23 February 2010. Written up as a novel, the relationship between Aβ and the α7 nicotinic acetylcholine receptor (α7 nAChR) has the makings of a bestseller—promise, peril, and mystery. Though earlier pages have insinuated sour interplay between the two proteins, the most recent chapter suggests they could have fruitful camaraderie during early-stage Alzheimer disease. Kelly Dineley, University of Texas Medical Branch, Galveston, and colleagues crossed α7 nAChR-null mice with the Tg2576 AD mouse model and found intensified learning and memory problems. Curiously, these α7 nAChR-deficient animals show premature evidence of a dodecameric Aβ oligomer that has been associated with early memory loss in that AD strain. Published in this week’s Journal of Neuroscience, the findings suggest that α7 nAChRs may be neuroprotective in early AD, possibly through effects on Aβ production and aggregation.
α7 nicotinic acetylcholine receptors (or α7 for short) have long captivated AD researchers and drug developers, perhaps more with intrigue than certainty. Upregulation of α7 has been linked to neuroprotection by several nicotinic receptor agonists (Jonnala and Buccafusco, 2001), and yet a number of studies suggest that Aβ peptides bind α7 receptors and that these interactions have harmful effects on tau (Wang et al., 2003) and NMDA-type glutamate receptors (Snyder et al., 2005). Nevertheless, several α7-targeting compounds are wending their way through the AD clinical pipeline (see ARF related conference story), and a recent report provides evidence that an α7 agonist can revive waning neurons by disrupting pre-existing Aβ-α7 complexes (Wang et al., 2009 and ARF related news story). Meanwhile, there are lingering concerns about whether and when such compounds would be effective, considering the controversial literature on whether α7 expression drops, remains stable, or increases during disease progression (see ARF related news story). This suggests, at the very least, that timing may be key for α7-targeted interventions, and the current study helps bear this out.
Given that genetic deficiency of α7 itself does not appear to influence motor or cognitive abilities in mice (Paylor et al., 1998), first author Caterina Hernandez and colleagues set out to test whether α7 loss would affect pathogenesis and memory loss in an AD model with early-stage disease. The researchers crossed α7 knockout mice with the well-characterized Tg2576 AD strain and analyzed the progeny at five months of age, when cognition is just starting to slip but plaques remain undetectable in the Tg2576 parent strain. They found that α7-deficient AD mice did worse in a test of hippocampal-dependent contextual fear learning, compared to Tg2576 animals with intact α7 receptors. In this behavioral test, mice learn to freeze in response to a tone that was previously paired with a foot shock. When this pairing was given twice in a test, AD mice do not learn the behavior as well as wild-type littermates. However, “if you pair the foot shock with the contextual conditioning five times during the training, the AD mice freeze to the same extent as wild-type animals. They’re rescued by the more robust training,” Dineley said, “whereas the α7-deficient AD mice were not rescued by the five-pair training.”
Compared with age-matched Tg2576 animals, α7-deficient AD mice also had modestly reduced numbers of hippocampal pyramidal and dentate gyrus neurons, and only about half as much choline acetyltransferase in terms of protein level and activity. Plaque load and total Aβ content in hippocampal tissues were also down in α7-deficient AD mice, though with greater enrichment of the more toxic peptide, Aβ1-42.
A bigger surprise came upon closer inspection of the Aβ aggregates in these mice. In immunoblots using the “OC” antibody that is specific for fibrillar amyloid oligomers (Kayed et al., 2007), the researchers detected 27 and 56 kDa proteins, the same molecular weight as Aβ hexamers and dodecamers. The Aβ12-mer, which appeared in the α7-knockout/Tg2576 mice but no other genotypes, could be the same as that originally described by Karen Ashe, Sylvain Lesné, and colleagues at University of Minnesota, Minneapolis. They identified a dodecamer as the soluble Aβ species responsible for early memory loss in Tg2576 mice as yet lacking brain Aβ deposition (Lesné et at., 2006 and ARF related news story). In that study, the dodecamer (dubbed Aβ*56, i.e., Aβstar56) showed up at six months of age but was absent in five-month-old Tg2576 mice. However, in Tg2576 animals without α7, it appears a dodecamer, as well as enhanced levels of a hexameric Aβ species, is now evident at five months of age, Dineley said.
Several reports suggest that Aβ*56 is present and associates with cognitive impairment in other AD transgenic lines, for example J20 (Cheng et al., 2007) and 3xTg (Oddo et al., 2006). Furthermore, Lesné and colleagues have evidence that the Aβ dodecamer may be clinically relevant. In biochemical studies of inferior temporal gyrus samples from 145 seniors with AD, mild cognitive impairment, or no cognitive decline, “we can easily detect Aβ*56, as well as other oligomers including dimers and trimers,” Lesné said of the soon-to-be-submitted data. “We see fluctuation of Aβ*56 levels with disease progression.” He and colleagues also examined samples from 49 of these participants, and could also detect all three oligomeric forms of Aβ. There has long been debate over whether these and other higher oligomeric forms may be artifacts created as brain homogenates are diluted for biochemical analysis, he said. “But it’s different with CSF because there’s no titration. All we do is take the samples and do immunoprecipitations in physiological conditions. For us, this is a fairly strong argument that the oligomers are truly present in the brain and not artificially created by extraction conditions.”
Dineley and colleagues are also continuing to pursue the oligomer story, based on their findings that Aβ aggregates seem to form more quickly in α7-deficient AD mice. “If α7 is present, how is it preventing the accumulation of the dodecamer with time? We’ve got in vitro studies addressing that right now,” she told ARF.
Aside from Aβ*56, the other big news from the current work is its apparent contradiction with a recent analysis showing α7 deficiency had beneficial effects in older AD mice of a different strain. In that study, Steve Heinemann, Gustavo Dziewczapolski, and colleagues at the Salk Institute in La Jolla, California, crossed α7 knockouts with PDAPP J9 AD mice, and found that lack of α7 rescues synaptic loss and long-term potentiation, and improves cognition in the AD mice (Dziewczapolski et al., 2009 and ARF related news story).
Though these findings seem at odds with Dineley’s, analyzing the two studies side by side could well be like comparing apples and oranges. First, the transgenes in the AD mice were different. The PDAPP J9 strain used in Heinemann’s investigation carries both Swedish (K670N, M671L) and Indiana (V717F) familial AD mutations, whereas the Tg2576 animals of the current study have only the Swedish mutation. Second, the UCSD researchers analyzed older (13 months and up) mice with advanced pathogenesis; the Texas group instead focused on three- to five-month-old mice with early-stage disease. The groups also used different behavioral tests—i.e., Morris water maze for the PDAPP mice, fear conditioning for the Tg2576 animals. Dineley said her group does plan to analyze older animals. Breeding has been a major hurdle, she said. The probability of getting the α7 knockout with the Tg2576 transgene is only one in eight, with the parent strains they use.
On the age issue, first author Dziewczapolski of the Heinemann lab said they had tried the water maze on a small group of PDAPP J9 mice of younger age (10 months), but their performance hardly differed from wild-type siblings. Hence, the researchers chose to analyze older animals with discernible cognitive deficits. Moreover, they gauged cognition using the water maze because none of their mice at this older age (13-16 months) showed deficits in fear conditioning.
Considering these differences, Dineley finds it “perfectly feasible” that 15-month-old α7-deficient mice could look the same on both AD transgenic backgrounds. “It may be that the rate of cognitive decline is different, but they end up at the same spot at very old age,” she said, noting her previous work showing that certain interventions, e.g., calcineurin inhibition, can reverse cognitive deficits in young (five months) but not older (12 months) Tg2576 mice (Taglialatela et al., 2009; Dineley et al., 2007). “These observations suggest that distinct molecular mechanisms play key roles during different ‘disease’ stages in mice,” she said.
Tampering with α7 receptors could also be time sensitive, suggests Darwin Berg of the University of California, San Diego. “Many lines of evidence point to α7 receptors as an attractive target for therapeutic intervention in Alzheimer disease, but this recent work makes clear the challenge posed by the different effects of the receptor,” he wrote in an e-mail to ARF (see full comment below).
Dineley agreed. “It may be that having α7 functional and present during early phases of the disease serves a neuroprotective function, whereas during late phases it may end up contributing to Aβ through prolonged interaction with the receptor,” she said.—Esther Landhuis.
Hernandez CM, Kayed R, Zheng H, Sweatt JD, Dineley KT. Loss of alpha7 Nicotinic Receptors Enhances beta-Amyloid Oligomer Accumulation, Exacerbating Early-Stage Cognitive Decline and Septohippocampal Pathology in a Mouse Model of Alzheimer's Disease. J Neurosci. 2010 Feb 17;30(7):2442-53. Abstract