16 November 2007. Could the failure of cell cycle regulation be an underlying cause of neurodegenerative disorders, including Alzheimer disease? High school biology teaches that the cell cycle, in which one mature cell divides into two daughter cells, is essential for the replication of eukaryotic cells. So how might cell cycle proteins influence neurons? After all, these long-lived cells are considered to be amitotic. On Saturday, 3 November, scientists convened at the 37th annual meeting of the Society for Neuroscience in San Diego, California, to address the diverse roles of cell cycle molecules and to discuss their impact on postmitotic neurons in a symposium dedicated to this topic. Beginning in the mid-1990s, an influx of research highlighted cell cycle reactivation in the pathogenesis of Alzheimer disease (AD), suggesting that the atypical re-entry of neurons into the cell cycle contributes to the neuronal loss seen early in its course (see ARF related news story and ARF news story). Until recently, many of the details of this process have remained a mystery.
Karl Herrup of Rutgers University, Piscataway, New Jersey, co-chaired the symposium with Lloyd Greene (see below). Herrup discussed the role of cyclin-dependent kinase 5 (Cdk5) in both normal and stressed neurons. Cdk5 has previously been implicated in the development of AD pathology at numerous levels. This kinase acts to phosphorylate the microtubule-associated protein tau, whose hyperphosphorylation results in the accumulation of neurofibrillary tangles (see ARF related news story). However, Cdk5 is generally portrayed as having no role in cell cycle regulation. Instead, the regulation of Cdk5 is reported to control both neuronal outgrowth and development. In an effort to better understand these mechanisms, Herrup examined the brains of Cdk5 knockout mice. The Cdk5 knockout mice demonstrated the unique characteristics of neurons out of position, neurons lacking differentiation markers, and neurons displaying embryonic markers, such as Nestin. Unexpectedly, in regions that should otherwise be postmitotic, such as the embryonic cortex, cells re-engaged in the cell cycle and then ultimately went on to die. Cdk5 appears to predominantly interact with cell cycle proteins in the nucleus. However, under conditions of stress, Cdk5 emigrates from the nucleus to the cytoplasm, acting as a signal that the cell should move along the pathway to division.
In a similar vein, Ruth Slack, University of Ottawa, Ontario, discussed new roles for the retinoblastoma (Rb) family of proteins in both neuronal differentiation and migration. Though the Rb protein (pRb) is known to be a key regulator of the cell cycle (acting to control entry into S phase), Slack wanted to determine whether pRb possesses an alternate function. The absence of pRb from the telencephalic region of the brain results in impaired development of nerve cells, in addition to abnormal cortical development. This suggests that, like Cdk5, pRb may play a role in the regulation of differentiation. Recent studies have demonstrated that the Rb/E2F pathway can act to coordinate novel functions beyond proliferation (Höglinger et al., 2007; McClellan and Slack, 2006). To examine this further, Slack compared the roles of p107 and pRb, proteins present in neuronal precursor cells and postmitotic neurons, respectively. Using BrdU incorporation as a cell cycle marker, Slack observed that p107 regulates both self-renewal of neural precursors and the rate of commitment to a neuronal fate. Once the precursor cell commits to a neuronal lineage, p107 exits, making way for the appearance of pRb. The interaction of E2F2/3 with pRb acts to regulate migration, demonstrating a unique mechanism by which Rb proteins regulate this process.
Azad Bonni, Harvard Medical School, provided a detailed update on the role of the Cdh1-anaphase promoting complex (APC) in the cell-intrinsic control of axonal growth and patterning. Cdh1-APC is a ubiquitin ligase that acts to ensure the correct progression of the cell cycle. The integration of developing neurons involves an ordered series of events, including polarization, growth, differentiation, and eventually death. Neuronal connectivity is predominantly regulated by extrinsic cues; however, recent evidence suggests that cell-intrinsic mechanisms also exist and can control distinct aspects of axonal, dendritic, and synaptic development. The effects of Cdh1-APC range from regulation of axon growth and patterning to synapse development and neuronal survival. This led Bonni to ask two questions: what are the downstream substrates of Cdh1-APC, and how is this process regulated? SnoN, a transcriptional regulator, is a substrate of neuronal Cdh1-APC, and studies in knockdown models demonstrate that SnoN promotes both axonal growth and serves a key role in TGF-β signaling in proliferating cells. TGF-β is well known to AD researchers as being an important growth factor that is increased in AD brain (Flanders et al., 1995). Additionally, the overproduction of TGF-β by astrocytes is neuroprotective, and TGF-β1 knockouts exhibit increased neuronal death (see ARF related news story). Bonni and colleagues took this a step further, demonstrating that TGF-β inhibits axonal growth by regulating the APC pathway to promote the ubiquitination and degradation of SnoN.
Tying everything together, Lloyd Greene, Columbia University, New York, discussed how cell cycle molecules regulate survival and death of postmitotic neurons in development and disease. Upon apoptotic stimulation, cyclin-dependent kinases become activated in neurons. This event triggers the sequential phosphorylation of Rb family members, and de-repression of transcription factors, such as Myb, which acts through the cooperative activation of the JNK/cJun and FOXO pathways, and the proapoptotic molecule BIM. Notably, BIM has been found to be elevated in the AD brain, in addition to pRb and other cell cycle molecules.
The cell cycle is activated in various disorders involving neuron death, and many researchers believe it may ultimately evolve into an avenue for therapeutic development. Though related on many levels, these disease states (e.g., Parkinson disease, ALS, stroke, spinal cord injury, etc.) are also very diverse. It is still an open question whether the pathways discussed in this symposium specifically tie into AD, or are of a more general nature. Interestingly, in a series of talks on Monday afternoon (slide sessions 444.4 to 6), Robert Vassar’s lab, Northwestern University, Chicago, outlined a link between Cdk5 and the upregulation of BACE1 in AD. Since increased BACE1 may be a major factor in the development of sporadic AD, there may be a relationship between both BACE1 and tau hyperphosphorylation at different levels in the pathogenic process. In his closing remarks, Karl Herrup reminded the attendees that, “as neuroscientists, we need to move beyond the description of the cell cycle that has been propagated by those in the cancer field, because the regulation of the cell cycle in the neuron is much more nuanced than in a cell simply growing in a dish.” Indeed, this symposium emphasized the atypical role of the cell cycle in neurons, revealing further evidence that all neurons may not leave mitosis for good. It is the examination of these subtleties within neurodegenerative pathways that will allow research to determine if what goes around actually comes around.—Rachel R. Ahmed.
Rachel Ahmed is a Ph.D. student at the University of Kentucky.