Presence of Xenogenic Mouse RNA in RPE and IPE Cells Cultured on Mitotically Inhibited 3T3 Fibroblasts

Abstract

Purpose.: Mitotically inhibited 3T3 fibroblasts are used as feeder layers to culture a variety of cells. However, transplantation of human cells cultured on mitotically arrested mouse cells poses potential risks, such as disease transfer and contamination with 3T3 cells. Bovine RPE and IPE cells were cultured on mitomycin-treated 3T3 fibroblasts, to examine cell characteristics and contamination by 3T3 products.

Methods.: IPE or RPE cells cultured on mitomycin-treated 3T3 fibroblasts were evaluated for adhesion, morphology, and tight junction formation by microscopy and immunohistochemistry. ROS phagocytosis was used to examine functional activity. Gene expression was evaluated by quantitative real-time PCR.

Results.: In the presence of 3T3 fibroblasts, primary IPE and RPE cells adhere, spread and acquire a hexagonal shape within 12 hours. When cultured on 3T3 fibroblasts, IPE and RPE cells exhibited stable expression of pigment epithelial genes, but expression of mouse collagen type I was also observed.

Conclusions.: Culturing IPE and RPE cells on mitomycin-treated 3T3 fibroblasts resulted in rapid adhesion and growth of primary pigment cells. However, the presence of potentially hazardous xenogeneic mRNA of mouse origin in the cultures limits the use of these cells for transplantation.

Since the 1970s, numerous investigators have shown that feeder cells, such as lethally treated 3T3 fibroblasts, can have profound effects on the growth characteristics of many cells in culture (e.g., epidermal keratinocytes, ^{1 –4} corneal epithelial cells, ^{5 –8} endothelial cells, ^9,10 enamel organ epithelial cells, ¹¹ salivary-gland cells, ^12,13 and middle ear epithelial cells ¹⁴ ). Cultivation of human keratinocytes and corneal epithelial cells on feeder layers of growth-inhibited mouse fibroblasts in FCS-supplemented medium is the standard culture technique to obtain sufficient numbers of keratinocytes for skin grafts ^4,15,16 and corneal epithelial cells for corneal reconstruction. ^{8,17 –20} The use of pigment epithelial cells has also been suggested for tissue engineering, ^{21 –26} as well as for gene therapeutic approaches ^{27 –30} to treat degenerative diseases such as age-related macular degeneration. 

RPE cells in culture are also influenced by the presence of other cells. ^{31 –33} For example, McKay and Burke ³² described increased proliferative activity by primary RPE cells when co-cultured with irradiated living cells. With passage, cultured RPE cells lose many differentiated characteristics. Therefore, it is difficult to determine whether results obtained with late-passage RPE cells or RPE cell lines represent the in vivo properties of RPE cells or the acquired properties characteristic of cultured RPE cells. Ideally, for the study of differentiated characteristics and for use in subretinal transplantation to treat degenerative diseases, IPE or RPE cells would be best expanded in primary cultures. A method that promotes fast adhesion and growth of differentiated primary IPE and RPE cells would be advantageous, especially for transplantation into the subretinal space, to ameliorate vision loss in diseases in which RPE cells degenerate. In this study, expansion of IPE and RPE cells on mitomycin-treated 3T3 fibroblasts resulted in primary cells that retain their functional and differentiated characteristics. However, even though mitomycin-treated 3T3 cells degenerate within 10 days in culture, products of the xenogeneic mouse fibroblasts persist in the IPE and RPE cell cultures, suggesting that transplantation of cells expanded on feeder layers of mouse fibroblasts could lead to immunologic rejection, graft failure, and possible disease transmission. 

Material and Methods

3T3 Fibroblasts

NIH-3T3 fibroblasts, derived from Swiss mouse embryos (DSMZ No: ACC 59), were cultured in Dulbecco's MEM/HAM's F-12 (DMEM/F12) containing 3.15 g/L glucose (Biochrom, Berlin, Germany) and supplemented with 10% fetal bovine serum (FBS; PAA Laboratories, Pasching, Austria), 80 U/mL penicillin, and 80 U/mL streptomycin (Lonza, Basel, Switzerland). The cells were passaged before reaching confluence at a ratio of 1:10 using 0.05% trypsin-0.02% EDTA (PAA Laboratories). Cell cultures were maintained at 37°C in a humidified atmosphere of 95% air and 5% CO₂. Conditioned medium was obtained from subconfluent 3T3 fibroblasts after 72 hours of culture. 

To determine the mitomycin concentration necessary to inhibit mitosis, we seeded 3T3 cells into 96-well plates at a concentration of 200,000 cells/mL, allowed them to attach for 24 hours, and then treated them with mitomycin (Medac GmbH, Hamburg, Germany) at a concentration of 10 to 100 μg/mL for 2 hours. The cells were then washed twice with balanced salt solution and cultured in complete medium for 10 days. 

Cell proliferation and viability were monitored with crystal violet every day beginning at day 2 of culture. Briefly, the medium was removed and 100 μL crystal violet solution (0.5% in 20% methanol) was added. After 10 minutes the cells were rinsed with water. After overnight drying, 50 μL sodium citrate solution (10 mM in 50% ethanol) was added to each plate and the absorbance measured at 595 nm with an ELISA reader. Each concentration was tested in 10 individual wells in triplicate experiments. 

Isolation of Iris Pigment Epithelial Cells

Porcine and bovine eyes were obtained from a local abattoir and brought to the laboratory within 2 hours of death. For isolation of IPE cells, the iris was dissected with scissors from the ciliary body. After incubation of the iris in 0.25% trypsin+0.02 g EDTA for 10 minutes, the IPE cells were isolated by gently brushing the posterior surface of the iris with a fire-polished glass spatula. The detached cells were centrifuged at 1000 rpm and the cell pellet was suspended in DMEM/F12 medium supplemented with 10% FBS, penicillin (80 U/mL), and streptomycin (80 U/mL). The cell suspension was plated into tissue culture dishes for proliferation or on chamber slides for immunohistochemistry at a density of 112,500 cells/cm². The cultures were maintained at 37°C in a humidified atmosphere of 95% air and 5% CO₂, and the medium was changed twice weekly. 

Isolation of RPE Cells

For RPE isolation, the anterior segment was cut approximately 3.5 mm posterior to the limbus, the vitreous and retina were removed, and the posterior eye cup was filled with 0.25% trypsin + 0.02 g EDTA and incubated at 37°C. After 10 minutes, the trypsin solution was removed, and the RPE dislodged by trituration in complete medium. The mixture was centrifuged at 1000 rpm and the pellet was resuspended in complete medium. The cell suspension was plated into tissue culture dishes for proliferation or on chamber slides for immunohistochemistry at a density of 112,500 cells/cm². The cultures were maintained at 37°C in a humidified atmosphere of 95% air and 5% CO₂, and the medium was changed twice weekly. 

Feeder Layer Co-Cultivation

Subconfluent 3T3 fibroblasts were treated with freshly prepared 50 μg/mL mitomycin for 2 hours at 37°C and seeded into culture dishes or on chamber slides at a density of 25,000 cells/cm². After 48 hours were allowed for the 3T3 cells to attach, freshly isolated porcine or bovine IPE or RPE cells were plated above the 3T3 fibroblasts. For the control, 3T3 cells were cultured without addition of pigment cells. 

Preparation and Labeling of Rod Outer Segments

Isolated porcine retinas were agitated in KCl buffer (0.3 M KCl, 10 mM HEPES, 0.5 mM CaCl₂, 1 mM MgCl₂, and 48% sucrose, pH 7.0) and the suspension was centrifuged at 7000 rpm for 5 minutes. The supernatant was filtered through a column filled with gauze, diluted with KCl buffer (1:1), and centrifuged again at 5000 rpm for 7 minutes. The isolated ROS were labeled with carboxy SNAFL-2 fluorescent dye (Molecular Probes, Leiden, Netherlands) using the technique described by Miceli et al. ³⁴ Briefly, 2 mg of ROS protein in 500 μL of sodium phosphate buffer (100 mM; pH 8.0) containing 100 mM NaCl were exposed to 100 μg of carboxy SNAFL-2 fluorescent dye in 10 μL of anhydrous dimethylformamide for 30 minutes at room temperature in the dark. The suspension was then centrifuged at 5000 rpm for 7 minutes and the precipitated ROS were washed three times with phosphate-buffered saline (PBS). The dual-wavelength fluorescent dye appears green-yellow in the acid form (pH 5) and yellow-orange in the alkaline form (pH 9). ³⁵  

Phagocytosis Assay

Confluent monolayers of porcine IPE or RPE cells cultured on feeder layers were incubated at 37°C with 150 μL of the pelleted porcine ROS freshly diluted in culture medium. After 4 hours, the cultures were washed five times with PBS to remove ROS that had not been phagocytized. The cells were trypsinized for 10 minutes with 0.25% trypsin containing 1 mM EDTA, centrifuged, washed with PBS, and used to determine the number of phagosomes. To analyze phagocytic activity, trypsinized cells were transferred to a Thoma chamber (Faust, Cologne, Germany) to which a few drops of 24 mM NaHCO₃ in PBS, containing 10% glycerol (pH 9) were added, and the chamber was covered with a glass coverslip. The cells were observed individually under a fluorescence microscope (Carl Zeiss Meditec, Oberkochen, Germany) with excitation at 450 nm and emission at 530 nm. Green-yellow fluorescent phagosomes were counted in 500 IPE and 500 RPE cells. For phase contrast the cells were photographed with a phase-contrast microscope. 

Identification of Tight Junctions by Immunohistochemistry

For immunohistochemistry freshly isolated porcine IPE and RPE cells were plated on 3T3 feeder layers grown in glass chamber slides. The cells were washed on ice with 4% paraformaldehyde in 0.2 M phosphate buffer (pH 7.0) for 30 minutes. Cell membranes were permeabilized by incubation with 0.5 M ammonium chloride and 0.25% Triton X-100 in PBS for 10 minutes. After the chamber slides were rinsed with PBS for 10 minutes, they were incubated with blocking buffer (20% normal goat serum and 5% bovine serum albumin in PBS) for 1 hour in a moist chamber at room temperature. The cells were washed with PBS and then reacted with a monoclonal anti-ZO1 antibody (tight-junction associated polypeptide; Biotrend, Cologne, Germany; dilution 1:800) for 1 hour. After 5 washes with PBS, the cells were reacted with secondary Cy3-conjugated antibody (Rockland Immunochemicals, Gilbertsville, PA). Cultures were again washed with PBS, counterstained with Hoechst 33342 dye (Molecular Probes; dilution 1:1000) at 37°C for 10 minutes, washed with PBS, and examined under a fluorescence microscope. 

Total RNA Isolation and cDNA Synthesis

Total RNA was extracted from confluent bovine IPE and RPE cells cultured in plastic six-well tissue culture plates (n = 3 independent cultures), from IPE and RPE cells cultured on feeder layers of mitomycin-treated 3T3 fibroblasts (n = 3 independent cultures), and from mitomycin-untreated 3T3 fibroblasts (RNeasy Mini Kit, in combination with the RNase-free DNase Set; Qiagen, Hilden, Germany) according to the protocol of the manufacturer. RNA concentration was determined by spectrophotometry in the UV range (UVette with a BioPhotometer; Eppendorf, Hamburg, Germany). Reverse transcription was performed on 1.0 μg total RNA on a commercial system (Promega, Madison, WI) with oligo(dT)₁₅ and random primers, according to the manufacturer's instructions. 

Primers

Using primer analysis software (NetPrimer; Premier Biosoft International, Palo Alto, CA) we selected gene-specific, intron-spanning primers suitable for quantitative real-time PCR (Table 1). Specificity for mouse collagen for the primer pair COL1A2 (accession number NM_007743; F: TCT GTC CTA GTC GAT GGC TG, R: CAC ACT GCT CTG ACC AAT CC) was determined by BLAST (Basic Local Alignment Search Tool) analysis. 

Table 1.

 

Primers Used for Quantitative Real-Time PCR

Gene Acc. Number	Primer sequence (5′-3′), Position	Melting Temp. (°C)	Annealing Temp. (°C)	Product Size (bp)
RPE65	F: AGT GAC CGA TTC AAG CCA TC, pos. 746	57.3	60.0	305
NM_174453	R: CCT TTC CAG CAA CAG AGA TC, pos. 1031
CRALBP	F: AAG ACA GTG CCT TCT TCC TG, pos. 454	57.3	62.0	245
NM_174451	R: GGT GAT TTC TTC AGA GTC CC, pos. 679
KRT8	F: CAG GCA ACT GTA TGA AGA GG, pos. 775	57.3	60.0	256
NM_001033610	R: CTC AGA AAT CTC CGT CTT CG, pos. 1011
KRT18	F: GTC TGT GGA GAG TGA CAT AC, pos. 631	57.3	60.0	216
XM_582930	R: ATC TTG CTA AGG TCC TGA GG, pos. 827
ZO1	F: AAC CGC ACA CAA ACC TGA CC, pos. 3649	59.4	60.0	282
XM_582218	R: GTC TCA ATG GAG CTC AGC AC, pos. 3911
PEDF	F: TTA CGG TAC GGC TTG GAT TC, pos. 768	57.3	60.0	268
AF017058	R: CAT CAA ACA GGG ATT GCA GC, pos. 1016
CD86	F: GTG TCA GCT CTC AAC AAC AGG, pos. 2352	59.8	62.0	273
NM_001038017	R: TTC AAG CAT GGC TAG CGT AGG, pos. 2604
BEST	F: TGG TGT CGT CCT TCG TTG AG, pos. 984	59.4	62.0	180
NM_001080245	R: TGT GTT CCG AGG GTG TCA TG, pos. 1144
HPRT1	F: TTG CCG ACC TGT TGG ATT AC, pos. 261	57.3		260
NM_001034035	R: CGC AAC CTT GAC CAT CTT TG, pos. 501
B2MG	F: CCA GCG TCC TCC AAA GAT TC, pos. 100	59.4		304
NM_173893	R: TGG TGC TGC TTA CAG GTC TC, pos. 384

PCR and DNA Sequence Analysis

Confirmation of 3T3 fibroblast RNA as well as verification of primer specificity was done by polymerase chain reaction (PCR) in a reaction volume of 25 μL containing diluted cDNA, corresponding to 25 ng of initial total RNA, 400 μM dNTP, 10 pmol of forward (F) and appropriate reverse (R) primer, 2 mM MgCl₂, 1 M betaine, and 0.5 U of polymerase (GoTaq Hot Start; Promega) in 1× PCR buffer. The cycling conditions were as follows: initial denaturation at 94°C for 2 minutes, followed by 35 cycles with denaturation at 94°C for 30 seconds, annealing at 58°C for 1 minute, and elongation at 72°C for 1 minute with a final extension at 72°C for 10 minutes. Amplicons were resolved on a 1.5% agarose gel and visualized by ethidium bromide staining. For sequence analysis the amplified PCR fragments were recovered by gel extraction (QIAqiuck Gel Extraction Kit; Qiagen) and analyzed (Eurofins MWG Operon, Ebersberg, Germany). 

Quantitative Real-Time PCR

The mRNA expression levels of various genes for bovine IPE and RPE cells cultured on plastic and on 3T3 feeder layers were quantified by real-time PCR on a thermal cycler (LightCycler 1.2 Instrument using the LightCycler FastStart DNA Master SYBR Green I kit; Roche Diagnostics, Mannheim, Germany) according to the manufacturer's protocol. Each run included the mRNA for one target gene (TG) and the two internal control genes (ICGs) HPRT1 and B2MG. Reactions were performed with diluted cDNA, corresponding to 20 ng initial total RNA, 4 mM MgCl₂, and a primer concentration of 0.1 and 0.5 μM, respectively. The following thermal cycler conditions were used: initial denaturation at 95°C for 10 minutes followed by 60 cycles with denaturation at 95°C for 10 seconds, annealing at 60°C for 8 seconds or at 62°C for 7 seconds, and elongation at 72°C for 15 seconds. Melting curve analysis of the PCR products was performed to confirm the amplification specificity from each primer pair. For evaluation of the crossing points (CPs) the resulting fluorescence curves were analyzed by means of the system software (LightCycler software 3.5.3; Roche Diagnostics). 

Even though analysis of the data revealed a constant expression for both internal control genes, the lowest SE was achieved with HPRT1 and thus gene expression levels of all target genes were normalized to the HPRT1 expression level. 

Data Analysis

To compare the gene expression levels of bovine IPE and RPE cells cultured on plastic with that of cells cultured on mitomycin-treated 3T3 fibroblasts, we used the 2^−ΔΔCT method, which describes the relative gene expression ratio. ^36,37 In this report, gene expression of bIPE and bRPE cells cultured on 3T3 fibroblasts is shown as the change in gene expression normalized to the internal control gene HPRT1 (equation 1: ΔCT = CT_{Target Gene} − CT_HPRT1) and relative to the gene expression of bIPE and bRPE cells cultured on plastic (equation 2: ΔΔCT = ΔCT_{NIH-3T3 co-culture} − ΔCT_{plastic-cultivation}). By definition, the ΔΔCT for cells cultured on plastic equals 0, resulting in a relative gene expression ratio of 1. For cells cultured on mitomycin-treated 3T3 fibroblasts, a ratio of 1 indicates a level of gene expression similar to cells cultured on plastic, whereas a ratio smaller than 1 indicates lower expression and a ratio larger than 1 indicates a higher level of gene expression. 

Precision of Quantitative Real-Time PCR

Intraseries precision was determined based on the analysis of a 10-fold cDNA synthesis series from bovine RPE cells cultured on a plastic substratum. For the high copy gene HPRT1 a mean CP of 21.60 with 0.56% coefficient of variation was determined. The mean CP value for RPE65, representing an absolute low copy gene for bRPE cultured on plastic, was 48.51 with a coefficient of variation of 4.05%. 

Interseries precision was evaluated by amplification of the internal control gene HPRT1 for different cDNA samples (Table 2). 

Table 2.

 

Precision of Quantitative Real-Time PCR

cDNA Sample	Mean Threshold Cycle	SD	Coefficient of Variation (%)
bRPE cultured on plastic	21.66	0.47	2.19
bIPE cultured on plastic	22.76	0.59	2.61
bRPE cultured on 3T3 fibroblasts	21.85	0.19	0.88
bIPE cultured on 3T3 fibroblasts	22.73	0.18	0.77

 Interseries precision evaluated by the amplification of HPRT1 throughout the different cDNA samples. The data represent the average of 3 cultures for each condition.

Statistical Analysis

A three-factor ANOVA model based on a split plot notation was fitted to the data. The factors were genes, culture conditions, and onset. Moreover, the nested effect, onset within culture, was used as the random effect and the culture by gene interaction was used in the model. Via this model, the ΔCT data for bIPE cells and bRPE cells were analyzed separately. Then two-way models were used to describe the differences for culture conditions for each gene separately. The analysis is rather explorative, so that the 5% significance level was set for orientation only (SAS software for Windows XP; SAS, Cary, NC). 

Results

3T3 Fibroblasts

Under the culture conditions used, nontreated control NIH-3T3 fibroblasts doubled approximately every 20 hours, whereas a 2-hour treatment with mitomycin resulted in a dose-dependent inhibition of proliferation. At mitomycin concentrations of 25 μg/mL or higher, proliferation was completely inhibited (Fig. 1). 

Figure 1.

 

Effect of mitomycin on the proliferation of 3T3 fibroblasts. Nontreated 3T3 fibroblasts proliferated exponentially. In the presence of 10 μg/mL mitomycin the cells still proliferated, but at a lower rate. However, the addition of 25 μg/mL or greater mitomycin concentrations resulted in the complete inhibition of proliferation.

Feeder Layer Co-cultivation

IPE and RPE cells cultured on a layer of mitomycin-treated 3T3 fibroblasts adhered, spread, and began to acquire a hexagonal shape within 12 hours of seeding (Figs. 2A, 2C) and reached confluence in 7 to 10 days, whereas cells cultured on a plastic substratum showed minimal evidence of attachment and spreading after 12 hours (Figs. 2B, 2D). Non–mitomycin-treated 3T3 cells cultured without the addition of pigment epithelial cells completely detached from the plastic dishes by day 10 of culture. 

Figure 2.

 

Phase-contrast micrographs of primary porcine RPE (A) and primary porcine IPE (C) cells cultured on a layer of mitomycin-treated 3T3 fibroblasts, 12 hours after seeding. Both the RPE and the IPE cells attached, spread, and started to form the typical hexagonal shape of epithelial cells (arrows). Phase-contrast micrographs of primary porcine RPE (B) and primary porcine IPE (D) cells cultured on a plastic substratum, 12 hours after seeding. Some cells appeared partially attached (arrows) and only an occasional cell showed the beginning of spreading (arrowheads). Original magnification, ×200.

Confirmation of NIH-3T3 Fibroblast RNA

The total RNA of bovine IPE and RPE cells co-cultured on mitomycin-treated 3T3 fibroblasts was analyzed by RT-PCR to detect the presence of xenogeneic fibroblast residues. The COL1A2 primer pair was designed such that it was specific for RNA of mouse origin only. As positive controls we used total RNAs of nontreated and mitomycin-treated 3T3 fibroblasts, and as negative controls we used total RNAs of bIPE and bRPE cells freshly isolated and cultured on a plastic substratum. 

Figure 3 clearly shows that RNA isolated from IPE (lanes 11, 12) and RPE (lanes 9, 10) cells cultured in the presence of mitomycin-treated 3T3 fibroblasts contained mouse collagen mRNA. Note that IPE and RPE cells cultured on plastic did not show a positive signal indicating that the COL1A2 primer pair was specific for mouse collagen. 

Figure 3.

 

PCR using the mouse-specific COL1A2 primer pair. PCR with COL1A2 primers verified the presence of residual mouse RNA in cultures of bovine RPE (lane 9) and IPE cells (lane 11) cultured on mitomycin-treated 3T3 fibroblasts. Total RNAs of 3T3 fibroblasts without mitomycin (lanes 5, 6) and with mitomycin treatment (lanes 7, 8) served as positive controls, whereas total RNAs of bIPE and bRPE cells freshly isolated (lanes 1, 2) as well as cultured on a plastic substratum (lanes 3, 4) were used as negative controls. Note that mouse-specific COL1A2 was not present in freshly isolated IPE or RPE but was present when the same cells were cultured on mitomycin-treated 3T3 fibroblasts.

Phagocytic Activity

Since ROS labeled with SNAFL-2 exhibit a bright green-yellow color in an acidic environment (ingested) and a pale yellow-orange color in a basic environment, ROS adherent to the outside of the cell membrane (basic environment) can be distinguished from ROS in the lysosomal compartment (acidic environment), where the ROS are transported after ingestion. When RPE and IPE cells are cultured on a feeder layer of mitomycin-treated 3T3 fibroblasts, 27% of the RPE cells and 21% of the IPE cells exhibited phagocytic activity (Table 3). The total number of phagosomes in 500 cells was 274 in RPE cells (0.55 phagosomes per cell) and 264 in IPE cells (0.52 phagosomes per cell). The phagocytic activity—that is the number of phagosomes per cell—of RPE cells cultured on a plastic substratum was twofold higher than that of RPE cells cultured on 3T3 feeder layers (1.1 vs. 0.55 phagosomes per cell) and the percentage of RPE cells containing phagosomes was also higher (32.4% on plastic vs. 27% on 3T3 fibroblasts). IPE cells also showed an increase in phagocytic activity when cultured on plastic surfaces (0.74 phagosomes per cell on plastic vs. 0.52 phagosomes per cell on 3T3 feeder layers), but there was no change in the percentage of cells containing phagosomes (Table 3). 

Table 3.

 

Phagocytic Activity of IPE and RPE Cells Cultured on Different Substrata

	RPE Cells + 3T3 Fibroblasts	IPE Cells + 3T3 Fibroblasts	RPE Cells on Plastic	IPE Cells on Plastic
Cells with phagosomes, n (%)	133 (27)	105 (21)	162 (32.4)	114 (22.8)
Average phagosomes per 500 cells	0.55 (n = 274)	0.52 (n = 264)	1.1 (n = 552)	0.74 (n = 372)

Immunohistochemistry

The formation of tight junctions was clearly evident in IPE and RPE cells cultured on a 3T3 feeder layer for 2 weeks (Figs. 4A, 4B). ZO1 immunostaining revealed the formation of an integrated monolayer by IPE and RPE cells with well demarked immunoreactivity at the cell margins indicating a pure epithelial cell culture. 

Figure 4.

 

Fluorescence micrograph of ZO1 immunostaining. Fluorescence micrograph of a porcine RPE cell monolayer (A) and porcine IPE cell monolayer (B) cultured on mitomycin-treated 3T3 fibroblasts for 2 weeks and stained with anti-ZO1 antibodies. Cell nuclei (blue) were counterstained with Hoechst dye. The complete staining of the cells' perimeter indicates tight junction formation. Original magnification, ×200.

Quantitative Real-Time PCR

Using the primers listed in Table 1, we observed single PCR products of the correct size for the appropriate genes, and sequence analysis verified the accuracy of the single amplicons when compared to the respective GenBank entries (Table 1, 1st column) (http://www.ncbi.nlm.nih.gov/Genbank; provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD). 

The PCR efficiency, calculated by the formula 10^(1/−S) − 1, ³⁸ resulted in a slope of 3.62, which is equivalent to an efficiency of 88.9% (data not shown). 

Gene Expression Profile in Cultured Bovine RPE and IPE Cells

The CPs from bIPE and bRPE cells cultured on a plastic substratum (P) and on mitomycin-treated 3T3 fibroblasts (3T3) are summarized in Table 4. For both bIPE and bRPE cells cultured on plastic or on 3T3 feeder layers PEDF and ZO1 belong to the group of high-copy genes (CP = 20–30), showing the lowest CP values, corresponding to the highest gene expression levels. KRT8 and CD86 belong to the group of low copy genes (CP = 30–40), whereas BEST, KRT18, and RPE65 belong to the group of absolute low-copy genes (CP = 40–50). For these genes no differences in CPs were observed between cells cultured on plastic or on 3T3 feeder layers. Both on plastic and on mitomycin-treated 3T3 fibroblasts, the CPs for bIPE and bRPE cells were similar for all analyzed genes, except for CRALBP and CD86. CRALBP, a low-copy gene in cultured bIPE cells (P = 36.9 ± 1.4; 3T3 = 39.3 ± 1.5), is expressed at higher levels than in bRPE cells (P = 44.0 ± 2.0; 3T3 = 44.5 ± 3.0), whereas CD86 is expressed at higher levels in bRPE cells (P = 32.5 ± 4.4; 3T3 = 31.3 ± 1.6) than in bIPE cells (P = 40.0 ± 2.3; 3T3 = 40.4 ± 3.0). 

Table 4.

 

CPs and Analysis of Gene Expression Levels by the 2^−ΔΔCP Method

Gene	Cell Type	Cultivation on Plastic				Co-cultivation on 3T3 Fibroblasts
		CP	ΔCP	ΔΔCP	2−ΔΔCP	CP	ΔCP	ΔΔCP	2−ΔΔCP
PEDF	RPE	28.04 ± 0.20	6.39 ± 0.17	0.00 ± 0.17	1.00 (0.89–1.12)	25.24 ± 0.91	3.42 ± 1.16	−2.96 ± 1.16	7.79 (3.47–17.47)
	IPE	29.88 ± 1.85	7.12 ± 2.06	0.00 ± 2.06	1.00 (0.24–4.18)	26.79 ± 0.70	3.99 ± 0.95	−3.13 ± 0.95	8.73 (4.53–16.82)
ZO1	RPE	31.13 ± 0.60	9.47 ± 0.68	0.00 ± 0.68	1.00 (0.62–1.61)	30.32 ± 0.47	8.50 ± 0.28	−0.98 ± 0.28	1.97 (1.62–2.39)
	IPE	29.85 ± 0.39	7.10 ± 0.33	0.00 ± 0.33	1.00 (0.79–1.26)	29.46 ± 0.17	6.66 ± 0.49	−0.43 ± 0.49	1.35 (0.96–1.90)
KRT8	RPE	33.43 ± 0.90	11.78 ± 1.01	0.00 ± 1.01	1.00 (0.50–2.02)	33.56 ± 0.87	11.74 ± 1.20	−0.03 ± 1.20	1.02 (0.45–2.34)
	IPE	34.55 ± 0.91	11.79 ± 1.10	0.00 ± 1.10	1.00 (0.46–2.15)	34.01 ± 1.06	11.21 ± 1.43	−0.58 ± 1.43	1.50 (0.56–4.02)
CD86	RPE	32.54 ± 4.40	11.14 ± 4.30	0.00 ± 4.30	1.00 (0.05–19.74)	31.32 ± 1.59	9.44 ± 1.75	−1.70 ± 1.75	3.24 (0.96–10.93)
	IPE	39.98 ± 2.32	17.48 ± 2.62	0.00 ± 2.62	1.00 (0.16–6.14)	40.43 ± 3.01	17.76 ± 3.51	0.28 ± 3.51	0.82 (0.07–9.37)
CRALBP	RPE	43.95 ± 1.96	22.55 ± 1.99	0.00 ± 1.99	1.00 (0.25–3.98)	44.54 ± 3.02	22.67 ± 3.41	0.12 ± 3.41	0.92 (0.09–9.76)
	IPE	36.89 ± 1.35	14.39 ± 1.66	0.00 ± 1.66	1.00 (0.32–3.15)	39.28 ± 1.52	16.61 ± 1.72	2.22 ± 1.72	0.21 (0.07–0.70)
BEST	RPE	44.34 ± 0.79	22.94 ± 0.57	0.00 ± 0.57	1.00 (0.67–1.49)	42.84 ± 1.08	20.96 ± 1.46	−1.97 ± 1.46	3.93 (1.43–10.82)
	IPE	44.24 ± 4.57	21.74 ± 4.83	0.00 ± 4.83	1.00 (0.04–28.51)	43.78 ± 1.01	21.11 ± 0.67	−0.63 ± 0.67	1.55 (0.97–2.46)
KRT18	RPE	48.08 ± 1.54	26.43 ± 1.81	0.00 ± 1.81	1.00 (0.28–3.51)	47.78 ± 0.65	25.96 ± 0.36	−0.46 ± 0.36	1.38 (1.07–1.77)
	IPE	43.67 ± 0.47	20.91 ± 0.34	0.00 ± 0.34	1.00 (0.79–1.27)	44.84 ± 1.94	22.04 ± 2.21	1.13 ± 2.21	0.46 (0.10–2.12)
RPE65	RPE	47.61 ± 1.78	25.96 ± 1.61	0.00 ± 1.61	1.00 (0.33–3.04)	47.35 ± 3.73	25.53 ± 4.09	−0.42 ± 4.09	1.34 (0.08–22.84)
	IPE	42.80 ± 1.92	20.05 ± 1.78	0.00 ± 1.78	1.00 (0.29–3.43)	46.87 ± 3.77	24.07 ± 3.33	4.02 ± 3.33	0.06 (0.01–0.62)

 The CP values were derived from quantitative real-time PCR analysis of bovine RPE and IPE cells cultured on a plastic substratum and on mitomycin-treated 3T3 fibroblasts to differentiate between high copy (CP between 20 and 30), low copy (CP between 30 and 40), and absolute low copy (CP between 40 and 50) genes. The 2^−ΔΔCP values represent the HPRT1-normalized gene expression of RPE and IPE cells cultured on mitomycin-treated 3T3 fibroblasts (ΔCP) relative to the HPRT1- normalized gene expression of RPE and IPE cells cultured on plastic (ΔΔCP). As customary, the 2^−ΔΔCP values for cells cultured on plastic are set to 1; values greater than 1 denote gene up regulation and values lower than 1 denote gene down regulation.

To analyze, whether cultivation on mitomycin-treated 3T3 fibroblasts has an effect on gene expression levels, we calculated the 2^−ΔΔCP values (Table 4, Fig. 5). Based on the 2^−ΔΔCP values, gene expression of PEDF was upregulated in the bIPE and bRPE cells cultured on mitomycin-treated 3T3 fibroblasts. Statistical analysis of the corresponding ΔCT data revealed that the difference approached significance at the 5% level (bRPE: F = 12.86, P = 0.0697; bIPE: F = 17.05, P = 0.0540). Gene expression of ZO1 was also upregulated and in bRPE cells the upregulation was significant at the 5% level (bRPE: F = 22.24, P = 0.0421). In bIPE cells gene expression of RPE65 was downregulated (F = 11.89, P = 0.0748) when the cells were cultured on mitomycin-treated 3T3 fibroblasts. 

Figure 5.

 

Analysis of gene expression performed by using the 2^−ΔΔCP method. Expression ratios of different genes of bovine RPE and IPE cells cultured on mitomycin-treated 3T3 fibroblasts normalized to the expression of cells cultured on a plastic substratum. Values greater than 1 denote upregulation and values lower than 1 denote downregulation of gene expression. Each column represents the median, minimum, and maximum and the 50th percentile of the data for three separate cultures.

Discussion

Ocular regenerative medicine has made remarkable progress in the past few years. ¹⁸ Transplantation of tissue engineered cell constructs has enabled treatment of sight-threatening complications of anterior segment diseases. Although approaches have been made to minimize contamination by avoiding serum and murine feeder layer techniques, ^39,40 most investigators still use mouse fibroblasts as an aid in cultivation of primary cells. ^{5 –8,20} Transplantation of autologous RPE and IPE cells has been advocated for the treatment of retinal degeneration. However autologous RPE and possibly IPE may have to be cultured to obtain a sufficient number of cells, also to genetically modify the cells before transplantation. Since both RPE and IPE cells are slow in attaching and starting to proliferate, a 3T3 fibroblast feeder layer would be of great advantage if it promoted cell attachment and proliferation. 

In our studies both IPE and RPE cells plated on mitotically inhibited 3T3 fibroblasts responded in a remarkable manner; the primary cells attached and began to acquire the characteristic hexagonal epithelial shape within 12 hours of seeding. Normally, these cells take a week or more to attach and spread on a plastic substratum. Areas of confluent monolayers were evident within 36 hours, whereas patches of confluent cells were apparent only after 5 to 7 days when cultured on a plastic substratum. In addition, both co-cultured RPE and IPE cells expressed characteristics of in vivo RPE cells (e.g., developed tight junctions and exhibited phagocytic activity). 

The phagocytic activity, a characteristic of RPE cells, was downregulated for RPE and IPE cells cultured on mitotically inhibited fibroblast feeder layers. Under these conditions RPE and IPE cells displayed almost equal phagocytic activity. However, on a plastic substratum RPE cells showed significantly greater phagocytic activity than IPE cells, indicating substantial downregulation of phagocytic activity in RPE cells by the mitomycin-treated 3T3 fibroblasts. 

It has been shown that 3T3 fibroblasts support epithelial cell growth and differentiation by releasing a number of soluble growth and/or maintenance factors, such as insulin-like growth factor (IGF)-I, ⁴¹ human growth factor (HGF), keratinocyte growth factor (KGF), ⁴² as well as antiapoptotic activity. ⁴³ These various factors and activities are present in medium conditioned by 3T3 fibroblasts and can support the growth and differentiation of keratinocytes, albeit at a lower level. ^3,44,45 In our studies fibroblast conditioned medium did not support the growth or differentiation of RPE and IPE cells; cell-to-cell contacts appeared necessary. In fact, 3T3-conditioned medium inhibited proliferation of RPE cells, confirming reports suggesting that the effect of the fibroblasts may be, at least partially, mediated through the cell surface. ⁴⁶  

ROS phagocytosis, development of tight junctions, and expression of RPE65 are characteristics of RPE and IPE cells and can therefore be regarded as markers of differentiation. Our studies have shown that mitomycin-treated 3T3 fibroblasts promote the growth of IPE and RPE cells in culture, while maintaining epithelial morphology and functional activity. To evaluate the effect of mitomycin-treated fibroblasts on gene expression, we have used quantitative real-time PCR to analyze the expression of genes important in RPE metabolism and function. Specifically, we analyzed the expression of RPE65 and CRALBP, which are involved in the metabolism of retinol; the keratin intermediate filaments encoding genes KRT8 and KRT18; the calcium-activated/calcium-dependent chloride channel encoding gene BEST; the neurotrophic and antiangiogenic PEDF encoding gene; and the gene CD86, which encodes a signaling molecule necessary for T cell activation. Gene expression analysis showed that compared to the cultivation on plastic, the genes for RPE65, CRALBP, and KRT18 were downregulated in the IPE cells cultured on 3T3 fibroblasts, whereas in the RPE cells, the same genes were either slightly upregulated or unchanged. It is of interest to note that the genes that were downregulated in the IPE cells cultured on 3T3 fibroblasts were expressed at higher levels in the IPE cells cultured on a plastic substratum than in the RPE cells cultured on a plastic substratum. The genes for PEDF and BEST were upregulated compared with the cultivation on plastic in both the RPE and the IPE cells. The only gene that was upregulated in the RPE cells cultured on mitomycin-treated 3T3 fibroblasts but remained unchanged or slightly downregulated in IPE cells was CD86. 

To examine whether RPE and IPE cells cultured on mitomycin-treated 3T3 fibroblasts are contaminated with mouse-derived xenogeneic products, we investigated the presence of the mRNA for collagen type 1, α2 chain. It was surprising that, after 16 days in culture, significant amounts of mouse collagen mRNA were present in the isolated total RNA of the IPE and RPE cells cultured on mitomycin-treated fibroblasts, even though in these studies a mitomycin concentration of 50 μg/mL was used, a dose that is twice the concentration (25 μg/mL) that inhibits proliferation completely (Fig. 1). The presence of mouse collagen mRNA suggests the persistence of metabolically active 3T3 fibroblasts in the cultures or that a horizontal gene transfer and translation of apoptotic DNA from 3T3 fibroblasts to IPE and RPE cells occurred. ⁴⁷  

The results presented here show that culturing IPE and RPE cells on mitomycin-treated 3T3 fibroblasts is useful in providing large quantities of primary cells that maintain their epithelial morphology and functional activity. However, the results should be interpreted with caution. In addition, transplantation of RPE or IPE cells cultured on 3T3 fibroblasts feeder layers in humans should not be contemplated because of the possible immunologic responses to xenogeneic proteins and graft failure. The results of our studies suggest that caution should be exercised in using cells cultured on xenologous feeder cell layers to re-engineer de novo corneal grafts, RPE-Bruch's membrane, and other tissues, that are intended for clinical use as replacement tissues in degenerative diseases. 

Footnotes

 Supported by a grant from the Interdisciplinary Center for Clinical Research IZKF Aachen within the faculty of Medicine at the RWTH Aachen University.

Footnotes

 Disclosure: S. Johnen, None; L. Wickert, None; M. Meier, None; A.K. Salz, None; P. Walter, None; G. Thumann, None

The authors thank Ralf-Dieter Hilgers (Institute of Medical Statistics, RWTH Aachen University, Aachen, Germany) for the statistical analysis.