Nest building and marble burying: Dopaminergic phenotyping of the deer mouse model of obsessive-compulsive disorder (OCD)

Abstract

Obsessive-compulsive disorder (OCD1) is a chronic, debilitating psychiatric condition that affects 3% of the global population and is characterized by obsessions, and compulsions. 40 – 60% of patients remain treatment-resistant to first-line treatment with selective serotonin reuptake inhibitors (SSRI’s2). There is evidence to indicate that the different phenotypes of OCD differ in their underlying neurobiology, specifically with regards to differences in dysfunction of the cortico-striatal-thalamo-cortical (CSTC3) circuit, which could explain the varying treatment resistance in patients. The CSTC-circuit is well documented to play a key role in the epidemiology of OCD. There are only a handful of pre-clinical studies that attempt to investigate the neural underpinnings of OCD, likely due to the lack of animal models that are representative of symptom heterogeneous obsessive-compulsive-like behaviour. In this regard, the different naturalistic compulsive-like phenotypes exhibited by deer mice, large nest building (LNB4) and high marble burying (HMB5), have proved useful for the purpose of this investigation. These behaviours are equally persistent, repetitive and seemingly purposeless, and are thus reminiscent of OCD symptomology. Moreover, LNB demonstrates therapeutic response to chronic high dose (50 mg/kg/day) oral SSRI treatment, while HMB remains refractory to such intervention. Therefore, the current investigation aimed to establish whether HMB and LNB may be founded within unique neurobiological processes as explored by means of pharmacological manipulation. Specifically, we hypothesized that LNB will be largely unresponsive to anti-dopaminergic and combination treatment, whereas HMB will respond to anti-dopaminergic treatment alone or in combination with escitalopram. Thus, the purpose of this study was to explore the differences in treatment-response of the two aforementioned compulsive-like behavioural phenotypes expressed by deer mice, i.e. LNB and HMB, by means of pharmacological intervention with either an SSRI alone, i.e. escitalopram, a low-dose anti-dopaminergic drug, i.e. flupentixol, and a combination of the two. 160 deer mice, male and female, were initially screened for marble burying behaviour by conducting the marble burying test (MBT6). Briefly, marble-burying cages consisted of nine glass marbles placed at equal distances from one another on a 5 cm layer of coarse river

sand. A two-zone paradigm was followed, i.e. the marbles were placed in one half of the cage only. The screening consisted of a 30-minute session per mouse over three consecutive nights. Manual scoring of the marble directed behaviour (MDB1) took place post-screening. Next, all animals were screened for nest building behaviour by providing them with an excess of pre-weighed cotton wool every day for 7 consecutive days. On each morning, the built nests were removed and the remaining cotton wool weighed. Animals were assigned to either of the two cohorts (LNB2/HMB3). HMB was defined based on the total number of marbles buried and the consistency of burying behaviour over three nights. LNB was defined based on the total quantity of cotton wool used over 7 days and the consistency of nest sizes over the separate trials. Following initial screening, the cohorts were divided into four treatment groups (n=6 per group), namely water, escitalopram (50 mg/kg/day), flupentixol (0,9 mg/kg/day), or a combination of the two drugs. The animals were treated for 28 days, after which post-treatment screening took place in the respective cohorts as described above. Our results demonstrate a significant post-treatment reduction in the number of marbles buried in the control group only (p = 0.007; d = 2.5). However, with respect to MDB, there was an overall statistically significant effect of time on the behavioural response observed following four weeks of treatment (p = 0.0001). Hence, although escitalopram seemed to reduce the MDB exhibited by HMB animals, this effect was masked by the influence of time-based adaptation. In contrast, while the LNB behaviour of control-treated animals exacerbated over time (p = 0.025), escitalopram reduced the average total nest size over time (p = 0.2; d = 2.0), while such reduction in nest size was less robust for the combination treatment (p = 0.98; d = 0.5). Flupentixol alone had no effect. While the LNB results were expected, the observations in HMB expressing animals would indicate that HMB may be representative of a highly treatment-resistant behavioural phenotype that should be interrogated in terms of its underlying neurobiology. Indeed, future exploration of HMB may potentially provide significant insight into the mechanisms underlying treatment-resistant persistent behaviours. LNB, on the other hand, appears to represent the classic, serotonergic model of OCD4. This confirms our hypothesis that LNB and HMB differ in underlying neurobiology.