Mouse embryonic fibroblast can be reprogrammed into induced pluripotent stem cells at low efficiency by the forced expression of transcription factors Oct-4, Sox-2, Klf-4, c-Myc. The rare intermediates of the reprogramming reaction are FACS isolated via labeling with antibodies against cell surface makers Thy-1.2, Ssea-1, and Epcam.
Mature cells can be reprogrammed to a pluripotent state. These so called induced pluripotent stem (iPS) cells are able to give rise to all cell types of the body and consequently have vast potential for regenerative medicine applications. Traditionally iPS cells are generated by viral introduction of transcription factors Oct-4, Klf-4, Sox-2, and c-Myc (OKSM) into fibroblasts. However, reprogramming is an inefficient process with only 0.1-1% of cells reverting towards a pluripotent state, making it difficult to study the reprogramming mechanism. A proven methodology that has allowed the study of the reprogramming process is to separate the rare intermediates of the reaction from the refractory bulk population. In the case of mouse embryonic fibroblasts (MEFs), we and others have previously shown that reprogramming cells undergo a distinct series of changes in the expression profile of cell surface markers which can be used for the separation of these cells. During the early stages of OKSM expression successfully reprogramming cells lose fibroblast identity marker Thy-1.2 and up-regulate pluripotency associated marker Ssea-1. The final transition of a subset of Ssea-1 positive cells towards the pluripotent state is marked by the expression of Epcam during the late stages of reprogramming. Here we provide a detailed description of the methodology used to isolate reprogramming intermediates from cultures of reprogramming MEFs. In order to increase experimental reproducibility we use a reprogrammable mouse strain that has been engineered to express a transcriptional transactivator (m2rtTA) under control of the Rosa26 locus and OKSM under control of a doxycycline responsive promoter. Cells isolated from these mice are isogenic and express OKSM homogenously upon addition of doxycycline. We describe in detail the establishment of the reprogrammable mice, the derivation of MEFs, and the subsequent isolation of intermediates during reprogramming into iPS cells via fluorescent activated cells sorting (FACS).
Embryonic stem (ES) cells are derived from the inner cell mass of blastocyst stage embryos1. Under appropriate culture conditions they self-renew and remain pluripotent. In 2006 Yamanaka and colleagues demonstrated that mature cells can be reprogrammed into so called induced pluripotent stem (iPS) cells by forced expression of the transcription factors Oct-4, Klf-4, Sox-2, c-Myc (OKSM) 2. iPS cells, like ES cells, can give rise to all cell types of the body, however, they are free of the ethical constraints surrounding the generation of ES cells3. Furthermore, iPS cells carry the promise of personalized regenerative medicine and hold tremendous potential for applications like disease modeling and in vitro drug screening4,5. In order for reprogramming technology to fulfill this potential, the basic mechanism of nuclear reprogramming needs to be fully understood. However, efforts to dissect the reprogramming pathway have been hampered by the fact that only a very small number of cells reprogram (0.1-1%). Successfully reprogramming fibroblasts have been reported to undergo a distinct series of events including a mesenchymal to epithelial transition 6-10 and, in the final stages of reprogramming, activation of the endogenous core pluripotency network 11-14. We and others 12,13,15-17 have recently identified a set of cell surface markers that allows for the separation of rare intermediates from the refractory bulk population. Reprogramming mouse embryonic fibroblasts (MEFs) undergo changes in the expression of Thy-1.2, Ssea1 and Epcam (among others) during the 2-week-long reprogramming process15. Early during reprogramming a subset of MEFs down-regulate expression of fibroblast identity marker (Thy-1.2) and then start expressing the pluripotency-associated marker Ssea-112. During the final stages of reprogramming Ssea1-positive cells reactivate endogenous pluripotency genes such as Oct-410-13,15. This last transition is marked on the cell surface by detectable expression of Epcam (see Figure 1) or in a later stage Pecam 15. Recently, O’Malley et al. reported the use of CD44 and iCAM1 as alternatives or complementary to Thy-1.2 and Ssea-1 for the identification of reprogramming intermediates. We have previously FACS extracted reprogramming intermediates from Day 0, Day 3, Day 6, Day 9 and Day 12 reprogramming cultures, as well as from established iPS cell lines based on these cell surface markers 15,18. For the below described reprogramming system and conditions we have shown at the single cell level that although the populations are quiet homogenous, there is a certain degree of heterogeneity in the identified intermediate populations. It should be noted that only a subset of cells within these populations are able to progress to the respective next stage of the reprogramming process and give rise to iPS cell colonies at different efficiencies, which have been extensively characterized previously15,19. Moreover, the reprogramming efficiency of these populations will depend as well on the re-plating and culture conditions. To increase experimental reproducibility we use a reprogrammable mouse strain that has been engineered to express a transcriptional transactivator (m2rtTA) under control of the Rosa26 locus and a polycistronic OKSM cassette under control of a doxycycline responsive promoter20,21. Using this mouse model circumvents the unwanted side effects of traditional viral methods of iPS cell generation, i.e., a heterogeneous starting population with cell to cell variability in number and location of integration sites of viral inserts. Two transgenic mouse strains (OKSM, m2rtTA), available as homozygous founder animals at the Jackson Laboratory, have to be crossed in order to establish the reprogrammable mouse model (see Figure 2). In this manuscript we describe in detail how to derive MEFs, generate iPS cells, and isolate the reprogramming intermediates at various stages of the conversion process by FACS.
1. Instrument Settings/Reagent Preparation/Genotyping
2. Generation of Mouse Embryonic Fibroblasts
3. Reprogramming of MEFs
NOTE: Pellet cells by centrifugation at 200 x g for 5 min at 4 °C and use normoxic tissue culture incubators for reprogramming. It can be beneficial to derive and expand MEFs under hypoxic conditions (5% oxygen, see discussion for more information).
4. Antibody Labeling
5. FACS Isolation of Intermediates
NOTE: Cells are sorted using a Fluorescence Activated Cell Sorter with 405 nm, 488 nm and 560 nm excitation lasers and a 100 µm nozzle.
After the dissection, disaggregation and plating of an E13.5 mouse embryo, a 10 cm dish is expected to reach confluency in approximately 1 to 2 days. At this stage, it is normal for the culture to contain some adherent pieces of tissue that have not been cellularized properly. These will disappear after passaging.
Upon doxycycline induction, reprogramming MEFs undergo distinct morphological changes. Around day 6, early colony-like patches should start to emerge (Figure 4C). These will continue to grow in size upon further culture (see Figure 4D,E). A good reprogramming experiment should result in >500 colonies per T75 that was initially seeded with 5 x 105 cells (Figure 4E). Established iPS cultures possess characteristic dome shaped colonies and should mostly be devoid of differentiated cells (Figure 4F). Occasionally, additional passaging of iPS cultures may be required to remove undifferentiated/partially reprogrammed cells; a confluent flask of cells can be split at a ratio of 1:10 onto a feeder layer of irradiated MEFs.
As mentioned above, morphological and molecular changes during reprogramming are reflected by changes in the FACS profiles for Thy-1.2, Ssea-1 and ultimately Epcam (see Figure 7A-F). As reported previously12,15, while MEFs are predominantly positive for Thy-1.2 and negative for the other markers, Day 3 cultures already look markedly different. A large proportion of cells have started to down-regulate the expression of Thy-1.2 and a very small subset of these Thy-1.2 negative cells have become positive for Ssea-1, the actual reprogramming intermediate for this time point. The number of reprogramming intermediates that can be extracted from Day 3 cultures is very unpredictable as the percentage of Ssea-1+/Thy-1.2- usually lies in a range between 1-10% of viable cells. On Days 6 and 9 an increased percentage of Ssea-1+ cells, usually well above 10%, can be detected. Around day 12 a subset of Ssea-1 positive cells can be detected that also label positive for Epcam. Day 12 cultures, like Day 3 cultures, represents a bottleneck in the purification of intermediates, as the percentage of Ssea-1.2+/Epcam+ cells is variable and usually falls in the range of only 2-4% of all live cells. The expected number of reprogramming intermediates presented in Table 1 only serves as a rough approximation. Established bona fide iPS cell cultures will be strongly positive for Ssea-1 and Epcam.
Figure 1. Surface marker changes during the reprogramming pathway: Down-regulation of fibroblast identity marker Thy-1.2 is followed by up-regulation of Ssea-1. The transition of a subset of Ssea-1 positive cells towards a pluripotent state is indicated by acquisition of Epcam around day 12. Please click here to view a larger version of this figure.
Figure 2. Schematic representation of the reprogrammable mouse model: m2rtTA expression is under control of the ubiquitously active Rosa26 locus. In the presence of doxycycline (dox) the m2rtTA protein binds to a tetracycline dependent promoter (tetOP) at the Collagen 1a1 (Col1a1) locus resulting in the expression of the four-factor cassette. Two bicistronic cassettes are linked by an internal ribosome entry site (IRES). Open reading frames for Oct-4/Klf-4 and Sox-2/c-Myc are fused by self-cleaving F2A and E2A sequences, respectively. Please click here to view a larger version of this figure.
Figure 3. Embryo dissection: (A) Transfer uterine horn into a 10 cm dish filled with 10 ml PBS and (B) cut into pieces containing one embryo. (C) Using forceps free the embryos from extra embryonic tissue and remove head, limbs, tail and internal organs. Please click here to view a larger version of this figure.
Figure 4. Morphological changes during reprogramming. Colony-like patches become apparent in the cultures from day 6 onwards. Bona fide iPSCs are characterized by a domed shaped morphology. (A) Day 0/MEFs, (B) Day 3, (C) Day 6, (D) Day 9, (E) Day 12, (F) iPS cell culture. Scale bar: 200 µm. Please click here to view a larger version of this figure.
Figure 5. Basic FACS setup. (A) Exclude debris with Side Scatter versus Forward Scatter Area blot. (B) Exclude aggregates from non-debris by gating on single cell population with Forward Scatter Area versus Forward Scatter High blot. (C) Exclude dead cells from single cell population by gating on PI low events with PI channel versus Forward Scatter Area blot. Please click here to view a larger version of this figure.
Figure 6. Gating. (A, B) Using the unlabeled control cells set up gates for Thy-1.2+/Ssea-1- cells, Thy-1.2-/Ssea-1- cells and Thy-1.2-/Ssea-1+ cells. (C) Use the unlabeled control cells to set gates for Epcam positive and Epcam negative cells. (D) Ssea-1+/Thy-1.2- cells can be separated into an Epcam positive and into an Epcam negative population. Please click here to view a larger version of this figure.
Figure 7. Thy-1.2 versus Ssea-1 FACS blots at various stages of the reprogramming process. (A) Day 0/MEFs, (B) Day 3, (C) Day 6, (D) Day 9, (E) Day 12, (F) iPS cell culture.
Table 1. Suggested seeding density, flask number and expected outcome for P1 MEFs of an embryo heterozygous for OKSM and m2rtTA. Please click here to view a larger version of this figure.
Day | Number of T75 flasks seeded on day 0 | Total number of Ssea1+ cells after FACS | Total number of Ssea1+/Epcam+ cells after FACS |
3 | 8 | 2 million | |
6 | 4 | 2 million | |
9 | 4 | 2 million | |
12 | 10 | 10 million | 2 million |
In order to successfully reprogram MEFs into iPS cells and to purify reprogramming intermediates at high quantity it is essential to be aware of factors that have an impact on the overall efficiency. In particular the batch of FBS used to supplement iPS media can have a detrimental effect. In general positive experiences were made with Embryonic stem (ES) cell qualified FBS but batch testing of sera from a variety of vendors might identify a cheaper alternative.
Another factor that impacts on reprogramming efficiency is the genotype of MEFs 15,20. Although MEFs derived from mice heterozygous (het) for both the OKSM and the m2rtTA locus do reprogram, homozygousity at one or both loci results in higher reprogramming efficiencies. Indeed, MEFs derived from the offspring of OKSM and m2rtTA crosses show an increase in reprogramming efficiency in the following order (OKSM/m2rtTA): het/het < homo/het < het/homo < homo/homo (please note that the numbers given in Table 1 are for het/het animals). On the rare occasion a male homo/homo animal is identified, a harem mating with multiple m2rtTA females can be set up. All offspring from these crosses will be het/homo for the OKSM/m2rtTA loci, respectively. In addition, the rapid genotyping of litters is recommended as larger litters are obtained when using younger mice for breeding (6-15 weeks of age).
Moreover, the use of low passage MEFs for reprogramming is emphasized 23. If MEFs were derived and expanded in a normoxic tissue culture incubator we strongly recommend to only use these cells up to passage 2. However, if the experimenter has access to a low oxygen incubator (5% oxygen) for the derivation and expansion of MEFs, passage numbers as high as 3 will still yield good results 23.
In case the experimenter encounters problems associated with reprogramming, as outlined, the most likely explanations are high passage numbers at the time of dox addition, incorrect genotype of the mice or the use of FBS that is not conducive for iPS cell generation. However, in rare cases we observed that MEFs were not able to reprogram even under ideal conditions. In these cases a spontaneous de novo deletion within the m2rtTA’s open reading frame was identified as the underlying reason.
This methodology was successfully adapted to extract Ssea-1+ intermediates via magnetic cell isolation (MACS). There is a clear time advantage in using MACS, however, the Ssea-1+ cell populations’ purity is reduced to 80-90%, rather than more than 95%, which depending on the type of experiment the cells are destined for might pose a problem. Thus, an option is to use MACS to enrich for Ssea-1+ cells followed by FACS to increase purity and/or fraction them into Ssea-1+/Epcam+ and Ssea-1+/Epcam- cells.
While the iPS cells produced by the outlined protocol and mouse model have been shown to be pluripotent and effectively contribute to all tissues of chimeric animals 15,20, a reduced capacity to produce embryos entirely composed of these iPS cells via tetraploid complementation has been demonstrated 24. Aberrant methylation patterns at the Dlk-Dio3 gene cluster have been identified as the underlying cause 24. However, addition of ascorbic acid at a concentration of 50 µg/ml to the media during reprogramming will produce tetraploid complementation competent iPS cells 24. Please be advised that an alternative reprogrammable mouse model engineered to express the reprogramming factors at a different stoichiometry can produce tetraploid competent iPS cells even in the absence of ascorbic acid treatment 25. However, in our hands cells from this strain reprogram at considerably lower frequency compared to the herein used mouse model.
The methodology described in this manuscript allows the separation of reprogramming intermediates from the refractory bulk of the population and should be seen as valuable tool to dissect and understand the reprogramming process, removing the previous limitations of having to use an unfractioned population for profiling (e.g., high signal noise from the refractory cells). The antibody combination described herein allows isolation of intermediates from mouse cells at high purity, however it is possible that additional cell surface markers will be uncovered in the future that will allow to obtain intermediates at even higher purity. Please note that Ssea-1 expression is not a hallmark of human induced pluripotent stem cells and as such this protocol cannot be used on reprogramming cells of human origin. In conclusion, reprogramming intermediates once isolated with the described method can be used for molecular analyses including expression profiling, chromatin immunoprecipitation, methylome analysis, and protein assays.
The authors have nothing to disclose.
We would like to acknowledge the financial support from the Monash Larkins Program as well as from a NHMRC CDF and a NHMRC project grant. Furthermore, we would like to thank Sue Mei Lim for her constructive suggestions, Edwina McGlinn for her support and the Monash Flowcore team (in particular Adam Dinsdale) for their help in producing the video.
Anti-mouse CD90.2 (Thy1.2) Pacific Blue | eBioscience | 48-0902-82 | |
Anti-Human/Mouse SSEA-1 Biotin | eBioscience | 13-8813 | |
Streptavidin PE-Cy 7 | eBioscience | 25-4317-82 | |
Anti- Mouse CD326(Epcam) Fitc | eBioscience | 48-5791-82 | |
Sodium pyruvate | Life Technologies | 11360-070 | |
MEM Non-Essential Amino Acids (NEAA) | Life Technologies | 11140-050 | |
Embryonic Stem Cell qualified Foetal Bovine Serum (FBS) | Life Technologies | 10439024 | |
Trypsin-EDTA (0.25%) | Life Technologies | 25200-056 | |
GlutaMAX | Life Technologies | 35050-070 | |
β-Mercaptoethanol | Life Technologies | 21985-023 | |
DMSO (Dimethyl Sulfoxide) | Sigma-Aldrich | D8418 | |
Penicillin/streptomycin | Life Technologies | 15140 | |
Propidium Iodide solution (PI) | Sigma-Aldrich | P4864-10ML | |
Gelatine from porcine skin | Sigma-Aldrich | G1890 | |
Doxycyclin hyclate | Sigma-Aldrich | 33429-100MG-R | |
Leukemia Inhibitory Factor (LIF) | Millipore | ESG1107 | |
DMEM High Glucose | Life Technologies | 11960-044 | |
DPBS (Dulbecco s phosphate buffer saline) | Life Technologies | 14190-144 | |
KnockOut DMEM | Life Technologies | 10829-018 | |
Irradiated MEFs | Life Technologies | S1520-100 | |
75-cm2 Tissue culture flasks | Corning/BD Bioscience | 430641 | |
15-ml centrifuge tubes | Corning/BD Bioscience | 430791 | |
50-ml centrifuge tubes | Corning/BD Bioscience | 430829 | |
Disposable surgical blades | Disposable surgical blades | 201 | |
Cryogenic vials | Corning/BD Bioscience | 430487 | |
70 µm Cell strainer | BD Bioscience | 352340 | |
Taq DNA Polymerase | Life Technologies | 10342-020 |