TOXICOLOGICAL SCIENCES 87(2), 451–459 (2005)
doi:10.1093/toxsci/kfi262
Advance Access publication July 20, 2005
Chronic Oral Treatment with 13-cis-Retinoic Acid (Isotretinoin)
or all-trans-Retinoic Acid Does Not Alter Depression-Like
Behaviors in Rats
Sherry A. Ferguson,*,1 F. Javier Cisneros,* B. Gough,* Joseph P. Hanig,† and Kimberly J. Berry*
*Division of Neurotoxicology, National Center for Toxicological Research/U.S. Food and Drug Administration, Jefferson, Arkasas 72079;
†Center for Drug Evaluation Research/U.S. Food and Drug Administration, Silver Spring, Maryland
Received June 14, 2005; accepted July 8, 2005
Oral treatment with the anti-acne drug Accutane (isotretinoin,
13-cis-retinoic acid) has been associated with suicide ideation and
depression. Here, depression-like behaviors (i.e., behavioral despair and anhedonia) were quantified in adult Sprague-Dawley
rats gavaged daily beginning at postnatal day (PND) 82 with 13-cisRA (7.5 or 22.5 mg/kg) or all-trans-retinoic acid (10 or 15 mg/kg ).
Tested at PND 130–131 in the Forced Swim Test, 7.5 mg/kg 13cis-RA marginally decreased immobility and slightly increased
climb/struggle durations whereas neither all-trans-retinoic acid
group differed from controls. Voluntary saccharin solution
(0.03%) intake at PND 102–104 and PND 151–153 was not
different from controls in any treated group, although all RAtreated groups had lower intakes. Swim speed in a water maze at
PND 180 was similar across groups, indicating no RA-induced
differences in physical ability. Open field activity was mildly
decreased at PND 91 in 7.5 mg/kg–treated males only, but it was
within the control range at PND 119, 147, and 175. Thus, at serum
levels similar to those in humans receiving the drug, chronic 13cis-RA treatment did not severely affect depression-like behaviors
in rats. These data do not substantiate the hypothesis of 13-cisRA-induced depression.
Key Words: isotretinoin; 13-cis-retinoic acid; all-trans-retinoic
acid; depression; anhedonia; locomotor activity; Forced Swim Test.
INTRODUCTION
Approved in 1982 to treat severe recalcitrant nodular acne,
13-cis-retinoic acid (13-cis-RA, isotretinoin, Accutane) is
described as ‘‘one of the greatest successes in dermatology’’
(Ellis and Krach, 2001). Soon after its approval, however,
reports emerged describing mood disorders, particularly depression and suicide ideation, associated with its use (Byrne
et al., 1998; Byrne and Hnatko, 1995; Hazen et al., 1983;
McCance-Katz and Price, 1992; O’Donnell, 2003; Scheinman
1
To whom correspondence should be addressed at Division of Neurotoxicology, NCTR/FDA, 3900 NCTR Road, Jefferson, AR 72079. Fax: 870–
543–7745. E-mail: [email protected].
Published by Oxford University Press 2005.
et al., 1990; Wooltorton, 2003; Wysowski et al., 2001).
Although many of those are case reports and there is human
evidence to the contrary (Ferahbas et al., 2004; Hersom et al.,
2003; Jacobs et al., 2001; Jick et al., 2000; Ng et al., 2002), the
American Academy of Dermatology recommended that significant efforts be expended to examine the possible association
between isotretinoin use and depression (Goldsmith et al.,
2004). Because of the potential confounds (e.g., racial and sex
biases) in 13-cis-RA use (Fleischer et al., 2003) and the
severity of the potential side effects, a definitive conclusion
regarding the ability of 13-cis-RA to induce depression may be
difficult to discern from clinical investigations. Such difficulties provide the justification for animal studies.
Two recent rodent studies illuminated potential mechanisms
of 13-cis-RA–induced depression (Crandall et al., 2004; Sakai
et al., 2004). In those studies, 13-cis-RA–treated mice exhibited suppressed hippocampal cell proliferation, neurogenesis, and survival at doses that approximated relevant human
serum levels. Decreased hippocampal neurogenesis and its
reversal by antidepressants are thought to be factors critically
involved in depression and the therapeutic efficacy of certain
drugs (Kempermann, 2002; Malberg et al., 2000; Malberg and
Schechter, 2005). Thus, a mechanism for depression potentially induced by 13-cis-RA treatment would appear to be
established. Should animal models also demonstrate behavioral
effects associated with 13-cis-RA treatment, this would lend
credence to their use in studies of depression in humans. As yet,
however, there are no studies directly linking hippocampal
neurogenesis and depressive behaviors in rodents.
Rodent depression measures generally evaluate behavioral
despair and/or anhedonia, two cardinal symptoms of human
depression, via (1) the Forced Swim Test (FST) and (2)
Voluntary Saccharin Intake. The FST, in which the rodent is
placed into an inescapable cylindrical tank of water, measures
duration of immobility or floating as the standard indicator of
depression (Porsolt et al., 1977). Decreased intake of highly
palatable solutions such as saccharin or sucrose solutions is
generally regarded as anhedonia. The FST and sucrose intake
452
FERGUSON ET AL.
have been pharmacologically validated (Nakamura and
Tanaka, 2001; Overstreet et al., 1995; Papp et al., 1996; Rex
et al., 2004; Sanchez and Meier, 1997), and the end points are
responsive to antidepressant treatment (Borsini and Meli, 1988;
Lucki, 1997; Papp et al., 1996). More importantly, however,
these tests have been shown to be sensitive to depressioninducing paradigms or treatments such as chronic mild stress
(Duncko et al., 2001; Tannenbaum et al., 2002; Willner, 1997)
and to rodent strains that are genetically prone to depression
(Ayensu et al., 1995; Overstreet et al., 1995; Pucilowski et al.,
1993; Yadid et al., 2000). Furthermore, increased immobility in
the FST has been demonstrated in relation to human pharmacological treatments such as neonatal cocaine or clomipramine exposure or adult MDMA treatment (Hansen-Trench and
Barron, 2005; Thompson et al., 2004; Vazquez-Palacios et al.,
2005). Other behaviors that have been associated with depression in rodents include decreased locomotor activity
(D’Aquila et al., 2000; Kopp et al., 1999) and loss of appetite
and body weight (Weiss, 1968).
Measures of depression in nonhuman animals are limited,
and the FST and voluntary saccharin intake are the most often
used. Other measures include learned helplessness, anhedonia
as measured by voluntary electrical brain stimulation, and tail
suspension. Willner has suggested that the learned helplessness
model may be more relevant to anxiety than to depression
(Willner, 1995), whereas Nestler et al. (2002) describe the
paradigm as more directly modeling posttraumatic stress
disorder. Decreased response rates for voluntary electrical
brain stimulation have been reported after exposure to either
chronic mild stress or inescapable shock (Nielsen et al., 2000;
Zacharko et al., 1983). However, there is some concern that
increased response rates, rather than decreased rates, may be
more reflective of anhedonia in that animals are responding more to achieve the same level of stimulation as prior
to the depression-inducing paradigm (i.e., inescapable shock)
(Anisman and Matheson, 2005). Tail suspension tests are not
typically conducted with rats.
The doses of 13-cis-RA (7.5 and 22.5 mg/kg) chosen for the
current study were based on previous research indicating that
7.5 mg/kg produces serum levels of 13-cis-RA comparable to
those of human Accutane users (Ferguson et al., unpublished
data). Groups treated with all-trans-RA were included because
13-cis-RA itself has low affinity for any of the RA receptors
(Crettaz et al., 1990; Kim et al., 1994) and its dermatological
effects are thought to be exerted via isomerization to all-transRA (Geiger et al., 1996). Doses for all-trans-RA were more
difficult to determine, as levels of this isomer are generally not
reported in studies of human Accutane users. However,
previous research had indicated that there were no body weight
or food/water intake alterations after a 7-day treatment period
in rats treated with the doses (10 and 15 mg/kg) used in the
present study.
Finally, both sexes were examined in our study to determine
sex-specific effects. Although there is no indication that 13-cis-
RA has such effects, there is ample evidence to suggest that
depression is much more common in women than in men
(Angst, 1998; Blazer et al., 1994; Ernst and Angst, 1992;
Fennig et al., 1994).
MATERIALS AND METHODS
Animals
Sixty weanling Sprague-Dawley rats (30 male and 30 female) were obtained
from the breeding colony at the National Center for Toxicological Research/
FDA and pair-housed with a same-sex rat in standard polycarbonate tub cages
lined with wood chip bedding on postnatal day (PND) 21 until PND 52, at
which time each was individually housed. Pelleted rat chow (NIH-31, Purina
Mills, St. Louis, MO) and water were provided ad libitum in the home cage.
Rats were maintained in vivariums at 22° ± 1°C (mean ± SEM) and 45–55%
humidity on a 12:12 hour light–dark cycle (lights on at 0700, off at 1900). All
animal procedures were approved by the NCTR Institutional Animal Care and
Use Committee.
Animal Treatments
Capsules (40 mg each) of 13-cis-RA (Accutane, Roche Pharmaceuticals,
Nutley, NJ) were obtained from Washington Wholesale Drug Exchange
(Savage, MD). Capsules were opened and their contents diluted in soybean
oil (ICN Biomedicals, Irvine, CA) to produce 5.25 and 15.75 mg/ml (7.5 and
22.5 mg/kg dose levels) suspensions. The all-trans-RA (Acros Organics,
Morris Plains, NJ) was similarly diluted to produce solutions to provide 7.0 and
10.5 mg/ml (10 and 15 mg/kg dose levels) suspensions. Suspensions were
prepared every 2 days and maintained at room temperature in amber glass
containers to prevent photoisomerization. All areas and rooms were fitted with
UV filters over fluorescent lights.
On PND 80–81, each rat was gavaged with soybean oil only to facilitate
adaptation to the gavage procedure. Rats were assigned to treatment groups
such that there were six animals/sex/dose group and gavaged daily with either
0 (soybean oil only), 7.5 (low), or 22.5 (high) mg/kg of 13-cis-RA, or 10 (low)
or 15 (high) mg/kg of all-trans-RA in a volume of 1.42 ml/kg beginning on
PND 82. All gavages were done in the morning hours and always prior to
behavioral testing.
Body Weight, and Food and Water Intake
Body weights and food intake were recorded daily beginning on PND 75 and
continued until sacrifice (i.e., at PND 185–189). Body weights were averaged
for each week. Food and water intake were standardized to body weight (i.e.,
g food/g body weight and ml water/g body weight) for each day and then
averaged for each week, including the 5 days prior to RA treatment. The week
prior to beginning RA treatment is expressed as baseline or week 0.
Behavioral Testing
Forced Swim Test (FST): procedure. Depression-like behaviors were
assessed on PND 130–131 using methodology similar to that originally
described by Porsolt et al. (1977), but modified by Cryan et al. (2005). Rats
were gavaged 30–100 min prior to testing and transported to the testing room
immediately prior to testing. Order of testing was randomized with respect to
treatment and sex. White noise of approximately 60 dB (AM radio set between
stations) buffered extraneous noise. The apparatus was a clear Plexiglas
cylinder (height ¼ 61 cm; diameter ¼ 29 cm) filled with 50 cm of fresh tap
water, maintained at 25° ± 1°C. White noise masked background noise. Each
rat was placed into the cylinder on 2 consecutive days for 10 min (PND 130) or
5 min (PND 131), after which time it was removed, gently towel dried, and
returned in its cage to the housing room. Early removal from the swim test (i.e.,
453
ISOTRETINOIN AND DEPRESSION-LIKE BEHAVIORS IN RATS
rescue) was necessary for only a few rats and only if it appeared that the animal
would not survive the session. Each test cylinder was emptied, rinsed, and
refilled after each individual to eliminate the potential for an alarm substance
pheromone to affect behavior of the subsequent rat (Abel, 1991, 1992).
Sessions were videotaped for later scoring. For each session, the total duration
of three behaviors was measured: (1) immobility or floating (i.e., making only
those movements necessary to keep the head above the water), (2) climb/
struggle (i.e., making active movements with the forepaws in and out of the
water, usually directed against the wall), and (3) swim (i.e., making active
swimming motions, more than necessary to merely keep head above the water,
typically in a horizontal posture). Frequency of dives and fecal boli were also
recorded. Two trained testers scored each test session. Prior to statistical
analyses, data from each of the two testers were averaged for each session.
Inter-rater reliabilities ranged from 0.74 for duration of swim to 0.83 for
duration of climb/struggle.
FST: behavioral data. Data from the six rescued rats (see Results) were
deleted prior to analyses because their behavioral durations would not be
comparable to those of rats that remained in the swim test for the entire session.
Saccharin Solution Intake
For 3 consecutive days during the periods PND 102–104 and PND 151–153,
intake of a sweet solution containing 0.03% saccharin (as the sodium salt,
1.46 mM solution) (ICN Biochemicals Inc., Aurora, OH) in water was
measured by placing two bottles on each rat’s cage, one containing only water
and the other containing the saccharin solution. The position of the bottles
remained the same throughout both tests and for all rats. Bottles were weighed
once daily. The amount consumed in milliliters per day was then divided by
body weight to yield ml/day/body weight.
Open Field Activity
Horizontal locomotor activity was assessed on PND 91, 119, 147, and 175.
Rats were gavaged from 30–190 min prior to testing. Activity was measured for
individual rats in a Plexiglas cube (46.5 3 46.5 3 46.5 cm) bisected with
photobeams. Session duration was 12 min, and each subject was tested in
lighted conditions during the light period of the animals’ circadian cycle. White
noise of approximately 60 dB (AM radio set between stations) buffered
extraneous noise. Activity was recorded as the total number of photobeam
breaks and duration of freezing (no photobeam breaks >1 min).
Swim Speed
Swim speed was assessed in each rat on PND 180. The apparatus was
a circular stainless-steel tank (183 cm interior diameter) with a black interior
filled to a depth of 34 cm with water that was made opaque by the addition of
powdered black paint. White noise of approximately 60 dB (AM radio set
between stations) buffered extraneous noise. Each rat was tested at one of four
starting locations randomly assigned for one trial of 120 s. Swim speed, percent
of trial in slow swim (% time slow), and percent of trial in thigmotaxis (% thig
time) were recorded by a video-tracking/computer digitizing system (HVS
Image, Hampton, UK). The apparatus and tracking system are those typically
used for Morris water maze testing. Here, however, there was no escape
platform.
Statistical Analyses
Effects of all-trans-RA or 13-cis-RA treatment were analyzed in separate
analyses of variance (ANOVAs). Body weight and food/water intake were
analyzed via repeated measures ANOVAs with treatment, sex, and week as
factors. The FST behaviors and open field activity were analyzed in separate
ANOVAs with treatment, sex, and day (day 1 or day 2 for FST and PND 91,
199, 147, or 175 for open field) as factors. Saccharin intake was analyzed in
separate ANOVAs with treatment, sex, day (day 1, 2, or 3 of each test), and test
(first test at PND 102–104, second test at PND 151–153) as factors. Finally,
swim speed was analyzed in separate ANOVAs with treatment and sex as
factors.
RESULTS
Body Weight, Food and Water Intake, and General
Health Observations
13-cis-RA. Although there was a significant interaction
between sex 3 week (F(15,450) ¼ 74.39, p < .0001) on body
weight, post-hoc analyses simply indicated that males weighed
more than females in all weeks. There was no main effect of
treatment, nor were there any significant interactions with
treatment. Food intake analyses indicated a significant interaction of sex 3 week (F(15,450) ¼ 1.74, p < .05; however,
females consumed more than males at each week as expressed as
g food/g body weight. Water intake analyses were similar in that
there was a significant sex 3 week interaction (F(15,449) ¼ 4.15,
p < .0001), and post-hoc analyses indicated that females drank
more at each week as expressed as ml water/g body weight.
All-trans-RA. Although there was a significant interaction
of treatment 3 sex 3 week (F(30,449) ¼ 2.02, p < .002) on
body weight, post-hoc analyses indicated no significant differences in any week between female controls and same-sex alltrans-RA groups; nor were there any significant differences
between male controls and same-sex all-trans-RA groups.
Food intake analyses indicated a significant interaction of sex
and week (F(15,448) ¼ 2.70, p < .0007; however, females
consumed more than males in each week, as expressed as g
food/g body weight. There was a significant interaction
between sex and week and the analysis of water intake
(F(15,448) ¼ 2.02, p < .02), and post-hoc analyses indicated
that females drank more than males during all weeks, except
weeks 2, 4, and 11.
Home cage observations indicated that some all-transRA–treated rats had bone fractures, an observation that was
later confirmed at sacrifice. Seven rats (one male and six
females, of which three were treated with 10 mg/kg and four
with 15 mg/kg) had fractures of the tibia or humerus. None of
these rats had been previously rescued during the FST, and all
had swim speed data within the range of controls. Similarly,
locomotor activity in the open field appeared unaffected by
these fractures. Full details of the fractures and the effect of
all-trans-RA on bone metabolism and density will be available
in a later publication (Hotchkiss et al., in press).
Forced Swim Test Behavior
13-cis-RA. Five male rats required rescue. These were two
control animals on day 1, one 7.5 mg/kg 13-cis-RA on day 1,
one control on day 2, and one 22.5 mg/kg 13-cis-RA on day 2.
No females required rescue. Figure 1 illustrates durations of
immobility and climb/struggle averaged over the two test days.
There was a significant interaction between sex and day
(F(1,25) ¼ 8.99, p < .007) and a marginally significant effect
of treatment (F(2,29) ¼ 2.83, p < .08) on duration of
immobility. Post-hoc tests indicated that both sexes were
significantly less immobile on day 2 than on day 1 (p < .05).
454
FERGUSON ET AL.
p < .02) and a marginally significant effect of treatment
(F(2,31) ¼ 3.29, p < .06) on duration of immobility. Post-hoc
tests indicated that both males and females exhibited less
immobility on day 2 than on day 1 (p < .05). The treatment
effect indicated that rats treated with 15 mg/kg all-transRA were somewhat more immobile than rats treated with
10 mg/kg. Analysis of duration of climb/struggle indicated
a significant interaction between sex and day (F(1,24) ¼ 7.19,
p < .02), and post-hoc tests indicated that females exhibited
much less climb/struggle behavior on day 2 than on day 1,
and less than males on day 2. Main effects of sex (F(1,31) ¼
8.39, p < .007) and day (F(1,24) ¼ 23.10, p < .0001) on
duration of swim indicated that females swam longer than
males and there was more swimming behavior exhibited on
day 1 than on day 2. A significant interaction of sex 3 day on
dive frequency (F(1,30) ¼ 5.63, p < .03) indicated that males
exhibited dives more frequently than females on day 1 ( p <
.05), and dive frequency for males was higher on day 1 than
on day 2 (p < .05). The significant effect of sex on number of
fecal boli (F(1,30) ¼ 12.67, p < .002) indicated that females
had higher numbers of fecal boli than males.
Saccharin Solution Intake
FIG. 1. Behavioral durations (mean ± SEM) in the FST (averaged over the
two test sessions) after 48 consecutive days of RA treatment. RA treatment did
not significantly alter duration of immobility or climb/struggle duration. Top:
Duration of immobility. Bottom: Duration of climb/struggle.
Analysis of the duration of climb/struggle indicated a significant interaction between sex and day (F(1,25) ¼ 8.64,
p < .007). Post-hoc tests indicated that females performed
significantly less climb/struggle on day 2 than on day 1 ( p <
.05), but the durations for males did not differ between days. A
main effect of day on duration of swim (F(1,25) ¼ 12.37, p <
.002) indicated less swimming behavior on day 2 than on day 1.
Analysis of frequency of dives indicated a significant interaction between treatment and day (F(2,25) ¼ 3.42, p < .05)
and although there was more diving on day 1 than on day 2, this
was significant only for the 22.5 mg/kg 13-cis-RA group ( p <
.05). There were no significant effects on number of fecal boli.
All-trans-RA. One male rat treated with 10 mg/kg all-transRA required rescue on day 2. No females required rescue.
There was an interaction between sex and day (F(1,24) ¼ 6.39,
13-cis-RA. Figure 2 illustrates saccharin in solution intake
by RA treatment and sex averaged over the two tests. There
were no significant treatment effects on voluntary saccharin
solution intake. A significant interaction between sex and day
(F(2,147) ¼ 3.89, p < .03) indicated that females consume
more saccharin on the first day of each test than do males on
the first day and that males consume more than females on the
second and third days of each test (p < .05). Intake on the
second day was also higher than intake on the third day for
females only ( p < .05); intake over the three days of each
test did not differ for males.
All-trans-RA. A main effect of sex (F(1,30) ¼ 13.54, p <
.0009) indicated increased saccharin intake in females while
a main effect of day (F(2,148) ¼ 9.95, p < .0001) indicated
higher intake on the first day than on the second and third days
(p < .05).
Open Field Activity
13-cis-RA. Figure 3 illustrates open field activity by RA
treatment, sex and age. There was a significant interaction of
treatment 3 sex 3 age (F(6,88) ¼ 3.71, p < .003) on total
activity, and post-hoc tests indicated that control males were
more active than 7.5 mg/kg 13-cis-RA males at PND 91 (p <
.05). Additionally, activity of control males at PND 91 was
higher than at PND 147 and 175 (p < .05). Analysis of freeze
duration indicated a similar interaction of treatment 3 sex 3 age
(F(6,89) ¼ 3.48, p < .004), and post-hoc tests reflected similar
results: lower durations of freezing behavior in control males at
PND 91 than in 7.5 mg/kg 13-cis-RA males of the same age, as
well as in control males at PND 147 and 175 (p < .05).
ISOTRETINOIN AND DEPRESSION-LIKE BEHAVIORS IN RATS
455
FIG. 2. Voluntary saccharin intake (0.03% solution) (mean ± SEM) (males, left; females, right) (averaged over the two tests [PNDs 102–104 and 151–153]).
Although all RA-treated groups consumed less than controls, there were no statistically significant differences. Females consumed significantly more than males
(ml/g body weight) (note scale differences) and there was significantly higher intake on day 1 than subsequent days.
All-trans-RA. There was a marginal effect of treatment on
total activity (F(2,32) ¼ 2.63, p < .09) which indicated that the
10 and 15 mg/kg all-trans-RA groups had somewhat less
activity overall than did controls. A significant main effect of
age (F(3,87) ¼ 15.66, p < .0001) indicated higher levels of total
activity at PND 91 in all groups and sexes than all other ages
and higher levels at PND 119 than at PND 147 or 175 ( p < .05).
Post-hoc tests of the main effects of treatment (F(2,32) ¼ 3.72,
p < .04), sex (F(1,32) ¼ 6.93, p < .02), and age (F(3,86) ¼
12.55, p < .0001) on freeze duration indicated several effects:
no treatment group was significantly different from control,
males had a lower duration of freezing than females (p < .05),
and freeze duration at PNDs 91 and 119 was lower than at
PNDs 147 and 175 (p < .05).
Swim Speed
13-cis-RA. There were no significant effects of 13-cis-RA
treatment on swim speed (m/s), % time slow, or % thigmotaxis
time (Table 1). There was a main effect of sex on % time slow
(F(1,31) ¼ 6.52, p < .02), which indicated increased duration
of time spent in slow swimming in females relative to males.
All-trans-RA. A main effect of sex on swim speed (F(1,32) ¼
4.94, p < .04), % time slow (F(1,32) ¼ 17.93, p < .0002), and
% thigmotaxis time (F(1,32) ¼ 4.33, p < .05) indicated slower
swim speeds and more time in slow swimming in females while
increased thigmotaxis time in males (Table 1).
DISCUSSION
FIG. 3. Open field activity across PNDs 90–180 for all RA treatment
groups (mean ± SEM) (males, top; females, bottom). At PND 91, the 7.5 mg/kg
13-cis-RA-treated males were significantly less active than same-sex controls.
Published reports suggesting the potential of Accutane to
induce depression in humans led to the proposed hypothesis
456
FERGUSON ET AL.
TABLE 1
Swimming abilities in 13-cis-RA and all-trans-RA treated rats
Swim speed (m/s)
% Time slow
% Thigmotaxis
time
Control males
7.5 mg/kg 13-cis-RA
males
22.5 mg/kg 13-cis-RA
males
10 mg/kg all-trans-RA
males
15 mg/kg all-trans-RA
males
0.2282 ± 0.0130
0.2482 ± 0.0115
13.0 ± 1.6
14.1 ± 1.5
70.2 ± 4.9
66.2 ± 2.6
0.2552 ± 0.0163
11.4 ± 1.2
63.1 ± 7.3
0.2400 ± 0.0083
13.6 ± 1.8
72.2 ± 6.3
0.2495 ± 0.0224
12.2 ± 1.3
65.5 ± 6.7
Control females
7.5 mg/kg 13-cis-RA
females
22.5 mg/kg 13-cis-RA
females
10 mg/kg all-trans-RA
females
15 mg/kg all-trans-RA
females
0.2182 ± 0.0128
0.2424 ± 0.0195
19.5 ± 2.7
14.7 ± 2.0
52.3 ± 5.8
67.3 ± 4.8
0.2483 ± 0.0116
17.2 ± 2.8
54.1 ± 5.9
0.2095 ± 0.0216
21.3 ± 3.5
64.3 ± 7.5
0.1865 ± 0.0283
29.9 ± 5.5
57.5 ± 8.1
Treatment group
that 13-cis-RA treatment in rats would cause increased
quantitative signs of depression in validated behavioral assessments. Contrary to that hypothesis, neither 7.5 nor 22.5 mg/kg
13-cis-RA treatment caused any significant signs of depression
in either the forced swim test (FST) or the measure of
anhedonia, voluntary saccharin intake. Groups treated with
all-trans-RA also were relatively unaffected by treatment.
Furthermore, neither RA-treated group exhibited alterations in
body weight or food and water intake.
These results suggest that there are no severe effects of
13-cis-RA on behavioral indices of depression. The FST
methodology used here was more complete than the typical
single measure of immobility and followed the methodology
proposed by Cryan et al. (2005) in that three behaviors
(immobility, climb/struggle, and swim) were measured for
each session. Measurement of behaviors other than immobility
is particularly important because drugs with different mechanisms of action can differentially alter behaviors (Lucki, 1997).
However, the direction of the effects observed were opposite
those originally predicted in that if 13-cis-RA treatment were
hypothesized to cause depression, then increased immobility
and decreased climb/struggle durations should have been
exhibited in the FST. Yet, this was not the case. Much has
been made of the sensitivity of the FST and whether immobility
represents an adaptive response to the environment (Hawkins
et al., 1978; Nishimura et al., 1988; Thierry et al., 1984) and
contributes to survival (Bruner and Vargas, 1994) or is a
reflection of ‘‘behavioral despair’’ (Porsolt et al., 1977). Some
investigators have questioned the ethological relevance of
these standard measures of depression (Mitchell and Redfern,
2005), suggesting that other tests may be more appropriate.
Regardless, neither RA-treated group exhibited severely altered FST behaviors. Furthermore, the lack of significant 13cis-RA effects on forced swim behaviors has recently been
replicated with rats treated with 7.5 and 30 mg/kg 13-cis-RA
(Ferguson, unpublished data).
Like the results of the FST, anhedonia as measured by
voluntary saccharin intake was not significantly affected by 13cis-RA or all-trans-RA treatment, although all treated groups
had lower intakes relative to controls. Locomotor open field
activity was only marginally affected in that males treated with
7.5 mg/kg had lower levels of activity than did control males
and only at the earliest test age (PND 91), a finding that is likely
a spurious effect. Finally, swim speed was unaffected by RA
treatment, even when such treatment (e.g., all-trans-RA)
caused bone fractures.
One relevant question is whether the tests as used here would
have been sensitive to RA treatment effects. Certainly,
significant effects of sex and session were apparent and in
the same direction as expected. For example, females are
reported to struggle more and exhibit decreased immobility in
the FST (Alonso et al., 1991; Barros and Ferigolo, 1998; Brotto
et al., 2000), and relatively similar behaviors have been noted
here. Dive frequency was less on the second FST session here
as has been reported (Barros and Ferigolo, 1998; Hawkins
et al., 1978). During the 3-day saccharin tests, all rats
consumed more on the first day and females consumed more
than males as previously described (Ferguson et al., 2000;
Flynn et al., 2001). Finally, decreased open field activity with
increasing age is a classical finding (Asano, 1986; Furchtgott
et al., 1961). Thus, irrespective of RA treatment, the significant
results are similar to those previously described.
The two measures of depression used here, FST and
voluntary saccharin intake, are thought to model ‘‘behavioral
despair’’ and anhedonia, respectively, How well these two
measures model the human symptoms of major depressive
disorder is intertwined with complex issues regarding the
utility of animal models of depression. For example, Matthews
et al. (2005) argued that nonhuman animals may be unable to
model many of the symptoms of depression. Recognizing the
complexity and co-morbidity associated with depression,
Frazer and Morilak (2005) suggest that animal models would
serve better to reflect three specific behavioral dimensions of
human depression, rather than attempting to model the entire
disorder. They propose that symptoms such as helplessness/
worthlessness be categorized into a negative affect dimension
while anhedonia be classified in the loss of positive affect
dimension. One advantage of this system is that the dimensions
are likely to be regulated by different neural systems, and
pharmacological treatments may affect one dimension while
having little or no effect on others. Similarly, treatments or
paradigms that induce depression in animal models may
increase symptomatology within one dimension only. If
behavioral despair as experienced in the FST is reflective of
ISOTRETINOIN AND DEPRESSION-LIKE BEHAVIORS IN RATS
helplessness/worthlessness, then the current study measured
two of the three dimensions proposed by Frazer and Morilak.
Such relatively global assessments make the conclusions even
more solid.
Despite the questions surrounding the validity of the model,
the nature of the findings was somewhat surprising. Although
there was not overwhelming evidence to suggest that 13-cisRA–induced behavioral alterations would appear, the information certainly was suggestive. For example, aged mice exhibit
reduced hippocampal expression of specific RA receptors
mRNAs relative to adult mice as well as impairments in spatial
learning and memory (Etchamendy et al., 2001). Further, RA
treatment (0.15 mg/kg, injected subcutaneously) reverses the
cognitive impairment those aged mice (Etchamendy et al.,
2001). However, 13-cis-RA treatment (1 mg/kg, injected
intraperitoneally) during young adulthood is associated with
both suppressed hippocampal cell proliferation and decreased
spatial learning and memory in mice (Crandall et al., 2004).
Thus, the lack of behavioral alterations in the current study was
surprising, particularly given the doses used here (7.5 and 22.5
mg/kg, oral gavage). The different routes of administration are
of little concern since we have documented the serum levels
achieved by these doses (Ferguson et al., unpublished data). It
might be hypothesized that 13-cis-RA treatment specifically
affects learning and memory while leaving depressive behaviors in rodents within control limits. However, we have recent
data indicating that spatial learning and memory in 13-cis-RA
treated rats is intact (Ferguson, unpublished data). Thus, at
present, species differences in RA sensitivity appear to be the
most likely candidate explanation, with rats much less sensitive
than mice.
Oral treatment of rats with 13-cis-RA that produced serum
levels similar to human therapeutic levels did not affect
traditional measures of behavioral despair and anhedonia.
Similarly, all-trans-RA had few effects on rat behavior, and
neither 13-cis-RA nor all-trans-RA affected body weight or
food and water intake. Open field locomotor activity was only
marginally affected. Although other end points might reveal
13-cis-RA effects in rats, it seems clear that RA does not
severely alter depression-like behaviors.
REFERENCES
Abel, E. L. (1991). Gradient of alarm substance in the Forced Swimming Test.
Physiol. Behav. 49, 321–323.
Abel, E. L. (1992). Response to alarm substance in different rat strains. Physiol.
Behav. 51, 345–347.
Alonso, S. J., Castellano, M. A., Afonso, D., and Rodriguez, M. (1991). Sex
differences in behavioral despair: Relationships between behavioral despair
and open field activity. Physiol. Behav. 49, 69–72.
Angst, J. (1998). Treated versus untreated major depressive episodes.
Psychopathology 31, 37–44.
Anisman, H., and Matheson, K. (2005). Stress, depression, and anhedonia:
Caveats concerning animal models. Neurosci. Biobehav. Rev. 29,
525–546.
457
Asano, Y. (1986). Characteristics of open field behavior of Wistar and SpragueDawley rats. Jikken Dobutsu 35, 505–508.
Ayensu, W. K., Pucilowski, O., Mason, G. A., Overstreet, D. H., Rezvani, A. H.,
and Janowsky, D. S. (1995). Effects of chronic mild stress on serum
complement activity, saccharin preference, and corticosterone levels in
Flinders lines of rats. Physiol. Behav. 57, 165–169.
Barros, H. M., and Ferigolo, M. (1998). Ethopharmacology of imipramine in
the forced-swimming test: Gender differences. Neurosci. Biobehav. Rev. 23,
279–286.
Blazer, D. G., Kessler, R. C., McGonagle, K. A., and Swartz, M. S. (1994). The
prevalence and distribution of major depression in a national community
sample: The National Comorbidity Survey. Am. J. Psychiatry 151, 979–986.
Borsini, F., and Meli, A. (1988). Is the forced swimming test a suitable model
for revealing antidepressant activity? Psychopharmacology (Berl.) 94,
147–160.
Brotto, L. A., Barr, A. M., and Gorzalka, B. B. (2000). Sex differences in
forced-swim and open-field test behaviours after chronic administration of
melatonin. Eur. J. Pharmacol. 402, 87–93.
Bruner, C. A., and Vargas, I. (1994). The activity of rats in a swimming
situation as a function of water temperature. Physiol. Behav. 55, 21–28.
Byrne, A., Costello, M., Greene, E., and Zibin, T. (1998). Isotretinoin therapy and
depression—Evidence for an association. Irish J. Psychol. Med. 15, 58–60.
Byrne, A., and Hnatko, G. (1995). Depression associated with isotretinoin
therapy. Can. J. Psychiatry 40, 567.
Crandall, J., Sakai, Y., Zhang, J., Koul, O., Mineur, Y., Crusio, W. E., and
McCaffery, P. (2004). 13-cis-retinoic acid suppresses hippocampal cell
division and hippocampal-dependent learning in mice. Proc. Natl. Acad. Sci.
U. S. A. 101, 5111–5116.
Crettaz, M., Baron, A., Siegenthaler, G., and Hunziker, W. (1990). Ligand
specificities of recombinant retinoic acid receptors RAR alpha and RAR
beta. Biochem. J. 272, 391–397.
Cryan, J. F., Valentino, R. J., and Lucki, I. (2005). Assessing substrates
underlying the behavioral effects of antidepressants using the modified rat
forced swimming test. Neurosci. Biobehav. Rev. 29, 547–569.
D’Aquila, P. S., Peana, A. T., Carboni, V., and Serra, G. (2000). Exploratory
behaviour and grooming after repeated restraint and chronic mild stress:
Effect of desipramine. Eur. J. Pharmacol. 399, 43–47.
Duncko, R., Kiss, A., Skultetyova, I., Rusnak, M., and Jezova, D. (2001).
Corticotropin-releasing hormone mRNA levels in response to chronic mild
stress rise in male but not in female rats while tyrosine hydroxylase mRNA
levels decrease in both sexes. Psychoneuroendocrinology 26, 77–89.
Ellis, C. N., and Krach, K. J. (2001). Uses and complications of isotretinoin
therapy. J. Am. Acad. Dermatol. 45, S150–S157.
Ernst, C., and Angst, J. (1992). The Zurich Study. XII. Sex differences in
depression. Evidence from longitudinal epidemiological data. Eur. Arch.
Psychiatry Clin. Neurosci. 241, 222–230.
Etchamendy, N., Enderlin, V., Marighetto, A., Vouimba, R. M., Pallet, V.,
Jaffard, R., and Higueret, P. (2001). Alleviation of a selective age-related
memory deficit in mice by pharmacologically-induced normalization of the
brain retinoid signalling. J. Neurosci. 21, 6423–6429.
Fennig, S., Schwartz, J. E., and Bromet, E. J. (1994). Are diagnostic criteria,
time of episode and occupational impairment important determinants of the
female:male ratio for major depression? J. Affect. Disord. 30, 147–154.
Ferahbas, A., Turan, M. T., Esel, E., Utas, S., Kutlugun, C., and Kilic, C. G.
(2004). A pilot study evaluating anxiety and depressive scores in acne
patients treated with isotretinoin. J. Dermatolog. Treat. 15, 153–157.
Ferguson, S. A., Flynn, K. M., Delclos, K. B., and Newbold, R. R. (2000).
Maternal and offspring toxicity but few sexually dimorphic behavioral
alterations result from nonylphenol exposure. Neurotoxicol. Teratol. 22,
583–591.
458
FERGUSON ET AL.
Fleischer, A. B., Jr., Simpson, J. K., McMichael, A., and Feldman, S. R. (2003).
Are there racial and sex differences in the use of oral isotretinoin for acne
management in the United States? J. Am. Acad. Dermatol. 49, 662–666.
Flynn, K. M., Delclos, K. B., Newbold, R. R., and Ferguson, S. A. (2001).
Behavioral responses of rats exposed to long-term dietary vinclozolin.
J. Agric. Food Chem. 49, 1658–1665.
Frazer, A., and Morilak, D. A. (2005). What should animal models of
depression model? Neurosci. Biobehav. Rev. 29, 515–523.
Furchtgott, E., Wechkin, S., and Dees, J. W. (1961). Open-field exploration as
a function of age. J. Comp. Physiol. Psychol. 54, 386–388.
Geiger, J. M., Hommel, L., Harms, M., and Saurat, J. H. (1996). Oral 13-cis
retinoic acid is superior to 9-cis retinoic acid in sebosuppression in human
beings. J. Am. Acad. Dermatol. 34, 513–515.
Goldsmith, L. A., Bolognia, J. L., Callen, J. P., Chen, S. C., Feldman, S. R.,
Lim, H. W., Lucky, A. W., Reed, B. R., Siegfried, E. C., Thiboutot, D. M.,
and Wheeland, R. G. (2004). American Academy of Dermatology Consensus
Conference on the safe and optimal use of isotretinoin: Summary and
recommendations. J. Am. Acad. Dermatol. 50, 900–906.
Nakamura, K., and Tanaka, Y. (2001). Antidepressant-like effects of aniracetam
in aged rats and its mode of action. Psychopharmacology (Berl.) 158,
205–212.
Nestler, E. J., Gould, E., Manji, H., Buncan, M., Duman, R. S., Greshenfeld,
H. K., Hen, R., Koester, S., Lederhendler, I., Meaney, M., et al. (2002).
Preclinical models: Status of basic research in depression. Biol. Psychiatry
52, 503–528.
Ng, C. H., Tam, M. M., Celi, E., Tate, B., and Schweitzer, I. (2002). Prospective
study of depressive symptoms and quality of life in acne vulgaris patients
treated with isotretinoin compared to antibiotic and topical therapy.
Australas. J. Dermatol. 45, 262–268.
Nielsen, C. K., Arnt, J., and Sanchez, C. (2000). Intracranial self-stimulation
and sucrose intake differ as hedonic measures following chronic mild
stress: Interstrain and interindividual differences. Behav. Brain Res. 107,
21–33.
Nishimura, H., Tsuda, A., Oguchi, M., Ida, Y., and Tanaka, M. (1988). Is
immobility of rats in the forced swim test ‘‘behavioral despair’’? Physiol.
Behav. 42, 93–95.
Hansen-Trench, L. S., and Barron, S. (2005). Effects of neonatal alcohol and/or
cocaine exposure on stress in juvenile and adult female rats. Neurotoxicol.
Teratol. 27, 55–63.
O’Donnell, J. (2003). Overview of existing research and information linking
isotretinoin (Accutane), depression, psychosis, and suicide. Am. J. Ther. 10,
148–159.
Hawkins, J., Hicks, R. A., Phillips, N., and Moore, J. D. (1978). Swimming rats
and human depression. Nature 274, 512–513.
Overstreet, D. H., Pucilowski, O., Rezvani, A. H., and Janowsky, D. S. (1995).
Administration of antidepressants, diazepam and psychomotor stimulants
further confirms the utility of Flinders Sensitive Line rats as an animal model
of depression. Psychopharmacology (Berl.) 121, 27–37.
Hazen, P. G., Carney, J. F., Walker, A. E., and Stewart, J. J. (1983).
Depression—A side effect of 13-cis-retinoic acid therapy. J. Am. Acad.
Dermatol. 9, 278–279.
Hersom, K., Neary, M. P., Levaux, H. P., Klaskala, W., and Strauss, J. S. (2003).
Isotretinoin and antidepressant pharmacotherapy: A prescription sequence
symmetry analysis. J. Am. Acad. Dermatol. 49, 424–432.
Hotchkiss, C. E., Latendresse, J. R., and Ferguson, S. A. (2005). Oral treatment
with 13-cis- and all-trans-retinoic acid in male and female adult rats. Bone
In press.
Jacobs, D. G., Deutsch, N. L., and Brewer, M. (2001). Suicide, depression, and
isotretinoin: Is there a causal link? J. Am. Acad. Dermatol. 45, S168–S175.
Jick, S. S., Kremers, H. M., and Vasilakis-Scaramozza, C. (2000). Isotretinoin
use and risk of depression, psychotic symptoms, suicide, and attempted
suicide. Arch. Dermatol. 136, 1231–1236.
Papp, M., Moryl, E., and Willner, P. (1996). Pharmacological validation of
the chronic mild stress model of depression. Eur. J. Pharmacol. 296,
129–136.
Porsolt, R. D., Le Pichon, M., and Jalfre, M. (1977). Depression: A new animal
model sensitive to antidepressant treatments. Nature 266, 730–732.
Pucilowski, O., Overstreet, D. H., Rezvani, A. H., and Janowsky, D. S. (1993).
Chronic mild stress-induced anhedonia: Greater effect in a genetic rat model
of depression. Physiol. Behav. 54, 1215–1220.
Rex, A., Schickert, R., and Fink, H. (2004). Antidepressant-like effect of
nicotinamide adenine dinucleotide in the forced swim test in rats.
Pharmacol. Biochem. Behav. 77, 303–307.
Kempermann, G. (2002). Regulation of adult hippocampal neurogenesis—
Implications for novel theories of major depression. Bipolar Disord. 4, 17–33.
Sakai, Y., Crandall, J. E., Brodsky, J., and McCaffery, P. (2004). 13-cis Retinoic
acid (Accutane) suppresses hippocampal cell survival in mice. Ann. N. Y.
Acad. Sci. 1021, 436–440.
Kim, Y. W., Sharma, R. P., and Li, J. K. (1994). Characterization of
heterologously expressed recombinant retinoic acid receptors with natural
or synthetic retinoids. J. Biochem. Toxicol. 9, 225–234.
Sanchez, C., and Meier, E. (1997). Behavioral profiles of SSRIs in animal
models of depression, anxiety and aggression. Are they all alike? Psychopharmacology (Berl.) 129, 197–205.
Kopp, C., Vogel, E., Rettori, M. C., Delagrange, P., and Misslin, R. (1999). The
effects of melatonin on the behavioral disturbances induced by chronic mild
stress in C3H/He mice. Behav. Pharmacol. 10, 73–83.
Scheinman, P. L., Peck, G. L., Rubinow, D. R., DiGiovanna, J. J., Abangan,
D. L., and Ravin, P. D. (1990). Acute depression from isotretinoin. J. Am.
Acad. Dermatol. 22, 1112–1114.
Tannenbaum, B., Tannenbaum, G. S., Sudom, K., and Anisman, H. (2002).
Neurochemical and behavioral alterations elicited by a chronic intermittent stressor regimen: Implications for allostatic load. Brain Res. 953,
82–92.
Lucki, I. (1997). The forced swimming test as a model for core and component
behavioral effects of antidepressant drugs. Behav. Pharmacol. 8, 523–532.
Malberg, J. E., Eisch, A. J., Nestler, E. J., and Duman, R. S. (2000). Chronic
antidepressant treatment increases neurogenesis in adult rat hippocampus.
J. Neurosci. 20, 9104–9110.
Malberg, J. E., and Schechter, L. E. (2005). Increasing hippocampal neurogenesis: A novel mechanism for antidepressant drugs. Curr. Pharm. Des. 11,
145–155.
Matthews, K., Christmas, D., Swan, J., and Sorrell, E. (2005). Animal models
of depression: Navigating through the clinical fog. Neurosci. Biobehav. Rev.
29, 503–513.
McCance-Katz, E. F., and Price, L. H. (1992). Depression associated with
vitamin A intoxication. Psychosomatics 33, 117–118.
Mitchell, P. J., and Redfern, P. H. (2005). Animal models of depressive illness:
The importance of chronic drug treatment. Curr. Pharm. Des. 11, 171–203.
Thierry, B., Steru, L., Chermat, R., and Simon, P. (1984). Searching-waiting
strategy: A candidate for an evolutionary model of depression? Behav.
Neural. Biol. 41, 180–189.
Thompson, M. R., Li, K. M., Clemens, K. J., Gurtman, C. G., Hunt, G. E.,
Cornish, J. L., and McGregor, I. S. (2004). Chronic fluoxetine treatment partly
attenuates the long-term anxiety and depressive symptoms induced by
MDMA (‘Ecstasy’) in rats. Neuropsychopharmacology 29, 694–704.
Vazquez-Palacios, G., Bonilla-Jaime, H., and Velazquez-Moctezuma, J. (2005).
Antidepressant effects of nicotine and fluoxetine in an animal model of
depression induced by neonatal treatment with clomipramine. Prog. Neuropsychopharmacol. Biol. Psychiatry 29, 39–46.
ISOTRETINOIN AND DEPRESSION-LIKE BEHAVIORS IN RATS
Weiss, J. M. (1968). Effects of coping responses on stress. J. Comp. Physiol.
Psychol. 65, 251–260.
Willner, P. (1995). Animal models of depression: Validity and applications. In
Depression and Mania: From Neurobiology to Treatment (G. Gessa, Fratta, W.,
Pani, L., Serra, G., Eds.), pp. 10–41. Raven Press, New York.
Willner, P. (1997). Validity, reliability and utility of the chronic mild stress
model of depression: A 10-year review and evaluation. Psychopharmacology
(Berl.) 134, 319–329.
Wooltorton, E. (2003). Accutane (isotretinoin) and psychiatric adverse effects.
Can. Med. Assoc. J. 168, 66.
459
Wysowski, D. K., Pitts, M., and Beitz, J. (2001). An analysis of reports of
depression and suicide in patients treated with isotretinoin. J. Am. Acad.
Dermatol. 45, 515–519.
Yadid, G., Nakash, R., Deri, I., Tamar, G., Kinor, N., Gispan, I., and Zangen, A.
(2000). Elucidation of the neurobiology of depression: Insights from a novel
genetic animal model. Prog. Neurobiol. 62, 353–378.
Zacharko, R. M., Bowers, W. J., Kokkinidis, L., and Anisman, H. (1983).
Region-specific reductions of intracranial self-stimulation after uncontrollable stress: Possible effects on reward processes. Behav. Brain Res. 9,
129–141.