Led by Paul Lombroso at Yale University, New Haven, Connecticut, researchers report that a protein phosphatase accumulates in mouse models of AD and mediates Amyloid beta (Aβ) toxicity. The buildup of the phosphatase, called striatal-enriched protein tyrosine phosphatase 61—or STEP61—is not due to overproduction, but failure of degradation by the ubiquitin proteasome system (UPS), the researchers show. In a second paper, a group led by Andrea LeBlanc at McGill University, Montreal, Canada, report that a ubiquitin-dependent ATPase that is crucial for the UPS is cleaved by caspase-6, a protease that LeBlanc’s research implicates in AD pathology. One cleavage fragment of the ATPase acts as an inhibitor of the UPS in cells, and the researchers show that this fragment is elevated in AD brains, hinting that it might impair the UPS in the disease as well. “The papers agree with the idea that an efficient working UPS is very relevant for preventing and/or delaying AD,” suggested Fred van Leeuwen, University of Maastricht, in an email to ARF.

Lombroso’s work connects STEP61 with Aβ-induced blockage of long-term potentiation, a form of synaptic plasticity. The phosphatase is localized to the post-synaptic compartment and is thought to act as a counter balance to synaptic strengthening by promoting internalization of glutamate receptor subunits. Previously, Lombroso and colleagues discovered that Aβ activates the phosphatase (see ARF news) and other researchers reported that the protein is elevated in J20 mice expressing mutant human amyloid precursor protein, or APP (see Chin et al., 2005). Now, joint first authors Pradeep Kurup, Yongfang Zhang and colleagues extend that observation to the widely-used Tg2576 mouse model and to the human brain. They found that between three and 12 months of age, STEP61 accumulates in the frontal cortex of the mice, and it seems elevated the AD brain as well. Comparison of postmortem cortical tissue extracts from five AD patients and five controls, revealed elevated levels of the phosphatase in the former. The findings suggest that STEP61 might play a role in AD pathology.

Testing this idea, Kurup and colleagues exposed rat and mouse cortical neurons to conditioned medium (CM) from Aβ-producing 7PA2 cells. The CM, previously shown to contain toxic forms of Aβ that block synaptic plasticity (see Walsh et al., 2002 and related ARF news), not only decreased the number of NR2B glutamate receptor subunits on the neurons, but increased the cellular content of STEP61. On the other hand, the conditioned medium had no effect on glutamate receptor subunits when added to neurons from STEP61 knockout mice. Overall, the findings indicate that the phosphatase is required for Aβ-induced internalization of glutamate receptors.

Curiously, neither inhibitors of transcription nor translation prevented Aβ-induced elevation of STEP61, which would seem to eliminate increased synthesis as an explanation for Aβ’s effects. “One other possible explanation for elevated STEP61 would be reduced degradation,” said Lombroso. That is, in fact, what the researchers found. By blocking the proteasome, the researchers increased STEP61 in cortical cells. They found that the phosphatase was ubiquitinated when cells were treated with Aβ. And they found increased levels of ubiquitinated STEP61 in the brain of 12-month-old Tg2576 mice compared to wild-type controls. These results point to a failure of the proteasome to clear STEP61 under an Aβ onslaught.

These discoveries would predict that reducing STEP61 levels in AD mouse models will prevent the loss of glutamate receptors and reverse cognitive deficits. Lombroso told ARF that he is investigating this possibility using a genetic approach. His lab is also screening for small molecules that might inhibit STEP61 phosphatase activity.

LeBlanc and colleagues’ study of the UPS was more direct. In a previous screen the scientists identified the valosin-containing ATPase, p97, which helps drive protein ubiquitination, as a potential substrate for caspase-6. Now, first author Dalia Halawani and colleagues report that caspase-6 cleaves both N- (residues 1-480) and C-terminal (residues 481-806) fragments of the ATPase in vitro. The N-terminal half was cleaved into at least four different polypeptides, and the C-terminal half at least two. Caspase-6 also cleaved the native protein, which forms a hexamer that is generally resistant to proteolysis.

To identify the exact sites of cleavage, the researchers separated caspase-6 cleaved fragments using liquid chromatography and analyzed the peptides using mass spectrometry. The size of the protein fragments was consistent with five major cleavage sites, the most consistent being the VAPD motif at residues 176-179. The researchers raised an antibody to the N-terminal fragment of the protein generated by proteolysis at this position and found that it specifically recognized the truncated protein, p97(1-179) but not full length. The antibody (anti-p97D179) stained hippocampal sections taken postmortem from AD patients and some MCI patients, indicating that caspase-6 cleavage of p97 occurs in the disease and possibly starts early in the progression of pathology. Of 15 AD hippocampal samples, anti-p97D179 immunoreactivity was least intense in those taken from patients with the highest MMSE scores and most intense in those with severest disease (MMSE <17).

LeBlanc thinks that since p97 is cleaved in AD, it could explain the high level of ubiquitin in Alzheimer’s neurons. “Some of it is on tau, but there are many other proteins that are ubiquitinated that accumulate in neurons in the Alzheimer’s brain and nobody really knows how it happens,” she said. Halawani found that expressing an ATPase-deficient mutant of p97, or the caspase-6 truncated ATPase, p97(1-179), caused accumulation of UPS substrates in N2a neuroblastoma cells, supporting the idea that dysfunctional p97 or caspase-6 cleavage of the protein jams up the UPS recycling system. Interestingly, anti-p97D179 staining in AD hippocampal tissue was mostly granular, “which is very similar to when you look at ubiquitinated products in Alzheimer’s,” said LeBlanc

What activates caspase-6? “We haven’t nailed that yet,” said LeBlanc. One possibility is through activation of the death receptor 6 by the N-terminal of APP, a pathway recently uncovered by Marc Tessier-Lavigne and colleagues at Genentech Inc., San Francisco, California (see related ARF news). DR6 can activate caspase-6 in axons. LeBlanc thinks caspase-6 activation might be the result of inflammation. “We found that if we block caspase-1, then caspase-6 is blocked as well. Since it has been known for a long time that interleukin 1-β is increased in Alzheimer’s, and caspase-1 is the IL1- β converting enzyme, then caspase-1 may be elevated as well,” she said.—Tom Fagan.

References:
Kurup P, Zhang Y, Xu J, Venkitaramani DV, Haroutunian V, Greengard P, Nairn AC, Lombroso PJ. Aβ-mediated NMDA receptor endocytosis in Alzheimer’s disease involves ubiquitination of the tyrosine phosphatase STEP61. The Journal of Neuroscience 2010 April 28;30:5948-5957. Abstract

Halawani D, Tessier S, Anzellotti D, Bennett DA, Latterich M, LeBlanc AC. Identification of caspase-6-mediated processing of the valosin containing protein (p97) in Alzheimer’s disease: a novel link to dysfunction in ubiquitin proteasome system-mediated protein degradation. The Journal of Neuroscience 2010 April 28;30:6132-6142. Abstract