On this page

2.1 Introduction
2.2 Neural differentiation
2.3 Cardiomyocyte differentiation
2.4 Definitive endoderm differentiation
2.5 Differentiation functional assays

Get your copy of the PSC Handbook

If you've found this chapter—Differentiation of Pluripotent Stem Cells—useful, you may be interested in getting your own copy of the entire PSC Handbook in convenient PDF format.

Request your free copy

2.1 Introduction

The generation of iPSCs is often an intermediate step to reach the real experimental goals. The purpose of PSCs in this case is to take advantage of the proliferative capacity and pluripotency of iPSCs to generate virtually unlimited numbers of mature, differentiated cell types including neurons, cardiomyocytes, beta cells, or conceivably any other cell type in the body. These PSC-derived cells can be used in a range of applications such as:

  • Modeling human embryonic development
  • As a source of difficult-to-isolate cells for basic research and disease modeling
  • Drug screening applications
  • Cell replacement therapy

The differentiation of PSCs to a specific lineage is obtained by timed exposure to specific conditions via growth factors, small molecules, and substrates that mimic the sequential events that occur during embryonic development.

Early differentiation protocols relied on the formation of embryoid bodies (EBs), which are 3D aggregates of cells that allowed for spontaneous differentiation. The differentiation from EBs in these earlier protocols could be biased by exposure to growth factors that promoted differentiation of one lineage over another (Figure 2.1).

Diagram showing differentiation of PSCs to ectoderm, mesoderm, and endoderm via an EB intermediate

Figure 2.1.  Differentiation of PSCs to different lineages via an EB intermediate.

More recently, differentiation protocols have become increasingly defined. Most bypass the EB formation step, which effectively created a black box within which signaling events controlling differentiation were poorly understood. 

Instead, recent protocols tend towards adherent culture, in which cells are exposed to a temporarily defined combination of small molecules. The replacement of growth factors with potent small molecules allows for differentiation that is not only more cost- effective but also more efficient.

As protocols have become more defined, the understanding of signaling events required to specify a given cell type has become increasingly complex. The focus in evaluating differentiation protocols now more typically revolves around the validity and functionality of the generated cells. How accurately does the iPSC-derived cell type recapitulate the behavior of the primary cell in vitro and in vivo? Does it express markers associated with the cell lineage? Does it perform as expected in functional assays? And in many cases, most importantly, does it integrate and function in vivo when transplanted into an animal model?

A final consideration pertains to the maturity of iPSC-derived cells. Cells derived via differentiation of PSCs will by default exhibit a fetal or neonatal phenotype. This may manifest itself via expression of fetal- associated markers, such as fetal globins in iPSC-derived erythrocytes or alpha-fetoprotein in iPSC-derived hepatocytes. Conversely, the expression of adult markers may be low or absent in iPSC-derived cells, such as cytochrome P450 levels in iPSC-derived hepatocytes. 

This may be a concern for drug screening and cell therapy applications or for researchers studying late-onset disorders such as neurodegenerative diseases (e.g., Alzheimer’s disease, Parkinson’s disease), since disease-specific phenotypes may not manifest themselves in fetal cells. 

Ongoing research in this field is exploring ways of aging and maturing iPSC-derived cell types in vitro, and it can be expected that more approaches to this problem will be uncovered in the near future. Until this time, keep this caveat in mind and plan around its impact on downstream research.

Some of the key considerations to take into account when developing a differentiation protocol, adapting a published protocol from the literature, or choosing a commercially available differentiation kit include:

  • High quality of cells—Does the final cell population express the markers associated with the cell type in vivo? Does it perform as expected in functional assays in vitro and in vivo?
  • A defined protocol—Does the protocol use defined media and substrates, or does it include components such as serum or BSA? Does it involve an EB formation step or co-culture with a stromal cell line? These factors can introduce variability into a differentiation protocol and make standardization and optimization difficult.
  • Speed—How quickly is the desired cell population obtained?
  • Efficiency—How high is the yield of the desired cell population? Are a considerable number of undesired “contaminating” cell types also obtained?
  • Reproducibility—Are cells and efficiencies obtained consistently across multiple experiments and among different users?
  • Robustness—Does the protocol work efficiently and consistently across multiple ESC and iPSC lines? Some protocols were developed with a small set of lines and adaptation to different lines may require significant optimization.
  • Cost—Does the protocol require significant amounts of expensive recombinant growth factors or specialized tissue culture plates?
  • User friendliness—How many different media are required and how often must cells be passaged or otherwise manipulated? Does the protocol involve labor-intensive picking steps, such as with neural rosettes?
  • Scalability—Can the protocol readily be scaled up for the production of high volumes of cells? Is the cost of media prohibitive? Is the culture system with respect to plate format or manual manipulation requirements not amenable to for larger scales?
  • Bankability—Can cells be frozen as mature cells or at an intermediate stage to establish a bankable population, or must they be derived fresh every time?
  • GMP compatibility—If there is interest in potential clinical applications, researchers may want to ask whether a protocol uses GMP-grade reagents or, if not, can readily be converted to GMP-grade conditions down the line. Starting with either a GMP-compatible protocol or one in which RUO reagents can readily be replaced with GMP versions can save significant time, effort, and cost that would be required to adapt and optimize protocols at a later time point when the project is ready to proceed to the clinic.

See the complete stem cell differentiation portfolio


2.2 Neural differentiation

iPSC-derived neural stem cells (NSCs) are particularly attractive due to their utility for a wide range of applications in disease modeling, drug discovery, and cell therapy.

At the NSC stage, cells are still mitotic and can be expanded and banked for later use. The multipotency of NSCs means that they possess the capacity to differentiate into multiple glial and neuronal subtypes depending on the maturation conditions to which they are exposed.

Early protocols for NSC induction relied on EB intermediates, stromal co-culture, and/or formation of rosette structures that required manual isolation prior to expansion. These protocols were poorly defined, inefficient, and labor- intensive. More recently, protocols relying on adherent cultures differentiated under defined conditions have been developed that allow for rapid and highly efficient induction of NSC populations.

PSC Neural Induction Medium

Gibco™ PSC Neural Induction Medium is a serum-free medium that provides high-efficiency neural induction of hPSCs (Figure 2.2) in only 7 days. Unlike other methodologies, use of PSC Neural Induction Medium does not require the intermediary step of EB formation, thus avoiding added time, labor, and variability (Figure 2.3). High-quality NSCs generated using PSC Neural Induction Medium have high expression of NSC markers and can be cryopreserved, expanded, and further differentiated into other neural cell types (Figure 2.4).

Learn more about neural differentiation systems

Four-panel immunofluorescence image showing expression of nestin and SOX2 in hNSCs as well as DAPI staining and an overlay image.
Bar chart of cells expressing pluripotentcy marker Oct4 vs neural markers nestin, SOX2, and SOX1.

Figure 2.2. At day 7 of neural induction using Gibco PSC Neural Induction Medium, H9 embryonic stem cell induced P0 NSCs were dissociated and re-plated on Geltrex coated plates overnight Cells were then fixed and stained with pluripotent marker Oct4 and neural markers including Nestin, Sox2, and Sox1.
(A) The re-plated P0 NSCs were positive for neural marker Nestin (green) and Sox2 (red) Cell nuclei were stained with DAPI (blue).
(B) The quantification of stained markers showed that less than 1% of P0 NSCs were positive for pluripotent marker Oct4 and more than 80% of P0 NSCs were positive for neural markers Nestin, Sox2, and Sox1.
 

PSC Neural Induction Medium methodology

Diagram of neural induction timeline showing cell splitting on day 0, induction on day 2, and ready-to-use NSCs on day 8.

Other methods of generating NSC

Diagram of neural induction timeline showing cell splitting on day 0, induction on day 5, and ready-to-use NSCs on day 12.

Figure 2.3. Unlike other methodologies, PSC Neural Induction Medium does not require the intermediary step of embryoid body (EB) formation which adds time, labor, and variability.
 

Three panel immunofluorescence imaging results showing expression of beta iii tubulin, GFAP, and GalC in NSCs.

Figure 2.4. Neural Stem Cells (NSCs) generated using PSC Neural Induction Medium have high expression of NSC markers and can be further differentiated into other neural cell types.


2.3 Cardiomyocyte differentiation

Few functional cellular behaviors are as impressive as the spontaneous rhythmic contractions of iPSC-derived cardiomyocytes. Human iPSC-derived cardiomyocytes serve as a particularly important system for studying inherited cardiomyopathies, as studies in animal models have largely been limited by significant differences in human and rodent cardiac electrophysiological properties. It should be noted, however, that iPSC-derived cardiomyocytes exhibit a fetal phenotype.

Applications of iPSC-derived cardiomyocytes include disease modeling, cell replacement therapy (e.g.,for following myocardial infarction), and, increasingly, cardiotoxicity screening during drug development.

PSC Cardiomyocyte Differentiation Kit

The Gibco™ PSC Cardiomyocyte Differentiation Kit consists of a set of serum-free and xeno-free media that enable efficient differentiation of hPSCs to contracting cardiomyocytes in as few as 8 days (Figure 2.5). Unlike other methods that require multiple components and longer assay duration, the PSC Cardiomyocyte Differentiation Kit can be used to generate cardiomyocytes from PSCs in a ready-to-use media format and in less time.

The kit consists of three 1X media that require no thawing or mixing, and each medium is used consecutively over a total of 14 days (Figure 2.6), resulting in functional cardiomyocytes that express relevant physiological markers, contract in culture, and can be subsequently maintained in culture for more than 15 days. 

Find out more about our culture systems for cardiac differentiation

Six panel graphs showing JMP Profiler tool results of confluency and marker expression vs cell density for three iPSC lines.

Figure 2.5. Efficiency across multiple PSC lines. Appropriate seeding density is crucial for optimal iPSC cardiomyocyte differentiation. PSC lines dissociated using TrypLE reagent-dissociated PSC lines were used for setup of these studies. For two lines derived using a CytoTune reprogramming kit-derived lines, BS2 iPSC was observed to be promiscuous at higher density. For the human episomal iPSC line, it was also found to be optimal at a specific density. For hESCs, H9 was observed to be promiscuous at various densities. The JMP Profiler tool identified optimal seeding densities for efficient differentiation of different PSC lines.

 

Cardiomyocyte Maintenance Medium 

Gibco™ Cardiomyocyte Maintenance Medium is a serum-free, xeno-free medium that is capable of maintaining cardiomyocytes that have been differentiated using the PSC Cardiomyocyte Differentiation Kit. This medium is included in the kit, but is also sold separately for researchers wanting to maintain differentiated cardiomyocytes in culture for extended periods of time.

Human Cardiomyocyte Immunocytochemistry Kit

The Invitrogen™ Molecular Probes™ Human Cardiomyocyte Immunocytochemistry Kit enables optimal image-based analysis of two key cardiomyocyte markers: NKX2.5 and TNNT2/cTnT. It is the only kit that offers superior imaging for cardiomyocytes in one box, with a complete set of primary and secondary antibodies, a nuclear DNA stain, and premade buffers to enable an optimized staining experiment.

Immunofluorescence staining of differentiated cardiomyocytes for NKX2.5, TNNT2/cTnt, and DAPI

Figure 2.6. iPSCs were differentiated for 14 days using the PSC Cardiomyocyte Differentiation Kit. The cells were stained using the Human Cardiomyocyte Immunocytochemistry Kit for the markers NKX2.5 (red) for early cardio mesoderm and TNNT2/cTnT (green) for cardiomyocytes, along with DAPI stain blue for nuclear DNA.


2.4 Definitive endoderm differentiation

Definitive endoderm encompasses an intermediate population of cells that gives rise to downstream lineages including pancreas, liver, and gut. As with many other lineages, it has been found that the generation of functionally relevant mature cell types is best achieved through a differentiation protocol that recapitulates the stepwise differentiation during embryonic development, including the passage through a definitive endoderm intermediate.

Downstream lineages have applications in modeling and cell therapy for a wide range of diseases, including diabetes for pancreatic beta cells and metabolic disorders for hepatocytes. iPSC-derived hepatocytes additionally have potential utility for hepatotoxicity studies during the drug discovery process.

Traditional protocols for definitive endoderm induction can be costly due to the requirement for activin and Wnt.

PSC Definitive Endoderm Induction Kit

The Gibco™ PSC Definitive Endoderm Induction Kit consists of two xeno-free media that enable efficient induction of hPSC to definitive endoderm. Unlike other methods that require multiple components and take 5 or more days, the PSC Definitive Endoderm Induction Kit enables generation of ≥90% CXCR4+/PDGFRα– definitive endoderm cells with only two components in just 2 days (Figure below).

Each medium is supplied in a 1X complete formulation, requiring no mixing of additional components, and the resultant definitive endoderm shows >90% high expression of key markers SOX17 and FOXA2 across multiple PSC lines (Figure 2.9) and is capable of differentiating to downstream lineages (Figure 2.10).

Learn more about our culture systems for definitive endoderm differentiation

Four panel flow cytometry dot plots showing definitive endoderm marker expression in induced hPSCs.

Figure 2.7. The PSC Definitive Endoderm Induction Kit produces definitive endoderm populations with high efficiency (≥90%) across hESC and iPSC lines, including cell lines reprogrammed using episomal vectors or CytoTune kits. Representative dot plots show CXCR4+/PDGFRα–- cell populations derived from various cell lines. For each experiment, unstained cells were used to set quadrant gates.

 

PSC Definitive Endoderm Induction Kit

Diagram of neural induction timeline showing plating on day 0, media additions on days 1 and 2, and ready-for-differentiation cells on day 3.

STEMdiff Definitive Endoderm Kit

Diagram of neural induction timeline showing adaptation on day 0, plating on day 5, media additions/changes on days 6, 7, 8, and 9, and ready-for-differentiation cells on day 10.

Figure 2.8. Compared to other differentiation protocols, the PSC Definitive Endoderm Induction Kit produces cells in up to 50% less time and requires no predifferentiation or mixing of media.
 

Three panel immunofluorescence imaging results showing expression of SOX17, FOXA2, and Oct4 in induced hESCs.

Figure 2.9. Immunocytochemistry of hESCs treated with the PSC Definitive Endoderm Induction Kit At day 3, induced cells were immunostained for the endodermal transcription factors SOX17 and FOXA2 and the pluripotent marker Oct4 Nuclei were counterstained with DAPI (blue) to assess total cell numbers.
 

Three panel immunofluorescence imaging results showing expression of various markers in midgut/hindgut, pancreatic endoderm, and liver bud progenitors.

Figure 2.10. Definitive endoderm can be differentiated to downstream lineages. H1 ESCs were treated with media from the PSC Definitive Endoderm Induction Kit and differentiated into functional cells: (A) midgut/hindgut (nuclei, blue; FOXA2, green; CDX2, red), (B) pancreatic endoderm (nuclei, blue; FOXA2, green; PDX1, red), and (C) liver bud progenitors (nuclei, blue; AFP, red) that express relevant physiological markers. Data credited to LT Ang, KM Loh, and B Lim of the Genome Institute of Singapore.
 

Useful tips

  • Neural differentiation:
    To further mature NSCs to specific downstream lineages such as oligodendrocytes, astrocytes, or neuronal subtypes, NSCs must be exposed to additional lineage-specific maturation factors.

    These conditions must be determined and optimized for each cell type. Key signaling pathways involved in lineage specification are summarized in (Figure 2.11)

    Culture conditions for NSC differentiation frequently also generate a contaminating population of neural crest, a highly proliferative and migratory population of cells that in vivo gives rise to a range of cell types including peripheral neurons and glia, bone, cartilage, and melanocytes. Neural crest contaminants can be identified by their expression of CD271 (NGFR/p75) and HNK-1. The maintenance medium for the PSC Neural Induction Medium does not promote the expansion of neural crest contaminants; however, NSC cultures generated via other protocols may exhibit significant levels of neural crest contamination. If these flat and highly migratory cells are observed in the dish, they should be removed by positive or negative selection to avoid overpopulation of the culture by these highly proliferative cells.

  • Cardiomyocyte differentiation:
    Singularizing PSCs for differentiation to cardiomyocytes allows better seeding and confluence estimates, resulting in more consistent results from well- to- well and overall better differentiation of difficult-to-differentiate lines. See a recommended protocol for plating singularized cells for cardiomyocyte differentiation.

  • Definitive endoderm differentiation:
    It is critical to use high-quality hPSCs (with minimal or no differentiated colonies) that are karyotypically normal, confirmed to exhibit pluripotency markers, and routinely passaged every 3 days for at least 3 passages before starting differentiation. Additionally, we recommend that the PSC line not be used past 100 passages.

    Depending on the cell reprogramming technology, incomplete conversion of adult cell types to an induced pluripotent state may lead to the generation of refractory cell lines that are unable to differentiate into some lineages. We recommend inclusion of a positive control cell line like the H7 or H9 hESC line to assess the ability of your iPSC line to differentiate into your cell type of interest.
Diagram showing differentiation NSCs to dopaminergic neurons, GABAergic neurons, glutamatergic neurons, motor neurons, oligodendrocytes, and type 2 astrocytes via various signaling pathways.

Figure 2.11. Summary of key signaling pathways regulating the differentiation of NSCs to specific neural and neuronal subtypes


2.5 Differentiation functional assays

iPSCs are powerful tools for disease modeling. They allow researchers to study disease-specific phenotypes in the disease-relevant cell type established from patient-specific iPSCs. The ease with which isogenic controls can be generated via gene editing further allows researchers to eliminate the effects of donor variability and, with high confidence, identify subtle disease-specific phenotypes. However, this requires the availability of assays to interrogate relevant phenotypes.

Neuronal functional and cell health assays

Neurons are a complex cell type amenable to a variety of cell type–specific assays. Most characteristically, the electrophysiological activity of iPSC-derived neurons can be measured via patch-clamp assays or using multi-electrode arrays to determine neuronal subtype–specific AP activity or to assess the effects of neurotoxic compounds.

NSCs can be subjected to a panel of assays compatible with high-throughput methods (Table 2.1) in the presence of various cell stressors to assess neural cell health (Figure 2.12). Additional levels of complexity can be obtained with all of these assays by co-culturing neurons and glial cells to isolate cell-autonomous from nonautonomous disease phenotypes or to determine neuroprotective effects of glia.

Table 2.1. Selected assays that can be used to measure different aspects of neural cell health.

Four panel line graph of fluorescence versus concentration of test compound for the PrestoBlue assay, CellEvent Caspase 3/7 assay, CellROX Green assay, and MitoSOX assay.

Figure 2.12. A panel of functional assays was used to assess the health of NSCs in response to various cell stressors. iPSC lines were derived from a Parkinson’s disease (PD)-affected donor (PD-3), one multiple systems atrophy (MSA)-affected donor, and two age-matched, healthy control individuals (Ctrl-1 and Ctrl-2), and differentiated into neural stem cell (NSC) populations using PSC Neural Induction Medium. The derived NSCs were expanded on CELLstart Substrate in Neural Expansion Medium for 7 passages followed by Gibco StemPro NSC SFM for another 4 passages. The NSCs were harvested and plated in 384-well assay plates coated with CELLstart Substrate, for evaluation by four high-throughput assays. A Tecan Safire™ reader was used to measure fluorescence. Representative results are shown for (A) the PrestoBlue assay on Ctrl-2, demonstrating the expected loss in metabolic activity with an increase in the concentration of stressors added, (B) the CellEvent Caspase-3/7 Green assay on MSA, demonstrating the expected increase in apoptosis with an increase in the concentration of stressors added, (C, D) the multiplexed CellROX Green assay and MitoSOX Red assay on PD-3, demonstrating the expected increase in oxidative stress with an increase in the concentration of stressors added.

 

Cardiomyocyte functional assays

The phenotypic and electrophysiological characteristics of iPSC-derived cardiomyocytes are comparable to their primary cell counterparts. The beating syncytium that spontaneously forms is particularly amenable to characterization and analysis. Contractions are accompanied by oscillating intracellular calcium levels that can be measured using calcium-sensitive dyes, and the response to cardiotoxic compounds can be quantified (Figure 2.13).

Cardiomyocyte beating patterns can also be measured using multi-electrode arrays and parameters such as peak count, peak frequency, peak amplitude, peak width, and decay time can be determined in disease versus control cells or in the presence or absence of cardiotoxic compounds (Figure 2.14).

Composite figure of Fluo-4 calcium fluorescence cell images, calcium images and transients measured in H9-derived cardiomyocytes.

Figure 2.13. Fluo-4 NW Calcium Assay imaging. (A) H9-derived cardiomyocytes were labeled with the Fluo-4 Calcium Imaging Kit, exchanged with Cardiomyocyte Maintenance Medium, and imaged on an EVOS FL Auto Imaging System at 4x magnification using liteCam™ HD software capturing 30 frames per second. (B) Calcium imaging of cells and no wash. Spontaneous calcium transients were imaged at 100 ms intervals with a TILL Polychrome (FEI Systems), and regions of interest were captured for plotting vs. time. Carrier control or drug containing solution was added as indicated from a 10X stock in Live Cell Imaging Solution.

 

Composite data figure showing voltage waveforms and spontaneous beat rate bar charts measured in PSC-derived cardiomyocytes.
Figure 2.14. Cardiomyocyte response to known cardiotoxicants, showing that cardioactive compounds can modulate PSC-derived cardiomyocyte contraction. After stabilization of electrode activity, spontaneously contracting cardiomyocytes plated at 4 x 104 cells/well on fibronectin-coated multielectrode arrays wells, signals at baseline and in response to drug treatment were averaged over a three minute period (A, B) Cardiomyocyte waveform under baseline or verapamil treatment conditions (C) Effect of verapamil, an L-type Ca2+ channel blocker, on the spontaneous beat rate of H9-derived cardiomyocytes; at the highest dose level, cardiomyocyte contraction ceased (D) Effect of isoproterenol, a beta-adrenergic receptor agonist, on the spontaneous beat rate of H9-derived cardiomyocytes.