Piroxicam-copper complexes: Inhibition of polymorphonuclear leukocyte migration toPseudomonas aeruginosa chemotactinsin vivo and superoxide dismutase-like activityin vitro

Agents Actions 38 ( 1993)

0065-4299/93/040196-06 $1.50+ 0.20/0 9 1993 Birkhfiuser Verlag, Basel

Piroxicam-copper complexes: Inhibition of polymorphonuclear leukocyte migration to Pseudomonas aeruginosa chemotactins in vivo and superoxide dismutase-like activity in vitro D. O. Sordelli 1-2, p. A. Font~m 1 and C. R. Amura ~ Departamento de Microbiologia, Parasitologia e Immunologia,Facultad de Medicina, Universidad de Buenos Aires, Argentina 2 Department of Pediatrics, Georgetown University School of Medicine, Washington, DC 20007

Abstract

Piroxicam-copper (Cu 2 +) complexes, formed spontaneously by mixing solutions of piroxicam and CuSO4 (1:1 Cu 2 + :piroxicam), inhibited the superoxide anion-catalyzed reduction of ferricytochrome C in a doserelated fashion. Addition of ethylenediaminetetraacetate to the mixture decreased in a dose-related manner the superoxide dismutase (SOD)-like activity of piroxicam Cu 2+. Piroxicam alone (10 5M, final concentration) did not display SODqike activity but 10 5 M Cu 2 + exhibited significant activity, similar to that of piroxicam Cu 2 +. Intraperitoneal treatment of mice with either 0.64 mg/kg piroxicam or its Cu 2 + complexes (0.64mg/kg piroxicam+0.12mg/kg Cu 2+) was equally effective in diminishing both the migration of polymorphonuclear leukocytes (PMNL) to the airways and the content of myeloperoxidase activity in the lungs, induced by aerosol challenge with Pseudomonas aeruginosa peptide chemotactins. Therefore, piroxicam-Cu 2+ complexes may provide both the anti-inflammatory activity of piroxicam plus the SOD-like activity of Cu 2 +. Introduction Activated polymorphonuclear leukocytes (PMNL) generate reactive oxygen intermediates, e.g. free radicals superoxide and hydroxyl, which play a dual role in host defenses against infection. On the one hand, they participate at different stages of bacterial inactivation and, on the other hand, they can mediate tissue damage [1, 2]. There is little doubt that the numerous P M N L migrating from circulation to the site of infection are associated with tissue damage [2, 3]. Treatment with nonAddress for correspondence: Dr. D. O. Sordelli, Dept. Microbiology FCM-UBA, Paraguay 2155 P-12, 1121 Buenos Aires, Argentina.

steroidal anti-inflammatory agents (NSAIA) aimed at preventing migration of excessive numbers of P M N L and consequent lung tissue damage during P. aeruqinosa lung infection has been suggested [4]. Studies performed in animal models revealed that the recruitment of P M N L to the lungs during acute P. aeruginosa pneumonia can be modulated by the NSA1As ibuprofen [4] and piroxicam [5, 6]. Moreover, mortality induced by lethal P. aeruginosa challenge was prevented in a dose-related fashion by piroxicam treatment [5]. The mechanisms involved in protection were attributed primarily to the inhibition of P M N L recruitment to the lungs induced by the NSAIA, although other less understood mechanisms related to antioxidant activity

Agents Actions38 (1993)

197

may be involved [7, 8]. Using pulse radiolysis to generate superoxide, it was shown that Cu 2+complexes of certain small molecules, including NSAIA, are able to dismutate superoxide, thus inactivating this important tissue damaging intermediate [9]. This investigation was aimed at evaluating whether piroxicam-Cu 2+ complexes can dismutate superoxide and whether these complexes retained the anti-inflammatory activity of piroxicam.

Materials and methods Bacterial chemotactins

The ferricytochrome C reduction method was utilized to determine the capacity of piroxicam-Cu 2+ to catalyze dismutation of superoxide. The superoxide anion was generated by oxidation ofxanthine by xanthine oxidase in glycine buffer, pH 9.8 [t2]. Inhibition of superoxide anion-catalyzed reduction of ferricytochrome C was evaluated by measuring the decline in As~ onm increase rate using a doublebeam spectrophotometer, and the results were recorded as units of SOD-like activity.

Animal experiments

Bacterial chemotactins were obtained by culturing P. aeruginosa Fisher immunotype 1 in VogelBonner defined medium [10]. Cultures were stopped in mid-log phase by centrifugation at 10000 x g, 4 ~ for 20 min and supernatants were sterilized by filtration through 0.2/am-pore filters. For chemotaxis assays, culture supernatants were ultrafiltered through a Centricon 3 membrane filter (Amicon Corp., Lexington, MA) at 2500 x g, 4 ~ The fraction with molecular size less than 3 kDa was reconstituted to the original volume with fresh culture medium. Chemotactic activity of P. aeruginosa culture supernatants was due almost exclusively to low molecular weight peptides with physical, chemical and functional characteristics similar, if not identical, to those exhibited by N-formyl-methionyl peptides [10].

Piroxicam and piroxicam

Superoxide dismutase-like activity of piroxicam and piroxicam-Cu 2+

Cu 2 +

complexes

Piroxicam was obtained from Pfizer (Buenos Aires, Argentina). The drug was dissolved in dimethylsulfoxide (Sigma Chemical Co., St. Louis, MO, USA), diluted in 0.1 N NaHCO3 and further diluted in Hanks balanced salt solution without calcium and magnesium. Piroxicam Cu 2+ complexes were formed by mixing equal volumes of 2• 10 4M piroxicam and 2• 10 *M Cu 2+ (CuSO4), to obtain the desired final concentrations in the SOD-like activity assay [11]. Piroxicam Cu 2 + complexes were prepared for parenteral administration to obtain doses of 0.64 mg/kg (piroxicam) and 0.12mg/kg (Cu2+). Intraperitoneal treatment with piroxicam (0.64 mg/kg) or its Cu 2 + complexes started five days before aerosol challenge.

The efficacy of parenteral treatment with piroxicam Cu 2+ was evaluated in a mouse model of aerosol challenge [13]. Briefly, 8-week-old Swiss mice were exposed to the chemotactin aerosol in a 1.81 volume chamber, where only the animal's nose was exposed to the chemotactin mist for 45 min, at 22-.-24 ~ Five ml volumes of sterile P. aeruginosa culture supernatant were consumed in each experiment. The actual chemotactin concentration attained in pulmonary fluids is unknown, but previous data from our laboratory revealed that aerosolization of culture supernatants of P. aeruginosa induces migration of P M N L into the mouse airways [13]. The effect of piroxicam Cu 2+ on P M N L migration was assessed by determining the total number of P M N L retrieved by lung lavage 4 h after aerosol challenge, according to a standard procedure described previously [14]. To estimate the total number of P M N L associated with the lungs, i.e. those which migrated into the airways, those trapped in the interstitium and those attached to the vasculature, the myeloperoxidase activity was determined in lung homogenate supernatants [15]. Briefly, mice were given a pentobarbital overdose and, before death, the lungs were perfused with saline to eliminate hemoglobin. The lungs were removed and homogenized in 5 ml sterile, icecold distilled water. The activity of myeloperoxidase in lung homogenate supernatant was determined spectrophotometrically by using 3,3', 5,5'tetramethylbenzidine as the enzyme substrate [16]. Anti-inflammation induced by the different compounds tested was monitored by means of the carrageenan paw test. Briefly, mice were inoculated subcutaneously in the left foot pad with 10 ~tl 1% carrageenan (Sigma). Mice were treated with

198

Agents Actions 38 (1993)

piroxicam Cu 2+, piroxicam or CuSO4, and 3h after carrageenan injection, they were sacrificed by pentobarbital overdose. Both hind paws were then carefully excised with a scalpel. The cut was made immediately below tibia and fibula, on the ankle joint, and each paw was weighed separately. Edema was measured as the difference in each mouse's right and left paw weight [17]. Statistical analysis Data with normal distribution are expressed as the arithmetic mean + SEM, and were c o m p a r e d statistically by the Student's t-test. Data with nonnormal distribution are expressed as medians (range), and were compared statistically by rank sum test. Calculations were made using the Epistat statistical package (T. Gustafson, Round Rock, TX) in a personal computer. P values lower than 0.05 were considered statistically significant. Results

Experiments performed in vitro demonstrated that piroxicam Cu 2 + complexes (1:1 Cu 2 + : piroxicam) inhibited the superoxide anion-catalyzed reduction of ferricytochrome C in dose-related fashion (Fig. 1). The SOD-like activity of p i r o x i c a m - C u 2+ was decreased, also in a dose-related fashion, by addition of ethylenediaminetetraacetate to the reaction

m

E 0.12.

i-- 0.080.100.060.04-

0.02O u3 O( 0

~ 0:5

medium (data not shown). Piroxicam alone (10-5 M, final concentration) did not exhibit SODlike activity but 10- s M Cu 2 + inhibited the reduction of ferricytochrome C at rates similar to those of piroxicam Cu 2 + (Fig. 2). The effect ofpiroxicam, Cu 2 + and p i r o x i c a m - C u 2+ on edema formation was assessed by injecting I % carrageenan into the mouse right footpad and by determining the differential weight between right and left hind paws. Intraperitoneal treatment with equivalent doses of piroxicam, Cu 2+ (CuSO4) or piroxicam-Cu 2 + produced significant and comparable decrease in paw edema in the mouse (Table t). Mice were challenged with an aerosol containing P. aeruginosa culture supernatant. Treatment with either piroxicam or p i r o x i c a m - C u 2+ did not increase the negligible number of P M N L in lung lavage fluid obtained from healthy, unchallenged mice (Table 2). Aerosol challenge with bacterial chemotactins induced migration of significant numbers of P M N L into the lower airways. The number of P M N L recovered from the lower airways of mice challenged and treated with either piroxicam or piroxicam Cu 2+ was significantly lower than that from untreated mice (Table 2). Inhibition of P M N L migration to the lungs after treatment with piroxicam Cu 2 + was confirmed by determining the myeloperoxidase activity in supernatants of lung homogenates. The activity of myeloperoxidase was decreased by 60% in mice challenged with the chemotactins and treated with piroxicam-Cu 2+, compared to that from challenged, untreated controls.

O

E 0.12- r--i

CONTROL B B PIROXICAM k'k-'k-R PIROXICAM-Cu 2+

0.08. 1:o

1:5

2:0

PIROXICAM-Cu 2 + COMPLEX CONCENTRATION

(x 10-5M) Figure 1 Dose-response curve of the SOD-like activity and the piroxicam-Cu2+ (1:1 Cu2+piroxicam) concentration. Each point depicts the arithmetic mean of triplicates from a representative experiment. The experiment was repeated 4 times with similar results.