Direct Identification and Enrichment of Retinal Stem Cells/Progenitors by Hoechst Dye Efflux Assay

Abstract

purpose. The present study describes a method for isolating neural stem cells/progenitors directly from the freshly dissociated embryonic retina (prospective identification) and compares their characteristics with those enriched from mitogen-exposed embryonic retinal cell culture.

methods. Cell dissociates from embryonic rat retina and mitogen-exposed embryonic retinal cultures were stained with Hoechst 33342 fluorescent dye. The emission patterns of cells were analyzed in both blue and red wavelength using flow cytometry to enrich cells that retained or excluded the dye. The phenotype characteristics and differentiation potential of enriched cells were analyzed by immunocytochemical, RT-PCR, and electrophysiological analyses.

results. The Hoechst dye efflux assay identified a minor population of cells, called side population (SP) cells, in fresh retinal dissociates. These cells that preferentially excluded the Hoechst 33342 fluorescent dye were proliferative and expressed both neural progenitor and retinal progenitor markers. The retinal SP cells generated functional neurons and glia and possessed the ability to differentiate along lineages of different late-born retinal cell types. Cells of similar phenotypes and potential were observed in the SP obtained from mitogen-exposed retinal culture.

conclusions. The Hoechst dye efflux assay represents an effective method for direct identification of retinal stem cells/progenitors. These results demonstrate that the prospectively isolated retinal stem cells/progenitors and those enriched as SP cells from mitogen-exposed retinal cell culture may be similar in their properties and potential.

Neural stem cells/progenitors that give rise to neurons and glia have been identified in different regions of the brain, including the retina. ¹ These cells can be maintained in a proliferative state in culture, usually in the presence of mitogens, epidermal growth factor (EGF), and/or fibroblast growth factor (FGF2). Withdrawal of mitogens from the culture and addition of serum or growth factors induce their differentiation into neurons and glia. ² The fact that cultured stem cells/progenitors can acquire site-specific phenotypes on transplantation suggests their potential usefulness in treating neurodegenerative changes. ^{3 4 5 6} However, in most cases, neural stem cells/progenitors have not been enriched directly from freshly dissociated neural tissue (prospective identification) and the question therefore remains of whether the potential that they display is inherent or acquired from prolonged exposure to mitogens used to maintain them in culture. This issue is a significant one for both the biology of stem cells and their use as therapeutic reagents. To investigate the inherent potential of neural stem cells/progenitors, characterization of properties is necessary that allows direct identification of neural stem cells/progenitors and study of their primitive characteristics over generations. With this objective in mind we have begun the identification and characterization of retinal stem cells/progenitors based on strategies developed for the isolation of hematopoietic stem cells (HSCs). 

There are currently two different approaches for prospective identification of HSCs. The first approach involves fluorescence-activated cell sorting (FACS), in which monoclonal antibodies to specific cell surface markers is used. This approach has been used successfully to isolate self-renewing, multipotent stem cells from the peripheral nervous system, with the low-affinity neurotrophin receptor p75 used as a surface marker. ⁷ More recently, specific monoclonal antibodies that recognize the surface markers, CD133 and 5E12, have been used for prospective identification of self-renewing and multipotent neural stem cells from human fetal brain. ⁸ Another approach to identify hematopoietic stem cells/progenitors is based on the ability of these cells to exclude selectively Hoechst 33342 dye, which can be enriched as a part of a minor population, called the side population (SP), by FACS. ⁹ This approach has been used to enrich neural progenitors from neurospheric culture. ¹⁰ We used the Hoechst dye exclusion assay for direct identification of neural stem cells/progenitors from freshly dissociated embryonic day (E)18 rat retina and demonstrate that their properties and potential are similar to those enriched from in vitro expanded E18 retinal cells. 

Methods

Dissociation of Embryonic Retina

Animals were handled in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Timed-pregnant E18 Sprague-Dawley rats were obtained from the supplier (SASCO, Wilmington, MA), and embryos were harvested in Hanks’ balanced salt solution (HBSS). Eyes were enucleated with minimum extraneous tissue and placed in HBSS in a separate Petri dish. The optic nerve and remaining mesenchymal tissues were carefully removed before the retina was isolated, to prevent the possible contamination of retina with brain tissues. The retina was carefully teased away from the retinal pigmented epithelium (RPE), and the central portion of the retina surrounding the optic nerve was removed and discarded. The isolated retina was collected in a sterile 15-mL polypropylene tube and dissociated into single cells, as previously described. ² Briefly, the retina was digested with 0.25% trypsin at 37°C for 10 minutes. Cells were washed with fresh retinal culture medium (RCM: DMEM-F12, N2 supplement [Gibco, Gaithersburg, MD], 2 mM l-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin) and triturated with a 1-mL pipette tip. Retinal dissociates were either cultured in the presence of EGF to expand progenitors or processed for the Hoechst dye efflux assay. 

Progenitor Cell Culture

E18 retinal dissociates were cultured in 75 cm² T-flasks at a density of 2 × 10⁴/cm² in the presence of 20 ng/mL EGF (Gibco) at 37°C in 5% CO₂ for 7 days. Most of the cells did not survive in this condition. A subset of cells generated individual colonies called neurospheres. Neurospheres were collected by centrifugation, dissociated into single cells, and processed for the Hoechst dye efflux assay. 

Hoechst Dye Efflux Assay

Dissociates from the fresh retina or mitogen-exposed culture of E18 retinal cells were resuspended in Hoechst Iscove’s modified Dulbecco’s medium (10⁶ cells/mL; IMDM) containing 2% fetal calf serum (FCS) at 4°C overnight, followed by staining with Hoechst 33342 (2.0 μg/mL) at 37°C for 30 minutes. When verapamil (Sigma) was used, cells were stained as just described, in the presence of 50 or 100 μM of the inhibitor. Cells were sorted by flow cytometry (FACStar Plus; BD Biosciences, Lincoln Park, NJ). The Hoechst dye was excited at 350 nm, and its fluorescence was measured at two wavelengths using a 485/22-nm band pass filter (485 BP22) and a 675-nm long pass edge optical filter (675 EFLP; Omega Optical, Inc., Brattleboro, VT). A 610-nm short-pass dichroic mirror was used to separate these emission wavelengths (Omega Optical, Inc.). First, a live gate was defined on the flow cytometer using Hoechst red and blue axes to exclude dead cells that are Hoechst red (very bright), red cells (no Hoechst stain), and debris. ⁹ After 10⁵ events were collected within the live gates, the SP and non-SP (NSP) cells were defined as Hoechst-low and -bright, respectively. The region between SP and NSP cells was defined as extra-SP or upper SP. ⁹ SP, NSP, and extra-SP cells were sorted into tubes (Eppendorf, Fremont, CA) containing 100% FBS. The sorted cells were cytocentrifuged for immunofluorescence analysis or cultured as described earlier and then subjected to immunofluorescence or electrophysiological analyses. 

Immunofluorescence Analysis

Immunofluorescence analysis on undifferentiated and differentiated SP and NSP cells was performed as previously described. ^{2 11} Briefly, cells were fixed in ice-cold 4% paraformaldehyde and incubated with cell-specific marker antibodies (Table 1) in blocking serum at 4°C overnight. After the incubation in species-specific IgG conjugated with CY3 and/or 7-amino-4-methyl-coumarin-3- acetic acid (AMCA) or FITC, cells were washed with PBS and examined by microscope. 

RT-PCR Analysis

Total RNA was isolated with a kit (Qiagen, Valencia, CA). cDNA (2 μM) was amplified with gene-specific primers (Table 2) by using the following step-cycle program (Robocycler; Stratagene, La Jolla, CA): denaturation at 94°C for 30 seconds, annealing at specific temperature (Table 2) for 35 seconds, and extension at 72°C for 40 seconds for 30 cycles, followed by a final extension at 72°C for 5 minutes. PCR products were resolved on 2.0% agarose gel against 100-bp DNA marker (MBI Fermentas, USA). 

Electrophysiological Analysis and Ca²⁺ Imaging

For electrophysiological and Ca²⁺ imaging studies, cells were plated on coverslips, placed in a chamber, and perfused on the stage of an upright, fixed-stage microscope (for electrophysiology, model BHWI; Olympus, Lake Success, NY; for imaging experiments, model E600FN; Nikon, Melville, NY) with an oxygenated solution containing NaCl, 140 mM; KCl, 5 mM; CaCl₂, 2 mM; MgCl₂, 1 mM; HEPES, 10 mM; glucose, 10 mM (pH 7.4). Experiments were performed at room temperature. For whole-cell recording, patch pipettes were pulled on a vertical puller (model PB-7; Narishige, Tokyo, Japan) from borosilicate glass pipettes (1.2 mm outer diameter, 0.95 mm inner diameter; Omega Dot; Stoelting Co., Chicago, IL) and had tips of 1 to 2 μm outer diameter with tip resistances of 6 to 12 MΩ. Pipettes were filled with a bathing solution containing KCH₃SO₄, 98 mM; KCl, 44 mM; NaCl, 3 mM; HEPES, 5 mM; EGTA, 3 mM; MgCl₂, 3 mM; CaCl₂, 1 mM; glucose, 2 mM; Mg-adenosine triphosphate (ATP), 1 mM; guanosine triphosphate (GTP), 1 mM; and reduced glutathione, 1 mM (pH 7.2). For Ca²⁺ imaging studies, cells were incubated for 45 minutes in Fura-2/AM (10 μM; Molecular Probes, Eugene, OR) plus nonionic detergent (10 μM; Pluronic F127; Biotium, Hayward, CA). After loading, cells were rinsed (1 mL/min) for 30 minutes by perfusion with the bathing solution. Test solutions were applied by bath perfusion. Cells were viewed through a 60× water-immersion objective (1.0 numerical aperture [NA]; Nikon), and fluorescence was stimulated with a 150-W xenon lamp attached to the microscope’s epifluorescence port by a liquid light guide (Sutter Instruments, San Raphael, CA), and excitation wavelengths (340 or 380 nm) were changed with a filter wheel (Lambda 10-2; Sutter Instruments). Fluorescence changes were monitored with a cooled charge-coupled device (CCD) camera (SensiCam; Cooke Corp., Auburn Hills, MI, using Imaging Workbench software; Axon, Inc., Burlingame, CA). Images were acquired every 5 to 10 seconds, using 2 × 2 binning with 0.5- to 1-second acquisition times. 

Coculture of SP Cells and Postnatal Retinal Cells

SP cells were plated on poly-d-lysine– and laminin-coated glass coverslips and cultured with postnatal day (PN)1 rat retinal cells across a 0.4-μm membrane (Millicell CM; Millipore, Milford, MA) as previously described. ² After 1 week, SP cells were fixed in 4% paraformaldehyde followed by immunocytochemical analyses. 

Results

Selective Exclusion of Hoechst Dye by a Subpopulation of Embryonic Retinal Cells

Retinal dissociates, obtained as previously described, ² were resuspended in 2 μg/mL Hoechst dye and incubated for 30 minutes at 37°C followed by FACS analysis using the dual Hoechst emission profile. FACS analysis revealed the presence of a subpopulation of retinal cells similar to HSC-SP (Fig. 1A 1C) . A similar result was obtained with cell dissociates obtained from fresh embryonic hippocampus (Fig. 1B) . The emergence of the retinal SP was extremely sensitive to dye concentration and the time of incubation in the dye. The retinal SP cells could not be discriminated from NSP cells that retained the dye if the concentration of the dye was greater than 2 μg/mL. The proportion of total retinal cells in the SP was ∼0.1%. To test the specificity of the subpopulation of retinal cells that selectively exclude the Hoechst dye, staining was performed in the presence of verapamil on an aliquot of retinal dissociates. It has been observed that the efflux of Hoechst dye from stem cells is verapamil sensitive and involves ABCG2 transporter. ^{9 25 26 27} Staining of retinal cells in the presence of verapamil eliminated the SP cells, and the effect of the drug was concentration dependent (Fig. 2) . To investigate the possibility that SP cells may be present due to contamination with blood during the isolation of retinal cells, expression of different hematopoietic lineage antigens by retinal cells from either the fresh tissue or neurospheres obtained from an EGF-exposed culture of E18 retinal cells was analyzed (Table 3) . The proportion of retinal cells expressing hematopoietic lineage markers was not substantial enough to account for the emergence of the SP cells from retinal dissociates. A significant number of retinal cells expressed Thy1 antigen (CD 90) because mature retinal ganglion cells are Thy1 positive. ²⁸ In contrast, nestin, a neuroectodermal stem cell marker, was expressed only by retinal dissociates and not by bone marrow cells (Table 3) . 

Expression of Retinal Progenitor Markers by Retinal SP Cells

To determine the nature of cells in the retinal SP, we investigated the proliferative potential of cells and the expression of different markers that characterize neural and retinal progenitors (Fig. 3) . The majority (88.2% ± 5.9%) of SP cells were proliferative, as they expressed Ki67 antigen, a cell proliferation marker. These cells expressed the neural progenitor markers nestin (91.1% ± 6.1%) and Notch1 (92.6% ± 6.4%) and the retinal progenitor marker Chx10 (86.4% ± 2.8%). There was no discrimination between nuclear and cytoplasmic distributions of specific antigens because stem cells/progenitors generally have a high nuclear-to-cytoplasmic ratio and because cells were cytocentrifuged for the purpose of immunocytochemical analysis, resulting in the loss of cytoplasmic processes. To determine whether SP cells represent distinct subpopulations, the expression of pan-neural and retinal progenitor markers was performed by double immunocytochemical analysis. Cells that expressed pan-neural markers, nestin, and/or Notch1, also expressed the retinal progenitor marker Chx10 (data not shown). 

RT-PCR analysis was used to compare the expression of cell-type–specific markers between SP and NSP cells (Fig. 4) . Transcripts corresponding to the general neural markers Map2 and glial fibrillary acidic protein (GFAP) and those encoding the retinal cell–specific markers opsin and syntaxin 1 were expressed in NSP cells, but not in SP cells, suggesting that the latter largely consists of undifferentiated cells. Results of RT-PCR analyses were corroborated by immunocytochemical analyses; while nestin (91.1% ± 6.1%) was expressed exclusively in SP, opsin (15.38% ± 5.5%) and syntaxin 1 (3.15% ± 0.9%) immunoreactivities were detected only in NSP cells. In addition, Ki67 positive cells (88.2% ± 5.9%) were found only in SP cells. 

Enrichment of Retinal SP Cells from Primary Neurospheres

To analyze the characteristics of progenitors exposed to mitogens, dissociates from E18 retina were cultured in the presence of EGF. In this culture condition, subsets of proliferating cells generated individual colonies called neurospheres (Figs. 5A 5B 5C) . Neurospheres were collected and dissociated, and cells were subjected to a Hoechst dye efflux assay. A robust SP was obtained that contained 1.0% of total retinal cells analyzed compared with 0.1% obtained from freshly dissociated retina (Fig. 5D) . These cells were not likely to be present because of contamination of neurospheric cells with blood cells, because the proportion of cells expressing hematopoietic lineage markers was insignificant (Table 3) . RT-PCR analysis showed transcripts corresponding to Map2, GFAP, opsin, and syntaxin in the NSP cells only, suggesting that the SP consists of undifferentiated cells (Fig. 5E) . Immunocytochemical analysis of these SP cells showed that similar to those directly isolated from E18 retina (Fig. 3) , they were Ki67 positive (88.3% ± 3.7%) and expressed both neural and retinal progenitor markers in similar proportions (nestin-positive cells: 92.1% ± 4.8%, Notch1-positive cells: 95.0% ± 5.5%, and Chx10-positive cells: 89.1% ± 5.5%) as those that were prospectively identified. 

To understand the proliferative nature and differentiation potential of retinal stem cells/progenitors, SP cells obtained from the primary neurospheres were cultured to high density (1.5–2.0 × 10⁴ cells/well) in a 96-well culture plate in the presence of EGF. A subset of the SP cells generated clonal secondary spheres (Fig. 6A) . Most of the cells in the NSP (Fig. 6B) , plated at the same density died. and few that survived did not generate neurospheres. Cells in the secondary neurospheres were proliferative and expressed neural progenitor markers (Figs. 6C 6D 6E 6F 6G 6H 6I 6J) . When EGF was withdrawn from the culture medium and supplemented with 1% serum, SP cells enriched from either freshly dissociated retina or mitogen-exposed neurospheres expressed the pan-neural markers, β-tubulin and GFAP, suggesting that a subset of SP cells is multipotent (Figs. 7 and 8) . The multipotent nature of SP cells was corroborated by RT-PCR analyses that showed that SP cells in differentiation conditions express transcripts corresponding to neuronal (Map2) and glial (GFAP) markers (Figs. 7F 8F) . 

Electrophysiological Analysis of Retinal SP Cells

Analyses of cell-type–specific antigenic markers and transcripts suggested that SP cells are multipotent and differentiate along neuronal and glial lineage. We wanted to know whether differentiated SP cells possess distinct electrophysiological properties in addition to expressing neural markers. This possibility was explored by electrophysiological analysis using whole-cell recording from SP cells in differentiation conditions (Fig. 9) . Voltage-dependent currents were evoked by a series of −20-mV voltage steps (150 ms) from −110 to +70 mV. The following phenotypes were observed based on current-voltage profiles: (1) cells that displayed neuronal features, such as a rapidly activating inward current evoked above −30 mV, attributed to sodium current (I_Na) and a sustained outward current attributed to potassium current (I_K; Figs. 9A 9B ); (2) cells that displayed glial features such as a prominent inward rectifying current at hyperpolarizing potential as observed in the Müller glia of primates ²⁹ (Figs. 9C 9D) ; and (3) cells that displayed a delayed, outward rectifying current uncharacteristic of either neurons or glia (data not shown). It is likely that these cells represent precursors in the process of differentiation, and the current–voltage profile displayed by them may be indicative of their immature nature. We further ascertained the functional differentiation of SP cells by analyzing the expression of N-methyl-d-aspartate (NMDA) and non-NMDA ionotropic glutamate, using Ca²⁺ imaging with Fura-2 (Fig. 10) . Application of kainic acid (KA, 30 μM) evoked an increase in the 340:380 ratio indicating an increase in intracellular [Ca²⁺] in three cells in the field (Fig. 10A2) . The changed ratio, within the circular region of interest, is plotted as a function of time (Fig. 10B) . After recovery from kainic acid, an Mg²⁺-free solution containing glycine (0.1 mM) was applied for 2 minutes, followed by application of the same solution with NMDA (0.1 mM) for 1 minute. Similar to kainic acid, NMDA evoked a Ca²⁺ increase in the cell, denoted by the circular region of interest (Figs. 10A3 10B) . Application of a high K⁺ solution to strongly depolarize the cell also evoked a Ca²⁺ increase in this cell as well as in two others in the field, suggesting the presence of voltage-gated Ca²⁺ channels. Thus, consistent with previous studies on retinal progenitor cells, differentiating conditions promote the expression of various physiological properties, including those commonly associated with neurons (e.g., voltage-dependent Na⁺ currents and ionotropic glutamate receptors) and glia (e.g., prominent inward rectifying currents). 

Generation of Retinal Neurons

The foregoing observations demonstrated that SP cells have neural potential. To determine whether they retain their ability to differentiate into retinal neurons, SP cells enriched from freshly dissociated E18 retina or neurospheres were cocultured with an excess of cells isolated from PN1 retina. We have shown that PN1 retinal cells facilitate the differentiation of cultured retinal progenitors into retinal neurons. ² Because the SP cells represented the late stage of retinal histogenesis, we examined their potential to generate late-born retinal neurons (rod photoreceptors, amacrine cells, and bipolar cells) and the Müller glia. We observed that retinal SP cells obtained from either source could express rod photoreceptor-specific (rhodopsin), bipolar cell-specific (PKC), amacrine cell-specific (syntaxin), and glia-specific (vimentin) markers when cocultured with PN1 cells for 5 to 7 days (Fig. 11) . This observation suggests that, besides possessing the potential to give rise to neurons and glia, SP cells enriched from either a freshly dissociated retina or neurospheres demonstrate a similar capability to differentiate along lineages of different late-born retinal cell types. 

Discussion

One of the most significant barriers in the study of the biology of neural stem cells/progenitors and in their therapeutic use is the lack of information about their inherent proliferative and differentiation potential. ¹ A significant proportion of information regarding the properties of neural stem cells/progenitors has emerged from in vitro studies. It is likely that these cells, exposed to high concentrations of mitogens to maintain them in culture, undergo changes and thereby acquire potential differences from their parents. Therefore, information regarding the inherent and acquired potential of these cells becomes important to shed light on mechanisms and the extent of their plasticity. This information is particularly important for therapeutic use of neural stem cells/progenitors in highly ordered and laminated sensory structures such as the retina, where the presence of undesirable neurons or glia due to heterologous differentiation may exacerbate rather than solve problems. 

In this study we have demonstrated that the Hoechst dye efflux assay can be used effectively for direct identification of retinal stem cells/progenitors and for cross-comparison of properties between populations of progenitors enriched from different sources. Because the number of dividing cells is greater in the developing retina at the stage of late neurogenesis (E18) than at early neurogenesis (E14), ³⁰ we performed the assay on the former to maximize the enrichment of proliferating progenitors. We observed that the dual-wavelength flow cytometric analysis of Hoechst-dye–stained E18 retinal cells yielded two basic populations of cells: the NSP and SP. The differentiated cells were largely distributed in the NSP. The stem cells/progenitors, in contrast, reside in SP that, like HSC SP, is verapamil sensitive. ⁹ The notion, that retinal SP cells represent stem cells/progenitors is supported by the fact that these cells are proliferative and express a number of markers that are characteristic of both neural (i.e., nestin) and retinal (i.e., Chx10) progenitors. However, SP cells account for a small proportion of nestin-positive cells present in either E18 retina or neurospheres (Table 3) . The rest of the proliferating nestin-positive cells found in an extra-SP region between the SP and upper NSP (data not shown) are likely to represent committed precursors in different stages of differentiation. In addition to dividing and expressing progenitor markers, retinal SP cells are undifferentiated, as ascertained by the absence of expression of neural cell– and retinal cell-type–specific markers. However, retinal SP cells are not homogeneous in the expression of progenitor markers. For example, although most cells expressed both nestin and Chx10, a subset of cells was observed that was nestin positive and Chx10 negative. Cellular heterogeneity has also been observed in HSCs and human cord blood SP cells. ^{9 31}  

The prospectively identified retinal progenitors appeared to be similar in characteristics and potential to progenitor cells enriched from EGF-exposed retinal neurospheres. This notion is supported by the following observations. First, both were distributed in the verapamil-sensitive SP. Second, the range and proportion of progenitor markers expressed by cells in both SPs were similar, suggesting that they represent similar progenitor populations and third, both are multipotent, capable of giving rise to neurons and glia in differentiation conditions. Cells from both SPs possessed the capacity to differentiate into late-born retinal neurons. The differentiated progeny, in addition to expressing cell-type–specific markers, were distinguishable on the basis of voltage-gated current profiles and the presence of ionotropic glutamate receptors, as demonstrated previously for cultured retinal stem cells/progenitors. ² These observations suggest that the differentiation of retinal SP cells involves several steps that are characterized by distinct molecular as well as electrophysiological properties. ²  

Retinal progenitors are selectively expanded when exposed to EGF, as suggested by a 10-fold increase in the proportion of the SP cells obtained from the neurospheres compared with those obtained from the fresh retina. Both progenitor populations failed to generate neurospheres when SP cells were seeded at a clonal density. Generation of secondary neurospheres was observed when SP cells enriched from the EGF-exposed neurospheres were cultured at high density. Such a culture condition was not possible with prospectively identified cells, because of their low number (0.1% vs. 1.0% of total cells). This suggests two possibilities: The proliferating cells isolated from retina are not stem cells but rather neural progenitors with a limited self-renewal property, or these cells are indeed stem cells, but conditions have not been identified that promote their self-renewal in vitro. The fact that clonal generation of neurospheres can be observed in high-density culture suggests that the self-renewal property of retinal stem cells/progenitors is a non–cell-autonomous process and therefore requires contributions from other cells. 

Taken together, our results suggest that the Hoechst dye exclusion assay represents an efficient method for the direct identification and enrichment of retinal progenitors and that progenitors enriched from mitogen-exposed retinal culture appear to be similar in characteristics and differentiation potential to those that are prospectively identified and therefore may be equally suitable for therapeutic purposes. 

 Supported by National Eye Institute, The Foundation for Fighting Blindness, Nebraska Research Initiative and Research to Prevent Blindness.

 Submitted for publication September 3, 2002; revised October 25 and November 27, 2002; accepted January 2, 2003.

 Disclosure: S. Bhattacharya, None; J.D. Jackson, None; A.V. Das, None; W.B. Thoreson, None; C. Kuszynski, None; J. James, None; S. Joshi, None; I. Ahmad, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

List of Antibodies Used for Immunoflourescence Analyses

Antibodies	Cell Types	Dilution	Source	References
CD4	Mature lymphocytes	1:50	Pharmingen*	12
CD8a	Mature lymphocytes	1:50	Pharmingen	13
CD11b	Neutrophils	1:50	Pharmingen	14
RP-1	Granulocytes	1:50	Pharmingen	15
CD45	Hematopoietic cells	1:50	Pharmingen	16
CD90	Thy1 Antigen	1:50	Pharmingen	17
Nestin	Neural progenitors	1:4	DSHB, †	18
Notch1	Progenitors	1:1	Ref. 19	19
Ki67	Proliferating cells	1:500	Pharmingen	20
Chx10	Retinal progenitors	1:250	Ref. 21	21
Map2	Neurons	1:400	Chemicon, †	22
βtubulin	Neurons	1:2000	Covance, §	23
GFAP	Astrocytes	1:100	Sigma, ∥	24
Syntaxin	Amacrine cells	1:50	Ref. 28	28
Rhodopsin	Photoreceptors	1:5000	Ref. 28	28

Table 2.

 

List of Primers and Their Respective Sequences Used for RT-PCR Analyses

Genes	Primer Sequences	Product Size (bp)	Genbank Accession Number	Temp (°C)
β-Actin	Forward: 5′GTGGGGCGCCCCAGGCACCA 3′	543	XMO37235	56
	Reverse: 5′CTCCTTAATGTCACGCACGATTTC 3′
Nestin	Forward: 5′TGGAGCAGGAGAAGCAAGGTCTAC 3′	295	NM012987	56
	Reverse: 5′TCAAGGGTATTAGGCAAGGGGG 3′
Opsin	Forward: 5′CATGCAGTGTTCATGTGGGA 3′	382	U22180	64
	Reverse: 5′AGCAGAGGCTGGTGAGCATG 3′
Syntaxin1	Forward: 5′AAGAGCATCGAGCAGAGCATC 3′	342	NM016801	60
	Reverse: 5′CATGGCCATGTCCATGAACAT 3′
Map2	Forward: 5′CGGACACAAGTAAAACTCCCCCAG 3′	329	X54100, M38462	56
	Reverse: 5′GTCAAACTGTTTCCTTCCCATC 3′
GFAP	Forward: 5′ ATCTGGAGAGGAAGGTTGAGTCG 3′	310	NM017009	58
	Reverse: 5′ TGGCGGCGATAGTCATTAGC 3′

Figure 1.

 

A subpopulation offreshly dissociated E18 retinal cells excluded Hoechst dye. Freshly isolated rat bone marrow (A), E18 hippocampal (B), and retinal (C) cells were stained with Hoechst 33342 fluorescent dye, and emission patterns of respective cell types were analyzed in both blue and red wavelength by flow cytometry. Boxed region: SP cells that selectively excluded the Hoechst 33342 dye.

Figure 2.

 

The exclusion of Hoechst dye by E18 retinal cells was verapamil sensitive. Freshly isolated E18 retinal cells were stained with Hoechst 33342 dye in the presence or absence of verapamil. The Hoechst dye staining and emission patterns of retinal cells revealed that the efflux of Hoechst dye by SP cells took place in the absence (A) and in the presence (B) of verapamil. The effect of verapamil on the Hoechst dye efflux was concentration dependent (C).

Table 3.

 

Expression of Hematopoietic and Neural Lineage Markers

Phenotypes	Bone Marrow (%)	E18 Retina (%)	Neurospheres (%)
CD4	16.41	0.32	0.38
CD8a	6.75	0.74	0.11
CD11b	15.38	0.68	0.37
RP-1	32.13	0.59	0.48
CD45	67.09	0.65	0.01
CD90	58.46	10.52	3.76
Nestin	0.21	4.25	45.95

Figure 3.

 

SP cells isolated directly from E18 retina were proliferative and expressed progenitor markers. The SP cells obtained from freshly dissociated E18 retina were cytocentrifuged and analyzed immunocytochemically for the expression of the cell proliferation marker Ki67 (A, B), the neural progenitor marker nestin (C, D), the general progenitor marker Notch1 (E, F), and the retinal progenitor marker Chx10 (G, H). Left: bright-field images. Nuclear and cytoplasmic distributions of specific antigens could not be distinguished, because stem cells/progenitors generally have a high nuclear-to-cytoplasmic ratio and because cells were cytocentrifuged, resulting in the loss of cytoplasmic processes. Magnification, ×200.

Figure 4.

 

SP cells isolated directly from E18 retina were undifferentiated. Freshly isolated E18 cells enriched as SP and NSP cells (A) were subjected to RT-PCR analysis to detect transcripts corresponding to progenitor- (nestin), neuron- (Map2), glia- (GFAP), and retina- (opsin and syntaxin 1) specific markers. (B) Cells expressing markers for differentiated phenotypes were confined to NSP, suggesting that SP cells are largely undifferentiated. Unsorted retinal cells were used as the control.

Figure 5.

 

SP cells isolated from mitogen-exposed retinal neurospheres expressed progenitor markers and were undifferentiated. Cells from E18 retina were cultured in the presence of EGF to obtain neurospheres consisting of dividing cells that incorporated 5-bromo-2-deoxyuridine (BrdU; B) and expressed nestin (C). (A) represents a bright field image. SP and NSP cells, isolated from neurospheres by Hoechst dye efflux assay (D), were subjected to RT-PCR analysis to detect transcripts corresponding to progenitor- (nestin), neuron- (Map2), glia- (GFAP), and retina- (opsin and syntaxin 1) specific cell markers (E). Cells expressing markers for differentiated phenotypes were confined to NSPs, suggesting that SP cells are undifferentiated. Unsorted retinal cells were used as the control. Magnification, ×200.

Figure 6.

 

SP cells can generate secondary neurospheres consisting of cells that proliferate and express progenitor markers. SP and NSP cells were cultured in high density on poly-d-lysine–coated substratum. Secondary neurospheres were generated by SP cells and not by NSP cells (A, B). SP-cell–derived secondary neurospheres were subjected to immunocytochemical analysis to detect progenitor markers. Cells in the secondary neurospheres were positive for the proliferation marker Ki67 (C, D) and expressed nestin (E, F), Notch 1 (G, H), and Chx10 (I, J). Magnification, ×200.

Figure 7.

 

SP cells isolated directly from E18 retina were multipotent. SP cells, isolated directly from E18 retina, were plated on poly-d-lysine– and laminin-coated substratum and cultured in a differentiation condition (1% serum and the absence of EGF) for 7 days. In this condition, cells were observed expressing (A, B) neuronal (β-tubulin) or (C, D) glial (GFAP) markers. Arrows: positive cells. (E) Quantification of marker-expressing cells. Immunocytochemical analysis of neural differentiation was corroborated by RT-PCR analysis of transcripts corresponding to Map2 and GFAP performed on SP and unsorted cells in differentiation conditions (F). Magnification, ×200.

Figure 8.

 

SP cells isolated from mitogen-exposed retinal neurospheres were multipotent. SP cells, isolated from mitogen-exposed retinal neurospheres, were plated on poly-d-lysine– and laminin-coated substratum and cultured in differentiation conditions (1% serum and the absence of EGF) for 7 days. In this condition, cells were observed expressing (A, B) neuronal (β-tubulin) or (C, D) glial (GFAP) markers. Arrows: positive cells. (E) Quantification of marker-expressing cells. Immunocytochemical analysis of neural differentiation was corroborated by RT-PCR analysis of transcripts corresponding to Map2 and GFAP performed on SP cells and unsorted neurospheric cells in differentiation condition (F). Magnification, ×200.

Figure 9.

 

SP cells displayed electrophysiological properties of differentiated cells. SP cells obtained from E18 retina were cultured in the presence of serum, and current–voltage relationships were obtained from a subset of cells by whole-cell recording. (A, C) Currents evoked by a series of 150-ms voltage steps applied from a holding potential of −70 mV. In the current–voltage plot in (B), both the maximal inward (▪) and outward (□) currents are plotted. The data show that both the sustained outward rectifying currents and the presumptive sodium current activated above −50 mV. The current–voltage plot in (D) shows that the currents were inwardly rectifying below −90 mV and outwardly rectifying above −50 mV. The cell in (A) and (B) exhibited an electrophysiological phenotype similar to glial Müller cells. The cell in (C) and (D) exhibited a neuronal phenotype.

Figure 10.

 

SP cells express ionotropic glutamate receptors. SP cells obtained E18 retina were cultured in the presence of serum and effects of kainic acid, NMDA, and high K⁺ solutions on intracellular Ca²⁺ levels were measured with Fura-2 in differentiated cells. (A) Bright-field image (top left) and a series of pseudocolor images illustrating 340:380 ratio levels in the control (A1), 30 μM kainic acid (A2), 100 μM NMDA (A3), and 140 mM K⁺ (A4). (B) Graph of ratio changes obtained from the circular region of interest indicated in (A). The numbers 1 to 4 in (B) indicate the time points at which the corresponding images in (A) were obtained.

Figure 11.

 

SP cells, isolated directly from E18 retina or mitogen-exposed retinal neurospheres differentiated into retinal neurons. SP cells from either source were cocultured in the presence of PN1 retinal cells across a membrane for 5 to 7 days followed by immunocytochemical analyses to detect retinal cell-type–specific markers. SP cells, enriched from either source, expressed marker for rod photoreceptor (Rhodopsin; A, B, I, J), amacrine cells (Syntaxin; C, D, K, L), bipolar cells (PKC; E, F, M, N), and Müller glia (Vimentin; G, H, O, P). Arrows: retinal cells of the cell type of interest. (Q) Relative proportion of specific retinal cell types in SP cells after coculture. Magnification, ×200.

The authors thank Roderick McInnes and Colin J. Barnstable for the Chx10 and RetP1 antibodies, respectively, and Linda Wilkie, Susan Brusnahan, and Tami Wattnem for technical assistance.