Neurobiology of Learning and Memory 87 (2007) 209–217

www.elsevier.com/locate/ynlme

Methylene blue facilitates the extinction of fear in an animal

model of susceptibility to learned helplessness
Kathryn M. Wrubel, Douglas Barrett, Jason Shumake,
S. Elizabeth Johnson, F. Gonzalez-Lima ¤
Department of Psychology and Institute for Neuroscience, University of Texas at Austin, Austin, TX 78712, USA

Received 29 June 2006; revised 22 August 2006; accepted 23 August 2006

Available online 2 October 2006

Abstract

The objectives were to (1) extend previous Wndings on fear extinction deWcits in male congenitally helpless rats (a model for susceptibil-
ity to learned helplessness) to female congenitally helpless rats, and (2) attempt a therapeutic intervention with methylene blue, a meta-
bolic enhancer that improves memory retention, to alleviate the predicted extinction deWcits. In the Wrst experiment, fear acquisition (four
tone-shock pairings in operant chamber) was followed by extinction training (60 tones in open Weld). Congenitally helpless rats showed
fear acquisition similar to controls but had dramatic extinction deWcits, and did not display the gradual extinction curves observed in con-
trols. Congenitally helpless rats demonstrated greater tone-evoked freezing as compared to controls in both the acquisition and extinction
contexts one week after extinction training, and also in the extinction probe conducted one month later. In the second experiment (which
began one month after the Wrst experiment) congenitally helpless subjects were further exposed to tones for 5 days, each followed by 4 mg/kg
methylene blue or saline IP, and had a fear renewal test in the acquisition context. Methylene blue administration improved retention of
the extinction memory as demonstrated by signiWcant decreases in fear renewal as compared to saline-administered congenitally helpless
subjects. The impaired ability to extinguish fear to a traumatic memory in congenitally helpless rats supports the validity of this strain as
an animal model for vulnerability to post-traumatic stress disorder, and this study further suggests that methylene blue may facilitate fear
extinction as an adjunct to exposure therapy.
© 2006 Elsevier Inc. All rights reserved.

Keywords: Extinction; Post-traumatic stress disorder; Congenitally helpless rats; Methylene blue; Fear conditioning; Metabolic enhancer; Animal model;
Disease susceptibility; Memory retention; Exposure therapy; Cytochrome oxidase

1. Introduction helpless phenotype, as opposed to the 5–20% of randomly

bred rats (Henn, Johnson, Edwards, & Anderson, 1985;
Inescapable electric shock prevents animals from subse- Lachman et al., 1992). Since epidemiological studies show
quently learning an escape response, a phenomenon termed that 40–50% of the risk of susceptibility to depression and
learned helplessness (Overmier & Seligman, 1967). Vulnera- 30% of the risk of susceptibility to post-traumatic stress
bility to learned helplessness is hereditable and can be disorder (PTSD) is genetic (Fava & Kendler, 2000; Sanders,
enhanced through selective breeding, as evidenced by the Detera-Wadleigh, & Gershon, 1999; True et al., 1993), an
creation of a strain of congenitally helpless rats (Henn & appropriate animal model of congenital vulnerability to
Vollmayr, 2005). Studies show that up to 95% of the con- these disorders could be very useful, especially in the devel-
genitally helpless oVspring typically show the learned opment of therapeutic treatments.
A recent paper reported behavioral characteristics of
male rats predisposed to learned helplessness (Shumake,
*
Corresponding author. Fax: +1 512 471 4728. Barrett, & Gonzalez-Lima, 2005). Compared to normal
E-mail address: gonzalez-lima@mail.utexas.edu (F. Gonzalez-Lima). rats, congenitally helpless rats demonstrated reduced

1074-7427/$ - see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.nlm.2006.08.009
210 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217

reward sensitivity, high novelty seeking, an increase in con- lower in subjects receiving MB than saline, and MB-treated
ditioned fear, and a deWcit in fear extinction. This behav- subjects also had a longer lasting eVect of extinction. There-
ioral phenotype appears similar in many respects to that of fore, the current study was undertaken to determine if con-
PTSD patients. For example, patients with PTSD show genitally helpless rats, which are resistant to extinction, can
stronger acquisition and reduced extinction of aversively likewise have their extinction learning improved by methy-
conditioned responses (Orr et al., 2000), and a personality lene blue.
proWle of low reward dependence, high novelty seeking, and
high behavioral inhibition (Richman & Frueh, 1996; Wang 2. Materials and methods
et al., 1997). Also, humans with PTSD show deWcits in
extinction of conditioned fears (Orr et al., 2000; Peri, Ben- 2.1. Subjects
Shakhar, Orr, & Shalev, 2000) and congenitally helpless
rats also demonstrated similar deWcits when compared with For the Wrst experiment, subjects were 23 female congenitally helpless
normal control rats (Shumake et al., 2005). Human studies rats (bred in our laboratory from breeding pairs obtained from the Central
Institute for Mental Health in Mannheim, Germany, courtesy of Dr. Fritz
have shown that females are more vulnerable to PTSD Henn), and 12 female Sprague–Dawley control rats obtained from Harlan
(with approximately twice the risk), and that their symp- (Houston, TX), all weighing approximately 300 g at the beginning of the
toms persist longer than those of males (NemeroV et al., experiment. The congenitally helpless subjects used in this study were bred
2006). Based on the human studies, one would expect con- in our laboratory and were not tested for susceptibility to learned helpless-
genitally helpless females to show greater extinction deWcits ness, as this would involve administration of foot shocks prior to acquisi-
tion training. Subjects were housed 2–3 per cage under standard
than males. The Wrst objective of the present study was to laboratory conditions with a 12 h light/dark cycle and free access to food
extend our previous Wndings on fear extinction deWcits to and water. Animal experimentation was approved by the University of
female congenitally helpless rats. Texas Institutional Animal Care and Use Committee. Male rats would
Methylene blue (MB), a metabolic enhancer, has been also have been used, but, because of the need for male subjects for another
shown to improve memory for extinction of fear condition- experiment and high infertility in our colony, we were unable to breed a
suYcient number of male subjects for this purpose. Rats were given daily
ing in normal rats (Gonzalez-Lima & Bruchey, 2004). The vaginal smears and began training on the Wrst day of their estrus cycle in
second objective of the present study was to determine if both experiments. For the second experiment, female congenitally helpless
MB could facilitate extinction in congenitally helpless sub- rats were subdivided into 2 groups: 12 were treated with MB and 11 were
jects. Congenitally helpless rats are postulated to be an ani- administered saline.
mal model of PTSD, demonstrating extinction deWcits
2.2. Apparati
observed in patients with the disorder. Since humans with
PTSD often go through extinction training to reduce their DiVerent apparati were utilized during extinction training and probes,
fears, we administered MB during extinction training to in order to parse out any eVects of contextual fear on freezing behavior.
determine if it would facilitate retention of extinction in The acquisition session was conducted in standard operant chambers and
congenitally helpless subjects. If so, MB may be a useful extinction sessions were conducted in open-Weld activity chambers in a
therapeutic adjunct to exposure therapy for patients with diVerent room.

PTSD to aid in retention of extinction of conditioned fears 2.2.1. Acquisition context (Context A)
to traumatic memories. Pavlovian tone foot-shock acquisition training was conducted in four
Methylene blue is a redox dye commonly administered operant conditioning chambers, each measuring 22 £ 25 £ 32 cm (MED
as an antidote for methemoglobinemia, a condition in Associates, St. Albans, VT) and enclosed in sound-attenuated boxes illu-
which (usually due to metabolic poisoning) the body is minated by red lights. The two sides of each chamber were aluminum, and
the front, back, and top were made of clear plexiglas. Tones were gener-
unable to convert methemoglobin to hemoglobin to allow ated by a Wavetek Sweep/Modulation Generator (Wavetek, San Diego,
oxygen transport (Bodansky & Gutmann, 1947; Bradberry, CA) and presented through speakers mounted at the top of each chamber.
2003; Clifton & Leiken, 2003; Etteldorf, 1951). MB is able The acoustic conditioned stimulus (CS) was a frequency-modulated tone
to reverse this process, and has been used safely in humans of 1–2 kHz, 2 sweeps/s, 15 s in duration, with an intensity of 68 dB, mea-
for over a century. The memory retention enhancing eVects sured at the center of the Xoor of the chamber with a decibel meter. The
unconditioned stimulus (US) was a mild foot shock of 0.5 mA, 0.75 s in
of MB were Wrst reported by Martinez, Jr. and colleagues duration, delivered through metal bars (separated by 1.2 cm) which
(1978), who discovered that low dose post-training admin- formed the Xoor of the chamber, and were wired to shock generators
istration of MB improved memory retention in an inhibi- (MED Associates). Stimulus presentations were controlled by computer
tory avoidance task. MB has also been shown to enhance programs, written by the experimenters using the MED-PC for Windows
memory retention in a spatial memory task, an object rec- programming language (MED Associates). A Bioclean solution (Stanbio
Laboratory, Boerne, TX) was placed in the tray beneath the chamber to
ognition task, and to aid in between-days habituation to a provide a distinct olfactory cue for the acquisition context.
familiar environment (Callaway, Riha, Bruchey, Munshi, &
Gonzalez-Lima, 2004; Riha, Bruchey, Echevarria, & Gonz- 2.2.2. Extinction context (Context B)
alez-Lima, 2005). More relevant to the present study is that Extinction training occurred in a diVerent context. Two open-Weld
Gonzalez-Lima and Bruchey (2004) reported that memory activity boxes measuring 31 £ 45 £ 45 cm (MED Associates) with Wber-
glass bottoms, clear plexiglas sides, and an open top were utilized for
retention of extinction of Pavlovian fear conditioning could extinction training. Horizontal activity was detected by arrays of infrared
be enhanced with post-extinction administration of MB in motion detectors (16 £ 16, 2.54 cm apart), with two arrays located 1 cm
normal rats. Total post-extinction freezing scores were above the Xoor of the chamber. Rearings were detected with a vertical-axis
 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217 211

Table 1
Design for Experiments 1 and 2
Experiment 1 Experiment 2
Day 1 Days 2–3 Day 9 Day 10 Day 1 Days 2–6 Day 7
Session Acquisition Extinction Probe-Acq Probe-Ext Probe-Acq Probe-Ext Probe-Acq
Context Context A Context B Context A Context B Context A Context B Context A
Stimuli 4 T!S 30 T/Day 4T 4T 4T 4 T/Day + MB or Saline 4T
T, 1–2 kHz 68 dB 15 s tone CS; S, 0.5 mA 0.75 s foot shock US; MB, methylene blue.

array positioned 13 cm above the surface of the Xoor to ensure that only Diestrus females received up to three additional days of
those rearing movements in which the subject’s forepaws left the ground habituation and handling.
would register as rearing counts. Open-Weld boxes were controlled by the
MED Associates Activity Monitor program (Version 5.10), which
The training procedure is presented in Table 1 under
recorded various parameters related to a subject’s ambulation (sequential “Experiment 1”. For acquisition training, all subjects were
horizontal beam breaks), immobility (no beam breaks), and rearing behav- placed in the conditioning chambers (Context A) and
iors (vertical beam breaks). Stereotypic movement is deWned by the com- received four tone-shock pairings over 15 min, with pseudo-
puter-controlled infrared beam detectors as movement without random inter-trial intervals averaging 3 min. Each 15-s tone
displacement, i.e., short movements that break a beam without the subject
entering into ambulation. The tone CS which was 1–2 kHz, 2 sweeps/s, and
CS co-terminated with the foot shock US.
15 s in duration, was digitally recorded from the tone generator used in the Extinction training occurred over two-consecutive days
acquisition context and presented through a computer speaker above the while female subjects were in the diestrus phase, to consis-
open Welds (measured at 68 dB in the center of the Xoor of each open Weld tently synchronize the phase of the cycle with the context,
with a decibel meter). such that animals would be tested in the same context at
the same phase of the estrus cycle (Birke & Archer, 1975)
2.3. Freezing scores
and avoid interference with conditioned freezing mea-
Freezing scores were utilized as an index of tone-evoked fear. Experi- sures in the open Weld (Context B, used for extinction
menters trained to score freezing with high inter-rater reliability recorded training). Subjects were placed in the open Weld, and 30
a freezing score during pre-CS and CS periods through direct observation. tones were presented over 60 min, with 2 min between
Freezing was operationally deWned as the subject having all four feet on each tone onset. The Wrst minute in the open-Weld context
the Xoor, with minimal head movements, and shallow, rapid breathing. A
score of 1 was recorded if the subject froze for the majority of a 3 s bin,
served as a baseline measure of activity (prior to the Wrst
with a maximum freezing score of 5 for complete freezing during the 15-s tone CS presentation).
pre-CS and CS periods. Immobility time was scored automatically as time Post-extinction probe trials occurred one week later in
spent with no new infrared beam breaks (no horizontal or vertical move- the acquisition context (when subjects were in estrus), in
ment). Immobility time overlapped with freezing scores, but whereas freez- order to measure context-dependent renewal of fear. Each
ing was scored in Wve 3-s bins by an experimenter, immobility time was
continuously scored by the computer. The two independent measures were
subject was returned to the operant chamber (Context A)
highly correlated (r D 0.81). and presented with four 15-s tones (in the absence of foot
shock) over 10 min, with 2 min intervals between each tone
2.4. Statistical analysis onset. Post-extinction probe trials (consisting of four tone
CS presentations, as described above) occurred in the
The SPSS 11.5 for Windows statistical software package was used for extinction context on the following day, when subjects were
statistical analysis. All data were analyzed with analyses of variance
(ANOVAs), and p-values were reported as Huynh–Feldt corrected values,
in diestrus, and served to measure retention of extinction
with an -value of .05 regarded as signiWcant. When warranted, simple- learning. Experimenters recorded freezing scores during
eVects tests of signiWcant interactions were performed and adjusted by pre-CS and CS periods in both contexts.
Bonferroni correction.
3.2. Results
3. Experiment 1: Extinction of Pavlovian fear conditioning in
the congenitally helpless female rat Helpless and control subjects showed similar baseline open-
Weld activity prior to extinction training. In contrast, they
3.1. Behavioral training showed dramatic group diVerences during extinction, with
congenitally helpless rats exhibiting a large extinction deWcit.
Subjects were handled every day for 7 days prior to the
start of training. During this time, each subject was habitu- 3.2.1. Baseline motor activity and excitatory conditioned
ated to the acquisition context (Context A) in the absence behavior in the open Weld
of tones or shocks for 1 h a day. Vaginal smears were used There were no diVerences in baseline motor activity mea-
to determine the estrus cycle phase for each individual sub- sures between congenitally helpless and control subjects
ject before training began. An estrus cycle of 4 days was during the Wrst minute of Extinction Session 1 (prior to the
conWrmed and training was staggered between subjects onset of the Wrst tone CS) in any of the parameters ana-
according to estrus cycle, such that the Wrst day of acquisi- lyzed, which included ambulatory time, stereotypic move-
tion training occurred while each subject was in estrus. ment time, immobility time, and rearing time (Table 2).
212 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217

Table 2 measures from each 15-s tone CS presentation were averaged

Motor activity (s) during baseline open Weld (1 min) in helpless and con- into six bins representing Wve tone presentations every 10 min.
trol females
There was a signiWcant group eVect on freezing behavior
Helpless n D 23 Control n D 12 during extinction training, F(1, 29) D 139.67, p < 0.001, with
Mean § SE Mean § SE
congenitally helpless subjects freezing much more than con-
Ambulatory time 9.7 § 0.8 9.6 § 1.2 trols. The degrees of freedom are 29 instead of 33 for the
Stereotypic movement 18.8 § 0.7 18.9 § 0.8
extinction sessions because of missing values from 4 out of
time
Immobility time 15.6 § 1.6 20.8 § 2.7 the 23 congenital helpless subjects. Control rats showed grad-
Rearing time 14.3 § 1.5 9.2 § 1.2 ual decreases in freezing, whereas congenitally helpless rats
Center time 15.0 § 1.1 11.9 § 1.7 show a paradoxical enhancement of freezing during the Wrst
No signiWcant group diVerences. session of extinction trials. There was a signiWcant
group £ trial £ session interaction, F(5, 145) D 4.13, p < 0.01.
There were no signiWcant diVerences between groups in There was also a signiWcant group £ trial interaction in ses-
time spent in the center of the open Weld during the Wrst min- sion 1, F(5, 145) D 10.92, p < 0.001, but not in session 2,
ute of Extinction Session 1 (Table 2), which is considered a F(5, 145) D 0.12. In session 1, there was a simple main eVect of
measure of fearfulness since timid subjects have been found trial for the control group, which decreased freezing,
to spend more time along the walls of the apparati (Treit & F(5, 55) D 9.56, p < 0.001; however, the helpless group showed
Fundytus, 1988). This indicates that the diVerences observed increased freezing, F(5, 100) D 2.75, p < 0.05. While control
in extinction behavior between congenitally helpless rats and rats showed similar rates of extinction in both sessions with
control subjects cannot be attributed to baseline strain diVer- smooth learning curves, congenitally helpless subjects dis-
ences in general motor activity or fearfulness. played no evidence of extinction in Extinction Session 1, and
To evaluate possible group diVerences in the strength of slowly began to extinguish freezing behavior in Extinction
the original excitatory conditioning, tone-evoked freezing Session 2 (Fig. 2A).
and motor activity measures during the Wrst tone after the
acquisition session were evaluated in the open Weld. There 3.2.3. Context-dependent renewal
were no signiWcant group diVerences in freezing time, immo- In order to assess the eVects of context-dependent fear
bility time, ambulatory time, rearing time or stereotypic renewal after extinction, the data from the four tone probes
movement time (Fig. 1). This indicates that both groups delivered on Day 9 in the acquisition context, Context A,
showed a similar strength of fear conditioning after the origi- were analyzed (Probe-Acq). Congenitally helpless subjects
nal acquisition session. The lack of a group eVect on the Wrst showed signiWcantly more freezing relative to controls,
extinction trial is consistent with the lack of an acquisition F(1, 33) D 230.94, p < 0.0001 in this session (Fig. 2B). The
eVect, but this may not entirely rule out an acquisition eVect. main reason for the large F value is that the group eVects of
helpless vs. control are extremely large.
3.2.2. Short-term extinction
Short-term extinction was deWned as the extinction behav- 3.2.4. Long-term extinction
ior seen in the Wrst two 1-h sessions of training (Days 2–3). Long-term extinction was deWned as extinction behavior
Data were evaluated with repeated measures ANOVA, with after 1 week, and was assessed by comparing freezing behav-
session and trial serving as within-subject factors. Behavioral ior during the Wrst four tones at the beginning of extinction
session 1 with freezing during the four probe trials on Day
10, which were conducted in the same context, the extinction
context B (Probe-Ext). Both control and congenitally help-
less rats showed long-term reductions in freezing as a conse-
quence of extinction training: main eVect of session for
control group, F(1,11) D 214.53, p < 0.001, and for helpless
group, F(1, 22) D 47.51, p < 0.001. However, congenitally help-
less subjects showed signiWcantly less long-term extinction
relative to controls: main eVect of group, F(1, 33) D 23.91,
p < 0.001 (Fig. 2B).

4. Experiment 2: The eVects of methylene blue on memory

retention of extinction of Pavlovian fear conditioning in
Fig. 1. Conditioned freezing and motor activity during the Wrst tone CS in congenitally helpless rats
extinction session 1 in the open Weld was assessed in the post-acquisition
probe, to investigate possible group diVerences in strength of the original
4.1. Behavioral training
excitatory conditioning. Freezing scores (0–5) were transformed to freez-
ing time (0–15 s) to match the 15-s timeframe of the open-Weld parameters.
There were no signiWcant group diVerences in freezing or ambulatory Subjects were the same 23 female congenitally helpless
measures. rats used in the Wrst experiment. The training procedure is
 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217 213

match the extinction protocol that the laboratory previ-

ously developed for MB testing and also to avoid a Xoor
eVect that more intensive extinction training would have
induced. Thirty minutes following each daily probe session
in the extinction context, subjects were injected with either
MB (4 mg/kg) or saline I.P. Freezing scores recorded during
pre-CS and CS presentations were recorded and analyzed
as previously described.
On Day 7, subjects were returned to the acquisition con-
text (Context A) for a post-treatment probe session consist-
ing of four 15-s tones (in the absence of foot shock) over
10 min, with 2 min intervals between each tone onset
(Probe-Acq Day 7). Freezing scores were recorded by an
experimenter blind to group assignment. Freezing scores to
the tone CS in pre-treatment (Day 1) and post-treatment
(Day 7) probe sessions recorded in the acquisition context
were compared between groups to determine if MB aided in
the facilitation of memory retention of extinction.

4.2. Results

Methylene blue treatment following extinction training

signiWcantly improved memory retention of extinction in
the acquisition context (Context A) in congenitally helpless
subjects. While MB did not improve extinction learning in
extinction context (Context B) sessions, it did signiWcantly
improve the generalization of that learning to the acquisi-
tion context. The relatively low levels of freezing in both
Fig. 2. (A) Extinction curves (means § SE). The conditioned response dur- groups at the start made it very diYcult to see an initial
ing Extinction Sessions 1 and 2 was measured in terms of freezing behav-
facilitation of extinction by MB. However, the MB-treated
ior. Each 1 h extinction session consisted of one tone CS every 2 min,
averaged by Wve tones for each 10-min bin for the repeated measures
rats showed suYcient freezing to the Wrst tone in the
ANOVA. SigniWcant diVerences (p < .01) between congenitally helpless renewal context, and the improvement is shown by an
and control groups were found at every time point in both sessions. acceleration of extinction in this context, when the tone is
¤
p < .001. (B) Freezing scores from post-extinction probe trials not followed by shock (Fig. 3).
(means § SE) in both the acquisition (Probe-Acq, Context A) and extinc-
tion (Probe-Ext, Context B) contexts. Each probe test consisted of four
tones in 10 min. Congenitally helpless subjects showed signiWcantly more
4.2.1. EVects of methylene blue treatment in the extinction
freezing relative to controls. ¤p < .0001. context (Context B) probe sessions
Freezing to the tone CS in the extinction context was
presented in Table 1 under “Experiment 2”. This experiment analyzed with repeated measures ANOVA, with the aver-
began one month after the context renewal probe trial con- age of the four trials from each extinction session serving as
ducted in Experiment 1. Congenitally helpless females were the repeated measure. No signiWcant eVects related to treat-
given daily vaginal smears to assess their estrus cycle. On ment were obtained (Fs < 1), but a signiWcant main eVect of
their Wrst day of estrus, they were returned to the acquisition session indicated that both MB and saline-administered
context (Context A) for probe trials consisting of four 15-s subjects froze less with each session, F(4, 84) D 9.50, p < .001.
tones (in the absence of foot-shock) over 10 min, with 2 min In addition to freezing, general activity measures recorded
intervals between each tone onset. These probe trials veriWed by the infrared arrays in the extinction context were analyzed
that all congenitally helpless subjects continued to show sig- using the same repeated measures design. There were no sig-
niWcant freezing one month after the end of Experiment 1. niWcant treatment-related diVerences in ambulation, rearing,
Subjects were matched based on these freezing scores and immobility, or stereotypic movement times, demonstrating
assigned to either the MB treatment or saline control group. that MB did not alter general motor activity (Table 3).
For each of the next 5 days, subjects were presented with
four 15-s tones (in the absence of foot shock) over 10 min, 4.2.2. EVects of methylene blue treatment in the acquisition
with 2 min intervals between each tone onset in the extinc- context (Context A) probe sessions
tion context (Context B, Probe-Ext Days 2–6). Extinction There were no group diVerences in pre-CS freezing in the
trials were conducted in a context diVerent from the origi- post-injection renewal test (Fig. 3B). DiVerences in freezing
nal acquisition context in order to reduce confounds result- to the tone CS in the acquisition context were analyzed
ing from contextual conditioning. The rationale was to with repeated measures ANOVA, with pre-treatment and
214 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217

post-treatment probes serving as repeated measures. A sig- Table 3

niWcant two-way (treatment £ probe) interaction was Post-injection average motor activity (s) in open Weld (600 s) in helpless
females
obtained, F(1, 21) D 21.59, p < .0001, indicating that congen-
itally helpless rats treated with MB showed signiWcantly MB n D 12 Saline n D 11
Mean § SE Mean § SE
reduced freezing in the acquisition context. Therefore, MB
facilitated fear extinction in these subjects (Fig. 3C). Ambulatory time 29 § 6 29 § 5
Stereotypic movement time 103 § 12 108 § 7
Immobility time 304 § 43 282 § 37
4.3. Discussion Rearing time 150 § 25 171 § 27
No signiWcant group diVerences.
Congenitally helpless female rats exhibited abnormally
high freezing in response to a tone formerly predictive of
electric shock, even after 60 presentations of the tone in the
absence of shock. This extinction deWcit could not be attrib-
uted to diVerences in the strength of the original excitatory
conditioning. While congenitally helpless females showed a
trend for higher freezing during the Wrst tone after extinc-
tion, they showed no signiWcant group diVerences in freez-
ing, immobility or any of the behaviors measured in the
open Weld. They failed to show the gradual decrement in
freezing characteristic of the normal extinction curves seen
in the control subjects during the extinction sessions, and
continued to show signiWcantly greater tone-evoked fear
one week after extinction training, when tested in both the
acquisition and extinction contexts as we hypothesized.
These Wndings are consistent with those previously
reported in the congenitally helpless male rat (Shumake
et al., 2005).
The behavioral deWcit observed in congenitally helpless
rats cannot be explained by a general suppression of motor
activity or by an overall increase in non-speciWc fearfulness,
which can be indicated by thigmotaxis, or time spent in the
periphery versus the center of the open Weld (Treit &
Fundytus, 1988). There were no baseline group diVerences
in open-Weld behavior prior to the onset of the fear-associ-
ated tones, in terms of motor activity or thigmotaxis.
This experiment demonstrated that congenitally helpless
rats show deWcits in both extinction performance and reten-
tion: an inherently neutral tone CS is capable of evoking
maladaptive fearful responses in the helpless subjects, even
outside of the context in which they experienced the aver-
sive shock US. Another animal model of post-traumatic
stress disorder is likewise characterized by the formation
and persistence of memories that evoke fear across multiple
Fig. 3. (A) Conditioned freezing (mean § SE) to the tone in Context B
contexts (Rau, DeCola, & Fanselow, 2005). Post-traumatic
(extinction/open Weld) averaged over 5 days (Probe-Ext Days 2–6) of MB
and saline injections (4 mg/kg I.P.) in congenitally helpless rats. No signiW-
stress disorder patients show increases in autonomic
cant eVects related to treatment were obtained (Fs < 1). (B) No signiWcant arousal (heart rate, skin conductance) in response to aver-
diVerences were found in freezing scores (mean § SE) during the pre-CS sive Pavlovian conditioning, and these conditioned
period in the acquisition context (Context A) between MB and saline- responses are resistant to extinction (Orr et al., 2000; Peri
treated groups of congenitally helpless rats (Fs < 1). (C) EVect of methy- et al., 2000). This phenomenon is known as paradoxical
lene blue on extinction of freezing to the tone in the acquisition context
(Context A). Pre-treatment (Probe-Acq Day 1) and post-treatment
enhancement (an increase in conditioned responding as a
(Probe-Acq Day 7) freezing scores can be seen in the congenitally helpless consequence of extinction training), and was demonstrated
subjects. Conditioned freezing (mean § SE) to the tone before and after by congenitally helpless subjects in the Wrst extinction ses-
daily post-training injections of 4 mg/kg I.P. of MB or saline is shown. sion, making an even stronger argument for their role as an
Subjects were matched into groups with respect to pre-injection freezing animal model of vulnerability to PTSD. In our previous
scores, then tested again after 5 days of tone presentations in the
extinction context. Conditioned freezing was signiWcantly reduced in con-
study (Shumake et al., 2005), helpless males showed
genitally helpless subjects receiving MB injections as compared to the approximately twice as much freezing as controls after
saline-injected group. ¤p < .0001. long-term extinction; whereas in the present study, helpless
 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217 215

females showed approximately three times more freezing brain metabolism in normal rats. Methylene blue success-
than controls in the long-term extinction probe. This con- fully improved Pavlovian fear extinction memory, and
curs with data from human studies showing that females resulted in higher metabolic activity in the frontal cortical
are more vulnerable to PTSD than males. regions previously implicated in extinction (Barrett et al.,
The brains of congenitally helpless rats have been com- 2003). In addition, this study found that the regional meta-
prehensively mapped using metabolic brain imaging tech- bolic increases were correlated with the degree of extinction
niques, and these subjects show extensive alterations in retention. The eVects of MB on the congenitally helpless
regional brain metabolism as compared to non-helpless subjects seen here, together with our prior Wndings, suggest
subjects (Shumake, Poremba, Edwards, & Gonzalez-Lima, that MB may be therapeutically useful in PTSD patients.
2000; Shumake, Edwards, & Gonzalez-Lima, 2001, 2002, Methylene blue’s metabolic eVect may facilitate the adap-
2003; Conejo-Jimenez, Gonzalez-Pardo, & Gonzalez-Lima, tive inhibition of the conditioned fear response provided by
2004; Shumake & Gonzalez-Lima, 2003). Congenitally frontal cortical regions during successful Pavlovian extinc-
helpless rats have reduced metabolism in prefrontal cortical tion. Moreover, this facilitatory eVect is seen when subjects
regions (Shumake et al., 2000), which are highly engaged in are returned to the threatening (acquisition) context,
behavioral extinction (Barrett, Shumake, Jones, & Gonz- although subjects never received MB after exposure to this
alez-Lima, 2003; Gonzalez-Lima & Bruchey, 2004; Milad & context.
Quirk, 2002; Nair, Berndt, Barrett, & Gonzalez-Lima, The memory retention eVects of low dose MB cannot be
2001a, Nair, Berndt, Barrett, & Gonzalez-Lima, 2001b). attributed to alterations in locomotor activity, motivation,
This could lead to persistent fear-related conditioned reward value, or fearfulness (Gonzalez-Lima & Bruchey,
responses to a Pavlovian conditioned tone, which may be 2004; Riha et al., 2005). Despite these reports, in order to
more resistant to extinction than in control rats. Subjects control for any non-speciWc or state-dependent learning
with higher prefrontal cortical activity are also more suc- eVects that may occur by giving pre-training MB, it has
cessful at inhibiting a conditioned emotional response (Bar- generally been administered following training in behav-
rett et al., 2003). ioral studies. Also, because MB enhances memory retention
Medial prefrontal cortex may be crucial for the inhibi- of the events preceding its administration, the time of its
tion of the conditioned response during extinction (Barrett injection follows the target memory task.
et al., 2003; Milad & Quirk, 2002; Quirk, Garcia, & Gonz- It is worth noting that MB was not given during the
alez-Lima, 2006; Sotres-Bayon, Cain, & LeDoux, 2006; original extinction training, but after spontaneous recovery
Weible, McEchron, & Disterhoft, 2000), and PTSD patients and re-extinction. It seems unlikely that the eVect of MB
show decreased metabolic activity in this region during might be speciWc to this phase of training, but diVerential
extinction (Bremner et al., 1999). In newborn congenitally eVects of drugs given in conjunction with initial extinction
helpless rats, brainstem regions are metabolically uncou- exposure vs. subsequent exposures have been reported with
pled from networks of frontal and limbic regions (Shumake passive avoidance (Bevilaqua, da Silva, Medina, Izquierdo,
et al., 2004). This functional decoupling between brainstem & Cammarota, 2005). Therefore, future experiments could
and forebrain regions may be indicative of a developmental test the eVects of MB when given during other phases of
disorder in which brainstem regions are removed from the training.
inhibition provided by the frontal cortex during successful Since MB has a half-life of 5–6.5 h (Peter, Hongwan,
Pavlovian extinction. A disruption of prefrontal–amygdala Kupfer, & Lauterburg, 2000), it is unlikely that the facilita-
interactions could also selectively impair Pavlovian extinc- tion of the extinction memory observed 24 h after the last
tion while leaving acquisition relatively intact (Phelps, Del- injection of MB reXects a continued direct action of the
gado, Nearing, & LeDoux, 2004; Sotres-Bayon, Bush, & drug. Rather, it is probably due to enhanced oxidative
LeDoux, 2004). energy metabolism (Riha et al., 2005) occurring at a critical
Successful fear extinction training in PTSD patients may time in memory consolidation because memory retention in
be facilitated by pharmacotherapy (Davis, Myers, Chhat- rats is not aVected by MB given 15 min before training, 6 h
wal, & Ressler, 2006). Methylene blue is an FDA-approved after training, or 15 min before testing (Martinez, Jr. et al.,
drug which can be safely used in humans. It acts as an elec- 1978). MB increases cytochrome oxidase enzymatic activity
tron carrier in mitochondria, improving cellular respiration in a use-dependent manner, with brain regions with the
(Visarius, Stucki, & Lauterburg, 1997), and enhancing neu- highest metabolic demand during memory consolidation
ronal metabolism. Administration of MB to congenitally after extinction showing the largest increases in cytochrome
helpless rats signiWcantly ameliorated their extinction deW- oxidase activity (Gonzalez-Lima & Bruchey, 2004).
cit in the fear-evoking acquisition context, as predicted in Studies with multiple learning paradigms including
our hypothesis. It did not enhance memory retention of appetitive and aversive tasks, object recognition memory
extinction during extinction sessions in Context B, possibly and long-term habituation show that MB works as a meta-
due to a Xoor eVect on conditioned freezing in the extinc- bolic memory enhancer by increasing cytochrome oxidase
tion context. activity (Callaway, Riha, Wrubel, McCollum, & Gonzalez-
Gonzalez-Lima and Bruchey (2004) demonstrated the Lima, 2002; Callaway et al., 2004; Gonzalez-Lima & Bru-
eYcacy of MB in Pavlovian fear extinction and on regional chey, 2004; Riha et al., 2005). Cytochrome oxidase is the
216 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217

terminal enzyme of the electron transport chain which is and Jason Shumake. The original breeding pairs, which
tightly coupled to neuronal metabolism and ATP produc- produced the oVspring used for this study, were obtained
tion. It is the rate-limiting enzyme in mitochondrial respira- from the Central Institute for Mental Health (Mannheim,
tion (Wong-Riley, 1989), and cytochrome oxidase increases Germany), courtesy of Dr. Fritz Henn and Dr. Barbara
are correlated with increased energy demand (Gonzalez- Vollmayr. We thank Abby Schindler for her assistance with
Lima & Cada, 1998). Electrons can be donated from data collection. Dr. Wrubel conducted this study in partial
reduced MB to enter the electron transport chain between fulWllment of the requirements for the Ph.D. degree at the
Complexes II and III (Visarius et al., 1997), resulting in University of Texas at Austin.
enzyme induction of cytochrome oxidase (Callaway et al.,
2004; Gonzalez-Lima & Bruchey, 2004) and elevated cellu- References
lar oxygen consumption (Riha et al., 2005; Visarius et al.,
1997). Therefore, by increasing mitochondrial respiration Barrett, D., Shumake, J., Jones, D., & Gonzalez-Lima, F. (2003). Metabolic
through cytochrome oxidase activity, MB can enhance the mapping of mouse brain activity after extinction of a conditioned emo-
amount of ATP available within a neuron in order to tional response. Journal of Neuroscience, 23, 5740–5749.
Berlau, D. J., & McGaugh, J. L. (2006). Enhancement of extinction mem-
improve memory retention and recall in certain cognitive ory consolidation: the role of the noradrenergic and GABAergic sys-
tasks. If administered post-training (during the time of tems within the basolateral amygdala. Neurobiology of Learning and
memory consolidation), MB has been shown to increase Memory, 86, 123–132.
cytochrome oxidase activity within brain regions utilized Bevilaqua, L. R. M., da Silva, W. N., Medina, J. H., Izquierdo, I., & Cam-
during a memory task (Gonzalez-Lima & Bruchey, 2004). marota, M. (2005). Extinction and reacquisition of a fear-motivated
memory require activity of the Src family of tyrosine kinases in the
An important consideration to make about the mech- CA1 region of the hippocampus. Pharmacology, Biochemistry and
anism of action of MB is that unlike other pharmacolog- Behavior, 81, 139–145.
ical treatments proposed for use in PTSD, MB does not Birke, L. I., & Archer, J. (1975). Open-Weld behaviour of oestrous and
appear to be working as an anxiolytic compound or on a dioestrus rats: evidence against an ‘emotionality’ interpretation. Ani-
speciWc neurotransmitter system (Berlau & McGaugh, mal Behavior, 23, 509–512.
Bodansky, O., & Gutmann, H. (1947). Treatment of methemoglobinemia.
2006; Chhatwal, Davis, Maguschak, & Ressler, 2005; Journal of Pharmacology, 90, 46–56.
Davis et al., 2006). Drugs working on pharmacologically Bradberry, S. M. (2003). Occupational methemoglobinemia: mechanisms
speciWc synapses have actions distributed throughout the of production, features, diagnosis and management including the use
brain, in relevant regions, as well as other regions that of methylene blue. Toxicology Reviews, 22, 13–27.
could lead to unwanted side eVects. Methylene blue selec- Bremner, J. D., Staib, L. H., Kaloupek, D., Southwick, S. M., Soufer, R., &
Charney, D. S. (1999). Neural correlates of exposure to traumatic pic-
tively enhances metabolism in the brain during memory tures and sound in Vietnam combat veterans with and without post-
formation, in regions related to the task, without the side traumatic stress disorder: a positron emission tomography study.
eVects of drugs acting on synaptic transmission else- Biological Psychiatry, 45, 806–816.
where in the brain. Methylene blue can be used to facili- Callaway, N. L., Riha, P. D., Bruchey, A. K., Munshi, Z., & Gonzalez-
tate extinction memory formation for a speciWc fear- Lima, F. (2004). Methylene blue improves brain oxidative metabolism
and memory retention in rats. Pharmacology, Biochemistry and Behav-
evoking stimulus, and can facilitate extinction training ior, 77, 175–181.
through repeated administrations without losing its Callaway, N. L., Riha, P. D., Wrubel, K. M., McCollum, D., & Gonzalez-
eVectiveness. Lima, F. (2002). Methylene blue restores spatial memory retention
Congenitally helpless rats showed somewhat enhanced impaired by an inhibitor of cytochrome oxidase in rats. Neuroscience
Letters, 332, 83–86.
acquisition and dramatically impaired extinction of fear
Charney, D. S., Deutch, A. Y., Krystal, J. H., Southwick, S. M., & Davis,
evoked by an aversive memory—a crucial part of the M. (1993). Psychobiologic mechanisms of posttraumatic stress disor-
behavioral phenotype associated with PTSD (Charney, der. Archives of General Psychiatry, 50, 294–305.
Deutch, Krystal, Southwick, & Davis, 1993). The congeni- Chhatwal, J. P., Davis, M., Maguschak, K. A., & Ressler, K. J. (2005).
tally helpless rat also shows diVerences in regional brain Enhancing cannabinoid neurotransmission augments the extinction of
metabolism which resemble biological abnormalities conditioned fear. Neuropsychopharmacology, 30, 516–524.
Clifton, J., & Leiken, J. B. (2003). Methylene blue. American Journal of
detected in PTSD patients (Shumake & Gonzalez-Lima, Therapeutics, 10, 289–291.
2003). Thus, congenitally helpless rats may be a useful Davis, M., Myers, K. M., Chhatwal, J., & Ressler, K. J. (2006). Pharmaco-
model for studying those biological factors which render a logical treatments that facilitate extinction of fear: relevance to psycho-
subset of individuals vulnerable to PTSD, and for testing therapy. NeuroRx, 3, 82–96.
Etteldorf, J. N. (1951). Methylene blue in the treatment of methemoglobi-
novel treatments such as methylene blue, which can facili-
nemia in premature infants caused by marking ink: a report of eight
tate the extinction of conditioned fear. cases. The Journal of Pediatrics, 38, 24–27.
Fava, M., & Kendler, K. S. (2000). Major depressive disorder. Neuron, 28,
Acknowledgments 335–341.
Gonzalez-Lima, F., & Bruchey, A. K. (2004). Extinction memory improve-
ment by the metabolic enhancer methylene blue. Learning and Mem-
This research was supported by National Institutes of
ory, 11, 633–640.
Health Grant R01 NS37755 to F. Gonzalez-Lima and Gonzalez-Lima, F., & Cada, A. (1998). Quantitative histochemistry of
Texas Consortium in Behavioral Neuroscience Fellowships cytochrome oxidase activity: theory, methods, and regional brain vul-
T32 MH65728 to Kathryn M. Wrubel, Douglas Barrett, nerability. In F. Gonzalez-Lima (Ed.), Cytochrome oxidase in neuronal
 K.M. Wrubel et al. / Neurobiology of Learning and Memory 87 (2007) 209–217 217

metabolism and Alzheimer’s disease (pp. 55–91). New York and Lon- behavior and brain oxygen consumption. European Journal of Pharma-
don: Plenum Press. cology, 511, 151–158.
Henn, F. A., Johnson, J., Edwards, E., & Anderson, D. (1985). Melancholia Sanders, A. R., Detera-Wadleigh, S. D., & Gershon, E. S. (1999). Molecular
in rodents: neurobiology and pharmacology. Psychopharmacology Bul- genetics of mood disorders. In D. S. Charney, E. J. Nestler, & B. S. Bun-
letin, 21, 443–446. ney (Eds.), Neurobiology of mental illness. New York: Oxford.
Henn, F. A., & Vollmayr, B. (2005). Stress models of depression: forming geneti- Shumake, J., Barrett, D., & Gonzalez-Lima, F. (2005). Behavioral charac-
cally vulnerable strains. Neuroscience and Biobehavioral Reviews, 29, 799–804. teristics of rats predisposed to learned helplessness: reduced reward
Lachman, H. M., Papolos, D. F., Weiner, E. D., Ramazankhana, R., Hart- sensitivity, increased novelty seeking, and persistent fear memories.
nick, C., Edwards, E., et al. (1992). Hippocampal neuropeptide Y Behavioural Brain Research, 164, 222–230.
mRNA is reduced in a strain of learned helpless resistant rats. Brain Shumake, J., Conejo-Jimenez, N., Gonzalez-Pardo, H., & Gonzalez-Lima,
Research: Molecular Brain Research, 14, 94–100. F. (2004). Brain diVerences in newborn rats predisposed to helpless and
Martinez, J. L., Jr., Jensen, R. A., Vasquez, B. J., McGuinness, T., & depressive behavior. Brain Research, 1030, 267–276.
McGaugh, J. L. (1978). Methylene blue alters retention of inhibitory Shumake, J., Edwards, E., & Gonzalez-Lima, F. (2001). Hypermetabolism
avoidance responses. Physiological Psychology, 6, 387–390. of paraventricular hypothalamus in the congenitally helpless rat. Neu-
Milad, M. R., & Quirk, G. J. (2002). Neurons in medial prefrontal cortex roscience Letters, 311, 45–48.
signal memory for fear extinction. Nature, 420, 70–74. Shumake, J., Edwards, E., & Gonzalez-Lima, F. (2002). Dissociation of
Nair, H. P., Berndt, J. D., Barrett, D., & Gonzalez-Lima, F. (2001a). Meta- septo-hippocampal metabolism in the congenitally helpless rat. Neuro-
bolic mapping of brain regions associated with behavioral extinction in science, 114, 373–377.
preweanling rats. Brain Research, 903, 141–153. Shumake, J., Edwards, E., & Gonzalez-Lima, F. (2003). Opposite meta-
Nair, H. P., Berndt, J. D., Barrett, D., & Gonzalez-Lima, F. (2001 b). Matu- bolic changes in the habenula and ventral tegmental area of a genetic
ration of extinction behavior in infant rats: large-scale regional interac- model of helpless behavior. Brain Research, 963, 274–281.
tions with medial prefrontal cortex, orbitofrontal cortex, and anterior Shumake, J., & Gonzalez-Lima, F. (2003). Brain systems underlying sus-
cingulate cortex. Journal of Neuroscience, 21, 4400–4407. ceptibility to helplessness and depression. Behavioral and Cognitive
NemeroV, C. B., Bremner, J. D., Foa, E. B., Mayberg, H. S., North, C. S., & Neuroscience Reviews, 2, 198–221.
Stein, M. B. (2006). Posttraumatic stress disorder: a state-of-the-science Shumake, J., Poremba, A., Edwards, E., & Gonzalez-Lima, F. (2000). Con-
review. Journal of Psychiatric Research, 40, 1–21. genital helpless rats as a genetic model for cortex metabolism in
Orr, S. P., Metzger, L. J., Lasko, N. B., Macklin, M. L., Peri, T., & Pitman, depression. Neuroreport, 11, 3793–3798.
R. K. (2000). De novo conditioning in trauma-exposed individuals with Sotres-Bayon, F., Bush, D. E., & LeDoux, J. E. (2004). Emotional persever-
and without post-traumatic stress disorder: a functional MRI study. ation: an update on prefrontal–amygdala interactions in fear extinc-
Journal of Abnormal Psychology, 109, 290–298. tion. Learning and Memory, 11, 525–535.
Overmier, J. B., & Seligman, M. E. (1967). EVects of inescapable shock Sotres-Bayon, F., Cain, C. K., & LeDoux, J. E. (2006). Brain mechanisms
upon subsequent escape and avoidance responding. Journal of Com- of fear extinction: historical perspectives on the contribution of pre-
parative and Physiological Psychology, 63, 28–33. frontal cortex. Biological Psychiatry, 60, 329–336.
Peri, T., Ben-Shakhar, G., Orr, S. P., & Shalev, A. Y. (2000). Psychophysio- Treit, D., & Fundytus, M. (1988). Thigmotaxis as a test for anxiolytic
logic assessment of aversive conditioning in posttraumatic stress disor- activity in rats. Pharmacology, Biochemistry and Behavior, 31, 959–962.
der. Biological Psychiatry, 47, 512–519. True, W. R., Rice, J., Eisen, S. A., Heath, A. C., Goldberg, J., Lyons, M. J.,
Peter, C., Hongwan, D., Kupfer, A., & Lauterburg, B. H. (2000). Pharma- et al. (1993). A twin study of genetic and environmental contributions
cokinetics and organ distribution of intravenous and oral methylene to liability for posttraumatic stress symptoms. Archives of General Psy-
blue. European Journal of Clinical Pharmacology, 56, 247–250. chiatry, 50, 257–264.
Phelps, E. A., Delgado, M. R., Nearing, K. I., & LeDoux, J. E. (2004). Visarius, T. M., Stucki, J. W., & Lauterburg, B. H. (1997). Stimulation of
Extinction learning in humans: role of the amygdala and vmPFC. Neu- respiration by methylene blue in rat liver mitochondria. Federation of
ron, 43, 897–905. European Biochemical Societies Letters, 412, 157–160.
Quirk, G. J., Garcia, R., & Gonzalez-Lima, F. (2006). Prefrontal mechanisms Wang, S., Mason, J., Charney, D., Yehuda, R., Riney, S., & Southwick, S.
in extinction of conditioned fear. Biological Psychiatry, 60, 337–343. (1997). Relationships between hormonal proWle and novelty seeking in
Rau, V., DeCola, J. P., & Fanselow, M. S. (2005). Stress-induced enhance- combat-related posttraumatic stress disorder. Biological Psychiatry, 41,
ment of fear learning: an animal model of posttraumatic stress disor- 145–151.
der. Neuroscience and Biobehavioral Reviews, 29, 1207–1223. Weible, A. P., McEchron, M. D., & Disterhoft, J. F. (2000). Cortical
Richman, H., & Frueh, B. C. (1996). Personality disorder symptomatology involvement in acquisition and extinction of trace eyeblink condition-
among Vietnam veterans with combat-related PTSD. Anxiety, 2, 286–295. ing. Behavioral Neuroscience, 114, 1058–1067.
Riha, P. D., Bruchey, A. K., Echevarria, D. J., & Gonzalez-Lima, F. (2005). Wong-Riley, M. T. (1989). Cytochrome oxidase: an endogenous metabolic
Memory facilitation by methylene blue: dose-dependent eVect on marker for neuronal activity. Trends in Neurosciences, 12, 94–101.