Described here is a simple workflow to differentiate endothelial cells from human pluripotent stem cells followed by a detailed protocol for their mechanical stimulation. This allows for the study of the developmental mechanobiology of endothelial cells. This approach is compatible with downstream assays of live cells collected from the culture chip after mechanical stimulation.
The heart is the first organ to be functionally established during development, thus initiating blood circulation very early in gestation. Besides transporting oxygen and nutrients to ensure fetal growth, fetal circulation controls many crucial developmental events taking place within the endothelial layer through mechanical cues. Biomechanical signals induce blood vessel structural changes, establish arteriovenous specification, and control the development of hematopoietic stem cells. The inaccessibility of the developing tissues limits the understanding of the role of circulation in early human development; therefore, in vitro models are pivotal tools for the study of vessel mechanobiology. This paper describes a protocol to differentiate endothelial cells from human induced pluripotent stem cells and their subsequent seeding into a fluidic device to study their response to mechanical cues. This approach allows for long-term culture of endothelial cells under mechanical stimulation followed by retrieval of the endothelial cells for phenotypical and functional characterization. The in vitro model established here will be instrumental to elucidate the intracellular molecular mechanisms that transduce the signaling mediated by mechanical cues, which ultimately orchestrate vessel development during human fetal life.
During embryonic development, the heart is the first organ to establish functionality1, with detectable contractions from the earliest stage of endocardial tube formation2. Circulation, along with the mechanical cues mediated by the flow of blood within the vessel, controls many crucial aspects of early development. Prior to fetal circulation establishment, the vasculature is organized into a primary capillary plexus; upon cardiac functioning, this plexus reorganizes into venous and arterial vasculature3. The role of mechanical cues in arteriovenous specification is reflected by the pan-endothelial expression of arterial and venous markers before blood flow initiation4.
Hemodynamic forces not only control the development of the vasculature itself but also play a fundamental role in the control of blood cell formation. Hematopoietic stem and progenitor cells (HSPCs) emerge from specialized endothelial cells called hemogenic endothelium5,6,7,8, present in different anatomical regions of the embryos exclusively in the early stage of development. Heart-deficient models, together with in vitro models, have demonstrated that mechanical cues instruct and increase HSPC production from the hemogenic endothelium9,10,11,12,13,14.
Different types of flow dynamics have been shown to differentially control the cell cycle15, known to be important in both hemogenic endothelium16,17 and arterial cell specification18. Altogether, mechanical cues are critical determinants of cell identity and function during development. Novel in vitro fluidic devices allow us to overcome the limitations involved with studying developmental mechanobiology during human blood development in vivo.
The overall goal of the protocol in this manuscript is to describe, step-by-step, the experimental pipeline to study the effect of shear stress on human endothelial cells derived in vitro from human induced pluripotent stem cells (hiPSCs). This protocol contains detailed instructions on the differentiation of hiPSCs into endothelial cells and their subsequent seeding into fluidic chips for the stimulation protocol. Using this, different in vitro-derived endothelial cells can be tested for their ability to sense the shear stress by analyzing their orientation in response to the flow. This will allow other laboratories to address questions about the response to shear stress and its functional consequences on different endothelial cell identities.
NOTE: All cell culture techniques must be performed under sterile conditions in a laminar flow hood and cells must be incubated at 37 °C in a humified atmosphere with 5% CO2. Instructions for all cytokine preparation for both maintenance (rhbFGF) and for the differentiation protocol (rhBMP4, rhVEGF, rhbFGF, rhIL6, rhFLT3L, rhIGF1, rhIL11, rhSCF, rhEPO, rhTPO, rhIL3) are in Supplementary Table S1.
1. Culturing of hiPSCs – thawing, maintenance, and freezing of cells
2. Differentiation of hiPSCs into endothelial cells
3. CD34+ cells isolation and seeding into the chip
NOTE: CD34+ cells are isolated via a positive isolation approach with a CD34 microbead kit (see Table of Materials), that contains CD34 microbeads conjugated to monoclonal mouse antibodies anti-human CD34 antibodies and FcR Blocking reagent (Human IgG). It is important to validate the efficiency of the column isolation by staining cells before and after the isolation for flow cytometry analysis, Below it is indicated when cells need to be taken for this analysis.
4. Application of continuous flow to endothelial cells – Aorta-on-a-chip
We describe here a protocol for the differentiation and mechano-stimulation of endothelial cells derived from hiPSCs that allows the study of their response to mechanical cues (Figure 1). This protocol results in the production of functionally mechanosensitive endothelial cells. We provide here representative results and describe the expected phenotype to assess how the cells respond to the cytokine stimulation during the differentiation.
Figure 1: Schematic of the differentiation and mechanical stimulation protocol. Schematic of the differentiation protocol showing the timing of the different mixes of cytokines, the CD34+ cell isolation, fluidic chip seeding, and final analysis of the mechanically stimulated cells. Please click here to view a larger version of this figure.
Culture of hiPSCs
It is important to start the protocol from hiPSCs that are growing correctly in self-renewal conditions. A good indication of the quality of the culture is the speed of their growth. After thawing, the cells might need 2-3 weeks to reach the correct phase of growth that will ensure good differentiation. When the cells can be passaged twice a week at the ratio of 1:6 reaching almost full confluency, this is the time that they are ready to be differentiated on the same day they need to be passaged.
Differentiation of hiPSCs into endothelial cells
The first step of the differentiation, consisting of the formation of embryoid bodies (EBs), is cell line-dependent and may need some optimization for the specific cell line in use. The dissociation described in protocol steps 1.3.2.2-1.3.2.4 can be modified by either reducing or extending the incubation with the dissociation reagent and the subsequent dissociation with the Pasteur pipette. Furthermore, other dissociation reagents can be used for this step in addition to the physical dissociation of the colonies with a cutting tool or a P100 pipette tip. EBs of good quality show a defined edge by day 2 of the differentiation and appear clear and bright when observed using a microscope; darker areas might indicate cell death within the EBs (Figure 2).
Figure 2: Embryoid bodies morphology. (A) Day 2 embryoid bodies showing well-defined outer edges and consistent size. (B) Day 2 embryoid bodies of poor quality showing extensive cell death leading to disaggregation of the structure. Scale bar = 500 µm. Please click here to view a larger version of this figure.
At day 2, the addition of CHIR99021 to the EBs inhibits the GSK-3 protein resulting in the activation of the Wnt pathway. Different cell lines have different responses to CHIR treatment, and this should be tested by quantifying the number of CD34+ cells obtained at day 8 by using different concentrations (Figure 3).
Figure 3: Endothelial cell differentiation with different CHIR treatments. Endothelial cell commitment quantified by flow cytometry at day 8 of differentiation by CD34 membrane expression, following CHIR treatment at day 2 at (A) 3 µM, (B) 5 µM, and (C) 7 µM. Flow cytometry data were obtained using five-laser cytometers and dedicated software (see Table of Materials). Please click here to view a larger version of this figure.
CD34+ cell isolation
It is important to validate that the CD34+ enrichment using the magnetic beads provides at least 80% CD34+ after elution of the column. To ensure sufficient purity, an aliquot of cells obtained from the magnetic isolation can be analyzed by flow cytometry making sure to use a different antibody clone than the one used for the magnetic enrichment. Here, the 4H11 clone was used and ~85% purity was achieved post enrichment (Figure 4).
Figure 4: Membrane expression of CD34 before and after enrichment by magnetic sorting. Day 8 dissociated embryoid bodies (grey) and cells after magnetic enrichment (green) were stained for CD34 expression and analyzed by flow cytometry, showing successful enrichment post sorting. Flow cytometry data were obtained using five-laser cytometers and dedicated software (see Table of Materials). Please click here to view a larger version of this figure.
Seeding cells into the fluidic channel
When seeding the cells in the fluidic channel, it is crucial to track the adhesion and proliferation of the endothelial cells. After seeding, the cells take ~5 h to fully adhere to the channel (Figure 5A). An alternative coating solution may also be tested to improve adhesion at this stage. To confirm that the tested cells are mechanosensitive and thus, able to respond to mechanical stimulation, the cell orientation can be tested over time. Cells before the stimulation show random orientation (Figure 5A and Figure 5C) and they reorient parallel to the direction of the flow (Figure 5B,C). The protocol described here allows for the collection of the cells from the channel to perform downstream analysis, for example, flow cytometry, for the study of their membrane immunophenotype, providing the endothelial identity of the stimulated cells (Figure 5D,E).
Figure 5: Mechanoresponsiveness of hiPSCs-derived endothelial cells. (A) Confluent layer of isolated CD34+ cells 48 h post seeding. (B) Reoriented layer of endothelial cells 3 days under dynamic culture. (C) Orientation analysis of the endothelial cells after 5 days of dynamic culture. (D) CD34 expression profile of cells cultured under flow for 5 days. (E) Percentage of CD34+ cells of cell population retrieved from the fluidic channel. Images were taken using an inverted in-incubator microscope; flow cytometry data were obtained using five-laser cytometers and dedicated software (see Table of Materials). Scale bars = 100 µm (A,B). Please click here to view a larger version of this figure.
Reagents | Stock concentration | Volume added | Final concentration |
Iscove’s Modified Dulbecco’s Medium (IMDM) | – | 333 mL | – |
Ham’s F-12 Nutrient mixture (F-12) | – | 167 mL | - |
N-2 supplement (100x) | 100 x | 5 mL | 1x |
B-27 supplement (50x) | 50 x | 10 mL | 1x |
Ascorbic acid | 10 mg/mL | 1.25 mL | 25 µg/mL |
α-Monothioglycerol (MTG) | 11.5 M | 19.5 µL | 448.5 µM |
Human Serum Albumin | 100 mg/mL | 2.5 mL | 0.5 mg/mL |
Holo-Transferrin | 100 mg/mL | 0.75 mL | 150 µg/mL |
Table 1: Composition and recipe for 500 mL of Serum-free Differentiation (SFD) medium.
Days of differentiation | Cytokine Mix | Cytokine | Final concentration |
Day 0 – 2 | Mix 1 | BMP4 | 20 ng/mL |
Day 2 | Mix 2 | CHIR99021 | 7 μM |
From day 3 onward | Mix 3 and 4 | VEGF | 15 ng/mL |
bFGF | 5 ng/mL | ||
From day 6 onward | Mix 4 | IL6 | 10 ng/mL |
FLT3L | 10 ng/mL | ||
IGF1 | 25 ng/mL | ||
IL11 | 5 ng/mL | ||
SCF | 50 ng/mL | ||
EPO | 3 U/mL | ||
TPO | 30 ng/mL | ||
IL3 | 30 ng/mL |
Table 2: Mixes of cytokines used for endothelial cell differentiation, days in which they are added to the SFD medium, and final concentration.
Shear Stress (dyn/cm2) | Time (h) |
0.5 | 1 |
1 | 1 |
1.5 | 1 |
2 | 1 |
2.5 | 1 |
3 | 1 |
3.5 | 1 |
4 | 1 |
4.5 | 1 |
5 | Until end of the experiment |
Table 3: Shear stress values for the dynamic culture and length of their application.
Supplementary Figure S1: Geometry and dimensions of the chip and tubing used for this protocol. Please click here to download this File.
Supplementary Figure S2: Step-by-step guide for the software controlling the air pump with a description of each step. Please click here to download this File.
Supplementary Figure S3: Guide for the orientation analysis using FIJI showing the drawing of the cell shape, elliptic fitting, and final measurement. Please click here to download this File.
Supplementary Table S1: Unit size, resuspension volume, and stock concentrations for cytokines used in differentiation protocol. Please click here to download this File.
The protocol that we describe here allows for the generation of mechanosensitive endothelial cells from human pluripotent stem cells and the study of their response to mechanical stimulation mediated by controlled shear stress. This protocol is entirely cytokine-based and fully compatible with GMP reagents for potential translation into the production of cells for cell therapy.
The derivation of hiPSCs provides scientists with an instrumental model for the early stages of embryonic development that enables the study of processes that are otherwise difficult to study in vivo24. In fact, human embryonic tissues available for research are collected from embryos lacking circulation, and this might have a significant impact on the molecular signature controlled by mechanical cues. The approach described here enables live-imaging and real-time study of cell response to shear stress. The combination of hiPSCs with fluidics provides a model of study that overcomes the limited availability and the inaccessibility of the developing fetal tissues when the initiation of circulation remodels and controls the establishment of the cardiovascular and blood system3,9,10,25.
A limitation of the protocol is that the endothelial cells derived from this protocol might not reflect the various identities of different endothelial cells present in the developing tissues. To overcome this limitation, a specific combination of cytokines might be needed during the differentiation process preceding the fluidic stimulation to obtain the desired identity or tissue-specific phenotype26. The isolation of endothelial subsets can be obtained using a more refined immunophenotype during the isolation step. This protocol isolates endothelial cells based only on the expression of CD34, thereby allowing for column isolation instead of fluorescence-activated cell sorting (FACS); this reduces cell death and the risk of contamination. Furthermore, this protocol is specifically designed to study the role of shear stress mediated by laminar flow. Alternative fluidic approaches will have to be employed to study the effect of other mechanical cues, such as stretching or compression, or other types of flow such as perturbed or disturbed flow.
We have previously shown that iPSC-derived endothelial cells mimic the heterogeneous arteriovenous cellular identities27 similar to that observed in the fetal dorsal aorta28,29,30. This is of particular importance in the context of vessel development and cellular specification, known to be controlled by blood circulation. Studies in different models showed that lack of circulation results in altered arteriovenous specification11,14,31. The mechanisms that connect mechanical cues with cell specification are still unknown and the pipeline described here allows for refined functional studies that could not be tested in vivo.
This pipeline describes the production and the stimulation of endothelial cells derived from hiPSCs using commercially available fluidic channels, avoiding the need for casting the devices as for the widely used polydimethylsiloxane (PDMS) devices12. Furthermore, the use of PDMS chips makes the collection of the stimulated cells particularly challenging, while with this protocol, the cells can be easily retrieved from the channel. This significantly improves the analytic power allowing for subsequent analysis such as proteomic and transcriptomic analyses, flow cytometry, and functional assays, which might need further culture or in vivo assays.
The authors have nothing to disclose.
This work was supported by the Research Advanced Grant 2021 from the European Hematology Association, the Global Research Award 2021 from the American Society of Hematology, and the Internal Strategy Support Fund ISSF3 funded by the Welcome Trust and the University of Edinburgh. We thank Fiona Rossi from the Flow Cytometry Facility for support in the flow cytometry analysis. For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) license to any Author Accepted Manuscript version arising from this submission.
0.6 Luer uncoated slide | ibidi | IB-80186 | |
25% BSA | Life Technologies | A10008-01 | |
6-well plates | Greiner Bio-one | 657160 | |
Accutase | Life Technologies | A1110501 | Cited as Dissociation reagent |
Ascorbic acid | Merck | A4544-100G | |
Aspiration pipette | Sardtedt | 86.1252.011 | |
B27 supplement | Life Technologies | 17504044 | Cited as Neuronal cell culture supplement (50x) |
BD FACS DIVA | BD Biosciences | Version 8.0.1 | Cited as flow cytometry software |
BD LSR Fortessa 5 Laser | BD Biosciences | ||
bFGF | Life Technologies | PHG0021 | |
CD34 Microbead kit | Miltenyil Biotec | 30-046-702 | |
CD34 PE clone 4H11 | Invitrogen | 12-0349-42 | |
CD34 PerCP-eFluor 710 clone 4H11 | Invitrogen | 44-0349-42 | |
Cellstar cell-repellent surface 6-well plates | Greiner Bio-one | 657970 | Cited as cell-repellent plate |
CHIR99021 | Cayman Chemicals | 13122-1mg-CAY | |
Cryostor CS10 cell cryopreservation | Merck | C2874-100ML | Cited as Cryopreservation solution |
Dimethyl Sulfoxide | VWR | 200-664-3 | Cited as DMSO |
DMEM/F-12 | Life Technologies | 10565-018 | |
DPB Ca2+ Mg2+ | Life Technologies | 14080055 | |
DPBS | Life Technologies | 14200075 | |
EASY Strainer 40 μm | Greiner Bio-one | 542040 | |
EDTA | Life technologies | 15575020 | |
FcR Blocking Reagent | Miltenyil Biotec | 130-059-901 | |
Fiji | Version 1.53c | ||
Flow Jo | Version 10.7.1 | Cited as flow cytometry sanalysis oftware | |
FLT3L | Peprotech | 300-19-10uG | |
Fluidic unit | ibidi | 10903 | |
GlutaMax | Life Technologies | 35050038 | Cited as L-glutamine supplement |
Ham F-12 | Life Technologies | 11765054 | |
Holo-transferrin | Merk | T0665-500MG | |
Human Serum Albumin | Fujifilm UK LTD | 9988 | |
Ibidi Pump system | ibidi | 10902 | Cited as Pump system |
IMDM | Life Technologies | 12440053 | |
Inverted microscope | ioLight/Thisle Scientific | IOL-IO-INVERT | Cited as inverted in-incubator microscope |
Lyophilised BSA | Merck | A2153-100G | |
MiniMACS Separator | Miltenyil Biotec | 130-042-102 | Cited as Magnetic separator |
MS Columns | Miltenyil Biotec | 130-042-201 | Cited as Magnetic column |
MTG | Merck | M6145-25ML | |
N2 supplement | Life Technologies | 17502048 | |
Notebook for pump system | ibidi | 10908 | |
Paraformaldehyde 37-41% | Fisher Chemicals | F/1501/PB15 | |
Pastette | Greiner Bio-one | 612398 | |
Pen/Strep | Gibco | 15070063 | |
Perfusion Set YELLOW/GREEN: 50 cm, ID 1.6 mm, 10 mL reservoirs | Ibidi | IB-10964 | Cited as Perfusion set |
Polystyrene Round Bottom Tubes | Falcon | 352008 | Cited as Flow cytometry tubes |
Prism 9 | Verison 9.4.0 | ||
Pump control software | ibidi | version 1.6.1 | Cited as Pump software |
ReLeSR | Stem cell tecchonologies | 5872 | Cited as Detaching solution |
rhBMP4 | R&D | 314-BP-010 | |
rhEPO | R&D | 287-TC-500 | |
rhIGF1 | Peprotech | 100-11-100uG | |
rhIL11 | Peprotech | 200-11-10uG | |
rhIL3 | Peprotech | 200-03-10uG | |
rhIL6 | R&D | 206-IL-010 | |
rhLaminin-521 | Life technologies | A29248 | Cited as Laminin |
rhSCF | Life Technologies | PHC2111 | |
rhTPO | R&D | 288-TPN-025 | |
rhVEGF | R&D | 293-VE-010 | |
RLT Lysis Buffer | Qiagen | 79216 | |
Serial Connector for µ-Slides: Sterile, Sterile | ibidi | IB-10830 | |
StemPro-CD34 SFM media | Life Technologies | 10639011 | Cited as Serum-Free media for CD34+ cells (SFM-34) |
StemPro-CD34 Nutrient Supplement | Life Technologies | 10641-025 | Cited as 34 nutrient supplement |
StemPro hESC SFM | Life Technologies | A1000701 | Cited as Culture media |
StemPro supplement | Life Technologies | A10006-01 | |
Vitronectin (VTN-N) recombinant human protein, truncated | Invitrogen | A31804 | |
Y-27632 dihydrochloride | Tocris | 1254 | Cited as iRock |
β-Mercaptoethanol | Gibco | 21985023 |