Derivation of Brain Endothelial-like Cells (iBECs) from hiPSCs and TEER Measurements
iBECs were derived using the method of Neal et al . Human iPSCs (hiPSCs) from the GM25256 line (Coriell Institute) were maintained on plates coated with Matrigel (Corning, cat no. 356230) in E8 Flex medium (Thermo Fisher Scientific, cat no. A28585-01). The day before differentiation was initiated, hiPSCs were dissociated into single cells with Accutase (Thermo Fisher Scientific, cat no. A1110501) and plated onto Matrigel-coated plates at a density of 15 x 103 cells/well in E8 Flex medium supplemented with 10 µM Rho-associated protein kinase (ROCK) inhibitor Y-27632 (R&D Systems, cat no. 1254). To initiate differentiation, the medium was changed to E6 (Thermo Fisher Scientific, cat no. A1516401) and E6 medium changes continued daily for 3 more days. Then, the medium was changed to human endothelial serum-free medium (HESFM, Thermo Fisher Scientific, cat no. 11111044) supplemented with 20 ng/mL basic fibroblast growth factor (bFGF; Peprotech, cat no. 100-18B), 10 µM retinoic acid (RA; Sigma, cat no. R2625), and 1% B27 supplement (Thermo Fisher Scientific, cat no. 17504044). After two days, iBECs were dissociated with Accutase and subcultured onto 24-well transwell permeable inserts (Corning, cat no. 3470) or tissue culture plates (Corning, cat no. 3513, 3548) coated with 1 mg/mL Collagen IV (Sigma, cat no. C5533) and 5 mM Fibronectin (Sigma, cat no. F1141) in HESFM + 20 ng/mL bFGF, 10 µM RA, and 1% B27 (day 0). 24 h later, the medium was changed to HESFM + 1% B27 without bFGF or RA, and resistance () values for monolayers of iBECs seeded on transwells were obtained for day 1 using an EVOM2 Voltohmmeter (World Precision Instruments, Sarasota Florida) coupled to an ENDOHM cup chamber. Transendothelial electrical resistance (TEER) was calculated by subtracting the resistance () value of a blank transwell and multiplying by the transwell surface area (0.33 cm2). TEER measurements occurred daily to confirm the integrity of the monolayers. Experiments were only conducted on iBECs with TEER > 1000 *cm2.
iBECs plated on 48-well plates were washed once with PBS (Thermo Fisher Scientific, cat no. 70011044) and fixed in a 1:1 methanol/acetone mixture for 15 min at 4°C. Wells were washed with PBS 3x for 5 min each, then blocked with 5% normal donkey serum (Jackson ImmumoResearch, cat no. 017-000-121) + 0.1% TX-100 (Sigma, cat no. X100) in PBS for 1 h at RT. Wells were washed 3x for 5 min each, then iBECs were incubated with primary antibody solutions (see Table 1 for dilutions) in phosphate-buffered normal antibody diluent (NAD) (Scytek, cat no. ABB500) overnight at 4°C. Wells were washed 3x for 5 min each, then incubated with secondary antibody solutions (all 1:200) and DAPI (Thermo Fisher Scientific, cat no. 62248) (1:5000) in NAD for 1 h at RT. Wells were washed 3x for 5 min each, then imaged using a Zeiss Axiovert 7. For each differentiation, 3 wells were designated per group and 3-4 images were taken per well. Mean fluorescence intensity (MFI) values were quantified using the Zen image analysis software and were corrected for differences in cell density via normalization to DAPI MFI.
BrdU Proliferation Assays
Analysis of 5-bromo-2'-deoxyuridine (BrdU) (Thermo Fisher Scientific, cat no. B23151) incorporation into the DNA was used to evaluate iBEC proliferation. For iBECs fixed on day 2, 10 µM BrdU was included with the medium change to HESFM + 1% B27 without bFGF or RA on day 1. For iBECs fixed on day 4, 5, 7, or 9, BrdU was added to a final concentration of 10 µM by diluting a 100 µM BrdU solution into the existing medium on day 3, 4, 6, or 8, respectively. 24 h after addition of 10 µM BrdU, cells were washed 1x quickly with PBS and fixed in ice-cold 1:1 methanol/acetone fixative for 15 min. at 4°C. Cells were permeabilized with 0.1% TX-100 in PBS for 20 min at RT. DNA strands were denatured with 2N HCl for 10 min at RT. Then, the IFA for was performed according to the procedure outlined above.
Protein Extractions and Immunoblots
iBECs plated on 12-well plates were washed in ice-cold PBS and scraped in a RIPA-like buffer (1% cytoplasmic stock, 1% NP-40, 0.5% Deoxycholate, 0.1% SDS) supplemented with protease (Sigma, cat no. P8340) and phosphatase (Sigma, cat no. P5726) inhibitors. Lysates were frozen at -80°C, then thawed on ice and centrifuged at 20000xg for 5 min at 4°C. The supernatant was saved and protein concentrations for each cell lysate were determined by Bradford assay (Thermo Fisher Scientific, cat no. 23200) using known concentrations of bovine serum albumin to create the protein standard curve. Samples were prepared for electrophoresis by mixing with 4X LDS Sample Buffer (Novex, cat no. NP007), 10X Sample Reducing Agent (Novex, cat no. NP0009), and molecular biology-grade water, then denatured at 70°C for 10 min. Protein mixtures containing 10 mg of protein were electrophoresed using ExpressPlus PAGE precast gels (GeneScript, cat no. M41210), then electrotransferred to nitrocellulose membranes (Invitrogen, cat no. IB301002) using the iBlot Dry Blotting System (Invitrogen). Nonspecific binding was blocked by incubating the membrane in Tris-buffered saline 0.1% Tween-20 (TBS-T) supplemented with 5% BSA (Sigma, cat no. A7030) at RT for 45 min. Primary antibody solutions (see Table 1 for dilutions) were prepared in 5% BSA/TBS-T and membrane incubations occurred overnight at 4°C. Membranes were washed with TBS-T three times for 5 min each, then incubated with horseradish peroxidase conjugated secondary antibodies (all 1:5000) at RT for 30 min. Then, membranes were washed with TBS-T three times for 5 min each followed by one TBS wash for 1 min. West Pico chemiluminescence reagent (Thermo Fisher Scientific, cat no. PI-34078) was applied, and bands of immunoreactive protein were visualized with the ImageQuant LAS4000 (Cytiva Life Sciences, formerly GE Healthcare Life Sciences) Western blot Imaging system or the Amersham™ ImageQuant 800 (Cytiva Life Sciences, formerly GE Healthcare Life Sciences). Quantification of the band volumes was performed with ImageQuant TL 1D Gel Analysis Software (Cytiva Life Sciences, formerly GE Healthcare Life Sciences).
Glucose Transport Assays
Transwells were distributed such that the mean TEER values were approximately equal among groups. The luminal and abluminal HESFM + 1% B27 mediums were refreshed, and cells were equilibrated in the incubator for 20 min. The input medium consisted of HESFM + 1% B27 containing 0.1 µCi 14C-2 deoxy-D-glucose (14C-DG, FW 164.16 g/mol) (Perkin Elmer) and 1 million CPM 99mTecnicium diethylenetriaminepentaacetic acid (99mTc-DTPA, FW 487.21 g/mol) per transwell. As 2-deoxy-D-glucose is transported at nearly the same rate as glucose and is trapped intracellularly after hexokinase phosphorylation, it is an appropriate tracer to measure transcellular glucose transport . The Pe for 99mTc-DTPA was used to measure non-specific para- and transcellular leakage. To evaluate the saturability of the iBEC glucose transport system, the input medium was supplemented with D (+)-Glucose (Sigma, cat no. G7528) or D-Mannitol (Sigma, cat no. M4125). BAY-876 (Tocris, cat no. 6199) was obtained as a powder and stored at -20°C in a glass vial. Stock solutions (2 mM) were prepared by diluting BAY-876 in DMSO (Sigma, cat no. D2650) (vehicle control) and were stored in 25 µL aliquots at -70°C. The input medium was supplemented BAY-876 to a final concentration of 2 µM to evaluate GLUT1 inhibition. For assays evaluating luminal-to-abluminal (blood-to-brain) transport, the 100 µL luminal (donor) chamber volume was switched to the input medium to initiate the assay, and 500 µL volumes of medium from the abluminal (receptor) chamber were collected and replaced with fresh, pre-warmed HESFM + 1% B27 medium after incubation times of 5, 10, 15, and 20 min. Of the 500 µL total volume collected, 200 µL was added to liquid scintillation vials containing 5 mL of EcoscintTM (National Diagnostics). The vials were dark-adapted for at least 4 days. Then, the radioactivity was counted in a Tricarb beta counter (Perkin Elmer) to measure 14C-DG levels. The radioactivity in the remaining 300 µL was counted in a Wizard 2 gamma counter (Perkin Elmer) to measure 99mTc-DTPA levels. For abluminal-to-luminal (brain-to-blood) transport assays, the 600 µL abluminal (donor) chamber volume was switched to an input medium to initiate the assay, and 90 µL volumes of medium from the luminal (receptor) chamber were collected and replaced with fresh, pre-warmed HESFM + 1% B27 medium after incubation times of 5, 10, 15, and 20 min. Of the 90 µL collection, 40 µL was counted in the beta counter and the remaining 50 µL was counted in the gamma counter. Permeability-surface area coefficient (Pe) calculations were performed according to the method of Dehouck et al., 1990 . Clearance was expressed as µL of radioactive tracer transported from the donor chamber to the receptor chamber, and was calculated from the initial level of radioactivity added to the donor chamber and the final level of radioactivity in the receptor chamber:
Clearance (µL) = [C]C x VC/[C]L
Where [C]L is the initial concentration of radioactivity in the donor chamber (in units of CPM/µL), [C]C is the concentration of radioactivity in the receptor chamber (in units of CPM/µL) and Vc is the volume of the receptor chamber in mL. The volume cleared was plotted vs. time, and the slope was estimated by linear regression. The slopes of clearance curves for the iBEC monolayer plus Transwell® membrane was denoted by PSapp, where PS is the permeability × surface area product (in µL/min). The slope of the clearance curve for a Transwell® membrane without iBECs was denoted by PSmembrane. The PS value for the iBEC monolayer (PSe) was calculated from 1 / PSapp = 1 / PSmembrane + 1 / PSe. The PSe values were divided by the surface area of the Transwell® inserts (0.33 cm2) to generate the endothelial permeability coefficient (Pe, in µL/min/cm2).
Recombinant human tumor necrosis factor (TNF)-α (R&D Systems, cat no. 210-TA) was obtained as a lyophilized powder and stored at -20°C in a glass vial. Stock solutions (100 µg/mL) were prepared by reconstituting in sterile PBS and were stored in 25 µL aliquots at -70°C. Stock solutions were diluted in the existing luminal and abluminal mediums for 100 ng/mL TNF-α treatments. After 16 h, TEER was measured.
Derivation of Neurons from hiPSCs and Conditioned Medium Treatments
We used previously established gene-edited hiPSC lines harboring the amyloid precursor protein (APP) Swedish mutation (Swe) and as well as the isogenic control described in Young et al., 2018 . This mutation corresponds to a 2bp change, GA to TC, in exon 16 of APP. We used cell heterozygous for this mutation, APPSwe/+, along with the isogenic APP wild-type, APPWT. Neurons from these cell lines were differentiated from hiPSCs as previously described in Knupp et al., 2020  and Rose et al., 2018 . hiPSC-neurons from the APPSwe/+ and APPWT lines were cultured in Neural Differentiation media (1:1 DMEM/F12 (Life Technologies, cat no. 11039047) + neurobasal media (GIBCO, cat no. 21103049), 0.5% N2 supplement (Thermo Fisher Scientific, cat no. 17502-048) 1% B27 supplement (Thermo Fisher Scientific, cat no. 17504-044), 0.5% GlutaMax (Thermo Fisher Scientific, cat no. 35050061), 0.5% insulin-transferrin-selenium (Thermo Fisher Scientific, cat no. 41400045), 0.5% NEAA (Thermo Fisher Scientific, cat no. 11140050), 0.2% β-mercaptoethanol (Life Technologies, cat no. 21985023), 0.2 μg/mL brain-derived neurotrophic factor (Eurotech, cat no. 450-02), 0.2 µg/mL glial-cell-derived neurotrophic factor (PeproTech, cat no. 450-10), and 0.5 M dbcAMP (Sigma Aldrich, cat no. D0260) and cultured on Matrigel (Corning, cat no. 356231) at 37°C in a 5% CO2 incubator. After 21 days of differentiation in Neural Differentiation media, conditioned medium (CM) was collected from each cell line following 72 hours of contact with the mixed neuronal cell population. Aβ peptide content in CM was quantified using an Aβ Triplex ELISA plate (Meso Scale Discovery, cat no. 151200E-2). For iBEC conditioned medium (CM) treatments, transwells were distributed among groups (APPWT or APPSwe/+) such that mean TEER values were approximately equal. The abluminal medium was replaced with CM collected from APPWT or APPSwe/+ neuronal cultures. After 24 h, TEER was measured.
The Prism 9.0 statistical software package was used for all statistical calculations (GraphPad Inc, San Diego, CA). Means are reported with their standard error (SE). For all figures, means are displayed with their SE and n. Linear regression lines and their slopes and intercepts were calculated using the Prism 9.0 software. Unpaired two-tailed t-tests were used to compare two means and analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test when more than two means were compared.