Activation of the β-Catenin Signaling Pathway and Its Impact on RPE Cell Cycle

Abstract

purpose. To investigate the effect of EGF, IGF-1, and VEGF on ARPE19 cell proliferation and differentiation.

methods. The gene expression of RPE-specific differentiation and proliferation markers and the transcriptional and translational activity of β-catenin signaling markers were measured by flow cytometry and RT-PCR.

results. The data showed a significant decrease in RPE65, CRALBP, and cytokeratin 18 in ARPE-19 cells stimulated with EGF and IGF-1. In addition, a significant decrease in GSK-3β and β-catenin was observed that was paralleled by an increase in cyclin D1 expression. Cell cycle studies revealed an increase in ARPE cells in the S-G₂/M-phase after treatment with EGF or IGF-1. VEGF, on the other hand, led to a reduction in cyclin D1 and to an increase in GSK 3β and β-catenin expression which was paralleled by an increase in RPE-specific differentiation markers.

conclusions. The data demonstrate the induction of proliferation by EGF and IGF-1 and upregulation of the β-catenin signaling pathway in ARPE-19 cells. The data suggest that activation of the β-catenin signaling pathway may be key in activating ARPE-19 cells by different growth factors.

Retinal degeneration often originates in dysfunctional retinal pigment epithelial (RPE) cells, which are quiescent, nondividing cells resting in the G₀-phase of the cell cycle. ¹  

One approach to the treatment of retinal degenerations is the transplantation of RPE cells. ² Autologous RPE cells have already been successfully transplanted in human eyes as a suspension ^{3 4} and as a patch. ⁵ Although recovery of vision has been observed in the animal model only limited improvement has been obtained in patients with AMD as transplantation of autologous RPE cells from aged donors may result in a senescent, dysfunctional RPE phenotype that will not have enough impact on recovering vision in patients with AMD. ^{6 7 8 9 10 11}  

Rejuvenated cells of autologous donors would be better candidates for obtaining higher success in transplantation. The first step of a rejuvenation process is the push back of cells into the G₁-phase of the cell cycle. Such a process would induce an active cell cycle for those cells, promoting proliferation and reducing differentiation. 

For this purpose, we chose different cell-specific differentiation markers (CRALBP, RPE65, cytokeratin 18), proliferation markers (cyclin D1), and markers of the β-catenin signaling pathway (β-catenin, GSK3β), to investigate proliferation versus differentiation in ARPE-19 cells on the transcriptional level via RT-PCR and translational level via flow cytometry. In addition we evaluated growth curves and performed cell cycle studies to determine whether IGF-1, VEGF, and EGF are candidates for promoting cell proliferation and restriction of differentiation via activation of the β-catenin signaling pathway, as IGF-1 and EGF are known to induce proliferation in other epithelial cell types ^{12 13} and VEGF has been found to stimulate vascular endothelial cell growth and endothelial cell survival. ¹⁴  

In a previous study, we evaluated mRNA expression patterns in ARPE-19 cells and found significant changes in GSK-3β and β-catenin gene expression. ¹⁵ Since changes of mRNA levels only poorly reflect actual protein patterns in cells, we also have evaluated protein expression levels by flow cytometry and have evaluated growth curves and cell cycle studies to comprehensively show the impact of the three growth factors on proliferation versus differentiation via activation of the β-catenin signaling pathway in this study. Compared to our previous study, we found similar mRNA expression patterns, which were verified by our cytometry data, growth curves, and cell cycle studies. 

Materials and Methods

Cell Culture

ARPE-19 cells were obtained from ATCC (Manassas, VA), ¹⁶ and cultured in Ham’s F10 with 10% FCS. ARPE-19 cells of passages 3 to 8 were used for growth factor stimulation. 

Stimulation of ARPE-19 Cells with EGF, IGF-1, or VEGF

Cells were plated out at a density of 1× 10⁶ per T₂₅ flask. Subconfluent cells were starved in 0% FCS medium for 30 minutes, followed by incubation with the different growth factors in 2% FCS medium for 24, 48, and 72 hours, and were harvested before cells reached confluence. 

Flow Cytometry

After growth factor treatment, cells were harvested and fixed and permeabilized with an intracellular staining kit (Fix and Perm; Biozol-An der Grub Bioresearch, Eching, Germany). Cells were incubated with primary antibodies against cytokeratin 8/18 (Dako Diagnostika, Hamburg, Germany), RPE65 (Abcam, Dresden, Germany), CRALBP (Abcam), cyclin D1 (New England Biolabs (Cell Signaling Technology distributor in Germany), Frankfurt am Maim, Germany), β-catenin (New England Biolabs), and GSK-3β (New England Biolabs). Thereafter, the cells were washed and incubated for 30 minutes with an FITC-labeled anti-mouse antibody (Caltag Medsystems, Hamburg, Germany). Cell sorting was performed on a flow cytometer (FACScan; BD Biosciences, Karlsruhe, Germany) and analyzed with the accompanying software (Cell Quest; BD Biosciences). 

Real-Time RT-PCR

Preparation of mRNA from ARPE-19 cells was performed (RNeasy kit; Qiagen, Hilden, Germany), followed by reverse transcription into cDNA with oligo-dT primers (Clontech, Palo Alto, CA). cDNA (5 μL) was amplified for 40 cycles with specific primers for G3PDH, cytokeratin 18, RPE65, CRALBP, β-catenin, cyclin-D1, and GSK-3β (Table 1 ; Refs. ^{13 15 17} ). PCR reactions containing SYBR-green were amplified on a real-time PCR system (Rotor gene 2000; Corbett Life Science, Sydney, NSW, Australia). 

Data were acquired as the cycle threshold (Ct) value, which denotes the starting cycle for the amplification. As an internal standard to normalize mRNA levels for differences in sample concentration and loading, amplification of G3PDH was used. Standard curves were constructed from standard reactions for each target gene and internal control by plotting Ct versus log cDNA dilution. Because the amplification efficiencies of target genes and internal control were equal, the relative change of target gene expression in stimulated ARPE-19 cells compared with unstimulated ARPE-19 cells (ΔCt calibration values) was calculated with the equation 2^−ΔΔCt, where ΔΔCt = ΔCt_{(calibration value)} − ΔCt_{(target value)}. ¹⁸ The ΔCt values were determined by subtracting the average G3PDH Ct from the average target gene Ct. The standard deviation of the difference was calculated from the standard deviations of target gene and β-actin values. 

After each real-time RT PCR, a melting profile and agarose gel electrophoresis of each sample were performed to rule out nonspecific PCR products and primer dimers. 

Cell Cycle Analyses

Cell cycle analysis was performed by standard protocols. Briefly, cells were fixed in ice-cold 70% ethanol (4°C, 30 minutes) followed by RNaseA digestion (37°C, 30 minutes). The cells were stained with propidium iodide for 5 minutes at room temperature. DNA content of the cells was measured by flow cytometry (FACScan; BD Biosciences, Franklin Lakes, NJ), and cell cycle distribution was calculated (CellQuest software; BD Biosciences). 

Statistics

Mean and SE were calculated. For statistical analyses, paired Student’s t-tests were performed. 

Results

Cell Morphology

After 24 and 48 hours of culture, ARPE-19 cells were subconfluent. 

Differentiation versus Proliferation

Protein Expression in Growth Factor–Stimulated ARPE19 Cells.

Significant changes in protein expression patterns of the evaluated proteins were observed after 24 hours of culture. 

After stimulation with IGF-1 for 48 hours, ARPE19 cells showed a significant decrease in cytokeratin 18 (44.85 ± 8.72 MCF [mean cell fluorescence] vs. 30.28 ± 13.32 MCF; P = 0.008), RPE65 (43.21 ± 10.44 MCF vs. 30.46 ± 14.98 MCF; P = 0.02), and CRALBP (42.73 ± 7.47 MCF vs. 35.74 ± 11.62 MCF; P = 0.03; Fig. 1 ). A significant increase in protein expression was found for cyclin D1 (37.66 ± 10.64 MCF vs. 43.33 ± 5.85 MCF; P = 0.38), whereas a decrease was demonstrated for β-catenin (42.88 ± 12.42 MCF vs. 34.44 ± 9.79 MCF; P = 0.03) and GSK-3β (42.97 ± 12.07 MCF vs. 27.14 ± 15.61 MCF; P = 0.03; Fig. 2 ). 

Stimulation of ARPE19 cells for 48 hours with EGF revealed a comparable change in protein expression pattern, as observed for IGF-1 (Figs. 1 2) . 

After stimulation of ARPE19 cells with VEGF, cytokeratin 18, RPE65, and CRALBP expression increased significantly (17.57 ± 2.30 MCF vs. 37.83 ± 15.14 MCF; P = 0.03; 19.59 ± 1.97 MCF vs. 43.89 ± 30.17 MCF; P = 0.02; 23.24 ± 10.02 MCF vs. 58.09 ± 35.39 MCF; P = 0.05), respectively (Fig. 1) . 

In contrast, cyclin D1 (37.00 ± 10.90 MCF vs. 33.35 ± 15.44 MCF; P = 0.38) and β-catenin and GSK-3β (20.49 ± 9.06 MCF vs. 33.66 ± 23.00 MCF; P = 0.09; 24.54 ± 10.85 MCF vs. 29.12 ± 16.62 MCF; P = 0.13, respectively) decreased (Fig. 2) . 

Expression of Target Gene mRNA in Growth Factor–Stimulated ARPE19 Cells.

No significant changes in transcription patterns of the evaluated markers were observed after 24 hours of culture. 

After EGF treatment, a highly significant decrease in the mRNAs of cytokeratin (20.67 ± 1.71 and 19.86 ± 0.77; P = 0.006), RPE65 (26.68 ± 0.52 and 26.24 ± 0.30; P = 0.0002), and CRALBP (28.67 ± 1.08 and 28.26 ± 0.09; P = 0.00002) was recorded, resulting in 2% cytokeratin, 1% RPE65, and 2% CRALBP mRNA remaining in relation to the control cells (Fig. 3) . Furthermore, a significant decrease in transcription of cyclin D1 (28.10 ± 1.88 and 27.00 ± 1.65; P = 0.005), β-catenin (24.93 ± 1.52 and 23.72 ± 0.86; P = 0.005), and GSK-3β (25.09 ± 1.60 and 23.90 ± 0.98; P = 0.006) was found, resulting in a residual of 0.001% cyclin D1, 2% β-catenin, and 2% GSK-3β mRNA. 

RT-PCR data measured after 48 hours’ IGF-1 stimulation represented a comparable but less significant change in transcription patterns, as observed for EGF (Fig. 3) . 

A decrease in transcription of the RPE-specific cytoskeleton marker cytokeratin (23.58 ± 5.42; P = 0.29) to residual 49% cytokeratin mRNA in relation to the control was found, as well as a decrease in the RPE-specific differentiation markers RPE65 (27.75 ± 0.87; P = 0.25) and CRALBP (34.70 ± 0.30; P = 0.02), resulting in a 35% decrease in RPE65 and a 96% decrease in CRALBP mRNA in relation to control subjects (Fig. 3) . In addition, a significant decrease in transcription of the Wnt-markers β-catenin (27.98 ± 2.52; P = 0.08) and GSK-3β (29.57 ± 5.00; P = 0.004) and a decrease in cyclin D1 mRNA (30.52 ± 2.61; P = 0.14) was detected, resulting in 38% β-catenin, 24% GSK-2β, and 16% cyclin D1 remaining related to G3PDH. 

RT-PCR data collected after VEGF incubation displayed a significant increase in cytokeratin (18.80 ± 0.29 and 20.60 ± 1.44; P = 0.03), RPE65 (26.79 ± 0.57 and 27.09 ± 0.30; P = 0.006), and CRALBP mRNA (28.61 ± 0.45 and 29.56 ± 0.71; P = 0.002), resulting in a residuary 160% cytokeratin, 420% RPE65, and 239% CRALBP mRNA in relation to G3PDH (Fig. 3) . In addition, a significant increase in β-catenin (28.61 ± 0.45 and 23.71 ± 1.01; P = 0.002), cyclin D1 (26.19 ± 0.40 and 26.97 ± 1.06; P = 0.01), and GSK-3β (22.98 ± 0.25 and 24.11 ± 0.44; P = 0.00004) mRNAs was identified, resulting in 253% β-catenin, 211% cyclin D1, and 211% GSK-3β mRNA. 

Cell Cycle Analyses

Cell cycle analyses shed light on the cells’ current cell cycle status due to growth factor treatment. After incubation in IGF-1 and EGF, we measured a significant decrease in cells in the in G₁/G₀-phase (50.44 ± 6.88; P = 0.03; 47.71 ± 5.18; P = 0.02) of the cell cycle, whereas cells located in G₂/M increased significantly (22.57 ± 2.69; P = 0.0001; 26.99 ± 3.79; P = 0.01) after a 48-hour incubation (Fig. 4) . No changes were observed in the S-phase (23.16 ± 4.12; P = 0.35; 22.64 ± 6.48; P = 0.32). VEGF treatment caused a significant increase in cells in the G₁/G₀ stage of the cell cycle (46.26 ± 4.14; P = 0.02), whereas cells in G₂/M significantly decreased because of VEGF incubation (23.76 ± 1.58; P = 0.01; Fig. 4 ). No changes were observed in the S-phase (20.60 ± 3.34; P = 0.27). 

Discussion

Flow cytometry and RT-PCR data were in agreement in confirming initialization of RPE cell proliferation via activation of the β-catenin signaling pathway, after incubation in EGF and IGF-1. 

IGF-1 is known to activate the canonic Wnt/β-catenin signaling pathway, ¹² leading to inactivation of GSK-3β and dephosphorylation of β-catenin in the cytoplasm and translocation of β-catenin into the cells’ nuclei. As a consequence, a decrease in β-catenin and GSK-3β transcription and translation can be observed, which was confirmed by our data. At its end, this signal transduction pathway is initiating the transcription of target genes as cyclin D1, ¹⁹ the main marker of the G₁-phase of the cell cycle. Cyclin D1 proteolysis and subcellular localization is known to be regulated by GSK-3β. ²⁰ As subcellular distribution of cyclin D1 is cell-cycle dependent, it accumulates in the nuclei of cells in the G₁-phase and exits the nucleus during S-phase, ²¹ resulting in an increase in cyclin D1 during G₁ and a decrease in the late G₁- and S-phases. Our cytometry data confirmed an increase in cyclin D1 protein in the cytoplasm of cells due to the induction of proliferation via activation of the β-catenin signaling pathway. In parallel, we measured a decrease in cyclin D1 mRNA transcription in the cells’ nuclei, as with this boost in proliferation, the cells passed the G₁ restriction point and headed forward into the S-G₂/M-phase of the cell cycle, where other cyclin-dependent kinases (i.e., cyclin E) bestride the cell cycle. Our cell cycle analyses confirmed this progress of cells into the active cell cycle, pointing out an increase in cells entering the S-G₂/M-phase accompanied by a decrease in cells cycling at the G₁ stage. Therefore, we investigated the downregulation of cyclin D1 transcription while the amount of cyclin D1 protein in the cytoplasm was still high. In addition we observed a decrease in the differentiation markers RPE65, CRALBP, and the cytoskeleton marker cytokeratin on the transcriptional and translational levels, highlighting the upregulation of cell proliferation after activation of the β-catenin signaling pathway via IGF-1 treatment. Our control experiments indicated that differentiation/cytoskeleton markers of RPE cells usually are already expressed at low levels during the proliferative phase of cells. However, induction of proliferation by growth factor treatment further reduced expression from this low level. One might speculate that this effect is even stronger in fully mature RPE cells. 

EGF is known to activate both the Ras/Raf/MEK/ERK ¹³ and the β-catenin signaling pathways. Recent published data show that the connecting link between the two pathways is Axin, a negative regulator of the canonic Wnt pathway/β-catenin signaling pathway, ²² which is responsible for downregulation of β-catenin signaling, acting in a negative feedback loop via Akt. ^{23 24} Our data showed that incubation of cells in EGF resulted in a decrease in β-catenin and GSK-3β transcription and translation comparable to IGF-1 data. EGF also is known to inhibit GSK-3β, leading to accumulation and translocation of β-catenin into the nucleus via activation of the β-catenin signaling pathway. ^{25 26} As demonstrated by our cyclin D1 data, target gene expression is boosted at the end of the β-catenin signaling cascade according to EGF treatment, indicating induction of cell proliferation. In addition, downregulation of RPE65, CRALBP, and cytokeratin were also a hint for cell proliferation rather than differentiation. 

VEGF data showed results that were the reverse of the EGF and IGF-1 findings. Analyses of flow cytometry data as well as RT-PCR findings showed that VEGF induced differentiation rather than proliferation. Cell cycle analyses supported the proposition that VEGF is not a proliferation factor in RPE cells. In recent studies, VEGF has been described to be a novel target of the canonic Wnt/β-catenin pathway in early neoplasia and has served to underscore the importance of angiogenesis in premalignant disease, ¹⁴ both supporting our conclusions. 

In all our experiments mRNA levels decreased much more steeply than did protein expression due to the longer half-life of protein compared with mRNA. Therefore, the suppressive effect could be seen more clearly in mRNA expression patterns rather than protein patterns. EGF is described to be the key factor in epithelial cells, ^{13 27} and it strongly modulates mRNA expression patterns. Therefore, mRNA expression is strongly inhibited in cells treated with EGF, whereas IGF-1 and VEGF show a more moderate downregulation of gene expression. In a previous study, we observed the same effect. ¹⁵  

ARPE-19 cells served as a model for studying RPE cell proliferation and differentiation in previous studies. This well-characterized RPE cell line expresses differentiation/cytoskeleton markers as well as proteins involved in proliferation of RPE cells. ^{16 28 29} The characteristics of this cell line therefore allow investigations of the modulatory effect of growth factors on gene expression during proliferation. Our data derived from ARPE-19 cells indicate how proliferation of RPE cells can be boosted and differentiation is inhibited even though these are already proliferating cells with minor differentiation abilities. 

With our data, we suggest how human senile RPE cells could attain re-entry into the cell cycle after activation of the β-catenin signaling pathway via EGF or IGF-1 treatment, leading to rejuvenation of these senile RPE cells. Our chosen differentiation, proliferation, and Wnt/β-catenin signaling markers offered a good overview of how proliferation of aged cells could be promoted and differentiation could be suppressed for obtaining the ideal human RPE phenotype for autologous transplantation. In a further experimental setup, we will verify our findings on human RPE cells obtained from elderly donors. As the β-catenin signaling pathway is part of the canonic Wnt pathway, ^{30 31 32} additional experiments could be conducted to further investigate which pathway upwards the β-catenin level is activated by these growth factors. 

 Submitted for publication November 11, 2008; revised March 12 and April 1, 2009; accepted June 25, 2009.

 Disclosure: K. Steindl-Kuscher, None; W. Krugluger, None; M. E. Boulton, None; P. Haas, None; K. Schrattbauer, None; H. Feichtinger, None; W. Adlassnig, None; S. Binder, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

PCR Primer Sequences

Marker	Forward Primer (5′–3′)	Reverse Primer (5′–3′)	PCR Product Length (bp)	Ref. Source
G3DPH	CCCATCAGGATCTTCCAG	CCTGCTTCACCACCTTCT	590	15
Cytokeratin 18	CACACAGTCTGCTGAGGTTG	TAAAGTCCTCGCCATCTTCC	332	15
RPE65	GTGTAGTTCTGAGTGTGGTG	CACAGAGGAAGTATGATTAT	369	15
GSK-3	AACTGCCCGACTAACAACAC	ATTGGTCTGTCCACGGTCTC	253	15
β-Catenin	TGCGGACTCAGAAGGAACTCATGAC	ACTAGTCGTGGAATGGCACC	162	15
CRALBP	TGGCAAAGTCAAGAAATCACC	CGTGGACAAAGACCCTCTCA	313	17
Cyclin-D1	CCATGGAACACCAGCTCC	GGAGCTGGTGTTCCATGG	270	13

Figure 1.

 

Flow cytometry data showing impact of EGF, IGF, and VEGF on translation of cytokeratin 18, RPE65, and CRALBP protein 48 hours after incubation. The RPE-specific differentiation markers CRALBP and RPE65 were significantly reduced due to treatment with IGF-1 and EGF, as well as the cytoskeleton marker cytokeratin 18. VEGF data showed the opposite effect, the translational increase in all three differentiation markers and therefore differentiation of cells. ☆Statistically significant.

Figure 2.

 

Flow cytometry showing the impact of EGF, IGF-1, and VEGF on the RPE-specific proliferation marker cyclin D1 and the Wnt-markers β-catenin and GSK-3β 48 hours after incubation. EGF- and IGF-1-treated cells offered an increase in the proliferation protein cyclin D1, whereas the negatively regulated Wnt proteins β-catenin and GSK-3β decreased, indicating cell proliferation. VEGF incubation caused reverse effect, indicating induction of differentiation and not proliferation of ARPE-19 cells. ☆Statistically significant.

Figure 3.

 

RT-PCR data showing the impact of IGF-1, EGF, and VEGF on transcription of differentiation markers (cytokeratin 18, RPE65, and CRALBP), Wnt markers (β-catenin and GSK 3β) and proliferation marker (cyclin D1) after 48-hour incubation. EGF and IGF-1 data indicated significant cell proliferation due to a decrease in the differentiation markers CRALBP and RPE65, the cytoskeletal marker cytokeratin 18, and the negatively regulated Wnt markers. The cell cycle marker cyclin D1 decreased significantly as well, indicating cells approaching the G₂/M checkpoint of the cell cycle and therefore proliferation of cells. VEGF data indicated cell differentiation rather than proliferation caused by an increase in differentiation markers, an increase in the negatively regulated Wnt markers, and a decrease in the proliferation marker cyclin D1. ☆Statistically significant.

Figure 4.

 

Cell cycle analyses showing the percentage of counted cells in the G₁/G₀- or S-G₂/M-phase of the cell cycle after a 48-hour treatment with EGF, IGF-1, or VEGF and intercalation of PI into DNA. EGF- and IGF-1-treated cells significantly indicate a decrease in cells in the G₁-phase of the cell cycle and an increase in cells in the S-G₂/M-phase. Cells incubated in VEGF offered a different picture. Cells cycling in the S-G₂/M-phase decreased significantly, whereas cells cycling in the G₁-phase significantly increased. ☆Statistically significant.