Chronic brain inflammation is associated with Alzheimer's disease (AD) and is classically attributed to amyloid plaque deposition. However, whether the amyloid pathology can trigger early inflammatory processes before plaque deposition remains a matter of debate. To address the possibility that a pre-plaque inflammatory process occurs, we investigated the status of neuronal, astrocytic, and microglial markers in pre- and post-amyloid plaque stages in a novel transgenic rat model of an AD-like amyloid pathology (McGill-R-Thy1-APP). In this model, we found a marked upregulation of several classical inflammatory markers such as COX-2, IL-1β, TNF-α, and fractalkine (CX3CL1) in the cerebral cortex and hippocampus. Interestingly, many of these markers were highly expressed in amyloid beta-burdened neurons. Activated astrocytes and microglia were associated with these Aβ-burdened neurons. These findings confirm the occurrence of a proinflammatory process preceding amyloid plaque deposition and suggest that Aβ-burdened neurons play a crucial role in initiating inflammation in AD.
Alzheimer's disease (AD) is the most frequent cause of cognitive decline and dementia, affecting more than 36 million people worldwide (World Alzheimer Report, 2012).
The principal pathologic features found in postmortem brains of AD sufferers are extracellular amyloid plaques, formed by aggregated amyloid-beta (Aβ) peptides, and neurofibrillary tangles composed of paired filaments of abnormally phosphorylated tau protein (Cuello et al., 2007, Grundke-Iqbal et al., 1986, Masters et al., 1985 and Selkoe, 2001).
At advanced stages of the pathology, in the presence of extracellular amyloid plaques, inflammation is an invariable component of the disease and might contribute to CNS dysfunction, injury, and ultimately loss of neurons (Haass and Selkoe, 2007). This overt inflammatory process is characterized by abundant reactive astrocytes and activated microglia surrounding amyloid plaques, by increased inflammatory molecules (cytokines and chemokines), and by the participation of the complement system (Akiyama et al., 2000).
The notion that inflammatory processes could play a role in the early AD pathology followed the discovery that patients with arthritis receiving lifelong treatment with nonsteroidal anti-inflammatory drugs (NSAIDs) had a lower incidence of AD (Andersen et al., 1995, McGeer et al., 1990 and Stewart et al., 1997); for a review: (Klegeris and McGeer, 2005). The data from these retrospective studies suggest that an inflammatory process should be present in the preclinical phases of the pathology, and that its inhibition could retard or prevent the onset of the clinical stages of the disease. In addition, in a prospective study, the NSAID naproxen diminished the incidence of AD in high-risk cohorts when given before the phase of subjective cognitive impairment (Breitner et al., 2011). However, prospective trials applying NSAIDs to clinically diagnosed AD cases have failed to slow down disease progression, suggesting that preventing inflammation at the clinical, more advanced stages is less likely to result in a beneficial outcome (Aisen et al., 2003, Martin et al., 2008 and Thal et al., 2005). The previously mentioned observations and the fact that minocycline, an anti-inflammatory tetracycline, corrects pre-plaque inflammation and reduces the amyloid component in a tg mouse model of AD (Ferretti et al., 2012a) would indicate that inflammation has a prominent disease-aggravating role in the initial phases of the disease. As most research efforts have been placed on characterizing the late, plaque-associated inflammatory response, little is known on the early inflammation occurring in preclinical AD stages.
The data on biomarkers emerging from familial and sporadic AD cases make a compelling case for the occurrence of a silent AD pathology starting decades before clinical diagnosis (Bateman et al., 2012 and Jack et al., 2013). As an acute injection of Aβ oligomers is sufficient to provoke the upregulation of inflammatory markers in a normal rat brain (Bruno et al., 2009), it is conceivable that, in the evolution of the AD pathology, a proinflammatory process is triggered by the early increase of soluble Aβ, before plaque deposition. We took advantage of a new transgenic (tg) rat model that reliably develops a progressive AD-like amyloid pathology to thoroughly characterize the inflammatory state of their brains at stages before and after amyloid plaque deposition.
The McGill-R-Thy1-APP tg rat model expresses the human APP751 transgene bearing the Indiana and Swedish mutations under the control of the murine Thy-1.2 promoter (Leon et al., 2010). As opposed to most tg mouse models containing multiple transgene copies, this rat model carries 1 copy per allele. This minimal transgenic charge is sufficient to reproduce an AD-like amyloid pathology and consequent cognitive impairments. Homozygous animals already display behavioral alterations at 3 months, a stage where Aβ immunoreactivity is limited to the intraneuronal compartment. The homozygous tg rats might display isolated amyloid plaques in the subiculum starting at the age of 6- to 8-month-old and invariably develop the full amyloid pathology as from 8- to 9-month-old.
We have used this animal model, closer to the human species than mice, to investigate whether a “pre-plaque” inflammatory process occurs before the overt extracellular aggregation of Aβ-amyloid peptides. Our investigations support the notion that Aβ-burdened neurons are key players in initiating this early inflammation. Such process would be akin to the earliest possible stages of the “silent AD” pathology in the human.
