Neural basis of motor symptoms in Alzheimer's disease: role of regional tau burden and cognition

Abstract

Introduction: With accumulating evidence that motor manifestations in Alzheimer's disease (AD) may emerge from AD pathology independent of other copathologies, we investigated the neural basis of motor dysfunction under the amyloid/tau/neurodegeneration (ATN) framework.

Methods: This study included 125 patients with AD, excluding individuals with severe leukoaraiosis or comorbid Lewy body features beyond Parkinsonism. Associations of ATN burden with motor dysfunction were tested using multivariate regression models, followed by mediation analyses exploring the cognitive contribution to these associations.

Results: Tau burden in the prefrontal, sensorimotor, and parietal regions was associated with motor dysfunction independent of amyloid or neurodegeneration. The effect of parietal tau on motor function was fully mediated by visuospatial dysfunction, whereas prefrontal/sensorimotor tau exerted direct effects without cognitive mediation.

Discussion: Increased tau burden in the sensorimotor and frontoparietal association cortices may elicit motor dysfunction in AD through either cognition-dependent or cognition-independent mechanisms, with effects depending on the affected regions.

Highlights: Tau burden was intimately associated with motor symptoms independent of Aβ or atrophy. Tau in sensorimotor and frontoparietal association cortices may elicit motor symptoms. Prefrontal/sensorimotor tau exerted cognition-independent effects on motor symptoms. Parietal tau indirectly influenced motor symptoms through visuospatial dysfunction. Parietal tau-related motor dysfunction may be partly explained by apraxic features.

Keywords: Alzheimer's disease; cognitive dysfunction; motor deficits; positron emission tomography; tau proteins.

FIGURE 1

Frequency and profiles of motor symptoms in AD. (A) Bar graphs showing percentage of presence of motor symptoms in each motor subdomain. (B) According to the classification scheme presented, participants were classified into three groups: no or subtle motor signs, mild parkinsonian signs, and overt parkinsonism. The severity of motor symptoms increased in accordance with the level of cognitive impairment in the order of amyloid‐negative controls, prodromal AD, and AD dementia (p for trend < 0.001). AD, Alzheimer's disease; UPDRS‐III, Unified Parkinson's Disease Rating Scale Part III.

FIGURE 2

Association of motor symptoms with ATN biomarkers. (A) Linear regression models were conducted to investigate the associations between UPDRS‐III total scores and global ATN biomarkers. Model 1 examined each ATN biomarker separately, while Model 2 included all three biomarkers simultaneously. (B) Serial mediation analysis of amyloid and motor symptom burden using two mediators (tau and neurodegeneration) reveals that the effect of amyloid on motor symptom is fully mediated via a tau‐dependent pathway. All analyses (both regression models and mediation analyses) were adjusted for age, sex, education years, disease duration, presence of APOE ε4 allele, white matter hyperintensity burden (total Fazekas score), and MRI scanner type. Abbreviations: Aβ, amyloid‐beta; FBB, florbetaben; FTP, flortaucipir; MRI, magnetic resonance imaging; SE, standard error; SUVR, standardized uptake value ratio; TIV, total intracranial volume; UPDRS‐III, Unified Parkinson's Disease Rating Scale Part III.

FIGURE 3

Associations between regional tau burden and motor symptom severity. Regional ¹⁸F‐Flortaucipir SUVR in the prefrontal, sensorimotor, lateral parietal, and medial parietal cortices revealed significant positive associations with UPDRS‐III total scores based on multivariate linear regression models. Age, sex, education years, disease duration, presence of APOE ε4 allele, white matter hyperintensity burden, MRI scanner type, global amyloid burden, and cortical gray matter volume were used as covariates. MRI, magnetic resonance imaging; P _FWE, family‐wise error corrected p value; UPDRS‐III, Unified Parkinson's Disease Rating Scale.

FIGURE 4

Mediation analyses of cognitive dysfunction on association between regional tau burden and motor symptom severity. (A and B) Effects of tau burden in lateral parietal (A) and medial parietal cortex (B) were fully mediated by visuospatial dysfunction. (C) Schematic illustration depicting divergent mechanisms for motor dysfunction elicited by regional tau. Prefrontal and sensorimotor tau directly influence motor dysfunction (cognition‐independent pathway), while medial and lateral parietal tau exert their effect on motor dysfunction through a cognition‐dependent pathway. Abbreviations: FBB, florbetaben; FTP, flortaucipir; SE, standard error; SUVR, standardized uptake value ratio; UPDRS‐III, Unified Parkinson's Disease Rating Scale.