2.1 Chemicals
3H-THK5117 (specific activity (SA) = 75 Ci/mmol), 3H-PiB: SA = 73 Ci/mmol and THK5117 from Novandi chemistry AB (Södertälje, Sweden). BTA-1 (2-(4´-Methylaminophenyl)benzothiazole) was purchased from Sigma-Aldrich, USA.) 3H-MK6240 and MK6240 were synthesized by Merck & Co (West Point, PA, USA; SA= 44 Ci/mmol). Immunocytochemistry used an AT8 mouse monoclonal phospho-Tau (epitope Ser202, Thr205) antibody (Invitrogen CA, USA) and Amylo-Glo ready to dilute (RTD, Biosensis, Therbarton, Australia) was used to visual amyloid.
2.2 Human brain tissues
Formalin-fixed paraffin-embedded brain sections (5-10 µm) from DS-AD (n=7), AD (n=9), and control (n=8) cases) were assessed. Tissues were obtained from: the Croatian Institute for Brain Research (Dr. Goran Simic), the Carroll A. Campbell Jr. Neuropathology Laboratory (CCNL) at the Medical University of South Carolina (MUSC, Dr. Steve Carroll and Dr. Eric Hamlett), and University of California, Irvine, Alzheimer's Disease Research Center (UCI-ADRC) and the Institute for Memory Impairments and Neurological Disorders (Dr. Ira Lott and Dr. Eric Doran). This research was completed under each University’s Institutional Review Board (IRB) with either the determination of Not Human Research (NHR) Criteria set forth by the Code of Federal Regulations (45CFR46) or as Human Subjects Research with the requirement for written informed consent from the participants or the participants Legally Authorized Representative per local statute (UCI-ADRC). The fetal tissue was obtained from the Croatian Brain Bank according to NIH regulations for fetal tissues. The brain tissue processing and neuropathological staging of AD and control cases was assessed by a board-certified neuropathologist and included assessment of CERAD plaque score and Braak stage according to NIA protocols [11]. To overcome potential inter-cohort variations in brain processing procedures, tissues from all cohorts were processed simultaneously by the same investigator for each experimental condition. In addition, amyloid and tau staining were performed by the same lab for all cases and each brain region to confirm neuropathological staging for every case (see Table 2 for case demographics.)
2.3 Autoradiography
Autoradiography was performed at Karolinska Institute using 5-10 µm thick paraffin sections from the MFG and HIPP from DS-AD, AD and age-matched control cases. Autoradiography was performed at room temperature (RT) after deparaffinization. For 3H-PIB, the sections were pre-incubated (15 min) with binding buffer (phosphate-buffered saline (PBS) + 1% BSA) and then 45 min with 1 nM of 3H-PiB in binding buffer + 10% ethanol. Non-specific binding (NSP) was determined by incubation with 10 µM of unlabeled BTA-1. For 3H-THK5117 and 3H-MK6240 sections were pre-incubated for 15 min in PBS containing 0.1% BSA followed by a 1-hour incubation with 3H-THK5117 (3-4 nM) or 3H-MK6240 (1 nM). NSP was assessed by incubation with unlabeled THK5117 (10 µM) or unlabeled MK6240 (1 µM). Binding was terminated by washing 3X5 min with cold binding buffer and distilled water. Sections were dried and apposed for 4 days for 3H-THK5117 and 7 days for 3H-PiB and 3H-MK6240 on a phosphoplate and read using a BAS-2500 imager (BAS-TR2040, Fuji imaging plate, Fujifilm, Tokyo, Japan) and analyzed using multigauge software, Fujifilm, Tokyo, Japan). Non-specific binding was subtracted from total binding to obtain specific binding densities for each tracer and brain region.
2.4 Immunostaining
To confirm that the Tau tracers were binding to regions containing Tau, AT8 immunostaining was performed on adjacent sections at Barrow Neurological Institute. Paraffin-embedded sections were deparaffinized and pre-treated for antigen retrieval using formic acid, as previously described [12]. Endogenous peroxidase activity was blocked by 1% hydrogen peroxide, followed by incubation overnight with the AT8 monoclonal antibody (1:1000, Invitrogen). After washing, sections were incubated with biotin-conjugated secondary antibodies, washed in PBS, incubated with streptavidin-horseradish peroxidase (HRP) complex and visualized using a 1 mg/ml diaminobenzidine (DAB) solution containing 0.02% H2O2, dehydrated and cover-slipped. Immunostained controls include deletion of the primary or secondary antibodies.
To confirm that the PIB binding was targeting amyloid, amyloid plaque staining was completed at University of Denver (DU) using Amylo-Glo RTD Staining reagent (Biosensis #TR-300-AG) following the manufacturer’s protocol. Briefly, rehydrated sections were placed in 70% ethanol, rinsed in distilled water, incubated in the Amylo-Glo RTD solution, rinsed in 0.9% saline solution for 5 minutes, briefly washed in distilled water and cover slipped with ProLong Diamond Antifade. Amyloid plaques labeled with Biosensis’ Amylo-Glo RTD were visualized using UV epifluorescent illumination attached to a Nikon Optiphot microscope (Eclipse 600). Each staining batch contained sections from all groups to avoid potential variability resulting from different processing protocols at the different brain banks.
2.5 Semi-quantitative data analysis
For autoradiography, the gray matter was used as a region of interest (ROI) and drawn manually both on the total and the non-specific autoradiogram using Multi Gauge V3.0 software (Fujifilm, Tokyo, Japan) [13]. Values were transformed from PSL/mm2 into fmol/mg using a tritium Standard and specific binding was determined by subtracting the non-specific binding from total binding. For the AT8 and Amylo-Glo stains, densitometric measures were obtained using the Image J density plugin, with a grayscale of 0-256 and background subtraction[14]. Three arbitrary frames were placed in the gray matter in each target area, mean density was obtained and background subtracted from a non-stained portion of each section within the ROI. An average density was obtained in each section and a mean value per brain region determined.
2.6 Statistical methods
Statistical analysis was performed using GraphPad Prism, version 6 (GraphPrism Software, La Jolla, CA, USA). Checking for outliers in each data set using Grubb’s test resulted in one outlier excluded from the Control group for Amylo-Glo staining density. One-way ANOVA followed by Tukey’s multiple comparisons test was used for the comparison between groups (AD, DS-AD, control) and regions (MFG and HIPP) for PET tracer binding. Significance was set at p < 0.05. Pearson correlations were used to assess associations between 3H-THK5117, 3H-MK6240 and AT8, and between 3H-PiB and Amylo-Glo.