AJP - GI Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Gastrointest Liver Physiol 277: G847-G854, 1999;
0193-1857/99 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Reuter, B. K.
Right arrow Articles by Wallace, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Reuter, B. K.
Right arrow Articles by Wallace, J. L.
Vol. 277, Issue 4, G847-G854, October 1999

Phosphodiesterase inhibitors prevent NSAID enteropathy independently of effects on TNF-alpha release

Brian K. Reuter and John L. Wallace

Department of Pharmacology and Therapeutics, Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada T2N 4N1


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Although the ability of nonsteriodal anti-inflammatory drugs (NSAIDs) to injure the small intestine has been well established in humans and animals, the mechanism involved in this type of injury has yet to be elucidated. The cytokine tumor necrosis factor-alpha (TNF-alpha ) has recently been demonstrated to play a critical role in the pathogenesis of NSAID-induced gastric damage. We therefore assessed the possibility that TNF-alpha is similarly involved in the pathogenesis of NSAID-induced small intestinal injury. Administration of multiple doses (n = 4) of diclofenac, but not a single dose, resulted in profound macroscopic damage in the intestine and significantly increased levels of TNF-alpha in intestinal tissue and bile. Pretreatment of rats with a phosphodiesterase inhibitor, pentoxifylline, theophylline, or rolipram, significantly attenuated the macroscopic intestinal ulceration produced by diclofenac administration. However, inhibition of TNF-alpha release with thalidomide or immunoneutralization with a polyclonal antibody directed against TNF-alpha failed to afford any protection. These results suggest that the cytokine TNF-alpha does not play a critical role in NSAID-induced small intestinal injury. Therefore, phosphodiesterase inhibitors mediate their protective effect through a mechanism independent of TNF-alpha synthesis inhibition.

ulcer; inflammation; enteritis; cyclooxygenase; nonsteroidal anti-inflammatory drugs; tumor necrosis factor-alpha


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

the ability of nonsteroidal anti-inflammatory drugs (NSAIDs) to cause ulceration of the stomach is well known. However, in the past two decades it has become apparent that these drugs also have the ability to produce extensive damage in the small intestine. Indeed, the incidence of NSAID-induced small intestinal injury has been suggested to be equivalent to, if not greater than, that of NSAID gastropathy. Bjarnason and colleagues (5) estimated that 60-70% of chronic NSAID users exhibit enteropathy, which is generally characterized by low-grade blood and protein loss. Additional studies conducted in rheumatoid arthritis patients using NSAIDs demonstrated that ~40% had small intestinal lesions or ulcers (24, 25). Furthermore, the incidence of lower bowel hemorrhage or perforation was twofold greater in patients taking anti-inflammatory drugs than in patients receiving other types of drugs (19).

Although research has been conducted to determine potential factors responsible for NSAID-induced small intestinal injury, the mechanism has yet to be elucidated. The majority of knowledge has been obtained primarily through the use of animal models. It is likely that the initial injury produced by NSAIDs is due to the topical irritant properties, and in the longer term this injury is exacerbated by enteric bacteria (26, 39). The role of enteric bacteria in NSAID-induced small intestinal injury has been substantiated by the observations that NSAID administration results in increased enteric bacterial numbers within the lumen of the gut and that the administration of broad-spectrum antibiotics reduces the severity of damage (17, 26, 39). Furthermore, germ-free rats have been found to be less susceptible to NSAID-induced injury than rats raised in conventional housing conditions (23, 28). The mechanism responsible for the initial intestinal injury has not been determined, but the topical irritant properties of NSAIDs and recurrent exposure of the intestinal mucosa to the drug through enterohepatic circulation are likely to be important. Evidence in support of this hypothesis is as follows: 1) NSAIDs that do not undergo enterohepatic circulation were found not to cause small intestinal injury (21, 26); 2) interruption of enterohepatic circulation by bile duct ligation or by cholestyramine administration significantly attenuated intestinal injury (6, 37, 39); and 3) when isolated intestinal segments (Thiry loops) were created and an NSAID was administered, the loops were spared of damage, whereas the anastomosed intestine was not (6). However, Yamada et al. (39) suggested that the presence of an NSAID in the lumen of the intestine is not in itself sufficient to produce damage. They found that the NSAID must be associated with bile or a component found in bile in order to be toxic to cultured intestinal epithelial cells (39).

Jackson and colleagues (14, 15) demonstrated that the cytokine tumor necrosis factor-alpha (TNF-alpha ) is a constitutive protein found in bile of various species, including rats. TNF-alpha may represent the biliary component that helps potentiate NSAID-induced small intestinal injury. TNF-alpha has been demonstrated to play a critical role in experimental NSAID gastropathy (2, 32), and recently a correlation between increased TNF-alpha levels and indomethacin-induced small intestinal injury has been demonstrated (4). We therefore wished to determine whether TNF-alpha plays an important role in mediating NSAID-induced small intestinal injury. We also wanted to establish whether bile is the key source of TNF-alpha and, if so, whether TNF-alpha represents the component of bile that combines with NSAIDs to produce the initial injury in the small intestine after NSAID administration. To answer these questions, we examined the effect of various TNF-alpha inhibitors, including phosphodiesterase inhibitors and thalidomide, and an anti-TNF-alpha antibody on NSAID-induced small intestinal injury. We also determined the effect of NSAID administration on bile and small intestinal tissue levels of TNF-alpha .


    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Male Wistar rats (250-300 g) were obtained from Charles River Breeding Farms (Montreal, PQ, Canada) and fed standard rodent chow ad libitum. All experimental procedures were approved by the Animal Care Committee of the University of Calgary.

NSAID-induced small intestinal injury. Small intestinal injury was induced by orogastric administration of diclofenac (10 mg/kg, n >=  10). Diclofenac was administered at 12-h intervals for 2 days (i.e., a total of 4 doses). The NSAID was initially dissolved in DMSO (5% by final volume) and then diluted in 0.5% carboxymethylcellulose. Twelve hours after the final dose of diclofenac, the rats were killed by cervical dislocation and the entire small intestine was removed. The intestine was opened along the antimesenteric border and scored for damage by an observer unaware of the treatment the rats had received. Scoring consisted of measuring the area of all the ulcers with digital calipers and summing the areas to give a final damage score. We previously showed by light microscopy that the damage produced in this model consists of ulcers that penetrate through the muscularis mucosae (26).

To determine the effects of TNF-alpha blockade on diclofenac-induced small intestinal injury, rats were pretreated with various inhibitors of TNF-alpha synthesis 30 min before each dose of diclofenac. Groups of rats (n = 4-7) received an intraperitoneal injection of pentoxifylline (50, 100, or 200 mg/kg), theophylline (50 mg/kg), rolipram (0.3, 1, or 3 mg/kg), or thalidomide (10, 20, or 50 mg/kg) 30 min before each dose of diclofenac. All the drugs have previously been shown to inhibit TNF-alpha release and/or synthesis to prevent NSAID-induced gastrointestinal injury at the doses used (2, 20, 29).

To further examine the role of TNF-alpha in diclofenac-induced small intestinal injury, rats (n = 5) were pretreated with 1.5 ml/kg ip of rabbit anti-recombinant mouse TNF-alpha antiserum. The antiserum has previously been shown to immunoneutralize rat TNF-alpha at the dose used (2, 7, 9, 11). Control rats (n = 4) received normal rabbit serum (NRS), which was heat inactivated (56°C for 1 h) and adjusted to a protein concentration equivalent to that of the anti-TNF-alpha antiserum. The anti-TNF-alpha antibody or NRS was administered 4 h before each dose of diclofenac (4 doses at 12-h intervals), and the rats were killed 12 h after the final dose of diclofenac. Small intestinal damage was assessed as described above.

Bile and tissue levels of TNF-alpha . Groups of rats (n = 4-7) received a single dose or multiple doses (4 doses at 12-h intervals) of vehicle, diclofenac (10 mg/kg), or nitrofenac (15 mg/kg). Nitrofenac is a nitric oxide (NO)-releasing derivative of diclofenac that we previously showed did not produce gastrointestinal injury, although it maintained its inhibitory effects on cyclooxygenase (26, 36). Nitrofenac was used to permit a comparison between ulcerogenic and nonulcerogenic NSAIDs and their effects on bile and tissue levels of TNF-alpha . Three hours after the final dose of vehicle or test drugs, the rats were anesthetized with pentobarbital sodium and the common bile duct was ligated with polyethylene tubing (PE-10, Clay Adams, Parsipany, NJ). Bile was collected for 1 h and then stored at -70°C for subsequent determination of TNF-alpha levels by a commercially available ELISA kit (Cytoscreen, Biosource, Camarillo, CA).

After a treatment protocol identical to that described above, groups of rats (n = 5-11) were killed by cervical dislocation 3 h after the final dose of vehicle or test drug and the entire small intestine was removed. Tissue samples (~100 mg/sample) were obtained from five regions: 1) ligament of Treitz, 2) one-fourth the distance to the ileocecal junction, 3) one-half the distance to the ileocecal junction, 4) three-fourths the distance to the ileocecal junction, and 5) 10 cm proximal to the ileocecal junction. Samples were immediately frozen in liquid nitrogen and stored at -70°C. Samples were prepared for determination of TNF-alpha levels following a method similar to that described by Ribbons et al. (27). Briefly, frozen samples were ground to a fine powder with a mortar and pestle prechilled with liquid nitrogen. Isotonic saline was then added to the ground tissue (1 µl/mg wet tissue wt), and the suspensions were mixed on a vortex. The samples were centrifuged at 10,000 g for 20 min at 4°C. Supernatants were collected and stored at -70°C until TNF-alpha and protein concentration levels were determined.

Inhibition of TNF-alpha synthesis/release. The ability of the phosphodiesterase inhibitors (pentoxifylline, theophylline, and rolipram) and thalidomide to inhibit TNF-alpha synthesis/release in vivo was determined by stimulating TNF-alpha production in rats via administration of lipopolysaccharide (LPS). LPS (5 mg/kg ip, Escherichia coli serotype 0127:B8) was administered to rats (n = 4-14/group) 30 min after the administration of vehicle (DMSO or saline), pentoxifylline (50 or 200 mg/kg), theophylline (50 mg/kg), rolipram (0.3, 1, or 3 mg/kg), or thalidomide (10, 20, or 50 mg/kg). Three hours after LPS the rats were anesthetized with pentobarbital sodium and 1 ml of blood was drawn from the descending aorta into a syringe containing sodium citrate (100 µl of 3.8% wt/vol in saline). The blood was transferred to plastic Eppendorf tubes and centrifuged at 16,000 g for 2 min. Supernatants were collected and stored at -70°C for subsequent determination of plasma TNF-alpha levels by ELISA.

TNF-alpha mRNA expression. TNF-alpha mRNA expression was measured using RT-PCR. Samples of the small intestine (full thickness) were taken from rats treated with a single administration of LPS (5 mg/kg) or with a single dose or multiple doses of vehicle or diclofenac (10 mg/kg). In rats that received multiple doses of diclofenac to induce intestinal injury, tissue samples were obtained from sites that exhibited macroscopically visible damage. Intestinal samples were obtained from a similar location in the rats without macroscopic damage (i.e., rats treated with vehicle, LPS, or a single dose of diclofenac). The tissue samples were immediately frozen in a 50% (wt/vol) guanidinium solution containing 26.4 mM sodium citrate (pH 7.0), 0.528% sarcosyl, and 0.0072% beta -mercaptoethanol. For each 100 mg of tissue, 1 ml of the guanidinium solution was used. Total RNA was isolated using the acid guanidinium isothiocyanate method, as described previously (8).

RT-PCR was carried out following a previously established method (11). TNF-alpha RT-PCR products were made using primers designed according to the published rat TNF-alpha sequence (18). The TNF-alpha primer sequences were as follows: 5'-CCA CCA CGC TCT TCT GTC TAC T-3' (upstream) and 5'-CCA CAC TTC ACT TCC GGT TCC T-3' (downstream). The expected length of this PCR product was 1,000 bp. The glyceraldehyde 3-phosphate dehydrogenase (GAPDH) RT-PCR product was made using primers described previously (38). Cycle tests indicated that amplification of TNF-alpha with GAPDH was optimal if the TNF-alpha gene was amplified for 31 cycles and the GAPDH gene was amplified for 18 cycles (data not shown). GAPDH upstream and downstream primers were therefore added to the PCR mixture during the hot start of cycle 14.

PCR products were run on a 1.65% agarose gel containing ethidium bromide, and a Polaroid picture of the gel was taken under ultraviolet light. The level of TNF-alpha mRNA expression was determined using a densitometer and National Institutes of Health software. Quantities of TNF-alpha were normalized according to control levels of GAPDH and expressed as percentage of control.

Biliary excretion of diclofenac. Bile samples were collected from rats that had been pretreated with an intraperitoneal injection of saline or pentoxifylline (200 mg/kg) 30 min before they received a single dose or multiple (n = 4) doses of diclofenac (10 mg/kg po). Bile was collected via cannulation of the common bile duct over a 1-h period at 1, 3, 6, and 12 h after diclofenac administration. Samples were stored at -20°C for subsequent analysis by reverse-phase HPLC.

Following a modified procedure that we have described previously (13), biliary levels of diclofenac were determined by HPLC. Briefly, a 100-µl sample of bile was deconjugated by the addition of 50 µl of 2 mol/l sodium hydroxide. The sample was then neutralized by addition of an equivalent amount of 2 mol/l hydrochloric acid. Finally, the pH of the sample was lowered to ~4.5 by addition of 250 µl of 1 mol/l potassium phosphate (monobasic). Mefenamic acid (25 µl, 100 µg/ml in acetonitrile) was utilized as an internal standard. The drugs were extracted with 1 ml of acetonitrile. Samples were then centrifuged for 30 min at 2,000 g, and the organic layer was transferred to clean glass tubes. The organic layer was evaporated to dryness (Speed Vac, Savant Instruments, Holbrook, NY) and then reconstituted in mobile phase consisting of acetonitrile and 0.3% acetic acid (60:40). A 100-µl aliquot was injected onto the HPLC system consisting of an HPLC instrument (model 1050, Hewlett-Packard, Palo Alto, CA) with a 5-µm C18 analytical column (HP Spherisorb ODS2, 250 × 4 mm, Hewlett-Packard). All analyses were performed at ambient temperature and at a flow rate of 1 ml/min. Diclofenac and mefenamic acid were measured at a detection wavelength of 275 nm and had retention times of 6 and 9 min, respectively. Concentration of diclofenac was determined by interpolation from calibration curves. Diclofenac calibration curves were constructed by addition of appropriate volumes of stock (100 µg/ml in methanol) or diluted stock solutions to yield final concentrations of 1.0-100 µg/ml.

Materials. Diclofenac sodium, mefenamic acid, pentoxifylline, and theophylline were obtained from Sigma Chemical (St. Louis, MO). Thalidomide and rolipram were obtained from Research Biochemicals International (Natick, MA). Nitrofenac was kindly provided by NicOx (Nice, France). The anti-TNF-alpha serum was kindly provided by Drs. Cory M. Hogaboam and Steven L. Kunkel (University of Michigan, Ann Arbor, MI). All other products were obtained from VWR Canlab (Mississauga, ON, Canada).

Statistical analysis. Values are means ± SE. Comparison between two experimental groups was performed using a Student's t-test. Comparison among three or more experimental groups was performed using a one-way ANOVA followed by a Dunnett's multiple comparison test or a Bonferroni post hoc test. An asociated probability (P value) of <5% was considered significant.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Bile and tissue levels of TNF-alpha . TNF-alpha levels in bile of rats treated with a single dose of diclofenac or nitrofenac were not significantly different from those of vehicle-treated rats (Fig. 1). However, after multiple administrations of diclofenac, a significant increase in the levels of TNF-alpha in bile was found. Diclofenac-treated rats exhibited TNF-alpha levels more than twice that of the vehicle-treated group (P < 0.01). In rats treated with multiple doses of nitrofenac, TNF-alpha levels in bile were not different from those in the vehicle group.


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 1.   Effect of nonsteroidal anti-inflammatory drug (NSAID) administration on levels of tumor necrosis factor-alpha (TNF-alpha ) in rat bile. Diclofenac (10 mg/kg) and nitrofenac (15 mg/kg; equimolar) were administered orally at 12-h intervals, and bile was collected 3 h after a single dose or multiple (n = 4) doses. Results are expressed as percentage of control of vehicle-treated group (n = 4-7/group). star star P < 0.01 vs. vehicle-treated group.

Similar results were obtained when small intestinal tissue TNF-alpha levels were measured (Fig. 2). Multiple doses of diclofenac, but not nitrofenac, doubled TNF-alpha levels in the small intestine compared with the vehicle group. In addition, indomethacin (also known to produce extensive small intestinal injury at the dose used) significantly (P < 0.01) increased tissue levels of TNF-alpha 12 h after a single administration of 10 mg/kg sc (192.2 ± 23.9% of control levels).


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 2.   Effect of NSAIDs on small intestinal tissue levels of TNF-alpha . Diclofenac (Dicl, 10 mg/kg) and nitrofenac (Nitr, 15 mg/kg) were administered orally at 12-h intervals for a total of 4 doses, and ileal tissue samples were obtained 3 h after final dose. A separate group of animals received a single administration of indomethacin (Indo, 10 mg/kg sc) 12 h before tissue collection. Data are expressed as percentage of control of vehicle (Veh)-treated group (n = 5-11/group). star  P < 0.05; star star P < 0.01 vs. vehicle-treated group.

Small intestinal tissue TNF-alpha mRNA expression. Alterations in tissue TNF-alpha mRNA expression were determined using semiquantitative RT-PCR. TNF-alpha mRNA expression was normalized to GAPDH mRNA expression, and results were expressed as percent control. Figure 3 shows the TNF-alpha mRNA expression for small intestinal tissue taken from rats treated with a single dose or multiple doses of diclofenac. No changes were seen in mRNA expression in tissue obtained from rats receiving a single dose of diclofenac or from tissue that appeared macroscopically normal after multiple doses of diclofenac compared with the vehicle-treated group. Despite increased TNF-alpha levels in bile and tissue, no significant changes in mRNA expression were seen in damaged tissue taken from rats treated with multiple doses of diclofenac. As a positive control, a separate group of rats was given LPS (5 mg/kg ip), and small intestinal tissue was collected 1 h later. LPS markedly increased small intestinal tissue TNF-alpha mRNA expression over control levels (P < 0.001).


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3.   Effect of a single dose or multiple (n = 4) doses of diclofenac (10 mg/kg) on expression of TNF-alpha mRNA in small intestinal tissue. Ileal tissue was collected 2 h after final dose of diclofenac. A separate group of rats was treated with lipopolysaccharide (LPS, 5 mg/kg ip) as a positive control, and tissue was collected 1 h later. Results are represented as percentage of vehicle-treated group (n = 5-14/group). star star star P < 0.001 vs. vehicle-treated group.

Inhibition of TNF-alpha synthesis/release. Multiple oral doses of diclofenac consistently resulted in extensive small intestinal injury, with an average damage score of 277 ± 36 (Fig. 4A). Rats receiving only the vehicle for diclofenac did not develop any small intestinal damage. Pretreatment with the phosphodiesterase inhibitors pentoxifylline and theophylline significantly attenuated the damage produced by diclofenac (Fig. 4A). Pentoxifylline dose dependently inhibited diclofenac-induced small intestinal injury, with a significant reduction in damage scores at 100 and 200 mg/kg (P < 0.05). Rats pretreated with theophylline (50 mg/kg) also exhibited significantly less small intestinal damage than the vehicle-treated group (P < 0.01).


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4.   Effect of phosphodiesterase inhibitor pretreatment on diclofenac-induced small intestinal injury (A) and plasma levels of TNF-alpha (B) after LPS administration. Phosphodiesterase inhibitors pentoxifylline (50-200 mg/kg ip) and theophylline (Theo, 50 mg/kg ip) were administered 30 min before diclofenac (10 mg/kg po) or LPS (5 mg/kg ip). Values are means ± SE (n = 4-14/group). star  P < 0.05; star star P < 0.01 vs. vehicle-treated group.

To ensure that the doses of pentoxifylline and theophylline were sufficient to inhibit TNF-alpha synthesis in vivo, rats were treated with LPS (5 mg/kg ip) to induce a significant and consistent increase in plasma TNF-alpha levels. LPS administration elicited an increase in plasma levels of TNF-alpha to 838 ± 254 pg/ml from plasma levels of <10 pg/ml in untreated rats. Pentoxifylline at 200 mg/kg (225 ± 22 pg/ml, P < 0.05), but not at 50 mg/kg (583 ± 182 pg/ml), substantially attenuated (~70% inhibition) the increase in plasma TNF-alpha levels induced by LPS (Fig. 4B). Plasma TNF-alpha levels in the theophylline-pretreated group were 130 ± 25 pg/ml (P < 0.05), representing a reduction in TNF-alpha levels of >80%. Thus pentoxifylline and theophylline only reduced diclofenac-induced intestinal damage at the doses that also significantly inhibited TNF-alpha synthesis.

In addition to examining the effect of pentoxifylline on LPS-induced increases in plasma TNF-alpha , its ability to inhibit the increase in bile TNF-alpha levels after diclofenac administration was also determined (Fig. 5). Administration of diclofenac significantly increased bile TNF-alpha levels compared with controls (P < 0.05). Pretreatment with pentoxifylline at 200 mg/kg, but not at 100 mg/kg, significantly reduced the TNF-alpha levels (Fig. 5).


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5.   Effect of pentoxifylline (Pentox) pretreatment on TNF-alpha levels in bile after diclofenac administration. Diclofenac (10 mg/kg po) was administered at 12-h intervals for a total of 4 doses. Pentoxifylline (100-200 mg/kg ip) was given 30 min before each dose of diclofenac. Data are shown as percentage of control (vehicle/saline group) and expressed as means ± SE (n = 4-17/group). star  P < 0.05 vs. vehicle/saline group. delta  P < 0.05 vs. diclofenac/saline group.

Figure 6 depicts the effects of various doses of rolipram, a specific type IV phosphodiesterase inhibitor (34), on intestinal damage (A) and LPS-induced plasma levels of TNF-alpha (B). Rolipram significantly attenuated LPS-induced increases in plasma TNF-alpha at all doses tested (Fig. 6B; P < 0.05). Rolipram administered at 0.3, 1.0, and 3.0 mg/kg reduced plasma TNF-alpha levels to a similar extent (75-78%). Although all three doses of rolipram significantly inhibited TNF-alpha synthesis, only the higher two doses (1.0 and 3.0 mg/kg) protected the small intestine from diclofenac-induced damage (Fig. 6A).


View larger version (15K):
[in this window]
[in a new window]
  		Fig. 6.   Effect of pretreatment with type IV phosphodiesterase inhibitor rolipram on diclofenac-induced small intestinal injury (A) and plasma levels of TNF-alpha (B) after LPS administration. Rolipram (0.3-3.0 mg/kg ip) was administered 30 min before diclofenac (10 mg/kg po) or LPS (5 mg/kg ip). Values are means ± SE (n = 4-14/group). star  P < 0.05; star star P < 0.01 vs. vehicle-treated group.

To further establish whether TNF-alpha blockade was required to afford protection against diclofenac-induced intestinal damage, two other studies were performed. Thalidomide is a well-characterized inhibitor of TNF-alpha that is structurally unrelated to the phosphodiesterase inhibitors (31). Thalidomide, at all doses tested (10, 20, and 50 mg/kg ip), had no protective effect in the diclofenac-induced small intestinal injury model (Fig. 7A). However, this compound was able to significantly inhibit LPS-induced increases in plasma TNF-alpha at the higher two doses (Fig. 7B).


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 7.   Effect of thalidomide pretreatment on diclofenac-induced small intestinal injury (A) and plasma levels of TNF-alpha (B) after LPS administration. Thalidomide (10-50 mg/kg ip) was administered 30 min before diclofenac (10 mg/kg po) or LPS (5 mg/kg ip). Values are means ± SE (n = 4-10/group). star  P < 0.05 vs. vehicle-treated group.

Similar results were obtained using an anti-TNF-alpha antibody (Fig. 8). Rats pretreated with the antibody developed intestinal injury that was not significantly different, in terms of severity, from that produced in the group pretreated with NRS.


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 8.   Effect of pretreatment with a polyclonal rabbit anti-mouse TNF-alpha antibody (Ab) on diclofenac-induced small intestinal injury. Polyclonal antibody was given 4 h before each dose of diclofenac (10 mg/kg po). Small intestinal damage was scored 12 h after final dose of diclofenac. Values are means ± SE (n = 4-5/group). star  P < 0.05; star star P < 0.01 vs. normal rabbit serum (NRS)/vehicle group; ns, not significant.

Biliary excretion of diclofenac. The protection afforded by the phosphodiesterase inhibitors could have been attributable to decreased diclofenac absorption and/or biliary excretion. To test this hypothesis, the concentration of diclofenac in bile was determined in rats pretreated with saline or pentoxifylline (200 mg/kg). Pretreatment with pentoxifylline resulted in decreased absorption of diclofenac after a single dose of drug (Fig. 9A). The area under the curve in the pentoxifylline-pretreated group was approximately one-half that of the saline-treated group. The major difference in excretion profiles occurred during the first 3 h, with the biliary levels being significantly lower (P < 0.01) at the 3-h time point (Fig. 9A). Although a significant difference occurred between the two groups after a single dose of drug, this difference was not seen when bile was collected after administration of multiple doses of diclofenac (Fig. 9B). The two diclofenac excretion profiles were very similar, with no significant differences between the two groups at any of the time points examined.


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 9.   Effect of pentoxifylline (200 mg/kg ip) pretreatment on biliary levels of diclofenac after a single dose (A) or multiple doses (B) of NSAID. Saline or pentoxifylline was administered 30 min before each dose of diclofenac. Bile was collected at 1, 3, 6, and 12 h after a single dose or multiple (n = 4) doses of diclofenac. Values are means ± SE (n >=  4/time point). star star P < 0.01 vs. pentoxifylline-pretreated group.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Santucci et al. (32) and Appleyard et al. (2) established a causal link between TNF-alpha and NSAID-induced gastric damage in rats. Recently, TNF-alpha has been postulated to play a role in NSAID-induced small intestinal injury. Bertrand and colleagues (3, 4) demonstrated that small intestinal tissue levels of TNF-alpha increased in rats treated with indomethacin or flurbiprofen. In addition, no increase in small intestinal tissue levels of TNF-alpha was found in animals treated with NO-flurbiprofen, which was found not to produce small intestinal injury. These authors also reported that the intestinal damage produced by indomethacin could be significantly attenuated by pretreatment with the specific type IV phosphodiesterase inhibitor RO-20-1724. From the above evidence, Bertrand and colleagues concluded that TNF-alpha plays a critical role in NSAID-induced small intestinal injury.

The results of the present study confirm, to some extent, the findings of Bertrand et al. (3, 4). We also found that rats treated with indomethacin or diclofenac developed severe intestinal injury, which was associated with a significant increase in small intestinal tissue levels of TNF-alpha . Furthermore, administration of nitrofenac (NO-diclofenac) did not produce small intestinal injury, as previously shown (26), or an increase in intestinal tissue levels of TNF-alpha . Finally, we demonstrated a significant attenuation of diclofenac-induced small intestinal damage by pretreatment with the phosphodiesterase inhibitors pentoxifylline, theophylline, and rolipram. However, the intestine was not protected from NSAID-induced injury if the rats were pretreated with another inhibitor of TNF-alpha synthesis, thalidomide, or with an antibody directed against TNF-alpha . These results could not simply be explained by an ineffective dose of drug, since thalidomide at 20 and 50 mg/kg was able to inhibit TNF-alpha synthesis induced by the administration of LPS, whereas the antibody has previously been shown to effectively immunoneutralize TNF-alpha in the rat at the dose used (2, 9, 11). Furthermore, the lowest dose of rolipram (type IV phosphodiesterase inhibitor) was unable to attenuate diclofenac-induced small intestinal damage, although it inhibited TNF-alpha synthesis as effectively as the doses that did afford protection (Fig. 6). We previously demonstrated (10) that rats treated with an NO-releasing derivative of naproxen have increased plasma levels of TNF-alpha but did not develop gastric or intestinal injury. Taken together, these findings suggest that TNF-alpha does not play a critical role in the pathogenesis of NSAID-induced small intestinal injury.

Our results and those of Bertrand et al. (4) suggest that the increases in tissue levels of TNF-alpha occurred in parallel with the development of damage. In the present study we saw increased levels of TNF-alpha in bile and tissue only after four doses of diclofenac; that is, when extensive intestinal damage was evident. In the study of Bertrand et al. (4), indomethacin produced a significant increase in intestinal tissue levels of TNF-alpha 8 h after its administration, the same amount of time required for a significant increase in intestinal damage to be observed. A feasible explanation for the increase in TNF-alpha levels in intestinal tissue, and possibly bile, is that the cytokine is produced as a consequence of the damage but does not contribute to the initiation of the damage.

Phosphodiesterase inhibitors have many effects besides inhibition of TNF-alpha synthesis that might account for the beneficial effects in experimental NSAID enteropathy. Indeed, there is renewed clinical interest in phosphodiesterase inhibitors because of their ability to act as vasodilators, antithrombotics, and anti-inflammatory and immunosuppressive agents (30, 34). The vasodilatory properties of phosphodiesterase inhibitors may be of particular relevance in the context of NSAID enteropathy. There have been various accounts of microcirculatory changes within the intestine of rats treated with NSAIDs (1, 16, 23). NSAID administration has been associated with intestinal villus shortening, decreased blood flow, and subsequent ulceration (1, 23). It is possible that the phosphodiesterase inhibitors are able to prevent the early ischemic event induced by NSAID administration. Phosphodiesterase inhibitors are known to be potent vasodilators and of clinical benefit in the treatment of occlusive vascular diseases (30). In addition, pentoxifylline has been shown to preserve and restore intestinal microvascular blood flow associated with hemorrhagic shock and bacteremia (12, 33). It appears that this effect is accomplished via a cAMP-dependent pathway. Studies utilizing other agents (opiate antagonists or beta -agonists) that increase cAMP levels have also found protection against indomethacin-induced small intestinal injury (1, 35).

In conclusion, TNF-alpha does not appear to play a critical role in the pathogenesis of NSAID-induced small intestinal injury. Phosphodiesterase inhibitors, but not thalidomide or an anti-TNF-alpha antibody, were able to protect against diclofenac-induced small intestinal damage. The results of the present study also suggest that the increase in TNF-alpha levels (tissue and biliary) occurs as a consequence of the injury and the ensuing inflammatory reaction.


    ACKNOWLEDGEMENTS

This work was supported by a grant from the Medical Research Council of Canada (MRC). J. L. Wallace is an MRC Senior Scientist and an Alberta Heritage Foundation for Medical Research Senior Scientist. B. K. Reuter is supported by an MRC Studentship.


    FOOTNOTES

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. §1734 solely to indicate this fact.

Address for reprint requests and other correspondence: J. L. Wallace, Dept. of Pharmacology and Therapeutics, University of Calgary, 3330 Hospital Dr. NW, Calgary, AB, Canada T2N 4N1 (E-mail: wallacej{at}ucalgary.ca).

Received 5 May 1999; accepted in final form 14 July 1999.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Anthony, A., A. K. Bahl, I. G. Oakley, C. F. Spraggs, A. P. Dhillon, M. A. Trevethick, C. K. Piasecki, R. E. Pounder, and A. J. Wakefield. The beta -adrenoceptor agonist CL316243 prevents indomethacin-induced jejunal ulceration in the rat by reversing early villous shortening. J. Pathol. 179: 340-346, 1996[Medline].

2.   Appleyard, C. B., D. M. McCafferty, A. W. Tigley, M. G. Swain, and J. L. Wallace. Tumor necrosis factor mediation of NSAID-induced gastric damage: role of leukocyte adherence. Am. J. Physiol. 270 (Gastrointest. Liver Physiol. 33): G42-G48, 1996[Abstract/Free Full Text].

3.   Bertrand, V., R. Guimbaud, P. Sogni, A. Lamrani, C. Mauprivez, J. P. Giroud, D. Couturier, L. Chauvelot-Moachon, and S. Chaussade. Role of tumour necrosis factor-alpha and inducible nitric oxide synthase in the prevention of nitro-flurbiprofen small intestine toxicity. Eur. J. Pharmacol. 356: 245-253, 1998[Medline].

4.   Bertrand, V., R. Guimbaud, M. Tulliez, C. Mauprivez, P. Sogni, D. Couturier, J. P. Giroud, S. Chaussade, and L. Chauvelot-Moachon. Increase in tumor necrosis factor-alpha production linked to the toxicity of indomethacin for the rat small intestine. Br. J. Pharmacol. 124: 1385-1394, 1998[Medline].

5.   Bjarnason, I., J. Hayllar, A. J. MacPherson, and A. S. Russell. Side effects of nonsteroidal anti-inflammatory drugs on the small and large intestine in humans. Gastroenterology 104: 1832-1847, 1993[Medline].

6.   Brodie, D. A., P. G. Cook, B. J. Bauer, and G. E. Dagle. Indomethacin-induced intestinal lesions in the rat. Toxicol. Appl. Pharmacol. 17: 615-624, 1970[Medline].

7.   Chensue, S. W., D. G. Remick, C. Shmyr-Forsch, T. F. Beals, and S. L. Kunkel. Immunohistochemical demonstration of cytoplasmic and membrane-associated tumor necrosis factor in murine macrophages. Am. J. Pathol. 133: 564-572, 1988[Abstract].

8.   Chomczynski, P., and N. Sacchi. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162: 156-159, 1987[Medline].

9.   Cooper, A. L., S. Brouwer, A. V. Turnball, G. N. Luheshi, S. J. Hopkins, S. L. Kunkel, and N. J. Rothwell. Tumor necrosis factor-alpha and fever after peripheral inflammation in the rat. Am. J. Physiol. 267 (Regulatory Integrative Comp. Physiol. 36): R1431-R1436, 1994[Abstract/Free Full Text].

10.   Davies, N. M., A. G. Roseth, C. B. Appleyard, W. McKnight, P. del Soldato, A. Calignano, G. Cirino, and J. L. Wallace. NO-naproxen vs. naproxen: ulcerogenic, analgesic and anti-inflammatory effects. Aliment. Pharmacol. Ther. 11: 69-79, 1997[Medline].

11.   Ferraz, J.-G. P., A. W. Tigley, C. B. Appleyard, and J. L. Wallace. TNF-alpha contributes to the pathogenesis of ethanol-induced gastric damage in cirrhotic rats. Am. J. Physiol. 272 (Gastrointest. Liver Physiol. 35): G809-G814, 1997[Abstract/Free Full Text].

12.   Flynn, W. J., H. G. Cryer, and R. N. Garrison. Pentoxifylline restores intestinal microvascular blood flow during resuscitated hemorrhagic shock. Surgery 110: 350-356, 1991[Medline].

13.   Guimbaud, R., V. Bertrand, L. Chauvelot-Moachon, G. Quartier, N. Vidon, J. P. Giroud, D. Couturier, and S. Chaussade. Network of inflammatory cytokines and correlation with disease activity in ulcerative colitis. Am. J. Gastroenterol. 93: 2397-2404, 1998[Medline].

14.   Hamilton, L. J., Y. Dai, and G. D. F. Jackson. Bile regulates the expression of major histocompatibility complex class II molecules on rat intestinal epithelium. Gastroenterology 113: 1901-1905, 1997[Medline].

15.   Jackson, G. D. F., Y. Dai, C. Fung, and P. G. C. Hansen. Tumour necrosis factor-alpha ---a constitutive protein of bile. In: Advances in Mucosal Immunology, edited by J. Mestecky, C. Blair, and P. L. Ogra. New York: Plenum, 1995, p. 1083-1086.

16.   Kelly, D., C. Piasecki, A. Anthony, A. P. Dhillon, R. E. Pounder, and A. J. Wakefield. Reversal and protection against indomethacin-induced blood stasis and mucosal damage in the rat jejunum by beta 3-adrenoceptor agonist. Aliment. Pharmacol. Ther. 12: 1121-1129, 1998[Medline].

17.   Kent, T. H., R. M. Cardelli, and F. W. Stamler. Small intestinal ulcers and intestinal flora in rats given indomethacin. Am. J. Pathol. 54: 237-249, 1969[Medline].

18.   Kirisits, M. J., D. Vardimon, H. W. Kunz, and T. J. Gill III. Mapping of the TNFA locus in the rat. Immunogenetics 39: 59-60, 1994[Medline].

19.   Langman, M. J. S., L. Morgan, and A. Worrall. Use of anti-inflammatory drugs by patients admitted with small or large bowel perforations and haemorrhage. Br. Med. J. 290: 347-349, 1985.

20.   Lifrak, E. T., R. A. Erickson, L. Halbur, and E. Quimbo. Role of tumor necrosis factor-alpha (TNF-alpha ) in NSAID-induced intestinal ulceration (Abstract). Gastroenterology 108: A149, 1995.

21.   Melarange, R., C. Gentry, M. Durie, C. O'Connell, and P. R. Blower. Gastrointestinal irritancy, anti-inflammatory activity, and prostanoid inhibition in the rat: differentiation of effects between nabumetone and etodolac. Dig. Dis. Sci. 39: 601-608, 1994[Medline].

22.   Melarange, R., G. Moore, P. R. Blower, M. E. Coates, F. W. Ward, and V. Ronaasen. A comparison of indomethacin with ibuprofen on gastrointestinal mucosal integrity in conventional and germ-free rats. Aliment. Pharmacol. Ther. 6: 67-77, 1992[Medline].

23.   Miura, S., M. Suematsu, S. Tanaka, H. Nagata, S. Houzawa, M. Suzuki, I. Kurose, H. Serizawa, and M. Tsuchiya. Microcirculatory disturbance in indomethacin-induced intestinal ulcer. Am. J. Physiol. 261 (Gastrointest. Liver Physiol. 24): G213-G219, 1991[Abstract/Free Full Text].

24.   Morris, A. J., R. Madhok, R. D. Sturrock, H. A. Capell, and J. F. MacKenzie. Enteropscopic diagnosis of small bowel ulceration in patients receiving non-steroidal anti-inflammatory drugs. Lancet 337: 520, 1991[Medline].

25.   Morris, A. J., L. A. Wasson, and J. F. MacKenzie. Small bowel enteroscopy in undiagnosed gastrointestinal blood loss. Gut 33: 887-889, 1992[Abstract/Free Full Text].

26.   Reuter, B. K., N. M. Davies, and J. L. Wallace. Nonsteroidal anti-inflammatory drug enteropathy in rats: role of permeability, bacteria, and enterohepatic circulation. Gastroenterology 112: 109-117, 1997[Medline].

27.   Ribbons, K. A., J. H. Thompson, X. P. Liu, K. Pennline, D. A. Clark, and M. J. S. Miller. Anti-inflammatory properties of interleukin-10 administration in hapten-induced colitis. Eur. J. Pharmacol. 323: 245-254, 1997[Medline].

28.   Robert, A., and T. Asano. Resistance of germ-free rats to indomethacin-induced intestinal lesions. Prostaglandins 14: 333-341, 1977[Medline].

29.   Salcedo, C., J. Puig, J. M. Palacios, and A. G. Fernandez. Phosphodiesterase 4 inhibition, a new mechanism to maintain and promote gastric and intestinal integrity against NSAID-induced injury in the rat (Abstract). Gastroenterology 114: A274, 1998.

30.   Samlaska, C. P., and E. A. Winfield. Pentoxifylline. J. Am. Acad. Dermatol. 30: 603-621, 1994[Medline].

31.   Sampaio, E. P., E. N. Sarno, R. Galilly, Z. A. Cohn, and G. Kaplan. Thalidomide selectively inhibits tumour necrosis factor production by stimulated human monocytes. J. Exp. Med. 173: 699-703, 1991[Abstract/Free Full Text].

32.   Santucci, L., S. Fiorucci, M. Giansanti, P. M. Brunori, F. M. Di Matteo, and A. Morelli. Pentoxifylline prevents indomethacin-induced acute gastric mucosal damage in rats: role of tumor necrosis factor-alpha . Gut 35: 909-915, 1994[Abstract/Free Full Text].

33.   Steeb, G. D., M. A. Wilson, and R. N. Garrison. Pentoxifylline preserves small-intestine microvascular blood flow during bacteremia. Surgery 112: 756-763, 1992[Medline].

34.   Thompson, W. J. Cyclic nucleotide phosphodiesterases: pharmacology, biochemistry and function. Pharmacol. Ther. 51: 13-33, 1991[Medline].

35.   Waisman, Y., H. Marcus, M. Ligumski, and G. Dinari. Modulation by opiates of small intestinal prostaglandin E2 and 3',5'-cyclic adenosine monophosphate levels and of indomethacin-induced ulceration in the rat. Life Sci. 48: 2035-2042, 1991[Medline].

36.   Wallace, J. L., B. Reuter, C. Cicala, and G. W. McKnight. A diclofenac derivative without ulcerogenic properties. Eur. J. Pharmacol. 257: 249-255, 1994[Medline].

37.   Wax, J., W. A. Clinger, P. Varner, P. Bass, and C. V. Winder. Relationship of the enterohepatic cycle to ulcerogenesis in the rat small bowel with flufenamic acid. Gastroenterology 58: 772-780, 1970[Medline].

38.   Wong, H., W. D. Anderson, T. Cheng, and K. T. Riabowol. Monitoring mRNA expression by polymerase chain reation: the "primer-dropping" method. Anal. Biochem. 223: 251-258, 1994[Medline].

39.   Yamada, T., E. Deitch, R. D. Specian, M. A. Perry, R. B. Sartor, and M. B. Grisham. Mechanisms of acute and chronic intestinal inflammation induced by indomethacin. Inflammation 17: 641-662, 1993[Medline].

Am J Physiol Gastroint Liver Physiol 277(4):G847-G854
0002-9513/99 $5.00 Copyright © 1999 the American Physiological Society


This article has been cited by other articles:


Home page
 GutHome page
S J H van Deventer
Small therapeutic molecules for the treatment of inflammatory bowel disease
Gut, May 1, 2002; 50(90003): iii47 - 53.
[Abstract] [Full Text] [PDF]

Home page
 Toxicol SciHome page
R. A. Sneed, J. P. Buchweitz, P. A. Jean, and P. E. Ganey
Pentoxifylline Attenuates Bacterial Lipopolysaccharide-Induced Enhancement of Allyl Alcohol Hepatotoxicity
Toxicol. Sci., July 1, 2000; 56(1): 203 - 210.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
N. Diaz-Granados, K. Howe, J. Lu, and D. M. McKay
Dextran Sulfate Sodium-Induced Colonic Histopathology, but not Altered Epithelial Ion Transport, Is Reduced by Inhibition of Phosphodiesterase Activity
Am. J. Pathol., June 1, 2000; 156(6): 2169 - 2177.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Reuter, B. K.
Right arrow Articles by Wallace, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Reuter, B. K.
Right arrow Articles by Wallace, J. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online