When is a kinase not a kinase? When it’s cyclin-dependent kinase 5 (Cdk5), perhaps. In neurons, Cdk5 puts the brakes on cell-cycle reentry, a poorly understood but potentially important facet of Alzheimer disease (AD). But as reported in the April 14 Journal of Neuroscience, Cdk5 kinase activity may have nothing to do with its cell cycle suppression. Led by Karl Herrup, researchers at Rutgers, The State University of New Jersey, Piscataway, report that the kinase blocks the cycle not through phosphorylation. Instead, it sequesters a nuclear transcription factor, simply by mass action. If confirmed, the findings reveal a novel, non-catalytic, neuroprotective function for Cdk5 that might be relevant to preventing neurodegeneration. “We think this has direct bearing on Alzheimer’s since we previously found that in AD patients, there is a loss of nuclear Cdk5,” first author Jie Zhang told ARF.
In a 65-year-old person’s brain, most neurons last entered the cell cycle as many years prior. Because neurons are terminally differentiated and lack the capacity to undergo a full cell cycle, any attempt they make to start the process is an exercise in futility, if not suicide. Autopsy findings suggest cells may head down this slippery slope in AD. Herrup’s lab previously reported increased cell-cycle events, including aberrant production of cyclins, in neurons from patients with either mild cognitive impairment or AD, compared to neurons from normal controls (see Yang et al., 2003). Work from Herrup’s and other labs linked this aberrant cyclin activation to increased DNA synthesis (see Mosch et al., 2007), while bi-nucleated neurons in AD brain tissue also suggest failed attempts at cell cycling (see Zhu et al., 2008).
These autopsy observations are supported by work in animal models, which suggest that some forms of tau protein precipitate cell cycle machinations (see Khurana et al., 2006 and related ARF Live Discussion). Since Cdk5 phosphorylates tau, and p25, a Cdk5 co-activator, induces neurodegeneration and is elevated in the brain in AD patients (see ARF related news story), one might think that Cdk5 instigates cell cycle reentry. Au contraire, the kinase, in fact, suppresses the process, which runs amok in neurons in Cdk5-negative animals.
To get a better handle on the role of Cdk5 in cell cycle suppression, Zhang and colleagues tested various green fluorescent protein (GFP)-tagged Cdk5 constructs for their ability to suppress cell-cycling in N2a neuroblastoma cells. Using newly made DNA (BrdU incorporation) as a marker, the authors found that Cdk5 prevented cell cycling only when it was expressed in the nucleus and only when it could bind to its co-activator p35; a mutant that lacks p35 binding failed to prevent BrdU incorporation. Surprisingly, a “kinase dead” mutant of the protein behaved just like wild-type, preventing BrdU uptake when it was expressed in the nucleus. The authors concluded that for cell cycle suppression, p35 binding but not kinase activity was essential. By electroporating Cdk5 constructs into neural progenitor cells in mouse embryos in utero, they found the same held true in vivo.
How does Cdk5, a bona fide kinase, pull off cell cycle suppression without an active kinase domain? For a possible answer, the researchers looked to the transcription factor E2F1, which, with its cofactor DP1, is a potent initiator of the cell cycle. Using a combination of immunoprecipitation experiments, Cdk5 and E2F1 knockout mice, and electrophoretic mobility shift assays, Zhang and colleagues found that the proteins bind and stabilize each other, and that Cdk5 and p35, together, prevent E2F1 from binding to regulatory elements on DNA. Delving deeper, they discovered that E2F1, DP1, and p35 form a tertiary complex in the cytosol, but in the nucleus Cdk5 replaces DP1 in this troika, emasculating the transcription factor. The findings fit in with previous work from the Herrup lab pointing to loss of nuclear Cdk5 as a key event in neuronal cell cycle reentry (see Zhang et al., 2008).
“Since the effect they see on the cell cycle does not require kinase activity, this is quite a totally different paradigm for Cdk5 function in the cell,” suggested Bruce Lamb, The Lerner Research Institute, Cleveland, Ohio. Lamb has collaborated with Herrup but was not involved in this study. It is not clear how this nuclear, non-kinase function of Cdk5 dovetails with its traditional kinase role. The findings do not rule out the latter, noted Lamb. Interestingly, however, Zhang and colleagues found that non-p35 co-activators of Cdk5, including p39 and the calpain-cleavage products of p35—p25 and p10—do not complex with Cdk5 and E2F1. This raises the possibility that p25, especially when it is elevated in conditions such as AD, may compete with p35 and E2F1 for the kinase. “That’s an attractive hypothesis,” said Lamb. “It suggests p25 may have a dual function, activating Cdk5 in the cytoplasm on the one hand, and then moving it from the nucleus where it is needed to prevent cell-cycle re-entry on the other.” That would deliver a double blow to neurons, making them vulnerable to cell-cycling and the ramifications of overactive Cdk5 kinase, including tau phosphorylation.
Herrup declined to comment on this hypothesis, but in an e-mail to ARF he noted that the story around Cdk5 kinase activity is probably very complex, with evidence for it being both neurotoxic (by destabilizing anti-apoptotic factors; see Gong et al., 2003) and neuroprotective. One neuroprotective aspect was revealed earlier this year by his own lab. In neurons exposed to amyloid-β, Cdk5 translocates from the nucleus to the cytoplasm, where it retards apoptosis (see Zhang et al., 2010).
“I think we will have to see how this story develops,” said Lamb. He suggested that it might be worth rethinking the biochemistry of p25 and re-examining postmortem tissue to distinguish cytoplasm and nuclear events. On the p25 aspect, Zhang thinks it might not be all about overactivating Cdk5. “We are working on this and have found interactions between p25 and other proteins,” he told ARF.
In the meantime, because of its neurotoxic effects, there is interest in developing Cdk5 inhibitors for treating neurodegenerative disorders (see ARF related news story). Zhang suggested that the right balance of Cdk5 might be more important. The Herrup lab has adopted a slightly different strategy, screening for small molecules that can retard translocation of the kinase to the cytosol, thereby keeping the brakes on the cell cycle.—Tom Fagan
- Cell Cycle Hypothesis Pedaling into Mainstream Acceptance? Results in Fly, Mouse Models Warrant a Second Look
- Enzyme Essential to Brain Development Found to Hyperphosphorylate Tau, Kill Neurons
- Picking Cdk5’s Pocket for New Kinase Inhibitors
- Yang Y, Mufson EJ, Herrup K. Neuronal cell death is preceded by cell cycle events at all stages of Alzheimer's disease. J Neurosci. 2003 Apr 1;23(7):2557-63. PubMed.
- Mosch B, Morawski M, Mittag A, Lenz D, Tarnok A, Arendt T. Aneuploidy and DNA replication in the normal human brain and Alzheimer's disease. J Neurosci. 2007 Jun 27;27(26):6859-67. PubMed.
- Zhu X, Siedlak SL, Wang Y, Perry G, Castellani RJ, Cohen ML, Smith MA. Neuronal binucleation in Alzheimer disease hippocampus. Neuropathol Appl Neurobiol. 2008 Aug;34(4):457-65. PubMed.
- Khurana V, Lu Y, Steinhilb ML, Oldham S, Shulman JM, Feany MB. TOR-mediated cell-cycle activation causes neurodegeneration in a Drosophila tauopathy model. Curr Biol. 2006 Feb 7;16(3):230-41. PubMed.
- Zhang J, Cicero SA, Wang L, Romito-Digiacomo RR, Yang Y, Herrup K. Nuclear localization of Cdk5 is a key determinant in the postmitotic state of neurons. Proc Natl Acad Sci U S A. 2008 Jun 24;105(25):8772-7. PubMed.
- Gong X, Tang X, Wiedmann M, Wang X, Peng J, Zheng D, Blair LA, Marshall J, Mao Z. Cdk5-mediated inhibition of the protective effects of transcription factor MEF2 in neurotoxicity-induced apoptosis. Neuron. 2003 Apr 10;38(1):33-46. PubMed.
- Zhang J, Li H, Herrup K. Cdk5 nuclear localization is p27-dependent in nerve cells: implications for cell cycle suppression and caspase-3 activation. J Biol Chem. 2010 Apr 30;285(18):14052-61. PubMed.
- Zhang J, Li H, Yabut O, Fitzpatrick H, D'Arcangelo G, Herrup K. Cdk5 suppresses the neuronal cell cycle by disrupting the E2F1-DP1 complex. J Neurosci. 2010 Apr 14;30(15):5219-28. PubMed.