B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

Ava i l ab l e on l i ne a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

An in vitro comparison of two different subpopulations ofretinal progenitor cells for self-renewal and multipotentiality

Jing Xiaa, 1, Hao Liub, 1, Xianqun Fana, Yamin Hua, Yidan Zhanga, Zhiliang Wanga,Xiaojian Zhoua, Min Luoa,⁎, Ping Gua,⁎aDepartment of Ophthalmology, Shanghai Ninth People's Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200011, ChinabH. Lee Moffitt Cancer Center and Research Institute, University of South Florida, Tampa, FL 33612, USA

A R T I C L E I N F O

⁎ Corresponding authors. Fax: +86 21 6313 560E-mail addresses: qiangson@sh163.net (MAbbreviations: RSCs/RPCs, retinal stem/pro

DAPI, 4′, 6-diamidino-2-phenylindole; PKC-aPBS, phosphate buffered saline; qPCR, quant1 These authors contributed equally to the

0006-8993/$ – see front matter © 2011 Elseviedoi:10.1016/j.brainres.2011.11.054

A B S T R A C T

Article history:Accepted 28 November 2011Available online 6 December 2011

Retinal progenitor cells (RPCs) show enormous potential for the treatment of retinaldegenerative diseases. It is well known that in vitro cultures of RPCs comprise suspensionspheres and adherent cells, but the differences between the two cell populations are notfully understood. In this study, cultured RPCs were sorted into suspension and adherentcells. Analyses of cell morphology, cell growth and retinal progenitor-related expressionmarkers were performed using quantitative polymerase chain reaction (qPCR) and immuno-cytochemistry to identify the proliferative and multipotent capacity of the cells in vitro. Thedata showed that both the suspension and adherent cells were maintained in an undifferen-tiated state, although the former exhibited a greater proliferative potential than the latter. Im-munocytochemistry analysis indicated that the two subsets of RPCs were able to differentiateinto different retinal cells in the presence of fetal bovine serum (FBS); the adherent cells weremore likely to differentiate toward the β3-tubulin-, AP2α- andMap2-positive neuronal lineage,while the suspension cells were more effective at differentiating into rod photoreceptors,which was consistent with the qPCR results. These findings suggest that adherent RPCs maybe a potential candidate for retinal interneuron or ganglion cell substitution therapies, where-as suspension RPCs may be preferred for photoreceptor cell replacement.

© 2011 Elsevier B.V. All rights reserved.

Keywords:RetinaProgenitor cellProliferationDifferentiation

1. Introduction

Many people lose their sight every year due to commonretina-related diseases, such as glaucoma, age-related macu-lar degeneration and diabetic retinopathy. These diseasesare all characterized by the loss of photoreceptors or other ret-inal neurons, leading to an irreversible decline in visual

6.. Luo), guping06@gmail.cgenitor cells; mRPCs, mo, protein kinase C-alpha;itative polymerase chainwork.

r B.V. All rights reserved

function. However, there are no effective treatments currentlyavailable to prevent the loss of retinal neurons. Stem cells arewidely regarded as cells with long-term self-renewal capabili-ties and an ability to generate special cell populations in agiven tissue. The isolation of different tissue-derived stemcells has attracted considerable attention for its potential asa treatment for various diseases. Fortunately, after further

om (P. Gu).use RPCs; FBS, fetal bovine serum; EGF, epidermal growth factor;AP2α, activator protein 2 alpha; GFAP, glial fibrillary acid protein;reaction

.

39B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

research on the properties of various types of stem cells,replacing lost neurons with stem cells may become a promis-ing way to treat people with retinal diseases. Recently, severalpopulations of stem cells or progenitor cells, including neuralstem cells, bone marrow-derived stem cells and embryonicstem cells, have been used for retinal transplantation(Jagatha et al., 2009; Sakaguchi et al., 2004; Tomita et al.,2006). However, previous studies have demonstrated that pro-genitor cells that are not derived from the eye have a very lim-ited ability to generate specific cell populations required forthe treatment of retinal diseases (Van Hoffelen et al., 2003).

Retinal stem/progenitor cells (RSCs/RPCs) have been identi-fied in the mammalian eye, and retinal stem/progenitor cellsof rodents, pigs and humans can be cultured in vitro (Coles etal., 2004; Gamm et al., 2005; Gu et al., 2007; Tropepe et al.,2000). In principle, RPCs could provide a source of new retina-specific cells (Livesey andCepko, 2001). Numerous retinal trans-plantation studies have demonstrated that mouse RPCs(mRPCs) may be the best cells for the restoration of vision bycell-replacement therapy. For example, previous studies haveshown that RPCs are able to attain functional integration intothe outer nuclear layer of the retina and express specificmarkers of photoreceptors, such as recoverin and rhodopsin(Lamba et al., 2009; MacLaren et al., 2006). Moreover, animalswith photoreceptor degeneration showed partial preservationof light sensitivity after the transplantation of RPCs (Klassen etal., 2004). Other experiments have also indicated that the differ-ent developmental stages and growth states of stem cells mayinfluence transplantation results (Cepko et al., 1996; Gamm etal., 2008; MacLaren et al., 2006; Reh, 2006; Wang et al., 2002;Young, 1985), thus, stem cells or progenitor cells with goodself-renewal and a specific differentiation potential are a keystep for transplantation therapy. Specifically, pre-selection ofRSCs/RPCs that show a highly preferential differentiation intospecific retinal neurons is a key factor for treating retinal disor-ders by cell transplantation.

During RPC expansion, an interesting phenomenon occursin which some cells grow as floating spheres and others adoptadherent growth; we describe these two types of cells as sus-pension cells and adherent cells, respectively. These twotypes of cells display different morphologies. Whether themorphological differences between these two subsets ofRPCs indicate different intrinsic characteristics of their prolif-erative capacity and differentiation potential remains unclear.In this study, we characterized the differences between sus-pension and adherent RPCs in vitro, which may be useful forthe treatment of retinal diseases.

Fig. 1 – Morphology and expansion potential of RPCs. The twocell subgroups were cultured and assessed on days 1 (A, B), 4(C, D) and 7 (E, F). The majority of the suspension cellsproliferated to form spheres, and the size of the spheresbecame larger over a 7-day period (A, C, E). Themajority of theadherent cells attached to the surface of the flask with two ormore short processes (B, D, F). The expansion potential ofthese two subgroupswas assessed through long-term culture;the suspension cells exhibited approximately 14.6% moreexpansion potential than the adherent cells (G). Scale bars:100 μm.

2. Results

2.1. Morphology and expansion potentials of suspensionand adherent RPCs

The treatment of two subsets of RPCs with the same cultureconditions resulted in different morphological changes. In theproliferation medium, the number of both suspension and ad-herent mRPCs increased with time. The suspension cells con-tinued to grow as floating spheres, which became larger insize, and only a few derived cells were found attached to the

flasks (Figs. 1-A, C and E). Under the proliferation condition,themajority of the adherent cells remained attached to the sur-face of the flask with two or more short processes, and only afew floating clusters were detected (Figs. 1-B, D and F).

The RPC proliferative capacity was examined using growthcurves. Both suspension and adherent RPCs were able to prolif-erate in vitro for up to one month, and suspension RPCs hadslightlymore expansion potential than adherent RPCs (Fig. 1-G).

2.2. Expression of progenitor and proliferation markers ofRPCs under proliferation conditions

To investigate if the self-renewal and expansion potential ofthe suspension RPCs were different from the adherent RPCs

40 B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

under the proliferation condition, the expression of retinalprogenitor markers (nestin and vimentin) and a cell prolifera-tion marker (Ki-67) was examined. The qPCR results showedthat the expression of nestin, vimentin and Ki-67 was not sig-nificantly different between the two subgroups (Fig. 2-A),which is in accordance with the immunofluorescence results.Most of the adherent and suspension RPCs were immunoreac-tive for nestin (87.6±7.3% and 90.3±7.5%, respectively) andvimentin (90±9.2% and 86±6.5%, respectively); there were nosignificant differences between these groups (Figs. 2-B, E, C,

Fig. 2 – Expression of progenitor and proliferation markers of RPvimentin and Ki-67 were not significantly different between theadherent cells than in suspension cells (A). RPCs were immunolanestin (B, E) and vimentin (C, F) and the proliferation marker Ki-6progenitor and proliferationmarkers; the expression of thesema(H). The percentage of positive cells were determined by dividingstained with DAPI. Five hundred to one thousand cells for each RData are expressed as the mean±SD of three independent exper

F and H). The percentage of Ki-67-positive cells was 60.7±6.8% in the adherent cells and 62.5±8.2% in the suspensionRPCs (Figs. 2-D, G and H). The differentiated cell markers, in-cluding β3-tubulin, Map2, protein kinase C alpha (PKC-α), rho-dopsin and glial fibrillary acidic protein (GFAP) were notdetected in either subpopulation of RPCs under proliferationconditions in vitro (data not shown). These results suggestedthat both suspension and adherent RPCs were in an undiffer-entiated state and possessed similar self-renewal and prolif-erative capacities.

Cs. qPCR results showed that the expressions of nestin,two subpopulations, while cadherin 4 was slightly higher inbeled with antibodies against the retinal progenitor markers7 (D, G) on day 4. The adherent cells continued to express

rkers was not significantly different from the suspension cellsthe number of immunopositive cells by the number of nucleiPC subgroup and each culture were counted in random fields.iments. Scale bars: 100 μm.

41B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

In addition, the gene expression of retinal adhesionmoleculecadherin 4 was tested. The qPCR results displayed that the ex-pression of cadherin 4 was slightly higher in adherent cellsthan in suspension cells, suggesting this adhesion moleculemay have an effect on their morphological differences (Fig. 2-A).

2.3. Differences inmorphology and gene expression betweenthe two subgroups of RPCs in the differentiation medium

Under differentiation conditions, suspension cells only occa-sionally extended short processes at day 1; subsequently, theyexhibited increasingly divergent morphologies. A higher densi-ty of short processes was observed at day 7 (Figs. 3-A, B and C).During the differentiation state, most of the adherent cells ex-tended short processes by the first day, and some of these pro-cesses became large with polygonal morphology. By day 7, themajority of the adherent cells exhibited two or more neurite-like long processes, which formed a network among the cellsand looked longer and thinner than those of the suspensioncells. Moreover, the adherent cells displayed smaller cell bodiesthan the suspension cells (Figs. 3-D, E and F).

To learn about the gene expression profile of the twomRPCsubsets, the expression of eight key genes involved in retinal

Fig. 3 – Morphology and gene expression of RPCs in the differentsuspension cells presented short and thick processes (A, B, C), wAdditionally, the cell bodies of the adherent cells appeared smallein the differentiation culture, the gene expression profiles of theexpression of β3-tubulin, Map2, AP2α, Brn3a, calbindin and PKCrhodopsin and GFAP was higher in suspension cells (G). The erroAbbreviations: GFAP, glial fibrillary acidic protein; PKC-α, protein

development was evaluated by qPCR analysis. The expressionof the rod photoreceptor marker rhodopsin and the glial cellmarker GFAP was increased in the suspension cells, whereasthe expression of the general neuronal makers β3-tubulin,Map2 and the retinal specific neuronal markers PKC-α (amarker for bipolar neurons), AP2α (a marker for amacrineneurons), calbindin (a marker for horizontal and ganglionneurons), Brn3a (a marker for ganglion neurons) were up-regulated in the adherent cells (Fig. 3-G). These results sug-gested that adherent mRPCs were more likely to differentiateinto ganglion cells and retinal interneurons including bipolar,amacrine and horizontal neurons, while suspension mRPCswere more effective at differentiating into rod photoreceptors.

2.4. Multipotentiality of the two subsets of cells asdetermined by immunocytochemistry

To explore if the two RPC subsets have different multipoten-tial capacities, the expression of several critical markers ofretinal cell development was further assessed through immu-nocytochemistry. RPCs were cultured for 7 days on coverslipscoated with laminin in the differentiation medium andstained with different cell-specific markers. The proportion

iation medium. At days 1 (A, D), 4 (B, E) and 7 (C, F), thehereas adherent cells displayed long and thin processes.r than those of the suspension cells (D, E, F). After seven dayssuspension and adherent cells were analyzed by qPCR. The-α was higher in the adherent cells, while the expression ofr bars show the standard deviation (n=3). Scale bars: 100 μm.kinase C alpha; AP2α, activator protein 2 alpha.

42 B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

of nestin- and Ki-67-positive cells was 12.4±4.1% and 15±3.4%in the suspension RPC cultures, respectively, whereas thisproportion was 10.5±2.5% and 13.6±4.1% in the adherentcells, respectively (Figs. 4-A, F, B and G). Furthermore, the per-centage of β3-tubulin-, Map2-, AP2α-, Brn3a-, calbindin- andPKC-α-immunoreactive cells was 29±6.5%, 24±4.8%, 8.2±1.7%, 3.5±1.3%, 8±1.9% and 26.7±3.3% in the suspensioncells, respectively, which was lower than in the adherentcells (41.2±3.0%, 31.7±3.9%, 11.7±2.7%, 5±2.2%, 10.5±2.9%and 28.7±4.2%, respectively) (Figs. 4-C, H, D, I, E, J, K, P, L, Q,M and R). Moreover, the proportion of rhodopsin- and GFAP-positive cells was 14.2±3.1% and 26.5±5.1% in the suspensioncells, respectively, which was more than in the adherent cells

Fig. 4 – Multipotentiality of the two subsets of cells. After sevenimmunostained with antibodies against nestin (A, F), Ki-67 (B, Gcalbindin (L, Q), PKC-α (M, R), rhodopsin (N, S) and GFAP (O, T). Thwere higher in the suspension cells than in the adherent cells, wcalbindin and PKC-α was lower in the suspension cells than in twas performed as described in Fig. 2-H. *P<0.05. Scale bars: 100

(9.3±2.3% and 24.5±3.5%, respectively) (Figs. 4-N, S, O and T).Collectively, our results suggest that both subgroups of RPCsmaintained their multi-differentiation potentials, whichwere capable of giving rise to the different cell types of the ret-ina; however, the two subgroups showed differing propensi-ties for differentiating into a specific cell type (Fig. 4-U).

3. Discussion

RPCs represent a subtype of tissue-specific multipotent cells,which are found throughout the neuraxis during developmentand into adulthood, and are often referred to as neural stem

days in the differentiation medium, the RPCs were fixed and), β3-tubulin (C, H), Map2 (D, I), AP2α (E, J), Brn3a (K, P),e levels of nestin-, Ki-67-, rhodopsin- and GFAP-positive cellshile the proportion of β3-tubulin, Map2, AP2α, Brn3a,he adherent cells (U). Quantification of immunoreactive cellsμm.

43B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

cells (Coles et al., 2004). These cells show great potential forthe treatment of retinal degenerative disease. Exploring thebiological characteristics of RPCs and obtaining RPCs with de-sirable features for transplantation are important for ideal ret-inal repair (Sun et al., 2009). In this study, we separated RPCsinto suspension and adherent cells and investigated theircharacteristics, which will be useful for further transplanta-tion research.

In cells cultured in the proliferation medium, the qPCRdata demonstrated that the adherent cells have a slightlylower expression of nestin and Ki-67 compared to the suspen-sion cells. This finding is consistent with the immunofluores-cence results, which showed expression levels of 90.3% fornestin and 62.5% for Ki-67 in the suspension RPCs as com-pared to the expression levels of 87.6% and 60.7% in the adher-ent RPCs, respectively. These data suggest that both theadherent and the suspension RPCs are maintained as undif-ferentiated progenitor cells and possess similar expansioncapacities.

Under differentiation conditions, the expression of rho-dopsin was significantly higher in the suspension cells thanin the adherent cells, indicating that the suspension cells pos-sessed a greater potential to differentiate into retinal rod pho-toreceptors and may be a better choice for photoreceptorreplacement therapy. However, the proportion of the photore-ceptors that differentiate from RPCs still needs to be improvedto reach an ideal transplantation efficiency (Inoue et al., 2010;Ohsawa and Kageyama, 2008). The low proportion of rod pho-toreceptors may be due to the following reasons. First, a previ-ous study showed that the retinal microenvironment and theintracellular cues of RPCs were crucial factors for the birth ofretina-specific cells (Zaghloul et al., 2005). It is possible thatthe addition of serum alone in the differentiated cultures inthis study may not be sufficient to induce their differentiationinto rhodopsin-positive photoreceptors. Secondly, Giordanoet al. (2007) reported that EGF had a negative effect on thenumber of cells expressing markers for rod photoreceptors.In the present study, EGF was added to the proliferation medi-um before differentiation, which may be one of the reasonsfor our findings. In addition, it is also possible that certainmarkers are only expressed during specific developmentalstages.

Our data show that adherent RPCs display a preference fordifferentiating into cells that express the markers β3-tubulin,Map2, AP2α, Brn3a, calbindin and PKC-α. These results arepartially in agreement with a study that demonstrated thatneurons generated in the regions around the progenitor cellsformed neurospheres (Lai et al., 2008). This may be partiallyexplained by the fact that repetitive separation of the adher-ent cells may partially select the RPCs that had already under-gone a transition into retinal interneuron or ganglion cellprogenitor states. The up-regulation of β3-tubulin, AP2α,Brn3a, calbindin andMap2 in the adherent RPCs is exciting be-cause it suggests that the adherent cells are a better candidatefor replacement therapy in the case of retinal interneuron organglion cell degeneration compared to the suspension cells.

In summary, this study revealed the proliferation and dif-ferentiation characteristics of suspension and adherent RPCsin vitro. We demonstrated that both cell subpopulations main-tain an undifferentiated progenitor state, although they

displayed diverse multipotentialities. The suspension cellsshowed a greater potential for photoreceptor differentiation,indicating their potential for the treatment of photoreceptordegenerative disorders, whereas the adherent cells exhibiteda greater capability for generating β3-tubulin-, AP2α-, Brn3a-,calbindin- and Map2-positive neurons, suggesting a potentialapplication in retinal interneuron or ganglion cell degenera-tion therapy. Further investigation of RPCs, in vitro and invivo, is necessary to achieve optimal replacement efficiencyprior to their clinical use as a treatment for these diseases.

4. Experimental procedures

4.1. Isolation and culture of mouse retinal progenitor cells

RPCs were derived from the neural retina of postnatal day oneGFP transgenic C57BL/6 mice. The GFP transgenic C57BL/6mice were produced with an enhanced GFP cDNA under thecontrol of a chicken beta-actin promoter and cytomegalovirusenhancer (gift fromMasaru Okabe, University of Osaka, Japan)(Gu et al., 2009). Briefly, retinal tissue was isolated anddigested with 0.1% type I collagenase (Invitrogen, Carlsbad,CA). The cell suspensions were passed through a 100-μm cellstrainer and centrifuged. Subsequently, a fire-polished pipettewas used to mechanically dissociate the cells into a single cellsuspension in fresh culture medium. The cell culture medi-um, termed “the proliferation medium”, consisted of ad-vanced DMEM/F12 (Invitrogen), 1% N2 neural supplement(Invitrogen), 2 mM L-glutamine (Invitrogen), 100 U/mlpenicillin-streptomycin (Invitrogen), and 20 ng/ml epidermalgrowth factor (recombinant human EGF, Invitrogen). Thecells were seeded into flasks at a density of 2×105 cells/mland were incubated at 37 °C with 5% CO2. The culturemediumwas changed every two days, and the cells were passaged atregular intervals of three to four days.

4.2. Separation of the suspension and adherent cells

After three passages, the RPCs were separated into two cellgroups by shaking the culture flasks slowly and gently. Thefloating cells were collected from the cell suspension in themedium. These cells were digested with trypsin (Invitrogen)and then mechanically dissociated into single cells using afire-polished pipette. The adherent cells were trypsinizedand suspended in a single cell suspension after the floatingcells were washed off. Both the floating cells and the adherentcells were cultured at a density of 2×105 cells/ml for 3–4 daysin the proliferation medium. The same procedures were ap-plied to these two subsets of cells five times to achieve puri-fied suspension cells and adherent cells for the subsequentstudies.

4.3. Proliferation and differentiation of retinal progenitor cellsin vitro

For the cell proliferation analysis, suspension and adherentcells dissociated into single cells were seeded in flasks at a den-sity of 2×105 cells/ml in the proliferationmedium for 7 days. Forthe cell differentiation analysis, the proliferation medium was

44 B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

replaced with the differentiation medium from which epider-mal growth factor (EGF) was removed and supplemented with5% fetal bovine serum (FBS) (Invitrogen). For both culture condi-tions, the medium was changed every two days. Images of thecells were captured with a fluorescence microscope (OlympusIX51, Japan), and total cellularmRNAwas extracted at predeter-mined time points on days 1, 4 and 7.

4.4. Growth curve of two subpopulation cells

The proliferative potentials of suspension and adherent cellswere assessed using a growth curve. These cells were digestedwith trypsin and mechanically dissociated into single cellsusing a fire-polished pipette, and then seeded in flasks at a densi-ty of 2×105 cells/ml. The culturemediumwas changed every twodays, and the cells were passaged at regular intervals of three tofour days for up to one month. The live cell numbers werecounted using a hemocytometer (Corning, Steuben County, NY).

4.5. Total RNA isolation and quality controls

Total RNA was isolated from the cultured cells at differenttime points. The RNA extraction was conducted using theRNeasy Mini Kit (Qiagen, Valencia, CA), according to the man-ufacturer's instructions. DNaseI was used to avoid genomicDNA contamination. The concentration and purity of thetotal RNA were determined spectrophotometrically atOD260 nm and OD280 nm. The samples with OD260/280 nmratios between 1.9 and 2.1 were used for cDNA synthesis.

4.6. Reverse transcription and quantitative polymerase chainreaction (qPCR)

One microgram of total RNA was reverse transcribed into cDNAin a final reaction volume of 10 μl using the PrimeScript™ RT re-agent kit (Perfect Real Time, TaKaRa, Dalian, China) (Chen et al.,2011). The resulting cDNAs and controls were diluted 20-fold innuclease-free water (Invitrogen) and used as templates forqPCR. Then, cDNA products were amplified using SYBR PremixEX Taq™ (TaKaRa); the total reaction mixture contained 10 μl of2× Power SYBR Green PCR Master Mix, ROX Reference Dye, 2 μl

Table 1 – Primers used for quantitative RT-PCR.

Genes Accession no. Forward(5′-3′)

Nestin NM_016701 aactggcacctcaagatgt tcaagVimentin NM_011701 tggttgacacccactcaaaa gctttKi-67 X82786 cagtactcggaatgcagcaa cagtcCadherin 4 NM_009867.2 atggttctgctgttcgttgtgt ggtcgMap2 NM_001039934 agaaaatggaagaaggaatgactg acatgβ3-tubulin NM_023279 cgagacctactgcatcgaca cattgPKC-α NM_011101 cccattccagaaggagatga ttcctAP2α NM_001122948.1 gccgtccacctagccaggga gattgBrn3a NM_011143.4 cgctctcgcacaacaacatga ttcttccalbindin NM_009788.4 tcaggatggcaacggataca aataaGFAP NM_010277 agaaaaccgcatcaccattc tcacaRhodopsin NM_145383 tcaccaccaccctctacaca tgatcβ-actin NM_007393 agccatgtacgtagccatcc ctctc

of cDNA, and 10 nM of gene-specific primers (Table 1). Subse-quently, qPCR was conducted using an Applied Biosystems 7500real-time PCR system (Applied Biosystems, Irvine, CA). Cyclingconditions for qPCR were as follows: initial denaturation at95 °C for 10min, followed by 40 cycles of 15 s at 95 °C and 1minat 60 °C. The β-actin genewas used as anendogenous control. Se-ries dilutions of cDNA at a ratio of 1:5 were used to establish astandard curve to determine primer efficiency. The quantitativePCR experiments were repeated three times, and the averagedCt values were obtained from three independent samples. Real-time PCR data were analyzed using the Pfaffl (2001) method,and the relative expression levels of the genes in adherent cellswere expressed as fold changes relative to those in the suspen-sion cells after normalizing to β-actin; the expression levels ofeach gene in the suspension cells were defined as 1.

4.7. Immunocytochemistry

Suspension and adherent RPCs were seeded onto glass cover-slips (VWR, West Chester, PA) coated with laminin (Sigma-Al-drich, Saint Louis, MO) in 12-well plates and cultured for sevendays. The cells were fed every two days and fixed in 4% (w/v)paraformaldehyde (Sigma-Aldrich) in 0.1 M phosphate-buffered saline (PBS; 2.68 mM KCI, 1.47 mM KH2PO4,135.60 mM NaCl and 8.10 mM Na2HPO4) for 15 min at roomtemperature. The cells were then washed with PBS and incu-bated in blocking buffer (PBS containing 10% (v/v) normalgoat serum (Invitrogen)), 0.3% TritonX-100 (Sigma-Aldrich)and 0.1% NaN3 (Sigma-Aldrich)) for 1 h. The coverslips werethen incubated in primary antibodies (Table 2) at 4 °C over-night. After washing with PBS, the coverslips were incubatedin fluorescent-conjugated secondary antibodies (AlexaFluor546 goat anti-mouse or goat anti-rabbit, 1:800 in PBS, BD)for 1 h at room temperature. After washing in PBS, the cell nu-clei were counterstained with Vectashield Mounting Mediumwith 4′,6-diamidino-2-phenylindole (DAPI) (Vector Laborato-ries, Burlingame, CA) for 5 min at room temperature. The con-trol samples were processed using the same protocol but withthe omission of the primary antibody. Immunoreactive cellswere visualized and imaged using a fluorescent microscope(Olympus BX51, Japan).

Reverse(5′-3′)

Annealing temperature(°C)

Product size(base pairs)

ggtattaggcaagggg 60 235tggggtgtcagttgt 60 269ttcaggggctctgtc 60 170tagtcttggtcctcctc 60 148gatcatctggtaccttttt 60 112agctgaccagggaat 60 152gtcagcaagcatcac 60 212ggccgcgagttcccc 60 208tcgccgccgttga 60 121gagcaaggtctgttcgg 60 169tcaccacgtccttgt 60 184caggtgaagaccaca 60 216agctgtggtggtgaa 60 152

Table 2 – Primary antibodies used for immunocytochemistry.

Antibodies Type Specificity in retina Source Dilution

Nestin Mouse monoclonal Progenitors reactive glia BD 1:200Vimentin Rabbit monoclonal, Progenitors reactive glia Epitomics 1:200Ki-67 Mouse monoclonal Proliferating cells BD 1:200Map2 Rabbit monoclonal, Neurons Epitomics 1:200β3-tubulin Mouse monoclonal Neurons Chemicon 1:100AP2α Mouse monoclonal Amacrine cells DSHB 1:600Brn3a Rabbit polyclonal Ganglion cells Millipore 1:500calbindin Mouse monoclonal Horizontal, ganglion cells Sigma 1:200PKC-α Mouse monoclonal Bipolar neurons BD 1:200GFAP Mouse monoclonal Glia Chemicon 1:200Rhodopsin Mouse monoclonal Photoreceptors (rods) Chemicon 1:100

45B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

4.8. Statistical analysis

The experimental statistics presented in this study areexpressed as themean±the standard derivation (SD). Each ex-periment was repeated at least three times unless otherwisespecified. Statistical analyses were performed using an un-paired Student's t-test, and the difference was considered sig-nificant when the P value≤0.05.

Acknowledgments

The authors are grateful to Dr. Henry Klassen and Dr. Michael J.Young for the original provenance of themouse retinal progen-itor cells. This research was supported by the Shanghai Munic-ipality Commission for Science and Technology (09PJ1407200),the Education Commission of Shanghai (11YZ47), the NationalNatural Science Foundations of China (81070737, 81070757),the Shanghai Leading Academic Discipline Project (S30205)and the Shanghai Public Health Bureau (2009075).

R E F E R E N C E S

Cepko, C.L., Austin, C.P., Yang, X., Alexiades, M., Ezzeddine, D.,1996. Cell fate determination in the vertebrate retina. Proc.Natl. Acad. Sci. U. S. A. 93, 589–595.

Chen, H., Fan, X., Xia, J., Chen, P., Zhou, X., Huang, J., Yu, J., Gu, P.,2011. Electrospun chitosan-graft-poly (varepsilon-caprolactone)/poly (varepsilon-caprolactone) nanofibrous scaffolds for retinaltissue engineering. Int. J. Nanomedicine 6, 453–461.

Coles, B.L., Angenieux, B., Inoue, T., Del Rio-Tsonis, K., Spence, J.R.,McInnes, R.R., Arsenijevic, Y., van der Kooy, D., 2004. Facileisolation and the characterization of human retinal stem cells.Proc. Natl. Acad. Sci. U. S. A. 101, 15772–15777.

Gamm, D.M., Nelson, A.D., Svendsen, C.N., 2005. Humanretinal progenitor cells grown as neurospheres demonstratetime-dependent changes in neuronal and glial cell fatepotential. Ann. N. Y. Acad. Sci. 1049, 107–117.

Gamm, D.M., Wright, L.S., Capowski, E.E., Shearer, R.L., Meyer, J.S.,Kim, H.J., Schneider, B.L., Melvan, J.N., Svendsen, C.N., 2008.Regulation of prenatal human retinal neurosphere growth andcell fate potential by retinal pigment epithelium and Mash1.Stem Cells 26, 3182–3193.

Giordano, F., De Marzo, A., Vetrini, F., Marigo, V., 2007. Fibroblastgrowth factor and epidermal growth factor differently affectdifferentiation of murine retinal stem cells in vitro. Mol. Vis.13, 1842–1850.

Gu, P., Harwood, L.J., Zhang, X., Wylie, M., Curry, W.J., Cogliati, T.,2007. Isolation of retinal progenitor and stem cells from theporcine eye. Mol. Vis. 13, 1045–1057.

Gu, P., Yang, J., Wang, J., Young, M.J., Klassen, H., 2009. Sequentialchanges in the gene expression profile of murine retinalprogenitor cells during the induction of differentiation. Mol.Vis. 15, 2111–2122.

Inoue, T., Coles, B.L., Dorval, K., Bremner, R., Bessho, Y.,Kageyama, R., Hino, S., Matsuoka, M., Craft, C.M., McInnes,R.R., Tremblay, F., Prusky, G.T., van der Kooy, D., 2010.Maximizing functional photoreceptor differentiation fromadult human retinal stem cells. Stem Cells 28, 489–500.

Jagatha, B., Divya, M.S., Sanalkumar, R., Indulekha, C.L.,Vidyanand, S., Divya, T.S., Das, A.V., James, J., 2009. In vitrodifferentiation of retinal ganglion-like cells from embryonicstem cell derived neural progenitors. Biochem. Biophys. Res.Commun. 380, 230–235.

Klassen, H.J., Ng, T.F., Kurimoto, Y., Kirov, I., Shatos, M., Coffey, P.,Young, M.J., 2004. Multipotent retinal progenitors expressdevelopmental markers, differentiate into retinal neurons,and preserve light-mediated behavior. Invest. Ophthalmol.Vis. Sci. 45, 4167–4173.

Lai, P., Tang, S., Yi, J., Zhu, X., Zou, Y., Xie, H., Gao, Y., Lin, S., 2008.Long-Term in vitro expansion of retinal progenitor cells byculturing intact neurospheres in monolayer. Ophthalmic Res.40, 291–297.

Lamba, D.A., Gust, J., Reh, T.A., 2009. Transplantation of humanembryonic stem cell-derived photoreceptors restores somevisual function in Crx-deficient mice. Cell Stem Cell 4, 73–79.

Livesey, F.J., Cepko, C.L., 2001. Vertebrate neural cell-fatedetermination: lessons from the retina. Nat. Rev. Neurosci. 2,109–118.

MacLaren, R.E., Pearson, R.A., MacNeil, A., Douglas, R.H., Salt, T.E.,Akimoto, M., Swaroop, A., Sowden, J.C., Ali, R.R., 2006. Retinalrepair by transplantation of photoreceptor precursors. Nature444, 203–207.

Ohsawa, R., Kageyama, R., 2008. Regulation of retinal cell fatespecification by multiple transcription factors. Brain Res. 1192,90–98.

Pfaffl, M.W., 2001. A new mathematical model for relativequantification in real-time RT-PCR. Nucleic Acids Res. 29, e45.

Reh, T.A., 2006. Neurobiology: right timing for retina repair.Nature 444, 156–157.

Sakaguchi, D.S., Van Hoffelen, S.J., Theusch, E., Parker, E., Orasky,J., Harper, M.M., Benediktsson, A., Young, M.J., 2004.Transplantation of neural progenitor cells into the developingretina of the Brazilian opossum: an in vivo system for studyingstem/progenitor cell plasticity. Dev. Neurosci. 26, 336–345.

Sun, X., Jiang, R., Zhang, Y., Chen, M., Xiang, P., Qi, Y., Gao, Q.,Huang, B., Ge, J., 2009. Gene expression and differentiationcharacteristics in mice E13.5 and E17.5 neural retinalprogenitors. Mol. Vis. 15, 2503–2514.

46 B R A I N R E S E A R C H 1 4 3 3 ( 2 0 1 2 ) 3 8 – 4 6

Tomita, M., Mori, T., Maruyama, K., Zahir, T., Ward, M.,Umezawa, A., Young, M.J., 2006. A comparison of neuraldifferentiation and retinal transplantation with bonemarrow-derived cells and retinal progenitor cells. Stem Cells24, 2270–2278.

Tropepe, V., Coles, B.L., Chiasson, B.J., Horsford, D.J., Elia, A.J.,McInnes, R.R., van der Kooy, D., 2000. Retinal stem cells in theadult mammalian eye. Science 287, 2032–2036.

Van Hoffelen, S.J., Young, M.J., Shatos, M.A., Sakaguchi, D.S., 2003.Incorporation of murine brain progenitor cells into the

developing mammalian retina. Invest. Ophthalmol. Vis. Sci.44, 426–434.

Wang, S.W., Mu, X., Bowers, W.J., Klein, W.H., 2002. Retinalganglion cell differentiation in cultured mouse retinalexplants. Methods 28, 448–456.

Young, R.W., 1985. Cell differentiation in the retina of the mouse.Anat. Rec. 212, 199–205.

Zaghloul, N.A., Yan, B., Moody, S.A., 2005. Step-wise specificationof retinal stem cells during normal embryogenesis. Biol. Cell97, 321–337.