Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

A fundamental task of the gastrointestinal (GI) epithelium is to function as a highly selective permeability barrier to prevent absorption of damaging substances (e.g., proinflammatory molecules) from the external environment into the mucosa. Increased GI barrier permeability, in contrast, can lead to the penetration of normally excluded luminal substances into the mucosa and cause the initiation or continuation of inflammatory processes and mucosal damage (Hollander, 1998; Banan et al., 1999; Keshavarzian et al., 1999). Not surprisingly, loss of mucosal barrier integrity has been implicated in the pathogenesis of systemic and GI disorders, including inflammatory bowel disease (IBD) as well as trauma and burn-induced inflammation (Hollander, 1992; Gardiner et al., 1995; Unno et al., 1996, 1998; Keshavarzian et al., 1999). For instance, loss of intestinal barrier permeability, which has been demonstrated in patients with IBD (Hollander, 1992; Gardiner et al., 1995, 1998), is now believed to be an integral factor in both the initiation and the perpetuation of the inflammatory cascade in IBD (Hollander, 1992, 1998; Keshavarzian et al., 1992, 2003). The underlying difficulty in managing this inflammatory disorder is due to a lack of preventive strategies, which is in turn due to our limited understanding of the specific mechanisms responsible for its pathophysiology. Accordingly, increasing our knowledge of the underlying mechanism of mucosal barrier disruption (hyperpermeability) should provide new insights into more effective treatment regimes for IBD.

The pathogenesis of mucosal barrier dysfunction in IBD remains unclear. However, recognition that a leaky gut barrier can cause intestinal inflammation and that oxidants can lead to this hyperpermeability as well as that maintaining a normal barrier function is key to intestinal health, has led to major advances in our understanding of the onset of inflammation in IBD. In animal models, for example, intestinal barrier hyperpermeability induced by the injection of bacterial endotoxin into the mucosa can elicit an oxidative and inflammatory condition similar to IBD (Yamada et al., 1993; Hermiston and Gordon, 1995). Thus, understanding the events underlying intestinal barrier hyperpermeability (barrier disruption) under oxidative, proinflammatory conditions is of essential clinical and biological value.

We previously showed using monolayers of human intestinal (Caco-2) cells that oxidants such as H₂O₂ cause loss of intestinal barrier permeability in part by disassembling the microtubule cytoskeleton and that cytoskeletal disruption is a key contributor to injury (Banan et al., 1998b, 1999, 2000a,b, 2001c,d, 2002c). We then showed that oxidants cause this hyperpermeability in large part through the activation of a specific isoform of protein kinase C (PKC) (Banan et al., 2002a).

The PKC family consists of at least 12 known isoenzymes that are classified into three subfamilies (Goodnight et al., 1995; Cho et al., 1998; Banan et al., 2001c, 2002a,c): classical PKC isoforms (, 1, 2, and ), novel PKC isoenzymes (, , , , and µ), and atypical PKC isoforms (, , and ). Intestinal epithelial (e.g., Caco-2) cells express multiple isoforms of PKC, including PKC- (Wang et al., 1996; Banan et al., 2001c, 2002a,c). We recently showed using naive type intestinal cells (Banan et al., 2002a) that oxidants induce the membrane translocation of the native -isoform of PKC (PKC-). Using stable transfection technology, we then found (Banan et al., 2002a) that the -isoform is needed for a substantial fraction of oxidant-induced monolayer barrier dysfunction. Despite the importance of the -isoform of PKC to intestinal hyperpermeability, the fundamental mechanism for the PKC--mediated, oxidant-induced disruption of monolayers remains unknown.

In other studies, we reported on the importance of the inducible nitric-oxide synthase (iNOS) as one of the major events in intestinal cytoskeletal and barrier disruption (Banan et al., 2000b, 2001a). Indeed, uncontrolled generation of iNOS-derived reactive nitrogen metabolites [e.g., NO (nitric oxide) and ONOO] is an important factor in tissue damage during IBD (Ramchilewitz et al., 1995; McKenizie et al., 1996; Keshavarzian et al., 2002). For example, we showed (Keshavarzian et al., 2002) that a number of these iNOS-dependent oxidative reactions, including cytoskeletal nitration, occur in mucosa of patients with IBD.

In view of the aforementioned, we hypothesized that PKC- not only causes iNOS and NO up-regulation and its injurious consequences but also that it is key to oxidant-induced nitration and disruption of the microtubule cytoskeleton and loss of intestinal epithelial monolayer barrier integrity under the oxidative stress of this up-regulation. Herein, we report a pathophysiological mechanism, up-regulation of iNOS and its deleterious oxidative consequences such as cytoskeletal oxidation and nitration, by the -isoform of PKC.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture. Caco-2 cells were obtained from American Type Culture Collection (Manassas, VA) at passage 15. This cell line was chosen for our studies because it forms monolayers that morphologically resemble small intestinal cells, with defined apical brush borders and a highly organized microtubule network upon differentiation (Gilbert et al., 1991). The utility and characterization of this cell line have been reported previously (Gilbert et al., 1991; Banan et al., 1998b).

Plasmids and Stable Transfection. The sense and dominant-negative plasmids of PKC- were constructed and then stably transfected by lipofectin (lipofectin reagent; Invitrogen, Carlsbad, CA), as we described previously (Banan et al., 2001d, 2002a). A tetracycline-responsive expression (TRE) system was used to overexpress the native PKC-. cDNA encoding the entire reading frame of PKC- was subcloned into the TRE vector creating TRE PKC-. Control conditions included vector alone (TRE-z). The dominant-negative PKC- plasmid was also constructed (Cho et al., 1998; Banan et al., 2002a). PKC- protein expression and activity were verified, respectively, by Western blot analysis of cell lysates or activity assay (see below). Clones were subsequently plated on cell culture inserts and allowed to form confluent monolayers and then used for experiments.

Experimental Design. First, postconfluent monolayers of parental Caco-2 cells were preincubated with oxidant (H₂O₂, 0-0.5 mM) or vehicle (isotonic saline) for 30 min. As we have shown previously (Banan et al., 2000a, 2001a), H₂O₂ at 0.5 mM disrupts microtubules and barrier integrity and up-regulates iNOS in these parental cells. These experiments were then repeated using monolayers composed of cells either overexpressing PKC- (i.e., TRE PKC-) or almost completely lacking PKC- activity (dominant-negative). In all experiments the following were assessed: 1) microtubule cytoskeletal stability (cytoarchitecture and tubulin assembly/disassembly); 2) barrier integrity (permeability); 3) PKC- subcellular distribution (membrane, cytoskeletal, and cytosolic); 4) PKC- activity (in vitro kinase assay); 5) iNOS activity and protein; 6) NO levels; 7) reactive nitrogen metabolite (RNM) levels (e.g., ONOO); 8) oxidative stress [dichlorofluorescein (DCF) fluorescence]; 9) tubulin nitration (nitrotyrosination); and 10) tubulin oxidation (carbonylation).

Fractionation and Western Immunoblotting of PKC. Differentiated cell monolayers grown in 75-cm² flasks were processed for the isolation of the cytosolic, membrane, and cytoskeletal fractions (Banan et al., 2001b,c, 2002a). Protein content of the various cell fractions was assessed by the Bradford (1976) method. For immunoblotting, samples (75 µg of protein/lane) were added to a standard SDS buffer, boiled for 5 min, and then separated on 7.5% SDS-PAGE. The immunoblotted proteins were visualized by enhanced chemiluminescence (Amersham Biosciences, Inc., Piscataway, NJ) and autoradiography (e.g., 1 h at 20°C), and subsequently analyzed by densitometry. The exposure times were adjusted to ensure linear responses. Under these immunological detection conditions, the chemiluminescence assay was linear between 25 and 100 µg of total protein. The identity of the PKC- band was confirmed as we described previously (Banan et al., 2002a). We also confirmed that overexpression of  or negative dominant inhibition of  did not affect the relative expression levels of other PKC isoforms (nor did it kill the Caco-2 cells).

Immunoprecipitation and PKC- Activity Assay. Immunoprecipitated PKC- was collected and processed for its ability to phosphorylate a synthetic peptide (Banan et al., 2002a). The extent of histone H1 phosphorylation was determined by scintillation counting of excised Coomassie Blue histone polypeptide bands. Counts for blanks were subtracted from the sample activity. Sample activity was also corrected for protein concentration (Bradford, 1976), and PKC- activity was reported as picomoles per minute per milligram of protein.

Assay of NOS Activity. Conversion of L-[³H]arginine (Amersham Biosciences, Inc.) to L-[³H]citrulline was measured in the cell homogenates by scintillation counting. Experiments in the presence of NADPH, without Ca²⁺ and with 5 mM EGTA, determined Ca²⁺-independent NOS (iNOS) activity (Banan et al., 2000b, 2001a).

Western Blot of the Level of Inducible Nitric-Oxide Synthase. After treatments, the cells were washed once with cold phosphate-buffered saline, scraped into 1 ml of cold phosphate-buffered saline, and harvested in a standard anti-protease cocktail. For immunoblotting, samples (25 µg of protein/lane) were separated on 7.5% SDS-PAGE. Membranes were visualized by enhanced chemiluminescence and autoradiography (Banan et al., 2001a).

Chemiluminescence Analysis of NO. NO production was assessed by a chemiluminescence procedure (Banan et al., 2000b, 2001a). Cells were homogenized by sonication and the endogenous nitrate (NO) and nitrite (NO), the metabolic degradation products of NO, were then reduced to NO using vanadium(III) (Sigma-Aldrich, St. Louis, MO) and HCl at 90°C before the measurement of NO concentration by a chemiluminescence analyzer (Sievers NOA 280 analyzer; Sievers Instruments, Inc., Boulder, CO). NO was expressed in micromolar concentration and calculated by comparison with the chemiluminescence of a standard solution of NaNO₂. The absolute NO values were reported as micromoles/1 × 10⁶ cells.

Determination of Cell Oxidative Stress. Oxidative stress was assessed by measuring the conversion of a nonfluorescent compound, 2',7'-dichlorofluorescein diacetate (DCFD) (Molecular Probes, Eugene, OR) into a fluorescent dye, DCF (Banan et al., 2000b, 2001a). Cells were preincubated with the membrane-permeable DCFD (10 µg/ml for 30 min) before the subsequent treatments. Fluorescent signals (i.e., DCF fluorescence) from samples were quantitated using a fluorescence multiplate reader (excitation wavelength of 485 nm; emission wavelength of 530 nm). The dependence of the assay on reactive oxygen metabolites production (e.g., O generation) was shown as we reported previously (Banan et al., 1999, 2000b, 2001a) by adding active superoxide radical scavenger, SOD, or as a control condition iSOD. Similarly, we previously showed the dependence of this assay on RNM production (e.g., NO or ONOO generation) by adding either an RNM scavenger (e.g., cysteine or urate) or an inhibitor of RNM biosynthesis (e.g., L-NIL).

Immunofluorescent Staining and High-Resolution Laser Scanning Confocal Microscopy of Microtubules. Cells from monolayers were fixed in cytoskeletal stabilization buffer (Banan et al., 1998a,b, 1999, 2000a,b, 2001b,c,d). Cells were subsequently processed for incubation with a primary antibody, monoclonal mouse anti--tubulin (Sigma-Aldrich) and then with a secondary antibody (fluorescein isothiocyanate-conjugated goat anti-mouse; Sigma-Aldrich). After staining, cells were observed with an argon laser ( = 488 nm) using a 63× oil immersion plan-apochromat objective, numerical aperture 1.4 (Carl Zeiss, Jena, Germany). The cytoskeletal elements were examined in a blinded manner for their overall morphology, orientation, and disruption.

Microtubule (Tubulin) Fractionation and Quantitative Immunoblotting of Tubulin Assembly. Polymerized (S2) and monomeric (S1) fractions of tubulin were isolated using a series of extraction and ultracentrifugation steps as we described previously (Banan et al., 1998b, 1999, 2001c). Fractionated S1 and S2 samples were then flash frozen in liquid N₂ and stored at 70°C until immunoblotting. For immunoblotting, samples (5 µg of protein/lane) were placed in a standard SDS sample buffer, boiled, and then subjected to PAGE on 7.5% gels. Standard (purified) tubulin loading controls (5 µg/lane) were run concurrently with each run. To additionally verify equal loading, blots were routinely stained with 0.1% India Ink in Tris-buffered saline/Tween 20 buffer. To quantify the relative levels of tubulin, the optical density (O.D.) of the bands corresponding to immunolabeled tubulin (i.e., immunocomplexed tubulin) was measured with a laser densitometer. More specifically, to determine the relative quantity of polymerized tubulin pool, the O.D. of the autoradiographic bands corresponding to tubulin pools on immunoblots was obtained. The percentage of polymerized tubulin cytoskeleton from the total pool of tubulin (S2 + S1) was then calculated using the following formula: percentage of polymerized tubulin = [(S2) divided by (S2 + S1)], where S2 + S1 is the total cellular tubulin.

Immunoblotting Determination of Protein Tubulin Oxidation and Tubulin Nitration. Oxidation and nitration of the microtubule (tubulin) cytoskeleton were assessed, respectively, by measuring protein carbonyl and nitrotyrosine formation (Banan et al., 2000a,b, 2001a). To avoid unwanted oxidation, all buffers contained 0.5 mM dithiothreitol and 20 mM 4,5-dihydroxy-1,3-benzene sulfonic acid (Sigma-Aldrich). Processing and film exposure were as in a standard Western blot protocol (Banan et al., 2000b, 2001a). The relative levels of oxidized or nitrated tubulin were then quantified by measuring, with a laser densitometer, the O.D. of the bands corresponding to anti-dinitrophenylhydrazone or anti-nitrotyrosine immunoreactivity. Immunoreactivity was expressed as the percentage of carbonyl or nitrotyrosine formation (O.D.) in the treatment group compared with the maximally oxidized or nitrated tubulin standards, which also served as loading controls. These tubulin loading controls (5 µg/lane) were run concurrently with corresponding treatment groups. To further verify equal loading of lanes, blots were routinely stained with 0.1% India Ink in Tris-buffered saline/Tween 20 buffer. Oxidized tubulin standard was prepared using purified tubulin (ICN Pharmaceuticals, Costa Mesa, CA) that was subsequently carbonylated by exposure to 0.5 mM H₂O₂ and 1 mM FeSO₄ (30 min at room temperature). Nitrated tubulin standard was prepared by reacting purified tubulin with 0.1 mM ONOO (30 min at 37°C). These oxidized standards were then precipitated with trichloroacetic acid followed by the decanting of supernatant and washed (three times) with trichloroacetic acid to remove excess oxidizing agents.

Determination of Barrier Permeability by Fluorometry. Status of the integrity of monolayer barrier function was assessed by a widely used technique that measures the apical-to-basolateral paracellular flux of markers such as fluorescein sulfonic acid (200 µg/ml; 0.478 kDa) as we (Banan et al., 1999, 2000a,b, 2001a,b,c,d, 2002a,c) and others (Unno et al., 1996) have described. After treatments, fluorescent signals from samples were quantitated by a fluorescence multiplate reader (FL 600; Bio-Tek Instruments, Winooski, VT).

Statistical Analysis. Data are presented as mean ± S.E.M. All experiments were carried out with a sample size of at least six observations per treatment group that were run in triplicate on two or three different days. Statistical analysis comparing treatment groups was performed using analysis of variance followed by Dunnett's multiple range test (Harter, 1960). Correlation analyses were done using Pearson's test for parametric analysis or, when applicable, Spearman's test for nonparametric analysis. p values < 0.05 were deemed statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

We initially confirmed our previous finding (Banan et al., 2002a) that parental intestinal cells (tTA parental) cotransfected with cDNA encoding both hygromycin resistance (for selection) and a tetracycline-responsive expression system for PKC- (TRE PKC-) stably overexpress the  (75-kDa)-isoform of PKC (~3.5-fold compared with parental type cells). In this TRE PKC- system, overexpression of native PKC- is achieved in the absence of tetracycline (TTX), whereas its presence reduces expression to the levels seen in the parental cell line. This overexpression by itself disrupts monolayer barrier integrity as well as potentiates oxidant-induced loss of barrier integrity (Banan et al., 2002a). Overexpression of PKC- at the levels used causes no cellular toxicity (0% cell death assessed by ethidium homodimer probe). In the current investigation, using both pharmacological and molecular biological interventions, we studied the underlying mechanism by which PKC- disrupts monolayer cytoskeleton and barrier function.

Stable Overexpression of PKC- Isoform Causes Oxidative Injury to the Cytoskeleton: Promotion of Both Tubulin Nitration and Carbonylation. Because PKC- induces monolayer barrier disruption, we surmised that this disruption might be due to the activation of oxidative pathways. We, thus, measured the "footprints" of 1) RNM formation, nitrotyrosine moieties; and 2) oxidation, carbonylation moieties. The 50-kDa tubulin molecule, the structural protein of microtubule cytoskeleton, was sequentially fractionated from cell monolayers and then purified and subsequently immunoblotted. In Caco-2 cells, PKC- over-expression by itself, in the absence of added oxidant, led to oxidative damage (Fig. 1A). In cells stably overexpressing PKC- (TRE PKC- and exposed to vehicle) tubulin cytoskeleton was both substantially nitrated and oxidized (~70-80%). Incubation of these same cells overexpressing PKC- with tetracycline (i.e., TRE PKC- + tetracycline), as might be expected, prevented tubulin nitration and oxidation. Parental type cells (those not overexpressing PKC-) exposed to vehicle (with or without tetracycline) also showed no excessive tubulin oxidation. These parental type cells, in contrast, had their tubulin nitrated and oxidized by oxidant (H₂O₂, 0.5 mM) (Fig. 1A). Moreover, incubation with oxidant significantly potentiated tubulin oxidation in cells overexpressing PKC- (TRE PKC-). This potentiation was prevented in the presence of tetracycline. Furthermore, as might be expected, transfection of the TRE-z vector alone did not cause tubulin oxidation and nitration. For instance, the percentage of tubulin that was nitrated was 0% for both vector-transfected cells exposed to vehicle and for parental cells exposed to vehicle, 0.73 ± 0.04% for vector-transfected cells exposed to H₂O₂ alone and 0.72 ± 0.02% for parental cells incubated in H₂O₂. Similarly, the percentage of tubulin that was carbonylated was 0% for both vector-transfected cells exposed to vehicle and for parental cells exposed to vehicle, 0.66 ± 0.01% for vector-transfected cells exposed to H₂O₂ alone and 0.64 ± 0.02% for parental cells incubated in H₂O₂. Thus, both vector-transfected and parental cells responded in a similar manner to either vehicle or H₂O₂.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   A, overexpression of PKC- causes both nitration (nitrotyrosination) and oxidation (carbonylation) injury to the tubulin cytoskeleton of Caco-2 monolayers assessed by immunoblotting analysis. A clone previously developed in our laboratory that was transfected with a TRE system for PKC- (i.e., TRE PKC-) and that overexpresses native PKC- by 3.5-fold was used. The intestinal cells stably overexpressing PKC- (TRE PKC-) or parental type (tTA parental) monolayers (those not overexpressing ) were incubated with or without oxidant H2O2 (0.5 mM) or vehicle. Cells overexpressing PKC- show nitration and carbonylation injury to 50-kDa tubulin. This injurious effect was inhibited in the presence of tetracycline (TRE PKC- + tetracycline). Parental monolayers exposed to vehicle (with or without tetracycline) showed no tubulin oxidation. These parental cells had their tubulin oxidized after oxidant challenge. Nitration or oxidation was normalized to a nitrated or oxidized purified tubulin standard, which also served as loading control (5 µg, shown in B and C). *, p < 0.05 versus vehicle-treated parental cells. +, p < 0.05 versus H2O2-treated parental cells. &, p < 0.05 versus. PKC--overexpressing ("TRE PKC-") cells exposed to vehicle or PKC--over-expressing cells incubated with tetracycline and exposed to oxidant ("TRE PKC- + tetracycline"); n = 6/group. B and C, representative immunoblots of the tubulin nitration (B) and tubulin oxidation (carbonylation, C) after treatments shown in A. The tubulin nitration (anti-nitrotyrosine) bands (B) or tubulin carbonylation (anti-DNP) bands (C) from left to right correspond to parental type cells exposed vehicle (a), parental type cells exposed vehicle + tetracycline (b), PKC--over-expressing cells exposed to vehicle (c), PKC--overexpressing cells exposed to vehicle + tetracycline (d), parental type cells exposed to H2O2 (e), parental type cells exposed to H2O2 + tetracycline (f), PKC--overexpressing cells exposed to H2O2 (g), PKC--overexpressing cells exposed to H2O2 + tetracycline (h); and corresponding 5 µg/lane of nitrated or oxidized tubulin loading control (standard, 50 kDa) (i). PKC- overexpression in transfected cells by itself causes both tubulin nitration and oxidation (lane c in both figures) as shown by increased tubulin oxidation band densities. This level is comparable with that of the parental cells exposed to oxidant (lane e). In transfected cells, which were treated with vehicle, the presence of tetracycline (used to inhibit  overexpression, lane d) maintains almost normal conditions, which is similar to that of the parental cells exposed to vehicle (lane a). Shown is a representative blot; n = 6/group.

PKC--Induced Disruption Involves Activation of iNOS-Driven Reactions: Up-Regulation of iNOS, NO, RNM (ONOO), and Oxidative Stress. We next probed potential mechanisms through which PKC- overexpression increases nitration and oxidation of cytoskeletal proteins. Specifically, we sought to determine whether the up-regulation of iNOS-driven pathways might be an essential mechanism for PKC--induced oxidative disruption. In initial studies, multiple clones of intestinal cells stably transfected with 1, 2, 3, 4, or 5 µg of TRE PKC- cDNA showed a dose-dependent increase in PKC- isoform protein levels (Fig. 2). In parallel, analysis of cell lysates from these stably transfected clones showed a dose-dependent activation of iNOS (L-[³H]citrulline formation) (Table 1). The clone stably transfected with 4 µg of TRE PKC- cDNA provided the maximum up-regulation of iNOS. Accordingly, we used this clone in the experiments described below.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Overexpression of PKC- protein in Caco-2 cells transfected with varying amounts (1, 2, 3, 4, or 5 µg) of cDNA for a TRE system for PKC- (i.e., TRE PKC-). The levels of PKC- isoform protein expression are dose dependently enhanced by increasing the amount of plasmid transfected. Cell monolayers were lysed, sonicated, and processed for SDS-PAGE using a monoclonal anti-PKC- antibody followed by a horseradish peroxidase-conjugated secondary antibody. Prestained molecular weights Mr 67,000 and 93,000 were run in adjacent lanes. Shown is a representative blot; n = 6/group.

View this table: [in this window] [in a new window]   TABLE 1 Effects of transfection of varying amounts of PKC--inducible plasmid or dominant-negative plasmid on both the iNOS activity and NO levels in intestinal monolayers Values are means ± S.E.M. Cells were stably transfeted with varying amounts of a TRE system to overexpress PKC- (1, 2, 3, 4, or 5 µg of plasmid). In separate studies, cells were stably transfected with varying amounts of a dominant-negative plasmid for PKC- (1, 2, 3, 4, or 5 µg of DNA) and then exposed to oxidant (H2O2, 0.5 mM). Monolayer iNOS activity and NO levels were assessed as described under Materials and Methods.

View larger version (48K): [in this window] [in a new window]   Fig. 3.   A, stimulatory effects of PKC- overexpression (4-µg clone) on the up-regulation of iNOS activity in Caco-2 monolayers assessed by L-[3H]citrulline formation. *, p < 0.05 versus corresponding vehicle-treated parental cells. +, p < 0.05 versus corresponding H2O2-treated parental cells. &, p < 0.05 versus corresponding PKC--overexpressing "TRE PKC-" cells exposed to vehicle or PKC--overexpressing cells incubated with tetracycline and exposed to oxidant "TRE PKC- + tetracycline". tTA parental, parental cells; TRE PKC-, cells transfected with TRE system for native PKC-; n = 6/group. B, representative Western blot for the stimulatory effects of PKC- overexpression (4-µg clone) on up-regulating iNOS protein levels in Caco-2 cell monolayers. The iNOS bands are from parental type cells exposed to vehicle (a), parental type cells exposed vehicle + tetracycline (b), PKC- over-expressing cells exposed to vehicle (c), PKC- overexpressing cells exposed to vehicle + tetracycline (d), parental type cells exposed to H2O2 (e), parental type cells exposed to H2O2 + tetracycline (f); PKC--overexpressing cells exposed to H2O2 (g), and PKC--overexpressing cells exposed to H2O2 + tetracycline (h). In parental type cells, H2O2 resulted in a large increase in the levels of iNOS protein (~130 kDa). In PKC--overexpressing cells, TRE PKC- exposed to vehicle, this up-regulation is also seen. In these transfected cells in the presence of tetracycline (to inhibit -overexpression), as expected, iNOS up-regulation is prevented. The region of gel shown was between the Mr 126,000 and 218,000 prestained molecular weights, which were run in adjacent lanes. ~, denotes approximate molecular weight for iNOS.

View larger version (62K): [in this window] [in a new window]   Fig. 4.   Concentrations of NO in the supernatant of homogenates of intestinal cells of transfected and parental type origin assessed by chemiluminescence analysis. As for effects on iNOS (Fig. 3), PKC- over-expression (4-µg clone) causes NO up-regulation. NO is overproduced in vehicle-treated PKC--overexpressing cells. As expected, tetracycline, which is used to inhibit  overexpression, prevented this injurious effect in these transfected cells. A dose of oxidant H2O2 (0.5 mM) that overproduces NO in parental type cells is also shown. Although PKC- over-expression by itself overproduces NO, note the potentiation of NO levels in these overexpressing cells when exposed to oxidant. Parental cells responded comparably with vehicle treatment (with or without tetracycline). These parental type cells show NO overproduction only after exposure to oxidant. *, p < 0.05 versus corresponding vehicle-treated parental cells. +, p < 0.05 versus corresponding H2O2-treated parental cells. &, p < 0.05 versus corresponding PKC--overexpressing "TRE PKC-" cells exposed to vehicle or PKC--overexpressing cells incubated with tetracycline and exposed to oxidant "TRE PKC- + tetracycline". tTA parental, parental cells; TRE PKC-, cells transfected with TRE system for native PKC-; n = 6/group.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Time course for the stimulation of the induction of iNOS and increases in NO in PKC--overexpressing (4-µg) clone TRE PKC-. For comparison, parental cells show similar, yet smaller, increases in iNOS and NO only when exposed to 0.5 mM H2O2 at zero time. Units are picomoles per milligram of protein for iNOS activity; 103 × optical density for iNOS protein levels; micromoles per 106 cells for NO levels.

View larger version (50K): [in this window] [in a new window]   Fig. 6.   A, oxidative stress in cell monolayers that is induced by PKC- overexpression (4-µg clone) as determined by the changes in DCF fluorescence intensity. *, p < 0.05 versus vehicle-treated parental cells. +, p < 0.05 versus H2O2-treated parental cells. &, p < 0.05 versus corresponding PKC--overexpressing "TRE PKC-" cells exposed to vehicle or PKC--overexpressing cells incubated with tetracycline and exposed to oxidant "TRE PKC- + tetracycline". tTA parental, parental cells. TRE PKC-, cells transfected with TRE system for native PKC-; n = 6/group. B, oxidative stress (DCF fluorescence) that is either induced by authentic oxidant (H2O2) in parental type cells or by overexpression of PKC- (TRE PKC-) in transfected clones is attenuated by the iNOS inhibitor L-NIL or by oxidant scavengers (L-cysteine or SOD). Monolayers were pretreated with 1 mM L-NIL, 1 mM L-cysteine, or 300 U/ml SOD (a superoxide anion scavenger) before shown conditions. *, p < 0.05 versus control condition (vehicle) in parental cells. +, p < 0.05 versus corresponding H2O2-treated parental cells or PKC--overexpressing "TRE PKC-" cells; n = 6/group. C, immunoblotting analysis showing the protective effects of iNOS inhibitor (L-NIL) or oxidant scavengers (L-cysteine or SOD) against nitration or carbonylation of polymerized tubulin in Caco-2 cells of either wild-type or transfected origin. Treatment and conditions as in B. Nitration and carbonylation was expressed as described under Materials and Methods. *, p < 0.05 versus control condition (vehicle) in parental cells. +, p < 0.05 versus corresponding H2O2-treated parental cells or PKC--overexpressing "TRE PKC-" cells; n = 6/group.

iNOS Inhibitor and Oxidant Scavengers Prevent Both Oxidative Stress and Tubulin Oxidation. To further show that RNM (and/or ROM) species are in fact involved in the up-regulation of oxidative stress and of tubulin oxidation, we then used several different scavengers of these reactive species and the inhibitor of iNOS (Banan et al., 1999, 2000b, 2001a). Figure 6B shows that pretreatment of monolayers of parental cells with either the RNM scavenger L-cysteine or the superoxide (O) scavenger SOD or the iNOS inhibitor L-NIL substantially attenuated H₂O₂-induced oxidative stress (DCF fluorescence). Similarly, pretreatment of monolayers overexpressing PKC- with these same scavengers or the inhibitor prevented the abnormally high levels of oxidative stress seen in these transfected clones. As might be expected, pretreatment with heat-inactivated iSOD did not protect (data not shown).

High Levels of Oxidative Stress in PKC- Overexpressing Cells Disrupt the Assembly of Tubulin and Architecture of the Microtubule Cytoskeleton. Because PKC- overexpression increased several measures of oxidative stress, including both tubulin nitration and oxidation, in our intestinal model, we then determined the state of microtubule cytoskeletal assembly and cytoarchitecture. PKC- overexpression causes disruption of the assembly of tubulin (Banan et al., 2002a). The cytoarchitecture of microtubule (tubulin-based) cytoskeleton in intestinal cells is shown in Fig. 7. High-resolution laser scanning confocal microscopy reveals (Fig. 7, A-D) that cells overexpressing PKC- (i.e., TRE PKC- without tetracycline) show a fragmented and collapsed microtubule cytoskeleton (Fig. 7C). In the presence of tetracycline (Fig. 7D), on the other hand, these cells show a normal and radial cytoarchitecture of the microtubule network. This intact distribution is similar to that of the parental cells exposed to vehicle (Fig. 7A). Parental cells show microtubule disruption after exposure to oxidant (Fig. 7B), which is comparable with that of the PKC--overexpressing cells (Fig. 7C).

View larger version (87K): [in this window] [in a new window]   Fig. 7.   Intracellular distribution of the microtubule (tubulin-based) cytoskeleton as captured by high-resolution laser scanning confocal microscopy (LSCM) in intestinal cell monolayers of transfected and parental type origin. Parental cells exposed to vehicle (A) show the intact microtubules, which disperse in a radial manner throughout the cytosol. Cells overexpressing PKC- (C), which were exposed to vehicle, show collapse, fragmentation, and disruption of the microtubule cytoarchitecture, resembling the parental cells exposed to 0.5 mM H2O2 (B). In the transfected cells, in the presence of tetracycline (D), which prevents PKC- overexpression, normal microtubule cytoskeleton is highly preserved. This appearance is indistinguishable from that of the parental cells exposed to vehicle. Scale bar, 25 µm. Shown is a representative photomicrograph; n = 6/group.

Constitutive Activation of the Overexpressed PKC- in Transfected Intestinal Cells Correlates with Several Different Indices of Oxidative Stress of iNOS and Its Injurious Consequences in Monolayers. Overexpression of the 75-kDa PKC- in intestinal cells leads to its distribution in mostly the particulate fractions (particulate includes membrane + cytoskeletal fractions) with a much smaller distribution in the cytosolic fractions, indicating the constitutive activation of -isoform (Banan et al., 2002a). As we also previously showed, there is a substantial increase in the activity levels of PKC- isoform in these transfected cells (i.e., TRE PKC- without tetracycline). In the presence of tetracycline (i.e., TRE PKC- + tetracycline), low or basal activation levels for this isoform are seen. Parental cells exposed to vehicle (with or without tetracycline) also show similar basal activation levels for PKC-. The activation levels for -isoform in the TRE PKC--overexpressing cells is further increased in the presence of oxidant, reaching near total levels possible for the 4-µg clone. When parental cells were exposed to oxidant, we also reported (Banan et al., 2002a) increased PKC- activation, but at much smaller levels than the cells overexpressing this isoform.

Dominant-Negative Inhibition of PKC- to Inactivate Native -Isoform and Its Prevention of Oxidant-Induced Oxidative Stress of iNOS Up-Regulation and Its Injurious Consequences. The above-mentioned findings indicate that PKC- by itself could play an essential role in cell monolayer disruption by oxidative stress of iNOS-driven reactions. To independently demonstrate a key role for PKC- in oxidant-induced iNOS up-regulation and consequent RNM-driven oxidative stress, we then used dominant-negative PKC- clones of Caco-2 cells that we have recently developed (Banan et al., 2002a). Using this dominant-negative approach for PKC-, we are capable of substantially reducing the steady-state activity levels for this isoform by 99.5% in these transfected cells. In these dominant-negative cells oxidant cannot increase the -isoform activity (only a ~0.5% increase).

View larger version (31K): [in this window] [in a new window]   Fig. 8.   A and B, prevention of the oxidant-induced iNOS up-regulation (A) and NO overproduction (B) in intestinal cell monolayers by the stable dominant-negative inhibition of PKC- activity. A dominant-negative transfected cell clone previously developed in our laboratory (see Materials and Methods), which almost completely lacks native PKC- activity, was grown as monolayers and subsequently exposed to 0.5 mM H2O2 or vehicle. Monolayer of parental type Caco-2 cells was treated in a similar manner. *, p < 0.05 versus vehicle. +, p < 0.05 versus. H2O2 in parental cells. Parental, parental tTA cells; dominant neg., dominant-negative inhibition of PKC- activity; n = 6/group. C, immunoblotting analysis showing the suppressive effects of the stable dominant-negative inactivation of PKC- on oxidant-induced tubulin nitration and oxidation in Caco-2 cells. This dominant inhibition is protective against oxidant-induced tubulin nitration and oxidation. Treatment and conditions as in A. *, p < 0.05 versus vehicle. +, p < 0.05 versus H2O2 in parental cells; n = 6/group.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the current study, using monolayers of intestinal epithelial cells as a model of gut barrier integrity, we demonstrated that the -isoform of PKC seems to play a key role in oxidant-induced iNOS up-regulation and the consequent oxidative injury to the integrity of the microtubule (tubulin-based) cytoskeleton and the intestinal epithelial barrier. A second conclusion is that overexpression and activation of PKC- by themselves seem to up-regulate iNOS and then disrupt the cytoskeletal network and the integrity of intestinal monolayers. The mechanism for the effects of PKC- isoform seems to be the oxidation and nitration of the subunit components of the microtubule cytoskeleton and the consequent disruption of microtubule assembly and cytoarchitecture. To our knowledge, this is the first time this mechanism has been ascribed to the disruption of epithelial cells. These conclusions are based on several independent lines of evidence as discussed below.

First, expression of PKC-, which we previously showed to cause H₂O₂-induced barrier hyperpermeability, induces an oxidant-like disruption, including oxidant-induced iNOS up-regulation. PKC- evokes a cascade of changes, including hyperactivation of iNOS, overproduction of NO, increases in RNM, and promotion of oxidative stress (DCF fluorescence). This oxidative disruption seems to require overexpression and constitutive activation of the PKC-. Second, overexpression of PKC- causes the footprints of oxidative injury (i.e., RNM formation) to the tubulin (50-kDa) protein of the microtubule cytoskeleton. Overexpression of PKC- increases the nitration (nitrotyrosination) of tubulin, promotes the oxidation (carbonylation) of tubulin, and disrupts normal-appearing microtubule cytoskeleton. Third, transfected cells that overexpress PKC- are more sensitive to the oxidative stress. Indeed, induction of PKC- overexpression potentiates the disruptive effects of the exogenously added oxidant. Fourth, dominant-negative to PKC-, which causes almost complete inactivation of native PKC-, substantially interfered with oxidant-induced promotion of the iNOS up-regulation (by about 64%), of nitration and carbonylation of tubulin, and of the instability of microtubules. Oxidant was also unable to overproduce NO or increase DCF fluorescence in these dominant-negative-transfected cells. Fifth, PKC- activation quantitatively correlates with increases in all outcomes indicating oxidative stress.

Using both transfected clones and parental type cells, we showed robust correlations between PKC- isoform activation and microtubule cytoskeletal oxidation (r = 0.89, p < 0.05) as well as several other parameters of oxidative stress and microtubule instability. These include tubulin nitration (RNM footprint) and PKC- activation (r = 0.90, p < 0.05); tubulin carbonylation (oxidation) and PKC- activation (r = 0.89, p < 0.05); and tubulin assembly (increase S1 monomer pool) and PKC- activation (r = 0.93, p < 0.05). Similar correlations are reached when oxidant-induced iNOS up-regulation and PKC- activation (r = 0.91, p < 0.05), or NO levels and PKC- activation (r = 0.89, p < 0.05), or DCF fluorescence levels and PKC- activation (r = 0.92, p < 0.05) are used. The high strength of these correlations indicates that PKC- activation is critical to the disruption induced by iNOS up-regulation and consequent oxidative stress to the assembly of the tubulin cytoskeleton and intestinal barrier function. In this view, activation of PKC- leads to the overgeneration of reactive nitrogen metabolites and instability of the microtubule cytoskeleton and barrier after oxidative stress of iNOS activation. Overall, our studies on the -isoform are consistent with a model in which enhanced activation of PKC- results in up-regulation of iNOS, increases in both NO and RNM levels, increases in both tubulin nitration and oxidation, and decreased assembly of polymerized tubulin pools and concomitant increases in monomeric tubulin pools, and subsequently leads to instability of the microtubule cytoskeleton and monolayer barrier integrity under proinflammatory conditions of oxidative stress. Other PKC isoforms may also be involved in this disruption; however, our findings indicate that such disruption is mediated, in large part (~60%), through the -isoform.

Our previous studies on the protective PKCs such as the atypical  (72-kDa)-isoform of PKC (Banan et al., 2002d) showed correlations that are inverse to those for PKC-. Specifically, PKC- activation correlated with protection against microtubule oxidation (r = 0.93, p < 0.05), protection against tubulin nitration (r = 0.92, p < 0.05), and protection against tubulin disassembly (increase S1 monomer pool) (r = 0.92, p < 0.05). Similarly, other robust correlations were seen when either tubulin assembly (increase S2 polymer pool) and PKC- activation (r = 0.92, p < 0.05) or percentage of normal microtubules and PKC- activation (r = 0.95, p < 0.05) were plotted. Furthermore, protection against oxidant-induced iNOS up-regulation and PKC- activation (r = 0.91, p < 0.05), NO levels and PKC- activation (r = 0.88, p < 0.05), and DCF fluorescence levels and PKC- activation (r = 0.90, p < 0.05) were also found. Thus, PKC- activation (unlike PKC-) is essential to protection against oxidative stress and iNOS up-regulation and the consequent disruption of the assembly of the tubulin cytoskeleton and of intestinal barrier function. Overall, our studies to date are consistent with a model in which activation of some PKC isoforms (e.g., ) protect intestinal barrier integrity and microtubules against oxidative stress of iNOS-driven reactions, whereas activation of other PKC isoforms (e.g., ) lead to intestinal injury through up-regulation of this same iNOS-dependent pathway.

Our findings using transfected cells are consistent with the known biochemical properties of PKC. All PKCs consist of N-terminal regulatory domains and C-terminal catalytic domains (Gopalakrishna and Jaken, 2000). In resting cells, PKC is mainly found in an inactive conformation. In this inactive form, PKC is mainly distributed in the soluble fraction and only loosely bound to membrane components. Regulatory domains of PKC isoforms vary from one subfamily to the next as well as among individual isoforms within a given subfamily (Cho et al., 1998; Mullin et al., 1998; Gopalakrishna and Jaken, 2000). Not surprisingly, differences among isozymes with respect to activation conditions and subcellular location indicate that individual PKC isoforms have distinct activation mechanisms as well as mediate distinct biological processes (Cho et al., 1998; Banan et al., 2001c, 2002a,c). Previous pharmacological reports (Wang et al., 1996; Banan et al., 2001b; Chang and Tepperman, 2001) have shown that PKC activation (translocation) is necessary for the observed effects of PKC. Specifically, the translocation of PKC from the cytosolic to the particulate fraction of the cell is a key step in its activation (Goodnight et al., 1995; Wang et al., 1996; Chang and Tepperman, 2001). Also, PKC- can cause disruption (hyperpermeation) of pig kidney (LLC-PK1) cell monolayers (Mullin et al., 1998). The effects of PKC activation in cellular models can sometimes be complex and may vary with different experimental settings and cell types. We recently reported (Banan et al., 2001c, 2002c) that the classical 1 (78-kDa) isoform of PKC and the atypical  (72-kDa) isoform of PKC are required for growth factor-induced protective effects on the intestinal epithelial barrier integrity. Thus, it seems that activating or mimicking different isoforms of PKC will have distinct effects on the GI epithelium, including both protection and disruption.

There are other reported effects of PKC-