Animals: TgF344-AD rats express two transgenes, APPswe and presenilin 1 with the Δ exon 9 mutation, driven by the mouse prion promoter on the F344 rat strain [9]. Four cohorts of rats, were utilized; 7 months, 10 months, 13 months, and 16 months of age, with a total of 20 rats utilized for this study (n = 5 for each cohort). Non-transgenic rats do not develop vascular amyloid and thus were not analyzed. All rats were bred in-house, and kept under a 12-hour light/dark cycle, with an ad libitum access to food and water. All procedures were approved by The Animal Care Committee of the Sunnybrook Research Institute. In addition, experimental procedures adhered to the standards and guidelines of the Canadian Council of Animal Care, and complied with the requirements of the Regulations of the Animals for Research Act. This study is reported in accordance with the Animal Research: Reporting of In Vivo Experiments guidelines.
Immunofluorescence staining: TgF344-AD rat brains were coronally sectioned at 40 µm and five sections per rat were selected between 1.28 mm and − 3.24 mm from Bregma. All sections were washed with phosphate-buffered-saline three times for 10 minutes, submerged in 1% Thioflavin S (ThioS) (Sigma-Aldrich, Canada, 1326-12-1) for 7 minutes, then washed with 70% ethanol. After three washes with 1x phosphate-buffered saline, sections were blocked in 5% goat serum and 0.5% triton in 1x phosphate-buffered saline for one hour. Primary antibody incubation overnight with rabbit anti-alpha smooth muscle actin antibody (1:1000) (Abcam, United Kingdom, ab5694) was conducted, the following day the sections were washed again with 1x phosphate-buffered saline and incubated with Alexa Fluor 594-conjugated goat-anti-rabbit secondary antibody (1:200) with lycopersicon esculentum lectin (1:200), DyLight 649 (Lectin) (Vector Laboratories, USA, DL-1178). Alternatively, 40 µm sections were washed three times with 1x tris-buffered saline for 10 minutes each, treated with 88% formic acid for eight minutes at room temperature (RT), followed by three 10 minutes washes of 1x tris-buffered saline with 0.25% triton and a one-hour block incubation containing 2% goat serum and 0.5% bovine serum albumin in 1x tris-buffered saline with 0.25% triton. Overnight primary antibody incubation was performed with rabbit anti-alpha smooth muscle actin antibody (1:1000) and one of five anti-Aβ antibodies (1:200 each); MOAB-2 (Abcam, U.K, ab126649), 4G8 (Biolegend, USA, 800701), 6E10 (Biolegend, USA, 803015), Aβ40 (Biolegend, USA, 805401), and Aβ42 (Biolegend, USA, 805501), followed by one of Alexa Fluor 488-conjugated goat anti-rabbit (1:200, Invitrogen, USA, A11008), Lectin 649 (1:200), and either the Alexa Fluor 555-conjugated goat anti-mouse IgG2b antibody (1:200, Invitrogen, USA, A21147) or Alexa Fluor 594-conjugated goat anti-mouse IgG antibody (1:200, Invitrogen, USA, A11032).
Image Analyses
For every transgenic rat, a total of 25 cortical penetrating arterioles and 25 cortical penetrating venules were randomly selected with n = 5 rats per cohort, thus analyzing 125 vessels of each type. Accumulation of arteriolar and venular amyloid was calculated using ImageJ software. Vascular amyloid was evaluated semi-quantitatively by categorizing penetrating arterioles and venules through five different grades based on severity – no coverage (grade 0), trace/scattered coverage (grade 1), mild (grade 2), moderate (grade 3), and severe coverage (grade 4) – as previously established [34, 44].
Laser Capture Microdissection
For laser capture microdissection rats at 12 and 16 months of age were utilized. Rat brains were coronally sectioned at 10 µm (Bregma 1.20 mm through − 6.04 mm), dehydrated using a gradient of ethanol, followed by xylene, and then airdried and stored at 4°C until use. The PALM microbeam microscope, together with the PALM RoboSoftware (Carl Zeiss Microscopy, Germany) was utilized to perform the laser capture microdissection. Using a freehand selection tool, three distinct amyloid aggregates were cut and isolated; parenchymal plaques (n = 1600 elements), arteriolar amyloid (n = 1900), and venular amyloid (n = 7700). Control samples were collected by isolating the non-plaque associated parenchyma (n = 850), arteriole (n = 1080) and venule (n = 3600) tissue devoid of vascular amyloid. Pooled samples were collected onto AdhesiveCap 500 centrifuge tubes (Carl Zeiss Microscopy, Germany), which were stored at -80°C prior to mass-spectrometry.
Mass Spectrometry
Proteins were extracted in 8 M urea in 50 mM TRIS (pH = 8.0) for 20 minutes at RT. Each sample was reduced for 1 hour in 10 mM ditiothreitol (DTT) at 60°C and alkylated 20 mM iodoacetamide for 45 minutes at RT. To decrease the concentration of urea to < 1 M, 50 mM ammonium bicarbonate was added. The samples were digested overnight using Trypsin (Pierce, Thermo Fisher Scientific, USA) at 37°C, then lyophilized on a SpeedVac (Thermo Fisher Scientific, USA), desalted using C-18 ziptip (Millipore, Canada), and lyophilized again. The peptides were resuspended in 0.1% formic acid and were separated using liquid chromatography, performed by an Easy-nLC 1200 nano-LC system (Thermo Fisher Scientific, USA) containing one analytical column comprising of the stationary phase, 2 µM C18 beads (Thermo Fisher Scientific, USA). The fragments were loaded onto the analytical column at a maximum pressure of 900 Bar at 60°C and were fractionated using a linear gradient containing 0.1% formic acid to 80% acetonitrile with 0.1% formic acid (Buffer B) at a flow rate of 250 nL/min. Eluted peptides were ionized with the Nanospray Flex ion source (Thermo Fisher Scientific, USA) and tandem mass spectrometry was completed by the Thermo Scientific Orbitrap Fusion Lumos Tribrid mass spectrometer. A full mass spectra was performed in the Orbitrap mass analyzer at a resolution of 120 000 FWHM and a dynamic exclusion filter was conducted, excluding ions with a mass tolerance window of 10 ppm for 10 seconds. Ions exceeding the minimum intensity threshold of 10 000 counts were selected for a data-dependent collision-induced ion trap (centroid datatype) fragmentation with a scan range of 200–1400 m/z and a collision energy set at 30%. The tandem mass spectrometry data were analysed against the Uniprot protein sequence FASTA files of the rat proteome with the addition of the human APP sequence. The MS-Amanda Proteome Discoverer (Research Institute of Molecular Pathology, Austria), Sequest (XCorr Only) (Thermo Fisher Scientific, USA), and X! Tandem (The GPM, thegpm.org) are the three databases used for the analysis and were searched with a fragment ion mass tolerance of 0.60 Da. Validation was completed by the Scaffold Software (version 4.10.0, Proteome Software Inc., USA). Any peptides exceeding a 95.0% probability threshold, established by the Peptide Prophet algorithm [22], and had at least two fragments identified were reported.
Statistical Analysis
Statistical analysis was performed using GraphPad Prism (Version 6.01, Graphpad Prism, USA) and data was described as the mean ± standard error of the mean (SEM). A one-way analysis of variance with a Holm-Sidak post hoc test was performed to analyze percent vascular amyloid coverage of arterioles and venules across all four age groups independently to allow for multiple comparisons (n = 5 per age cohort). Paired two-tailed t-tests were performed to compare percent arteriolar and venular amyloid within an age cohort (n = 5 per age cohort). The rate of arteriolar and venular amyloid accumulation was assessed using a nonlinear regression analysis. Separate one-way ANOVAs were used to analyze the proportion of amyloid-positive arterioles and venules in three regions of the cortex with disease progression; somatosensory, motor, and the cingulate cortices (n = 5 per age cohort). No data were excluded in our results.