Induction of Retinal Pigment Epithelial Cells from Monkey iPS Cells

Abstract

Purpose.: The induced pluripotent stem (iPS) cell is expected to be a powerful tool for research and development in regenerative medicine. Previously, the authors reported that human iPS cells differentiated into retinal cells, including photoreceptors and retinal pigment epithelial cells. In this study, they produced iPS cell lines from monkeys to investigate their ability to differentiate into retinal cells.

Methods.: To generate iPS cells, the fibroblasts derived from cynomolgus monkey abdominal skin were infected with retroviruses carrying Oct3/4, Sox2, Klf4, and c-Myc genes and then were cultured on STO feeder cells. Next, the established iPS cells were cultured with the conditioned medium of PA6 cells to induce RPE cells. The properties of the differentiated RPE cells were analyzed.

Results.: Approximately 1 month after viral infection, some epithelial-like colonies appeared among the fibroblasts. These colonies were morphologically similar to the cynomolgus embryonic stem (ES) cell and expressed ES cell-specific markers. By producing teratomas in SCID mice, these cells were confirmed to have the ability to differentiate into three germ layers. In addition, the RPE cells induced from the monkey iPS cells had characteristic polygonal shapes and pigments. These cells expressed RPE cell-specific markers such as RPE65, CRALBP, Bestrophin 1, and MERTK and exhibited phagocytotic function in vitro.

Conclusions.: The RPE cells derived from monkey skin with iPS cell technology can be used for autologous or allogeneic transplantation to test the possibility of immune rejection and to evaluate their function in vivo with the same techniques that will be used in clinical trials.

The retina is a sensor for light and a processor of visual information. Retinal damage causes permanent loss of vision; however, recent studies in regenerative medicine have given us the hope of rescuing visual function. One of the strategies is transplantation of retinal cells, especially photoreceptors or RPE cells. The pluripotent embryonic stem (ES) cell is one of the possible cell sources for transplantation because it is derived from the inner cell mass of the embryo in the blastocyst stage and has the potential to generate all types of cells in the body. ^1,2 Previously, we established induction methods for retinal cells from mouse, monkey, and human ES cells. A stromal cell-derived inducing activity (SDIA) method in which ES cells were cocultured with PA6 cells, a mouse stromal cell line, provided RPE cells, ³ and a serum-free floating culture of embryoid body-like aggregates method, which was characterized by suspension culture without serum, produced retinal progenitors and mature photoreceptors as well as RPE cells. ^4,5 Furthermore, we previously reported that RPE cells induced from cynomolgus monkey ES cells by the SDIA method could rescue photoreceptor cells after transplantation into the subretinal spaces of Royal College of Surgeons rats, which is a photoreceptor-degeneration model with impaired RPE function. ⁶ However, in clinical situations, ES cells might cause rejection because they are always allogeneic cells to the recipients. 

Five years ago, Takahashi and Yamanaka ⁷ reported that they produced brand-new pluripotent stem cells by inducing 4 genes (Oct3/4, Sox2, Klf4, and c-Myc) into embryonic and adult mouse skin cells. They named them induced pluripotent stem (iPS) cells. ⁷ It is easier to obtain iPS cells than it is to obtain ES cells, not only technically but also ethically because iPS cells are derived from somatic cells whereas ES cells are derived from embryos. In addition, iPS cells can be produced from a patient's own cells, making it possible to avoid rejection from cell transplantation. Whereas it was first reported that there were some differences in iPS cells from ES cells in the pattern of gene expression and methylation and no contribution to chimera formation, the following year they reported that iPS cells of the second generation overcame these problems by changing the screening method. ⁸ Furthermore, the generation of human iPS cells was reported by two groups concurrently during the same year. ^9,10 Subsequently, various species of iPS cells have been described, such as rat, ^11,12 rabbit, ¹³ dog, ¹⁴ and pig. ^15,16 As for primates, iPS cells of rhesus monkeys (rhesus macaque) ¹⁷ and marmosets ^18,19 have been described. 

Autologous and allogeneic transplantation systems are important for the development of regenerative medicine, especially for the study of immune reactions. In this regard, iPS cell technology has an advantage. Furthermore, as for retinal research, only humans and monkeys have a fovea in the retina in which high visual resolution can be obtained in mammals. In this report, we describe the production of iPS cell lines from cynomolgus monkeys (crab-eating macaques), with which we have been acquainted regarding results of ES cell studies. Moreover, the generated monkey iPS cells could differentiate into various types of cells including RPE cells. These cells will be useful in future research studies to develop retinal regeneration methodologies. 

Methods

Animal Experiments

All animal experiments were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the RIKEN Center for Developmental Biology Ethical Committee (no. AH20–02) for monkey iPS cell generation or (no. AH18-05) for teratoma formation assay. 

Preparation of Monkey Fibroblasts

Monkey fibroblasts were cultured from the skin of cynomolgus monkeys (Japan Biological Science, Tokyo, Japan). The skin was dissected with a scalpel into ∼5-mm squares, and the squares were cultured in MF-start medium (Toyobo, Osaka, Japan) on gelatin-coated dishes in a humidified atmosphere of 5% CO₂ and 95% air at 37°C (all cultures were performed under the same conditions). After fibroblasts crawled out of the skin (approximately 5 days after the culture started), the culture medium was changed to DMEM containing 10% FBS (Biological Industries, Kibbutz Beit-Haemek, Israel). For passage, the cells were washed with PBS and incubated with trypsin-EDTA solution at 37°C. Then the cells were washed with DMEM containing 10% FBS and were split in a 1:3 ratio. 

Establishment of Monkey iPS Cells

For producing iPS cells, the monkey fibroblasts were infected with retroviruses containing the iPS-factor genes (pMXs-hOCT3/4 [addgene: 17217], pMXs-hSOX2 [addgene: 17218], pMXs-hKLF4 [addgene: 17219], and pMXs-hc-MYC [addgene: 17220]) twice for 48 hours. The viral supernatant was produced in a GP2–293 packaging cell (Retro-X Universal Packaging System; Clontech, Mountain View, CA) culture transfected with pVSV-G and the retroviral vectors described. On the next day of the second infection, the virus-containing medium, was replaced with fresh DMEM containing 10% FBS. Six days after the viral infection, the fibroblasts were trypsinized and transferred onto mitomycin C-treated STO feeder layers at a concentration ranging from 5 × 10⁴ to 5 × 10⁵ cells/100-mm dish. The following day, the medium was changed to primate ES cell-maintenance medium (DMEM/F12 containing 20% KnockOut Serum Replacement [KSR; Invitrogen, Carlsbad, CA], 0.1 mM nonessential amino acids, and 0.1 mM 2-mercaptoethanol) supplemented with 5 ng/mL basic fibroblast growth factor (bFGF; Wako Chemicals, Osaka, Japan), 10 μM Y-27632, and 1 mM valproic acid sodium salt. Approximately 3 to 4 weeks after gene transduction, iPS cell colonies appeared among the feeder layers. Then they were picked up manually, dissociated mechanically, and transferred onto fresh STO feeder layers for expansion. 

Established monkey iPS cells were maintained with standard protocols for primate/human ES cells. Briefly, the cells were cultured on mitomycin C-treated STO feeders in primate ES cell-maintenance medium. For passage, iPS cells were dissociated with 0.25% trypsin-1 mg/mL collagenase IV (BD Biosciences, Franklin Lakes, NJ), and some were transferred onto fresh feeder layers. 

Teratoma Formation

To generate teratomas, 10⁶ iPS cells dissociated enzymatically were injected subcapsularly into the testes of 7- to 8-week-old SCID mice (Oriental Yeast, Tokyo, Japan). Three months later, the testes with teratomas were fixed, paraffin-embedded, and sectioned for hematoxylin and eosin or Alcian blue staining. 

Induction and Expansion of RPE Cells

For RPE cell induction, iPS cells were dissociated into cell clumps using trypsin/collagenase IV and cultured on gelatin-coated dishes in differentiation medium (GMEM containing 10% KSR, 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, and 0.1 mM 2-mercaptoethanol) with PA6 cell feeder (standard SDIA method) or cultured in the supernatant of PA6 cells cultured with the differentiation medium (modified SDIA method). Induced RPE colonies were picked up manually and cultured on laminin (Sigma, St. Louis, MO)–coated dishes in DMEM/F12 supplemented with B27 (Invitrogen) and 10 ng/mL bFGF for more expansion. ²⁰ When the culture reached confluence, cells were dispersed with trypsin-EDTA and reseeded. 

Induction of Neural Retinal Progenitors

iPS cells were enzymatically dissociated into cell clumps and cultured in suspension in differentiation medium with SB431542 and CKI-7, which are inhibitors against Nodal and Wnt signaling, respectively. ²¹ After 2 weeks of suspension culture, the cells were transferred to laminin-coated dishes and were cultured in DMEM/F12 supplemented with B27 and 10 ng/mL bFGF for 2 weeks. 

Immunocytochemistry

Cells were fixed with 3.7% formaldehyde for 15 minutes at 4°C and then permeabilized with 0.2% Triton X-100 for 10 minutes After 1-hour blocking with 5% goat serum, cells were incubated with primary antibodies overnight at 4°C and subsequently with secondary antibodies for 1 hour at room temperature. 

Reverse Transcriptase-Polymerase Chain Reaction

Total RNA was extracted with reagent (Trizol; Invitrogen) and treated with RNase-free DNase I (Roche, Indianapolis, IN). cDNA was synthesized with reverse transcriptase (SuperScript III; Invitrogen), and PCR reactions were performed with EX Taq DNA polymerase (Takara Bio, Shiga, Japan). The PCR products were separated by electrophoresis on 2% agarose gels and detected under UV illumination. The gene-specific primers used in this research are listed in Table 1. 

Table 1.

 

Primers Used for RT-PCR

Primers		Sequences
Exo-Oct3/4	Forward	CCCCAGGGCCCCATTTTGGTACC
Exo-Sox2	Forward	GGCACCCCTGGCATGGCTCTTGGCTC
Exo-Klf4	Forward	ACGATCGTGGCCCCGGAAAAGGACC
Exo-c-Myc	Forward	CAACAACCGAAAATGCACCAGCCCCAG
Exo-genes	Reverse	TTATCGTCGACCACTGTGCTGCTG
Endo-Oct3/4	Forward	CAGATCAGCCACATTGCCCAG
	Reverse	CAAAAGCCCTGGCACAAACTCT
Endo-Sox2	Forward	GGTTACCTCTTCCTCCCACTCC
	Reverse	CCTCCCATTTCCCTCGTTTT
Endo-Klf4	Forward	TTTTCGGTTTTGGCTTCGTTTC
	Reverse	GTCCAGGTCCAGGAGATCGTTG
Endo-c-Myc	Forward	GCGTCGTGGGAAGGGAGATAC
	Reverse	CACCGAGTCGTAGTCGAGGTCATA
Nanog	Forward	CCTATGCCTGTGATTTGTGGG
	Reverse	AGGTTGTTTGCCTTTGGGAC
GAPDH	Forward	ACCACAGTCCATGCCATCAC
	Reverse	TCCACCACCCTGTTGCTGTA
Bestrophin 1	Forward	TAGAACCATCAGCGCCGTC
	Reverse	TGAGTGTAGTGTGTATGTTGG
RPE65	Forward	TCCCCAATACAACTGCCACT
	Reverse	CCTTGGCATTCAGAATCAGG
MERTK	Forward	TCCTTGGCCATCAGAAAAAG
	Reverse	CATTTGGGTGGCTGAAGTCT
CRALBP	Forward	GAGGGTGCAAGAGAAGGACA
	Reverse	TGCAGAAGCCATTGATTTGA

Phagocytosis Assay

Confluent RPE cells were incubated with fluorescent 1-μm polystyrene microspheres (FluoSpheres; Invitrogen) at a concentration of 10⁷ particles/mL media for 6 hours, with or without anti-MERTK antibody. Then the cells were washed with PBS three times and fixed with 3.7% formaldehyde, followed by observation and counting of the remaining microspheres under the fluorescence microscope. 

Results

Generation of Monkey iPS Cells

To produce monkey iPS cells, a culture of skin fibroblasts from cynomolgus monkeys was performed (Figs. 1A, 1B), and then cells were infected with retroviruses carrying the reprogramming genes (Oct3/4, Sox2, Klf4, c-Myc). Six days after viral infection, the fibroblasts were transferred onto mitomycin C-treated STO feeder layers, and the next day the culture medium was changed to primate ES cell medium containing basic FGF with valproic acid, with the goal of increasing efficiency of the iPS colony formation. Approximately 1 month after viral infection, iPS cell colonies that had a monkey ES cell-like morphology appeared among the monkey fibroblasts and feeder cells (Figs. 1C–H). They were picked up manually and transferred to fresh feeder layers for expansion. 

Figure 1.

 

Generation of cynomolgus monkey iPS cells. (A, B) Primary monkey fibroblast culture. The fibroblasts expanded from a piece of abdominal skin tissue (A) and grew exponentially (B). (C–H) Morphology of colonies of monkey iPS cell lines and ES cell lines. The iPS cells derived from skin fibroblasts (C–G) were morphologically similar to the cynomolgus ES cell line, CMK6 (H). (I) The iPS cell lines we established had a normal chromosome number—42—and a healthy karyotype. Scale bars, 200 μm.

Most chromosomes of the generated iPS cell lines had the normal number of chromosomes for monkeys—42—and healthy karyotypes (Fig. 1I). To check on the expression of ES cell marker genes in the iPS cells, immunostaining and RT-PCR analyses were performed. Immunostaining analysis revealed that these iPS cells expressed monkey ES cell markers (alkaline phosphatase, Oct3/4, Nanog, SSEA-4) (Fig. 2A). Expression of SSEA-3 was negative in these cells, unlike human iPS cells (data not shown). The results on ES cell-marker expression meant that these iPS cells remained in an undifferentiated state as ES cells. In RT-PCR analysis, these iPS cells also expressed the ES cell marker genes from an endogenous locus (Fig. 2B). On the other hand, with some exceptions, most exogenous genes that were induced to generate iPS cells were not expressed. Although a clone, Cyn54a, expressed fewer endogenous c-Myc and Nanog genes detected by RT-PCR (Fig. 2B), it appeared to express enough Nanog protein by immunostaining and to be in an undifferentiated state (Fig. 2A). 

Figure 2.

 

Monkey iPS cells expressed ES cell markers. (A) The iPS cells also expressed monkey ES cell markers (alkaline phosphatase, Oct3/4, Nanog, SSEA-4). Scale bar, 200 μm. (B) Monkey iPS cells expressed endogenous mRNA of Oct3/4, Sox2, Klf4, c-Myc, and Nanog, although the exogenous transgenes that were forced to express mRNA to reprogram the fibroblasts were mostly silenced in the established lines.

Pluripotency and Retinal Differentiation of Monkey iPS Cells

To check whether these iPS cells could differentiate into tissues of three germ layer origins, teratoma formation analysis was performed. The undifferentiated iPS cells were dissociated into cell clumps by trypsin and collagenase IV and then were transplanted into the testes of SCID mice. Approximately 2 to 3 months after cell injections in mice, every cell line that was tested formed teratomas in the testes. When these teratomas were sectioned and stained with hematoxylin-eosin or Alcian blue for visualization of the tissues, they included various types of tissues derived from all three germ layers (Figs. 3A–I). 

Figure 3.

 

Monkey iPS cells differentiated into various types of cells, including retinal cells. (A–I) The monkey iPS cells differentiated into various cell types of three germ layer origins when they were transplanted into the testes of SCID mice. Sections of teratoma tissues were treated with hematoxylin and eosin stain (A–H) or Alcian blue stain (I). RPE (A), sweat gland (D), and neural tissue (G) derived from the ectoderm; muscle (B), cartilage (E), and adipose tissue (H) derived from the mesoderm; airway-like epithelium (C), salivary gland (F), and gut-like epithelium (I) derived from the endoderm. (J–O) Monkey iPS cells differentiated into retinal cells. (J–M) RPE colonies appeared when they were cultured with a mouse stromal cell line, PA6 (derived from Cyn46a, L, M) or its conditioned medium (derived from Cyn54a, J, K). (N, O) Using floating culture followed by adherent culture on laminin, monkey iPS cells (Cyn46a) formed neural spheres, with some cells expressing the neural retinal progenitor marker, Pax6. Scale bars, 100 μm.

Previously, we were successful with retinal cell induction, including RPE cells and photoreceptors from monkey ES cells and human ES/iPS cells. The expectation was that monkey iPS cells should also differentiate into these retinal cells using the same or similar methods. Therefore, the monkey iPS cells were next examined for the ability to differentiate into retinal cells. To generate RPE cells, these iPS cell lines were dissociated into cell clumps and cocultured with PA6 cells in GMEM/10% KSR (standard SDIA method) or were cultured in PA6-conditioned medium (modified SDIA method). Approximately 1 month after the start of differentiation culture, some pigmented colonies of RPE cells appeared on the culture dishes (Figs. 3J–M). Then these RPE colonies were picked up manually and cultured on laminin-coated dishes in DMEM/F12-Ham with B27 supplement and basic FGF for more expansion. In addition, the monkey iPS cells formed spheres, and some cells in those spheres expressed Pax6, a neural retinal progenitor marker, when the iPS cells were cultured in suspension, followed by adherent culture on a laminin plate (Fig. 3N, 3O). 

Characterization of Monkey iPS-Derived RPE Cells

RPE cells induced from the monkey iPS cells showed clear polygonal morphology, mostly hexagonal, even after five passages (Fig. 4A). To characterize them, the expression of RPE specific marker genes was examined. In immunostaining analysis, the RPE cells expressed the tight junction component of ZO-1 and N-cadherin along their polygonal shape (Figs. 4B–D). It is known that N-cadherin is a dominant cadherin in mature RPE cells, so this result indicates that the pigmented cells derived from the iPS cells were mature RPE cells. Additionally, RT-PCR analysis revealed that they expressed RPE-specific genes (RPE65 and CRALBP), which are involved in the metabolism of visual pigments, Bestrophin 1, which is a chloride channel on the plasma membrane of RPE cells, and MERTK, which is involved in the phagocytosis of photoreceptor outer segments (Fig. 4E). In this way, the RPE cells induced from the iPS cells had many characteristics of mature RPE cells in vivo. 

Figure 4.

 

Marker expression of RPE cells induced from monkey iPS cells during culture passage. (A–D) Subcultured RPE cells derived from a monkey iPS clone, Cyn46a. (A) Bright-field image. (B–D) Immunocytochemistry image against ZO-1 (B), N-cadherin (C), and the merged image (D). The RPE cells increased exponentially and still had melanin pigments, normal morphology, and marker expression after five passages. Scale bars, 100 μm. (E) Induced RPE cells expressed mature RPE marker genes, which are essential for RPE functions in vivo.

It is well known that RPE cells phagocytose outer segments of photoreceptors, which convert light signals into electric signals. Phagocytosis by RPE cells keeps the outer segments fresh and eventually enables the photoreceptors to survive in a healthy state. Therefore, with the use of polystyrene microbead–conjugated fluorescent dye, we determined whether the RPE cells induced from monkey iPS cells had phagocytosis activity as a function of mature RPE cells. The diameter of a microbead is 1 μm, which is roughly equivalent to that of the outer segment of a rod photoreceptor. Fluoro beads were applied to RPE cells to make the cells phagocytose the beads. After 6 hours' culture, the cells were washed, and the remaining beads inside the cells were counted under a microscope. RPE cells derived from monkey iPS cells phagocytosed many beads, and this phagocytosis activity was inhibited by an inhibitory antibody against MERTK, ²² which is necessary for RPE cells to phagocytose rod outer segments (P < 0.001, Student's t-test; Fig. 5). In contrast, HeLa S3 cells, which were used as general epithelial cells in our study, hardly showed phagocytosis activity. Thus, the RPE cells derived from the monkey iPS cells by our method had a specific function as mature RPE cells. 

Figure 5.

 

Phagocytotic activity of monkey iPS-RPE cells. Induced monkey RPE had phagocytosis capacity. The intake of microspheres was evident and inhibited by the addition of anti–MERTK antibodies, although general epithelial cells (HeLa S3 cells) had little phagocytosis activity. This indicated that induced monkey RPE cells from fibroblasts by iPS cells had a unique function that is observed in RPE cells in vivo. Error bars, SD; n = 3.

Discussion

In this study, we first established cynomolgus monkey iPS cell lines from skin fibroblasts and then induced RPE cells from the iPS cells. The RPE cells appeared not only to show the expression of RPE cell-specific markers but also phagocytotic activity, which is one essential function of RPE cells in vivo to maintain the health and survival of photoreceptors. These functional RPE cells can be used in transplantation research to analyze an in vivo effect on RPE-impaired animal models and immune reactions such as the inflammatory reaction or rejection. 

Monkey iPS cell lines were produced by the induction of reprogramming genes using retroviral infection with a pantropic VSV-G envelope glycoprotein. The method we performed was one of a general method used for the induction of human iPS cells, but the efficiency of iPS cell colony formation was very low. We obtained five colonies from 10⁷ monkey skin fibroblasts. This meant that the efficiency in monkey cells was approximately three orders of magnitude lower than that in human cells using the same protocol (∼0.01%). It was a concern that the transgenes used for producing monkey iPS cells included human cDNA instead of monkey cDNA. However, we obtained many mouse iPS cell colonies when we used the same human genes for reprogramming of mouse fibroblasts (unpublished data, November 2009). Accordingly, we thought that the difference in the animal species of these transgenes did not lead to low colony formation. Furthermore, monkey cells do not also always have low reprogramming efficiency because some groups produce iPS cells of the rhesus monkey or marmoset at the rate of ∼0.03%. ^17,18 Further study and improvement will be necessary for large-scale research. 

We generated RPE cells from monkey iPS cells not only by coculture with PA6 cells (standard SDIA method) ³ but also by culture with conditioned medium of PA6 cell culture (modified SDIA method). This was in contrast to a previous result in that the conditioned medium failed to induce the neuronal marker TuJ-positive cells. ²³ In the previous report, the physical contact between ES cells and PA6 cells was not essential to induce neurons, but the conditioned medium could not elicit a significant induction. Therefore, they suggested that their results might be mediated by cell surface-tethered factors, labile-soluble factors, or both. We thought the factors for RPE induction in PA6 culture were different from the induction factors for other neurons, such as dopaminergic neurons. In addition, it is known that RPE cell generation sometimes occurs spontaneously from ES/iPS cell culture even without any additional factors. ²⁴ Considering this, the effector molecules for RPE cell induction in PA6-conditioned medium might not promote the induction directly but could work to make a convenient environment for RPE cell differentiation, such as by the repression of other types of cells. With further research, RPE cell induction efficiency is expected to be achieved. 

We are proceeding with autologous and allogeneic transplantation in monkeys using the iPS cell-derived RPE cells described here. These experiments are important not only so we can study the function of a graft in an animal that has foveae and maculae, as do humans, but to confirm the safety of this transplantation method before the clinical trial because the procedure for autologous transplantation experiments in monkeys is essentially same as that for clinical transplantation in human patients. 

Footnotes

 Supported by a Grant-in-Aid for the Leading Project (MT) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

Footnotes

 Disclosure: S. Okamoto, None; M. Takahashi, None

The authors thank Yu Wataoka, Kazuyo Sadamoto, and Akane Nomori for technical assistance; Fumitaka Osakada, Yasuhiko Hirami, and Chie Ishigami for technical advice; and Kazutoshi Takahashi, Koji Tanabe, and Shinya Yamanaka for instruction in iPS cell generation.