Reports conflict regarding the effect of nitric oxide (NO) on intestinal epithelium. In chronic injury, NO appears detrimental by combining with reactive oxygen to form potent-free radicals. In contrast, inhibition of NO synthesis after acute injury exacerbates damage and inflammation. Recent studies have disclosed constitutive expression of inducible NO synthase (iNOS) by normal intestinal epithelia, yet little attention has been given to the role of iNOS in acute epithelial repair. We studied the local effects of iNOS on early epithelial repair of porcine ileal mucosa injured by deoxycholate within Ussing chambers. iNOS was constitutively expressed by the villous epithelium, and after deoxycholate injury, iNOS was expressed by injured and detaching enterocytes. Selective inhibition of iNOS abolished increases in NO synthesis and villous reepithelialization after injury. Exogenous l-arginine rescued baseline reepithelialization from NOS inhibitors but was only capable of stimulating additional repair in the presence of serum. These results demonstrate that iNOS-derived NO is a key mediator of early villous reepithelialization following acute mucosal injury.
the single columnar epithelial lining of the small intestine is the first line of defense against the translocation of luminal bacteria, antigenic peptides, or endotoxin into body fluids. After acute mucosal injury, three local events culminate in restoration of epithelial continuity and normal permeability including 1) villus contraction, which reduces the total and denuded surface area for repair (9,32); 2) migration of enterocytes to seal the exposed basement membrane (33); and 3) closure of leaky epithelial intercellular spaces and tight junctions (11). These events are locally regulated by mediators arising from a complex network of nerves, immune effector cells, structural and contractile fibroblasts, endothelial cells, and extracellular matrix within the underlying lamina propria. A critical role for these lamina propria mediators has only begun to be fully appreciated (3, 29).
With the use of a variety of models of intestinal injury, induction of inducible nitric oxide (NO) synthase (iNOS) has been associated with large and sustained amounts of NO synthesis that often perpetuates mucosal damage (2, 30, 50). A majority of these studies has examined the effects of NO in the presence of established mucosal inflammation or after prolonged durations of exposure in which coexisting oxidative stress is anticipated to be high. Under such conditions, NO has been demonstrated to exacerbate injury by combining with superoxide to generate more potent free radicals such as peroxynitrite (2, 31, 47, 50). Thus increased NO production alone may not be responsible for tissue damage.
In contrast, studies performed during the acute phase of intestinal injury have shown that iNOS may ameliorate mucosal damage. Inhibition of iNOS in the early phase of chemical or surgical gut injury has been demonstrated to exacerbate damage and inflammation and delay intestinal repair (6, 7, 26, 27, 53). Recently, constitutive expression of iNOS has been described in normal intestinal epithelia from several species (16, 30, 52). These epithelia appear capable of rapid increases in synthesis of iNOS and NO after acute gut injury (2, 17, 30). This unique capacity to constitutively and rapidly upregulate NO synthesis suggests that mucosal iNOS may play an important role in acute epithelial defense and repair. Nevertheless, little is known about the early and local mechanisms of iNOS action in acute epithelial repair, due, in part, to the complexity of the in vivo models studied and the multitude of physiological effects influenced by NO such as blood flow and leukocyte infiltration (20, 35).
We have developed an in vitro model of early epithelial repair using intact porcine ileal mucosa injured by deoxycholate within Ussing chambers. The model retains the physiological complexity of interactions between the lamina propria and overlying epithelium while eliminating the influences of leukocyte emigration and mucosal blood flow on mechanisms of NO action. With the use of the parameters of barrier function and histomorphometry, the influence of iNOS on local mechanisms that affect epithelial repair, including villus contraction, epithelial cell migration, and tight junction permeability, could be discerned. In the present study, we demonstrate, for the first time, that constitutive iNOS expression and NO synthesis by intestinal epithelium mediate early epithelial repair following acute mucosal injury.
All studies were approved by the North Carolina State University Institutional Animal Care and Use Committee. Six- to eight-week-old female Yorkshire crossbred pigs were fed a commercial pelleted diet and had free access to water. Pigs were allowed a 4-day minimum acclimatization period before study. Pigs were sedated with xylazine (1.5 mg/kg im; Fort Dodge, Fort Dodge, Iowa) and ketamine (11 mg/kg im; Fort Dodge) and then euthanatized by intravenous pentobarbital sodium overdose (Vortech Pharmacia, Dearborn, MI). Approximately 30 cm of the distal ileum were immediately removed and incised along the antimesenteric border to expose the luminal mucosa.
Ussing chamber studies.
Ileal mucosa was rinsed and stripped from the seromuscular layer in oxygenated (95% O2-5% CO2) Ringer solution and mounted in a 3.14-cm2 aperture Ussing Chambers. Tissues from each pig were mounted in 12 individual Ussing chambers, each of which received a different treatment. Tissues were bathed on the serosal and mucosal sides with 10 ml Ringer solution containing (in mM) 142 Na+, 5 K+, 1.25 Ca2+, 1.1 Mg2+, 124 Cl−, 25 HCO , 1.65 HPO , and 0.3 H2PO . The serosal bathing solution contained 10 mM glucose and was osmotically balanced on the mucosal side with 10 mM mannitol. Bathing solutions were oxygenated (95% O2-5% CO2) and circulated in water-jacketed reservoirs maintained at 37°C. The spontaneous potential difference (PD) was measured using Ringer-agar bridges connected to calomel electrodes, and the PD was short circuited through Ag-AgCl electrodes using a voltage clamp that corrected for fluid resistance. Resistance (R; Ω · cm2) was calculated from the spontaneous PD and short-circuit current (I sc). If the spontaneous PD was between −1.0 and 1.0 mV, tissues were current clamped at ±100 μA for 5 s and the PD was recorded. Tissues were allowed to equilibrate for 15 min before injury [time 0–15 (min)]. After 15 min of injury (time 15–30), I sc and PD were recorded every 15–30 min for 180 min (time 30–210). The 210-min total duration of study was chosen based on observations of increasing mucosal permeability of uninjured control tissue within the Ussing chamber after this time period.
In vitro deoxycholate-induced mucosal injury.
We sought a method for acute, in vitro injury of mucosa that resulted in a uniform degree of epithelial damage but which also allowed repair by restitution during the time frame of the experiment. Because prior studies have characterized the effects of bile salts on mucosal epithelium (1, 8, 11, 15), tissues were injured by addition of deoxycholate (Sigma; St. Louis, MO) to the mucosal reservoir of the Ussing chamber, after which the bile salt was washed out and replaced with normal Ringer's solution. A preliminary time (15, 30, and 60 min) and dose (6, 12, and 24 mM)-response study showed that a 15-min luminal exposure to 6 mM bile salt resulted in an injury that partially recovered R and permselectivity over the study period and therefore allowed evaluation of treatments that could either facilitate or inhibit recovery. To standardize the severity of injury, tissues exceeding 1 standard deviation of the mean Rof 22 ± 6 Ω · cm2 (i.e., <16 or >28 Ω · cm2) immediately after bile salt washout and before addition of treatments (time 30) were discarded from analysis. Because of these necessary criteria, treatment group sizes are not equal. For each pig, two chambers were designated for uninjured control and bile salt-injured control tissue to be studied in parallel with treatments. Studies were performed over a 3-yr period. Investigations of baseline epithelial repair in the presence of NO inhibitors were performed independently and using a different group of animals than those used to study stimulated repair in the presence ofl-arginine (ARG) and serum. To account for variation between groups of animals over time, each study included its own set of control data.
Isotopic flux studies.
These studies were performed at the same time as electrical measurements were recorded. The mucosal-to-serosal flux of mannitol was determined by addition of 3H-labeled mannitol (2 μCi, 6.6 mM; Dupont; Boston, MA) to the mucosal reservoir after bile salt washout. After a 15-min equilibration period, standards were taken from the mucosal reservoir. Two successive 60-min flux periods (from 90 to 210 min) were performed by taking paired samples from the serosal reservoir beginning 60 min after bile salt washout. After samples were removed, an equal volume and composition of Ringer solution containing appropriate treatments was replaced. Samples were counted for3H in a liquid scintillation counter (LKB Wallac; Turku, Finland), and flux of the mannitol from mucosa to serosa was calculated using standard equations.
Tissues were removed from the Ussing chamber at 30 (immediately after bile salt washout), 45, 75, 150, and 210 min (end of study). After tissues were fixated in formalin, they were sectioned (5 μm) and stained with hematoxylin and eosin. Three sections from each tissue were examined. With the use of an ocular micrometer, the following measurements were recorded and then averaged for five well-oriented villi: villus height measured from the crypt opening to the villus tip, crypt depth, linear length of villus perimeter, and linear length of denuded villus. The percentage of epithelialized villus surface was calculated directly from linear measurements taken of the epithelialized vs. denuded villus perimeter and was used as an index of epithelial restitution. Villi were considered well oriented if the adjacent crypt lumen was patent to the level of the muscularis mucosa. All measurements were performed without knowledge of the treatment administered. To account for reductions in villus height and surface area that result from the stripping and mounting of mucosa (24), measurements were compared with chambered, but uninjured, tissue fixed at matching time points.
Assessment of ARG and NO effects on barrier recovery after deoxycholate injury.
Treatments were added to tissues immediately after replacement of the bile salt by Ringer solution (time 30). Treatments included ARG (added mucosal and serosal; Sigma) and heat-inactivated fetal bovine serum (FBS; 1% mucosal and serosal; Mediatech; Herndon, VA). The dose of ARG was chosen based on its peak migratory effect on cultured porcine intestinal epithelial cells [IPEC-J2 (46)] and high-physiological concentration compared with intraluminal ARG in neonatal pigs [2 mM (51)]. All treatments extending into the millimolar range were osmotically balanced by addition to both sides of the mucosal preparation.
Endogenous NO formation was inhibited in the presence and absence of ARG using the nonselective and reversible NOS inhibitorN G-nitro-l-arginine methyl ester (l-NAME; added mucosal and serosal; Sigma) or the selective and irreversible iNOS inhibitors aminoguanidine orl-N6-(1-iminoethyl)lysine (l-NIL; added mucosal and serosal; Sigma). Dose range of l-NAME was chosen based on prior observations that three times morel-NAME than ARG was required for adequate competitive inhibition of mucosal NOS within the Ussing chamber (54). Prior studies have shown that Ca2+-independent (inducible) NOS activity is selectively inhibited at comparable concentrations ofl-NIL and aminoguanidine used in the present study (2, 23, 34). Polyamine formation was prevented in the presence and absence of ARG by blockade of ornithine decarboxylase (ODC) activity using the suicide inhibitor α-difluoromethylornithine (DFMO; added mucosal and serosal; Marion Merrell Dow Research; Cincinnati, OH).
Total NO2 + NO3 concentration accumulated in the mucosal reservoir over the entire course of deoxycholate injury, and repair was measured by means of fluorometric reversed-phase high-performance liquid chromatography, as described previously (22), and by conversion of NO3 to NO2 by nitrate reductase with detection of NO2using a commercial kit (Griess Assay; Cayman Chemical, Ann Arbor, MI).
Immunohistochemistry for iNOS (polyclonal rabbit anti-NOS II; Transduction Laboratories, Lexington, KY) was performed using formalin-fixed, 4–6 μm sections of mucosa removed after 0, 30, 75, and 210 min in the Ussing chamber in the presence or absence of bile salt injury. Tissues were deparaffinized by immersion in xylene, rehydrated in a graded series of ethanol, and hydrated to buffer (phosphate-buffered saline, pH 7.4). Tissues were treated with 3% H2O2 in methanol for 10 min at 4°C to quench endogenous peroxidase and blocked for 30 min at room temperature with nonimmune goat serum. A commercial kit was used for blocking endogenous avidin and biotin activity (Zymed Laboratories, avidin/biotin blocking kit; San Francisco, CA). Antibody was diluted 1:50 with diluent before incubation with tissue for 60 min at RT. Sections were immunostained using a commercially available, broad-spectrum streptavidin-biotin-peroxidase system with 3-amino-9-ethylcarbazole as the chromogen (Zymed Laboratories, Histostain-SP broad spectrum). Sections were counterstained with hematoxylin and eosin, and coverslips were mounted with an aqueous mounting media. Negative control sections were not treated with primary antibody.
Control and bile salt-injured mucosa was removed from the Ussing chamber after 15, 25, 30, 75, 120, and 210 min. The epithelium and lamina propria were scraped from each tissue sample using a glass slide and immediately immersed in 2 ml of chilled RIPA buffer [0.15 M NaCl, 50 mM Tris (pH 7.2), 1% deoxycholic acid, 1% Triton X-100, and 0.1% SDS] containing protease inhibitors. This mixture was sonicated on ice and centrifuged at 10,000 g for 10 min at 4°C. The supernatants were saved, and their protein concentrations were determined (Dc protein assay; Bio-Rad; Hercules, CA). Samples were mixed with an equal volume of 2× SDS-PAGE sample buffer, boiled for 4 min, and loaded on a 6% SDS-polyacrylamide gel. Electrophoresis was carried out according to standard protocols. Proteins were transferred to a nitrocellulose membrane [Hybond enhanced chemiluminescence (ECL); Amersham Life Science; Birmingham, UK] using an electroblotting minitransfer apparatus according to the manufacturer's protocol. Membranes were blocked at RT for 1 h in Tris-buffered saline plus 0.05% Tween 20 (TBST) and 5% powdered milk. Membranes were incubated overnight (1:5,000 polyclonal rabbit anti-NOS II; Transduction Laboratories) in primary antibody. After membranes were washed three times each with TBST, they were incubated for 1 h with horseradish peroxidase-conjugated secondary antibody (1:7,000; Santa Cruz Biotechnology; Santa Cruz, CA). After the membranes were washed three additional times for 5 min each with TBST, they were developed for visualization of protein by addition of ECL reagent according to the manufacturer's instructions (Amersham; Princeton, NJ). Densitometric analysis of immunoblots was performed using commercially available software (Scanalytics; Fairfax, VA).
Data are reported as means ± SE. For all analyses,P < 0.05 was considered significant. One-way or repeated-measures ANOVA and a post hoc Tukey's test and paired or unpaired t-tests were used to compare differences between treatment and control tissues (n = number of pigs receiving treatment; SigmaStat, Jandel Scientific; San Rafael, CA).
In vitro model of acute mucosal injury and epithelial repair.
Mucosal injury was induced by a 15-min luminal exposure to 6 mM deoxycholate within the Ussing chamber. The injury resulted in an immediate decrease in transepithelial electrical R (48 ± 2% of control) and increased permeability of the mucosa to3H-labeled mannitol (350 ± 30% of control;n = 18, flux period 90–150 min). After removal of the deoxycholate (time 30), partial recovery of R(80 ± 2% of control) and a significant decrease in3H-labeled mannitol permeability (270 ± 28% of control, n = 18, flux period 150–210 min;P < 0.05, 1-way ANOVA) were observed over a 180-min period of repair (time 30–210; Fig.1 A).
To quantify epithelial repair, the percent epithelialized villous surface was calculated using linear measurements taken from the villi of tissues fixed in the presence and absence of deoxycholate injury (times 0, 30, 45, 75,150, and 210; Fig. 1 B). Ussing-chambered but uninjured tissue remained covered by confluent epithelium and retained 99.3 ± 13% of initial (poststripping) villus height and 101.7 ± 9.5% of initial crypt depth for over 210 min if left unperturbed within the chamber. Immediately after removal of deoxycholate (time 30), histology revealed an acute villous contraction and sloughing of enterocytes from the villus tips. Enterocyte losses continued initially, resulting in maximal injury at time 75. By time 210, there was partial repair of the denuded villi by migrating flattened to cuboidal absorptive cells (Fig. 1 C). Early studies demonstrated that treated tissues differed only by the magnitude of R and epithelialization achieved by the endpoint of each experiment. Therefore, treatment effects on R and villous reepithelization are presented hereafter at the 210-min period only.
Endogenous NO synthesis by iNOS mediates villous reepithelialization.
Deoxycholate injury resulted in increased synthesis of NO (NO2 + NO3; control = 2.6 ± 0.55; injured = 8.4 ± 3.6 μM, n = 14 each at 210 min). Synthesis of NO by the injured tissue was equally inhibited by the nonselective reversible NOS inhibitorl-NAME (5 mM; 1.34 ± 0.07 μM) and the iNOS-selective irreversible inhibitor l-NIL (5 μM; 0.56 ± 0.25 μM, n = 4 each at 210 min;P < 0.05, paired t-test), demonstrating that the increase in NO synthesis resulting from deoxycholate injury was mediated by iNOS.
Each inhibitor entirely abolished villous reepithelialization and recovery of R after deoxycholate injury while having no effect on uninjured control mucosa (Fig.2). Exogenous ARG (10 mM) rescued the effect of reversible total NOS inhibition (l-NAME) but not the irreversible inhibition of iNOS alone (l-NIL), demonstrating that villous reepithelialization was mediated by iNOS (Fig. 2). Identical inhibitory effects on villous reepithelialization were observed using a second iNOS selective inhibitor aminoguanidine (1 mM; 62 ± 3% confluent, n = 10). Blockade of endogenous ARG conversion to polyamines by inhibition of ODC (α-DFMO; 0.01, 0.1, 1, 2, and 5 mM; n = 3–6 each) was without effect on baseline recovery (data not shown).
Exogenous ARG serum enhances villous reepithelialization.
Exogenous ARG increased NO synthesis after deoxycholate injury (injured = 2.4 ± 1.3 μM; injured + ARG = 5.2 ± 0.78 μM, n = 3 each at 210 min) but alone did not significantly stimulate reepithelialization beyond basal levels (Fig. 3). However, in the presence of serum (FBS; 1% heat inactivated), ARG stimulated reepithelialization to nearly complete restoration of epithelial confluency, R, and 3H-labeled mannitol permeability by 210 min of repair (Fig. 3; control = 0.32 ± 0.03 μmol/cm2 · h, n = 22; injured control = 0.63 ± 0.04 μmol/cm2 · h,n = 21; ARG = 0.58 ± 0.07 μmol/cm2 · h, n = 5; ARG-serum = 0.41 ± 0.06 μmol/cm2 · h,n = 5 at 210 min; P < 0.05, 1-way ANOVA). Importantly, serum alone did not have a significant effect on recovery of R or degree of villous reepithelialization compared with injured control tissues (serum = 0.62 ± 0.04 μmol/cm2 · h, n = 13; Fig. 3). Such observations suggested that serum components were rate limiting for optimal repair by ARG.
The stimulated repair mediated by ARG serum was dose dependently inhibited by l-NAME (5, 10, and 15 mM; n = 5 each), and inhibition was reversed by rescue with excess ARG (5, 10, and 15 mM; n = 9 each; Fig.4). Selective inhibition of iNOS alone (30 μM l-NIL) had similar effects to the total NOS inhibitor (10 mM l-NAME) in reversing the effect of ARG serum on R and reepithelialization (Fig. 3). These observations were also seen with aminoguanidine (1 mM; 61 ± 3% confluent, n = 11). At the doses of l-NAME and l-NIL chosen for these studies, baseline R, but not baseline reepithelialization, was also significantly inhibited. Inhibition of polyamine synthesis (DFMO; 0.1, 1, 2, and 5 mM,n = 5 each) was without effect on ARG serum-stimulated recovery (data not shown).
iNOS expression accompanies early epithelial injury and repair.
Immunohistochemistry and Western analysis were performed to characterize iNOS protein expression by control and deoxycholate-injured mucosa. In control mucosa, iNOS was detected by immunohistochemistry within mature villous enterocytes and not crypt epithelium. Expression of iNOS, as determined by immunohistochemistry and Western analysis, appeared unchanged for over 210 min in the absence of injury. Commensurate with deoxycholate injury immunohistochemistry revealed focal iNOS expression by injured and detaching enterocytes along the apical villi (Fig.5). Over the recovery period (30–210 min), considerable numbers of villous enterocytes underwent detachment resulting in severe villous atrophy and reepithelialization of the villi by crypt epithelium. Coincident with these epithelial losses and their replacement by crypt epithelium was an ongoing decline in mucosal iNOS expression (Figs. 5 and 6).
Experimental models of epithelial repair that integrate reparative events within the context of the native lamina propria and its overlying epithelium are complex and infrequently reported (10,32, 33). With the use of intact ileal mucosa in an in vitro model of acute gut injury, the present studies were able to discern the local mechanisms of iNOS-derived NO on early epithelial repair before the onset of mucosal inflammation and independent of NO effects on blood flow. Our results have shown that early villous reepithelialization is dependent on iNOS-derived NO. ExogenousARG rescued baseline reepithelialization from NOS inhibitors but was only capable of stimulating additional repair in the presence of serum. It is likely that ARG and serum worked additively to promote a significant increase in epithelial repair, although different underlying mechanisms for the role of iNOS and serum factors in basal and stimulated epithelial repair are also possible.
The constitutive isoforms of NOS produce picomolar amounts of NO, have been described in enteric nerves, enterocytes, and endothelium of the intestine, and are associated with preservation of the mucosal barrier (19, 25, 43). iNOS is transcriptionally regulated in response to proinflammatory cytokines, can be induced in numerous cell types, produces greater and more sustained amounts of NO, and often perpetuates mucosal injury. Reports conflict regarding the effect of increased NO synthesis on intestinal epithelium, perhaps attributed to differences in experimental models, duration of NO exposure, and extent of associated inflammation and oxidative stress (2, 19, 40, 47,50). In models of chronic intestinal inflammation, NO has been demonstrated to exacerbate injury by combining with superoxide to generate more potent free radicals such as peroxynitrite (30). However, salutory effects of iNOS-derived NO have been shown in several models of acute, chemically induced injury (6, 26, 27). Thus increased NO production alone may not be responsible for tissue damage. There has been surprisingly little attention given to the early and independent mechanisms of NO action on epithelial repair after acute gut injury.
Recently, constitutive expression of iNOS has been described in normal respiratory, renal tubular, and intestinal epithelia (14, 16, 30,48, 52). Consistent with the high level of enzymatic activity of iNOS, such epithelia appear capable of nanomolar conversion of ARG to NO (14, 48). In the intestine, constitutive iNOS expression has been described in the mature enterocytes of the ileal villi in several species including mice, guinea pigs, and pigs (16, 30, 52). Studies have further demonstrated that iNOS mRNA and protein synthesis by these epithelia can be acutely upregulated within 30 min of injury both in vivo and in cell culture (2, 17). The mechanism underlying this acute capacity for iNOS expression is unknown, although protein synthesis from a preexisting mRNA pool or activation/assembly of iNOS has recently been hypothesized (2). The increased vulnerability of villous epithelium to injury coupled with an ability to constitutively and rapidly upregulate NO synthesis suggests the intriguing possibility that iNOS plays an important role in acute epithelial defense or repair.
The present studies demonstrate that iNOS is contitutively expressed by ileal epithelium and mediates a significant increase in endogenous NO synthesis after acute deoxycholate injury. These observations are in agreement with prior observations of an immediate and sustained increase in luminal NO2 after in vivo deoxycholate perfusion of porcine ileum (31). We have further demonstrated using two different mechanism-based, selective, and irreversible inhibitors of the iNOS isoform (l-NIL and aminoguanidine) (5, 13) that inducible NO synthesis mediates early villous reepithelialization. Indeed, in the presence of iNOS inhibitors, villi did not reepithelialize. Exogenous ARG could restore basal reepithelialization by outcompeting the total NOS inhibitor (l-NAME), whereas after selective and irreversible inhibition of iNOS (l-NIL), no response to ARG was demonstrated. These observations suggest that early reepithelialization was mediated exclusively by the iNOS isoform. In a related study of laser-wounded renal tubular epithelial cells, iNOS was expressed along the leading edge of the migrating monolayer, ARG-stimulated migration, and repair was abolished after treatment with antisense iNOS oligonucleotides (38). Such observations suggest that iNOS synthesis by the epithelium alone may be sufficient to mediate these early reepithelialization events.
Exogenous ARG had a modest effect on NO formation by the injured mucosal epithelium but alone was incapable of stimulating additional recovery of barrier function and epithelialization within the 210-min experimental period. These observations suggested that delayed synthesis of additional factors was rate limiting for optimal response to NO. Recent cell culture studies have demonstrated permissive effects of NO on growth factor-stimulated epithelial cell migration. Growth factor-stimulated cell migration was blocked by treatment with NO synthesis inhibitors, whereas the addition of ARG, NO donors, or cGMP (the NO second messenger) enhanced migration (12,36-38, 49, 55). Recently Goligorsky et al. (12) demonstrated that NO reduces the tractional forces exerted by focal adhesions on the extracellular matrix, resulting in a permissive effect on growth factor-directed migration. A requirement for endogenous growth factor synthesis could account for the requirement of serum in addition to ARG for stimulation of acute villous reepithelialization of intact mucosa. Alternatively, ARG and serum may have contributed additively to achieve significant stimulation of epithelial repair. NO-mediated decreases in cellular adhesion may explain why greater iNOS expression by injured and detaching enterocytes at the apical villus was associated with epithelial detachment. It is possible that synthesis of NO at the apical villus leads repair by promoting loss of injured enterocytes while facilitating migration of the adjacent epithelium. Immunohistochemical detection of iNOS expression by injured and sloughing enterocytes was not associated with a significant increase in iNOS protein by Western analysis. This may reflect the transient residence and focal expression of iNOS by the injured apical enterocytes relative to the entire mucosal preparation from which the protein was derived. Our observation that crypt epithelium did not express iNOS constitutively is in agreement with the findings of others (16, 30, 48). Replacement of the villous enterocytes by crypt epithelium is likely responsible for the ongoing decline in mucosal iNOS protein (Western analysis and immunohistochemistry) seen with advancing repair.
The mediator of the ARG-serum effect appeared to be through NO formation rather than via increased polyamine production, both of which can occur through ARG metabolism. Thus l-NAME, a competitive inhibitor of NOS, completely inhibited the ARG-enhanced effect, whereas DFMO, a specific inhibitor of ODC, was without effect. Furthermore, we were able to reverse the l-NAME blockade with increased ARG concentrations, thereby arguing against a nonspecific effect of the inhibitor. Preliminary studies with piglet intestinal cells in culture also have shown that ARG increases NO production but has no effect on polyamine levels during a 24-h incubation (unpublished observation). De novo synthesis of polyamines from ARG may not play a significant role in polyamine content of porcine enterocytes because of low basal ODC activity (4). Furthermore, studies providing exogenous polyamines to wounded cells have been unable to stimulate cell migration beyond control levels (28). Therefore, it is likely that the stimulated migratory event seen after ARG in our studies can be attributed to NO.
We have not determined the factor(s) in serum responsible for synergistic effects with ARG. Serum contains as many as 20 growth regulatory peptides and cytokines (44), and spray-dried serum and colostrum concentrates are gaining popularity as a source of growth factors for stimulating gastrointestinal repair (21, 41,45). That a single growth factor is responsible for the effects of serum on epithelial repair is unlikely. Indeed, 10% fetal calf serum stimulates migration of intestinal epithelial cells to a greater extent than epidermal growth factor (EGF), transforming growth factor (TGF)-β, or hepatocyte growth factor alone (56). Of the growth factors known to promote epithelial repair, only a few are found in sufficient amounts in serum to promote exogenous reparative effects. Likely candidates include TGF-β and insulin-like growth factor (IGF)-I, because both have been identified as important serum factors mediating intestinal epithelial cell migration (18), whereas EGF concentrations in serum are low (42). We have previously shown that bovine serum concentrate (BSC) enhances the motogenic effect of ARG on migrating porcine jejunal cell monolayers (IPEC-J2) (46). The motogenic effect of BSC is reduced some 67% by anti-TGF-β and 62% by anti-IGF-I antibody (45). In endothelium, both TGF-β and IGF-I stimulate cell migration in an NO-dependent manner (38, 39, 49). Their abundance in serum and NO-dependent promigratory effects make TGF-β and IGF-I likely candidates for serum factors responsible for synergism with ARG in the present study.
In the present study, we demonstrate that constitutively expressed iNOS and NO synthesis by villous epithelium mediate early epithelial repair following an acute mucosal injury. Selective inhibition of iNOS abolished early villous reepithelialization after acute deoxycholate injury. Exogenous ARG alone was not capable of stimulating reepithelialization beyond basal levels except in the presence of added serum. The effect of iNOS on basal and ARG serum-stimulated epithelial repair appears to be mediated locally insofar as the model used was unfettered by changes in blood flow or leukocyte numbers. These findings suggest that in acute gut injury, before the onset of inflammation, iNOS-derived NO is the key mediator of early epithelial repair.
We thank Dr. G. Wu, Texas A&M Univ., for high-performance liquid chromatography analyses and M. Armstrong and M. Gray for technical assistance.
This study was supported by grants from the United States Department of Agriculture [Grant 9702239 (to R. A. Argenzio)] and National Institute of Diabetes and Digestive and Kidney Diseases [Center for Gastrointestinal Biology and Disease Grant DK-34987 and K08 DK-02868–01 (J. L. Gookin)].
An abstract of this work was presented at the American Gastroenterological Association annual meetings in San Diego, CA (May 2000) and Atlanta, GA (May 2001).
Address for reprint requests and other correspondence: J. L. Gookin; Dept. of Anatomy, Physiological Sciences, and Radiology, School of Veterinary Medicine, North Carolina State Univ., 4700 Hillsborough St., Raleigh, NC 27606 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published February 27, 2002;10.1152/ajpgi.00005.2001
- Copyright © 2002 the American Physiological Society