A metagenomic microbiome investigation in the deer mouse : a translational approach to obsessive-compulsive disorder

Abstract

Obsessive-compulsive disorder (OCD)1 is characterized by persistent, intrusive and often anxiety-provoking thoughts, i.e. obsessions and/or ritualistic behaviors (compulsions) that are expressed in an attempt to reduce the level of anxiety experienced. Although OCD is a common psychiatric illness that results in significant impairment in the normal functioning of patients, current pharmacotherapeutic strategies yield suboptimal results. Indeed, up to 40% of patients demonstrate treatment refractory symptomology to first-line intervention, i.e. chronic high-dose selective serotonin reuptake inhibitors (SSRIs)2, while a further 40 – 60% of such cases also remain non-responsive to augmentation strategies. While previous efforts at developing effective treatment have generally been aimed at modulating brain neurotransmitter function within the cortico-striatal-thalamic-cortical (CSTC)3 circuitry, recent research highlighted a possible role for dysbiosis, i.e. unstable changes in the gut microbiota, in psychiatric pathology. Although the relationship between dysbiosis and OCD is still largely unknown and taking into account that the exact neurobiological constructs underlying OCD have not yet been elucidated, investigations of the gut microbiota and its possible involvement in compulsive-like behavior, may provide valuable insight. In fact, the gut microbiota may potentially contribute to obsessive-compulsive pathology in meaningful ways, e.g. via modulation of immune responses in the central nervous system, alteration of neurotransmitter concentration and via indirect actions on the hypothalamus-pituitary axis (HPA)4. Therefore, the current project aimed to apply a validated animal model of naturalistically developing compulsive-like behavior, i.e. large nest building (LNB)5 to investigate whether such behavior can potentially be associated with alterations in the gut microbiota compared to normal nest building (NNB)6 controls. Further, as LNB has previously been shown to respond to chronic high-dose oral treatment with escitalopram, a clinically used SSRI, we wanted to establish whether the same treatment regimen would affect the composition of the gut microbiota differently in LNB, compared to NNB animal. We demonstrate here that the composition of the gut microbiota in LNB animals is significantly different from that in the NNB cohort. As LNB transpires naturally over the course of development and given that animals included in this investigation have been randomly selected without litter bias, indicates that the difference observed in microbiota composition naturally parallels the differences observed in behavioral expression. Further, we found Robinsoniella, a gram-negative, spore-forming and non-motile bacterial genus to be more abundant in LNB animals. That Robinsoniella was found to be more abundant in LNB1 compared to NNB2 mice, may provide some valuable direction for continued exploration in studies relating to the underlying role of the GBA3 in OCD4. However, in terms of causality, it needs to be determined whether an underlying neuropsychiatric construct that may be unique to LNB animals, i.e. alterations in neurotransmitter signaling or anxiogenic stress, elicited adaptive changes in the microbiota that is different from what is observed in the NNB cohort. It may well be possible that the microbiota composition of LNB animals can exert a bottom-up influence on the behavioral expression of LNB animals via nerve pathways or immunological signaling. This remains to be established in this model. Although our findings pertaining to the response of the gut microbiota to escitalopram intervention are not statistically significant (Yano et al., 2015), the adaptation of the microbiota in LNB animals trended towards being more extensive compared to what was found in the NNB cohort. Considering that escitalopram is known to have antimicrobial effects, it is important to highlight that dysbiosis will result from changes in the inherent abundance of different gut microbiota strains. It may therefore be of potential value if future investigations consider the antimicrobial actions of SSRI5 administration as a possible contributing factor to changes in central nervous system processes. Taken together, the data presented here provide for the first time in an investigation of OC6-like behavior in animals evidence that altered microbial composition parallels the manifestation of a naturally developing compulsive-like phenotype. Further, we also highlight a possible association between adaptations in the microbiota composition and escitalopram intervention. Future investigations into a possible causal role of the gut microbiota in the etiology of compulsive phenotypes, are warranted. Specifically, the relationship between compulsive phenotypes, physiological and psychological stress, vagus nerve signaling and immune alterations on the one hand and adaptations in the microbiota of normal and compulsive-like deer mice on the other hand, needs further elucidation. Further, it would be valuable to characterize the behavioral response in LNB deer mice both in the presence and absence of microbiota to establish a clear mechanism for its potent behavioral effect, as reported earlier. By providing a clearer roadmap for future investigation, such studies could possibly contribute to a better understanding of the neurobiology underlying OCD1 that may ultimately lead to the development of better treatment.