Parvalbumin interneurons and perineuronal nets in ageing and neurodegeneration

A fine balance of excitation and inhibition is required for normal cortical and hippocampal function. Parvalbumin (PV) interneurons are a distinct subtype of GABAergic interneuron with unique properties, including the perineuronal net (PNNs), which is a specialised extracellular matrix structure that commonly surrounds the soma and proximal dendrites. PV interneurons and PNNs are key regulators of the balance of excitation and inhibition as they provide a timing infrastructure for the cortex and hippocampus and enable coherent coordination throughout regions. In doing so, PV interneurons drive cortical oscillations in the gamma band frequency (40-90Hz); these oscillations are associated with higher order processes like sensory integration, learning, and memory. PV interneurons and PNNs likely play a role in facilitating effective integration of information in the cortex and hippocampus. Therefore, any disruption of these cellular and extracellular structures is likely to perturb their function. Indeed, there is evidence for alterations in the expression and function of PV and the PNN in a range of disorders and diseases that disrupt cognition and behaviour.
The aim of this thesis was to examine changes in PV interneurons and PNNs in ageing and neurodegeneration. Of particular interest were conditions in which pathophysiology was accompanied by disruptions to the physical structure of the cortex and hippocampus, including amyloidogenesis and traumatic brain injury (TBI). To this end, a series of experiments were established investigating the impact of amyloid accumulation in APP/PS1 mice and their wildtype C57BL/6J counterparts, and the effect of traumatic brain injury (TBI) in wildtype C57BL/6J mice.
First, amyloid-β plaque load, PV percentage area and PNN percentage area were quantified in 6-, 12-, 18-, and 24-month-old APP/PS1 mice and age-matched wildtype C57BL/6J. There was an increase in plaque load from 6 to 24 months in APP/PS1 mice in primary motor cortex, somatosensory barrel field cortex (S1BF), primary visual cortex, and dorsolateral entorhinal cortex. However, there were no significant differences between wildtype and APP/PS1 mice in the percentage area of PV and of wisteria floribunda lectin (WFA)-positive PNNs in the aforementioned regions. Similarly, there was no main effect of age, indicating that neither normal ageing, nor the accumulation of amyloid pathology had a detectable effect on the overall quantity of PV or WFA-positive PNNs present in select cortical regions.
Next, a cohort of APP/PS1 mice were subject to whisker trimming and sleep disruption intended to alter cortical plasticity and reduce sleep quality, which are both factors thought to be implicated in AD. Both sleep disruption and whisker trimming have independently been shown to alter PV density and PNN expression in rodents and affect plaque load in APP/PS1 mice. Male and female mice were assigned to one of four treatment groups (control, whisker trimming only, sleep disruption only, whisker trimming and sleep disruption) and aged to 6 or 9 months of age. Plaque load significantly increased between 6 and 9 months-of-age, while PV percentage area was significantly lower in 9-month-old animals, compared to 6-month?old animals. This age-related reduction in PV expression contradicts the findings reported in Chapter 3 where no change in PV was observed with age. This study was comparatively better powered and may have been better able to detect changes in PV percentage area. Additionally, both males and females were included here, meaning that sex differences could play a role in this reduction. Disrupting sleep increased WFA-positive PNNs, but only in 6- month-old animals. In the whisker trimming group, disrupted sleep was associated with a higher percentage area of PNN, with no differences between disrupted and normal sleep for animals in the sham whisker-treatment groups. These findings suggest that sleep disruption may impact PNNs in APP/PS1 mice, independent of effects on PV expression and not mediated by changes in plaque load.
TBI can cause persistent cognitive changes and result in ongoing neurodegeneration through both physical and physiological disruption of the cortex and hippocampus. To determine the effects of a mild to moderate TBI on PV interneurons and PNNs in the cortex and hippocampus, a midline fluid-percussion injury model was employed in male and female wildtype C57BL/6J mice. Brains were collected at 3 hours, 1, 3, or 7 days post-injury and compared to naïve animals. The density of PV-expressing cells was reduced in S1BF, hippocampal field cornu Ammonis 3 (CA3), and dentate gyrus (DG), however this recovered to levels indistinguishable from naïve animals by 7 days post-injury. In CA1, female mice demonstrated a significantly greater density of PV-expressing cells relative to male mice, irrespective of experimental time-point. The difference between the sexes was greatest at 1 day post injury, which may point to a role for diffuse TBI in exacerbating existing sex differences. Additionally, for PV percentage area measurements, there was an interaction between time and sex in DG, where female mice had a significantly higher percentage area of PV compared to males at 1 and 3 days post-injury. These findings demonstrate that both PV and PNN expression may be changed in the days following TBI and that these changes depend on sex, region, and time-course post-injury. These changes would be expected to have functional consequences for inhibition and plasticity and may contribute to the cognitive and behavioural changes observed following TBI.
Together these findings provide insight into the gross effects of neurodegeneration on PV interneurons and PNNs. The results indicate that the accumulation of amyloid pathology, which disrupts both the physical structure and physiological function of the cortex and hippocampus alone may not be sufficient to disrupt the presence of PNNs or alter the expression of PV. However, in the same mouse model, the addition of protocols that alter cortical plasticity and disrupt sleep can induce changes in the expression of the PNN, without impacting PV. However, more acute insults like TBI do appear to impact both PV interneurons and PNNs on a relatively short timescale. Therefore, further work may be warranted in exploring the contributions of PV interneurons and their PNNs in the cognitive impairment observed following TBI.