3 December 2003. This meeting report is anchored by Kelly Dineley, with individual presentations written by the respective investigators.
At the 33rd Annual Meeting of the Society for Neuroscience in New Orleans, approximately 100 scientists gathered to discuss informally what roles the interaction between Aβ and the nicotinic acetylcholine receptor (nAChR) play in normal cognitive function and Alzheimer's disease. This meeting was prompted by the increasing number of reports of high-affinity interactions between Aβ peptides and nAChRs. The type of interaction (agonist vs. antagonist; competitive vs. non-competitive) varies, as does the subtype (α7 vs. α4/β2) and location (presynaptic vs. postsynaptic vs. somatic) of the nAChR involved.
After a brief introduction, David Sweatt, Baylor College of Medicine, Houston, Texas, posed the following six questions which were addressed through open discussion. I have answered them based upon that dialogue:
1. Does Aβ interact with nAChRs? Yes. There are several examples in the literature, as well as approximately 12 abstracts submitted to the present SfN meeting. Which receptors? There are reports for α4/β2 and α7-containing neuronal nAChRs, as well as the muscle nAChR.
2. Does Aβ activate or inhibit nAChRs? There is evidence in the literature for both. On the surface these reports appear contradictory; however, the biological preparations and methods of detection vary among laboratories. These differences indicate that the cell population, subcellular location, accessory proteins, and posttranslational modifications may influence the net effect of Aβ on nAChRs.
3. Is this interaction good or bad? Possibly both, depending on the concentration and aggregation state of the peptide. Sally Frautschy, University of California, Los Angeles, reported that infusion of Aβ peptide into cerebral ventricles of rodents initially improves cognitive performance, followed later by impairment (as Aβ accumulates, in theory; see ARF related news story). Clearly, further study is needed.
4. What is the normal physiological role of Aβ-nAChR interaction? It is noteworthy that non-demented brain contains picomolar concentration of Aβ. a concentration sufficient for interaction with nAChRs. This suggests that Aβ-nAChR interaction has a biological function under normal physiologic conditions. Published studies indicate that Aβ-nAChR interaction leads to second messenger system activation, intracellular Ca2+ increases, and increased neurotransmitter release. Other studies report that Aβ interaction with nAChRs leads to receptor antagonism, indicating that the precise location and conditions under which receptor-peptide interaction takes place can lead to quite varied outcomes. Thus far, it appears to be a complicated relationship, and we will gain clarity only through further research.
In addition, a role in Aβ metabolism was cited, since the literature has documented nAChR-mediated endocytosis. The role of astrocytes and a possible link to inflammation have precedence in the literature, as well. Finally, since it appears that muscle nAChRs and Aβ interact, there are implications yet to be explored for muscle function and peripheral amyloid disease.
5. What are the therapeutic opportunities based on Aβ-nAChR interaction? Discussion of this question mainly led to more questions: Should we block the interaction? What if nAChRs serve to remove Aβ from other deleterious interactions? Should we provide more nAChR-like binding sites as a decoy for Aβ-nAChR interaction? Is there a basal level of good Aβ-nAChR interaction? Our understanding of Aβ-nAChR interaction is too inchoate to decide these issues.
6. What important future directions might this research take? One suggestion was to define clearly where Aβ and nAChRs interact. Based on known nAChR and Aβ biology, participants urged investigation of potential intracellular interactions between the two proteins, in addition to the well-documented extracellular contact.
An additional point was to encourage a more accurate structural description of the β-amyloid used for study. Thus far, most investigators report that they prepare freshly solubilized, nonfibrillar Aβ for experimentation. However, few report direct investigations into the structure of their Aβ mixtures. Below are summaries of some individual presentations.
Kelly Dineley, University of Texas Medical Branch, Galveston, presented preliminary data obtained from cognitive testing of a new mouse model for altered nAChR function and β-amyloid overproduction. The model is a genetic cross between mice null for α7 nAChRs and the Tg2576 strain. Tg2576 mice exhibit an associative learning deficit at five months of age as measured with the rodent fear-conditioning paradigm. Coincident with this is an upregulation of α7 nAChRs and dysregulation of ERK MAPK signaling in the hippocampus of these animals (see also SfN abstract 945.13). Proper ERK MAPK signaling is necessary for rodent fear learning. Five-month-old Tg2576 are impaired in the contextual test for fear learning 24 hours following two pairs of conditioned stimulus (CS:cue) and unconditioned stimulus (US:footshock). This impairment is overcome with a more rigorous training schedule of five pairs of CS-US. Five-month-old animals lacking α7 nAChRs which also produce excessive β-amyloid are further impaired in the contextual test for fear learning in that five pairs of CS-US fail to impart memory of the context in which CS-US were delivered during training 24 hours prior. These results indicate that the relationship between β-amyloid and α7 nAChRs is more complex than a simple ligand-receptor interaction. If that were the case, one would expect that knocking out α7 nAChRs would alleviate the contextual learning deficit induced by excessive β-amyloid production. Since the deficit worsens, it implies that α7 nAChRs may be important for buffering the toxic effects of elevated β-amyloid. These animals are currently being evaluated for plaque load, Aβ oligomer load, and total Aβ burden, as well as for inflammatory status.
Jerryl Yakel, NIEHS, Research Triangle Park, North Carolina
We found that Aβ1-42 inhibits whole-cell and single-channel nAChR currents from rat CA1 stratum radiatum interneurons in hippocampal slices at concentrations as low as 100 nM. This inhibition appears specific for neuronal nAChRs, because Aβ1-42 had no effect on glutamate or serotonin 5-HT3 receptors, and may be a direct effect on the channel rather than an indirect effect via a signal transduction cascade. In addition, the magnitude of Aβ1-42 inhibition was dependent on the subtype of nAChR. When investigating the block of the single-channel currents by Aβ1-42, the non-α7/62 pS channel was significantly more inhibited (i.e., 54 percent) than was the α7-containing 38 pS channel. In addition to the full-length Aβ1-42 peptide, we previously showed that the fragment of Aβ1-42 including amino acid residues 12-28, Aβ12-28, also inhibited these channels. Recently, we showed that propionyl-amyloid β-protein (31-34) amide (Pr-Aβ31-34), a small fragment of Aβ1-42 which previously was reported to block the neurotoxic effects associated with Aβ1-42 on cholinergic neurons of the rat magnocellular nucleus basalis in vivo, and Aβ31-35, also blocked nAChRs, in a rapid and dose-dependent manner with a similar potency to Pr-Aβ31-34.
Darwin Berg, University of California, San Diego
We've found that Aβ specifically and preferentially blocks the α7-nAChR response both in chick ciliary ganglion neurons and in rat hippocampal neurons (Liu et al., 2001). The blockade is reversible, voltage-independent, and non-competitive. The peptide does not block α3-containing nAChRs under our conditions or non-cholinergic ionotropic receptors. So there does appear to be specificity; we did not test it on α4/β2-nAChRs. Interestingly, the blockade can be compensated by using specific albumins to potentiate the remaining response (Conroy et al., 2003). The potentiation appears to depend on specific sequences within the albumin rather than on absorbed components, and is likely to involve the extracellular N-terminal domain of the α7-nAChR. Synthetic compounds devised to mimic the albumin effect may have significant therapeutic value. We have not seen Aβ-induced currents attributable to α7-nAChR activation, though we tested the peptide under a variety of conditions and concentrations with and without potentiators present. It is quite possible that Aβ interacts with α7-nAChRs in a variety of ways, depending on the cell type and the history of the receptor, vis-a-vis posttranslational modification status. Recently, we have identified postsynaptic molecular scaffolds associated with α7-nAChRs in neurons. The scaffolds involve PDZ-containing proteins and mediate calcium-dependent downstream signaling as exemplified by nicotinic regulation of gene expression. The scaffolds introduce a new complexity to possible consequences of Aβ/α7-nAChR interactions, since the composition of the scaffold varies with receptor subtype and host cell. Further, the receptor is subject to rapid nicotine-induced SNARE-dependent trafficking, presenting the possibility of receptor-mediated Aβ internalization. These findings suggest numerous mechanisms by which Aβ may impact nicotinic signaling in the nervous system.
Robert Nichols, Drexel University College of Medicine, Philadelphia
We assessed the effect of Aβ peptides on nicotine-evoked changes in presynaptic Ca2+ level via confocal imaging of isolated presynaptic nerve endings from rat hippocampus and neocortex. Picomolar Aβ1-42 was found to directly evoke sustained increases in presynaptic Ca2+ via nAChRs. The direct effect of Aβ was found to be sensitive to α-bungarotoxin, mecamylamine, and dihydro-β-erythroidine, indicating involvement of both α7-containing nAChRs and non-α7-containing nAChRs. Prior depolarization strongly attenuated subsequent Aβ-evoked responses in a manner dependent on amplitude of the initial presynaptic function. Together, these results suggest that the sustained increases in presynaptic Ca2+ evoked by Aβ may underlie disruptions in neuronal signaling via nAChRs.
Jie Wu, Barrow Neurological Institute, Phoenix, Arizona
We employed patch-clamp techniques to elucidate acute effects of Aβ1-42 on human α4/β2 nAChRs that are heterologously expressed in SH-EP1 cells. One nM Aβ1-42 reduced both peak and steady-state components of nicotine-induced currents, accelerated acute desensitization, and slowed channel relaxation. Aβ1-42 modulates α4/β2 nAChR-mediated currents in a concentration-dependent manner, and needs appropriate Aβ pretreatment. Comparison of effects of Aβ1-42 on α4/β2 and α7 nAChR-mediated currents indicates that at pathological concentration (1nM), Aβ1-42 significantly modulates α4/β2 nAChRs function without affecting α7 nAChRs. At pharmacological concentrations (>100 nM), Aβ1-42 modulates both α4/β2 nAChRs and α7 nAChRs, with more dramatic effects on α7 nAChRs.