Updated 18 February 2010.
3 February 2010. Death and taxes are reputedly inevitable, though death and receptors may be more interesting to Alzforum readers. Both were in season at this month’s Keystone symposium, “Alzheimer’s Disease Beyond Aβ,” held 10-15 January 2010 at Copper Mountain, Colorado. Frank Longo, of Stanford University School of Medicine, California, reviewed evidence for the part the neurotrophin receptor p75 plays in amyloid-β toxicity, arguing that small-molecule p75 ligands may protect against AD. Marc Tessier-Lavigne of Genentech Inc., San Francisco, California, updated the audience on a topic that first made a splash at last year’s Keystone symposium (see ARF related news story). It is about a neurodegenerative cascade the N-terminal end of the amyloid precursor protein (N-APP) can touch off when it binds to death receptor 6 (DR6), a member of the tumor necrosis factor superfamily. The two stories are related because Tessier-Lavigne’s data suggest that N-APP may bind to p75 as well.
This neurotrophin receptor has been linked to AD for some time. For example, as Longo noted at Copper Mountain, p75, like its close relative DR6, is expressed in brain areas most vulnerable in AD, including the entorhinal cortex, hippocampus, and basal forebrain, and expression is elevated in patients with the disease. The receptor mediates Aβ-induced cell death according to data from Elizabeth Coulson’s group at the University of Queensland, Brisbane, Australia (see Sotthibundhu et al., 2008), and recently Longo and colleagues reported that p75 mediates Aβ-induced neuritic dystrophy, one of the key pathological findings in AD (see Knowles et al., 2009). However, other reports suggest p75 can also be protective in AD (see Bengoechea et al., 2009).
Much of the early work Longo described stemmed from the use of p75 knockout (KO) mice. Rather than having the full gene knocked out, these animals are missing the third exon and hence the majority of the receptor. Primary neurons from these knockouts survive in culture and resist Aβ42, showing less dystrophy when exposed to the peptide than do wild-type neurons. Neuritic dystrophy is milder in offspring from APP transgenic mice (with Swedish and London mutations) crossed with p75 KO mice. Those offspring also have fewer plaques than the parent human APP transgenic strain.
After summarizing some of his lab’s recent findings, Longo outlined a strategy to develop small-molecule p75 ligands to treat AD. The work is being done at Pharmatrophix, a startup company Longo co-founded. The company received initial support from the Alzheimer’s Drug Discovery Foundation (see marketwire story) and has since entered a partnership with Elan.
Longo and colleagues have developed small-molecule ligands that bind to p75, block Aβ-induced neurodegeneration, and prevent some of the downstream cascades associated with Aβ toxicity, such as activation of Cdk5, GSK3β, Jnk, tau phosphorylation, and inhibition of Akt (see Yang et al., 2008). Some of the molecules developed by Pharmatrophix are active in the picomolar range, Longo said, adding that he has begun testing them in both hippocampal slices and in vivo in mouse models of AD.
In cooperation with Mike Shelanski’s lab at Columbia University, New York, Longo found that the compounds can protect against Aβ-induced loss of dendritic spines in hippocampal slices. Together with Ottavio Arrancio, also at Columbia, the researchers found that the p75 ligands rescue LTP deficits in tissue from APP/PS1 mice. Several of these compounds get into the brain, Longo said. Though they do not affect plaques or the levels of Aβ as judged by ELISA, Longo reported that they do reduce neuritic dystrophy. The compounds rescue spine loss in pyramidal neurons, as well as deficits in novel object recognition and contextual fear conditioning exhibited by APP transgenic mice. In normal mice, the compounds also seem to have some benefit, Longo noted, preventing age-related loss of basal forebrain cholinergic neurons and shortening of neurite outgrowth.
Shortening and retraction of axons is also one of the consequences of activating DR6. When Tessier-Lavigne reported last year that N-APP activated the death receptor pathway, his data were limited to embryonic neurons deprived of growth factor support. Though the work introduced a potential new role for APP in neurodevelopment, its relevance to post-embryonic tissues, and to Alzheimer disease (AD) in particular, was unclear. Tessier-Lavigne’s group has since extended those studies to juvenile and adult neurons, and presented some of his newest data at the Copper Mountain conference. At present, the bottom line is that the N-APP/DR6 pathway may regulate synaptic structure and axon growth in adult tissue, but the question of whether it spells doom in the form of dementia remains open.
Using a transcranial window to examine neurons in 60-day-old mice, Tessier-Lavigne and Genentech colleagues Dara Kallop and Robby Weimer found that the density of dendritic spines is higher in DR6 KO animals than in DR6 heterozygotes. A similar phenotype was recently reported in APP knockout (KO) animals compared to APP heterozygotes by Jochen Herms and colleagues (Bittner et al., 2009). The apparent similarity in the phenotypes raises the possibility that N-APP’s activation of DR6 may help regulate synaptic structure and possibly plasticity.
Evidence that the signaling pathway controls axon growth and regeneration comes from both in vitro and in vivo work. Tessier-Lavigne showed that N-APP shortens axons in postnatal sensory and cortical neurons, and that this depends on activating DR6 (again the DR6.1 antibody or DR6 knockout blocked the effect). In these neurons, N-APP turned on caspase 6, which mediates DR6 signaling in embryos, suggesting that the same signaling pathways are at work in developing and developed neurons. To test if DR6 regulates axons in vivo, the scientists turned to an axon lesion paradigm where after transection, some axons regenerate but almost never extend through the lesion site. The scientists found more extensive regeneration in DR6 knockout animals. Six weeks after the injury, the DR6 KO animals had less dying back of axons than did control animals, and some of the regenerating axons actually crossed the lesion scar. Together, this evidence suggests that DR6 modulates axon growth in adult neurons, possibly through activation by APP, Tessier-Lavigne said.
How N-APP/DR6 relate to Alzheimer’s remains up for grabs; certainly there is no direct evidence yet that the death receptor pathway is causal or linked to AD. That said, Tessier-Lavigne cited some observations to support the idea that might be relevant to dementia. As far back as 1989, Greg Cole, then at the University of California, San Diego, with colleagues including the late Tsunao Saitoh, found that antibodies against the N-terminus of APP detected a unique peptide in the brain of AD patients and also a protein of 35 KDa (the size of the N-APP fragment) in soluble fractions from AD brain but not controls (Cole et al., 1989). The same scientists went on to show that antibodies against the APP N-terminus reacted with plaques (Cole et al., 1991). Those findings suggest that N-APP levels may be higher in AD brain. In addition, DR6 expression could help explain why neurodegeneration in AD is limited to specific regions of the brain despite the ubiquitous nature of APP. DR6 is highly expressed in areas where disease pathology is thought to occur earliest, such as the hippocampus and the entorhinal cortex, but sparsely expressed in areas spared in AD, such as the striatum. The gene for DR6 maps to a susceptibility locus for late-onset AD on chromosome 6, and it activates caspase 6, which occurs in plaques and tangles and is elevated in both general aging and in mild cognitive impairment. “It appears that in AD the ligand [N-APP] is there, the receptor is there, and the pathway might be active,” concluded Tessier-Lavigne.
There may also be more to the N-APP story than activation of DR6. Though the binding is less tight, Tessier-Lavigne and colleagues previously reported that N-APP does interact with the neurotrophin receptor p75, another member of the TNF superfamily, and Tessier-Lavigne thought this partnership might be physiologically relevant. He noted that N-APP can induce axon shortening in cerebellar neurons that do not express DR6 but do have p75, and the inhibition is relieved by interfering with p75. “It is possible that DR6 and p75 do a similar job,” he said. Whether Pharmatrophix’s p75 inhibitors might prevent N-APP-dependent neurodegeneration is an open question at this point.—Tom Fagan.