Martins et al. 2006

Glutamate regulates retinal progenitors cells proliferationduring development

Rodrigo A. P. Martins,1,2 Rafael Linden2 and Michael A. Dyer1,3

1Department of Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA2Laboratorio de Neurogenese, Instituto de Biofısica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, Rio de Janeiro,RJ, Brazil3Department of Ophthalmology, University of Tennessee Health Science Center, Memphis, TN 38163, USA

Keywords: AMPA ⁄ kainate, cyclin E ⁄ Cdk2, neurotransmitters, NMDA, retrovirus

Abstract

The precise coordination of cell cycle exit and cell fate specification is essential for generating the correct proportion of retinal celltypes during development. The decision to exit the cell cycle is regulated by intrinsic and extrinsic cues. There is growing evidencethat neurotransmitters can regulate cell proliferation and cell fate specification during the early stages of CNS development prior tothe formation of synaptic connections. We found that the excitatory neurotransmitter glutamate regulates retinal progenitor cellproliferation during embryonic development of the mouse. AMPA ⁄ kainate and N-methyl-d-aspartate receptors are expressed inembryonic retinal progenitor cells. Addition of exogenous glutamate leads to a dose-dependent decrease in cell proliferation withoutinducing cell death or activating the p53 pathway. Activation of AMPA ⁄ kainate receptors induced retinal progenitor cells toprematurely exit the cell cycle. Using a replication-incompetent retrovirus to follow the clonal expansion of individual retinal progenitorcells, it was observed that blockade of AMPA ⁄ kainate receptors increased the proportion of large clones, showing that modulation ofendogenous glutamatergic activity can have long-term consequences on retinal cell proliferation. Real time reverse transcriptase-polymerase chain reaction and immunoblot analyses demonstrated that glutamate does not alter the levels of the mRNA and proteinsthat regulate the G1 ⁄ S-phase transition. Instead, the activity of the Cdk2 kinase is reduced in the presence of glutamate. These dataindicate that glutamate regulates retinal progenitor cell proliferation by post-translational modulation of cyclin ⁄ Cdk2 kinase activity.

Introduction

The mammalian retina is composed of six neuronal types (rod, cone,horizontal, bipolar, amacrine and ganglion cells) and one glial cell type.During development, multipotent progenitor cells generate thosedifferent cell types in a conserved birth order (Young, 1985).The proliferation and the cell cycle exit of these progenitor cells mustbe precisely regulated to generate a functional retina that contains theappropriate number and proportion of each cell type (Dyer & Cepko,2001b). Deregulated proliferation of progenitor cells during retinaldevelopment can lead to blindness as a result of degeneration (Ma et al.,1998), dysplasia (Nakayama et al., 1996; Dyer &Cepko, 2000a; Levineet al., 2000) or retinoblastoma (Zhang et al., 2004a).

Both cell-intrinsic and -extrinsic factors regulate progenitor cellproliferation during CNS development (Dyer & Cepko, 2001a).Prox1, Rb, Six3 and Chx10 are examples of intrinsic transcriptionalregulators that contribute to proliferation regulation during retinaldevelopment (Zhu et al., 2002; Dyer, 2003; Marquardt, 2003; Zhanget al., 2004b). Similarly, extracellular molecules such as growthfactors and neurotransmitters have been implicated in the extrinsicregulation of cell proliferation in the developing CNS, including theretina (Carey et al., 2002; Pearson et al., 2002, 2005; Nguyen et al.,

2003). Interestingly, a role of neurotransmitters in cell fate regulationwas also proposed from studies in developing mouse retina (Young &Cepko, 2004).The excitatory neurotransmitter glutamate regulates cell prolifera-

tion in different regions of the CNS. In the developing cerebral cortex,glutamate has an anti-proliferative effect upon neuronal progenitorsthrough the AMPA ⁄ kainate subtype of ionotropic receptor (LoTurcoet al., 1995; Haydar et al., 2000). Interestingly, it was reported thatN-methyl-d-aspartate (NMDA) receptor activation has the oppositeeffect upon neuronal progenitor cells of the striatum (Luk et al., 2003).This finding suggests that signaling through different classes ofglutamatergic receptors may have distinct effects upon the cell cyclemachinery that controls neuronal progenitors cells proliferation.Advances have been made in the understanding of the intracellular

machinery that regulates cell cycle progression during retinal devel-opment (Dyer & Cepko, 2001a). It was shown that the cyclin-dependent kinases inhibitors p19 (Cdkn2d), p27 (Cdkn1b) and p57(Cdkn1c) control cell cycle exit of retinal progenitor cells (Levineet al., 2000; Dyer & Cepko, 2001b; Cunningham et al., 2002). Theimportance of cyclin D1 (Ccnd1) as a positive regulator of cell cycleprogression in the retina has been demonstrated in related studies(Fantl et al., 1995; Sicinski et al., 1995; Ma et al., 1998). Some studiesindicated mechanisms of how an extrinsic cue communicates with thecell cycle machinery during the development of the retina (Close et al.,2005). However, it is still unclear whether neurotransmitters regulate

Correspondence: Dr M.A. Dyer, 1Department of Development Neurobiology, as above.E-mail: [email protected]

Received 7 February 2006, revised 9 May 2006, accepted 14 May 2006

European Journal of Neuroscience, Vol. 24, pp. 969–980, 2006 doi:10.1111/j.1460-9568.2006.04966.x

ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd

retinal progenitor cells proliferation and, if so, how neurotransmitter-activated pathways connect to the cell cycle machinery that controlsretinal progenitor cells proliferation.In the present study, we investigated whether glutamate regulates

retinal progenitor cell proliferation during mouse development. Wefound that glutamate is an anti-proliferative signal within develop-ing retinal tissue that induces cell cycle exit of retinal progenitorcells.

Materials and methods

Materials

TRIzol� reagent and SuperScript� II cDNA synthesis kit were fromInvitrogen (Carlsbad, CA, USA). AMPA, kainate, NBQX, NMDA andMK-801 were from Tocris Bioscience (Ellisville, MO, USA). Normalserum, biotin-conjugated secondary antibodies and Vectastain ABC kitwere from Vector Laboratories (Burlingame, CA, USA), anti-GluR1,anti-GluR5 ⁄ 6 ⁄ 7 and anti-NR1 antibodies were from Chemicon(Temecula, CA, USA). Antibodies against cyclin D3, cyclin E, p19,p27, p57, Cdk2, Cdk4 were from Santa Cruz Biotechnologies (SantaCruz, CA, USA). Anti- bromodeoxyuridine (BrdU) antibody, [3H]-thymidine, ECL system, protein A-agarose and [c-32P]-ATP werefrom Amersham (Piscataway, NJ, USA).

Animals and retinal explant cultures

C57 BL ⁄ 6 mice were purchased from Charles River Laboratories(Wilmington, MA, USA). The procedures for maintaining explants ofmouse retinae in culture have been described previously (Dyer &Cepko, 2000a, b). Animals were anesthetized with carbon dioxide andkilled by cervical dislocation. Tissue dissociation was also performedas described previously (Morrow et al., 1998). Detailed protocols areavailable at: http://www.stjude.org/faculty/0,2512,407-2030-10417,00.html

RNA extraction, cDNA synthesis and real-time reversetranscriptase-polymerase chain reaction (RT-PCR) analysis

Retinae were removed from staged embryonic mice at embryonic day(E)14.5 and E17.5, from newborn mice at postnatal day (P)0, P3, P6and P12, and from adult mice. RNA was extracted using TRIzol�

reagent according to the manufacturer’s instructions and reversetranscribed using the SuperScript� II cDNA synthesis kit. Real-timeRT-PCR primers and probes for GluR1 (Gria1), KA2 (Grik5), NR1(Grin1), GLAST-1 ⁄ EAAT1 (Slc1a3), GLT-1 ⁄ EAAT2 (Slc1a2), EAA-C1 ⁄ EAAT3 (Slc1a1), EAAT4 (Slc1a6), cyclin D1 (Ccnd1), cyclin D3(Ccnd3), cyclin E1 (Ccne1), p107 (Rbl1), Cdk2, p19 (Cdkn2d), p27(Cdkn1b), p53 (Trp53), p21 (Cdkn1a), p19 ⁄ Arf (Cdkn2a), Bax,MDM2, cdc25, E2f1 and GAPDH (Gapd) were designed using PrimerExpress software (ABI) (Reporter-FAM, Quencher-BHQ, non-fluor-escent). The sequences of these primers are shown in Supplementarymaterial, (Table S1). For clarity, we have included both the traditionalnames and official gene names of the different glutamate receptors andtransporters in the Supplementary material (Table S2). The cDNAamplifications were performed in 20-lL reactions containing 200 nm

of primers, 200 nm of probe and TaqMan Universal PCR Mastermix(Applied Biosystems) on an ABI 7900 HT Sequence DetectionSystem (Applied Biosystems). The following PCR parameters wereused: incubation at 50 �C for 2 min, followed by 95 �C for 10 min,and 40 cycles of 95 �C for 15 s and 60 �C for 1 min.

Immunohistochemistry and antibodies

Dissociated cells were fixed in paraformaldehyde [4% in phosphate-buffered saline (PBS)], washed and treated with hydrogen peroxide(1% in PBS) before incubation in blocking solution (PBS containing0.025% Triton X-100 and 2% normal serum). Normal donkey serumwas used for the anti-GluR5 ⁄ 6 ⁄ 7 (mouse monoclonal, 1 : 5000).Normal goat serum was used for the anti-GluR1 (rabbit polyclonal,1 : 100) and anti-NR1 (rabbit polyclonal, 1 : 100) antibodies. Biotin-conjugated secondary antibodies (donkey anti-mouse IgG, goat anti-rabbit IgG) were used at a dilution of 1 : 500 in blocking solution.After secondary antibody binding, the dissociated cells were incubatedwith an avidin–biotin–peroxidase complex (Vectastain ABC) and thendetected with Cy-3 tyramide (Perkin Elmer, Wellesley, MA, USA)according to the manufacturers’ instructions. For nuclear staining,dissociated cells were incubated with either SYTOX green or DAPI(Molecular Probes, 1 : 20 000). Labeled cells were visualized using aZeiss Axioplan 2 microscope, and images were captured with anAxiocam digital camera (Zeiss, Thornwood, NY, USA).

BrdU and [3H]-thymidine labeling

To label S-phase retinal progenitor cells, we incubated cultured retinalexplants or freshly dissected retinae in explant culture mediumcontaining 10 lm BrdU (Boehringer Mannheim, Germany) or [3H]-thymidine (5 lCi ⁄ mL; 89 Ci ⁄ mmol) for the indicated times at 37 �C.Autoradiography and BrdU detection were carried out as describedpreviously (Dyer & Cepko, 2001b). Detailed protocols are availableat: http://www.stjude.org/faculty/0,2512,407-2030-10417,00.html

Apoptosis analysis

Dissociated cells or retinal sections (14 lm thick) obtained on a Leicacryostat (Leica, Bannockburn, IL, USA) were stained using theTUNEL apoptosis system (Promega, Madison, WI, USA) according tothe manufacturer’s instructions; however, for detection, we used Cy3-tyramide rather than the colorimetric substrate. Stained retinal sectionswere imaged using a Leica TLSNT confocal microscope, and thepercentage of labeled nuclei was determined from micrographs.Analysis of cell viability and annexin staining was done by FACSusing the Viacount kit (Guava Technologies, Hayward, CA, USA)according to the manufacturer’s instructions.

Retroviral labeling and lineage analysis

Retinal progenitor cells were infected with the replication-incompetentretrovirus NIN-E and their lineage was analysed. These methods havebeen extensively described elsewhere (Dyer & Cepto, 2001b).

Electroporation and FACS

Purified DNA was resuspended in HBSS at 1 lg ⁄ lL. Electroporationconsisted of five pulses of 25 V for 50 ms each with 950 ms recoveryperiods. For FACS purification of electroporated cells, retinae weredissociated as described previously (Dyer & Cepko, 2001b),resuspended in culture medium, and sorted using vYFP fluorescenceon a Becton-Dickson FACS system (Rockville, MD, USA).

Immunoblots and Cdk2-associated kinase activity assay

For immunoblots, total retinae protein was extracted with RIPAbuffer containing protease and phosphatase inhibitors (Sigma,

970 R. A. P. Martins et al.

ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 24, 969–980

St Louis, MO, USA). The concentration of lysates was determined byBradford assay. Lysate (30 lg) was electrophoresed in polyacryla-mide–sodium dodecyl sulfate (SDS) gels and then transferred tonitrocellulose membranes. Primary antibodies [cyclin D1 (#2926),pCdk2 (#2561) (Cell Signaling Technology, Beverly, MA, USA),cyclin D3 (#sc182), cyclin E (#sc481), p19 (#sc1063), p27 (#2552),p57 (#sc8298), Cdk2 (#sc163), Cdk4 (#sc260)] and horseradishperoxidase-conjugated secondary antibodies were used at a dilutionof 1 : 1000. The Amersham ECL system was used according to themanufacturer’s instructions. Cdk2-associated kinase activity wasdetermined as reported previously (Matsushime et al., 1994), withthe following modifications. Retinal explants were homogenized inlysis buffer [in mm: HEPES, 50, pH 7.4; NaCl, 150; EGTA, 2.5;EDTA, 1; 0.1% Tween20, Na3VO4, 0.1; NaF, 1; dithiothreitol (DTT),1; and a protease inhibitor cocktail]. Cell lysates (500 lg) wereimmunoprecipitated with a rabbit polyclonal antibody against Cdk2(#sc163) and protein A-agarose for 3 h at 4 �C. The beads werewashed four times with the lysis buffer and twice with the kinaseassay buffer (in mm: HEPES, 50, pH 7.4; MnCl2, 50; DTT, 1).Kinase reaction mixtures contained 5 lg histone H1 (UpstateBiotechnology, Lake Placid, NY, USA), 20 lm ATP and 10 lCi[c-32P]-ATP. SDS-loading buffer was added to stop the reaction,samples were heated at 95 �C for 3 min, and proteins were resolvedon SDS–polyacrylamide gels. Autoradiography was performed tovisualize the levels of phosphorylated histone H1.

Results

Glutamate receptors are expressed in retinal progenitor cells

To test whether the glutamate receptors are expressed in developingretina, we performed real-time PCR with TaqMan probes specific forrepresentative subunits of AMPA, kainate and NMDA glutamatereceptors on cDNA prepared from seven stages of mouse retinaldevelopment. Real-time PCR analysis of GluR1 (Gria1), KA2 (Grik5)and NR1 (Grin1) showed that AMPA, kainate and NMDA glutamatereceptors are expressed in the developing mouse retina (Fig. 1A). Allthree genes studied (GluR1, KA2 and NR1) were detected at theearliest embryonic stage analysed (E14.5). Of the subunits studied, theNMDA receptor subunit NR1 was expressed at the highest levels witha peak of expression at P0. The AMPA receptor subunit GluR1 alsopresented significant levels of expression in both embryonic andpostnatal stages. To obtain further evidence for the role of glutamatesignaling in the embryonic retina, we studied the expression ofglutamate transporters by real-time PCR analysis using probes forEAAT1–4 (Slc1a3, Slc1a2, Slc1a1 and Slc1a6). Our results confirmedprevious reports (Pow & Barnett, 1999, 2000; Reye et al., 2002) thatseveral glutamate transporters are also expressed in the embryonic andpostnatal mouse retina (Fig. 1B).

Previous studies reported the expression of glutamate receptors inthe embryonic retinal tissue (Watanabe et al., 1994; Zhang et al.,1996). The presence of immunopositive cells in the ventricular zone ofthe retina suggested that glutamate receptors may be expressed byretinal progenitor cells. To directly test if glutamate receptor proteinsare expressed in proliferating embryonic retinal progenitor cells(E14.5), we immunostained [3H]-thymidine-labeled embryonic retinalcells using antibodies that specifically recognize GluR1, GluR5 ⁄ 6 ⁄ 7or NR1, and then overlaid the cells with autoradiographic emulsion(Fig. 1C). We scored the proportion of GluR1-, GluR5 ⁄ 6 ⁄ 7- or NR1-immunopositive cells at E14.5 (Fig. 1D) as well as the proportion ofimmunopositive cells in S-phase at the time of labeling ([3H]-thymidine-labeled) (Fig. 1E). These data confirmed that glutamate

receptors of the AMPA, kainate and NMDA subtype are expressed inproliferating retinal progenitor cells.

Glutamate is an endogenous anti-proliferative signal inthe developing retina

To test whether glutamate regulates retinal progenitor cell proliferationin the developing retina, we cultured E14.5 retinae in the presence ofdifferent concentrations of glutamate for 16–20 h. At the end of theculture period, we pulse-labeled the retinal progenitor cells in S-phasewith BrdU for 2 h. Retinae were then dissociated, immunostained forBrdU (Fig. 2A), and the proportion of BrdU+ cells was scored(Fig. 2B). The proportion of S-phase cells was significantly reducedwhen retinal explants were treated with 50, 100 or 500 lm glutamate(Fig. 2B). Similar results were obtained with E17.5 and P0 retinae(data not shown). Previous studies in the cortex demonstrated that 50–300 lm glutamate depolarized cells, induced calcium influx anddecreased the proliferation of cortical progenitor cells (LoTurco et al.,1995).Excessive activation of glutamatergic receptors may induce excito-

toxic cell death (Olney et al., 1971). To test whether the observedreduction in BrdU-labeled retinal cells after glutamate treatment wascaused by increased cell death, we measured the proportion of viablecells by flow cytometry (Fig. 2C). There was no difference in theproportion of viable cells in either untreated samples or in samplestreated with 200 lm or 500 lm glutamate (Fig. 2D). Analysis ofapoptosis using an assay of annexin binding and flow cytometryprovided additional evidence for the absence of apoptotic cell death.The proportion of apoptotic cells was 6.3% in the untreated sampleand 6.8% in the glutamate-treated samples (data not shown). Next, wescored the proportion of apoptotic cells using a TUNEL assay inretinal sections. Treatment with 500 lm glutamate for differentperiods (3, 6, 9, 12, 16 and 24 h) did not change the proportion ofTUNEL+ (apoptotic) cells at any time point tested (Fig. 2E).It has been shown that glutamate receptor activation can induce the

p53 pathway in neural tissue that may either result in excitotoxic celldeath (Uberti et al., 1998, 2000) or cell cycle arrest (Ghiani et al.,1999). To test whether the p53 pathway is activated in developingretina following glutamate receptor activation, we treated E14.5retinae with glutamate and analysed the expression of p53 targetgenes. Using real-time RT-PCR, we found that there was no inductionof the p53 target genes Mdm2, p21 (Cdkn1a) or Bax after treatmentwith glutamate for 1, 4 or 16 h (Fig. 2F, data not shown). Takentogether, these data demonstrate that the reduction in BrdU+ retinalprogenitor cells following glutamate exposure is not due induction ofcell death, and suggest that glutamate receptor activation does notinduce the p53 ⁄ p21 pathway of cell cycle arrest.To identify which receptor subtype mediates glutamate-induced

reduction in the proportion of retinal progenitor cells in S-phase, wetreated retinal explants with glutamatergic pharmacological agonists orantagonists. Embryonic retinal progenitor cells treated with AMPA orkainate (100 lm) exhibited a significant reduction in the proportion ofBrdU+ cells that was similar to the reduction seen after glutamatetreatment (Fig. 2G). In contrast, treatment with NMDA (100 lm) didnot alter the proportion of BrdU+ cells (Fig. 2G). In addition, co-incubation of retinal explants with glutamate and the AMPA ⁄ kainateantagonist, NBQX, prevented the glutamate-induced decrease inBrdU+ cells (Fig. 2H). Co-incubation with glutamate and the NMDAantagonist, MK-801, had no effect. These results indicate that theeffect of exogenous glutamate is mediated by AMPA ⁄ kainatereceptors.

Glutamatergic regulation of retinal histogenesis 971


Fig.2.

Glutamatergicsignalingmodulates

retinalproliferation.

(A)Representativepictures

ofbrom

odeoxyuri-

dine

(BrdU)im

munostainingin

dissociatedcellsfrom

cultured

E14.5

retina.(B)E14.5

retinalexplants

were

cultured

for20

h,then

BrdU

(10

lm)was

addedfor2h.

Treatmentwithglutam

ate(50–500

lm)decreasedthe

proportion

ofBrdU+

cells.

(C)FA

CS

quantification

oftheproportion

ofviable

cells.

(D)Glutamate(200–

500

lm)didnotaffect

thecellviabilityof

E14.5

retinalexplants.(E)The

proportion

ofTUNEL+cells,

asquantified

inconfocal

serial

sections

ofretinalexplants,was

unchangedafterglutam

atestim

ulation(500

lm

for

20h).(F)E14.5

retinalexplants

weretreatedwithglutam

atefor16

h.Real-timePCR

analysis

revealed

nosignificant

change

intheexpression

ofanymoleculeexam

ined.(G

)Treatmentof

retinalexplantswithAMPA

orkainate(100

lm)reducedtheproportion

ofBrdU+cellsin

retinalexplants

similar

toglutam

ate.

N-m

ethy

l-d-

aspartate(N

MDA,100

lm)hadno

effect.(H

)Co-incubation

withNBQX

(100

lm),bu

tno

tMK-801

(25

lm),

preventedtheglutam

ate-mediatedreductionof

BrdU+cells.

(I)Treatmentof

E14.5

retinalexplants

for3days

withNBQX

(100

lm)increasedtheproportion

ofBrdU+.In

contrast,MK-801

hadno

effect,suggesting

that

endogeno

usglutam

atereducesprogenitor

cellproliferationthroug

hAMPA

⁄kainate

receptors.Error

bars

indicate

SEM.Scale

bars:10

lm

(A).

Fig.1.

Glutamatereceptorsandtransporters

areexpressedduring

retinaldevelopm

ent.

Expressionof

genesthat

encode

glutam

atereceptor

subunits

(A)andglutam

ateup

take

transporters

(B)was

determ

ined

byreal

timeRT-PCR

atsevenstages

ofmouse

retinal

developm

ent.Graphsshow

expression

ofGluR1,

KA2,

NR1(A

),EAAT1,

2,3and4(B)

norm

alized

byGpi-1

andrelative

toexpression

levelsatE14

.5.(C–E

)Toverify

theprotein

expression

ofglutam

atereceptorssubunits

inretinalprogenitor

cells,E14.5

explantswere

cultured

inthepresence

of[3H]-thym

idinefor1h,

dissociated,

stained

forglutam

ate

receptorssubunits

GluR1,

GluR5

⁄6⁄7

orNR1(red),andoverlaid

withautoradiographic

emulsion.Nuclear

counterstainingisshow

nin

green.

Graphsshow

theproportion

ofcells

expressing

thereceptor

subunitsscored

inthetotalsample(D

)andin

the[3H]-labeledcells

(E).

Error

bars

indicate

SEM.Scale

bars:10

lm(C).



To test for a role of glutamate as an endogenous regulator of retinalprogenitor cells proliferation, we maintained retinal explants in culturein the presence of NBQX or MK-801 alone for 16–20 h. Theproportion of S-phase cells in these treated samples did not differ fromthat of controls (data not shown). But, interestingly, when theincubation period was extended to 3 days, cells maintained in thepresence of NBQX showed a significant, dose-dependent increase inthe proportion of S-phase cells (Fig. 2I and data not shown). MK-801(2.5–25 mm) had no significant effect.

Glutamate induces cell cycle exit of retinal progenitors

The reduction in BrdU-labeled cells after glutamate exposure may bethe result of either premature cell cycle exit or changes in the kineticsof cell cycle progression. To distinguish between these two possibil-ities, we pulse labeled the retinal progenitors cells in S-phase withBrdU and then treated the retinal explants with glutamate. To ensurethat BrdU+ cells had progressed through the S-phase of the cell cycle(Alexiades & Cepko, 1996), we waited 9 h before adding [3H]-thymidine to the culture medium (Fig. 3A). At different time points(11, 15, 19 and 24 h), the retinae were dissociated, immunostained forincorporated BrdU and overlaid with autoradiographic emulsion to

detect [3H]-thymidine. BrdU+ cells, [3H]-thymidine+ cells anddouble-positive cells were scored in glutamate-treated and untreatedsamples (Fig. 3B). As expected, the proportion of BrdU+ cells wasindistinguishable between glutamate-treated or untreated samples at11 h. This is because retinal explants were labeled with BrdU prior toglutamate addition. Typically, as cells progress through the G2 and Mphases and undergo cytokinesis, the proportion of BrdU+ cells slightlyincreases (Dyer & Cepko, 2000a, 2001a). In the untreated retinalexplants, there was an increase in the proportion of BrdU+ cells at24 h, but this did not occur in the glutamate-treated samples (Fig. 3C).Cumulative [3H]-thymidine-labeling data indicated that glutamate-

treated samples contained a significantly smaller proportion of labeledcells than did untreated samples at 19 h (P ¼ 0.017) and 24 h(Fig. 3D). These results are consistent with an increase in theproportion of retinal progenitor cells exiting the cell cycle exit afterglutamate treatment. If this is true, then the proportion of BrdU+, [3H]-thymidine+ cells would also be reduced at 19 and 24 h, because fewerBrdU+ cells would progress all the way through the cell cycle, re-enterthe S-phase and incorporate [3H]-thymidine. Our data showed that thiswas indeed the case, the proportion of BrdU+, [3H]-thymidine+ cellswas significantly reduced in the glutamate-treated samples at 19 h(P ¼ 0.019) and 24 h (Fig. 3E). Taken together, these data suggestthat glutamate receptor activation leads to an increase in the proportion

Fig. 3. Kinetics of glutamate-induced cell cycle exit. (A) In order to label S-phase cells, E14.5 retinal explants were cultured for 1 h in the presence ofbromodeoxyuridine (BrdU, 10 lm). Then, explants were washed, culture medium was changed, and glutamate (500 lm) was added. After 9 h, [3H]-thymidine wasadded to the culture medium, and the experiment was ended at the time points indicated. (B) Representative picture of BrdU- and [3H]-thymidine labeling ofdissociated retinal cells. (C) The proportion of BrdU+ cells increased with cell cycle progression. Glutamate-treated samples exhibited a smaller increase. (D, E)Treatment of retinal explants with glutamate decreased the proportion of [3H]-thymidine+ cells and [3H]-thymidine+ BrdU+ cells. The kinetics of glutamate’s anti-proliferative effect and the decrease in the proportion of double-labeled cells suggests that glutamate induces cell cycle exit in a proportion of retinal progenitor cells.Different y-axis scales were used to allow the visualization of the differences. Error bars indicate SEM. Scale bar: 10 lm (B).



of retinal progenitor cells that exit the cell cycle with little, if any,effect on cell cycle length.

Glutamatergic signaling regulates the clonal expansion ofretinal progenitor cells

To test the effects of changes in glutamatergic signaling on individualretinal progenitor cells throughout retinal histogenesis, we performedclonal analysis using the NIN-E retrovirus (Fig. 4A). This replication-incompetent retrovirus infects retinal progenitor cells and expressesnuclear LacZ, which is ideal for scoring clone size (Fig. 4B). E14.5mouse retinal explants were infected with NIN-E and cultured for10 days either in the presence or absence of NBQX. Retinal explantswere then stained with X-gal, and the number of cells in each clonewas scored in serial sections. By scoring over 450 clones, we obtaineda distribution of clone sizes ranging from 1 to 20 cells. If the anti-proliferative effect of glutamate occurs through the AMPA ⁄ kainateglutamate receptor as our pharmacological experiments suggest(Fig. 2), then exposure of retinal explants to the AMPA ⁄ kainateantagonist NBQX should increase clone size. NIN-E-infected retinalexplants maintained in culture for 10 days in the presence of NBQX(100 lm) underwent a significant (P ¼ 0.008) increase in theproportion of large clones (‡ 5 cells) and a decrease in the proportionof small clones (1 cell clones) compared with untreated controls(Fig. 4C). As expected, glutamate treatment had the converse effectleading to an increase in the proportion of small clones at the expenseof large clones (data not shown).

Cell-autonomous effect of AMPA glutamate receptors

The expression data (Fig. 1) combined with the pharmacological data(Fig. 2) suggested that the premature cell cycle exit of retinal

progenitor cells is primarily mediated by the AMPA ⁄ kainate subtypeof glutamate receptors. However, those data did not distinguishbetween cell-autonomous and non-cell-autonomous effects of glutam-ate signaling. To test whether regulation of retinal progenitor cellproliferation by glutamate is cell-autonomous, we ectopicallyexpressed GluR1 (Gria1) in E14.5 retinae and analysed their effecton glutamate signaling. In addition, we analysed previously charac-terized dominant-negative form of the GluR1 (Gria1) (M599R)receptor (Robert et al., 2002).Briefly, E14.5 retinae were transfected with plasmids expressing the

different glutamate receptors and a venus YFP (vYFP) reporter geneby squarewave electroporation (Matsuda & Cepko, 2004) (Fig. 5A).Seventy-two hours later, proliferating retinal progenitor cells werelabeled with [3H]-thymidine for 1 h. Culture media was changed;retinal explants were washed and then cultured for an additional 23 hin the presence or absence of glutamate. To determine the proportionof retinal progenitor cells that were still dividing, a 1-h BrdU pulsewas performed (Fig. 5A). Retinae were then dissociated, and thevYFP+ cells were purified by FACS (Fig. 5B). The proportions of[3H]-thymidine+ cells and BrdU+, [3H]-thymidine+ cells in eachsample were determined (Fig. 5C).Ectopic expression of GluR1 (Gria1) significantly reduced the

proportion of [3H]-thymidine+ retinal progenitor cells, and glutamateexposure further reduced the proportion of [3H]-thymidine+ cells(Fig. 5D). To determine if retinal progenitor cells labeled with [3H]-thymidine prior to glutamate treatment prematurely exited the cellcycle, we scored the proportion of BrdU+, [3H]-thymidine+ cells.Consistently, glutamate receptor activation also decreased the propor-tion of double-labeled cells (Fig. 5E). These data were consistent withpremature cell cycle exit (see Fig. 3) in response to glutamatesignaling through the AMPA glutamate receptors. In addition,glutamate response was blocked when the dominant-negative GluR1receptor, M599R, was ectopically expressed (Fig. 5D and E). These

Fig. 4. Effect of glutamatergic signaling modulation on clonal expansion of retinal progenitor cells. (A) E14.5 retinal explants were infected with a replication-incompetent retrovirus NIN-E that encodes nuclear-localized b-galactosidase. After 10 days in culture, explants were stained with X-gal and sectioned. The effect ofglutamatergic signaling modulation on the clonal expansion of retinal progenitors was examined by quantifying the number of clones and the number of cells in eachclone, as exemplified in (B). (C) Graph showing the proportion of clones containing the indicated number of cells. Retinae were cultured for 10 days in thepresence of NBQX (100 lm). Pharmacological blockade of the AMPA ⁄ kainate receptor increased the proportion of clones containing few cells and decreased theproportion of larger clones. Error bars indicate SEM. Scale bars: 10 lm (B).



findings confirm that signaling through the AMPA receptor subtype isthe primary mechanism by which glutamate regulates retinal progen-itor cell proliferation during development.

Mechanism of glutamate-mediated cell cycle exit

To begin to determine the mechanism underlying glutamate-mediatedcell cycle exit in retinal progenitor cells, we performed microarrayanalysis using RNA prepared from E14.5 retinae cultured in thepresence of glutamate for 16 h. RNA from untreated E14.5 retinaecultured in parallel were used as a control. The microarray was acDNA mouse retinal microarray with 13 000 total cDNAs including abroad collection of cell cycle genes. None of the cell cycle genesassociated with the regulation of retinal progenitor cell proliferationwas found to be either up- or downregulated after the 16-h glutamatetreatment (data not shown). To confirm these data, we performed

real-time PCR using TaqMan probes to several genes that encode keycell cycle proteins found in the developing retina, i.e. p27 (Cdkn1b),cyclin D1 (Ccnd1), cyclin D3 (Ccnd3) and cyclin E1 (Ccne1). Nosignificant difference was seen in the mRNA levels of these genesafter 4 or 16 h of glutamate treatment (Fig. 6A and B). Taken together,these results suggest that the premature cell cycle exit induced byglutamate may be mediated by post-transcriptional modification of thecell cycle machinery rather than by direct changes in gene expression.To test if the protein levels of the cell cycle regulators change after

glutamate treatment, we performed immunoblot analysis for cyclin D1(Ccnd1), cyclin D3 (Ccnd3), p27 (Cdkn1b), p57 (Cdkn1c) and p19(Cdkn2d). There was no significant change in the expression of any ofthese proteins after 4, 8 or 16 h of glutamate treatment (Fig. 6C; datanot shown). We also analysed the expression of the major cyclin-dependent kinases in the developing retina (Cdk2 and Cdk4) after 4 or24 h of glutamate treatment (Fig. 6E; data not shown) and found nochange on the levels of either protein. Next, we examined the

Fig. 5. Cell-autonomous effect of glutamatergic signaling on the proliferation of retinal progenitors. (A) E14.5 retinal explants were electroporated with indicatedplasmids and cultured for 3 days in vitro. S-phase cells were labeled by adding [3H]-thymidine to the culture medium for 1 h. Then, explants were washed, culturemedium was changed and glutamate (500 lm) added. One day later, retinal explants were pulsed with bromodeoxyuridine (BrdU, 10 lm) for 1 h. (B, C) Dissociatedretinal cells were purified by FACS, plated, stained for BrdU and overlaid with autoradiographic emulsion. Nuclear counterstaining is shown in blue. Graphs showthe proportions of [3H]-thymidine+ (D) or BrdU+ [3H]-thymidine+ cells (E). Overexpression of GluR1 decreased the proportion of [3H]-thymidine+ and double-positive cells, indicating that increased glutamatergic activity induced a premature cell cycle exit. Error bars indicate SEM. Scale bar: 10 lm (C).



Fig. 6. Glutamate effect on the expression of molecules that regulate the cell cycle. (A and B) E14.5 retinal explants were treated with glutamate for 4 or 16 h.Real-time PCR analysis revealed no significant change in the expression of any molecule examined. (C) This finding was confirmed by immunoblot analysis ofcyclin D1, cyclin D3, p19, p27 and p57 protein expression after treatment with glutamate (500 lm) for 16 h. (D) Immunoblot analysis of Cdk2 and Cdk4 proteinexpression after treatment with glutamate (500 lm) for 4 or 24 h. (E) Phosphorylation of Cdk2 at the threonine 160 (Thr 160) residue was assayed by immunoblotafter a 30-min glutamate treatment. (F, G) Glutamate (500 lm) effect on Cdk2 kinase activity in embryonic E14.5 retinas cultured for 4 h or 16 h. Cdk2-associatedkinase activity was determined after Cdk2 immunoprecipitation and assessment of in vitro phosphorylation of histone H1 substrate. (C–F) Immunoblots and Cdk2activity assays were replicated in four independent experiments with similar results; a representative image of each is shown. (G) Error bars indicate SEM.



phosphorylation status of the threonine 160 residue of Cdk2, which isrequired for activation of the kinase and progression through the cellcycle. Cdk2 phosphorylation was slightly reduced after glutamatetreatment (Fig. 6E). To test whether this reduction had an effect onkinase activity, we performed a Cdk2 kinase assay using lysatesprepared from E14.5 retinae treated with glutamate for 4 or 16 h, andcompared the kinase activity to that of lysates prepared from untreatedretinae. The level of substrate (histone H1) phosphorylation byimmunoprecipitated Cdk2 was reduced after glutamate treatment(Fig. 5F and G). Our results are consistent with a reduction in Cdk2kinase activity after glutamate receptor activation.

Discussion

In this study, we present several lines of evidence that glutamateregulates retinal progenitor cell proliferation during development.Gene expression studies showed that glutamate receptors and uptaketransporters are expressed during embryonic development of themouse retina before synaptogenesis. The expression of several classesof glutamate receptors in embryonic retinal progenitors was confirmedby immunohistochemical analysis. Using the in vitro retinal explantculture system, we demonstrated that activation of glutamate receptorsreduces cell proliferation during embryonic retinal developmentwithout affecting cell death or activating the p53 cell cycle arrestpathway. A long-term analysis of the clonal expansion of the retinalprogenitor cells also indicated an anti-proliferative role of glutamateduring retinal development. We did not detect effects of glutamatereceptor activation on the expression of genes or proteins known toregulate the cell cycle machinery in retinal progenitor cells. Instead, inaccordance with the anti-proliferative effects observed, glutamatereceptor activation significantly decreased Cdk2 kinase activity.These data are consistent with a model in which glutamate releasedin the retinal tissue leads to increased retinal progenitor cell cycle exit.We hypothesize that this may be a mechanism to coordinate cell cycleexit of retinal progenitors with neuronal differentiation duringretinogenesis.

Retinal progenitor cells express glutamate receptors

Previous studies have demonstrated the gene expression of glutamatereceptors in the developing mouse retina. The expression of theAMPA ⁄ kainate subunits GluR1, GluR4 and KA2 was shown to startat about E14.5, while that of GluR2, GluR3, GluR5, GluR6, GluR7and KA1 were first detected at about E16.5. All of these genes werepersistently expressed during postnatal development of the mouseretina (Zhang et al., 1996). NMDA receptor subtypes were alsodetected during the early stages of mouse retinal development; NR1and NR2B subunits were detected as early as E15.5 (Watanabe et al.,1994). Using real-time RT-PCR analysis, we confirmed the expressionof different genes that encode representative subunits of AMPA,kainate and NMDA glutamate receptors during mouse retinaldevelopment. The expression of glutamate uptake transporters duringretinal development has also been studied previously. EAAT1, EAAT2(GLT-1a) and EAAT5 were first detected at birth in rat retinae, andtheir level of expression increased at about the first week of postnataldevelopment (Pow & Barnett, 1999, 2000; Reye et al., 2002).Interestingly, the other EAAT2 splice variant, GLT-1B, is expressed asearly as E14.5 (Reye et al., 2002). In accordance with previousstudies, we detected the expression of EAAT1 mRNA during postnataldevelopment. We extended these findings by showing that glutamateuptake transporters mRNA are also expressed during embryonic

development of the mouse retina, suggesting that the extracellularavailability of glutamate may be precisely regulated during embryonicretinal development.More importantly, immunostaining analysis using specific antibod-

ies against the subunits GluR1 (AMPA), GluR5 ⁄ 6 ⁄ 7 (kainate) andNR1 (NMDA) demonstrated that all three classes of glutamateionotropic receptors are expressed in retinal progenitor cells of thedeveloping retina. Previous studies also demonstrated the expressionof glutamate receptors in proliferative regions of the developing CNS.Expression of functional AMPA receptors has been observed incortical progenitor cells in vitro and in vivo (LoTurco et al., 1995;Maric et al., 2000), and similar results were obtained in embryonicrabbit and chick retinae (Wong, 1995; Pearson et al., 2002). We haveextended those earlier studies, providing a precise quantification of theproportion of cells expressing glutamatergic receptors and directlydemonstrating that proliferating [3H]-thymidine-labeled retinal pro-genitors cells express different classes of glutamate receptors. Our dataprovide evidence that glutamate receptors are expressed in prolifer-ating retinal progenitor cells of the developing mouse retina.

Endogenous activation of glutamate receptors inhibits retinalprogenitor cells proliferation

Activation of glutamate receptors has been previously shown tomodulate cell proliferation of cortical and cerebellar progenitor cellsduring development (Yuan et al., 1998; Haydar et al., 2000; Luk et al.,2003). In addition, it has been reported that activation of metabotropicglutamate receptors regulates the proliferation of progenitor cells fromthe adult forebrain (Giorgi-Gerevini et al., 2005).The data presented here provide for the first time evidence that

glutamate is an anti-proliferative factor in the developing mouseretina. Three independent experimental approaches support thisconclusion. First, activation of glutamatergic receptors by glutamateor AMPA ⁄ kainate receptor agonists induced cell cycle exit. Moreimportantly, inactivation of the AMPA ⁄ kainate increased the propor-tion of cells in the S-phase. Second, long-term blockade ofAMPA ⁄ kainate receptors altered the pattern of clonal expansion ofretinal progenitor cells, increasing the proportion of large clones.Finally, overexpression of the AMPA receptor subunit GluR1decreased the proliferation of retinal progenitor cells in a cell-autonomous manner, and the dominant negative point mutant form ofthe GluR1 receptor (M599R) blocked this effect.Addition of exogenous glutamate at concentrations that are known

to activate the different classes of glutamate receptors decreased theproportion of proliferating retinal progenitor cells. This effect could bea consequence of retinal progenitor cell death, cell cycle lengtheningor cell cycle exit. However, the highest concentration of glutamate thatwas used (500 lm) did not change the proportions of viable orapoptotic cells in comparison with the untreated samples. Accordingly,previous studies have demonstrated that similar concentrations ofglutamate (or its agonists) activate glutamatergic receptors expressedby progenitor cells inducing membrane depolarization and calciuminflux without any toxic effect (LoTurco et al., 1995; Wong, 1995;Pearson et al., 2002). Moreover, both the kinetics of the anti-proliferative effect of glutamate and the decrease in the proportion ofprogenitor cells that continued in the cell cycle (Fig. 3E) indicate thatglutamate induces retinal progenitors to exit the cell cycle. However,we cannot rule out minor alterations in progenitor cell cycle lengthafter glutamate receptor activation.Consistent with these findings, pharmacological inactivation of the

AMPA ⁄ kainate receptors increased the proportion of cells entering the



S-phase. The increased proportion of BrdU-labeled cells after NBQXtreatment suggested that receptor activation by endogenous glutamatesignals retinal progenitor cells to exit the cell cycle. Indeed, clonalexpansion of retinal progenitor cells infected with a replication-incompetent retrovirus supports this proposal. Long-term treatmentwith NBQX increased the proportion of larger clones and reduced theproduction of smaller clones. These findings are consistent withendogenous glutamate driving retinal progenitor cells out of the cellcycle through the activation of AMPA ⁄ kainate receptors. Consistently,previous studies indicated that glutamate is abundant in the extracel-lular milieu of developing retinal tissue (Fletcher & Kallionatis, 1997).Our data add to the previously described anti-proliferative effect ofother neurotransmitters upon progenitor cells of embryonic mouseretina (Young & Cepko, 2004). However, preliminary data indicatethat, different from taurine, glutamate receptor activation has no effecton the cell fate regulation. No difference in the proportion of anyretinal cell type was observed after culturing E14.5 mouse retinas for10 days in the presence or absence of glutamate (data not shown).These data suggest that glutamate receptor activation may specificallyregulate signaling pathways that are exclusively involved in cell cycleprogression and not cell fate specification.In addition to the pharmacological data, the effect of misexpression

of wild-type and mutant forms of AMPA receptor subunits indicatesthat GluR1 is the primary receptor subunit involved in the anti-proliferative effect of glutamate upon retinal progenitor cells. Over-expression of the dominant-negative GluR1 receptor (M599R)completely blocked glutamate-induced decrease in retinal progenitorcells proliferation. However, it is noteworthy that the decrease in theproportion of BrdU+ cells induced by glutamate is smaller then theproportion of retinal progenitor cells expressing the GluR1 protein(Fig. 1D). It is possible that some retinal progenitor cells may have tointegrate other anti-proliferative signals to exit the cell cycle. This maybe another extrinsic factor, such as neurotransmitter systems (e.g.acetylcholine, taurine, c-aminobutyric acid, ATP), or an intrinsiccomponent capable of regulating retinal progenitor cells responsivityto environmental factors.Based on the effect of the AMPA ⁄ kainate pharmacological

inhibitor, NBQX, one would expect that genetic inactivation of thisreceptor by overexpression of the dominant-negative GluR1 receptor(M599R) would also increase cell proliferation in untreated retinae.However, the proportion of BrdU+ cells in retinae ectopicallyexpressing GluR1(M599R) did not increase compared with the vectorcontrol retinae. One possible explanation for this result is that ectopicexpression of GluR1 (M599R) does not inactivate all of theendogenous GluR1-containing receptors. That is, some receptorsmay not have incorporated the dominant negative subunit. Anotherpossibility is that the effect of glutamate on retinal progenitor cellproliferation is mediated by a combination of GluR1-containingreceptors and non-GluR1-containing receptors. Therefore, ectopicexpression of GluR1 (M599R) would only inactivate a subset of thefunctional glutamate receptors on retinal progenitor cells. Thepharmacological agent (NBQX) inactivates all AMPA ⁄ kainate recep-tors. This would explain why the pharmacological data with NBQX(Fig. 2) differ from the data using GluR1 (M599R).One interesting, but unresolved, aspect of glutamate-mediated

regulation of retinal histogenesis is the determination of the source ofthis neurotransmitter within the embryonic retinal tissue. It was shownthat postmitotic cells (e.g. ganglion cells) may regulate retinalprogenitors proliferation by secreting signals that either promote orinhibit progenitor cells proliferation (Belliveau & Cepko, 1999;Gonzalez-Hoyuela et al., 2001; Wang et al., 2005). It is possible thatglutamate secreted by such early-born differentiating retinal neurons

signals retinal progenitor cells to generate more postmitotic cells.However, additional studies will be required to determine if retinalganglion cells are the source of glutamate that controls retinalprogenitor cells proliferation.

Mechanisms of glutamate-induced cell cycle exit

Many studies have demonstrated glutamate-mediated regulation ofcell proliferation during CNS development. However, the glutamate-mediated cell cycle mechanisms that control the proliferation ofneuronal progenitor cells have not been identified.The functional importance of cyclin D1 and the Cdk inhibitors p27,

p57 and p19 in the control of the G1 ⁄ S-phase transition in retinalprogenitor cells has been well documented (Fantl et al., 1995;Sicinsky et al., 1995; Dyer & Cepko, 2001b; Tong & Pollard, 2001;Cunningham et al., 2002). Although, different extrinsic factors havebeen shown as regulators of retinal progenitor cells proliferation(Lillien & Cepko, 1992; Young & Cepko, 2004; Close et al., 2005;Wang et al., 2005), only recently the cell cycle mechanisms thatmediate those functions started to be elucidated. It was suggested thattumor growth factor b mediates its anti-proliferative effect through theregulation of p27 expression (Close et al., 2005). It was also reportedthat cyclin D1 expression is regulated by Shh and may have a role inShh-induced cell proliferation (Wang et al., 2005). We did not detectglutamate-induced changes in the gene expression levels of cell cycle-related molecules. In addition, immunoblot analysis demonstrated thatthe protein expression of cyclin D1, D3, Cdk2, Cdk4, p19, p27 andp57 was not changed by glutamate receptor activation. These datastrongly suggest that glutamate may induce cell cycle exit throughpost-transcriptional mechanisms.Phosphorylation of Rb protein by the cyclin E-Cdk2 complex is

crucial for progenitor cells to progress from G1 to S-phase (Sherr &Roberts, 1999). Therefore, Cdk2 kinase activity is a potential target forsignals driving progenitor cells to exit the cell cycle. Activation ofAMPA ⁄ kainate glutamate receptors was shown to induce cell cyclearrest in glial progenitor cells by inhibiting Cdk2 activity (Ghiani &Gallo, 2001). Our data similarly showed that glutamate receptoractivation decreased Cdk2-associated activity from embryonic retinalprogenitors. However, we believe that the inhibition of Cdk2 activityin the retina is not part of a cell cycle arrest or an excitotoxic response,because the gene expression of different targets of the p53 pathway,including the cyclin kinase inhibitor p21, was not altered afterglutamate receptor activation (Fig. 2). Together with the unchangedlevels of mRNA and protein expression of known cell cycleregulators, the observed decrease in Cdk2 activity supports the ideathat glutamate-induced cell cycle exit may be mediated by a post-transcriptional inhibition of this kinase. This adds to the well-knowntranscriptionally mediated regulation of cyclin D1 expression bymitogenic factors (Lobjois et al., 2004; Wang et al., 2004), as amechanism of regulation of the cell cycle of CNS progenitor cells byextrinsic factors. These data also provide evidence for the molecularmechanism involved in the anti-proliferative effect of glutamate in thedeveloping retina.

Supplementary material

The following supplementary material may found on www.blackwell-synergy.comTable S1. Primers and probes used for real-time RT-PCR analysis.Table S2. Glutamate receptors and transporters expressed in thedeveloping mouse retina.



Acknowledgements

We thank J.R. Howe (Yale University School of Medicine) for the glutamatereceptors cDNA, and J. Gray, J. Zhang, L. Dabo, J. Nilson dos Santos and J. F.Tiburcio for technical assistance and expertise. This work was supported bygrants from the National Institutes of Health, Cancer Center Support from theNational Cancer Institute, Research to Prevent Blindness and the AmericanLebanese Syrian Associated Charities (ALSAC) (to M.A.D.), and by grantsfrom the Conselho Nacional de Desenvolvimento Cientifico e Tecnologico(CNPq), Fundacao de Amparo a Pesquisa do Estado do Rio de Janeiro (Faperj)and Pronex (to R.L.). R.A.P.M. was supported by fellowships from CNPq andCoordenacao de Aperfeicoamento de Pessoal de Nıvel Superior (CAPES); R.L.is a fellow of The John Simon Guggenheim Foundation. M.A.D. is a PewScholar.

Abbreviations

BrdU, bromodeoxyuridine; DTT, dithiothreitol; E, embryonic day; NMDA,N-methyl-d-aspartate;P, postnatal day;PBS,phosphate-buffered saline;RT-PCR,reverse transcriptase-polymerase chain reaction; SDS, sodium dodecyl sulfate.

References

Alexiades, M.R. & Cepko, C. (1996) Quantitative analysis of proliferation andcell cycle length during development of the rat retina. Dev. Dyn., 205, 293–307.

Belliveau, M. & Cepko, C. (1999) Extrinsic and intrinsic factors control thegenesis of amacrine and cone cells in the rat retina. Development, 126, 555–566.

Carey, R.G., Li, B. & DiCicco-Bloom, E. (2002) Pituitary adenylate cyclaseactivating polypeptide anti-mitogenic signaling in cerebral cortical progeni-tors is regulated by p57Kip2-dependent CDK2 activity. J. Neurosci., 22,1583–1591.

Close, J.L., Gumuscu, B. & Reh, T. (2005) Retinal neurons regulateproliferation of postnatal progenitors and Muller glia in the rat retina viaTGFb signaling. Development, 132, 3015–3026.

Cunningham, J.J., Levine, E.M., Zindy, F., Goloubeva, O., Roussel, M.F. &Smeyne, R.J. (2002) The cyclin-dependent kinase inhibitors p19 (Ink4d) andp27 (Kip1) are coexpressed in select retinal cells and act cooperatively tocontrol cell cycle exit. Mol. Cell. Neurosci., 19, 359–374.

Dyer, M.A. (2003) Regulation of proliferation, cell fate specification anddifferentiation by the homeodomain proteins Prox1, Six3, and Chx10 in thedeveloping retina. Cell Cycle, 2, 350–357.

Dyer, M.A. & Cepko, C.L. (2000a) Control of Muller glial cell proliferationand activation following retinal injury. Nat. Neurosci., 3, 873–880.

Dyer, M.A. & Cepko, C.L. (2000b) p57 (Kip2) regulates progenitor cellproliferation and amacrine interneuron development in the mouse retina.Development, 127, 3593–3605.

Dyer, M.A. & Cepko, C.L. (2001a) Regulating proliferation during retinaldevelopment. Nat. Rev. Neurosci., 2, 333–342.

Dyer, M.A. & Cepko, C.L. (2001b) p27Kip1 and p57Kip2 regulateproliferation in distinct retinal progenitor cell populations. J. Neurosci., 21,4259–4271.

Fantl, V., Stamp, G., Andrews, A., Rosewell, I. & Dickson, C. (1995) Micelacking cyclin D1 are small and show defects in eye and mammary glanddevelopment. Genes Dev., 9, 2364–2372.

Fletcher, E.L. & Kallionatis, M. (1997) Localisation of aminoacid neuro-transmitters during postnatal developmental of the retina. J. Comp. Neurol.,380, 449–471.

Ghiani, C.A., Eisen, A.M., Yuan, X., DePinho, R.A., McBain, C.J. & Gallo, V.(1999) Neurotransmitter receptor activation triggers p27 (Kip1) and p21(CIP1) accumulation and G1 cell cycle arrest in oligodendrocyte progenitors.Development, 126, 1077–1090.

Ghiani, C.A. & Gallo, V. (2001) Inhibition of cyclin E-cyclin-dependent kinase2 complex formation and activity is associated with cell cycle arrest andwithdrawal in oligodendrocyte progenitor cells. J. Neurosci., 21, 1274–1282.

Giorgi-Gerevini, V., Melchiorri, D., Battaglia, G., Ricci-Vitiani, L., Ciceroni,C., Busceti, C.L., Biagioni, F., Iacovelli, L., Canudas, A.M., Parati, E., DeMaria, R. & Nicoletti, F. (2005) Endogenous activation of metabotropicglutamate receptors supports the proliferation and survival of neuralprogenitor cells. Cell Death Differ., 12, 1124–1133.

Gonzalez-Hoyuela, M., Barbas, J. & Rodriguez-Tebar, A. (2001) The auto-regulation of retinal ganglion cells cell number. Development, 128, 117–124.

Haydar, T.F., Wang, F., Schwartz, M.L. & Rakic, P. (2000) Differentialmodulation of proliferation in the neocortical ventricular and subventricularzones. J. Neurosci., 20, 5764–5774.

Levine, E., Close, J., Fero, M., Ostrovsky, A. & Reh, T. (2000) p27 (Kip1)regulates cell cycle withdrawal of late multipotent progenitor cells in themammalian retina. Dev. Biol., 219, 299–314.

Lillien, L. & Cepko, C.L. (1992) Control of proliferation in the retina: temporalchanges in responsiveness to FGF and TGF alpha. Development, 115, 253–266.

Lobjois, V., Benazeraf, B., Bertrand, N., Medevielle, F. & Pituello, F. (2004)Specific regulation of cyclins D1 and D2 by FGF and Shh signalingcoordinates cell cycle progression, patterning, and differentiation duringearly steps of spinal cord development. Dev. Biol., 273, 195–209.

LoTurco, J.J., Owens, D.F., Heath, M.J., Davis, M.B. & Kriegstein, A.R.(1995) GABA and glutamate depolarize cortical progenitor cells and inhibitDNA synthesis. Neuron, 15, 1287–1298.

Luk, K.C., Kennedy, T.E. & Sadikot, A.F. (2003) Glutamate promotesproliferation of striatal neuronal progenitors by an NMDA receptor-mediatedmechanism. J. Neurosci., 23, 2239–2250.

Ma, C., Papermaster, D. & Cepko, C.L. (1998) A unique pattern ofphotoreceptor degeneration in cyclin D1 mutant mice. Proc. Natl Acad.Sci. USA, 95, 9938–9943.

Maric, D., Liu, Q.Y., Grant, G.M., Andreadis, J.D., Hu, Q., Chang, Y.H.,Barker, J.L., Joseph, J., Stenger, D.A. & Ma, W. (2000) Functionalionotropic glutamate receptors emerge during terminal cell division and earlyneuronal differentiation of rat neuroepithelial cells. J. Neurosci. Res., 61,652–662.

Marquardt, T. (2003) Transcriptional control of neuronal diversification in theretina. Prog. Retin. Eye Res., 22, 567–577.

Matsuda, T. & Cepko, C.L. (2004) Electroporation and RNA interference in therodent retina in vivo and in vitro. Proc. Natl Acad. Sci. USA, 101, 16–22.

Matsushime, H., Quelle, D.E., Shurtleff, S.A., Shibuya, M., Sherr, C.J. & Kato,J.Y. (1994) D-type cyclin-dependent kinase activity in mammalian cells.Mol. Cell. Biol., 14, 2066–2076.

Morrow, E.M., Belliveau, M.J. & Cepko, C.L. (1998) Two phases of rodphotoreceptor differentiation during rat retinal development. J. Neurosci., 18,3738–3748.

Nakayama, K., Nakayama, N., Ishida, N., Shirane, M., Inomata, A., Inoue, T.,Shishido, N., Horii, I., Loh, D. & Nakayama, K. (1996) Mice lacking p27(Kip1) display increased body size, multiple organ hyperplasia, retinaldysplasia, and pituitary tumors. Cell, 85, 707–720.

Nguyen, L., Malgrange, B., Breuskin, I., Bettendorff, L., Moonen, G.,Belachew, S. & Rigo, J.M. (2003) Autocrine ⁄ paracrine activation of theGABA(A) receptor inhibits the proliferation of neurogenic polysialylatedneural cell adhesion molecule-positive (PSA-NCAM+) precursor cells frompostnatal striatum. J. Neurosci., 23, 3278–3294.

Olney, J.W., Ho, O.L. & Rhee, V. (1971) Cytotoxic effects of acidic andsulphur containing amino acids on the infant mouse central nervous system.Exp. Brain Res., 14, 61–76.

Pearson, R., Catsicas, M., Becker, D. & Mobbs, P. (2002) Purinergic andmuscarinic modulation of the cell cycle and calcium signaling in the chickretinal ventricular zone. J. Neurosci., 22, 7569–7579.

Pearson, R.A., Dale, N., Llaudet, E. & Mobbs, P. (2005) ATP released via gapjunction hemichannels from the pigment epithelium regulates neural retinalprogenitor proliferation. Neuron, 46, 731–744.

Pow, D.V. & Barnett, N.L. (1999) Changing patterns of spatial buffering ofglutamate in developing rat retinae are mediated by the Muller cell glutamatetransporter GLAST. Cell Tissue Res., 297, 57–66.

Pow, D.V. & Barnett, N.L. (2000) Developmental expression of excitatoryamino acid transporter 5: a photoreceptor and bipolar cell glutamatetransporter in rat retina. Neurosci. Lett., 280, 21–24.

Reye, P., Sullivan, R. & Pow, D.V. (2002) Distribution of two splice variants ofthe glutamate transporter GLT-1 in the developing rat retina. J. Comp.Neurol., 447, 323–330.

Robert, A., Hyde, R., Hughes, T.E. & Howe, J.R. (2002) The expression ofdominant-negative subunits selectively suppresses neuronal AMPA andkainate receptors. Neuroscience, 115, 1199–1210.

Sherr, C.J. & Roberts, J.M. (1999) CDK inhibitors: positive and negativeregulators of G1-phase progression. Genes Dev., 13, 1501–1512.

Sicinski, P., Donaher, J.L., Parker, S.B., Li, T., Fazeli, A., Gardner, H., Haslam,S.Z., Bronson, R.T., Elledge, S.J. & Weinberg, R.A. (1995) Cyclin D1provides a link between development and oncogenesis in the retina andbreast. Cell, 82, 621–630.

Tong, W. & Pollard, J.W. (2001) Genetic evidence for the interactions of cyclinD1 and p27 (Kip1) in mice. Mol. Cell. Biol., 21, 1319–1328.



Uberti, D., Belloni, M., Grilli, M., Spano, P. & Memorandum, M. (1998)Induction of tumour-suppressor phosphoprotein p53 in the apoptosis ofcultured rat cerebellar neurones triggered by excitatory amino acids. Eur. J.Neurosci., 10, 246–254.

Uberti, D., Grilli, M. & Memorandum, M. (2000) Induction of p53 in theglutamate-induced cell death program. Amino Acids, 19, 253–261.

Wang, Y., Dakubo, G., Thurig, S., Mazerolle, C. & Wallace, V. (2005) Retinalganglion cell-derived sonic hedgehog locally controls proliferation and thetiming of RGC development in the embryonic mouse retina. Development,132, 5103–5113.

Wang, C., Li, Z., Fu, M., Bouras, T. & Pestell, R.G. (2004) Signaltransduction mediated by cyclin D1: from mitogens to cell proliferation:a molecular target with therapeutic potential. Cancer Treat Res., 119,217–237.

Watanabe, M., Mishina, M. & Inoue, Y. (1994) Differential distributions of theNMDA receptor channel subunit mRNAs in the mouse retina. Brain Res.,634, 328–332.

Wong, R.O. (1995) Effects of glutamate and its analogs on intracellular calciumlevels in the developing retina. Vis. Neurosci., 12, 907–917.

Young, R.W. (1985) Cell proliferation during postnatal development of theretina in the mouse. Brain Res., 353, 229–239.

Young, T.L. & Cepko, C.L. (2004) A role for ligand-gated ion channels in rodphotoreceptor development. Neuron, 41, 867–879.

Yuan, X., Eisen, A.M., McBain, C.J. & Gallo, V. (1998) A role for glutamateand its receptors in the regulation of oligodendrocyte development incerebellar tissue slices. Development, 125, 2901–2914.

Zhang, J., Gray, J., Wu, L., Leone, G., Rowan, S., Cepko, C.L., Zhu, X., Craft,C.M. & Dyer, M.A. (2004b) Rb regulates proliferation and rod photoreceptordevelopment in the mouse retina. Nat. Genet., 36, 351–360.

Zhang, C., Hammassaki-Britto, D.E., Britto, L.R. & Duvoisin, R.M. (1996)Expression of glutamate receptor subunit genes during development of themouse retina. Neuroreport, 8, 335–340.

Zhang, J., Schweers, B. & Dyer, M.A. (2004a) The first knockout mouse modelof retinoblastoma. Cell Cycle, 3, 952–959.

Zhu, C.C., Dyer, M.A., Uchikawa, M., Kondoh, H., Lagutin, O.V. & Oliver, G.(2002) Six-3 mediated auto repression and eye development requires itsinteraction with members of the Grouch-related family of co-repressors.Development, 129, 2835–2849.



Martins et al. 2006

Documents

Transcript of Martins et al. 2006