An experimental model of Braak’s pretangle proposal for the origin of Alzheimer’s disease: the role of locus coeruleus in early symptom development

Abhinaba Ghosh,
Sarah E. Torraville,
Bandhan Mukherjee,
Susan G. Walling,
Gerard M. Martin,
Carolyn W. HarleyEmail author and
Qi YuanEmail author View ORCID ID profile

Alzheimer's Research & Therapy201911:59

https://doi.org/10.1186/s13195-019-0511-2

Received: 8 March 2019
Accepted: 6 June 2019
Published: 3 July 2019

Abstract

Background

The earliest brain pathology related to Alzheimer’s disease (AD) is hyperphosphorylated soluble tau in the noradrenergic locus coeruleus (LC) neurons. Braak characterizes five pretangle tau stages preceding AD tangles. Pretangles begin in young humans and persist in the LC while spreading from there to other neuromodulatory neurons and, later, to the cortex. While LC pretangles appear in all by age 40, they do not necessarily result in AD prior to death. However, with age and pretangle spread, more individuals progress to AD stages. LC neurons are lost late, at Braak stages III–IV, when memory deficits appear. It is not clear if LC hyperphosphorylated tau generates the pathology and cognitive changes associated with preclinical AD. We use a rat model expressing pseudohyperphosphorylated human tau in LC to investigate the hypothesis that LC pretangles generate preclinical Alzheimer pathology.

Methods

We infused an adeno-associated viral vector carrying a human tau gene pseudophosphorylated at 14 sites common in LC pretangles into 2–3- or 14–16-month TH-Cre rats. We used odor discrimination to probe LC dysfunction, and we evaluated LC cell and fiber loss.

Results

Abnormal human tau was expressed in LC and exhibited somatodendritic mislocalization. In rats infused at 2–3 months old, 4 months post-infusion abnormal LC tau had transferred to the serotonergic raphe neurons. After 7 months, difficult similar odor discrimination learning was impaired. Impairment was associated with reduced LC axonal density in the olfactory cortex and upregulated β1-adrenoceptors. LC infusions in 14–16-month-old rats resulted in more severe outcomes. By 5–6 months post-infusion, rats were impaired even in simple odor discrimination learning. LC neuron number was reduced. Human tau appeared in the microglia and cortical neurons.

Conclusions

Our animal model suggests, for the first time, that Braak’s hypothesis that human AD originates with pretangle stages is plausible. LC pretangle progression here generates both preclinical AD pathological changes and cognitive decline. The odor discrimination deficits are similar to human odor identification deficits seen with aging and preclinical AD. When initiated in aged rats, pretangle stages progress rapidly and cause LC cell loss. These age-related outcomes are associated with a severe learning impairment consistent with memory decline in Braak stages III–IV.