Graduation Year

2020

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Medical Sciences

Major Professor

Danielle Gulick, Ph.D.

Committee Member

Paula Bickford, Ph.D.

Committee Member

Rex Philpot, Ph.D.

Committee Member

Lynn Wecker, Ph.D.

Committee Member

Kevin Nash, Ph.D.

Keywords

circadian rhythms, learning and memory, mouse model, schizophrenia

Abstract

Cognitive and circadian dysfunction are common aspects of neuropsychiatric disease that are often overlooked in disease models and treatment strategies. In this work, we first evaluated the potential for cognitive enhancement in healthy adult mice by targeting a kinase involved in regulation of the molecular circadian clock. We then developed and characterized a model of neuropsychiatric disease that could present a target for rescuing circadian and cognitive deficits using this treatment.

Time-of-day effects have been noted in a wide variety of cognitive behavioral tests, and perturbation of circadian rhythms impairs hippocampus-dependent learning and memory. We first asked whether casein kinase 1 (CK1) δ/ε inhibition might also be used to improve cognitive performance in the absence of disease. The oscillation of both clock genes and learning and memory-related genes in the hippocampus suggests that these rhythms might be targetable.

We used cognitive behavioral testing to determine whether chronic, short-term treatment with the pharmacological CK1δ/ε inhibitor PF-670462 could improve performance. Our behavioral results revealed improved performance in both Morris water maze and radial arm water maze, as well as contextual but not cued fear conditioning. Our molecular results also show a mild increase in phosphorylated extracellular signal-regulated kinase in the hippocampus of mice treated with PF-670462. While the hippocampus is essential for learning and memory, the amygdala also plays a role in both types of fear-based associative learning. We found an increase in activity-regulated cytoskeleton-associated (Arc) protein expression in the right amygdala of CK1i-treated mice exposed to an acute, negative learning experience. Previous studies show a critical role for Arc expression in synaptic plasticity and memory consolidation, lateralized to the right amygdala for negatively-valenced learning events.

Our molecular results suggest that CK1δ inhibition may interact with learning and memory systems in the hippocampus and amygdala to improve light cycle performance on behavioral tasks dependent on these areas of the brain, even in the absence of major circadian disruption. It is possible that this is due to a shift in the time-of-day efficiency of these processes resulting from the period-lengthening effects of CK1δ inhibition, or from stabilization of period proteins in the nucleus and subsequent stabilization of circadian rhythms.

After establishing the potential for PF-670462 to improve cognition, we next sought to develop a relevant model of neuropsychiatric disease that might benefit from a treatment option that addresses cognitive and circadian components.

The etiology of schizophrenia (SCZ) is complex, polygenic, and highly variable among individuals. Cognitive impairment is a core component of disease symptomology and is indicative of patient outcomes. This study evaluated a novel mouse model with disruptions in two putative SCZ susceptibility genes with roles in neurogenesis and synaptic plasticity, Disrupted-in-schizophrenia-1 (DISC1) and reelin (RELN). Heterozygous, haploinsufficient reeler mice were bred with mice expressing dominant-negative c-terminal truncated human DISC1 to produce offspring with both mutations: a transgenic model of impaired neurogenesis and neuropsychiatric disease. We evaluated the construct and face validity of the model, demonstrating SCZ-relevant behavioral alterations and underlying neuropathology.

HRM/DISC1 mice display some SCZ-relevant behavioral phenotypes. Prepulse inhibition of the acoustic startle response is used as a translatable measure of sensorimotor gating in both human and animal testing. We found a reduction in baseline startle response in HRM/DISC1 mice, as well as decreased prepulse inhibition. We found subtle alterations in circadian rhythms in HRM/DISC1 mice, including longer circadian periods during a phase shift and reduced bouts of voluntary wheel-running activity.

HRM/DISC1 mice have subtle cognitive and social behavioral alterations, displaying poor performance on water maze and social recognition tasks, and have no apparent changes in affect. The most striking behavioral findings are reduced exploratory behavior in the open field, decreased mobility in other tests, and reduction in bouts of voluntary wheel-running activity. These effects limit the ability of many of our behavioral assays to detect a potential cognitive phenotype.

One of the most consistent clinical finding in SCZ patients is abnormal gamma band oscillations. Gamma oscillations (~25-100hz) are generated and synchronized through inhibitory feedback by fast-spiking parvalbumin+ interneurons onto pyramidal neurons during a range of cognitive processes. We found a reduction in the total number of parvalbumin+ cells in the medial PFC of HRM/DISC1 mice. Reduced inhibitory output by PV+ GABAergic interneurons onto PFC pyramidal neurons could lead to behavioral deficits in the HRM/DISC1 mouse model.

Administration of the anxiolytic diazepam unexpectedly had paradoxical effects on HRM/DISC1 mice, failing to rescue the deficit in exploratory behavior we found in the open field test. As diazepam acts directly on GABAA receptors, we followed up on this finding by investigating the level of GABAA receptor subtype expression, finding increased expression of A1-containing GABA receptors in the prefrontal cortex of HRM/DISC1 mice.

Defects in adult hippocampal neurogenesis have been reported in patients with SCZ and in multiple animal models. We characterized the rate of neurogenesis, and the survival and maturation of adult-born neurons in HRM/DISC1 mice. While we found a no changes in the number of immature neurons in the hippocampus of HRM/DISC1 mice, these cells had reduced dendritic outgrowth. We also found an increased number of BrdU+ cells in the in the dentate gyrus of the dorsal hippocampus of the HRM/DISC1 brain, a lower proportion of which were NeuN-, indicating differentiation into non-neuronal cell types. Taken together, these results indicate a similar rate of neurogenesis, but suggest that the processes underlying the survival and maturation of adult-born neurons in the hippocampus are altered in the HRM/DISC1 brain.

Overall, we identified alterations in neuronal maturation and dendritic outgrowth in the hippocampus, alterations in GABAergic cells in the prefrontal cortex, and a complex behavioral profile best described by intact affect, reduced voluntary mobility, a strong sensorimotor gating deficit, and subtle changes in cognition and circadian rhythms. These alterations may present a target for treatment with chronotherapeutic interventions like CK1 inhibition.

Included in

Neurosciences Commons

Share

COinS