Stereotypy and spontaneous alternation in deer mice and its
response to anti-adenosinergic intervention
Geoffrey de Browuer1 | Jaco Engelbrecht1 | Daniel C. Mograbi2,3 | Lesetja Legoabe1 |
Stephan F. Steyn1 | De Wet Wolmarans1
Edited by Joshua Plotkin and Cristina Ghiani. Reviewed by Annalisa Scimemi and Jill Crittenden.
Center of Excellence for Pharmaceutical
Sciences, Department of Pharmacology,
Faculty of Health Sciences, North-West
University, Potchefstroom, South Africa
Department of Psychology, Pontifícia
Universidade Católica – Rio (PUC-Rio), Rio
de Janeiro, Brazil
Institute of Psychiatry, Psychology &
Neuroscience, King’s College London,
London, UK
Correspondence
De Wet Wolmarans, Center of Excellence
for Pharmaceutical Sciences, Department of
Pharmacology, Faculty of Health Sciences,
North-West University, Potchefstroom,
South Africa.
Email: [email protected]
Abstract
Repetitive behavioral phenotypes are a trait of several neuropsychiatric disorders,
including obsessive-compulsive disorder (OCD). Such behaviors are typified by complex interactions between cognitive and neurobiological processes which most likely
contribute to the suboptimal treatment responses often observed. To this end, exploration of the adenosinergic system may be useful, since adenosine-receptor modulation has previously shown promise to restore control over voluntary behavior and
improve cognition in patients presenting with motor repetition. Here, we employed
the deer mouse (Peromyscus maniculatus bairdii) model of compulsive-like behavioral
persistence, seeking to investigate possible associations between stereotypic motor
behavior and cognitive flexibility as measured in the T-maze continuous alternation
task (T-CAT). The effect of istradefylline, a selective adenosine A2A receptor antagonist at two doses (10 and 20 mg kg−1 day−1) on the expression of stereotypy and
T-CAT performance in high (H) and non-(N) stereotypical animals, was investigated in
comparison to a control intervention (six groups; n = 8 or 9 per group). No correlation
between H behavior and T-CAT performance was found. However, H but not N animals presented with istradefylline-sensitive spontaneous alternation and stereotypy,
in that istradefylline at both doses significantly improved the spontaneous alternation scores and attenuated the stereotypical expression of H animals. Thus, evidence
is presented that anti-adenosinergic drug action improves repetitive behavior and
spontaneous alternation in stereotypical deer mice, putatively pointing to a shared
psychobiological construct underlying naturalistic stereotypy and alterations in cognitive flexibility in deer mice.
KEYWORDS
adenosine, animal model, cognitive flexibility, compulsive, deer mouse, istradefylline
1 | INTRODUCTION
Obsessive-compulsive disorder (OCD) is a debilitating neuropsychiatric
condition with a global 12-month prevalence rate of 1%–3% (APA, 2013;
Coles et al., 2018). OCD is characterized by persistent and intrusive
thoughts, images, or urges (obsessions) that cause significant distress
and anxiety. Generally, patients also present with repetitive, ritualized,
and neutralizing behaviors (compulsions), which are expressed to reduce
the level of distress experienced. OCD can be diagnosed based on the
presence of either obsessions or compulsions, or a combination of both;
as such, a definitive role for anxiety in its presentation remains disputed
(Abramowitz & Jacoby, 2015; Bartz & Hollander, 2006).
2 | de BROWUER et al.
OCD responds favorably, though suboptimally, to chronic, highdose treatment with serotonin reuptake inhibitors (SRIs), for example
clomipramine, and selective SRIs (SSRIs), for example fluoxetine and
escitalopram (Albert et al., 2018) which constitutes first-line treatment. However, treatment resistance remains an obstacle, in which
case several approaches can be followed, for example increasing the
dose of the current SRI/SSRI used, augmentation of SRI/SSRI therapy with low-dose anti-dopaminergic agents, for example risperidone, haloperidol, or clozapine (Albert et al., 2018). An accumulating
body of evidence points to the likelihood that different obsessivecompulsive (OC) phenotypes may be associated with unique perturbations in underlying psychobiology (Parkes et al., 2019; Raines
et al., 2015; Rosso et al., 2012; Torres et al., 2016; Wheaton
et al., 2016; Williams et al., 2013), which are proposed to underlie inconsistent treatment response. These findings highlight a need for a
better understanding of the potential interactions between cognitive
and neurobiological processes and how they may contribute to OC
symptom manifestation.
One area of research which is afforded noteworthy attention
in literature focuses on the construct of cognitive flexibility, which
broadly refers to the ability of an organism to adjust its behavioral
engagement based on changing circumstances to ensure an optimal outcome (Braem & Egner, 2018). Patients with OCD present
with deficits in cognitive flexibility (Bradbury et al., 2011; Gruner &
Pittenger, 2017; Vaghi et al., 2017). However, findings show some
degree of inconsistency, which directed research toward the study
of associations between specific OC traits and cognitive dysfunction
(Abramovitch et al., 2019; Robbins et al., 2012; Rufer et al., 2006).
Further, taking into consideration that some phenotypes, that is
sexual and aggressive thoughts, respond less to cognitive behavioral interventions than others, for example contamination/washing (Thorsen et al., 2018), a distinct role for implicit versus explicit
thought processing underlying treatment response is also implied.
Classic OC neurobiological theory implicates cortico-striatalthalamic-cortical (CSTC) circuit dysfunction as the primary neurobiological construct of OCD (Milad & Rauch, 2012; Shephard et al., 2021).
Collectively, these brain structures play important roles in the planning, gating, execution, and termination of goal-directed behavior and
normal executive functioning (Gruner & Pittenger, 2017). The clinical
efficacy of SRI/SSRI intervention in OCD can likely be ascribed to the
functional opposition of dopaminergic activity within the CSTC circuit
(Goddard et al., 2008), which is organized as a behaviorally activating
direct and a behaviorally inactivating indirect pathway (Markarian
et al., 2010; Nambu et al., 2002; Vaghi et al., 2017). Nevertheless,
taking into consideration the varying treatment response observed in
different patients, recent evidence for the contribution of other brain
areas has also been presented. Of special interest is the amygdala, of
which smaller physical volume has been linked to symptom severity
within the sexual/aggressive thought and safety/checking dimensions
(Pujol et al., 2004; Wood & Ahmari, 2015).
Against this background, investigations into adenosinergic processes in OCD might be informative. Although not as well-researched
as serotonin and dopamine, adenosine may be a significant role player
in the neuropathology of OCD since it acts as an important modulator of serotonergic and dopaminergic action (Chen, 2014). Further,
adenosine receptors, that is A1 and A2, are found in several central
nervous system structures, including the CSTC circuits and the amygdala (Stockwell et al., 2017). Whereas A1 receptors are commonly coexpressed with D1 receptors on the direct pathway of the CSTC circuit,
A2 receptors—most notably so the A2A subtype—are co-expressed with
D2 receptors on the indirect pathway (Stockwell et al., 2017). The actions of adenosine in the central nervous system are complex and can
at most be understood as “resting-state dependent” (Fuxe et al., 2010;
Rau et al., 2015). Here, we will explore the effects of the selective adenosine A2A receptor antagonist, istradefylline, on cognitive flexibility
and stereotypical expression.
Although a detailed review of the adenosinergic system falls beyond the scope of the current paper, some key aspects of the A2A
receptor must be highlighted. First, when co-expressed with another
receptor subtype, for example D2/A2A, a functionally antagonistic
interaction is found (Ferre et al., 1991), where activation of the adenosine receptor reduces the activity and/or binding affinity of the coexpressed receptor for its endogenous ligand and vice versa. Second,
although being abundantly expressed in the indirect dorsal and ventral striatopallidal pathways, the A2A receptor is also located in other
extrastriatal forebrain regions, that is the amygdala and hippocampus
(Stockwell et al., 2017). From a biobehavioral point of view, this is significant since such distinct neuroanatomical distribution is associated
with opposing behavioral and anxiety-like effects. For example, Shen
et al. (2008) found that extrastriatal A2A receptor (eA2A) blockade (and
knockout) attenuated the ambulatory and rearing responses of mice
dosed with the dopamine transporter inhibitor, cocaine; in contrast,
such behavior was aggravated by the specific blocking or knocking out
of striatal A2A receptors (sA2A). Moreover, eA2A, but not sA2A blockade
has been shown to maintain synaptic plasticity (Costenla et al., 2011).
Notwithstanding, the pro-cognitive effects of and improvement of
motor control following broad A2A receptor antagonism are well established in patients with Parkinson’s disease (Chen & Cunha, 2020).
Significance
Stereotypical expression in deer mice is repetitive, persistent, and seemingly purposeless. Such characteristics
are typical of clinical compulsive symptomology, which
is marked by suboptimal treatment response to currently
available serotonergic drug interventions. Given the complex psychobiological interactions underlying compulsivity, we aimed to explore the relationship between naturally
occurring stereotypical behavior and cognitive flexibility
in deer mice. We then sought to investigate the potential
effects of istradefylline, a selective adenosine A2A receptor antagonist which has shown promise as a pro-cognitive
drug in humans, on stereotypy and cognitive flexibility in
this species.
| de BROWUER et al. 3
Here, we aim to explore the effects of istradefylline, a highly
selective adenosine A2A receptor antagonist, in a naturalistic animal model of behavioral persistence that is reminiscent of clinical
compulsivity, that is the deer mouse (Peromyscus maniculatus bairdii).
Since the model is wholly naturalistic (the persistent behaviors expressed by the deer mice require no provocation by pharmacological, genetic, or behavioral means; Scheepers et al., 2018), it presents
research with a unique avenue for studying the potential underlying
associations between behavioral persistence and dysfunctional cognitive processes. Briefly, laboratory-housed deer mice exhibit a series of seemingly purposeless and repetitive motor stereotypies that
can be categorized into vertical and horizontal stereotypies, that is
jumping and pattern running, respectively (Hadley et al., 2006; Korff
et al., 2008; Scheepers et al., 2018). Such stereotypies are expressed
in 40%–45% of animals of both sexes from the age of 10 weeks and
manifest in a waxing and waning nature over the course of a single dark cycle (Scheepers et al., 2018). Therefore, the behavior is
reminiscent of clinical OCD in terms of being characterized by bouts
of excessive stereotypical engagement. With respect to its predictive validity, H behavior shows favorable, albeit not full response to
chronic, but not subchronic, high-dose (50 mg kg−1 day−1) oral escitalopram exposure (Scheepers et al., 2018).
Building on previous studies, the present investigation will seek to
elucidate the potential relationship between the expression of stereotypy and cognitive flexibility as assessed in the T-maze continuous alternation task (T-CAT). The T-CAT is a validated and widely applied test
of cognitive flexibility in rodents and is often used to identify changes
in processes related to working memory and the natural tendency of
rodents to explore a novel environment (Deacon & Rawlins, 2006). We
will further assess the potential effects of chronic istradefylline exposure on T-CAT performance and stereotypy.
2 | MATERIALS AND METHODS
2.1 | Animals
Since only 40%–45% of deer mice express H behavior, 100 deer mice
of both sexes (aged 12 weeks at the onset of experimentation) were
randomly sourced from the North-West University (NWU) vivarium without litter or cage bias for initial stereotypical assessment.
Animals were selected according to a random numbering and cage
allocation system which ensures adequate separation of litter mates
at weaning, as well as cage mates at adulthood. Prior to the onset of
stereotypy screening for selection (see following paragraph), all animals were group-housed four to six same-sex animals per home cage
until they reached 12 weeks of age, the age at which stereotypical behavior is fully established (Figure 1; Wolmarans et al., 2013).
However, from the onset of stereotypy screening onwards, animals
were single-housed to monitor drug intake (see below). Cages [35 cm
(l) × 20 cm (d) × 13 cm (h); Techniplast® S.P.A., Varese, Italy] were
individually ventilated, isolated from harsh noise and environmental interference, and prepared with a 3 cm-thick layer of ground
corncob as bedding substrate. Cages were maintained at 23°C at a
relative humidity of 55% and were kept on a 12-hr light/dark cycle
(06:00 hr/18:00 hr). Food and water (or drug solutions; see below)
were available ad lib throughout the course of the investigation.
Cages were cleaned weekly but inspected daily to monitor animal
welfare. Before the onset of investigation, all aspects of this work
were approved by the AnimCare Research Ethics Committee of the
NWU (approval number NWU-00575-19-A5).
2.2 | Baseline characterization of stereotypical
behavior and allocation to exposure groups
Deer mice were categorized as N or H according to a previously
published protocol (Wolmarans et al., 2013). Briefly, each of the
100 animals that were initially selected for behavioral screening
underwent three separate 12-hr nocturnal activity assessments,
conducted 48 hr apart (Figure 1). This is necessary to establish a
robust average baseline stereotypy count since animals acclimatize to
the testing environment over the course of the first two assessments.
On any specific assessment day, animals were moved from their
housing environment to the behavioral screening room which is
located on the same floor of the vivarium and maintained according
to the same environmental parameters as the housing room. Each
animal was introduced to its own automated animal activity testing
cage [21 cm (w) × 21 cm (l) × 35 cm (h); Accuscan® Inc., Columbus,
OH, USA] which is constructed from clear, translucent Plexiglas®
FIGURE 1 Graphical layout of study timeline. N: animals
selected based on the criterium for normal stereotypy (Table 1);
H: animals selected based on the criterium for high stereotypical
activity (Table 1); T-CAT, T-maze continuous alternation task
4 | de BROWUER et al.
and fitted with position-detecting infrared beams. Corncob was
provided in quantities enough to cover the floor of the test cages
but also ensuring that it does not interrupt the scoring of behavioral
data by means of infrared beam detection. Food and water (or drug
solutions) were provided ad lib throughout the testing phase. Animals
were introduced to the testing cages by 16:00 and habituated for
at least 2 hr before assessment commenced at the onset of the
dark cycle at 18:00. Each assessment session was conducted over
12 hr and test cages were cleaned with a disinfectant registered for
use in animal research facilities (F10® SCXD, Health and Hygiene
Products®, Randburg, South Africa) after each assessment.
Behavioral data were scored using Fusion® software (Accuscan®
Inc., Columbus, OH, USA). The number of vertical beam interruptions generated by the apparatus was used as a measure of vertical
activity, while the number of clockwise and anticlockwise cage revolutions was applied to evaluate the expression of pattern running
(Wolmarans et al., 2013). Following the three baseline assessments,
animals were divided into N and H groups according to a previously
published protocol (Wolmarans et al., 2013) which was developed
to consider both behavioral intensity and the time spent executing
stereotypical behaviors. As such, an average of the nine highest individual 30-min stereotypy counts generated by each animal and the
number of 30-min H stereotypical bouts generated across all three
baseline assessments were applied to determine behavioral severity.
Cutoff values for N and H behavior are provided in Table 1. Further,
animals had to express at least six H bouts across all three trials to be
selected for inclusion in the H cohort. Importantly, H animals could
demonstrate stereotypical behavior in either or both behavioral topographies to be classified as H. However, if both topographies were
expressed to an H level in a single 30-min bout, that is if an animal
both jumped and completed cage revolutions to the extent that such
behavior qualified as an H bout for the respective phenotypes within
the same 30-min bout, such a bout was counted only once. From
these data, 30 N and 30 H expressing animals, 30 of each sex (N:
13 males and 17 females; H: 17 males and 13 females), were initially
selected for inclusion in the remainder of the study (Figure 1). The
remaining 40 animals not used for further study were either diverted
to other unrelated, ethically approved behavioral studies, or euthanized. Based on extensive prior investigation in this model system
(please see Scheepers et al. (2018) for a review) as well as a statistical
power analysis, subjects were subsequently divided into the following drug exposure groups (initially, n = 10 per drug exposure group
per phenotype; both sexes included in each exposure group; group
numbers changed to 8–9 animals per group, depending on successful completion of the T-CAT assessment after drug exposure): (a)
vehicle only (N: three males and seven females; H: five males and
five females), (b) lower dose istradefylline (10 mg kg−1 day−1; Orr
et al. (2018); N: four males and six females; H: seven males and three
females), and (c) higher dose istradefylline (20 mg kg−1 day−1; Orr
et al. (2018); N: six males and four females; H: five males and five
females), both of which were constituted in the same vehicle solution as used in the vehicle-exposed group. Importantly, although this
work was designed to include animals of each sex in both behavioral
cohorts, we could not distribute the sexes equally between the different drug exposure groups, since animals are bred in successive
batches from different breeding pairs. As such, the between-batch
yield of N and H male and female animals differed slightly between
batches. The dual-tiered dosing schedule for istradefylline was designed to account for the possibility that opposing behavioral effects
may be observed at a lower and higher dose (Bastia et al., 2005; Orr
et al., 2018; Pourcher et al., 2012). Following 28 days of drug exposure and subsequent T-CAT testing, the same mice, already acclimatized to the testing environment, were reassessed for stereotypy
over a single 12-hr session.
2.3 | Drugs
Since istradefylline is insoluble in water, it was constituted for
oral administration via a drinking vehicle which consisted of a 2%
sucrose solution containing 2% dimethyl sulfoxide (DMSO), 1.5%
polyethoxylated castor oil (Kolliphor® EL), and 1.5% mineral oil
(all obtained from Sigma-Aldrich®, South Africa; Orr et al. (2018)).
The same solution was administered without the addition of drug
to control-exposed animals. Importantly, deer mice drink fluid at
an average rate of 0.25 ml/g/24 hr (de Brouwer, Fick, et al., 2020).
Although this has not been shown to change as a function of drug
exposure or additives added to the drinking water (de Brouwer, Fick,
et al., 2020; de Brouwer, Harvey, et al., 2020), the average daily fluid
intake per cage was monitored by weighing the residual amount
of liquid left on each following day (Supplementary Information).
Istradefylline was constituted in concentrations that ensured the
delivery of the appropriate doses to animals over a 24-hr period (4
and 8 mg/100 ml for the 10 and 20 mg kg−1 day−1 doses, respectively).
Orally administered istradefylline has a bioavailability of 60.8% in
rats, shows good blood–brain barrier permeation, and has a high
affinity for the A2A receptor (Ki
- 150 nM), making it suitable for
chronic oral administration as applied here (Müller, 2013). Drug
solutions were provided uninterrupted over 28 days and were
freshly constituted every other day.
TABLE 1 Cutoff criteria for N and H behavioral expression during a 30-min bout across both stereotypies (Wolmarans et al., 2013)
Behavioral topography N H
Pattern running <150 cage revolutions >200 cage revolutions
Vertical jumping <500 vertical beam breaks >2,000 vertical beam breaks
Abbreviations: H, high stereotypical; N, non-stereotypical.
| de BROWUER et al. 5
2.4 | Spontaneous alternation testing (T-CAT)
2.4.1 | Apparatus
The T-CAT apparatus used here was modified from the setup of
Deacon and Rawlins (2006) to accommodate the screening of
stereotypy-expressing deer mice. Briefly, the apparatus consisted
of a T-shaped maze constructed from black steel (walls; 250 mm
tall) and black, infrared-translucent Plexiglas® (floors) (Figure 2).
All arms were 400 mm long and 75 mm wide. Each of the arms
as well as the starting area (100 × 75 mm) at the foot of the stem
could be closed off from the rest of the maze with manually operated doors (Deacon & Rawlins, 2006). For the first trial (defined
as an arm choice) of each animal only, a divider was inserted between the two arms which extended 100 mm into the stem of the
T-maze (originating against the wall of the transverse arms and
centrally between the two arms) to force animals to enter either
the left or right arm without being presented with an opportunity to reverse the initial arm choice (Deacon & Rawlins, 2006).
All sessions were video-recorded, and since the maze was constructed on an infrared backlight, ambulatory activity was scored
by means of Ethovision® XT14 software (Noldus® Information
Technologies, The Netherlands). Following each T-CAT assessment session, the maze was cleaned with F10® SCXD (Health and
Hygiene Products®, Randburg, South Africa) to prevent any influence of previously associated odors on the alternating behavior of
subsequently assessed mice (Deacon & Rawlins, 2006).
2.4.2 | The procedure
As the current investigation aimed to assess the cognitive
performance of mice that already presented with different degrees
of motor stereotypy, animals were not pretrained in the T-CAT
before commencing experimentation (Deacon & Rawlins, 2006).
Since deer mice are nocturnal animals, the test was conducted
during the dark cycle and only after 19:00, so that at least 1 hr of
the dark cycle lapsed before the onset of testing. For each animal,
the test began with a single seemingly “forced” trial during which
the divider was left in place. This was followed by 14 free choice
trials (the divider being removed). At the beginning of the 15-
trial assessment session, mice were confined to the start box of
the maze for 30 s after which the door was raised, and the mouse
allowed to freely enter and explore the maze. Following its entry
into one of the arms, that is with all four paws, the arm choice was
recorded, the nonentered arm closed off and the mouse allowed to
return to the starting area at its own leisure. Here it was once again
confined for 30 s before being allowed to enter the arena again.
Mice were allowed to make up to 15 arm choices or explore the
maze for 35 min, whichever came first. The following parameters
were measured throughout the assessment period: percentage
alternation, that is the number of alternating arm choices out of the
maximum of 15 trials [(alternations/completed trials) × 100], the
time to first entry (s), and the average time per trial (s). Assessment
sessions not completed within 35 min were stopped and the
percentage alternation was calculated based on the number of
trials that had been completed. Mice that completed fewer than six
trials within 35 min were excluded from further analysis. As such,
the final numbers of animals that were included in the data analysis
were as follows: (a) vehicle only (N: two males and seven females;
H: four males and four females), (b) lower dose istradefylline
(10 mg kg−1 day−1; Orr et al. (2018); N: two males and six females;
H: seven males and two females), and (c) higher dose istradefylline
(20 mg kg−1 day−1; Orr et al. (2018); N: six males and two females; H:
four males and four females). Each animal was assessed only once
in the T-CAT after 28 days of vehicle or drug exposure.
2.4.3 | Statistical analyses
An initial power analysis was performed with G*Power® (version
3; Universität Kiel, Germany). An a priori test was set at an F-value
FIGURE 2 Schematic layout of the T-CAT apparatus used in this investigation
6 | de BROWUER et al.
of 0.4, α = 0.05, and 80% power (numerator df = 3). A subsequent
sensitivity analysis supported the use of at least 8–10 animals which
were preselected for stereotypical expression and as used elsewhere
(Shoji et al., 2012), per group.
Statistical analyses were performed with IBM® SPSS® (version
27; IBM® Corporation, USA), while GraphPad® Prism® (version
8.4; GraphPad® Software, USA) was used for the graphical representation of data. To establish the degree of correlation between
stereotypical expression and T-CAT alternation, time to first entry
and total time per trial, respectively, in the T-CAT, Spearman’s
rank-order correlations were run (data not represented graphically). Ordinary two-way analysis of variance (two-way ANOVA)
was applied to determine the effect that cohort and drug exposure (independent variables) had on the different behavioral parameters analyzed in the T-CAT, that is percentage alternation
(Figure 3), time to first entry (Figure 4a), and total time per trial
(Figure 4b) (dependent variables). Since selected N and H mice
presented with varying degrees of stereotypy before the onset of
drug exposure, one-way analysis of covariance (ANCOVA) was applied to determine the effect of drug exposure (independent variable) on the expression of postexposure stereotypy (dependent
variable) after controlling for the baseline expression of stereotypy (Figures 5 and 6). Then, paired t-tests (Gaussian-distributed
data) or Wilcoxon’s signed-rank tests (non-Gaussian-distributed
data) were run to establish the differences between the pre- and
postexposure expression of stereotypy in animals of the same
cohort and exposure groups. Data were assessed for normality
by means of Shapiro-Wilk. Grubb’s test was used to identify and
remove outliers (α = 0.05). All analyses were followed up with a
Bonferroni post hoc test where applicable. Statistical significance
was set at p < 0.05 and reported as Bonferroni-adjusted values,
while all significant and other noteworthy differences as mentioned were informed with effect magnitude calculations, that is
Cohen’s d-value reported with a 95% confidence interval and partial eta-squared value for ANOVA analyses.
3 | RESULTS
3.1 | T-CAT performance
3.1.1 | Association between T-CAT performance and
motor stereotypy in vehicle-exposed control N and H
expressing mice
Spearman’s correlations were run in all vehicle-exposed control
mice to establish the relationships between the time that animals
engaged in H and N behavior, and the percentage spontaneous
arm alternation, time to first arm entry in the T-CAT, and the total
time per trial in the T-CAT, respectively. No statistically significant
correlation existed between the time spent indulging in either H or N
behavior and the percentage alternation [H: rs
(17) = −0.03, p = 0.90;
N: rs
(17) = 0.06, p = 0.81], time to first entry [H: rs
(17) = −0.08,
p = 0.76; N: rs
(17) = −0.01, p = 0.97] and total time per trial [H:
rs
(17) = −0.13, p = 0.63; N: rs
(17) = −0.08, p = 0.76], respectively. We
further analyzed the relationships between the average and highest
intensity of stereotypical expression across both stereotypical
phenotypes and the percentage spontaneous alternation displayed
by N and H animals, respectively. Here also, no significant correlations
were demonstrated (data not shown).
FIGURE 3 Spontaneous alternation behavior displayed by N and H animals in the T-CAT following 28 days of drug exposure. Data
are representative of the individual alternation scores generated. Error bars represent mean ± 95CI. Ordinary two-way ANOVA followed
by *Bonferroni multiple comparison tests was applied. Significance is indicated compared to vehicle-exposed control H mice. N: animals
selected based on the criterium for normal stereotypy (Table 1); H: animals selected based on the criterium for high stereotypical activity
(Table 1); Ctrl: control; Istra 10: istradefylline 10 mg kg−1 day−1; Istra 20: istradefylline 20 mg kg−1 day−1; group sizes n = 8–9; T-CAT, T-maze
continuous alternation task. Purple points indicate female animals. Where the number of purple points disagrees with the number of
male or female animals in the text, data points are overlaid and cannot be indicated by means of coloring
| de BROWUER et al. 7
A two-way ANOVA revealed no statistically significant drug*cohort interaction [F(2,44) = 1.35; p = 0.27; Figure 3]. However, a significant main effect of drug exposure [F(2,44) = 4.25; p = 0.02], but
not cohort (p = 1.0), was shown. With respect to the arm alternation
behavior of H animals, both concentrations of istradefylline were
associated with significant and robust improvements in alternation
scores compared to control (10 mg kg−1 day−1: p = 0.02; d = 1.3 [0.3;
2.4]; 20 mg kg−1 day−1: p = 0.03; d = 1.4 [0.3; 2.5]).
No significant interaction between cohort and drug exposure was shown with respect to the time taken to make a first arm
entry [F(2,43) = 1.92, p = 0.16; Figure 4a]. Further, neither cohort
(p = 0.32) nor drug exposure (p = 0.6) had a significant main effect
on the results.
With respect to the total time taken to complete an individual
trial, no significant interaction between cohort and drug exposure
[F(2,43) = 0.39; p = 0.68] was shown (Figure 4b). Also, neither cohort (p = 0.6) nor drug exposure (p = 0.26) significantly influenced
the results.
3.2 | Postexposure expression of stereotypy
With respect to the percentage time spent engaging in no stereotypical activity (Figure 5a), that is bouts during which stereotypy counts
of zero (0) were generated, and after adjusting for preexposure behavioral variation (Table 2), statistically significant differences were
only identified in the postexposure values of H [F(2,21) = 5.73,
p = 0.01, 휂2
p = 0.35] and not N animals [F(2,21) = 1.78, p = 0.19, 휂2
p =
0.15]. In line with this, only istradefylline 10 mg kg−1 day−1 (p = 0.01)
increased the percentage of time during which H animals expressed
no stereotypical behavior, relative to vehicle-exposed control H
mice. The effect of istradefylline 20 mg kg−1 day−1 (p = 0.052) narrowly missed statistical significance. Further, paired tests were
conducted to determine the effect of the various drug exposures
(pre- vs. postexposure) on the expression of no stereotypical behavior by H animals within the respective exposure groups. In this regard, both 10 mg kg−1 day−1 (z = 2.68, p = 0.01) and 20 mg kg−1 day−1
(t(7) = 2.41, p = 0.05) increased the percentage time spent expressing
FIGURE 4 Time to first entry (a) and average total time per trial (b) in N and H animals in the T-CAT following 28 days of drug exposure
(a single outlier identified and removed in the H Ctrl and N Istra 10 groups, respectively). Data are representative of the individual alternation
scores generated. Error bars represent mean ± 95CI. Ordinary two-way analysis of variance was applied. N: animals selected based on the
criterium for normal stereotypy (Table 1); H: animals selected based on the criterium for high stereotypical activity (Table 1); Ctrl: control;
Istra 10: istradefylline 10 mg kg−1 day−1; Istra 20: istradefylline 20 mg kg−1 day−1; group sizes n = 8–9; T-CAT, T-maze continuous alternation
task. Purple points indicate female animals. Where the number of purple points disagrees with the number of male or female animals in the
text, data points are overlaid and cannot be indicated by means of coloring
8 | de BROWUER et al.
non-stereotypical behavior compared to the respective preexposure
values.
As per Figure 5b, which represents the percentage of time that
animals of the N and H cohorts spent expressing H behavior, statistically significant postexposure group differences were observed in
the behavior of H animals, after adjusting for preexposure behavioral
variations [F(2,21) = 7.26, p = 0.004, 휂2
p = 0.41] (Table 2). Again, only
istradefylline 10 mg kg−1 day−1 (p = 0.004) and not 20 mg kg−1 day−1
(p = 0.052) reduced the amount of time animals spent expressing H behavior, compared to their vehicle-exposed control H expressing counterparts. Paired comparisons further indicated that
both istradefylline 10 mg kg−1 day−1 (t(8) = 4.13, p = 0.003) and
FIGURE 5 Percentage time that N and H animals expressed no stereotypical activity (a) and H activity (b). Data are representative of the
unadjusted individual mean behavioral expression of each animal over the full dark cycles of the three preexposure (closed dots) and the
single postexposure (open dots) assessment. Error bars represent mean ± 95CI. One-way analysis of covariance based on adjusted means as
provided in Table 2 with the respective preexposure values set as covariate was applied, followed by *Bonferroni post hoc testing; ^
Paired
t- or #
Wilcoxon’s signed-rank tests of the pre- versus postexposure behavioral expression within each cohort and exposure groups. N:
animals selected based on the criterium for normal stereotypy (Table 1); H: animals selected based on the criterium for high stereotypical
activity (Table 1); Group sizes are reported in Table 2; Ctrl: control; Istra 10: istradefylline 10 mg kg−1 day−1; Istra 20: istradefylline
20 mg kg−1 day−1; Pre-Rx, preexposure values; Post-Rx, post-exposure values. Purple points indicate female animals. Where the number of
purple points disagrees with the number of male or female animals in the text, data points are overlaid and cannot be indicated by means of
coloring
| de BROWUER et al. 9
20 mg kg−1 day−1 (t(7) = 3.21, p = 0.01) reduced the percentage of
time that H animals of these exposure groups expressed stereotypical behavior, compared to their respective preexposure behavioral
expression.
Here, we analyzed both the whole night (represented as a 12-hr
value on the figures) average intensity of stereotypical expression,
that is the average stereotypy count generated across all twentyfour 30-min long dark cycle bouts (Figure 6a,c), as well as changes
in the average highest expression of stereotypy (Figure 6b,d) after
controlling for the preexposure degree of behavioral expression. For
the latter parameter, we analyzed the average intensity of the three
highest stereotypical bouts (thus represented as a 90-min value on
the figures) generated over the dark cycle since the postexposure
assessment consisted of a single night only.
There were significant differences in the average expression
of vertical stereotypy between the different postexposure groups
[F(2,21) = 8.25, p = 0.002, 휂2
p = 0.44; Figure 6a; Table 3]. Both concentrations of istradefylline lowered average vertical stereotypy
counts, relative to their vehicle-exposed control counterparts
(10 mg kg−1 day−1: p = 0.004; 20 mg kg−1 day−1: p = 0.009). These
reductions were also significant in terms of the pre- versus postexposure expression of H stereotypy by both istradefylline-exposed
groups (10 mg kg−1 day−1: z = −2.55, p = 0.01; 20 mg kg−1 day−1: t(7)
= 3.64, p = 0.01).
Statistically significant postexposure differences were also identified between the average highest vertical stereotypy scores generated by H animals [F(2,21) = 5.65, p = 0.01, 휂2
p = 0.35; Figure 6b;
Table 3]. Here, only istradefylline 20 mg kg−1 day−1 (p = 0.01) and
not 10 mg kg−1 day−1 (p = 0.051) exposed animals presented with
a reduction in the highest scores generated compared to vehicleexposed control mice. Still, both the lower (t(8) = 4.37, p = 0.002)
and the higher (t(7) = 3.44, p = 0.01) concentrations of istradefylline reduced the highest vertical stereotypy scores compared to the
respective preexposure values which were generated by the same
animals.
With respect to Figure 6c, the whole night average horizontal
stereotypy scores were significantly different between the various
exposure groups [F(2,21) = 5.10, p = 0.02, 휂2
p = 0.33; Table 3]. In this
FIGURE 6 Average dark cycle (12 hr) vertical (a) and horizontal (c) activity as well as the average highest (90 min) vertical (b) and
horizontal (d) stereotypical expression, displayed by H deer mice. Data are representative of the unadjusted individual mean behavioral
expression over the full dark cycles of the three preexposure and the single postexposure assessment (a and c), as well as the average of
the nine highest preexposure and the three highest postexposure 30-min stereotypy scores (b and d). Error bars represent mean ± 95CI.
One-way analysis of covariance based on adjusted means as provided in Table 3 with the respective preexposure values set as covariate was
applied, followed by *Bonferroni post hoc testing (postexposure comparisons between different exposure groups); ^
Paired t- or #
Wilcoxon’s
signed-rank tests of the pre- versus postexposure behavioral expression within exposure groups. Group sizes are reported in Table 3; Ctrl:
control; Istra 10: istradefylline 10 mg kg−1 day−1; Istra 20: istradefylline 20 mg kg−1 day−1; Pre-Rx, preexposure values; Post-Rx, postexposure
values. Purple points indicate female animals. Where the number of purple points disagrees with the number of male or female animals in
the text, data points are overlaid and cannot be indicated by means of coloring
10 | de BROWUER et al.
regard, low-dose istradefylline-exposed animals generated lower average horizontal stereotypy scores, compared to animals exposed to
higher dose istradefylline (p = 0.02). Similarly, only the lower dose
of istradefylline decreased the average horizontal stereotypy scores
generated over the whole dark cycle, relative to the preexposure behavior of the same animals (z = −2.67, p = 0.01).
In terms of the three highest postexposure horizontal stereotypy
scores generated (Figure 6d), no significant differences between
groups for said behavior, after controlling for preexposure behavioral variations were identified [F(2,21) = 3.30, p = 0.06, 휂2
p = 0.24;
Table 3]. In terms of pre- versus postexposure comparisons within
each exposure group, istradefylline 10 mg kg−1 day−1 (z = −2.55,
p = 0.01) and 20 mg kg−1 day−1 (z = −2.38, p = 0.02) reduced this
measure compared to the respective preexposure values generated
by the same animals.
4 | DISCUSSION
Although there have been significant advancements in our understanding of OCD, treatment resistance remains a common clinical dilemma (Brakoulias et al., 2017). Psychobiological research is
increasingly pointing to distinct patterns of dysfunction that may
underlie specific OC symptom dimensions, potentially contributing
to poor treatment response in some. To this extent, explorations of
the adenosinergic system may be a valuable target for investigation.
Repetitive behavioral phenotypes are a trait of several neuropsychiatric disorders, including OCD. A total of 40%–45% of deer mice of
both sexes exhibit such behaviors, which are persistent over time
but resemble human compulsivity in terms of its waxing and waning nature over the course of a single dark cycle. Since no study has
TABLE 2 Unadjusted and adjusted means (and variability) for
the postexposure percentage of time that animals engaged in no
and high stereotypical activity with the preexposure percentages as
a covariate
Post-RX group n
Unadjusted Adjusted
Mean SD Mean SE
N animals expressing N behavior (Figure 2a)
Ctrl 9 27.89 20.21 28.33 5.47
Istra 10 8 33.13 9.60 32.57 5.83
Istra 20 8 42.88 15.80 42.94 5.76
H animals expressing N behavior (Figure 2a)
Ctrl 8a 13.54 10.62 15.33 5.65
Istra 10 9 40.74 14.40 40.31 5.20
Istra 20 8 37.50 21.36 36.20 5.59
H animals expressing H behavior (Figure 2b)
Ctrl 8a 30.21 25.76 26.84 4.95
Istra 10 9 3.24 7.15 2.09 4.54
Istra 20 8 3.13 4.85 7.76 5.01
Outlier identified and removed from group; N animals did not express
any H behaviors prior to drug exposure.
TABLE 3 Unadjusted and adjusted means (and variability) for the average and highest postexposure vertical and horizontal stereotypy
scores generated by H animals with the preexposure values as a covariate
Post-RX group n
Unadjusted Adjusted
Mean SD Mean SE
Average vertical stereotypy (12 hr; Figure 3a)
Ctrl 8a 933.67 668.32 914.22 138.44
Istra 10 9 250.55 210.89 218.80 131.30
Istra 20 8 192.82 176.66 247.98 141.34
Highest average vertical stereotypy (90 min; Figure 3b)
Ctrl 8a 2,539.67 1,866.99 2,582.22 384.05
Istra 10 9 1,194.89 669.70 1,213.48 361.87
Istra 20 8 921.79 958.55 858.33 384.40
Average horizontal stereotypy (12 hr; Figure 3c)
Ctrl 8a 81.39 96.80 57.65 19.21
Istra 10 9 16.33 28.91 12.36 17.14
Istra 20 8 66.48 53.63 94.68 19.63
Highest average horizontal stereotypy (90 min; Figure 3d)
Ctrl 8a 172.00 171.09 143.50 31.42
Istra 10 9 54.56 74.71 36.76 29.10
Istra 20 8 21.13 23.17 69.65 33.23
Outlier identified and removed from group.
| de BROWUER et al. 11
explored the potential associations between cognitive flexibility and
stereotypy, or the role of the adenosinergic system in the model
before, we set out to elucidate the potential relationship between
cognitive flexibility, as assessed in the T-maze, and stereotypical expression in deer mice and sought to assess how both parameters
would respond to anti-adenosinergic intervention.
Here, we showed mainly that (a) no correlation exists between
the time spent engaging in either N or H behavior and the degree of
spontaneous alternation expressed by deer mice, (b) istradefylline
significantly improved the alternating behavior of H expressing animals, and (c) both doses of istradefylline reduced the expression of
stereotypical behavior in H, but not N deer mice.
With respect to our first main finding, we report that no noteworthy relationship seems to exist between the time spent engaging in stereotypical expression and arm alternation in the T-CAT.
We also analyzed whether the expression of no, that is bouts
during which no stereotypical behavior was expressed, and H behavior, respectively, correlated with putative measures of anxiety
in the T-CAT, that is time to first entry or overall time needed to
complete the T-CAT assessment (Estanislau & Morato, 2005), neither of which showed significant correlations. The data presented
here must be interpreted against the background of prior preclinical and clinical literature. Alternation in the T-maze is normally
used to test for various measures of cognitive flexibility and can
be applied in tests of spatial learning, spontaneous exploration, or
cue-directed behavioral engagement (Deacon & Rawlins, 2006). As
such, the influence of drug exposure T-maze behavior is differentially applied to draw test-specific conclusions. For example, it has
been shown that serotonin (5HT) 1A receptor agonists, for example
8-hydroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT) reduces spontaneous alternation in the T-CAT (Yadin et al., 1991), while anticholinergic drug exposure has been shown to impair cue-directed learning
in learned T-maze arm choice (Spangler et al., 1986). Conversely,
the T-maze can also be applied to screen for drugs with potential
pro-cognitive effects (Andriambeloson et al., 2014). Although prior
studies with istradefylline have been conducted in learned T-maze
behavior (Yohn et al., 2015), we aimed to assess the effects of this
drug on naturalistic and spontaneous alternation and thus applied a
T-CAT paradigm. To this end, our findings are informative. Although
cognitive deficits have been illustrated in OCD, findings remain inconsistent, likely due to different inclusion criteria that are applied
with respect to symptom dimension, comorbidity, gender, and age
(Abramovitch et al., 2019). For example, while it has been shown
that patients presenting with collecting compulsions show deficits
in response inhibition, set shifting, spatial orientation, and feedback learning (Morein-Zamir et al., 2014), OCD patients in general
seem not have broad impairments in cognitive performance (Olley
et al., 2007). Therefore, although the collective body of research
indicates that perturbations in cognitive performance characterize
OCD, these are likely phenotype and trait specific. Ironically, our
experimental design could well have fell victim to the same obstacles faced in clinical research with respect to the “inclusion criteria”
applied here. Importantly, since our H group consisted of animals
that expressed stereotypical behavior above specific cutoff values (Table 1), we cannot exclude the possibility that an alternative,
intensity-orientated approach to selecting animals for inclusion and
making use of different tests of anxiety may yield a different result,
highlighting a need for future study.
Notwithstanding, a different perspective on cognitive flexibility
and its association with stereotypy was borne from the second main
finding of this work. Specifically, arm choice alternation in H, but not
N animals increased under the influence of istradefylline, compared
to control exposure (Figure 3). However, neither concentration of
istradefylline affected the time to first entry (Figure 4a) or total
time per trial (Figure 4b), highlighting some key aspects for consideration. First, since N animals did not present with drug-sensitive
arm alternation behavior, the identification and selection of these
subjects as a suitable normal behavioral control is confirmed with
respect to both the motor and cognitive domains. Second, arm alternation in H mice and its improvement following istradefylline exposure is likely not an artifact of an anxiolytic effect, since neither
time to first entry, nor the total time per trial was altered. Further,
given that istradefylline has shown dose-dependent differences in
terms of its effect on memory impairment in models of Alzheimer’s
disease (Orr et al., 2018) and considering that performance in the
T-CAT partially relies on the processes of working memory (Deacon
& Rawlins, 2006), we employed a two-tiered dosing regimen. Still, we
show that both doses had similar effects on the spontaneous alternation behavior of H mice. As such, it could be postulated that compared to other tasks that specifically assess working memory, that is
the Morris water maze (Orr et al., 2018), deer mouse performance in
the T-CAT may be founded less on the processes of working memory
and more on cognitive constructs related to habit formation and/or
behavioral inflexibility. Still, this finding might relate to behavioral
disinhibition that could be expected from striatal A2A antagonism
(Fuxe et al., 2010). However, that istradefylline also markedly attenuated the expression of stereotypy (see below) and that the behavior
of N animals remained unaffected across all measures disagree with
such a conclusion.
In terms of stereotypical behavior, we show here that istradefylline administration significantly attenuated the expression of H
stereotypy, irrespective of the dose used (Figures 5 and 6). Not
only did istradefylline increase the amount of time that H animals
refrained from engaging in stereotypical behavior at all (Figure 5a),
but it also reduced the amount of time that H animals engaged
in the expression of H behavior (Figure 5b). This reciprocal relationship between the two variables might seem obvious; however,
a closer look at stereotypical expression is needed. Since stereotypy manifests on a binary scale from none to high, the reduction
in the time spent engaging in H behavior was not only reduced
but paralleled by an increase in the number of bouts that animals
could spent engaging in normal rodent activity, for example cage
exploration. This is true, since animals did not simply engage in
a lower degree of stereotypy that would otherwise qualify as intervals of N activity; rather, they did not engage in stereotypical
activity at all. This result can also not be explained by locomotor
12 | de BROWUER et al.
inhibition since the general activity of N animals exposed to istradefylline was indistinct from that of vehicle-exposed control
mice (Figure 5a). Moreover, istradefylline also broadly attenuated
the average (Figure 6a,c) and highest (Figure 6b,d) preexposure
stereotypical intensity displayed by subjects of the same exposure
groups, while these values were also significantly lower compared
to those generated by vehicle-exposed control H mice.
Such robust effects of istradefylline on the expression of stereotypy should be afforded closer attention. As alluded to earlier,
since istradefylline is an allosteric potentiator of dopaminergic
action at the D2 receptor (Fuxe et al., 2010), it could well have
been expected that the expression of motor stereotypy would
be exacerbated (Carr et al., 2001). To the contrary, our results
suggest that H mice are better able to exert control over stereotypical expression. Although this conclusion disagrees with theories that align with current treatment approaches (Fineberg &
Craig, 2007; Marazziti & Consoli, 2010), our data are congruent
with clinical reports from studies in Parkinson’s disease, indicating A2A antagonism to improve control over involuntary behavior,
thereby facilitating the execution of complex voluntary motor action (Chen, 2014; Chen & Cunha, 2020). A possible perspective on
this model could therefore be that rather than being reminiscent
of compulsive-like repetition, stereotypy in H deer mice might
more reflect inadequate control over motor behavior. While such
a notion would require further study, it should be highlighted that
the waxing and waning nature of drug-naïve stereotypy, the attenuated response thereof to environmental enrichment (Hadley
et al., 2006), and the way animals engage in the T-CAT would disagree with this theory. Our results could also be related to the
proposed role of anxiety in the manifestation of compulsivity.
We have previously argued that the waxing and waning nature of
stereotypical expression seems representative of clinical compulsivity in that mice express H behavior during some 30-min bouts
in the dark cycle only (Wolmarans et al., 2013). Such a temporal
pattern of behavioral expression could suggest that episodes of
mounting anxiety precede the execution of potentially anxiolytic
stereotypical bouts. Since amygdalar and hippocampal A2A antagonism is associated with anxiolytic-like effects in rodent models
(Wei et al., 2014), our findings align conceptually well with this
theory. While our results pertaining to the putative T-CAT-related
anxiety measures reported here do not support such a conclusion,
further investigation is warranted, especially since T-CAT performance and stereotypy might differentially associate with underlying anxiety.
This study was not without some noteworthy shortcomings
which require discussion. First, we initially selected 10 animals
for each drug exposure group based on the baseline stereotypy
assessment. This was also done with due cognizance of the importance to distribute animals as far as it is possible, equally between the different sexes. However, we did not foresee that some
of these would not complete the postexposure T-CAT testing
and thus the group sizes ultimately employed here (n = 8–9) may
have been too small to detect meaningful correlations between
stereotypy and cognitive performance. Second, following the
exclusion of certain animals from the data analysis following the
T-CAT assessment, group numbers were notably skewed between
the sexes, which prevented us from drawing sex-related conclusions. That said, from Figures 3–6, it is evident that, as it stands,
the behavior of both sexes is represented across the reported
ranges. This finding is broadly in line with our previous research
(de Brouwer, Fick, et al., 2020; Wolmarans et al., 2013), but would
need substantial future investigation to confirm. Still, given that
the animals included in this investigation were clustered based on
behavioral expression, that is without considering sex as a potential biological variable, the chosen group sizes were determined to
be adequate for comparisons of stereotypical expression. Third, we
did not employ other behavioral assessments that can accurately
highlight perturbations in specific neurocognitive constructs, that
is anxiety, habit formation, and behavioral disinhibition. In extension, this work also did not include neurobiological methods to
accurately identify the central targets of istradefylline’s actions.
Collectively, this impeded our ability to draw conclusions relating
to the biological locus of istradefylline’s action.
5 | CONCLUSION
In this work, we explored the relationship between stereotypical
behavior cognitive flexibility as reflected by spontaneous alternation
behavior in the T-CAT and stereotypy and investigated the effect
of anti-adenosinergic drug exposure on both parameters. Although
no correlation between the two constructs was shown, arm
alternation percentage improved and stereotypy in H animals
was reduced following chronic istradefylline exposure. As such,
these findings propose a common construct that could underlie
perturbations across both domains, which may be founded on
dysfunctional adenosinergic signaling. Future research in deer mice
is therefore directed toward divulging the potential role of striatal
versus extrastriatal adenosinergic involvement in altered cognitive
performance and the expression of H stereotypy. Also, since it is
uncertain to what extent the various neuroanatomical targets of
istradefylline’s actions contributed to our results, further research
should explore this question. Last, our data are congruent with
others in showing a definitive role for A2A receptor antagonism in
the improvement of cognitive flexibility. This not only highlights our
understanding of spontaneous stereotypy in deer mice, but also
sheds light on the potential relationship between compulsive-like
behavioral persistence and cognitive processing.
CONFLICT OF INTEREST
The authors declare no conflicts of interest pertaining to the current
publication.
AUTHOR CONTRIBUTIONS
Geoffrey de Brouwer – Investigation; Formal Analysis; Writing
– original draft; Writing – review & editing. Jaco Engelbrecht
| de BROWUER et al. 13
- Conceptualization; Data curation; Formal Analysis; Investigation;
Methodology; Project administration; Writing – original draft. Daniel
C. Mograbi – Conceptualization; Methodology; Supervision; Writing
– review & editing. Lesetja Legoabe – Conceptualization; Funding
acquisition; Resources; Writing – review & editing. Stephan F Steyn
- Conceptualization; Formal Analysis; Supervision; Writing – review
& editing. De Wet Wolmarans – Conceptualization; Data curation;
Formal Analysis; Funding acquisition; Investigation; Methodology;
Supervision; Writing – original draft; Writing – review & editing.
DECLARATION OF TRANSPARENCY
The authors, reviewers and editors affirm that in accordance to the
policies set by the Journal of Neuroscience Research, this manuscript
presents an accurate and transparent account of the study being reported and that all critical details describing the methods and results
are present.
PEER REVIEW
The peer review history for this article is available at https://publons.
com/publon/10.1002/jnr.24867.
DATA AVAILABILITY STATEMENT
Data available on request from the authors. The data that support
the findings of this study are available from the corresponding
author upon reasonable request.
REFERENCES
Abramovitch, A., McCormack, B., Brunner, D., Johnson, M., & Wofford,
N. (2019). The impact of symptom severity on cognitive function in
obsessive-compulsive disorder: A meta-analysis. Clinical Psychology
Review, 67, 36–44. https://doi.org/10.1016/j.cpr.2018.09.003
Abramowitz, J. S., & Jacoby, R. J. (2015). Obsessive-compulsive and related disorders: A critical review of the new diagnostic class. Annual
Review of Clinical Psychology, 11, 165–186. https://doi.org/10.1146/
annurev-clinpsy-032813-153713
Albert, U., Marazziti, D., Di Salvo, G., Solia, F., Rosso, G., & Maina, G.
(2018). A systematic review of evidence-based treatment strategies
for obsessive-compulsive disorder resistant to first-line pharmacotherapy. Current Medicinal Chemistry, 25(41), 5647–5661. https://doi.
org/10.2174/0929867325666171222163645
Andriambeloson, E., Huyard, B., Poiraud, E., & Wagner, S. (2014).
Methyllycaconitine-and scopolamine-induced cognitive dysfunction: Differential reversal effect by cognition-enhancing drugs.
Pharmacology Research & Perspectives, 2(4), e00048. https://doi.
org/10.1002/prp2.48
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (Vol. 5.). Author.
Bartz, J. A., & Hollander, E. (2006). Is obsessive–compulsive disorder
an anxiety disorder? Progress in Neuro-Psychopharmacology and
Biological Psychiatry, 30(3), 338–352. https://doi.org/10.1016/j.
pnpbp.2005.11.003
Bastia, E., Xu, Y.-H., Scibelli, A. C., Day, Y.-J., Linden, J., Chen,
J.-F., & Schwarzschild, M. A. (2005). A crucial role for forebrain adenosine A 2A receptors in amphetamine sensitization. Neuropsychopharmacology, 30(5), 891–900. https://doi.
org/10.1038/sj.npp.1300630
Bradbury, C., Cassin, S. E., & Rector, N. A. (2011). Obsessive beliefs
and neurocognitive flexibility in obsessive-compulsive disorder.
Psychiatry Research, 187(1–2), 160–165. https://doi.org/10.1016/j.
psychres.2010.11.008
Braem, S., & Egner, T. (2018). Getting a grip on cognitive flexibility.
Current Directions in Psychological Science, 27(6), 470–476. https://
doi.org/10.1177/0963721418787475
Brakoulias, V., Starcevic, V., Belloch, A., Brown, C., Ferrao, Y. A.,
Fontenelle, L. F., Lochner, C., Marazziti, D., Matsunaga, H., Miguel,
E. C., Reddy, Y., do Rosario, M. C., Shavitt, R. G., Shyam Sundar, A.,
Stein, D. J., Torres, A. R., & Viswasam, K. (2017). Comorbidity, age
of onset and suicidality in obsessive–compulsive disorder (OCD):
An international collaboration. Comprehensive Psychiatry, 76, 79–86.
https://doi.org/10.1016/j.comppsych.2017.04.002
Carr, K. D., Kim, G.-Y., & de Vaca, S. C. (2001). Rewarding and locomotoractivating effects of direct dopamine receptor agonists are augmented by chronic food restriction in rats. Psychopharmacology,
154(4), 420–428. https://doi.org/10.1007/s002130000674
Chen, J.-F. (2014). Adenosine receptor control of cognition in normal and
disease. International Review of Neurobiology, 119, 257–307.
Chen, J.-F., & Cunha, R. A. (2020). The belated US FDA approval of the
adenosine A 2A receptor antagonist istradefylline for treatment of
Parkinson’s disease. Purinergic Signalling, 16, 167–174.
Coles, M. E., Wirshba, C. J., Nota, J., Schubert, J., & Grunthal, B. A.
(2018). Obsessive compulsive disorder prevalence increases with
latitude. Journal of Obsessive-Compulsive and Related Disorders, 18,
25–30. https://doi.org/10.1016/j.jocrd.2018.04.001
Costenla, A. R., Diógenes, M. J., Canas, P. M., Rodrigues, R. J., Nogueira,
C., Maroco, J., Agostinho, P. M., Ribeiro, J. A., Cunha, R. A., & de
Mendonça, A. (2011). Enhanced role of adenosine A2A receptors in the modulation of LTP in the rat hippocampus upon ageing. European Journal of Neuroscience, 34(1), 12–21. https://doi.
org/10.1111/j.1460-9568.2011.07719.x
de Brouwer, G., Fick, A., Lombaard, A., Stein, D. J., Harvey, B. H., &
Wolmarans, D. W. (2020). Large nest building and high marbleburying: Two compulsive-like phenotypes expressed by deer mice
(Peromyscus maniculatus bairdii) and their unique response to serotoninergic and dopamine modulating intervention. Behavioural Brain
Research, 393, 112794. https://doi.org/10.1016/j.bbr.2020.112794
de Brouwer, G., Harvey, B. H., & Wolmarans, D. W. (2020). Naturalistic
operant responses in deer mice (Peromyscus maniculatus bairdii) and
its response to outcome manipulation and serotonergic intervention.
Behavioural Pharmacology, 31(4), 343–358. https://doi.org/10.1097/
FBP.0000000000000536
Deacon, R. M., & Rawlins, J. N. P. (2006). T-maze alternation in the rodent.
Nature Protocols, 1(1), 7–12. https://doi.org/10.1038/nprot.2006.2
Estanislau, C., & Morato, S. (2005). Prenatal stress produces more behavioral alterations than maternal separation in the elevated plus-maze
and in the elevated T-maze. Behavioural Brain Research, 163(1), 70–77.
https://doi.org/10.1016/j.bbr.2005.04.003
Ferre, S., Von Euler, G., Johansson, B., Fredholm, B. B., & Fuxe, K.
(1991). Stimulation of high-affinity adenosine A2 receptors decreases the affinity of dopamine D2 receptors in rat striatal membranes. Proceedings of the National Academy of Sciences of the United
States of America, 88(16), 7238–7241. https://doi.org/10.1073/
pnas.88.16.7238
Fineberg, N. A., & Craig, K. J. (2007). Pharmacological treatment for
obsessive–compulsive disorder. Psychiatry, 6(6), 234–239. https://
doi.org/10.1016/j.mppsy.2007.04.001
Fuxe, K., Marcellino, D., Borroto-Escuela, D. O., Guescini, M., FernándezDueñas, V., Tanganelli, S., Rivera, A., Ciruela, F., & Agnati, L. F. (2010).
Adenosine–dopamine interactions in the pathophysiology and treatment of CNS disorders. CNS Neuroscience & Therapeutics, 16(3), e18–
e42. https://doi.org/10.1111/j.1755-5949.2009.00126.x
Goddard, A. W., Shekhar, A., Whiteman, A. F., & McDougle, C. J. (2008).
Serotoninergic mechanisms in the treatment of obsessive–compulsive
14 | de BROWUER et al.
disorder. Drug Discovery Today, 13(7–8), 325–332. https://doi.
org/10.1016/j.drudis.2007.12.009
Gruner, P., & Pittenger, C. (2017). Cognitive inflexibility in obsessivecompulsive disorder. Neuroscience, 345, 243–255. https://doi.
org/10.1016/j.neuroscience.2016.07.030
Hadley, C., Hadley, B., Ephraim, S., Yang, M., & Lewis, M. H. (2006).
Spontaneous stereotypy and environmental enrichment in deer mice
(Peromyscus maniculatus): Reversibility of experience. Applied Animal
Behaviour Science, 97(2–4), 312–322. https://doi.org/10.1016/j.appla
nim.2005.08.006
Korff, S., J. Stein, D., & H. Harvey, B. (2008). Stereotypic behaviour in
the deer mouse: Pharmacological validation and relevance for obsessive compulsive disorder. Progress in Neuro-Psychopharmacology
and Biological Psychiatry, 32(2), 348–355. http://www.scopus.com/
inward/record.url?eid=2-s2.0-38749088723&partnerID=40&
md5=5c216806576b56a7bc85afa60d0e631e https://doi.org/
10.1016/j.pnpbp.2007.08.032
Marazziti, D., & Consoli, G. (2010). Treatment strategies for obsessivecompulsive disorder. Expert Opinion on Pharmacotherapy, 11(3), 331–
343. https://doi.org/10.1517/14656560903446948
Markarian, Y., Larson, M. J., Aldea, M. A., Baldwin, S. A., Good, D.,
Berkeljon, A., Murphy, T. K., Storch, E. A., & McKay, D. (2010).
Multiple pathways to functional impairment in obsessive–compulsive
disorder. Clinical Psychology Review, 30(1), 78–88. https://doi.
org/10.1016/j.cpr.2009.09.005
Milad, M. R., & Rauch, S. L. (2012). Obsessive-compulsive disorder:
Beyond segregated cortico-striatal pathways. Trends in Cognitive
Sciences, 16(1), 43–51. https://doi.org/10.1016/j.tics.2011.11.003
Morein-Zamir, S., Papmeyer, M., Pertusa, A., Chamberlain, S. R., Fineberg,
N. A., Sahakian, B. J., Mataix-Cols, D., & Robbins, T. W. (2014). The
profile of executive function in OCD hoarders and hoarding disorder. Psychiatry Research, 215(3), 659–667. https://doi.org/10.1016/j.
psychres.2013.12.026
Müller, T. (2013). Suitability of the adenosine antagonist istradefylline for
the treatment of Parkinson’s disease: Pharmacokinetic and clinical
considerations. Expert Opinion on Drug Metabolism & Toxicology, 9(8),
1015–1024. https://doi.org/10.1517/17425255.2013.795541
Nambu, A., Tokuno, H., & Takada, M. (2002). Functional significance of the cortico–subthalamo–pallidal ‘hyperdirect’ pathway.
Neuroscience Research, 43(2), 111–117. https://doi.org/10.1016/
S0168-0102(02)00027-5
Olley, A., Malhi, G., & Sachdev, P. (2007). Memory and executive functioning in obsessive-compulsive disorder: A selective review. Journal
of Affective Disorders, 104(1–3), 15–23. https://doi.org/10.1016/j.
jad.2007.02.023
Orr, A. G., Lo, I., Schumacher, H., Ho, K., Gill, M., Guo, W., Kim, D. H.,
Knox, A., Saito, T., Saido, T. C., Simms, J., Toddes, C., Wang, X., Yu,
G.-Q., & Mucke, L. (2018). Istradefylline reduces memory deficits in
aging mice with amyloid pathology. Neurobiology of Disease, 110, 29–
36. https://doi.org/10.1016/j.nbd.2017.10.014
Parkes, L., Tiego, J., Aquino, K., Braganza, L., Chamberlain, S. R.,
Fontenelle, L. F., Harrison, B. J., Lorenzetti, V., Paton, B., Razi, A.,
Fornito, A., & Yücel, M. (2019). Transdiagnostic variations in impulsivity and compulsivity in obsessive-compulsive disorder and
gambling disorder correlate with effective connectivity in corticalstriatal-thalamic-cortical circuits. Neuroimage, 202, 116070. https://
doi.org/10.1016/j.neuroimage.2019.116070
Pourcher, E., Fernandez, H. H., Stacy, M., Mori, A., Ballerini, R., &
Chaikin, P. (2012). Istradefylline for Parkinson’s disease patients
experiencing motor fluctuations: Results of the KW-6002-US-018
study. Parkinsonism & Related Disorders, 18(2), 178–184. https://doi.
org/10.1016/j.parkreldis.2011.09.023
Pujol, J., Soriano-Mas, C., Alonso, P., Cardoner, N., Menchón, J. M.,
Deus, J., & Vallejo, J. (2004). Mapping structural brain alterations in
obsessive-compulsive disorder. Archives of General Psychiatry, 61(7),
720–730. https://doi.org/10.1001/archpsyc.61.7.720
Raines, A. M., Allan, N. P., Oglesby, M. E., Short, N. A., & Schmidt,
N. B. (2015). Examination of the relations between obsessive–
compulsive symptom dimensions and fear and distress disorder
symptoms. Journal of Affective Disorders, 183, 253–257. https://doi.
org/10.1016/j.jad.2015.05.013
Rau, A. R., Ariwodola, O. J., & Weiner, J. L. (2015). Postsynaptic adenosine A2A receptors modulate intrinsic excitability of pyramidal cells in the rat basolateral amygdala. International Journal of
Neuropsychopharmacology, 18(6), pyv017. https://doi.org/10.1093/
ijnp/pyv017
Robbins, T. W., Gillan, C. M., Smith, D. G., de Wit, S., & Ersche, K. D.
(2012). Neurocognitive endophenotypes of impulsivity and compulsivity: Towards dimensional psychiatry. Special Issue: Cognition in
Neuropsychiatric Disorders, 16(1), 81–91. https://doi.org/10.1016/j.
tics.2011.11.009
Rosso, G., Albert, U., Asinari, G. F., Bogetto, F., & Maina, G. (2012).
Stressful life events and obsessive–compulsive disorder: Clinical
features and symptom dimensions. Psychiatry Research, 197(3), 259–
264. https://doi.org/10.1016/j.psychres.2011.10.005
Rufer, M., Fricke, S., Moritz, S., Kloss, M., & Hand, I. (2006). Symptom dimensions in obsessive–compulsive disorder: Prediction of cognitivebehavior therapy outcome. Acta Psychiatrica Scandinavica, 113(5),
440–446. https://doi.org/10.1111/j.1600-0447.2005.00682.x
Shen, H.-Y., Coelho, J. E., Ohtsuka, N., Canas, P. M., Day, Y.-J., Huang, Q.-
Y., Rebola, N., Yu, L., Boison, D., Cunha, R. A., Linden, J., Tsien, J. Z.,
& Chen, J.-F. (2008). A critical role of the adenosine A2A receptor in
extrastriatal neurons in modulating psychomotor activity as revealed
by opposite phenotypes of striatum and forebrain A2A receptor
knock-outs. Journal of Neuroscience, 28(12), 2970–2975. https://doi.
org/10.1523/JNEUROSCI.5255-07.2008
Shephard, E., Stern, E. R., van den Heuvel, O. A., Costa, D. L. C., Batistuzzo,
M. C., Godoy, P. B. G., Lopes, A. C., Brunoni, A. R., Hoexter, M. Q.,
Shavitt, R. G., Reddy, Y. C. J., Lochner, C., Stein, D. J., Simpson, H.
B., & Miguel, E. C. (2021). Toward a neurocircuit-based taxonomy
to guide treatment of obsessive–compulsive disorder. Molecular
Psychiatry. https://doi.org/10.1038/s41380-020-01007-8
Shoji, H., Hagihara, H., Takao, K., Hattori, S., & Miyakawa, T. (2012).
T-maze forced alternation and left-right discrimination tasks for assessing working and reference memory in mice. Journal of Visualized
Experiments, 60, 3300. https://doi.org/10.3791/3300
Spangler, E. L., Rigby, P., & Ingram, D. K. (1986). Scopolamine impairs
learning performance of rats in a 14-unit T-maze. Pharmacology
Biochemistry and Behavior, 25(3), 673–679. https://doi.
org/10.1016/0091-3057(86)90158-9
Stockwell, J., Jakova, E., & Cayabyab, F. S. (2017). Adenosine A1 and A2A
receptors in the brain: Current research and their role in neurodegeneration. Molecules, 22(4), 676.
Thorsen, A. L., Kvale, G., Hansen, B., & van den Heuvel, O. A. (2018).
Symptom dimensions in obsessive-compulsive disorder as predictors of neurobiology and treatment response. Current Treatment
Options in Psychiatry, 5(1), 182–194. https://doi.org/10.1007/s4050
1-018-0142-4
Torres, A. R., Fontenelle, L. F., Shavitt, R. G., Ferrão, Y. A., do Rosário,
M. C., Storch, E. A., & Miguel, E. C. (2016). Comorbidity variation in
patients with obsessive–compulsive disorder according to symptom
dimensions: Results from a large multicentre clinical sample. Journal
of Affective Disorders, 190, 508–516. https://doi.org/10.1016/j.
jad.2015.10.051
Vaghi, M. M., Vértes, P. E., Kitzbichler, M. G., Apergis-Schoute, A. M., van
der Flier, F. E., Fineberg, N. A., Sule, A., Zaman, R., Voon, V., Kundu,
P., Bullmore, E. T., & Robbins, T. W. (2017). Specific frontostriatal
circuits for impaired cognitive flexibility and goal-directed planning
| de BROWUER et al. 15
in obsessive-compulsive disorder: Evidence from resting-state functional connectivity. Biological Psychiatry, 81(8), 708–717. https://doi.
org/10.1016/j.biopsych.2016.08.009
Wei, C. J., Augusto, E., Gomes, C. A., Singer, P., Wang, Y., Boison, D.,
Cunha, R. A., Yee, B. K., & Chen, J.-F. (2014). Regulation of fear
responses by striatal and extrastriatal adenosine A2A receptors
in forebrain. Biological Psychiatry, 75(11), 855–863. https://doi.
org/10.1016/j.biopsych.2013.05.003
Wheaton, M. G., Sternberg, L., McFarlane, K., & Sarda, A. (2016). Selfconcealment in obsessive-compulsive disorder: Associations with
symptom dimensions, help seeking attitudes, and treatment expectancy. Journal of Obsessive-Compulsive and Related Disorders, 11, 43–
48. https://doi.org/10.1016/j.jocrd.2016.08.002
Williams, M. T., Mugno, B., Franklin, M., & Faber, S. (2013). Symptom dimensions in obsessive-compulsive disorder: Phenomenology and treatment outcomes with exposure and ritual prevention. Psychopathology,
46(6), 365–376. https://doi.org/10.1159/000348582
Wolmarans, D., Brand, L., Stein, D. J., & Harvey, B. H. (2013). Reappraisal
of spontaneous stereotypy in the deer mouse as an animal model of
obsessive-compulsive disorder (OCD): Response to escitalopram treatment and basal serotonin transporter (SERT) density. Behavioral Brain
Research, 256, 545–553. https://doi.org/10.1016/j.bbr.2013.08.049
Wolmarans, D. W., Scheepers, I. M., Stein, D. J., & Harvey, B. H. (2018).
Peromyscus maniculatus bairdii as a naturalistic mammalian model of
obsessive-compulsive disorder: Current status and future challenges.
Metabolic Brain Disease, 33(2), 443–455. https://doi.org/10.1007/
s11011-017-0161-7
Wood, J., & Ahmari, S. E. (2015). A framework for understanding the
emerging role of corticolimbic-ventral striatal networks in OCDassociated repetitive behaviors. Frontiers in Systems Neuroscience, 9,
171. https://doi.org/10.3389/fnsys.2015.00171
Yadin, E., Friedman, E., & Bridger, W. H. (1991). Spontaneous alternation behavior: An animal model for obsessive-compulsive disorder?
Pharmacology Biochemistry and Behavior, 40(2), 311–315. https://doi.
org/10.1016/0091-3057(91)90559-K
Yohn, S. E., Thompson, C., Randall, P. A., Lee, C. A., Müller, C. E., Baqi,
Y., Correa, M., & Salamone, J. D. (2015). The VMAT-2 inhibitor tetrabenazine alters effort-related decision making as measured by the
T-maze barrier choice task: Reversal with the adenosine A 2A antagonist MSX-3 and the catecholamine uptake blocker bupropion.
Psychopharmacology, 232(7), 1313–1323. https://doi.org/10.1007/
s00213-014-3766-0
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the
Supporting Information section.
Transparent Peer Review Report
Transparent Science Questionnaire for Authors
Supplementary Information
How to cite this article: de Browuer G, Engelbrecht J,
Mograbi DC, Legoabe L, Steyn SF, Wolmarans DW.
Stereotypy and spontaneous alternation in deer mice and its
response to anti-adenosinergic intervention. J Neurosci Res.
2021;00:1–15. https://doi.org/10.1002/jnr.24867