An exploration into the bio-behavioural characteristics of the Ndufs4 (knockout) mice : a novel preclinical perspective on psychiatric conditions

Abstract

The brain and its associated disorders as a field of study are extremely complex, even more so when research strives to create models that mimic human psychiatric conditions. Given that psychiatric illnesses are associated with several risk factors (Kendler, 2019), attempting to pinpoint an overarching etiology has proven to be a difficult task. The link between mitochondrial dysfunction, neuropsychiatric, and neurodegenerative disease has increasingly been elucidated in recent years. As to why the mitochondrial dysfunction paradigm in psychiatric disease research eluded the scientific community up until recently, might be explained by the subtle nature of these dysfunctions. Dysfunctional mitochondria can appear in seemingly random anatomical areas, but can be masked by the compensating functional mitochondria, which creates a threshold effect. Essentially, a phenotypically healthy person can live a normal life with an unknown number of dysfunctional mitochondria, because of the larger functional percentile of the total mitochondrial mass, which compensates for the dysfunctional portion. Consequently, mitochondrial dysfunction could escape clinical psychiatric screening until overt symptoms appear when a stressor affects the existing threshold. The recent increase in research output largely stems from the fact that mitochondrial and neurodegenerative diseases often show pathophysiologic overlap, which is especially true considering that both disease categories present with comorbidities such as anxiety, depression, psychoses, etc. In this case, preclinical research is necessary to better define and understand the subset of cases where mitochondrial dysfunction forms part of the etiology and pathophysiology associated with psychiatric diseases. In this work, the Ndufs4 mouse was used as a point of departure to explore theories of mitochondrial dysfunction within the pre-clinical domain. Specifically, anxiety and depressive-like behavioral screening was employed with applicable neurochemical analyses. Thirty-six (n = 36) Ndufs4 mice, derived from which progenitor strain, were screened to identify three genotypes i.e., Ndufs4-/- knockout (KO), Ndufs4+/- heterozygous (HET), and Ndufs4+/+ wildtype. Male and female mice of each genotype were exposed to a battery of behavioral tests and neurochemical analyses. The behavioral tests were performed at an age where KO mice are phenotypically most comparable to controls, which allowed us to screen a seemingly healthy subject for behavioral and neurological deficits caused by mitochondrial dysfunction. The main goal of this study design and the repetitive nature of behavioral layout was to measure changes in the behavioral constructs and use combinations of these behavioral changes to address multiple constructs linked to psychiatric diseases. Firstly, between postnatal days (PND) 28 and 30, mice were subjected to the open field test to measure baseline locomotor activity. Here, the KO mice constantly covered less distance than the controls, yet the distance moved declined at a comparable rate between the three genotypes, over three open field trials. Similar results were

observed in the elevated plus maze (in terms of distance moved) on PND33, together with comparable time spent in the open arms of the maze between the different genotypes. Conversely, KO mice displayed increased risk resilience (risk-engaging) behavior in the mirror box test on PND34. As for depressive-like behavior, KO mice displayed a transient depressive-like phenotype between PNDs31 and 35. Initially, these animals were more immobile in the tail suspension test (PND31) than their controls yet displayed comparable behavior in the forced swim test on PND35. Taken together, the KO mice displayed risk resilience in the mirror box, that aligns with the suspected shift in biphasic behavior from depressive-like to non-depressive-like behavior. The fact the typical pharmacological interventions increase serotonin levels and decrease oxidative stress whilst alleviating depressive behavior, support the proposed like between behavioral and correlating neurochemical shifts, within the current study. This behavior is supported by increased hippocampal and cortical serotonin levels, and an improved redox state (increased GSH/GSSG) in the KO mice at PND36 that may have been caused by the dysfunctional mitochondria. These neurochemical deficits are supported by increased kynurenine/tryptophan turnover and decreased kynurenic acid/kynurenine ratios. These findings confirmed that Ndufs4 KO mice display bio-behavioral deviances between PND28 and 35, compared to WT mice. Considered together, our findings highlight the value of a rodent model founded on mitochondrial dysfunction for future research investigating mitochondrial contributions to neuropsychiatric illness.