Transgenic mouse model

PDAPP and hTau APP KI male and female mice were used in our experiments, including digital spatial profiling and immunohistochemistry (IHC) analyses. PDAPP colony (line 13,388) was established through an inbreeding exercise wherein mice were inbred from selected litters that maintained decreased variability in both soluble and insoluble Aβ. The plaque deposition phenotype of the inbred PDAPP line 13,388 was similar to the originally described PDAPP colony [11]. The pathology of APP KI mice is characterized by predictable plaque development with progressive increases in parenchymal amyloid pathology as well as mild but detectable cerebral amyloid angiopathy [24]. PDAPP mice received weekly subcutaneous injections of 3D6 (anti-Aβ1-x,IgG2b) (25 mg/kg) or twice-weekly subcutaneous injections of IgG (IgG2a isotype) (50 mg/kg) for three months. hTau APP KI mice received weekly subcutaneous injections of biotynlated-3D6 (anti-Aβ1-x,IgG2b) (25 mg/kg) or weekly subcutaneous injections of IgG (IgG2a isotype) (25 mg/kg) for one month. For dosing experiments, 23-to 26-month-old were utilized. This age was selected due to the extensive vascular amyloid accumulation at 23–26 months of age in the PDAPP and hTau APP KI models. All experiments were performed in accordance with the Institutional Animal Care and Use Guidelines for Eli Lilly.

Brain sections immunofluorescence

Mice were anesthetized with avertin and perfused with ice-cold PBS. Brains were carefully removed from the cranium to avoid separating the meninges and fixed in 4% (wt/vol) paraformaldehyde (PFA) for 24 h followed by cryoprotection in 30% (wt/vol) sucrose in PBS solution. hTau APP KI brains were subsequently frozen and coronally sectioned (10 μm thick) using a freezing/sliding microtome. PDAPP coronal brain Sects. (10 μm thick) were paraffin embedded, deparaffinized in xylene, rehydrated in ethanol (EtOH) and washed with deionized water. Then, sections were heated in low pH 1 × citrate buffer for 30 min. After washing in PBS for 5 min twice, the sections were blocked with a solution of 5% goat serum and 0.01% Triton X-100 in PBS for 1 h at room temperature (RT). Sections were then incubated overnight at 4 °C with the following antibodies, each diluted 1:100 in blocking solution: anti-Fibrinogen ( PA1-85,429, ThermoFisher), anti-SMA (PA585070, ThermoFisher), anti-CD 31 (14–0311-82, ThermoFisher), anti-VEGF (19,003–1-AP, ThermoFisher), anti-Cldn5 (34–1600, ThermoFisher), anti-GFAP (13–030-0, ThermoFisher), anti-AQP4 (50–173-0968, ThermoFisher), anti-CD19 (14–0194-82, ThermoFisher), anti-CD3 (PIMA514524, Fisher), anti-CD4 (14–9766-82, ThermoFisher), anti-CD8 (14–0808-82, ThermoFisher), anti-Laminin (NB300-144,NovusBio) anti-MHC II (107,610, Citeab), anti-ERTR7 (NB100-64932, Novus) and anti-CD36 (18,836–1-AP, ThermoFisher). The next day, the sections were washed 3 times in PBS and incubated with the following secondary antibodies,1 h at room temperature, diluted 1:100 in blocking solution: goat anti-mouse IgG 555 (A31570, ThermoFisher), goat anti-mouse IgG 647 (A32728,ThermoFisher), goat anti-rabbit IgG 555 (A32732,ThermoFisher), goat anti-rabbit IgG 647 (A32733,ThermoFisher), goat anti-rat IgG 555 (A21434,ThermoFisher), goat anti-rat IgG 647 (A48272,ThermoFisher), streptavidin 488 (S32354, ThermoFisher). Slides were then crosslinked with 4% pfa for 5 min followed by two 5-min washes in deionized PBS. Amyloid was stained with X-34 (SML1954, 1:10,000) prepared in 50% EtOH in TBS for 10 min at RT, followed by two 5-min washes in 50% EtOH and two 5-min washes in PBS. Finally, the sections were washed in PBS and mounted with ProLong Diamond Antifade medium (P36961, ThermoFisher).

Microhemorrhage analysis

Slides were mounted with six μm thick coronal tissue sections for detection of vascular bleeds marked by hemosiderin deposits by incubating in 2% potassium ferrocyanide (EMS, 26,613–01) in 2% hydrochloric acid for 30 min followed by two 5-min washes in PBS. Slides utilized for microhemorrhage analysis were counter stained with nuclear fast-red (N8002, Sigma-Aldrich) for 5 min followed by a 5-min wash in PBS. Finally, the sections were washed in PBS and mounted with ProLong Diamond Antifade medium (P36961, ThermoFisher). Whole brain slices were scanned using the Zeiss AxioScan Z.1 with a 20X objective (0.8NA). Images were then analyzed for region of interest defined along images of tissue edges (100 μm wide) and Hemosiderin+ signal was thresholded to quantify percent coverage of stain relative to ROI using HALO v3.5 image analysis suite (Indica Labs).

GeoMx digital spatial profiling

Assay was performed on 10 μm thick FFPE (Formalin-fixed paraffin-embedded) coronal brain sections from IgG- and 3D6-treated PDAPP mice. Sections were mounted onto charged slides and underwent baking at 65 °C in a drying oven for 2 h prior to deparaffinization. Slides were deparaffinized and rehydrated by incubating in CitriSolv (Decon Labs) 3 times for 5 min each followed by 2 washes of 100% ethanol for 5 min and 2 washes of 95% Ethanol for 5 min and 2 washes of distilled water for 5 min each. The following steps were conducted on the BOND RXm (Leica Biosystems®, Baden-Wurttemberg, Germany) instrument. Slides were removed from the BOND RXm and were incubated overnight at 4 °C with a multiplexed cocktail of primary antibodies each of which are conjugated to unique ultraviolent-photocleavable oligonucleotide tags (GeoMx Immune Profile Core and Immune Cell Typing modules containing 31 targets, including 6 control probes). Additional fluorescently labeled antibodies were used as morphology markers to select regions of interest (ROIs). In the current study, the morphology markers included β-amyloid (MOAB-2, Novus Biologicals, NBP2-13075AF594), and laminin (Novus Biologicals, NB300-144AF532) to visualize β-amyloid deposition and neuro-vasculature, respectively. Following overnight incubation, slides were washed 3 times in 1X tris-buffered saline with Tween-20 (TBS-T) for 10 min each. Slides were then post-fixed in 4% paraformaldehyde (ThermoFisher, FB002, R37814) for 30 min at room temperature followed by two washes of TBS-T for 5 min each. For nuclear staining, slides were incubated in 1:10 dilution of Nuclear Stain Morphology Kits (DNA dye, SYTO 13), for 15 min at room temperature. Slides were washed with TBS-T for 5 min and loaded onto the GeoMx™ DSP according to the manufacturer’s instructions. In total, 7 mice (IgG treated n = 3 and 3D6 treated n = 4) were analyzed using NanoString GeoMx™ DSP technology. For each mouse, 12 CAA containing leptomeningeal and 12 CAA containing penetrating vessels were selected as ROIs for high-resolution multiplex profiling, comprising a total of 24 ROIs per animal. A 100-μm in diameter circle was selected as the center ROI surrounding each blood vessel with β-amyloid deposition. The specific 100 μm ROIs for molecular profiling were then sequentially processed by the microscope automation, as previously described [25,26,27]. In brief, ROIs were selectively illuminated with UV light to release the indexing oligos by coupling UV LED light with a double digital mirror device module. Following each UV illumination cycle, the eluent was collected from the defined region and transferred to an individual well of a microtiter plate. Once all ROIs were processed, pools of released indexing oligos were hybridized to NanoString optical barcodes for digital counting based on the manufacturer's directions. Hybridizations were performed at 65 °C overnight in a thermocycler. After hybridization, samples were processed using the nCounter Prep Station and Digital Analyzer as per manufacturer instructions. Data were normalized to technical controls and areas. Digital counts were normalized with negative controls (Rt IgG2b, Rb IgG, and Rt IgG2a).

RNAscope in situ hybridization

In short, 5 μm thick FFPE fixed coronal mounted tissue sections were serially dehydrated in 50%, 70%, 95%, 100% and 100% ethanol for 5 min each. In between all pretreatment steps, tissue sections were briefly washed with ultra-pure water. Incubation, wash, and probe application periods were performed by utilizing the RNAscope® LS Multiplex Fluorescent Reagent kit system (Advanced Cell Diagnostics) and RNAscope® LS 4-plex Ancillary Kit for Multiplex Fluorescent Reagent (Advanced Cell Diagnostics) on the BOND RX (Leica). Mounted slices were treated with pretreat solution 3 (protease reagent) for 15 min at 40 °C. Slides were then incubated with custom mouse CD74 RNAscope® probes (catalog #437,508-C4, ACD), Mrc1 (Catalog #437,518-C1, ACD) Trem2 (Catalog #404,118-C2, ACD). Following in situ hybridization, the sections were processed for immunohistochemistry. Briefly, following the blocking step with 5% normal goat serum (NGS) in PBS for 1 h at room temperature, post-hybridized slides were incubated 1:100 dilution with a 555-fluorescent conjugated antibody (NBP2-13075R, Novus) or 647-fluorescent conjugated antibody (NBP2-13075AF647, Novus) against amyloid in the presence of 2% NGS in PBS overnight at 4 °C. Brain sections were rinsed with PBS three times and incubated for 5 min in PBS with DAPI solution (1:50,000) for counterstained nuclei. Whole brain slices were scanned using the Zeiss AxioScan Z.1 with a 20X objective (0.8NA). Images were then analyzed for region of interest defined around vessels (100 μm circle) or along tissue edges (100 μm wide) and CD74+, Mrc1+, and Trem2+ signal was thresholded to quantify percent coverage of stain relative to ROI using HALO v3.5 image analysis suite (Indica Labs).

Microscopy and image analysis

For image analysis of mouse brain sections, we used ImageJ software v1.53 (NIH) to create one index that represented changes in the number of Fibrinogen+,SMA+,PECAM-1+, VEGF+,Cldn5+,GFAP+, AQP4+, CD19+,CD3+,CD4+, CD8+, MHC II+,or CD36+ pixels divided by the total number of pixels in the image, expressed as ( +) area %. To ensure the representativeness of our study's findings, we applied random selection when choosing animals from the total population for subsequent immunohistological investigations. A subgroup of 3D6 and IgG treated PDAPP mice (6 animals each) was randomly selected across microhemorrhage groups and cerebral cortices were examined using a SP8 confocal microscope (Leica), 63 × objective with a 0.15 μm z-step. We analyzed 8–10 vascular amyloid deposits from both the leptomeninges and penetrating vessels per animal across the entire coronal sections, using two brain sections per mouse. We quantified X-34+ area across the leptomeninges, focusing on a 100 μm diameter region along the entire leptomeningeal surface, while the vascular amyloid+ area percentage was assessed based on 3D6 immunoreactivity. Furthermore, we quantified the leptomeningeal diameters in microns by measuring from the border of the tissue edge to the pial surface. Results are shown as mean ± SEM of subgroup analysis n = 6 (mice). For colocalization quantification, images were analyzed in Imaris v8.4 (Bitplane) to determine relative colocalization coefficients (colocalization ratio) using Manders’ coefficients [28].

Statistics and reproducibility

The details about experimental design and statistics used in different data analyses performed in this study are given in the respective sections of results and methods. Sample sizes were determined based on previous publications. Significance in NanoString experiments was assessed employing a two-stage linear step-up procedure by Benjamin, Krieger, and Yekutieli through a multiple t test (unpaired t test on each row) [29]. The adjusted p-value threshold was set at alpha 0.05, utilizing the Holm-Sidak correction method. The results are represented as -log10 adjusted p-values for clarity. Additionally, ratio builder generated ratios of relative expression of target probes, calculating the ratio of each probe in every segment relative to the average of one or more segments chosen as the baseline. Pearson correlation analysis was used to analyze the relationship between different immune cells in Spotfire software and P < 0.05 was considered statistically significant. For imaging experimental modalities utilized in this study, a minimum of six independent biological replicates were employed to ensure robust and reliable results. Investigators were blinded for staining experiments. GraphPad Prism v10 was used to perform all statistical analyses. Statistical significance between groups was calculated using a student’s t-test (two-tailed). Data is presented as the mean ± SEM unless otherwise stated. *, **, *** and **** denote p < 0.05, p < 0.01, p < 0.001 and p <  < 0.0001, respectively. No other statistical comparisons were significant unless otherwise noted.