Main

NO is a free radical that can act as both a biologic messenger and cytotoxic agent(1, 2). NO is produced by a family of isoenzymes collectively termed NOSs of which three general isoforms exist, namely brain, endothelial, and macrophage NOS. The gastrointestinal epithelium of adult animals is able to synthesize NO via a constitutively expressed Ca2+-dependent isoform and after administration of bacterial endotoxin LPS via an inducible Ca2+-independent isoform (referred to for ease of reading as cNOS and iNOS, respectively)(3). NO synthesized by cNOS is involved in the regulation of gastric mucus and bicarbonate secretion and maintenance of intestinal epithelial integrity(4, 5). In contrast NO synthesized by iNOS is associated with gastrointestinal mucosal damage and injury to gastric and intestinal cells(3, 6). Furthermore, induction of iNOS activity is associated with an increase in intestinal epithelial permeability(7) and has been observed in colonic mucosal biopsies from patients with active ulcerative colitis(8). These findings suggest that overproduction of NO by iNOS is an important factor in the development of gastrointestinal mucosal inflammation.

The entire gastrointestinal tract of the newborn rat is functionally immature at birth and during the first 2 postnatal weeks(9). Changes in the activity of digestive enzymes that occur postnatally in the rat occur prenatally in other mammals including humans, which include a reduction in lactase and an increase in sucrase activity(10). In addition to alterations in digestive enzyme activity, the composition of proinflammatory enzymes in the neonatal rat intestine have also recently been examined, with xanthine oxidase and myeloperoxidase activity being maximal during the first 10 and 15 d of life and thereafter declining(11).

Levels of NO in excess of that required to activate guanylate cyclase can inhibit glycolysis, the mitochondrial respiratory chain, and DNA replication(1). Furthermore, the relatively small amounts of NO produced by the constitutive enzymes have been implicated as scavengers of oxidants(12), whereas the large amounts of NO produced by the inducible isozyme can interact with other reactive oxygen metabolites, such as superoxide, resulting in the propagation of the highly reactive species, peroxynitrite. This potent oxidizing agent can initiate lipid peroxidation and thus produce extreme membrane damage(13). Finally, the rodent gastrointestinal mucosa appears to have a differential susceptibility to damage, the stomach of the neonate(<20 d old) being most resilient and the lower bowel being the most sensitive, characteristics which appear to alter during development(14, 15). Similarly in the human neonate, intestinal immaturity is a major risk factor in the etiology of the colonic inflammation associated with conditions such as neonatal necrotizing enterocolitis(16).

Therefore, the aim of this study is to examine NOS activity during the development of the neonatal rat colon and to determine whether intestinal immaturity predisposes the neonatal rat to LPS-induced NO-mediated epithelial damage.

METHODS

Animals. Pregnant female Sprague-Dawley rats in late gestation were purchased from Canada Breeding Labs (St. Constant, Quebec). After arrival, cages were checked twice daily for pups. The day of birth was designated as d 0, and experiments were performed at various ages thereafter. Litter size was restricted to 12 pups per dam which was performed at 2 d postpartum. Pups of between 0 and 25 d of age were reared with their mother, who was fed and allowed water ad libitum. Weaning in our laboratory was observed to occur between 18 and 21 d of age. All animals were kept in a temperature-controlled room (22 ± 1 °C) and maintained on a 12-h light-dark cycle.

Pups of either sex from various litters of equal age were randomized to four experimental groups and received the following treatments: 1) control (vehicle); 2) Escherichia coli LPS (serotype 0111:B4, Sigma Chemical Co., St. Louis, MO; 3 mg/kg, i.p., in sterile saline) 4 h before sacrifice; 3) dexamethasone (2 mg/kg, i.p., in sterile saline) 1 h before administration of LPS. In a further group of experiments, animals were treated with the selective iNOS inhibitor aminoguanidine (25 mg/kg, i.p.) at the same time as LPS administration. All animals were treated so that the time of sacrifice would fall between 1400 and 1500 h to eliminate any potential diurnal effects, and the volume of agents administered was always ≤100 μL.

Determination of NOS activity. Rat pups of between 10 and 25 d of age were killed by cervical dislocation. A midline incision was made to reveal the peritoneal contents, and whole thickness samples of colon equidistant between the cecum and rectum were rapidly removed. The luminal contents were flushed out, and the tissue was washed in ice-cold saline, blotted dry, and placed on ice. Ice-cold homogenization buffer (pH 7.4) consisting of N- 2-hydroxyethylpiperzine-N'-2-ethanesulfonic acid 10 mM, sucrose 0.32 M, EDTA 0.1 mM, DTT 1 mM, soya bean trypsin inhibitor 10μg/mL, leupeptin 10 μg/mL, and aprotinin 2 μg/mL was quickly added to produce a final tissue content of 33% wt/vol. Enzyme activity was released by homogenization (Ultra-Turrax) on ice for 15 s at full speed, and cell debris was removed by centrifugation at 10 000 × g for 20 min at 4°C. Conversion of [14C]arginine to the NO co-product citrulline was determined as described previously(6). Briefly, 20 μL of tissue homogenate supernatant were incubated for 10 min at 37 °C with 50 μL of substrate buffer (pH 7.4) consisting of (final concentrations) 30 mM potassium phosphate, 150 μM CaCl2, 0.7 mM MgCl2, 15 μM L-[14C]arginine (800 000 dpm/mL; ICN, Mississauga, Ontario, Canada), 0.1 mM NADPH, and 10 mM L-valine. Unconverted L-[14C]arginine was removed by addition of 0.5 mL of a 1:1 suspension of Dowex (AG 50W-8; Sigma Chemical Co.) in water. Distilled water (1 mL) was added to each tube, and the resin was allowed to settle for 20 min before removal of 975 μL of the liquid phase for estimation of product formation by liquid scintillation counting. Product formation that was inhibited by removal by Ca2+ from the substrate buffer by addition of EGTA (1 mM) and by the incubation with the NOS inhibitor NG-monomethyl-L-arginine (300 μM; Sigma Chemical Co.) was used as an index of constitutive Ca2+-dependent NOS activity. Inducible Ca2+-independent NOS activity was measured as that activity which was inhibited by in vitro incubation with NG-monomethyl-L-arginine but not by EGTA(17).

Determination of lipid peroxidation. The tissue concentration of MDA was measured to provide an index of lipid peroxidation according to a method previously described by Yagi(18). Briefly, whole thickness samples of distal colon were homogenized at full speed(Ultra-Turrax) on ice in Earle's balanced salt solution for 15 s, producing a final tissue concentration of 100 mg/mL. The homogenate was then sonicated for 5 s, and 10 μL were removed for spectrophotometric determination of the protein content using the Bio-Rad protein assay and BSA as the protein standard. Trichloroacetic acid (5%, 1.8 mL) was added to 200 μL of homogenate before centrifugation at 3000 × g at 4 °C for 20 min. A 1.5-mL aliquot of the supernatant was added to 1.5 mL of freshly prepared thiobarbituric acid (0.6%), and the mixture was vortexed for 5 s and boiled for 10 min. The colored product was extracted by addition of 2 mL of n- butanol before centrifugation at 3000 × g for 15 min at ambient temperature. The fluorescence of the upper n-butanol layer was measured at an excitation wavelength of 515 nm and emission wavelength of 553 nm using tetramethoxypropane (Sigma Chemical Co.) as the standard. Tissue MDA levels are presented as nanomoles/mg of tissue protein.

Histologic assessment of mucosal damage. Mucosal injury was also examined histologically. Whole thickness sections of distal colon were fixed in neutral-buffered formalin (10% in PBS; pH 7.4), processed routinely, cut longitudinally (4-μm thickness), and stained with hematoxylin and eosin. Sections were examined by light microscopy by a naive observer in a randomized manner, and a scoring system was used where 0 = normal appearance; 1 = vasocongestion; 2 = submucosal edema with erythrocyte extravasation; 3 = submucosal edema and epithelial disruption.

Statistical analysis. Results are presented as mean ± SEM with n equal to the number of animals. Data were analyzed by a suitable form of t test or by a one-way or two-way analysis of variance and an appropriate multiple comparison test where p < 0.05 was considered significant.

RESULTS

cNOS activity. In control animals which had received vehicle only, cNOS activity in the distal colon homogenates steadily increased from 10 d of age up to maximal activity at 20 d of age (Fig. 1). After weaning, which was observed to occur in our laboratory between 18 and 21 d of age, the cNOS activity in 25-d-old rat pups declined to the levels observed in 10-d-old preweaned animals.

Figure 1
figure 1

Age dependency of Ca2+-dependent constitutive NOS activity in tissue homogenates (33.33% wt/vol), from rat colon. Data are presented as means ± SEM (n = 7-8) and analyzed by one-way analysis of variance and Duncan's multiple range test where significance between groups (p < 0.05) is denoted by dissimilar lowercase letters.

Full size image

iNOS activity. Tissue homogenates of distal colon excised from animals which had received bacterial LPS (3 mg/kg, i.p.) 4 h before sacrifice possessed significant (p < 0.05 versus control) iNOS activity between 10 and 25 d of age (Fig. 2). LPS-induced iNOS activity was observed in all age groups examined, although there was no significant differences among these values (Fig. 2).

Figure 2
figure 2

Effect of dexamethasone pretreatment upon Ca2+-independent inducible NOS activity in colon homogenates (33.33% wt/vol) removed from 10-25 d old rats. Animals were administered either LPS (3 mg/kg, i.p.) 4 h before sacrifice (□) or dexamethasone (2 mg/kg, i.p.) 1 h before administration of LPS (▪). Data are presented as mean ± SEM(n = 7-8). One-way analysis of variance indicated that there was no significant effect of age on iNOS activity, whereas dexamethasone treatment significantly reduced iNOS activity as determined by Duncan's multiple range test. **p < 0.01 for differences between dexamethasone treatment and the respective LPS alone group.

Full size image

Pretreatment of rat pups with the steroidal antiinflammatory dexamethasone(2 mg/kg, i.p.) 1 h before administration of LPS, which at this dose and regime has previously been shown to inhibit the expression of iNOS(17), attenuated the induction of iNOS activity in the colon excised from rats aged between 10 and 25 d (Fig. 2).

Effect of LPS administration on lipid peroxidation. The MDA content of whole tissue homogenates was measured to provide an index of lipid peroxidation. Administration of LPS (3 mg/kg, i.p.) 4 h before sacrifice, caused a significant increase in the MDA content of colonic tissue removed from both 10 and 25-d-old animals (Fig. 3). However, the MDA content of colonic tissue removed from 25-d-old animals was significantly less (p < 0.05) than that observed in 10-d-old animals (Fig. 3). The increase in the MDA content associated with the administration of LPS to both 10- and 25-d-old animals could be attenuated by pretreatment with the steroidal antiinflammatory agent, dexamethasone (Fig. 3). Furthermore, administration of the selective iNOS inhibitor, aminoguanidine, at the same time as LPS also attenuated the LPS-associated increase in lipid peroxidation (Fig. 3). When administered alone, without LPS, neither agent had any effect upon the MDA content compared with colons from untreated animals which was 0.08± 0.01 and 0.09 ± 0.01 nmol of MDA/mg of protein (n = 6) for dexamethasone and aminoguanidine treatments, respectively, in 10-d-old animals and 0.11 ± 0.02 and 0.10 ± 0.02 nmol of MDA/mg of protein (n = 6) for dexamethasone and aminoguanidine treatments, respectively, in 25-d-old animals.

Figure 3
figure 3

Lipid peroxidation, as measured by the concentration of MDA in samples of colon removed from 10-d-old animals (□) and 25-d-old animals (▪). Animals were administered LPS (3 mg/kg, i.p.) 4 h before sacrifice, dexamethasone (DEX; 2 mg/kg, i.p.) 1 h before LPS, or aminoguanidine (AG; 25 mg/kg, i.p.) at the same time as LPS. Data are presented as mean ± SEM (n = 6-13). Two-way analysis of variance indicated significant differences between 10- and 25-d-old animals.#p < 0.05 for difference between 10- and 25-d-old LPS-treated animals by t test for unpaired data. **p < 0.01 and *p < 0.05 for difference from untreated animals and†p < 0.01 for difference from animals treated with LPS alone within each age group as determined by one-way analysis of variance and Duncan's multiple range test.

Full size image

Histologic damage. Whole thickness sections of distal colon were examined for evidence of histologic abnormalities. The appearance of the serosa, muscularis, and mucosa of the distal colon from 10-d-old untreated animals was unremarkable (Fig. 4A), similarly the histologic damage score was not significantly different from zero (Fig. 5). However, the colons from 10-d-old animals treated with LPS presented with gross evidence of inflammation, which manifested in vasocongestion, submucosal edema, erythrocyte extravasation into intercellular spaces, and epithelial disruption (Fig. 4B). In conjunction, there was a significant (p < 0.05) increase in the histologic damage score (Fig. 5), which was attenuated by pretreatment with dexamethasone (Fig. 5) and presented as essentially microscopically intact sections (Fig 4C). Sections of colonic tissue from 25-d-old animals treated with LPS presented with considerably less microscopic damage than that seen in tissue removed from 10-d-old animals (Fig. 4D), which was confirmed by measurement of the histologic damage score (Fig. 5). The LPS-associated histologic damage in sections from 25-d-old animals, as in sections from 10-d-old animals, was attenuated by pretreatment with dexamethasone (Fig. 5).

Figure 4
figure 4

Hematoxylin and eosin staining of rat colon from(A) 10-d-old control animal; (B) 10-d-old animal treated with LPS (3 mg/kg, i.p.) 4 h before sacrifice; (C) 10-d-old animal pretreated with dexamethasone (2 mg/kg, i.p.) 1 h before administration of LPS; and (D) 25-d-old animal treated with LPS (3 mg/kg, i.p.) 4 h before sacrifice. Arrows indicate areas of damage where a = epithelial disruption, b = vasocongestion, c = erythrocyte extravasation, and d = submucosal edema. Bar represents 50 μm.

Full size image
Figure 5
figure 5

Histologic damage score in sections of colon removed from 10-d-old animals (□) and 25-d-old animals (▪). Animals were administered LPS (3 mg/kg, i.p.) 4 h before sacrifice, dexamethasone(DEX; 2 mg/kg, i.p.) 1 h before administration of LPS or aminoguanidine (AG; 25 mg/kg, i.p.) at the same time as administration of LPS. Data are presented as mean ± SEM (n = 5-6). Two-way analysis of variance indicated significant differences between 10- and 25-d-old animals. #p < 0.05 for difference between 10- and 25-d-old LPS-treated animals by t test for unpaired data.**p < 0.01 for difference from untreated animals and†p < 0.01 for difference from animals treated with LPS alone within each age group as determined by one-way analysis of variance and Duncan's multiple range test.

Full size image

DISCUSSION

In this study we have demonstrated that cNOS activity is detectable in the colon of both preweaned (younger than 18 d of age) and postweaned (older than 21 d of age) rat pups. As far as we are aware this is the first report of ontogenic changes in cNOS activity in the colon of neonatal rats. Because cNOS activity significantly increased between 10 and 20 d of age, this may suggest that NO synthesized by cNOS may be involved in the functional changes, such as digestive enzyme development and tissue permeability, which occur during the period of weaning(9, 10). Indeed, ontogenic changes in cNOS activity have previously been reported in enterocytes isolated from the pig small intestine(19). A similar trend in cNOS activity was observed in this current study with the exception that, in the pig enterocytes, cNOS activity continued to rise postweaning, whereas in the rat, colonic cNOS activity fell postweaning. Other groups have reported similar ontogenic changes in neuronal NOS activity in both the rat and guinea pig forebrain and cerebellum(20). The significance of these alterations in cNOS activity is currently unclear. However, changes in cNOS activity may be related to the rapid replacement after birth of fetal cells by adult type colonocytes(21), which may be down-regulated after weaning.

As well as being involved in mediating physiologic processes, sustained production of NO can produce cytotoxic effects in a variety of mammalian cells, including cells of the gastrointestinal epithelium(22–24). Administration of LPS to adult rats in vivo results in the induction of iNOS activity, which has been detected between 3 and 24 h after administration(17). Furthermore, NO produced by iNOS has been implicated in the pathology of intestinal inflammation and may serve as an important mediator of diseases such as ulcerative colitis(8). The results of the present study confirm and extend previous findings in adult animals by demonstrating that iNOS activity can be induced, after administration of LPS, in the colon of neonatal as well as adult rats. In addition, we have demonstrated that, although there are small differences in the magnitude of LPS-induced iNOS activity related to the postnatal age of the animal, these effects were not found to be statistically significant. It is possible therefore that differences in the susceptibility to injury observed in animals aged between 10 and 25 d of age may be related to additional factors including changes in antioxidant capacity. This possibility is currently under investigation.

It is conceivable that NO is not the final mediator of mucosal damage, because interactions between NO and other cytotoxic reactive free radicals such as the superoxide anion have been reported(13). Musemeche et al.(11), demonstrated that both myeloperoxidase and xanthine oxidase activity follow similar ontogenic changes as iNOS activity and are maximal in the intestine of neonatal rats between 5 and 15 d of age. Xanthine oxidase catalyzes the reaction of hypoxanthine(metabolic product of ATP metabolism) and oxygen to form the superoxide anion, which sets the stage for further free radical generation and which under conditions of high NO or nitrite/nitrate availability (e.g. after LPS treatment) will include the peroxynitrite anion. Peroxynitrite is a highly reactive oxidizing agent, which has been reported to initiate lipid peroxidation and thus produce extreme cellular membrane damage(13). Indeed, peroxynitrite has previously been used to produce an animal model of colitis(25). The ability of aminoguanidine to inhibit LPS-induced lipid peroxidation suggests the involvement of iNOS, and subsequently NO, in this response.

In this study, the age at which maximal iNOS activity was detected coincided with the age at which the greatest degrees of lipid peroxidation and histologic damage were observed. These findings suggest that the immature rat colon is highly susceptible to the detrimental effects of LPS-induced iNOS activity. The suggestion that NO is involved in mediating this damage is supported by our findings that the steroidal antiinflammatory agent dexamethasone, which inhibits the expression of iNOS and the selective iNOS inhibitor aminoguanidine, increases in both lipid peroxidation and histologic damage. These findings provide strong evidence to suggest that induction of iNOS activity is indeed involved in mediating the LPS-induced damage.

In summary, the results of the present study suggest that cNOS and subsequently NO may be important during development of the neonatal rat colon. Furthermore, overproduction of NO after LPS administration may, at least in part, be responsible for the observed increase in histologic damage and lipid peroxidation. We have also demonstrated that the colon of the immature preweaned rat is more susceptible to LPS-induced damage than the colon of postweaned animals. The precise reasons for this greater susceptibility to damage are currently unclear.