- Email: [email protected]
Gastroprotective and ulcer healing effects of nitric oxide-releasing non-steroidal anti-inflammatory drugs
ALIMENTARYTRACT
DlGESTLlllER DIS 2000;32:593-94
Gastroprotective and ulcer healing effects of nitric oxide-releasing non-steroidal anti-inflammatory drugs T. Brzozows ki P.Ch. Konturekl S. J. Konturek Z. Sliwowski D. Drozdowicz S. KwiecieA R. Pajdo A. Ptak M. Pawlik E. Hahn’
hW#E Department of Fuysiology,Jegiell~t-tisn UniversitySchoolof Medicine, Crecow, Poland; l Repzzrtmentof Medicine 1, Uffiversiityof Er&ngen-IVurembe~, Erlangen,Germany. Thisst&y was supported by InterdisciplineryCenCerfar Clinical Inves@etion,Erlingen-NWemberg University,Erlehgen,Germany. , lwdrawhw Prof. S.J. Kontorek,Depament of Physiology, Jegieifonien University School of Medicine, 16 Gmegorzeeke
Str., 31-531 Cracow, P&end. FBX 0043-
42-212578.
Background 6; Aim. /Vew class of nitric oxide-releasing non-steroidal anti-inflammatory drugs was shown to inhibit cyclooxygenase and prostaglandin generation without causing mucosal damage but whether these agents are capable of affecting gastric mucosal damage induced by strong irritants and healing of chronic gastric ulcers remains to be studied. In this investigation, effects of nitric oxide-releasing aspirin and nitric oxide-releasing naproxen were compared with those of native agents on gastric lesions provoked by 100% ethanol and on healing of chronic acetic acid ulcers. Results. Both, nitric oxide-releasing aspirin and naproxen dosedapendently attenuated ethanol-induced damage and produced a significant rise in gastric blood flow but did not delay healing of gastric ulcers while native aspirin and naproxen had no influence on ethanol-induced gastric damage but significantly prolonged ulcer healing, reduced gastric blood flow and suppressed mucosal generation of prostaglandin Ez. The gastroprotective and hyperaemic effects of both nitric oxide-non-steroidal anti-inflammatory drugs were completely abolished by ODQ, an inhibitor of guanylyl Cyclase-cGMP system but not influenced by suppression of nitric oxidesynthase with L-NNA. The damaging effects of native acetyl salicylate acid or naproxen were aggravated by acidification of these non-steroidal anti-inflammatory drugs but the exogenous acid added to nitric oxide-acetyl salicylate acid or nitric oxide-naproxen failed to influence their effect. Despite inhibiting of PGEz generation, both nitric oxide-releasing derivatives and native aspirin and naproxen failed to affect expression of cyclooxygenase-I mRNA but upregulated the cyclooxygenase2 mRNA. Concurrent inhibition of cyclooxygenase-2 by selective inhibitor NS-398 which by itself delayed ulcer healing and attenuated the gastric blood flow at ulcer margin, significantly worsened the effects of these nitric oxidenon-steroidal anti-inflammatory drugs and their parent drugs on ulcer healing and the gastric blood flow at the ulcer margin. Conclusions. I] Coupling of nitric oxide to aspirin or naproxen attenuates ethanol-induced damage, possibly due to an increase in gastric microcirculation mediated by excessive release and action of nitric oxide that probably compensates for PG deficiency induced by non-steroidal anti-inflammatory drugs; and 21 nitric oxide-nonsteroidal anti-inflammatory drug, unlike classic non-steroidal anti-inflammatory drugs, does not affect intact gastric mucosa and fails to delay the healing of pre-existing ulcers.
E-me;/:[email protected] Submitted March 15, 2000. RevisedJuly 2U, 2oo0, AcceptedJuly 23 2iX70.
Digest
Liver
Dis 2000;32:583-94
Key words: aspirin; cyclooxygenase; gastric ide: nitric oxide synthase; ulcer healing
blood
flow;
gastroprotection;
naproxen;
nitric
ox-
583
NO-NSAID in gastroprotection
and ulcer healing
Introduction Non-steroidal anti-inflammatory drugs (NSAIDs) such as aspirin (ASA) or naproxen are widely used but the major limitation of their clinical application are serious side-effects, including damage of gastrointestinal mucosa, aggravation of stress ulcerations and exacerbation of pre-existing gastric ulcerations lms.Recently, a new class of NSAID has been developed by addition of nitric oxide (NO) moiety to the native NSAID 6-‘o. The rationale behind this strategy is that NO released from these derivatives exerts a beneficial effect on gastric mucosa by enhancing the mucosal defense ability and prevention of pathogenic events resulting from the suppression of prostanoid synthesis such as the reduction in mucosal microcirculation and the leukocyte-endothelial adherence “-‘3. Earlier studies revealed that endogenous NO released from vascular endothelium, sensory nerves or gastric epithelium co-operates with prostaglandins in the maintenance of gastric mucosa integrity and microcirculation I4 Is. In contrast to native NSAID, their NO-releasing derivatives such as NO-aspirin (NO-ASA) or NO-naproxen were found to exhibit lower gastric toxicity despite similar inhibition of both COX-1 and COX-2 (cyclooxygenase) activity in the gastric mucosa 5 ‘O12.The major importance of NO in the prevention of mucosal damage or in preservation of ulcer healing is supported by previous studies showing that both endogenous NO released by capsaicin Is or NO originating from L-arginine, a substrate for NO-synthase (NOS) 16, or that released from glyceryl trinitrate ” exert gastroprotective activity and accelerate healing mainly due to the maintenance of blood flow around the ulcer and enhancement of angiogenesis. The present study was designed; 1) to compare the effect of topical application of aspirin or naproxen and their NO-derivatives on gastric mucosal integrity and blood flow; 2) to assess the possible gastroprotective effects of NO-ASA and NO-naproxen, and for comparison of S-nitroso-penicillamine (SNAP), a donor of NO I’, added to native NSAID on acute gastric injury provoked by 100% ethanol; 3) to examine the influence of NO-ASA and NO-naproxen and their native compounds as well as SNAP, on the healing of chronic gastric ulcerations with or without the concurrent treatment with NS-398, a highly selective COX-2 inhibitor I” lx; and 4) to evaluate the expression of COX-1 and COX-2 in gastric mucosa with chronic gastric ulcers treated with native NSAID and their NO derivatives.
Material and Methods Male Wistar rats, weighing 180-220 g and fasted for 24 h before the study were employed in all investigations. 584
Rats were divided into series A for studies on production of acute gastric lesions and series B for studies with chronic gastric ulcers. Production of gastric lesions and measurement of gastric bloodflow (GBF) Gastric lesions were produced by an intragastric (ig) application of 1.5 ml of 100% ethanol by means of a metal orogastric tube as described in detail elsewhere Is 16. Native NSAIDs such as ASA, indomethacin and naproxen were obtained from Sigma Chemical Co. (St. Louis, MO, USA) while NO-aspirin (NCX 4016; 2(Acetyloxy)benzoic acid acid 3-(nitrooxymethyl) phenyl ester) and NO-naproxen (HCT 3012; (S)-6Metoxy-a-methyl-2-naphtalenacetic acid 4-(nitrooxy)butyl ester were kindly provided by Mrs. Nathalie Baudry from NicOx S.A., Sophia Antipolis, France. Native ASA (25-100 mg/kg) or naproxen (12.5 -50 mg/kg) and equimolar doses of NO-ASA (40-160 mg/kg) or NO-naproxen (18-72 mg/kg) were introduced ig in order to compare the damage produced by native NSAID and their NO derivatives and to examine the effect of pretreatment with NO-NSAID and their parent drugs applied ig 30 min prior to exposure of the gastric mucosa to 100% ethanol on the mucosal lesions induced by this corrosive agent. One hour after ig application of NSAID, their NO derivatives or ethanol, the animals were anaesthetised with ether, their stomachs were exposed and the GBF was measured in the oxyntic gland area by means of local Hr-gas clearance technique. To evaluate the effect of ASA, naproxen and their NO derivatives on GBF, the groups of animals were anaesthetised with ether and the abdomen was opened by midline incision. The stomach was exposed to assess the blood flow using the H2-gas clearance technique as described elsewhere I9*“. Briefly, the GBF was measured in the gastric mucosa not involved in the gastric lesions or at the ulcer base, ulcer margin and in the intact mucosa in rats with pre-existing ulcers using double electrodes of electrolytic regional blood flowmeter (Biotechnical Science, Model RBF-2, Osaka, Japan) inserted through the serosa into the mucosa. One for these electrodes was used for the local generation in the mucosa of H2 and another for measurement of tissue this Hz. With this method, the H2 generated locally is carried out by flow of blood, while the polarographic current detector records decreasing tissue HZ The clearance curve of tissue HZ was used to calculate an absolute blood flow rate (ml/min/lOO g) in the oxyntic gland area as described elsewhere l”. The measurement was made in three areas of the gastric oxyntic mucosa and the mean values of these measurements were calculated and expressed as percent changes from those recorded in vehicle-treated control animals. Using this technique, the results of the repeated blood
flow gave highly reproducible results and the inter- and intraassay variability of these measurements were 3.5% and 5.8%, respectively. The area of gastric lesions was measured by planimetry (Morphomat, Carl Zeiss, Berlin, Germany) by two investigators, under blinded conditions, as described elsewhere I*. Production
of gastric ulcers
Gastric ulcers were produced in 150 fasted rats of series B, using our modification 21 of the acetic acid method originally proposed by Okabe et al. **. With the animals under ether anaesthesia, the stomach was exposed and 75 yl of acetic acid was poured through the plastic mould (6 mm diameter) onto the serosal surface of the anterior wall of the stomach just proximal to the antral gland area for 25 s. This produced immediate necrosis of the entire mucosa and submucosa within the area where the acetic acid was applied, i.e., about 28 mm?. The excess of acetic acid was then removed and the serosa was gently washed with tap saline. Our previous studies documented that these ulcers become chronic within 2-3 days and heal completely within 2-3 weeks without perforation or penetration to the surrounding organs as described in the original technique of Okabe et al. ?2. After the application of acetic acid, the animals were allowed to recover from anaesthesia and received only water on the day of the operation (Day 0). Then, they were divided into various groups and received normal chow and water ad libitum for the next 3, 9 and I5 days. To evaluate the effects of native ASA or naproxen and their NO-releasing derivatives, the animals were anaesthetised with ether and the abdomen was opened and the stomach was exposed to assess the GBF at ulcer margin, ulcer crater and in the contralateral intact mucosa using the Hz-gas clearance technique as described above I7 *O.The stomach was then removed and pinned open for the determination of the area of gastric ulcers by planimetry. The half of the stomach with gastric ulcer in rats with or without administration of vehicle (control), ASA, naproxen or their NO-derivatives were taken during autopsy and immediately fixed in 10% formalin, embedded in paraffin and stained with haematoxylin and eosin for the histological assessment of the quality of ulcer healing. Coded specimens of mucosa stained with haematoxylin and eosin were evaluated at 260x magnification under blinded conditions. The other half of the stomach was either used immediately for PGE? generation or immediately frozen in liquid nitrogen and then stored at -80°C until use for mRNA expression of COX-I and COX-2 at the ulcer margin and contralateral intact mucosa. Experimental
groups and treatments
Series A consisting of 80 rats was used to determine the acute effects of topical administration of acidified ASA
or naproxen with those of NO-ASA or NO-naproxen against the mucosal lesions provoked by 100% ethanol. Series A included the following groups that received ig: 1) vehicle (saline or 0.2 N HCl); 2) ASA (25-100 mg/kg) or naproxen (12.5-50 mg/kg); 3) NO-ASA (40160 mg/kg) or NO-naproxen (18-72 mg/kg) followed 30 min later by ig application of 100% ethanol (1.5 ml). The time of duration of the protective effect of NONSAID was studied in animals pretreated with NOASA (80 mg/kg ig) or NO-naproxen (72 mg/kg ig) and then followed 30, 60, 120, 240 and 360 min later by ig application of 100% ethanol. In certain groups of rats treated with NO-NSAID, the effect of pretreatment with NG-nitro-L-arginine (L-NNA) to suppress the NO[4,3,synthase I5 or with lH-[1,2,4]oxidiazolo alquinoxalin-l-one (ODQ), which is a highly specific inhibitor of guanylyl cyclase 28, was determined in order to establish the involvement of the NO and guanylyl cyclase-cGMP system in the gastroprotection induced by these NO derivatives of NSAID. Series B rats with gastric ulcers were treated daily with ASA and naproxen (50 mg/kg ig) or with NO-ASA and NO-naproxen introduced ig in equimolar doses of 80 mg/kg and 72 mg/kg, respectively, without or with pretreatment with specific COX-2 inhibitor, N-(2-cyclohexyloxy-4-nitrophenyl)-methansulfonamide (NS398, Taisho Pharmaceutical Co., Tokyo, Japan) at a dose of 5 mg/kg ig. The animals were killed 1 h after ethanol administration or at day 3, 9 and 15 after ulcer induction and the area of gastric ulcerations was determined by planimetry, the GBF was measured by the H2-gas clearance method as described above and the gastric luminal nitrite/nitrate content was assessed by a specific kit based on Griess’ method. As already mentioned, the half of the stomach which was not used for histology, was employed for determination of the generation of PGE? by RIA and COX-1 and COX-2 mRNA by RT-PCR. Determination of PGE2 generation and luminal NO release
in gastric
mucosa
For ex vivo determination of PGE?, the mucosa was processed, according to the method described elsewhere 18. Briefly, a portion of the fundic mucosa (about 100 mg) excised from rats treated with vehicle, acidified ASA, naproxen and NO-ASA or NOnaproxen was placed immediately (after excision) in an Eppendorf plastic tube containing 1 .O ml of 10 mM phosphate buffer (pH 7.4) and minced with fine scissors for 15 s. The samples were then incubated in a shaking bath (37°C) for 20 min and then centrifuged 30 s at 9000 g. The supematant was collected and frozen at -20°C for subsequent determination of PGE? using radioimmunoassay (PGE?-RIA kit, Amersham, Munich, Germany). 585
NO-NW0
in gastmprotectinn
and ulcer healing
For the measurement of NO generation in the gastric lumen, 1 ml of saline was introduced into the stomach and the luminal content was collected to determine the total nitrite/nitrate concentration using a commercially available kit purchased from Cayman (Ann Arbour, MI, USA) as described in detail elsewhere *O. Histological evaluation of gastric mucosa The halves of the stomach treated with ig application of this NO-NSAID or their native agents, naproxen or acidified ASA without or with each of the tested irritants such as ethanol were used for histology. For this purpose, the mucosal specimens were fixed in formalin, then embedded in paraffin and finally stained with haematoxylin and eosin (H & E). The quantitative histology was performed using a Nikon microscope equipped with Microplan II. The disrupted surface of the mucosal strips denuded of epithelium and the deeper necrotic lesions penetrating the mucosa were measured and expressed as a percent of total strip length. Reverse-transcriptase polymerase chain reaction for detection of messenger RNA (mRNA) for COX-1 and cox-2 Gastric mucosal samples were collected (at -8OC) for determination of expression of COX- 1 and COX-2 mRNA by reverse-transcriptase polymerase chain reaction (RT-PCR). Total RNA was isolated from the gastric mucosal samples by guanidinum isothiocyanate/phenol chloroform using single step extraction kit from Stratagene according to the original procedure described by Chomczynski et al. 23, as in our previous studies 24. Single-stranded cDNA was generated from 5 ug of cellular RNA using StrataScriptTM reverse transcriptase (Stratagene, La Jolla, USA) and oligo (dT) primers. The resulting complementary DNA (cDNA) was used as a template for subsequent PCR. After the reverse transcription, the transcriptase activity was destroyed by heating, and the cDNA were stored at -20°C until PCR. The primer sequences were designed according to the published cDNA sequence for the rat cyclooxygenases 25-27.The COX-1 primer sequences were as follows: up-stream, 5’-AGC CCC TCA TTC ACC CAT CAT TT; downstream, 5’-CAG GGA CGC CTG TTC TAC GG. The expected length of this PCR product was 561 bp. The COX-2 primer sequences were as follows: upstream, 5’-ACA ACA TTC CTT CCT TC; downstream, 5’-CCT TAT TTC CTT TCA CAC C. The expected length of this PCR product was 201 bp. Concomitantly, amplification of control rat p-actin (Clon Tech, Palo Alto, CA, USA) was performed on the same samples to ascertain RNA integrity. DNA amplification was carried out under the following conditions; denaturation at 94°C for 1 min, anneal588
ing at 60°C for 45 s, and extension at 72°C for 45 s. To maximise amplification specificity, Taq DNA polymerase was added to the PCR mixture during the hot start of cycle 1. The amplification of PCR was performed at 30 cycles. Each PCR-product (8 ,ul) was electrophoresed on 1.5% agarose gel stained with ethidium bromide, and then visualised under UV light. Location of predicted PCR product was confirmed by using a loo-base pair ladder (Gibco BRL/Life Technologies, Eggenstein, Germany) as standard marker. To avoid PCR contamination, PCR reactions were prepared in a dedicated area used only for PCR and the PCR products were opened in a laminar flow hood separated from the PCR preparation area. The intensity of bands was quantified using semiquantitative densitometry analysis as described in detail elsewhere l9 24. The COX-1 and COX-2 signals were standardised against the p-actin signal for each sample and results expressed as COX-1 and COX-2/pactin mRNA ratio. Amplification was performed using the DNA thermal cycler (Perkin-Elmer-Cetus, Norwalk, USA) and the cycles were established in each separate PCR reaction. The PCR product was electrophoresed on 1.5% agarose gel stained with ethidium bromide, and then visualised under ultraviolet (UV). Expression of the product was quantified using video image analysis system (LKB Ultrascan, Pharmacia, Sweden) as described in detail elsewhere l9 24. Statistical analysis Results are expressed as means f SEM. For statistical analysis, the nonparametric Mann-Whitney U and Kruskal-Wallis tests for comparisons were applied where appropriate, with values of less than 0.05 (~~0.05) taken as statistically significant. Results
EfSect of NO-ASA and NO-naproxen of gastric lesions induced by ethanol and alteration in GBF: PGEI generation and luminal nitrite/nitrate content Figure 1 shows the area of gastric lesions and the alterations in GBF and luminal nitrite/nitrate concentration in rats exposed to 100% ethanol without or with the pretreatment with vehicle (saline) or NO-ASA introduced ig in graded doses ranging from 40 mg/kg up to 160 mg/kg. Mean GBF in the intact gastric mucosa was 52&7 ml/mm/100 g (taken as lOO%), while the mean gastric nitrite/nitrate content was 3.1kO.5 umol/l and these values were not significantly affected following ig application of vehicle (saline). Administration of 100% ethanol to vehicle pretreated rats reduced the blood flow to about 55% and the gastric nitrite/nitrate value
T. Brzozowski et al.
was similar to that recorded in intact mucosa, while causing the appearance of gastric lesions of about 7826 mm* (Fig. 1). Pretreatment with NO-ASA applied topically, significantly reduced the area of lesions induced by 100% ethanol. With graded concentrations of NOASA administered before 100% ethanol, the area of gastric lesions was significantly attenuated and a significant increase in the GBF and nitrite/nitrate content starting with a dose of 80 mg/kg of NO-ASA was observed and the dose reducing, by 50%, the ulcer area IDso averaged about 95 mg/kg. The protective effect of NO-ASA was accompanied by a significant rise in GBF and luminal nitrite/nitrate content (Fig. 1). NOnaproxen given ig also reduced dose-dependently the area of gastric lesions caused by 100% ethanol with the threshold diminution occurring at a dose of 36 mgkg and with the IDso averaging 63 mg/kg of NO-naproxen (Fig. 2). Pretreatment with NO-naproxen resulted in a significant rise in the GBF and luminal nitrite/nitrate contents. In contrast, acidified ASA or naproxen failed to influence the gastric lesions induced by 100% ethanol and accompanying changes in GBF and luminal nitrite/nitrate levels (Figs. 1, 2). The generation of PGE2 in the gastric mucosa averaged 156+ 18 rig/g of wet tissue weight and this was not significantly influenced by exposure of gastric mucosa to 100% ethanol (Table I). Native ASA or naproxen introduced ig in a standard concentration of 50 mg/kg before 100% ethanol significantly inhibited the generation of PGE2 in the gastric mucosa as compared to that measured in rats with ethanol alone. Mucosal content of PGE2 was also suppressed in rats pretreated with NO-ASA or NO-naproxen despite the fact that NO-ASA- or NO-naproxen dose-dependently attenuated the area of gastric lesions induced by 100% ethanol (Table I, Figs. 1, 2). Topical application of 100% ethanol resulted in a significant increase in superficial and deep necrosis and this was not significantly affected by pretreatment with native ASA or naproxen. In la!& I. Generation of Pi& in gastric mucosa end area of superficial and deep histalogical damage in intact rats and those pretreated with vehicle Net?), AS4 and nepmxen @I m&kg Lg.1 or NO-ASA f60 mglkg Lg.1 and NO-naproxen (72 m@kg Lg.1 before exposure to ‘KiC% ethanol. Histology results are expressed as peme-nw of damaged area in mucusal strip length. Results ara mean k !3EM of 6 - 6 rats.
Intact Veh AS A Napmxen NO-ASA NO-napmxen
156*16 145zkl1 3%4” 58~t7” 47r6 56r5
0.71to.03 43k3.2 46ct4.2 421t2.8* 21*2.2’+ 24k7.8”’
0 21*0.8 22~0.6 20a0.4 5*0.6”+ 7*0.4*+
’ Signihnt dwge as canpsred to value oL%%ined in intact re& +signh%xnCdwge 8s wmpai-ed M due obtained in IW& pretreate with native ASA or nepmxen. Abbrevi&ms: see list. 587
NO-NSRID in gastroprotection
and ulcer healing
contrast, the pretreatment with NO-ASA or NOnaproxen significantly attenuated the area of superficial and deep necrosis provoked by ethanol as compared to that measured in rats pretreated with native ASA or naproxen. Figure 3 shows the duration time of the protective effects of NO-NSAID against ethanol damage in comparison with the effect of their parent NSAID. This protection lasted up to 4 hours upon NO-ASA (80 mg/kg ig) or NO-naproxen (72 mg/kg ig) administration while native ASA or native naproxen had no influence on the ethanol-induced lesions at any time.
[email protected]. Area of ethanol-induced gastric lesions and gastric blood flow fG8F1 in rats treated with vehicle Winal, NO-ASA f8Clmgkg igl or NO-naproxen f72 mg&g igl without or with the pretreatment with LNNA (20 m$kg ivl. Means f SEM of 6-8 rats. ’ Significant change as compared to those obtained in vehicle-treated gestric mucosa with or without L-NNA pretreatment. + Significant change as compared to value obtained in vehicle-treated gastric mucosa without pretreatment with L-NNA,
so-...
I
Fig. 3. Time duration of protective effect of vehicle fsalinel, NO-ASA I80 mgkg igl or NO-napmxen (72 mgkg igl in comparison with native ASA or native napmxen f50 mgkg igl against gastric lesions induced by 100% ethanol. Means f SEM of 6-8 rats. * Significant change as compared to those obtained in vehicletreated gastric mucosa
Effect of suppression of NO-synthase and inhibition of guanylyl cyclase-cGMP system with ODQ on the gastroprotection exhibited by NO-ASA and NO-naproxen Figure 4 shows the effect of NO-ASA and NO-naproxen introduced ig in a standard dose of 80 mg/kg and 72 mg/kg, respectively, on the area of gastric lesions induced by 100% ethanol and accompanying changes in the GBF in rats without and with pretreatment with LNNA, an inhibitor of NOS. L-NNA by itself applied at a dose of 20 mg/kg iv failed to affect the ethanol damage but produced a significant drop in the GBF as compared to that in vehicle pretreated rats given 100% ethanol. NO-ASA and NO-naproxen significantly attenuated the area of acute gastric lesions or the drop in GBF caused by 100% ethanol and these effects were not influenced by pretreatment with L-NNA. In contrast, pretreatment with ODQ, which is a highly specific inhibitor of guanylyl cyclase 28, completely re588
Fi3.5. Area of ethanol-induced gastric lesions and gastric blood flow [GBFI in rats treeted with vahide faatinal, NO-ASA f80 n-q/kg igl or with NO-naproxen f72 mgkg igl without or with pratreatment with OOQ (10 mg/kg i.g.1. Means f S&l of 6-6 rats. *Significant change as compared to those obtained in v&&la-treatad gastric rnucosa. ‘significant ohange as comparad to value obtained in rats without OOO pretreatment.
versed the protection and the rise in the GBF evoked by NO-ASA and NO-naproxen (Fig. 5). Effect of addition of exogenous acid to native and NO-releasing NSAID on gastric damage and alteration in GBF As shown in Figure 6, native ASA or native naproxen produced gastric lesions and reduced GBF as compared
T. Brzomwski
AeA NmoxaN +SNM +SNAP
et al.
MA NAROXEN
Fi& 0. Mean araa of gastric lesions and changes in GBF induced by native ASA and native naproxen Iboth 50 mgkg ig1 or NO-ASA [80 m&!&g ig) and NG-naproxen I72 m@kg igl without or with addition of exogenous HM. Mean k SEM of 68 rats. *Significant change as compared to vslue obteinad in vehicle-control animals. + A significant changa as compared to valueobtained in rats without acidification.
Fig. 7. Mean area of gastric lesions and changes in G8F induced by native ASA or native naproxen or their NO-derivatives NO-ASA and NO-naproxen without or with addition of S-nitroso-penicillamine CiNAP, 5 mgkg igl , Mean + SEM of 6-8 rats. * Significant change as compared to the value obtained in vehicle-control animals or those pretreated with native ASA or naproxan.
to those measured in the vehicle-control gastric mucosa without luminal acidification. In contrast, NO-ASA and NO-naproxen failed to induce gastric lesions and this was accompanied by a significant rise in the GBF as compared to that obtained in ASA- or naproxentreated mucosa. The addition of 1 ml of 0.2 N HCl to the stomach by itself failed to produce gastric lesions but reduced the GBF, to about 90% of the intact value. It resulted in a remarkable worsening of gastric lesions induced by native NSAID and greater drop in the GBF compared to those recorded with these agents in gastric mucosa without addition of exogenous acid. Addition of exogenous acid to NO-ASA or NO-naproxen failed to produce gastric lesions and did not induce any greater alteration in the GBF than that observed in the gastric mucosa treated with NO-ASA or NO-naproxen without the combination with exogenous acid.
Effect of vehicle or native ASA and naproxen and their NO-releasing derivatives or SNAP with or without NS398 on the area of gastric ulcers and gastric blood flow (GBF) at ulcer margin Figure 8 shows the time-course of ulcer healing in rats following ig administration of vehicle, native ASA and naproxen applied daily in a dose of 50 mg/kg and NOASA, NO-naproxen and SNAP applied daily in a single
Effect of S-nitroso-penicillamine (SNAP) alone or combined with ASA and naproxen on gastric lesions and the GBF As shown in Figure 7, pretreatment of the gastric mucosa with SNAP (5 mg/kg ig), a potent donor of NO, significantly reduced the area of ethanol lesions and this was accompanied by a significant rise in the GBF. In contrast, native ASA or naproxen failed to affect ethanol lesions and the accompanying drop in the GBF induced by this irritant. Addition of SNAP to native ASA or NOnaproxen afforded the protection and the rise in the GBF to an extent similar to that observed after the administration of NO-ASA or NO-naproxen introduced ig at a dose of 80 mg/kg and 72 mg/kg, respectively.
NOPROXE
O
J
-+
i
3
9
II
DAVS AFTER ULCRR INDllCllON
Fig. 8. Area of chronic gestrii ulcers in rats treated by vehicle, native ASA and native naproxan or their NO-derivatives NO-ASA and NOnaproxen and SNAP, a donor of NO, during period of 3,9 end 15 days upon ulcer induction. fvlaan f SEM of 6-8 rats. “significant change as compared to value obtained at day 0. + Significant change as compared to value obtained in vehicle-control gastric mucosa at day 9 and day 15 upon ulcer induction.
588
NO-NSAID in gaatroprotection
and ulcer healing
dose of 80 mg/kg, 72 mg/kg and 5 mg/kg, respectively. The alterations in these ulcers are expressed as a change in the area of these ulcers as determined at day 0, 3, 9 and 15 upon their induction. In rats treated with vehicle, a significant reduction in the area of these ulcers was observed from an initial size of about 28 mm2 to 16.2k3.1 mm* and 10.9k2.5 mm2 at day 9 and 15, respectively. Treatment with SNAP resulted in a significant acceleration of the ulcer healing at day 9 and day 15 as compared to the respective values recorded in rats treated with vehicle. Native ASA or naproxen failed to significantly affect the area of gastric ulcers at day 3 upon ulcer induction but significantly delayed ulcer healing at day 9 and day 15, while NO-AS A or NO-naproxen did not interfere with the ulcer healing as compared to that measured in vehicle-control. The GBF in the non-ulcerated antro-oxyntic mucosa of rats treated with vehicle averaged 48 ml/min- 100 g (taken as 100%) being significantly reduced (by about 15%) at the ulcer margin at day 15 when compared to that in the contralateral intact mucosa (Fig. 9). Native ASA or naproxen significantly delayed ulcer healing at day 15 and this was accompanied by the significant decrease in the GBF at the ulcer margin as compared to the respective values recorded in vehicle-treated gastric mucosa. In contrast, SNAP significantly accelerated the healing of gastric ulcers at day 15 and increased the GBF at the ulcer margin to a significantly greater extent than that recorded in intact gastric mucosa. Treatment with NO-ASA and NO-naproxen accelerated ulcer healing in a similar fashion to vehicle and this was accompanied by a rise in the GBF at ulcer margin to a significantly higher extent than that in gastric mucosa treated with native NSAID and similar to that observed in vehicle-treated rats. Co-administration of NS-398, a highly selective COX2 inhibitor with vehicle, ASA and naproxen, or NOASA and NO-naproxen, produced a significant increase in the area of gastric ulcers as compared with the respective values obtained in rats treated with vehicle or NO-NSAID without addition of COX-2 inhibitor (Fig. 9). Only the acceleration of ulcer healing and accompanying hyperaemia at the ulcer margin induced by SNAP were significantly attenuated in rats with the concurrent treatment with NS-398. The increase in the area of gastric ulcers induced by native ASA and naproxen was significantly enhanced by the addition of COX-2 inhibitor to these NSAIDs. Likewise, the concurrent treatment with NS-398 significantly delayed the ulcer healing and decreased the GBF at the ulcer margin as compared to those treated with NO-ASA and NO-naproxen alone (Fig. 9). PGE2 generation in intact non-ulcerated gastric mucosa of the vehicle-treated rats at day 15 after ulcer induction averaged 9427 uglg wet tissue weight and this was sig590
nificantly increased in the ulcerated gastric mucosa excised from the margin of the gastric ulcer (Fig. 10). ASA and naproxen or NO-ASA and NO-naproxen inhibited mucosal PGE2 generation in the intact non-ulcerated gastric mucosa, the extent being similar to that
ULChDAIrD MUCOSA
Fig. 18. Generation of PGE2 in ulcerated [ulcer margin1 and non-ulcerated Wactl gastric mucosa of rats treated 15 days with native ASA, native naproxen, NO&A, NO-napruxen and SNAR a donor of NO. Mean & SEM of S-8 rate. * Significant change es compared to value obtained in vehicle-control k&act1 gaetric mucosa. + Significant change as compared to va&s obtained in the non-ulcerated gastric mucosa. ++Significent change as compared to vehicle-Weated uloereted mucosa.
1. Inozowski
measured in ulcerated gastric mucosa. In contrast, SNAP significantly raised the generation of PGE2 in the non-ulcerated and ulcerated gastric mucosa as compared to those obtained in vehicle-treated mucosa. RT-PCR expression of COX-1 and COX-2 mRNA in gastric mucosa of rats treated with native ASA or naproxen and their NO-derivatives The expression of COX-1 mRNA was detected as a strong signal in gastric mucosa treated with vehicle, whereas COX-2 mRNA was undetectable in this mucosa (Fig. 11 B,C). The intense signals for COX-1 mRNA was preserved in gastric mucosa exposed to native ASA and naproxen as well as in those treated with NOASA or NO-naproxen (Fig. 11 B). In contrast, COX-2 mRNA was strongly upregulated in the gastric mucosa at the ulcer margin and the mucosa whether treated with native ASA and naproxen or as well as with NOASA and NO-naproxen (Fig. 11 C). The mRNA expression for p-actin was well preserved in each sample tested (Fig. 11 A). The ratio of COX-1 mRNA over p-
et al.
actin mRNA was not significantly different in rats treated with native ASA and naproxen or their NO-derivatives (Fig. 12). In contrast, the ratio of COX-2 mRNA over p-actin showed a significant rise already at the ulcer margin and in the gastric mucosa of rat treated with ASA and naproxen or NO-ASA and NO-naproxen (Fig. 13). The ratio of COX-2 over p-actin was significantly increased in rats treated with native ASA and its NO-derivative as compared to those treated with vehicle, native naproxen and NO-naproxen.
e
1.01
FQ.W. Assessmentof ratio C;OX-1mRNA&eetin mRNAexpression aspreeetlWinFig. l1,MeanitMQfi%?rata
Fi& H. ti&ssm RNAexpreesiwtfor @-actin,CCJX-1end COX-E tnRNAat ay 1k upon ulcer induatt~ in rc+@with in it%aotgastric mocasaMW 1) and uk%rsti mucpse fuioer mclrginltreatad wit1 whir&t kse 2J, rte&e A8A &me 31, netim neprvxsnkne 4J, NO ASA km 51 ad ~n~~x~ Itme 6). I% is rtegstivacontmi &a* terl. M-eke msrkerDhlA,AITQW - expectedPCRproduct (bp1.
F** 1% Assessmentof r&i2 COX-2rn~~otin,rn~ expmeeion as preeent&iin Fig.11. Mesr~$ SE%!of 6+8r&s;.“$i&@fi@sot oheege es compersdto v&usoh&&ad in iotai gsst& We5 +SignMsnt changeee compsr& to Joe o&a&d in rata trek&d with vehide, netiveneproxenand NO-n~xen
NO-NSAID in gastropratection
and ulcer healing
Discussion Our study demonstrates that the new class NSAID, namely NO-ASA and NO-naproxen, by themselves do not produce macroscopic mucosal damage despite the PGEz suppression similar to that observed with their parent ASA and naproxen. In clear contrast, topical application of ASA or naproxen, particularly when acidified, results in gastric lesions and this effect is accompanied by a significant drop in GBF and marked suppression of PGEz generation. Both NO-NSAIDs dosedependently attenuated lesions induced by 100% ethanol and this protective effect of NO-NSAIDs was accompanied by a rise in GBF and a significant increase in nitrite/nitrate content in the gastric lumen. Cotreatment with SNAP, a potent source of NO, added to native ASA or naproxen, also affords the protection, hyperaemia and rise in the luminal NO release similar to those observed with NO-NSAID. Suppression of the specific NO-sensitive guanylyl cyclase-cGMP pathway by the pretreatment with ODQ, dramatically reversed the protection and the rise in the GBF induced by NO-releasing of NSAIDs suggesting that NO released from these NO-derivatives of NSAIDs contributes to their beneficial effects in the gastric mucosa involving intracellular guanylyl cyclase and the rise in cGMP level. In the chronic ulcer studies, treatment with ASA or naproxen administered ig for 15 days caused an expected delay in ulcer healing and decreased both GBF at ulcer margin and PGE2 synthesis in the gastric mucosa. In contrast, NO-ASA and NO-naproxen did not delay the healing rate of chronic gastric ulcers and GBF at ulcer margin as compared to those in the vehicle-treated controls. Treatment with NS-398, a highly selective COX-2 inhibitor, which by itself prolonged ulcer healing and decreased GBF at the ulcer margin, reversed the favourable effects of NO-NSAIDs and increased the deleterious effects of native ASA and native naproxen on ulcer healing and GBF at the ulcer margin. Paradoxically, COX-2 specific inhibitors that are widely proposed as an attractive therapeutic development in the treatment of rheumatoid arthritic and osteoarthritis, partly because they were shown to spare the COX-1 isoform responsible for gastrointestinal protection and failed to cause peptic ulcers or bleeding 29, appear, in our hands, to exert much less favourable effects on healing of pre-existing ulcers than simple NO-NSAID, blocking both COX-1 and COX-2. Further studies, particularly clinical trials with novel NO-NSAID which are in progress, should provide an answer as to whether these relatively simple agents are more beneficial to the gastrointestinal tract than the so-called “Coxibs” recently designed for this purpose. Furthermore, we found that COX- 1 mRNA was detected in the 592
intact gastric mucosa at the ulcer margin and in the mucosa treated with NO-ASA, NO-naproxen and their parent NSAID. In contrast the COX-2 mRNA was not detected in rat mucosa treated with vehicle (control) but was upregulated already at the ulcer margin in rats treated for 15 days with vehicle and both, native ASA and naproxen or their NO-derivatives. This suggests that the enhanced mucosal expression of COX-2 at the ulcer margin contributes to ulcer healing and that NO released from gastric-sparing NO-NSAIDs derivatives can compensate with NO for PG deficiency induced by NSAID. As shown in previous studies, NSAID exhibit potent anti-inflammatory and analgesic actions but the use of these drugs, in humans, is limited due to untoward effects such as formation of gastric mucosal bleeding erosions, enhancement of the ulcerogenic response to various stimuli and impairment of healing of pre-existing ulcers l 2 I2 30. This deleterious action of conventional NSAIDs was attributed to their topical irritating effect, activation of neutrophils, fall in the microcirculation, enhancement of motility induced by these agents and the reduction in mucosal generation of PGEz 4 I2 30. In contrast, newly developed NO-releasing NSAID, constructed by adding an nitroxy-butyl moiety to aspirin or naproxen, were reported to attenuate side-effects of native NSAID including gastrointestinal toxicity while exerting anti-thrombotic effects comparable to their parent NSAID 5 6 lo. Moreover, it was shown that NOreleasing NSAID by themselves exhibit only minimal ulcerogenic properties in the gastrointestinal tract, despite exerting a potent anti-inflammatory and analgesic action, similar to native NSAIDs 9 lo. The NSAIDs examined in our present study, such as aspirin and naproxen, were ulcerogenic by themselves in the stomach and their damaging action was strongly enhanced by addition of the exogenous acid that was employed to mimic the fate of both NSAIDs such as those occurring in highly acidic conditions in the stomach. Moreover, both non-specific COX-inhibitors inhibited PGE;! generation confirming previous observations that suppression of COX and subsequent deficiency of endogenous PG in the gastric mucosa could account, at least in part, for their damaging effect. In contrast, NO-releasing ASA and NO-releasing naproxen by themselves were devoid of ulcerogenic properties and failed to enhance gastric damage after acidification of these drugs, despite the sustained ability of these agents to suppress PGE:! generation similar to that observed after application of their parent drugs. Furthermore, pretreatment with NO-ASA and NO-naproxen resulted in the attenuation of ethanol-induced damage and significantly raised the GBF and these effects were counteracted by ODQ, an inhibitor of NO-dependent guanylyl cyclase 28 but remained unaffected by inhibition of NO-synthase
by L-NNA. The mechanism of protection and accompanying hyperaemia induced by NO-releasing NSAIDs is not yet clear but could be due to excessive local NO released from NO-derivatives of NSAIDs. This is supported by the fact that a considerable amount of NO metabolites were detected in the luminal content of the stomach of rats pretreated with NO-releasing NSAID. Furthermore, SNAP, a potent NO donor, which by itself significantly reduced the gastric damage provoked by ethanol, added to native ASA or naproxen resulted in protection and hyperaemia comparable to those exhibited by NO-releasing NSAIDs suggesting that, indeed, NO released from these compounds plays a crucial role in this protection and accompanying hyperaemia. Since gastrointestinal ulcerations are associated with the use of all NSAIDs, the new strategy for treatment of inflammatory states includes novel series of NSAID that consist of an NSAID-linked either to a NO-releasing moiety 3’-33or to phospholipids such as zwitterronic acid 34. The rationale behind the development of NONSAIDs that spare the gastric mucosa was that NO released from these compounds would counteract two events that occur subsequent to the suppression of PG synthesis by the NSAID, namely the reduced GBF and an increased adherence of neutrophils to the vascular endothelium of the gastric microcirculation I1 l7 3336. Both these events have been suggested as critical in the pathogenesis of experimental NSAID-gastropathy 3o3536. Recently, Fiorucci et al. 3’ have shown that administration of aspirin resulted in enhancement of apoptosis rate via upregulation of the caspase system mediated by tumour necrosis factor-alpha (TNFa) and these effects were counteracted by NO-derivatives of NSAID. It was suggested that NO-NSAID such as NO-ASA can spare the gastric mucosa and inhibit caspase activity, at least in part, through the cGMP-dependent pathway 31 including the release of NO from these NSAIDs. We documented that both native aspirin and naproxen significantly impaired the healing of chronic gastric ulcers and attenuated both the GBF at the ulcer margin and the PGE2 generation in the intact as well as ulcerated gastric mucosa. This inhibition of COX-1 and COX-2 activity could contribute to the prolongation of ulcer healing by NSAID, since previous studies have shown that indomethacin, a non-selective COX inhibitor and NS398 or rofecoxib, both highly specific inhibitors of COX-2, delayed healing of pre-existing chronic gastric ulcers and ischaemia-reperfusion-induced acute erosions progressing into chronic ulcers 24.The importance of COX-2 for ulcer healing was recently emphasised by the observation that expression of COX-2 mRNA is increased after induction of chronic gastric ulcers suggesting that COX-2 expression and PG-derived from COX-2 may be crucial in the mechanism of ulcer healing 2436. Furthermore, it was demonstrated that aspirin
upregulated COX-2 mRNA, possibly due to the deficiency of endogenous PG induced by this NSAID 37. Our results are in keeping with these findings since we showed COX-2 mRNA to be upregulated at the ulcer margin in vehicle control rats and in those treated throughout a period of 15 days with aspirin and naproxen while under the same experimental conditions, mRNA for COX-1 expression remained at the same level. It is, therefore, rational to assume that PGs derived from COX-2 located at the ulcer margin may be crucial in the healing response of gastric ulcer. This hypothesis is supported by the fact that treatment with NS-398 increased the delay in ulcer healing caused by classic NSAID and significantly impaired that exhibited by NO-releasing NSAID and NO donor such as SNAP. This suggests that COX-2 derivatives play an important role in the effect of NO-NSAID on ulcer healing and that NO derived, for instance, from NO-NSAID activates the COX-2 pathway as suggested by Salvemini et al. 38. Interestingly, treatment with NO-ASA and NOnaproxen failed to significantly delay the ulcer healing despite the fact that these NO-derivatives of NSAID suppressed COX-activity in the non-ulcerated and ulcerated gastric mucosa as effectively as their parent NSAID. Thus, it is possible that NO released from NOreleasing NSAID may counteract the potential mucosal impairment resulting from the PG deficiency due to the COX inhibition. This hypothesis is supported by our previous observations that L-arginine, a substrate for NO-synthase or NO donors such as glyceryl trinitrate, accelerate ulcer healing by enhancement in the microcirculation at the ulcer margin and angiogenesis I6 “. On the other hand, administration of NO-synthase inhibitors such as L-NNA or L-NAME delayed healing of acute gastric injury and prolonged the healing of chronic gastric ulcers I’. In summary, we found that NO-releasing aspirin and naproxen, unlike their native substances, while inhibiting PGE2 generation in the gastric mucosa, are not ulcerogenic by themselves, fail to impair the healing of pre-existing ulcers and rather exhibited gastroprotective activity against corrosive substance such as ethanol suggesting that NO released from these compounds has a beneficial influence on the gastric mucosa and may compensate for the deficiency of endogenous Pdinduced by these NSAIDs. -
list
of abbreviations
ASA: Aspirin; GBF: Gastric blood flow; COX: cyclooxyganase; NO: Nitric oxide; NOS: Nitric oxide synthase; NSAIO: Non-steroidal enti-inflammatory drug; SEM: Standard error of the mean; SNAP: S-Nitroso-penicillamine; ig: Intragastric; TNFa: Tumour necrosis factoralpha; UGIB: Upper gastrointestinal bleeding; Veh: Vehicle
593
MO-MAID
in Sastropratection
and ulcer healing
References i Lanza FL. Endoscopic studies of gastric and duodenal injury after the use of ibuprofen, aspirin and other nonsteroidal anti-inflammatory agents. Am J Med 1984;13:19-24. 2 Levi S, Goodlad RA, Lee CY, Stamp G, Walport MJ, Wright NA, et al. Inhibitory effect of non-steroidal anti-inflammatory drugs on mucosal cell proliferation associated with gastric ulcer healing. Lancet 1990;336:840-3. 3 Konturek PK, Brzozowski T, Konturek SJ, Dembinski A. Role of epidermal growth factor, prostaglandin, and sulfbydryls in stressinduced gastric lesions. Gastroenterology 1990;99:1607-15. 4 Wang JY, Yamasaki S, Takeuchi K, Okabe S. Delayed healing of acetic acid-induced gastric ulcer in rats by indomethacin. Gastroenterology 1989;96:393-402. 5 Wallace JL, Reuter B, Cicala C, M&night W, Grishman M, Cirino G. A diclofenac derivative without ulcerogenic properties. Eur J Pharmacol 1994;257:249-55. 6 Davies NM, Rosenth AG, Appleyard CB, M&night W, Del Soldato P, Calignano A, et al. NO-naproxen vs. naproxen: ulcerogenic, analgesic and anti-inflammatory effects. Aliment Pharmacol Ther 1997;11:69-79. ’ Elliott SN, M&night W, Cirino G, Wallace JL. A nitric oxide-releasing non steroidal anti-inflammatory drug accelerates gastric ulcer healing in rats. Gastroenterology 1995;109:524-30. 8 Wallace JL, MC Knight W, Del Soldato P, Bavdoun AR, Chino G. Anti-thrombotic effects of a nitric oxide-releasing, gastric sparing aspirin derivative. J Clin Invest 1995;96:2711-8. 9 Wallace JL, Chino G, M&night GW, Elliott SN. Reduction of gastrointestinal injury in acute endotoxic shock by flurbiprofen nitroxybutylester. Eur J Pharmacol 1995;280:63-8. lo Takeuchi K, Suzuki K, Yammamoto H, Araki H, Mizoguchi H, Ukawa H. Cyclooxygenase-2 selective and nitric oxide-releasing nonsteroidal anti-inflammatory drugs and gastric mucosal responses. J Physiol Pharmacol 1998;49:501-5. I1 Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: Physiology, pathophysiology and pharmacology. Pharmacol Rev 1991;43:109-42. I2 Vane JR, Botting RM. New insights into the mode of action of anti-inflammatory drugs. Inflamm Res 1995;44:1-10. I3 Gaboury J, Woodman R, Granger DN, Reinhardt P, Kubes P. Nitric oxide prevents leukocyte adherence; role of superoxide. Am J Physiol 1993;265:H862-7. I4 Whittle BJR, Lopez-Bolmonte J, Moncada S. Regulation of gastric mucosal integrity by endogenous nitric oxide: interactions with prostanoids and sensory neuropeptides in the rat. Br J Pharmacol 1990;99:607- 11. I5 Brzozowski T, Drozdowicz D, Szlachcic A, Pytko-Polonczyk J, Majka J, Konturek SJ, et al. Role of nitric oxide and prostaglandins in gastroprotection induced by capsaicin and papaverine. Digestion 1993;54:24-3 1. I6 Brzozowski T, Konturek SJ, Sliwowski Z, Drozdowicz D, Laczek M, Kedra D. Role of arginine, a substrate for nitric oxide synthase in gastroprotection and ulcer healing. J Gastroenterol 1997;32:442-52. I7 Konturek SJ, Brzozowski T, Majka J, Pytko-Polonczyk J, Stachura J. Inhibition of nitric oxide synthase delays healing of chronic gastric ulcers. Eur J Pharmacol 1993;239:215-7. ‘* Konturek PCh, Brzozowski T, Pierzchalski P, Kwiecien S, Pajdo R, Hahn EG, et al. Activation of genes for snasmolvtic neotide, transforming growth factor alpha and for- cyclobxygenase- 1 (COX- 1) and COX-2 durins gastric adautation to as&in damage in rats. Aliment Pharmacol%er 1998;1&2:767-77. * I9 Brzozowski T, Konturek PCh, Pajdo R, Duda A, Pierzchalski P,
594
Bielanski W, et al. Leptin in gastroprotection induced by cholecystokinin or by a meal. Role of vagal and sensory nerves and nitric oxide. Eur J Pharmacol 1999;374:263-76. *” Brzozowski T, Konturek PCh, Konturek SJ, Pajdo R, Drozdowicz D, Kwiecien S, et al. Acceleration of ulcer healing by cholecystokinin (CCK): role of CCK-A receptors, somatostatin, nitric oxide and sensory nerves. Reg Pept 1999;82: 19-33. *’ Konturek SJ, Brzozowski T, Garlicki J, Majka J, Stachura J, Nauert CM. Intragastric pH in the gastroprotective and ulcer healing activity of aluminium-containing antacids. Digestion 1991;49:140-50. ** Okabe S, Pfeiffer CJ, Roth ILA. A method for experimental penetrating gastric and duodenal ulcers in rats. Digestion 1987;38:103-13. 23 Chomczynski J, Sac&i N. A single step method of RNA isolation by acid guanidinum thiocyanate. Anal Biol Chem 1987;162:156-9. 24 Brzozowski T, Konturek PCh, Konturek SJ, Sliwowski Z, Drozdowicz D, Stachura J, et al. Role of prostaglandin-generated by cyclooxygenase-1 and cyclooxygenase-2 in healing of ischemia-reperfusion induced gastric lesions. Eur J Pharmacol 1999;385:47-61. 25 Feng L, Sun W, Xia Y, Tang WW, Chanmugan P, Soyoda E, et al. Cloning two isofonns of rat cyclooxygenase; differential regulation of their expression. Arch Biochem Biophys 1992;307:361-8. 26 Kennedy BP, Chan CC, Culp SA, Cronlish WA. Cloning and expression of rat prostaglandin endoperoxide synthase (cyclooxygenase)-2 cDNA. Biochem Biophys Res Commun 1993;197:494-500. *’ O’Banion MK, Winn VD, Young DA. Cloning and functional activity of a glucocorticoid-regulated inflammatory cyclooxygenase. Proc Nat1 Acad Sci 1992;89:4888-92. ** Lefebre RA. Influence of a selective guanylate cyclase inhibitor, and of the contraction level on nitrergic relaxations in the gastric mucosa. Br J Pharmacol 1998; 124: 1439-48. 29 Furst DE. Pharmacology and efficacy of cyclooxygenase (COX) inhibitors. Am J Med 1999;107:185-228. 3o Vane JR, Botting RM. Anti-inflammatory drugs and their mecbanism of action. Inflamm Res 1998;47:78-87. 31 Fiorucci S, Antonelli E, Santucci L, Morelli 0, Miglietti M, Federici B, et al. Gastrointestinal safety of nitric oxide-derived aspirin is related to inhibition of ICE-like cysteine proteases in rats. Gastroenterology 1999;116:1089-106. 32 Takeuchi K, Suzuki K, Yamamoto H, Araki H, Mizoguchi H, Ukawa H. Cyclooxygenase-2 selective and nitric oxide-releasing nonsteroidal anti-inflammatory drugs and gastric mucosal responses. J Physiol Pharmacol 1998;49:501-13. 33 Takeuchi K, Yasuhiro T, Asada Y, Sugawa Y. Role of nitric oxide in pathogenesis of aspirin-induced gastric mucosal damage in rats. Digestion 1998;59:298-307. 34 Anand BS, Romero JJ, Sanduje SK, Lichtenberger LM. Phospholipid association reduces the gastric mucosal toxicity of aspirin in human subjects. Am J Gastroenterol 1999;94:1818-22. 35 Wallace JL, M&night W, Wilson TL, Del Soldato P, Cirino G. Reduction of shock-induced gastric damage by a nitric oxide-releasing aspirin derivative: role of neutrophils. Am J Physiol 1997;273:1246-51. 36 Mizuno H, Sakamoto Ch, Matsuda K, Wada K, Kichida T, Noguchi H, et al. Induction of cyclooxygenase-2 in gastric mucosa lesions and its inhibition by the specific antagonist delays healing in mice. Gastroenterology 1997;12:387-97. 37 Davies NM, Sharkey KA, Asfaha S, MacNaughton WK, Wallace JL. Aspirin causes rapid up-regulation of cyclooxygenase-2 expression in the stomach of rats. Aliment Pharmacol Ther 1997;11:1101-8. 38 Salvemini D, Currie MG, Mollace V. Nitric oxide-mediated cyclooxygenase activation. J Clin Invest 1996;97:2562-8.
DlGESTLlllER DIS 2000;32:593-94
Gastroprotective and ulcer healing effects of nitric oxide-releasing non-steroidal anti-inflammatory drugs T. Brzozows ki P.Ch. Konturekl S. J. Konturek Z. Sliwowski D. Drozdowicz S. KwiecieA R. Pajdo A. Ptak M. Pawlik E. Hahn’
hW#E Department of Fuysiology,Jegiell~t-tisn UniversitySchoolof Medicine, Crecow, Poland; l Repzzrtmentof Medicine 1, Uffiversiityof Er&ngen-IVurembe~, Erlangen,Germany. Thisst&y was supported by InterdisciplineryCenCerfar Clinical Inves@etion,Erlingen-NWemberg University,Erlehgen,Germany. , lwdrawhw Prof. S.J. Kontorek,Depament of Physiology, Jegieifonien University School of Medicine, 16 Gmegorzeeke
Str., 31-531 Cracow, P&end. FBX 0043-
42-212578.
Background 6; Aim. /Vew class of nitric oxide-releasing non-steroidal anti-inflammatory drugs was shown to inhibit cyclooxygenase and prostaglandin generation without causing mucosal damage but whether these agents are capable of affecting gastric mucosal damage induced by strong irritants and healing of chronic gastric ulcers remains to be studied. In this investigation, effects of nitric oxide-releasing aspirin and nitric oxide-releasing naproxen were compared with those of native agents on gastric lesions provoked by 100% ethanol and on healing of chronic acetic acid ulcers. Results. Both, nitric oxide-releasing aspirin and naproxen dosedapendently attenuated ethanol-induced damage and produced a significant rise in gastric blood flow but did not delay healing of gastric ulcers while native aspirin and naproxen had no influence on ethanol-induced gastric damage but significantly prolonged ulcer healing, reduced gastric blood flow and suppressed mucosal generation of prostaglandin Ez. The gastroprotective and hyperaemic effects of both nitric oxide-non-steroidal anti-inflammatory drugs were completely abolished by ODQ, an inhibitor of guanylyl Cyclase-cGMP system but not influenced by suppression of nitric oxidesynthase with L-NNA. The damaging effects of native acetyl salicylate acid or naproxen were aggravated by acidification of these non-steroidal anti-inflammatory drugs but the exogenous acid added to nitric oxide-acetyl salicylate acid or nitric oxide-naproxen failed to influence their effect. Despite inhibiting of PGEz generation, both nitric oxide-releasing derivatives and native aspirin and naproxen failed to affect expression of cyclooxygenase-I mRNA but upregulated the cyclooxygenase2 mRNA. Concurrent inhibition of cyclooxygenase-2 by selective inhibitor NS-398 which by itself delayed ulcer healing and attenuated the gastric blood flow at ulcer margin, significantly worsened the effects of these nitric oxidenon-steroidal anti-inflammatory drugs and their parent drugs on ulcer healing and the gastric blood flow at the ulcer margin. Conclusions. I] Coupling of nitric oxide to aspirin or naproxen attenuates ethanol-induced damage, possibly due to an increase in gastric microcirculation mediated by excessive release and action of nitric oxide that probably compensates for PG deficiency induced by non-steroidal anti-inflammatory drugs; and 21 nitric oxide-nonsteroidal anti-inflammatory drug, unlike classic non-steroidal anti-inflammatory drugs, does not affect intact gastric mucosa and fails to delay the healing of pre-existing ulcers.
E-me;/:[email protected] Submitted March 15, 2000. RevisedJuly 2U, 2oo0, AcceptedJuly 23 2iX70.
Digest
Liver
Dis 2000;32:583-94
Key words: aspirin; cyclooxygenase; gastric ide: nitric oxide synthase; ulcer healing
blood
flow;
gastroprotection;
naproxen;
nitric
ox-
583
NO-NSAID in gastroprotection
and ulcer healing
Introduction Non-steroidal anti-inflammatory drugs (NSAIDs) such as aspirin (ASA) or naproxen are widely used but the major limitation of their clinical application are serious side-effects, including damage of gastrointestinal mucosa, aggravation of stress ulcerations and exacerbation of pre-existing gastric ulcerations lms.Recently, a new class of NSAID has been developed by addition of nitric oxide (NO) moiety to the native NSAID 6-‘o. The rationale behind this strategy is that NO released from these derivatives exerts a beneficial effect on gastric mucosa by enhancing the mucosal defense ability and prevention of pathogenic events resulting from the suppression of prostanoid synthesis such as the reduction in mucosal microcirculation and the leukocyte-endothelial adherence “-‘3. Earlier studies revealed that endogenous NO released from vascular endothelium, sensory nerves or gastric epithelium co-operates with prostaglandins in the maintenance of gastric mucosa integrity and microcirculation I4 Is. In contrast to native NSAID, their NO-releasing derivatives such as NO-aspirin (NO-ASA) or NO-naproxen were found to exhibit lower gastric toxicity despite similar inhibition of both COX-1 and COX-2 (cyclooxygenase) activity in the gastric mucosa 5 ‘O12.The major importance of NO in the prevention of mucosal damage or in preservation of ulcer healing is supported by previous studies showing that both endogenous NO released by capsaicin Is or NO originating from L-arginine, a substrate for NO-synthase (NOS) 16, or that released from glyceryl trinitrate ” exert gastroprotective activity and accelerate healing mainly due to the maintenance of blood flow around the ulcer and enhancement of angiogenesis. The present study was designed; 1) to compare the effect of topical application of aspirin or naproxen and their NO-derivatives on gastric mucosal integrity and blood flow; 2) to assess the possible gastroprotective effects of NO-ASA and NO-naproxen, and for comparison of S-nitroso-penicillamine (SNAP), a donor of NO I’, added to native NSAID on acute gastric injury provoked by 100% ethanol; 3) to examine the influence of NO-ASA and NO-naproxen and their native compounds as well as SNAP, on the healing of chronic gastric ulcerations with or without the concurrent treatment with NS-398, a highly selective COX-2 inhibitor I” lx; and 4) to evaluate the expression of COX-1 and COX-2 in gastric mucosa with chronic gastric ulcers treated with native NSAID and their NO derivatives.
Material and Methods Male Wistar rats, weighing 180-220 g and fasted for 24 h before the study were employed in all investigations. 584
Rats were divided into series A for studies on production of acute gastric lesions and series B for studies with chronic gastric ulcers. Production of gastric lesions and measurement of gastric bloodflow (GBF) Gastric lesions were produced by an intragastric (ig) application of 1.5 ml of 100% ethanol by means of a metal orogastric tube as described in detail elsewhere Is 16. Native NSAIDs such as ASA, indomethacin and naproxen were obtained from Sigma Chemical Co. (St. Louis, MO, USA) while NO-aspirin (NCX 4016; 2(Acetyloxy)benzoic acid acid 3-(nitrooxymethyl) phenyl ester) and NO-naproxen (HCT 3012; (S)-6Metoxy-a-methyl-2-naphtalenacetic acid 4-(nitrooxy)butyl ester were kindly provided by Mrs. Nathalie Baudry from NicOx S.A., Sophia Antipolis, France. Native ASA (25-100 mg/kg) or naproxen (12.5 -50 mg/kg) and equimolar doses of NO-ASA (40-160 mg/kg) or NO-naproxen (18-72 mg/kg) were introduced ig in order to compare the damage produced by native NSAID and their NO derivatives and to examine the effect of pretreatment with NO-NSAID and their parent drugs applied ig 30 min prior to exposure of the gastric mucosa to 100% ethanol on the mucosal lesions induced by this corrosive agent. One hour after ig application of NSAID, their NO derivatives or ethanol, the animals were anaesthetised with ether, their stomachs were exposed and the GBF was measured in the oxyntic gland area by means of local Hr-gas clearance technique. To evaluate the effect of ASA, naproxen and their NO derivatives on GBF, the groups of animals were anaesthetised with ether and the abdomen was opened by midline incision. The stomach was exposed to assess the blood flow using the H2-gas clearance technique as described elsewhere I9*“. Briefly, the GBF was measured in the gastric mucosa not involved in the gastric lesions or at the ulcer base, ulcer margin and in the intact mucosa in rats with pre-existing ulcers using double electrodes of electrolytic regional blood flowmeter (Biotechnical Science, Model RBF-2, Osaka, Japan) inserted through the serosa into the mucosa. One for these electrodes was used for the local generation in the mucosa of H2 and another for measurement of tissue this Hz. With this method, the H2 generated locally is carried out by flow of blood, while the polarographic current detector records decreasing tissue HZ The clearance curve of tissue HZ was used to calculate an absolute blood flow rate (ml/min/lOO g) in the oxyntic gland area as described elsewhere l”. The measurement was made in three areas of the gastric oxyntic mucosa and the mean values of these measurements were calculated and expressed as percent changes from those recorded in vehicle-treated control animals. Using this technique, the results of the repeated blood
flow gave highly reproducible results and the inter- and intraassay variability of these measurements were 3.5% and 5.8%, respectively. The area of gastric lesions was measured by planimetry (Morphomat, Carl Zeiss, Berlin, Germany) by two investigators, under blinded conditions, as described elsewhere I*. Production
of gastric ulcers
Gastric ulcers were produced in 150 fasted rats of series B, using our modification 21 of the acetic acid method originally proposed by Okabe et al. **. With the animals under ether anaesthesia, the stomach was exposed and 75 yl of acetic acid was poured through the plastic mould (6 mm diameter) onto the serosal surface of the anterior wall of the stomach just proximal to the antral gland area for 25 s. This produced immediate necrosis of the entire mucosa and submucosa within the area where the acetic acid was applied, i.e., about 28 mm?. The excess of acetic acid was then removed and the serosa was gently washed with tap saline. Our previous studies documented that these ulcers become chronic within 2-3 days and heal completely within 2-3 weeks without perforation or penetration to the surrounding organs as described in the original technique of Okabe et al. ?2. After the application of acetic acid, the animals were allowed to recover from anaesthesia and received only water on the day of the operation (Day 0). Then, they were divided into various groups and received normal chow and water ad libitum for the next 3, 9 and I5 days. To evaluate the effects of native ASA or naproxen and their NO-releasing derivatives, the animals were anaesthetised with ether and the abdomen was opened and the stomach was exposed to assess the GBF at ulcer margin, ulcer crater and in the contralateral intact mucosa using the Hz-gas clearance technique as described above I7 *O.The stomach was then removed and pinned open for the determination of the area of gastric ulcers by planimetry. The half of the stomach with gastric ulcer in rats with or without administration of vehicle (control), ASA, naproxen or their NO-derivatives were taken during autopsy and immediately fixed in 10% formalin, embedded in paraffin and stained with haematoxylin and eosin for the histological assessment of the quality of ulcer healing. Coded specimens of mucosa stained with haematoxylin and eosin were evaluated at 260x magnification under blinded conditions. The other half of the stomach was either used immediately for PGE? generation or immediately frozen in liquid nitrogen and then stored at -80°C until use for mRNA expression of COX-I and COX-2 at the ulcer margin and contralateral intact mucosa. Experimental
groups and treatments
Series A consisting of 80 rats was used to determine the acute effects of topical administration of acidified ASA
or naproxen with those of NO-ASA or NO-naproxen against the mucosal lesions provoked by 100% ethanol. Series A included the following groups that received ig: 1) vehicle (saline or 0.2 N HCl); 2) ASA (25-100 mg/kg) or naproxen (12.5-50 mg/kg); 3) NO-ASA (40160 mg/kg) or NO-naproxen (18-72 mg/kg) followed 30 min later by ig application of 100% ethanol (1.5 ml). The time of duration of the protective effect of NONSAID was studied in animals pretreated with NOASA (80 mg/kg ig) or NO-naproxen (72 mg/kg ig) and then followed 30, 60, 120, 240 and 360 min later by ig application of 100% ethanol. In certain groups of rats treated with NO-NSAID, the effect of pretreatment with NG-nitro-L-arginine (L-NNA) to suppress the NO[4,3,synthase I5 or with lH-[1,2,4]oxidiazolo alquinoxalin-l-one (ODQ), which is a highly specific inhibitor of guanylyl cyclase 28, was determined in order to establish the involvement of the NO and guanylyl cyclase-cGMP system in the gastroprotection induced by these NO derivatives of NSAID. Series B rats with gastric ulcers were treated daily with ASA and naproxen (50 mg/kg ig) or with NO-ASA and NO-naproxen introduced ig in equimolar doses of 80 mg/kg and 72 mg/kg, respectively, without or with pretreatment with specific COX-2 inhibitor, N-(2-cyclohexyloxy-4-nitrophenyl)-methansulfonamide (NS398, Taisho Pharmaceutical Co., Tokyo, Japan) at a dose of 5 mg/kg ig. The animals were killed 1 h after ethanol administration or at day 3, 9 and 15 after ulcer induction and the area of gastric ulcerations was determined by planimetry, the GBF was measured by the H2-gas clearance method as described above and the gastric luminal nitrite/nitrate content was assessed by a specific kit based on Griess’ method. As already mentioned, the half of the stomach which was not used for histology, was employed for determination of the generation of PGE? by RIA and COX-1 and COX-2 mRNA by RT-PCR. Determination of PGE2 generation and luminal NO release
in gastric
mucosa
For ex vivo determination of PGE?, the mucosa was processed, according to the method described elsewhere 18. Briefly, a portion of the fundic mucosa (about 100 mg) excised from rats treated with vehicle, acidified ASA, naproxen and NO-ASA or NOnaproxen was placed immediately (after excision) in an Eppendorf plastic tube containing 1 .O ml of 10 mM phosphate buffer (pH 7.4) and minced with fine scissors for 15 s. The samples were then incubated in a shaking bath (37°C) for 20 min and then centrifuged 30 s at 9000 g. The supematant was collected and frozen at -20°C for subsequent determination of PGE? using radioimmunoassay (PGE?-RIA kit, Amersham, Munich, Germany). 585
NO-NW0
in gastmprotectinn
and ulcer healing
For the measurement of NO generation in the gastric lumen, 1 ml of saline was introduced into the stomach and the luminal content was collected to determine the total nitrite/nitrate concentration using a commercially available kit purchased from Cayman (Ann Arbour, MI, USA) as described in detail elsewhere *O. Histological evaluation of gastric mucosa The halves of the stomach treated with ig application of this NO-NSAID or their native agents, naproxen or acidified ASA without or with each of the tested irritants such as ethanol were used for histology. For this purpose, the mucosal specimens were fixed in formalin, then embedded in paraffin and finally stained with haematoxylin and eosin (H & E). The quantitative histology was performed using a Nikon microscope equipped with Microplan II. The disrupted surface of the mucosal strips denuded of epithelium and the deeper necrotic lesions penetrating the mucosa were measured and expressed as a percent of total strip length. Reverse-transcriptase polymerase chain reaction for detection of messenger RNA (mRNA) for COX-1 and cox-2 Gastric mucosal samples were collected (at -8OC) for determination of expression of COX- 1 and COX-2 mRNA by reverse-transcriptase polymerase chain reaction (RT-PCR). Total RNA was isolated from the gastric mucosal samples by guanidinum isothiocyanate/phenol chloroform using single step extraction kit from Stratagene according to the original procedure described by Chomczynski et al. 23, as in our previous studies 24. Single-stranded cDNA was generated from 5 ug of cellular RNA using StrataScriptTM reverse transcriptase (Stratagene, La Jolla, USA) and oligo (dT) primers. The resulting complementary DNA (cDNA) was used as a template for subsequent PCR. After the reverse transcription, the transcriptase activity was destroyed by heating, and the cDNA were stored at -20°C until PCR. The primer sequences were designed according to the published cDNA sequence for the rat cyclooxygenases 25-27.The COX-1 primer sequences were as follows: up-stream, 5’-AGC CCC TCA TTC ACC CAT CAT TT; downstream, 5’-CAG GGA CGC CTG TTC TAC GG. The expected length of this PCR product was 561 bp. The COX-2 primer sequences were as follows: upstream, 5’-ACA ACA TTC CTT CCT TC; downstream, 5’-CCT TAT TTC CTT TCA CAC C. The expected length of this PCR product was 201 bp. Concomitantly, amplification of control rat p-actin (Clon Tech, Palo Alto, CA, USA) was performed on the same samples to ascertain RNA integrity. DNA amplification was carried out under the following conditions; denaturation at 94°C for 1 min, anneal588
ing at 60°C for 45 s, and extension at 72°C for 45 s. To maximise amplification specificity, Taq DNA polymerase was added to the PCR mixture during the hot start of cycle 1. The amplification of PCR was performed at 30 cycles. Each PCR-product (8 ,ul) was electrophoresed on 1.5% agarose gel stained with ethidium bromide, and then visualised under UV light. Location of predicted PCR product was confirmed by using a loo-base pair ladder (Gibco BRL/Life Technologies, Eggenstein, Germany) as standard marker. To avoid PCR contamination, PCR reactions were prepared in a dedicated area used only for PCR and the PCR products were opened in a laminar flow hood separated from the PCR preparation area. The intensity of bands was quantified using semiquantitative densitometry analysis as described in detail elsewhere l9 24. The COX-1 and COX-2 signals were standardised against the p-actin signal for each sample and results expressed as COX-1 and COX-2/pactin mRNA ratio. Amplification was performed using the DNA thermal cycler (Perkin-Elmer-Cetus, Norwalk, USA) and the cycles were established in each separate PCR reaction. The PCR product was electrophoresed on 1.5% agarose gel stained with ethidium bromide, and then visualised under ultraviolet (UV). Expression of the product was quantified using video image analysis system (LKB Ultrascan, Pharmacia, Sweden) as described in detail elsewhere l9 24. Statistical analysis Results are expressed as means f SEM. For statistical analysis, the nonparametric Mann-Whitney U and Kruskal-Wallis tests for comparisons were applied where appropriate, with values of less than 0.05 (~~0.05) taken as statistically significant. Results
EfSect of NO-ASA and NO-naproxen of gastric lesions induced by ethanol and alteration in GBF: PGEI generation and luminal nitrite/nitrate content Figure 1 shows the area of gastric lesions and the alterations in GBF and luminal nitrite/nitrate concentration in rats exposed to 100% ethanol without or with the pretreatment with vehicle (saline) or NO-ASA introduced ig in graded doses ranging from 40 mg/kg up to 160 mg/kg. Mean GBF in the intact gastric mucosa was 52&7 ml/mm/100 g (taken as lOO%), while the mean gastric nitrite/nitrate content was 3.1kO.5 umol/l and these values were not significantly affected following ig application of vehicle (saline). Administration of 100% ethanol to vehicle pretreated rats reduced the blood flow to about 55% and the gastric nitrite/nitrate value
T. Brzozowski et al.
was similar to that recorded in intact mucosa, while causing the appearance of gastric lesions of about 7826 mm* (Fig. 1). Pretreatment with NO-ASA applied topically, significantly reduced the area of lesions induced by 100% ethanol. With graded concentrations of NOASA administered before 100% ethanol, the area of gastric lesions was significantly attenuated and a significant increase in the GBF and nitrite/nitrate content starting with a dose of 80 mg/kg of NO-ASA was observed and the dose reducing, by 50%, the ulcer area IDso averaged about 95 mg/kg. The protective effect of NO-ASA was accompanied by a significant rise in GBF and luminal nitrite/nitrate content (Fig. 1). NOnaproxen given ig also reduced dose-dependently the area of gastric lesions caused by 100% ethanol with the threshold diminution occurring at a dose of 36 mgkg and with the IDso averaging 63 mg/kg of NO-naproxen (Fig. 2). Pretreatment with NO-naproxen resulted in a significant rise in the GBF and luminal nitrite/nitrate contents. In contrast, acidified ASA or naproxen failed to influence the gastric lesions induced by 100% ethanol and accompanying changes in GBF and luminal nitrite/nitrate levels (Figs. 1, 2). The generation of PGE2 in the gastric mucosa averaged 156+ 18 rig/g of wet tissue weight and this was not significantly influenced by exposure of gastric mucosa to 100% ethanol (Table I). Native ASA or naproxen introduced ig in a standard concentration of 50 mg/kg before 100% ethanol significantly inhibited the generation of PGE2 in the gastric mucosa as compared to that measured in rats with ethanol alone. Mucosal content of PGE2 was also suppressed in rats pretreated with NO-ASA or NO-naproxen despite the fact that NO-ASA- or NO-naproxen dose-dependently attenuated the area of gastric lesions induced by 100% ethanol (Table I, Figs. 1, 2). Topical application of 100% ethanol resulted in a significant increase in superficial and deep necrosis and this was not significantly affected by pretreatment with native ASA or naproxen. In la!& I. Generation of Pi& in gastric mucosa end area of superficial and deep histalogical damage in intact rats and those pretreated with vehicle Net?), AS4 and nepmxen @I m&kg Lg.1 or NO-ASA f60 mglkg Lg.1 and NO-naproxen (72 m@kg Lg.1 before exposure to ‘KiC% ethanol. Histology results are expressed as peme-nw of damaged area in mucusal strip length. Results ara mean k !3EM of 6 - 6 rats.
Intact Veh AS A Napmxen NO-ASA NO-napmxen
156*16 145zkl1 3%4” 58~t7” 47r6 56r5
0.71to.03 43k3.2 46ct4.2 421t2.8* 21*2.2’+ 24k7.8”’
0 21*0.8 22~0.6 20a0.4 5*0.6”+ 7*0.4*+
’ Signihnt dwge as canpsred to value oL%%ined in intact re& +signh%xnCdwge 8s wmpai-ed M due obtained in IW& pretreate with native ASA or nepmxen. Abbrevi&ms: see list. 587
NO-NSRID in gastroprotection
and ulcer healing
contrast, the pretreatment with NO-ASA or NOnaproxen significantly attenuated the area of superficial and deep necrosis provoked by ethanol as compared to that measured in rats pretreated with native ASA or naproxen. Figure 3 shows the duration time of the protective effects of NO-NSAID against ethanol damage in comparison with the effect of their parent NSAID. This protection lasted up to 4 hours upon NO-ASA (80 mg/kg ig) or NO-naproxen (72 mg/kg ig) administration while native ASA or native naproxen had no influence on the ethanol-induced lesions at any time.
[email protected]. Area of ethanol-induced gastric lesions and gastric blood flow fG8F1 in rats treated with vehicle Winal, NO-ASA f8Clmgkg igl or NO-naproxen f72 mg&g igl without or with the pretreatment with LNNA (20 m$kg ivl. Means f SEM of 6-8 rats. ’ Significant change as compared to those obtained in vehicle-treated gestric mucosa with or without L-NNA pretreatment. + Significant change as compared to value obtained in vehicle-treated gastric mucosa without pretreatment with L-NNA,
so-...
I
Fig. 3. Time duration of protective effect of vehicle fsalinel, NO-ASA I80 mgkg igl or NO-napmxen (72 mgkg igl in comparison with native ASA or native napmxen f50 mgkg igl against gastric lesions induced by 100% ethanol. Means f SEM of 6-8 rats. * Significant change as compared to those obtained in vehicletreated gastric mucosa
Effect of suppression of NO-synthase and inhibition of guanylyl cyclase-cGMP system with ODQ on the gastroprotection exhibited by NO-ASA and NO-naproxen Figure 4 shows the effect of NO-ASA and NO-naproxen introduced ig in a standard dose of 80 mg/kg and 72 mg/kg, respectively, on the area of gastric lesions induced by 100% ethanol and accompanying changes in the GBF in rats without and with pretreatment with LNNA, an inhibitor of NOS. L-NNA by itself applied at a dose of 20 mg/kg iv failed to affect the ethanol damage but produced a significant drop in the GBF as compared to that in vehicle pretreated rats given 100% ethanol. NO-ASA and NO-naproxen significantly attenuated the area of acute gastric lesions or the drop in GBF caused by 100% ethanol and these effects were not influenced by pretreatment with L-NNA. In contrast, pretreatment with ODQ, which is a highly specific inhibitor of guanylyl cyclase 28, completely re588
Fi3.5. Area of ethanol-induced gastric lesions and gastric blood flow [GBFI in rats treeted with vahide faatinal, NO-ASA f80 n-q/kg igl or with NO-naproxen f72 mgkg igl without or with pratreatment with OOQ (10 mg/kg i.g.1. Means f S&l of 6-6 rats. *Significant change as compared to those obtained in v&&la-treatad gastric rnucosa. ‘significant ohange as comparad to value obtained in rats without OOO pretreatment.
versed the protection and the rise in the GBF evoked by NO-ASA and NO-naproxen (Fig. 5). Effect of addition of exogenous acid to native and NO-releasing NSAID on gastric damage and alteration in GBF As shown in Figure 6, native ASA or native naproxen produced gastric lesions and reduced GBF as compared
T. Brzomwski
AeA NmoxaN +SNM +SNAP
et al.
MA NAROXEN
Fi& 0. Mean araa of gastric lesions and changes in GBF induced by native ASA and native naproxen Iboth 50 mgkg ig1 or NO-ASA [80 m&!&g ig) and NG-naproxen I72 m@kg igl without or with addition of exogenous HM. Mean k SEM of 68 rats. *Significant change as compared to vslue obteinad in vehicle-control animals. + A significant changa as compared to valueobtained in rats without acidification.
Fig. 7. Mean area of gastric lesions and changes in G8F induced by native ASA or native naproxen or their NO-derivatives NO-ASA and NO-naproxen without or with addition of S-nitroso-penicillamine CiNAP, 5 mgkg igl , Mean + SEM of 6-8 rats. * Significant change as compared to the value obtained in vehicle-control animals or those pretreated with native ASA or naproxan.
to those measured in the vehicle-control gastric mucosa without luminal acidification. In contrast, NO-ASA and NO-naproxen failed to induce gastric lesions and this was accompanied by a significant rise in the GBF as compared to that obtained in ASA- or naproxentreated mucosa. The addition of 1 ml of 0.2 N HCl to the stomach by itself failed to produce gastric lesions but reduced the GBF, to about 90% of the intact value. It resulted in a remarkable worsening of gastric lesions induced by native NSAID and greater drop in the GBF compared to those recorded with these agents in gastric mucosa without addition of exogenous acid. Addition of exogenous acid to NO-ASA or NO-naproxen failed to produce gastric lesions and did not induce any greater alteration in the GBF than that observed in the gastric mucosa treated with NO-ASA or NO-naproxen without the combination with exogenous acid.
Effect of vehicle or native ASA and naproxen and their NO-releasing derivatives or SNAP with or without NS398 on the area of gastric ulcers and gastric blood flow (GBF) at ulcer margin Figure 8 shows the time-course of ulcer healing in rats following ig administration of vehicle, native ASA and naproxen applied daily in a dose of 50 mg/kg and NOASA, NO-naproxen and SNAP applied daily in a single
Effect of S-nitroso-penicillamine (SNAP) alone or combined with ASA and naproxen on gastric lesions and the GBF As shown in Figure 7, pretreatment of the gastric mucosa with SNAP (5 mg/kg ig), a potent donor of NO, significantly reduced the area of ethanol lesions and this was accompanied by a significant rise in the GBF. In contrast, native ASA or naproxen failed to affect ethanol lesions and the accompanying drop in the GBF induced by this irritant. Addition of SNAP to native ASA or NOnaproxen afforded the protection and the rise in the GBF to an extent similar to that observed after the administration of NO-ASA or NO-naproxen introduced ig at a dose of 80 mg/kg and 72 mg/kg, respectively.
NOPROXE
O
J
-+
i
3
9
II
DAVS AFTER ULCRR INDllCllON
Fig. 8. Area of chronic gestrii ulcers in rats treated by vehicle, native ASA and native naproxan or their NO-derivatives NO-ASA and NOnaproxen and SNAP, a donor of NO, during period of 3,9 end 15 days upon ulcer induction. fvlaan f SEM of 6-8 rats. “significant change as compared to value obtained at day 0. + Significant change as compared to value obtained in vehicle-control gastric mucosa at day 9 and day 15 upon ulcer induction.
588
NO-NSAID in gaatroprotection
and ulcer healing
dose of 80 mg/kg, 72 mg/kg and 5 mg/kg, respectively. The alterations in these ulcers are expressed as a change in the area of these ulcers as determined at day 0, 3, 9 and 15 upon their induction. In rats treated with vehicle, a significant reduction in the area of these ulcers was observed from an initial size of about 28 mm2 to 16.2k3.1 mm* and 10.9k2.5 mm2 at day 9 and 15, respectively. Treatment with SNAP resulted in a significant acceleration of the ulcer healing at day 9 and day 15 as compared to the respective values recorded in rats treated with vehicle. Native ASA or naproxen failed to significantly affect the area of gastric ulcers at day 3 upon ulcer induction but significantly delayed ulcer healing at day 9 and day 15, while NO-AS A or NO-naproxen did not interfere with the ulcer healing as compared to that measured in vehicle-control. The GBF in the non-ulcerated antro-oxyntic mucosa of rats treated with vehicle averaged 48 ml/min- 100 g (taken as 100%) being significantly reduced (by about 15%) at the ulcer margin at day 15 when compared to that in the contralateral intact mucosa (Fig. 9). Native ASA or naproxen significantly delayed ulcer healing at day 15 and this was accompanied by the significant decrease in the GBF at the ulcer margin as compared to the respective values recorded in vehicle-treated gastric mucosa. In contrast, SNAP significantly accelerated the healing of gastric ulcers at day 15 and increased the GBF at the ulcer margin to a significantly greater extent than that recorded in intact gastric mucosa. Treatment with NO-ASA and NO-naproxen accelerated ulcer healing in a similar fashion to vehicle and this was accompanied by a rise in the GBF at ulcer margin to a significantly higher extent than that in gastric mucosa treated with native NSAID and similar to that observed in vehicle-treated rats. Co-administration of NS-398, a highly selective COX2 inhibitor with vehicle, ASA and naproxen, or NOASA and NO-naproxen, produced a significant increase in the area of gastric ulcers as compared with the respective values obtained in rats treated with vehicle or NO-NSAID without addition of COX-2 inhibitor (Fig. 9). Only the acceleration of ulcer healing and accompanying hyperaemia at the ulcer margin induced by SNAP were significantly attenuated in rats with the concurrent treatment with NS-398. The increase in the area of gastric ulcers induced by native ASA and naproxen was significantly enhanced by the addition of COX-2 inhibitor to these NSAIDs. Likewise, the concurrent treatment with NS-398 significantly delayed the ulcer healing and decreased the GBF at the ulcer margin as compared to those treated with NO-ASA and NO-naproxen alone (Fig. 9). PGE2 generation in intact non-ulcerated gastric mucosa of the vehicle-treated rats at day 15 after ulcer induction averaged 9427 uglg wet tissue weight and this was sig590
nificantly increased in the ulcerated gastric mucosa excised from the margin of the gastric ulcer (Fig. 10). ASA and naproxen or NO-ASA and NO-naproxen inhibited mucosal PGE2 generation in the intact non-ulcerated gastric mucosa, the extent being similar to that
ULChDAIrD MUCOSA
Fig. 18. Generation of PGE2 in ulcerated [ulcer margin1 and non-ulcerated Wactl gastric mucosa of rats treated 15 days with native ASA, native naproxen, NO&A, NO-napruxen and SNAR a donor of NO. Mean & SEM of S-8 rate. * Significant change es compared to value obtained in vehicle-control k&act1 gaetric mucosa. + Significant change as compared to va&s obtained in the non-ulcerated gastric mucosa. ++Significent change as compared to vehicle-Weated uloereted mucosa.
1. Inozowski
measured in ulcerated gastric mucosa. In contrast, SNAP significantly raised the generation of PGE2 in the non-ulcerated and ulcerated gastric mucosa as compared to those obtained in vehicle-treated mucosa. RT-PCR expression of COX-1 and COX-2 mRNA in gastric mucosa of rats treated with native ASA or naproxen and their NO-derivatives The expression of COX-1 mRNA was detected as a strong signal in gastric mucosa treated with vehicle, whereas COX-2 mRNA was undetectable in this mucosa (Fig. 11 B,C). The intense signals for COX-1 mRNA was preserved in gastric mucosa exposed to native ASA and naproxen as well as in those treated with NOASA or NO-naproxen (Fig. 11 B). In contrast, COX-2 mRNA was strongly upregulated in the gastric mucosa at the ulcer margin and the mucosa whether treated with native ASA and naproxen or as well as with NOASA and NO-naproxen (Fig. 11 C). The mRNA expression for p-actin was well preserved in each sample tested (Fig. 11 A). The ratio of COX-1 mRNA over p-
et al.
actin mRNA was not significantly different in rats treated with native ASA and naproxen or their NO-derivatives (Fig. 12). In contrast, the ratio of COX-2 mRNA over p-actin showed a significant rise already at the ulcer margin and in the gastric mucosa of rat treated with ASA and naproxen or NO-ASA and NO-naproxen (Fig. 13). The ratio of COX-2 over p-actin was significantly increased in rats treated with native ASA and its NO-derivative as compared to those treated with vehicle, native naproxen and NO-naproxen.
e
1.01
FQ.W. Assessmentof ratio C;OX-1mRNA&eetin mRNAexpression aspreeetlWinFig. l1,MeanitMQfi%?rata
Fi& H. ti&ssm RNAexpreesiwtfor @-actin,CCJX-1end COX-E tnRNAat ay 1k upon ulcer induatt~ in rc+@with in it%aotgastric mocasaMW 1) and uk%rsti mucpse fuioer mclrginltreatad wit1 whir&t kse 2J, rte&e A8A &me 31, netim neprvxsnkne 4J, NO ASA km 51 ad ~n~~x~ Itme 6). I% is rtegstivacontmi &a* terl. M-eke msrkerDhlA,AITQW - expectedPCRproduct (bp1.
F** 1% Assessmentof r&i2 COX-2rn~~otin,rn~ expmeeion as preeent&iin Fig.11. Mesr~$ SE%!of 6+8r&s;.“$i&@fi@sot oheege es compersdto v&usoh&&ad in iotai gsst& We5 +SignMsnt changeee compsr& to Joe o&a&d in rata trek&d with vehide, netiveneproxenand NO-n~xen
NO-NSAID in gastropratection
and ulcer healing
Discussion Our study demonstrates that the new class NSAID, namely NO-ASA and NO-naproxen, by themselves do not produce macroscopic mucosal damage despite the PGEz suppression similar to that observed with their parent ASA and naproxen. In clear contrast, topical application of ASA or naproxen, particularly when acidified, results in gastric lesions and this effect is accompanied by a significant drop in GBF and marked suppression of PGEz generation. Both NO-NSAIDs dosedependently attenuated lesions induced by 100% ethanol and this protective effect of NO-NSAIDs was accompanied by a rise in GBF and a significant increase in nitrite/nitrate content in the gastric lumen. Cotreatment with SNAP, a potent source of NO, added to native ASA or naproxen, also affords the protection, hyperaemia and rise in the luminal NO release similar to those observed with NO-NSAID. Suppression of the specific NO-sensitive guanylyl cyclase-cGMP pathway by the pretreatment with ODQ, dramatically reversed the protection and the rise in the GBF induced by NO-releasing of NSAIDs suggesting that NO released from these NO-derivatives of NSAIDs contributes to their beneficial effects in the gastric mucosa involving intracellular guanylyl cyclase and the rise in cGMP level. In the chronic ulcer studies, treatment with ASA or naproxen administered ig for 15 days caused an expected delay in ulcer healing and decreased both GBF at ulcer margin and PGE2 synthesis in the gastric mucosa. In contrast, NO-ASA and NO-naproxen did not delay the healing rate of chronic gastric ulcers and GBF at ulcer margin as compared to those in the vehicle-treated controls. Treatment with NS-398, a highly selective COX-2 inhibitor, which by itself prolonged ulcer healing and decreased GBF at the ulcer margin, reversed the favourable effects of NO-NSAIDs and increased the deleterious effects of native ASA and native naproxen on ulcer healing and GBF at the ulcer margin. Paradoxically, COX-2 specific inhibitors that are widely proposed as an attractive therapeutic development in the treatment of rheumatoid arthritic and osteoarthritis, partly because they were shown to spare the COX-1 isoform responsible for gastrointestinal protection and failed to cause peptic ulcers or bleeding 29, appear, in our hands, to exert much less favourable effects on healing of pre-existing ulcers than simple NO-NSAID, blocking both COX-1 and COX-2. Further studies, particularly clinical trials with novel NO-NSAID which are in progress, should provide an answer as to whether these relatively simple agents are more beneficial to the gastrointestinal tract than the so-called “Coxibs” recently designed for this purpose. Furthermore, we found that COX- 1 mRNA was detected in the 592
intact gastric mucosa at the ulcer margin and in the mucosa treated with NO-ASA, NO-naproxen and their parent NSAID. In contrast the COX-2 mRNA was not detected in rat mucosa treated with vehicle (control) but was upregulated already at the ulcer margin in rats treated for 15 days with vehicle and both, native ASA and naproxen or their NO-derivatives. This suggests that the enhanced mucosal expression of COX-2 at the ulcer margin contributes to ulcer healing and that NO released from gastric-sparing NO-NSAIDs derivatives can compensate with NO for PG deficiency induced by NSAID. As shown in previous studies, NSAID exhibit potent anti-inflammatory and analgesic actions but the use of these drugs, in humans, is limited due to untoward effects such as formation of gastric mucosal bleeding erosions, enhancement of the ulcerogenic response to various stimuli and impairment of healing of pre-existing ulcers l 2 I2 30. This deleterious action of conventional NSAIDs was attributed to their topical irritating effect, activation of neutrophils, fall in the microcirculation, enhancement of motility induced by these agents and the reduction in mucosal generation of PGEz 4 I2 30. In contrast, newly developed NO-releasing NSAID, constructed by adding an nitroxy-butyl moiety to aspirin or naproxen, were reported to attenuate side-effects of native NSAID including gastrointestinal toxicity while exerting anti-thrombotic effects comparable to their parent NSAID 5 6 lo. Moreover, it was shown that NOreleasing NSAID by themselves exhibit only minimal ulcerogenic properties in the gastrointestinal tract, despite exerting a potent anti-inflammatory and analgesic action, similar to native NSAIDs 9 lo. The NSAIDs examined in our present study, such as aspirin and naproxen, were ulcerogenic by themselves in the stomach and their damaging action was strongly enhanced by addition of the exogenous acid that was employed to mimic the fate of both NSAIDs such as those occurring in highly acidic conditions in the stomach. Moreover, both non-specific COX-inhibitors inhibited PGE;! generation confirming previous observations that suppression of COX and subsequent deficiency of endogenous PG in the gastric mucosa could account, at least in part, for their damaging effect. In contrast, NO-releasing ASA and NO-releasing naproxen by themselves were devoid of ulcerogenic properties and failed to enhance gastric damage after acidification of these drugs, despite the sustained ability of these agents to suppress PGE:! generation similar to that observed after application of their parent drugs. Furthermore, pretreatment with NO-ASA and NO-naproxen resulted in the attenuation of ethanol-induced damage and significantly raised the GBF and these effects were counteracted by ODQ, an inhibitor of NO-dependent guanylyl cyclase 28 but remained unaffected by inhibition of NO-synthase
by L-NNA. The mechanism of protection and accompanying hyperaemia induced by NO-releasing NSAIDs is not yet clear but could be due to excessive local NO released from NO-derivatives of NSAIDs. This is supported by the fact that a considerable amount of NO metabolites were detected in the luminal content of the stomach of rats pretreated with NO-releasing NSAID. Furthermore, SNAP, a potent NO donor, which by itself significantly reduced the gastric damage provoked by ethanol, added to native ASA or naproxen resulted in protection and hyperaemia comparable to those exhibited by NO-releasing NSAIDs suggesting that, indeed, NO released from these compounds plays a crucial role in this protection and accompanying hyperaemia. Since gastrointestinal ulcerations are associated with the use of all NSAIDs, the new strategy for treatment of inflammatory states includes novel series of NSAID that consist of an NSAID-linked either to a NO-releasing moiety 3’-33or to phospholipids such as zwitterronic acid 34. The rationale behind the development of NONSAIDs that spare the gastric mucosa was that NO released from these compounds would counteract two events that occur subsequent to the suppression of PG synthesis by the NSAID, namely the reduced GBF and an increased adherence of neutrophils to the vascular endothelium of the gastric microcirculation I1 l7 3336. Both these events have been suggested as critical in the pathogenesis of experimental NSAID-gastropathy 3o3536. Recently, Fiorucci et al. 3’ have shown that administration of aspirin resulted in enhancement of apoptosis rate via upregulation of the caspase system mediated by tumour necrosis factor-alpha (TNFa) and these effects were counteracted by NO-derivatives of NSAID. It was suggested that NO-NSAID such as NO-ASA can spare the gastric mucosa and inhibit caspase activity, at least in part, through the cGMP-dependent pathway 31 including the release of NO from these NSAIDs. We documented that both native aspirin and naproxen significantly impaired the healing of chronic gastric ulcers and attenuated both the GBF at the ulcer margin and the PGE2 generation in the intact as well as ulcerated gastric mucosa. This inhibition of COX-1 and COX-2 activity could contribute to the prolongation of ulcer healing by NSAID, since previous studies have shown that indomethacin, a non-selective COX inhibitor and NS398 or rofecoxib, both highly specific inhibitors of COX-2, delayed healing of pre-existing chronic gastric ulcers and ischaemia-reperfusion-induced acute erosions progressing into chronic ulcers 24.The importance of COX-2 for ulcer healing was recently emphasised by the observation that expression of COX-2 mRNA is increased after induction of chronic gastric ulcers suggesting that COX-2 expression and PG-derived from COX-2 may be crucial in the mechanism of ulcer healing 2436. Furthermore, it was demonstrated that aspirin
upregulated COX-2 mRNA, possibly due to the deficiency of endogenous PG induced by this NSAID 37. Our results are in keeping with these findings since we showed COX-2 mRNA to be upregulated at the ulcer margin in vehicle control rats and in those treated throughout a period of 15 days with aspirin and naproxen while under the same experimental conditions, mRNA for COX-1 expression remained at the same level. It is, therefore, rational to assume that PGs derived from COX-2 located at the ulcer margin may be crucial in the healing response of gastric ulcer. This hypothesis is supported by the fact that treatment with NS-398 increased the delay in ulcer healing caused by classic NSAID and significantly impaired that exhibited by NO-releasing NSAID and NO donor such as SNAP. This suggests that COX-2 derivatives play an important role in the effect of NO-NSAID on ulcer healing and that NO derived, for instance, from NO-NSAID activates the COX-2 pathway as suggested by Salvemini et al. 38. Interestingly, treatment with NO-ASA and NOnaproxen failed to significantly delay the ulcer healing despite the fact that these NO-derivatives of NSAID suppressed COX-activity in the non-ulcerated and ulcerated gastric mucosa as effectively as their parent NSAID. Thus, it is possible that NO released from NOreleasing NSAID may counteract the potential mucosal impairment resulting from the PG deficiency due to the COX inhibition. This hypothesis is supported by our previous observations that L-arginine, a substrate for NO-synthase or NO donors such as glyceryl trinitrate, accelerate ulcer healing by enhancement in the microcirculation at the ulcer margin and angiogenesis I6 “. On the other hand, administration of NO-synthase inhibitors such as L-NNA or L-NAME delayed healing of acute gastric injury and prolonged the healing of chronic gastric ulcers I’. In summary, we found that NO-releasing aspirin and naproxen, unlike their native substances, while inhibiting PGE2 generation in the gastric mucosa, are not ulcerogenic by themselves, fail to impair the healing of pre-existing ulcers and rather exhibited gastroprotective activity against corrosive substance such as ethanol suggesting that NO released from these compounds has a beneficial influence on the gastric mucosa and may compensate for the deficiency of endogenous Pdinduced by these NSAIDs. -
list
of abbreviations
ASA: Aspirin; GBF: Gastric blood flow; COX: cyclooxyganase; NO: Nitric oxide; NOS: Nitric oxide synthase; NSAIO: Non-steroidal enti-inflammatory drug; SEM: Standard error of the mean; SNAP: S-Nitroso-penicillamine; ig: Intragastric; TNFa: Tumour necrosis factoralpha; UGIB: Upper gastrointestinal bleeding; Veh: Vehicle
593
MO-MAID
in Sastropratection
and ulcer healing
References i Lanza FL. Endoscopic studies of gastric and duodenal injury after the use of ibuprofen, aspirin and other nonsteroidal anti-inflammatory agents. Am J Med 1984;13:19-24. 2 Levi S, Goodlad RA, Lee CY, Stamp G, Walport MJ, Wright NA, et al. Inhibitory effect of non-steroidal anti-inflammatory drugs on mucosal cell proliferation associated with gastric ulcer healing. Lancet 1990;336:840-3. 3 Konturek PK, Brzozowski T, Konturek SJ, Dembinski A. Role of epidermal growth factor, prostaglandin, and sulfbydryls in stressinduced gastric lesions. Gastroenterology 1990;99:1607-15. 4 Wang JY, Yamasaki S, Takeuchi K, Okabe S. Delayed healing of acetic acid-induced gastric ulcer in rats by indomethacin. Gastroenterology 1989;96:393-402. 5 Wallace JL, Reuter B, Cicala C, M&night W, Grishman M, Cirino G. A diclofenac derivative without ulcerogenic properties. Eur J Pharmacol 1994;257:249-55. 6 Davies NM, Rosenth AG, Appleyard CB, M&night W, Del Soldato P, Calignano A, et al. NO-naproxen vs. naproxen: ulcerogenic, analgesic and anti-inflammatory effects. Aliment Pharmacol Ther 1997;11:69-79. ’ Elliott SN, M&night W, Cirino G, Wallace JL. A nitric oxide-releasing non steroidal anti-inflammatory drug accelerates gastric ulcer healing in rats. Gastroenterology 1995;109:524-30. 8 Wallace JL, MC Knight W, Del Soldato P, Bavdoun AR, Chino G. Anti-thrombotic effects of a nitric oxide-releasing, gastric sparing aspirin derivative. J Clin Invest 1995;96:2711-8. 9 Wallace JL, Chino G, M&night GW, Elliott SN. Reduction of gastrointestinal injury in acute endotoxic shock by flurbiprofen nitroxybutylester. Eur J Pharmacol 1995;280:63-8. lo Takeuchi K, Suzuki K, Yammamoto H, Araki H, Mizoguchi H, Ukawa H. Cyclooxygenase-2 selective and nitric oxide-releasing nonsteroidal anti-inflammatory drugs and gastric mucosal responses. J Physiol Pharmacol 1998;49:501-5. I1 Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: Physiology, pathophysiology and pharmacology. Pharmacol Rev 1991;43:109-42. I2 Vane JR, Botting RM. New insights into the mode of action of anti-inflammatory drugs. Inflamm Res 1995;44:1-10. I3 Gaboury J, Woodman R, Granger DN, Reinhardt P, Kubes P. Nitric oxide prevents leukocyte adherence; role of superoxide. Am J Physiol 1993;265:H862-7. I4 Whittle BJR, Lopez-Bolmonte J, Moncada S. Regulation of gastric mucosal integrity by endogenous nitric oxide: interactions with prostanoids and sensory neuropeptides in the rat. Br J Pharmacol 1990;99:607- 11. I5 Brzozowski T, Drozdowicz D, Szlachcic A, Pytko-Polonczyk J, Majka J, Konturek SJ, et al. Role of nitric oxide and prostaglandins in gastroprotection induced by capsaicin and papaverine. Digestion 1993;54:24-3 1. I6 Brzozowski T, Konturek SJ, Sliwowski Z, Drozdowicz D, Laczek M, Kedra D. Role of arginine, a substrate for nitric oxide synthase in gastroprotection and ulcer healing. J Gastroenterol 1997;32:442-52. I7 Konturek SJ, Brzozowski T, Majka J, Pytko-Polonczyk J, Stachura J. Inhibition of nitric oxide synthase delays healing of chronic gastric ulcers. Eur J Pharmacol 1993;239:215-7. ‘* Konturek PCh, Brzozowski T, Pierzchalski P, Kwiecien S, Pajdo R, Hahn EG, et al. Activation of genes for snasmolvtic neotide, transforming growth factor alpha and for- cyclobxygenase- 1 (COX- 1) and COX-2 durins gastric adautation to as&in damage in rats. Aliment Pharmacol%er 1998;1&2:767-77. * I9 Brzozowski T, Konturek PCh, Pajdo R, Duda A, Pierzchalski P,
594
Bielanski W, et al. Leptin in gastroprotection induced by cholecystokinin or by a meal. Role of vagal and sensory nerves and nitric oxide. Eur J Pharmacol 1999;374:263-76. *” Brzozowski T, Konturek PCh, Konturek SJ, Pajdo R, Drozdowicz D, Kwiecien S, et al. Acceleration of ulcer healing by cholecystokinin (CCK): role of CCK-A receptors, somatostatin, nitric oxide and sensory nerves. Reg Pept 1999;82: 19-33. *’ Konturek SJ, Brzozowski T, Garlicki J, Majka J, Stachura J, Nauert CM. Intragastric pH in the gastroprotective and ulcer healing activity of aluminium-containing antacids. Digestion 1991;49:140-50. ** Okabe S, Pfeiffer CJ, Roth ILA. A method for experimental penetrating gastric and duodenal ulcers in rats. Digestion 1987;38:103-13. 23 Chomczynski J, Sac&i N. A single step method of RNA isolation by acid guanidinum thiocyanate. Anal Biol Chem 1987;162:156-9. 24 Brzozowski T, Konturek PCh, Konturek SJ, Sliwowski Z, Drozdowicz D, Stachura J, et al. Role of prostaglandin-generated by cyclooxygenase-1 and cyclooxygenase-2 in healing of ischemia-reperfusion induced gastric lesions. Eur J Pharmacol 1999;385:47-61. 25 Feng L, Sun W, Xia Y, Tang WW, Chanmugan P, Soyoda E, et al. Cloning two isofonns of rat cyclooxygenase; differential regulation of their expression. Arch Biochem Biophys 1992;307:361-8. 26 Kennedy BP, Chan CC, Culp SA, Cronlish WA. Cloning and expression of rat prostaglandin endoperoxide synthase (cyclooxygenase)-2 cDNA. Biochem Biophys Res Commun 1993;197:494-500. *’ O’Banion MK, Winn VD, Young DA. Cloning and functional activity of a glucocorticoid-regulated inflammatory cyclooxygenase. Proc Nat1 Acad Sci 1992;89:4888-92. ** Lefebre RA. Influence of a selective guanylate cyclase inhibitor, and of the contraction level on nitrergic relaxations in the gastric mucosa. Br J Pharmacol 1998; 124: 1439-48. 29 Furst DE. Pharmacology and efficacy of cyclooxygenase (COX) inhibitors. Am J Med 1999;107:185-228. 3o Vane JR, Botting RM. Anti-inflammatory drugs and their mecbanism of action. Inflamm Res 1998;47:78-87. 31 Fiorucci S, Antonelli E, Santucci L, Morelli 0, Miglietti M, Federici B, et al. Gastrointestinal safety of nitric oxide-derived aspirin is related to inhibition of ICE-like cysteine proteases in rats. Gastroenterology 1999;116:1089-106. 32 Takeuchi K, Suzuki K, Yamamoto H, Araki H, Mizoguchi H, Ukawa H. Cyclooxygenase-2 selective and nitric oxide-releasing nonsteroidal anti-inflammatory drugs and gastric mucosal responses. J Physiol Pharmacol 1998;49:501-13. 33 Takeuchi K, Yasuhiro T, Asada Y, Sugawa Y. Role of nitric oxide in pathogenesis of aspirin-induced gastric mucosal damage in rats. Digestion 1998;59:298-307. 34 Anand BS, Romero JJ, Sanduje SK, Lichtenberger LM. Phospholipid association reduces the gastric mucosal toxicity of aspirin in human subjects. Am J Gastroenterol 1999;94:1818-22. 35 Wallace JL, M&night W, Wilson TL, Del Soldato P, Cirino G. Reduction of shock-induced gastric damage by a nitric oxide-releasing aspirin derivative: role of neutrophils. Am J Physiol 1997;273:1246-51. 36 Mizuno H, Sakamoto Ch, Matsuda K, Wada K, Kichida T, Noguchi H, et al. Induction of cyclooxygenase-2 in gastric mucosa lesions and its inhibition by the specific antagonist delays healing in mice. Gastroenterology 1997;12:387-97. 37 Davies NM, Sharkey KA, Asfaha S, MacNaughton WK, Wallace JL. Aspirin causes rapid up-regulation of cyclooxygenase-2 expression in the stomach of rats. Aliment Pharmacol Ther 1997;11:1101-8. 38 Salvemini D, Currie MG, Mollace V. Nitric oxide-mediated cyclooxygenase activation. J Clin Invest 1996;97:2562-8.