Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
, 76 (Pt B), 254-279

Obsessive-compulsive Disorder: Insights From Animal Models

Affiliations
Review

Obsessive-compulsive Disorder: Insights From Animal Models

Henry Szechtman et al. Neurosci Biobehav Rev.

Abstract

Research with animal models of obsessive-compulsive disorder (OCD) shows the following: (1) Optogenetic studies in mice provide evidence for a plausible cause-effect relation between increased activity in cortico-basal ganglia-thalamo-cortical (CBGTC) circuits and OCD by demonstrating the induction of compulsive behavior with the experimental manipulation of the CBGTC circuit. (2) Parallel use of several animal models is a fruitful paradigm to examine the mechanisms of treatment effects of deep brain stimulation in distinct OCD endophenotypes. (3) Features of spontaneous behavior in deer mice constitute a rich platform to investigate the neurobiology of OCD, social ramifications of a compulsive phenotype, and test novel drugs. (4) Studies in animal models for psychiatric disorders comorbid with OCD suggest comorbidity may involve shared neural circuits controlling expression of compulsive behavior. (5) Analysis of compulsive behavior into its constitutive components provides evidence from an animal model for a motivational perspective on OCD. (6) Methods of behavioral analysis in an animal model translate to dissection of compulsive rituals in OCD patients, leading to diagnostic tests.

Keywords: Animal model; Basal ganglia; Deer mouse; Endophenotypes; Nucleus accumbens core; Obsessive-compulsive disorder; Orbitofrontal cortex; Quinpirole; Security motivation system; Striatum.

Conflict of interest statement

Conflict of interests

Brian H. Harvey has participated in advisory boards and received honoraria from Servier®, and has received research funding from Servier® and Lundbeck®. BHH acknowledges that opinions, findings and conclusions or recommendations expressed in any publication generated by National Research Foundation (NRF) supported research are those of the authors, and that the NRF accepts no liability whatsoever in this regard. Remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Optogenetically induced increase in perseverative grooming. A) Channelrhodopsin (ChR2) is expressed in the ventromedial orbitofrontal cortex (VO), and a chronic fiberoptic implant in ventromedial striatum (VMS) is used to stimulate ChR2+ terminals projecting from VO. B) ChR2+ VO-VMS projections were stimulated 5 min/day for 6 days (T7 to T12). C) 6 days of stimulation led to a progressive increase in grooming 1 h post stimulation. D) 2 wk after repeated stimulation (T28), ChR2+ animals still demonstrate significantly increased grooming (Groomchronic *P < 0.03) despite no intervening stimulation. E) Progressive increase in evoked firing rate over course of multiple days. Z-scores indicate activity before vs. after light. (Modified from Ahmari et al., 2013).
Fig. 2
Fig. 2
MK-801 significantly increased daily mean (±SEM) water drinking across days in the schedule-induced polydipsia (SIP) paradigm. Experimental groups received saline (1.0 ml/kg) or the NMDA receptor blocker MK-801 (0.5 mg/kg) twice daily for 7 days followed by a 4-daywashout prior to the beginning of testing. Control groups received the same drug treatments but instead of receiving one food pellet each minute according to the fixed time schedule during daily 2-h sessions, they received 120 pellets in a dish placed next to the feeder cup in the test chamber. Only the experimental groups showed SIP and the MK-801 group drank more. *Analysis of variance revealed a significant 3-way interaction [group (MK-801 and saline) × condition (experimental and control) × day], F(1,36) = 5.88, p = 0.02. MK-801 Experimental and Saline Experimental groups did not differ significantly on day 1, t(22) = 0.98, p = 0.34, but by day 21 the MK-801 Experimental group was drinking more, t(22)=3.30, p = 0.004. From Hawken et al. (2011).
Fig. 3
Fig. 3
Y-maze test used to identify different response strategies in rat. During the training phase, food-restricted rats are started in the same arm on each trial and learn to choose the arm baited with a food pellet (Reward). On probe test trials, rats are started in the arm that was neither the usual start arm nor the arm where a food pellet was found. At the choice point, a right turn reflects a habit learning (striatal) strategy and a left turn reflects a place learning (hippocampal) strategy. No food reward is provided on probe trials.
Fig. 4
Fig. 4
Number of animal that used response (habit) or place-learning strategies in groups pre-treated for 5 days with amphetamine (AMPH; 1.5 mg/kg) or saline. † = significantly greater proportion than expected by chance in binomial probability test. From Gregory et al. (2015).
Fig. 5
Fig. 5
The heterogeneous nature of deer mouse stereotypy. Deer mouse stereotypy is heterogeneous within a given population of animals, with approximately 45% of animals classified as having high stereotypic behavior (HSB), 41% as having low stereotypic behavior (LSB), and 16% as being non-stereotypic (NSB). In this graph, deer mice are compared to C57Bl/6 mice as control. From Korff et al. (2008).
Fig. 6
Fig. 6
Differential response of deer mouse stereotypy to chronic fluoxetine and desipramine treatment. Effect of treatment with 20mg/kg fluoxetine, 20mg/kg desipramine and saline on stereotypic behavior of deer mice. Baseline (untreated) stereotypic activity for each treatment group (solid bars) is provided for high stereotypic behavior (H) mice. Data represent the average of three behavioral assessment sessions forthe baseline score and a once-off measurement for the treatment altered score (open bars), and expressed as the mean ± SEM. The number of animals (n) is shown below the indicated drug treatment. Locomotor effects following the various drug treatments were minimal (data not shown). *p<0.05 end-point vs baseline analysis for each treatment group (Student’s t-test). #p<0.05 end-point analysis compared to post-saline treatment (Dunnett’s test). From Korff et al. (2008).
Fig. 7
Fig. 7
Cortical but not striatal glutathione redox imbalance is correlated with severity of stereotypy in deer mice. Comparative oxidized (GSSG; top panel) and reduced (GSH; bottom panel) glutathione in the frontal cortex and striatum of non-stereotypic (NS), low stereotypic (LSB) and high stereotypic (HSB) deer mice (n = 20,16 and 24, respectively) are shown; **p<0.01, Bonferroni). Also shown are appropriate correlations between stereotypy count and GSSG or GSH in all animals (n = 60). From Guldenpfennig et al. (2011).
Fig. 8
Fig. 8
cAMP-PDE4 signaling in stereotypic deer mice, and response to fluoxetine. Top panel: Frontal cortical cAMP levels (A) and PDE4 enzyme activity (B) in low stereotypic (LSB) and high stereotypic (HSB) deer mice compared to non-stereotypic (NS) mice. Significant differences versus control NS mice are indicated by an asterisk (one-way ANOVA followed by the Tukey test; p<0.05). Data are expressed as mean ±S.E.M. Bottom panel: Effect of chronic fluoxetine or saline treatment (×21 days) on cAMP levels and PDE4 activity in the frontal cortex of HSB mice. Significant differences versus control SAL are indicated by an asterisk (Students t-test; p < 0.05). Data shown represent the mean ±S.E.M. From Korff et al. (2009).
Fig. 9
Fig. 9
Experimental set-up and test for compulsive checking. (a.) The open field apparatus with 4 objects on it. (b.) Subdivision of the open field into 25 places. The software algorithm assigns the positions of x,y coordinates of a stop within these locales. (c.) Test for compulsive checking on the 8th injection of quinpirole (0.5 mg/kg). Rats are said to show compulsive checking behavior when their performance is significantly different from saline controls on all 4 measures: frequency of checking (# of stops in key locale); length of check (mean duration in seconds of stay in key locale); recurrence time of checking (mean duration in seconds of return times to key place); and, # of stops before returning to check (mean number of places visited between returns to key locale). *p < 0.05 vs saline controls. Modified from Alkhatib et al. (2013).
Fig. 10
Fig. 10
Performance on criteria measures of compulsive checking behavior shown by groups of rats with lesion to the basolateral amygdala (BLA), nucleus accumbens core (NAc), orbital frontal cortex (OFC) or sham lesion. Blue bars represent groups with chronic saline treatment (left cluster of each panel) and red bars represent groups with chronic quinpirole treatment (right cluster of bars of each panel). Solid fill bars in top row show effect of NAc lesion on frequency of checking and length of check while those in the bottom row show effect of OFC lesion on recurrence of checking and stops before checking. * P < 0.05 vs. sham controls, BLA lesion, and OFC lesion groups treated chronically with saline; ** P<0.05 vs every group treated chronically with saline; *** P<0.05 vs. every other group; ## P<0.05 vs. every group treated chronically with quinpirole as well as sham controls and NAc groups treated chronically with saline. Modified from Dvorkin et al. (2010). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 11
Fig. 11
Performance on criteria measures for compulsive checking behavior shown by groups of sham controls and NAc core lesion rats treated with saline or DPAT. Open bars, sham controls injected with saline; right hatch, sham controls injected with DPAT; gray filled bars, NAc core lesion rats injected with saline; color filled bars, NAc core lesion rats injected with DPAT. * main effect of lesion; # main effect of drug. From Tucci et al. (2014a).
Fig. 12
Fig. 12
Duration of negative feedback signal as measured by time to next checking bout in rats treated chronically with saline (blue open circles) and quinpirole (red solid squares) during the course of treatment to induce compulsive checking. From Dvorkin et al. (2006). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 13
Fig. 13
A) the degree of anti-compulsive effects in the signal attenuation (SA) and quinpirole (QNP) model following high-frequency deep brain stimulation (DBS) to brain targets of the cortical-basal ganglia-thalamo-cortical circuit (CBGTC) ÷ = no effect; B) the CBGTC loop, including the differential neuropathology between the two models with respect to striatal neurotransmitter systems and cellular arrangement. DBS applied to the STN elicits symptom-comprehensive effects, whereas the NAc and GPe selectively reduce compulsivity in the QNP and SA model, respectively. 5-HT, serotonin; DA, dopamine; EP, entopeduncular nucleus; GPe, external globus pallidus; GPi, internal globus pallidus; LGP, lateral globus pallidus; NAc, nucleus accumbens; SNr, substantia nigra pars reticulata; STN, subthalamic nucleus.
Fig. 14
Fig. 14
The method of tracing the trajectories of locomotion and scoring the behavior in stopping places. The figure is based on data from Zadicario et al. (2007). a. Trajectories of traveling of two rats after the 18th injection of 0.5 mg/kg quinpirole (top) or saline (bottom). The trajectories represent the activity during the 60 min after the injection in a 2 × 2 m arena. As shown, the quinpirole rats traveled repeatedly the same paths in a restricted portion of the arena whereas the lesser activity of the saline rats spans over the entire arena, with seldom passing the same paths. b. Behavior of a quinpirole rat during 12 visits to the bottom right arena corner at which a small box was placed. Each row represents one visit, and the characters represent the following behaviors: A—arriving at a diagonal direction; B—snout contact with the box; C—climbing on top of the box; D—performing a large lateral turn; H—departure from the corner to the left. As shown, there is high regularity in the behavior of the rat over repeated visits to the corner.
Fig. 15
Fig. 15
Excerpt from the patient’s diary a. describing the ritual of turning on the TV. Behavior comprised systematic traveling between the bathroom and the TV, which could be schematically depicted as shown b. In the description of the acts when at the TV c., each visit to the TV during the ritual is depicted along one row, and each circle represents one act (similar acts are depicted in the same color).
Fig. 16
Fig. 16
The sequence of acts performed by a control individual (upper box) and an OCD patient (bottom box) as they lock and walk away from their car. Large circles depict common acts and small circles depict idiosyncratic acts. As shown, the control individual had only one idiosyncratic act and no repetition of acts, whereas the OCD patient had numerous idiosyncratic acts, repetition of common acts, and a long “tail” of idiosyncratic acts at the end of the task.
Fig. 17
Fig. 17
The number (mean ± SEM) of common acts (open bars) and idiosyncratic acts (gray bars) in the repertoire of acts (repetitions excluded) of 43 OCD rituals (bottom) and their non-OCD controls (top). The number of common acts performed in OCD and control rituals was identical (open bars). However, the number of idiosyncratic acts in OCD was three-fold that of controls (gray bars). The overall act repertoire was almost twice as large in OCD as in control rituals. Moreover, in the controls there were more common than idiosyncratic acts, whereas in the OCD patients it was the opposite: more idiosyncratic than common acts. (Based on data from Eilam et al. (2012).

Similar articles

See all similar articles

Cited by 12 articles

See all "Cited by" articles
Feedback