Induction of inverted morphology in brain organoids by vertical-mixing bioreactors

Ethics statement

Generation and use of human iPSC were approved by the Ethics Committee of each institute. All methods were performed in accordance with the approved guidelines. Formal informed consent was obtained from a subject.

Maintenance of human iPSCs

Human iPSCs were maintained on a recombinant fragment of laminin-511 (iMatrix-511^TM, Nippi, Tokyo, Japan) by using StemFit AK02N medium (Ajinomoto, Tokyo, Japan). We used 201B7 iPSC line derived from fibroblasts of a healthy subject⁴¹ and APP1E111 iPSC line derived from fibroblasts of a patient with familial Alzheimer’s disease²⁹.

Generation of brain organoids by orbital mixing

iPSCs were cultured up to 80% confluency (typically 10 days after passage). iPSC colonies were dissociated into single cells after 4 min of incubation in 0.5 x Tryple Select/0.25 mM EDTA (Thermo Fisher Scientific, Waltham, MA) at 37 °C and suspended in StemFit AK02N with 10 µM of Y-27632 (Nacalai Tesque, Inc., Kyoto, Japan) to count the number and viability.

iPSCs were transferred into an embryoid body (EB) formation medium (EB medium) consisting of DMEM/F12 (Thermo Fisher Scientific) with 20% Knockout Serum replacement (Thermo Fisher Scientific), 3% Fetal bovine serum (FBS) (Thermo Fisher Scientific), 1% Glutamax (Thermo Fisher Scientific), 1% non-essential amino acids mix (NEAA, Thermo Fisher Scientific), 1% penicillin/streptomycin (Thermo Fisher Scientific), 4 ng/ml basic FGF (Wako Chemicals, Osaka, Japan), and 50 µM Y-27632. Dissociated 9000 alive iPSCs in EB medium were disseminated into single wells of U-bottom 96-well plates (ultra-low attachment type, NunclonTM SpheraTM microplates, 96U-Well Plate (174729), Thermo Fisher Scientific). The U-bottom 96-well plates were centrifuged at 200 × g for 3 min to make iPSCs aggregate quickly at the bottom of the well, and were kept in the incubator under a condition of 5% CO₂ at 37 °C. We defined the day of dissemination as Day 0. On Day 4, we replaced the medium with EB formation medium without basic FGF or Y-27632. On Day 6, we replaced the medium with neural induction medium consisting of DMEM/F12 with 1% N2 supplement (Thermo Fisher Scientific), 1% GlutaMAX, 1% NEAA, 1 μg/ml heparin (Nacalai Tesque, Inc.), and 1% penicillin/streptomycin. On Day 11, the outside of EBs became brighter and showed smooth edges. This appearance indicates that EBs contained radial organization. EBs were transferred into cold droplets of Matrigel^TM (Corning, Corning, NY) on a sheet of Parafilm (Parafilm® M PM996, Bermis, WI) with small 3-mm dimples in a 10-cm petri dish and were incubated for 20 min in an incubator at 37 °C to allow Matrigel polymerization. After the polymerization step, the EB-Matrigel droplets were removed from the Parafilm sheet and transferred into Neuroepithelial expansion medium, which consisted of a 1:1 mixture of DMEM/F12 and Neurobasal medium (Thermo Fisher Scientific) with 0.5% N2 supplement, 1% B-27 supplement without vitamin A Thermo Fisher Scientific), 1% GlutaMAX, 0.5% NEAA, insulin 2.5 μg/ml (I9278, Sigma), 1% penicillin/streptomycin, and 0.1% Amphotericin B (Thermo Fisher Scientific). On Day 15, 16 EB-Matrigel droplets were transferred into a 6-cm dish with 6 ml of Differentiation medium which consisting of 1:1 mixture of DMEM/F12 and Neurobasal medium with 0.5% N2 supplement, 1% B-27 supplement without AO (Thermo Fisher Scientific), 1% GlutaMAX, 0.5% NEAA, insulin 2.5 μg/ml, 1% penicillin/streptomycin, and 0.1% Amphotericin B. All 6-cm dishes were horizontally rotated on an orbital shaker (Cell Shaker, CS-LR 0081704-000, Taitec, Saitama, Japan), equipped inside the incubator, at a rotating speed of 60 rpm⁴². The culture medium was refreshed every 3 to 4 days.

From Day 40, EB-Matrigel droplets were cultured in differentiation medium with additional 1% Matrigel (growth factor reduced type, 354230, Corning). From Day 70, EB-Matrigel droplets were cultured in differentiation medium with additional 2% Matrigel (growth factor reduced type) and 2% B27 supplement without AO.

Culture brain organoids by vertical mixing

To establish vertical mixing under tight regulation of a stable temperature, pH, and dissolved O₂ concentration, we utilized the vertically mixing bioreactor system HiD 4×4 (Satake Co. Ltd, Tokyo, Japan), with the controlling system of cultivation condition, S-BOX×02 (Satake) (Supplementary Fig. 1).

iPSCs were cultured up to 80% confluence (typically 10 days after passage). iPSC colonies were dissociated into single cells after 4 min of incubation in 0.5 x Tryple Select/0.25 mM EDTA (Thermo Fisher Scientific) in a 37 °C incubator, and suspended in StemFit AK02N with 50 µM of Y-27632 (Nacalai Tesque, Inc.) for number counting and viability determination.

Dissociated 2.5 × 10⁶ cells in 250 ml of StemFit AK02N plus 50 µM of Y-27632 were disseminated into a single-use bottle (Supplementary Fig. 1) specialized for suspension culture under continuous vertical mixing of the HiD 4×4 system (Satake) at a setting of 60 mm/s speed and 15-mm stroke with continuous air flow of 30 ml/s. After 7 days of cultivation in StemFit AK02N medium, disseminated iPSCs formed spheres of homogeneous size (~100–150-μm diameter), and the medium was refreshed with EB formation medium as described above. We defined the day of medium refreshment as Day 0. On Day 6, we replaced the medium with Neural induction medium, as described above. On Day 17, we replaced the medium with Neuroepithelial expansion medium, as described above. On Day 25, we replaced the medium with Differentiation medium, also as described above. Sphere cultivation in the vertical mixing system was continued up to Day 90, with weekly medium change under tight regulation of pH 6.5–7.5, O₂ concentration 20%, and temperature 32–34 °C.

Histological and immunohistochemical analysis

Tissues were fixed in 4% paraformaldehyde for 15 min (1-mm organoids) or 30 min (3-mm organoids) at room temperature, followed by three times washing in PBS for 10 min. Tissues were allowed to sink in 30% sucrose overnight and were then embedded in O.C.T. compound (Sakura Finetek, Tokyo, Japan) and quickly frozen in liquid N₂. Frozen tissues were cut into 12-µm slices by cryostat (CM1850, Leica Biosystems, Wetzlar, Germany) at −18 to −20 °C. For immunohistochemistry, sections were permeabilized in 0.5% Triton-X100/PBS (0.5% PBST) for 30 min at room temperature, and were then blocked in blocking solution consisting of 0.1% PBST with 10% normal donkey or goat serum for 2 h at room temperature. Sections were then incubated with primary antibodies in blocking solution at 4 °C overnight. These antibodies were used: anti-SOX2 (MAB2018, R&D System, 1:1000), anti-β3-Tubulin (D71G9, Cell Signaling Technology, 1:500), anti-MAP2 (ab5392, Abcam, 1:3000), anti-TBR2 (ab23345, Abcam, 1:1000), anti-CTIP2 (ab18645, Abcam, 1:1000), anti-N-cadherin (C3865, Sigma, 1:1000), anti-ARL13B (17711-1-AP, Proteintech, 1:400), anti-Pericentrin (ab28144, Abcam, 1:200), anti-SOX2 (14-9811-82, eBioscience, 1:400), anti-NKX2.1(MAB5460, Merk, 1:500), anti-GABA (A2052, Sigma, 1:300), anti-VGLUT1(135303, Synaptic Systems, 1:150), anti-PROX1(ab199359, Abcam, 1:500), anti-Aβ oligomer specific antibody NU1^29,43 (1:500). After washing with 0.1% PBST four times, 15 min/time, samples were incubated with secondary antibodies of Alexa Flour 405/488/546/594 or 647 conjugates (Invitrogen, 1:1000) for 2 h at room temperature while protected from light. Then, samples were washed with 0.1% PBST four times, 15 min/time, and mounted with Prolong Gold mounting (ProLong™ Gold Antifade Mountant, Thermo Fisher Scientific) and kept stable in the dark at room temperature for about 20 h. Data were observed under fluorescent microscopy (confocal FV1000, Olympus or IN Cell analyzer 6000, GE Healthcare).

ImageJ software (NIH, USA) was used to measure the percentage of SOX2-positive cells in the peripheral area of brain organoids and the direction of primary cilia.

Computational fluid dynamic analysis

Physical flow simulations in a cell culture dish/bottle were performed using CFD software, and ANSYS Fluent 2019 R3 (ANSYS, Inc.) was used for calculation. For verification of the turbulence model, we performed the Realizable κ-ε model. For gas–liquid boundary tracking method with the cell culture dish, we used the VOF model. For liquid boundary tracking method with a bottle, we used the Slip wall boundary condition.

For calculating the drag force, we used the following equation:

D: drag (N); ρ_f: fluid density (kg/m³); µ_f: fluid viscosity (Pa-s); V: Relative velocity of fluid and particle (m/s); S: Cross-section of particle (m²); C_D: Drag coefficient (–); Re: Reynolds number (–); u: particle movement speed (m/s); d: particle size (m).

MEA recordings

The organoids were each incubated with accumax for 15 min at 37 °C and dissociated with slow pipetting to 2–3 clumps, and cell mixtures from independent individual organoids including clumps and dissociated cells were cultured on 64-channel MEA chips (MED-R515A; Alpha Med Scientific) coated with Polyethyleneimine (Sigma) and Laminin-511 (Nippi) at 37 °C in 5% CO₂. For culture on MEAs, cell mixture from one organoid was plated on electrodes of a dish as a droplet with 100 μl of medium for 2 h, and then 1 ml of medium was added. Half of the medium was changed every 2 days. After 6 weeks, spontaneous extracellular field potentials were acquired using a 64-channel MEA system (MED64-Basic; Alpha Med Scientific).

Single-cell RNA sequencing

Dissociated cells were resuspended in PBS containing 1% BSA, immediately followed by a library preparation targeting single cells using the Chromium Single Cell 3′ Reagent Kit v3 (10×Genomics) and Chromium Controller (10×Genomics) according to the manufacturer’s instructions. Three organoids generated by orbital mixing and three organoids generated by vertical mixing were analyzed. Two thousand cells from each organoid were targeted, and a total of 12,000 cells were analyzed. The library was sequenced on a HiSeq2500 TruSeq SBS v3 reagent. Cell-specific FASTQ files were generated by deconvolution of UMIs and cell barcodes using bcl2fastq 2.20.0.422 (Illumina). Alignment to the human reference genome GRCh38 and UMI counting were conducted by Cell Ranger v3.1.0 pipeline (10×Genomics). Uniform manifold approximation and projection (UMAP) implemented in the Seurat package v3.2.3 was conducted using 1st to 10th principal components after filtering out the cells with a high number of detected genes (≥9000) and RNA molecules (≥60,000), and with a high percentage of mitochondrial genes (>60%). The differentially expressed genes were identified using the Wilcoxon rank-sum test and heat map was plotted using the pheatmap R package version 1.0.12.

Statistics and reproducibility

Results were analyzed using student’s two-tailed t-test or one-way ANOVA followed by Dunnett’s post hoc test to determine statistical significances of the data. Differences were considered significant at p < 0.05. Analyses were performed using GraphPad Prism software version 8.0 (GraphPad Software, San Diego, CA).

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.