Alzheimer’s Protein Links Nerve Cell Death and Cancer

A study in a fruit fly model of Alzheimer’s disease provides an intriguing connection between brain cell death and the abnormal activation of cell division. The lab of Mel Feany, HMS associate professor of pathology, reports in the Feb. 7 Current Biology that the protein tau causes neurons to die by turning on the cell cycle through a pathway previously linked to cancer.

Deadly division. When cells activate the cell cycle abnormally, it usually leads to cell proliferation and cancer. But Mel Feany’s lab found that when the cell cycle is turned on in mature neurons in fruit flies, it can lead to apoptosis and neurodegeneration (diagram). In a fruit fly model of Alzheimer’s disease, the accumulation of tau caused degeneration of the eye (images). The team could rescue the damage by blocking the cell cycle. Raising expression of cell cycle proteins increased the damage.

Normal adult neurons do not divide. But previous studies on the brains of people who died of Alzheimer’s disease found that the affected neurons had activated proteins involved in the cell cycle and cell division. It has not been clear, however, whether these markers were simply a sign of sickness in the cells or whether cell division was a cause of the disease process.

Feany’s team, led by Vikram Khurana, a research fellow in the Harvard Center for Neurodegeneration and Repair, used the simplicity of a fly model to address this question. The flies express human tau, a protein that along with amyloid-beta accumulates in the brain as part of Alzheimer’s disease and other neurodegenerative disorders.

The team found that cell cycle proteins were activated in dying neurons in this model. They could reduce signs of neurodegeneration by genetically blocking steps in the cell cycle or treating flies with chemicals that did the same. Conversely, boosting the expression of cell cycle proteins enhanced cell death. The researchers found that cell cycle activation plays a role in neurodegeneration in both the brain and the retina of flies and in flies expressing either the normal tau protein or a mutant version found in human familial dementia.

The study showed that tau activates the cell cycle through the protein kinase TOR (target of rapamycin), which is part of a widely studied pathway that promotes cell growth and has been implicated in cancer and other diseases. Activating the cell cycle through the TOR pathway leads adult neurons to undergo apoptosis. In fact, the team could induce cell death simply by expressing cell cycle activators in fly brains, even without tau. TOR and cell cycle activation, however, may not be a common feature of all neurodegenerative diseases; they were not found to play a role in two other fly models of neurodegeneration that do not involve tau.

If neurons die in Alzheimer’s and other tau-related diseases by entering the cell cycle, Feany explained, “both TOR and cell cycle proteins are fantastic therapeutic targets” for researchers to investigate, especially since agents targeting the TOR pathway have already been developed for cancer. For Khurana, the most exciting aspect of the finding is the link between cancer and neurodegeneration, “the understanding that these two disease groups that are very common in the aging population are related.”

—Courtney Humphries

Stem Cells and Immune Memory Cells Share Self-renewal Program

Stem cells are not the only cells that have the ability to renew themselves in perpetuity: memory B and T cells are fully differentiated cells that have regained the ability to self-renew throughout life. In doing so, they help the body retain a signature of past infections so it can more rapidly respond if it meets the same pathogen again. It has been speculated that the self-renewal of memory cells and hematopoietic stem cells (HSCs) in the bone marrow might involve a common cellular program, but so far little direct proof has supported this notion.

A study in the Feb. 21 Proceedings of the National Academy of Sciences offers data from DNA chips that support the idea that self-renewal works through a few common pathways. The lab of Diane Mathis and Christophe Benoist, HMS professors of medicine at Joslin Diabetes Center, teamed up with the lab of Irving Weissman, professor of pathology and developmental biology at Stanford University, to pinpoint a set of genes whose expression overlaps in these self-renewing cells. Further research will need to investigate whether the correlations between the groups bear out.

The team collected gene expression data from different cell populations generated by several labs. They first winnowed down the data from each population of cells to only those properties that make self-renewing cells unique. For both B and T cells, they isolated the subsets of gene transcripts that were enriched only in memory cells. For stem cells, they used only those genes with higher expression in long-term HSCs, which are capable of lifelong self-renewal, versus short-term HSCs that have a much more limited capacity.

John Luckey, clinical fellow in pathology at Brigham and Women’s Hospital and one of the study’s first authors, said, “There’s a subset of genes that seems to be conserved among all three populations.” But while nearly all of the genes highly expressed in both B and T cells were also found in HSCs, far more genes were shared between HSCs and only one of the memory cell populations. Luckey speculates that stem cells may contain several redundant pathways responsible for self-renewal, and memory B and T cells turn on different pathways to regain the capacity.

—Courtney Humphries

Molecule Shapes Up Cell Organelle

How do organelles in the cell get their shape? The lab of Tom Rapoport, HMS professor of cell biology, began looking at this question in the endoplasmic reticulum, part of which forms a network of tight tubules. “Here we have a lipid bilayer that is highly curved,” Rapoport said, a structure that is energetically very costly. “How do you maintain this high curvature?”

His team, led by postdoc Gia Voeltz, developed an in vitro system for creating a tubule network out of membranes from Xenopus eggs. Using this system, they could test molecules that would inhibit tubule formation. A particular class of compounds that modify cysteines was able to keep tubules from forming. It turned out that the molecules were targeting a member of the reticulon family, Rtn4a. Reticulons are abundant in all eukaryotic cells and had previously been associated with the endoplasmic reticulum. Rapoport’s team showed that Rtn4a is found exclusively in tubular ER, a unique feature for a protein.

Overexpressing Rtn4a caused more tubules to form. When the team knocked out the reticulon proteins in yeast, however, ER tubules still formed unless the cell was stressed, suggesting that reticulons are not the only player at work. The team then identified the missing component as a protein called DP1. When they removed the reticulons and the yeast homologue of DP1, the ER formed sheets instead of tubules.

“We think we have identified at least the major components required to shape the tubular ER,” Rapoport said. How the proteins function is still unknown, but Voeltz and Rapoport speculate they may work together to form a structural lattice that shapes the membrane. And their structure suggests they may induce curvature by forming wedges in the outer lipid layer, giving it a larger surface than the inner layer.

The study, which appears in the Feb. 10 Cell, adds a new page to the story of Rtn4a, which also goes by the name Nogo-A. Several years ago, Nogo-A was identified as a potential inhibitor of regrowth in axons of the central nervous system, after an antibody to Nogo-A was found to promote the growth of neurites in culture. The problem, Rapoport said, was “an antibody is supposed to bind on the cell surface. But Nogo-A is actually inside the cell” and would not be exposed on the outer surface even if it sat in the membrane. Animal knockouts of Nogo-A have offered inconclusive answers about its role in regeneration. Rapoport believes that the protein probably does not play a role at the cell membrane. His team, however, has seen that overexpressing Rtn4a/Nogo-A in cultured cells can cause long outgrowths to form, and it may be that Nogo-A has a still uncharacterized role in axon formation.

—Courtney Humphries

top