Withdrawal from chronic amphetamine induces Depressive-Like behavioral effects

in rodents

John F. Cryan,1,2 Daniel Hoyer,2 Athina Markou.1,2

1 Department of Neuropharmacology (JFC, AM), Scripps Research Institute, La

Jolla, California, USA.
2 Nervous System Research (JFC, DH), Novartis Pharma AG, Basel, Switzerland.

Received 3 June 2002; received in revised form 15 August 2002; accepted 4

September 2002.

RESUMEN

Background
Amphetamine withdrawal and major depression share many behavioral
commonalties in humans. Therefore, the examination of the behavioral effects of
amphetamine withdrawal in rodents may provide insights into the neurobiological
mechanisms underlying both disorders and aid in the development of animal
models of depression that are sensitive to antidepressant agents.

Methods
We examined the behavioral effects of withdrawal from chronic continuous infusion
of amphetamine (via minipump) in three behavioral paradigms: the intracranial self-
stimulation (ICSS) procedure in rats, the modified forced swim test in rats, and the
tail suspension test in mice.

Results
Amphetamine withdrawal resulted in a prolonged (5 day) deficit in brain reward
function as assessed by elevations in ICSS thresholds. Using a similar regimen of
amphetamine administration, we examined the behavioral effects of withdrawal in a
modified rat forced swim test. Animals that were treated with the highest dose of
amphetamine (10 mg/kg/day) exhibited increased climbing behavior and
decreased immobility 24 hours after withdrawal; by the 48-hour testing time point,
this effect had dissipated. In contrast, animals that had been pretreated with 5
mg/kg/day amphetamine exhibited a pronounced increase in immobility indicative
of an increase in “depressive-like” behavior, coupled with decreases in swimming
and climbing. In the mouse tail suspension test, both regimens of amphetamine
pretreatment induced increases in immobility scores, also indicative of “depressive-
like” behavior, 24 hours following withdrawal.

Conclusions
Withdrawal from chronic amphetamine administration results in behavioral changes
that may be analogous to some aspects of depression in humans, such as reward
deficits (i.e., elevations in brain reward thresholds) and behaviors opposite to those
seen after treatment with antidepressant drugs, such as decreased immobility in
the forced swim test and the tail suspension test.

INTRODUCCIÓN
Diminished interest or pleasure in rewarding stimuli (anhedonia) is one of the core
symptoms of both depression and psychostimulant withdrawal [American
Psychiatric Association 1994], which suggests that there are common
neurobiological correlates underlying the manifestations of these disorders [Barr et
al, Lynch and Leonard 1978, Markou and Kenny and Markou et al 1998].
Therefore, the characterization of the behavioral changes associated with
withdrawal from drugs of abuse can be used as an animal model of the symptom of
anhedonia with construct, convergent, and predictive validities [Geyer and Markou
1995 and Kokkinidis et al 1986]. The use of the intracranial self-stimulation (ICSS)
paradigm has provided investigators with a reliable and quantitative behavioral
readout that enables the assessment of reduced brain reward function following
withdrawal from various drugs of abuse, including cocaine [Markou and Koob
1991], amphetamine [Harrison et al 2001, Kokkinidis et al 1980 and Paterson et al
2000], ethanol [Schulteis et al 1995], morphine [Schulteis et al 1994], and nicotine
[Epping-Jordan et al 1998 and Harrison et al 2001].

The forced swim test, as originally described by [Porsolt et al 1977 and Porsolt et
al 1978], is the most widely used pharmacologic model for assessing
antidepressant activity in the rodent laboratory, largely because of its ease of use,
reliability across laboratories, and ability to detect a broad spectrum of
antidepressants [Cryan et al 2002a]. The development of immobility when rats are
placed in an inescapable cylinder of water is thought to reflect either a failure of
persistence in escape-directed behavior (i.e., behavioral despair) or the
development of passive behavior that disengages the animal from active forms of
coping with stressful stimuli [Lucki 1997].

Lucki and colleagues modified the traditional forced swim test [Detke et al 1995
and Lucki 1997] and demonstrated that the test reveals specific behavioral
components of active behaviors—namely swimming, which is sensitive to
serotonergic compounds such as the selective serotonin (5-HT) reuptake inhibitors
and 5-HT receptor agonists, and climbing, which is sensitive to tricyclic
antidepressants and drugs with selective effects on catecholaminergic
transmission [Cryan and Lucki 2000, Cryan et al 2002a, Cryan et al 2002b and
Detke et al 1995]. These modifications include increasing the water depth to 30 cm
from traditional depths of 15–18 cm and using a time sampling technique to rate
the predominant behavior over a 5-sec interval.

The tail suspension test is theoretically similar to the forced swim test and also has
the ability to detect a broad spectrum of antidepressants [Mayorga and Lucki 2001
and Steru et al 1985]. Briefly, mice (rats are rarely used) are suspended by their
tails for 6 min, and the amount of time they adopt an immobile position is
determined. Antidepressants reduce the time the animals are observed to be
immobile. Although both the forced swim test and tail suspension test are similar in
the constructs that they purport to assess, it is becoming clear that they are
probably different in terms of the biological substrates that underlie the observed
behaviors [Bai et al 2001]. Accordingly, the use of both tests often can give both
complementary and converging data on potential antidepressant drugs [Bai et al
2001, Conti et al 2002 and Porsolt 2000]. One interpretation of the behavioral
immobility observed in the forced swim and tail suspension tests is that immobility
behavior allows for adaptive retraction from the inescapable stress of forced
swimming or tail suspension, which is interrupted with bouts of escape-motivated
activity. Together these alternating behavioral responses comprise a coping
strategy [Thierry et al 1984] in which immobility behaviors represent the
psychological concept of “entrapment” described in clinical depression [Dixon
1998, Gilbert and Allan 1998 and Lucki et al 2001].

Although both the forced swim and the tail suspension tests initially were
introduced to detect the effects of antidepressant compounds, interventions known
to be involved in the susceptibility or induction of major depression in humans
induce an increase in immobility. These manipulations include a genetic
predisposition [Vaugeois et al 1996], genetic alterations to noradrenergic [Sallinen
et al 1999 and Schramm et al 2001] or opioid receptors [Filliol et al 2000],
exposure to early life stressors [Papaioannou et al 2002] or to prenatal stress
[Alonso et al 2000], pharmacologic alterations of noradrenergic neurotransmission
[Stone and Quartermain 1999], being in the postpartum state [Galea et al 2001], or
deprivation of dietary tryptophan [Blokland et al 2002]. In addition, withdrawal from
both morphine and phencyclidine, which in humans has been associated with
depressive-like behavior, recently have been shown to increase immobility in the
forced swim test in rats and mice respectively [Anraku et al 2001 and Noda et al
2000]. Furthermore, increases in immobility in the forced swim test also have been
reported in some models of depression such as the Flinders rat model [Overstreet
et al 1995], neonatal clomipramine administration [Hansen et al 1997
andVelazquez-Moctezuma and Diaz Ruiz 1992], and social isolation [Heritch et al
1990], further supporting the use of this parameter to detect depressive-like
behavior and indicating the etiologic validity of these paradigms [Geyer and
Markou 1995].

Because psychostimulant withdrawal precipitates depressive-like behaviors in

humans [Barr et al], we investigated whether withdrawal from chronic
amphetamine also can induce immobility in the forced swim and the tail
suspension tests. We used a regimen of continuous amphetamine or saline
exposure administered through subcutaneous osmotic minipumps. The minipumps
were removed after 6–7 days of exposure to amphetamine, and the animals’
behavior in the ICSS procedure, the forced swim test, and the tail suspension tests
was assessed. We hypothesized that withdrawal from chronic amphetamine
administration would result in behavioral alterations with relevance to depression
that would be detectable in all three paradigms tested. In the case of ICSS, we
predicted an elevation in brain reward thresholds reflecting diminished interest in
the rewarding electrical stimuli, whereas in both the modified forced swim and tail
suspension tests, we expected an increase in immobility reflecting decreased
active coping with the situation. As described earlier, both diminished interest or
pleasure in rewarding stimuli and decreased active coping are constructs
characterizing depression in humans.

Methods and materials

Animals
Male Wistar rats (Charles River, Hollister, CA), weighing 275–350 g for the ICSS
study and 175–225g (Charles River, Portage, IN) for the forced swim test study,
were housed in pairs with food and water available ad libitum, except during
testing, in a temperature and humidity controlled vivarium (21°C). Rats were
maintained on a 12-hour reverse light–dark cycle with lights on at 6 . DBA/2Ha
mice (Harlan, Dublin, VA) were used in the tail suspension test study and were
housed in the same manner as the rats, except with four mice per cage. All
experimental procedures occurred during the dark cycle, were approved by the
Institutional Animal Care and Use Committee of the Scripps Research Institute,
and were in accordance with the Association for the Assessment and Accreditation
of Laboratory Animal Care (AAALAC) guidelines. Animals were allowed to
habituate to their new environment for at least 1 week before the start of any
procedure, during which time they were handled at least twice.

ICSS electrode implantation

When the subjects reached a minimal weight of 325 g, they were anesthetized with
an isoflurane–oxygen vapor mixture (1–3% isoflurane) and secured in a stereotaxic
frame (David Kopf Instruments, Tujunga, CA). Stainless steel bipolar electrodes,
with a diameter of .25 mm (model MS303/2, Plastics One, Roanoke, VA) cut to 11
mm in length, were implanted into the medial forebrain bundle at the level of the
posterior lateral hypothalamus (AP −.5 mm from bregma; ML ± 1.7 mm; and DV
−8.3 mm from dura with the incisor bar at + 5 mm above the interaural line). Dental
acrylic (Co.-Oral-Ite, Diamond Springs, CA) anchored the electrode to four
stainless steel screws embedded in the skull.

ICSS procedure
Brief electrical stimulation of the posterior lateral hypothalamus is reinforcing as
indicated by the fact that rats perform an operant (turn a wheel) to receive the
electrical stimuli [Markou and Koob 1992 and Stellar and Stellar 1985]. A discrete-
trial ICSS procedure was used that provides one with current-intensity thresholds,
a measure of reward [Kornetsky and Esposito 1979, Markou and Koob 1992 and
Markou and Koob 1993]. The procedure used (for details see [Harrison et al 2001,
Markou and Koob 1992, Markou and Koob 1993 and Paterson et al 2000] was a
modification of an ICSS procedure described previously [Kornetsky and Esposito
1979]. The duration of each ICSS session was approximately 30 min. Stable
baseline thresholds (defined as 10% of the mean on 5 consecutive days) were
established for all rats before pump implantation was initiated.

Osmotic pump implantation and removal

Subjects were anesthetized with an isoflurane and oxygen vapor mixture (1–3%
isoflurane) and prepared with subcutaneous osmotic minipumps (Alzet model 2
ML1 for rats or model 2001 for mice; Alza Corporation, Palo Alto, CA) along the
back, parallel to the spine, with the flow meter directed posteriorly. Pumps were
filled with either physiologic saline or amphetamine solution (Sigma Aldrich, St.
Louis, MO). The concentration of the latter was adjusted according to animal
weight and pumping rate to deliver a dose of 5 or 10 mg/kg/day. The wound was
closed with stainless steel wound clips and covered with a povidone and iodine
antiseptic ointment. Pumps were surgically removed 6 (rat) or 7 days (mice) later
under isoflurane anesthesia. The wounds were reclipped and treated with the
antiseptic ointment.

Modified forced swim test

The modified forced swim test was carried out as described previously [Cryan and
Lucki 2000 and Cryan et al 2002a], [Cryan et al 2002b]. Briefly, rats were taken
from the vivarium and immediately placed individually for 15 min in Pyrex cylinders
(21 × 46 cm; Fisher Scientific, Tustin, CA), which were filled with water to a depth
of 30 cm. The testing room was bright, and two bright (60-watt) lamps provided
extra illumination above the cylinders. The rats were removed 15 min later, dried,
and placed in their home cage. Twenty-four hours after their first exposure, the
animals were again placed in the swim apparatus for 5 min, and behaviors were
monitored from above by video camera for subsequent analysis. The rater of the
behavioral patterns was blind with respect to the experimental conditions being
scored. A time sampling technique was employed whereby the predominant
behavior in each 5-sec period of the 300-sec test was recorded. Climbing behavior
consisted of upward-directed movements of the forepaws along the side of the
swim chamber. Swimming behavior was defined as movement (usually horizontal)
throughout the swim chamber, which also included crossing into another quadrant.
Immobility was assigned when no additional activity was observed other than that
required to keep the rat’s head above water.

Because most previous studies using the modified forced swim test were carried
out in Sprague–Dawley rats under a normal light cycle [Lucki 1997], the first study
characterized the effects of two antidepressants, the selective serotonin reuptake
inhibitor paroxetine (20 mg/kg; generously provided by SmithKline Beecham,
Harlow, UK) and the norepinephrine reuptake inhibitor desipramine (15 mg/kg;
Sigma, St. Louis, MO) in Wistar rats under a reverse light cycle setting. Although
Wistar rats are often not as sensitive as Sprague–Dawley rats to the effects of
antidepressants in the forced swim test [Porsolt et al 1978], it was deemed
important for us to use a strain and a light cycle that would enable cross
comparisons with our ICSS procedure. Drug doses were selected from previous
experiments showing robust effects at these doses [Detke et al 1995].
Antidepressants or saline were administered subcutaneously three times at 1, 5,
and 23.5 hours before the test session.

In the second forced swim test experiment, a separate group of naïve rats was
prepared with minipumps containing 0, 5, or 10 mg/kg/day amphetamine for 6
days. Twenty-four hours following withdrawal, the animals were individually placed
in the testing chambers for a 15-min session for what is usually called the “pretest”
session. The first 5 min were videotaped and scored; 24 and 48 hours later, the
animals were replaced in the swim chambers for 5 min, and their behavior was
monitored from above by video.

The tail suspension test

The tail suspension test was carried out in mice as described previously [Conti et al
2002, Mayorga and Lucki 2001 and Steru et al 1985]. All experimental testing
sessions were conducted between 12 and 6 , with mice randomly assigned to
treatment conditions and tested in counterbalanced order. Naïve mice were
prepared with minipumps containing 0, 5, or 10 mg/kg/day amphetamine for 7
days. Twenty-four hours following withdrawal, mice were individually taken from
the vivarium to an adjoining room and immediately suspended by the tail to a
horizontal ring-stand bar (distance from the FLOOR = 30 cm) using adhesive tape
affixed 2 cm from the tip of the tail. The testing room was brightly lit, and two bright
(60-watt) lamps provided extra illumination above the ringstand. Typically, mice
demonstrated several escape-oriented behaviors interspersed with bouts of
immobility of increasing length as the session progressed. A 6-min test session
was employed that was recorded by a video camera positioned in front of the tail
suspension test apparatus. Video tapes were scored by a trained observer who
was blind to the experimental conditions. The behavioral measure scored was the
duration of “immobility,” defined as the time when the mouse was not engaged in
escape-like behaviors. The animals were retested for 3 more consecutive days
following withdrawal from amphetamine to investigate whether the test was
sensitive to temporal effects of amphetamine withdrawal, repeated testing, or both.
Statistics

For ICSS, a within-subjects repeated measures analysis of variance (ANOVA) was

carried out on threshold values that were expressed as a percentage of the
baseline thresholds assessed during the 5 days before minipump implantation. For
the forced swim test, a repeated-measures ANOVA was carried out on each of the
three parameters: immobility, swimming, and climbing. For the tail suspension test,
a one-way ANOVA was conducted on the data from day 1 only; data from the
subsequent testing days were not analyzed further because repeated testing
resulted in changes in the control group’s behavior that made data interpretation
difficult (see Results). Statistically significant effects were followed where
appropriate with Fisher’s individual comparison tests. The level of significance was
set at p < .05.

Results

ICSS thresholds
Administration of amphetamine (10 mg/kg/day) followed by its termination resulted
in prolonged alterations in ICSS thresholds for the amphetamine-treated rats
compared with saline-treated animals as indicated by a significant time ×
amphetamine administration interaction [F(15,240) = 13.964, p < .001]. Upon
exposure to amphetamine, rats had significantly lower thresholds compared with
saline-treated rats (p < .05). Upon pump removal, amphetamine-treated animals
exhibited elevated brain reward thresholds compared with saline-treated rats. The
threshold elevations exhibited by the amphetamine-treated rats lasted for 5 days.
Raw mean 5-day baseline thresholds with standard errors of the mean before any
manipulation were 135 μA (± 51) for saline-treated animals and 118 μA (± 44) for
amphetamine-treated animals (see Figure 1).

Figure 1. The effects of withdrawal from chronic amphetamine on intracranial self

stimulation reward thresholds. Rats administered amphetamine (10 mg/kg/day; n =
10) exhibited a marked lowering in their brain reward thresholds compared with
those of saline-treated animals (n = 10). This threshold lowering is indicative of the
rewarding effects of amphetamine. Upon pump removal, the amphetamine-
pretreated animals exhibited significant elevations in brain reward threshold that
lasted for 5 days. All data points represent mean values with vertical lines
indicating 1 SEM. *Groups that differed significantly from saline-treated animals, p
< .05.

Forced swim test

In the first forced swim test study, there was a significant effect of antidepressant
treatment on immobility [F(2,25) = 10.10, p = .001], swimming [F(2,25) = 27.34, p <
.001], and climbing [F(2,25) = 24.76, p < .001] behavior in Wistar rats during the
dark-cycle testing. Post hoc analysis revealed that both the norepinephrine
reuptake inhibitor desipramine and the selective serotonin reuptake paroxetine
lowered immobility levels compared with saline-treated control animals. Paroxetine
treatment only increased swimming behavior, whereas desipramine treatment only
increased climbing behavior compared with saline treated control animals. These
data are consistent with previous data generated with Sprague–Dawley rats that
were tested during the light cycle [Detke et al 1995]; see Figure 2).

Figure 2. The effects of antidepressants in the modified forced swim test (as
developed by Lucki and coworkers, [Lucki 1997] in Wistar rats. Animals
administered either the selective serotonin reuptake inhibitor paroxetine (20 mg/kg,
three times; n = 10) or the norepinephrine reuptake inhibitor desipramine (15
mg/kg, three times times; n = 9) had a reduction in immobility time in the modified
forced swim test compared with saline-treated animals (n = 9). Paroxetine-treated
animals had a significant increase in swimming behavior, whereas desipramine
treatment induced a significant increase in climbing behavior. All bars represent
mean values with vertical lines indicating 1 SEM. *Groups that differed significantly
from saline-treated animals, p < .05.

In the second forced swim test study, there was a significant interaction effect of
amphetamine pretreatment with time tested, on climbing behavior [F(4,50) = 2.59,
p < .05] and a corresponding trend toward a significant interaction effect of
amphetamine pretreatment with time on immobility scores [F(4,50) = 2.32, p = .
070]. There was a main effect of amphetamine withdrawal on immobility scores
[F(2, 25) = 27.34, p < .05] and a corresponding trend toward a significant effect on
climbing behavior [F(2,25) = 3.09, p = .063]. In addition, there was a main effect of
time on immobility [F(2,50) = 5.86, p = .005] and climbing behavior [F(2,50) = 6.15,
p = .004]. There was no overall significant main or interaction effect of
amphetamine withdrawal on swimming behavior. Planned comparisons revealed
that withdrawal from amphetamine (5 mg/kg/day) produced an increase in
immobility scores at both the 48- and 72-hour time points following withdrawal.
These changes in immobility were brought about by significant decreases in
climbing and nonsignificant decreases in swimming behaviors at these time points.
Interestingly, withdrawal from 10 mg/kg/day amphetamine produced an
antidepressant-like effect 24 hours following withdrawal followed by increases in
immobility scores on subsequent exposures to the swim test. This decrease in
immobility corresponds to a significant increase in climbing behavior; however,
planned comparisons showed that animals undergoing withdrawal from both doses
of amphetamine exhibited increases in immobility at both 48 (10 mg/kg; p = .051)
and 72 hours compared with their behavior 24 hours following withdrawal. This
increase in immobility corresponded to significant decreases in climbing and
nonsignificant decreases in swimming behavior at these time points (see Figure 3).

Figure 3. The effects of withdrawal from chronic amphetamine in the modified

forced swim test (as developed by Lucki and coworkers, [Lucki 1997] in Wistar
rats. Animals administered 5 mg/kg/day amphetamine (n = 10) exhibited a
significant increase in immobility scores 48 hours following withdrawal compared to
saline-pretreated animals (n = 8). In contrast, animals administered 10 mg/kg/day
amphetamine (n = 10) exhibited a significant decrease in immobility time 24 hours
following withdrawal, which corresponded to a significant increase in climbing
behavior. Animals pretreated with both doses of amphetamine exhibited a marked
increase in immobility on the second and third exposure compared with the first
exposure at 24 hours following withdrawal. This increase in immobility reflected
significant decreases in climbing behavior and nonsignificant decreases in
swimming behavior. All bars represent mean values with vertical lines indicating 1
SEM. *Groups that differed significantly from saline-treated animals, p < .05.
#Groups that differed significantly from the corresponding 24-hour withdrawal time
point, p < .05.

Tail suspension test

One animal climbed its tail and was excluded from any further analyses. Such tail
climbing has been reported previously in various strains of mice [Mayorga and
Lucki 2001]. Twenty-four hours following minipump removal, there was a significant
effect of amphetamine pretreatment on immobility scores [F(2,50) = 6.15, p = .004].
Post hoc analysis revealed that both groups of animals pretreated with
amphetamine (5 mg/kg/day or 10 mg/kg/day) exhibited increased immobility scores
compared with saline-treated control subjects (see Figure 4). During the
subsequent testing days, there was no increase in immobility in animals pretreated
with amphetamine compared with those pretreated with saline; however, this
apparent lack of effect was due to an increase in immobility levels in the saline
pretreated animals following reexposure to the test. In contrast, animals in a state
of amphetamine withdrawal maintained constant immobility levels throughout the
4-day testing period.

Figure 4. The effects of withdrawal from chronic amphetamine or saline

administration on behavior in the mouse tail suspension test. Animals administered
amphetamine (5 mg/kg/day; n = 10; 10 mg/kg/day; n = 10) exhibited a significant
increase in immobility scores 24 hours following withdrawal compared with saline-
pretreated animals (n = 10). All bars represent mean values with vertical lines
indicating 1 SEM. *Groups that differed significantly from saline-treated animals, p
< .05.

Discussion

Withdrawal from chronic amphetamine resulted in marked behavioral changes in

all three paradigms that may be interpreted as behaviors that are analogous to
aspects of depression. These changes include a pronounced elevation in ICSS
thresholds and an increase in immobility in both the forced swim test at certain
doses of amphetamine and the tail suspension test at all doses tested. These data
indicate that psychostimulant-withdrawal-induced depression is a robust
phenomenon that is detectable across a variety of converging behavioral tasks,
albeit in a differential temporal fashion. The alterations in behavior assessed by
ICSS are both quantitatively and qualitatively consistent with previous data using a
similar protocol, including the use of minipump to deliver amphetamine [Paterson
et al 2000]. This consistency in the data suggests that this experimental design and
method of delivering the psychostimulant produces a robust and reliable reward
deficit. It should be noted that the purpose of such an administration schedule is
not to mimic human amphetamine abuse phenomenologically per se, but rather to
induce behavioral deficits that may have relevance to both drug and non-drug-
induced depression.

Withdrawal from chronic amphetamine administration, similar to that used in the

ICSS study, produced differential temporal effects on active and passive behaviors
in the forced swim test. These effects were qualitatively opposite to those seen
with the antidepressants paroxetine and desipramine, which lowered immobility
with corresponding increases in swimming and climbing behavior, respectively.
Animals pretreated with amphetamine (5 mg/kg/day) had a marked increase in
immobility scores 48 hours following withdrawal that was still evident 72 hours
following withdrawal. This increase was associated with significant decreases in
climbing behavior and nonsignificant decreases in swimming behavior indicative of
a recruitment of both catecholaminergic, and to a lesser extent serotonergic,
pathways in the mediation of this “depressive-like” behavior.

Interestingly, following 24 hours of withdrawal, on the first exposure to the test

apparatus (often referred to as the pretest) the highest dose tested (10 mg/kg)
counterintuitively produced an antidepressant-like effect, which was associated
with an increase in the catecholaminergically mediated climbing behaviors. These
changes were no longer present 24 hours later in the test session or at the 72 hour
time point. This antidepressant-like behavior of the animals pretreated with the
higher dose of amphetamine on the first swim test exposure is difficult to explain
but may be due to a residual effect of the psychostimulant; however, it is clear from
the ICSS data that animals at this time point of withdrawal are undeniably in an
anhedonic-like state. Such anhedonia however has been reported even before the
cessation of a psychostimulant binge in humans [Gawin and Kleber 1986]. Of
further interest, it is clearly evident that animals in withdrawal from amphetamine
(at both doses) have a similar relative increase in immobility 48 and 72 hours
following withdrawal compared with their behavior on the initial exposure (24 hours
following withdrawal), unlike saline-pretreated control rats. This increase in
immobility also corresponded to significant decreases in climbing and
nonsignificant decreases in swimming behaviors at these time points.

It should be noted that a preexposure to the swim apparatus usually is needed to

reliably induce immobility, and hence assess antidepressant-like behavior, in the
original rat forced swim test [Borsini et al 1989]; however, the results of our study
indicate that this induction test may not be necessary because there was little
change in immobility levels of saline-treated animals over the repeated testing
sessions. The reason for this apparent redundancy of the induction swim test may
lie in the modified design of the swim test we employed. Unlike the original method
described by [Porsolt et al 1977 and Porsolt et al 1978], we used a greater water
depth in our cylinders (30 cm compared with 15–20 cm), which results in
substantially more active behaviors on both exposures (see [Detke and Lucki
1996]. Similarly, it has been shown that mice repeatedly tested in the forced swim
test, in a large testing apparatus with increased water depth, also maintain
consistent immobility scores [Conti et al 2002]. Of further interest, no preexposure
was required for the behavioral effects of short- and long-term continuous
treatment with various antidepressants to be manifested in the modified forced
swim test (Cryan and Lucki, unpublished observations).

Our data are in general agreement with previous data showing that withdrawal
from a chronic injection regimen of amphetamine resulted in increased immobility
in the mouse forced swim test [Kokkinidis et al 1986]. In contrast, recent data
[Hedou et al 2001] failed to show any behavioral effect of withdrawal from either
amphetamine or cocaine in an automated rat forced swim test. A differential
injection schedule may account for the differences between these studies.
Although in our ICSS paradigm we have never used the once-daily amphetamine
(1.5 mg/kg) injection schedule used by Hedou and colleagues, we have
demonstrated previously that the effects of withdrawal from amphetamine
administered through minipumps [Paterson et al 2000] are quantitatively much
greater (on the order of 50%) than those from both an escalating injection regimen
[Harrison et al 2001] and a once-daily injection regimen [Lin et al 2000]. In
conclusion, these differences suggests that differential neurochemical adaptations
occur depending on the nature of the amphetamine exposure and hence
withdrawal. Thus, it is possible that the amphetamine administration regimen of
[Hedou et al 2001] does not deliver high enough doses to induce a robust
withdrawal effect as assessed by the forced swim test. It is also possible that the
intermittent injections whereby drug concentrations rise rapidly and then fall to zero
levels allows the various receptors to recover, whereas in the case of drug
administration through subcutaneous minipumps, the receptors are continuously
bombarded with the endogenous agonist(s) with no opportunity to recover, as a
consequence of amphetamine’s actions on monoamine release mechanisms.

In the modified forced swim test using Wistar rats, we have shown that the
selective serotonin reuptake inhibitor paroxetine increases swimming behaviors,
whereas the norepinephrine reuptake inhibitor desipramine increases climbing
behavior; both compounds decrease immobility. Although many other research
groups (for example, [Kelliher et al, Molina-Hernandez and Tellez-Alcantara 2001,
Reneric et al 2001 and Stogner and Holmes 2000]) in addition to that of Lucki and
colleagues [Detke et al 1995 and Lucki 1997] have used the modified swim test
with much success, it was important to show that under our laboratory conditions
(during the dark cycle) in the strain of rat most commonly used in our laboratory
(Wistar), standard antidepressants induce expected behavioral changes. Our data
are consistent with previous data in Sprague–Dawley rats and further demonstrate
the reliability of the modified forced swim test paradigm [Cryan et al 2002a and
Detke et al 1995].

Withdrawal from chronic amphetamine also produced robust increases in

immobility 24 hours following withdrawal in the tail suspension test in mice,
indicative of an increase in “depressive-like” behavior. It is notable that the
withdrawal-induced increase in immobility was evident on the first testing in the tail
suspension test, whereas it was not detectable until the second exposure in the
modified rat forced swim test. Interestingly, for unknown reasons and unlike in the
original rat forced swim test, antidepressant-like activity also is detectable following
one test exposure in the mouse tail suspension test. Our data confirm that
detection of “depressive-like” behavioral changes in the tail suspension test can be
detected on the first test. Furthermore, following repeated testing in the tail
suspension test, saline-pretreated animals (but not animals in amphetamine
withdrawal) showed habituation to the test and, on repeated testing, had the same
immobility level as the animals in amphetamine withdrawal (data not shown).
These findings suggest that this paradigm, at least in the DBA/Ha mouse strain,
under these given conditions may not be valid under repeated testing schedules.
This effect exemplifies the clear advantage of the ICSS paradigm, which allows
repeated testing within the same animal and enables the investigator to obtain a
temporal readout of neurobiological changes that are relevant to the withdrawal
phenomenon and, accordingly, to depression. Nonetheless, our data confirm that
there are other behavioral readouts that can also detect psychostimulant
withdrawal-induced “depression” in rodents. Although the effects on these other
procedures may not be as long lasting as the effects on brain reward thresholds,
the forced swim test and tail suspension test offer fast and robust readouts.
Moreover, the increases in immobility in both the forced swim test and tail
suspension test following a withdrawal regimen that produces robust effects in the
ICSS paradigm further confirm the utility of these tests in understanding both drug-
induced and non-drug-induced depressions.

These data further demonstrate that manipulations that induce depressive-like

behavior in humans (in this case, psychostimulant withdrawal), also can increase
immobility in the forced swim test and tail suspension, which further validates the
utility of these paradigms beyond being simple antidepressant-screening tools. Our
data also are consistent with recent studies by Carlezon and colleagues [Pliakas et
al 2001] demonstrating that overexpression of the transcription factor cAMP
response element-binding protein in the nucleus accumbens induced a
psychostimulant-withdrawal-like state as assessed using conditioned place
aversion behavior in the rat. The same manipulation also increased immobility
behavior in the rat forced swim test, further reinforcing the link between
psychostimulant withdrawal and behavioral depression as assessed in the forced
swim test.

The analysis of behavior in animals that are manipulated to be in a “depressed-

like” state intuitively is a more desirable starting point for the behavioral,
neurochemical, and genetic analysis of the effects of potential antidepressant
drugs in animal models [Cryan et al 2002a]. Our data support the investigation of
such compounds in both the tail suspension test (24 hours; 5 and 10 mg/kg/day
amphetamine) and forced swim test (48 hours; 5 mg/kg/day amphetamine)
following withdrawal from chronic amphetamine. Analysis of the effects of
antidepressants in the ICSS paradigm following amphetamine and nicotine
withdrawal has shown that it is one of the few paradigms that can detect the
therapeutic effects of early-onset antidepressant drug combinations [Harrison et al
2001]. Whether such combinations can be detected using other paradigms
following drug withdrawal remains to be tested.

In conclusion, our data show that withdrawal from amphetamine administered via
subcutaneous osmotic minipump results in marked differential time-dependent
behavioral changes in three paradigms that are used routinely to assess brain
reward function and antidepressant action. The manifestation of such changes
follows a different temporal pattern in each of the tests. Such data confirm that
psychostimulant withdrawal and depression converge at multiple behavioral levels.
Further studies are needed to confirm whether the depressive-like behaviors are
reversed by antidepressant combinations in each of the parameters. Moreover,
given the commonality between drug withdrawal and depression, these tests may
serve as appropriate behavioral tools that provide converging evidence in the study
of the underlying neurobiology of depression in addition to being tests for
antidepressant drug actions. [Lucki 2001]