In Part I, we describe a neuron-enriched culture system, and use this model, to show that non-neuronal cells are required for cell cycle-related neurodegeneration mediated by inflammatory challenges. Chronic inflammation associated with activated microglia and reactive astrocytes plays an important role in the pathogenesis of neurodegenerative diseases such as Alzheimer’s. Both in vivo and in vitro studies have demonstrated that inflammatory responses to immune challenges contribute to cell cycle-related neuronal death. In order to investigate the role of glial cells, especially astrocytes in this phenomenon, a method is described that allows the removal of non-neuronal cells in primary cultures while without affecting neuronal growth. We adopted previously reported methods usi...[ Read more ]

In Part I, we describe a neuron-enriched culture system, and use this model, to show that non-neuronal cells are required for cell cycle-related neurodegeneration mediated by inflammatory challenges. Chronic inflammation associated with activated microglia and reactive astrocytes plays an important role in the pathogenesis of neurodegenerative diseases such as Alzheimer’s. Both in vivo and in vitro studies have demonstrated that inflammatory responses to immune challenges contribute to cell cycle-related neuronal death. In order to investigate the role of glial cells, especially astrocytes in this phenomenon, a method is described that allows the removal of non-neuronal cells in primary cultures while without affecting neuronal growth. We adopted previously reported methods using the thymidine analog, 5-Fluoro-2’-deoxyuridine (FdU), and found the minimum treatment that was sufficient to substantially deplete the non-neuronal cells that normally overgrow the neurons in primary cultures. Cell cycle and glial markers confirm the loss of ~99% of all microglia, astrocytes and oligodendrocyte precursor cells (OPCs). With this milder treatment, neither apparent neuronal loss nor any morphological defects are observed at DIV15-21; both pre- and post-synaptic markers are retained in the purified neuron culture. Furthermore, neurons in FdU-treated cultures remained responsive to excitotoxicity induced by glutamate application. The new model system was used to show that the absence of glial cells attenuates neuronal cell cycle events and neuronal cell death, perhaps because NFκB protein and the mRNA of several cytokine receptors are significantly changed.

In Parts II and III, we investigate the roles of the transcription factor E2F1 in cell cycle-related neuronal death. Adult CNS neurons are fully differentiated and permanently post-mitotic. Despite this, neurons are apparently induced to re-enter a cell cycle during various neurodegenerative diseases, including Alzheimer’s disease (AD). The mechanisms underlying the induction of this lethal cell cycle remain elusive. E2F1 is a transcription factor that both promotes cell cycle progression and also regulates apoptosis and the DNA damage responses. The multiple roles of E2F1 make it possible that E2F1 works as a link between cell cycle and cell death in neurons. In fact, elevated expression of E2F1 has been noted previously in the neurons of brain from AD patients and mouse models. In vitro, oligomeric beta-amyloid (AβO) significantly increased the expression level of E2F1 as well as other cell cycle-related proteins in primary cortical neurons. Overexpression of E2F1 robustly drives the initiation of cell cycling in primary neurons, yet E2F1-overexpressing neurons do not replicate their DNA (assessed by BrdU incorporation). Instead they die a relatively short time after E2F1 overexpression. This unexpected observation could be explained by a DNA damage response and G1/S checkpoint arrest. We also showed that DNA binding activity, but not its transactivation activity, is required for E2F1 to trigger cell cycle initiation and cell death in neurons.

We confirmed that DNA damage triggered cell cycle-related neuronal death is associated with aberrant E2F1 expression, but also showed that the cell cycle activity is restricted in early cell cycle stages. No DNA replication was observed, ruling out the possibility that E2F1 acts to drive G1/S transition in this model. This led us to hypothesize that cell cycle activation occurs as a response to DNA damage. To further dissect the functions of E2F1 in neurons, we performed ChIP-seq with E2F1 antibodies from purified primary neuronal cultures and confirmed its major roles in regulating cell cycle-related genes. Nevertheless, upon DNA damage, E2F1 tends to bind to the promoters of genes associated with DNA repair, suggesting a novel role of E2F1 in regulating the DNA damage response. Taken together, we report a canonical function of E2F1 in promoting cell cycle and apoptosis in post-mitotic neurons and a novel role in regulating DNA damage response dependent on cellular environment.