Abstract Body

Preventive therapies for AD are not yet available, and disease-modifying therapies targeting amyloid-β plaques in symptomatic stages of AD have only just been approved in the United States. These treatments are antibodies that are costly and require intravenous infusion. Ultimately, the development of orally bioavailable small molecule drugs would be ideal for the cost-effective and convenient treatment or prevention of AD. Small-molecule inhibitors of β-site amyloid precursor protein (APP)-cleaving enzyme 1 (BACE1) reduce the production of amyloid- β peptide and are among the most advanced orally bioavailable drug candidates for AD. However, to date all clinical trials of BACE inhibitors were either concluded without benefit or discontinued owing to futility or the occurrence of adverse effects. Adverse effects included early, mild cognitive impairment that was associated with all but one inhibitor; preliminary results suggest that the cognitive effects are non-progressive and reversible. These discontinuations have raised questions regarding the suitability of BACE1 as a drug target for AD. In this presentation, I will discuss the status of BACE inhibitors and suggest ways in which the results of the discontinued trials can inform the development of future clinical trials of BACE inhibitors and related secretase modulators as preventative therapies or used in combination with anti-amyloid immunotherapies. I will also discuss our efforts to determine a dosage of BACE inhibitor that sufficiently reduces Abeta production without cognitive impairment in wild-type and PDAPP amyloid pathology model mice. How cleavage of other BACE1 substrates correlates with Abeta lowering and cognition in mice will also be presented. The goal of these experiments is to test the possibility that low levels of BACE1 inhibition could avoid adverse effects while achieving efficacy for AD.

Abstract Body

APP endocytosis is a key link between aging and Alzheimer’s disease (AD) as normal neuronal aging increases the intracellular production of β-amyloid, due to an upregulation of the amyloid precursor protein endocytosis Moreover prominent alterations of the endocytic and lysosomal pathways have been identified in AD.

Here we describe a novel approach to inhibit b-secretase cleavage site of APP (BACE1)by blocking the substrate rather than the total enzymatic activity. This approach overcomes some of the limitations presented by BACE1 inhibition as he shares the processing of other vital non-APP substrates .We developed antibodies directed to the β-secretase cleavage site of APP (BBS1) to interfere with endocytic pathway of BACE1 . The antibodies are supposed to block the cleavage site of BACE1 and interfere with APP-BACE interaction, exploiting APP presence at the cell surface prior to internalization into early endosomes.

Administration of these antibodies to the cellular and transgenic mice models of AD resulted in a considerable decrease -56%- in intracellular Aβ load

Treatment of 3xTg-AD mice with BBS1 not only improves the cognitive functions and lowers the levels of total Aβ as expected, but induce a considerable effect in TAU pathology (80% reduction in phosphorylated tau levels and 56% reduction in tangles.) In addition, the treatment moderates the inflammatory response in mice and affects the expression levels of p53 and pGSK3β involved in neuronal apoptosis.

This approach demonstrates that inhibiting Aβ production via endocytic pathway without affecting BACE1 processing of other substrates is possible and suggest a clinical value for the treatment of AD

Aims

Unfavorable outcomes for Alzheimer disease trials suggest that the complexity of BACE1 inhibition extends beyond APP cleavage by BACE1 and Abeta generation. To date, inhibitors could reduce Abeta burden and plaque load, yet have failed to improve cognitive function and/or exhibited severe side effects. Work from our laboratory shows that the amyloidolytic activity of BACE1 can cleave Abeta species as an amyloid degrading enzyme. Here we tested the pharmacological inhibition of BACE1 to examine the effect on its amyloidolytic versus amyloidogenic activity.

Methods

We mechanistically assessed BACE1 activity in the presence of a series of BACE1 inhibitors under substrate and enzyme overexpression conditions, i.e. in APP, C99 and BACE1 transfected SH-SY5Y cells. Expression and cleavage products were verified by Western blotting and quantified by ELISA and/or multiplexed assays.

Results

When BACE1 is present in abundance, BACE1 inhibitors elevated Abeta40 and Abeta42 levels in a concentration dependent manner before Abeta34 levels were lowered and total Abeta generation was abandoned. Thus, in stable BACE1 overexpressing cells, pharmacological inhibition acts first upon the amyloidolytic enzymatic activity of BACE1. In APP-C99 overexpressing cells, in which the initial BACE1-mediated beta-site cleavage of APP is bypassed, primarily Abeta34 levels were decreased upon BACE1 inhibition. Since the inhibitor treatments affected the amyloidolytic activity of BACE1 first and foremost, existing BACE1 inhibitors may impair BACE1-mediated Abeta clearance at concentrations wherein amyloidogenic activity is still detectable.

Conclusions

The differential effect of BACE1 inhibitors on Abeta suggests that the substrate to enzyme ratio is the major factor affecting the balance between the amyloidogenic and amyloidolytic activities of BACE1. Since BACE1 levels are increased in the AD brain, a moderate slowdown of amyloid clearance could cause an increase in Abeta42 that may ultimately worsen symptoms.

Aims

Background. Endolysosomal abnormalities are early cytopathological changes observed in Alzheimer’s disease, well before amyloid plaques appear. Several of these defects are recapitulated in presenilin (PSEN) deficient cells and neurons of which the underlying derailed mechanism(s) remain debated. In this study, we aimed to determine (i) the chronology of events leading to endolysosomal dysfunction and (ii) the contribution of a derailed APP proteolysis.

Methods

Methods. We used CRISPR/Cas9 to generate cell lines deficient for both PSENs and PSENs/APP triple KO lines. This allowed us to re-introduce APP fragments using lentiviral technology and explore their contribution to endolysosomal dysfunctions. We used functional endolysosomal readouts and combined this with advanced imaging, including super-resolution (live) imaging and EM to evaluate endolysosomal dyshomeostasis, including their communication through membrane contact sites (MCSs).

Results

Results. When γ-secretase activity is chronically inhibited, the earliest defects include a lysosomal calcium imbalance and cholesterol accumulation, causing a subsequent collapse of endolysosomal compartments that affected normal endosome recycling and maturation as well as cargo sorting and turnover, including of APP. The decrease in lysosomal calcium content in PSEN deficient cells originated from defective calcium re-filling from the endoplasmic reticulum (ER). A concomitant accretion of cholesterol instigated a cascade leading to endolysosomal demise. Furthermore, transmission EM and live super-resolution imaging demonstrated that MCSs between lysosomes and ER are morphologically altered leading to decreased lysosomal motility and prolonged lingering at the ER. Whereas KO of APP significantly contributed to the restoration of endolysosomal homeostasis, reversely, defects re-appeared when introducing APP fragments.

Conclusions

Conclusions. Collectively, γ-secretase-dependent cytopathogenic changes suggest a surveillance role for presenilin/γ-secretase activity in lysosomal function. Herein we reveal a novel role for APP in the homeostatic regulation of inter-organellar communication.

Abstract Body

Presenilin/γ-secretase allosteric modulators: too late for precision medicine in Alzheimer’s disease ?

Two successful clinical phase III trials with anti-amyloid antibodies suggest that preventing amyloid plaque deposition may effectively prevent the entire disease process. Antibody therapy may, however, not be the optimal as we will discuss in the context of an evaluation of an earlier small molecule approach: the ‘gamma secretase modulator (GSM)’ drug class. Recent breakthroughs in understanding of the structure-function, mechanism of action, and the diverse biology of the γ-secretases show that they are responsible for the production of a spectrum of Aβ peptides including a minor population of long Aβ peptides that seeds amyloid plaque formation. The gamma secretase allosteric modulator (GAM) subclass of GSMs bind to the enzyme in such a manner that alters the size spectrum of Aβ peptides in favour of the shorter, non-amyloidogenic forms of the peptide. GAM have no quantitative effect on the proteolytic processing and signalling properties of other prominent γ-secretase substrates, such as Notch, although a complete analysis of all γ-secretase substrates is lacking. GAMs present, in theory, a precision medicine approach to the prevention of amyloid deposition: the accurate targeting of a single component of a complex cell biological signalling complex.

Aims

Alzheimer’s disease is characterized pathologically by cerebral deposition of 42-residue amyloid beta-peptide (Abeta42), proteolytically produced from amyloid precursor protein (APP) by beta- and gamma-secretases. Although mutations in APP and presenilin, the catalytic component of gamma-secretase, cause familial Alzheimer’s disease (FAD), a role for Abeta42 as the primary disease driver remains controversial. We aimed to elucidate the effects of FAD mutations on gamma-secretase processing and mechanism and on synaptic integrity.

Methods

METHODS: Effects of FAD mutations on all proteolytic steps of gamma-secretase cleavage of APP C99 substrate were quantified by mass spectrometry. The structure of active protease bound to substrate mimetic inhibitor was determined by cryoelectron microscopy (cryoEM). Effects of FAD mutations on the enzyme-substrate interaction were tested by Gaussian-accelerated molecular dynamics (GaMD) simulations and fluorescence lifetime imaging microscopy (FLIM) in cultured cells. Effects of FAD mutations on lifespan and synaptic integrity were tested in transgenic C. elegans lines that co-express C99 and PSEN1 in neurons.

Results

RESULTS: FAD mutations were consistently deficient in early proteolytic events, not later events that produce secreted Abeta peptides. CryoEM revealed that the substrate mimetic trapped gamma-secretase at the transition state for intramembrane proteolysis, and this structure closely aligns with activated enzyme-substrate complex captured by GaMD. In silico simulations and FLIM in cultured cells support stabilization by FAD mutations of enzyme-substrate complexes. Neuronal expression of C99 and/or presenilin-1 in C. elegans led to age-dependent synaptic loss only when one of the transgenes carried an FAD mutation. Designed mutations that stabilize the enzyme-substrate complex and block proteolysis likewise led to synaptic loss.

Conclusions

CONCLUSIONS: Collectively, these findings implicate the stalled process—not the released products—of gamma-secretase cleavage of substrates in synaptic degeneration caused by FAD mutations.

Abstract Body

Alzheimer’s disease (AD) is characterized by the accumulation and deposition of Abeta. Whereas Aβ is produced by amyloidogenic APP processing, APP processing along the competing non-amyloidogenic pathway results in the secretion of neurotrophic and synaptotrophic APPsalpha. Our previous studies in knockout mice revealed that physiologically APP family proteins are essential to mediate synapse formation, synaptic plasticity and cognition. Increasing evidence suggests that synaptic functions of the amyloid precursor protein (APP), may be carried out by its large secreted ectodomain (APPs). Indeed, brain-specific conditional knockout mice lacking APP and the related APLP2 showed a severe synaptic phenotype that is rescued by intracranial re-expression of APPsalpha. Towards moving these findings into a therapeutic context, we demonstrated that APPsalpha reduces plaque pathology in transgenic AD mice and rescues Aβ-dependent impairments leading to a restoration of spine density, synaptic plasticity and memory. Notably, we could recently show that APPsalpha also mitigates Tau-induced deficits in two models of tauopathy, by reducing GSK3b and CDK5 kinase activity. From a mechanistic and therapeutic perspective it is crucial to identify the minimal functional domain(s) of APPsalpha. Towards this end, we have developed a novel AAV vector that allows efficient in vivo expression of the short 16 amino acid long C-terminal CTa16 domain, that distinguishes APPsalpha from APPsbeta. Strikingly, intracranial AAV-CTa16 application potently restored LTP and spine density in conditional APP/APLP2 double knockout mice. Moreover, when applied in aged THY-Tau22 mice, AAV-CTa16 also rescued spine density deficits in the presence of established Tau pathology. Thus, APPsalpha and CTa16 belong to the few molecules exhibiting therapeutic effects in mice, both for Aβ- and Tau-dependent synaptic impairments.

Abstract Body

In Alzheimer disease, besides amyloid β-peptides, APP-derived C-terminal fragments, including the C99 generated upon β-secretase cleavage could well participate in the pathology. Thus, our laboratory demonstrated that C99 could alter lysosomal/autophagic process as well as mitochondrial health (1,2). Our data indicated that the C99-linked toxic phenotype was likely due to its propensity to aggregate and that this was potentiated upon γ-secretase inhibitor treatment. Of note, we documented the fact that C99 aggregates could accumulate in exosomal vesicles (3), thus supporting the view of an exosomal transfer of C99 aggregates that could support its propensity to spread. However, the possibility to visualize C99 neo-aggregates formation and to follow its cell-to-cell transfer in dishes as well as in vivo was lacking.

Here we describe novel probes aimed at following neo-aggregation of C99 based on bimolecular fluorescence complementation. We designed inducible C99 constructs harboring either the N-terminus or C-terminus parts of the Venus protein. Interestingly, only complementary interaction of C99 constructs could reconstitute a fully functional Venus and thus, yield fluorescence. We will present the full characterization of our novel inducible constructs by both biochemical, pharmacological and high-resolution microscopy and we will document their in situ localization and influence on the autophagic pathway.

1- Lauritzen I et al. (2016) Acta Neuropathologica, 132, 257-276.

2- Vaillant-Beuchot L et al. (2021) Acta Neuropathologica, 141, 39-65.

3- Lauritzen I et al. (2019) Translational Neurodegeneration, Dec 5;8:35. doi: 10.1186/s40035-019-0176-6