Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

In mammals, intestinal secretion may be induced by several mechanisms: 1) the stimulation of active transcellular transport of anions and fluid into the lumen, as generated by cholera toxin, vasoactive intestinal peptide, or prostaglandin E₂; 2) an increase in epithelial paracellular PER that has been observed in the presence of mucosal inflammation or epithelial disruption (Farack and Loeschke, 1984); and/or 3) an increase in the intestinal blood flow also enhances intestinal secretion (Hansen et al., 1998). Opioids have been demonstrated to reduce intraluminal accumulation of fluid under basal (control) conditions, and in response to increased active secretion in various species, in vivo (Farmer and Burks, 1991; Lemcke et al., 1991) and in vitro (Schulzke et al., 1990; Sheldon et al., 1990). This effect of opioids has been reported to be related to the stimulation of the active Na⁺ and Cl absorption across the intestinal mucosa in rabbit (Binder et al., 1984) and guinea pig (Kachur et al., 1980). Functional and binding studies have demonstrated that the antisecretory action of opioids is produced by binding to different subtypes of OR, which may vary in the different animal species. Thus, in the rat, µ- and -OR located in myenteric and submucosal plexus of the gastrointestinal tract would be involved (Bagnol et al., 1997), whereas in mice, µ-, -, and -OR placed in the submucosal plexus (Sheldon et al., 1990), as well as OR (µ and ) present in the brain (Shook et al., 1989) and the spinal cord (Lemcke et al., 1991) are involved. In the guinea pig, -OR located on the nerve terminals of submucosal plexus (Kachur et al., 1980; Mihara and North, 1986), and µ- and -OR located in the cryptal enterocytes (Lang et al., 1996) of the gut could regulate water and electrolyte secretion. The relative contribution of the peripheral and central components in the overall response to systemic opioids is not completely characterized, but it is known to be dependent upon the route and the dose of administration.

Opioid effects on intestinal secretion and PER during inflammation are not well established. Using an experimental model of acute intestinal inflammation in mice, we have recently reported a significant increase in the inhibitory effects of µ-OR agonists on intestinal secretion and PER during acute inflammation (Valle et al., 2000). The aim of the present investigation was to evaluate and compare the effects of specific OR agonists on PER during acute and chronic intestinal inflammation in mice. Our working hypothesis was that the degree or intensity of the inflammatory process would modify the effects of opioids on intestinal PER, in a similar manner as that demonstrated by our group when evaluating gastrointestinal motility, where the potency of both µ- and -opioids was distinctly enhanced according to the degree of inflammation (Puig and Pol, 1998).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male Swiss CD-1 mice weighing 25 to 30 g were used in all experiments. The study protocol was approved by the local Committee of Animal Use and Care of our institution, in accordance with the International Association for the Study of Pain guidelines on ethical standards for investigations in animals. Mice were housed under 12-h light/dark conditions in a room with controlled temperature (22°C) and humidity (66%). Animals had free access to food and water and were used after a minimum of 4 days acclimatization to the housing conditions. All experiments were conducted between 9:00 AM and 5:00 PM.

Intestinal Inflammation. Two types of intestinal inflammation (acute and chronic) were used in our study. Acute inflammation was induced by the p.o. administration of a single dose (0.05 ml) of CO; control animals received the same volume of p.o. SS. Mice in the chronic treatment group were gavaged with a second dose of CO or SS (0.05 ml) 24 h after the first one (Fig. 1). In both instances, mice were fasted for 18 h before CO or SS administration, except for free access to water, which was available for the duration of the study. Chronic animals had access to food and water for a period of 6 h between the two doses of CO and for another one of 54 h after the second dose of CO; then, they were fasted again for 18 h before surgery. In the acute-CO group PER was evaluated 3 h after CO or SS, and in the chronic-CO PER was evaluated 96 h after the second dose of CO. These time points were selected based on previous studies from our laboratory demonstrating that they were the times of maximal epithelial injury in both models of intestinal inflammation. Morphological changes in the acute (Pol et al., 1995) and chronic models (Puig and Pol, 1998) of intestinal inflammation were demonstrated by electron and optical microscopy, respectively. No differences between controls and mice with acute-inflammation were observed under light microscopy; however, electron microscopy examination demonstrated an increased number of clear vesicles in the cytoplasm of the enterocytes, swollen mitochondria with disrupted cristae, and enlarged spaces filled with granular material in the extravascular compartment of the villi. During chronic inflammation, light microscopy showed a clear disruption of the mucosa, with a massive infiltration of lymphocytes within the submucosa. Preparations from acute and chronic saline-treated animals examined under light (chronic) and electron (acute) microscopy did not show morphological changes in the jejunum.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Experimental design showing the sequence of CO or SS administration, and the time (h) of evaluation of permeability in acute and chronic treatment groups.

Intestinal Permeability. PER was assessed by measuring the blood-to-lumen transfer of ⁵¹Cr-EDTA, according to the procedure used in our laboratory (Pol et al., 1999; Valle et al., 2000). Animals were surgically manipulated 2 or 95 h after the first dose of CO for the acute and chronic model, respectively. Mice were laparotomized under ether anesthesia, and both renal pedicles ligated to prevent rapid urinary excretion of ⁵¹Cr-EDTA. Animals were then allowed to recover for a period of 40 min and, at that time 4 µCi of ⁵¹Cr-EDTA was injected into the circulation via the right vein of the tail (i.v.). Forty-five minutes later, mice were sacrificed by cervical dislocation, the small intestine was removed, and the intestinal lumen washed with 0.4 ml of saline. Levels of radioactivity were determined in a gamma counter (1282 Compugamma, LKB-WALLAC, Pharmacia Ibérica, S.A., Barcelona, Spain) and results (cpm) expressed as the percentage of the dose administered.

Intracerebroventricular Injection. The i.c.v. injections were carried out into the left lateral ventricle of ether-anesthetized mice. Control animals received the same volume of vehicle. Injections were performed using a Hamilton microliter syringe (Microdispenser Socorex; PANREAC S.A., Barcelona, Spain) fitted with a 26-gauge needle, according to the method of Haley and McCormick (1957). The site of injection was 2 mm caudal and 2 mm lateral to the bregma, and 3 mm in depth from the skull surface.

Groups of Experiments. The groups of experiments performed are shown in Table 1. All drugs, regardless of the route of administration, were evaluated in control conditions (SS) and during acute and chronic inflammation. Initially, we tested the effects of the conventional (central and peripheral OR binding) opioid agonists morphine, fentanyl, DPDPE, and U50,488H, injected s.c.; we also determined their reversibility by specific antagonists (naloxone, -FNA, naltrindole, and MR-2266). The peripheral component of the effects was established using opioids that do not freely cross the blood-brain barrier (BBB) and by the i.c.v. administration of conventional opioids. We used 1) the s.c. administration of peripherally acting OR agonists (PL017, ICI-204,448), 2) the s.c. administration of conventional OR agonists in the presence of a peripherally acting OR antagonist (NX-ME), and 3) the i.c.v. administration of conventional OR agonists.

Drugs. We used morphine hydrochloride (Alcaiber S.A., Madrid, Spain); fentanyl (Syntex Latino, Madrid, Spain); U50,488H, DPDPE, naltrindole, ICI-204,448, NX-ME, and -FNA (Research Biomedical Incorporated, Wayuland, MA); MR-2266, a gift from Boehringer-Ingelheim, Mannheim, West Germany; naloxone hydrochloride (Sigma Chemical Co., St. Louis, MO); and PL017 (Peninsula Laboratories, Belmont, CA). All drugs were dissolved in sterile pyrogen-free 0.9% sodium chloride just before use. In our study, we will designate as "conventional opioids" those that cross the BBB and access the central nervous system according to their physicochemical characteristics, mainly lipid solubility and molecular weight. On the contrary, peripherally acting opioids (agonists and antagonists) are those that have limited accessibility into the central nervous system.

Data Analysis. The inhibitory effects of the opioid agonists were calculated as the percentage of inhibition of PER in an opioid-treated animals (test PER) compared with the mean PER obtained in the corresponding control group of vehicle-treated mice (n = 6-8) using the following equation: % inhibition = [vehicle PER  test PER)/(vehicle PER)] × 100.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Acute and Chronic Administration of CO on PER. PER was assessed in acute- and chronic-treated mice with p.o. CO or SS (n = 6-8). In these experiments, ⁵¹Cr-EDTA was injected i.v., 2 h 45 min or 95 h 45 min following the first dose of CO for the acute and chronic treatments, respectively. Figure 2 shows that the blood-to-lumen transfer of ⁵¹Cr-EDTA was increased 2.5 times (P < 0.001, Student's t test) during acute inflammation (0.40 ± 0.05% in SS versus 1.02 ± 0.03% in CO). Likewise, chronic inflammation induced a 3.2-fold increase in PER (1.27 ± 0.07%, P < 0.001, Student's t test), compared with the respective control group (0.39 ± 0.04%). Statistically significant differences were also observed when acute and chronic treatments with CO (but not SS) were compared (P < 0.05, Student's t test).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Effects of acute and chronic treatment with saline () or croton oil () on percentage of permeability to 51Cr-EDTA. Results are expressed as mean values ± S.E. from six to eight mice. ***P < 0.001 compared with saline (Student's t test). Statistical differences were observed between acute and chronic administration of croton oil (P < 0.05, Student's t test).

Inhibitory Effects of µ-OR Agonists on PER. The effects of morphine (mixed µ/-OR agonist) and fentanyl (selective µ-OR agonist) on PER were evaluated in control animals and during acute and chronic inflammation (6-8 animals/dose). Morphine induced a dose-related inhibition of PER in both acute- and chronic-treated animals (Fig. 3A). Because the curves obtained in acute and chronic SS-treated mice were superimposed, the latter is represented as control. In all instances, dose-response curves showed coefficients of correlation close to 1 and were parallel without significant differences in their slopes (acute: SS, 47.6 ± 2.7 and CO, 41.9 ± 1.6; chronic: SS, 45.3 ± 2.8 and CO, 45.7 ± 1.7). During acute and chronic inflammation, the curves were shifted to the left, demonstrating an increase in the effect. ED₅₀ values were obtained from each of the four dose-response curves as a measure of potency (Table 2). The results show that the inhibitory potency of morphine was 3.8-fold increased during acute and 8.7-fold during chronic inflammation; in all curves, E_max values ranged between 87 and 94%. Fentanyl also induced a dose-related inhibition of PER in all experimental conditions (Fig. 3B). As with morphine, dose-response curves showed coefficients of correlation close to 1, with no significant differences in their slopes (acute: SS, 53.7 ± 5.2 and CO, 60.1 ± 3.5; chronic: SS, 56.9 ± 5.8 and CO, 46.5 ± 3.5); the curves obtained in acute and chronic SS-treated mice were superimposed and those ones obtained during inflammation appeared shifted to the left in a parallel manner. The ED₅₀ values were 2.0 and 4.3 times decreased in acute and chronic CO-treated groups, respectively (Table 2); E_max values were not significantly different among themselves (range 89.5-100.1%).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Dose-related inhibition of permeability in mice by morphine (A) or fentanyl (B) in controls (chronic saline, CR-SS) and during acute (AC-CO) and chronic (CR-CO) croton oil. Each point represents the mean ± S.E. expressed in mg/kg from six to eight mice. *P < 0.05, when acute and chronic croton oil were compared with controls (Student-Newman-Keuls test). Results obtained in acute saline animals were not significantly different from those observed in chronic saline mice.

Inhibitory Effects of - and -OR Agonists on PER. We evaluated the inhibitory effects of DPDPE (a selective -OR agonist) and U50,488H (a selective -OR agonist) on PER. DPDPE produced dose-related inhibitions of PER in acute and chronic SS- and CO-treated mice (Fig. 4A; n = 6-8 animals/dose); in all instances, dose-response lines showed coefficients of correlation close to 1, even though the slopes obtained in control groups (SS-acute: 15.2 ± 2.1, SS-chronic: 17.4 ± 3.4) were significantly lower than those obtained in the groups with inflammation (CO-acute: 30.0 ± 0.5, CO-chronic: 37.8 ± 5.2); thus, curves obtained in SS- and CO-treated animals were not parallel. In controls (SS), the curves obtained were superimposed and showed a maximal inhibition of 50.4 to 52.6% (using a dose range of 0.01-20 mg/kg). The lines obtained in acute and chronic CO-treated mice were parallel among themselves and shifted to the left from the controls, with E_max values of 91.9 and 85.5%, respectively. Comparative analysis of the results demonstrated a significant effect of treatment and dose on percentage of inhibition of PER (P < 0.05, two-way ANOVA), but no significant interaction was observed between the two factors. The responses obtained after the administration of each individual dose were analyzed by Student's t test or one-way ANOVA, whenever applicable. The results show that acute and chronic treatment with CO significantly increased the inhibitory effects of DPDPE on PER at all doses tested (P < 0.05). In fact, low doses of DPDPE (0.003 and 0.005 mg/kg) that did not produce an inhibition of PER in SS-treated animals had a substantial effect during acute and chronic inflammation. In an attempt to compare the relative potencies of DPDPE in the absence and presence of inflammation, we used the ratios of the calculated ED₅₀ values (Table 3); the results show that the doses required to produce a 50% inhibition of the maximal effect were 4.7 and 11.1 times lower than control, during acute and chronic inflammation (Table 3). Since the dose-response curves to DPDPE were not parallel, the ED₅₀ values were also obtained using polynomial quadratic regression analysis, which eliminates normalization of the results; the resulting ED₅₀ values were saline (SS-CR) 0.109 ± 0.02, acute inflammation (AC-CO) 0.022 ± 0.007, and chronic inflammation (CR-CO) 0.009 ± 0.002 mg/kg. These values are similar to those obtained using linear regression analysis (Table 3).

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Inhibitory effects of subcutaneous DPDPE (A) and U50,488H (B) on PER in controls (chronic saline, CR-SS) and during acute (AC-CO) and chronic croton oil (CR-CO). Each point represents the mean ± S.E. expressed in mg/kg from six to eight mice. *P < 0.05, when acute and chronic croton oil were compared with controls (Student-Newman-Keuls test). Results obtained in acute saline did not significantly differ from those observed in chronic saline mice.

Antagonism of the Inhibitory Effects of µ-, -, and -OR Agonists by Opioid Antagonists. To evaluate the specificity of the observed responses in the presence of intestinal inflammation, the effects of µ-, -, and -OR agonists were assessed after the administration of naloxone (0.1 mg/kg), naltrindole (1 mg/kg), and MR-2266 (3 mg/kg). The doses of the opioid antagonists were selected on the basis of previous studies reporting selective blockage of the different types of OR (Magnam et al., 1982; Portoghese et al., 1988). In these experiments we tested the effects of the ED₅₀ values of the agonists (obtained from their respective dose-response curves), in the presence of the above-mentioned doses of the antagonists (n = 6-8 mice/dose). Our results show that in all experimental conditions the effects of morphine and fentanyl were completely reversed by naloxone, whereas those of DPDPE and U50,488H were antagonized by naltrindole and MR-2266, respectively. These results demonstrate that the enhanced effects of the agonists during acute and chronic inflammation are mediated by specific OR.

Reversibility of the Inhibitory Effects of Morphine by -FNA. To approximate the fraction of the total number of µ-ORs that mediate the enhanced response to morphine during acute and chronic inflammation, we used -FNA, a competitive nonreversible µ-OR antagonist. For this purpose, the inhibitory effects of 3 mg/kg morphine were evaluated in the presence of increasing doses of i.p. -FNA (5-200 µg) using six to eight animals per dose tested. -FNA was administered 150 min before morphine to avoid its initial agonist effects (Ward et al., 1982). In all experimental conditions, the inhibitory effects of morphine on PER were antagonized by -FNA in a dose-dependent manner (Fig. 5). The effects of the treatment and those of the -FNA were analyzed by two-way ANOVA. The results show a significant effect of the two factors (P < 0.001) and of their interaction (P < 0.01) on percentage of inhibition of PER. When the results were compared according to the individual doses of -FNA used, significant differences were observed between acute SS- and CO-treated mice (P < 0.05, Student's t test) at 5 and 10 µg, but not at 20 and 50 µg of the antagonist. During chronic inflammation, significantly higher doses of -FNA were required to antagonize the effects of morphine at each dose point (P < 0.05, Student-Newman-Keuls). These results suggest that a similar population of OR would mediate the observed response in SS- and acute CO-treated animals, whereas a recruitment of OR may occur in the presence of chronic inflammation. We did not examine the saline control during acute inflammation because the inhibitory effects produced by morphine during acute and chronic inflammation at all doses tested were similar.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Antagonism of the inhibitory effects of subcutaneous morphine (3 mg/kg) by increasing doses of intraperitoneal -FNA (µg) in control animals (chronic saline, CR-SS) and during acute (AC-CO) or chronic croton oil (CR-CO). Each point represents the mean value obtained from six to eight mice. Different letters (a-c) indicate significant differences between treatment groups (CR-SS, AC-CO, CR-CO) (P < 0.05). Comparison for statistical analysis for the three groups was carried out using the Student-Newman-Keuls test, whereas at the doses where only two values were available a Student's t test was used.

Peripheral Component of the Opioid Effects on PER during Intestinal Inflammation. To evaluate the role of peripheral (intestinal) OR in the enhanced response to opioids during acute or chronic inflammation, the effects of two peripherally acting agonists, PL017 (µ) and ICI-204,448 (), were assessed using six to eight animals per dose. Peripheral -OR agonists were not used because they were not commercially available. PL017 produced a dose-related inhibition of PER in SS-treated animals and during acute and chronic inflammation. In all cases, dose-response curves showed coefficients of correlation close to 1 and similar slopes (acute: SS, 36.9 ± 4.0 and CO, 37.0 ± 2.2; chronic: SS, 39.4 ± 4.0 and CO, 40.5 ± 5.9). The curves obtained during inflammation were shifted to the left in a parallel manner from those obtained in controls. When ED₅₀ values were calculated, a 2.2- and 5-fold increase in the potency of PL017 was observed during acute and chronic inflammation, respectively (Table 4); E_max values ranged between 73.7 and 80.8%. Statistical analysis of results show that the inhibitory effects of PL017 increased significantly during acute and chronic inflammation (P < 0.05, Student-Newman-Keuls); significant differences were also observed between these two groups of study. The peripheral -OR agonist ICI-204,448 revealed a similar inhibitory profile to U50,488H. When administered at a dose range of 3 to 30 mg/kg, the E_max values were 46.7 and 48.5% for controls, whereas the E_max values were 49.5 and 50.3% in the acute and chronic CO groups, respectively. ED₅₀ values decreased 4.2- and 5.8-fold in the presence of acute and chronic inflammation; the analysis of the results showed that treatment with CO produced a significant increase in the potency of ICI-204,448 (P < 0.05, Student-Newman-Keuls), but no differences between acute and chronic inflammation. These results show that the enhanced effects of µ- and -OR agonists have an important peripheral component.

View larger version (31K): [in this window] [in a new window]   Fig. 6.   Effects of subcutaneous vehicle and NX-ME (0.3 mg/kg) on the inhibitory action of the ED50 values of subcutaneous morphine (CR-SS = 6.98 mg/kg; AC-CO = 1.87 mg/kg; CR-CO = 0.80 mg/kg) (A); fentanyl (CR-SS = 0.052 mg/kg; AC-CO = 0.025 mg/kg; CR-CO = 0.012 mg/kg) (B), and DPDPE (CR-SS = 0.09 mg/kg; AC-CO = 0.021 mg/kg; CR-CO = 0.008 mg/kg) (C). Results shown on percentage of PER of 51Cr-EDTA. Each column represents the mean ± S.E. from six to eight mice. For each treatment, *P < 0.05 when compared with the other groups (control permeability without drugs, and the effect of the agonist plus naloxone-methiodide; Student-Newman-Keuls test). CR-SS, chronic saline; AC-CO, acute croton oil; and CR-CO, chronic croton oil.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Dose-related inhibition of permeability (%) by intracerebroventricular administration of morphine, fentanyl, and DPDPE during chronic inflammation. Each point represents the mean value ± S.E from six to eight mice (in nmol).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The effects of opioids on intestinal PER have been evaluated under physiological conditions and in response to different secretagogues, but their response during intestinal inflammation is not well established. The present study reports the effects of receptor-specific opioids on PER in two models of intestinal inflammation (acute and chronic) in mice. In both instances we used CO as a proinflammatory agent, whose effects have been demonstrated in the cornea of the rabbit (Villena et al., 1999), the skin of the rat (Blazso et al., 1999), and the small intestine of mice (Pol et al., 1994; Puig and Pol, 1998). In our previous studies distinct morphological changes could be established after acute (a single dose of CO) and chronic (two doses) administration of the inflammatory agent, thus permitting the evaluation of the effects of opioids during two different stages of the inflammatory process.

In the present study we used ⁵¹Cr-EDTA to evaluate the transfer of fluid and electrolytes to the intestinal lumen. ⁵¹Cr-EDTA is a chemically stable hydrophilic chelate of small molecular size that is not metabolized in the tissues, resists bacterial degradation (Travis and Menzies, 1992; Bjarnason, 1995), and can permeate the intestinal mucosa through paracellular shunts by diffusion and/or solvent transport (Peeters et al., 1994). When blood-to-lumen transfer of ⁵¹Cr-EDTA is assessed in animals, the ligature of the renal pedicles is mandatory to avoid the urinary excretion of the marker. In our experimental conditions renal exclusion induced a mild (acute) renal insufficiency reflected by an increase in plasma urea; however, the metabolic irregularity did not alter the inhibitory effects of opioids on intestinal secretion or PER (Valle et al., 2000).

Another methodological aspect that we would like to point out is that the precise mechanism/s by which CO increases intestinal PER is unknown. Our results suggest that CO may induce an increase in the passive filtration across the paracellular pathway. Similar increases in fluid secretion and PER have been reported in vivo in response to deoxycholic acid, an agent that has been demonstrated to reduce the resistance of the paracellular pathway and increase electrolyte and fluid secretion within the intestinal lumen by this route (Goerg et al., 1983; Farack and Loeschke, 1984).

All OR agonists tested in our model of CO-induced inflammation showed an increase in potency during chronic > acute inflammation when drugs were administered s.c. These results suggest that during inflammation, the OR involved in the modulation of fluid and electrolyte transport are sensitized or up-regulated. Interestingly, the highest increase (11 times) in potency was observed during chronic inflammation with DPDPE, followed by morphine (8.8 times), a mixed µ/-OR agonist. On the basis of these results we would like to postulate that -OR play the most relevant role inhibiting PER during chronic inflammation. However, the precise location of the -OR involved in the inhibition of intestinal PER in our model cannot be established from the present experiments. Most probably, OR located on the submucous and myenteric plexuses, as well as those found on sensory neurons that modulate mesenteric blood flow (Li and Duckles, 1991) could be involved in the enhanced response to -opioids during intestinal inflammation.

The effects mediated by fentanyl (µ-OR agonist) and U50,488H (-OR agonist) were also enhanced during inflammation, although to a lower degree, suggesting that these receptors are also sensitized during inflammation. Previous studies from our laboratory have shown that the increased potency of opioids on gastrointestinal transit occurs only in the presence of inflammatory diarrhea; however, the p.o. administration of castor oil (an agent that induces hypersecretory diarrhea without inflammation) did not alter the potency of the opioids (Pol et al., 1996; Pol and Puig, 1997). In the present study, we tested the inhibitory effects of the ED₅₀ values of morphine and fentanyl on PER in animals that (instead of CO) received intragastric castor oil (data not shown), without significant changes in their effects. Thus, the potency of opioids is only enhanced in the presence of inflammatory diarrhea.

Due to the availability of -FNA, a competitive nonreversible µ-OR antagonist, we could evaluate the performance of the OR during acute and chronic inflammation. We assessed the effects of a fixed dose of morphine (3 mg/kg) in the presence of increasing doses of -FNA, a drug that binds covalently to µ-OR and thereby decreases the number of available OR (Mjanger and Yaksh, 1991). In our study, the antagonism of the effects of morphine in control animals and during acute inflammation was similar in the presence of high (but not low) doses of -FNA. These results suggest that the actual number of functional OR is unaltered during acute inflammation, and that the enhanced effects may be related to other factors such as a decrease in local pH (Pol and Puig, 1997), which could increase opioid efficacy to inhibit adenylyl cyclase (Selley et al., 1993), and/or the disruption of the perineurium, facilitating the access of agonists to OR (Antonijevic et al., 1995).

During chronic inflammation, significantly higher doses of -FNA were required to antagonize morphine effects, suggesting an increase in the number of functional µ-OR. This fact could be related to an enhanced expression of the gene that codifies for µ-OR in neurons located in submucosal or myenteric plexus or in mucosal epithelial cells in this experimental condition (Pol and Puig, 1999). The most relevant results of the present study are the enhanced inhibitory effects of the -OR agonists during intestinal inflammation. The increase was greater than that obtained for µ-OR agonists in the same experimental conditions. When the dose-response curves to DPDPE were analyzed, it was observed that in control conditions (no inflammation) E_max values hardly reached 50%, whereas during inflammation they were above 80%. The increase in E_max values suggests an augmentation in the number of functional -OR, but this possibility could not be tested pharmacologically, due to the unavailability of specific nonreversible -OR antagonists. A potential increase in the number of -OR during chronic inflammation (due to an enhanced gene expression or synthesis of the receptor protein), would not exclude a sensitization of intestinal -OR or an enhanced externalization of cytoplasmatic -OR (neuronal or not).

Another characteristic of the dose-response curves for DPDPE was that the control lines were not parallel to those obtained during inflammation, suggesting that a different subpopulation of -OR, not characterized in the intestine, would become functional (or available) during intestinal inflammation. Since dose-response relationships were not parallel, we estimated the relative potencies of DPDPE on the basis of the ED₅₀ values obtained using two methods of analysis: a linear regression of the straight segments of the curves, and a quadratic polynomial analysis; the ED₅₀ derived from both methods of analysis were similar, thus validating the comparison of the potencies of DPDPE in the different experimental conditions.

The -OR do not seem to be highly involved in the regulation of PER during inflammation. The increased inhibitory effects of U50,488H during acute and chronic inflammation differ with its effects on intestinal transit where no significant changes were observed during acute (Pol et al., 1994) or chronic (Puig and Pol, 1998) inflammation. The discrepancy suggests a different purpose or localization of -OR when regulating motility or PER.

The inhibitory effects of the agonists were reversed by specific antagonists, demonstrating that the enhanced effects of µ-, -, and -OR agonists during inflammation are mediated by interaction with specific OR.

The potency of peripherally acting µ- and -OR agonists in both models of inflammation was similarly increased compared with conventional opioids, suggesting that in both instances, the effects are mediated by peripheral OR. This assumption is supported by 1) the complete antagonism of µ- and -OR agonists by NX-ME during inflammation, and 2) the fact that the potency of i.c.v. morphine, fentanyl, and DPDPE remained unaltered in the presence of peripheral (intestinal) inflammation. These experiments suggest that the enhanced effects of the tested opioid agonists, when administered s.c., are mediated by peripheral OR. In control conditions, µ-OR agonists demonstrated a higher potency when administered i.c.v., a finding previously reported on intestinal transit (Shook et al., 1987; Pol et al., 1999) or secretion (Shook et al., 1989; Jiang et al., 1990). On the contrary, the potency of s.c. DPDPE on PER was much higher than that observed after i.c.v. administration, a fact that suggests a low diffusion of DPDPE in nervous tissue or a low density of central -OR involved in the physiological modulation of PER.

It has been recently reported that the permeability of the BBB to serum proteins increases after the peripheral injection of Freund's adjuvant in mice (Rabchevsky et al., 1999), an effect that was observed after 2 to 3 weeks of the treatment. Thus, the possibility that CO-induced inflammation alters the permeability of the BBB, although unlikely due to the differences in the extent and duration of the inflammatory process, cannot be excluded at present. Other investigators have been unable to establish changes in BBB permeability after high doses of i.p. endotoxin in the rat (Bickel et al., 1998). Thus, until further evidence is available it would be reasonable to assume that inflammatory changes induced by the administration of two doses of CO may not significantly alter the permeability of the BBB to peripherally acting opioids. Another aspect that has not been tested in our study is the possibility that peripheral inflammation may alter the pharmacokinetics (distribution) of opioids, inducing their accumulation in the injured tissue.

In summary our results show that intestinal inflammation increases the potency of opioid agonists on the inhibition of intestinal PER with a receptor-specificity of - > µ- > -OR, and that the effects are more prominent during chronic inflammation. The enhanced effects could be related to a sensitization or up-regulation of OR located in the neuronal or intestinal epithelial cells that participate in the modulation of PER during inflammation.