|Home | About | Journals | Submit | Contact Us | Français|
This protocol describes an EDTA-based passaging procedure to be used with chemically defined E8 medium that serves as a tool for basic and translational research into human pluripotent stem cells (iPSCs). In this protocol, passaging one six-well or 10 cm plate of cells takes about 6–7 min. This enzyme-free protocol achieves maximum cell survival without enzyme neutralization, centrifugation, or drug treatment. It also allows for higher throughput, requires minimal material and limits contamination. Here we describe how to produce a consistent E8 medium for routine maintenance and reprogramming and how to incorporate the EDTA-based passaging procedure into human induced PSC (iPSC) derivation, colony expansion, cryopreservation and teratoma formation. This protocol has been successful in routine cell expansion, and efficient for expanding large-volume cultures or a large number of cells with preferential dissociation of PSCs. Effective for all culture stages, this procedure provides a consistent and universal approach to passaging human pluripotent stem cells in E8 medium.
The study of human embryonic stem cells (ESCs), the first established human PSC lines 1,2, led to the derivation of iPSCs from human somatic cells3–5. Human ESCs and iPSCs are both capable of generating cells from all three germ lines and have thus gathered tremendous interest for their practical and scientific values. Maintained in cell culture, both are affected by different culture conditions and cell handling techniques.
Cell culture conditions for human ESCs and iPSCs have evolved from feeder-dependent and feeder-free medium to defined medium on defined extracellular matrix (ECM) 1,2,6–9. Improved understanding of self-renewal and differentiation means that growth media and ECM are becoming increasingly defined and simplified. At the same time, multiple cell handling techniques have evolved to facilitate PSC research. However, no single technique can be used at every stage of culture, from expansion through differentiation and cryopreservation.
For regular cell culture practices, cells are often individualized during passaging to achieve even distribution and uniform treatments. However, human ESCs survive poorly after individualization (i.e., being made single cell), because these cells are more sensitive to treatments and are prone to cell death, a fact that has made the development of a universal dissociation method particularly challenging. In routine experiments, dissociation methods are chosen on the basis of either cell survival or sensitivity. In regular expansion, cell survival is the priority. ESCs and iPSCs are usually passaged as aggregates with enzymatic dissociation, with collagenase used for culture on feeder cells 1,2 and Dispase used for culture on feeder-free cells8. Mechanical approaches, such as cell scrapers and other passaging tools, have also been developed to dissociate cells as aggregates. Furthermore, in a differentiation or transfection experiment, TryPLE and Accutase can be used to individualize ESCs, but poor survival often leads to abnormal karyotypes 10,11,12; small chemicals, such as Rho-associated protein kinase (ROCK) inhibitors, must be used to boost cell survival in this process13. All these methods require specialized tools or reagents that are costly for long-term or large-scale experiments. At the same time, the consistency of enzymatic methods is usually affected by the quality of enzymes from batch to batch. Given the variability of these methods, it is highly desirable to find a universal approach for all purposes.
The quality of culture conditions is also crucial to the maintenance and expansion of the PSCs. The medium components related to feeder cells or animal products often greatly affect the consistency of the cell culture, which could be even more problematic when cells have potential applications in translational research. To improve the consistency of cell culture, we systematically analyzed the functions of individual medium components, and finally formulated a xeno-free, chemically defined stem cell medium E8, which contains eight essential factors for human ESCs14. With minimal modification of E8, human iPSCs can be directly derived from skin biopsies in chemically defined conditions. This culture system could have great applications in future PSC research. In companion with the development of cell culture medium, a consistent xeno-free dissociation method is important to fulfilling the potential of the defined culture conditions.
We previously studied cell death mechanisms after individualization and found that myosin-actin dependent contraction leads to cell death, and that cell-cell adhesions promote survival by inhibiting the contraction15. Cells survived by reforming small aggregates in the first few hours after dissociation. This observation provided the theoretical foundation for a method that generates aggregates strong enough to survive, but small enough to be accessed by growth factors or transfection reagents. In combination with the development of chemically defined cell culture, we set out to devise a universal dissociation technique for the expansion, cryopreservation, and experimentation of human PSCs in defined conditions. We found that, after a specific EDTA treatment, cells can be partially dissociated to generate small aggregates that survive as well as Dispase-treated cells, and which are more accessible to various treatments. When used with the chemically defined E8 medium, this dissociation method can be used to handle human PSCs during derivation, expansion, and preservation. In this protocol we describe how to derive iPSCs from fibroblasts as an example of how to use our EDTA dissociation method and defined chemical culture conditions in each stage of the iPSC derivation process.
Conventionally, human PSCs are passaged as aggregates with enzymes, with individualized cells typically dying from actomyosin contraction. However, individualized cells can survive through the inhibition of ROCK-actin-myosin pathway components. Alternatively, cells can survive naturally through reaggregation of adjacent cells, indicating that higher local density could lead to reassociation efficiency independent of drug treatment (Fig. 1a). We found that EDTA treatment could partially dissociate human ESCs, and that ESCs were easily washed off with medium (Supplementary Fig. 1a). Loose adhesions between the cells generated very high local density beneficial to cell survival. Most cells were in small aggregates, which attached to a Matrigel-coated plate within minutes (5 min), spread in 2 h and survived as colonies (24 h; Fig. 1b). EDTA-dissociated ESCs survive significantly better than individualized cells by TryPLE, and the survival efficiency is similar to passaging cells as large aggregates using a Dispase protocol (Fig. 1c and Supplementary Methods). Cell survival when using this EDTA method is comparable to that observed in cells treated by ROCK inhibitor (Supplementary Fig. 1b), and it is E-cadherin dependent (Supplementary Fig. 1c). This indicates that cells harvested by EDTA survive without drug treatment, but require direct cell-cell adhesion.
Because of the survival efficiency of EDTA dissociation, we have used it extensively in long-term culture expansion of human PSCs, on more than 50 different human ESCs and iPSCs in different feeder-free media, including TeSR and E8. In each of these cases, normal karyotypes were effectively maintained in long-term culture. The longest iPSC culture conducted in our research lasted more than 51 passages and longer than 6 months, and the cells maintained normal karyotypes (Supplementary Table 1 lists some of these ESC and iPSC lines). Cells dissociated by EDTA efficiently survived not only on Matrigel and vitronectin surfaces, but also on synthetic surface independent of ROCK inhibitors (Supplementary Fig. 1d). Because of its simplicity and consistency, this protocol has been used for our routine cell culture practices14–17, and other groups have successfully used EDTA dissociation to maintain human PSCs18.
EDTA dissociation is also useful in transfection and differentiation experiments (Supplementary Fig. 2 and Supplementary Methods). In transfection experiments, EDTA-dissociated cells were transfected efficiently while maintaining high survival rate (Supplementary Fig. 2a–c). At the same time, these cells were sensitive to growth factor-induced differentiation (Supplementary Fig. 2d). We hypothesize that EDTA dissociation enables small aggregates to survive better than individualized cells, and that the use of this method allows easier access to reagents used in differentiation experiments when compared with conventional aggregates collected with enzymes such as Dispase. The EDTA method can also be used to harvest PSCs for effective teratoma formation in immune-deficient mice. EDTA dissociation has been successfully used for all culture stages of human PSCs.
EDTA dissociation also improves the handling of PSCs mixed with differentiated cells. In reprogramming experiments iPSC colonies are mechanically isolated for colony expansion, and in the process they are often contaminated with unreprogrammed somatic cells. Conventionally, colony picking is used to enrich stem cells, but this approach is prone to bacterial and fungal contamination, and is not practical when too many lines are involved. We were able to avoid contamination by combining the advantages of E8 medium and the EDTA dissociation method. First, defined E8 medium allows stem cells to grow faster than fibroblasts, needing additional factors such as hydrocortisone. Even without colony isolation, PSCs could potentially outgrow somatic cells in a few passages with high passaging efficiency. Second, we found that PSCs respond to EDTA differently from somatic cells, such as fibroblasts. EDTA treatment preferentially harvests ESC colonies, leaving most fibroblast cells on the original plate (Supplementary Fig. 3a,b). In comparison, TrypLE ubiquitously dissociates both ESCs and fibroblasts.
These observations led us to use EDTA dissociation for iPSC derivation and colony expansion, which requires both human PSCs and somatic cells. In a conventional iPSC derivation procedure, cells are often passaged as individual cells, with ROCK inhibitors boosting iPSC survival before individual colonies are picked. However, we have used EDTA differential dissociation and harvested iPSCs with high survival rates without the help of a ROCK inhibitor (Supplementary Fig. 3c); we found that most EDTA-dissociated cells expressed the stem cell marker SSEA-4 (Supplementary Fig. 3d). EDTA dissociation thus enables the enrichment of potential iPSCs in an overcrowded reprogramming culture in which a secondary passaging is needed. We have also used EDTA dissociation to expand isolated iPSC colonies. We have often left differentiated cells in the original plate and found that stem cells reached a high purity in one or two passages. In addition, in EDTA-based cell expansion, we routinely maintained established ESC and iPSC lines with high purity (i.e., >95% of the population is positive for the stem cell marker Oct4) without enrichment by manual picking. It is possible that spontaneous differentiation is suppressed by the easy access to growth factors by small aggregates created by EDTA passaging, and that differential dissociation allows potential enrichment of stem cells after each passage while leaving some differentiated cells behind.
In addition to preferential dissociation, EDTA passaging has another advantage: it is a quick procedure with minimal opportunity for contamination. The procedure can be performed in a biosafety cabinet, and it does not require enzymatic neutralization and centrifugation. We are thus able to passage 24 individual lines in 15 minutes with minimal risk of contamination. In addition, as EDTA treatment does not damage the ECM, a portion of the iPSCs can be maintained in the original wells after dissociation, providing a backup or duplicate plate for maintenance or further characterizations.
This protocol describes how to produce chemically defined stem cell medium E8 with the formulas listed in Tables 1 and and2.2. To maintain consistency in the cell culture, we recommend making large quantities of medium each time, storing it in the freezer and performing a batch test before applying the medium in real experiments. A base medium, which can be used for all stages of stem cell culture, is first prepared according to Table 1 (see also Supplementary Table 2). Full E8 medium can be prepared with additional self-renewal essential factors (Table 2). The medium can also be prepared on a small scale with the reagents listed in Tables 1 and and22.
When the E8 medium is ready, we test it with H1 ESCs and one control fibroblast iPSC line. EDTA dissociation is used to passage the cells for at least five passages, and cells are then analyzed with flow cytometry and karyotyping. After the E8 medium passes the test, the frozen stock can be used for routine cell culture expansion and other experiments. We find that OCT4-positive cells consistently compose >95% of the whole population when we use the EDTA passaging method with E8 medium.
After E8 components pass the quality test, a set of E8-based media are used to reprogram human skin fibroblast cells. After transduction, cells are first cultured in E8-based reprogramming medium 1, and then in reprogramming medium 2 (Table 2). We usually perform the reprogramming experiment in a six-well plate so that multiple treatments can be applied to the cells in the same experiment (Fig. 2). Sodium butyrate treatment is usually used to improve reprogramming efficiency.
Reprogramming cells usually show up about 3–5 d after transduction, and the colonies mature 20–30 d after transduction. During this period, cells usually need to be passaged to avoid overconfluence, which can inhibit the emergence of true iPSC colonies. The EDTA method is applied to dissociate cells after the emergence of reprogramming cells, which preferentially enrich potential iPSCs.
The procedure has been successfully used in lentiviral-, sendai virus- and episomal DNA-based reprogramming. For simplicity, we used commercially available STEMCCA polycistron lentivirus in this protocol. This procedure could also be modified to reprogram human umbilical vein endothelial cells (HUVECs) and adipocyte cells.
When iPSCs are mature, individual colonies are mechanically isolated into a 24-well plate for colony expansion. Cells are subsequently expanded with the EDTA method, and can be cryopreserved in 3–5 d. The cells can be expanded and should be characterized by alkaline phosphatase staining (APS) and other appropriate characterizations. After further confirmation by karyotyping and immunostaining for pluripotency markers, validated iPSC lines can be used for teratoma formation with the EDTA method.
Add 500 μl 0.5 M EDTA (pH 8.0) stock into 500 ml calcium/magnesium-free PBS. Add 0.9g of NaCl and adjust the osmolarity to 340 mOsm. Sterilize the solution by filtration, and store it at 4°C for up to 6 months.
Critical: To achieve the least disturbance of cells during dissociation, the osmolarity of EDTA solution is designed to be the same as that of E8 medium.
See Box 1 for Matrigel Preparation
Y-27632 is the most commonly used ROCK inhibitor, and it can be dissolved in H2O or DMSO. Dissolve the Y-27632 in sterile H2O or sterile DMSO with a final concentration of 10 mM (1000x), then aliquot and store it at −80°C. The solution is stable for at least 1 year.
Critical: With the EDTA protocol, ROCK inhibitor is not as essential as it is in enzymatic protocols. In routine maintenance of iPSC lines, ROCK inhibitor treatment is optional. However, if the stem cell cultures are too confluent, or have not been passaged for 4 d or more, ROCK inhibitor can greatly increase cell survival. If most of the cells are individualized, ROCK inhibitor treatment is recommended.
Dissolve 500 mg of Holo-transferrin in 50 mL of Dulbecco’s PBS and filter to sterilize. This makes a 1000x (10 mg/ml) stock that can be divided into 500-μl aliquots and frozen at −80°C, at which temperature it is stable for at least 1 year.
One milligram of FGF should be dissolved in 10 mL of sterile 0.05% HSA in PBS. This 1000x FGF (0.1 mg/ml) should be divided into 500-μl aliquots and frozen at −80°C, at which temperature it is stable for at least 1 year.
TGFβ1 should be resuspended to a concentration of 1.74μg/ml (1000x) in sterile 0.05% HSA in 4 mM HCl in PBS. This is then aliquotted into 500-μl aliquots and frozen at −80°C; it is stable at this temperature for at least 1 year.
A measure of 3.625 mg of hydrocortisone is dissolved in 1 ml of sterile water (10mM stock). This solution is then divided into 50-μl aliquots and frozen at −80°C; it is stable at this temperature for at least 1 year.
A measure of 110 mg of sodium butyrate is dissolved in 10 ml of sterile water and filtered (100 mM stock). This is then divided into 500-μl aliquots and frozen at −80°C; it is stable at this temperature for at least 1 year.
Add 2 ml of sterile DMSO and 20 μl 10 mM ROCK inhibitor stock into 8 ml E8 medium to make 2x cryopreservation medium for iPSCs in colony expansion. Medium can be stored at 4°C for up to 1 week.
Dissolve the polybrene in PBS to 1000x (6 μg/ml) and filter to sterilize. Freeze the solution in small aliquots at −80°C and store them for up to 1 year before use. The final concentration on cells during infection will be 6 ng/ml.
Pour 50 liters of DMEM-F12 medium into a suitable container, and stir it on a stir plate. Save the empty bottles. Add 3,200 mg L-ascorbic acid and 970μl sodium selenite (0.7 mg/ml stock) into medium and mix well. Adjust the pH to 7.4 with 10 N NaOH. Add NaCl to adjust the osmolarity to 340 mOsm. Aliquot the medium back into the original bottles. The final composition of the medium should contain 64 mg/l L-ascorbic acid and 13.6μg/l sodium selenite. The details of how to make this medium are also given in Table 1. This base E8 medium should be frozen and stored at −20°C, it is stable at this temperature for at least 1 year.
Thaw one bottle (500 ml) of E8 base medium. Add 500μl of 1000x holo-transferrin, 500μl of 1000x FGF, 500μl of 1000x TGFβ1 and 1 ml 10 mg/ml insulin. Filter the E8 (TGFβ1) medium through a 500-ml Millipore Stericup filter. Store the medium at 4°C before use. The medium is stable for 2–3 weeks at 4°C. Do not warm this medium in a 37°C water bath before use, as it will shorten the life of the growth factors. Note that full E8 medium can now be purchased as Essential 8 media from Life Technologies (cat. no. A14666SA) or from Stem Cell Technologies (cat. no. 05840) as TeSR-E8. The details of how to make this medium are also given in Table 2.
Caution: These purchased media are only for stem cell maintenance. They contain TGF-β1, which can interfere with reprogramming.
Prepare 500 ml of reprogramming medium with the same formula as Full E8 medium, but leave out the TGF-β1. In place of the TGF-β1, either add 50μl 10,000s hydrocortisone (reprogramming Medium 1) or 500μl of 1000x sodium butyrate (reprogramming media 2) before filtering.
The details of how to make these media are also given in Table 2. Media can be stored at 4°C for up to 2 weeks.
Add 25 mg HAS to 50 ml of PBS and filter to sterilize. Store it for 1–2 months at 4°C or make aliquots and store them at −20°C for 1 year.
Dissolve 25 mg of HAS in 50 ml of 4 mM HCl in PBS and filter to sterilize. Store it for 1–2 months at 4°C or make aliquots and store them at −20°C for 1 year.
Troubleshooting advice can be found in Table 3.
Chemically defined E8 medium provides ideal cell culture conditions for human PSC research. By modifying a few growth factors, we are able to create cell culture conditions for human iPSC derivation and expansion (Tables 1 and and2).2). EDTA dissociation also provides a simple and efficient way to handle stem cells in different circumstances (Table 4). This combination of medium and handling method enables effective culture expansion and other treatments. All these experiments can be performed without drug treatment, but ROCK inhibitor is included in this protocol for those who prefer its use.
When human fibroblasts are reprogrammed in the E8-based feeder-free system, transformed cells are visible 5–6 d after transduction. It takes 20–30 d before true iPSC colonies mature enough for colony expansion (Fig. 2c). Overcrowded fibroblast cells often suppress the emergence of iPSC colonies, and thus EDTA dissociation could be used to differentially enrich iPSCs.
After mechanical isolation of iPSC colonies, EDTA dissociation enables effective expansion and cryopreservation in 1–2 weeks, markedly decreasing the time needed to preserve the cells. Owing to its simplicity, we are able to passage 24 individual lines from a 24-well plate in 15 minutes.
Expanded iPSCs should be first confirmed by APS staining and by pluripotency marker staining using flow cytometry or immunostaining (Fig. 2d,e and Supplementary Fig. 4b). Teratoma formation assays should be performed on selected iPSC lines by injecting cells into SCID mice. Teratomas usually emerge after 4 weeks, and are ready to be analyzed 6–7 weeks after injection.
This work was supported by NHLBI, NIH Common Fund through the Center for Regenerative Medicine (to G.C. and J.B.), the Charlotte Geyer Foundation, the Morgridge Institute for Research, NIH Grant UO1ES017166 (to J.A.T.), NIH contract RR-05-19 (to J.A.T.), NIH contract No. HHSN309200582085C (to J.J.) and private funds from the Wisconsin Alumni Research Foundation (to J.J.). We thank M. Boehm, T. Finkel and M. Rao for their suggestions. We thank K. Eastman for editorial assistance.
ContributionsG.C. and J.A.T. conceived the experiments and supervised the project; G.C. developed dissociation protocol; J.B. and G.C. performed the reprogramming experimental procedure in the paper; G.C. and D.R.G. demonstrated preferential dissociation of EDTA; N.G, J.J., D.R.G. and G.C. performed long-term culture; G.C., D.R.G. and J.B. performed teratoma formation assay; L.I.S. performed immunostaining and EDTA sequential dissociation imaging; J.B. and G.C. wrote the paper.
Competing Financial interests
J.A.T. is a founder, stockowner, consultant, and board member of Cellular Dynamics International. He also serves as scientific advisor to and has financial interests in Tactics II Stem Cell Ventures.