Why is Alzheimer's bad for neurons? One answer might be cell cycle re-entry

Alzheimer's research is an ongoing field. Although we know a lot more than we used to, we still don't entirely understand why the accumulation of proteins in Alzheimer's disease kills neurons or renders them non-functional. One intriguing part of the explanation may be offered by Varvel et al. who show that the active proteins in Alzheimer's disease (more on this in a second) cause neurons to re-enter the cell cycle in a mouse model of the disease.

Introduction

I have variously described Alzheimer's -- and other neurodegenerative diseases -- as a disease of molecular crud. The pathological hallmark of the disease is the accumulation of proteinaceous plaques around dying neurons in certain regions of the brain composed mostly of a peptide called A-Beta. For years we have known that A-Beta was important for Alzheimer's disease, but we didn't know exactly whether it was the crud aggregates that killed neurons.

More recently, research into Alzheimer's has focused on the activity of A-Beta oligomers. It turns out that A-Beta that is in large aggregates that we can see is not particularly toxic, nor is A-Beta floating around in its singular form. Instead, it is when many A-Beta molecules get together to form an oligomer that it gets its toxicity. (More about the production of A-Beta here.) The significant issue here is that A-Beta oligomer production often precedes the appearance of large visible aggregates, and it is a much better predictor of decline in cognitive capacity than those aggregates. (The evidence for A-Beta oligomers as the factors causing neurodegeneration is summarized here.)

The question then becomes: how do A-Beta oligomers kill neurons? Many answers have been forwarded, mostly focusing on disarranging the signaling between neurons and creating excitotoxicity. Varvel et al. provide exciting evidence that one of the ways A-Beta oligomers is bad for neurons is by causing them to re-enter the cell cycle.

For a long-time we thought that there were no new neurons in the brain once you reach adulthood. This is actually not true. There are not many new neurons, but we do have evidence for ongoing neurogenesis in certain parts of the brain -- although its significance in normal function and disease is still under dispute. On the other hand, already differentiated neurons are pretty quiescent when it comes to cell growth. These "adult" neurons are not supposed to divide; they are only supposed to sit there and do the business of signaling, thinking, etc. On the whole, they are quite good at this business of not dividing, as is evidenced by the fact that neuronally-derived cancers are exceptionally rare.

So any evidence that we might have to something caused the neurons to re-enter the cell cycle and try to divide is bad news. Adult neurons are probably incapable of doing so, and even if they were it would definitely impair their function.

This is where Varvel et al. come in. They present evidence in vitro and from a mouse model of Alzheimer's that neurons subjected to A-Beta oligomers are attempting to re-enter the cell cycle. If this is happening, it might explain why the neurons are dying or not working properly.

Experimental methods and Results

Varvel et al. use a mouse model of Alzheimer's disease to show that some neurons in older mice are starting to express cell cycle proteins called cyclins. Mice do not naturally get Alzheimer's disease, but you can create a mouse model of the disease by genetically modifying the mouse so that it overexpresses the A-Beta precursor protein (APP). Over-expression of APP causes more expression of A-Beta which means more A-Beta oligomers. There are many of these mouse models; the particular one that they use is called R1.40.

Cyclins are proteins that are involved in facilitating the cell cycle. Cyclins are active at different points in the cycle to activate processes like DNA replication and the production of proteins for cell division, all in a defined temporal sequence. (More on the cell cycle here.)

The key point is that adult neurons should never be expressing cyclins. The presence of cyclins in any mature neurons would suggest that they are trying to re-enter the cell cycle. Bad news.

Varvel et al. stained parts of the R1.40 mice brains to show that these transgenic mice have neurons that express cyclins. Moreover, cyclin expression in these mice precedes the creation of visible A-Beta plaques (the hallmark of any good mouse model is that these appear eventually) and follows a similar the sequence of neuron death in this model. (In Alzheimer's disease, brain regions are not affected at the same time. Some are affected earlier than others.)

Here is an example of their data for a protein called Cyclin A (Figure 1 from the paper).

This figure is dense, so let's take it step by step.

In this figure, the second column (B,E, and H) is staining for a protein called NeuN (in red). NeuN is a marker for mature neurons; the point of staining for it is to confirm that the cells you are looking at are actually neurons. The first column (A,D, and G) is staining for Cyclin A (in green). The first row is an R1.40 mouse at 4 months of age. Note how there is no staining for Cyclin A. In contrast, the second row is an R1.40 mouse that is 6 months of age. See how the green Cyclin A appears. The third row is a normal mouse that is also 6 months of age. See how there is little or no Cyclin A. This indicates that it is not the age of the mouse but rather the genetic alteration that has caused the Cyclin A production. Neurons in these transgenic mice are expressing cyclins at 6 months of age.

The third column (C,F, and I) is a merger of the image in the first two columns. (It also adds a blue color that is staining for all nuclei.) Wherever you see whitish it is because the areas of green and red overlap -- indicating that cyclin A is being expressed in neurons in the transgenic mice aged 6 months. The insets into the third column use a technique called FISH to identify DNA synthesis. The green spots on the inset are places where DNA is being synthesized. DNA is being synthesized in all the animals, but only in the 6 month-old R1.40 mice is it being synthesized in neurons.

The authors perform this same experiment for cyclins D (although the rats in these experiments were older -- 12 months).

These experiments provide evidence that the transgenic animals have neurons that are trying to re-enter the cell cycle, and that this aberrant behavior presents with age.

Another key question is: can we attribute this to A-Beta oligomers -- or is it some other effect in the genetically-modified mice? The authors provide two pieces of evidence that it is A-Beta oligomers.

First, they show that in the R1.40 mice additionally modified to lack a protein called BACE, there are no cyclins and no cell cycle re-entry. BACE is a protein that is necessary for the production of A-Beta from its precursor APP. No BACE means no A-Beta. So mechanistically the cell cycle re-entry looks like it is based on A-Beta.

Additionally, they show that A-Beta oligomers but not monomers given to neurons in vitro causes an increase in DNA synthesis. This is quantified below (from Figure 6):

These figures show a dose response curve for A-Beta oligomers and monomers for BrdU incorporation. BrdU is a compound that is incorporated as new DNA is synthesized. An increase in incorporation strongly suggest that a cell is replicating its DNA to eventually divide. This in vitro evidence suggests that A-Beta acts on neurons to trigger cell-cycle re-entry.

Significance and Caveats

This papers is significant because it shows one way that A-Beta oligomers disorder the normal function of neurons in a mouse model of Alzheimer's. Mature neurons should not be re-entering the cell cycle, and that they are suggests that they will be impaired in their normal functions. Further, the appearance of cell cycle proteins followed by neurons trying to re-enter the cell cycle preceded the appearance of visible A-Beta aggregates by 6-8 months -- providing additional evidence that it is A-Beta oligomers and not the aggregates that are actually toxic.

This is a brilliant observation that recapitulates a clinical finding in Alzheimer's brains. Neurons attempting to re-enter the cell cycle have been observed in the brains of patients with Alzheimer's. The use of this mouse model to observe adverse cell cycle event in these neurons may allow us to understand why they happen and try to develop ways to prevent them -- in essence, to roll back the clock.

That being said, I want to add to caveats to this paper -- that is besides the normal caveat that mice are not humans. First, the authors provide an interesting correlation with neuron death that the neurons are re-entering the cell cycle. But the mechanism of why this is bad for the neurons is not clear. This is not a defect in the paper itself. We just don't know whether cell cycle re-entry is bad for neurons. I think that one could reasonably assume that it is, but you have to check that.

Second, there are many other mechanisms for A-Beta oligomers to be bad for neurons including binding to NMDA receptors and inducing oxidative stress, disordering insulin signaling, and calcium dysregulation. I do not intend to suggest that cell cycle re-entry is the only way that A-Beta hurts neurons. It may not even be an important way.

This is certainly a very interesting theory, though, and certainly merits further study.

Hat-tip: Faculty of 1000
N. H. Varvel, K. Bhaskar, A. R. Patil, S. W. Pimplikar, K. Herrup, B. T. Lamb (2008). A Oligomers Induce Neuronal Cell Cycle Events in Alzheimer's Disease Journal of Neuroscience, 28 (43), 10786-10793 DOI: 10.1523/JNEUROSCI.2441-08.2008