Molecular Vision 2006; 12:1-14 <http://www.molvis.org/molvis/v12/a1/> Received 23 May 2005 | Accepted 7 October 2005 | Published 5 January 2006

Download Reprint

Gene expression profile of cultured adult compared to immortalized human retinal pigment epithelium

Hui Cai, Lucian V. Del Priore
 
 

Department of Ophthalmology, Harkness Eye Institute, Columbia University, New York, NY

Correspondence to: Lucian V. Del Priore, MD, PhD, Robert L. Burch III Scholar, Department of Ophthalmology, Harkness Eye Institute of Columbia University, 635 West 165^th Street, New York, NY; Phone: (212) 305-2923; FAX: (212) 342-1724; email: ldelpriore@yahoo.com

Abstract

Purpose: ARPE19 is a spontaneously immortalized cell line of human retinal pigment epithelium (RPE) that is used widely to draw inferences about the behavior of adult human RPE (ahRPE). We used DNA microarray analysis to compare the gene expression profiles of these two cell types.

Methods: Second-passage cultured ahRPE from four human donors (age range 48-82 years) and ARPE19 cultured to confluence in five dishes were used for this DNA microarray study. Total RNA was isolated and first- and second-strand complimentary DNA was synthesized using standard techniques. Biotin-labeled antisense complimentary RNA was produced by an in vitro transcription reaction. Target hybridization, washing, staining, and scanning probe arrays were done following an Affymetrix GeneChip Expression Analysis Manual. Microarray data were normalized and statistical techniques were used to determine the presence or absence of expression of individual genes within ARPE19 and ahRPE, and their relative expression levels.

Results: Hierarchic clustering analysis demonstrated that the gene expression profile of ahRPE and ARPE19 samples cluster into two distinct groups with no discernable overlap. The expression of 5,634±65 gene probes (out of 12,600 on microarray Human U95Av2 chip) was detected in ARPE19 cells compared to 5,580±84 genes in ahRPE cells from four human donor eyes. Thirty-five genes are expressed exclusively in ahRPE and nine genes exclusively in ARPE19 cells. Fifty additional genes have a threefold increase and 40 genes have a threefold decrease in expression level in ahRPE compared to ARPE19. There was no clear difference in the global expression level of genes known to be related to phagocytosis, angiogenesis, or apoptosis.

Conclusions: There are significant differences in the gene expression profile of ahRPE compared to ARPE19, and with some genes exclusively being expressed in one group and other genes being upregulated or downregulated by threefold. Caution should be exercised when generalizing results obtained from ARPE19 to the behavior of ahRPE.

Introduction

The retinal pigment epithelium (RPE) forms a hexagonal monolayer of cells between the choriocapillaris and outer retina in the human eye. Proper RPE function is necessary to maintain the integrity of the outer blood-retinal barrier and function of photoreceptors. Several significant ophthalmic disorders have been associated with early RPE dysfunction, including age-related macular degeneration, some forms of retinitis pigmentosa, and other peripheral tapetoretinal degenerations [1,2]. Primary and cultured adult human RPE (ahRPE) have been used to study the physiology of the RPE in health and disease and to determine the cellular and molecular events responsible for RPE dysfunction [3-7]. However there are some limitations in studying RPE harvested from human donors, including donor-to-donor variation [8,9]. ARPE19 is a spontaneous immortalized RPE cell line obtained initially from a single human donor; due to its immortality, this cell line has been studied extensively over the last decade to obtain important insights into RPE cell biology [10-13]. Many of the known functions of ahRPE can be performed by ARPE19, including vitamin A metabolism and phagocytosis of outer segment material [10,14].

Despite these facts, there are clearly some differences between ahRPE and ARPE19 that should be considered when results with the immortalized cell line are extrapolated to the behavior of primary or cultured ahRPE. For example, the promoter of α5 integrin, which is an important molecule for RPE phagocytosis, is regulated differently between these two cell types [15]. Little is known about the overall gene expression profile of these two cell types despite extensive prior study of their properties. In the present study, we use microarray techniques to compare the gene expression profile of ahRPE to that of ARPE19, and demonstrate that there are some significant differences between these two cell types. This study provides an important reference database for subsequent studies in which the results of studies on ARPE19 are extrapolated to RPE.

Methods

Preparation of adult human retinal pigment epithelium cultures

Human eyes obtained from the National Disease Research Interchange (NDRI, Philadelphia, PA) were processed within 48 h of donor death. The ages and other information regarding donor eyes are shown in Table 1. Four samples from human donor eyes were used; given the expense and availability of human tissue, small sample sizes have been used in the past to generate data on gene expression within human tissue [16-18]. Primary RPE cell cultures were prepared from the posterior poles of the human cadaver eyes as described previously [19]. Upon receipt in the laboratory, eyes were cleaned of extraocular tissue. The anterior segment structures, vitreous, and retina were removed, leaving an eyecup with RPE on the inner surface. For these studies, 500,000 primary ahRPE cells were collected with trypsin from each pair of globes harvested from four individual human donors (age range 48 to 82 years, Table 1). The cells were incubated in a humidified atmosphere of 5% CO₂ and 95% air at 37 °C and maintained in Dulbecco's modified Eagle's medium (DMEM H16; Gibco BRL, Carlsbad, CA) supplemented with 15% fetal bovine serum (FBS; Gibco BRL), 100 IU/ml penicillin G, 100 mg/ml streptomycin, 5 mg/ml gentamicin, 2.5 mg/ml amphotericin B and 1 ng/ml human recombinant basic fibroblast growth factor (bFGF, Gibco BRL) in a 35x10 mm culture dish (Becton Dickinson Labware, Franklin Lakes, NJ) to promote RPE cell growth. The medium was changed every other day. Cells became confluent in about 14 days and confluent cultures were passaged by trypsinization to a 60 mm culture dish (Becton Dickinson Labware). For the microarray analysis about 10⁶ RPE cells were harvested from second passage cultures before synchronization by serum depletion and bFGF removal for 24 h and total RNA was isolated for the DNA microarray study.

Cytokeratin labeling

Cells were stained using a pancytokeratin antibody to verify that all cells were of epithelial origin [20,21]. For this purpose, ahRPE cells in a culture dish were rinsed in phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde for 30 min, and washed again with PBS. The cells were treated for 1 h at room temperature with 3% bovine serum albumin (Sigma Chemical, St. Louis, MO) in PBS to block nonspecific binding sites. The cells were then incubated at 37 °C for 1 h with an fluorescein-isothiocyanate (FITC)-conjugated monoclonal antipan cytokeratin antibody to cytokeratins 5, 6, and 8 (Sigma). The cells were washed thrice with PBS and examined under a fluorescence microscope. An irrelevant isotypic IgG primary antibody (antihuman von Willebrand antibody, Sigma) coupled with an FITC-conjugated secondary antibody was also used and showed no background staining. All of the harvested cells were positive for pancytokeratin indicating the cells were of epithelial origin.

Preparation of immortalized ARPE19

Immortalized human RPE (ARPE19) obtained from ATCC (Manassas, VA) were cultured and propagated in DMEM (Gibco BRL) containing 10% FBS, 100 IU/ml penicillin G, 100 μg/ml streptomycin, 100 ug/ml gentamicin, 2.5 μg/ml amphotericin B (Gibco BRL). The cells were incubated in a humidified atmosphere of 5% CO₂ and 95% air at 37 °C, and the culture medium was changed every other day. Cultures were plated in 60 mm tissue culture-treated dishes. Five dishes of 500,000 ARPE19 were cultured for about 10 days to confluence. Cells were synchronized by serum depletion for 24 h, and total RNA was isolated for the DNA microarray study.

Isolation of total RNA from RPE cells

RPE (either ARPE19 or ahRPE) were harvested by trypsinization, and RNA was isolated and purified using a Qiagen RNeasy Mini Kit (Qiagen Co., Valencia, CA) according to manufacturer's instructions. RPE cells (about 10⁶ cells) were disrupted, and total RNA was isolated using a QIA shredder and RNeasy Mini Kit. Briefly, 600 μl of lysing buffer (RLT) was added to cells in a 1.5-ml microfuge tube and cell lysate was loaded onto a QIA shredder spin column and spun for 2 min at 13,000 rpm. The homogenized lysate was then mixed with 600 μl of 70% ethanol and applied to an RNeasy mini spin column and centrifuged for 15 s at 13,000 rpm. Next 700 μl of buffer RW1 and buffer RPE were added and spun sequentially for washing twice. Then 60 μl RNase-free water was used to elute total RNA from RNeasy column. All total RNA used in the experiments were relatively pure (A260/A280>1.9). Total RNA was stored at -80 °C for later use.

DNA microarray experiments

Affymetrix U95Av2 chips (Affymetrix Inc., Santa Clara, CA) were used in the experiments. Each chip contained 12,600 human gene probes and has 1,090 RPE-related gene probes, and gene probes for many other genes related to basic cell functions such as cell proliferation and differentiation. The advantages of using such an array include the large number of genes that are measured, and that no selection bias arises from the use of pathway-specific arrays that are designed to look at proliferation genes, apoptosis genes, etc.

A T7-(dT)₂₄ oligomer, superscript reverse transcriptase II and DNA Polymerase I (Gibco BRL) were used for first-strand and second-strand cDNA synthesis using total RNA as templates. Double-stranded cDNA was cleaned with Phase Lock Gels-Phenol/Chloroform extraction and ethanol precipitation. Biotin-labeled antisense cRNA was produced by an in vitro transcription reaction (ENZO BioArray High Yield RNA Transcript Labeling Kit; Affymetrix Inc.) and incubated with fragmentation buffer (Tris-acetate, KOAc and MgOAcat; 94 °C for 35 min). Target hybridization, washing, staining, and scanning probe arrays were done following an Affymetrix GeneChip Expression Analysis Manual.

Quality controls, definitions of gene presence or absence, and statistical analysis

For quality control, the U95Av2 DNA microarray chips used includes 20 housekeeping gene probes to measure the consistency of the hybridization signals from their 3', middle, and 5' fragment of these mRNA coding regions [22]. The definitions of presence and absence of gene expression are defined with Affymetrix GCOS 1.2 statistical algorithm [23] and further filtered, with "presence" defined as a scan densitometry reading of >50 and "absence" defined as a scan densitometry reading of <50. For comparison of the different levels of gene expression between ahRPE and ARPE19, the changes were considered to be significant if there was a greater than threefold increases or decreases in the expression level of these two cell types and if the changes were statistically different (p<0.01, student's t-test). Gene expression analyses, including global normalization and scaling were performed using the Affymetrix GCOS 1.2, Array Assist 3.01 (Stratagene, La Jolla, CA) software.

Results

Quality control assessment

After 10 days in culture, ahRPE cells appeared healthy and confluent (data not shown). All nine DNA chips passed quality control using the hybridization signals from 3', middle, and 5' fragment of mRNA of 20 housekeeping genes coded in Affymetrix DNA chips.

Hierarchic clustering analysis

Clustering analysis is a statistical technique that is used to sort heterogeneous samples into several distinct clusters; samples within the cluster have more of a relationship to one another than samples from different clusters [24,25]. Hierarchical clustering yields a branching or tree diagram with the branches indicating the relationship of samples within the cluster to other samples both within and outside the clusters [26]. Hierarchical cluster analysis of the gene expression profile of the cells demonstrates that the gene expression profile of ahRPE and ARPE19 cluster into two distinct groups with no discernable overlap (Figure 1).

Expression profiles of adult RPE and ARPE19 cells

The expression of 5,634±65 gene probes (range: 5,554 to 5,726) out of 12,600 gene probes on microarray Human U95Av2 chip is detected in five ARPE19 cell samples, in comparison to detection of only 5,580±84 gene probes (range: 5,508 to 5,695) in ahRPE cells from four human donor eyes (Figure 2). More genes were detected among ARPE19 cells but this difference was not statistically significant (student t-test, p=0.312). To minimize the ambiguity for the data analyzed, the study does not include any genes whose expression levels are defined as marginal presence.

Genes detected exclusively in either cell type

We then determined the list of genes expressed in only one cell type, using stringent criteria in which genes of interest were detected as present in all samples from one cell type and not present in any of the samples from the other cell type, and in which the expression level was consistently greater than 50 in densitometry readings. Thirty-five genes were expressed only in ahRPE and nine genes were expressed only in ARPE19 cells (Figure 2). The list of these genes and their reported functions are in Table 2 and Table 3. Of the 2,500 most abundantly expressed genes on Affymetrix U95A chip in adult RPE cells, four genes were not expressed in ARPE19 (Table 2). As for the nine uniquely expressed ARPE19 genes, none were contained within the 2,500 most abundant RPE gene list. There were 2,437 (97.5%) genes present within the list of the 2,500 most abundant genes in each group.

Different expression levels

In addition to analysis of present or absent genes in ahRPE or ARPE19, the relative expression levels of the genes present in all nine ahRPE and ARPE19 samples (4,479 genes) are also compared. After normalization of all nine samples there are 50 genes in ahRPE samples whose expression levels are increased >3 fold compared to ARPE19 (Table 4). There are 40 genes whose expression levels in ahRPE are decreased >3 fold compared to ARPE19 (Table 5). When we applied the same criteria to the list of the 2,500 most abundant genes, there are 35 genes expressed at higher levels in adult RPE and seven genes expressed higher levels in ARPE19 cells (Table 4, Table 5). We also compared the expression levels of genes related to known RPE functions [27]. We were able to identify a total of 23 genes involved in phagocytosis, 142 for neurogenesis, and 40 for angiogenesis on Affymetrix U95Av2 DNA microarray chip [27]. Interestingly, none of these genes showed a differential expression between these two cell types, and thus did not show up in our differentially expressed gene list, suggesting that expression of genes responsible for major RPE functions is similar between ahRPE and ARPE19.

Discussion

ARPE19 is a spontaneously immortalized cell line isolated in the early 1990s that has been used extensively to study RPE cell biology [10,12,15,28]. Our recent literature search on public domain database (PubMED, updated on April 23, 2005) found more than 300 published papers regarding RPE cell biology studies in 2004 alone, and approximately 1/8 of these employed ARPE19. In many ways ARPE19 exhibit behavior that is similar to the behavior of adult human RPE. For example, ARPE19 form polarized epithelial monolayers with tight junctions and show barrier properties. Both types of RPE cells express RPE-specific markers CRALBP and RPE65, and both require the integrin receptor α(v)β5 for the binding and internalization of rod outer segments for phagocytosis [10,14]. Despite the extensive use of ARPE19 to understand RPE cell biology, to our knowledge the similarities and differences between the gene expression profiles of these cell types has not been characterized comprehensively elsewhere. Our study demonstrates that the global gene expression profiles of ARPE19 and ahRPE clusters into two distinct groups with no discernible overlap on the hierarchical clustering analysis. Simultaneously, there is remarkable consistency in the gene expression profile within the ARPE19 subgroup and ahRPE subgroup, respectively, in the distribution and number of genes expressed. A larger number of genes appear expressed in ARPE19 than in ahRPE even though this difference was not statistically significant. There are 35 genes detectable in ahRPE but are missing in ARPE19. A careful consideration of these genes suggests that many of these genes play an important role in the structure and function of the RPE (Table 2). For instances, tenascin, which is expressed in ahRPE but not in ARPE19, has shown to be involved in RPE cell migration [29,30].

Selenoprotein P, an antioxidant protein that protects eyes of experimental animals from oxidative damage, may be important to the development and pathogenesis of age-related macular degeneration [31]. Mitogen-activated protein kinase 4 is involved in the stress-activated kinase signaling pathway that leads to RPE cell death [32]. Guanine nucleotide binding protein (G-protein), retinoic acid receptor responder, and dipeptidylpeptidase 4 are also expressed solely in ahRPE, and the biological effect of the absence of these major regulators of cell proliferation and differentiation in ARPE19 is not known [33-35].

Additional differences in gene expression between these two cell types arise when we compare the absolute expression levels within ARPE19 and ahRPE. Several cell differentiation-related genes [36-38], including NDRG family member 4, interferon γ-inducible protein 16, and CD24 antigens, are upregulated in ahRPE cells. The role of these genes in the proper function of the retina and RPE are not known. It is possible that the low expression levels of the differentiation related genes in ARPE19 compared with ahRPE cells may hamper the ability of ARPE19 to maintain a state of differentiation, which is consistent with the observation that proliferation-related genes are expressed at a higher level in an immortalized cell line [39,40]. Simultaneously, proliferation-related genes, such as serine/threonine kinase 6 and V-erb-a erythroblastic leukemia viral oncogene homolog 4, are upregulated in ARPE19 [41,42]. We noted no differences in the expression levels of known genes involved in phagocytosis, neurogenesis, and angiogenesis on the Affymetrix U95A DNA chip, thus suggesting that genes responsible for these RPE functions are preserved in ARPE19 cell line.

It is known that the gene expression profile of cells can be altered by changes in the surrounding environment, including changes in the culture medium [15], passage number [43], and contact with retinal outer segment [44]. Previous workers have demonstrated that results obtained can also depend upon the platform used to study the gene expression profile [45]. In the current study, we use culture conditions that are similar to those commonly employed to culture RPE, although some caution should be exercised in generalizing these results to other culture conditions or to the expression profile of RPE in vivo.

Basic fibroblast growth factor (bFGF) was removed from the medium by washing 24 h before collecting cultured adult RPE in the experiment. We cannot exclude the possibility that bFGF may have residual effects on the primary RPE culture, but most published data studying primary RPE use bFGF as a supplement.

There are remarkable similarities but significant differences in the gene expression profile of cultured adult and immortalized ARPE cells, and it is important to note that some specific genes are only expressed in one of these two groups. These studies suggest caution should be exercised when generalizing results obtained from ARPE19 to results that would be obtained with adult RPE.

Acknowledgements

Supported by the Robert L. Burch III Fund, the Macula Society, the Macula Foundation, the Foundation Fighting Blindness, and unrestricted funds from Research to Prevent Blindness.

References

1. Eagle RC Jr. Mechanisms of maculopathy. Ophthalmology 1984; 91:613-25.

2. Jackson GR, Owsley C, Curcio CA. Photoreceptor degeneration and dysfunction in aging and age-related maculopathy. Ageing Res Rev 2002; 1:381-96.

3. Ishida BY, Bailey KR, Duncan KG, Chalkley RJ, Burlingame AL, Kane JP, Schwartz DM. Regulated expression of apolipoprotein E by human retinal pigment epithelial cells. J Lipid Res 2004; 45:263-71.

4. Edwards RB, Brandt JT, Hardenbergh GS. A 31,000-dalton protein released by cultured human retinal pigment epithelium. Invest Ophthalmol Vis Sci 1987; 28:1213-8.

5. Tezel TH, Del Priore LV. Serum-free media for culturing and serial-passaging of adult human retinal pigment epithelium. Exp Eye Res 1998; 66:807-15.

6. Del Priore LV, Geng L, Tezel TH, Kaplan HJ. Extracellular matrix ligands promote RPE attachment to inner Bruch's membrane. Curr Eye Res 2002; 25:79-89.

7. Tezel TH, Del Priore LV. Reattachment to a substrate prevents apoptosis of human retinal pigment epithelium. Graefes Arch Clin Exp Ophthalmol 1997; 235:41-7.

8. Garg TK, Chang JY. Oxidative stress causes ERK phosphorylation and cell death in cultured retinal pigment epithelium: prevention of cell death by AG126 and 15-deoxy-delta 12, 14-PGJ2. BMC Ophthalmol 2003; 3:5.

9. Zarbin MA. Analysis of retinal pigment epithelium integrin expression and adhesion to aged submacular human Bruch's membrane. Trans Am Ophthalmol Soc 2003; 101:499-520.

10. Dunn KC, Aotaki-Keen AE, Putkey FR, Hjelmeland LM. ARPE-19, a human retinal pigment epithelial cell line with differentiated properties. Exp Eye Res 1996; 62:155-69.

11. Alizadeh M, Wada M, Gelfman CM, Handa JT, Hjelmeland LM. Downregulation of differentiation specific gene expression by oxidative stress in ARPE-19 cells. Invest Ophthalmol Vis Sci 2001; 42:2706-13.

12. Bejjani RA, BenEzra D, Cohen H, Rieger J, Andrieu C, Jeanny JC, Gollomb G, Behar-Cohen FF. Nanoparticles for gene delivery to retinal pigment epithelial cells. Mol Vis 2005; 11:124-32 <http://www.molvis.org/molvis/v11/a14/>.

13. King RE, Kent KD, Bomser JA. Resveratrol reduces oxidation and proliferation of human retinal pigment epithelial cells via extracellular signal-regulated kinase inhibition. Chem Biol Interact 2005; 151:143-9.

14. Finnemann SC, Bonilha VL, Marmorstein AD, Rodriguez-Boulan E. Phagocytosis of rod outer segments by retinal pigment epithelial cells requires alpha(v)beta5 integrin for binding but not for internalization. Proc Natl Acad Sci U S A 1997; 94:12932-7.

15. Proulx S, Landreville S, Guerin SL, Salesse C. Integrin alpha5 expression by the ARPE-19 cell line: comparison with primary RPE cultures and effect of growth medium on the alpha5 gene promoter strength. Exp Eye Res 2004; 79:157-65.

16. Wistow G, Bernstein SL, Wyatt MK, Fariss RN, Behal A, Touchman JW, Bouffard G, Smith D, Peterson K. Expressed sequence tag analysis of human RPE/choroid for the NEIBank Project: over 6000 non-redundant transcripts, novel genes and splice variants. Mol Vis 2002; 8:205-20 <http://www.molvis.org/molvis/v8/a27/>.

17. Hollborn M, Tenckhoff S, Jahn K, Iandiev I, Biedermann B, Schnurrbusch UE, Limb GA, Reichenbach A, Wolf S, Wiedemann P, Kohen L, Bringmann A. Changes in retinal gene expression in proliferative vitreoretinopathy: glial cell expression of HB-EGF. Mol Vis 2005; 11:397-413 <http://www.molvis.org/molvis/v11/a47/>.

18. Chowers I, Gunatilaka TL, Farkas RH, Qian J, Hackam AS, Duh E, Kageyama M, Wang C, Vora A, Campochiaro PA, Zack DJ. Identification of novel genes preferentially expressed in the retina using a custom human retina cDNA microarray. Invest Ophthalmol Vis Sci 2003; 44:3732-41.

19. Ho TC, Del Priore LV. Reattachment of cultured human retinal pigment epithelium to extracellular matrix and human Bruch's membrane. Invest Ophthalmol Vis Sci 1997; 38:1110-8.

20. Del Priore LV, Tezel TH. Reattachment rate of human retinal pigment epithelium to layers of human Bruch's membrane. Arch Ophthalmol 1998; 116:335-41.

21. Tezel TH, Del Priore LV. Repopulation of different layers of host human Bruch's membrane by retinal pigment epithelial cell grafts. Invest Ophthalmol Vis Sci 1999; 40:767-74.

22. Hubbell E, Liu WM, Mei R. Robust estimators for expression analysis. Bioinformatics 2002; 18:1585-92.

23. Liu WM, Mei R, Di X, Ryder TB, Hubbell E, Dee S, Webster TA, Harrington CA, Ho MH, Baid J, Smeekens SP. Analysis of high density expression microarrays with signed-rank call algorithms. Bioinformatics 2002; 18:1593-9.

24. Krajewski P, Bocianowski J. Statistical methods for microarray assays. J Appl Genet 2002; 43:269-78.

25. Raychaudhuri S, Sutphin PD, Chang JT, Altman RB. Basic microarray analysis: grouping and feature reduction. Trends Biotechnol 2001; 19:189-93.

26. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci U S A 1998; 95:14863-8.

27. Liu G, Loraine AE, Shigeta R, Cline M, Cheng J, Valmeekam V, Sun S, Kulp D, Siani-Rose MA. NetAffx: Affymetrix probesets and annotations. Nucleic Acids Res 2003; 31:82-6.

28. Abcouwer SF, Marjon PL, Loper RK, Vander Jagt DL. Response of VEGF expression to amino acid deprivation and inducers of endoplasmic reticulum stress. Invest Ophthalmol Vis Sci 2002; 43:2791-8.

29. Sanchez-Lopez A, Cuadros MA, Calvente R, Tassi M, Marin-Teva JL, Navascues J. Radial migration of developing microglial cells in quail retina: a confocal microscopy study. Glia 2004; 46:261-73.

30. Zagzag D, Shiff B, Jallo GI, Greco MA, Blanco C, Cohen H, Hukin J, Allen JC, Friedlander DR. Tenascin-C promotes microvascular cell migration and phosphorylation of focal adhesion kinase. Cancer Res 2002; 62:2660-8.

31. Flohe L. Selenium, selenoproteins and vision. Dev Ophthalmol 2005; 38:89-102.

32. Hecquet C, Lefevre G, Valtink M, Engelmann K, Mascarelli F. Activation and role of MAP kinase-dependent pathways in retinal pigment epithelium cells: JNK1, P38 kinase, and cell death. Invest Ophthalmol Vis Sci 2003; 44:1320-9.

33. DiSepio D, Ghosn C, Eckert RL, Deucher A, Robinson N, Duvic M, Chandraratna RA, Nagpal S. Identification and characterization of a retinoid-induced class II tumor suppressor/growth regulatory gene. Proc Natl Acad Sci U S A 1998; 95:14811-5.

34. Shi WY, Skeath JB. The Drosophila RCC1 homolog, Bj1, regulates nucleocytoplasmic transport and neural differentiation during Drosophila development. Dev Biol 2004; 270:106-21.

35. Lendeckel U, Arndt M, Wolke C, Reinhold D, Kahne T, Ansorge S. Inhibition of human leukocyte function, alanyl aminopeptidase (APN, CD13) and dipeptidylpeptidase IV (DP IV, CD26) enzymatic activities by aqueous extracts of Cistus incanus L. ssp. incanus. J Ethnopharmacol 2002; 79:221-7.

36. Lachat P, Shaw P, Gebhard S, van Belzen N, Chaubert P, Bosman FT. Expression of NDRG1, a differentiation-related gene, in human tissues. Histochem Cell Biol 2002; 118:399-408.

37. Wei W, Clarke CJ, Somers GR, Cresswell KS, Loveland KA, Trapani JA, Johnstone RW. Expression of IFI 16 in epithelial cells and lymphoid tissues. Histochem Cell Biol 2003; 119:45-54.

38. Huang LR, Hsu HC. Cloning and expression of CD24 gene in human hepatocellular carcinoma: a potential early tumor marker gene correlates with p53 mutation and tumor differentiation. Cancer Res 1995; 55:4717-21.

39. Fernandes AM, Hamburger AW, Gerwin BI. Dominance of ErbB-1 heterodimers in lung epithelial cells overexpressing ErbB-2. Both ErbB-1 and ErbB-2 contribute significantly to tumorigenicity. Am J Respir Cell Mol Biol 1999; 21:701-9.

40. Grasso AW, Wen D, Miller CM, Rhim JS, Pretlow TG, Kung HJ. ErbB kinases and NDF signaling in human prostate cancer cells. Oncogene 1997; 15:2705-16.

41. Ewart-Toland A, Briassouli P, de Koning JP, Mao JH, Yuan J, Chan F, MacCarthy-Morrogh L, Ponder BA, Nagase H, Burn J, Ball S, Almeida M, Linardopoulos S, Balmain A. Identification of Stk6/STK15 as a candidate low-penetrance tumor-susceptibility gene in mouse and human. Nat Genet 2003; 34:403-12.

42. Schmucker J, Ader M, Brockschnieder D, Brodarac A, Bartsch U, Riethmacher D. erbB3 is dispensable for oligodendrocyte development in vitro and in vivo. Glia 2003; 44:67-75.

43. Wang XF, Cui JZ, Nie W, Prasad SS, Matsubara JA. Differential gene expression of early and late passage retinal pigment epithelial cells. Exp Eye Res 2004; 79:209-21.

44. Chowers I, Kim Y, Farkas RH, Gunatilaka TL, Hackam AS, Campochiaro PA, Finnemann SC, Zack DJ. Changes in retinal pigment epithelial gene expression induced by rod outer segment uptake. Invest Ophthalmol Vis Sci 2004; 45:2098-106.

45. Rogojina AT, Orr WE, Song BK, Geisert EE Jr. Comparing the use of Affymetrix to spotted oligonucleotide microarrays using two retinal pigment epithelium cell lines. Mol Vis 2003; 9:482-96 <http://www.molvis.org/molvis/v9/a61/>.