Markedly reduced toxicity of a hydrogen sulphide-releasing derivative of naproxen (ATB-346)

Abstract

Background and purpose: Hydrogen sulphide is an important mediator of gastric mucosal defence. The use of non-steroidal anti-inflammatory drugs continues to be limited by their toxicity, particularly in the upper gastrointestinal tract. We evaluated the gastrointestinal safety and anti-inflammatory efficacy of a novel hydrogen sulphide-releasing derivative of naproxen, ATB-346 [2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester].

Experimental approach: The ability of ATB-346 versus naproxen to cause gastric damage was evaluated in healthy rats and in rats with compromised gastric mucosal defence. Effects on the small intestine and on the healing of ulcers were also assessed. The ability of ATB-346 to inhibit cyclooxygenase-1 and 2 and to reduce inflammation in vivo was also evaluated.

Key results: ATB-346 suppressed gastric prostaglandin E(2) synthesis as effectively as naproxen, but produced negligible damage in the stomach and intestine. In situations in which the gastric mucosa was rendered significantly more susceptible to naproxen-induced damage (e.g. ablation of sensory afferent nerves, inhibition of endogenous nitric oxide or hydrogen sulphide synthesis, co-administration with aspirin, antagonism of K(IR)6.x channels), ATB-346 did not cause significant damage. Unlike naproxen and celecoxib, ATB-346 accelerated healing of pre-existing gastric ulcers. In a mouse airpouch model, ATB-346 suppressed cyclooxygenase-2 activity and inhibited leukocyte infiltration more effectively than naproxen. ATB-346 was as effective as naproxen in adjuvant-induced arthritis in rats, with a more rapid onset of activity. Unlike naproxen, ATB-346 did not elevate blood pressure in hypertensive rats.

Conclusions and implications: ATB-346 exhibits anti-inflammatory properties similar to naproxen, but with substantially reduced gastrointestinal toxicity.

Figure 1

Chemical structures of two hydrogen sulphide-releasing derivatives of naproxen (ATB-345 and ATB-346). ATB-345, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-(5-thioxo-5H-[1,2]dithiol-3-yl)-phenyl ester; ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester.

Figure 2

Effects of naproxen and two hydrogen sulphide-releasing naproxen derivatives (ATB-345 and ATB-346) in a model of zymosan-induced inflammation in the mouse. Samples of inflammatory exudates and of blood were collected 6 h after injection of zymosan into a preformed airpouch on the back of the mice. The test drugs were administered orally at a dose of 30 µmol·kg⁻¹ 1 h prior to zymosan administration. The test drugs significantly reduced infiltration of leukocytes into the airpouch (A), PGE₂ levels in the exudates, which are indicative of cyclooxygenase-2 activity (B), and whole blood thromboxane synthesis, which is indicative of cyclooxygenase-1 activity (C). Gastric damage was observed in the mice treated with naproxen, but not in the other groups (D). *P < 0.05 versus all other groups. ^αP < 0.05 versus the naproxen-treated group; ^ΨP < 0.05 versus the other groups treated with one of the test drugs (anova and Dunnett's Multiple Comparison test). Each group consisted of five to six mice. ATB-345, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-(5-thioxo-5H-[1,2]dithiol-3-yl)-phenyl ester; ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester; PG, prostaglandin; TXB₂, thromboxane B₂.

Figure 3

The reduction of adjuvant arthritis-associated paw swelling by orally administered naproxen and ATB-346 (both at 4 µmol·kg⁻¹ twice-daily). Freund's complete adjuvant was administered on day 0, after the initial paw volume measurement. Results are shown as the summed volumes of the two hindpaws. Treatment with the test drugs or vehicle was carried out from days 7 to 21 after adjuvant administration. Naproxen significantly reduced paw oedema at day 21 (*P < 0.05 vs. the vehicle-treated group), but not at day 14. ATB-346 significantly reduced paw oedema at days 14 and 21 (*P < 0.05 vs. the vehicle-treated group). Each group consisted of six to seven rats (data were compared using anova and Dunnett's Multiple Comparison test). ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester.

Figure 4

Oral administration of naproxen caused haemorrhagic damage in the stomach that increased in severity in a dose-dependent manner. In contrast, ATB-346 administration caused markedly less gastric damage at all doses tested; the stomach appeared normal with doses of 30 to 120 µmol·kg⁻¹). At the dose of 2740 µmol·kg⁻¹, there was significantly more damage with naproxen than with ATB-346 (P < 0.001; Student's t-test). Each group consisted of at least five rats. ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester.

Figure 5

ATB-346 spares the stomach of injury in circumstances in which gastric mucosal defence is impaired. (A) Administration of aspirin at a dose that itself did not cause detectable gastric damage (10 mg·kg⁻¹) resulted in a significant (*P < 0.05; anova and Dunnett's Multiple Comparison test) increase in the severity of gastric damage when co-administered with naproxen (60 µmol·kg⁻¹) or celecoxib (30 µmol·kg⁻¹), but not with ATB-346 (60 µmol·kg⁻¹). (B) Abalation of sensory afferent nerves by neonatal capsaicin treatment resulted in a significant increase (*P < 0.05; Student's t-test) in the severity of naproxen-induced gastric damage, but an equimolar dose (60 µmol·kg⁻¹) of ATB-346 did not cause significant gastric damage in capsaicin-ablated rats. Each group consisted of five to six rats. ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester; Veh, vehicle.

Figure 6

ATB-346 protected the stomach from injury in circumstances in which gastric mucosal defence was impaired. Administration of L-NAME (15 mg·kg⁻¹; A), an inhibitor of nitric oxide synthase, or BCA (50 mg·kg⁻¹; B), an inhibitor of hydrogen sulphide synthesis, significantly increased the severity of gastric damage induced by naproxen (shaded columns; 60 µmol·kg⁻¹; *P < 0.05; Student's t-test), but significant damage was not observed when the rats were treated with an equimolar dose of ATB-346 (solid columns). Pretreatment with glibenclamide (10 mg·kg⁻¹; C), an antagonist of ATP-sensitive potassium (K_IR6.x) channels, also significantly increased the severity of naproxen-induced gastric damage, but rats cotreated with ATB-346 did not develop significant gastric injury. Each group consisted of five to six rats. ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester; BCA, β-cyanoalanine; L-NAME, N^G-nitro-L-arginine methylester; Veh, vehicle.

Figure 7

ATB-346 protected the small intestine from damage and bleeding. Twice-daily oral administration of naproxen (90 µmol·kg⁻¹) for 5 days resulted in extensive ulceration and bleeding in the small intestine (A). The latter was evident upon examination of the intestine at the end of the study, and the severity of the bleeding is further exemplified by the significant decrease in haematocrit in naproxen-treated rats (B). Treatment with an equimolar dose of ATB-346 resulted in markedly less intestinal damage (*P < 0.05; anova and Mann–Whitney test) and did not significantly affect the haematocrit, as compared with that measured before drug administration. Each group consisted of six rats. ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester.

Figure 8

Effects of naproxen, ATB-346 and celecoxib on healing of gastric ulcers in mice. Mice were treated twice-daily with one of the test drugs, or vehicle (orally), from days 3 to 7 post ulcer induction. The extent of healing with each treatment is shown. While naproxen (60 µmol·kg⁻¹) and celecoxib (30 µmol·kg⁻¹) significantly impaired ulcer healing as compared with the vehicle-treated group, ATB-346 (60 µmol·kg⁻¹) significantly enhanced ulcer healing. Each group consisted of six to seven mice (*P < 0.05 vs. the vehicle-treated group; anova and Dunnett's Multiple Comparison test). ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester.

Figure 9

Unlike conventional NSAIDs (diclofenac and naproxen at 60 and 90 µmol·kg⁻¹ respectively), equimolar doses of hydrogen sulphide-releasing derivatives of these drugs (ATB-337 and ATB-346 respectively) did not significantly elevate mean arterial blood pressure in rats with hypertension induced by addition of L-NAME to the drinking water (400 mg·L⁻¹). The basal blood pressure (after 2 weeks of consumption of the supplemented drinking water) was 153 ± 5 mmHg (n= 32), with no significant differences in basal blood pressure among the treatment groups. Results are shown as the change in mean arterial blood pressure over the course of 1 h after an i.p. bolus injection of the test drug or vehicle. Each group consisted of five to seven rats (anova and Dunnett's Multiple Comparison test). ATB-337, (2-(2,6-dichloro-phenylamino)-phenyl)-acetic acid 4-(3-thioxo-3H-[1,2]dithiol-4-yl)-phenyl ester; ATB-346, 2-(6-methoxy-napthalen-2-yl)-propionic acid 4-thiocarbamoyl-phenyl ester; L-NAME, N^G-nitro-L-arginine methylester; NSAIDs, non-steroidal anti-inflammatory drugs.