Viewpoint pubs.acs.org/chemneuro
N‑Terminal Hypothesis for Alzheimer’s Disease Brian Murray, Bhanushee Sharma, and Georges Belfort* Howard P. Isermann Department of Chemical and Biological Engineering and the Center for Biotechnology & Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, New York 12180-3590, United States ABSTRACT: Although the amyloid (abeta peptide, Aβ) hypothesis is 25 years old, is the dominant model of Alzheimer’s disease (AD) pathogenesis, and guides the development of potential treatments, it is still controversial. One possible reason is a lack of a mechanistic path from the cleavage products of the amyloid precursor protein (APP) such as soluble Aβ monomer and soluble molecular fragments to the deleterious effects on synaptic form and function. From a review of the recent literature and our own published work including aggregation kinetics and structural morphology, Aβ clearance, molecular simulations, longterm potentiation measurements with inhibition binding, and the binding of a commercial monoclonal antibody, aducanumab, we hypothesize that the N-terminal domains of neurotoxic Aβ oligomers are implicated in causing the disease. KEYWORDS: Aβ peptide, N-terminus, fragments, multivalent interactions, Alzheimer’s disease
W
and A2V Aβ1−42 mutant.2 Blocking the N-terminus with sequence specific antibodies prevents LTP deficit whereas blocking the central hydrophobic core (CHC) and C-terminus have minimal effects. N-terminally extended fragments of APP induce LTP deficits only when the N-terminus is included in the fragment (Aη-α compared with Aη-β, Figure 1).5 (4) Aducanumab: Aducanumab, currently in Phase 3 trials, binds to the N-terminus (residues 3−6) of Aβ1−42 oligomers not monomers. (5) N-Truncated Aβ: After aminopeptidase degradation of the N-terminal D and A amino acids, the remaining glutamic acid (E) can be cyclized in a dehydration reaction with glutaminyl cyclase forming N-terminally truncated, pyroglutamylated (pE) forms of Aβ. These species (AβpE3−42, AβpE11−42, AβpE3−40, and AβpE11−40 fragments) are strongly associated with AD, are more stable, abundant and toxic than Aβ1−42 and Aβ1−40, and have been proposed as initiators of AD pathogenesis. Additionally, since monomer is not neurotoxic while dimers and trimers are and aducanumab only binds to oligomers, this suggests that at least two N-termini are necessary for LTP deficit induction. Thus, the mechanism of neurotoxicity could be due to multivalent interactions between the N-termini (≥2) of Aβ oligomers with receptors such as the glutamate or Nmethyl-D-aspartate receptors. Taken together, these collective findings strongly suggest that the N-terminus of Aβ oligomers could be determinant for AD. However, this supporting information is not without limitations. The role of decreased and increased APP cleavage toward protection and enhancement of AD, respectively, is unclear. While the protective mutation, A673T, causes a reduction of the production of Aβ by about 50%,1 it was recently reported that decreased β-secretase cleavage correspondingly increases proteolysis by η-secretase at a recently discovered APP cut-site (Figure 1).5,6 Interestingly, these N-
ith the announcement less than 3 months ago that Lilly’s drug, solanezumab, has failed in a large phase III clinical trial, it seems reasonable to ask why this antibody that binds to the amino acid region 16−26 of the abeta (Aβ) peptide failed! If our N-terminal hypothesis is valid, then binding to amino acid region 1−14 residues is preferred rather than 16−26 residues and it predicts that aducanumab (binds to residues 3−6, Biogen) rather than solanezumab is preferable. Historically, much emphasis has been placed on hydrophobic interactions that drive amyloid aggregation as potential hotspots for therapeutic targets (i.e., the central hydrophobic core (CHC) and the hydrophobic C-terminal, or residues 17− 24 and 32−42, respectively). Here, we present the evidence for the relevance of amyloid-β’s N-terminus (residues 1−14), the least hydrophobic domain of the peptide (Figure 1): (1) APP Fragmentation: The first discovered protective mutation of APP (A673T) or Aβ (A2T) against AD reduces BACE1 cleavage at the β-secretase position (N-terminus of Aβ) resulting in a drop in Aβ concentration of about 50%.1 A causative mutation at the same location (A673 V) or Aβ (A2V) results in an increase in Aβ concentration of about 100%. (2) Biophysical Properties: The synthetic A2T and A2V mutants of Aβ1−42 increase the aggregation lag time prior to the onset of fibril formation compared with wild type Aβ1−42 by a factor of ∼1.5 and ∼8, respectively.2 Aggregate morphology of the N-terminal mutants (A2T and A2V) is altered compared with wild type Aβ1−42.2 Molecular simulations show the importance of N-terminus on monomer folding and lowering of the formation of neurotoxic β-hairpin structures for the A2T Aβ1−42 mutant.3 Recently reported experimental structures also show a flexible/exposed N-terminal region (Aβ1−14) in the disease relevant Aβ1−42 fibril.4 The results of Das et al.3 provide a possible rational for protection by the A2T mutant by demonstrating that N-terminal contacts with the central hydrophobic core could inhibit formation of the doublehorseshoe-like cross-β-sheet structure of the Aβ1−42 fibril.4 (3) Long-Term Potentiation: Long-term potentiation (LTP) deficit, which correlates with memory and learning, is reduced for the A2T Aβ1−42 mutant, in comparison with the wild type © XXXX American Chemical Society
Received: January 27, 2017 Accepted: January 30, 2017
A
DOI: 10.1021/acschemneuro.7b00037 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Viewpoint
ACS Chemical Neuroscience
Figure 1. APP fragmentation scheme. Amyloid precursor protein (APP770) fragmentation proceeding along two possible pathways of fragmentation, the β path and the recently discovered η path. The β path produces Aβ monomer by β-secretase cleavage between residues 671 and 672 of APP770, and subsequent γ-secretase cleavage to produce Aβ of length 39−43. Additionally, the β path also forms a 16 residue N-terminal fragment of Aβ by additional cleavage by α-secretase between the positions 687 and 688 (β-α, colored in light blue). Alternatively, the η path, driven by η-secretase cleavage between residues 579 and 580, produces two main N-terminally extended fragments, one that contains the N-terminus of Aβ, Aη-α, and one that does not, Aη-β. Species reported to be neurotoxic are highlighted by the red striped background (adapted from Willem et al.5).
terminally extended APP fragments induce LTP deficits to a similar degree as synthetic Aβ1−42 dimers.5,6 There is the potential that η-secretase cleavage occurs much less frequently in vivo than in vitro. Also, extending the Aη-α and Aη-β fragments to N-terminus (Aβ 1−14 ) containing sAPPα (APP18−687/688), and N-terminus (Aβ1−14) lacking sAPPβ (APP18−671/672) fragments are reported be neuroprotective and toxic, respectively. A direct relationship between the biophysical properties of Aβ and the pathogenesis of AD has been notoriously difficult to establish. It is well-known that Aβ aggregates into oligomers, fibrils, and eventually plaques in vivo, and that the oligomers are believed to be the most neurotoxic of these species. However, since amyloid plaques have been repeatedly shown to not correlate with AD symptoms as well as other biomarkers, it is unclear whether increasing or decreasing the overall rate of monomer aggregation into plaques translates into increased or decreased AD pathogenesis. There remain several questions regarding the LTP measurements with respect to Aβ’s N-terminus as being necessary for LTP deficit induction. Experiments, reporting that recombinant Aη-α induced in vitro LTP deficits whereas recombinant Aη-β did not, lacked explicit identification of the concentration of the fragments used in the experiment. Additionally, the same group reported that both synthetic Aη-α and Aη-β induced comparable LTP deficits, thus calling into question the necessity of the N-terminus. Additionally, in light of the neurotoxicity of Aη-α and Aη-β and their size being approximately the same as that of Aβ dimers, there are concerns regarding the conclusions of the experiments that demonstrated blocking the N-terminus of Aβ with a sequence specific antibody prevents LTP deficits whereas both CHC and
C-terminus sequence specific antibodies were less effective. There is the potential that these ∼8 kDa molecules, originally believed to be Aβ dimers, were actually N-terminally extended fragments such as Aη-α, in which case, the effects produced by the three domain specific antibodies against Aβ remain inconclusive in identifying the segment of the protein most responsible for LTP deficit induction. To complicate this further, synthetic β-α (residues 1−16 of Aβ) have been reported not only to prevent the LTP deficits caused by synthetic Aβ dimers, but also to enhance in vitro LTP deficit on their own. Clearly, further work needs to be directed toward identifying the fragment of Aβ oligomers and the N-terminally extended species responsible for inducing LTP deficits at physiologically relevant concentrations. Despite the five main pieces of evidence from our and other reports in the literature implicating the “N-Terminal Hypothesis” for AD, substantial concerns can be raised. Future work should be directed toward identification of the residues directly responsible for neurotoxicity of Aβ oligomers and N-terminally extended APP fragments for the purpose of therapeutic development of inhibitors against a precise molecular target.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Georges Belfort: 0000-0002-7314-422X Notes
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors declare no competing financial interest. B
DOI: 10.1021/acschemneuro.7b00037 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Viewpoint
ACS Chemical Neuroscience
■
ACKNOWLEDGMENTS Dr. Sri V. Ranganathan is thanked for valuable comments. G.B. acknowledges general support from his endowed Institute Chair at RPI.
■
REFERENCES
(1) Jonsson, T., Atwal, J. K., Steinberg, S., Snaedal, J., Jonsson, P. V., Bjornsson, S., Stefansson, H., Sulem, P., Gudbjartsson, D., Maloney, J., Hoyte, K., Gustafson, A., Liu, Y., Lu, Y., Bhangale, T., Graham, R. R., Huttenlocher, J., Bjornsdottir, G., Andreassen, O. A., Jonsson, E. G., Palotie, A., Behrens, T. W., Magnusson, O. T., Kong, A., Thorsteinsdottir, U., Watts, R. J., and Stefansson, K. (2012) A mutation in APP protects against Alzheimer’s disease and age-related cognitive decline. Nature 488 (7409), 96−9. (2) Murray, B., Sorci, M., Rosenthal, J., Lippens, J., Isaacson, D., Das, P., Fabris, D., Li, S., and Belfort, G. (2016) A2T and A2V Aβ Peptides Exhibit Different Aggregation Kinetics, Morphology, Structure and LTP Inhibition. Proteins: Struct., Funct., Genet. 84 (4), 488−500. (3) Das, P., Murray, B., and Belfort, G. (2015) Alzheimer’s Protective A2T Mutation Changes the Conformational Landscape of the Aβ1−42 Monomer Differently Than Does the A2V Mutation. Biophys. J. 108 (3), 738−747. (4) Walti, M. A., Ravotti, F., Arai, H., Glabe, C. G., Wall, J. S., Bockmann, A., Guntert, P., Meier, B. H., and Riek, R. (2016) Atomicresolution structure of a disease-relevant Abeta(1−42) amyloid fibril. Proc. Natl. Acad. Sci. U. S. A. 113, E4976−E4984. (5) Willem, M., Tahirovic, S., Busche, M. A., Ovsepian, S. V., Chafai, M., Kootar, S., Hornburg, D., Evans, L. D., Moore, S., Daria, A., Hampel, H., Muller, V., Giudici, C., Nuscher, B., Wenninger-Weinzierl, A., Kremmer, E., Heneka, M. T., Thal, D. R., Giedraitis, V., Lannfelt, L., Muller, U., Livesey, F. J., Meissner, F., Herms, J., Konnerth, A., Marie, H., and Haass, C. (2015) η-Secretase processing of APP inhibits neuronal activity in the hippocampus. Nature 526 (7573), 443−7. (6) Welzel, A. T., Maggio, J. E., Shankar, G. M., Walker, D. E., Ostaszewski, B. L., Li, S., Klyubin, I., Rowan, M. J., Seubert, P., Walsh, D. M., and Selkoe, D. (2014) Secreted amyloid β-proteins in a cell culture model include N-terminally extended peptides that impair synaptic plasticity. Biochemistry 53 (24), 3908−3921.
C
DOI: 10.1021/acschemneuro.7b00037 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
