Ex vivo expanded autologous limbal epithelial cells on amniotic membrane using a culture medium with human serum as single supplement

Please download to get full document.

View again

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.
 5
 
  Ex vivo expanded autologous limbal epithelial cells on amniotic membrane using a culture medium with human serum as single supplement
Share
Transcript
  Ex vivo  expanded autologous limbal epithelial cells on amniotic membrane usinga culture medium with human serum as single supplement q Aboulghassem Shahdadfar a , 1 , Kristiane Haug a , 1 , Meeta Pathak a , Liv Drolsum a , Ole Kristoffer Olstad b ,Erik O. Johnsen a , Goran Petrovski c , Morten C. Moe a , 2 , * , Bjørn Nicolaissen a , 2 a Center for Eye Research, Department of Ophthalmology, Oslo University Hospital and University of Oslo, Norway b Department of Medical Biochemistry, Oslo University Hospital, Norway c Department of Ophthalmology, Medical and Health Science Center, University of Debrecen, Hungary a r t i c l e i n f o  Article history: Received 9 August 2011Accepted in revised form 29 January 2012Available online 7 February 2012 Keywords: stem celllimbal epitheliumexplant culturehuman serumcornealimbal stem cell de 󿬁 ciency a b s t r a c t In patients with limbal stem cell de 󿬁 ciency (LSCD), transplantation of   ex vivo  expanded human limbalepithelial cells (HLECs) can restore the structural and functional integrity of the corneal surface.However, the protocol for cultivation and transplantation of HLECs differ signi 󿬁 cantly, and in mostprotocols growth additives such as cholera toxins, exogenous growth factors, hormones and fetal calf serum are used. In the present article, we compare for the 󿬁 rst time human limbal epithelial cells (HLECs)cultivated on human amniotic membrane (HAM) in a complex medium (COM) including fetal bovineserum to a medium with human serum as single growth supplement (HSM), and report on our  󿬁 rstexaminations of HLECs expanded in autologous HSM and used for transplant procedures in patients withLSCD. Expanded HLECs were examined by genome-wide microarray, RT-PCR, Western blotting, and forcell viability, morphology, expression of immunohistochemical markers and colony forming ef  󿬁 ciency.Cultivation of HLECs in HSM produced a multilayered epithelium where cells with markers associatedwith LESCs were detected in the basal layers. There were few transcriptional differences and comparablecell viability between cells cultivated in HSM and COM. The  p63  gene associated with LESCs wereexpressed 3.5 fold more in HSM compared to COM, and Western blotting con 󿬁 rmed a stronger p63 a band in HSM cultures. The cornea-speci 󿬁 c keratin CK12 was equally found in both culture conditions,while there were signi 󿬁 cantly more CK3 positive cells in HSM. Cells in epithelial sheets on HAMremaining after transplant surgery of patients with LSCD expressed central epithelial characteristics, anddissociated cells cultured at low density on growth-arrested  󿬁 broblasts produced clones containing21    12% cells positive for p63 a  ( n  ¼  3). In conclusion, a culture medium without growth additivesderived from animals or from animal cell cultures and with human serum as single growth supplementmay serve as an equivalent replacement for the commonly used complex medium for  ex vivo  expansionof HLECs on HAM.   2012 Elsevier Ltd. All rights reserved. 1. Introduction Slow-cycling limbal epithelial stem cells (LESCs) found withinthe basal cell layer of the limbal epithelium are responsible forcontinuously renewing the entire corneal epithelium, and thusensuring a transparent cornea (Ahmad et al., 2010; Cotsarelis et al., 1989; Davanger and Evensen, 1971; Dua et al., 2005; Majo et al., 2008). When the limbal area is partially or totally damaged, lim-bal stem cell de 󿬁 ciency (LSCD) occurs, a condition characterized bycorneal ingrowth of conjunctival epithelium, neovascularization,recurrent epithelial defects, scarring, chronic in 󿬂 ammation, painand reduced vision (Tseng, 1996). In such cases, grafting of limbaltissue or  ex vivo  expanded human limbal epithelial cells (HLECs)can restore the structural and functional integrity of the cornealsurface (Notara et al., 2010; Shortt et al., 2007). While the use of  autologous limbal fragments depends on a healthy contralateraleye,  ex vivo  autologous expansion of HLECs can be used to treat q Grant information: Supported by the Research Council of Norway, the Blind-emissionen IL, the Norwegian Association of the Blind and Partially Sighted, theFaculty of Medicine University of Oslo and Oslo University Hospital. *  Corresponding author. Center for Eye Research, Department of Ophthalmologyand Norwegian Center for Stem Cell Research, Oslo University Hospital, Kirkeveien,0407 Oslo, Norway. Tel.:  þ 47 22 11 80 80; fax:  þ 47 22 11 99 89. E-mail address:  m.c.moe@medisin.uio.no (M.C. Moe). 1 These authors contributed equally to this work. 2 Co-senior authors. Contents lists available at SciVerse ScienceDirect Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer 0014-4835/$  e  see front matter    2012 Elsevier Ltd. All rights reserved.doi:10.1016/j.exer.2012.01.013 Experimental Eye Research 97 (2012) 1 e 9  patients with bilateral disease, as long as some healthy limbaltissue is present. Alternatively, epithelial cells from other sourcessuch as the conjunctiva (Tanioka et al., 2006), the oral mucosa(Nishida et al., 2004), or allogenic HLECs from a cadaveric or living relative donor can be transplanted (Shortt et al., 2007), the latter requiring long-term postoperative immunosuppression (Dayaet al., 2005; Shortt et al., 2007). Since Pellegrini et al. (1997) published the clinical trans-plantation of   ex vivo  expanded HLECs in two patients, this tech-nique has become a routine treatment for ocular surfacereconstruction in patients with LSCD in several clinics (Ahmadet al., 2010; Sangwan et al., 2006; Shortt et al., 2007). However, the protocol for cultivation and transplantation differ signi 󿬁 cantly.These protocols include the use of explants - or cell culture, the useof mouse 3T3 feeder cell layer, as well as different carriers for cellexpansion and transplantation (Di Girolamo et al., 2009;Mariappan et al., 2010; Pellegrini et al., 2010; Shortt et al., 2007). The use of HAM as substrate has been suggested to be bene 󿬁 cialsince it is easily obtained, and serves as a strong biodegradable,hypoimmunogenic and relatively easily manipulated carrier. Inaddition, it facilitates the growth and expansion of HLECs withoutthe need of 3T3 feeder cells and may have a positive in 󿬂 uence onthe long-term survival of LESCs (Lee and Tseng,1997; Meller et al., 2002; Shortt et al., 2007, 2008). The composition of the medium is also essential for the cultureof HLECs. To achieve successful cell culturing conditions, fetalbovine serum (FBS), in addition to various hormones and growthfactors,hasbeenincludedinmostculturemethodsfortreatmentof LSCD (Shortt et al., 2007). However, these animal derived products carry a potential risk of transmission of animal viruses, prions andforeign proteins that may initiate xenogeneic immune responses.Therefore, using a culture medium completely free of animalproducts could be bene 󿬁 cial.  Ex vivo , 1 e 10% of human serum issuitable for cultivation of HLECs (Di Girolamo et al., 2007;Mariappan et al., 2010; Zakaria et al., 2010), and transplantation of  HLECs expanded on HAM in an autologous serum-based mediahave been shown to be successful in treatment of LSCD (Kolli et al.,2010; Meller et al., 2010; Nakamura et al., 2006; Shimazaki et al., 2007). However, in these studies an epithelial media containingvarious growth factors, cholera toxin and hormones were used.Only one group that we are aware of has previously used a culturemedium with autologous serum as single growth supplement, andthey applied a contact lens-based technique (Di Girolamo et al.,2009). In the present study,  ex vivo  expanded HLECs on HAM ina commonly used complex medium containing FBS and other non-human derived products is compared to a culture medium withhuman serum as single growth supplement. We also report on our 󿬁 rst examinations of   ex vivo  expanded autologous HLECs main-tained on HAM in medium with autologous serum and used intransplant procedures of patients with LSCD. 2. Materials and methods  2.1. Preparation of human serum AllreagentswerepurchasedfromSigma e Aldrich(St.Louis,MO)unless otherwise stated. Thirty ml blood was obtained from each 4healthy voluntary donors (comparative  ex vivo  studies) or frompatients undergoing clinical transplantations of HLECs. From eachdonor, suf  󿬁 cient venous blood was drained into 10 ml vacutainertubes without anticoagulants (BD, Plymouth, U.K.) and allowed toclot.Subsequently,thebloodwascentrifugedat1800  gfor15minat 4   C. The serum from each donor was collected and passedthrough 0.22  m m pore size  󿬁 lters and aliquots of the sterile serumwere stored at   20   C. For the comparative  ex vivo  studies, equalvolume of serum from each of the 4 donors was pooled.  2.2. Culture medium 2.2.1. Human serum medium (HSM) DMEM/F12 (Invitrogen, Carlsbad, CA), Penicillin/Streptomycin(100 U/ml), amphotericin B (2.5  m g/ml) and 10% pooled humanserum (comparative  ex vivo  studies using HLEC derived from donoreyes) or 10% autologous serum (using HLECs derived from patientswith LSCD).  2.2.2. Complex medium (COM) DMEM/F12 (Invitrogen), Penicillin/Streptomycin (100 U/ml),amphotericin B (2.5  m g/ml), 5% FBS, EGF (2 ng/ml, R&D Systems,MN), ITS (insulin 5  m g/ml, transferrin 5  m g/ml and sodium selenite5 ng/ml), cholera toxin A (30 ng/ml, Biomol International, LP),dimethylsulfoxid (DMSO, 0.5%), hydrocortisone (15  m M), genta-micin (50  m g/ml).  2.3. Explant culture All experiments were conducted in accordance with theDeclaration of Helsinki and all tissue harvesting was approved bythe Local Committees for Medical Research Ethics. For thecomparative  ex vivo  studies, human corneoscleral tissue wasobtained from limbal rings of cadaveric donors, available afterpenetrating keratoplasty, and preserved in Optisol-GS (Baush&-Lomb Inc., NY) at 4   C. Each ring ( n ¼ 5) was divided in 8 samples.Corneal limbal epithelial tissue (1.5    2 mm) from patients withLSCD scheduled for transplant surgery were derived from healthylimbal areas in the contralateral or in the same eye ( n  ¼  3). Thetissue was treated with 1.1 U/ml Dispase II in Mg 2 þ and Ca 2 þ -freeHanks ’  balanced salt solution (HBSS) at 37   C for 10 min, thereafterrinsed in HSM or COM (Meller et al., 2002; Raeder et al., 2007). Human amniotic membranes (HAM) were preserved according tothe method described by Lee and Tseng (Lee and Tseng, 1997). A formal Institutional Review Board approval and informed consentfrom the donor of the HAM were obtained. The HAM was cry-opreserved in 50% (v/v) glycerol and media. After thawing, a pieceof the HAM was placed on a Netwell plate and sutured in sixcorners. The limbal biopsy was placed with the epithelium facingdown on the basement membrane surface of the HAM and allowedto attach. All cultures were incubated at 37   C and 5% CO 2 . Theculture mediumwas changed every 2 e 3 days. For the comparative ex vivo  experiments, four pieces from each of the Eye Bank donoreyes were cultured inparallel in either HSM or COM. Samples frompatients with LSCD were cultured in medium using autologousserum.  2.4. Colony forming assay Colony forming assays were performed by dissociating HAM-attached HLECs and seeding at clonal concentrations(3   10 3 cell/cm 2 ) (Kolli et al., 2010) on a feeder layer of CRL2429 human  󿬁 broblast (ATCC, Manassas, VA) growth-arrested by 40 Gyirradiation.Colonyformationwasmonitoreddailyandstainedwith0.5% Rhodamine or immunostained after 10 days of culture.  2.5. Assay for viability/cell death analysis Cell viability/death was assessed by the Annexin-V-FITCApoptosis Detection Kit (MBL, Woburn, MA) according to manu-facturer ’ s recommendations; proportion of stained Annexin-V  þ and Annexin-V  þ /Propidium iodide (PI) þ cells was determined by  A. Shahdadfar et al. / Experimental Eye Research 97 (2012) 1 e 9 2  󿬂 uorescence activated cell sorter (FACS) analysis on BD Bioscience(San Diego, CA)  󿬂 ow cytometer (Petrovski et al., 2007).  2.6. Immunohistochemistry Samples were  󿬁 xed in 4% fresh formaldehyde and embedded inparaf  󿬁 n. Three micrometer sections were cut and stained usingLabVision Autostainer360 (Lab Vision Corporation, VT) and visu-alized using a standard peroxidase technique (UltravisionOne HRPsystem). The following monoclonal antibodies were used; Cytko-keratin 3 (CK3, 1:500, ImmuQuest), Cytokeratin 12 (CK12, 1:400,Santa Cruz Biotechnology, CA), p63 a  (1:500, Santa Cruz Biotech-nology), Ki-67 (1:200, Thermo Scienti 󿬁 c), Vimentin (1:200, Neo-Markers) and mouse anti-human ABCG2 (1:80). The positiveimmunoreaction of the primary antibody was detected bya secondary antibody conjugated with peroxidase-labeled polymerwith diaminobenzidine (DAB) (Utheim et al., 2009) or the  󿬂 uores-cent markers Cy3 (1:1000) and Alexa Fluor 488 (1:500, Invitrogen).Hoechst (1:500, Invitrogen) was used for nuclear staining. Sectionswere also stained with hematoxylin & eosin (H&E) for morpho-logical examination.  2.7. Real-time RT-PCR TotalRNAwasextractedfromcellsusingQiazolreagent(Qiagen,Hilden, Germany). Following DNase treatment (Ambion, Austin,TX), RNA was quanti 󿬁 ed by spectrophotometry (Nanodrop, Wil-mington, DE). Reverse transcription (RT) was performed using theHigh Capacity cDNA Archive Kit (Applied Biosystems, Abingdon,U.K.) with 200 ng total RNA per 20  m l RT reaction. ComparativeRelativeQuanti 󿬁 cationwas performed using the StepOnePlus Real-Time RT-PCR system (Applied Biosystems) and Taqman GeneExpression assays following protocols from the manufacturer(Applied Biosystems) (Table 1). The data were analyzed by 2  DD Ct method as the fold change in gene expression and normalized toGAPDH as endogenous reference gene and relative to COM, whichwas arbitrarily chosen as calibrator and equals one. All sampleswere run in triplicates (each reaction: 2.0  m l cDNA, total volume20  m l). The thermo cycling parameters were 95   C for 10 min fol-lowed by 40 cycles of 95   C for 15 s and 60   C for 1 min.  2.8. Affymetrix gene expression pro  󿬁 ling  100ngoftotalRNAwassubjectedtotheGeneChipHTOne-CyclecDNA Synthesis Kit and GeneChip  HT IVT Labeling Kit followingthe manufacturer ’ s (Affymetrix, Santa Clara, CA) recommendedprotocol for whole transcriptgeneexpression analysis. Labeled andfragmented single-stranded DNA was hybridized to the AffymetrixGeneChip Human Gene 1.0 ST Arrays. The signal intensities weredetectedbyHewlettPackardGeneArrayScanner30007G(HewlettPackard,PaloAlto,CA).TheCEL  󿬁 leswereimportedintoArrayAssistExpression Software (v5.2.0; Iobion Informatics LLC, La Jolla, CA)and normalized using the RMA (Robust Multichip Average)algorithm in Array Assist to calculate relative signal values for eachprobe set (Utheim et al., 2009).  2.9. Western blotting  Total protein was prepared from frozen samples (Invitrogen).Proteins were mixed in sample-loading buffer (Tris buffer pH 6.8,2% SDS, 10% sucrose, and protease inhibitors), boiled for 10 min,centrifuged, and protein concentration in the clari 󿬁 ed lysates wasdetermined using the BCA Protein Assay kit (Termo Fisher Scien-ti 󿬁 c, Rocford, IL). Equal amounts of protein in cell lysates wereseparated by 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions and electrotransferred to a poly-vinylidine di 󿬂 uoride membrane (Millipore, Billerica, MA) themembraneswereblockedwith5%skimmilkinPBScontaining0.1%Tween 20, incubated with the primary anti-human ABCG2 anti-body (Abcam, Cambridge, U.K.) and anti-human p63 a  antibody(Santa Cruz) for 2 h, washed 3 times, and incubated respectivelywith the anti-rabbit or anti-mouse IgG conjugated to horseradishperoxidase for 1 h. Finally, the membranes were washed 3 timesand protein bands were detected using enhanced chem-illuminescence reagent (Amersham Biosciences, Sweden). Themembrane was stripped for re-blotting with  b -actin antibody ascontrol.  2.10. Transmission electron microscopy (TEM) The HAM-attached HLECs were  󿬁 xed in 2% glutaraldehyde incacodylatebuffer(pH7.4)overnightat4  C,post 󿬁 xedin1%osmiumtetroxide, and dehydrated through a graded series of ethanol up to100%. The tissues were then immersed in propylene oxide for20 min and embedded in Epon (Electron Microscopy Sciences,Hat 󿬁 eld, PA). Ultra-thin sections (60 e 70 nm thick) were cut ona Leica Ultracut Ultramicrotome UCT (Leica, Wetzlar, Germany),stained with uranyl acetate and lead citrate and examined usinga Tecnai12 transmission electron microscope (Phillips, Amsterdam,the Netherlands) (Moe et al., 2009).  2.11. Statistics The results are presented as mean    SD. Differences betweengroups were tested with one-sample  t  -tests or two-tailed inde-pendent sample  t  -tests. Data that were not normally distributedwere normalized by log transformation. The signi 󿬁 cance level wasset to  p < 0.05, and data were analyzed using SPSS Version 14.0. 3. Results  3.1. Comparison of ex vivo expansion in human serum medium andcomplex medium A total of 40 HCLEC explants on HAM from 5 donors werecultured in parallel in either HSM or COM (20 explants in eachgroup). Under both culture conditions, the epithelial cells migratedfrom the limbal edges to form an epithelial sheet of 1 e 5 cell layerswith basal column-shaped cells and super 󿬁 cial  󿬂 attenedsquamous-like cells.  3.1.1. Comparative transcriptional pro  󿬁 ling  To determine differences between HLECs cultured in HSM andCOM at the transcription level, we compared the global geneexpression pro 󿬁 le using microarray from three different donors.Intensitypro 󿬁 lesofthelog 2 transformedsignalvaluesofthe28,869transcripts (vertical columns) are shown in Fig. 1A. Of these, 188transcripts were differentially expressed more than 2 fold change  Table 1 Primers used for Real-time RT-PCR.Gene name Gene symbol  Alias  Taqman assay IDATP-binding cassette sub family G2  ABCG2 BCRP   HS01053790_m1Cytokeratin 3  KRT3 CK3  HS00365080_m1Gap junction protein  a 1, 43 Kda  GJA1 CX43  HS00748445_s1Occludin  OCLN   e  HS00170162_m1Glyceraldehyde-3-phosphatedehydrogenase GAPDH GAPD  HS99999905_m1Tumor protein P63  TP63 P63  HS00978340_m1  A. Shahdadfar et al. / Experimental Eye Research 97 (2012) 1 e 9  3  ( n  ¼  3,  p  <  0.01) and 85 transcripts exhibited more than 3 foldchange (  p  <  0.01) between the two culture conditions. This indi-cates relatively low transcriptional differences between the HSMand COM culture conditions. Table 2 shows 40 top genes compar-atively expressed in HLECs cultured in HSM compared to COMwhich were differentially expressed more than 5 fold change(  p  <  0.01). These genes were mostly involved in extracellularmatrix and protein binding activities, while a few genes wererelated to cell stemness and control of cornea integrity such as  ALDH1A1  and  DKK2 , respectively (Ahmad et al., 2008;Mukhopadhyay et al., 2006).  3.1.2. Cell viability/death pro  󿬁 ling  The viability of HLECs cultured in HSM and COM was alsocomparable (HSM 79.7    8.1 and COM 84.9    5.1) with a compa-rable but small percentage of cells undergoing early or reversible(annexin V  þ ) apoptosis (HSM 12.0    3.9 and COM 4.1   1.1) andsimilarly small percentage undergoing late or irreversible (annexinV  þ /PI þ ) apoptosis or secondary necrosis (HSM 7.8    3.7 and COM11.0   4.2) (Fig.1D,  n ¼ 3,  p > 0.05).  3.1.3. Expression of stem cell  e  and differentiation associatedmarkers Next, we performed Real-time RT-PCR, western blot andimmunohistochemicalanalysisof selected markersassociated withstemness and differentiation in HLECs. Even though no speci 󿬁 cmarkers for LESCs have been identi 󿬁 ed ( Joseph et al., 2004;Schlotzer-Schrehardt and Kruse, 2005), the ATP-binding cassettetransporter G2 (ABCG2), which is the molecular determinant of theside population phenotype (Zhou et al., 2001), is currently one of  the leading candidate LESC markers (Watanabe et al., 2004).In addition, the transcription factor p63 (Yang et al., 1999), and inparticular the isoform  D Np63 a  that is by far the most abundant inthe corneal epithelium (Robertson et al., 2008), has been linked tostemness and also clinical outcome after transplantation of HLECs(Rama et al., 2010; Schlotzer-Schrehardt and Kruse, 2005). We found that there were a tendency for higher  ABCG2  expression(  p ¼ 0.18) and a 3.5 fold increased  p63  expression after cultivationin HSM than in COM, respectively ( n  ¼  5) (Fig. 1B). Western blotresults con 󿬁 rmed an increase in p63 a  protein expression of explants cultured in HSM compared to COM (Fig. 1C). Immuno-histochemical analysis revealed abundant ABCG2 staining of theplasma membrane in both HSM and COM cultures, and eventhough there were a signi 󿬁 cant increase in the p63 a  mRNAexpression, no signi 󿬁 cant difference in the number of cells withnuclearp63 a stainingwasfound(65  22and79  13%ofHSMandCOM cultures, respectively,  n  ¼  3) (Fig. 2). Immunohistochemicalanalysisof theintermediate 󿬁 lamentVimentin,thatalsoisfoundinLESCs (Schlotzer-Schrehardt and Kruse, 2005), showed equalexpressionpatterns of basally located cells co-stained with p63 a  inthe two culture conditions (Fig. 2), and the proliferation marker Ki-67 did not indicate any major differences in the proliferativecapacityofHLECscultivatedinHSMandCOM,withaKi-67indexof 28  4% and 28  8%, respectively ( n ¼ 3) (Fig. 2).Of the differentiation-associated markers (Schlotzer-Schrehardtand Kruse, 2005), Real-time RT-PCR showed a tendency for thatboth the gap-junction marker  connexin 43  (  p ¼ 0 . 09 ) and the tight- junction marker  occludin  (  p  ¼  0.13) were upregulated in cultureswith HSM compared to COM, however these analysis did not reachstatistical signi 󿬁 cance (Fig. 1B,  n  ¼  5). While the amount of cellspositive for the differentiation marker cytokeratin 12 (CK 12) wasnot statistically different between HSM and COM (48    21% and54  23%,respectively, n ¼ 3),CK3positivecellswerealmostabsentin the COM cultures (2    4%), whereas 13    2% of the cells in the Fig.1.  Differential gene and protein expression levels of human limbal epithelial cells (HLECs) from a validated microarray study using cells from three different donors which werein parallel expanded in a culture mediumwith pooled human serum as single supplement (HSM) or a complex medium (COM) containing fetal bovine serum and other non-humanproducts. Intensity pro 󿬁 les of the log 2  transformed signal values of the 28,869 transcripts (vertical columns) are shown. Red and blue colors indicate high and low expression,respectively (A). Real-time RT-PCR analysis of the expression of selected genes associated with stemness (  ABCG2  and  p63 ) and differentiated corneal epithelial cells ( connexin43 ,  CK3 and  occludin ) in HLECs cultivated in HSM relative to COM, which was arbitrarily chosen as calibrator and equals one (B). Western blotting of ABCG2 and p63 a  in HSM andCOM explant cultures. The membrane was stripped for re-blotting with  b -actin antibody as control(C). Representative cell viability/death analysis of the HLECs grown in HSM andCOM  e  distribution of viable, annexin V  þ and annexin V  þ /propidium iodide þ cells after 2 weeks of cultivation (D).  A. Shahdadfar et al. / Experimental Eye Research 97 (2012) 1 e 9 4  HSM cultures, mostly in super 󿬁 cial layers, stained for this cornealepithelial marker ( n ¼ 3,  p < 0.05) (Fig. 2) even thoughRT-PCRdatadid not show any upregulation of CK3 in HSM (Fig.1B).  3.2. HLECs from patients with LSCD expanded ex vivo in autologoushuman serum medium and used for transplant surgery Two weeks after the autologous limbal biopsy, the HLECsexpanded in autologous HSM on HAM had grown to form a sheetcovering most of the culture plate. Examination of epithelial sheetson HAM remaining after transplant procedures revealed that cellsexpanded in autologous HSM expressed central epithelial charac-teristics, including  󿬂 at super 󿬁 cial cells, basal cells with a morecuboidal shape attached to the HAM basement membrane andnumerous desmosomal junctions between adjacent epithelial cells(Fig.3A e C).Inaseparateexperiment,wetestedwhetherexpandedHLECsretainedapopulationofcolonyformingcells.HAM-attachedHLECs were dissociated and cultivated at low density on growth-arrested human  󿬁 broblasts. Of the three epithelial sheets tested,epithelial clones (Kolli et al., 2010; Pellegrini et al., 1999) with a smooth outline appeared in all cases after 10 days of culture(Fig. 3D). In these clones, 21   12% ( n  ¼  3) of the cells were p63 a positive (Fig. 3E). 4. Discussion In establishing tissue equivalents for transplantation, the idealmethod should 1) be approved and safe with respect to diseasetransmission and 2) be able to recapitulate the tissue of srcin afterintegration. For the corneal epithelium, the latter should includeboth LESCswith abilityofself-renewaland targeted differentiation,as well as differentiated epithelial cells able to protect the ocularsurface (Rama et al., 2010; Schlotzer-Schrehardt and Kruse, 2005; Shortt et al., 2008). Our culture medium uses human serum assingle growth supplement, and is free of both animal derivedproductsandothergrowthsupplementssuchasexogenousgrowthfactors, hormones and cholera toxin. The present article indicatesthat HLECs can be expanded on HAM  ex vivo  using this novelmethod without losing the potential to generate a healthy epithe-lium. Although there are no exact ways to identify LESCs at present(Lyngholmetal.,2008;Robertsonetal.,2008;Schlotzer-Schrehardt andKruse,2005),therewereseveralindicationsofLESCspresentinthe HLECs expanded with the current protocol, including 1)expression of markers found in LESCs such as p63 a , ABCG2 andVimentin, 2) a multilayered epithelium with  󿬂 attened apical cellsand basal cuboidal-shaped cells attached to the HAM basementmembranewith intercellularjunctional complexes, and 3) retainedcolony forming capacity.  Table 2 Mosthighlyoverexpressedtranscripts(foldchange  5, and  p < 0.01,  n ¼ 3)ofhumanlimbal epithelial cells(HLECs) culturedwith humanserumassingle growth supplement(HSM) compared to a complex medium (COM) containing fetal bovine serum and other non-human products. If a gene was represented with more than one probe inAffymetrixgeneexpressionpro 󿬁 linglist,theaverageofdifferentialexpressionwasselected.ValuesrepresentmedianregulationofgeneexpressioninHSMculturescomparedto COM.Gene symbol Gene description Fold change Regulation Molecular function POSTN   Periostin speci 󿬁 c factor 59 Up Protein binding TAGLN   Transgelin 23 Up Actin binding BGN   Biglycan 20 Up Extracellular matrix constituent COL3A1  Collagen type III 20 Up Extracellular matrix constituent COL1A2  Collagen type 1 19 Up Extracellular matrix constituent SPP1  Secreted phosphoprotein1 15 Down Cytokine activity CDH11  Cadherin 11 13 Up Calcium ion binding TNC   Tenascin C 11 Up Receptor binding DCN   Decorin 11 Up Protein binding COL6A3  Collagen type VI 11 Up Serine inhibitor PDGFRA  PDGF receptor 11 Up Nucleotide binding IGFBP5  IGF- binding protein 5 11 Up Insulin-like growth factor binding COL1A1  Collagen, type I, alpha 1 11 Up Extracellular matrix constituent FABP4  Fatty acid binding protein 10 Up Transporter activity VCAN   Versican 10 Up Calcium ion binding C1S   Complement component 9 Up Rhodopsin-like receptor activity SULF1  Sulfatase 1 8 Up Arylsulfatase activity THBS2  Thrombospondin 2 8 Up Cell adhesion FBN1  Fibrillin 1 8 Up Transmembrane receptor activity GLIPR1  GLI pathogenesis-related1 8 Up  e CCDC80  Coil domain containing 80 8 Up  e MFAP5  Fibrillar associated protein 5 8 Up Extracellular matrix constituent CDH2  Cadherin 2 (N-cadherin) 8 Up Calcium ion binding binding PRRX1  Paired related homeobox 1 8 Up Transcription factor activity RGS4  Regulator of protein signal 4 7 Up Signal transducer activity  AEBP1  AE binding protein 1 7 Up Transcription factor activity FAP   Fibroblast activation protein 7 Up Metalloendopeptidase CCL2  Chemokine ligand 2 7 Up G-protein-coupled receptor COL12A1  Collagen, type XII, alpha 1 7 Up Structural molecule activity  AMTN   Amelotin 7 Up Protein binding COL5A1  Collagen type V 7 Up Heparin binding DKK2  Dickkopf homolog 2 7 Up Multicellular development FBLN5  Fibulin 5 7 Up Transmembrane receptor activity HMCN1  Hemicentin 1 7 Up Transmembrane receptor  ACTG2  Actin gamma 2 6 Up Nucleotide binding MMP13  Matrix metallopeptidase 6 Up Metalloendopeptidase activity LUM   Lumican 6 Up Extracellular matrix constituent TGFB2  Transforming growth factor 2 5 Up Growth factor activity THY1  Thy-1 cell surface antigen 5 Up Protein bindingALDH1A1 Aldehyde dehydrogenase 1 5 Down Cellular enzyme activity  A. Shahdadfar et al. / Experimental Eye Research 97 (2012) 1 e 9  5
Related Search
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
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x