Assessments of cognitive abilities in a mouse model of Parkinson’s disease with a touch screen test

Neuroanatomical and cognitive biomarkers of alpha-synuclein propagation in a mouse model of synucleinopathy prior to onset of motor symptoms

Significant evidence suggests that misfolded alpha-synuclein (aSyn), a major component of Lewy bodies, propagates in a prion-like manner contributing to disease progression in Parkinson's disease (PD) and other synucleinopathies. In fact, timed inoculation of M83 hemizygous mice with recombinant human aSyn preformed fibrils (PFF) has shown symptomatic deficits after substantial spreading of pathogenic alpha-synuclein, as detected by markers for the phosphorylation of S129 of aSyn. However, whether accumulated toxicity impact human-relevant cognitive and structural neuroanatomical measures is not fully understood. Here we performed a single unilateral striatal PFF injection in M83 hemizygous mice, and using two assays with translational potential, ex vivo magnetic resonance imaging (MRI) and touchscreen testing, we examined the combined neuroanatomical and behavioral impact of aSyn propagation. In PFF-injected mice, we observed widespread atrophy in bilateral regions that project to or receive input from the injection site using MRI. We also identified early deficits in reversal learning prior to the emergence of motor symptoms. Our findings highlight a network of regions with related cellular correlates of pathology that follow the progression of aSyn spreading, and that affect brain areas relevant for reversal learning. Our experiments suggest that M83 hemizygous mice injected with human PFF provides a model to understand how misfolded aSyn affects human-relevant pre-clinical measures and suggest that these pre-clinical biomarkers could be used to detect early toxicity of aSyn and provide better translational measures between mice and human disease.

Neuropsychiatric and Cognitive Deficits in Parkinson's Disease and Their Modeling in Rodents

Parkinson’s disease (PD) is associated with a large burden of non-motor symptoms including olfactory and autonomic dysfunction, as well as neuropsychiatric (depression, anxiety, apathy) and cognitive disorders (executive dysfunctions, memory and learning impairments). Some of these non-motor symptoms may precede the onset of motor symptoms by several years, and they significantly worsen during the course of the disease. The lack of systematic improvement of these non-motor features by dopamine replacement therapy underlines their multifactorial origin, with an involvement of monoaminergic and cholinergic systems, as well as alpha-synuclein pathology in frontal and limbic cortical circuits. Here we describe mood and neuropsychiatric disorders in PD and review their occurrence in rodent models of PD. Altogether, toxin-based rodent models of PD indicate a significant but non-exclusive contribution of mesencephalic dopaminergic loss in anxiety, apathy, and depressive-like behaviors, as well as in learning and memory deficits. Gene-based models display significant deficits in learning and memory, as well as executive functions, highlighting the contribution of alpha-synuclein pathology to these non-motor deficits. Collectively, neuropsychiatric and cognitive deficits are recapitulated to some extent in rodent models, providing partial but nevertheless useful options to understand the pathophysiology of non-motor symptoms and develop therapeutic options for these debilitating symptoms of PD.

Touchscreen cognitive testing: Cross-species translation and co-clinical trials in neurodegenerative and neuropsychiatric disease

Translating results from pre-clinical animal studies to successful human clinical trials in neurodegenerative and neuropsychiatric disease presents a significant challenge. While this issue is clearly multifaceted, the lack of reproducibility and poor translational validity of many paradigms used to assess cognition in animal models are central contributors to this challenge. Computer-automated cognitive test batteries have the potential to substantially improve translation between pre-clinical studies and clinical trials by increasing both reproducibility and translational validity. Given the structured nature of data output, computer-automated tests also lend themselves to increased data sharing and other open science good practices. Over the past two decades, computer automated, touchscreen-based cognitive testing methods have been developed for non-human primate and rodent models. These automated methods lend themselves to increased standardization, hence reproducibility, and have become increasingly important for the elucidation of the neurobiological basis of cognition in animal models. More recently, there have been increased efforts to use these methods to enhance translational validity by developing task batteries that are nearly identical across different species via forward (i.e., translating animal tasks to humans) and reverse (i.e., translating human tasks to animals) translation. An additional benefit of the touchscreen approach is that a cross-species cognitive test battery makes it possible to implement co-clinical trials-an approach developed initially in cancer research-for novel treatments for neurodegenerative disorders. Co-clinical trials bring together pre-clinical and early clinical studies, which facilitates testing of novel treatments in mouse models with underlying genetic or other changes, and can help to stratify patients on the basis of genetic, molecular, or cognitive criteria. This approach can help to determine which patients should be enrolled in specific clinical trials and can facilitate repositioning and/or repurposing of previously approved drugs. This has the potential to mitigate the resources required to study treatment responses in large numbers of human patients.

Cognitive flexibility deficits in rats with dorsomedial striatal 6-hydroxydopamine lesions tested using a three-choice serial reaction time task with reversal learning

Lesions of the dorsomedial striatum elicit deficits in cognitive flexibility that are an early feature of Parkinson's disease (PD), and presumably reflect alterations in frontostriatal processing. The current study aimed to examine deficits in cognitive flexibility in rats with bilateral 6-hydroxydopamine lesions in the dorsomedial striatum. While deficits in cognitive flexibility have previously been examined in rodent PD models using the cross-maze, T-maze, and a food-digging task, the current study is the first to examine such deficits using a 3-choice serial reaction time task (3-CSRT) with reversal learning (3-CSRT-R). Although the rate of acquisition in 3-CSRT was slower in lesioned compared to control rats, lesioned animals were able to acquire a level of accuracy comparable to that of control animals following 4 weeks of training. In contrast, substantial and persistent deficits were apparent during the reversal learning phase. Our results demonstrate that deficits in cognitive flexibility can be robustly unmasked by reversal learning in the 3-CSRT-R paradigm, which can be a useful test for evaluating effects of dorsomedial striatal deafferentation and interventions.

Optimizing reproducibility of operant testing through reinforcer standardization: identification of key nutritional constituents determining reward strength in touchscreens

Reliable and reproducible assessment of animal learning and behavior is a central aim of basic and translational neuroscience research. Recent developments in automated operant chamber technology have led to the possibility of universal standard protocols, in addition to increased translational potential, reliability and accuracy. However, the impact of regional and national differences in the supplies of available reinforcers in this system on behavioural performance and inter-laboratory variability is an unknown and at present uncontrolled variable. Therefore, we aimed to identify which constituent(s) of the reward determines reinforcer strength to enable improved standardization of this parameter across laboratories. Male C57BL/6 mice were examined in the touchscreen-based fixed ratio (FR) and progressive ratio (PR) schedules, reinforced with different kinds of milk-based reinforcers to directly compare the incentive values of plain milk (PM, high-calorie: high-fat/low-sugar), strawberry-flavored milk (SM, high-calorie: low-fat/high-sugar), and semi-skimmed low-fat milk (LM, low-calorie: low-fat/low-sugar) on the basis of differences in caloric content, sugar/fat content, and flavor. Use of a higher caloric content reward was effective in increasing operant training acquisition rate. Total trial number completed in FR and breakpoint in PR were higher using the two isocaloric milk products (PM and SM) than the lower caloric LM, with comparable outcomes between PM and SM conditions, suggesting that total caloric content determines reward strength. Analysis of within-session changes in response rate revealed that overall outputs in FR and PR primarily depend on the response rate at the initial phase of a session, which itself was dependent on reinforcer caloric content. Interestingly, the rate of satiation, indicated by decay in response rate within a FR session, was highest when reinforced with SM, suggesting a rapid satiating effect of sugar. The key contribution of reward caloric content to operant performance was confirmed in a multi-laboratory study using the touchscreen 5-choice serial reaction time task (5-CSRTT) reinforced by two isocaloric milk-based liquid rewards with different countries of origin, which yielded consistent performance parameters across sites. Our results indicate that milk-based liquid reinforcer standardization can be facilitated by matching caloric content across laboratories despite regional or national differences in other non-caloric aspects of the reinforcers.