AD is devastating, costly, incurable, and increasing in prevalence in aging populations around the world. Familial forms of AD (FAD) are caused by autosomal dominant mutations in three genes, human amyloid precursor protein (hAPP), presenilin 1 (PS1), and presenilin 2 (PS2), and by duplication of wild-type hAPP. The vast majority of these genetic alterations (~150 have been described since 1991) are associated with increased amyloid-β (Aβ) levels, providing strong human genetic evidence implicating this peptide in the pathogenesis of AD. Although not identical, familial and sporadic AD are clinically and pathologically similar, suggesting that they share common mechanisms of cognitive dysfunction.
Numerous lines of evidence suggest that Aβ peptides contribute to AD pathogenesis, but the exact in vivo mechanisms by which Aβ leads to cognitive decline are unknown. To study complex human diseases such as AD, we primarily use transgenic mouse models that recapitulate key aspects of the cognitive dysfunction and pathology of the human condition. Using these genetic tools, we aim to identify key mechanisms of hAPP/Aβ-induced cognitive dysfunction relevant to human AD. We are particularly interested in the relationships between cognitive decline, abnormal neuronal activity, GABAergic dysfunction, synaptic depression, and molecular and circuit alterations in brain regions critical for learning and memory. Our multidisciplinary approach includes genetic and pharmacologic manipulations in vivo, learning-dependent behavioral tests, long-term EEG recordings, slice physiology, in situ hybridization, immunohistochemistry, and western blots.
Epileptiform activity, compensatory responses, and cognitive decline
AD results in progressive dismantling of glutamatergic synapses, circuits, and neurons associated with a build-up of amyloid Aβ and tangle pathology. In vitro and in vivostudies from many laboratories have demonstrated that high levels of Aβ effectively suppress transmission strength or plasticity at specific glutamatergic synapses. However, the net effects on overall neuronal activity at the network level were unexplored until we discovered that transgenic models of AD with high levels of Aβ have generalized epileptiform activity and nonconvulsive seizures involving cortical and hippocampal networks. These findings indicate that synaptic depression and aberrant network activity coexist in AD (Fig. 1) and might be mechanistically related. Importantly, humans with AD, particularly early-onset AD or FAD, have an increased incidence of convulsive seizures. Such aberrant neuronal activity has been widely interpreted as a secondary process resulting from advanced stages of neurodegeneration. However, the above findings challenge this notion, raising the possibility that aberrant neuronal activity in AD is a primary upstream mechanism of high levels of Aβ.
Figure 1. Aβ can affect neuronal activity at multiple levels of complexity. High levels of Aβ depress excitatory synaptic transmission and impair synaptic plasticity at the level of specific synapses (left), but elicit epileptiform activity and seizures at the network level.
Figure 2. Model of Aβ-induced cognitive dysfunction. (A) High levels of Aβ induce epileptiform activity and compensatory inhibitory responses to counteract overexcitation, which may contribute to AD-related cognitive deficits. (B) Aβ-dependent circuit remodeling in the dentate gyrus of hAPP mice. These alterations likely reflect compensatory inhibitory responses to aberrant excitatory neuronal activity.
Our working model of Aβ-induced cognitive dysfunction proposes that high levels of Aβ lead to aberrant neuronal activity and compensatory inhibitory responses involving learning and memory circuits (Fig. 2), which may critically contribute to cognitive decline in AD mouse models and possibly in humans with AD. Aβ-induced epileptic activity likely triggers a variety of compensatory inhibitory responses in hippocampal circuits to counteract imbalances in network activity (Fig 2). However, these responses may also interfere with a number of normal neuronal and synaptic functions that are required for learning and memory. Hippocampal compensatory inhibitory responses may include depletion of calcium- and synaptic-regulated proteins (Fig. 2) such as calbindin, reduced levels of Arc/Arg3.1 and Fos, GABAergic sprouting, and ectopic expression of inhibitory neuropeptides. Most of these alterations are tightly associated with learning and memory deficits in independent lines of hAPP transgenic mice.
More importantly, experimental manipulations that prevent seizure activity and compensatory responses in hAPP mice also prevent cognitive deficits in these models. Confirming that Aβ-induced aberrant excitatory neuronal activity is causally related to hippocampal remodeling and cognitive decline would provide key insights into the pathogenesis of AD and open new therapeutic avenues.
- Does epileptiform or aberrant network activity contribute to cognitive dysfunction in AD and mouse models?
- Are synaptic depression and aberrant excitatory neuronal activity mechanistically related?
- What are the mechanisms of Aβ-induced epileptiform activity?
- Are inhibitory GABAergic interneurons more susceptible to Aβ-induced synaptic depression than glutamatergic neurons?
- Is hAPP/Aβ part of a negative feed back loop controlling neuronal activity?
- What is the relation between the pathological hallmarks of AD—amyloid Aβ plaques and tangles—and cognitive dysfunction?