A pharmacological and behavioural investigation of anxiety in the deer mouse model of obsessive- compulsive disorder

Abstract

Obsessive-compulsive disorder (OCD)1 is a chronic, debilitating psychiatric disorder with a global lifetime prevalence rate of 2.0%-2.9% and which generally manifests in early adulthood. The condition is characterised by two broad symptom groups, i.e., obsessions (disturbing and intrusive thoughts) and compulsions (persistent, repetitive, and overt behavioural routines). Although selective serotonin reuptake inhibitors (SSRIs)2 are currently used as the first line pharmacotherapeutic intervention for OCD, only 40%-60% of patients respond to treatment. The adjunctive treatment options for SSRI- refractory OCD include low dose anti-dopaminergic drugs, i.e. quetiapine or risperidone. Still, only 30% of SSRI-refractory patients respond to these interventions. Conceptually, compulsions can be viewed as an active coping mechanism recruited to attenuate the level of distress caused by the relevant obsession. In that regard, compulsions are 'goal-directed', although the goal in this instance does not speak to a realistic, attainable endpoint. It is unclear whether a sense of mounting anxiety provokes anxiolytic compulsive routines or whether patients struggle to refrain from engaging in excessive behaviours that in turn evoke a sense of anxiety. Importantly, OCD patients tend to overestimate and inflate the actual significance of the consequences which they feel may potentially arise should they not act on the obsession. Given the overly repetitive and often uncontrolled expression of compulsions, such behaviours are also appraised against the background of impulsivity, which exists on the opposite end of a continuum of excessive and inappropriate behaviours. Impulsive features are often described in OCD, pointing to a shared neurobiological and psychopathological architecture. Over the past decade our laboratory has carried out several studies to characterise, validate, develop, and scrutinise the naturalistic, persistent behavioural phenotypes expressed by deer mice (Peromyscus maniculatus bairdii) as a model of compulsive-like behavioural persistence. Spontaneous stereotypy, large nest building (LNB)3 and high marble-burying (HMB)4 are three behaviourally heterogeneous and persistent behaviours expressed by deer mice. Though previous studies indicated that chronic, high dose escitalopram (50 mg/kg/day) attenuates LNB behaviour and spontaneous stereotypy, HMB remains treatment-refractory to such intervention. While LNB behaviour is persistent and repetitive, the phenotype has not yet been studied in terms of the potential role that anxiety may play in its manifestation. In fact, since the immediate and long-

term temporal relationship between compulsive symptom manifestation and anxiety is not fully understood, it is unknown if anxiogenic manipulation of an environment, i.e. assessment of nesting behaviour under anxiogenic circumstances, will exacerbate the expression of LNB1, or whether the expression of LNB is a more predetermined and inflexible behavioural phenotype. As such, this work sought to determine whether the expression of normal (NNB)2 and LNB behaviour would adapt differently when assessed in conditions where mice would have to overcome their natural fear of open spaces to indulge in excessive nesting behaviour. We also aimed to determine how such behaviour, as measured under the aforementioned circumstances, would respond to known anti-compulsive and anxiolytic interventions. Deer mice (182, both sexes, aged 10 - 12 weeks at the onset of investigation) were screened for nest building behaviour, and subsequently divided into two primary behavioural cohorts i.e. NNB and LNB. Each cohort was then further subdivided into three (3) drug exposure groups [n = 10 per cohort per exposure group; all drugs administered via the drinking water; control, chronic escitalopram (50 mg/kg/day), and sub-acute lorazepam (2 mg/kg/day)]. Following selection and grouping, water (control) or escitalopram were administered for 28 days as well as during the four days of post- exposure testing. Lorazepam, as a sub-acute intervention, was only administered during the four nights of behavioural testing. Thus, mice in this group also received regular drinking water during the 28-day exposure period. After 28 days of control or escitalopram exposure, the primary behavioural investigation commenced. To this end, mice were placed into novel mirror chambers, which consisted of a safe, enclosed dark space with an internal nesting material hopper as well as a white-floored, mirror-walled open space containing an external nesting material hopper. The internal hopper contained enough material to construct a nest of typical size (as determined by analyses of home cage nesting behaviour), while an excess of material was available in the external hopper. Thus, to engage in LNB behaviour, mice had to enter the open, aversive space. The nesting behaviour of each mouse that was included in the drug (or control) exposure phase was assessed in this manner for four consecutive nights, all while being videotaped for assessment of ambulatory activity in the open space. To test the unincentivized exploration of the mirror chamber, two additional groups, i.e. one NNB and one LNB group (n = 8, both untreated) were assessed in the mirror chamber for four consecutive nights. However, no nesting material was available. Prior to commencement of this work, the study was approved by the AnimCare Research Ethics Committee of the North-West University (approval number: NWU-00574-19-A5).

The main findings of this study were the following: 1) LNB1 behaviour expressed by deer mice is an inflated, but goal-directed behavioural phenotype which remains stable, irrespective of the context in which it is assessed, 2) LNB mice find an open field arena to be less aversive compared to the behaviour or their NNB2 expressing counterparts, although this apparent unrestrained exploration of the open space is reduced when mice are able to indulge in the expression of LNB, and 3) escitalopram and lorazepam reduced the nesting behaviour of LNB mice, while differentially affecting the open field behaviours of NNB and LNB expressing mice. Since LNB animals presented with an inflated motivational drive to engage in nesting behaviour, irrespective of the potential negative outcomes, we have been able to confirm the excessive, persistent, and seemingly purposeless nature of LNB. LNB animals were also more likely to explore an environment associated with potential danger, and although it could be explained as differences in the underlying anxiety-state in NNB vs. LNB animals, there might be another explanation, i.e. that LNB mice exhibit impulsive-like, risk-taking behaviour. This may be true, since the expression of LNB not only responded to chronic escitalopram exposure, but also to sub-acute lorazepam; this taking into consideration that the same drug interventions differentially affected ambulatory activity in the open field. In conclusion, we confirmed that an anxiogenic environment does not deter LNB behaviour in deer mice and that excessive nest building is an inherent and stable behavioural phenotype that is displayed by LNB animals, irrespective of anxiety-related contextual circumstance. The potentially impulsive- like, risk-taking behaviour observed in LNB animals when they are unable to carry out compulsive-like nest building, is drastically attenuated with the introduction of nesting material, i.e. when being able to express LNB behaviour. In this case it appears that engaging in excessive nesting, distracts from impulsive-like exploration of the open field area. However, future studies are needed to further elucidate the full extent of the relationships between compulsive-, anxiety- and impulsive-like traits in