.
[20] Willner P, Muscat R, Papp M. Chronic mild stress-induced anhedonia: a realistic
animal model of depression. Neurosci Biobehav Rev 1992;16(4):525–34.
[21] Yalcin I, Aksu F, Belzung C. Effects of desipramine and tramadol in a chronic mild
stress model in mice are altered by yohimbine but not by pindolol. Eur J Pharmacol
2005;514(2–3):165–74.
[22] Gambarana C, Scheggi S, Tagliamonte A, Tolu P, De Montis MG. Animal models for
the study of antidepressant activity. Brain Res Brain Res Protoc 2001;7(1):11–20.
[23] Surget A, Belzung C. Unpredictable chronic mild stress in mice. In: Kalueff AV,
LaPorte JL, editors. Experimental models in neurobehavioral research. New York:
Nova Science; 2008. p. 79–112.
[24] Ducottet C, Aubert A, Belzung C. Susceptibility to subchronic unpredictable stress
is related to individual reactivity to threat stimuli in mice. Behav Brain Res
2004;155(2):291–9.
[25] Willner P. Chronic mild stress (CMS) revisited: consistency and behavioural–
neurobiological concordance in the effects of CMS. Neuropsychobiology 2005;52(2):
90–110.
546 E. Isingrini et al. / Physiology & Behavior 103 (2011) 540–546
[26] d'Audiffret AC, Frisbee SJ, Stapleton PA, Goodwill AG, Isingrini E, Frisbee JC.
Depressive behavior and vascular dysfunction: a link between clinical depression
and vascular disease? J Appl Physiol 2010;108(5):1041–51.
[27] Ayensu WK, Pucilowski O, Mason GA, Overstreet DH, Rezvani AH, Janowsky DS.
Effects of chronic mild stress on serum complement activity, saccharin preference,
and corticosterone levels in Flinders lines of rats. Physiol Behav 1995;57(1):
165–9.
[28] Griebel G, Stemmelin J, Scatton B. Anxiolytic- and antidepressant-like effects of
the non-peptide vasopressin V1b receptor antagonist, SSR149415, suggest an
innovative approach for the treatment of stress-related disorders. Proc Natl Acad
Sci USA 2002;99(9):6370–5.
[29] Sudom K, Turrin NP, Hayley S, Anisman H. Influence of chronic interleukin-2
infusion and stressors on sickness behaviors and neurochemical change in mice.
Neuroimmunomodulation 2004;11(5):341–50.
[30] Mineur YS, Prasol DJ, Belzung C, Crusio WE. Agonistic behavior and unpredictable
chronic mild stress in mice. Behav Genet 2003;33(5):513–9.
[31] Ducottet C, Belzung C. Correlations between behaviours in the elevated plus-maze
and sensitivity to unpredictable subchronic mild stress: evidence from inbred
strains of mice. Behav Brain Res 2005;156(1):153–62.
[32] Pothion S, Bizot JC, Trovero F, Belzung C. Strain differences in sucrose preference
and in the consequences of unpredictable chronic mild stress. Behav Brain Res
2004;155(1):135–46.
[33] Behrendt D, Ganz P. Endothelial function. From vascular biology to clinical
applications. Am J Cardiol 2002;90(10C):40L–8L.
[34] Kamper EF, Chatzigeorgiou A, Tsimpoukidi O, Kamper M, Dalla C, Pitychoutis PM,
et al. Sex differences in oxidant/antioxidant balance under a chronic mild stress
regime. Physiol Behav 2009;98(1–2):215–22.
[35] Li JM, Kong LD, Wang YM, Cheng CH, Zhang WY, Tan WZ. Behavioral and
biochemical studies on chronic mild stress models in rats treated with a Chinese
traditional prescription Banxia-houpu decoction. Life Sci 2003;74(1):55–73.
[36] Belmaker RH, Agam G. Major depressive disorder. N Engl J Med 2008;358(1):
55–68.
[37] Mangos GJ, Walker BR, Kelly JJ, Lawson JA, Webb DJ, Whitworth JA. Cortisol inhibits
cholinergic vasodilation in the human forearm. Am J Hypertens 2000;13(11):
1155–60.
[38] Iuchi T, Akaike M, Mitsui T, Ohshima Y, Shintani Y, Azuma H, et al. Glucocorticoid
excess induces superoxide production in vascular endothelial cells and elicits
vascular endothelial dysfunction. Circ Res 2003;92(1):81–7.
[39] Broadley AJ, Korszun A, Abdelaal E, Moskvina V, Deanfield J, Jones CJ, et al.
Metyrapone improves endothelial dysfunction in patients with treated depres-
sion. J Am Coll Cardiol 2006;48(1):170–5.
[40] Grippo AJ, Moffitt JA, Johnson AK. Cardiovascular alterations and autonomic
imbalance in an experimental model of depression. Am J Physiol Regul Integr
Comp Physiol 2002;282(5):R1333–41.
[41] Harris KF, Matthews KA. Interactions between autonomic nervous system activity
and endothelial function: a model for the development of cardiovascular disease.
Psychosom Med 2004;66(2):153–64.