Neural stem cell properties of Müller glia in the mammalian retina: Regulation by Notch and Wnt signaling

Neural stem cell properties of Müller glia in the mammalian retina: Regulation by Notch and Wnt signaling

Please download to get full document.

View again

of 20
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Funny & Jokes

Publish on:

Views: 12 | Pages: 20

Extension: PDF | Download: 0

   Neural stem cell properties of Müller glia in the mammalian retina:Regulation by Notch and Wnt signaling Ani V. Das, Kavita B. Mallya, Xing Zhao, Faraz Ahmad, Sumitra Bhattacharya,Wallace B. Thoreson, Ganapati V. Hegde, Iqbal Ahmad ⁎  Department of Ophthalmology and Visual Sciences, DRC 4034, 98-5840 Nebraska Medical Center, University of Nebraska Medical Center,Omaha, NE 68198-5840, USA Received for publication 31 May 2006; revised 16 July 2006; accepted 25 July 2006Available online 29 July 2006 Abstract The retina in adult mammals, unlike those in lower vertebrates such as fish and amphibians, is not known to support neurogenesis. However,when injured, the adult mammalian retina displays neurogenic changes, raising the possibility that neurogenic potential may be evolutionarilyconserved and could be exploited for regenerative therapy. Here, we show that Müller cells, when retrospectively enriched from the normal retina,like their radial glial counterparts in the central nervous system (CNS), display cardinal features of neural stem cells (NSCs), i.e., they self-renewand generate all three basic cell types of the CNS. In addition, they possess the potential to generate retinal neurons, both in vitro and in vivo. Wealso provide direct evidence, by transplanting prospectively enriched injury-activated Müller cells into normal eye, that Müller cells haveneurogenic potential and can generate retinal neurons, confirming a hypothesis, first proposed in lower vertebrates. This potential is likely due tothe NSC nature of Müller cells that remains dormant under the constraint of non-neurogenic environment of the adult normal retina. Additionally,we demonstrate that the mechanism of activating the dormant stem cell properties in Müller cells involves Wnt and Notch pathways. Together,these results identify Müller cells as latent NSCs in the mammalian retina and hence, may serve as a potential target for cellular manipulation for treating retinal degeneration.© 2006 Elsevier Inc. All rights reserved.  Keywords:  Müller cells; Stem cells; Retina; Progenitors; Notch signaling; Wnt signaling; Chemical injury Introduction Two recent developments have major impact on our understanding of brain development and the potential of treating neurodegenerative changes. The first is the reaffirma-tion of long-standing observations that the adult brain harbors proliferating populations of cells and that neurons are bornthroughout life, particularly in the subventricular zone (SVZ)of the lateral ventricle and subgranular layer (SGL) of thedentate gyrus of the hippocampus (Alvarez-Buylla and Lim,2004). The second is that glia perform dual functions of  providing homeostatic support and participating in neurogene-sis (Alvarez-Buylla et al., 2001; Doetsch, 2003; Gotz and Barde, 2005). A variety of approaches demonstrated that SVZand SGL astrocytes possess NSC properties, and are the primary source of neurogenesis in the lateral ventricle-rostralmigratory zone and hippocampus (Doetsch et al., 1999; Seriet al., 2001). Radial glia, popularly known for providingscaffolds for migrating neuroblasts, were demonstrated toserve as NSCs in the embryonic brain and a prevalent sourceof cortical projection neurons (Malatesta et al., 2003; Noctor et al., 2001). Recently, SVZ radial glia in the adult brain wereobserved to be the source of SVZ astrocytic stem cells(Merkle et al., 2004).Unlike the SVZ and SGL, active neurogenesis has not beendetected in the normal adult mammalian retina. However,neurogenic changes have been observed in injured retina, andone of the sources of injury-induced neurogenesis is thought to Developmental Biology 299 (2006) 283 – ⁎  Corresponding author. Fax: +1 402 559 3251.  E-mail address: (I. Ahmad).0012-1606/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.ydbio.2006.07.029   be the major glial cell type of the retina, the Müller cells(Braisted et al., 1994; Ooto et al., 2004; Reh and Fischer, 2001).Are Müller cells latent NSCs in retina, that can be activated byextrinsic cues, including those that are injury-induced, to proliferate and generate new neurons? To answer this question,we purified Müller cells from the rodent retina and demon-strated that they display cardinal features of NSCs in vitro, i.e.,they self-renew, and generate both neurons and glia, in additionto generating retinal neurons in conducive culture conditions.Müller cells, identified and enriched prospectively as side population (SP) cells, when transplanted into retina generatedlamina-specific retinal neurons, thus providing a direct proof of their neurogenic potential. In addition, our results suggested that  Notch and Wnt pathways acted in concert to regulate the stemcell properties of Müller cells. This raises the possibility of treating degenerating retinas from within, by targeted manip-ulation of these cells. Materials and methods  Müller cell culture Enrichment of Müller cells was done according to a previously describedmethod (Hicks and Courtois, 1990). Briefly, eyes from postnatal (PN) days 10 – 21 rats were enucleated and incubated overnight in DMEM. Eyecups weretransferred to dissociation solution (DMEM containing 0.1% Trypsin and 70 U/ ml collagenase) and incubated at 37°C in CO 2  incubator for 1 h. Eyecups werewashed with DMEM and the retina was dissected out with care to avoidcontamination from RPE and ciliary epithelium. The retina was mechanicallydissociated into small aggregates and cultured in DMEM containing 10% FBSfor 8 – 10 days. The floating retinal aggregates and debris were removed leaving purified flat cell population of Müller cells attached to the bottom of the dish.Cells were trypsinized and cultured in DMEM containing 10% FBS for another 5daystogeta furtherpurifiedpopulation.Cellsweredissociated usingTrypsin – EDTA and cultured in DMEM/F12 (containing 1× N2 supplement (GIBCO),2 mM  L -glutamine, 100 U/ml penicillin, 100 μ g/ml streptomycin) supplementedwith FGF2 (10 ng/ml; Collaborative Research) and EGF (20 ng/ml;Collaborative Research) at a density of   ∼ 5×10 4 cells/cm 2 for 5 – 7 days togenerate neurospheres. To remove microglia, cell dissociates were subjected toimmunopanning using Mac-1 antibody (Barres et al., 1988;Meucci et al., 1998).Toexaminetheirneurosphere-formingability,microgliawaspurifiedfromretinaaccording to a previously described method (Roque and Caldwell, 1993) andcultured in the same medium mentioned above for 4 – 6 days. To ascertain theclonal generation of neurospheres by co-culturing green (GFP-expressing) andnon-green cells (Zhao et al., 2002), Müller cells were enriched from the retina of the green [constitutively expressing the green fluorescent protein (GFP)] andwild-type mice as described above. LDA analysis was carried out as previouslydescribed (Ahmad et al., 2004; Das et al., 2004). Cells were diluted to give aninitial concentration of 5000 cells/ml from which serial dilutions in 200  μ laliquots were plated in individual wells of a 96-well plate. Culture was carriedout for 7 days, after which the fraction of wells not containing neurospheres for each cell-plating density was calculated. The negative logarithm of fraction of negative wells was plotted against the number of cells plated/well to provide astraightlineinasemilogarithmplot.ThezerotermofPoissonequation(  F  0 = e −  x ,where  F  0  is the fraction of well without neurospheres and  x  is the mean number ofcell/well)predictsthatwhen37%oftestwellsarenegative,thereisan averageof one stem cell/well. To examine the differentiation potential of Müller cell-derived stem cells/progenitors, neurospheres were exposed to 10  μ M BrdU(Sigma) for the final 48 h to tag the dividing cells and plated on poly- D -lysine(500  μ g/ml) and laminin (5  μ g/ml) coated 12-mm glass coverslips. In order to promote differentiation, mitogens were substituted with brain-derived neuro-trophic factor (BDNF; 1 ng/ml), retinoic acid (RA; 1  μ M) and 1% FBS andculture was continued for 5 – 7 days. Cells were fixed using cold 4% paraformaldehyde for immunocytochemical analysis.  FACS analysis In order to find out the purity of enriched Müller cells, we carried out FACSanalysis using antibodies specific to Müller glia such as glutamine synthetase(GS) and vimentin. Briefly, cells in the monolayer culture were dissociated intosingle cells and fixed using ice-cold ethanol. After washing with 1× PBScontaining 1% BSA, cells (2×10 6 ) were incubated with appropriate dilution of GS/vimentin (Table 1) in 1× PBS containing 1% BSA for 1 h at 4°C. Prior toantibody incubation, cells were blocked for non-specific binding in 1× PBScontaining 0.1% Triton-X100 for 30 min at 4°C. After antibody incubation, cellswere washed and incubated in PBS-BSA solution containing appropriatesecondary antibodies linked to FITC for 1 h at 4°C. Cells were washed withPBS-BSA and resuspended in PBS-BSA for FACS analysis.  Immunocytochemical analysis Immunocytochemical analysis was carried out for the detection of BrdUand/or cell-specific markers as previously described (Zhao et al., 2002). Briefly, paraformaldehyde-fixed cells were incubated in PBS containing 5% NGS and 0,0.2 or 0.4% Triton-X100 followed by an overnight incubation in antibodies at 4°C. The list of antibodies used is given in Table 1. Cells were examined for epifluorescence following incubation in IgG conjugated to Cy3/FITC. Imageswere captured using cooled CCD-camera (Princeton Instruments) and Openlabsoftware (Improvision).  RT-PCR analysis Isolation of total RNA and cDNA synthesis were carried out as previouslydescribed (James et al., 2003). Briefly, 4  μ g of RNAwas transcribed into cDNAin total volume of 50  μ l. Specific transcripts were amplified with gene-specificforward and reverse primers using a step cycle program on a Robocycler (Stratagene). Amplifications were carried out for 25 cycles so that they remainedwithin the range of linearity to yield a semi-quantitative estimation of changes inthe relative levels of transcripts. Products were visualized by ethidium bromidestaining after electrophoresis on 2% agarose gel. Gene-specific primers used for RT-PCR analyses are given in Table 2. Co-culture experiments BrdU-labeled neurospheres derived from enriched Müller cells werecollected, washed extensively to remove excess BrdU and co-cultured onTable 1List of antibodies used Name Species Dilution CompanyVimentin Mouse 1:10 HybridomaGS Rabbit 1:5000 Sigma Nestin Mouse 1:4 HybridomaSox2 Rabbit 1:500 Chemicon Notch1 Rat 1:1 Gift  β -Tubulin Rabbit 1:1000 BabcoGFAP Rabbit 1:100 SigmaO4 Mouse 1:30 ChemiconMap2 Mouse 1:200 ChemiconCalbindin Rabbit 1:10,000 Gift Pax6 Rabbit 1:1000 BabcoChx10 Rabbit 1:250 Gift Rx Mouse 1:250 Gift CD31 Mouse 1:50 ChemiconMac-1 Rabbit 1:100 Accurate Chemical and Corp.Brn3b Rabbit 1:300 BabcoOpsin Mouse 1:1000 Gift PKC Rabbit 1:1000 SigmaSyntaxin Mouse 1:100 Gift BrdU Rat 1:100 Accurate Chemical and Corp.284  A.V. Das et al. / Developmental Biology 299 (2006) 283  –  302  Table 2List of primers used for the RT-PCR analysis Name Sequence Annealing temp (°C) Size (bp) Acc. NoVimentin Forward: 5 ′ AAGGCACTAATGAGTCCCTGGAG3 ′  56 251 NM031140Reverse: 5 ′ GTTTGGAAGAGGCAGAGAAATCC3 ′ Glutamine synthetase (GS) Forward: 5 ′ TCACAGGGACAAATGCCGAG3 ′  58 362 M96152Reverse: 5 ′ GTTGATGTTGGAGGTTTCGTGG3 ′ CRALBP Forward: 5 ′ CTGAGTTTGGAGGAATCTTGC3 ′  54 150 XM217702Reverse: 5 ′ TGGATTTGGGGGAGAGTTC3 ′ Opsin Forward: 5 ′ CATGCAGTGTTCATGTGGGA 3 ′  64 422 U22180Reverse: 5 ′ AGCAGAGGCTGGTGAGCATG 3 ′ mGluR6 Forward: 5 ′ CACAGCGTGATTGACTACGAG3 ′  56 317 D13963Reverse: 5 ′ CTCAGGCTCAGTGACACAGTTAG3 ′ HPC1 Forward: 5 ′ AAGAGCATCGAGCAGCAGAGCATC3 ′  60 342 NM016801Reverse: 5 ′ CATGGCCATGTCCATGAACAT3 ′ Brn3b Forward: 5 ′ GGCTGGAGGAAGCAGAGAAATC 3 ′  60 141 AF390076Reverse: 5 ′ TTGGCTGGATGGCGAAGTAG 3 ′ β -Tubulin Forward: 5 ′ TGCGTGTGTACAGGTGAATGC3 ′  54 250 NM139254Reverse: 5 ′ AGGCTGCATAGTCATTTCCAAG3 ′ GFAP Forward: 5 ′ ATCTGGAGAGGAAGGTTGAGTCG3 ′  58 310 NM017009Reverse: 5 ′ TGGCGGCGATAGTCATTAGA3 ′ PLP Forward: 5 ′ CGGGTGTGTCATTGTTTGGG3 ′  58 310 NM030990Reverse: 5 ′ ACAGGTGGAAGGTCATTTGGAAC3 ′  Notch1 Forward: 5 ′ TCTGGACAAGATTGATGGCTACG3 ′  56 329 NM008714Reverse: 5 ′ CGTTGACACAAGGGTTGGACTC3 ′  Nestin Forward: 5 ′ TGGAGCAGGAGAAGCAAGGTCTAC3 ′  56 295 NM012987Reverse: 5 ′ TCAAGGGTATTAGGCAAGGGGG3 ′ Pax6 Forward: 5 ′ CCATCTTTGCTTGGGAAATCC3 ′  56 310 NM_013001Reverse: 5 ′ TCATCCGAGTCTTCTCCATTGG3 ′ Musashi1 Forward: 5 ′ TGAAAGAGTGTCTGGTGATGCG3 ′  52 309 NM148890Reverse: 5 ′ GCCTGTTGGTGGTTTTGTCG3 ′ Sox2 Forward: 5 ′ AGGGCTGGGAGAAAGAAGAG3 ′  56 179 NM_011443Reverse: 5 ′ GGAGAATAGTTGGGGGGAAG3 ′ ABCG2 Forward: 5 ′ GATGGCTACACTCACTCACAAAG3 ′  54 325 NM_011920Reverse: 5 ′ ACTGAATCTAACCCAAGGAAGG3 ′ Clusterin Forward: 5 ′ CCTCCAGTCCAAGATGCTCAAC 58 292 NM_053021Reverse: 5 ′ TTTCCTGCGGTATTCCTGTAGCCarbonic anhydrase Forward: 5 ′ TTGCCAATGGAGACCGACAG 58 233 NM_019291Reverse: 5 ′ TGAGCCCCAGTGAAAGTGAAACCD31 Forward: 5 ′ AAGAGCAACTTCCAGACCGTCC 58 222 NM_031591Reverse: 5 ′ AAGCACCATTTCATCTCCAGACTGTyrosinase Forward: 5 ′ TCAGTCTATGTCATCCCCACAGG 56 252 NM_011661Reverse: 5 ′ GTTCTCATCCCCAGTTAGTTCTCGAth5 Forward: 5 ′ TGGGG(I)CA(GA)GA(CT)AA(GA)AA(GA)3 ′  52 231 AF071223Reverse: 5 ′ CAT(I)GG(GA)AA(I)GG(CT)TC(I)GG(CT)TG3 ′ Ath3 Forward: 5 ′ CGACTGGCAAGGAACTACATCTG3 ′  50 629 NM007501Reverse: 5 ′ ACTAATGCTCAGGGGTGGTGTG3 ′ Otx2 Forward: 5 ′ GAGAGGAGGTGGCACTGAAAATC3 ′  56 391 U96488Reverse: 5 ′ CCCCCAAAGTAGGAAGTTGAGC3 ′  NeuroD Forward: 5 ′ AAGAGGAGGAGGAAGAGGAGGAG3 ′  58 337 AF107728Reverse: 5 ′ TTGGTAGTGGGCTGGGACAAAC3 ′ CyclinA Forward: 5 ′ TACACACACGAGGTAGTGACGCTG3 ′  56 307 X60767Reverse: 5 ′ CCAAGCCGTTTTCATCCAGG3 ′ CyclinD1 Forward: 5 ′ ACACCAATCTCCTCAACCGACC3 ′  56 383 NM171992Reverse: 5 ′ GTTCACCAGAAGCAGTTCCATTTG3 ′ CyclinD3 Forward: 5 ′ CGAAACCACACCCCTGACTATTG3 ′  60 201 NM012766Reverse: 5 ′ TGCTTTTTGACCAGTGCCTGC3 ′ Delta1 Forward: 5 ′ CGACCTCGCAACAGAAAAC3 ′  56 397 NM032063Reverse: 5 ′ CGACCTCGCAACAGAAAAC3 ′ Hes1 Forward: 5 ′ GCTTTCCTCATCCCCAATG3 ′  56 224 NM024360Reverse: 5 ′ CGTATTTAGTGTCCGTCAGAAGAG3 ′ Crx Forward: 5 ′ CCTCACTATTCGGTCAATGCC3 ′  58 346 NM_021855Reverse: 5 ′ ATGTGCCTGCCTTCCTCTTC3 ′ Lef1 Forward: 5 ′ CACACAACTGGCATCCCTCATC3 ′  56 205 NM130429Reverse: 5 ′ TACACTCGGCTACGACATTCGC3 ′ Fzd1 Forward: 5 ′ TGGTTTCGTGTCGCTCTTCC 3 ′  60 223 XM216082 (continued on next page) 285  A.V. Das et al. / Developmental Biology 299 (2006) 283  –  302   poly- D -lysine, laminin coated 12-mm glass coverslips with E3 chick/PN1 rat retinal cells in 1% FBS for 5 – 7 days, across 0.4  μ m membrane (Millipore). At the end of culture, cells were fixed with 4% paraformaldehyde for 15 min at 4°Cfollowed by immunocytochemical analysis for various retinal neuronal markers.Media were changed every other day and neurospheres were allowed todifferentiate for 5 – 7 days. To obtain E3 chick retinal cells, fertilized hens eggs(SPAFAS, Wilmington, MA) were incubated in a humidified chamber at 37°Cfor 3 days. Embryos were staged according to Hamburger and Hamilton (1951).Retina were dissected out from eyes and dissociated into single cells as previously described (James et al., 2003).  Hoechst dye efflux assay Hoechst dye efflux assay was carried as previously reported (Ahmad et al.,2004; Bhattacharya et al., 2003). Briefly, dissociated retinal cells wereresuspended in Iscove's modified Dulbecco's medium (IMDM) at a concentra-tion of 10 6 cells/ml and incubated at 4°C overnight followed by staining withHoechst 33342 (4  μ g/ml) at 37°C for 60 min and sorted on a FACStar Plus (BDBiosciences, Lincoln Park, NJ) cell sorter. Hoechst dye was excited at 350 nm,and fluorescence was measured at two wavelengths using a 485 BP22 (485-nm bandpass filter) and a 675 EFLP (675-nm long-pass edge filter) optical filter (Omega Optical, Brattleboro, VT). The SP region was defined on the flowcytometer on the basis of its fluorescence emission in both blue and redwavelengths. Dead cells and debris were excluded by establishing a live gate onthe flow cytometer using forward versus side scatter. The sorted SP cells were processed for immunocytochemical and RT-PCR analyses. The specificity of theassay was ascertained by incubating cells with 150  μ M of verapamil during thestaining stage.  In vivo activation and enrichment of Müller cells PN21 rats were used for this experiment. The rats were anesthetized withketamine and xylazine and right eyes were injected with 0.5  μ l – 10  μ l of amixed solution of neurotoxins [kainate (40 mM) and NMDA (4 mM)], growthfactors (FGF2, 20 ng/eye; insulin, 1 μ g/eye) and BrdU (1 μ g/eye) by intravitrealinjection using a glass micropipette attached to a 10- μ l Hamilton syringe as previously described (Zhao et al., 2005; Chacko et al., 2003). The left eyes were injected with the vehicle and were used as controls. Rats were givenintraperitoneal injections of BrdU (0.12 mg/g body weight) everyday untilthe end of the experiment on 4th day. Eyes were enucleated. Some eyes were processed for immunohistochemical analysis and retina from the rest wassubjected to Hoechst dye efflux assay as described above. To examine the roleof Notch signaling on the in vivo activated Müller cells, DAPT (25  μ M), a  γ -secretase inhibitor (James et al., 2004), was injected along with the growth factors. To examine the involvement of Wnt signaling in the in vivo activationof Müller cells, Wnt2b and FzdCRD conditioned medium was injectedintravitreously. Wnt2b and FzdCRD conditioned medium was obtained bytransiently transfecting pEF-Wnt2b vector (Kubo et al., 2003) and Mfz8CRD-IgG (Hsieh et al., 1999) in 293 cells by using Effectene transfection kit (Qiagen). One day after transfection, cells were transferred to serum-freeDMEM/F-12, and the medium was collected after an additional 24 h. Controlconditioned medium was obtained from untransfected 293T cells. Tocompensate for the artifactual differences in levels of cell-specific transcriptsin the injured retina due to cell loss, examination of levels of transcripts by PCR was carried out after normalization of levels of beta actin transcripts in inuredand uninjured retina.  Brain and retinal transplantation Transplantation of GFP + cells from Müller cell-derived neurospheres wascarried out as previously described (Vicario-Abejon et al., 1995; Zhao et al.,2002). Briefly, to implant cells in the hippocampus of the PN2 rat pups ( n =4)animals were anesthetized by hypothermia. Each animal was placed on aminiaturized hypothermic stereotaxic instrument (Stoelting Co.). A hole wasdrilled in the skull using a dental burr at 1.7 mm lateral to midline and 1.2 mm posterior to bregma. A pulled glass pipette with an external diameter of  ∼ 70 μ mwas placed  ∼ 1.4 mm below dura to deliver 0.5  μ l of cell suspension over a period of 2 min. The surgical area was cleaned and the incision was closed withSuperglue. The animal was revived on a 37°C heating pad and returned to itsmother. Intravitreal transplantation of GFP + cells from Müller cell-derivedneurospheres cells or SP cells was performed in PN1 rats ( n =3) as previouslydescribed (Chacko et al., 2003). Briefly, animals were anesthetized with 20 – 25  μ l of a 1:1 dilution of ketamine (60 mg/ml) and xylazine (8 ng/ml), injectedintraperitoneally. A 30-gauge needle was inserted into the eye near the equator.The needle was retracted, and a glass micropipette connected to a 10  μ lHamilton syringe was inserted through the wound to deliver 50,000 – 100,000cells in a 1 – 2  μ l volume. Animals were sacrificed 1 week after transplantation.Eyes were enucleated and processed for immunohistochemical analysis. The SPcells were labeled with CFDA-SE as per manufacturer's instructions (Molecular Probes) for tracking purpose. In order to determine the lamina-specificincorporation of grafted cells, retinal sections (75 sections) from different transplantation experiments were examined and total cells integrated in a particular section were counted. Next, the number of cells incorporated in eachlamina was counted to determine the lamina-specific proportion of cells withrespect to total cell integrated.  Brain overlay culture To analyze whether Müller cells differentiate into physiologically activeneuronal cells in vivo, we carried out brain slice overlay culture according toa previously described method (Benninger et al., 2003; Polleux et al., 1998). Briefly, PN1 pups were sacrificed by decapitation, brains removed in ice-coldHBSS, and placed on a clean Whatman filter paper. A solution of low-melting agarose was poured on top of the brain and was allowed to solidify by placing the filter on a cooled surface. 300- μ m-thick coronal sections werecut using a mechanical tissue chopper (Stoelting Co.). Slices were collectedusing a flat brush and cultured in a 6-well plate on millicell filter (Millipore)covered with a film of slice culture medium (DMEM containing 25% HBSSand 10% FBS). Müller cell-derived neurospheres from GFP-transgenic micewere dissociated and resuspended in slice culture medium. Approximately1×10 5 cells were added onto and around the slices and cultured for 3 – 4 days. At the end of culture, the slices were analyzed for the presence of GFP + cells and used for electrophysiological and immunohistochemicalanalyses.Table 2 ( continued  ) Name Sequence Annealing temp (°C) Size (bp) Acc. NoReverse: 5 ′ AGGGCAAGGGATGGCATAAC3 ′ Fzd2 Forward: 5 ′ CGCTTCTACTTTCTTCACGGTCAC3 ′  58 211 NM172035Reverse: 5 ′ GCCTTCTTTCTTAGTGCCCTGC3 ′ Fzd4 Forward: 5 ′ TGCTTCATCTCTACCACCTTCACTG3 ′  60 282 NM022623Reverse: 5 ′ CAGAATAACCCACCAAATGGAGC3 ′ Ki67 Forward: 5 ′ GAGCAGTTACAGGGAACCGAAG3 ′  58 262 X82786Reverse: 5 ′ CCTACTTTGGGTGAAGAGGCTG3 ′  p27Kip1 Forward: 5 ′ GACTTTCAGAATCATAAGCCCCTG3 ′  58 261 NM031762Reverse: 5 ′ TGGACACTGCTCCGCTAACC3 ′ β -Actin Forward: 5 ′ GTGGGGCGCCCCAGGCACCA 3 ′  50 548 XM_037235Reverse: 5 ′  CTCCTTAATGTCACGCACGATTTC 3 ′ 286  A.V. Das et al. / Developmental Biology 299 (2006) 283  –  302   Electrophysiological analysis For electrophysiological studies, brain slices with GFP + cells from Müller cell-derivedneurosphereswereplaced inachamber,andperfusedon thestageof an upright, fixed-stage microscope (for electrophysiology, model BHWI;Olympus, Lake Success, NY; for imaging experiments, model E600FN; Nikon,Melville, NY) with an oxygenated solution containing NaCl, 140 mM; KCl,5 mM; CaCl 2 , 2 mM; MgCl 2 , 1 mM; HEPES, 10 mM; glucose, 10 mM (pH 7.4).Experiments were performed at room temperature. For whole-cell recording, patch pipettes were pulled on a vertical puller (model PB-7; Narishige, Tokyo,Japan) from borosilicate glass pipettes (1.2-mm outer diameter, 0.95-mm inner diameter; with internal filament (World Precision Instruments) and had tips of 1-to 2- μ m outer diameter with tip resistances of 6 to 12 M Ω . Pipettes were filledwith a bathing solution containing KCH 3 SO 4 , 98 mM; KCl, 44 mM; NaCl,3 mM; HEPES, 5 mM; EGTA, 3 mM; MgCl 2 , 3 mM; CaCl 2 , 1 mM; glucose,2 mM; Mg-adenosine triphosphate (ATP), 1 mM; guanosine triphosphate(GTP),1 mM; andreducedglutathione,1 mM (pH 7.2)(Das et al., 2005b; Jameset al., 2003). GFP-positive cells were identified under epifluorescence and patched for whole-cell recording.  Adenovirus infection To eliminate the immigrant astrocytes from the Müller cell culture, weinfected the purified Müller cells with Adgfap2Tk vector for 24 h and exposedthemtoGanciclovir(GCV; 150 μ g/ml) (Das etal., 2006;Vandier etal., 2000) for  another 24 h. GCV specifically eliminate Tk-expressing cells. Cells werecollected, washed and then cultured in the presence of FGF2 to generateneurospheres. Results  Enriched Müller cells display properties of NSCs To examine whether or not Müller cells possess NSC properties we carried out neurosphere assay (Doetsch et al.,1999; Laywell et al., 2000; Merkle et al., 2004) on Müller cells, enriched from rodent retina, ranging from PN10 to PN21. Weadopted a monolayer culture protocol that has been demon-strated to enrich Müller cells to the range of greater than 95%,free from neuronal, fibroblastic or astrocytic contaminations(Hicks and Courtois, 1990). In our hands, approximately 95%of cells in the monolayer culture were Müller cells as revealed by the proportion of cells expressing vimentin and glutaminesynthetase (GS) immunoreactivities (Figs. 1A –  N). To further ascertain the enrichment and purity of monolayer Müller cellculture, we examined the expression of cell-specific transcripts.RT-PCR analysis revealed that cells in culture expressed a battery of transcripts characteristic of Müller cells such as  GS,CRALBP, Vimentin, Clusterin and Carbonic anhydrase  (Black-shaw et al., 2004) (Fig. 1O). In contrast, transcripts corresponding to rod photoreceptors ( opsin ), bipolar cells( mGluR6  ), amacrine cells ( Syntaxin 1 ), retinal ganglion cells or RCGs (  Brn3b ), endothelial cells ( CD31 ) and retinal pigmentedepithelium (RPE)/pigmented ciliary epithelium ( tyrosinase )were not detected suggesting that the monolayer culture wasenriched for Müller cells and not contaminated with theabovementioned cells (Fig. 1P). However, low levels of transcripts characteristic of microglia (  Iba1 ) were detected,suggesting the presence of microglia in the enriched Müller cellculture. Subsequent immunocytochemical analysis revealed the presence of microglia in the monolayer culture in the range of 2 – 4%, which could be removed by immunopanning (see below).Having determined the purity of the culture, we first examined the proliferating and self-renewing abilities of enriched Müller cells. When these cells were cultured in the presence of FGF2 for 4 – 5 days, a small subset (0.5% – 1%) of cells proliferated to generate spheres of an average size of 200  μ M. The size of neurospheres was dependent on culturetime [Fig. 2A (5 days in culture) versus Fig. 2P (2 days in culture)]. The spheres displayed classic features of neuro-spheres; the majority (72.4±2.6%) of cells in spheresincorporated BrdU, attesting to their proliferative nature andexpressed general (Notch1; 83.0±3.0) and neural (Sox2:74.0±3.4; Nestin: 86.0±1.8, Musashi: 73.0±4.3) stem cellmarkers (Figs. 2A – F and G). The majority (88.0±1.5%) of  proliferating cells co-expressed Nestin and Notch, suggestingthat these cells did not represent distinct sub-populations of cells but rather a single population of neural progenitors, expressingmultiple neural progenitor markers (Figs. 2H – L). Although theabsence of cells expressing CD31, retinal neuron-specificmarkers and tyrosinase ensured that endothelial cells, retinalneurons, RPE/pigmented ciliary epithelium, respectively, werenot the source of neurospheres, a possibility remained that thesource could be the contaminating microglia. To examine this possibility, we purified microglia from retina (Roque andCaldwell, 1993) and cultured them identically like enrichedMüller cells in the presence of FGF2. At the end of 7 days inculture, neurospheres were not detected (Fig. 2M). In addition,we observed that enriched Müller cells, following the removalof microglia by immunopanning, still formed neurospheres,when exposed to FGF2 for 2 days (Fig. 2 N). Together, theseobservations suggested that enriched Müller cells and not contaminating microglia were the source of neurospheres. It could be argued that immigrant astrocytes, as a possible minor contaminant in the monolayer culture, could be the source of neurospheres. To address this issue, we adapted an approachof selectively ablating GFAP-expressing astrocytes in themonolayer culture, followed by examining the generation of neurospheres (Das et al., 2006; Morshead et al., 2003). Cells in the monolayer culture, before their exposure to FGF2, wereinfected with adenoviral vector   Adgfap2Tk   in which the HSV-Tk gene is driven by GFAP promoter. Following infectionand exposure to ganciclovir (GCV) for 24 h, cells werecultured in the presence of FGF2 for 1 – 2 days. GCV isconverted into a toxic triphosphate form in the presence of HSV-Tk, killing dividing cells by incorporating in replicatingDNA (Vandier et al., 2000). We observed the generation of neurospheres in culture of   Adgfap2Tk   infected cells and thosethat were infected with the control virus, providing evidenceagainst immigrant astrocytes as the source of neurospheres(Fig. 2O).To address the issue that neurospheres generated in bulk culture may not have clonal origin but arise due to cellaggregation, we co-cultured green [obtained from greenfluorescent protein (GFP)-expressing mice] and non-green(obtained from wild-type mice) enriched Müller cells in the presence of FGF2. The culture was carried out for 2 days since 287  A.V. Das et al. / Developmental Biology 299 (2006) 283  –  302
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks