Molecular Vision 2005; 11:133-142 <http://www.molvis.org/molvis/v11/a15/> Received 25 October 2004 | Accepted 11 February 2005 | Published 18 February 2005

Download Reprint

Reduction of Pnn by RNAi induces loss of cell-cell adhesion between human corneal epithelial cells

Jeong-Hoon Joo, Roman Alpatov, Gustavo C. Munguba, Moira R. Jackson, Marguerite E. Hunt, Stephen P. Sugrue
 
 

Department of Anatomy and Cell Biology, University of Florida College of Medicine, Gainesville, FL

Correspondence to: Dr. Stephen P. Sugrue, Department of Anatomy and Cell Biology, University of Florida College of Medicine, PO Box 100235, Gainesville, FL, 32610-0235; Phone: (352) 392-3569; FAX: (352) 392-3305; email: sugrue@anatomy.med.ufl.edu

Purpose: Pinin (Pnn/DRS/memA) plays an important role in regulating cell-cell adhesion of corneal epithelial cells. In the nucleus, Pnn interacts with both transcriptional repressor and pre-mRNA processing machinery. Here we investigated the consequences of "knocking down" Pnn expression with short hairpin RNAi (shRNAi) on the corneal epithelial cell phenotype.

Methods: Cultured human corneal epithelial (HCE-T) cells were cotransfected with a shRNAi-expressing construct containing an inverted repeat of a Pnn specific 21 nucleotide sequence (Pnn shRNAi) and a GFP vector as a marker of transfected cells. After 24-48 h, cells were fixed and immunostained with antibodies against Pnn, keratin, desmoplakin, desmoglein, E-cadherin, ZO-1, SR-proteins, and SRm300. To demonstrate specificity of the Pnn knock down, a rescue vector was designed by incorporating three conservative nucleotide substitutions within the Pnn-shRNAi targeting sequences of the full length Pnn-GFP construct, thus generating a Pnn construct to produce mRNA that Pnn shRNAi could not target (Pnn-CS3-GFP).

Results: HCE-T cells were cotransfected with Pnn shRNAi and GFP vectors and after 24 and 48 h exhibited significantly reduced immunostaining for Pnn. Western blot analyses of Pnn and E-cadherin protein expression in cells transfected with Pnn-shRNAi and GFP vectors revealed marked reduction in levels of both proteins compared to those observed in cells transfected with GFP alone. The cells receiving Pnn-shRNAi appeared to be less adherent to neighboring nontransfected cells, often exhibited altered cell shape, downregulated cell adhesion and cell junction molecules, and escaped from the epithelium. The Pnn shRNAi transfected cells exhibited fewer keratin filaments anchored to desmosomes and a concurrent increase in the perinuclear bundling of filaments. SR proteins and SRm300 showed an altered distribution in the Pnn knock down cells. Cotransfection of Pnn-CS3-GFP with Pnn shRNAi demonstrated that the conservatively mutated Pnn could maintain cell-cell adhesion.

Conclusions: Our results indicate that knocking down Pnn expression leads to a loss of epithelial cell-cell adhesion, changes in cell shape, and movement of Pnn shRNAi transfected cells out of the epithelium. We suggest that Pnn plays an integral role in the establishment and maintenance of epithelial cell-cell adhesion via its activity within nuclear multi-protein complexes.

Introduction

Competent corneal epithelial barrier function is dependent upon the maintenance of cell-cell and cell-matrix adhesions. We have previously demonstrated that Pnn, a 140 kDa phosphoprotein, is expressed in corneal epithelium and, subsequent to debridement wounding, desmosome associated Pnn is absent from the cells at the migrating edge of the closing wound and returns only when the wound is closed and new desmosomes have been formed [1]. Furthermore, over-expressing Pnn in an in vitro wounded epithelial monolayer drives the cells to a more epithelial phenotype with increased E-cadherin expression and abrogates migration of cells and wound closure [1]. Yeast two hybrid studies showed that Pnn binds to keratins 8, 18, and 19 via its amino terminal [2], while the carboxyl portion of Pnn binds to nuclear proteins involved in splicing and transcriptional repression [3,4], providing evidence that Pnn may be one of a growing number of proteins that are thought to play dual roles in both cytoskeletal anchorage at the membrane and in nuclear events.

Endogenous and exogenously expressed Pnn are observed in the nucleus, diffusely throughout the nucleoplasm, and in more prominent nuclear speckles. Nuclear speckles are reservoirs from which proteins involved in transcription and/or pre-mRNA processing are recruited. There is compelling evidence that Pnn may play a role in pre-mRNA splicing. Not only does it interact with RNPS1, a key factor in the exon-junction complex, but it may impact alternative pre-mRNA splicing by decreasing the use of distal splice sites [5,6]. Pnn also interacts with a subset of serine-arginine (SR) rich proteins; SRp75, SRm300, and a novel protein identified in our laboratory, SRrp130 [3]. Moreover, it appears to associate preferentially with spliced mRNA [7] leading to the suggestion that it may be involved in mRNA export. We have recently shown that Pnn interacts with the transcriptional co-repressor, c-terminal binding protein (CtBP), via a PEDLS domain found near Pnn's c-terminal, and Pnn relieves CtBP mediated repression of the E-cadherin promoter [4]. In addition, Dellaire and coworkers [8] demonstrated that Pnn may be part of a repressor complex including PRP4 kinase and N-CoR. Thus, Pnn appears to play a role in the regulation, perhaps the coordination of transcription and pre-mRNA splicing, and desmosomal stability, raising the intriguing possibility that these processes may be intricately interlinked.

In addition to the upregulation of E-cadherin, cDNA array analysis following Pnn over-expression revealed changes in cell cycle related genes such as p21^cip/waf, CDK4, CPR2, and several cell migration and invasion regulatory genes such as RhoA, CDK5, TIMP1, MMP7, and EMMPRIN [9]. The net effect of these changes is to drive the cell to a more epithelial phenotype with enhanced cell-cell adhesion and inhibition of cell proliferation and migration, indicative of a potential tumor suppressor. Indeed, the human Pnn gene localizes within the 14q13 region, believed to be a potential tumor suppressor locus. Aberrant methylation of CpG islands of the Pnn promoter has been observed in some tumors displaying decreased/absent Pnn expression [10].

To determine whether the loss of Pnn would result in a loss of cell-cell adhesion, we designed a shRNAi construct to drive RNAi mediated depletion of Pnn mRNA. Our results indicate that knocking down Pnn expression leads to a loss of epithelial cell-cell adhesion, change in cell shape, and movement of Pnn shRNAi transfected cells out of the epithelium. We also demonstrate that addition of exogenous "rescue" vector (Pnn-CS3-GFP), which contained Pnn sequences harboring three conservative substitutions in the region to which the Pnn shRNAi was designed, effectively replaces the endogenous Pnn and protects from loss of cell-cell adhesion. These data support the contention that Pnn may function as a key regulator in the establishment and maintenance of epithelial cell-cell adhesion.

Methods

Cell culture and transfections

The SV-40 immortalized human corneal epithelial cell line (HCE-T, RCB1384) was generously provided by Kaoru Araki-Sasaki (Osaka University, School of Medicine, Osaka, Japan). Corneal epithelial cells were cultured in DMEM/F12 media containing 200 U/ml each of penicillin and streptomycin, 5% FBS (Cellgro, Mediatech, Herndon, VA), 0.1 μg/ml cholera toxin, 5 μg/ml insulin, 0.5% dimethyl sulphoxide (all Sigma-Aldrich, St Louis, MO) and 10 ng/ml human epidermal growth factor (Invitrogen-Gibco, Gaithersburg, MD). HCE-T cells were transfected at 60% confluency using 3 μl of 1 μg/ml 25 kDa branched polyethylenimine, pH 7.0 (Sigma-Aldrich) per 1 μg of DNA [11]. HCE-T cells, after transfection with Pnn shRNAi and GFP vectors, were also plated at 90% confluency in 24 well filter inserts with 8.0 μm pore size PET track-etched membrane (BD FALCON^TM, BD Biosciences, Bedford, MA). After the first 12 h, the cells were placed at the air-liquid interface for 2-5 days followed by fixation using 4% paraformaldehyde. The epithelia were then embedded in OCT and cryostat sectioned for immunostaining. HCE-T cell line retrovirally transduced to stably express epitope tagged Pnn (Pnn-Flag-HA) was established as previously described [4].

Immunocytochemistry

HCE-T cells were grown and transfected on cover slips. After 24 and 48 h they were fixed in methanol at -20 °C for 2 min and washed in phosphate buffered saline (PBS). Primary antibodies diluted in PBS were applied for 1 h at room temperature. These included monoclonal antibodies recognizing E-cadherin and desmoglein (BD Transduction Labs., San Diego, CA), desmoplakin and β-catenin (American Research Products, Belmont, MA), Keratin (SigmaAldrich), keratin 18 (Neomarkers, Freemont, CA), ZO-1, occludin, and SR proteins (Zymed, South San Francisco, CA). Polyclonal antibodies recognizing GFP, and ZO-1 (Molecular Probes, Eugene, OR) and SRm300 (generous gift from B. J. Blencowe, University of Toronto, Ontario) were also utilized. Pnn 2.2.143 mAB in hybridoma supernatant was applied undiluted [4]. Rhodamine conjugated secondary antibodies (Chemicon International, Temecula, CA) were applied for 1 h and, after further PBS washes, cover slips were mounted using Vectashield containing DAPI (Vector Laboratories, Inc., Burlingame, CA). Specimens were examined using a Leica DMR microscope and images were recorded using IP Lab software (Scanalytics inc., Fairfax, VA).

Quantification of immunofluorescence and western blot data

Nuclear pixel brightness intensities were averaged for 20 cells per experimental group and expressed as mean±standard deviation using IP Lab software (Scanalytics Inc., Fairfax, VA). In SRm300 experiments, the mean intensity of brightness and area of abnormally large speckles were obtained simultaneously. The intensity of western blot bands was quantified by densitometry and normalized to α-tubulin levels in each lane.

FACS selection of cells expressing GFP

HCE-T cells were transfected with Pnn shRNAi and GFP vectors or GFP vector alone. After 24 h cells were trypsinized and sorted by fluorescence activated cell sorting (FACS) to isolate transfected (GFP positive) cells. These cells were then lysed for western blot analysis.

Immunoblotting

Cells were lysed in RIPA buffer (150 mM NaCl, 1% NP-40, 0.5% deoxycholine, 0.1% SDS and 50 mM Tris, pH 8.0), containing Complete^TM protease inhibitors (Roche Diagnostics, Mannheim, Germany), 200 μM phenylmethylsulfonyl fluoride, and 10 mM β-mercaptoethanol (both Sigma-Aldrich) at 4 °C for 15 min and extracts were spun at 14,000 rpm. Supernatants were then boiled in SDS sample buffer, run on 4-15% gradient polyacrylamide gels and transferred to Immobilon polyvinylidene difluoride membrane. After blocking in 5% nonfat milk in PBST (PBS containing 0.1% Tween 20), membranes were incubated in 2% milk containing primary antibodies against E-cadherin, GFP, α-tubulin (Sigma-Aldrich), or undiluted Pnn 2.2.143 hybridoma supernatant for 1 h. After six 5-min washes in PBST, membranes were incubated in secondary antibodies conjugated to HRP (Amersham Biosciences, Arlington Heights, IL) for 30 min, washed, and proteins of interest were visualized using the ECL detection system according to manufacturer's instructions (Amersham Biosciences).

Construction of Pnn shRNAi vector

As a means of exploring the function of Pnn by knocking down its expression in mammalian cells, we constructed an shRNAi expressing vector harboring a U6 promoter and an inverted repeat of Pnn specific 21 nucleotide sequences. Nucleotides 919-939 of the human Pnn coding region encompassing amino acids 307-313 (GKVAQRE) of human Pnn were selected after BLAST analysis (NCBI). Forward and reverse oligos of Pnn specific sequences with appropriate restriction sites were annealed and ligated into the parental plasmid BS/U6 [12] (gift from Y. Shi, Harvard University) between ApaI and HindIII or HindIII and EcoR1 sites. The vector was verified by sequencing.

Construction of Pnn shRNAi rescue vector

A construct containing a GFP-tagged full length human Pnn sequence (HpGFP) was modified by sequential introduction of three conservative substitutions in the shRNAi-targeted region using the QuickChange XL Site Directed Mutagenesis Kit (Stratagene, La Jolla, CA) as per the manufacturer's instructions. The T at 921, G at 927, and T at 930 were mutated to A, C, and A, respectively, ensuring that the amino acids encoded for were identical to those in wild type Pnn (namely G, V, and A). After sequence verification, this Pnn-CS3-GFP construct was cotransfected into cells with the Pnn shRNAi construct to determine whether the cells would express and use the exogenous full length Pnn.

Results

HCE-T cells exhibit well developed adhesion complexes

HCE-T cells both grown on coverslips, or at air-media interface on filter inserts, stably retained their epithelial morphology and phenotype. Immunostaining these cells for adhesion and junction related components demonstrated extensive cell-cell adhesion with an abundance of desmosomes and an apically placed zonula occludens junction (Figure 1). This model system has been shown to be an appropriate tissue equivalent upon which to develop relevant in vitro methods to explore corneal epithelial cell barrier function and physiology [13].

Pnn shRNAi mediated interference abrogates Pnn expression in human corneal epithelial cells

HCE-T cells were cotransfected with Pnn shRNAi and GFP vectors and after 24 and 48 h cells were fixed and Pnn was then detected with mAB 2.2.143. Pnn was observed in nuclear speckles in untransfected cells (mean relative intensity 66.81±13.59), while immunostaining of Pnn in transfected (GFP containing) cells was significantly reduced (11.68±7.74) indicating effective RNAi mediated Pnn mRNA degradation and abrogation of Pnn protein synthesis (Figure 2A,B). In addition, the cells receiving the Pnn-shRNAi appeared to be less adherent to neighboring nontransfected cells and often exhibited altered cell shape with elongated spindle shaped morphology and long cellular processes, which extend either between or on the top of adjacent cells. Transfection of an irrelevant shRNAi (CDK2) did not affect Pnn expression, cell-cell adhesion, nor cell shape (data not shown).

RNAi knockdown of Pnn results in alteration of the distribution of known Pnn associated proteins

Examination of HCE-T cells cotransfected with Pnn shRNAi and GFP vectors revealed a dramatic alteration in keratin filament organization. The Pnn shRNAi transfected cells exhibited fewer keratin filaments anchored to desmosomes and a concurrent increase in the perinuclear bundling of filaments (Figure 2C). The cells also appeared more rounded and somewhat out of the plane of the epithelium, again exhibiting elongated cellular processes that extend between neighboring cells. Within the nucleus, both the family SR proteins and specifically SRm300 showed an altered distribution in the Pnn knock down cells. SR proteins, when visualized with an antibody, which recognizes a motif expressed within a subset of approximately 20 non-snRNP splicing factors, exhibited a typical nuclear speckle distribution. However, in the Pnn knockdown cells the SR staining demonstrated larger speckles that appeared smoother with fewer nucleoplasmic extensions (Figure 2D). SRm300, a nuclear speckle protein that has been shown to specifically bind to Pnn [3], exhibited an increase in the size (95.15±36.42 compared to 234.7±75.45 pixels) and staining intensity (82.38±21.07 compared to 218.1±16.29) of the speckles (Figure 2E). These data suggest that Pnn may play fundamental roles in the organization of the keratin filament array and the nuclear speckle compartment.

Pnn-RNAi expression resulted in reduction in levels of Pnn and E-cadherin proteins

Previously, we have shown that exogenous expression of Pnn in HEK-293 and MDCK cells was associated with an increase in cell-cell adhesion and E-cadherin expression levels [9]. In order to investigate whether knocking down Pnn expression levels with RNAi might affect E-cadherin levels, HCE-T cells cotransfected with Pnn shRNAi and GFP or GFP alone, were analyzed by western blotting. Both Pnn-shRNAi/GFP and GFP only cells were sorted by FACS at 24 h post-transfection. Approximately 7.5x10⁴ GFP positive cells from each sample were sorted and subsequently lysed in SDS sample buffer and resolved on SDS PAGE. Western blot analyses of Pnn and E-cadherin protein expression in cells transfected with Pnn-shRNAi and GFP vectors revealed a marked reduction in levels of both proteins compared to those observed in cells transfected with GFP alone (Figure 3). Levels of α-tubulin illustrate equal loading of total protein between lanes. These data indicate that the RNAi methodology was indeed effective in knocking down its target and that the expression of Pnn was tightly linked to the expression of E-cadherin.

Knock down of Pnn results in loss of E-cadherin, the reduction of several cell junction associated proteins, and dramatic changes in cell shape and cell-cell associations

Interfering with Pnn expression by shRNAi induced dramatic changes in corneal epithelial cell-cell contacts and, in turn, cell shape. The Pnn-RNAi transfected cells appeared to weaken their contacts with neighboring cells, downregulate cell adhesion and cell junction molecules and escape from the epithelium. E-cadherin immunostaining was reduced between adjacent transfected cells (Figure 4A), and the cells appeared to be crawling over the neighboring cells. Sectioning of the transfected epithelia revealed that the cells expressing Pnn-RNAi appeared on top of the epithelia, often exhibiting long cellular processes (Figure 4B,D,E). Desmosomal and tight junctional integrity also appeared to be compromised in transfected cells with reduced levels of desmoplakin (Figure 4C) and desmoglein (Figure 4D) and the presence of gaps between neighboring transfected cells, while membrane localization of the tight junction associated protein ZO-1 was also reduced and more punctate (Figure 4E,F). Reductions in all these junctional proteins resulted in weaker cell-cell adhesion and an alteration from an epithelial to a more elongate cell morphology, which impacted negatively on epithelial monolayer integrity.

Pnn-CS3-GFP reverses the effects of Pnn shRNAi

To ascertain whether the aforementioned changes in cell phenotype were due to a reduction in Pnn specifically, in contrast to a secondary indirect RNAi-phenomenon, we sought to express a Pnn rescue construct containing three conservatively mutated residues (CS3-Pnn), to which the Pnn shRNAi could not anneal. Cotransfection of this rescue CS3-Pnn in the presence of Pnn shRNAi would address whether exogenous CS3-Pnn can protect from the effects of Pnn-RNAi (Figure 5A). Wild type (HpGFP) and mutated Pnn-GFP containing one (CS1), two (CS2), or three (CS3) conservative substitutions were transfected along with Pnn-shRNAi. Pnn shRNAi successfully abrogated exogenous expression of wild type (ratio of Pnn:α-tubulin intensities-0.2), partially diminished expression of CS1-Pnn (intensity ratio 0.18) and CS2-Pnn (intensity ratio 0.2), but was unable to prevent synthesis of Pnn encoded by the CS3-Pnn construct (intensity ratio 2.0, Figure 5B). The Pnn-CS3-GFP construct or HpGFP construct were then transfected into HCE-T cells with or without Pnn shRNAi. After 24 h cells were lysed as described, run on SDS PAGE, and western blotted for GFP and/or Pnn. Pnn shRNAi abrogated production of wt-HpGFP, but did not abrogate synthesis of Pnn-CS3-GFP protein (Figure 5C). Cotransfection of Pnn-CS3-GFP with Pnn shRNAi into HCE-T cells that stably expressed Pnn-Flag-HA [4] was then performed. After 24 h cells were fixed and antibodies against HA, E-cadherin or desmoplakin were applied. Pnn-CS3-GFP protein was observed in nuclear speckles in transfected cells whereas Pnn-Flag-HA exhibited the RNAi mediated knockdown. Furthermore, levels of E-cadherin and desmoplakin in Pnn-CS3-GFP cells appeared no different than untreated or GFP transfected controls (Figure 5D). These data suggest that the changes in cell shape and adhesion complexes observed after transfection of Pnn-shRNAi vector into HCE-T cells were indeed specifically attributable to knocking down Pnn.

Discussion

RNA interference (RNAi) silences gene expression by the sequence specific degradation of mRNAs. Employing shorter interfering RNA (siRNA) duplexes <30 bp long circumvents both the induction of dsRNA induced cell death and nonspecific gene silencing initially observed with longer sequences [14]. siRNAs can be constitutively expressed in mammalian cells by using vectors that contain RNA polymerase type III promoters such as the U6 small nuclear RNA promoter and drive the expression of short hairpin RNA (shRNAi). We employed this technology to specifically knock down expression of endogenous Pnn in HCE-T cells. Cotransfection of GFP with Pnn shRNAi enabled us to identify and observe transfected cells uncoupling from their neighbors and to select transfected cells to assess protein expression levels.

Transfection of Pnn-shRNAi resulted in reduced levels of both Pnn and E-cadherin proteins 24 h post-transfection, lending further support to our earlier observations that Pnn can modulate the expression of E-cadherin [4,9]. One mechanism by which Pnn may influence E-cadherin transcription is by interacting with the transcriptional co-repressor CtBP via its PEDLS motif [4]. Knocking down Pnn may deplete the amount of Pnn available to relieve CtBP mediated repression of E-cadherin expression. Additionally, we have shown loss of Pnn expression in transitional cell carcinoma, renal cell carcinoma, and some metastatic ovarian tumors. In some cases the loss of Pnn expression has been correlated to aberrant methylation of CpG islands in the promoter of the Pnn gene [10]. E-cadherin expression is reduced and its promoter hypermethylated in transitional cell carcinoma [15], while loss of E-cadherin in renal cell carcinoma is prognostically relevant [16]. It is tempting to speculate that loss of Pnn expression may be a crucial step in the progression of various epithelial tumors. The RNAi data presented here indeed suggest that the RNAi mediated downregulation of Pnn is associated with a decrease of cell-cell adhesion, cell shape change, and epithelial cell dysadhesion, all key steps in carcinoma progression.

The formation and stability of adherens junctions and desmosomes are intricately interlinked [17]. Membrane association of both desmoplakin and desmoglein appeared to be reduced and more punctate in HCE-T cells transfected with Pnn shRNAi, perhaps indicative of either a remodeling of, or a reduction in, desmosomal junctional complexes. Consistent with these observations, keratin filament membrane anchorages were sparse and intermediate filament bundles were condensed to the perinuclear region. Similarly, it has been reported that truncation of desmoglein or absence of desmoplakin both result in aberrant anchorage of keratin filaments and condensation of keratin filaments to perinuclear bundles [18,19]. Whether Pnn functions by interacting directly with keratins [2] or by affecting nuclear function, whereby the loss of Pnn in turn down regulates cell-cell adhesion (E-cadherin) and thereby secondarily affects the desmosome, remains to be resolved. Pnn's role may be similar to the dual location and function that has been reported for plakophillin 1 [20]. Finally, the loss of Pnn may impact signaling pathways involving the β-catenin/TCF signaling pathway and thereby have significant consequences on cell-cell adhesion and gene expression.

Knocking down Pnn expression also disrupted membrane association of ZO-1. ZO-1 provides an essential link between occludins and claudins and the actin cytoskeleton, and is salient to tight junction stability and barrier function. ZO-1 is found in the apical layers of both human [21] and rabbit [22] corneal epithelia, and between wing and basal cells where its association with paxillin may serve to reinforce these attachments [22]. N-terminal truncation mutants of ZO-1 disrupt corneal epithlelial cell morphology and suggest a partial transformation to a mesenchymal phenotype [13]. These findings all suggest that Pnn is crucial in the maintenance of epithelial integrity and barrier function. Loss of Pnn expression results in decreased cell-cell adhesion and, perhaps, a more invasive cell, consistent with a role for Pnn as a tumor suppressor or a potential inhibitor of epithelial-mesenchymal transformation.

Down regulation of Pnn led to the accumulation of SR proteins, specifically SRm300 with which Pnn has been shown to interact [3], within nuclear speckles or splicing factor compartments (SFCs). Both SR and SR related proteins [23] are involved at specific stages of spliceosome formation. SRm300 is a nuclear matrix antigen thought to act as a coactivator of pre-mRNA splicing with SRm160 [24]. After transfection of Pnn shRNAi, speckles appeared larger with fewer extensions into the nucleoplasm, reminiscent of those observed after inhibition of transcriptional initiation [25,26]. Reduction in Pnn levels may impact speckle associated kinases as SR protein phosphorylation appears to be a prerequisite for recruitment from SFCs to active sites of transcription [27,28]. Interestingly, Pnn interacts with PRP4K, which is a kinase associated to both coactivator and corepressor complexes [8]. Further support for the interaction between Pnn and the transcriptional corepressor CtBP is observed when transfection of HCE-T cells with CtBP shRNAi results in a partial translocation of Pnn from the nuclear speckle to the cytoplasm [4]. Yeast-two-hybrid analyses indicate that Pnn can interact with BS69 (unpublished results) which can repress both E1A and Ets-2 mediated transcription [29,30]. Thus, Pnn may be a key player in the essential coordination of chromatin mediated regulation of transcription and pre-mRNA processing [7,31]. It is tempting to speculate that this co-ordination might specifically alleviate repression of transcription and processing of a cassette of epithelial related genes and/or repress mesenchymal associated genes.

The conservatively mutated Pnn-CS3-GFP vector is a powerful tool for examining the role of Pnn in specific intracellular processes and its interaction with other proteins. Cotransfection of Pnn shRNAi and Pnn-CS3-GFP resulted in the knockdown of both endogenous and wild type transfected Pnn and forced the cells to use the conservatively mutated exogenous GFP tagged Pnn (CS3). Pnn-CS3-GFP rescued epithelial integrity by maintaining the epithelial phenotype and junctional complexes within Pnn-shRNAi transfected cells. These data confirmed the specificity of Pnn-shRNAi. The strategy of forcing corneal epithelial cells to use the exogenously provided construct, by co-tranfecting shRNAi with rescue vector, will be valuable in combination with the introduction of further mutations and/or deletions in specific regions of the Pnn-CS3-GFP construct. Such strategies will enable further examination of the role of Pnn in corneal epithelial homeostasis and in wound healing. Furthermore, modulation of the level of Pnn expression may be a useful tool in cases where re-establishment of corneal epithelial barrier integrity cannot be achieved.

These data presented in this paper support the contention that Pnn may function as a key regulator in the establishment and maintenance of epithelial cell-cell adhesion. We suggest that Pnn influences epithelial cell-cell adhesion via its activity within nuclear multi-protein complexes. Thus Pnn may functionally bridge transcriptional repression and mRNA processing and thereby have profound consequences on gene expression of the corneal epithelial cell. Future investigations will focus on whether the dramatic influence Pnn exerts on epithelial behavior is strictly through the regulation of E-cadherin or whether Pnn exerts its affect on other epithelial gene expression also.

Acknowledgements

This work was supported by National Institutes of Health grant EY007883 and a Vision Center Core Grant EY008571.

References

1. Shi Y, Tabesh M, Sugrue SP. Role of cell adhesion-associated protein, pinin (DRS/memA), in corneal epithelial migration. Invest Ophthalmol Vis Sci 2000; 41:1337-45.

2. Shi J, Sugrue SP. Dissection of protein linkage between keratins and pinin, a protein with dual location at desmosome-intermediate filament complex and in the nucleus. J Biol Chem 2000; 275:14910-5.

3. Zimowska G, Shi J, Munguba G, Jackson MR, Alpatov R, Simmons MN, Shi Y, Sugrue SP. Pinin/DRS/memA interacts with SRp75, SRm300 and SRrp130 in corneal epithelial cells. Invest Ophthalmol Vis Sci 2003; 44:4715-23.

4. Alpatov R, Munguba GC, Caton P, Joo JH, Shi Y, Shi Y, Hunt ME, Sugrue SP. Nuclear speckle-associated protein Pnn/DRS binds to the transcriptional corepressor CtBP and relieves CtBP-mediated repression of the E-cadherin gene. Mol Cell Biol 2004; 24:10223-35.

5. Wang P, Lou PJ, Leu S, Ouyang P. Modulation of alternative pre-mRNA splicing in vivo by pinin. Biochem Biophys Res Commun 2002; 294:448-55.

6. Sakashita E, Tatsumi S, Werner D, Endo H, Mayeda A. Human RNPS1 and its associated factors: a versatile alternative pre-mRNA splicing regulator in vivo. Mol Cell Biol 2004; 24:1174-87. Erratum in: Mol Cell Biol. 2004; 24:3068.

7. Li C, Lin RI, Lai MC, Ouyang P, Tarn WY. Nuclear Pnn/DRS protein binds to spliced mRNPs and participates in mRNA processing and export via interaction with RNPS1. Mol Cell Biol 2003; 23:7363-76.

8. Dellaire G, Makarov EM, Cowger JJ, Longman D, Sutherland HG, Luhrmann R, Torchia J, Bickmore WA. Mammalian PRP4 kinase copurifies and interacts with components of both the U5 snRNP and the N-CoR deacetylase complexes. Mol Cell Biol 2002; 22:5141-56.

9. Shi Y, Simmons MN, Seki T, Oh SP, Sugrue SP. Change in gene expression subsequent to induction of Pnn/DRS/memA: increase in p21(cip1/waf1). Oncogene 2001; 20:4007-18.

10. Shi Y, Ouyang P, Sugrue SP. Characterization of the gene encoding pinin/DRS/memA and evidence for its potential tumor suppressor function. Oncogene 2000; 19:289-97.

11. Demeneix B, Behr J, Boussif O, Zanta MA, Abdallah B, Remy J. Gene transfer with lipospermines and polyethylenimines. Adv Drug Deliv Rev 1998; 30:85-95.

12. Sui G, Soohoo C, Affar el B, Gay F, Shi Y, Forrester WC, Shi Y. A DNA vector-based RNAi technology to suppress gene expression in mammalian cells. Proc Natl Acad Sci U S A 2002; 99:5515-20.

13. Ryeom SW, Paul D, Goodenough DA. Truncation mutants of the tight junction protein ZO-1 disrupt corneal epithelial cell morphology. Mol Biol Cell 2000; 11:1687-96.

14. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 2001; 411:494-8.

15. Horikawa Y, Sugano K, Shigyo M, Yamamoto H, Nakazono M, Fujimoto H, Kanai Y, Hirohashi S, Kakizoe T, Habuchi T, Kato T. Hypermethylation of an E-cadherin (CDH1) promoter region in high grade transitional cell carcinoma of the bladder comprising carcinoma in situ. J Urol 2003; 169:1541-5.

16. Morell-Quadreny L, Rubio J, Lopez-Guerrero JA, Casanova J, Ramos D, Iborra I, Solsona E, Llombart-Bosch A. Disruption of basement membrane, extracellular matrix metalloproteinases and E-cadherin in renal-cell carcinoma. Anticancer Res 2003; 23:5005-10.

17. Garrod DR, Merritt AJ, Nie Z. Desmosomal cadherins. Curr Opin Cell Biol 2002; 14:537-45.

18. Hanakawa Y, Amagai M, Shirakata Y, Sayama K, Hashimoto K. Different effects of dominant negative mutants of desmocollin and desmoglein on the cell-cell adhesion of keratinocytes. J Cell Sci 2000; 113:1803-11.

19. Vasioukhin V, Bauer C, Degenstein L, Wise B, Fuchs E. Hyperproliferation and defects in epithelial polarity upon conditional ablation of alpha-catenin in skin. Cell 2001; 104:605-17.

20. Schmidt A, Langbein L, Rode M, Pratzel S, Zimbelmann R, Franke WW. Plakophilins 1a and 1b: widespread nuclear proteins recruited in specific epithelial cells as desmosomal plaque components. Cell Tissue Res 1997; 290:481-99.

21. Ban Y, Cooper LJ, Fullwood NJ, Nakamura T, Tsuzuki M, Koizumi N, Dota A, Mochida C, Kinoshita S. Comparison of ultrastructure, tight junction-related protein expression and barrier function of human corneal epithelial cells cultivated on amniotic membrane with and without air-lifting. Exp Eye Res 2003; 76:735-43.

22. Sugrue SP, Zieske JD. ZO1 in corneal epithelium: association to the zonula occludens and adherens junctions. Exp Eye Res 1997; 64:11-20.

23. Blencowe BJ, Bowman JA, McCracken S, Rosonina E. SR-related proteins and the processing of messenger RNA precursors. Biochem Cell Biol 1999; 77:277-91.

24. Blencowe BJ, Bauren G, Eldridge AG, Issner R, Nickerson JA, Rosonina E, Sharp PA. The SRm160/300 splicing coactivator subunits. RNA 2000; 6:111-20.

25. Spector DL, O'Keefe RT, Jimenez-Garcia LF. Dynamics of transcription and pre-mRNA splicing within the mammalian cell nucleus. Cold Spring Harb Symp Quant Biol 1993; 58:799-805.

26. Misteli T, Caceres JF, Spector DL. The dynamics of a pre-mRNA splicing factor in living cells. Nature 1997; 387:523-7.

27. Misteli T, Spector DL. The cellular organization of gene expression. Curr Opin Cell Biol 1998; 10:323-31.

28. Sacco-Bubulya P, Spector DL. Disassembly of interchromatin granule clusters alters the coordination of transcription and pre-mRNA splicing. J Cell Biol 2002; 156:425-36.

29. Masselink H, Bernards R. The adenovirus E1A binding protein BS69 is a corepressor of transcription through recruitment of N-CoR. Oncogene 2000; 19:1538-46.

30. Wei G, Schaffner AE, Baker KM, Mansky KC, Ostrowski MC. Ets-2 interacts with co-repressor BS69 to repress target gene expression. Anticancer Res 2003; 23:2173-8.

31. Maniatis T, Reed R. An extensive network of coupling among gene expression machines. Nature 2002; 416:499-506.