Co-morbidity and systemic inflammation as drivers of cognitive decline: new experimental models adopting a broader paradigm in dementia research

Abstract

Dementia prevalence increases with age and Alzheimer's disease (AD) accounts for up to 75% of cases. However, significant variability and overlap exists in the extent of amyloid-β and Tau pathology in AD and non-demented populations and it is clear that other factors must influence progression of cognitive decline, perhaps independent of effects on amyloid pathology. Coupled with the failure of amyloid-clearing strategies to provide benefits for AD patients, it seems necessary to broaden the paradigm in dementia research beyond amyloid deposition and clearance. Evidence has emerged from alternative animal model approaches as well as clinical and population epidemiological studies that co-morbidities contribute significantly to neurodegeneration/cognitive decline and systemic inflammation has been a strong common theme in these approaches. We hypothesise, and discuss in this review, that a disproportionate inflammatory response to infection, injury or chronic peripheral disease is a key determinant of cognitive decline. We propose that detailed study of alternative models, which encompass acute and chronic systemic inflammatory co-morbidities, is an important priority for the field and we examine the cognitive consequences of several of these alternative experimental approaches. Experimental models of severe sepsis in normal animals or moderate acute systemic inflammation in animals with existing neurodegenerative pathology have uncovered roles for inflammatory mediators interleukin-1β, tumour necrosis factor-α, inducible nitric oxide synthase, complement, prostaglandins and NADPH oxidase in inflammation-induced cognitive dysfunction and neuronal death. Moreover, microglia are primed by existing neurodegenerative pathology to produce exaggerated responses to subsequent stimulation with bacterial lipopolysaccharide or other inflammatory stimuli and these insults drive acute dysfunction and negatively affect disease trajectory. Chronic co-morbidities, such as arthritis, atherosclerosis, obesity and diabetes, are risk factors for subsequent dementia and those with high inflammatory status are particularly at risk. Models of chronic co-morbidities, and indeed low grade systemic inflammation in the absence of specific pathology, indicate that interleukin-1β, tumour necrosis factor-α and other inflammatory mediators drive insulin resistance, hypothalamic dysfunction, impaired neurogenesis and cognitive function and impact on functional decline. Detailed study of these pathways will uncover important mechanisms of peripheral inflammation-driven cognitive decline and are already driving clinical initiatives to mitigate AD progression through minimising systemic inflammation.

Figure 1

Inflammatory co-morbidities damage the brain. Severe (that is, severe sepsis) or prolonged systemic inflammation (that is, diabetes, atherosclerosis, obesity, arthritis), even when superimposed on the normal healthy brain (left: intact synaptic integrity and normal ramified microglia shown), can activate microglia and contribute to changes deleterious for cognitive function and thus increase dementia risk. Strength of induction of inflammatory mediators is shown in the dashed box and echoed by the red gradient. Similarly, when superimposed upon the already pathological brain (right: comprising β-amyloidosis, synaptic loss, neuronal death (green apoptotic nuclei in red-labelled neurons) and microglial activation), even relatively mild/moderate acute systemic inflammation can switch the phenotype of primed microglial cells to produce robust exacerbation of central nervous system (CNS) inflammation and to produce damage in the brain, which can contribute to long-term cognitive decline. Severe or prolonged inflammation superimposed on the already pathological brain is predicted to have even more deleterious consequences for trajectory of decline. Figure adapted from [106] and used with permission of Cambridge University Press. BDNF, brain-derived neurotrophic factor.

Figure 2

Recognition of microbial products and alarmins to induce systemic inflammation and impacts on the brain. Pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs or alarmins) induce systemic inflammatory mediators in multiple tissues of the body after infection, surgery, injury or arthritis. Although some aspects of the pathways shown remain unclear, it is clear that all conditions can bring about elevated systemic inflammatory mediators and that these can signal to the brain via well established routes, including direct neural activation via afferent nerves and activation of inflammatory cells in circumventricular organs lacking a patent blood–brain barrier, allowing secretion of inflammatory mediators into the brain parenchyma and activation of soluble mediators at the brain endothelium. Direct impacts on brain pathology or on cognitive function have been shown for all of these insults. Dashed arrows indicate that though these mediators are the result of inflammatory stimulation in the tissues/joints, they also contribute to the ongoing inflammation in those tissues. HMGB1, high mobility group box-1; IFN, interferon; IL, interleukin; LPS, lipopolysaccharide; NO, nitric oxide; PGN, peptidoglycan; ROS, reactive oxygen species; TNF, tumour necrosis factor.

Figure 3

Altered trajectories. Cognitive function may decline via stepwise decrements upon a declining baseline due to the cumulative effect of multiple acute systemic inflammatory events (SIEs; shown as lightning strikes, with corresponding acute decrements shown on the blue trajectory) but may also progress more rapidly due to the ongoing effects of chronic inflammatory co-morbidities (black, dashed trajectory) such as those discussed herein. The prediction is that underlying pathology such as amyloid beta may not manifest as dementia, or will manifest significantly later (disease without (w/o) acute SIEs, red trajectory), without the influence of these co-morbid factors (data based on [22,23,45]).

Figure 4

Inflammatory metabolic syndrome. This schematic summarises the key inflammatory stimuli arising from excessive nutrient intake, the main tissues experiencing inflammatory changes, the predominant inflammatory mediator output of these tissues and the impact of these changes on propagation of the metabolic syndrome and associated risk for Alzheimer’s disease. In particular it has emerged that hypothalamic inflammation produces hypothalamic dysfunction, which further disrupts central nervous system regulation of appetite and energy expenditure. Dashed arrows indicate that though these mediators are the result of inflammatory stimulation in the tissues/joints, they also contribute to the ongoing inflammation in those tissues. AGE, advanced glycation end products; CRP, C reactive protein; ER, endoplasmic reticulum stress; FFA, free fatty acids; IL, interleukin; LDL, low density lipoprotein; NO, nitric oxide; ROS, reactive oxygen species; tumour necrosis factor.