Mensch M, Dunot J, Yishan SM, Harris SS, Blistein A, Avdiu A, Pousinha PA, Giudici C, Busche MA, Jedlicka P, Willem M, Marie H. Aη-α and Aη-β peptides impair LTP ex vivo within the low nanomolar range and impact neuronal activity in vivo. Alzheimers Res Ther. 2021 Jul 8;13(1):125. PubMed.
Recommends
Please login to recommend the paper.
Comments
Institut de Pharmacologie Moléculaire et Cellulaire
Backtrack on η-secretase-derived APP fragment-related toxicity
In the October 15, 2015, issue of Nature, Willem and colleagues reported a novel proteolytic cleavage of the Aβ precursor protein (APP, Willem et al., 2015). Specifically, the authors proposed that an η cleavage of APP by membrane-bound matrix metalloproteases such as MT5-MMP, and secondary cleavages of the η-secretase cleavage product (CTFη) by α-secretase and β-secretase produced cleavage fragments that had differential effects on neuronal function, with the α-secretase product (Aηα) being neurotoxic, whereas the β-secretase product (Aηβ) was not. The implication was that β-secretase inhibitors would contribute to disease progression by shifting the balance of CTFη cleavage toward the ostensibly more neurotoxic Aηα and away from the non-neurotoxic Aηβ fragment.
The pathophysiological processing of bAPP has been studied in depth from multiple points of view, including molecular genetics. Thus, multiple missense mutations associated with autosomal-dominant forms of familial AD have been described in genes involved in APP processing, including APP itself or the enzymes that cleave it. These robust, well-replicated, and widely accepted observations led to the empirical view that interfering with Aβ would ultimately prevent, slow, or arrest AD progression. However, most Aβ-directed clinical trials consistently failed.
In this context, Willem et al. drew considerable attention. The authors suggested that the CTFη fragment may be relevant to AD pathogenesis because it accumulated in dystrophic neurites in various AD mouse models as well as in human AD brains. A question then arose as to whether CTFη itself and/or its α-secretase- (Aηα) and β-secretase-derived fragments (Aηβ) were neurotoxic. Willem and colleagues reported that preincubation of hippocampal slices with Aηα-enriched conditioned media reduced long-term potentiation and suppressed neuronal activity. Overall, this interesting data strongly supported a potential pathogenic combined action of the MT5-MMP/α-secretase duo in AD pathology.
A key aspect of the study was the observation that unlike Aηα, Aηβ had little if any influence on synaptic plasticity. Willem et al. pointed out two major implications. First, their data would require a rethinking of the roles of the different APP-linked proteolytic events in the pathogenesis of AD. Second, it would require a revision of prevailing strategies to therapeutically target these cleavage events. Specifically, Willem et al. suggested that α-secretase behaved as a pathogenic protease, while BACE1 behaved as a nonpathogenic protease that favored the production of nontoxic Aηβ fragments, and should no longer be considered a therapeutic target in AD.
If true, these results could potentially have explained the lack of efficacy of BACE1 inhibitors in prior clinical trials and would diminish enthusiasm for the development of new BACE1 blocking agents. However, the study's conclusions remained puzzling because they are seemingly contradicted by the bulk of genetic, neuropathological, and cellular evidence indicating that interfering with BACE reduces Aβ production and associated neurotoxicity (Hampel et al., 2020).
Particularly troublesome in this regard are the accepted and widely replicated analysis of two human genetic variants: the Swedish mutation, which increases BACE1 proteolytic activity and triggers an aggressive familial form of AD, and the Icelandic mutation, which lowers BACE1 activity and reduces the risk of developing AD. These biological effects are in contradiction with Willem et al., 2015, which predicts that the 673 APP mutation would impair β-secretase-mediated production of Aηβ, while enhancing the α-secretase-mediated production of the neurotoxic Aηα fragment and thereby enhanced risk of AD.
These concerns seem to have been overlooked when the paper was published. Regardless, the biological robustness of the conclusions by Willem et al. is now directly questioned in this new study. The same authors (bar Dr. Haass) report that Aηα and Aηβ similarly impair LTP when examined using the same in vivo electrophysiological approaches and in vivo two-photon calcium imaging as in their original study. The explanation for the discrepancy with the earlier study carried out by the same groups is unclear. Mensch et al. note that recombinant Aηα and Aηβ were prepared by similar procedures, but suggest that the discrepant results could be related to the fact that, in the prior work, they did not precisely quantify the peptides, and that this could have led to a putative effect of Aηβ being missed. However, in their Nature article, Willem et al. reported a particularly careful analysis of Aη fragments using coupled liquid chromatography/mass-spectrometry, and Mensch et al. referred to the Nature paper to underline the fact that the purification procedure was identical in both papers.
Regardless, it is far from clear that minor differences in sample preparation could account for the very distinct functional outcomes reported in the original Nature article. This is particularly so because Aηα and Aηβ peptides in this new study are reported to be active at the low nanomolar range. An additional possibility raised by the authors in their recent paper concerns differences in o-glycosylation, purity, or degradation of the fragments. These explanations are also not compelling. Differences in the purity or the presence of degradation products would presumably have been detected by the similar mass-spec/liquid chromatography procedure described in their first paper. O-glycosylation occurs by the integration of glycan(s) on the hydroxyl group on either threonine or serine. A close examination of the sequence on Aηα that is absent from Aηβ indeed reveals a single serine residue that would have yielded o-glycosylated species with a delta MW of 200 to >800 Da, which are easily detectable by mass-spec/liquid chromatography.
Given the significant unmet need for therapies for this disease, it appears essential to avoid discrepant and puzzling data before inappropriately dismissing still-valid targets.
References:
Willem M, Tahirovic S, Busche MA, Ovsepian SV, Chafai M, Kootar S, Hornburg D, Evans LD, Moore S, Daria A, Hampel H, Müller V, Giudici C, Nuscher B, Wenninger-Weinzierl A, Kremmer E, Heneka MT, Thal DR, Giedraitis V, Lannfelt L, Müller U, Livesey FJ, Meissner F, Herms J, Konnerth A, Marie H, Haass C. η-Secretase processing of APP inhibits neuronal activity in the hippocampus. Nature. 2015 Oct 15;526(7573):443-7. Epub 2015 Aug 31 PubMed.
Baranger K, Marchalant Y, Bonnet AE, Crouzin N, Carrete A, Paumier JM, Py NA, Bernard A, Bauer C, Charrat E, Moschke K, Seiki M, Vignes M, Lichtenthaler SF, Checler F, Khrestchatisky M, Rivera S. MT5-MMP is a new pro-amyloidogenic proteinase that promotes amyloid pathology and cognitive decline in a transgenic mouse model of Alzheimer's disease. Cell Mol Life Sci. 2016 Jan;73(1):217-36. Epub 2015 Jul 23 PubMed.
Hampel H, Vassar R, De Strooper B, Hardy J, Willem M, Singh N, Zhou J, Yan R, Vanmechelen E, De Vos A, Nisticò R, Corbo M, Imbimbo BP, Streffer J, Voytyuk I, Timmers M, Tahami Monfared AA, Irizarry M, Albala B, Koyama A, Watanabe N, Kimura T, Yarenis L, Lista S, Kramer L, Vergallo A. The β-Secretase BACE1 in Alzheimer's Disease. Biol Psychiatry. 2020 Feb 13; PubMed.
IPMC
University Munich
η-secretase-derived APP fragments inhibit neuronal activity
In the paper presenting Aη peptides for the first time (Willem et al., 2015), we discussed already the discrepancy commented on by Frédéric Checler, and wrote therein: “Although Aη may be involved in the modulation of neuronal activity and synaptic plasticity, the differential bioactivity of recombinant Aη-α and Aη-bη is currently unclear. One may speculate that the longer Aη-α is more stable, probably due to unknown post-translational modifications. This would be consistent with our observation that, in contrast to cell-produced Aη-bη, 100 nM of synthetic Aη-bη inhibits LTP.” As a matter of fact, we continued to clarify this weakness of our finding and improved the synthesis to yield better quality of recombinant protein. Last year, in Mensch et al. (2021), we published the data obtained with the comparison of synthetic and recombinant peptides of the two species, which lead us to the conclusion that the N-terminal part of Aη is harboring the activity responsible for LTP impairment.
The important message of the Mensch et al. paper is that the C-terminal elongation of Aη-α, the part overlapping with the N-terminus of Abη, is not relevant for the regulatory function of Aη on LTP, but that the active site is located within the N-terminal domain common to both the short and long forms of the Aη peptide.
As such, we propose a competition of the amounts of different APP processing products, because several enzymes compete for shedding. Also, as Aη peptides can modulate LTP within the very low nanomolar concentration range, we propose a role of these peptides in a physiological setting.
In contrast, in AD, while we believe that the predominant aggregation and accumulation of Abη is triggering many secondary effects, we also observed higher levels of CTF-η at dystrophic neurites (Willem et al., 2015). However, we currently still do not know what could be the consequence of this increased CTF-η level in AD brains.
References:
Willem M, Tahirovic S, Busche MA, Ovsepian SV, Chafai M, Kootar S, Hornburg D, Evans LD, Moore S, Daria A, Hampel H, Müller V, Giudici C, Nuscher B, Wenninger-Weinzierl A, Kremmer E, Heneka MT, Thal DR, Giedraitis V, Lannfelt L, Müller U, Livesey FJ, Meissner F, Herms J, Konnerth A, Marie H, Haass C. η-Secretase processing of APP inhibits neuronal activity in the hippocampus. Nature. 2015 Oct 15;526(7573):443-7. Epub 2015 Aug 31 PubMed.
Mensch M, Dunot J, Yishan SM, Harris SS, Blistein A, Avdiu A, Pousinha PA, Giudici C, Busche MA, Jedlicka P, Willem M, Marie H. Aη-α and Aη-β peptides impair LTP ex vivo within the low nanomolar range and impact neuronal activity in vivo. Alzheimers Res Ther. 2021 Jul 8;13(1):125. PubMed.
Institut de Pharmacologie Moléculaire et Cellulaire
In their response to my comment, Marie and Willem wrote “In the paper presenting Aη peptides for the first time (Willem et al., 2015), we discussed already the discrepancy commented on by Frédéric Checler and wrote therein: ‘Although Aη may be involved in the modulation of neuronal activity and synaptic plasticity, the differential bioactivity of recombinant Aη-α and Aη-β is currently unclear.’” In my comment, although I indeed discussed the genetic grounds suggesting the unlikeliness of initial data reported in the Willem et al. Nature paper, the main discrepancy underlined was not about the distinct bioactivity of Aη-α and Aη-β but about the irreproducibility and discrepancy between data described in two papers by the same authors. Thus, Marie and Willem’s statement that “the differential bioactivity of recombinant Aη-α and Aη-β is currently unclear” should be read as “the differential bioactivity of Aη-α and Aη-β was wrong.” This is important information for the scientific community working in the field of APP processing and putative associated pathogenic fragments.
References:
Willem M, Tahirovic S, Busche MA, Ovsepian SV, Chafai M, Kootar S, Hornburg D, Evans LD, Moore S, Daria A, Hampel H, Müller V, Giudici C, Nuscher B, Wenninger-Weinzierl A, Kremmer E, Heneka MT, Thal DR, Giedraitis V, Lannfelt L, Müller U, Livesey FJ, Meissner F, Herms J, Konnerth A, Marie H, Haass C. η-Secretase processing of APP inhibits neuronal activity in the hippocampus. Nature. 2015 Oct 15;526(7573):443-7. Epub 2015 Aug 31 PubMed.
Make a Comment
To make a comment you must login or register.