Targeting Synaptic Dysfunction in Alzheimer's Disease

Websites

The Alzheimer's Association
The Alzheimer's Association presents statistics, basic facts about the disease, as well as information about grants and schedules for ICAD, the annual meeting on dementia research.

The Alzheimer Research Forum
The Alzheimer Research Forum offers resources for scientists who study the molecular mechanisms of this neurodegenerative disease. In particular, a few journal articles about changes in synaptic function were chosen as "papers of the week," with relevant discussions, such as a 2009 paper about A-β related synaptotoxicity, and a discussion of evidence that oligomers interfere with NMDA receptor function.

Journal Articles

Haroutunian

Haroutunian V, Schnaider-Beeri M, Schmeidler J, et al. Role of the neuropathology of Alzheimer disease in dementia in the oldest-old. Arch. Neurol. 2008;65(9):1211-7.

Katsel P, Tan W, Haroutunian V. Gain in brain immunity in the oldest-old differentiates cognitively normal from demented individuals. PLoS One 2009;4(10):e7642.

Mattson

Calabrese V, Cornelius C, Dinkova-Kostova AT, et al. Cellular stress responses, the hormesis paradigm, and vitagenes: novel targets for therapeutic intervention in neurodegenerative disorders. Antioxid. Redox. Signal 2010;13(11):1763-811.

Kapogiannis D, Mattson MP. Disrupted energy metabolism and neuronal circuit dysfunction in cognitive impairment and Alzheimer's disease. Lancet Neurol. 2011;10(2):187-98. Erratum in: Lancet Neurol. 2011;10(2):115.

Mattson MP. The impact of dietary energy intake on cognitive aging. Front. Aging Neurosci. 2010;2:5.

Stutzmann

Bruno AM, Huang JY, Bennett DA, et al. Altered ryanodine receptor expression in mild cognitive impairment and Alzheimer's disease. Neurobiol. Aging 2011.

Goussakov I, Miller MB, Stutzmann GE. 2010. NMDA-mediated Ca(2+) influx drives aberrant ryanodine receptor activation in dendrites of young Alzheimer's disease mice. J. Neurosci. 2010;30(36):12128-37.

Arancio

Puzzo D, Staniszewski A, Deng SX, et al. Phosphodiesterase 5 inhibition improves synaptic function, memory, and amyloid-beta load in an Alzheimer's disease mouse model. J. Neurosci. 2009;29(25):8075-86.

Hempstead

Hempstead BL. Commentary: Regulating proNGF action: multiple targets for therapeutic intervention. Neurotox. Res. 2009;16(3):255-60.

Teng KK, Felice S, Kim T, Hempstead BL. Understanding proneurotrophin actions: Recent advances and challenges. Dev. Neurobiol. 2010;70(5):350-9. Review.

Di Paolo

Chang-Ileto B, Frere SG, Chan RB, et al. Synaptojanin 1-mediated PI(4,5)P2 hydrolysis is modulated by membrane curvature and facilitates membrane fission. Dev. Cell 2011;20(2):206-18.

Di Paolo G, Kim TW. Linking lipids to Alzheimer's disease: cholesterol and beyond. Nat. Rev. Neurosci. 2011;12(5):284-96. Review.

Strittmatter

Gimbel DA, Nygaard HB, Coffey EE, et al. Memory impairment in transgenic Alzheimer mice requires cellular prion protein. J. Neurosci. 2010;30(18):6367-74.

Gunther EC, Strittmatter SM. β-amyloid oligomers and cellular prion protein in Alzheimer's disease. J. Mol. Med. 2010;88(4):331-8. Review.

Klann

Ma T, Hoeffer CA, Wong H, Massaad CA, et al. Amyloid β-induced impairments in hippocampal synaptic plasticity are rescued by decreasing mitochondrial superoxide. J. Neurosci. 2011;31(15):5589-95.

Massaad CA, Washington TM, Pautler RG, Klann E. Overexpression of SOD-2 reduces hippocampal superoxide and prevents memory deficits in a mouse model of Alzheimer's disease. Proc. Natl. Acad. Sci. USA 2009;106(32):13576-81.

Wennogle

He G, Luo W, Li P, et al. Gamma-secretase activating protein is a therapeutic target for Alzheimer's disease. Nature 2010;467(7311):95-8.

Grant Support

This event was funded in part by the Life Technologies™ Foundation.

Speakers:
Vahram Haroutunian, Mount Sinai School of Medicine
Mark P. Mattson, National Institute on Aging

Highlights

• Changes in the synapse in patients with Alzheimer's disease (AD) appear to affect all components of the synaptic machinery.
• Synaptic abnormalities are present in AD patients of all ages, even when other disease hallmarks such as plaques and tangles are less apparent.
• Caloric restriction reduces amyloid-β deposits in an animal model; intermittent fasting also protects against synaptic dysfunction.
• In animal and cell models, exercise promotes mechanisms that repair DNA damage.
• The insulin-sensitizing drug exendin-4 might slow cognitive decline.

The synapse in the oldest old

The synapse is the business end of a neuron, the place where its most important transactions take place. Presynaptically—the term refers to the cell producing a signal—neurotransmitters are packaged into vesicles that are transported to the cell membrane where they dock, fuse with the membrane, release their contents, and are recycled into new vesicles. On the presynaptic side, dysfunctions might involve any or all of these processes.

To examine changes in presynaptic function in brains affected by Alzheimer's disease, Vahram Haroutunian of the Mount Sinai School of Medicine conducted gene expression and messenger RNA (mRNA) analyses in samples of AD brains, focusing on the superior temporal gyrus (STG). Brains with vascular damage, inflammatory changes, or from patients with a life history of depression were excluded. To obtain a measure of synaptic integrity, Haroutunian's group compared mRNA and protein levels of several important synaptic players, including syntaxin-1, SNAP-25, complexin, synaptobrevin, and synaptophysin, in brains from controls and dementia patients. They found an overall 25% decrease in mRNA and protein levels of these gene products. The broad decline suggests the entire synapse was compromised in the disease. Protein and gene expression fall off with disease severity.

Much of the machinery of the synapse is affected in Alzheimer's disease.

Microarray studies of brain samples reveal that dementia takes a different form in the brains of the "oldest old," those over the age of 85. The brains of people over that age who are equally demented as their younger counterparts have fewer plaques, less amyloid-β-40 and amyloid-β-42 (the two most common amyloid-β species) but synaptic pathology is comparable. In "young-old" patients under age 85, synaptic protein levels are the best correlate of dementia and neuropathology; in older patients, synaptic mRNA levels are more reflective of these changes. One possibility is that in these oldest patients, pathology is more localized, so that neurons projecting to the STG are less compromised than in younger patients.

Diet, exercise, and brain cell survival

Cognitive challenges, exercise, and reductions in dietary intake are powerful modulators of neuronal and synaptic vulnerability, perhaps because they activate adaptive stress mechanisms, suggested Mark Mattson of the National Institute of Aging in a review of studies from his group.

Most of the neurons that degenerate in Alzheimer's disease are glutamatergic, meaning that they are excitatory. The activity they induce in a healthy young brain puts mild stress on cells, encouraging the neurons to release neurotrophic factors that promote synaptic plasticity and cell survival. Presynaptic release of glutamate, when combined with growth factors, also promotes synaptogenesis. An excitatory imbalance may develop with aging that promotes amyloid-β aggregation and synaptic dysfunction; exercise and energy intake reduction may counteract these effects.

In experiments manipulating energy intake in a widely-used triple-transgenic AD mouse model that develops amyloid-β pathology, animals on a calorie-restricted diet of 30% normal rations for one year developed lower levels of amyloid-β than normally fed animals or those subjected to alternate-day fasting. However, synapses were preserved in the alternate-day-fasted animals despite their relatively high amyloid-β levels, suggesting that reducing amyloid-β is not required to prevent synaptic dysfunction. In another model, intermittent fasting protected young but not old animals against neurological deficits caused by an artificial stroke, a result that may be related to the more robust upregulation of brain-derived neurotrophic factor (BDNF) and other neurotrophins in younger brains.

Recent work establishes that BNDF upregulates a DNA repair enzyme in neurons in culture, and that exercise upregulates DNA repair mechanisms in vivo. "If you're using neurons, or exercising, you're increasing the neurons' ability to repair DNA damage," said Mattson. In another experiment, both calorie restriction and exercise increased the synapses in normal animals and in an animal model of diabetes, a condition that can accelerate the deterioration of cognitive function and synaptic plasticity. Exendin-4, prescribed as the diabetes drug Byetta for its ability to increase insulin sensitivity, also increases BDNF production and protects against neurodegeneration in AD models. It is being tested in early AD to slow cognitive decline.

Speakers:
Grace E. Stutzmann, Rosalind Franklin University/Chicago Medical School
Ottavio Arancio, Columbia University Medical Center
Barbara L. Hempstead, Weill Cornell Medical College
Gilbert Di Paolo, Columbia University

Highlights

• Calcium signaling is disrupted in early Alzheimer's disease (AD), an effect that appears to be related to ryanodine receptor (RyR) function.
• Chronic treatment with RyR blockers selectively restored normal calcium signaling in AD mice.
• Increasing CREB phosphorylation via phosphodiesterase (PDE) inhibition might offer an effective AD therapeutic strategy.
• A novel PDE5 inhibitor restores normal synaptic function and memory following amyloid-β administration in slices and in vivo.
• Neurotrophins have potent effects on synaptic activity and neuronal survival.
• ProNGF, a neurotrophin precursor, may mediate cell death and dysfunction.

Manipulating calcium signals at the synapse

At the synapse, one of the most important and energy-intensive jobs is the effective regulation of ionic calcium inflow and efflux. This balance is compromised in Alzheimer's disease (AD), with related effects on signaling and plasticity. Disruptions in early calcium signaling—that is to say, calcium release from the endoplasmic reticulum through ryanodine receptor (RyR) and inositol triphosphate (IP3R) channels—were described by Grace Stutzmann of the Rosalind Franklin University and Chicago Medical School. In AD, these changes may contribute to synaptic loss. Evidence from slice preparations of AD mouse models suggests that alterations in calcium balance change neurotransmitter release and deplete vesicles, which could over time contribute to a breakdown in synaptic structure and function. Excitability increases in the postsynaptic neuron, and can be synergistically potentiated with caffeine, which activates RyR.

In AD mouse models (3xTg on this chart), treatment with the RyR-activator caffeine potentiates calcium signaling in dendritic spines (right column, right chart).

Data from experiments using postmortem human brains and from mouse models suggest that the RyR2 isoform is upregulated in the disease state. Manipulating RyR function and calcium signaling has different effects in normal and AD mouse models. Neurotransmission in normal animals does not appear to be affected by RyR blockade but does change in AD model animals, suggesting that this mechanism is engaged differently during baseline activity in normal and AD models.

Blocking RyR prevents long-term potentiation (LTP) and long-term depression (LTD), two contrasting cellular plasticity mechanisms associated with learning and memory. In a commonly used triple-transgenic AD mouse model, RyR blockade enhanced LTD and caused LTP to be converted into LTD. Chronic treatment with RyR blockers such as dantrolene normalized calcium signaling and plasticity in AD-model mice, suggesting that this receptor system might offer a suitable target for AD therapeutics. "We hope that by blocking Ry receptors at early presymptomatic stages, we could generate preventative therapeutic options," said Stutzmann.

The potential of CREB

Interfering directly with amyloid-β production is not the only way to restore synaptic plasticity in AD, and may not be the most effective therapeutic strategy, pointed out Ottavio Arancio of Columbia University Medical Center. He described work on targets downstream, in learning and memory mechanisms, particularly the transcription factor CREB (cAMP-response element binding protein). During induction of synaptic plasticity, CREB is phosphorylated, a process that is reduced by high levels of amyloid-β and tau. Conversely, the induction of two pathways that increase CREB phosphorylation via the inhibition of phosphodiesterase 4 reduce amyloid-β-related memory deficits and restore LTP in a transgenic AD mouse model. Encouragingly, memory rescue is also seen in older mice that had already developed significant neuropathology.

Even in transgenic mice loaded with amyloid-β, phosphodiesterase 4 inhibition by rolipram restored LTP (top line).

This observation suggests a new strategy for developing AD therapeutics; inhibiting one of two phosphodiesterases, PDE4 and PDE5, in order to enhance the effects of CREB-phosphorylating kinases. In mouse models, such PDE inhibitors re-establish normal CREB phosphorylation, alleviate memory loss, and restore LTP. No current PDE5 inhibitors are selective enough to be used in elderly populations, but Arancio described a novel PDE5-inhibiting compound with high potency and selectivity that rescues LTP in amyloid-β-treated hippocampal slices. His group is currently modifying the compound to improve druggability.

The promise and pitfalls of growth factors

Neurotrophins such as nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) have essential roles in synaptogenesis, modulating arborization and pruning of dendrites, and in synaptic plasticity, but the complex effects of these factors and their precursors are only partly understood. While neurotrophins promote neuronal survival and differentiation, the precursor form of nerve growth factor (NGF), may contribute to cell dysfunction or death, explained Barbara Hempstead of Weill Cornell Medical College.

In a variety of animal model systems and some SNP studies in humans, alterations in BDNF affect memory, anxiety, depression, and aggression. "Even modest increases of BDNF can have behavioral consequences," said Hempstead. BDNF is reduced in the brains of AD patients, and gene therapy delivery of BDNF in animal models improves memory and rescues neurons from lesion. A common genetic variant in the prodomain of BDNF in humans affects hippocampal memory and motor learning.

Gene therapy trials in humans are already investigating the therapeutic benefit of delivering nerve growth factor to the nucleus basalis in AD patients, but because of its invasive and technologically challenging nature, gene therapy may not be the best solution. An alternative approach would activate the TrkB receptor, which responds to neurotrophins, and various groups are now devising compounds to mimic the loop structures of BDNF.

At high concentrations, proneurotrophins, the precursor compounds of neurotrophins, paradoxically activate cell death mechanisms. ProNGF may also be expressed in neurodegenerative disease and even during mild cognitive impairment. Preventing proNGF actions may offer another therapeutic strategy; Hempstead suggested that given the complex protein structure, antibody-mediated blockade might be a good tactic to explore.

Speakers:
Stephen M. Strittmatter, Yale University School of Medicine
Eric Klann, New York University
Lawrence P. Wennogle, Intra-Cellular Therapeutics

Highlights

• Low-abundance lipids are highly relevant to core processes in Alzheimer's disease (AD).
• Oligomers of amyloid-β bind to postsynaptic prion protein.
• Interactions between amyloid-β and prion protein may trigger changes in the phosphorylation and redistribution of NMDA receptors.
• Reactive oxygen species cause neuronal damage in AD.
• Overexpressing superoxide dismutase-2, which decreases superoxide, rescues deficits in memory and synaptic plasticity in AD model animals.
• A compound that blocks gamma-secretase activating protein reduces amyloid-β without affecting processing of the important protein Notch.
• PDE1 inhibitors could act as cognitive enhancers by increasing dopamine D1 receptor activity.

Lipids at the synapse

The third pathological hallmark of Alzheimer's disease, as noted by Alois Alzheimer in 1907, is lipid imbalance. Normal lipid function is essential to maintaining membrane properties, synaptic function and well-regulated signaling, but various lipids whose presence is important in the brain are dysregulated in AD and changes in their levels (both increases and decreases) can have powerful effects despite their low abundance. Cholesterol has received most of the attention, but metabolism of synaptojanin and phospholipase D is also altered in AD, said Gilbert Di Paolo of Columbia University. "The evidence is in the literature, phospholipases somehow have gone wild in AD," he said.

An unbiased "lipidomics" approach offers a way to identify previously unrecognized players in AD pathology. A comparison of lipid profiles of several mouse AD models and human AD brain samples, executed with HLPC and electrospray-ionization mass spectrometry, points to changes in diacylglycerol (DAG), the membrane component phosphatidylinositol 4,5 bisphosphate (Pi(4,5)p2 or PIP2), lysobisphosphatidic acid (LBPA), cholesterol esters, and other lipids.

The PIP2 products IP3 and DAG have a number of roles in neurons including stimulating the release of calcium from the ER, and regulating ion channels, cell adhesion processes, and exocytosis and endocytosis of synaptic vesicles. Experiments have suggested that amyloid-β42 leads to a decrease in PIP2 levels at the cell surface but also activates phospholipase C (PLC) metabolism of PIP2 to IP3 and DAG. To examine the effects of restoring the levels of PIP2 at the membrane, the team looked in a system where the conversion of PIP2 into PIP is absent because of the lack of the lipid phosphatase synaptojanin. Synaptojanin has been shown to be important in vesicle trafficking at the synapse. Slice experiments with a synaptojanin heterozygous knockout mouse show protection against PIP2-deficiency and LTP depression brought on by amyloid-β42, the most toxic version of the oligomer. Cognitive deficits are rescued when AD model mice are crossed with this knockout mouse, despite the fact that amyloid-β42 levels remain high, suggesting a downstream effect. Aβ42 perturbs dendritic spines, promoting "weak" signaling spines, and blocking synaptojanin also rescues this effect.

Amyloid and the prion connection

Plaques of amyloid-β may not be the most toxic forms of the protein; recently, attention has shifted to oligomers, which may be involved with cell surface binding in a way that leads to synaptic impairment. More specifically, said Stephen Strittmatter of Yale University School of Medicine, oligomers bind to dendrites in a way that suggests a receptor-like mechanism is at work. An unbiased genome-wide screen of cDNAs from mouse brain identified post-synaptic cellular prion protein (PrP-C), the normal version of the glycoprotein affected by prion disease. Several labs have confirmed that amyloid-β oligomers bind to prion protein.

The interaction of oligomeric assemblies with PrP activates Fyn kinase in a dose-dependent manner in transfected cells and in primary cortical neurons in culture. Fyn activation in turn phosphorylates the NR2B subunit of NMDA receptors, which are involved in learning and memory. NMDA receptor function and trafficking are affected by phosphorylation. When cells were preincubated with amyloid-β, NMDA receptor responses increased initially in response to calcium signaling, and subsequently were suppressed in a manner that correlates with surface expression of NMDA receptors. amyloid-β also drove NMDA receptors toward the surface.

PrP seems to be required for the amyloid-β related suppression of LTP, although not all groups have reached this same finding. Human AD brain extracts also inhibit LTP in vivo in a way mediated by PrP. Crossing AD model mice to PrP knockout animals rescued memory deficits, as did treating AD model mice with a high-dose PrP antibody.

The Roles of ROSs

Reactive oxygen species such as superoxide are another cellular stress that may be involved in the pathology of aging as well as AD and other neurodegenerative diseases. Reducing superoxide through the overexpression of superoxide dismutase 2 (SOD-2), can prevent memory deficits in one mouse model of AD, and rescue LTP impairments, said Eric Klann of New York University. SOD-2 overexpression also reduces superoxide levels in these mice, and reduces amyloid plaque formation by half.

Because injecting SOD-2 into the brain is not a likely therapeutic strategy, Klann's group explored MitoQ, a mitochondrial-targeted antioxidant. MitoQ prevented some deleterious effects of amyloid-β, such as increases in mitochondrial superoxide and LTP blockade. "This suggests that ongoing production of superoxide from mitochondria causes synaptic dysfunction, and implies that just by decreasing superoxide you can restore some degree of synaptic plasticity," said Klann.

Two more targets

Gamma-secretase, the enzyme that cleaves amyloid precursor protein into amyloid-β, is a challenging target because it is such a promiscuous protease with many physiologically important substrates. Intra-Cellular Therapeutics is pursuing an alternate strategy to interfere with amyloid-β production that involves targeting γ-secretase activating protein (GSAP), explained Lawrence Wennogle. GSAP apparently facilitates the interaction between γ-secretase and APP. Knocking down amyloid-β with GSAP siRNA does not affect cleavage of Notch, an essential protein. In collaboration with Rockefeller's Paul Greengard, Wennogle's group has identified a lead candidate that interferes with GSAP and does not affect Notch. "We've taken promiscuous γ-secretase and added specificity to this system," said Wennogle.

Another Intra-Cellular Therapeutics development program relevant to synaptic activity involves inhibitors for phosphodiesterase 1 (PDE1), which shuts down dopamine signaling in the brain. Inhibiting PDE1 might increase dopamine D1 receptor activity and have cognitive-enhancing effects. In an animal model, their lead compound enhanced all phases of memory, including acquisition, consolidation, and retrieval. It is currently in Phase 2 trials for cognitive enhancement in schizophrenia.

The synapse seems to be the place where dementia "happens," as AD brings on a wide range of marked changes in both animal models and human patients. As complex as this region is, it also offers a wide range of possible targets—a rich and promising region for AD therapeutics.

What protects the brains of the nondemented "oldest old"—those over 85?

Why do the brains of the "oldest old" show less plaque pathology than younger brains, even at the same level of AD symptom severity?

Do mild stresses like cognitive challenge or exercise induce an adaptive stress response?

Do environmental interventions such as restricting caloric intake or exercising promote neuronal resilience via neurotrophins?

Does early synaptic loss underlie cognitive deficits?

What role does proNGF play in neurodegeneration and cell death?

What anomalies in membrane lipid metabolism are caused by amyloid-β?

Does prion protein interact with other amyloid-β binding proteins?

Is prion protein necessary for the LTP-suppressing affects of amyloid-β?

What is the precise molecular mechanism by which synaptojanin inhibition prevents LTP impairment?

Can reducing reactive oxygen species in the aged brain restore synaptic plasticity?