In vivo near-infrared imaging of a FRET-based γ-secretase biosensor in the mouse cortex revealed that individual neurons differ widely in γ-secretase activity yet neurons with similar activity form spatial clusters: the activity of a neuron correlates positively, albeit modestly, with the average activity of its five nearest neighbors, and this correlation is abolished by the γ-secretase inhibitor DAPT, suggesting that γ-secretase activity is regulated cell non-autonomously. The mechanistic basis of this intercellular coordination is entirely unidentified. The problem is to determine which signal and pathway — secreted Aβ42 acting through product feedback inhibition, local oxygen concentration acting through HIF-1α, or other microenvironmental cues — synchronizes γ-secretase activity across neighboring neurons, whether the coordination is mediated by direct neuron-to-neuron communication or by glial cells, how the six possible γ-secretase complex isoforms contribute to the spatial pattern, and whether the clustering has functional consequences for neuronal survival, synaptic plasticity, or Alzheimer's disease progression. A rigorous demonstration that the reported correlation instead arises from an experimental or pharmacological artifact would equally resolve the source question.

Background

γ-Secretase is a four-component intramembrane aspartyl protease complex (presenilin PSEN1 or PSEN2 together with nicastrin, APH-1, and PEN-2) that cleaves the transmembrane stubs of more than 150 substrates, including the amyloid precursor protein (APP) C-terminal fragment C99, whose processing releases amyloid-β (Aβ) peptides that accumulate in Alzheimer's disease (AD) brains, and Notch receptors. Because rodents encode two presenilin isoforms (PSEN1, PSEN2) and three APH-1 isoforms (Aph1a, Aph1b, Aph1c), a single neuron can in principle express six distinct γ-secretase complexes with different activities, and genetic ablation of individual subunits produces non-equivalent phenotypes, from embryonic lethality (PSEN1, Aph1a) to viability (PSEN2, Aph1b/c). Work with genetically encoded Förster resonance energy transfer (FRET) reporters — biosensors in which cleavage of a C99-based substrate changes the ratio of donor to acceptor emission — established that γ-secretase activity is heterogeneous from cell to cell in cultured neurons, and single-cell immunoassays likewise showed that individual neurons secrete different amounts of Aβ. What remained unknown until recently was how γ-secretase activity is regulated in the intact brain in vivo. Hou et al. (2024) addressed this by delivering an adeno-associated virus expressing C99 720-670, a near-infrared (NIR) ratiometric FRET biosensor of γ-secretase, to the somatosensory cortex of living mice and imaging neuronal cell bodies through a cranial window by NIR confocal microscopy (one-photon imaging in the NIR range, chosen over standard multiphoton microscopy for lower phototoxicity and cost). They confirmed sensor validity in vivo by showing dose-dependent accumulation of endogenous APP C-terminal fragments and a 22% biosensor-ratio decrease after systemic administration of DAPT, a γ-secretase inhibitor. Their central unexplained observation is spatial: the 720/670 ratio of a neuron correlates positively with the mean ratio of its five nearest neighbors, forming clusters of neurons with similar activity, and this correlation is abolished by DAPT — implying that γ-secretase activity in one neuron can influence γ-secretase activity in neighboring neurons, a putative 'cell non-autonomous' regulation whose molecular mechanism the authors state remains elusive. Two candidate mechanisms have been proposed but not tested in this context: (i) secreted Aβ42 acting as an intercellular negative signal, based on the demonstration that Aβ42 exerts product feedback inhibition on γ-secretase and impairs downstream signaling (Zoltowska et al., 2024); and (ii) local oxygen concentration, based on findings that hypoxia-inducible factor-1α (HIF-1α), the transcription factor activated under hypoxia, binds and activates the γ-secretase complex and that hypoxia/hypoperfusion boosts Aβ production (Villa et al., 2014; Alexander et al., 2022). Whether the coordinating signal passes directly between neurons or through glial cells is untested — the published imaging used a neuron-specific promoter, so glia were not monitored — and the authors explicitly propose familial-AD PSEN1 knock-in and PSEN1 conditional-knockout mice as the next experiments to test the Aβ42-feedback hypothesis in vivo. The functional stakes are high because conditional γ-secretase loss in adult neurons causes age-dependent synaptic dysfunction, neuroinflammation, and neuronal loss, so spatially coordinated γ-secretase regulation could shape regional vulnerability in AD.

Problem Statement

Explain the cell non-autonomous spatial correlation of γ-secretase activity observed among neighboring neurons in the living mouse cortex: identify the molecular signal(s) and pathway(s) — for example secreted Aβ42 acting via product feedback inhibition, local oxygen concentration acting via HIF-1α, or other microenvironmental cues — that synchronize γ-secretase activity across neighboring neurons, determine whether this coordination reflects direct neuron-to-neuron communication or mediation by glial cells, establish how the composition of γ-secretase complexes (PSEN1 vs PSEN2, Aph1a/b/c isoforms) contributes to the observed cell-by-cell heterogeneity and spatial clustering, and establish whether the clustering is functionally consequential for neuronal survival, synaptic plasticity, or disease progression in vivo. A rigorous demonstration that the reported neighbor correlation arises from a non-biological artifact — such as differential accessibility of DAPT to cells, biosensor-expression covariates, or imaging-analysis confounds — would refute the source's cell non-autonomous interpretation. The four facets — the synchronizing signal and pathway, the direct-versus-glial routing of the coordination, the isoform-resolved contribution to the spatial pattern, and the functional consequence of the clustering — are independent sub-questions, each refutable on its own.

References

  1. Hou SS, Ikegawa Y, Kwon Y, Wieckiewicz N, Houser MCQ, Lundin B, Bacskai BJ, Berezovska O, Maesako M. Recording γ-secretase activity in living mouse brains. eLife. 2024;13:RP96848. doi:10.7554/eLife.96848. https://doi.org/10.7554/eLife.96848 (bioRxiv preprint: doi:10.1101/2024.01.31.578105, https://doi.org/10.1101/2024.01.31.578105)
  2. Yokomizo M, Sadek M, Williams E, Houser MCQ, Wieckiewicz N, Torres S, Berezovska O, Maesako M. A unique subpopulation of wild-type neurons recapitulating familial Alzheimer's disease phenotypes. Cell Death & Disease. 2025;16(1):604. doi:10.1038/s41419-025-07934-0. https://doi.org/10.1038/s41419-025-07934-0
  3. Williams E, Houser MCQ, Torres S, Wieckiewicz N, Sadek M, Yokomizo M, Maesako M. Endogenous γ-Secretase Is Linked to Phagocytic Activity in Microglial Cells. Sensors. 2025;25(11):3298. doi:10.3390/s25113298. https://doi.org/10.3390/s25113298
  4. Zoltowska KM, Das U, Lismont S, Enzlein T, Maesako M, Houser MCQ, Franco ML, Özcan B, Moreira DG, Karachentsev D, Becker A, Hopf C, Vilar M, Berezovska O, Mobley W, Chávez-Gutiérrez L. Alzheimer's disease linked Aβ42 exerts product feedback inhibition on γ-secretase impairing downstream cell signaling. eLife. 2024;12:RP90690. doi:10.7554/eLife.90690. https://doi.org/10.7554/eLife.90690
  5. Alexander C, Li T, Hattori Y, Chiu D, Frost GR, Jonas L, Liu C, Anderson CJ, Wong E, Park L, Iadecola C, Li YM. Hypoxia Inducible Factor-1α binds and activates γ-secretase for Aβ production under hypoxia and cerebral hypoperfusion. Molecular Psychiatry. 2022;27:4264–4273. doi:10.1038/s41380-022-01676-7. https://doi.org/10.1038/s41380-022-01676-7
  6. Villa JC, Chiu D, Brandes AH, Escorcia FE, Villa CH, Maguire WF, Hu C-J, de Stanchina E, Simon MC, Sisodia SS, Scheinberg DA, Li Y-M. Nontranscriptional role of Hif-1α in activation of γ-secretase and notch signaling in breast cancer. Cell Reports. 2014;8:1077–1092. doi:10.1016/j.celrep.2014.07.028. https://doi.org/10.1016/j.celrep.2014.07.028
  7. Houser MC, Hou SS, Perrin F, Turchyna Y, Bacskai BJ, Berezovska O, Maesako M. A Novel NIR-FRET Biosensor for Reporting PS/γ-Secretase Activity in Live Cells. Sensors. 2020;20(21):5980. doi:10.3390/s20215980. https://doi.org/10.3390/s20215980
  8. Maesako M, Sekula NM, Aristarkhova A, Feschenko P, Anderson LC, Berezovska O. Visualization of PS/γ-secretase activity in living cells. iScience. 2020;23:101139. doi:10.1016/j.isci.2020.101139. https://doi.org/10.1016/j.isci.2020.101139
  9. Serneels L, Dejaegere T, Craessaerts K, Horré K, Jorissen E, Tousseyn T, Hébert S, Coolen M, Martens G, Zwijsen A, Annaert W, Hartmann D, De Strooper B. Differential contribution of the three Aph1 genes to γ-secretase activity in vivo. PNAS. 2005;102:1719–1724. doi:10.1073/pnas.0408901102. https://doi.org/10.1073/pnas.0408901102
  10. Saura CA, Choi S-Y, Beglopoulos V, Malkani S, Zhang D, Rao BSS, Chattarji S, Kelleher RJ III, Kandel ER, Duff K, Kirkwood A, Shen J. Loss of presenilin function causes impairments of memory and synaptic plasticity followed by age-dependent neurodegeneration. Neuron. 2004;42:23–36. doi:10.1016/S0896-6273(04)00182-5. https://doi.org/10.1016/S0896-6273(04)00182-5