Supplement Material and Methods Notch reporter assay The cells were seeded in 96 well microtiter plates in cell culture medium (Iscove without phenol red, plus 10% (v/v) fetal bovine serum) to about 80% confluency and the test compounds were added at a 10x concentration in 1/10 volume of medium without FCS containing 8% DMSO (final concentration of DMSO was kept at 0.8% v/v). After 20 hrs incubation at 37°C and 5% CO2 in a humidified incubator the microtiter plates were allowed to equilibrate to room temperature. 45l of SteadyGlo Mix (Promega) were added and incubated another 20’ at room temperature. Chemoluminescence was measured in a SpectraMax Paradigm Multi-Mode Microplate Reader using a Glow Luminescence cartrige (Molecular Devices). Protein binding Protein binding of RO5506284 in mouse plasma was determined by equilibrium dialysis on a high throughput dialysis (HTD) plate with 12–14 K dialysis membrane (HTDialysis, LLC, Gales Ferry; CT). RO5506284 was added to pooled C57Bl/6 mouse plasma from commercial suppliers (0.6 mL) to yield a final concentration of 1 mM and then distributed in triplicates (0.15 mL per well) into an HTD plate. After 5 h equilibration against 0.15 mL phosphate buffer (Na 2HPO4/KH2PO4, pH 7.4), 10 µL and 90 µL were collected from the plasma and the buffer dialysis compartment, respectively. The concentration of RO5506284 was determined by LC–MS/MS. The plasma unbound fraction was calculated as described in 1. 1 P-glycoprotein Transporter-mediated efflux of RO5506284 mediated by mouse p-glycoprotein was studied in vitro in a cellular system overexpressing Mdr1a (LLC-PK1 type cells provided by Dr. A. Schinkel, Netherlands Cancer Institute, Amsterdam, The Netherlands). The method has previously been described 2. In short, culture medium was removed from the apical and basolateral compartments of the 96- insert plate and replaced with medium without phenol red in the receiver side shortly before initiation of the experiment. The measurement of transcellular transport was started by adding to the donor side culture medium without phenol red containing 1 µM test compound. The transport experiment was performed in both apical-to-basolateral and basolateral-to-apical directions in triplicates on a Tecan (Maennedorf, Switzerland) robotic pipetting device. The inserts were incubated at 37oC and 5% CO2 and 20 µL samples were taken from both the donor and receiver sides after 3.5 h of incubation. Test compound concentrations were measured by LC–MS/MS and the permeability rates and efflux ratio was calculated as described 2. Estimation of pharmacological in vivo potency The pharmacokinetic (PK) and pharmacodynamics (PD) parameters for RO5506284 were estimated using the plasma concentration- and the effect-time (relative Aβ42 levels in mice brain) profiles generated at different dose levels. Estimation of the PK and PD parameters was performed sequentially. First, an open 2-compartmental PK model with first order absorption and a dose-dependent elimination rate constant was fit to the concentration-time data of doses ranging from 3 to 100 mg/kg (Supplement Figure 1). The pharmacokinetics of RO5506284 are characterized by an apparent volume and clearance of 1.22 L/kg and 1.90 mL/min/kg, respectively. The pharmacokinetic properties of RO5506284 result in rapid changes in the 2 concentration time profiles (see Figure 2B) evident by a short terminal half-life of 2.2 h. Then, an indirect response model type I (inhibition of production) was fit to the effect-time data. The relevant PD parameters kout, Imax and IC50 are 1.08 h-1, 0.78 and 1340 ng/mL (referring to total plasma concentration), respectively. The turnover rate constant (kout) is consistent with that of Aβ40 in mice 3. A (% o f c trl S E M ) 150 b r a in A 4 2 100 *** 50 *** *** *** *** 0 v e h ic le 2h 4h 6h 8h 10h 24h 1 0 0 m g /k g Supplement Figure 1: Time dependent reduction of brain A42 after a single dose of 100 mg/kg p.o. in young 3 month-old, pre-amyloid Tg mice. Each bar represents the mean of N=4 animals. (***) statistically significant at p<0.001 (tested by one-way ANOVA, Dunnett's multiple comparisons test) GSM plasma level assessment Plasma samples (50 µL) were added to a 96-deep well plate and treated for protein precipitation with addition of 150 µL acetonitrile/methanol (1:1) containing the internal standard of a known concentration of RO5506284. After mixing and centrifugation (10 min, 5889 x g) 100 µL portions of the supernatant were transferred to another 96deep well plate, with addition of 400 µL volumes of H2O/methanol (1:1), whereupon the plate was vortexed gently. The deep well plate was sealed and transferred to the autosampler, for LC-MS analysis of 12 µL samples in order to quantify GSM levels. 3 Chromatographic analyses were performed using an Ascentis Express C18 column at a flow rate of 0.7 mL/min. The mobile phase was a mixture of two solvents: (A) 10 mM ammonium acetate, 0.05% formic acid and (B) methanol, 10 mM ammonium acetate, 0.05% formic acid. Mass spectrometric detection was carried out on an API3000 triple quadrupole mass spectrometer using selected reaction monitoring (SRM) in the negative ion mode. The calibration standards in the range of 0.25-5000 ng/mL were prepared in mouse plasma; the lower limit of quantification (LLOQ) was 5 ng/mL, using 50 µL plasma. Animals Transgenic APP-Swe mice were backcrossed with C57BL/6 mice to generate a line with less than 5% DBA2 background, as determined by microsatellite genotyping. Mice heterozygous for the transgene start to accumulate Aβ in brain at the age of approximately nine months, and at around the age of 12 months proceed to develop histologically visible Aβ plaques, located mainly in the cortical mantle. Aβ extraction from brain and quantification In brief, brain hemispheres were homogenized in nine volumes of 1% diethylamine, 50 mM sodium chloride, incubated for three hours on ice, and then centrifuged for one hour at 100,000 x g (4°C). Portions of supernatants were aliquoted and stored at -80 °C until assayed. To extract the insoluble Aβ pool, the centrifugation pellet was resuspended in urea buffer (9 M urea, 50 mM Tris/HCl (pH 7.4)). After incubation for 30 minutes at room temperature, the sample was centrifuged for one hour at 55,000 x g, and the supernatant (urea fraction) was stored at -80 °C until further analysis. The amounts of Aβ38, Aβ40 and Aβ42 in brain extracts were determined by a sandwich 4 immunoassay as described previously 4, except that the blocking buffer was changed to 5% BSA, 60 mM NaCl, 0.1% Tween20, 50 mM Tris/HCl (pH 7.4). The plates were washed in the same buffer lacking BSA. For quantification, samples of the urea fraction were diluted 1:10 in assay buffer (0.5% BSA, 60 mM NaCl, 0.1% Tween20, 50 mM Tris/HCl pH 7.4). Aβ standards were prepared in assay buffer supplemented with 10% urea buffer and 6% FBS. Aβ-PET In brief, mice were anesthetized with isoflurane (1.5%, delivered at 3.5 L/min) and received an injection of 11.5 ± 2.6 MBq [18F]-florbetaben to a tail vein. After placement in the aperture of the Inveon DPET (Preclinical Solutions, Siemens Healthcare Molecular Imaging), a single frame emission recording was obtained in the interval 30-60 min p.i., followed by a 15-min transmission scan using a rotating [57Co] point source. The image reconstruction procedure consisted of a 3D ordered subset expectation maximization (OSEM, 4 iterations) and a 3D maximum a posteriori (MAP, 32 iterations) at a zoom factor of 1.0, and with scatter, attenuation, and decay corrections, resulting in a final voxel dimension of 0.78 x 0.78 x 0.8 mm. The origin of the Aβ-PET images was concealed from the operator for further analyses. Manual rigid-body coregistration of emission images to a 3T magnetic resonance imaging (MRI) mouse head template, followed by re-alignment of individual Aβ-PET images on an Aβ-PET template (baseline scan) or individual AβPET baseline image was accomplished using the PMOD fusion tool (version 3.5; PMOD Technologies Ltd.). VOI-based partial volume effect correction was subsequently performed with a full-width-at-half-maximum (FWHM) of 1.72 mm, and a ten-region VOI-mask, which includes two neocortical VOIs, one for cerebellum, one containing the rest of the cerebrum, and six extracerebral VOIs. For VOI-based 5 analyses, a cerebellar VOI comprising 64 mm³ and two bilateral frontal cortex VOIs comprising 15 mm³ each, were employed for calculation of Aβ-PET cortex-tocerebellum standardized-uptake-value-ratios (SUVRCTX/CBL), as described earlier 5 (Supplement Figure 2). Supplement Figure 2: MRI atlas in axial (upper image), sagittal (middle image), and coronal plane (lower image) with the superimposed cortical VOIs in orange, used for PET analyses. Histochemical analyses Cerebral hemispheres randomly selected for histochemistry were fixed by immersion in 4% paraformaldehyde at 4 °C for two days. A mean of five representative 50 µm thick slices per animal were then cut in the sagittal plane about 1.5 mm from the midline using a vibratome (VT 1000 S, Leica, Wetzlar, Germany). The slices were permeabilized overnight in 2% Triton X-100 in phosphate buffered saline (PBS; pH 7.4) at room temperature. The unbound dye was removed by three washing steps with PBS, and the slices were then mounted on microscope slides with fluorescent 6 mounting medium (Dako, Germany). 3D image stacks for each hemisphere were acquired on an anepi-fluorescence microscope (Axio Imager.M2 with ApoTome.2, Jena, Zeiss, Germany). Imaging of the whole slice was performed in tile scan mode, which allows automatic stitching of an array of fields of view. For this imaging, the methoxy-X04 was excited at 405 nm, and the emitted light was collected from 410 to 585 nm. The area and number of plaques were automatically counted using Imaris software (Imaris 7. 6.5; Bitplane, Zurich) in a region (0.27 ± 0.04 mm³) defined to match the frontal cortex VOI as applied for Aβ-PET image analysis. 7 Results Protein binding and p-glycoprotein RO5506284 exhibited a free fraction in mouse plasma of 0.47% and a mouse pglycoprotein efflux ratio of 1.5. This efflux ratio suggests that RO5506284 is not a substrate for mouse p-glycoprotein and therefore unlikely exposed to blood-brain barrier efflux 2. Safety aspects of the GSM in the chronic treatment arm During the course of the study, one animal of the TG-VEH group died at the age of 12 months due to interstitial pneumonia prior to start of treatment and the baseline Aβ-PET-scan. One animal of the TG-GSM group was lost at the age of 16.5 months due to lymphoma and two animals of the TG-VEH group died at the age of 17/17.5 months due to unknown reasons. In the WT-GSM group, two animals were lost during anaesthesia induction for baseline or 16 month follow-up Aβ-PET-scans, with no specific pathological finding. One animal of the WT-VEH group died at 15 months due to interstitial pneumonia. Therefore no treatment-specific side effect could be detected. 8 Correlation analyses across modalities Supplement Figure 3: Correlations across modalities indicate excellent agreement between A) final Aβ-PET estimates and plaque load, and B) insoluble Aβ 42 levels and plaque load. 9 Supplement References: 1. Caruso A, Alvarez-Sanchez R, Hillebrecht A, Poirier A, Schuler F, Lave T et al. PK/PD assessment in CNS drug discovery: Prediction of CSF concentration in rodents for Pglycoprotein substrates and application to in vivo potency estimation. Biochem Pharmacol 2013; 85(11): 1684-1699. 2. Schwab D, Fischer H, Tabatabaei A, Poli S, Huwyler J. Comparison of in vitro P-glycoprotein screening assays: recommendations for their use in drug discovery. J Med Chem 2003; 46(9): 1716-1725. 3. Lu Y, Barton HA, Leung L, Zhang L, Hajos-Korcsok E, Nolan CE et al. Cerebrospinal fluid betaAmyloid turnover in the mouse, dog, monkey and human evaluated by systematic quantitative analyses. Neurodegener Dis 2013; 12(1): 36-50. 4. Page RM, Baumann K, Tomioka M, Perez-Revuelta BI, Fukumori A, Jacobsen H et al. Generation of Abeta38 and Abeta42 is independently and differentially affected by familial Alzheimer disease-associated presenilin mutations and gamma-secretase modulation. J Biol Chem 2008; 283(2): 677-683. 5. Brendel M, Delker A, Rotzer C, Boning G, Carlsen J, Cyran C et al. Impact of partial volume effect correction on cerebral beta-amyloid imaging in APP-Swe mice using [(18)F]florbetaben PET. Neuroimage 2014; 84: 843-853. 10