11 September 2013. Some research suggests that amyloid β oligomers bind cellular prion protein (PrPc) at the cell surface, but how could that wreak havoc within the cell? In the September 4 Neuron, researchers led by Stephen Strittmatter of Yale University School of Medicine, New Haven, Connecticut, place the blame on the metabotropic glutamate receptor 5 (mGluR5). Overexpression of mGluR5 enabled PrPc-bound Aβ oligomers to activate a molecular cascade that shriveled spines in cultured cells, and treatment with an mGluR5 inhibitor rescued cognition and restored synaptic density in two Alzheimer’s disease mouse models. The study “helps develop a coherent story from many previously identified pieces of the puzzle,” noted William Klein of Northwestern University in Evanston, Illinois, in an email to Alzforum. Moreover, a paper in last month's issue of Glia suggests that Aβ’s effects on mGluR5 extend beyond neurons. Antoine Triller of INSERM in Paris, France, and colleagues reported that in astroctyes, too, mGluR5s cluster around Aβ oligomers at the surface, where they seem to cause an outpouring of ATP, which renders nearby neurons hyperactive (Shrivastava et al., 2013 ).
Much research blames Aβ oligomers, as opposed to amyloid plaques, for neurodegeneration and memory loss in AD. Scientists have isolated Aβ oligomers from patient brain (see Shankar et al., 2008 ) and spinal fluid (Klyubin et al., 2008 ), and have shown they disrupt memory when injected into rodents (see Reed et al., 2011 ). In 2009, Strittmatter and colleagues reported that Aβ oligomers bind PrPc (see ARF related news), and later added that the Aβ-PrPc interaction activates the kinase fyn, which shrinks spines and kills neurons by making them hyper-responsive to glutamate signaling (see ARF related news). Other research indicates that fyn drives Aβ toxicity in AD models (see Chin et al., 2005 ) and suggests the kinase may link Aβ with pathological tau hyperphosphorylation (Larson et al., 2012 ). Tau may also help Aβ carry out its dirty work by targeting fyn to the synapse (see ARF related news). Meanwhile, it remains unclear whether PrPc-bound Aβ oligomers are the prime source of toxicity in AD, suggested Marco Prado of Robarts Research Institute in Ontario, Canada. While some studies besides Strittmatter’s seem to show that PrPc is required for Aβ’s harmful effects (see ARF related news), others suggest that PrPc has little bearing on Aβ toxicity (see ARF related news).
Nevertheless, given that PrPc interacts with Aβ oligomers at the cell surface while fyn sits within the neuron, first author Ji Won Um and colleagues looked for a molecule that might couple PrPc with fyn, a src family kinase that becomes active when phosphorylated. They screened 61 likely candidates, all transmembrane synaptic proteins, and found that only one, mGluR5, fit the bill. When transfected into HEK293 cells cultured with synthetic Aβ oligomers—by some standards a model quite removed from Alzheimer’s disease and relying on non-physiological concentrations of Aβ oligomers (1 μM)—mGluR5 activated fyn as assessed by anti-phospho src family kinase immunoblotting. Immunofluorescence and immunoprecipitation experiments with transfected COS-7 cells showed mGluR5 associating with PrPc and fyn but not with Aβ oligomers.
Besides activating fyn, PrPc-bound oligomers also increased intracellular calcium—another consequence of mGluR5 engagement—as determined by increased current through transmembrane chloride channels in Xenopus oocytes. In mouse cortical neurons, AD brain extracts elicited more calcium flux than did extracts from age-matched controls, and the more Aβ bound to mouse PrPc, the bigger the calcium response, the researchers reported. Compounds that block mGluR5 activity prevented the calcium currents, as did the use of mGluR5- or PrPc-deficient neurons. Furthermore, inhibiting mGluR5 curbed the spine loss typically seen when mouse hippocampal neuronal cultures are spiked with Aβ oligomers. In vivo, twice-daily injections of the mGluR5 antagonist MTEP (3-((2-Methyl-4-thiazolyl)ethynyl)pyridine) http://en.wikipedia.org/wiki/MTEP for 10 days erased memory deficits in 8-9-month-old APP/PS1 and 3xTg AD mice. At that age, these animals have rampant amyloid deposits in the brain, and while the authors did not check for changes in amyloid status, Strittmatter noted that the treatment was short and unlikely to have cleared Aβ deposits.
The paper confirms and ties together past studies suggesting that metabotropic glutamate receptors interact with PrPc (see Beraldo et al., 2011) and bind and activate fyn (Heidinger et al., 2002). Prior work had also shown that Aβ oligomers induce clustering of mGluRs (see ARF related news), and that mGluR5 antagonists can block Aβ-induced long-term potentiation, a cellular correlate of learning and memory (see ARF related news). “But it was unclear (from those studies) how this was happening and how mGluR5 fit in,” Strittmatter told Alzforum. “(The current study) defines a specific molecular pathway.”
The findings “point to the idea that mGluR5 is an interesting target for drug development,” Strittmatter said. His lab is trying to develop compounds that preferentially disrupt interactions between mGluR5 and Aβ while preserving physiologic glutamate signaling. In addition, the researchers have launched a Phase 1 trial of a fyn kinase inhibitor (saracatinib, or AZD0530) in people with mild-to moderate AD. AZD0530 is a repurposed drug that was ineffective in Phase 2 cancer trial, Strittmatter said. Several companies have developed mGluR5 antagonists (Roche’s RG7090) or negative allosteric modulators (Novartis’ AFQ056 and Addex Therapeutics’ dipraglurant) for Fragile X syndrome, an inherited form of mental retardation (see ARF related news). Experts were not aware of any mGluR5 compounds being developed for AD at present.—Esther Landhuis
Um JW, Kaufman AC, Kostylev M, Heiss JK, Stagi M, Takahashi H, Kerrisk ME, Vortmeyer A, Wisniewski T, Koleske AJ, Gunther EC, Nygaard HB, Strittmatter SM. Metabotropic Glutamate Receptor 5 is a Coreceptor for Alzheimer Aβ Oligomer Bound to Cellular Prion Protein. Neuron. 4 Sep 2013;79:887-902. Abstract .