24 June 2013. How does Aβ poison synapses? According to a paper in the June 17 Proceedings of the National Academy of Sciences, oligomers of the peptide may harm these structures indirectly by releasing a flood of glutamate from astrocytes. This excess glutamate triggers extrasynaptic NMDA-type glutamate receptors on neurons, leading to toxic downstream signaling and synapse loss. Researchers led by Stuart Lipton and Juan Piña-Crespo at the Sanford-Burnham Medical Research Institute and Stephen Heinemann at the Salk Institute, La Jolla, California, report that the FDA-approved Alzheimer’s drug memantine blocks extrasynaptic glutamate signaling and partially protects synapses in a transgenic mouse model of Alzheimer's disease. Even better, NitroMemantine, a more potent version of the drug that is currently in development, maintained synapses at wild-type levels, the authors report. “To my knowledge, this is the first full protection of synapses,” Lipton said. Lipton is an author on worldwide memantine patents.
“What is so interesting about this paper is that it provides new mechanistic insight into how the amyloid protein leads to synaptic loss,” said Hilmar Bading at the University of Heidelberg, Germany. He was not involved in the work.
Prior work has shown that extrasynaptic NMDA receptors play the villain in the brain, antagonizing the beneficial effects of their synaptic counterparts. Whereas activation of synaptic receptors promotes memory formation and neuron survival, the extrasynaptic variety triggers synapse loss and cell death (see ARF related news story). Lipton and colleagues previously showed that memantine selectively blocks extrasynaptic glutamate signaling while mostly sparing synaptic glutamatergic transmission (see ARF related news story). Meanwhile, research implicated excess glutamate in mediating the toxic effects of Aβ on extrasynaptic receptors (see ARF related news story; ARF news story), but the origin of the glutamate was unknown.
To investigate this, the authors used a Fluorescence Resonance Energy Transfer (FRET)-based glutamate sensor developed by Roger Tsien at the University of California, San Diego. In this system, reporter molecules fluoresce when they bind glutamate (see Hires et al., 2008). This probe offers unprecedented temporal and spatial resolution, Lipton said. Co-first authors Maria Talantova, Sara Sanz-Blasco, Xiaofei Zhang, and Peng Xia, all at Sanford-Burnham, plated astrocytes on top of cells containing the sensor probe. Sensor cells also expressed the cell adhesion molecule neuroligin to ensure they made close contact with astrocytes. In some cultures, neurons were present as well. The authors then added Aβ42 to the mixed neuronal and glia rat cerebrocortical cultures or to pure astrocyte cultures. In some experiments they used oligomerized synthetic Aβ42, in others picomolar concentrations of Aβ prepared from human postmortem brain. Both preparations caused astrocytes to spew out glutamate, as visualized by FRET fluorescence at the astrocyte membrane.
Glutamate release required the activation of astrocytic α7 nicotinic acetylcholine receptors (α7nAChR), the authors report. When these receptors were inhibited with α-bungarotoxin, very little of the neurotransmitter spilled out in response to Aβ. Astrocytes from α7nAChR knockout mice also kept glutamate in check. To see if the results held in vivo, the authors performed microdialysis on hAPP J20 transgenic mice. They found that transgenics had elevated glutamate in brain interstitial fluid compared to controls. When hAPP mice were crossed with α7 nicotinic knockouts, however, glutamate levels in the offspring matched those in wild-type mice.
Excessive glutamate increased extrasynaptic currents while decreasing synaptic ones. The authors wondered what downstream effects this would have. They examined hippocampal slices from the J20 mice and found that extrasynaptic activation phosphorylated tau and turned on caspase-3, events implicated in the loss of dendritic spines (see, e.g., ARF related news story; ARF news story; ARF news story; and ARF news story). Treating slices with memantine protected only some spines, while NitroMemantine kept them at wild-type levels in the presence of Aβ. Likewise, in triple transgenic mice treated from six to nine months of age, NitroMemantine maintained a higher spine density than did memantine. Mice treated with NitroMemantine, but not memantine, also improved in the novel object recognition test, a measure of hippocampal function. In ongoing work, the authors are doing additional behavioral tests, Lipton said.
NitroMemantine consists of memantine conjugated to nitroglycerin. The modified drug transfers a nitric oxide group to the NMDA receptor and thus attenuates its activity. NitroMemantine selectively binds extrasynaptic receptors over synaptic, Lipton claimed. Panorama Research Inc., a small biotech firm in Sunnyvale, California, owns the compound. The company is studying the pharmacokinetic and toxicological properties of the drug and looking for a pharmaceutical partner to take the drug into clinical trials, Lipton said. It is unclear at what stage of AD the treatment would work best. Memantine blocks excessive glutamate activity better than normal physiological activity and works more effectively at later stages of disease. This raises the possibility that NitroMemantine could treat moderate or severe AD, as long as neurons are still alive, Lipton suggested.
However, Kelly Dineley at the University of Texas Medical Branch, Galveston, noted that J20 mice probably model preclinical disease or mild cognitive impairment (MCI). This implies that the treatment might benefit people with MCI, Dineley said. Intriguingly, Dineley and colleagues previously reported that Aβ42 acts on astrocytic acetylcholine receptors to stimulate glutamate release in hippocampal slices from Tg2576 mice (see ARF related news story). This finding dovetails with Lipton’s cell culture results.
Bading noted that extrasynaptic NMDA receptors also cause damage and cell death in stroke and Huntington’s disease. Under ischemic conditions, glutamate floods the extracellular space, while in HD, NMDA receptors abandon the synapse for extrasynaptic regions (see ARF related news story; ARF news story). “The common endpoint is the extrasynaptic receptor,” Bading said. This suggests that NitroMemantine has the potential to treat these conditions as well, he added. Memantine is currently in a Phase 2 trial for HD.—Madolyn Bowman Rogers.
Talantova M, Sanz-Blasco S, Zhang X, Xia P, Akhtar MW, Okamoto SI, Dziewczapolski G, Nakamura T, Cao G, Pratt AE, Kang YJ, Tu S, Molokanova E, McKercher SR, Hires SA, Sason H, Stouffer DG, Buczynski MW, Solomon JP, Michael S, Powers ET, Kelly JW, Roberts A, Tong G, Fang-Newmeyer T, Parker J, Holland EA, Zhang D, Nakanishi N, Chen HS, Wolosker H, Wang Y, Parsons LH, Ambasudhan R, Masliah E, Heinemann SF, Piña-Crespo JC, Lipton SA. Aβ induces astrocytic glutamate release, extrasynaptic NMDA receptor activation, and synaptic loss. Proc Natl Acad Sci USA. 2013 Jun 17. Abstract