SEARCH

SEARCH BY CITATION

Keywords:

  • ES cell;
  • eye;
  • retina;
  • lens;
  • Pax6

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Embryonic stem cells have the potential to give rise to all cell lineages when introduced into the early embryo. They also give rise to a limited number of different cell types in vitro in specialized culture systems. In this study, we established a culture system in which a structure consisting of lens, neural retina, and pigmented retina was efficiently induced from embryonic stem cells. Refractile cell masses containing lens and neural retina were surrounded by retinal pigment epithelium layers and, thus, designated as eye-like structures. Developmental processes required for eye development appear to proceed in this culture system, because the formation of the eye-like structures depended on the expression of Pax6, a key transcription factor for eye development. The present culture system opens up the possibility of examining early stages of eye development and also of producing cells for use in cellular therapy for various diseases of the eye. Developmental Dynamics 228:664–671, 2003. © 2003 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The therapeutic use of cells for the correction of disease, generally referred to as cell therapy, is expected to provide an opportunity to cure untreatable diseases caused by cell damage or dysfunction; such therapy has already been put to practical use, for example, in the case of bone marrow transplantation (Weissman, 2000). For the general source of therapeutic cells, those differentiated from adult stem cells such as neuronal stem cells, mesodermal stem cells of the bone marrow, hematopoietic stem cells, and keratinocyte stem cells are now being applied or planned to be applied (Weissman et al., 2001; Keller, 2002, and the references therein). An alternative choice of the stem cell is embryonic stem (ES) cells isolated from the inner cell mass of the preimplantation embryo (Evans and Kaufman, 1981; Martin, 1981; Brook and Gardner, 1997). As ES cells continue to proliferate in an undifferentiated state in vitro, an unlimited stem cell source or its derivatives, thus, may be secured. It is also a potential benefit that ES cells may be genetically manipulated to permit the selective differentiation and/or isolation of a specific cell type.

Because of their potential to give rise to virtually all cell lineages when introduced into a blastocyst, ES cells also differentiate into a variety of cell types in vitro when appropriate environmental cues are reproduced in the culture system (O'Shera, 1999; Prelle et al., 2002; Turksen, 2002; Wichterle et al., 2002; Kim et al., 2002). In these efforts, culture systems have been refined to selectively induce specific cell types and not to generate other cell types by using a combination of growth factors and forced expression of lineage-specific transcription factors. However, as observed during embryogenesis, blastomeres are programmed to form an integrated set of functional cells called tissues or organs. Therefore, it is possible that developmental programs are also effective to induce tissue-related structures from ES cells in vitro. In fact, in the first step of the differentiation of a specific cell type, ES cells are induced to form embryoid bodies (EBs) containing a variety of cell types or their precursors. However, structures corresponding to tissues or organs that require the proper organization of different lineage cells are generally not formed in vitro (O'Shera, 1999). We and others previously showed that ES cells cultured on a monolayer of feeder cells collectively called stromal cells differentiate into a variety of cell types (Nakano et al., 1994; Yamane et al., 1997, 1999; Kawasaki et al., 2000, 2002). In these culture systems, proper combinations of additives, including growth factors and hormones, preferentially induced specific cell types. In this report using a stromal cell-dependent culture system, we unexpectedly found that cells expressing specific phenotypic markers representative of lens cells and neural and pigmented retina were induced from single ES cells in a closely associated cluster of differentiated cells.

Although many protocols have been introduced to induce neuronal cells or their precursors and successfully applied to regenerate the affected central or peripheral nervous systems, the induction of the structures formed by partly organized clusters of differentiated cells has not been shown before. Efficient and reproducible induction of retinal cells including pigmented retina and photoreceptor cells from ES cells has great potential in cell-replacement therapy for currently untreatable retinal disorders.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Morphology of the Eye-Like Structures Induced From ES Cells

By using the PA6 stromal cell line as a feeder layer to support the differentiation of ES cells, we observed evidence of the induction of organized structures that appeared to contain pigmented retinal cells (Fig. 1A). As shown in Figure 1B,C, and H, typical structures derived from ES cells placed on the PA6 cell layer resembled a vertical section of the vertebrate eye: a refractile cell mass containing fibrous structures was found adjacent to a cluster of pigmented epithelium-like cells. For comparison, a longitudinal section of adult mouse eye is presented in Figure 1L. These structures, designated as eye-like structures, were induced from different ES cell lines, including D3, CCE, and J1 and from several of our independently established ES cell lines from B6 mouse embryos. In day 7 cultures (Fig. 1D), the ES cell-derived colonies were simple aggregates; whereas after day 9, globular structures and retinal pigmented epithelium (RPE) -like cells with weak pigmentation emerged (Fig. 1E), and then at day 11 or later, distinctive eye-like structures were formed (Fig. 1F). On day 11, higher magnification images of highly pigmented cells showed polygonal cell structures containing relatively large melanin granules (Fig. 1G), suggesting that these cells constitute RPE. When these RPE-like cells were derived from ES cell lines established from transgenic mouse embryos containing a fusion construct formed with the transcriptional control sequence of the pigment cell-specific Dct (DOPAchrome tautomerase) gene and the lacZ reporter (Jordan and Jackson, 2000), they expressed lacZ activity, further supporting the view that these cells represent RPE (Fig. 1H). A cross-section of the eye-like structure (Fig. 1I) indicated that the refractile cell mass consisted of multiple cell layers, whereas the marginal RPE layer formed a single cell layer. Enhanced green fluorescent protein (EGFP) signals were observed in these structures when they were induced from ES cells that constitutively expressed EGFP, thus making it highly unlikely that PA6 stromal cells contributed to these structures (Fig. 1J,K).

thumbnail image

Figure 1. Induction of eye-like structures from embryonic stem (ES) cells. A,B: Colonies derived from single ES cells form characteristic eye-like structures at day 11 of the culture (indicated by arrowheads in A and shown at higher magnification in B). C: The culture was stained with crystal violet, and a lens-like structure (arrowhead) was intensively stained. A cluster of pigmented cells is indicated by the asterisk. D: At day 7 of the culture, most of the colonies were simple aggregates of cells and did not contain pigmented cells. E,F: By day 9, a reflective, round structure (arrowhead) and a cluster of pigmented cells (arrow) has emerged in some colonies (E), and then at day 11 these structures display an eye-like appearance (F). G: A higher-magnification image of the pigmented cells in the structure is also shown. H: Also, eye-like structures induced from ES cells established from a Dct-lacZ transgenic mouse were stained with X-gal. Note that the pigmented cells and their surrounding cells were stained. I: A cross-section of the eye-like structure is also shown. The boundary between single-layered pigmented cells and the multi-layered cell mass is shown by the arrowhead. J,K: ES cells constitutively expressing enhanced green fluorescent protein (EGFP) were induced to form eye-like structures (J), and almost all cells constituting these structures were EGFP positive (K). L: A longitudinal section of an adult Balb-c mouse eye is shown for comparison. Scale bars = 1.5 mm in A, 400 μm in B–F,H–K, 20 μm in G, 1 mm in L.

Download figure to PowerPoint

Expression of Marker Genes of Lens and Retina in the Eye-Like Structures

Lens-specific protein αA-crystallin (Sawada et al., 1993) was found immunocytochemically to be expressed in a small cluster of cells in the refractile cell mass of the eye-like structure (Fig. 2A,B). Other lens-specific proteins, αB-crystallin (Fig. 2C,D) and β-crystallin (Fig. 2E,F), were detected in greater abundance in the cell masses. Cells expressing some retinal marker proteins (Haverkamp and Wassle, 2000) were also detected in the cell mass close to the RPE cluster, i.e., Brn3b (Fig. 2G) and syntaxin (Fig. 1H), which are ganglion cell- and amacrine cell-specific proteins, respectively. Synaptophysin (Fig. 2I) and vesicular γ-aminobutyric acid (GABA) transporter (Fig. 2J), which are known to be expressed in several retinal subtype cells, were also expressed. Rhodopsin and recoverin, both specifically expressed in photoreceptor cells, were expressed in these cell masses, as shown in Figure 2K,L, and were mostly colocalized (Fig. 2M). On the other hand, αB-crystallin and the retinal marker protein calbindin were detected in separate compartments of the eye-like structure, as shown in Figure 2O,P. The cell mass forming the eye-like structure contained multiple cell layers, as shown in Figure 1I. These layers seem to represent the layer structure of the retina to some extent, as most of the cells expressing specific retinal or lens marker proteins formed cognate clusters of cells, as shown in Figure 2N–Q. In Figure 2P, the red syntaxin-positive cell layer was observed to extend beneath the green rhodopsin-positive layer (the extended layer is out of focus in this photo); and in Figure 2Q, the red calbindin-positive layer was observed under the green rhodopsin-positive layer. Expression of crystallin subtypes and of phosducin, a retinal photoreceptor specific signal transduction molecule, was also detected by reverse transcription-polymerase chain reaction (RT-PCR; Fig. 2R). These observations show that lens and retinal cell differentiation was induced under these culture conditions. Individual cell types were not randomly distributed but clustered to a certain extent, indicative of the layered structures observed in the retina.

thumbnail image

Figure 2. Expression of eye-specific molecular markers in the induced eye-like structures. α-A crystallin (fluorescent image in A and brightfield image in B), α-B crystallin (C,D), and β-crystallin (anti-β6, E,F), were immunostained. Fluorescent images of Brn3b (G), syntaxin (H), synaptophysin (I), and vesicular γ-aminobutyric acid transporter (J) are also presented. Double immunostaining for rhodopsin (K, green fluorescence) and recoverin (L, red fluorescence) was done, and the merged image (M) shows almost complete colocalization of these two photoreceptor-specific proteins. The merged image of another photoreceptor-specific protein, peripherin (red), indicates its colocalization with rhodopsin (green, N). Double immunostaining for syntaxin (red) and rhodopsin (green, O) and its magnified part (P) indicates mostly separate localization of these markers. Rhodopsin (green) and syntaxin (red) also indicate separate localization (Q). R: mRNAs of crystallin subtypes and rhodopsin were analyzed by reverse transcriptase-polymerase chain reaction. Lane 1, day-10 culture of PA6 only; lane 2, day-10 culture of embryonic stem cells with PA6; lane 3, day-10 whole embryo. Scale bars = 400 μm in A–J,N,O, 40 μm in K–M,P,Q.

Download figure to PowerPoint

Optimal Culture Conditions for the Induction of the Eye-Like Structures

To determine the optimal conditions for the induction of these eye-like structures, we then varied the culture conditions that had been designed previously for the induction of melanocytes from ES cells. The number of ES cells plated was fixed at 500 per well in six-well dishes, because the number of eye-like structures increased in proportion to the cell number when 100 to 1,000 ES cells were used. Then the effects of basic fibroblast growth factor (bFGF), cholera toxin (CT), and dexamethasone (Dex) were tested at the concentrations of 20 pM, 10 pM, and 10 nM, respectively. As shown in Figure 3A, the presence of all three of these factors throughout the culture period resulted in the highest number and percentage of eye-like structures per culture dish. Increasing the concentration of bFGF to 100 pM, of CT to 40 pM, and of Dex to 20 nM caused no marked change compared with the effects of the lower concentrations; and further increases in each concentration decreased the efficiency of the induction of the eye-like structures. We then tested the effect of the timing of the addition of these factors in various combinations. In the presence of bFGF, Dex was most effective when added from day 3 of the culture (data not shown), rather than when it was present throughout the culture period. In the presence of bFGF (throughout the culture period) and Dex (from day 3), the addition of CT from day 0 to 3 caused the highest rate of induction of the eye-like structures (Fig. 3B); whereas changing the time of bFGF addition did not cause any increase (data not shown). Based on these data, our standard conditions were fixed as described in the Experimental Procedures section.

thumbnail image

Figure 3. Effects of basic fibroblast growth factor (bFGF), dexamethasone (Dex), and cholera toxin (CT) on the induction of eye-like structures from embryonic stem (ES) cells. A: Each factor was present (+) or not (−) in the cultures for 11 days, after which the number of eye-like structures containing pigmented retinal cells was scored. B: Each factor was present during the indicated period, and the number of eye-like structures containing pigmented retinal cells was scored. From 500 ES cells inoculated, an average of 200 colonies was induced; and 69% of these colonies formed the eye-like structures under the best conditions, which was the case for the data shown in B.

Download figure to PowerPoint

Developmental Characterization of the Eye-Like Structures

We next tested the expression of genes known to be pivotal for normal eye development. Pax6, a paired-type homeodomain protein, is a major regulator of eye development and is required for the formation of eye structures, the maintenance of lens competence, and the multipotency of retinal cells (Grindley et al., 1995; Marquardt et al., 2001). As shown in Figure 4A, Pax6 mRNA was detected from day 8 of the culture period, when no obvious eye-like structures had yet formed. Expression of other transcription factors such as Sox2, c-Maf, and Prox1, all known to be important for lens development (Ogino and Yasuda, 2000; Kamachi et al., 2001), was also detected from day 8 (Fig. 4A). mRNAs of the soluble signaling molecule bone morphogenic protein 4 (BMP-4), important for lens induction (Ogino and Yasuda, 2000), and lens-specific γ-crystallin (Sawada et al., 1993) were detected on day 8. These expression patterns suggest that transcription factors governing eye development are expressed in this culture system in a time-dependent manner.

thumbnail image

Figure 4. Genetic control and consequences of formation of embryonic stem (ES) cell-derived eye-like structures. A: Reverse transcriptase-polymerase chain reaction detection of expression of genes affecting eye development. As a control, PA6 cells were cultured without ES cells (indicated as P) for the indicated time under the same conditions as used for the induction of eye-like structures from ES cells with PA6 (indicated as E). B:Dct-lacZ ES cells were induced to differentiate and stained by X-gal on the day indicated. On day 7, most colonies were lacZ negative; and on day 8, approximately 70% of the total colonies became lacZ positive, and this lacZ-positive population was maintained on day 9. On day 9, pigmented cells expressing lacZ could be first recognized as indicated by the arrowheads in the enlarged photo. Total number of lacZ-positive cells/six-well plate at the indicated culture days are shown in the histogram. C: The phenotype of Pax6Sey/Sey (Pax6−/−) mice. Note the complete lack of eyes in the Pax6−/− mouse. D: Immunocytochemical staining for Pax6 in eye-like structures induced from Pax6+/− ES cells. E,F: From Pax6+/− ES cells, normal eye-like structures were formed (E), whereas only cell aggregates composed mainly of β-III tubulin-positive neurons (data not shown) were induced from Pax6−/− cells (F). From 100 to 1,000 ES cell-derived structures were checked for each of the 7 Pax6+/− and 3 Pax6−/− ES cell lines. A total of 1,369 eye-like structures containing retinal pigmented epithelium were observed in 5,150 colonies scored from Pax6+/− cell lines, whereas no eye-like structure was detected in 2,935 colonies from the Pax6−/− cell lines. Scale bars = 20μm.

Download figure to PowerPoint

To test more precisely the relationship between gene expression and morphology of the eye-like structures, we prepared Dct-lacZ ES cells and analyzed them for the expression of Dct-driven lacZ activity during induction in culture. As shown in Figure 4B, clusters of lacZ-positive cells emerged precisely on day 8 of the culture before the mature RPEs. On day 9, RPEs containing melanin granules differentiated from these lacZ-positive cells. The number of these clusters of lacZ-positive cells did not change after day 8, and all of the lacZ-positive clusters finally differentiated into the eye-like structures. The accurate time-dependent expression of the Dct gene in the precursors of RPEs further suggests that developmental programs governing the normal developmental process also work in the present ES cell culture. In fact, pigmentation of the RPEs start from E11.5 to 12.5, and this timing well correlate with the in vitro development provided that ES cells represent E 3.5 inner cell masses.

Finally, the genetical basis of the development of the eye-like structures was investigated by using ES cell lines established from the embryos of a Pax6 mutant (Pax6Sey, Hill et al., 1991) in which eye structures are absent in the homozygote (Fig. 4C, shown as Pax6-/-). Expression of Pax6 protein was detected throughout the cell mass adjacent to the RPE (Fig. 4D). Whereas seven of seven Pax6Sey/+ ES cell lines developed eye-like structures and a fully mature RPE (Fig. 4E), and all three of the three independent Pax6Sey/Sey ES cell lines formed smaller cell aggregates and no RPE (Fig. 4F). The developmental deficiency of Pax6Sey/Sey ES cells appeared to be restricted to eye development, because hematopoietic cells, cardiac muscle cells, neuronal cells, and melanocytes were induced normally (data not shown). This effect of the mutation of Pax6, whose function in eye development is well conserved throughout the animal kingdom (Gehring and Ikeo, 1999), indicates that these eye-like structures were formed according to the normal developmental program for the vertebrate eye.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

In this report, we designated structures formed by the cells corresponding to pigmented retina, lens, and neural retina as eye-like structures. However, unlike the normal eye, which is composed of a separate lens and an organized layered structure of individual retinal cell types, most of the ocular cell types except RPEs were more or less mixed within the multilayered cell masses induced from ES cells. Despite this apparent difference, we would like to emphasize that most of the cell masses forming eye-like structures contained various ocular cell types, including lens and retina as well as RPEs. Each cell type formed mutually exclusive cell clusters in the eye-like structures, indicative of the layered structure in vivo, as shown in Figure 2P,Q. In addition, stable and efficient induction of various ocular cell types from a single undifferentiated ES cell may exclude the possibility of the stochastic and sporadic nature for the generation of observed cell types during the growth of the cell masses. In these respects, unlike the embryoid bodies, which contain several cell types of all three germ layers, our observed structures were highly concentrated and partly organized structures representing at least a part of the normal eye structure. In fact, the existence of synaptophysin concentrated in specific regions suggests synapse formation between neuronal cells, as observed in the neural retina. The dependence of the formation of the eye-like structures on Pax6, a master gene for eye morphogenesis, also strongly supports our notion.

Because some markers used for the neural retina may be expressed in other nervous systems, it is possible that other cell types not related to the ocular system may exist in the culture. In fact, PA6, originally developed for use in in vitro long-term bone marrow cultures, was used for the selective induction of neuronal cells other than neural retina (Kawasaki et al., 2000), although the culture conditions used were different. Nonetheless, markers for lens and photoreceptor cells are fairly exclusive for the eye, and cells positive for these markers were concentrated in the eye-like structure, as indicated in Figure 2. It also should be noted that the induced cells might not be fully mature. Based on observations by immunoelectron microscopy, some rhodopsin-positive cells showed an elongated shape indicative of immature precursors for photoreceptor cells. However, specific intracellular structures such as photoreceptor discs were not observed even after day 30 of the culture (data not shown).

The presently observed eye-like structures may not necessarily reflect the product of the normal eye development program but may have resulted from partial transdifferentiation into lens cells after induction of the pigment or neural retina cells from ES cells (Eguchi and Kodama, 1993). Pax6-dependent induction of the eye-like structures may just indicate Pax6-dependent induction of RPEs followed by their transdifferentiation into other cell types. When lacZ-positive RPE precursors were detected suddenly on day 8 among Dct-lacZ ES cells, we noticed that the refractile cell mass containing lens and neural retina had already been formed, suggesting that transdifferentiation from RPEs or their precursors might not be the case. Still, it could be that other cell types participated as the parental cells for transdifferentiation. Detailed analyses of the timing and expression of each cell type in the eye-like structures might be necessary.

In addition to the intrinsic genetic program for eye development, a series of reciprocal tissue interactions between the head ectoderm and the region of neuroectoderm called the optic cup is required for eye development (Grainger, 1992; Ogino and Yasuda, 2000). Because the induction rate of the eye-like structures among the ES cell-derived colonies was very high and the structures induced showed organized morphology to some extent, it is likely that these interactions occur even in the simple culture system described here, just as recently suggested for the induction of pancreatic islet-like cell masses secreting insulin (Lumelsky et al., 2001) and gut organ-like structures (Yamada et al., 2002), both from embryoid bodies. At present, we could not describe the factors responsible for the induction of eye-like structures. Although Figure 3A apparently indicates bFGF as an indispensable factor for the induction of the eye-like structures, we could not induce the eye-like structures without PA6, even in the presence of bFGF. Combinational stimuli of PA6-derived factors and bFGF may be necessary for the efficient induction of the eye-like structures. The present culture system should be useful for examining early stages of eye development as well as for producing cells for use in cellular therapy for various eye diseases.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Cell Culture

ES cell lines and the PA6 stromal cell line were maintained and prepared for in vitro differentiation as described (Yamane et al., 1999). For differentiation, 500 to 1,000 trypsinized ES cells were inoculated into six-well plates previously seeded with PA6 cells and differentiated in a-Minimal Essential Medium supplemented with 10% fetal calf serum under 5% CO2/95% air at 37°C. Unless otherwise mentioned, incubation with 20 pM bFGF and 10 nM Dex was started at day 0 and day 3, respectively; and that with 10 pM cholera toxin was done from day 0 to 3 only. The medium was changed twice a week. In some cases, gelatin-coated coverslips placed in six-well dishes were used for the culture.

Establishment of ES Cell Lines From Mouse Embryos

The detailed protocol was described elsewhere (Tokunaga and Tsunoda, 1992). In brief, blastocysts were prepared from Dct-lacZ transgenic mice or Pax6Sey mice. The ES cells were isolated and maintained on a primary mouse embryonal fibroblast cell feeder layer in Dulbecco's Modified Eagle's Medium supplemented with 17.5% Knockout Serum Replacement (GibcoBRL), 10−4 M 2-mercaptoethanol, nonessential amino acid solution (GibcoBRL), and human LIF (20 ng/ml). Pax6Sey/Pax6Sey ES cells were genotyped by RT-PCR using 5′gcttcagctgaagtcgcatctg3′ and 5′aggaaggctttgtggaggcgcaga3′ as primers. The genomic DNA fragment containing Pax6Sey was cut out by using DdeI enzyme.

Establishment of GFP-Expressing ES Cell Line

cDNA encoding EGFP was placed under the CAG promoter of the PCXN2 vector, in which the neor sequence was inserted into the pCAGGS expression vector. D3 ES cells were transfected with the DNA by electroporation. Among 50 independent G418-resistant clones, #24, which expressed EGFP after differentiation to hematopoietic cells and melanocytes, was used for further analysis.

Immunocytochemistry

Mouse anti-αA crystallin (2G2B6), anti-αB (2G6B6) crystallin, and anti-β crystallin (β5 and β6) monoclonal antibodies (culture supernatant, RIKEN Cell Bank), goat anti-Brn3b polyclonal antibody (1:200 dilution, Santa Cruz), rabbit anti-syntaxin (HPC-1) polyclonal antibody (1:500, a gift from Dr. K. Akagawa), mouse anti-synaptophysin monoclonal antibody (1:50, Neo Markers), rabbit anti-vesicular GABA transporter polyclonal antibody (1:200, Synaptic System), rabbit anti-rhodopsin polyclonal antibody (1:1,500, LSL), rabbit anti-recoverin polyclonal antibody (1:1,000, a gift from Dr. Karl-W. Koch), rabbit anti-peripherin polyclonal antibody (1:100, Chemicon), mouse anti-calbindin monoclonal antibody (1:500, Sigma), and rabbit anti-Pax6 polyclonal antibody (1:1,000, a gift from Dr. A. Kawakami) were used as primary antibodies. Cultures were fixed with 4% paraformaldehyde for 10 min at 4°C. For staining with Pax6 and αA-crystallin antibodies, the cultures were fixed with cold methanol for 15 min at 4°C and with acetone for 10 min at room temperature, respectively. Fixed cells were permeabilized and incubated with primary antibodies for 40 min at room temperature and with secondary antibodies labeled with fluorescein isothiocyanate, Alexa Fluor 488, or Alexa Fluor 546 for 40 min at room temperature.

RT-PCR

Isogen reagent (Wako) was added to the cultures to extract total RNA. One microgram of total RNA was subjected to reverse transcription by using oligo (dT) and reverse transcriptase, and then one tenth of the product was used for PCR.

Primer sequences (forward and reverse) and the length of the amplified products were as follows: α A-crystallin: 5′-GACTGTTCGACCAGTTCTTCGG-3′, 5′-GAAGGTCAGCATGCCATCAGC-3′ (365, 434bp); β A3/A1-crystallin: 5′-TCGATGGGATGCCTGGAG-3′, 5′-GGATTTTTGGAACTTTTGCTCTAT-3′ (386 bp); γ E/F-crystallin: 5′-CTGCTGGATGCTCTATGAGC-3′, 5′-CGTGGAAGGAGTGGAAGTCAC-3′ (252 bp); phosducin: 5′-GTGCATGCAGGATATGCATC-3′, 5′-CTCTAGGTCATGTATCTCTC-3′ (420 bp); G3PDH: 5′-CACCACCATGGAGAAGGCCGGG-3′, 5′-GTGTAGCCCAAGATGCCTTCA-3′ (525 bp); Pax6: 5′-CCCACAGCCCACCACACCTGTCTC-3′, 5′-GGTGAAATGAGTCCTGTTGAAGTG-3′ (300 bp); Sox2: 5′-TGCAGGAGCAGCTGGGCTAC-3′, 5′-TGGGCTATGTGCAGTCTACT-3′ (366 bp); c-Maf: 5′-GTGCAGCAGAGACACGTCCT-3′, 5′-CAACTAGCAAGCCCACTC-3′ (272bp); Prox1: 5′-GGACGGTAGGGACAGCATCT-3′, 5′-TGGAAAAGGCATCATGGCAT-3′ (190 bp); and Bmp-4: 5′-TGTGAGGAGTTTCCATCACGAAGAA-3′, 5′-AGTTCTTATTCTTCTTCCTGGACCG-3′ (570 bp). Amplification conditions were 94°C for 4 min followed by 35 cycles of 94°C for 30 sec, 60°C for 1 min, and 72°C for 1 min.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Drs. Ian Jackson and Siobhan Jordan for providing Dct-lacZ mice, Hiromi Okuyama for establishing GFP-expressing D3ES cell lines, and Drs. Shin-ichi Nishikawa and Satomi Nishikawa for PA6 cell lines, Dr. K.W. Koch for anti-recoverin antibody, Dr. Atsushi Kawakami for anti-Pax6 antibody, and Dr. Kimio Akagawa for anti-syntaxin antibody. We also thank Drs. Mizuho Sato and Shin-ichi Aota for technical assistance, Dr. Akira Hara for providing Figure 1L, and Dr. Kiyokazu Agata for technical advice.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES