Abstract
Putra M, Rao NS, Gardner C, Liu G, Trommater J, Bunney M, Gage M, Bassuk AG, Hefti M, Lee G, Thippeswamy T. Brain Commun. 2024 Sep 19;6(5):fcae327. doi:10.1093/braincomms/fcae327. PMID: 39355003; PMCID: PMC11444080. Epilepsy and Alzheimer's disease share some common pathologies, such as neurodegeneration, seizures, and impaired cognition. However, the molecular mechanisms of these changes are still largely unknown. Fyn, a Src-family non-receptor tyrosine kinase (SFK), and its interaction with tau in mediating brain pathology in epilepsy and Alzheimer's disease can be a potential therapeutic target for disease modification. Although Fyn and tau pathology occurs in both Alzheimer's disease and epilepsy, the dynamics of Fyn–tau and PSD95–NR2B interactions affected by seizures and their impact on brain pathology in epilepsy have not been investigated. In this study, we demonstrate a significant increase of Fyn–tau interactions following seizure induction by kainate in both acute and chronic rodent models and in human epilepsy. In the early phase of epileptogenesis, we show increased Fyn/tau/NR2B/PSD95/neuronal nitric oxide synthase complexes after status epilepticus and a postsynaptic increase of phosphorylated tau (pY18 and AT8), Fyn (pSFK-Y416), NMDAR (pNR2BY1472), and neuronal nitric oxide synthase. Hippocampal proximity ligation assay and co-immunoprecipitation revealed a sustained increase of Fyn–tau and NR2B–PSD95 complexes/binding in rat chronic epilepsy at 3 months post-status epilepticus. Enhanced Fyn–tau complexes strongly correlated with the frequency of spontaneously recurring convulsive seizures and epileptiform spikes in the chronic epilepsy model. In human epileptic brains, we also identified increased Fyn–tau and NR2B–PSD95 complexes, tau phosphorylation (pY18 and AT8), and Fyn activation (pSFK-Y416), implying the translational and therapeutic potential of these molecular interactions. In tau knockout mice and in rats treated with a Fyn/SFK inhibitor, saracatinib, we found a significant reduction of phosphorylated Fyn, tau (AT8 in saracatinib-treated), NR2B and neuronal nitric oxide synthase and their interactions (Fyn–tau and NR2B–PSD95 in saracatinib-treated group; NR2B–PSD95 in tau knockout group). The reduction of Fyn–tau and NR2B–PSD95 interactions in the saracatinib-treated group, in contrast to the vehicle-treated group, correlated with the modification in seizure progression in the rat chronic epilepsy model. These findings from animal models and human epilepsy provide evidence for the role of Fyn–tau and NR2B–PSD95 interactions in seizure-induced brain pathology and suggest that blocking such interactions could modify the progression of epilepsy.
Commentary
Epilepsy and Alzheimer's disease (AD) share many behavioral and pathological features, such as increased neuroinflammation, gliosis, and temporal lobe atrophy (see Lehmann et al 1 ). AD is thus a useful disease state in which to uncover novel molecular drivers of seizures, especially in older adults. Adults with AD have a high incidence of comorbid seizures that inversely correlates to age of onset. Moreover, seizures can worsen functional and cognitive decline in affected AD cases.1,2 There are currently ∼7 million Americans with AD, and this number is expected to double by 2060. 3 There is thus an urgent need to develop treatments that can meaningfully slow AD progression. Since seizures can worsen cognitive decline, addressing seizures in those with AD may be an untapped disease-modifying treatment opportunity. Despite significant overlap in pathological features, the mechanism behind greater seizure incidence among those with AD is poorly understood. Three deterministic risk genes are associated with early-onset AD: amyloid precursor protein (APP), presenilin 1, and presenilin 2. All facilitate the production and cleavage of APP into amyloid-beta (Aβ), but variants in these genes have also been shown to have Aβ-independent mechanisms that also contribute to AD pathology. Mechanistic studies at the intersection of AD and epilepsy have primarily focused on mouse models that overproduce APP, accelerating Aβ plaque deposition. This singular focus on the contributions of Aβ to seizure onset and susceptibility has left other contributing factors, such as tauopathy and neuroinflammation, relatively underexplored, despite the fact that they correlate better with cognitive decline than Aβ levels. 4 While epilepsy itself is not generally considered a tauopathy, high levels of hyperphosphorylated tau are associated with a number of epilepsy syndromes, including Dravet syndrome and temporal lobe epilepsy (TLE). 5 Thus, therapies that selectively inhibit tau hyperphosphorylation represent a potentially promising treatment strategy for epilepsy, but further understanding how aberrant tau phosphorylation promotes hyperexcitability is still needed.
Putra et al 6 address this gap by assessing mechanisms of tau pathology through its interactions with Fyn in the kainic acid (KA) model of TLE. In rodents, repeated systemic low-dose KA administration results in acute seizures and status epilepticus (SE). KA-induced SE can lead to one of the most widely used preclinical models of spontaneous recurrent seizures (SRS) and acquired TLE. 7 This approach allowed Putra and colleagues to assess the dynamic alterations in Fyn–tau interactions during both acute and chronic stages of epileptogenesis. TLE is the most common form of epilepsy in adults and is characterized by SRS that are often resistant to current antiseizure medications (ASMs). 8 Human TLE is characterized by neurodegeneration, hippocampal sclerosis, mossy fiber sprouting, and alterations in hippocampal networks. Similarly, acute systemic KA administration to laboratory rodents consistently evokes hippocampal pyramidal cell loss, reactive gliosis, alterations in hippocampal synaptic plasticity, mossy fiber sprouting, and ASM resistance.7,8 Thus, Putra and colleagues sought to investigate whether tau, which has been shown to impact both long-term depression and long-term potentiation via its role in AMPA-type glutamate receptor (AMPAR) trafficking and NMDA-type glutamate receptor (NMDAR) stabilization at the synapse, 9 interacts in any way with Fyn. Notably, the Fyn tyrosine kinase, a member of the Src-family kinases, is an important modulator of synaptic plasticity in the CA1 region of the hippocampus via the tight regulation of the function of NMDARs.
In this study, Putra et al show that hippocampal Fyn–tau interactions are increased following KA-induced SE. They also demonstrate that Fyn–tau subsequently forms a complex with NMDA receptors, PSD95, and nNOS, stabilizing these receptors at the synapse and potentially increasing NMDA receptor-mediated current. These findings open a door to the development of new pharmacological treatments that can lower the likelihood of these interactions. Despite decades of research leading to the development of over 30 FDA-approved ASMs, approximately one-third of patients with epilepsy are considered drug resistant. Thus, there remains a need for the development of novel ASMs. Evidence for the therapeutic potential of Fyn–tau-targeting compounds is bolstered by the author's use of multiple animal models of TLE in addition to human brain samples from patients with epilepsy. This is especially exciting considering that these samples were donated from patients undergoing resection surgery for treatment of drug-resistant TLE. While there was no available data on seizure severity to correlate to Fyn–tau coupling in the human samples, there is compelling evidence from the rat KA-induced TLE model to suggest that the extent of Fyn–tau interaction correlates with the frequency of seizures. This rat model also allowed for the testing of a Fyn/Src-family non-receptor tyrosine kinase inhibitor, saracatinib, as a potential therapeutic option. Selective kinase inhibitors are emerging as a new class of ASMs. In particular, mTOR inhibitors have been shown to be successful in the treatment of epilepsy. In Putra et al, saracatinib effectively reduced SRS frequency in post-KA-SE animals, demonstrating its potential as a novel ASM. Based on these findings, it will be interesting to see subsequent cognitive behavioral data, how cognitive performance in the KA-induced rodent models of TLE correlates with Fyn–tau interaction, and if treatment with saracatinib will minimize any epilepsy-associated cognitive deficits. Whether other AD-associated risk factors, such as the presenilins or APP, demonstrate similar alterations in Fyn–tau interactions or whether saracatinib or other glutamate receptor modulating strategies, as employed herein, can beneficially improve seizure-related outcomes in these AD-related models or clinical cases is presently unknown.
Overall, these results demonstrate that targeting Fyn–tau interactions could be an effective treatment for TLE. Given that tau pathology occurs in both epilepsy and AD, it is possible that treatments focused on limiting this interaction could prove effective for both diseases. Small molecule therapies for symptomatic seizure control in people with epilepsy have historically targeted ion channels. Despite their obvious role in the propagation of seizures and development of epilepsy, this paper highlights that many other molecular targets may hold relevance upstream of the direct modulation of ion channel function. Similarly, the Aβ hypothesis has dominated AD research for more than three decades, but treatments focused on Aβ clearance have translated into therapies that provide only minor, if any, tangible improvements in cognitive function. 10 Novel molecular targets that can be gleaned from syndrome-specific models of chronic neurological disorders like AD, which are associated with high-frequency SRS, will be pivotal in developing pharmacological therapies that may meaningfully treat the underlying pathological mechanisms leading to seizures in people with epilepsy, especially for older adults with late-onset epilepsy as occurs in AD.
