BEHAVIOURAL

BRAIN
RESEARCH
ELSEVIER Behavioural Brain Research73 (1996) 43-46

Detection of serotonergic and noradrenergic antidepressants in the rat

forced swimming test: the effects of water depth
M i c h a e l J. D e t k e a, Irwin Lucki b, c,,
a Department of Psychology, University of Pennsylvania, 3815 Walnut Street, Philadelphia, PA 19104-6196 USA
b Department of Psychiatry, University of Pennsylvania, 3600 Market Street, Room 808, Philadelphia, PA 19104-2649 USA
c Department of Pharmacology, University of Pennsylvania, 3600 Market Street, Room 808, Philadelphia, PA 19104-2649 USA

Abstract

A new scoring technique is described that measures active behaviors of rats in the forced swimming test, a test that predicts
antidepressant drug effects. The technique distinguishes the effects of selective serotonin reuptake inhibitors, which reduce
immobility and increase swimming behavior, from selective norepinephrine reuptake inhibitors, which reduce immobility and
increase climbing behavior. The magnitude of behavioral effects described for each drug (i.e., reduced immobility for both drugs,
increased swimming for fluoxetine, increased climbing for desipramine) was greater when testing was conducted at the deeper
30-cm rather than the shallow 15-cm water depth. Results obtained with the technique demonstrate that selective serotonin
reuptake inhibitors are not false negatives in the rat forced swimming test, as previously thought.

Keywords: Antidepressants;Forced swimmingtest; Fluoxetine;Desipramine;Serotonin;Norepinephrine;Behavior;Rat

Behavioral responses produced by antidepressant
drugs in animals serve as useful endpoints for determin-
ing potential neural mechanisms associated with their
therapeutic effects. The forced swimming test (FST),
also known as the 'behavioral despair' test [1,2], is a
behavioral test which predicts the clinical efficacy of
antidepressant drugs. The FST consists of placing rats
in a cylinder of water from which they cannot escape
for a 15-rain 'pretest', followed 24 h later by a 5-min
swimming 'test'. The pretest is a stressor which is thought
to induce a state of behavioral despair, since the animals
become progressively more immobile. If antidepressant
drugs are administered between the pretest and test
periods, they decrease the duration of immobility in the
test, i.e., they make the rats more active. The FST, when
conducted using rats, is sensitive to all of the major
classes of antidepressant drugs, as well as to electrocon-
vulsive therapy and REM sleep deprivation, whereas
few other psychoactive drugs elicit antidepressant-like
effects in the FST [3]. One major problem, however, is
that selective serotonin reuptake inhibitors (SSRIs), an
important and widely used class of antidepressant drugs,
have been suggested to be inactive in this test [4-6].
The lack of effect of SSRIs has served to limit the
* Correspondingauthor. Fax:215-573-2149.
potential predictive validity of the FST as a behavioral
test for antidepressant drugs and limit the behavioral
procedures that are available for investigating the neural
mechanisms underlying the effects of SSRIs.
Recently, the effects of different types of antidepressant
drugs were reevaluated in the FST using a new procedure
for scoring active behaviors (swimming and climbing
(a.k.a., thrashing)) as well as immobility in the FST [7].
Using this procedure, different behavioral patterns in
the FST were identified for antidepressants which selec-
tively inhibit the uptake of either norepinephrine (NE)
or serotonin (5-HT). NE uptake inhibitors, such as
desipramine and maprotiline, were shown to reduce
immobility and selectively increase climbing without
affecting swimming. In contrast, SSRIs, such as fluoxe-
tine, sertraline and paroxetine, also reduced immobility
but increased swimming without affecting climbing. This
demonstration showed that the SSRIs were effective in
the FST, despite the fact that they had previously been
thought to be false negatives in this test. This result also
demonstrated that a more refined analysis of a purely
behavioral assay could be used to distinguish drugs with
a common therapeutic outcome, but which act on dis-
tinct neurotransmitter systems.
One issue that concerned us was describing testing
conditions that might optimize demonstration of the

0166-4328/96/$9.50 © ElsevierScienceB.V. All rights reserved

SSSDI 0166-4328(96)00067-4
44 Michael J. Detke, Irwin Lucki/Behavioural Brain Research 73 (1996) 43 46

behavioral effects of SSRIs, since previous studies had ANOVA was statistically reliable (/9<0.05). The first
failed to detect the effects of SSRIs in the FST. One analysis utilized a three-factor (pretest water depth, test
difference between most previous studies that examined water depth, and drug) ANOVA to analyze each beha-
SSRIs in the FST and that of Detke et al [7] was that vior. The pretest water depth did not affect any of the
the latter used deeper water in the swimming tanks. The behaviors, nor did it interact with test water depth, drug,
previous studies had used only 15-17-cm deep water, a or the combination of the two, on any of the behaviors;
depth allowing rats to stand on the bottom of the thus it was dropped from subsequent analyses, and
cylinders, whereas the 30-cm deep water used by Detke further references to water depth will always concern
et al. did not allow rats to stand on the cylinder bottom. the depth of water during the test session.
Thus, it was hypothesized that the greater water depth The effects of desipramine and fluoxetine at different
may have encouraged the rats to swim and/or climb water depths are illustrated in Fig. 1. At both water
more, and this may have accentuated the differences in depths, desipramine and fluoxetine reduced immobility,
these parameters produced by the appropriate antide- fluoxetine increased swimming, and desipramine
pressant treatments [7]. increased climbing. This is exactly the same pattern that
The current study was undertaken to explicitly test was reported previously [7].
the above hypothesis concerning the effects of water However, there were several differences between the
depth on behaviors in the FST in response to different different water depths. Two-factor (test water depth,
antidepressants. Separate groups of rats received either drug) ANOVAs showed that there was less immobility
saline, the prototypical SSRI fluoxetine, or the selective IF( 1,64)= 123.13, p < 0.0001 ] more swimming
N E uptake inhibitor desipramine [8,9]. Doses were [F(1,63)=73.11, p<0.0001] and more climbing
chosen to produce behavioral effects of similar magni- IF(1,64) = 35.14, p < 0.0001 ] at the greater water depth.
tude: 10mg/kg subcutaneous for desipramine and In addition, the differentiation of drug effects on climbing
20 mg/kg subcutaneous for fluoxetine (obtained as a gift was greater in 30 cm of water, as demonstrated by a
from Eli Lilly, Indianapolis IN). Within each of these
groups, half of the rats were given a pretest in either 5O7
15-cm or 30-cm deep water. After the corresponding 15 cm Water Depth [~] FSal
luoxinetnileeDeSpiramnie
drug treatments, half of each of these groups were then
40
given a test in either 15-cm or 30-cm deep water. The
£
results of this study demonstrate that swimming and
climbing behaviors are exhibited more when the rats are 30
tested in deeper water, and that the differences in beha-
viors produced by specific classes of antidepressants are ¢o
20
magnified when rats are tested in deeper water.
Male Sprague-Dawley rats weighing approximately
200 g were placed in individual glass cylinders 46 cm 10
tall × 20 cm in diameter containing 23-25°C water 15 or
30 cm deep. Two swim sessions were conducted: a 15-rain
pretest followed 24 h later by a 5-min test. Drug treat- Immobility Swimming Climbing
ments were administered 23, 5 and 1 h prior to the test. 5(I
Test sessions were videotaped from above and later 30 ¢m Water Depth ~ FuloxetnleDeSpiramInieSaLnie
scored as follows: at the end of each 5-s period during
the test session, the scorer would rate the rat's behavior 40
at that time, as one of the following. (1) Immobility: a
rat was making only those movements necessary to keep 30
its head above water. (2) Swimming: a rat was making
active swimming motions, e.g., moving around in the
20
cylinder. (3) Climbing: a rat was making vigorous move-
ments with its forepaws in and out of the water, usually
directed against the walls. All of the behavior scoring l0 ¸
was done by a single rater, who was blind to the
treatment condition. It has been previously shown that
this method is highly reliable, and compares well to Immobility Swimming Climbing
timing the durations of the behaviors [7].
Fig. 1. Mean(+SEM) counts of immobility, swimming and climbing
The results were analyzed using ANOVAs, and then
behaviors when sampled every 5 s during the 5-min FST testing
Dunnett's test was used for comparisons between the period, in response to saline, desipramine and fluoxetine. Group n's
drug groups and the saline control group when the are 12 each. Differences in comparison to saline: *p<0.05, **p<0.01.
 Michael J. Detke, Irwin Lucki/Behavioural Brain Research 73 (1996) 43-46
significant drug×water depth interaction (F=6.66,
p<0.0025). The differentiation of the drug effects on
swimming was not reliably greater by the conventional
statistical threshold, but there was a trend toward such
an effect (F=2.68, p=0.076). In addition, as is evident
from the figure, the magnitudes of the behavioral effects
described for each drug (i.e., reduced immobility for
both drugs, increased swimming for fluoxetine, increased
climbing for desipramine) were all greater at 30 cm of
water depth.
The results reported here are consistent with those of
Borsini et al. [10,11] who showed that there was less
immobility in the FST when testing at 30-cm depth than
at 15-cm depth, and that there was little difference
between 15-cm and 30-cm pretest depths. This latter
finding suggests that the magnitude of the stressor (i.e.,
water depth) used to induce the state of behavioral
despair in the pretest was relatively unimportant within
the range of parameters tested here. The present study
varied both test depths and antidepressant drug treat-
ment within the same study, which had not previously
been done. The fact that pretest depth was varied within
the same study allowed us to exclude the possibility that
the pretest and test depths interacted in any way, such
as might be possible due to the subjects' familiarity with
a specific situation. Such contextual learning effects have
been reported with other behavioral paradigms 1-12].
The present study further extended these findings to
include the description of active behaviors in the FST,
the effects of an SSRI, and differentiation between behav-
ioral effects produced by an SSRI and a selective NE
uptake inhibitor. Most importantly, this study demon-
strated that the greater depth of water in the FST test
session helped to enhance and clarify the behavioral
effects of both of the tested types of antidepressants.
The results of the present study support the impor-
tance of distinguishing the active behaviors shown by
rats in the FST. The wide differences between laborato-
ries in the immobility times reported for rats receiving
vehicle (140-280 s out of 300 [13,14]), may be due to
the failure to score swimming as a separate behavior.
Unlike climbing, which is a very distinct behavior (inter-
rater reliability correlation=0.97), swimming involves
less vigorous movement and is more likely to be confused
with immobility (inter-rater reliability correlations = 0.81
and 0.90, respectively). When immobility and swimming
are not distinguished, but are collapsed, a decrease in
immobility and an increase in swimming would be likely
to cancel each other out. This may be another reason
(in addition to the use of shallow water) for why the
effects of SSRIs were not detected in many previous
studies.
While certain aspects of the pharmacology of the
effects of DMI and related compounds in the rat have
been discerned e.g., [15,16], relatively little is known
about the pharmacological and anatomical substrates
45
for the antidepressant effects of SSRIs. Having identified
a reliable behavioral assay for SSRI antidepressants
using the FST, it is now possible to study further the
anatomical and pharmacological components of the
5-HT system that specifically mediate these effects. For
example, the demonstration that 5-HTxA receptor ago-
nists such as 8-OH-DPAT and gepirone produce antide-
pressant-like effects in the FST [17] and that the pattern
of behaviors resembles that of the SSRIs [7] suggests
that 5-HTta receptors may be involved in mediating the
antidepressant effects of SSRIs. The role played by other
subtypes of 5-HT receptors in mediating this pattern of
behaviors remains to be examined. In addition, the
specific anatomic substrates of the antidepressant-like
effect of SSRIs in the FST can now be studied, by using
techniques such as local intracerebral injection of com-
pounds, followed by the use of the behavioral assay.
Both of these areas of investigation may provide a great
deal of information about the mechanisms by which the
SSRIs mediate specific antidepressant-like functions in
animals, and may provide us with insights as to the
mechanisms by which they mediate therapeutic effects
in human patients.
The distinction between noradrenergic and serotoner-
gic antidepressants in the FST presented here provides
an intriguing parallel between the results of animal
behavioral tests and some recent findings in the human
clinical literature on depression. The present studies
distinguish the NE and 5-HT systems by demonstrating
that they produce separate behaviors in an analog of
depression. Similarly, Delgado et al. [18, 19] have shown
that the NE and 5-HT neurotransmitter systems produce
distinguishable contributions to therapeutic efficacy in
human depressed patients. In their studies, dietary deple-
tion of 5-HT precursors led to clinical relapse in
depressed patients who were successfully treated with
SSRIs, but not in those treated with NE-selective uptake
inhibitors. Conversely, blockade of NE synthesis led to
clinical relapse in depressed patients who were success-
fully treated with NE-selective uptake inhibitors, but
not in those treated with SSRIs. Others have also begun
to make inroads into the specifics of the distinctions
between the two neurotransmitter systems in depression.
For example, Katz et al. [20] have shown that changes
in levels of the metabolites of NE in patients treated for
depression are correlated with changes in psychomotor
symptoms, while changes in 5-HT metabolite levels
correlate with improvement of mood and anxiety symp-
toms. These convergent results between animal and
clinical studies may lead to better understanding of the
separate contributions of NE and 5-HT systems in the
biological treatment of depression.
This research was supported by USPHS grants MH
18902, MH 36262, and GM 07170, and the Benjamin
and Mary Siddons Measey foundation.
46 Michael J. Detke, Irwin Lucki/Behavioural Brain Research 73 (1996) 43 46
References
[ 1] Porsolt, R.D., Anton, G., Deniel, M. and Jalfre, M., Behavioral
despair in rats: A new model sensitive to antidepressant treat-
ments, Eur. J. Pharmacol., 47 (1978) 379 391.
[2] Porsolt, R.D., Le Pichon, M. and Jalfre, M., Depression: A new
animal model sensitive to antidepressant treatments, Nature, 266
(1977) 730-732.
[3] Borsini, F. and Meli, A., Is the forced swimming test a suitable
model for revealing antidepressant activity? Psychopharmacology,
94 (1988) 147-160.
[4] Porsolt, R.D., Behavioral despair: Present status and future per-
spectives. In B.E. Leonard and P.J. Spencer (Eds.), Antidepres-
sants: Thirty Years On, CNS Publishers, London, 1990, pp. 85-94.
[5] Porsolt, R.D. and Lenegre, A., Behavioral models of depression.
In J.M. Elliot, D.J. Heal and C.A. Marsden (Eds.), Experimental
Approaches to Anxiety and Depression, John Wiley and Sons, New
York, 1992, pp. 73-85.
[6] Porsolt, R.D., Lenegre, A. and McArthur, R.A., Pharmacological
models of depression. In Animal Models in Psychopharmacology,
Advances in Pharmacological Science, Birkh~iuser Verlag, Basel,
1991, pp. 137-159.
[7] Detke, M.J., Rickels, M. and Lucki, I., Active behaviors in the
rat forced swimming test differentially produced by serotonergic
and noradrenergic antidepressants, Psychopharmacology, 121
(1995) 66-72.
[8] Bolden-Watson, C. and Richelson, E., Blockade by newly-devel-
oped antidepressants of biogenic amine uptake into rat brain
synaptosomes, Life Sci., 52 (1993) 1023-1029.
[9] Richelson, E. and Pfenning, M., Blockade by antidepressants and
related compounds of biogenic amine uptake into rat brain syn-
aptosomes: Most antidepressants selectively block norepineph-
rine uptake, Eur. J. Pharmacol., 104 (1984) 277-286.
[-10] Borsini, F., Lecci, A., Sessarego, A., Frassine, R., and Meli, A.,
Discovery of antidepressant activity by forced swimming test may
depend on pre-exposure of rats to a stressful situation, Psycho-
pharmacology, 97 (1989) 183-188.
[11] Borsini, F., Volterra, G. and Meli, A., Does the behavioral
'despair' test measure 'despair'? Physiol. Behav., 38 (1986)
385-386.
[12] Detke, M.J., Extinction of sequential conditioned inhibition,
Anita. Learning Behav., 19 (1991) 345-354.
[13] Armario, A., Gavalda, A. and Marti, O., Forced swimming test
in rats: Effect of desipramine administration and the period of
exposure to the test on struggling behavior, swimming, immobil-
ity and defecation rate, Eur. J. Pharmacol., 158 (1988) 207 212.
[ 14] Gorka, Z., Wojtasik, E., Kwiatek, H. and Maj, J., Action of seroto-
ninmimetics in the behavioral despair test in rats, Comm. Psycho-
pharmacol., 3 (1979) 133 136.
El5] Cervo, L., Grignaschi, G. and Samanin, R., ~2-adrenoceptor
blockade prevents the effect of desipramine in the forced swim-
ming test, Eur. Z Pharmacol., 175 (1990) 301-307.
[16] Kitada, Y., Miyauchi, T., Kanazawa, Y., Nakamichi, H. and
Satoh, S., Involvement of ~-and fll-adrenergic mechanisms in the
immobility-reducing action of desipramine in the forced swim-
ming test, Neuropharmacology, 22 (1983) 1055-1060.
[17] Lucki, I., Singh, A. and Kreiss, D.S., Antidepressant-like behav-
ioral effects of serotonin receptor agonists, Neurosci. Biobehav.
Rev., 18 (1994) 85-95.
[18] Delgado, P.L,, Miller, H.L., Salomon, R.M., Licinio, J., Hen-
inger, G.R., Gelenberg, A.J. and Charney, D.S., Monoamines and
the mechanism of antidepressant action: Effects of catecholamine
depletion on mood of patients treated with antidepressants,
Psychopharmacol. Bull., 29 (1993) 389-396.
[19] Delgado, P.L., Price, L.H., Miller, H.L., Salomon, R.M., Licinio,
J., Krystal, J.H., Heninger, G.R. and Charney, D.S., Rapid sero-
tonin depletion as a provocative challenge test for patients with
major depression: Relevance to antidepressant action and the
neurobiology of depression, Psychopharmacol. Bull., 27 (1991)
321 330.
[20] Katz, M.M., Maas, J.W., Frazer, A., Koslow, S.H., Bowden,
C.L., Berman, N., Swann, A.C. and Stokes, P.E., Drug-induced
actions on brain neurotransmitter systems and changes in the
behaviors and emotions of depressed patients, Neuropsycho-
pharmacology, 11, (1994) 89 100.