Animals
To generate the tg-UppSwe mouse line, cDNA of human APP (hAPP) with the APPSwe (KM670/671NL) [25] and the APPUpp (690-695Δ) [27] mutations was inserted into the murine Thy-1323-cassette, as previously described for the generation of the tg-Swe and tg-ArcSwe mouse lines [20]. Tg-UppSwe mice will therefore be fully comparable to tg-ArcSwe and tg-Swe in terms of any potential promoter induced effects. The linearized DNA was injected into oocytes at the Karolinska Center for Transgene Technologies (KCTT, Karolinska Institute, Stockholm, Sweden), resulting in nine founders of both sexes. The tg-UppSwe mouse line was established from one male founder by heterozygous breeding on a C57/BL6J-BomTac background. Upon further breeding, mice of both sexes from generation 6–8 were used in the study. For comparison, sex- and age-matched wildtype (wt) littermates were included. Presence of the human transgene was confirmed by PCR, using two sets of primer pairs that framed the Thy-1 basal promoter region and the APP coding region. Copy numbers of the human APP gene inserted in the mouse genome were assessed using TaqMan real-time PCR. In short, DNA was extracted from mouse brain tissue using DNeasy Blood & Tissue Kit (Qiagen, Germany). Taqman human APP assay Hs01255859_cn, together with an internal copy number reference (mouse transferrin receptor gene, Tfrc), was used according to the manufacturer’s instructions (Thermo Fisher Scientific, USA). Quantitative PCR (qPCR) was performed on the StepOnePlus™ Real-Time PCR System and results were analyzed by the CopyCaller v2.1 software (Thermo Fisher Scientific, USA). Tg-UppSwe mice were compared with tg-ArcSwe mice, expressing hAPP with the APPArc (E693G) and APPSwe mutations, and with tg-Swe mice, harboring only the APPSwe mutation [20]. In addition, wt mice were used for generation of primary cell cultures. All mice were bred on a C57/BL6J-BomTac background. The total number of animals included in the study is given in Table 1. All procedures were approved by the Stockholm North or Uppsala County Animal Ethics boards (N174-15, C85-16, 5.8.18-13350/17, 5.8.18-20401/20), following the rules and regulations of the Swedish Animal Welfare Agency, and were in compliance with the European Communities Council Directive of 22 September 2010 (2010/63/EU).
Brain tissue homogenization and Aβ ELISAs
Fresh frozen brain tissue (cerebrum) was homogenized at a 1:5 tissue:buffer ratio, using a Precellys Evolution homogenizer (Bertin Technologies, Montigny-Le-Bretonneux, France) to sequentially extract TBS, TBS with 0.5% Triton X-100 (TBS-T) and formic acid (FA) soluble fractions of Aβ. After homogenization in TBS, samples were centrifuged at 16,000× g for 1 h at 4 °C (TBS16K). For a subset of experiments, a fraction of this supernatant was further centrifuged at 100,000× g followed by collection of the supernatant (TBS100K). Homogenization of the remaining tissue pellet was repeated according to the same procedure with TBS-T, then followed by FA (Additional file 1: Fig. S1).
Brain extracts were analyzed with Aβ1-40 and Aβ1-42 ELISAs, as previously described [27]. In brief, 96-well half-area plates were coated over night with 50 ng/well of anti-Aβ40 (custom production, Agrisera) or anti-Aβ42 (700254, Thermo Fisher Scientific, USA), then blocked with 1% BSA in PBS for 3 h at RT. TBS16K, TBS-T and FA brain extracts were diluted in ELISA incubation buffer (PBS, 0.1% BSA, 0.05% Tween) and incubated over night at 4 °C, followed by detection with 0.5 µg/ml biotinylated 3D6 and HRP-conjugated streptavidin (Mabtech AB, Nacka, Sweden). Signals were developed with K Blue Aqueous TMB substrate (Neogen Corp., Lexington, KA, US). Plates were developed and read with a spectrophotometer at 450 nm.
The TBS and TBS-T extracts were also analyzed for soluble Aβ aggregates using two different sandwich ELISAs. The first, based on 3D6 as both capture and detection antibody, allows for detection of Aβ aggregates from the size of a dimer, but not for monomeric Aβ. The second preferentially detects larger soluble Aβ aggregates as it utilizes the Aβ protofibril selective antibody mAb158 for capture and 3D6 for detection [23]. Ninety-six-well half-area plates were coated over night with 3D6 (50 ng/well) and blocked with 1% BSA in PBS for 3 h at RT. Brain extracts were diluted in ELISA incubation buffer (PBS, 0.1% BSA, 0.05% Tween) and incubated over night at 4 °C, followed by 3D6-biotin detection and development as above.
Immunohistochemistry and thioflavin S staining
Right hemispheres from fresh frozen mouse brains were sectioned at 20 µm. Next, the sections were fixed with 4% paraformaldehyde and treated with pre-heated citrate buffer, pH 6.3, for 30 min followed by 70% formic acid for 5 min. Aβ was visualized with anti-Aβ40 (Agrisera, Umeå, Sweden), anti-Aβ42 1:1000 (700254, Thermo Fisher Scientific, USA), mAb158 or 3D6 1:1000 (in-house expression); activated astrocytes with anti-GFAP 1:200 (Abcam, Cambridge, UK); and microglia with antibodies against Iba1 (1:200) (Wako chemicals, Richmond, VA) and TREM2 1:200 (AF1729; R&D, Abingdon, UK). For colorimetric staining the Vector NovaRED™ horse radish peroxidase (HRP) substrate kit (Vector Laboratories, Burlingame, CA) was used for detection while for fluorescent staining Alexa secondary antibodies were used (Thermo Fisher Scientific, USA). For thioflavin S (ThS) staining, sections were pretreated in 95% and 70% ethanol (3 min in each), and quickly rinsed in water before they were incubated in 0.1% ThS for 10 min. Finally, the sections were briefly washed in 80% ethanol and water, dehydrated in ethanol, cleared in xylene and mounted with DPX.
Array tomography
Fresh brain tissue was collected from an 18-month-old tg-UppSwe mouse [11, 14]. Small tissue blocks containing cortex were fixed in 4% paraformaldehyde and 2.5% sucrose in 20 mM phosphate buffered saline pH 7.4 (PBS) for 3 h. Samples were dehydrated through cold graded ethanol of ascending strengths and embedded into LR White Resin (Electron Microscopy Sciences, Hatfield, PA, USA), which was allowed to polymerize overnight at 53 °C. Resin embedded tissue blocks were cut into array ribbons of 70 nm thick sections using an ultracut microtome (Leica, Wetzlar, Germany) equipped with a Jumbo Histo Diamond Knife (Diatome, Hatfield, PA) and collected onto gelatin coated coverslips. For detection of colocalization between pathological proteins and synapses, array ribbons were immunostained with a primary antibody against post-synapses (PSD95), pre-synapses (synaptophysin) and with the OC polyclonal antibody against fibrillar Aβ [12]. Sections were counterstained with 0.01 mg/mL 4′-6-diamidino-2-phenylindole (DAPI). For each experiment, a short extra ribbon was used as a negative control. Images were obtained on serial sections using an AxioImager Z2 epifluorescent microscope (Carl Zeiss, Oberkochen, Germany) with a 10× objective for tile scans and 63× 1.4 NA Plan-Apochromat objective for high resolution images. Images were acquired with a CoolSnap digital camera and AxioImager software with array tomography macros (Carl Zeiss). Images from each set of serial sections were converted into image stacks and aligned using the ImageJ plug-in, MultiStackReg (courtesy of Brad Busse and P. Thevenaz, Stanford University) [39]. Regions of interest within the cortical neuropil were chosen (10 μm2) and their proximity to plaque edges recorded (< 20 μm from a plaque edge considered “near” plaques and > 20 μm from a plaque edge considered “far” from plaques). Image stacks were then binarised using thresholding algorithms in ImageJ. For synaptic staining, image stacks were binarised using an ImageJ script that combines different thresholding algorithms in order to select both high and low intensity synapses in an automated and unbiased manner. To examine pathological protein presence at the synapse, thresholded images were processed and analyzed in MATLAB to remove background noise.
Antibody production and radiochemistry
For PET imaging with antibody-based ligands (immunoPET), the bispecific brain penetrating Aβ antibodies RmAb3D6-scFv8D3 [5] and RmAb158-scFv8D3 [9] were used. While 3D6 binds to the N-terminus of Aβ [1], mAb158 preferentially binds to soluble Aβ protofibrils and to some extent also to Aβ fibrils [2]. Both bispecific antibodies actively enter the brain via receptor mediated transcytosis, using the TfR binding domain scFv8D3. The antibodies were produced recombinantly in Expi293 cells, according to previously described procedures [4] and radiolabeled with iodine-124 (124I) for PET imaging or with iodine-125 (125I) for ex vivo studies [36]. In brief, for 124I-labeling, a [124I]iodide stock solution (Advanced Center Oncology Macerata, Montecosaro, Italy) was pre-incubated for 15 min with half a volume of 50 µM NaI and then neutralized with 0.5% HAc and PBS. After adding 90 µg antibody, the reaction was initiated by the addition of 40 µg Chloramine-T (Sigma Aldrich, Stockholm, Sweden) and, after 120 s, quenched with 80 µg sodium metabisulfite (Sigma Aldrich). For 125I labeling, a [125I]iodide stock solution (PerkinElmer Inc., Waltham, MA, USA) was mixed with 40 µg of antibody in PBS and the reaction was initiated by the addition of 5 µg Chloramine-T (Sigma Aldrich) and, after 90 s, quenched with 10 µg sodium metabisulfite (Sigma Aldrich). The radiolabeled antibody was purified from non-reacted [125I]iodide with Zeba spin desalting columns (7K MWCO, 0.5 mL, ThermoFisher, Uppsala, Sweden). The molar activity of [124I]RmAb3D6-scFv8D3 and [124I]RmAb158-scFv8D3 was 135 MBq/nmol and 133 MBq/nmol, respectively.
The amyloid PET radioligand [11C]PiB, formulated in 10% ethanol in PBS, was synthesized as previously described with minor modifications to adapt the procedure to our in-house built synthesis device (TPS) [13].
PET imaging
Eighteen months old tg-UppSwe, tg-ArcSwe, tg-Swe and wt mice underwent PET imaging with [11C]PiB or with either of the two antibody radioligands, [124I]RmAb3D6-scFv8D3 or [124I]RmAb158-scFv8D3 (n = 3 per ligand and genotype). For [11C]PiB-PET, mice were injected with 13.2 ± 2.9 MBq radioligand and PET data acquired between 40–60 min after injection was used for all subsequent analyses. For immunoPET, mice were given 0.2% NaI in the drinking water to reduce thyroidal uptake of 124I. The following day, the mice were injected with 8.7 ± 1.6 MBq of [124I]RmAb3D6-scFv8D3 or [124I]RmAb158-scFv8D3, corresponding to an antibody dose of 2.3 nmol/kg body weight. Four days after antibody injection, mice were PET scanned for 60 min with either a Triumph Trimodality System (TriFoil Imaging, Chatsworth, CA) or a nanoScan system PET/MRI (Mediso, Budapest, Hungary). PET scans performed with the Triumph system were reconstructed with a 3-dimensional ordered-subsets expectations maximization, with 20 iterations. The PET data acquired with the Mediso system were reconstructed using a Tera-TomoTM 3D algorithm (Mediso) with four iterations and six subsets. Each mouse underwent a CT scan following PET. All subsequent image processing was performed with Amide version 1.0.4. The CT and PET data were manually aligned with a T2-weighted mouse brain atlas [22] for quantification of activity in the cerebrum.
Single injection immunotherapy and brain distribution
To investigate potential acute treatment effects and to further assess brain distribution of the two antibodies used for immunoPET, 18-months-old tg-UppSwe mice were injected with PBS (n = 4), or with a therapeutic dose (32 nmol/kg) of RmAb3D6-scFv8D3 (n = 5) or RmAb158-scFv8D3 (n = 5). The bispecific antibody preparations were supplemented with trace amounts (1.2 nmol/kg; 18 ± 1.6 MBq/kg) of [125I]RmAb3D6-scFv8D3 or [125I]RmAb158-scFv8D3 for quantification. After three days, the mice were euthanized by intracardiac perfusion. Radioactivity was quantified in brain and blood as well as in TBS16K, TBS-T and FA extracts of homogenized brain.
Ex vivo antibody brain distribution
After immunoPET imaging or administration of antibodies at a therapeutic dose, the mice underwent intracardiac perfusion with 20 ml 0.9% NaCl during 2.5 min. The brains were then isolated and separated into right and left hemispheres, followed by a further division of the left hemisphere into cerebrum and cerebellum. To assess the concentration of antibody in the brain tissue, radioactivity was quantified in the isolated brain regions using a gamma counter (Wizard 1480 Wizard™, Wallac Oy, Turku, Finland) and expressed as % of injected dose per gram brain tissue (%ID/g brain). To visualize brain distribution of radiolabeled antibody, 20 µm cryosections from the right hemisphere were exposed to phosphor imaging plates (MS, MultiSensitive, PerkinElmer, Downers Grove, IL) for 7 days. Plates were scanned with a Cyclone Plus phosphor imager (PerkinElmer, Waltham, MA) at 600 dpi resolution. Radioactivity distribution was visualized with ImageJ using a royal lookup table and combined with Aβ42 immunostaining of an adjacent brain section.
Astrocyte cultures and Aβ uptake studies
To study interactions between Aβ and primary mouse astrocyte cultures, cerebral cortices of wt mice (n = 3) were dissected from embryonal day (E14) mice in Hank’s buffered salt solution (HBSS) supplemented with 50 U/ml Penicillin, 50 mg/ml Streptomycin, and 8 mM Hepes buffer (ThermoFisher Scientific). The cortices were centrifuged in fresh HBSS for 3 min at 150× g and then resuspended and dissociated into a homogenous solution. Any remaining blood vessels were allowed to sediment for 10 min. The supernatant was transferred to a new tube and centrifuged for 5 min at 150× g. The cell pellet was carefully resuspended in DMEM/F12 GlutaMax cell culture medium. The embryonic cortical stem cells were allowed to expand as neurospheres in DMEM/F12 GlutaMax medium (Invitrogen) supplemented with B27 (Invitrogen), 100 U/ml penicillin, 100 μg/ml streptomycin, 8 mM HEPES buffer, 10 ng/ml bFGF (Invitrogen, diluted in 10 mM Tris–HCl (pH 7.6) + 0.1% BSA and PBS) and 20 ng/ml EGF (BD biosciences), dissolved in MQ water). The cells were then dissociated and plated as a monolayer at a density of 3 × 104 cells/cm2 on cover glasses coated with poly-L-ornithine and laminin. The following day, the growth factors were removed to start differentiation, resulting in co-cultures containing ~ 75% astrocytes after 7 days. Differentiated cell cultures were exposed to 0.1 μM Cy3-labeled sonicated fibrils of Aβ1-42Upp, Aβ1-42Arc or Aβ1-42wt. Control cultures received culture medium without Aβ. After 24 h, the cells were washed three times in cell culture medium and the cover slips were transferred to new culture dishes.
Preparation and Cy3 labeling of Aβ fibrils
To induce fibrillization, synthetic Aβ (200 µM in NaOH) was diluted in 2× PBS to a concentration of 100 µM and incubated on a shaker at 1500 rpm and 37 °C for 24 h. Tween-20 was then added to a final concentration of 0.01%.
For the labeling process, a Cy3AM antibody labeling kit (GE Healthcare, PA33000) was used. The fibrils were gently mixed in the coupling buffer by vortexing and then supplemented with Cy3. The mixture was incubated for 1 h at RT in the dark, then purified from unreacted Cy3 with dialysis in PBS with 0.01% Tween-20 for 2 h. The resulting Cy3-Αβ fibrils were diluted in sterile PBS to a final concentration of 0.5 mg/ml and sonicated at 20% amplitude, 1 s on/off pulses for 1 min (#VCX130, Vibra Cell sonicator, Sonics, CT, USA).
Immunocytochemistry
The cells were fixed for 15 min at RT with 4% paraformaldehyde, washed twice with PBS and permeabilized and blocked with 0.1% Triton X-100 (both from Sigma-Aldrich) and 5% normal goat serum (NGS, BioNordika) in PBS for 30 min at RT. Primary antibodies, diluted in 0.1% Triton X-100 with 0.5% NGS, were added and left to incubate for 1–4 h at RT or overnight at 4 °C. The cells were then washed 3 × 10 min with PBS before incubation with secondary antibodies (diluted in 0.1% Triton X-100 and 0.5% NGS) for 45 min at 37 °C or 1 h at RT. Cover slips were mounted onto microscope glass slides using VECTASHIELD hard set mounting medium with DAPI (DAKO). Imaging was performed using a Zeiss Observer Z1 Microscope, and the images were visualized with the Zen 2012 software and representative 40× images were captured. The primary antibodies used were chicken anti-Glial Fibrillary Acidic Protein (GFAP, 1:200, Abcam) and rabbit anti-lysosome-associated membrane protein-1 (LAMP-1, 1:200, Abcam). The secondary antibodies applied were AlexaFluor 488 (rabbit, 1:200, Thermofisher), and AlexaFluor 647 (chicken, 1:200, Thermofisher).
GFAP ELISA
Quantification of GFAP in tg-UppSwe brain extracts was performed with a sandwich ELISA, as previously described [26]. Ninety-six-well half-area plates were coated over night with 25 ng/well of anti-GFAP antibody GA5 (Sigma Aldrich), then blocked with 1% BSA in PBS for 2 h at RT. The TBS16K and TBS-T brain extracts were diluted in ELISA incubation buffer (PBS, 0.1% BSA, 0.05% Tween) and incubated over night at 4 °C, followed by 1 h incubation with 0.5 µg/ml polyclonal anti-GFAP (Dako, Z0334). Signals were detected with 0.5 µg/ml biotinylated goat anti-rabbit antibody in combination with HRP-conjugated streptavidin and K Blue Aqueous TMB substrate and read at 450 nm as above. A standard curve of recombinant GFAP (in-house produced [24]) was used for quantification.
TREM2 ELISA
Soluble TREM2 (sTREM2) was detected with a sandwich ELISA, performed in a similar manner as for the assays described above. Ninety-six-well half-area plates were coated over night with 25 ng/well of anti-TREM2 antibody AF1729 (R&D, Abingdon, UK), then blocked with 1% BSA in PBS for 2 h at RT. The TBS16K and TBS-T brain extracts were diluted in ELISA incubation buffer (PBS, 0.1% BSA, 0.05% Tween) and incubated over night at 4 °C, followed by detection with 0.25 µg/ml biotinylated anti-TREM2 BAF1729 (R&D), HRP-conjugated streptavidin and K Blue Aqueous TMB substrate and read at 450 nm with a spectrophotometer. A standard curve of recombinant TREM2 was used for quantification.
Statistics
Data was analyzed using GraphPad Prism (version 6 and 7, San Diego, CA). Comparisons of three or more groups were analyzed by one-way ANOVA for single datasets and by two-way ANOVA for multiple datasets, followed by Tukey’s post hoc test. A p value threshold of 0.05 was used for assessment of the statistical significance. Values are shown as means ± SD.
Add Comment