Method Article
* Wspomniani autorzy wnieśli do projektu równy wkład.
This protocol presents a method for generating vascular organoids using human pluripotent stem cells.
Vascular organoids derived from human induced pluripotent stem cells (hiPSCs) recapitulate the cell type diversity and complex architecture of human vascular networks. This three-dimensional (3D) model holds substantial potential for vascular pathology modeling and in vitro drug screening. Despite recent advances, a key technical challenge remains in reproducibly generating organoids with consistent quality, which is crucial for downstream assays and applications. Here, a modified protocol is presented that improves both the homogeneity and reproducibility of vascular organoid generation. The modified protocol incorporates the use of microwells and the CEPT cocktail (chroman 1, emricasan, polyamines, and the integrated stress response inhibitor, trans-ISRIB) to improve embryoid body formation and cell survival. Differentiated, mature vascular organoids generated using this protocol are characterized by whole-mount 3D immunofluorescence microscopy to analyze their morphology and complex vasculature. This protocol enables the production of high-quality vascular organoids in a scalable manner, potentially facilitating their use in disease modeling and drug screening applications.
In vitro microphysiological models, including organoid and tissue-on-a-chip systems, have emerged over the past decade1,2,3. These three-dimensional (3D), human cell-derived systems address the limitations in conventional two-dimensional (2D) cell culture and animal models, such as the lack of many components in the physiological microenvironment and genetic differences across species, respectively. They provide more physiological insights and are more suitable for disease modeling and drug screening than conventional models. Among the microphysiological models, organoids derived from human induced pluripotent stem cells (hiPSCs) have been proven to recapitulate the architectural features of native human tissues effectively and have been developed to mimic different human organs, including the brain4,5, kidney6, liver7,8 and retina9. Here, we particularly focus on the generation of vascular organoids from hiPSCs and outline a streamlined protocol that aims to minimize the variations between different organoid samples and batches, thereby improving the overall reproducibility.
Vasculature plays a crucial role in maintaining physiological homeostasis across all human organs. The formation of vasculature involves the processes of vasculogenesis and angiogenesis, orchestrated through the interactions between endothelial and mural (e.g., pericytes and vascular smooth muscle cells) cells and influenced by various stimuli10. Penninger, Wimmer and colleagues developed the first vascular organoid generation method11,12 that mimics these two processes. With the organoids generated from this method, their group12 and us13,14 have demonstrated the potential of vascular organoids for modeling vascular malfunctions in diseases. For example, in our recent works13,14, distinct phenotypes were observed, such as vessel sprouting and mural cell composition variations, between disease-mimicking vascular organoids and their wild-type counterparts. These examples highlight that the vascular organoid model could serve as a platform for drug screening by mimicking disease-related vascular malfunctions.
Built upon the initial vascular organoid generation works11,12, a revised protocol is presented that includes the use of microwell to facilitate homogenous embryoid body (EB) formation, a critical factor in minimizing the variations often seen within the same organoid generation batch. Additionally, the CEPT cocktail, a combination of chroman 1, emricasan, polyamines and the integrated stress response inhibitor (trans-ISRIB)15, is employed to improve the viability of hiPSCs and differentiated cells during the EB formation process. This modification is mainly to reduce the variations between different organoid samples and batches. Uniform vascular organoids can be produced with this roughly two-week-long protocol, where vasculogenesis is induced to help endothelium organize into primitive tubular networks on Day 1, followed by angiogenesis induction on Day 5 to develop complex vascular networks in organoids. On Day 12-15, the generated organoids should show mature vascular networks composed of both endothelium and mural cells.
Figure 1 presents an overview of the steps involved in this protocol. Detailed information regarding the reagents and materials utilized in this protocol is given in the Table of Materials. It is recommended that all the media should be prewarmed at 37 °C before use unless specified otherwise. All the cell culture materials and supplies should be either autoclaved or sterile, and cell handling should be done in a biosafety cabinet. Additionally, the pH value of the collagen I mixture should be carefully adjusted to pH 7.3 to avoid possible solidification issues.
1. Human iPSC maintenance
2. EB formation (Day 0)
3. Vascular differentiation (Day 1 - 5)
4. Hydrogel embedding (Day 5)
5. 3D Immunofluorescent assay
Figure 1A presents the scheme of this protocol, including the key components used in each stage. The differentiation started with the EB formation, followed by mesoderm induction and vascular lineage specification (Figure 1B-D). The aggregates grew larger and less spherical over time. Organoids started showing rough edges on Day 5. After embedded in the collagen I-basement membrane matrix mixture, vascular endothelial cells and pericytes sprouted and outgrew to form complex vascular networks as early as Day 7 (Figure 1E). As the main purpose of this modified method is to minimize the variations between different generation batches, we measured the EB size from two independent batches, and both size measurements and the statistical analysis confirmed the consistency and uniformity with this method (Figure 1F).
To validate the vascular differentiation, organoids were collected on Day 5 before embedding and stained with vascular markers, including CD31, ERG1 and PDGFRβ. The organoids expressed CD31, ERG1 and PDGFRβ (Figure 2), indicating the presence of vascular endothelial cells and pericytes.
After tissue clearing, the vascular networks were analyzed with whole-mount immunofluorescence microscopy (Figure 3A). Mature organoids were stained with CD31, ERG1 and PDGFRβ on Day 17 (Figure 3B, sample in the gel block) and Day 28 (Figure 3C, the sample was retrieved from the gel block and culture individually in a 96-well plate; note that the vascular network at the edges wrapped the spheroids instead of radially expanding after retrieved from the gel block).
Figure 1: Vascular organoid generation. (A) Scheme of the steps and timeline for vascular organoid generation. (B-E) Representative bright field images of vascular organoids at different stages. Scale bars: 100 µm. (F) Statistical analysis in EB size from two different EB generation batches using the microwell method presented in this work. For EB generation, 1,000 cells per microwell were used, and for Batches 1 and 2, n = 42 and n = 40, respectively. Significance is presented as ns, not significant. Please click here to view a larger version of this figure.
Figure 2: Immunostaining of vascular organoids before gel embedding. Representative fluorescent images of organoids stained with the endothelial markers (A) CD31 (green) and (B) ERG1 (green) as well as (C) the pericyte marker, PDGFRβ (green), respectively. Nuclei were counterstained with DAPI (blue). Scale bars: 50 µm. Please click here to view a larger version of this figure.
Figure 3: Immunofluorescent characterizations of the vascular network in vascular organoids. (A) A representative image shows gel-embedded vascular organoids in the glass dish after tissue clearing. The organoids were transparent but visible under a 385 nm UV light. (B) A representative 3D whole-mount image of vascular organoid on Day 17. The vascular organoid was stained with DAPI (blue), the vascular endothelial cell marker, CD31 (green) and the pericyte marker, PDGFRβ (magenta). (C) A representative whole-mount 3D imaging of a retrieved vascular organoid on Day 28, stained with the endothelial markers CD31 (red) and ERG1 (green). Scale bars: 150 µm. Please click here to view a larger version of this figure.
hiPSC expansion medium | 500 mL |
mTeSR Plus basal medium | 400 mL |
mTeSR Plus supplement | 100 mL |
Gentamicin (50 mg/mL) | 1 mL |
Table 1: Composition of hiPSC expansion medium.
hiPSC seeding medium | 20 mL |
PSC expansion medium | 20 mL |
Chroman 1 (100 µM) | 10 µL |
Emricasan (1 mM) | 30 µL |
polyamine supplement (1000x) | 20 µL |
trans-ISRIB (1 mM) | 14 µL |
Table 2: Composition of hiPSC seeding medium.
N2B27 medium | 100 mL |
DMEM/F-12, GlutaMAX supplement | 48.6 mL |
NeuroBasal medium | 48.6 mL |
GlutaMAX supplement (100x) | 0.5 mL |
MEM Non-Essential Amino Acids Solution (NEAA, 100x) | 0.5 mL |
2-Mercaptoethanol (1000x) | 0.1 mL |
Gentamicin (50 mg/mL) | 0.1 mL |
N2 supplement (100x) | 0.5 mL |
B27 supplement minus vitamin A (50x) | 1 mL |
Heparin, 4 kU/mL | 0.1 mL |
Table 3: Composition of N2B27 medium.
Mesoderm induction medium (MIM) | 3 mL |
N2B27 medium | 3 mL |
BMP4 (100 ng/µL) | 0.9 µL |
CHIR99021 (12 mM) | 3 µL |
Table 4: Composition of mesoderm induction medium (MIM).
Vascular differentiation medium 1 (VDM1) | 3 mL |
N2B27 medium | 3 mL |
Forsklin (12 mM) | 0.5 µL |
VEGFA (100 ng/µL) | 3 µL |
Table 5: Composition of vascular differentiation medium 1 (VDM1).
Collagen I mixture | 5 mL |
Collagen I (3.61g/mL) | 2.77 mL |
10x DMEM/F12 medium | 0.5 mL |
Sodium bicarbonate, 7.5% | 0.5 mL |
HEPES, 1 M | 0.5 mL |
H2O | 0.73 mL |
NaOH, 1M | 75 µL |
Table 6: Composition of collagen I mixture.
Collagen I - basement membrane matrix mixture | 6.5 mL |
Collage I mixture | 5 mL |
Matrigel (10 mg/mL) | 1.5 mL |
Table 7: Composition of collagen I-basement membrane matrix mixture.
Vascular differentiation medium 2 (VDM2) | 5 mL |
DMEM/F-12, GlutaMAX™ supplement | 3.67 mL |
Gentamicin (50 mg/mL) | 5 µL |
MEM Non-Essential Amino Acids Solution (NEAA, 100X) | 50 µL |
2-Mercaptoethanol (1000x) | 5 µL |
Heparin, 4 kU/mL | 5 µL |
Knockout Serum Replacement | 0.75 mL |
FBS | 0.5 mL |
VEGFA (100 ng/µL) | 5 µL |
FGF2 (100 ng/µL) | 5 µL |
Table 8: Composition of vascular differentiation medium 2 (VDM2).
Blocking buffer | 50 mL |
Donkey serum | 2.5 mL |
Tween 20 | 0.25 mL |
Triton X-100 | 0.25 mL |
Sodium deoxycholate solution (1%, wt/v) | 0.5 mL |
PBS | 46.5 mL |
Table 9: Composition of blocking buffer for immunostaining.
The described protocol includes two key modifications for homogenous and reproducible generation of high-quality vascular organoids. The first modification is the use of a microwell plate to enable better control of the EB uniformity. As a starting point of vascular differentiation, the size of EBs is critical as it regulates the stem cell fate and differentiation efficacy towards different lineages. Variations in EB size usually lead to heterogeneous organoids with inconsistent structure and organization16,17. The conventional EB generation method by lifting hiPSC colonies11,12 makes it difficult to control the cell numbers for individual EBs and usually results in heterogeneous EB formation. In this protocol, hiPSCs are dissociated into single-cell suspension and reaggregated in microwells16,18. By adjusting the seeding cell numbers, the size of EBs can be optimized to ensure consistent and effective mesodermal differentiation. For microwell seeding, 500-1,000 cells are used, and this can yield relatively small EBs (with a diameter of 100-200 µm). This method could reproducibly generate uniform EBs with desired sizes from batch to batch (Figure 1F) and among various cell lines13. Since Day 0, the morphology of the EB has changed over time during the differentiation: the shape becomes less spherical, and the surface becomes rougher (Figure 1), but the volume does not increase significantly. The 3D vasculature morphology depends on the embedding density and the sizes of the organoids embedded. Too-large (>500 µm) organoids usually result in insufficient vascular cell differentiation, while too-small organoids tend to have underdeveloped vasculature.
The second modification is incorporating the CEPT cocktail to improve the hiPSC viability during the EB formation. The cell viability is crucial in addition to the quality of hiPSC colonies during this process. When hiPSCs are dissociated into a single-cell suspension, the ROCK inhibitor Y27632 is usually added to improve cell viability. However, Y27632 has been shown suboptimal for hiPSCs in low density and may cause undesired membrane stress, while the CEPT cocktail (composed of chroman 1, emricasan, polyamines and trans-ISRIB) could significantly reduce cell stress and improve the viability during this EB formation process with hiPSCs in a lower density15.
Prior to organoid embedding in the gel, the organoids are maintained in the microwell plate, and the medium change should be gentle to avoid possible disruption. It would be better to use a 200 µL pipette to change the medium for the first 5 days. All the media used for vascular organoid generation must be prewarmed before use. Especially after gel embedding, the use of cold media could partially liquify the basement membrane matrix and affect the 3D hydrogel niche for vascular network formation. The embedding density and even distribution of the organoids are also critical factors for vascular network formation. Dense organoid embedding could potentially lead to excessive fusion of vascular networks, while low embedding density could lead to insufficient network formation. Seeding 60-80 organoids per well and gently moving the plate back and forth immediately after adding them to the first layer of the 3D gel can better distribute these organoids.
This protocol can be further modified for different applications. For example, blood-brain barrier (BBB) cues can be applied to further restrain the vascular endothelial cells from the formation of the BBB interface19,20,21,22,23,24,25. The 3D hydrogel composition can be adjusted as well. For example, fibrin and vitronectin are commonly used for vascular network formation in vitro25,26,27,28,29,30,31,32. The basement membrane matrix without collagen I may be used for embedding, which may simplify the embedding process and provide different mechanical stimuli.
Although the vascular organoids generated from this protocol could be potentially used for disease modeling or drug screening applications, one limitation is that these organoids only recapitulate mostly the capillary vasculature with few larger vessels. For research focusing on larger vessels, organoid generation may require additional mechanical stimuli24,31,32,33. Connecting the vascular organoids with recirculation, such as transplantation to animals34 or micro-/macro-fluidic devices with perfusion pumps33, could potentially address this.
The authors declare no competing financial interest.
We would like to acknowledge the technical support from the Confocal and Specialized Microscopy Shared Resource of Herbert Irving Comprehensive Cancer Center at Columbia University, funded in part through NIH Center Grant (P30CA013696). This work is supported by NIH (R21NS133635, Y.-H.L.; UH3TR002151, K. W. L.). Figure 1A was created using BioRender.
Name | Company | Catalog Number | Comments |
2-Mercaptoethanol (1000x) | Gibco | 21985023 | |
Accutase | Sigma-Aldrich | A6964 | |
Alexa Fluor 488 AffiniPure F(ab')2 Fragment Donkey Anti-Rabbit IgG (H+L) | Jackson Immuno Research Labs | 711-546-152 | reconstitute with 0.25 mL 50% glycerol, store at -20 °C for up to 1 year, dilution ratio for use: 1:1000 |
Alexa Fluor 647 AffiniPure F(ab')2 Fragment Donkey Anti-Goat IgG (H+L) | Jackson Immuno Research Labs | 705-606-147 | reconstitute with 0.25 mL 50% glycerol, store at -20°C for up to 1 year, dilution ratio for use: 1:1000 |
B27 supplement minus vitamin A (50x) | Gibco | 12587010 | thaw at 4 °C, aliquot and store at -20 °C, avoid freeze-thaw cycle |
BMP4 | ProSpec | CYT-081 | reconstitute with sterile ddH2O at a concentration of 100 ng/µL, aliquot and store at -20 °C |
CD31 antibody | abcam | ab28364 | dilution ratio: 1:400 |
Centrifuge | Eppendorf | 022626001 | |
CHIR99021 | Tocris Bioscience | 4423/10 | make the stock solution at 12 mM with DMSO, and store at -20 °C for up to 1 year. |
Chroman 1 | Tocris | 7163/10 | make the stock solution at 100 µM with DMSO, and store at -20 °C for up to 1 year. |
Collagen I | Corning | 40236 | |
Confocal Microscope | Nikon | AXRMP/Ti2 | |
Corning Elplasia Plates | Corning | 4441 | |
Countess II FL automated cell counter | Thermo Fisher | AMQAF1000 | |
DMEM/F-12, GlutaMAX supplement | Gibco | 10565018 | |
DMEM/F-12, powder | Gibco | 12500062 | make 10x stock solution with biograde ddH2O and sterilize it with a 0.2 µm filter. Store at 4 °C. |
Edi042A | Cedars Sinai | ||
Emricasan | Medchemexpress LLC | HY1039610MG | make the stock solution at 1 mM with DMSO, and store at -20 °C for up to 1 year. |
ERG antibody | Cell Singali Technology | 97249 | dilution ratio: 1:400 |
Fetal Bovine Serum - Premium | Atlanta Biologicals | S11150 | thaw at 4 °C and aliquot, store at -20 °C for up to five years, or store at 4 °C for up to a month |
FGF2 | ProSpec | CYT557 | reconstitute with sterile ddH2O at a concentration of 100 ng/µL, aliquot and store at -20 °C |
Fluorodish Cell Culture Dish | World Precision Instruments | FD35-100 | |
Forskolin | Tocris Bioscience | 1099 | reconstitute with DMSO at a concentration of 12 mM, aliquot and store at -20 °C |
Gentamicin (50 mg/mL) | Gibco | 15710064 | |
GlutaMAX supplement (100x) | Gibco | 35050061 | |
Heparin, 100 kU | MilliporeSigma | 375095100KU | reconstitute with sterile ddH2O at a concentration of 4 kU/mL |
HEPES (1 M) | Gibco | 15630080 | |
Incubator | Thermo Scientific | 13998123 | |
Knockout Serum Replacement | Gibco | 10828028 | thaw at 4 °C, aliquot and store at -20°C, avoid freeze-thaw cycle |
Matrigel, GFR Basement Membrane Matrix | Corning | 356230 | thaw at 4 °C, aliquot and store at -80°C, avoid freeze-thaw cycle |
MEM Non-Essential Amino Acids Solution (NEAA, 100x) | Gibco | 11140050 | |
Molecular Biology Grade Water | Corning | 46000CV | |
mTeSR plus kit | STEMCELL Technologies | 1001130 | thaw at 4 °C, aliquot and store at -20°C, avoid freeze-thaw cycle |
N2 supplement (100x) | Gibco | 17502048 | thaw at 4 °C, aliquot and store at -20°C, avoid freeze-thaw cycle |
NaOH solution (1 M) | Cytiva HyClone | SH31088.01 | |
NeuroBasal medium | Gibco | 21103049 | |
NIS-Elements, ver 5.42 | Nikon | ||
Normal Donkey Serum | Jackson Immuno Research Labs | 17000121 | reconstitute with 10 mL sterile ddH2O, store at 4 °C for up to two weeks |
paraformaldehyde, 20% | Electron Microscopy Sciences | 15713S | dilute with PBS |
PBST, 20x | Thermofisher | 28352 | |
PDGFRβ antibody | R&D | AF385 | dilution ratio: 1:400 |
polyamine supplement (1000x) | Sigma-Aldrich | P8483 | |
Rapiclear clearing reagent, 1.49 | SUNJIN LAB | RC149001 | |
Sodium Bicarbonate (7.5%) | Gibco | 25080094 | |
Sodium deoxycholate monohydrate | Thermofisher | J6228822 | dissolve with ddH2O for 1% (wt/v) stock solution |
TBST, 20x | Thermofisher | 28360 | |
trans-ISRIB | Tocris Bioscience | 5284/10 | make the stock solution at 1 mM with DMSO, and store at -20°C for up to 1 year. |
Tritin X-100 | Sigma-Aldrich | T8787-50ML | |
Trypan Blue Stain (0.4%) | Thermo Fisher | 15250061 | |
Tween 20 | Sigma-Aldrich | P7949-100ML | |
VEGF-A | ProSpec | CYT-116 | reconstitute with sterile ddH2O at a concentration of 100 ng/µL, aliquot and store at -20°C |
Zapytaj o uprawnienia na użycie tekstu lub obrazów z tego artykułu JoVE
Zapytaj o uprawnieniaThis article has been published
Video Coming Soon
Copyright © 2025 MyJoVE Corporation. Wszelkie prawa zastrzeżone