Low-molecular-weight inhibitors of cell differentiation enable efficient growth of mouse iPS cells under feeder-free conditions

Abstract

Embryonic stem cells and induced pluripotent stem (iPS) cells are usually maintained on feeder cells derived from mouse embryonic fibroblasts (MEFs). In recent years, the cell culture of iPS cells under serum- and feeder-free conditions is gaining attention in overcoming the biosafety issues for clinical applications. In this study, we report on the use of multiple small-molecular inhibitors (i.e., CHIR99021, PD0325901, and Thiazovivin) to efficiently cultivate mouse iPS cells without feeder cells in a chemically-defined and serum-free condition. In this condition, we showed that mouse iPS cells are expressing the Nanog, Oct3/4, and SSEA-1 pluripotent markers, indicating that the culture condition is optimized to maintain the pluripotent status of iPS cells. Without these small-molecular inhibitors, mouse iPS cells required the adaptation period to start the stable cell proliferation. The application of these inhibitors enabled us the shortcut culture method for the cellular adaptation. This study will be useful to efficiently establish mouse iPS cell lines without MEF-derived feeder cells.

Introduction

The reprogramming of somatic cells to generate induced pluripotent stem (iPS) cells was achieved by expressing four defined transcription factors: Oct3/4, Sox2, Klf4, and c-Myc (Takahashi and Yamanaka 2006). Such iPS cells have major potential in the study and therapy of human diseases because they are capable of self-renewal and can generate all the three primary germ layers: ectoderm, mesoderm, and endoderm. iPS cells are usually maintained on feeder cells that are derived from mouse embryonic fibroblasts (MEFs) (Takahashi and Yamanaka 2006; Takahashi et al. 2007; Yu et al. 2007). However, several feeder-free conditions were developed recently especially for human embryonic stem (ES) cells and iPS cells to ensure biosafe human clinical applications (Xu et al. 2001; Amit et al. 2004; Rosler et al. 2004; Lu et al. 2006; Navarro-Alvarez et al. 2008; Miyazaki et al. 2008; Sun et al. 2009; Kitajima and Niwa 2010; Rodin et al. 2010; Hayashi et al. 2010; Fukusumi et al. 2013). Animal-derived components are relatively high risk for potential infections in clinical regenerative medicine. In addition to regenerative medicine, iPS cells are also expected to play an important role in drug screening. However, the mixed culture condition of iPS cells and the need of feeder cells make the evaluation of the effects of drugs difficult. Furthermore, the establishment of serum- and feeder-free culture conditions is significant to overcome the ethical issues related to animal experiments.

Furue et al. (2005) previously developed a serum-free, chemically defined medium called “ESF7” (now available as ESF-C), which maintains mouse ES cells on collagen-coated dishes. In this manuscript, we show that mouse iPS cells can be cultured in feeder-free conditions as well as on feeder cells using the ESF-C medium containing different kinds of low-molecular-weight cell differentiation inhibitors.

Materials and methods

Mouse iPS cell induction

A human papilloma virus-derived E6 gene was introduced to MEFs (E6-MEFs) to extend the cell proliferation until cellular senescence (Yamamoto et al. 2003). Then, STEMCCA-loxP lentiviral vector (STEMCCA means a “stem cell cassette”), which is composed of Yamanaka 4 factors and self-cleaving 2A peptides, was introduced into E6-MEFs to generate iPS cells (Sommer et al. 2009).

Cell culture

MEFs and feeder cells were cultured in MEF medium consisting of Dulbecco’s modified Eagle’s medium (DMEM; #08459-35, Nacalai Tesque, Kyoto, Japan) supplemented with 10 % fetal bovine serum (FBS; #12483-020, Invitrogen, Carlsbad, CA, USA) and 100× antibiotic–antimycotic mixed solution (#02892-54, Nacalai Tesque). The experimental conditions for the established mouse iPS cells are as follows: (1) mouse iPS cells were cultured on MEF feeder cells in mouse iPS cell medium [DMEM with 15 % Knockout Serum Replacement (referred to as KSR, #10828-028, Invitrogen), 0.22 mM 2-mercaptoethanol (#21438-82, Nacalai Tesque), 100 × MEM Non-essential Amino Acids Solution (#139-15651, Wako Pure Chemical Industries, Osaka, Japan), 100× antibiotic–antimycotic mixed solution and 1,000× leukemia inhibitory factor (LIF, Human, recombinant, Culture Supernatant, #125-05603, Wako Pure Chemical Industries)]; (2) mouse iPS cells were cultured on collagen (Cellmatrix Type I-A, Nitta Gelatin, Osaka, Japan)-coated dishes in ESF-C medium (#2004-05, Cell Science and Technology Institute, Sendai, Japan) containing 1,000 × LIF; (3) the same cell culture of point (2) containing three low-molecular-weight inhibitors [1.5 μM CHIR99021 (#163-24001, Wako Pure Chemical Industries), 0.5 μM PD0325901 (hereinafter the combination of these two inhibitors are referred to as 2i for short, #13034, Cayman Chemical, Ann Arbor, MI, USA), and 0.5 μM Thiazovivin (referred to as Tzv, #04-0017, Stemgent, Cambridge, MA, USA)]. The preparation of MEFs and feeder cells were described in our previous report (Donai et al. 2013).

Population doubling (PD) assay

Mouse iPS cells were seeded at a density of 5.0 × 10⁴ cells per well in a 6-well plate. We evaluated three independent mouse iPS clones. The passages of the mouse iPS cells were carried out until one well of the 6-well plate became confluent. The cell number was counted using an automatic cell counter (Countess, Invitrogen), and the cell growth was evaluated with triplicated experiments. The passage process was repeated up to the 6th passage. Cell PD was obtained from the number of cells using the formula described in a previous report as follows: PD = log₂(a/b), where ‘a’ is the number of the cells counted in passage and ‘b’ represents the number of seeded cells (Qin et al. 2012). The results of PD assay are shown as means with standard deviations. Statistical significance was evaluated using Student’s t test (P < 0.05).

Embryoid bodies (EBs) formation

To form EBs, 2–3 × 10⁶ iPS cells were seeded into a Petri dish (for bacterial culture). After 3–4 days, floating EBs were picked up and transferred to 0.1 % gelatin-coated 24-well plates and cultured for 3–4 additional days. EB medium contained DMEM with 15 % FBS, 100× antibiotic–antimycotic mixed solution, 0.22 mM 2-mercaptoethanol and 100× MEM non-essential amino acids solution.

Alkaline phosphatase (AP) staining and Immunocytochemistry

Detailed method for detection of AP activity and pluripotent markers with immunocytochemistry was previously described in our publication (Donai et al. 2013). EBs on gelatin-coated dishes were stained with βIII tubulin monoclonal antibody (1:600 dilution, #MAB1637, Millipore, Billerica, MA, USA), 4′,6-diamidino-2-phenylindole (DAPI) solution (1:1,000 dilution, #340-07971, Wako Pure Chemical Industries) and Alexa Fluor 568 (1:500 dilution, #A21124, Invitrogen).

Reverse transcription (RT) and quantitative real-time polymerase chain reaction (PCR)

Total RNA was extracted from cells confluent in a 6-well plate and a Petri dish using the TaKaRa FastPure RNA Kit (#9190, TaKaRa Bio, Shiga, Japan) according to the manufacturer’s protocol. Complementary DNA (cDNA) synthesis and PCR were performed in a similar way as in our previous report (Donai et al. 2013).

Results and discussion

We initially investigated whether mouse iPS cells could maintain their specific morphologies in each culture condition. Figure 1 shows the cell morphology and alkaline phosphatase (AP) staining under the conditions 1 and 2. In condition 1, using feeder cells as control, the cells showed mouse ES cell-like morphology, characterized by larger nuclei, scant cytoplasm, dome-shaped colonies, and strong positive AP staining (Fig. 1a, b). In condition 2 (collagen coating + ESF-C), mouse iPS cells showed colonies with scattering morphologies (Fig. 1c), and relatively weak AP activity (Fig. 1d). Then, we compared the cell proliferation between condition 1 and 2 using the corresponding PD values. The cell numbers in condition 1 exponentially increased (Fig. 1e, white circle) and showed aggressive growth potential represented by >1.5 population doubling per day (Fig. 1f, open bar). In condition 2, the cell growth of mouse iPS cells was remarkably delayed (Fig. 1e, black circle). The delayed iPS cell growth restored after passage three, suggesting that the cellular adaptation of mouse iPS cells to the feeder-free condition was necessary to ensure a stable cell growth. The delayed cell growth in the early passages of mouse iPS cells was apparent when the PD value was normalized by the culture day (PD per day, Fig. 1f). Furthermore, although we tried to maintain the three independent mouse iPS clones under feeder-free conditions, one of the mouse iPS clones did not show cell growth and was unable to survive in ESF-C medium (data not shown).

Fig. 1.

Morphology of mouse iPS cell colonies in condition 1 and condition 2. Condition 1, on feeders in mouse iPS cell medium with LIF; Condition 2, on collagen-coated dishes (feeder-independent) in ESF-C medium (serum-free) with LIF. White bar = 50 μm. a, b Morphology of mouse iPS cell colonies (a) and AP staining (b), respectively, in condition 1, showing compact colonies and strong reactivity. c, d Morphology of mouse iPS cell colonies (c) and AP staining (d), respectively, in condition 2. Shape scattering of colonies and weak AP signals are evident. e The population doublings under condition 1 and condition 2. n = 3 × 3 f Population doubling per day. Open and closed bars correspond to condition 1 and 2, respectively. The cell growth was partly recovered at late passages in condition 2

In condition 3, we added three inhibitors (CHIR99021, PD0325901 and Thiazovivin) to the ESF-C medium. CHIR99021 is reported as the inhibitor of glycogen synthase kinase 3β (GSK-3β) signaling, which has a large impact to maintain pluripotency of ES and iPS cells effectively (Sato et al. 2004; Besser 2004; Polychronopoulos et al. 2004; Ying et al. 2008). PD0325901 targets mitogen-activated protein kinase (MAPK/ERK kinase) (Sebolt-Leopold and Herrera 2004; Bain et al. 2007). In combination with CHIR99021, PD0325901 affects the self-renewality of ES and iPS cells to prevent cell differentiation (Lin et al. 2009). Thiazovivin acts as a Rho-associated Kinase inhibitor, which can improve the survival of ES and iPS cells from single cell dissociation (Lin et al. 2009; Xu et al. 2010). The iPS cell colonies were more rounded with clearer boundaries than those obtained under condition 2 (Fig. 2a). In agreement with the improved morphology, the reactivity against AP of the cells in the ESF-C in the presence of these three inhibitors as well as on the feeder cells became strong (Fig. 2b). Furthermore, mouse iPS cells showed enhanced cell proliferation compared to those in condition 2 (Fig. 2c). Although the cell proliferation speed in condition 3 was significantly lower than that in condition 1, the PD value of condition 3 was similar to that of condition 1. As shown in Fig. 2d, the effect of these low-molecular inhibitors (2i + Tzv) became more apparent with increasing PD per day. Based on these results, we concluded that 2i + Tzv enabled us to efficiently maintain mouse iPS cells in the chemically defined medium and in feeder-free conditions without requiring the cellular adaptation period. We evaluated the expression of pluripotent markers of mouse iPS cells under condition 3 by immunocytochemistry (Fig. 2e). The detection of the expression of markers in mouse iPS cells was in good agreement with that of mouse ES cells. Briefly, the mouse iPS cell colonies in condition 3 showed immunoreactivity against Nanog, Oct3/4, and SSEA-1 antibodies. From these data, we concluded that our mouse iPS cells maintained a pluripotent state even in the absence of serum and feeder cells.

Fig. 2.

The small-molecule inhibitors improve the morphology of mouse iPS cell colonies, their proliferation, and the expression of pluripotent markers. Condition 3, on collagen-coated dishes (feeder-independent) in ESF-C medium (serum-free) with LIF, 2i and Tzv. White bar = 50 μm. a, b Morphology of mouse iPS cell colonies and AP staining, respectively, under condition 3. The cells formed dome-shaped colonies with clear edges and displayed strong AP activity. c The number of population doubling under condition 1 and condition 3. The growth speed in condition 3 showed enhanced proliferation similar to that of condition 1. d Population doubling per day. Mouse iPS cells showed good cell proliferation starting from early passages. Open and closed bars represent condition 1 and 3, respectively. e Expression of pluripotent markers in mouse iPS cells. Nanog- and Oct 3/4-positive staining, which could be overlapped with DAPI blue counterstaining, was detected in the nuclei. Stage-specific embryonic antigen (SSEA)-1 protein localizes at the cell surface. White bar = 100 μm. (Color figure online)

To validate whether iPS cells have differentiation capacity in feeder-independent culture, we tried to form EBs after incubation in condition 1 and 3. We could not detect any difference in EBs formation between condition 1 and 3 (Fig. 3a, b). After EBs formation, we detected lower expression levels of stem cell-specific genes compared with pluripotent state (Fig. 3c). Furthermore, we also detected βIII tubulin (ectoderm) marker by immunostaining (Fig. 3d). These findings indicate that differentiation capacity is still intact even in chemically-defined and serum-free condition.

Fig. 3.

In vitro differentiation of iPS cells. a, b Morphology of EBs floating in a Petri dish after 4 days incubation. Left panel (a) EBs derived from iPS cells cultivated under condition 1; Right panel (b) EBs derived from iPS cells cultivated under condition 3. White bar 100 μm. c Detection of stem cell-specific genes by real-time quantitative polymerase chain reaction (RT-qPCR) between iPS cells and EBs in condition 3. White column iPS cells (undifferentiated state); Black column EBs (differentiated state). n = 3 d immunostaining of βIII tubulin (ectoderm) marker in EBs cultivated 4 days on gelatin-coated dishes derived from condition 3. White bar 50 μm

In this study, we demonstrated the efficient expansion of mouse iPS cells on collagen-coated dishes under serum- and feeder-free conditions in combination with three small-molecule cell differentiation inhibitors. We could not observe any morphological or cell proliferation change until passage 10. We also confirmed the effect of the inhibitors on the cell growth of mouse ES cell line E14tg2a. They expectedly showed growth pattern similar to that of iPS cells under the presence of these inhibitors (data not shown). Collagen coating is a simple method and does not require special techniques. This method provides a useful means for using mouse iPS cells to assess novel drugs and animal-derived components without adaptation periods.

Abbreviations

ES

Embryonic stem

iPS

Induced pluripotent stem

MEFs

Mouse embryonic fibroblasts

SSEA

Stage specific embryonic antigen

HPV

Human papilloma virus

STEMCCA

Stem cell cassette

DMEM

Dulbecco’s modified Eagle’s medium

KSR

Knockout serum replacement

LIF

Leukemia inhibitory factor

2i

2 Inhibitors (CHIR99021 and PD0325901)

Tzv

Thiazovivin

GSK-3β

Glycogen synthase kinase 3β

MAPK

Mitogen-activated protein kinase

ROCK

Rho-associated Kinase

PD

Population doubling

EBs

Embryoid bodies

AP

Alkaline phosphatase

DAPI

4′,6-Diamidino-2-phenylindole

RT

Reverse transcription

PCR

Polymerase chain reaction

cDNA

Complementary DNA