Dataset overview

This dataset contains the underlying data files associated with the figures in the manuscript. The main data package is organized inside Figures.zip which unzips to theFigures folder. Within this folder, each subfolder corresponds to one main figure or a grouped set of related main and supplementary figures.

The dataset includes raw image files, GraphPad Prism analysis files, SPSS files, MATLAB code, behavioral analysis files, raw CSV files, and supporting image-analysis outputs. Raw CSV files are included wherever applicable to provide the numerical values used for graph generation and statistical analysis in an open, unformatted, and reusable format.

General file organization

The primary folder is:

Figures/

Within Figures/, the data are organized by figure:

Figure 1/
Figure 2/
Figure 3, S2, S3/
Figure 4/
Figure 5/
Figure 6, S4/
Figure 7, S5/
Figure 8/

Each figure folder contains some or all of the following file types:

• Raw image files, including .png, .tif, and raw immunoblot or immunostaining source images.
• GraphPad Prism files, where statistical analyses and graph generation were performed.
• Raw CSV files containing the numerical values used for graphs and statistical analyses.
• SPSS files for correlation analyses, where applicable.
• MATLAB code for image analysis, where applicable.
• Adobe Illustrator or Photoshop files when these files were used for figure assembly or merged image visualization.

General analysis notes

GraphPad Prism was used for graph generation and statistical analysis unless otherwise specified. For grouped comparisons, values were organized by experimental or diagnostic group in Prism and analyzed using the statistical test indicated in the corresponding Prism file.

Raw CSV files are provided in addition to Prism, SPSS, or other software-specific files so that the data can be accessed without proprietary software.

Immunoblot densitometry was quantified using Fiji. Raw immunoblot images were opened in Fiji, bands corresponding to each protein of interest were selected, and densitometry values were measured. The resulting values were exported and used for graph generation and statistical analysis in GraphPad Prism.

For RT-PCR analyses, GAPDH was used as the normalization control. For each sample, delta Ct was calculated as:

delta Ct = target gene Ct - GAPDH Ct

Delta delta Ct was calculated by comparing each sample or group to the appropriate control group:

delta delta Ct = delta Ct of sample - mean delta Ct of control group

Fold change was calculated using:

fold change = 2^(-delta delta Ct)

For all RT-PCR-related CSV files in this dataset, the reported fold change values correspond to 2^(-delta delta Ct) values.

Human CSF sample and de-identification note

All human CSF samples were received as de-identified specimens from Cleveland Clinic and Johns Hopkins Medicine. No direct participant identifiers are included in these files. The datasets include only limited, non-identifiable information required for analysis, including diagnostic group and clinically reported CSF biomarker values where available.

Figure 1 data files

Overview

The Figure 1 folder contains the underlying numerical data and analysis files used to generate Figure 1A, 1B, 1C, 1D, 1E, and 1F. The GraphPad Prism analyses for grouped comparisons are included in:

Figure 1/Figure 1.prism

Raw CSV files containing the numerical values used for these analyses are included in:

Figure 1/Raw CSV files

These CSV files provide the underlying tabular data used to generate the corresponding Prism and SPSS analyses and are included to support accessibility, reuse, and downstream analysis.

Figure 1A and Figure 1D: PAR ELISA data

The Prism file Figure 1.prism contains PAR ELISA values measured in CSF samples. Samples were received as de-identified specimens from Cleveland Clinic and Johns Hopkins Medicine.

Data were organized in Prism by diagnostic group. PAR ELISA values were pasted into separate groups corresponding to Control, MCI, and AD. Group comparisons were performed using one-way ANOVA.

Data organization in Prism:

Control: PAR ELISA values from samples classified as Control.
MCI: PAR ELISA values from samples classified as MCI.
AD: PAR ELISA values from samples classified as AD.

Figure panels supported by this file:

Figure 1A: PAR ELISA values from the Johns Hopkins Medicine cohort, grouped by diagnostic category.
Figure 1D: PAR ELISA values from the Cleveland Clinic cohort, grouped by diagnostic category.

Figure 1B and Figure 1E: CSF Aβ42/Aβ40 ratio data

The Prism file Figure 1.prism contains the clinically reported CSF Aβ42/Aβ40 ratio values. These values were obtained from de-identified clinical information provided by Cleveland Clinic and Johns Hopkins Medicine.

Data were organized in Prism by diagnostic group. Aβ42/Aβ40 ratio values were pasted into separate groups corresponding to Control, MCI, and AD. Group comparisons were performed using one-way ANOVA.

Data organization in Prism:

Control: Aβ42/Aβ40 ratio values from samples classified as Control.
MCI: Aβ42/Aβ40 ratio values from samples classified as MCI.
AD: Aβ42/Aβ40 ratio values from samples classified as AD.

Figure panels supported by this file:

Figure 1B: CSF Aβ42/Aβ40 ratio values from the Johns Hopkins Medicine cohort, grouped by diagnostic category.
Figure 1E: CSF Aβ42/Aβ40 ratio values from the Cleveland Clinic cohort, grouped by diagnostic category.

Figure 1C and Figure 1F: SPSS correlation analysis files

Figure 1C and Figure 1F were analyzed using SPSS. Separate SPSS files are provided for the Johns Hopkins Medicine and Cleveland Clinic datasets. For each analysis, files are provided in .sav and .spv formats. The .sav files contain the SPSS dataset, and the .spv files contain the corresponding SPSS output and graph.

SPSS files are labeled using the following format:

1C_[name of the analysis]_[source institution]
1F_[name of the analysis]_[source institution]

Data were entered into SPSS in tabular format. Each row represents one de-identified sample. The values used for the correlation analysis were pasted into SPSS along with the corresponding diagnostic group assignment.

Diagnostic group coding:

1 = Control
2 = MCI
3 = AD

Correlation analyses were performed in SPSS using the pasted values for each source institution. The analysis was used to assess the relationship between PAR ELISA values and clinically reported CSF Aβ42/Aβ40 ratio values.

Scatter plots were generated in SPSS. Each dot represents one de-identified sample. Dot colors were assigned according to the diagnostic group code.

Figure panels supported by these files:

Figure 1C: Correlation analysis for the Johns Hopkins Medicine cohort.
Figure 1F: Correlation analysis for the Cleveland Clinic cohort.

Figure 2 and Supplementary Figure S1 data files

Overview

Figure 2 and Supplementary Figure S1 are related and are organized together in the Figure 2 folder. This folder contains raw image files, raw numerical data files, and the GraphPad Prism analysis file used to generate Figure 2 and Supplementary Figure S1.

The Prism file is named:

Figure 2/Figure 2, S1.prism

Raw CSV files containing numerical values used for the graphs are included in:

Figure 2/Raw CSV files

Raw western blot images

Raw western blot images are organized into subfolders corresponding to the relevant figure panel number. Folder names follow the figure panel format:

2[letter]
S1[letter]

The raw western blot image folders include:

2A
S1A
S1B
S1E

Each folder contains the raw western blot image files used for the corresponding figure panel.

Except this folder

S1I- Compiled western blot image in png format- as it was developed using an x-ray film

Raw immunostaining images

Raw immunostaining images are organized into subfolders corresponding to the relevant figure panel number. These folders contain the raw image files used for the immunostaining panels in Figure 2 and Supplementary Figure S1.

The raw immunostaining image folders include:

2D
S1G

Raw CSV files

Raw CSV files contain the numerical data used to generate each graph in Figure 2 and Supplementary Figure S1. CSV files are labeled according to the corresponding figure panel number using the format:

2[letter]
S1[letter]

When densitometry data from the same western blot image are used to generate more than one graph, the related quantification files are grouped together. For example, if quantification for PAR and γH2AX comes from the same Figure 2A western blot image, the densitometry data used for Figure 2B and Figure 2C are organized together as:

2B, 2C

This organization indicates that the related graph data were derived from the same western blot source image but were quantified as separate outcome measures.

Figure 3, Supplementary Figure S2, and Supplementary Figure S3 data files

Overview

Figure 3, Supplementary Figure S2, and Supplementary Figure S3 are organized together because the experiments and analyses are related. The folder Figure 3, S2, S3 contains raw image files, raw numerical data files, and the Prism analysis file used to generate these figure panels.

The Prism file is named:

Figure 3, S2, S3/Figure 3, S2, S3.prism

Raw CSV files are included in:

Figure 3, S2, S3/Raw CSV files

Supplementary Figure S2A: Breeding scheme

The breeding scheme image is included in:

Figure 3, S2, S3/S2A

This file shows the breeding strategy used to generate the experimental mouse genotypes used for the Figure 3-related analyses.

Supplementary Figure S2B: Agarose gel genotyping image

The raw agarose gel genotyping image is included as a .tif file in:

Figure 3, S2, S3/S2B

This folder contains the source genotyping image used to identify the mouse genotypes included in the study. The genotypes represented include:

WT
5xFAD
PARP1 knockout
5xFAD-PARP1 knockout

Supplementary Figure S2C: Immunoblot genotype validation

The raw immunoblot image showing genotype validation is included in:

Figure 3, S2, S3/S2C

This immunoblot shows protein-level validation of the genotype groups used in the study. The genotypes represented include:

WT
5xFAD
PARP1 knockout
5xFAD-PARP1 knockout

Supplementary Figure S2D: RT-PCR data

Raw RT-PCR data for Supplementary Figure S2D are included in the corresponding CSV files in:

Figure 3, S2, S3/Raw CSV files

These data were used to assess genotype-related transcript levels. The RT-PCR values used for analysis are also included in:

Figure 3, S2, S3/Figure 3, S2, S3.prism

The values reported in the RT-PCR CSV files are 2^(-delta delta Ct) fold change values.

Figure 3A: ThioS plaque immunostaining images

Raw ThioS plaque immunostaining images are included in:

Figure 3, S2, S3/3A

This folder contains the raw immunostaining image files used for Figure 3A.

Figure 3B: ThioS plaque quantification

The numerical quantification data corresponding to Figure 3A are provided as CSV files labeled 3B in:

Figure 3, S2, S3/Raw CSV files

The 3B CSV files contain raw numerical values used to quantify ThioS-positive plaque burden or related plaque measurements shown in Figure 3B. These values are also included in Figure 3, S2, S3.prism.

Figure 3C: Aβ40 serial extraction data

The CSV files for Figure 3C contain Aβ40 measurements from serially extracted brain fractions. The extraction fractions include:

TBS
TBSX
70% formic acid

These files contain raw numerical values used to compare Aβ40 levels across the relevant genotype groups and extraction fractions. The same values are included in Figure 3, S2, S3.prism.

Figure 3D: Aβ42 serial extraction data

The CSV files for Figure 3D contain Aβ42 measurements from serially extracted brain fractions. The extraction fractions include:

TBS
TBSX
70% formic acid

These files contain raw numerical values used to compare Aβ42 levels across the relevant genotype groups and extraction fractions. The same values are included in Figure 3, S2, S3.prism.

Supplementary Figure S3A: Immunoblot source files

The immunoblot source files for Supplementary Figure S3A are included in:

Figure 3, S2, S3/S3A

This folder contains the raw immunoblot image files used for Supplementary Figure S3A.

Supplementary Figure S3B and Supplementary Figure S3C: Immunoblot quantification

The quantification data for Supplementary Figure S3B and Supplementary Figure S3C are included in:

Figure 3, S2, S3/Figure 3, S2, S3.prism

The same numerical values are also provided as CSV files labeled S3B, S3C in:

Figure 3, S2, S3/Raw CSV files

Figure 4 data files

Overview

The Figure 4 folder contains raw immunostaining image files, raw numerical data files, MATLAB code, and the Prism analysis file used to generate the Figure 4 panels. Raw image files are organized into figure panel-specific folders. Raw CSV files containing numerical values used for the graphs are included in:

Figure 4/Raw CSV Files

Raw immunostaining images

Raw immunostaining image files are provided as .png and/or .tif files in their respective figure panel folders.

The raw immunostaining image folders include:

4A
4C
4E
4G

Raw CSV files

Raw CSV files are labeled according to the corresponding figure panel number and contain raw numerical data used for graph generation and statistical analysis.

The raw CSV files include data for:

4B
4D
4E
4F
4H

Figure 4B: PSD95 density around Aβ plaques

Figure 4B contains PSD95 density measurements around Aβ plaques generated from the raw immunostaining images shown in Figure 4A. These data were quantified using the MATLAB script:

MATLAB CODE FOR 4B.m

The MATLAB script identifies Aβ-positive plaque regions, expands the plaque boundary, masks the Aβ signal from the PSD95 channel, and calculates PSD95-positive area or density within the defined peri-plaque region.

The input images were paired using image file name flags for Aβ and PSD95 channels. The script prompts the user to select the input image directory, output directory, Aβ image flag, and PSD95 image flag. The output includes quantified PSD95 measurements and processed mask images generated during the analysis.

The MATLAB analysis pipeline for Figure 4B was as follows:

1. Raw Aβ and PSD95 immunostaining images were placed in the same input folder.
2. Aβ and PSD95 images were identified using user-defined filename flags.
3. The PSD95 signal was extracted from the appropriate image channel.
4. Aβ-positive plaque regions were thresholded using an automated thresholding approach.
5. The Aβ plaque mask was dilated to define the peri-plaque region.
6. The Aβ signal was excluded from the PSD95 channel to measure PSD95 signal surrounding, rather than within, the plaque.
7. PSD95-positive area or density was calculated within the defined peri-plaque region.
8. Quantified values were exported as 4B.csv for graph generation and statistical analysis.

The raw numerical values for Figure 4B are provided in:

Figure 4/Raw CSV Files

Figure 4D and Figure 4H: Percent area quantification

Figure 4D and Figure 4H contain percent area quantification data generated from the corresponding raw immunostaining images. Percent area was quantified using an ImageJ-based analysis pipeline.

The ImageJ analysis pipeline for percent area quantification was as follows:

1. Raw immunostaining images were opened in ImageJ.
2. Images were converted to the appropriate grayscale or single-channel format for analysis.
3. A consistent threshold was applied across images within each analysis to identify positive immunostaining signal.
4. The threshold-positive area was measured within the defined ROI.
5. Percent area was calculated as:

percent area = threshold-positive area / total ROI area × 100

1. The resulting values were exported and used for graph generation and statistical analysis.

The raw percent area values for Figure 4D and Figure 4H are provided in:

Figure 4/Raw CSV Files

Figure 4E: Quantification data

Raw numerical values used to generate Figure 4E are provided as CSV files in:

Figure 4/Raw CSV Files

These files contain the quantification values used for graph generation and statistical analysis in Prism.

Figure 4F: Cell count data

Figure 4F contains cell count data generated using Stereo Investigator software. Counts were performed using the Optical Fractionator workflow.

The counting output included the estimated population using the user-defined section thickness and the total number of markers counted. These values were exported and used for graph generation and statistical analysis.

Stereo Investigator output fields included:

Estimated population using user-defined section thickness: Estimated total population calculated by the Optical Fractionator workflow.
Total markers counted: Total number of counted markers used to generate the population estimate.

The raw numerical values for Figure 4F are provided in:

Figure 4/Raw CSV Files

Figure 5 data files

Overview

The Figure 5 folder contains raw immunostaining image files, raw numerical data files, and the Prism analysis file used to generate the Figure 5 panels. Raw image files are organized into figure panel-specific folders and genotype-specific subfolders. Raw CSV files containing numerical values used for the graphs are included in:

Figure 5/Raw CSV Files

The Prism file is named:

Figure 5/Figure 5.prism

Figure 5A: GFAP and IBA1 immunostaining images

Raw immunostaining images for Figure 5A are organized into two main subfolders:

Figure 5/5A/GFAP
Figure 5/5A/IBA1

The GFAP image folder contains four genotype-specific subfolders:

WT
PARP1ko
FAD
FADPARP1KO

Each genotype-specific folder contains the corresponding GFAP, DAPI, and Aβ channel images as .png files. Each folder also includes a Photoshop MERGED .psd file containing the merged image used for visualization.

The IBA1 image folder contains four genotype-specific subfolders:

WT
PARP1ko
FAD
FADPARP1KO

Each genotype-specific folder contains the corresponding IBA1, DAPI, and Aβ channel images as .png files. Each folder also includes a Photoshop MERGED .psd file containing the merged image used for visualization.

Figure 5B and Figure 5C: GFAP and IBA1 percent area quantification

Raw CSV files containing the numerical values used for Figure 5B and Figure 5C are included in:

Figure 5/Raw CSV Files

The CSV files are labeled according to the corresponding figure panel:

5B: GFAP percent area quantification
5C: IBA1 percent area quantification

The percent area values were generated from the corresponding raw immunostaining images using a Fiji-based image analysis pipeline.

The Fiji analysis pipeline for percent area quantification was as follows:

1. Raw immunostaining images were opened in Fiji.
2. The relevant channel was selected for analysis.
3. Images were converted to the appropriate grayscale or single-channel format.
4. A consistent threshold was applied across images within each analysis to identify positive GFAP or IBA1 immunostaining signal.
5. The threshold-positive area was measured within the defined ROI.
6. Percent area was calculated as:

percent area = threshold-positive area / total ROI area × 100

1. The resulting percent area values were exported as CSV files and used for graph generation and statistical analysis in Prism.

Figures 5D-5H: RT-PCR fold change data

Figures 5D, 5E, 5F, 5G, and 5H contain RT-PCR data analyzed as fold change values. Numerical values for these panels are included in:

Figure 5/Raw CSV Files
Figure 5/Figure 5.prism

The values reported in the RT-PCR CSV files are 2^(-delta delta Ct) fold change values.

Figure 6 and Supplementary Figure S4 data files

Overview

Figure 6 and Supplementary Figure S4 contain behavioral testing data and are organized together because the panels are related. The Figure 6, S4 folder contains behavior track plots, raw numerical data files, and the Prism analysis file used to generate the behavior graphs.

Raw CSV files are included in:

Figure 6, S4/Raw CSV Files

Behavior track plots

Behavior track plots for the Morris water maze and Y-maze were copied directly from ANY-maze software into an Adobe Illustrator file. Open field track plots were copied directly from PAS software into an Adobe Illustrator file. Because these track plots were copied directly from the behavior software into Illustrator, no separate source image version is available.

Morris water maze data

The Morris water maze data are included in CSV files labeled:

6B, S4A, S4B (Morris water maze)

These files contain raw behavioral measurements used for the Morris water maze analyses.

Figure panels supported by these files:

Figure 6B: Escape latency.
Supplementary Figure S4A: Distance traveled in the target quadrant.
Supplementary Figure S4B: Platform entries.

Y-maze data

The Y-maze data are included in CSV files labeled:

6E, 6F (Y-maze)

Figure panels supported by these files:

Figure 6E: Arm entries.
Figure 6F: Percent alternation.

Open field data

The open field data are included in CSV files labeled:

S4D, S4E (Open field)

Figure panels supported by these files:

Supplementary Figure S4D: Percent time or activity in the periphery.
Supplementary Figure S4E: Distance traveled.

Figure 7 and Supplementary Figure S5 data files

Overview

Figure 7 and Supplementary Figure S5 are organized together because the analyses are related. The folder Figure 7, S5 contains raw immunoblot image files, raw numerical data files, and the Prism analysis file used to generate Figure 7 and Supplementary Figure S5.

The Prism file is named:

Figure 7, S5/Figure 7, S5.prism

Raw CSV files are included in:

Figure 7, S5/Raw CSV Files

Figure 7A: hAPP ELISA data

The CSV file labeled 7A contains the hAPP ELISA values used to generate Figure 7A. These values are also included in:

Figure 7, S5/Figure 7, S5.prism

Figure 7B: Immunoblot source image

The raw immunoblot image used for Figure 7B is included in:

Figure 7, S5/7B

This folder contains the immunoblot source file used for the densitometry analyses shown in Figure 7C-7L and Supplementary Figure S5A-S5B.

Figure 7C-7L and Supplementary Figure S5A-S5B: Immunoblot densitometry data

The CSV files labeled 7C-7L, S5A, S5B contain densitometry values quantified from the Figure 7B immunoblot. These values are also included in:

Figure 7, S5/Figure 7, S5.prism

The graphs generated from the Figure 7B immunoblot densitometry are organized in the following order:

Figure 7C: FL-APP
Figure 7D: APP-CTFα + APP-CTFβ
Figure 7E: APP-CTFα
Figure 7F: APP-CTFβ
Figure 7G: sAPPα
Figure 7H: sAPPβ
Figure 7I: BACE1
Figure 7J: PSEN1
Figure 7K: Nicastrin
Figure 7L: NEP2/neprilysin
Supplementary Figure S5A: PSEN2
Supplementary Figure S5B: IDE

Supplementary Figure S5A and Supplementary Figure S5B were included in the supplementary figures due to space limitations in the main figure.

Figure 8 data files

Overview

The Figure 8 folder contains the Prism analysis file and raw numerical data file used to generate Figure 8.

The Prism file is named:

Figure 8/Figure 8.prism

The folder contains one raw CSV file with the fold change values used for Figure 8. This file contains the numerical data used for graph generation and statistical analysis in Prism.

Data organization

The fold change values were organized in Prism according to the corresponding experimental groups shown in Figure 8. The same numerical values are provided in the CSV file to support accessibility, reuse, and downstream analysis.

The data in the Figure 8 CSV file are organized in the following order:

BACE1
PSEN1
PSEN2
Nicastrin
IDE
NEP2/neprilysin

Abbreviations

AD: Alzheimer’s disease
ANOVA: Analysis of variance
APP: Amyloid precursor protein
APP-CTF: APP C-terminal fragment
Aβ: Amyloid-beta
BACE1: Beta-secretase 1
CSV: Comma-separated values
CSF: Cerebrospinal fluid
Ct: Cycle threshold
DAPI: 4′,6-diamidino-2-phenylindole
ELISA: Enzyme-linked immunosorbent assay
FAD: 5xFAD
Fiji: Fiji image analysis software
FL-APP: Full-length APP
GAPDH: Glyceraldehyde 3-phosphate dehydrogenase
GFAP: Glial fibrillary acidic protein
hAPP: Human amyloid precursor protein
IBA1: Ionized calcium-binding adapter molecule 1
IDE: Insulin-degrading enzyme
ImageJ: ImageJ image analysis software
MATLAB: Matrix Laboratory
MCI: Mild cognitive impairment
MWM: Morris water maze
NEP2: Neprilysin 2
PAR: Poly(ADP-ribose)
PARP1: Poly(ADP-ribose) polymerase 1
PARP1ko: PARP1 knockout
PAS: Open field behavior analysis software used for this study
PSD: Photoshop document
PSD95: Postsynaptic density protein 95
PSEN1: Presenilin 1
PSEN2: Presenilin 2
ROI: Region of interest
RT-PCR: Reverse transcription polymerase chain reaction
SPSS: IBM SPSS Statistics
TBS: Tris-buffered saline
TBSX: Tris-buffered saline with detergent
ThioS: Thioflavin S
WT: Wild-type
γH2AX: Gamma H2A histone family member X