• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS ONE. 2008; 3(8): e3000.
Published online Aug 20, 2008. doi:  10.1371/journal.pone.0003000
PMCID: PMC2500194

The Cellular Prion Protein PrPc Is Involved in the Proliferation of Epithelial Cells and in the Distribution of Junction-Associated Proteins

Neil Hotchin, Editor

Abstract

Background

The physiological function of the ubiquitous cellular prion protein, PrPc, is still under debate. It was essentially studied in nervous system, but poorly investigated in epithelial cells. We previously reported that PrPc is targeted to cell–cell junctions of polarized epithelial cells, where it interacts with c-Src.

Methodology/Findings

We show here that, in cultured human enterocytes and in intestine in vivo, the mature PrPc is differentially targeted either to the nucleus in dividing cells or to cell–cell contacts in polarized/differentiated cells. By proteomic analysis, we demonstrate that the junctional PrPc interacts with cytoskeleton-associated proteins, such as gamma- and beta-actin, alpha-spectrin, annexin A2, and with the desmosome-associated proteins desmoglein, plakoglobin and desmoplakin. In addition, co-immunoprecipitation experiments revealed complexes associating PrPc, desmoglein and c-Src in raft domains. Through siRNA strategy, we show that PrPc is necessary to complete the process of epithelial cell proliferation and for the sub-cellular distribution of proteins involved in cell architecture and junctions. Moreover, analysis of the architecture of the intestinal epithelium of PrPc knock-out mice revealed a net decrease in the size of desmosomal junctions and, without change in the amount of BrdU incorporation, a shortening of the length of intestinal villi.

Conclusions/Significance

From these results, PrPc could be considered as a new partner involved in the balance between proliferation and polarization/differentiation in epithelial cells.

Introduction

The cellular prion protein (PrPc) is a ubiquitous glycoprotein anchored to the outer leaflet of the plasma membrane, in raft domains, through a glycosylphosphatidylinositol (GPI) moiety [1]. Its central role in transmissible spongiform encephalopathies has been clearly demonstrated for many years [2][4] and efforts have been made to determine its biological role apart from pathological situations. Although many cells and tissues, such as blood lymphocytes, muscle, heart, kidney, digestive tract and skin, express PrPc [5][9], most of the studies concerning its physiological function have been performed on nerve cells. In these models, it has been established that PrPc binds copper [10] and can homodimerize [11] or interact with other proteins, among which are synapsin Ib, Grb2, Pint1, LRP/LR, and N-CAMs [12][19]. It has also been reported that PrPc, via interaction with phosphorylated Fyn [20], participates in cell redox homeostasis through ROS production [21]. In addition, it has been shown that multiple biochemical changes occur in prion protein knockout mice. They include increased levels of NF-κB and COX-IV and decreased levels of p53 and Cu, Zn superoxide dismutase activity, along with an increased neuronal sensitivity to oxidative stress in cultured cells isolated from these mice [22].

Much less is known about the role of PrPc in extra-neuronal tissues. In epithelial cells, PrPc was reported to be directed to basolateral membranes of MDCK and FRT epithelial cells [23]. We have shown that PrPc is expressed in enterocytes [24], which are highly polarized epithelial cells of the intestinal epithelium. Interestingly, we showed that, in polarized/differentiated enterocytes, PrPc is targeted to the lateral junctional complexes of adjacent cells where it interacts with Src kinase [24]. This tyrosine kinase is known targeted to cell–cell junctions where it phosphorylates substrates that induce adhesion turnover and actin remodeling [25]. Such a localization of PrPc, was also observed in human keratinocytes [24] and in endothelial cells [26], opening questions about the role of PrPc in cell–cell junctions of physiological barriers.

To address this question, we focused our study on intestinal epithelium and on enterocytes, the major cell population of this epithelium. Intestinal epithelium undergoes a rapid renewal throughout life (for review see [27]). Such a process requires a continuous coordination between proliferation, differentiation and death programs, along with a remodeling of cell-matrix and cell–cell contacts responsible for cell polarization. In crypts are localized stem cells and dividing cells, which migrate up to the villus where differentiation takes place. In the present work, we have analyzed whether the sub-cellular localization of PrPc varied in relation with cell proliferation, cell polarization and the state of cell–cell junctions in human intestinal epithelium in vivo and in the human Caco-2/TC7 enterocytes [28], which reproduce in culture the sequence of division and polarization/differentiation. In this enterocyte model, we characterized the partners of PrPc in cell–cell junctions. Finally, the impact of the invalidation of PrPc on the distribution of cell–cell junctions-associated proteins, the process of cell proliferation and the morphology of the intestinal epithelium was analyzed.

Materials and Methods

Reagents

Except when indicated, all chemicals were purchased from Sigma (Sigma-Aldrich, Saint Quentin Fallavier, France). Mouse 12F10 (against peptide 142–160) and SAF32 (against peptide 79–92) anti-human PrPc monoclonal antibodies were obtained from S.P.I BIO (Montigny le Bretonneux, France). Rabbit anti-human Ki67 polyclonal antibody and rat anti-mouse E-cadherin monoclonal antibody (ECCD-2) were purchased from Zymed Laboratories (San Francisco, CA). Rabbit anti-human Src (sc-18), and goat anti-human poly (ADP-ribose) polymerase (PARP) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse anti-human calnexin, desmoglein, plakoglobin and annexin A2 monoclonal antibodies were purchased from BD Biosciences (Erembodegem, Belgium). Rabbit anti-human PrPc (Ab 703), anti-pan desmoglein, anti-desmoplakin, anti phospho-S10-Histone H3 (Ab5176) polyclonal antibodies and rat anti-BrdU monoclonal antibody were purchased from Abcam (Cambridge, UK). Secondary CY2-, CY3- and CY5-labelled antibodies were purchased from Jackson Immuno-Research (West Grove, PA). F-actin was labelled with phalloidin-FITC. Endoglycosidase F was purchased from VWR (Fontenay sous bois, France). The biotinylated pro-aerolysin bacterial toxin [29] was a kind gift from Gisou van der Goot (Ecole Polytechnique de Lausanne, CH-1015 Lausanne, Switzerland).

Cell culture

All culture media were purchased from Gibco, Invitrogen Life Technologies (Cergy Pontoise, France). Caco-2/TC7 cells [28] were cultured with high glucose DMEM (Dulbecco's modified Eagle's medium) Glutamax I supplemented with 20% heat inactivated (56°C, 30 min) fetal calf serum (AbCys, Paris, France), 1% non-essential amino acids, penicillin (100 IU/ml) and streptomycin (10 µg/ml) in a 10% CO2/air atmosphere. The medium was changed every day. Depending on experiments, cells were plated on 1 µm pore size microporous PET filters (Falcon, BD Biosciences, Franklin lakes, NJ), or in plastic flasks (Falcon) or on glass lamellae (Polylabo, Strasbourg, France).

Cells treatments

Cycloheximide treatment

When indicated, the cells were treated with cycloheximide (10 µM final concentration).

siRNA transfection

siRNA corresponding to the human Prnp gene from codon 399 to 417 was synthesized by MWG Biotech (Ebersberg, Germany). The specific human siRNA sequence used was: 5′-GCC-GAG-UAA-GCC-AAA-AAC-CTT-3′ (sense). Cells were seeded at 5000 cell/cm2 on plastic or glass lamellae. siRNAs were mixed with Oligofectamine reagent (Invitrogen Life Technologies) for 15 min and Opti-MEM medium without serum was added according to the manufacturer's instructions. The final concentration of siRNA was 400 nM. After incubation for 6 hours at 37°C, Opti-MEM supplemented with 60% fetal calf serum was added to reach a final 20% serum concentration. A mouse PrPc siRNA sequence (5′-GCC-CAG-CAA-ACC-AAA-AAC-CTT-3′), inefficient on human PrPc RNA, and a scramble siRNA (5′-CCG-AGA-AGU-AAA-GCC-AAC-CTT-3′) were used as controls along with cells incubated with Oligofectamine reagent only. After 24 h, the medium was changed for the standard medium and the cells from all conditions were maintained in culture for the indicated times, the medium being changed every day.

GFP-PrPc plasmid construct and cell transfection

pEGFP-mouse PrPc plasmid [30] was obtained from MA Prado. GFP-PrPc was originally under the control of CMV promoter in this plasmid. To allow the expression of GFP-PrPc in differentiated cells, the corresponding sequence was subcloned into pGL2basic vector (Promega France) under the control of SV40 promoter. Caco-2/TC7 cells were transfected on day 2 with oligofectamine (Invitrogen, France) according to the manufacturer's instructions. After selection with G418 (Gibco BRL, France), transformed cells were allowed to expand prior to sorting, on the basis of GFP fluorescence, in an ALTRA cell sorter (Beckman Coulter, France).

Immunofluorescence analysis

Cells were washed twice with PBS containing 1 mM CaCl2 and 0.5 mM MgCl2, and fixed for 30 min with 4% paraformaldehyde (wt/vol) in PBS at 4°C. After an extensive washing in 150 mM glycine in PBS (PBS-glycine), the cells were permeabilized by incubation for 30 min in 0.1% Triton X-100 in PBS and washed in PBS-glycine followed by PBS plus 1% BSA. Cells were incubated for 60 min at room temperature with primary antibodies in PBS supplemented with 1% BSA, washed with PBS and stained with secondary antibodies in PBS with 1% BSA for 60 min at room temperature in the dark. After extensive washing in PBS, cells were mounted in Fluoprep (BioMérieux, Marcy l'Etoile, France), and examined by epifluorescence microscopy (Axiophot microscope connected to Axiocam camera using Axiovision 4.5 software; Carl Zeiss). Confocal microscopy (LSM 510 microscope; Carl Zeiss, Jena, Germany) was used for the observation of cells cultured on microporous filters, which are strongly autofluorescent and generate excessive background in epifluorescence. X–Z planes resulted from 0.1 µm stacks.

Tissue analysis

PrPc knock out mice [2] backcrossed on C57Bl6 and their wild type C57Bl6 counterparts were purchased from CDTA (Orléans, France). After removal of intestine in wild type or PrPc knock out mice, 2 cm segments of duodenum and jejunum were cut, gently flushed with PBS and opened longitudinally. Rolled segments were frozen in cryo-embedding media (OCT) and stored at −80°C until cryostat sectioning (10–20 µm). Sections were applied onto gelatine-coated glass slides, fixed in paraformaldehyde solution (4%), permeabilized with Triton before DAPI staining and then mounted in fluoprep solution.

Mitotic crypt cells were labeled with an anti-phosphoS10-Histone H3, using the same protocol as for cell immuno-labeling. To label proliferating intestinal crypt cells in S-phase, PrPc knock out and wild-type C57Bl6 mice were given an intraperitoneal injection of 5-bromo-2′-deoxyuridine (BrdU; Sigma; 120 mg/kg body weight) 90 min before sacrifice. Paraffin sections of alcohol-formalin-acetic acid (AFA)-fixed jejunum were incubated for 30 min in 2.5 N HCl before processing for immunofluorescent labeling with anti-BrdU antibody.

Sections from paraffin-embedded human jejunum were sequentially treated with xylene (2×5 min), 100% EtOH (2×5 min), 95% EtOH (1×5 min) and then rinsed with water. Antigens retrieval was performed by boiling slides in 10 mM citrate buffer (pH 6) for 10 min. After washes in PBS, sections were fixed and then the same protocol as described above for immunofluorescence was used.

Immunoelectron microscopy

Caco-2/TC7 cells, grown on Thermanox coverslips (Agar scientific), were fixed with acetone. After incubation with 12F10 monoclonal antibody, gold (1 nm particles)-labelled goat anti mouse IgG (Amersham Biosciences) were used as secondary antibodies and the resulting signal was enhanced by the Intense TM M silver enhancement kit (Amersham Biosciences). After alcohol-graded dehydratation, sections were embedded in Epon and ultrathin sections were analyzed (Jeol 100 CX II). For desmosomal structure analyses, intestine was flushed with cold 0.1 M Phosphate buffer (pH 7.4), ligatured and filled with cold 2.5% glutaraldhehyde/0.5% tannic acid in 0.1M cacodylate buffer at pH 7.4 for 2H. Fine samples of intestine were cut and fixed by 0.6% glutaraldhehyde/0.5% tannic, which stains and preserve the ultrastructure of phospholipids [31] in the same buffer overnight at 4°C. Postfixation was carried out in 2% osmic acid in 0.1 M Phosphate buffer for 1h at 4°C. Samples were then dehydrated in graded alcohol and embedded in Epon resin (Poly/Bed 812, Polysciences Inc.Warrington, PA). Ultrathin sections of around 65 nm were counterstained with uranyl acetate (30 min, 40°C) and lead citrate (10 min, 25°C) using an LKB 2168 ultrostainer. Observations were made using a JEOL CX100 equipped with a Gatan Digital camera (3.11.0) and the migrographs were processed with Gatan software.

Sub-cellular fractionation

Preparation of detergent-insoluble membranes on sucrose gradient

Caco-2/TC7 cells were homogenized on ice for 30 min in 2 ml of 10 mM Tris-HCl pH 8, 150 mM NaCl buffer (TN) containing 1% Triton X100 or in 2 ml of 20 mM Tris-HCl pH 7.8, 250 mM sucrose, 1 mM CaCl2 and 1 mM MgCl2 without detergent [32]. Anti-protease cocktail and anti-phosphatases (orthovanadate and beta-glycerophosphate) were added in both conditions. The cell homogenate was then adjusted to 40% sucrose by addition of 2 ml sucrose (80% in TN). The resulting 4 ml were covered with 4 ml of 30% sucrose and 4 ml of 5% sucrose and centrifuged for 16 h (39,000 rpm, 4°C) in a SW-41 rotor (Beckman Instruments, Gagny, France). Sequential 1 ml fractions were then collected from the top of the tube and the turbid fraction containing the floating detergent-insoluble membranes (fraction 4) was adjusted to 11 ml in TN buffer and centrifuged in a SW-41 rotor (35000 rpm, 1h). The pellet was dissolved in TN buffer containing 1% NP40, anti-proteases and anti-phosphatases and stored at −80°C until immunoprecipitations.

Nuclei and crude membrane preparations

Proliferating or polarized/differentiated Caco-2/TC7 cells were washed twice in 10 mM Tris-HCL pH 7.5 containing 20 mM KCl, 2 mM CaCl2, 2 mM MgCl2 and 0.2 mM spermidine (TKCM buffer) and scrapped in TKCM containing 1% Triton X-100, 1 mM phenylmethylsulfonylfluoride (PMSF), anti-proteases and anti-phosphatases. Nuclei were pelleted by centrifugation at 1000g for 10 min at 4°C and supernatants corresponding to cytosolic and membrane proteins were stored at –80°C until analysis. The pellets were washed in TKCM buffer and nuclear proteins were extracted with 2M NaCl in TKCM buffer for 1h at 4°C. Excess NaCl was removed by overnight dialysis against PBS at 4°C.

Immunoprecipitation and immunoblotting analyses

The raft fractions were immunoprecipitated with rabbit anti-PrPc (Ab 703) polyclonal antibodies, or non-specific rabbit immunoglobulins or anti-pan desmoglein antibodies coupled to protein A-sepharose 4B (Amersham Biosciences, Orsay, France). For SDS-PAGE, samples were boiled for 10 min in Laemmli buffer (2.5% SDS final concentration) and fractionated under reducing conditions in polyacrylamide gel. Proteins were transferred onto nitrocellulose membranes (Bio-Rad), blocked 2h with 5% non-fat dried milk in TBS/0.1% Tween 20 (TBST). After two washes in TBST, membranes were incubated (1h at room temperature) with specific antibodies. After three washes in TBST, the membranes were incubated (1h at room temperature) with peroxidase-labelled (HRP) secondary antibodies (Amersham Biosciences) in TBST. After three washes in TBST, bound antibodies were detected by chemiluminescent method (ECL, Amersham Biosciences). The quantitative analyses were performed with a high performance calibrated imaging densitometer (Bio-Rad GS-800) using PD Quest and Image Quant 5.2 softwares.

GPI anchor detection

After SDS-PAGE of immunoprecipitated materials and transfer onto nitrocellulose, membranes were incubated two hours in a binding buffer (50 mM NaH2PO4/0.3% Tween 20) before addition of biotinylated pro-aerolysin bacterial toxin. Biotinylated proteins were detected by blotting with HRP-conjugated streptavidin.

Endo F treatment

Nuclear proteins and rafts extracts were treated with Endoglycosidase F (10 units/50 µg proteins) in 40 mM sodium phosphate buffer, pH 7.5 containing 0.4% SDS, 20 mM DTT and 0.8% NP40 for 16 hours at 37°C before immunoblot analysis.

MS analysis

SDS/PAGE separation and protein digestion

Raft fractions were immunoprecipitated with anti PrPc antibodies and separated on 4–12% SDS/polyacrylamide gels. After staining with colloidal Commassie blue (G250, Bio-Rad), the visualized bands were cut into slices of 1 mm. Gel slices were then reduced, alkylated and subjected to digestion with trypsin (Roche Diagnostics) as already described [33]. Extracted peptides were dried and solubilized in solvent A (95/5 water/acetonitrile in 0.1% (w/v) formic acid). The total digestion product of a gel slice was used per liquid chromatography-tandem MS (MS/MS) analysis.

Liquid chromatography-MS/MS analysis

The extracted peptides were concentrated and separated on a LC-Packing system (Dionex S.A.) coupled to the nano-electrospray II ionisation interface of a QSTAR Pulsar i (Applied Biosystems) using a PicoTip (10 µm i.d., New Objectives, Woburn, MA). The MS/MS data was searched twice by using MASCOT (Matrix Science, London) and PHENYX (Geneva Bioinformatics S.A.) softwares on internal servers, first without taxonomic restriction to reveal the presence of proteins of interest and mammalian contaminants, then again the National Center for Biotechnology Information Human database (National Library of Medicine, Bethesda). All data are manually verified in order to minimise the errors in protein identification and/or characterization.

Statistical analysis

Statistical analyses were performed using student's t test.

Results

The cellular prion protein is localized in the nucleus in dividing cells and in cell–cell junctions in polarized epithelial cells

We analyzed, by immunofluorescence and immunoelectron microscopy, the distribution of PrPc or of GFP-PrPc in exponentially growing or polarized Caco-2/TC7 enterocytes. Representative images of PrPc, E-cadherin and DAPI labeling of the nuclei in exponentially growing Caco-2/TC7 cells (day 3) are shown in Figure 1A. When cells have not yet established well-defined adherens junctions, as shown by the poor expression of E-cadherin at cell–cell contacts (left panel), PrPc was mainly detected intracellularly. Interestingly, this staining co-localized with DAPI labeling, in the nucleus (middle panel). Immunodetected PrPc appeared as dots that were distributed around the nucleolus (right panel). This localization was confirmed by immunoelectron microscopy where the PrPc signal appeared accumulated in the nucleus (Fig. 1B, N) and systematically excluded from the nucleolus (Fig. 1B, *). The nuclear localization of the transfected mouse GFP-PrPc at this stage of the culture further strengthened the results obtained for the endogenous protein (Fig. 1C).

Figure 1
Expression and localization of PrPc in proliferating or differentiated/polarized Caco-2/TC7 cells.

In confluent and polarized Caco-2/TC7 cells (day 10), when E-cadherin-dependent junctions were established, PrPc was detected at the lateral membrane, and did not co-localize with DAPI in 90–95% of the cells (Fig. 1D). GFP-PrPc was also found targeted to the lateral membranes of polarized cells (Fig. 1E). This PrPc localization was previously revealed by immunoelectron microscopy [24]. As cells were not synchronized, PrPc was found in the nucleus in the few dividing cells (5–10%) that were still present within the confluent cell layer. In all conditions, a significant proportion of trafficking PrPc, which corresponded to approximately 30–40% of the total protein, was also observed in the cytoplasm.

The cellular prion protein is differently localized in proliferative and differentiated compartments of human intestinal epithelium

The sub-cellular localization of PrPc was analyzed in human intestinal epithelium and compared in crypts and villi, which correspond to the proliferative and differentiated compartments, respectively. In crypts, PrPc was found to co-localize with Ki-67, a nuclear marker of dividing cells (Fig. 2A). By contrast, in the crypt-villus transition compartment (Fig. 2B) and in villi (Fig. 2C), i.e. as soon as the process of cell division is stopped and the differentiation takes place, the nuclei, visualized by DAPI, were found devoid of Ki67 and of PrPc, which was visualized in the cytoplasm and in the lateral membranes of adjacent cells.

Figure 2
Expression and localization of PrPc and Ki67 in normal human intestinal epithelium.

Nuclear and junctional PrPc isoforms exhibit similar post-translational modifications but differ in their stability

The differential sub-cellular localization of PrPc could result from different molecular properties of the protein. Thus, we first compared the stability of PrPc when localized in the nucleus or in the lateral membranes. The amount of PrPc was determined by western blot, in nuclear extracts from proliferative cells (day 3) and membrane PrPc-containing raft domains from differentiated Caco-2/TC7 cells (day 10), at different times after inhibiting translation by cycloheximide. Results reported in Figure 3A show that PrPc is much more stable when localized in the nucleus, where degradation could not be observed over a 3 hour period, than at the membrane, where 50% of the protein were degraded at 30 min.

Figure 3
Biochemical characterization of membrane- and nucleus-associated PrPc isoforms.

PrPc is submitted to post-translational modifications such as N-linked glycosylation and addition of a GPI anchor. Endo F treatment of nuclear or raft extracts resulted in an equivalent shift of PrPc bands, demonstrating that nuclear and raft PrPc are similarly N-glycosylated (Fig. 3B). Analysis of the presence of a GPI anchor was performed on PrPc immunoprecipitated by the specific polyclonal rabbit antibodies Ab 703, using the biotinylated pro-aerolysin bacterial toxin, which recognizes GPI anchors [29]. A specific band corresponding to the presence of a GPI anchor on PrPc was detected in both nuclear and raft extracts (Fig. 3C, middle panel), at the same molecular weight as the PrPc signal (Fig. 3C upper panel), i.e. at 30 kDa. The absence of contamination of the nuclear and raft extracts with proteins derived from the other compartment was verified. Calnexin protein, which is known to be regularly buoyed with rafts [34] and the poly-(ADP-ribose) polymerase (PARP) protein that is exclusively expressed in the nucleus were used as markers of the respective compartments (Fig. 3C, lower panels).

Junctional PrPc is part of a complex involving desmosome-associated proteins and c-Src, and interacts with the structural proteins actin and annexin A2

A proteomic analysis was undertaken to determine the partners of PrPc in the junctional domains of enterocytes. Results presented in table 1 showed that the membrane PrPc interacts with five desmosome-associated proteins, among which desmoglein, plakoglobin and desmoplakin, and with gamma- and beta-actin and annexin A2, a structural protein that is known to participate in the regulation of actin cytoskeleton dynamics in junctions of epithelial cells [35]. The interaction with the desmosomal proteins and with annexin A2 was corroborated by western blots after purification of rafts in the presence of detergent and immunoprecipitation of PrPc by the specific rabbit polyclonal antibodies Ab 703 (Fig. 4A). Same results were obtained with rafts purified, after sucrose gradients, from cell extracts prepared without detergent (not shown). The absence of interaction between E-cadherin and PrPc [24] was confirmed, since E-cadherin was recovered exclusively in the immunoprecipitation supernatant (Fig. 4B). Interestingly, besides the already reported interaction of c-Src with the junctional PrPc [24], the immunoprecipitation of raft fraction with the anti-transmembrane protein desmoglein antibody revealed a complex including c-Src, desmoglein, and PrPc (Fig. 4C), while, as expected, desmoglein and E-cadherin did not interact.

Figure 4
Immunodetection of PrPc-associated desmosomal proteins in differentiated Caco-2/TC7 cells.
Table 1
PrPc partners in rafts.

PrPc invalidation impairs the sub-cellular localization of junction-associated proteins and desmosome structure

Based on its interaction with desmosomal proteins, studies were undertaken to determine whether PrPc could be involved in the organization of cell–cell junctions. Enterocytes were thus treated with human PrPc-siRNA, before the onset of cell polarity and the formation of cell–cell junctions. A kinetic analysis of PrPc levels in transfected cells, by immunofluorescence, showed that in 50 to 60% of the cells the levels of the endogenous PrPc were dramatically decreased 2 days (not shown) or 3 days after transfection (Fig. 5A) as compared to control cells, i. e. cells transfected with an inefficient mouse PrPc-siRNA or with a scramble siRNA or cells incubated with Oligofectamine only, and returned to the control levels from 4 days after transfection (not shown).

Figure 5
Effects of PrPc invalidation on the sub-cellular localization of junctional PrPc partners and on desmosome structure.

In a first attempt, the junctional state was assessed 3 and 7 days after transfection by the analysis of the expression and localization of E-cadherin, which does not interact with PrPc but is the most studied marker of cell–cell junctions. Three days after transfection (i.e 6 days after seeding), junctional complexes were already formed in the center of expanding cell clusters. In cells where PrPc expression was specifically impaired, the E-cadherin still localized to the lateral membranes of adjacent cells but its labeling intensity was decreased (Fig. 5A). Moreover, in the same fields, the cells appeared enlarged as compared with the three control conditions, as clearly noticeable in phase contrast picture (Fig. 5A). When the effect of human siRNAs on the expression of PrPc was no longer observed (day 7, Fig. 5A), the size of the cells and the amount of the junctional E-cadherin were rescued. We then analyzed the expression and the sub-cellular localization of the junctional partners of PrPc by immunofluorescence (Fig. 5B). To better compare cells that still expressed PrPc (40–50%) or not (50–60%), fields that combined the two cell populations of human PrPc-siRNA transfected cells are shown (Fig 5B, right panels) along with pictures representative of the results obtained in the three control conditions (Fig. 5B, left panels). Cells exhibiting a net decrease of PrPc levels were systematically enlarged. In these cells, the amount and/or the sub-cellular localization of the junctional partners of PrPc were modified: c-Src was found essentially intracellular, desmoglein, plakoglobin and desmoplakin labelings were lowered, and numerous actin stress fibers could be visualized in large cells that no longer expressed PrPc.

The role of PrPc on structural organization of cell–cell junctions was further analyzed in intestinal epithelium of PrPc knock out mice. Ultrastructural analysis revealed drastic changes in the length of the desmosomal plaque (Fig. 5C upper panel), which was not compensated by their number (not shown). Quantification of the length of desmosomes indicated a distribution of their size concentrated between 83 to 134 nm in knock out mice instead of 100 to 183 nm in wild type mice (Fig. 5C lower panel).

PrPc invalidation impairs completion of cell division and results in a shortening of intestinal villi

When compared to the three control conditions, we noticed an enlargement of human PrPc siRNA-treated dividing enterocytes (Fig. 5A) that could reflect the impact of PrPc-expression on cell proliferation. Analysis of cell growth (Fig. 6A) showed an arrest of the increase in cell numbers between 2 and 3 days after transfection, i.e during the period when PrPc expression was significantly decreased by human siRNAs. Surprisingly, DAPI staining of the nuclei showed that growth arrest was paralleled with a net increase of the number of mitosis, without apoptosis, in cells that did not express PrPc anymore (Fig. 6B). In parallel, the morphological examination of the intestinal epithelium of PrPc knock out mice revealed a net decrease in the length of the villi in both duodenum and jejunum segments in comparison with their wild type counterparts (Fig. 6C). To analyze the impact of PrPc expression on cell proliferation within the intestinal epithelium, PrPc knock out and wild type mice were pulse-labeled with BrdU for 1.5 hour prior to sacrifice. No significant difference in the number of BrdU-labeled cells in sections of jejunal crypts was observed between wild type and knock out mice (Fig. 6D), suggesting that S phase was not affected. By contrast, labeling of the mitosis-associated phospho-H3 revealed a net increase of mitotic cells in the crypt compartments of intestine from PrPc knock out mice as compared with wild type mice, while no obvious change in the size of the crypts was observed.

Figure 6
Effects of PrPc invalidation on the completion of cell division and length of intestinal villi.

Discussion

Our present results demonstrate that the cellular prion protein, PrPc, exhibits a dual distribution between the nucleus, in actively dividing cells, and cell–cell adhesion sites in polarized/differentiated cells. Interestingly, the membrane PrPc interacts with desmosomal proteins as well as with actin and actin-binding proteins at cell–cell junctions. Furthermore, we show that down regulation of PrPc expression in Caco-2/TC7 enterocytes lead to a complex pattern of alterations in both cell architecture and completion of the cell division process. These results are strengthened by the analysis of the intestinal epithelium of PrPc knock out mice, in which intestinal villi were found shortened and the size of enterocyte desmosomes decreased as compared to wild type mice.

Contrary to abnormal prion proteins [36][38], a targeting of the normal PrPc isoform to the nucleus has been rarely reported, [39], [40]. We demonstrate here that in intestinal epithelial cells, such a nuclear targeting is observed only in actively dividing cells, both in cultured enterocytes and in the intestinal epithelium in vivo. The characterization of the nuclear pattern of PrPc isoforms revealed that it is similar to that of membrane-associated PrPc in terms of glycosylation and presence of a GPI anchor. Analysis of protein stability shows a much longer half-life of PrPc in the nucleus than in plasma membrane, where junctional proteins are rapidly recycled. Nuclear PrPc stability could rely on the association with particular sub-nuclear compartments, such as PML bodies [41], a localization compatible with the pattern of nuclear PrPc staining that we observed. Contrasting with the misfolded protein [38], we show an exclusion of normal PrPc from the nucleolus. Altogether, our results asked the question of its biological role in the nucleus of dividing cells. Upon PrPc down regulation in cultured enterocytes, we observed modifications of cell morphology and an arrest of cell growth. This observation was consistent with the decreased villus size of the intestinal epithelium that we observed in PrPc null mice. In several mouse models, it has been shown that reduced intestinal crypt cell proliferation is associated with shorter villi [42], [43]. The growth arrest observed in PrPc siRNA-transfected cells was paralleled by an increase of mitotic cells. Further analysis of cell cycle perturbations was rendered difficult by the moderate transfection efficiency of Caco-2/TC7 cells by siRNA (50%). Nevertheless, the absence of S phase overt perturbation in crypts from PrPc null mice, as shown by BrdU incorporation experiments, along with the important increase of the number of mitotic cells in the crypt compartment suggest that PrPc invalidation would affect the mitosis process. PrPc has been shown to associate with tubulin [44] [45]. In addition, desmoplakin, that we identified as a PrPc partner, participates to the organization of microtubules in keratinocytes, through the recruitment at cell–cell junctions of a centrosomal protein, which is required for microtubule anchoring [46]. However, it cannot be concluded yet whether all the phases of mitosis or more particularly the last step of cytokinesis are slowed down through PrPc invalidation.

Another important finding of our study is the identification of desmosomal proteins as PrPc partners at cell–cell junctions. We had already identified c-Src kinase as a partner of PrPc in polarized enterocytes and demonstrated that PrPc does not interact with E-cadherin [24], asking the question of the partner that could link PrPc, which is anchored in the outer leaflet of the lateral membrane, to Src, which is localized in the inner one. Our data are consistent with the existence of a molecular complex in which the transmembrane desmosomal cadherin desmoglein is this link between PrPc and c-Src. Involvement of PrPc in the regulation of desmosome organization, and more generally in cell architecture, is further supported by our demonstration of alterations in the junctional targeting of Src and desmosomal proteins upon PrPc down-regulation by siRNA during the establishment of enterocyte cell–cell contacts. Importantly, absence of PrPc in vivo results in alterations in the size of desmosomes in intestinal epithelial cells. This structural perturbation could reflect modifications of assembly and/or stability of the desmosomal-protein complex in the absence of PrPc, as described for invalidation or mutation of other desmosome-associated proteins, such as desmoglein, plakoglobin or plakophilin families, in murine models or human pathologies (for review, see [47]). Desmosomal proteins have emerged as adhesion molecules that not only play critical structural roles but are also involved in signaling pathways [48]. Our data further support the hypothesis of the involvement of PrPc, in association with its desmosomal partners, in a signaling platform of the junctional state.

In conclusion, our results indicate that the normal cellular prion isoform PrPc, by its differential nuclear or junctional localization, must be considered as a potential actor of the balance between proliferation and polarization/differentiation of epithelial cells, through interaction with c-Src and with desmosome- and cytoskeleton-associated proteins. As such PrPc could be involved in the homeostasis of renewing epithelia.

Acknowledgments

Human intestine sections were obtained from Pr E. Leteurtre (Service d'Anatomie et de Cytologie Pathologiques, Pôle Biologie Pathologie, parc Eurasanté, CHRU, avenue Oscar Lambret, 59037 Lille cedex; Faculté de médecine Henri Warembourg, 59045 Lille cedex; Centre de Recherche Jean-Pierre AUBERT - JPARC Inserm U837, Equipe “Mucines, différenciation et cancérogenèse épithéliales”, Place de Verdun, 59045 Lille cedex). GFP-PrPc plasmid was obtained from M.A. Prado (Departamento de Bioquimica-Imunologia, Instituto de Ciencias Biologicas, Universidade Federal de Minas Gerais, Belo Horizontz, Brazil), and the biotinylated pro-aerolysin bacterial toxin was a kind gift from Gisou van der Goot (Ecole Polytechnique de Lausanne, CH-1015 Lausanne, Switzerland). We thank Julie Biscan (UMR S 872, CRC, Paris) for cryostat sections of mouse intestine. Confocal and electron microscopy analyzes were performed using the facilities of Centre de Recherche des Cordeliers, UMR S 872.

Footnotes

Competing Interests: The authors have declared that no competing interests exist.

Funding: E. Morel was recipient of a Ministère National de la Recherche et de la Technologie fellowship. S. Fouquet was recipient of a grant from F. Aupetit association. This work was supported in part by grants from INSERM (ATC Prion) and from Ministère de la recherche (GIS Prion).

References

1. Prusiner SB, Scott MR, DeArmond SJ, Cohen FE. Prion protein biology. Cell. 1998;93:337–348. [PubMed]
2. Bueler H, Aguzzi A, Sailer A, Greiner RA, Autenried P, et al. Mice devoid of PrP are resistant to scrapie. Cell. 1993;73:1339–1347. [PubMed]
3. Prusiner SB. The prion diseases. Brain Pathol. 1998;8:499–513. [PubMed]
4. Weissmann C, Bueler H, Fischer M, Sailer A, Aguzzi A, et al. PrP-deficient mice are resistant to scrapie. Ann N Y Acad Sci. 1994;724:235–240. [PubMed]
5. Bendheim PE, Brown HR, Rudelli RD, Scala LJ, Goller NL, et al. Nearly ubiquitous tissue distribution of the scrapie agent precursor protein. Neurology. 1992;42:149–156. [PubMed]
6. Cashman NR, Loertscher R, Nalbantoglu J, Shaw I, Kascsak RJ, et al. Cellular isoform of the scrapie agent protein participates in lymphocyte activation. Cell. 1990;61:185–192. [PubMed]
7. Horiuchi M, Yamazaki N, Ikeda T, Ishiguro N, Shinagawa M. A cellular form of prion protein (PrPC) exists in many non-neuronal tissues of sheep. J Gen Virol. 1995;76(Pt 10):2583–2587. [PubMed]
8. Pammer J, Cross HS, Frobert Y, Tschachler E, Oberhuber G. The pattern of prion-related protein expression in the gastrointestinal tract. Virchows Arch. 2000;436:466–472. [PubMed]
9. Sugaya M, Nakamura K, Watanabe T, Asahina A, Yasaka N, et al. Expression of cellular prion-related protein by murine Langerhans cells and keratinocytes. J Dermatol Sci. 2002;28:126–134. [PubMed]
10. Brockes JP. Topics in prion cell biology. Curr Opin Neurobiol. 1999;9:571–577. [PubMed]
11. Meyer RK, Lustig A, Oesch B, Fatzer R, Zurbriggen A, et al. A monomer-dimer equilibrium of a cellular prion protein (PrPC) not observed with recombinant PrP. J Biol Chem. 2000;275:38081–38087. [PubMed]
12. Gauczynski S, Hundt C, Leucht C, Weiss S. Interaction of prion proteins with cell surface receptors, molecular chaperones, and other molecules. Adv Protein Chem. 2001;57:229–272. [PubMed]
13. Gauczynski S, Peyrin JM, Haik S, Leucht C, Hundt C, et al. The 37-kDa/67-kDa laminin receptor acts as the cell-surface receptor for the cellular prion protein. Embo J. 2001;20:5863–5875. [PMC free article] [PubMed]
14. Hundt C, Peyrin JM, Haik S, Gauczynski S, Leucht C, et al. Identification of interaction domains of the prion protein with its 37-kDa/67-kDa laminin receptor. Embo J. 2001;20:5876–5886. [PMC free article] [PubMed]
15. Martins VR, Mercadante AF, Cabral AL, Freitas AR, Castro RM. Insights into the physiological function of cellular prion protein. Braz J Med Biol Res. 2001;34:585–595. [PubMed]
16. Rieger R, Edenhofer F, Lasmezas CI, Weiss S. The human 37-kDa laminin receptor precursor interacts with the prion protein in eukaryotic cells. Nat Med. 1997;3:1383–1388. [PubMed]
17. Schmitt-Ulms G, Legname G, Baldwin MA, Ball HL, Bradon N, et al. Binding of neural cell adhesion molecules (N-CAMs) to the cellular prion protein. J Mol Biol. 2001;314:1209–1225. [PubMed]
18. Shmakov AN, Bode J, Kilshaw PJ, Ghosh S. Diverse patterns of expression of the 67-kD laminin receptor in human small intestinal mucosa: potential binding sites for prion proteins? J Pathol. 2000;191:318–322. [PubMed]
19. Spielhaupter C, Schatzl HM. PrPC directly interacts with proteins involved in signaling pathways. J Biol Chem. 2001;276:44604–44612. [PubMed]
20. Mouillet-Richard S, Ermonval M, Chebassier C, Laplanche JL, Lehmann S, et al. Signal transduction through prion protein. Science. 2000;289:1925–1928. [PubMed]
21. Schneider B, Mutel V, Pietri M, Ermonval M, Mouillet-Richard S, et al. NADPH oxidase and extracellular regulated kinases 1/2 are targets of prion protein signaling in neuronal and nonneuronal cells. Proc Natl Acad Sci U S A. 2003;100:13326–13331. [PMC free article] [PubMed]
22. Brown DR, Nicholas RS, Canevari L. Lack of prion protein expression results in a neuronal phenotype sensitive to stress. J Neurosci Res. 2002;67:211–224. [PubMed]
23. Sarnataro D, Paladino S, Campana V, Grassi J, Nitsch L, et al. PrPC is sorted to the basolateral membrane of epithelial cells independently of its association with rafts. Traffic. 2002;3:810–821. [PubMed]
24. Morel E, Fouquet S, Chateau D, Yvernault L, Frobert Y, et al. The cellular prion protein PrPc is expressed in human enterocytes in cell-cell junctional domains. J Biol Chem. 2004;279:1499–1505. [PubMed]
25. Frame MC, Fincham VJ, Carragher NO, Wyke JA. v-Src's hold over actin and cell adhesions. Nat Rev Mol Cell Biol. 2002;3:233–245. [PubMed]
26. Viegas P, Chaverot N, Enslen H, Perriere N, Couraud PO, et al. Junctional expression of the prion protein PrPC by brain endothelial cells: a role in trans-endothelial migration of human monocytes. J Cell Sci. 2006;119:4634–4643. [PubMed]
27. Stappenbeck TS, Wong MH, Saam JR, Mysorekar IU, Gordon JI. Notes from some crypt watchers: regulation of renewal in the mouse intestinal epithelium. Curr Opin Cell Biol. 1998;10:702–709. [PubMed]
28. Chantret I, Rodolosse A, Barbat A, Dussaulx E, Brot-Laroche E, et al. Differential expression of sucrase-isomaltase in clones isolated from early and late passages of the cell line Caco-2: evidence for glucose-dependent negative regulation. J Cell Sci. 1994;107(Pt 1):213–225. [PubMed]
29. Abrami L, Fivaz M, Glauser PE, Parton RG, van der Goot FG. A pore-forming toxin interacts with a GPI-anchored protein and causes vacuolation of the endoplasmic reticulum. J Cell Biol. 1998;140:525–540. [PMC free article] [PubMed]
30. Lee KS, Magalhaes AC, Zanata SM, Brentani RR, Martins VR, et al. Internalization of mammalian fluorescent cellular prion protein and N-terminal deletion mutants in living cells. J Neurochem. 2001;79:79–87. [PubMed]
31. Kalina M, Pease DC. The preservation of ultrastructure in saturated phosphatidyl cholines by tannic acid in model systems and type II pneumocytes. J Cell Biol. 1977;74:726–741. [PMC free article] [PubMed]
32. Macdonald JL, Pike LJ. A simplified method for the preparation of detergent-free lipid rafts. J Lipid Res. 2005;46:1061–1067. [PubMed]
33. Fevrier B, Vilette D, Archer F, Loew D, Faigle W, et al. Cells release prions in association with exosomes. Proc Natl Acad Sci U S A 2004 [PMC free article] [PubMed]
34. Rouvinski A, Gahali-Sass I, Stav I, Metzer E, Atlan H, et al. Both raft- and non-raft proteins associate with CHAPS-insoluble complexes: some APP in large complexes. Biochem Biophys Res Commun. 2003;308:750–758. [PubMed]
35. Benaud C, Gentil BJ, Assard N, Court M, Garin J, et al. AHNAK interaction with the annexin 2/S100A10 complex regulates cell membrane cytoarchitecture. J Cell Biol. 2004;164:133–144. [PMC free article] [PubMed]
36. Crozet C, Vezilier J, Delfieu V, Nishimura T, Onodera T, et al. The truncated 23-230 form of the prion protein localizes to the nuclei of inducible cell lines independently of its nuclear localization signals and is not cytotoxic. Mol Cell Neurosci. 2006;32:315–323. [PubMed]
37. Gu Y, Hinnerwisch J, Fredricks R, Kalepu S, Mishra RS, et al. Identification of cryptic nuclear localization signals in the prion protein. Neurobiol Dis. 2003;12:133–149. [PubMed]
38. Mange A, Crozet C, Lehmann S, Beranger F. Scrapie-like prion protein is translocated to the nuclei of infected cells independently of proteasome inhibition and interacts with chromatin. J Cell Sci. 2004;117:2411–2416. [PubMed]
39. Rybner C, Finel-Szermanski S, Felin M, Sahraoui T, Rousseau C, et al. The cellular prion protein: a new partner of the lectin CBP70 in the nucleus of NB4 human promyelocytic leukemia cells. J Cell Biochem. 2002;84:408–419. [PubMed]
40. Hosokawa T, Tsuchiya K, Sato I, Takeyama N, Ueda S, et al. A monoclonal antibody (1D12) defines novel distribution patterns of prion protein (PrP) as granules in nucleus. Biochem Biophys Res Commun. 2008;366:657–663. [PubMed]
41. Villagra NT, Navascues J, Casafont I, Val-Bernal JF, Lafarga M, et al. The PML-nuclear inclusion of human supraoptic neurons: a new compartment with SUMO-1- and ubiquitin-proteasome-associated domains. Neurobiol Dis. 2006;21:181–193. [PubMed]
42. Pinto D, Gregorieff A, Begthel H, Clevers H. Canonical Wnt signals are essential for homeostasis of the intestinal epithelium. Genes Dev. 2003;17:1709–1713. [PMC free article] [PubMed]
43. Simmen FA, Xiao R, Velarde MC, Nicholson RD, Bowman MT, et al. Dysregulation of intestinal crypt cell proliferation and villus cell migration in mice lacking Kruppel-like factor 9. Am J Physiol Gastrointest Liver Physiol. 2007;292:G1757–1769. [PubMed]
44. Nieznanski K, Nieznanska H, Skowronek KJ, Osiecka KM, Stepkowski D. Direct interaction between prion protein and tubulin. Biochem Biophys Res Commun. 2005;334:403–411. [PubMed]
45. Dong CF, Shi S, Wang XF, An R, Li P, et al. The N-terminus of PrP is responsible for interacting with tubulin and fCJD related PrP mutants possess stronger inhibitive effect on microtubule assembly in vitro. Arch Biochem Biophys. 2008;470:83–92. [PubMed]
46. Lechler T, Fuchs E. Desmoplakin: an unexpected regulator of microtubule organization in the epidermis. J Cell Biol. 2007;176:147–154. [PMC free article] [PubMed]
47. Holthofer B, Windoffer R, Troyanovsky S, Leube RE. Structure and function of desmosomes. Int Rev Cytol. 2007;264:65–163. [PubMed]
48. Dusek RL, Godsel LM, Green KJ. Discriminating roles of desmosomal cadherins: beyond desmosomal adhesion. J Dermatol Sci. 2007;45:7–21. [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science
PubReader format: click here to try

Formats: