This pathway has already drawn the interest of AD researchers. Autophagy enables cells to recycle molecules and dispose of cellular garbage in vesicles known as autophagosomes. Autophagosomes fuse with lysosomes, and their contents are then broken down by enzymes. The disruption of autophagy leads to neurodegeneration in mouse models (see ARF related news story on Hara et al., 2006 and Komatsu et al., 2006) and in cultured neurons (see ARF related news story on Boland et al., 2008), indicating its importance in keeping neurons healthy. AD-causing presenilin-1 mutations have been shown to hamper lysosomal degradation (see ARF related news story on Lee et al., 2010). Manipulation of a crucial autophagic regulatory protein, beclin 1, can promote or hinder Aβ clearance (see ARF related news story on Pickford et al., 2008; and see Jaeger et al., 2010). Likewise, recent work has shown that stimulating autophagy can improve symptoms in AD mice (see ARF related news story).

Most of these studies have examined the roles of proteins in autophagic regulation. Walter and colleagues approached the problem from a different direction, looking instead at the effect of membrane lipids. These molecules drew their attention because lipid levels have been shown to increase during aging and in AD (see Han, 2005; Yamamoto et al., 2008; Ariga et al., 2008). Several genetic disorders known as lysosomal storage disorders (LSDs) are distinguished by abnormal accumulations of membrane lipids, and involve failure of autophagy (see Settembre et al., 2008) and some AD-like features (see Auer and Jacobson, 1995 and Jeyakumar et al., 2003).

To examine the role of sphingolipids, first author Irfan Tamboli developed two cellular models. In the first, Tamboli and colleagues fed neuroblastoma cells an excess of glycosphingolipids; in the second, the authors used primary fibroblasts from patients with various LSDs such as Niemann-Pick’s, Tay-Sachs's, and Sandhoff’s diseases. Processing of amyloid precursor protein was altered in both of these lipid-rich cell models, leading to high levels of C-terminal APP fragments (APP-CTFs) produced by β- and α-secretase cleavage. Significantly, cells with an excess of a different kind of lipid, ceramide, did not amass APP fragments, demonstrating a specific role for sphingolipids. In radiolabeling experiments, Tamboli and colleagues showed that this overabundance of APP-CTFs occurred because the fragments were less efficiently degraded. The surplus APP-CTFs built up in autophagosomes. These vesicles seemed to fuse normally with lysosomes in the lipid-rich cells, but the degradation process stalled. Electron microscope images showed that the resulting autophagolysosomes were filled with dense material, similar to that seen in cells where autophagy is inhibited.

Additionally, Tamboli and colleagues found that the presence of sphingolipids increased the activity of γ-secretase, which clips Aβ from APP-CTFs. Unsurprisingly, the combination of more substrate and more enzyme activity led to higher levels of the Aβ product, both intracellular and secreted, in lipid-rich cells. The negative effects of sphingolipids did not stop there: The molecules also seemed to cause a greater induction of autophagic vesicles, leading to more APP-CTFs stuck in vesicles and available for Aβ generation. Finally, the authors showed that trying to promote autophagy by starving the cultured cells did not overcome the autophagic block.

One of the burning questions is how sphingolipids might produce these effects. Walter suggests two possible routes by which lipids might block the final stage of autophagy. One possibility is that sphingolipids could lessen the activity of vacuolar ATPase, the proton pump in lysosomes that maintains their acidic internal environment. Research on LSDs has provided some evidence in support of this interaction, Walter said. The loss of acidity would shut down or dampen lysosomal enzymes, causing autophagy to grind to a halt. To test this idea, Walter and colleagues plan to measure the activity level of these enzymes in lipid-fed cells. If it is normal, it would lend support instead to the second hypothesis—that the lipids in the vesicle membranes simply get in the way, keeping APP-CTFs away from lysosomal enzymes.

On a side note, vacuolar ATPase was also featured in another paper in this week’s Journal of Neuroscience. Researchers in Germany and Belgium report that several inhibitors of vacuolar ATPase boost progranulin levels by increasing the pH of acidic compartments, suggesting this might be a strategy to explore for treating forms of frontotemporal dementia marked by progranulin deficiency (see ARF related news story on Capell et al., 2011).

An intriguing implication of these data is that excess membrane lipids might contribute to the development of AD. Walter points out that in Niemann-Pick's disease, a fatal cholesterol storage disorder, the brains of patients develop tau tangles similar to those seen in Alzheimer’s. If excess lipids can trigger both tau and Aβ aggregation, it is possible that an age-related accumulation of membrane lipids could set off the initial events of AD, Walter speculated. He would like to test this idea using mouse models of LSDs. He plans to cross these mice with APP transgenic mice, and see if the defect in lipid degradation increases tau or amyloid pathology.

Ralph Nixon, at the Nathan Kline Institute for Psychiatric Research, Orangeburg, New York, calls this “an intriguing paper that adds strong support for the importance of primary lysosomal dysfunction as pathogenically relevant to AD.” The lysosomal disease field is tremendously interested in the significance of the genetic disorders to AD, Nixon said, with researchers discovering numerous AD pathologies in young LSD brains. The field is also changing its concept of what the lysosome does, Nixon said. Rather than seeing it as simply an endstage garbage disposal, scientists now recognize it as having a complex job, releasing many molecules important for cell health. “This is a dynamic process that is almost certainly dysregulated by Aβ, and possibly other AD-related substances like ApoE,” he said.

This study only sharpens interest in the treatment possibilities of modifying autophagy. “It’s a whole new world of therapeutic targets,” Nixon said, adding that vesicles and their components have largely been ignored as drug targets. “The finding that manipulating autophagy is therapeutic in AD models is encouraging evidence that this is a promising direction for therapeutic research,” added Nixon, though he advised caution. In cells where autophagy is blocked at the lysosome stage, merely inducing more autophagy may do more harm than good, he said. He suggests that the interventions will have to be carefully tailored to the specific pathology.—Madolyn Bowman Rogers.

Reference:
Tamboli IY, Hampel H, Tien NT, Tolksdorf K, Breiden B, Mathews PM, Saftig P, Sandhoff K, Walter J. Sphingolipid storage affects autophagic metabolism of the amyloid precursor protein and promotes Abeta generation. J Neurosci. 2011 Feb 2;31(5):1837-49. Abstract