Rho/ROCK Signaling in Regulation of Corneal Epithelial Cell Cycle Progression

Abstract

purpose. The authors’ previous study showed that the expression of a Rho-associated serine/threonine kinase (ROCK) is regulated during cell cycle progression in corneal epithelial cells. The present study was conducted to determine whether and how Rho/ROCK signaling regulates cell cycle progression.

methods. Rabbit corneal epithelial cells (RCECs) in culture were arrested in the G₀ phase of the cell cycle by serum deprivation and then allowed to re-enter the cell cycle in the presence or absence of the ROCK inhibitor (Y27632) in serum-supplemented medium. The number of cells in the S phase, the relative levels of specific cyclins and CDKs and their intracellular distribution, and the relative levels of mRNAs were determined by BrdU labeling, Western blot and immunocytochemical analyses, and real-time RT-PCR, respectively.

results. ROCK inhibition delayed the progression of G₁ to S phase and led to a decrease in the number of RCECs entering the S phase between 12 and 24 hours from 31.5% ± 4.5% to 8.1% ± 2.6%. During the cell cycle progression, protein and mRNA levels of cyclin-D1 and -D3 and cyclin-dependent kinases CDK4 and CDK6 were significantly lower, whereas the protein levels of the CDK inhibitor p27^Kip1 were higher in ROCK-inhibited cells. Intracellular mRNA or protein levels of cyclin-E and protein levels of CDK2 were not significantly affected, but their nuclear translocation was delayed by ROCK inhibition.

conclusions. ROCK signaling is involved in cell cycle progression in RCECs, possibly by upregulation of cyclin-D1 and -D3 and CDK4, -6, and -2; nuclear translocation of CDK2 and cyclin-E; and downregulation of p27^Kip1.

Corneal epithelium is a self-renewing tissue that is maintained by centripetal migration of the differentiated corneal epithelial cells derived from stem cells located at the limbus. Once the differentiated cells enter the cell cycle, they undergo a limited number of cell divisions before following the pathway of terminal differentiation. Currently, the mechanisms of the regulation of cell cycle entry and progression in corneal epithelial cells are not well understood. Most of the reported studies on corneal epithelial cell proliferation have dealt with wound-healing responses. However, more recently, since the discovery of cell cycle regulatory proteins, their role in the regulation of epithelial homeostasis and repair has received attention. ^{1 2 3 4 5} Studies of cultured mammalian cells have shown that the progression through the cell cycle is regulated by several different stage-specific cyclin-dependent kinases (CDKs) in association with specific cyclins and CDK inhibitors (CDKIs). ^{6 7} Cell cycle progression through each stage requires different sets of cyclins, CDKs, and CDKIs. Regulation of the abundance of specific cell-cycle-regulatory proteins during the progression of the cell cycle is controlled by programmed synthesis and degradation. Progression through the G₁ and G₁/S phase is regulated by CDK4, CDK6, and CDK2, their associated cyclin-D1, -D3, and -E, and CDKIs p27^Kip1 and p21^Cip1. 

We had previously reported that increased expression of ROCK-1 (Rho-associated coiled coil kinase) was one of the changes associated with limbal to corneal epithelial transition. ⁸ ROCK-I (ROKβ) and -II (ROKα) are among the effectors of RhoA that have been most extensively characterized. ^{9 10 11 12} ROCK phosphorylates myosin light-chain phosphatase and thereby inactivates it. ROCK also phosphorylates myosin light chains and thus increases myosin ATPase activity, leading to the assembly of actomyosin filaments. Other ROCK substrates include LIM kinase 1 and 2 (LIMK-1 and -2), which are involved in the stabilization of actin filaments. Our in vitro studies of cultured rabbit corneal epithelial cells (RCECs) indicated that ROCK-I expression is regulated during the progression of the cell cycle. ¹³ Increased ROCK has been associated with metastasis of some cancers and tumor cell proliferation. ^{14 15 16 17 18} ROCK inhibition with a specific inhibitor of ROCK, Y27632, has been shown to inhibit proliferation of several different cell types in vitro. ^{17 19 20 21 22 23 24 25 26 27 28 29 30 31 32} Based on these findings, we speculated that ROCK-I is involved in the regulation of G₁ and G₁/S progression in corneal epithelial cells. To test this hypothesis, we evaluated the effects of ROCK-inhibition on G₁/S progression and expression of G₁ stage-specific cyclins and CDKs in RCE. 

Materials and Methods

Cell Culture

All procedures involving rabbits were performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Corneas with the adjacent limbus were excised from rabbit eyes (Pel-Freez Biologicals, Rogers, AK) and used for growing the primary cultures (P₀) of RCECs in SHEM (supplemented hormonal epithelial medium), ³³ according to Ebato et al. ³⁴ Cells in P₀ were subcultured using 0.25% trypsin/EDTA (Invitrogen-Gibco, Grand Island, NY), into 60-mm tissue culture dishes or four-well chamber tissue culture slides (Nalge-Nunc International, Naperville, IL) at a density of 2 × 10⁴ cells/cm² and incubated for 24 hours. To arrest the P1 cells in G_0, the medium was replaced with serum-free (SF) Dulbecco’s modified Eagle’s medium (DMEM) for 60 to 62 hours. The medium was replaced with SF DMEM (controls) or with SF DMEM with 10 μM Y27632 (ROCK inhibitor). After 6 hours, the media were replaced with DMEM/F12 supplemented with 10% serum, with or without (controls) 10 μM Y27632. At specific time intervals, the cells were processed for protein and RNA extraction or immunocytochemical analyses. 

Cell Proliferation Assay

Cells in the G₁/S or S phase were identified by staining for Ki67 (a proliferative nuclear antigen) using monoclonal anti-Ki67 antibody (Zymed, South San Francisco, CA), as described previously. ¹³ BrdU incorporation was used to identify RCECs in the S phase of the cell cycle. Six hours before the analysis of the cells, BrdU (10 μM) was added to the media in the chamber slides. After they were rinsed with phosphate-buffered saline (PBS), the cells were fixed and stained with Alexa 488-conjugated anti-BrdU according to the manufacturer’s protocol (Invitrogen, Carlsbad, CA). The cells were then counterstained for nuclei with 5 μg/mL of propidium iodide (PI) in PBS for 30 seconds, rinsed with PBS, and mounted (Immuno-mount; Shandon, Pittsburgh, PA). Confocal images were captured and recorded in commercial software (LaserSharp; Bio-Rad Laboratories, Richmond, VA) via a liquid light acquisition system (Radiance; Bio-Rad Laboratories) and the number of total (PI-stained) and BrdU-positive cells were counted in three different fields containing 200 to 300 cells using imaging software (MetaMorph; Molecular Devices, Sunnyvale, CA). 

Immunocytochemical Analysis

For actin filament staining, the cells in the chamber slide were rinsed with PBS, fixed with paraformaldehyde-lysine-periodate (PLP) and permeabilized with PIPES buffer containing 0.2% Triton X-100, as described previously. ³⁵ The fixed and permeabilized cells were reacted with Alexa fluor 546 phalloidin (Invitrogen-Molecular Probes) at a 1:50 dilution for 30 minutes, rinsed with PBS and then mounted Immuno-mount; Shandon). For immunofluorescence staining of cyclin-D1, -D3, and -E1; CDK2, -4, and -6; and p27^Kip1, the cells were fixed with cold methanol (−20°C) for 5 minutes, blocked with 5% heat-inactivated goat serum and reacted with the primary and secondary antibodies and PI (nuclear stain), as described previously. ³⁶ Primary antibodies included monoclonal anti-cyclin-D1, -D3, and -E (HE12); CDK2; and p27^Kip1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and monoclonal anti-CDK4 and -6 (Chemicon/Millipore Corp., Billerica, MA), and the secondary antibodies were Alexa 488-conjugated goat anti-mouse IgG (1:2500; Invitrogen-Molecular Probes Inc.). Fluorescent z-stack images were collected at 0.25-μm intervals with a confocal scanning laser system (Radiance 200; Bio-Rad) attached to an inverted microscope (IX70; Olympus, Tokyo, Japan), with the same settings used for the comparisons. Semiquantitative analysis of the relative intensities of fluorescence staining was performed with the image analysis software (MetaMorph; Molecular Devices). 

Western Blot Analyses

RCECs in the tissue culture dishes were rinsed three times with PBS and then extracted in RIPA buffer (1 mM dibasic sodium phosphate, 1.7 mM monobasic sodium phosphate, 150 mM NaCl [pH 7.4], 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 0.03 TIU/mL aprotinin (Sigma-Aldrich, St. Louis, MO), 1 μM sodium orthovanadate, and 100 μg/mL phenylmethylsulfonyl fluoride [PMSF]) using the protocol recommended by Santa Cruz Biotechnology, Inc. For Western blot analyses aliquots of cell extracts containing 20 to 60 μg of proteins were subjected to SDS-PAGE. Protein bands on SDS-PAGE were electrophoretically transferred to membranes (Immobilon-P; Millipore Corp.), stained with 0.1% Coomassie blue 250 in 40% methanol and 1% acetic acid for 1 minute, destained one to three times with 50% methanol and 1% acetic acid, scanned for densitometric comparisons of relative concentrations of proteins per lane, destained with methanol, and then subjected to Western blot analysis using the same primary antibodies that were used for immunocytochemical analysis. The secondary antibodies were HRP-conjugated goat anti-mouse antibodies (Santa Cruz Biotechnology, Inc.). The immunoreactive bands were detected with Western chemiluminescence reagents (Immobilon; Millipore Corp.), according to the manufacturer’s protocols. The relative differences in the densities of Coomassie-stained protein bands and the bands on the x-ray films were estimated with NIH Image-J analysis software (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). 

Quantitative RT-PCR

Cells were lysed and extracted in RLT buffer, and RNA was isolated (RNeasy Mini kit; Qiagen, Valencia, CA). Quantitative RT-PCR for the mRNAs of cyclin-D1, cyclin-D3, CDK4, CDK6, and p27^Kip1 was performed with SYBR Green RT-PCR core reagents (Applied Biosystems, Inc. [ABI], Foster City, CA), according to the manufacturer’s instructions. The reactions were performed (7700 Detection System; ABI) for 40 cycles of 15 seconds at 95°C and 60 seconds at 60°C after initial incubations for 10 minutes at 95°C. The reaction mixture contained 1× SYBR Green PCR buffer, 3.0 mM MgCl₂; 100 μM each of dATP, dCTP, dGTP, dUTP, and dTTP; and 0.02 U/μL polymerase (AmpliTaq Gold; ABI), 1.0 μL of cDNA and forward and reverse primers (Table 1)at optimized concentrations. Since the rabbit DNA sequences encoding the genes of interest were not known, partial sequences of the rabbit genes were first determined from corresponding PCR-amplified rabbit cDNAs, generated using PCR primers with cross-species homologous sequences. The primers (Table 1)used for real-time RT-PCR were then designed from the partial sequences of rabbit genes (GenBank ID in Table 1 ) after confirming their close cross-species homology. Gene-specific C_t values were standardized based on 18S C_t values obtained for each cDNA. Values shown represent the mean ± SD of the results of analyses of the RNA samples (run in duplicate or triplicate) from three separate experiments. After amplification, a dissociation curve for each reaction was generated to confirm the absence of nonspecific amplification. Gel electrophoresis confirmed the amplified product to be of the expected size. 

Statistical Analysis

All data are presented as the mean ± SD. Statistical analysis of data from three or more separate experiments was performed with repeated-measures ANOVA. The differences were considered significant at P ≤ 0.05. 

Results

Influence of ROCK Activity on G₁/S Cell-Cycle Progression

All the experiments were performed with RCECs in P1. The cells growing in the SHEM were arrested in the G₀ phase by serum starvation for 60 hours and then induced to enter the cell cycle by replacing the media with DMEM/F12 with 10% FBS, with and without 10 μM Y27632. The number of total cells in the G₁/S and S phases (Ki67 positive) in the nonsynchronized cultures derived from different groups of eyes, varied from 20% to 30%. After serum deprivation for 60 hours, only 0.5% to 2% of cells were in the S phase, as determined by Ki67-positive staining. The cells in SHEM contained actin filaments organized cortically as well as some (stress fibers) traversing through the cytoplasm. ROCK inhibition by treatment with Y27632 was evident from a significant reduction in the stress fibers. The effect of ROCK inhibition on the G₁/S progression was determined from BrdU incorporation to identify the cells that entered the S phase at different time intervals (Fig. 1) . The percentage of cells in the S phase at different times after replacing the SF medium with FBS supplemented medium is shown in Figure 2 . ROCK inhibition resulted in a reduction in the number of cells entering the S phase between 0 and 24 hours after the removal of G₀ block. However, between 24 and 36 hours, approximately 12% ± 3% of the ROCK-inhibited cells entered the S phase, indicating a delay in the G₁/S progression. 

Effects of ROCK Inhibition on G₁ Regulatory Proteins

The effects of ROCK inhibition on the expression of cyclin-D1, -D3, and -E, the regulators of G₁ to S progression in the corneal epithelial cells, were analyzed by semiquantitative Western blot analyses of these proteins in the cell extracts. When G₀ arrested epithelial cells were allowed to enter the cell cycle by changing the culture medium to FBS-supplemented DMEM/F12, intracellular protein levels of cyclin-D1 and -D3 were upregulated by severalfold (Fig. 3) . However, in the presence of Y27632 their levels were significantly lower (P < 0.05) at 18, 24, and 36 hours than those in the controls (Fig. 3) . Two protein bands (M_r ∼48 and 51 kDa) reacted with anti-cyclin-E1 monoclonal antibody. The levels of these proteins were not significantly different in the controls and ROCK-inhibited cells at different time points (Fig. 3) . However, there was an increase in the 51-kDa band and decrease in the 48-kDa band at 36 hours (P < 0.05). The two differently migrating bands of cyclin-E1 may correspond to differentially phosphorylated or alternatively spliced forms of cyclin-E1. 

Immunocytochemical analyses indicated that the intensities of staining of cyclin-D1 and- D3 in the nuclei were higher in the control cells than in the Y27632-treated cells at 12, 18, 24, and 36 hours during G₁ progression (results of 12 and 24 hours shown in Fig. 4 ). Semiquantitative analyses of the fluorescence intensities in the nuclei indicated that the average intensities of nuclear staining for cyclin-D1 and -D3 (in approximately 50 cells per field) were 2.1- and 4.5-fold lower, respectively, in the ROCK-inhibited cells than in the controls at 24 hours. Based on the immunostaining pattern, most of cyclin-D3 was localized in the nuclei of both control and ROCK-inhibited cells. Cyclin-D1 was also detectable in the cytoplasm of the ROCK-inhibited cells (Fig. 4) . At 12, 18, and 24 hours, cyclin-E1 was detectable in the nuclei of the control cells; however, it was located more in the cytoplasm of the ROCK-inhibited cells, mostly in the perinuclear regions and less in the nuclei (Fig. 5) . By 36 hours, cyclin-E1 was detectable in the nuclei of the ROCK-inhibited cells. This finding suggested that there was a delay in the nuclear translocation of cyclin-E on ROCK inhibition. 

To determine whether the altered levels of these G₁ cyclins in Y27632-treated cells were due to the inhibition of the transcription of their mRNAs, relative levels of their mRNAs were analyzed by real-time RT-PCR. The relative levels of cyclin-D1 mRNAs were 3.8 ± 0.3- and 3.9 ± 1.1-fold lower (P < 0.05) in the cells treated with Y27632 for 18 and 36 hours, respectively, than in the control untreated cells (Fig. 6) . Cyclin-D3 mRNA levels were 2.8 ± 0.2- and 2.7 ± 0.2-fold lower (P < 0.05) in cells treated with Y27632 for 18 and 36 hours, respectively, than in the control untreated cells (Fig. 6) . Cyclin-E1 mRNA levels were not significantly affected by the inhibition of ROCK activity (Fig. 6) . 

ROCK inhibition also decreased the intracellular protein levels of cyclin-dependent kinases CDK4 and CDK6 (subunits of cyclin-D1 and -D3) but not of CDK2 (subunit of cyclin-E) during G₁ progression (Fig. 7 , *P < 0.05). Immunocytochemical analyses showed intensities of staining of nuclear CDK4 and -6 were significantly lower in the cells treated for 12, 18, 24, and 36 hours with Y27632 than in the control nontreated cells (results of 12- and 24-hour treatment are shown in Fig. 8 ). The average intensities of nuclear staining for CDK4 and -6 were 1.6- and 2.1-fold lower, respectively, in the ROCK-inhibited cells than in the controls at 24 hours. Unlike CDK4, CDK6 was also located in the cytoplasm of the ROCK inhibited cells (Fig. 8) , suggesting either inhibition of nuclear translocation or increased export of this protein. At 12, 18, and 24 hours, immunostaining of CDK2 was localized in the nuclei of most of the control cells; however, it was present in the cytoplasm (concentrated in perinuclear regions) of most of the ROCK-inhibited cells between 12 and 24 hours and was translocated to nuclei by 36 hours (Fig. 5) . 

The relative levels of the mRNAs encoding CDK4 were 2.2 ± 0.06- and 1.5 ± 0.08-fold lower in the cells treated with Y27632 for 18 and 36 hours, respectively, than in the control untreated cells (Fig. 9) . The relative levels of CDK6 mRNA were 1.76 ± 0.26- and 2.1 ± 0.6-fold lower (P < 0.05) in the cells treated with Y27632 for 18 and 36 hours, respectively, than in the control untreated cells (Fig. 9) . Similarly, mRNAs of CDK2 were 1.7 ± 0.3- and 1.5 ± 0.24-fold lower (P < 0.05) in the cells treated with Y27632 for 18 and 36 hours, respectively, than in the control untreated cells (Fig. 9) . 

The intracellular levels of the CDK inhibitor, p27^Kip1, were increased when ROCK activity was inhibited during the cell-cycle progression, as evident from the semiquantitative Western blot analysis (Fig. 10) . The intensities of nuclear staining of p27^Kip1 were significantly higher in the cells treated with Y27632 for 12, 18, 24, and 36 hours (results of 12 and 24 hours are shown in Fig. 11 ). The average intensity of nuclear staining for p27^Kip1 was 2.2-fold higher in the ROCK-inhibited cells than in the control at 24 hours. However, the levels of mRNA encoding p27^Kip1 were not significantly different in the cells treated with Y27632 (Fig. 9) . 

Discussion

In the past few years, the role of Rho/ROCK signaling in proliferation has become evident from the antiproliferative effects of the ROCK inhibitor, Y27632, documented in many normal and tumor cell types in vitro. ^{14 16 19 32 37 38 39 40} However, the mechanisms of Rho/ROCK signaling pathways are not well understood. ROCK inhibition has been reported to produce a decrease in cyclin-D1 expression and an increase in p27^Kip1, but no change in p21^Cip1 in hepatocytes. In cardiomyocytes, ROCK inhibition results in decreased expression of cyclin-D3, CDK6 and p27^Kip119, whereas ROCK activation was found to increase cyclin-D1 and p21^cip1 and decrease cyclin-A and p27^Kip1 in NIH 3T3 cells. ⁴¹ These differences in the effects of ROCK inhibition in different cell types suggest that the downstream effects of Rho/ROCK signaling may be cell-type specific. Our previous studies showed that ROCK expression is regulated during different phases of the cell cycle in corneal epithelial cells. In the present study we investigated whether the progression of G₁/S phase was dependent on ROCK activity. 

When RCECs in culture were arrested in G₀ by serum deprivation and then allowed to re-enter the cell cycle by providing serum in the culture medium these cells progressed through the G₁ phase and entered into the S phase within 24 hours. However, the inhibition of ROCK activity delayed the progression into the G₁ phase and entry into the cell cycle by approximately 12 hours. The expression of distinct types of cyclins and fluctuation in their levels regulates the activity of specific CDKs and the progression of the cell cycle. The activation of quiescent cells with growth factors induces the expression of D types of cyclins, including D1, D2, and D3. ⁴² D-type cyclins form complexes with CDK4 and -6, whose major substrate is retinoblastoma protein (Rb). For the entry of cells into the S phase, transcription factor E2F bound to hypophosphorylated retinoblastoma protein (Rb) has to be released. E2F encodes the products required for progression of the cell cycle through S, G₂, and M phases. ⁴³ Activation of cyclin-E/CDK2 completes phosphorylation of Rb and leads to progression of the cell cycle into the S phase. In the present study, ROCK inhibition was found to downregulate the levels of proteins as well as the mRNAs of cyclin-D1, cyclin-D3, CDK4, and CDK6. These findings suggest that ROCK activity increases the expression of these proteins during the progression of G₁/S by upregulating the transcription of the genes encoding these proteins. ROCK inhibition of RCE cells did not affect the intracellular levels of CDK2 or cyclin-E during G₁ and G₁/S progression, but it delayed their nuclear translocation. Therefore, delayed entry of ROCK-inhibited cells into the S phase may be explained by delayed translocation of cyclin-E/CDK2 into the nuclei. Although as a general rule, CDK genes are constitutively expressed and are relatively stable, ROCK inhibition was found to decrease mRNA and the protein levels of CDK4 and -6 and mRNA levels of CDK2. However, cyclin genes have been known to regulate expression and degradation, and ROCK inhibition was found to downregulate mRNA and protein levels of cyclin-D1 and -D3. 

The activities, intracellular translocation, and degradation of cyclins and CDKs are regulated by their specific posttranslational modifications by phosphorylation. ^{44 45} After the CDKs assemble with their cyclins, their full activity is achieved by phosphorylation of the CDKs on a conserved threonine residue. Ubiquitin-mediated degradation of cyclins is initiated by phosphorylation of specific threonine residues. The D types of cyclins are localized to the nucleus during the G₁ phase of the cell cycle and transported to the cytoplasm as the cells enter the S phase. In the present study, cyclin-D1 and -E and CDK6 and -2 were present in the cytoplasm and at significantly lower levels in the nuclei of the ROCK-inhibited cells during the G₁ phase of the cell cycle. Whether this was due to the inhibition of the translocation of these proteins to the nuclei or to the induction of their export from the nuclei remains to be investigated. Whether ROCK plays a role in the phosphorylation of these proteins at specific sites and thereby either promotes activation or degradation of these cyclins and CDK2 is not known. It is interesting to note that ROCK inhibition led to a delay in the translocation of cyclin-E and CDK2 to the nuclei. 

The activity of cyclin-E/CDK2 has been known to be inhibited by the CDKIs p21^Cip1 and p27^Kip1, however, in low concentrations these CDKIs coactivate cyclin-CDKs. In the present study, although ROCK inhibition during G₁ progression upregulated protein levels of p27^Kip1, it did not affect the levels of mRNA. Although increases in the levels of p27^Kip1 have been reported in hepatocytes, an opposite effect is reported in cardiomyocytes and mouse 3T3 fibroblasts. ^{17 19} Neither of these studies determined whether the levels of mRNA encoding p27^Kip1 were regulated by ROCK. Our observation that ROCK inhibition increased the levels of p27^Kip1 without significantly altering the levels of its mRNA suggested that upregulation of p27^Kip1 possibly resulted from its decreased degradation in the ROCK-inhibited cells. Ubiquitin-mediated degradation of p27^Kip1 occurs during the G₀/G₁ as well as the G₁/S and the S phases of the cell cycle. Therefore, ROCK may be important in the mediation of this process during early entry of cells into the G₁ and S phases. Degradation of p27^Kip1 by the ubiquitin/proteosome pathway is initiated by its phosphorylation. During the G₁/S transition and in the S phase, p27^Kip1 is catalyzed at Thr187 by cyclin-E/CDK2 ^{46 47} and is degraded in the nucleus by binding to Skp2, the F-box protein component of an SCF ubiquitin ligase. ^{48 49} However, major phosphorylation of p27^Kip1 occurs at Ser-10 during the G₀/G₁ transition and is necessary for its export from the nucleus to cytoplasm, where it is degraded. ⁵⁰ In response to mitogenic stimulation in corneal endothelial cells, phosphorylation of p27^Kip1 at Ser-10 followed by its nuclear export and degradation ubiquitinated pp27^Kip1 Ser-10 in the cytoplasm occurs earlier than phosphorylation of p27^Kip1 at Thr-187 followed by degradation of ubiquitinated pp27^Kip1 Thr-187 in the nucleus. ⁵¹ These kinetic studies suggest the existence of two populations of p27^Kip1 that get phosphorylated at Ser-10 and Thr-187, respectively. Involvement of ROCK in the regulation of posttranslational modification and degradation of p27^Kip1 remains to be investigated. 

The findings suggest that the decreased levels of cyclin-D1 and -D3 and CDK4 and -6 and increased levels of p27^Kip1 in ROCK-inhibited RCECs and delayed nuclear translocation of CDK2 and cyclin-E may have resulted in a delay in hyperphosphorylation of Rb, the release of E2F, and the entry of the cells into the S phase. In conclusion, Rho/ROCK signaling influenced G₁/S progression in RCECs, at least in part, by upregulation of the transcription of cyclin-D1 and -D3 and CDK4 and -6 and by decreasing the levels of p27^Kip1. 

 Supported by National Eye Institute Grant EY03263 (NS), Core grant EY008098; Research to Prevent Blindness, and The Eye and Ear Foundation (Pittsburgh, PA).

 Submitted for publication April 23, 2007; revised September 15 and October 4, 2007; accepted November 21, 2007.

 Disclosure: J. Chen, None; E. Guerriero, None; K. Lathrop, None; N. SundarRaj, 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.

 

Primers Used in Quantitative RT-PCR

Primer Names	GenBank ID	Oligonucleotide Sequence
18S	X00640	Forward: 5′-CTCAACACGGGAAACCTCAC-3′
		Reverse: 3′-ACCACCCACAGAATCGAG AA-3′
Cyclin D1	DQ845180	Forward: 5′-ATGGAACTGCTCCTGGTGAA-3′
		Reverse: 5′-GGAGAGGAAGTGCTCGATGA-3′
Cyclin D3	DQ845182	Forward: 5′-TACTGGATGCTGGAGGTGTG-3′
		Reverse: 5′-GGTAGCGATCCAGGTAGTTCA-3′
Cyclin E1	EU137106	Forward: 5′-CAGGTTGCGTACCTGAATGA-3′
		Reverse: 5′-CGACATCCAGGACACAGAGA-3′
CDK2	EU137107	Forward: 5′-GGTCCTGCACCGAGATCTAA-3′
		Reverse: 5′-ACCACCTCGTGGGTGTAAGT-3′
CDK4	DQ845183	Forward: 5′-CGCTTACACCTGTGGTTGTG-3′
		Reverse: 5′-CACAGGCGTGGCATATGTAG-3′
CDK6	DQ845184	Forward: 5′-TGGTACCGAGCTCCAGAAGT-3′
		Reverse: 5′-CAAATATGCAGCCGACACTC-3′
p27kip1	DQ845181	Forward: 5′-CGCCTGCAGAAACCTCTTC-3′
		Reverse: 5′-CCATGTCTCTGCAGTGCTTCT-3′

Figure 1.

 

BrdU incorporation into corneal epithelial cells. RCECs in passage one (P1) were arrested in G₀ by serum deprivation and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were treated for 6 hours with BrdU before they were fixed and subjected to indirect immunofluorescence staining with anti-BrdU antibody and nuclear counterstaining with PI.

Figure 2.

 

Effects of ROCK-inhibition on the progression of cells from G₀ to S phase. RCECs (P1), arrested in G₀, were allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were treated with BrdU for 6 hours before the analysis. A significantly lower percentage of total Y27632-treated cells than nontreated control cells entered the S phase between 12 and 24 hours (*P < 0.05).

Figure 3.

 

Western blot analyses of cyclin-D1, -D3, and -E. RCECs in passage 1 (P1) were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were extracted in RIPA buffer at the times indicated, and the extracts containing 20 μg of total proteins were loaded per lane for Western blot analysis. Top: representative Western blot. Bar graph, the mean ± SD of the relative intensities of the bands in the Western blots (from three independent experiments), normalized with densities of major protein bands in the blots stained with Coomassie blue before the analyses (*P < 0.05).

Figure 4.

 

Immunocytochemical analyses of the effects of ROCK-inhibition on cyclin-D1 and -D3 expression and distribution in RCECs. RCECs in passage 1 (P1), were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were fixed, permeabilized, and stained with anti-cyclin-D1 or -D3 antibody by an indirect immunofluorescence technique. There were significantly lower intensities of nuclear fluorescence in the Y27632-treated cells stained for cyclin-D1 and -D3 than in the control.

Figure 5.

 

Immunocytochemical analyses of the effects of ROCK-inhibition on cyclin-E1 and CDK2 expression and distribution in RCECs. RCECs in passage 1 (P1), were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were fixed, permeabilized, and stained with anti-cyclin-E1 and anti-CDK2 antibodies by an indirect immunofluorescence technique. Staining was noted in the cytoplasm and significantly lower intensities of nuclear staining were seen in Y27632-treated cells at 24 hours than in the control cells, with increased nuclear staining in Y27632-treated cells at 36 hours.

Figure 6.

 

Relative concentrations of mRNA transcripts of cyclin-D1, -D3, and -E1. RCECs in passage 1 (P1) were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). Total RNA was extracted from the cells at 18 and 36 hours, and the relative concentrations of mRNA transcripts of cyclin-D1, -D3, and -E1 were analyzed with real-time RT-PCR with values normalized to levels of 18S rRNA. The values presented in the bar graph are the mean ± SD of relative mRNA levels compared with those of G₀ cells (as 100) in three experiments. Reduced mRNA levels of cyclin-D1 and -D3 were noted in the cells treated with Y27632 (*P < 0.05).

Figure 7.

 

Western blot analysis of CDK4, -6, and -2. RCECs in passage 1, arrested in G₀, were allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were extracted in RIPA buffer at the times indicated, and the extracts containing 20 μg of total proteins per lane were loaded for Western blot analysis. Top: representative Western blot; bottom: mean ± SD of the relative intensities of the bands in the Western blots (from three independent experiments), normalized with densities of major protein bands in corresponding lanes in the blots stained with Coomassie blue before the analyses (*P > 0.05).

Figure 8.

 

Immunocytochemical analyses of the effects of ROCK-inhibition of CDK4 and -6 expression and distribution in corneal epithelial cells. RCECs in passage 1 (P1), were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were fixed, permeabilized, and stained with anti-CDK4 or -6 antibodies by an indirect immunofluorescence technique. Significantly lower intensities of nuclear fluorescence were noted in the Y27632-treated cells.

Figure 9.

 

Relative concentrations of mRNA transcripts of CDK4, -6, and -2 and p27^Kip1. RCECs in passage 1 (P1) were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). Total RNA was extracted from the cells at 18 and 36 hours and relative concentrations of mRNA transcripts of CDK4, -6, and -2 and p27^Kip1 were analyzed with real-time RT-PCR with values normalized to 18S rRNA levels. The values presented in the bar graph are the mean ± SD of relative mRNA levels compared with those of G₀ cells (as 100) in three experiments (*P < 0.05).

Figure 10.

 

Western blot analyses of p27^Kip1. RCECs in P1 were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were extracted in RIPA buffer at the times indicated, and the extracts containing 20 μg of total proteins per lane were loaded for Western blot analysis. Top: representative Western blot; bottom: the mean ± SD of the relative intensities of the bands in the Western blots (from three independent experiments), normalized with densities of major protein bands in corresponding lanes in the blots stained with Coomassie blue before the analyses (P < 0.05).

Figure 11.

 

Immunocytochemical analyses of the effects of ROCK inhibition on p27^Kip1 expression and distribution in RCECs. RCECs in passage 1 (P1) were arrested in G₀ and then allowed to re-enter the cell cycle in medium containing 10% serum, with or without 10 μM Y27632 (ROCK inhibitor). The cells were fixed, permeabilized, and stained with anti-p27^Kip1 antibody by an indirect immunofluorescence technique. Significantly higher intensities of nuclear fluorescence were noted in the Y27632-treated cells than in the control.

The authors thank Cindy Stone, Lisa DiCesare, and Ryan Eberwine for technical help.