Sans Presenilin-2, Lysosomes Struggle. Synapses, Memory Circuits Erode

Both presenilin-1 or the less-studied presenlin-2 can act as the catalytic subunit of γ-secretase, the enzyme responsible for churning out Aβ peptides. A new study, published November 29 in Nature Communications, zeros in on PSEN2’s unique contribution to amyloidosis and the synaptic havoc that unfolds in AD. Scientists led by Wim Annaert at KU Leuven in Belgium report that in an APP knock-in mouse, crippling PSEN2—either by saddling it with a familial AD mutation or by deleting it entirely—not only exacerbated amyloidosis, but also derailed synaptic signaling in hippocampal circuits critical for working memory. Profound lysosomal dysfunction was deemed the primary culprit underlying these synaptic snafus, in keeping with PSEN2 residing within endolysosomal compartments. The findings support the concept of the endolysosomal system as ground zero in AD.

  • In APPNL-G-F mice, removing or mutating PSEN2 exacerbated amyloid pathology.
  • It derailed synaptic signaling in the hippocampal CA3 region, hampering working memory.
  • Synaptic deficits attributed to disturbance in endolysosomal flow.

Scott Small of Columbia University in New York called the study “beautiful in its scientific elegance.” He wrote that it addresses two nested “where” questions about AD’s neurodegenerative trigger: Is it inside or outside the neuron, and, if inside, whereabouts? “By extensively and comprehensively investigating the effects of PSEN2, which together with APP, PSEN1, and SORL1, makes up  AD’s ‘causal quartet,’ the manuscript supports the following answers: Inside the neuron; and in the endosomal recycling pathway,” Small wrote.

Gunnar Gouras of Sweden’s Lund University wrote that the study offers numerous insights into how PSEN2 influences amyloidosis, endolysosomes, synaptic function, and memory. “Such comprehensive studies that encompass many diverse methods to provide new synaptic-cellular mechanistic insights are rare in our field,” he noted.

PSEN1 and PSEN2 are equally capable of unleashing Aβ peptides, but they do so within different cellular haunts. While PSEN1 prefers to bask on the cell surface and sometimes ventures into the early endosomes, PSEN2 contains a sorting sequence that strictly restricts it to late endosomes and lysosomes. As such, PSEN2 generates an intracellular pool of Aβ peptides, which skew toward longer, aggregation-prone varieties when familial AD mutations hobble its processive peptidase activity (Jun 2016 news).

To investigate the unique contribution of PSEN2 to amyloidosis and other aspects of AD pathogenesis, first author Anika Perdok and colleagues crossed APPNL-G-F knock-in mice with PSEN2 knockouts, or with mice expressing PSEN2 harboring the familial AD (FAD) mutation N141L. They hypothesized that knocking out PSEN2 would counteract amyloidosis, while FAD-PSEN2 would exacerbate it. What they found surprised them: both conditions aggravated amyloidosis.

Amyloid Aggravator. Amyloid plaques crowd the hippocampus of 6-month-old APP NL-G-F mice (top). In PSEN2-KOs (middle), plaques are larger and more diffuse, while in FAD-PSEN2 mice, they are smaller and more numerous (bottom). [Courtesy of Perdok et al., Nature Communications, 2024.]

Compared to APP-KI controls, both 9-month-old PSEN2-KO and FAD-PSEN2 mice had a heftier plaque burden, and more dystrophic neurites in the hippocampus and cortex. Plaques were larger and more diffuse in the knockouts than the mutants. In the former, APP processing fell entirely on the shoulders of wild-type PS1, skewing the peptide pool to shorter, secreted varieties. Annaert thinks this explains the shift toward looser and larger Aβ plaques in the absence of PSEN2. In contrast, expression of FAD-PSEN2 promoted a profusion of smaller, more compact plaques; Aβ peptides were longer, and a greater proportion of them were inside the cell, in keeping with PSEN2’s localization within endolysosomes.  

Behavioral deficits in APP-KI mice tend to be mild. Would tweaking either expression of function of PSEN2 change that? Yes, and in a specific way. While all three genotypes of mice performed similarly to wild-type animals on multiple behavioral and spatial memory tests, they faltered in tests of working memory. APP-KI had a subtle deficit, which was exacerbated markedly in PSEN2-KO and FAD-PSEN2 mice. Fittingly, the scientists found that PSEN2 was most highly expressed in pyramidal neurons in the CA3 region of the hippocampus. These neurons receive input from mossy fibers, a circuit that drives working memory.

The scientists then investigated the cellular, biochemical, and electrophysiological consequences of PSEN2 loss or dysfunction on synaptic signaling within hippocampal circuitry. While they found differences in how PSEN2-KO versus FAD-PSEN2 altered the mechanics of synaptic signaling, the result was essentially the same: a waning of synaptic plasticity in hippocampal circuits. Alterations in trafficking of synaptic vesicles, and a drop in surface levels of critical pre- and post-synaptic receptors, seemed to blame (image below).

Could changes in endolysosomal trafficking explain the synaptic breakdown in the PSEN2 mice? To investigate, Perdok and colleagues measured the accumulation of Aβ peptides, as well as APP β-C-terminal fragments in intracellular vesicles. A substrate for γ-secretase, β-CTF  accumulation has been tied to endolysosomal congestion (Jul 2023 news; Apr 2024 news). The scientists found that, relative to late endosomes and lysosomes in APP-KI hippocampal neurons, these compartments in PSEN2-KO mice had less Aβ, and more β-CTF. In PSEN2-FAD mice, both Aβ and β-CTF were elevated in these vesicular compartments.

The build-up of these APP products corresponded with a breakdown of endolysosomal function. The vesicles contained less calcium, a finding that jibes with Annaert’s recent report that excess β-CTF stems the flow of the ion from the endoplasmic reticulum into the lysosome (Apr 2024 news). The scientists also monitored the evolution of autophagosomes into autolysosomes, by tagging the autophagic membrane marker LC3 with two fluorescent tags—GFP and mCherry. In autophagosomes, the two combine to yield yellow fluorescence, but once merged with lysosomes, the acidic environs quenches GFP, leaving only red fluorescence (image below). In this way, they spotted a glut of autophagosomes and autolysosomes choking neurons in both PSEN2 models, indicating a slowdown in lysosomal digestion and trafficking (image below).

Constipated Lysosomes. In primary hippocampal neurons from APPNL-G-F mice (top), autophagosomes (pink) and autolysosomes (green) accumulate in PSEN2 knockouts (middle) and in PSEN mutants (bottom). [Courtesy of Perdok et al., Nature Communications, 2024.]

“These findings link the aberrant accumulation of APP metabolic fragments, due to either PSEN2KO or FADPSEN2, to impaired late endosomal/lysosomal and autophagic functions, which could in turn affect synaptic receptor recycling/trafficking underlying the observed synaptic defects,” the authors wrote.  

Annaert told Alzforum that the findings drive home the point that problems with lysosomal degradation and trafficking compromise synaptic signaling quickly, and that PSEN2 is important for maintaining lysosomal flow (model below). In the case of FAD-PSEN2, accumulation of toxic intracellular Aβ peptides, as well as β-CTF, likely contribute to the lysosomal congestion. When PSEN2 is missing entirely, a build-up of β-CTF is likely the prime instigator.

Without PSEN2. When PSEN2 is mutated or deleted, endolysosomal trafficking defects and Aβ accumulation derail the proper trafficking of synaptic vesicles and receptors (top). This impairs synaptic signaling and working memory (bottom left). Calcium influx into lysosomes slows, thwarting vesicular trafficking (bottom right). [Courtesy of Perdok et al., Nature Communications, 2024.]

Highlighting this dichotomy, Claudia Almeida of the Universidade Nova de Lisboa, in Portugal, wondered about the relative contributions of Aβ accumulation and PSEN2 function to endolysosomal and synaptic dysfunction. “It would be interesting to investigate whether improving endolysosomal function, as we did in aged neurons, recovers synapses,” she wrote. Almeida found that restoring the acid pH of lysosomes corrected age-related synaptic decline (Burrinha et al., 2023).—Jessica Shugart 

Comments

  1. This manuscript directly addresses two important, but related, questions that are actively, and sometimes heatedly, discussed in meetings. 

    1. The neurodegenerative process is the direct driver of brain dysfunction and the disease’s debilitating symptoms. So, is it AD’s extracellular pathology (amyloid plaques) or intracellular pathology (endolysosomal disruption) that is the primary driver of neurodegeneration? Is the trigger inside or outside the neuron?

    The paper provides additional evidence that PSEN2, like the other causal AD mutations in APP, PSEN1, and SORL1, causes intraneuronal endolysosomal pathology independent of extracellular amyloid pathology. It then elegantly explains the neurobiology of how this intracellular pathology leads to profound synaptic dysfunction, independent of extracellular pathology. This aspect of the manuscript is, therefore, generally supportive of the view that it is the intracellular pathology that might trigger the neurodegenerative process, and might explain why clearing plaques in patients has, thus far, shown little amelioration of neurodegeneration. I say, “thus far,” since I am hoping that for my patients we will see benefits on neurodegenerative readouts and stronger benefits on clinical readouts when the drugs are tested in earlier stages of disease.

    2. Even if endolysosomal disruptions are the trigger, the problem then is to reconcile that many neurodegenerative diseases target the same intracellular system, as we are becoming increasingly aware. The solution might be that neurodegenerative disorders differentially target separate components of the endolysosomal network, which after all is comprised of multiple molecularly and functionally distinct organelles and endowed with multiple trafficking routes. So, which specific component of the endolysosomal network is differentially targeted by AD?

    While having many effects, the paper localizes PSEN2’s main effect to a specific endosomal pathway required for the normal recycling of glutamate receptors to the post-synaptic neuronal surface. In fact, there are numerous endosomal recycling pathways, and so the paper can be even more precise: It is specifically the retromer-dependent recycling pathway that regulates glutamate receptor recycling. The idea that AD, compared to other neurodegenerative disorders, differentially targets this specific pathway is supported by other causal genes, notably, SORL1, and maybe even APP (see for example Chen et al., 2023).

    References:

    Chen XQ, Sawa M, Becker A, Karachentsev D, Zuo X, Rynearson KD, Tanzi RE, Mobley WC. Retromer Proteins Reduced in Down Syndrome and the Dp16 Model: Impact of APP Dose and Preclinical Studies of a γ-Secretase Modulator. Ann Neurol. 2023 Aug;94(2):245-258. Epub 2023 Apr 29 PubMed.

    View all comments by Scott Small

  2. This is another high-level study by the Annaert lab and collaborating groups in Leuven, providing numerous novel insights into how presenilin 2 (PSEN2) KO and a common familial AD PSEN2 mutation affect synapses/physiology, dystrophic neurites/plaques, endolysosomes, and behavior when crossed with, and compared to, the APP NL-G-F KI mouse model of AD. They describe intriguing differences between PSEN2 KO and the PSEN2 FAD mutation. Such comprehensive studies in experimental AD research that encompass many diverse methods to provide new synaptic-cellular mechanistic insights are rare in our field.

    Many questions arise from this excellent work. We are given intriguing insights into the synaptic functions of PSEN2. Remarkably, PSEN2 KO worsens amyloid pathology in the APP KI mice, which is ascribed to the increased ability of PSEN1 to secrete Aβ in the setting of PSEN2 KO. The PSEN2 FAD mutation causes elevated intracellular Aβ. PSEN2 KO is not clearly linked with human AD, while FAD PSEN1 and 2 mutations of course both induce AD pathogenesis. It remains unclear why PSEN1 and 2 mutations seem to similarly induce AD pathology when they have different effects on intra- and extra-pools of Aβ, although they both do lead to longer Aβ peptides.

    View all comments by Gunnar Gouras

  3. The PSEN2 mutation is associated with familial early onset Alzheimer’s disease and is predicted to be pathogenic, supported by evidence from transgenic animals. Previous work from Win Annaert nicely tied PSEN2 activity to late endosomes and production of intracellular Aβ42. These data relate to our initial observations that late endosomes become dysfunctional due to accumulation of Aβ42 in endosomes (Almeida et al. 2006). 

    In this latest excellent paper, the Annaert and Gutierrez labs show that crossing a knock-in of the most common PSEN2 mutation, N141K, with a KI of three human APP familial mutations aggravates the AD-like phenotype, proving causality. The mice recapitulate aggravated amyloidosis and memory deficits, likely due to the PSEN2 loss of function in synapses and the endolysosomal system, as the PSEN2 KO phenotype is similar. 

    One wonders about the relative contribution of the PSEN2 mutation to endolysosome and synapse dysfunction, whether via a direct effect of PSEN2 function or via Aβ production. It would be interesting to investigate whether improving endolysosomal function, as we did in aged neurons (Burrinha et al. 2023), recovers synapses.

    This excellent manuscript builds on the proposal by us and others that rescuing the endolysosomal system would be an exciting therapeutic target with the potential to treat Alzheimer’s disease—at least for carriers of PSEN2 mutations. Further work should be undertaken in this direction.

    References:

    Almeida CG, Takahashi RH, Gouras GK. Beta-amyloid accumulation impairs multivesicular body sorting by inhibiting the ubiquitin-proteasome system. J Neurosci. 2006 Apr 19;26(16):4277-88. PubMed.

    Burrinha T, Cunha C, Hall MJ, Lopes-da-Silva M, Seabra MC, Guimas Almeida C. Deacidification of endolysosomes by neuronal aging drives synapse loss. Traffic. 2023 Aug;24(8):334-354. Epub 2023 May 23 PubMed.

    View all comments by Claudia Almeida

  4. In this study, the authors examined the pathological significance of PSEN2 in a mouse model of AD: APP knock-in mice, in which Aβ plaques accumulate, were crossed with a PSEN2 knockout or with mice that had the PSEN2 familial AD mutation, N141I, knocked in, to examine effects on Aβ production and Aβ plaque accumulation.

    Interestingly, the deposition of Aβ plaques was exacerbated in PSEN2 N141I knock-in and PSEN2 knockout mice, indicating that inhibition of γ-secretase activity of PSEN2 leads to worsening Aβ pathology. At the same time, synaptic, endolysosomal, and behavioral abnormalities were observed in these mice, indicating that partial inhibition of PSEN2 exacerbates AD pathology.

    The study elucidated that various functional abnormalities of PSEN are involved in AD pathophysiology. Moreover, this is an important paper in that it demonstrates the unprecedented concept that changes in PSEN2 expression may affect the process of AD pathogenesis in vivo.

    View all comments by Taisuke Tomita

Make a Comment

To make a comment you must login or register.

References

Mutations Citations

  1. PSEN2 N141I

News Citations

  1. Too Basic: APP β-CTF's YENTPY Motif Binds Proton Pump, Thwarts Lysosomes
  2. APP C-Terminal Fragments Stifle Calcium Flow from ER to Lysosomes

Paper Citations

  1. Burrinha T, Cunha C, Hall MJ, Lopes-da-Silva M, Seabra MC, Guimas Almeida C. Deacidification of endolysosomes by neuronal aging drives synapse loss. Traffic. 2023 Aug;24(8):334-354. Epub 2023 May 23 PubMed.

External Citations

  1. Jun 2016 news

Further Reading

Primary Papers

  1. Perdok A, Van Acker ZP, Vrancx C, Sannerud R, Vorsters I, Verrengia A, Callaerts-Végh Z, Creemers E, Gutiérrez Fernández S, D'hauw B, Serneels L, Wierda K, Chávez-Gutiérrez L, Annaert W. Altered expression of Presenilin2 impacts endolysosomal homeostasis and synapse function in Alzheimer's disease-relevant brain circuits. Nat Commun. 2024 Nov 29;15(1):10412. PubMed.

Annotate

To make an annotation you must Login or Register.