NUP intensity and coverage are decreased in App KI neurons.

(A) Representative confocal images nuclei of WT and App KI hippocampal neurons (DIV 7-28). Cells were co-labelled with antibodies against pan-NPCs (red), MAP2 (cyan), and Hoechst (blue). A heatmap for pan-NPC expression is included. Scale bar at 5 µm.

(B-C) Volumetric quantification of intensity and coverage of NPCs in WT and App KI nuclei (DIV 7-28). Bar graph is plotted as percent time-matched WT neurons for each condition across independent experiments (n=45 neurons/group across 3 replicates for DIV 7, 21, and 28; n=90 neurons/group across 6 replicates for DIV 14).

(D) Representative confocal images of brain sections in hippocampal CA1 pyramidal cell layer from WT and App KI mice at different ages (4, 7 and 13 months). Slices were labelled with antibodies against NPCs (red), Hoechst (blue). Scale bar at 20 µm. Inset image is a magnification of the area outlined by a white dotted box.

(E) Quantification of pan-NPC intensity from individual neuronal nuclei in (D) as outlined in methods. Group data plotted as percent change from average age-matched WT (4, 7 and 13 months.; n=120 neurons/group across 3 replicates).

(F) Zeiss AiryScan confocal images of hippocampal neurons (DIV 14) labelled with pan-NPCs antibodies (red) after linear deconvolution. Representative spot masks and nearest neighbour distances shown. Scale bars are 1 µm (left) and 0.5 µm (right).

(G) Quantification of total NPC puncta per nuclei from WT and App KI neurons from time-matched neuronal cultures (n=14 nuclei/group averaged across 2 replicates).

(H-I) Quantification of average distance between nearest neighbours (1, 3, 5 and 9 nearest neighbours) for individual NPC puncta. The average distance for each nearest neighbour is plotted as individual line graphs for WT and App KI neurons across time-matched cultures (DIV 7-21) and also in (I) where it is normalized against the average signal from DIV 7 WT neurons (n=14 nuclei/group across 2 replicates). Significance values in (I) represent change from DIV 7 for each genotype.

For all graphs values in blue represent WT and values in red represent App KI. Unless otherwise stated, all data are represented as mean ± SEM. Depending on distribution of data, significance testing between WT and App KI samples were done using Mann-Whitney U test (B, E), unpaired t-test (C, G, H,) or 2-way ANOVA (I). For all statistics significance is as follows: not significant (ns), < 0.05 (*), < 0.001 (**), < 0.0001 (***), < 0.0001 (****/@@@@). For complete statistical output, refer to Supp. Table 1-Fig. 1.

NUP98 and NUP107 are reduced in App KI neurons.

(A) Representative confocal images for WT and App KI neurons (DIV 7-21) immunolabeled with antibodies against NUP98 (green), MAP2 (cyan), and Hoechst (blue). Scale bar at 5 µm.

(B) Volumetric quantification of NUP98 intensity in WT and App KI nuclei (DIV 7-21). Bar graph is plotted as percent time-matched WT neurons for each condition across independent experiments (n=45 neurons/group across 3 replicates for DIV 7, and 21; n=75 neurons/group across 5 replicates for DIV 14).

(C) Same as in (A) except neurons are labelled with NUP107 (Magenta). Scale bar at 5 µm.

(D) Same as in (B) except quantifications are done for NUP107 (n=45 neurons/group across 3 replicates for DIV 7, and 21; n=75 neurons/group across 5 replicates for DIV 14)

(E) IHC of brain slices from App KI (2, 4, 7, 13 and 18 months) and age-matched WT mice. Images show hippocampal CA1 pyramidal layer neurons. Sections were labelled with antibodies against NUP98 (green), NUP107 (magenta) and Hoechst (blue). Scale bar at 10 µm.

(F-G) Quantification of NUP98 and NUP107 intensities from individual neurons shown in (E). Imaging parameters and methodology of cell selection are outlined in methods. Group data plotted as percent age-matched WT for each age group (2, 4, 13, and 18 months-n=90 neurons/group across 3 animals; 7 months-n=210 neurons/group across 7 animals).

(H) Western blots of purified forebrain nuclear extracts from 2- and 14-month-old WT and App KI animals. Blots were immunoassayed for antibodies against NUP98, NUP107, LAMIN-B1, and Histone H3. For fold change quantification, all bands were normalised against H3 and analysed with one sample t-test (N=5 for 2-months or N=4 for 14-months) paired experiments from WT and App KI animals). Dotted line on Y axis indicate a fold change of 1.0. For all graphs values in blue represent WT and values in red represent App KI.

For all graphs, data was reported as mean ± SEM. Depending on distribution of data, significance testing were performed using Mann-Whitney U test (B, D, F, G) or one sample t-test (2H). For all statistical significance is as follows: not significant (ns), < 0.05 (*), < 0.01 (**), < 0.001 (***), < 0.0001 (****). For complete statistical analysis, refer to Supp. Table 1-Fig. 2.

Increase in intracellular Aβ correlates with loss of NUPs.

(A) Representative ICC images of DIV 14 App KI cultured neurons stained with two Aβ antibodies MOAB2 (red) and 82E1 (red), MAP2 (cyan) and Hoechst (blue) showing intracellular Aβ accumulations. Scale bar at 5 µm.

(B) IHC confocal image of CA1 pyramidal cell region of WT and App KI 2-months mice stained with MOAB2 (red) and Hoechst (blue). Scale bar at 20 µm.

(C) Quantification of Aβ with MOAB2 and 82E1 antibodies in DIV 7-28 cocultured neurons. Values shown as percent change from time-matched WTs (n=45 cells/group across 3 replicates for MOAB2 (DIV 7, 21, and 28) for 82E1 (DIV 7-28); n=90 cells/group across 6 replicates for MOAB 2 (DIV 14).

(D) Experimental timeline for inhibition of APP cleavage using DAPT. Media from cocultures were partially replaced from DIV 7-10 to remove pre-existing Aβ, then DAPT was added (10 uM) every 24 h. from DIV 10-14.

(E) ICC of WT and App KI neurons incubated with DMSO (control) or DAPT stained for NPCs (red), MAP2 (cyan), and Hoechst (dark blue). Scale bar at 5 µm.

(F-H) Quantification of intensity (F), coverage (G) of NPCs and intracellular Aβ (H) in WT and App KI neurons exposed to DMSO and DAPT. Values shown as a percent change from time-matched WT neurons exposed to DMSO (n=45 cells/group across 3 replicates).

(I) Experimental timeline for addition of exogenous Aβ addition to WT cocultures. Different preparations of synthetic Aβ was added to cultures from DIV 10-14.

(J) ICC images of WT neurons exposed to monomeric, oligomeric and fibrillar Aβ stained for NPCs (red), MAP2 (cyan) and Hoechst (dark blue). Scale bar at 5 µm.

(K-M) Quantification of intensity (K) and coverage (L) of NPCs and intracellular Aβ (M) in WT and App KI cocultures exposed to synthetic Aβ. Values shown as a percent change from WT (n=45 cells/group across 3 replicates).

(N) Timeline for media swap experiment. Conditioned media derived from WT and App KI cocultures were swapped and incubated for 2 days (DIV 19-21) before ICC.

(O) Confocal images of WT and App KI neurons exposed to WT conditioned media (wtm) and App KI conditioned media (appm). Cells were co-stained with pan-NPCs (red), MAP2 (cyan), and Hoechst (dark blue). Scale bar at 5 µm.

(P-R) Quantification of intensity (P) and coverage (Q) of NPCs and intracellular Aβ intensity (R) in WT and App KI cultures exposed to wtm or appm. Values shown as a percent change from WT culture exposed to WT media (WT+wtm; n=45 cells/group across 3 replicates).

For all graphs values in blue represent WT and values in red represent App KI. All data are reported as mean ± SEM. Unless otherwise stated, all significance tests were conducted with Mann-Whitney U test (C) or 1-way ANOVA (Kruskal Wallis; F, H, K, L, M, P, Q, R. Tukey; G). For all statistics significance is as follows: not significant (ns), < 0.05 (*), < 0.01 (**), < 0.001 (***), < 0.0001 (****). For complete statistical profiles for each experiment, refer to Supp. Table 1-Fig. 3

Nucleocytoplasmic protein compartmentalization is impaired in App KI neurons.

(A) Schematic of FRAP experiment in transfected neurons expressing EGFP. EGFP is photobleached in the nucleus and recovery of fluorescence is tracked in the nucleus (magenta dotted line).

(B) Time course for EGFP FRAP experiment and imaging parameters.

(C) Representative static images with time stamp (t=0, 7 and 350 s) of WT and App KI neurons captured during FRAP. Black dotted line represents the photobleached ROI. Scale bar at 10 µm.

(D) FRAP recovery curve for EGFP in the nucleus. All data points are normalized against baseline values prior to photobleaching. Shaded region (yellow) signifies photobleaching period. Dotted vertical line indicates all time points post t = 105 s that are p< 0.05 between WT and App KI (n=8 neurons/group across 2 replicates).

(E) Schematic of FRAP experiment in transfected neurons expressing 4xEGFP. 4xEGFP is photobleached and fluorescence recovery is tracked in the nucleus (magenta dotted line).

(F) Time course for 4xEGFP FRAP experiment and imaging parameters. Image capture was performed at two frame rates: 1 frame/sec (t=35-120s) followed by 2 frames/min (t=121-3600 s).

(G) Representative static images with time stamps (t=0, 35, and 3600 s) of WT and App KI neurons captured during FRAP. White dotted line represents the photobleached region. Scale bar at 10 µm. Inset image showing over-saturated signal to highlight 4xEGFP nuclear accumulation. Scale bar at 10 µm.

(H) FRAP recovery curve for 4xEGFP in the nucleus. All data points are normalized against baseline values prior to photobleaching. Shaded region (yellow) signifies photobleaching period. Dotted vertical line indicates all time points post t = 67 s that are p< 0.05 between WT and App KI (n=10 neurons/group across 3 replicates).

(I) Schematic of FRAP experiment for transfected neurons expressing NLS-4xEGFP. Whole cell photobleaching was performed and fluorescence recovery tracked in the nucleus and cytoplasm (magenta dotted lines).

(J) Time course for NLS-4xEGFP FRAP experiment and imaging parameters.

(K) Representative static images with time stamps (t=0, 65 and 600 s) of WT and App KI neurons captured during FRAP. Images are intentionally oversaturated (for the nucleus) to show differences in cytoplasmic NLS-4xEGFP. Scale bar at 10 µm. Inset image shows NLS-4xEGFP signal in the nucleus at normal saturation levels to capture differences in the nucleus. Scale bar is 5 µm.

(L) Recovery curve for NLS-4xEGFP in the cytoplasm. All data points are normalized against baseline values prior to photobleaching. Shaded region (yellow) indicates photobleaching period. Dotted vertical line indicates all time points post t=289 s that are p< 0.05 between WT and App KI (n=10 neurons/group across 3 replicates).

(M) Quantification of nuclear NLS-4xEGFP post-photobleaching. All values are normalized against data point at t=65 s (first time point post-photobleaching). Dotted vertical line indicates all time points after t=70 s that are p<0.05 between WT and App KI (n=10 neurons over 3 replicates). For all graphs, values in blue represent WT and values in red represent App KI. All graphs are reported as mean values ± SEM. For all recovery curves, significance tests performed with Mann-Whitney U test. For complete statistical profiles for each experiment, refer to Supp. Table 1-Fig. 4.

The nuclear permeability barrier is compromised in App KI mice.

(A) Micrograph of 2- and 13-months old WT and App KI coronal forebrain sections immunoassayed with MOAB-2 (white). Scale bar is 500 µm.

(B) Confocal images of purified nuclei from 2- and 14-months WT and App KI mice incubated with different sized fluorescence-conjugated dextran. Images show nuclei extracted from 2- and 14-months old animals in the presence of 500 kDa Rhodamine-dextran (red), 70 kDa FITC-dextran (green) and Hoechst nuclear dye (blue). Scale bar at 20 µm. Magnified images show single nuclei. Scale bar at 5 µm.

(C-D) Quantification of nuclear intensities for Rhodamine-dextran and FITC-dextran. Group data values shown as percent WTs (n=300 nuclei/group across 3 replicates). For all graphs values in blue represent WT and values in red represent App KI. All data are represented as mean ± SEM. Significance tested by Mann-Whitney U test as follows: not significant (ns), < 0.05 (*), < 0.01 (**), < 0.001 (***), < 0.0001 (****). For complete statistical analysis, refer to Supp. Table 1-Fig. 5.

Importin-mediated active transport is impaired in App KI neurons.

(A) Schematic of FRAP experiment for neurons expressing NLS-4xEGFP. Partial photobleaching and recovery is performed on the nucleus to track NLS-mediated nuclear import as outlined by the magenta dotted lines.

(B) Time course for NLS-4xEGFP FRAP experiment and imaging parameters. Image capture was performed at two frame rates: 1 frame/sec (t=5-120s) followed by 2 frames/min (t=121-2760 s).

(C) Representative images with time stamp (t=0, 5, 60 and 2760 s) of WT and App KI neurons during FRAP. Scale bar at 10 µm.

(D) FRAP recovery curve for NLS-4xEGFP in the nucleus. All data points are normalized against baseline values prior to photobleaching. Shaded region (yellow) signifies photobleaching period. Dotted vertical line indicates all time points between t=6 s and t=1980 s are p< 0.05 between WT and App KI (n=10 neurons/group across 3 replicates: Mann-Whitney U Test).

(E) WT and App KI hippocampal neurons (DIV 7-28) showing Importin β1 (IMP-β1; red) in WT and App KI neurons. Cells were co-labelled with MAP2 (cyan), and Hoechst (blue). Scale bar at 5 µm.

(F-G) Average intensity (F) and coverage (G) of IMP-β1 in the nuclei. All values are reported as percent WT (n=45 neurons/groups per group across 3 replicates).

(H) WT and App KI hippocampal neurons (DIV 14) showing RanGAP1 (red) nuclear rim localization in WT and App KI neurons. A line profile of RanGAP1 signal from a single confocal plane is included for each genotype. Cells were co-labelled with MAP2 (cyan), and Hoechst (blue). Scale bar at 5 µm.

(I) Average RanGAP1 intensity in the nuclei as in (H). All values are reported as percent WT (n=45 neurons/groups per group across 3 replicates).

For all graphs values in blue represent WT and values in red represent App KI. All graphs are reported as mean values ± SEM. All significance testing were performed with Mann-Whitney U test: not significant (ns), < 0.05 (*), < 0.01 (**), < 0.001 (***), < 0.0001 (****). For complete statistical profiles for each experiment, refer to Supp. Table 1-Fig. 6.

App KI neurons are more susceptible to necroptosis due to loss of NUPs

(A) Representative live imaging bright field image of WT and App KI neurons exposed to TSZ cocktail containing 0 or 100 ng TNF-α. Cells stained with Hoechst (blue), propidium iodide (red), and FITC-Annexin V (green). Cells with red are necroptotic cells, cells with green signal are apoptotic cells, brightfield used to identify neurons for quantification. Scale bar at 50 µm.

(B) Quantification of necroptotic cell of WT and App KI neurons exposed to TSZ cocktail with 0 or 100 ng TNF-α in the presence of Necrostatin 1 stable (Nec1s; N=3 replicates; [#] WT TSZ 0 ng v. WT TSZ 100 ng; [@] App KI TSZ 0 ng v. App KI TSZ 100 ng; [*] WT TSZ 100 ng v. App KI TSZ 100 ng).

(C) Concentration curve of WT and App KI neurons exposed to increasing concentrations of TNF-α (0-100 ng) showing percent necroptotic cell death (N=3 replicates; [#] significance tests for WT TSZ 0 ng v. WT TSZ TNF-α; [@] significance tests for App KI TSZ 0 ng v. App KI TSZ TNF-α; [*] significance tests for WT TSZ TNF-α v. App KI TSZ TNF-α for individual concentrations).

(D) ICC of WT and App KI neurons exposed to TSZ cocktail with 0-100 ng TNF-α stained for pRIPK3 (magenta), pMLKL (green), MAP2 (cyan), and Hoechst (dark blue). Scale bar at 5 µm.

(E-F) Quantification of somatic pRIPK3 (E) and pMLKL (F). Values reported as percent change from WT-0 ng TNF-α (n= 45 neurons/group across 3 replicates; [#] significance tests for all groups v. WT TSZ 0 ng TNF-α; [*] significance tests for WT TSZ TNF-α v. App KI TSZ TNF-α samples at specific TNF-α concentrations).

(G) Quantification of percent necroptotic cell death for WT and App KI neurons exposed too TSZ in the presence of LMB (50 nM; 2 h.) or GppNHp (500 nM; 2 h.) (N=3 replicates; [#] significance tests for WT groups v. WT TSZ 100 ng TNF-α; [@] significance tests for all App KI groups v. App KI TSZ 100 ng TNF-α; [*] significance tests for WT TSZ TNFα + LMB/GppNHp v. App KI TSZ TNF-α + LMB/GppNHp).

(H) Graph showing percent change in cell death for WT and App KI neurons with LMB and GppNHp compared to their respective 100 ng TNF-α controls (N= 3 replicates; [*] WT TSZ TNFα + LMB/GppNHp v. App KI TSZ TNF-α + LMB/GppNHp). Significance tested by unpaired t-test.

(I) Representative pseudo-coloured confocal images and heatmap of neurons exposed to TSZ in the presence of LMB and GppNHp and labelled with pRIPK3 and pMLKL. Nuclei outlined in dotted magenta shapes Scale bar at 5 µm.

(J-K) Somatic quantifications of pRIPK3 (J) and pMLKL (K) from I (n= 45 neurons/group across 3 replicates; [#] significance tests for all groups v. TSZ TNF-α 0 ng; [*] significance tests for indicated comparisons between groups).

(L) Nucleus to cytoplasmic ratio for neurons exposed to LMB showing nuclear accumulation of pRIPK3 and pMLKL in WT neurons. Significance testing performed by Mann Whitney U test (n= 45 neurons/group across 3 replicates).

For all graphs, values in blue represent WT and values in red represent App KI. All graphs are reported as mean values ± SEM. Significance testing with 2-way ANOVA (Tukey; B, C, G) or 1-way ANOVA (Kruskal-Wallis; E, F, J, K) unless otherwise stated. For all statistics significance is as follows: not significant (ns), < 0.05 (*), < 0.01 (**), < 0.001 (***), < 0.0001 (****). For complete statistical profiles for each experiment, refer to Supp. Table 1-Fig. 7.

NPC dysfunction in App KI neurons

A diagram modelling the reduction of NUPs and potential loss of NPCs in App KI neurons contributing toward a breakdown of the permeability barrier and disruption of nucleocytoplasmic compartmentalization and protein transport. Localization of proteins < 60 kDa (magenta) or > 60 kDa, with (green) or without (yellow) a nuclear localization sequence (NLS) are altered in App KI nuclei.