Effect of all-trans retinoic acid on the barrier function in human retinal pigment epithelial cells

Effect of all-trans retinoic acid on the barrier function in human retinal pigment epithelial cells

Biochemical and Biophysical Research Communications 407 (2011) 605–609 Contents lists available at ScienceDirect Biochemical and Biophysical Researc...

468KB Sizes 0 Downloads 19 Views

Biochemical and Biophysical Research Communications 407 (2011) 605–609

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

Effect of all-trans retinoic acid on the barrier function in human retinal pigment epithelial cells Junbo Rong, Shuangzhen Liu ⇑ Department of Ophthalmology, Xiangya Hospital, Central South University, Changsha, People’s Republic of China

a r t i c l e

i n f o

Article history: Received 5 March 2011 Available online 21 March 2011 Keywords: ARPE-19 cells All-trans retinoic acid Tight junction Barrier function

a b s t r a c t To investigate the effects of all-trans retinoic acid (atRA) on the barrier function in human retinal pigment epithelial cells, ARPE-19 cells were cultured on the filters as monolayer with atRA being added in the apical side. The change of epithelial permeability was observed from the measurement of transepithelial electrical resistance (TER), permeability assay, and Western Blot analysis. We discovered that atRA promoted the epithelial barrier function in vitro, and its bioavailability regulates the epithelial barrier, which is accompanied by altering expression of tight junctions (TJ)-associated proteins. Our study indicates that atRA provides barrier-positive elements to the RPE cell. Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction Depending on the intercellular tight junctions (TJs), the retinal pigment epithelium (RPE), as a polarized monolayer located between the photoreceptors of the neurosensory retina and the choroidal capillary bed, forms a highly selective and regulatable permeability barrier (i.e., the outer blood–retina barrier (oBRB) that is crucial for maintaining the microenvironment of the sensory retina and the choriocapillaris [1]. A number of TJ-associated proteins have been identified, including the cytoplasmic anchor proteins zonula occludens (ZOs) and the transmembrane protein Occludin [2]. The Claudins family of transmembrane proteins has also been identified as a critical component of this tight junction barrier function [3]. The expression and distribution of tight junction proteins are generally considered to be responsible for the permeability of RPE. However, the permeability and the selectivity of tight junctions, differing among the different RPE lines, may be regulated by many other factors [4–7]. All-trans retinoic acid (atRA), a derivative of vitamin A and metabolite of the visual transduction pathway, is an important signaling molecule in most tissues and has major impacts on both normal and diseased eye [8,9]. Based on studies of myopia models in various species, both the concentration of atRA and the expression of Retinoic acid receptor-beta (RAR-b) have exhibited the bidirectional and reversible changes in retina, choroid, and sclera. atRA has been suggested as a chemical signal involved in ocular growth regulation [10–14]. Both fundus changes of high myopia in clinical and the changes of RPE and choroidal thickness in different animal myopic models suggest that the integrity and barrier function of ⇑ Corresponding author. E-mail address: [email protected] (S. Liu). 0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.03.080

RPE have changed during the development of myopia [15–19]. atRA may involve in this process. Recent studies have showed regulatory effect of atRA on RPE morphology, secretion, and contractility [20– 23], however, the report on the functions of atRA in RPE permeability is still rare. Hence, we report our investigation about the effects of atRA on the barrier function and the expression of tight junction proteins in human retinal pigment epithelial cells by measurement of the transepithelial electrical resistance (TER), permeability assay, and Western Blot analysis. 2. Materials and methods 2.1. Routine conditions for cell culture and passage All experiments were carried out using ARPE-19 provided by Tongping, M.D. from the Department of Ophthalmology of the Second Affiliated Hospital of Central South University, China. Cells were routinely cultured in a DMEM/F12 (1:1) high glucose (4.5 g/l) medium containing 10% FBS, 2 mM L-glutamine, and 100 IU/mL penicillin/streptomycin. The medium was changed twice a week. All cultures were maintained at 37 °C in 5% (vol/vol) CO2 humidified atmosphere. Cells were routinely passaged at a confluence inbetween 80% and 90% and at a split ratio ranging from 1:3 to 1:6 by dissociation in 0.05% trypsin–0.02% ethylene diamine tetraacetic acid. 2.2. Retinoid handling A stock solution was prepared by dissolving all-trans retinoic acid, atRA (Sigma–Aldrich, St. Louis, MO, USA), in ethanol at a concentration of 10 3 M, which was handled in yellow or dimmed


J. Rong, S. Liu / Biochemical and Biophysical Research Communications 407 (2011) 605–609

light, and kept at 20 °C. The medium for cell treatment was always prepared freshly. When the cells were treated with different concentrations of atRA, each experiment was involved in the same amount of ethanol with the final concentration in the medium never exceeding 0.01% (vol/vol). 2.3. Culture of ARPE-19 cells on microporous filters ARPE-19 cells, between passages 16 and 19, were grown on microporous filters (Millicell–PET, Millipore, USA) with 1.0 lm pore size and 6.5 mm inner diameter, supported by 24-well culture plates. Culture started at 80,000 RPE cells/well (4  105 cells/mL) in a medium with 10% FBS. The volume on the apical side was 0.2 mL, while 1.2 mL on the basolateral side. The fluid pressure was the same in the two chambers. After cells being attached, medium was changed on the following day. On the third day, cells reached early confluence, and then serum concentration of the culture medium was reduced to 1%. After eight days of culture, atRA was added to culture medium at the different final concentration 10 6 M, 10 7 M and 10 8 M. Cells cultured in medium including only ethanol were used as controls. The corresponding media were changed every 2 days. 2.4. Measurement of barrier function of RPE 2.4.1. Measurement of transepithelial electrical resistance TER measurements were performed at the 4th, 8th, 12th, 16th, 20th, 24th and 28th day, respectively, using an EVOM voltmeter with ENDOHM-12 (World Precision Instruments, Sarasota, FL, USA). Microporous filters were removed from the incubator and placed at room temperature (RT) for 30 min equilibration before measurements. Net TER was calculated by subtracting the resistance of a blank filter from the measured value and normalized by the area of the monolayer expressed in standard units of ohms per square centimeter (X cm2). Each experiment was repeated with at least three different filters, and twice for each.

solubilize proteins and then centrifugalized at 14,000 g for 10 min. Supernate was harvested and quantitated by the BCA kit. A total of 20 lg protein was in electrophoresis by 7.5% SDS–PAGE and transferred onto PVDF membrane in semi-dry system. After blocking in solution containing 5% skimmed milk powder for 1 h at room temperature, the membrane was incubated with specific antibodies (SantaCruz Biotechnology, SantaCruz, CA) against occludin (rabbit anti-occludin; 1:500), claudin-1 (rabbit anti-claudin-1; 1:200), ZO-1 (mouse anti-ZO-1; 1:500) and b-actin (mouse anti-b-actin; 1:500) at 4 °C overnight. Then membrane was washed extensively and then reacted correspondingly with goat anti-rabbit or mouse horseradish peroxidase-conjugated secondary antibody (Pierce; Thermo Scientific, Rockford, IL, USA), which was washed again after that. Proteins were vosialized using the enhanced chemiluminescence detection system (ECL Plus; Beyotime, Institution of biotechnology, JiangSu, China), which was recorded by exposure to autoradiograph films (Eastman Kodak, Rochester, NY). b-Actin was used as a loading control in experiments of cell-associated proteins. 2.6. Statistical analysis Multivariate analysis was performed using the Kruskal-Wallis and post hoc tests. P < 0.05 was considered significant. 3. Results 3.1. Effect of atRA on TER After the serum concentration in the culture medium was reduced to 2%, the TER of ARPE-19 cells grown on the microporous membranes reached from 25 to 30 X cm2 at the 12th day and remained this level in the following 2 weeks (Fig. 1). In the 20th day, treatment with atRA at both 10 7 M and 10 8 M increased TER significantly (Fig. 1, ⁄). But in the 28th day, only atRA at10 7 M

2.4.2. Permeability assay The permeability of the ARPE-19 monolayer was determined by measuring the apical to basolateral movements of horseradish peroxidase (HRP; 40 kDa; Sigma–Aldrich, Poole, UK). At the 16th day and 24th day after the culture began, HRP was added to the apical compartment in a 50 lg/mL concentration. At 5, 15, 30 and 60 min, respectively, after the addition of HRP, 50 ll of fluid was collected from the basolateral compartment of each filter and the same volume of the appropriate medium was added to compensate the medium removed. HRP was determined colorimetrically by reacting 20-ll samples with 150 ll of freshly made substrate (400 lg/ ml O-phenylenediamine in 0.05 M citric acid and 0.1 M phosphate with 0.012% hydrogen peroxide at pH 5.0). The reaction was terminated with the addition of 50 ll of 0.25 M sulfuric acid. And the optical density was determined using microplate reader (Stat Fax-2100, Awareness, Palm City, USA). The concentration in the standard medium was used as the background concentration in each experiment. The readings for the tracer were then converted into microgram per milliliter by comparison with standard curves. Three cultures were used for each measurement. 2.5. Western Blot analysis Measurements were performed at the 16th and 24th day, respectively. Cells cultured on the microporous membranes were lysed with 100 ll of ice-cold modified RIPA buffer (50 mM Tris– HCl, 1% Triton X-100, 0.2% SDS, 1 mM dithiothreitol, 2 mM EGTA, 4 mM EDTA, 2 mM sodium orthovanadate, 100 mM NaCl and fresh protease inhibitors). Samples were incubated at 4 °C for 30 min to

Fig. 1. Transepithelial electrical resistance (TER) of confluent ARPE-19 cells in relation to the concentration of atRA. Vertical axis is the TER, and the horizontal axis is the time. Significant differences are shown by ⁄ and + (n = 6, p < 0.001, compared with control), Error bars: maximum and minimum. (d) atRA 0 M, control; (j) atRA 10 6 M; (.)atRA 10 7 M; () atRA 10 8 M.

J. Rong, S. Liu / Biochemical and Biophysical Research Communications 407 (2011) 605–609


measurements. The concentration of HRP in the basal compartment increased linearly over time (Fig. 2). The concentration of HRP in the 10 7 M atRA-supplemented medium was significantly lower than that in the control medium, at 5, 10, 30 min, respectively, after adding the test molecules (Fig. 2, ⁄). 10 8 M atRA showed the similar effect on the permeability to HRP at 10, 30 min, respectively, but showed no statistical difference. Corresponding to the TER value, higher concentration atRA (10 6 M) led to the higher permeability to HRP. 3.3. Western Blot analysis

Fig. 2. Effects of atRA on the permeability to a horseradish peroxidase (HRP) at the 24th day. The concentrations of HRP are those in the lower compartment. Permeability was significantly lower in cultures under 10 7 M atRA. (n = 6, p < 0.05, at 5, 10, 30 min, ⁄). The similar effect was showed under 10 8 M atRA, but showed no statistical difference. The opposite effect was observed in cultures under higher concentration atRA (10 6 M).

shows this ability obviously (Fig. 1, +). In sharp contrast, atRA at higher concentration (10 6 M) caused the decrease of TER. Within a certain range of concentration (10 7 to 10 8 M), atRA dosedependently maintained the increased TER in ARPE-19 cells during a long-term culture.

In accordance with our observations made with TER and permeability, notably, we discovered significant differences in TJ-associated proteins expression between different atRA conditions (Fig. 3). At the 16th day, higher protein expression was observed in samples grown at 10 7 M of atRA for ZO-1 (0.32 ± 0.01 vs 0.13 ± 0.02; n = 6, p < 0.05) and Occludin (0.49 ± 0.08 vs 0.35 ± 0.08; n = 6, p < 0.05). At the 24th day this difference was even more evident for ZO-1 (0.72 ± 0.04 vs 0.27 ± 0.03; p < 0.05), but not obvious for Occludin (0.63 ± 0.08 vs 0.68 ± 0.07; p > 0.05). The similar differences were observed in samples grown at 10 8 M of atRA, however, they were not significant. By contrast, at the 24th day the expression of Claudin-1 is lower in atRA condition than in control medium. 4. Discussion 4.1. ARPE-19 developed stable TER when placed on microporous filters in long-term culture

3.2. Effect of atRA on permeability The permeability assays were carried out with cells cultured for 24 days under the same conditions as those used for the TER

ARPE-19 is a primary cultured RPE cell line from the globes of a 19-year-old male donor. As distinguished from other spontaneously arising human RPE lines, it is diploid exhibiting apparently

Fig. 3. Western Blot analysis, samples were taken from 16-and 24-day-old cultures with different atRA supplementation. The expressions of Occludin (65-kDa) and ZO-1 (225-kDa) were higher in the 10 7 M atRA supplemented medium than in the control medium. Conversely, the Claudin-1 (22-kDa) expression in control medium was higher than that in the atRA medium. (p < 0.05, ⁄).


J. Rong, S. Liu / Biochemical and Biophysical Research Communications 407 (2011) 605–609

normal karyology, while most of others are aneuploid. In spite of the unknown nature of the event conferring immortality, ARPE19 is a valuable source of human RPE cells, which is not transformed and possesses structural and functional properties of RPE cells in vivo [24]. In our study, ARPE-19 displayed a highly epithelial morphology, forming a hexagonal cobblestone monolayer when placed on microporous filters. After 12 days culturing in low serum medium, TER of ARPE-19 monolayer was stable and slightly higher than 25 X cm2. This matches the other research results [6]. 4.2. Barrier function of ARPE-19 monolayer was enhanced by atRA Accumulated evidences support that atRA may involve in the establishment of epithelial integrity in different tissues, [25–27]. Effects of atRA on the cultured RPE had been studied in various species, but mainly focusing on the treatment of proliferative vitreoretinopathy (PVR). The atRA has been reported to inhibit the proliferation of RPE without evident cytotoxic effects over a very wide range of concentrations (10 9 to 10 5 M) spanning a period of 2–12 days [21]. However, in our study a significant cytotoxic effect of atRA on the ARPE-19 could be detected in the concentration of 10 6 M. These contradictory findings may attribute to the difference between RPE from the donor eye and from the PVR membrane that is already at active proliferation status. Only a limited number of reports have been published on investigating the role of atRA on the permeability of RPE monolayer. Chicken primary RPE cells with concentrations of 10 8 M and 10 9 M atRA in MODS was reported to have no obvious effect in morphology and TER [28]. Our results indicated that concentrations of 10 7 M and 10 8 M atRA show the ability to enhance barrier function of ARPE-19 monolayer by increasing TER and reducing HRP flux. The gap between different results could be due to the variation of the species and the cultural conditions. 4.3. Under the treatment of atRA, expression of TJ-associated proteins changed accordingly with the altering of permeability Like other epitheliums, the cells of ARPE-19 monolayer are bound to each by a circumferential band of junctional complex. The tight junctions (TJs) that are synthesized and assembled during epithelial differentiation are the most apical structures of the junctional complex. They serve as a barrier to control the flow of solutes and fluid from the choroidal vasculature into the outer retina and to regulate the passage of ions and small molecules through the paracellular pathway. As being studied as the key proteins, occludin and the Claudins are linked to the cytoskeleton by the intracellular membrane-associated guanylate kinase homologs ZO-1, ZO-2 and ZO-3, ZO-1, claudin-1 [29]. Based on the observation in F9 cells, in colitis model, and in some cancer tissues [25,30,31], The atRA is believed to be an obligatory component in the differentiation of epithelial cells that leads to the establishment of epithelial integrity. These evidences prompted us to examine the hypothesis that atRA may play an important role in forming functional TJs and in maintaining the integrity of RPE. Under the treatment of atRA, we discovered that the expression of TJ-associated proteins changed accordingly with the alteration of permeability. The significant up-regulation of ZO1 and occludin expression, observed in cultures treated with atRA, suggests that atRA shows barrier function in a process involving a specific increase in these tight junction proteins. However, no significant difference was observed on claudin-1. We infer that other claudins such as claudin-11 and -12 may play more important roles in the process [32].

4.4. RPE and myopia Eye growth is sensitive to visual experience and altered by both form-deprivation (FD) and optical defocus. Observations of local growth modulation certainly imply that the neural retina itself has to be the source of growth-regulating signals. The primary targets of growth regulation are the choroidal and scleral layers of the eye that demarcate the boundary of the posterior vitreous chamber. Thus the RPE, interposed between the retina and the choroid, is likely to be critical in relaying retinal growth signals to the choroid and sclera by relaying critical growth signals to the choroid, secreting signal molecules into the choroid, or modulating fluid exchange between the retina and choroid [33,34]. Myopia is mostly caused by excessive elongation of the vitreous chamber of the eye. Anatomical studies of RPE showed that some of the observed changes might simply reflect a flattening of RPE cells to accommodate the enlarging vitreous chamber without cell number’s alteration [35]. At the same time, Choroidal thickness can be altered by visual manipulation [36]. These changes move the retina towards the eye’s altered plane of focus and are presumed to involve mechanisms that underlie normal developmental emmetropization. There are biochemical and fluid changes that may subserve choroidal thickness changes. Available data show that RPE ion and fluid transport are altered in form-deprived and recovering eyes [18]. The enlargement of the vitreous chamber seen in FDM is likely a consequence of retention of fluid in the vitreous chamber too. And it is supported by studies showing an increased amount of liquid vitreous with normal hyaluronate in occluded chick eyes, which would be consistent with a reduction in retinochoroidal flow during deprivation [37]. However, the change of RPE barrier function that may influence the net transfer of fluid across RPE during the process of ocular growth is generally neglected by myopia researchers. A possible mediator of the eye growth changes is all-trans retinoic acid (atRA) [11]. It is reported that retinal-atRA increased in myopic eyes with accelerated elongation and was lower in eyes with inhibited elongation. Previous reports have shown the functional atRA can modulate the permeability of various cell types in vitro, and the impaired atRA signaling leads to the disruption of functional TJs [38]. In our study, ARPE-19 cells were cultured on the filters as monolayer and atRA was added in the apical side to simulate the increasing atRA in retina during myopia development. A change of epithelial permeability was observed. We discovered that atRA promoted the epithelial barrier function in vitro and its bioavailability regulates the epithelial barrier accompanied by altering expression of TJ-associated proteins. Though the establishment and regulation of epithelial permeability maybe the result of a cascade of events triggered by visual conditions, our present study provides evidence that cellular atRA bioavailability is an important candidate for determining the blood–retinal barrier (BRB) integrity. It is interesting that, in tree shrews, the inward permeability of blood–retinal barrier (BRB) increases as eyes become myopic [16]. However, the fact that the BRB changes were found only 45 days after FD suggests that the abnormal BRB function is secondary to the myopia development rather than the cause of myopia. Although it cannot be concluded by current study that the atRA effect is direct on the epithelial integrity, our observation may support the plausible hypothesis that: With the evolution of myopia, the vitreous chamber was elongated, while RPE cell numbers were not altered. The expansion of individual RPE cell and the increment of TJs between epithelial cells were mostly like the responses to the pulling force applied on RPE cells. The fact of atRA up-regulation in retina and the elevation of the barrier function of RPE layer might be a negative feedback regulation to ocular growth, which led to the depletion of fluid flow across the RPE and the thickness

J. Rong, S. Liu / Biochemical and Biophysical Research Communications 407 (2011) 605–609

decreasing of choroid. However, with the evolution of myopia, the vitreous chamber elongated excessively, RPE layer could not maintain the epithelial integrity. As a result, the balance was broken so that chorioretinal degeneration was observed in high myopia. Our present study indicates that atRA provides barrier-positive elements to the RPE cell. We speculate that the disorder of RPE barrier function during myopia evolution is likely not the cause of myopia, but just the pathological phenomena of refractive error. Acknowledgment This research was supported by the Nature Science Foundation of China (No. 81070752). References [1] O. Strauss, The retinal pigment epithelium in visual function, Physiological Review 85 (2005) 845–881. [2] N.N. Salama, N.D. Eddington, A. Fasano, Tight junction modulation and its relationship to drug delivery, Advanced Drug Delivery Reviews 58 (2006) 15– 28. [3] K. Turksen, T.C. Troy, Barriers built on claudins, Journal of Cell Science 117 (2004) 2435–2447. [4] M. Jin, E. Barron, S. He, S.J. Ryan, D.R. Hinton, Regulation of RPE intercellular junction integrity and function by hepatocyte growth factor, Investigative Ophthalmology and Visual Science 43 (2002) 2782–2790. [5] L. Bai, Z. Zhang, H. Zhang, X. Li, Q. Yu, H. Lin, W. Yang, HIV-1 Tat protein alter the tight junction integrity and function of retinal pigment epithelium: an in vitro study, BMC Infectious Diseases 8 (2008) 77. [6] T. Abe, E. Sugano, Y. Saigo, M. Tamai, Interleukin-1ß and barrier function of retinal pigment epithelial cells (ARPE-19): aberrant expression of junctional complex molecules, Investigative Ophthalmology and Visual Science 44 (2003) 4097–4104. [7] T.A. Bailey, N. Kanuga, I.A. Romero, J. Greenwood, P.J. Luthert, M.E. Cheetham 1, Oxidative stress affects the junctional integrity of retinal pigment epithelial cells, Investigative Ophthalmology and Visual Science 45 (2004) 675–684. [8] D.I. Axel, A. Frigge, J. Dittmann, H. Runge, I. Spyridopoulos, R. Riessen, R. Viebahn, K.R. Karsch, All-trans retinoic acid regulates proliferation, migration, differentiation, and extracellular matrix turnover of human arterial smooth muscle cells, Cardiovascular Research 49 (2001) 851–862. [9] M. Osanai, M. Petkovich, Expression of the retinoic acid-metabolizing enzyme CYP26A1 limits programmed cell death, Molecular Pharmacology 67 (2005) 1808–1817. [10] M. Bitzer, M. Feldkatmper, F. Schaeffel, Visually induced changes in components of the retinoic acid system in fundal layers of the chick, Experimental Eye Research 70 (2000) 97–106. [11] S.A. McFadden, M.H. Howlett, J.R. Mertz, Retinoic acid signals the direction of ocular elongation in the guinea pig eye, Vision Research 44 (2004) 643–653. [12] J.R. Mertz, J. Wallman, Choroidal retinoic acid synthesis: a possible mediator between refractive error and compensatory eye growth, Experimental Eye Research 70 (2000) 519–527. [13] D. Troilo, D.L. Nickla, J.R. Mertz, J.S. Rada, Change in the synthesis rates of ocular retinoic acid and scleral glycosaminoglycan during experimentally altered eye growth in marmosets, Investigative Ophthalmology and Vision Science 47 (2006) 1768–1777. [14] D. Yan, X. Zhou, X. Chen, F. Lu, J. Wang, D. Hu, J. Qu, Expression of retinoid acid receptors in human scleral fibroblasts, regulation of growth of fibroblasts by retinoic acid, Zhonghua Yan Ke Za Zhi 43 (2007) 750–753. [15] K. Kobayashi, K. Ohno-Matsui, A. Kojima, N. Shimada, K. Yasuzumi, T. Yoshida, S. Futagami, T. Tokoro, M. Mochizuki, Fundus characteristics of high myopia in children, Japanese Journal of Ophthalmology 49 (2005) 306–311. [16] N. Kitaya, S. Ishiko, T. Abiko, F. Mori, H. Kagokawa, M. Kojima, K. Saito, A. Yoshida, Changes in blood–retinal barrier permeability in form deprivation myopia in tree shrews, Vision Research 40 (2000) 2369–2377. [17] D. Troilo, D.L. Nickla, C.F. Wildsoet, Choroidal thickness changes during altered eye growth and refractive state in a primate, Investigative Ophthalmology and Vision Science 41 (2000) 1249–1258.


[18] H. Liang, S.G. Crewther, D.P. Crewther, B.M. Junghans, Structural and elemental evidence for edema in the retina, retinal pigment epithelium, and choroid during recovery from experimentally induced myopia, Investigative Ophthalmology and Vision Science 45 (2004) 2463–2474. [19] L.F. Hung, J. Wallman, E.L. Smith 3rd, Vision-dependent changes in the choroidal thickness of macaque monkeys, Investigative Ophthalmology and Vision Science 41 (2000) 1259–1269. [20] Y.C. Chang, Y.H. Kao, D.N. Hu, L.Y. Tsai, W.C. Wu, All-trans retinoic acid remodels extracellular matrix and suppresses laminin-enhanced contractility of cultured human retinal pigment epithelial cells, Experimental Eye Research 88 (2009) 900–909. [21] W.C. Wu, D.N. Hu, S. Mehta, Y.C. Chang, Effects of retinoic acid on retinal pigment epithelium from excised membranes from proliferative vitreoretinopathy, Journal of Ocular Pharmacology and Therapeutics 21 (2005) 44–54. [22] T. Yasunari, N. Yanagihara, T. Komatsu, M. Moriwaki, K. Shiraki, T. Miki, Y. Yano, S. Otani, Effect of retinoic acid on proliferation and polyamine metabolism in cultured bovine retinal pigment epithelial cells, Ophthalmic Research 31 (1999) 24–32. [23] H. Kishi, E. Kuroda, H.K. Mishima, U. Yamashita, Role of TGF-beta in the retinoic acid-induced inhibition of proliferation and melanin synthesis in chick retinal pigment epithelial cells in vitro, Cell Biology International 25 (2001) 1125–1129. [24] K.C. Dunn, A.E. Aotaki-Keen, F.R. Putkey, L.M. Hjelmeland, ARPE-19, a human retinal pigment epithelial cell line with differentiated properties, Experimental Eye Research 62 (1996) 155–169. [25] M. Osanai, N. Nishikiori, M. Murata, H. Chiba, T. Kojima, N. Sawada, Cellular retinoic acid bioavailability determines epithelial integrity: role of retinoic acid receptor alpha agonists in colitis, Molecular Pharmacology 71 (2007) 250–258. [26] S. Komiya, M. Shimizu, J. Ikenouchi, S. Yonemura, T. Matsui, Y. Fukunaga, H. Liu, F. Endo, S. Tsukita, A. Nagafuchi, Apical membrane and junctional complex formation during simple epithelial cell differentiation of F9 cells, Genes to Cells 10 (2005) 1065–1080. [27] D. Telgenhoff, S. Ramsay, S. Hilz, P. Slusarewicz, B. Shroot, Claudin 2 mRNA and protein are present in human keratinocytes and may be regulated by all-transretinoic acid, Skin Pharmacology and Physiology 21 (2008) 211–217. [28] K. Konari, N. Sawada, Y. Zhong, H. Isomura, T. Nakagawa, M. Mori, Development of the blood–retinal barrier in vitro: formation of tight junctions as revealed by occludin and ZO-1 correlates with the barrier function of chick retinal pigment epithelial cells, Experimental Eye Research 61 (1995) 99–108. [29] J. Miyoshi, Y. Takai, Molecular perspective on tight-junction assembly and epithelial polarity, Advanced Drug Delivery Reviews 57 (2005) 815–855. [30] H. Kubota, H. Chiba, Y. Takakuwa, M. Osanai, H. Tobioka, G. Kohama, M. Mori, N. Sawada, Retinoid X receptor alpha and retinoic acid receptor gamma mediate expression of genes encoding tight-junction proteins and barrier function in F9 cells during visceral endodermal differentiation, Experimental Cell Research 263 (2001) 163–172. [31] H. de Thé, Altered retinoic acid receptors, FASEB Journal 10 (1996) 955–960. [32] S. Prasad, R. Mingrino, K. Kaukinen, K.L. Hayes, R.M. Powell, T.T. MacDonald, J.E. Collins, Inflammatory processes have differential effects on claudins 2, 3 and 4 in colonic epithelial cells, Laboratory Investigation 85 (2005) 1139– 1162. [33] A.M. Westbrook, S.G. Crewther, H. Liang, J.A. Beresford, M. Allen, I. Keller, D.P. Crewther, Formoguanamine-induced inhibition of deprivation myopia in chick is accompanied by choroidal thinning while retinal function is retained, Vision Research 35 (1995) 2075–2088. [34] Y. Seko, Y. Tanaka, T. Tokoro, Scleral cell growth is influenced by retinal pigment epithelium in vitro, Graefes Archive for Clinical and Experimental Ophthalmology 232 (1994) 545–552. [35] A.M. Harman, R. Hoskins, L.D. Beazley, Experimental eye enlargement in mature animals changes the retinal pigment epithelium, Visual Neuroscience 16 (1999) 619–628. [36] D.L. Nickla, J. Wallman, The multifunctional choroid, Progress in Retinal and Eye Research 29 (2010) 144–168. [37] R.L. Pickett-Seltner, J.G. Sivak, J.J. Pasternak, Experimentally induced myopia in chicks: morphometric and biochemical analysis during the first 14 days after hatching, Vision Research 28 (1988) 323–328. [38] K.S. Hayes, A.J. Bancroft, R.K. Grencis, Immune-mediated regulation of chronic intestinal nematode infection, Immunological Reviews 201 (2004) 75–88.