The Role of Cerebellum in Alzheimer’s Disease: A Forgotten Research Corner

Alzheimer’s disease (AD) is the most common cause of dementia, progressively degrading cognitive and social-emotional function [1]. The primary pathological features of AD include the accumulation of amyloid-β (Aβ) peptide, forming amyloid plaques in the extracellular space, and the aggregation of hyperphosphorylated tau protein into neurofibrillary tangles within neurons. These pathological changes contribute to neuronal dysfunction, cortical atrophy, and eventual cognitive decline [1]. While much AD research has focused on disease-associated changes of the cortex and hippocampus in the cerebrum; nevertheless, the role of the cerebellum has been largely overlooked [2]. Recently clinical and neuroimaging data has demonstrated the cerebellum’s involvement in cognition, language, and emotion, with further neuropathological and magnetic resonance imaging evidence revealing significant cerebellar alterations in AD [3]. Dr. Le’s team recently published a compelling paper in Alzheimer’s and Dementia, demonstrating that cerebellar electrophysiology and sleep-wake cycles were altered in APP^swe/PS1 ^ΔE9 mice [2]. Their findings suggest that changes in cerebellar electroencephalogram (EEG) and sleep-wake cycles may occur before cognitive decline [2]. Cerebral EEG has been used to determine whether patients with mild cognitive impairment could develop AD in long-term studies [4], and recently cerebellar EEG has been successfully employed in clinical studies [5].

Using the APP^swe/PS1^ΔE9 mouse model of AD, Yu et al. found that the power spectrum density of cerebellar EEG increased at 3, 6, and 9 months of age, with significant alterations during sleep-wake cycles, starting as early as 3 months of age [2]. According to their observations, Aβ deposits began to appear in the cerebrum around 6 months and in the cerebellum around 9 months of age, respectively [2]. Meanwhile, the AD mice developed impaired cognitive abilities at 6 months of age. There was no Aβ deposition in the brain nor cognitive impairment in the AD mice at 3 months of age [2].

However, even at this early stage, changes in cerebellar EEG and sleep-wake cycles were already present, suggesting that sleep-related electrophysiological alterations may precede cognitive impairment and the neuropathological onset of AD (Fig. 1).

It is well known that non-cognitive disorders such as sleep disturbances, as well as mood and personality changes often occur in the early disease course of AD. Epidemiological studies have shown that up to 45% of patients with AD suffer from sleep disturbances during the preclinical stage of the disease [6]. The cerebellum contributes to sleep regulations by communicating with the cerebral cortex via the superior cerebellar peduncle [7]. There is also evidence of the interactions between the cerebellum and the hippocampus during sleep [8]. The cerebellum receives inputs from systems regulating arousals and wake-sleep cycles, such as cholinergic inputs from the pedunculopontine nucleus [9] and projects to several brain regions that control arousals and sleep in turn [10]. Notably, increased Purkinje cells activity is accompanied by decreased activity in deep cerebellar nuclei neurons at the non-rapid eye-movement (NREM) sleep-wakefulness transition [11]. These findings provide in vivo electrophysiological evidence that the cerebellum has the potential to actively regulate the sleep-wakefulness transition [11].

OPEN IN VIEWER

Fig. 1

Cerebellar EGG and sleep-wake cycles significantly changed in APP^swe/PS1 ^ΔE9 mice as early as 3 months of age, before cognitive impairment at 6 months of age and Aβ deposits in the cerebellum at 9 months of age. Soluble human Aβ₄₀ and Aβ₄₂ increased rapidly in the cerebellum, cerebral cortex, and hippocampus at 9 months of age, and they were lower in the cerebellum than in the hippocampus and cerebral cortex at all ages. High expression of IDE and NEP may account for the resistance of cerebellum to Aβ accumulation in AD. Astrocytes and microglia were activated around Aβ depositions in the cerebral cortex and hippocampus from 6 months of age, and in the cerebellum only when senile plaques appeared in the cerebellar cortex at 9 months of age.

The role of 4pt the cerebellum in AD has been neglected for a long time, whereas some evidence in recent years has shown that cerebellar integrity is of importance in memory, language, and constructional praxis [1]. Cerebella tissues from 10 patients with severe AD and 10 age- and gender-matched controls showed a significant 12.7% reduction in total cerebellar volume in AD [12]. In addition, Olivito et al. reported a progression of cerebellar gray matter volume change continuously from the early to late clinical stages of AD [3]. It has been reported that the vermis and posterior lobe of the cerebellum are affected in the early stage of AD, and the anterior lobe becomes involved as the disease progresses [1].

The cerebellum is often regarded as a balance and motor control coordinate center. There is a clear segregation between “sensorimotor” and “cognitive” cerebellum, with the latter, in the posterior lobe (lobules VIIa, Crus I, and II), connecting to the prefrontal and posterior-parietal cortices related to cognitive and emotional functions [13]. White matter volume declines more rapidly than grey matter volume in the cerebellum and extends to the anterior lobe in patients with AD [14], supporting the hypothesis that connectivity deficits play an important role in the pathophysiology of AD [15]. The cerebral and cerebellar cortex, as well as the surrounding subcortical nuclei, are highly interconnected and function as a well-integrated system in both mice and primate species [16]. The cerebellum-cerebrum connection displays notable species differences between mice and humans, including size, complexity, cognitive functions, and connectivity patterns [17]. It is suggested that the cerebellum may also regulate cognitive functions primarily through interactions with different cortical regions. The dentate nucleus (DN) is the largest cerebellar nucleus and the major output channel to the cerebral cortex [13]. AD patients often show increased functional connectivity (FC) between DN and the lateral temporal lobe region [3]. The lower memory performance in AD may be associated with FC changes within specific functional modules of the cerebellar cortex, thus indicating the cerebellum contribution to AD pathophysiology and typical AD memory dysfunction [3].

The brainstem is another essential part of the cerebro-cerebellar network, within which the inferior olivary nuclei and the locus coeruleus are crucial for the modulatory input of the entire cerebellum (Fig. 1) [1]. The locus coeruleus has been postulated to be the initial site of tau pathology, which may potentiate amyloid pathology. Noradrenalin, a neuromodulator produced in the locus coeruleus, modulates cerebellar learning via its actions on other neurotransmitters. Noradrenalin promotes Aβ clearance by which the brainstem influences the Aβ metabolite in the cerebellar neurons [1].

The neuroinflammatory patches and glial cell proliferation constitute two main histopathological features of AD [18]. Commercially available AD transgenic and emerging mouse strains modeling late-onset AD have significantly advanced our understanding of AD pathology and potential therapeutic interventions [19]. In the present study by Yu et al., astrocytes and microglia were activated around Aβ depositions in the cerebral cortex and hippocampus at 6 and 9 months of age, whereas activated astrocytes and microglia were found in the cerebellum only when senile plaques appeared in the cerebellar cortex at 9 months of age (Fig. 1) [2]. Recent data confirms that Aβ is not the cause but the consequences of microglial activation and inflammation [20]. It appeared that glial fibrillary acidic protein (GFAP)-positive astrocytes in the granule cell layer could not migrate to the molecular layer of the cerebellum [2]. Although the soluble human Aβ₄₀ and Aβ₄₂ increased significantly in the cerebellum, cerebral cortex, and hippocampus at 9 months of age, they were lower in the cerebellum than in the hippocampus and cerebral cortex at all ages (Fig. 1) [2]. The insulin-degrading enzyme (IDE) and neprilysin (NEP) are proteases that degrade Aβ, whose levels are significantly higher in the cerebellum compared to AD-vulnerable regions such as the hippocampus and cortex in both AD mice and human [21]. Neuronal overexpression of IDE reduces Aβ levels and greatly retards cerebral plaque formation, which is even able to rescue the premature lethality present in APP^Swe/Ind transgenic mice [22]. High expression of IDE and NEP may account for the resistance of the cerebellum to Aβ accumulation in AD (Fig. 1) [21].

Studies suggest that tau pathology can spread to the cerebellum from other brain regions in advanced stages of the disease [23]. Falla et al. characterized patients with PS1-E280A mutant showed deposition of hyperphosphorylated tau (pTau) in the cerebellum. The presence of pTau in the cerebellum of PS1-E280A patients underscores the relevance of cerebellar involvement in AD and might be correlated to clinical phenotype [24]. The involvement of the cerebellum is quite variable depending on the mutations, region, and the cell type analyzed. More experiments are needed to understand the role of specific mutations and the involvement of tau in the pathology of animal models of AD [25].

Historically, EEG studies primarily focus on the cerebrum and little is known about the characteristics of cerebellar EEG in AD. The recent research conducted by Dr. Le’s team has shed light on altered cerebellar EEG and sleep-wake cycles prior to the cognitive impairment and neuropathological onset of AD, which provides a new perspective to study the cerebellum’s role in AD (Fig. 1) [2]. Additionally, investigating cerebellar changes in other transgenic mouse strains of the cerebellum’s contribution to AD pathology may provide valuable insights into disease progression and cognitive decline. However, it is noted that interpreting the results with any mouse model requires caution as they do not reflect the entire pathology of the disease.

Plenty of evidence has confirmed the involvement of the cerebellum in the pathogenesis of AD. Nevertheless, the specific biochemical and electrophysiological mechanisms underlying the cerebellum’s participation in AD pathology require further thorough investigation.