The ubiquitous metabolic intermediary and nucleoside adenosine is a “master regulator” in all living systems. Under baseline conditions adenosine kinase (ADK) is the primary enzyme for the metabolic clearance of adenosine. By regulating the availability of adenosine, ADK is a critical upstream regulator of complex homeostatic and metabolic networks. Not surprisingly, ADK dysfunction is involved in several pathologies, including diabetes, epilepsy, and cancer. ADK protein exists in the two isoforms nuclear ADK-L, and cytoplasmic ADK-S, which are subject to dynamic expression changes during brain development and in response to brain injury; however, gene expression changes of the Adk gene as well as regulatory mechanisms that direct the cell-type and isoform specific expression of ADK have never been investigated. Here we analyzed potential gene regulatory mechanisms that may influence Adk expression including DNA promoter methylation, histone modifications and transcription factor binding. Our data suggest binding of transcription factor SP1 to the Adk promoter influences the regulation of Adk expression.

Hypersynchronous neuronal excitation manifests clinically as seizure (ictogenesis), and may recur spontaneously and repetitively after a variable latency period (epileptogenesis). Despite tremendous research efforts to describe molecular pathways and signatures of epileptogenesis, molecular pathomechanisms leading to chronic epilepsy remain to be clarified. We hypothesized that epigenetic modifications may form the basis for a cellular memory of epileptogenesis, and used a primary neuronal cell culture model of the rat hippocampus to study the translation of massive neuronal excitation into persisting changes of epigenetic signatures and pro-epileptogenic target gene expression. Increased spontaneous activation of cultured neurons was detected 3 and 7 days after stimulation with 10 μM glutamate when compared to sham-treated time-matched controls using calcium-imaging in vitro. Chromatin-immunoprecipitation experiments revealed short-term (3 h, 7 h, and 24 h) and long-term (3 d and 2 weeks) changes in histone modifications, which were directly linked to decreased expression of two selected epilepsy target genes, e.g. excitatory glutamate receptor genes Gria2 and Grin2a. Increased promoter methylation observed 4 weeks after glutamate stimulation at respective genes suggested long-term repression of Gria2 and Grin2a genes. Inhibition of glutamatergic activation or blocking the propagation of action potentials in cultured neurons rescued altered gene expression and regulatory epigenetic modifications. Our data support the concept of a cellular memory of epileptogenesis and persisting epigenetic modifications of epilepsy target genes, which are able to turn normal into pro-epileptic neurons and circuits.

Understanding seizure development requires an integrated knowledge of different scales of organization of epileptic networks. We developed a model of “epilepsy-in-a-dish” based on dissociated primary neuronal cells from neonatal rat hippocampus. We demonstrate how a single application of glutamate stimulated neurons to generate spontaneous synchronous spiking activity with further progression into spontaneous seizure-like events after a distinct latency period. By computational analysis, we compared the observed neuronal activity in vitro with intracranial electroencephalography (EEG) data recorded from epilepsy patients and identified strong similarities, including a related sequence of events with defined onset, progression, and termination. Next, a link between the neurophysiological changes with network composition and cellular structure down to molecular changes was established. Temporal development of epileptiform network activity correlated with increased neurite outgrowth and altered branching, increased ratio of glutamatergic over GABAergic synapses, and loss of calbindin-positive interneurons, as well as genome-wide alterations in DNA methylation. Differentially methylated genes were engaged in various cellular activities related to cellular structure, intracellular signaling, and regulation of gene expression. Our data provide evidence that a single short-term excess of glutamate is sufficient to induce a cascade of events covering different scales from molecule- to network-level, all of which jointly contribute to seizure development.