Vol. 60, No. 6
INFECTION AND IMMUNITY, June 1992, p. 2432-2437
0019-9567/92/062432-06$02.00/0
Copyright C 1992, American Society for Microbiology
Protective Effects of a Leukotriene Inhibitor and a Leukotriene Antagonist on Endotoxin-Induced Mortality in Carrageenan-Pretreated Mice MASANORI OGATA,1* TAKAHIRO MATSUMOTO,' MASAYUKI KAMOCHI,1 SHIN-ICHI YOSHIDA,2 YASUO MIZUGUCHI,2 AND AKIO SHIGEMATSU' Department of Anesthesiology' and Department of Microbiology, 2 School of Medicine, University of Occupational and Environmental Health, Kitakyushu 807, Japan Received 18 November 1991/Accepted 17 March 1992
The leukotrienes and tumor necrosis factor (TNF) play an important role in the pathophysiology of septic shock, in which hypotension, leukopenia, thrombocytopenia, and hemoconcentration are observed. This study was performed to examine the effects of a 5-lipoxygenase inhibitor (AA-861), a selective leukotriene receptor antagonist (ONO-1078), and a cyclooxygenase inhibitor (indomethacin) on endotoxin-induced mortality and TNF production in mice. Mice were injected intraperitoneally with carrageenan (5 mg per mouse), which we previously reported as an effective priming agent for lipopolysaccharide (LPS)-induced TNF production and mortality (M. Ogata, S. Yoshida, M. Kamochi, A. Shigematsu, and Y. Mizuguchi, Infect. Immun. 59:679-683, 1991). The indicated doses of AA-861, ONO-1078, indomethacin, or controls were administrated subcutaneously 30 min before LPS (50 ,ig per mouse) provocation. The mortality of mice was significantly decreased by pretreatment with AA-861 (P < 0.001) or ONO-1078 (P < 0.01) but not by pretreatment with indomethacin. The 50%o lethal dose of LPS in the mice treated with dimethyl sulfoxide or ethanol was 32 or 33 ,g, respectively, and it increased to 83 ,Ig with AA-861 or 59 ,ug with ONO-1078, respectively. Neither AA-861 nor ONO-1078 suppressed LPS-induced TNF production in sera. Treatment with AA-861 significantly decreased the leukopenia and thrombocytopenia, and ONO-1078 significantly decreased the hemoconcentration and thrombocytopenia. The role of endogenous TNF was also examined in the carrageenan-pretreated mice. Treatment with 2 x IV U of rabbit anti TNF-o antibody intravenously 2 h before LPS challenge significantly suppressed the LPS-induced TNF activity and decreased the mortality. Therefore, both leukotrienes and TNF play important roles in endotoxin-induced shock and mortality.
peritoneal polymorphonuclear leukocytes (36). This compound strongly inhibited allergic bronchoconstriction in guinea pigs and moderately reduced CAR-induced paw edema and pleurisy in rats (3). 4-Oxo-8-[p-(4-phenylbutyloxy)benzoylamino]-2-tetrazol-(5-yl)-4H-1-benzopyran hemihydrate (ONO-1078) is an LTC4D4 receptor antagonist that has been shown to selectively antagonize LT-induced bronchoconstriction in guinea pig and human bronchial smooth muscles (23). The effect was 10 to 100 times stronger than that of the LT antagonist FPL-55712 (1). We therefore investigated the effect of AA-861, ONO-1078, and indomethacin on LPS-induced TNF production and mortality in CAR-pretreated mice. The present study shows that AA-861 and ONO-1078 cannot reduce the plasma TNF levels but can significantly reduce LPS-induced mortality of CAR-pretreated mice. (Part of this study was presented at the Second International Congress on the Immune Consequences of Trauma, Shock and Sepsis Mechanisms and Therapeutic Approaches, Munich, Germany, 1991.)
The tumor necrosis factor (TNF) is an important cytokine that mediates endotoxin shock and causes multiple organ damage (5). It has been reported that patients with peritonitis or burns are apt to go into septic shock and to develop multiple organ failure (13). We previously reported that in mice pretreated intraperitoneally with carrageenan (CAR), a sulfated polygalactose, even a low dose of endotoxin (50 ,ug per mouse) caused enormous TNF production in serum and death (24). It was supposed that inflammation played an important role in the enhancement of lipopolysaccharide (LPS)-induced TNF production and mortality caused by CAR pretreatment, because CAR is used as an inflammatory agent as well as a reticuloendothelial system blocker. These findings suggest that the inflammation plays an important role in the development of low-dose endotoxin-induced mortality and multiple organ failure and that CAR-pretreated mice provide a useful model to analyze the pathophysiological mechanism. There is increasing recent evidence that inflammatory mediators such as leukotrienes (LTs) (8-10, 33) and cyclooxygenase products (2, 20) are involved in the pathophysiology of endotoxemia. It was reported that LT inhibitors suppressed LPS-induced TNF production and mortality in galactosamine-treated mice (30). 2-(12-Hydroxydodeca-5,10-dinyl)-3,5,6-trimethyl-1,4-benzoquinone (AA-861) is a selective 5-lipoxygenase inhibitor that was reported to inhibit 5-lipoxygenase from guinea pig *
MATERIALS AND METHODS Animals. Male ddY mice were obtained from the Seiwa Experimental Animal Co., Oita, Japan. All mice used were 7 to 8 weeks of age. Mice were housed in groups of 10 and were allowed food and water ad libitum. Reagents. Phenol-extracted Escherichia coli (0127:B8) LPS was purchased from Difco Laboratories, Detroit, Mich. Iota-carrageenan (lot 59C-0328) and indomethacin (lot 1l1F-
Corresponding author. 2432
VOL. 60, 1992
0099) were purchased from Sigma, St. Louis, Mo. Both LPS and CAR were dissolved in pyrogen-free physiological saline (Otsuka Pharmaceutical Co., Naruto, Japan). The CAR solution was autoclaved at 121°C for 15 min before use. AA-861 was a gift from Takeda Pharmaceutical Co., Osaka, Japan. ONO-1078 was kindly given by Ono Pharmaceutical Co., Osaka, Japan. AA-861, indomethacin, and ONO-1078 were dissolved in 10% dimethyl sulfoxide (DMSO), saline, and 10% ethanol, respectively, at the concentrations indicated below. DMSO, saline, and ethanol were used alone as controls. Rabbit anti TNF-a polyclonal antibody (IP-400; neutralizing activity, 106 U/ml) and recombinant murine TNF-ao were purchased from Genzyme, Cambridge, Mass. Induction of endotoxin shock and treatment with drugs. Endotoxin shock was induced in mice as previously described (24). In brief, CAR (5 mg in 0.5 ml of physiological saline) was injected intraperitoneally (i.p.) as a priming agent 24 h before LPS challenge. LPS (50 p,g in 0.5 ml of physiological saline) was injected intravenously into the tail vein as an inducing agent. The indicated doses of AA-861, ONO-1078, saline, DMSO, or ethanol were administrated subcutaneously (s.c.) in a volume of 1 ml into the backs of mice 30 min before the LPS provocation. Both drugs were injected s.c., because CAR i.p. pretreatment caused peritonitis. To examine the role of endogenous TNF in CARpretreated mice, 2 x 105 U of rabbit anti-TNF-a antibody or normal serum of rabbit in 0.2 ml was injected intravenously (i.v.) before the LPS challenge. TNF assay. At intervals after LPS injection, the blood was collected by cardiac puncture and kept at room temperature for clotting. The serum samples were stored at -80°C until they were used for the TNF assay. TNF activity was assayed by determining cytotoxicity to L929 cells colorimetrically as previously reported (27). The titer of TNF, expressed in units per milliliter, is the reciprocal of the dilution necessary to cause lysis of 50% of the cells. In our assay, 1 U/ml was equivalent to 0.63 pg of recombinant murine TNF-a per ml. Hematological studies. CAR (5 mg per mouse) was injected i.p., and mice were challenged with LPS (50 p.g per mouse) i.v. 24 h later. Hematocrit, leukocyte, and platelet measurements were made 30 min and 4 h after this challenge. The blood was collected by cardiac puncture into a heparincoated syringe. The count of leukocytes and platelets was determined by using an automatic analyzer (Sysmex Platelet Counter PL-110; Toa Medical Electrical Co., Kobe, Japan). Hematocrits were determined after centrifugation at 12,000 x g for 5 min by standard methods. The hematological measurements of various experimental groups were compared 30 min and 4 h after LPS injection. After LPS or saline injection, all mice received no water and food for the rest of this experiment. Mortality. The mortality rates were determined by counting the numbers of dead mice at 6, 18, 24, 36, 48, and 72 h after the administration of endotoxin; 10 mice were used in each drug treatment group (Fig. 1). The cumulative percent mortality in mice receiving each dose of LPS was determined by counting the dead mice at 72 h after LPS injection (Fig. 2). Statistics. A comparison of the physiological variables between groups was carried out with a two-factor repeatedmeasures analysis of variance. When indicated, further analysis was performed with the unpaired t test. The survival curves were analyzed by the Kaplan Meier statistical method. The effects of various treatments on endotoxininduced lethality were analyzed by using the chi-square test.
LEUKOTRIENES IN LPS-INDUCED MORTALITY
2433
TABLE 1. Protective effect of AA-861 or ONO-1078 on the LPS (50 p.g)-induced mortality of CAR-pretreated mice"
AA-861
Treatment"
Dose
(s.c.)
(mmol/kg) 20 10 5
Mortality rate (no. dead/total) Expt Expt Total 1 2
0/10C
1/10C
1/20d
6/20 8/20 16/20 AA-861 control (DMSO) 1/10c 4/20c 40 ONO-1078 6/20 2/10 20 8/20 4/10 10 8/10 16/20 ONO-1078 control (ethanol) 9/10 ND 40 Indomethacine 8/10 ND Indomethacine control (saline) a CAR (5 mg per mouse) was injected i.p. 24 h before LPS (50 pg per 3/10 4/10 8/10 3/10 4/10 4/10 8/10 9/10 8/10
3/10 4/10 8/10
mouse) was injected i.v. ND, not done. b Each treatment was done 30 min before the LPS administration. c P < 0.05 versus the control. d P < 0.01 versus the control.
A significant difference was presumed at a probability value of
INFECTION AND IMMUNITY, June 1992, p. 2432-2437
0019-9567/92/062432-06$02.00/0
Copyright C 1992, American Society for Microbiology
Protective Effects of a Leukotriene Inhibitor and a Leukotriene Antagonist on Endotoxin-Induced Mortality in Carrageenan-Pretreated Mice MASANORI OGATA,1* TAKAHIRO MATSUMOTO,' MASAYUKI KAMOCHI,1 SHIN-ICHI YOSHIDA,2 YASUO MIZUGUCHI,2 AND AKIO SHIGEMATSU' Department of Anesthesiology' and Department of Microbiology, 2 School of Medicine, University of Occupational and Environmental Health, Kitakyushu 807, Japan Received 18 November 1991/Accepted 17 March 1992
The leukotrienes and tumor necrosis factor (TNF) play an important role in the pathophysiology of septic shock, in which hypotension, leukopenia, thrombocytopenia, and hemoconcentration are observed. This study was performed to examine the effects of a 5-lipoxygenase inhibitor (AA-861), a selective leukotriene receptor antagonist (ONO-1078), and a cyclooxygenase inhibitor (indomethacin) on endotoxin-induced mortality and TNF production in mice. Mice were injected intraperitoneally with carrageenan (5 mg per mouse), which we previously reported as an effective priming agent for lipopolysaccharide (LPS)-induced TNF production and mortality (M. Ogata, S. Yoshida, M. Kamochi, A. Shigematsu, and Y. Mizuguchi, Infect. Immun. 59:679-683, 1991). The indicated doses of AA-861, ONO-1078, indomethacin, or controls were administrated subcutaneously 30 min before LPS (50 ,ig per mouse) provocation. The mortality of mice was significantly decreased by pretreatment with AA-861 (P < 0.001) or ONO-1078 (P < 0.01) but not by pretreatment with indomethacin. The 50%o lethal dose of LPS in the mice treated with dimethyl sulfoxide or ethanol was 32 or 33 ,g, respectively, and it increased to 83 ,Ig with AA-861 or 59 ,ug with ONO-1078, respectively. Neither AA-861 nor ONO-1078 suppressed LPS-induced TNF production in sera. Treatment with AA-861 significantly decreased the leukopenia and thrombocytopenia, and ONO-1078 significantly decreased the hemoconcentration and thrombocytopenia. The role of endogenous TNF was also examined in the carrageenan-pretreated mice. Treatment with 2 x IV U of rabbit anti TNF-o antibody intravenously 2 h before LPS challenge significantly suppressed the LPS-induced TNF activity and decreased the mortality. Therefore, both leukotrienes and TNF play important roles in endotoxin-induced shock and mortality.
peritoneal polymorphonuclear leukocytes (36). This compound strongly inhibited allergic bronchoconstriction in guinea pigs and moderately reduced CAR-induced paw edema and pleurisy in rats (3). 4-Oxo-8-[p-(4-phenylbutyloxy)benzoylamino]-2-tetrazol-(5-yl)-4H-1-benzopyran hemihydrate (ONO-1078) is an LTC4D4 receptor antagonist that has been shown to selectively antagonize LT-induced bronchoconstriction in guinea pig and human bronchial smooth muscles (23). The effect was 10 to 100 times stronger than that of the LT antagonist FPL-55712 (1). We therefore investigated the effect of AA-861, ONO-1078, and indomethacin on LPS-induced TNF production and mortality in CAR-pretreated mice. The present study shows that AA-861 and ONO-1078 cannot reduce the plasma TNF levels but can significantly reduce LPS-induced mortality of CAR-pretreated mice. (Part of this study was presented at the Second International Congress on the Immune Consequences of Trauma, Shock and Sepsis Mechanisms and Therapeutic Approaches, Munich, Germany, 1991.)
The tumor necrosis factor (TNF) is an important cytokine that mediates endotoxin shock and causes multiple organ damage (5). It has been reported that patients with peritonitis or burns are apt to go into septic shock and to develop multiple organ failure (13). We previously reported that in mice pretreated intraperitoneally with carrageenan (CAR), a sulfated polygalactose, even a low dose of endotoxin (50 ,ug per mouse) caused enormous TNF production in serum and death (24). It was supposed that inflammation played an important role in the enhancement of lipopolysaccharide (LPS)-induced TNF production and mortality caused by CAR pretreatment, because CAR is used as an inflammatory agent as well as a reticuloendothelial system blocker. These findings suggest that the inflammation plays an important role in the development of low-dose endotoxin-induced mortality and multiple organ failure and that CAR-pretreated mice provide a useful model to analyze the pathophysiological mechanism. There is increasing recent evidence that inflammatory mediators such as leukotrienes (LTs) (8-10, 33) and cyclooxygenase products (2, 20) are involved in the pathophysiology of endotoxemia. It was reported that LT inhibitors suppressed LPS-induced TNF production and mortality in galactosamine-treated mice (30). 2-(12-Hydroxydodeca-5,10-dinyl)-3,5,6-trimethyl-1,4-benzoquinone (AA-861) is a selective 5-lipoxygenase inhibitor that was reported to inhibit 5-lipoxygenase from guinea pig *
MATERIALS AND METHODS Animals. Male ddY mice were obtained from the Seiwa Experimental Animal Co., Oita, Japan. All mice used were 7 to 8 weeks of age. Mice were housed in groups of 10 and were allowed food and water ad libitum. Reagents. Phenol-extracted Escherichia coli (0127:B8) LPS was purchased from Difco Laboratories, Detroit, Mich. Iota-carrageenan (lot 59C-0328) and indomethacin (lot 1l1F-
Corresponding author. 2432
VOL. 60, 1992
0099) were purchased from Sigma, St. Louis, Mo. Both LPS and CAR were dissolved in pyrogen-free physiological saline (Otsuka Pharmaceutical Co., Naruto, Japan). The CAR solution was autoclaved at 121°C for 15 min before use. AA-861 was a gift from Takeda Pharmaceutical Co., Osaka, Japan. ONO-1078 was kindly given by Ono Pharmaceutical Co., Osaka, Japan. AA-861, indomethacin, and ONO-1078 were dissolved in 10% dimethyl sulfoxide (DMSO), saline, and 10% ethanol, respectively, at the concentrations indicated below. DMSO, saline, and ethanol were used alone as controls. Rabbit anti TNF-a polyclonal antibody (IP-400; neutralizing activity, 106 U/ml) and recombinant murine TNF-ao were purchased from Genzyme, Cambridge, Mass. Induction of endotoxin shock and treatment with drugs. Endotoxin shock was induced in mice as previously described (24). In brief, CAR (5 mg in 0.5 ml of physiological saline) was injected intraperitoneally (i.p.) as a priming agent 24 h before LPS challenge. LPS (50 p,g in 0.5 ml of physiological saline) was injected intravenously into the tail vein as an inducing agent. The indicated doses of AA-861, ONO-1078, saline, DMSO, or ethanol were administrated subcutaneously (s.c.) in a volume of 1 ml into the backs of mice 30 min before the LPS provocation. Both drugs were injected s.c., because CAR i.p. pretreatment caused peritonitis. To examine the role of endogenous TNF in CARpretreated mice, 2 x 105 U of rabbit anti-TNF-a antibody or normal serum of rabbit in 0.2 ml was injected intravenously (i.v.) before the LPS challenge. TNF assay. At intervals after LPS injection, the blood was collected by cardiac puncture and kept at room temperature for clotting. The serum samples were stored at -80°C until they were used for the TNF assay. TNF activity was assayed by determining cytotoxicity to L929 cells colorimetrically as previously reported (27). The titer of TNF, expressed in units per milliliter, is the reciprocal of the dilution necessary to cause lysis of 50% of the cells. In our assay, 1 U/ml was equivalent to 0.63 pg of recombinant murine TNF-a per ml. Hematological studies. CAR (5 mg per mouse) was injected i.p., and mice were challenged with LPS (50 p.g per mouse) i.v. 24 h later. Hematocrit, leukocyte, and platelet measurements were made 30 min and 4 h after this challenge. The blood was collected by cardiac puncture into a heparincoated syringe. The count of leukocytes and platelets was determined by using an automatic analyzer (Sysmex Platelet Counter PL-110; Toa Medical Electrical Co., Kobe, Japan). Hematocrits were determined after centrifugation at 12,000 x g for 5 min by standard methods. The hematological measurements of various experimental groups were compared 30 min and 4 h after LPS injection. After LPS or saline injection, all mice received no water and food for the rest of this experiment. Mortality. The mortality rates were determined by counting the numbers of dead mice at 6, 18, 24, 36, 48, and 72 h after the administration of endotoxin; 10 mice were used in each drug treatment group (Fig. 1). The cumulative percent mortality in mice receiving each dose of LPS was determined by counting the dead mice at 72 h after LPS injection (Fig. 2). Statistics. A comparison of the physiological variables between groups was carried out with a two-factor repeatedmeasures analysis of variance. When indicated, further analysis was performed with the unpaired t test. The survival curves were analyzed by the Kaplan Meier statistical method. The effects of various treatments on endotoxininduced lethality were analyzed by using the chi-square test.
LEUKOTRIENES IN LPS-INDUCED MORTALITY
2433
TABLE 1. Protective effect of AA-861 or ONO-1078 on the LPS (50 p.g)-induced mortality of CAR-pretreated mice"
AA-861
Treatment"
Dose
(s.c.)
(mmol/kg) 20 10 5
Mortality rate (no. dead/total) Expt Expt Total 1 2
0/10C
1/10C
1/20d
6/20 8/20 16/20 AA-861 control (DMSO) 1/10c 4/20c 40 ONO-1078 6/20 2/10 20 8/20 4/10 10 8/10 16/20 ONO-1078 control (ethanol) 9/10 ND 40 Indomethacine 8/10 ND Indomethacine control (saline) a CAR (5 mg per mouse) was injected i.p. 24 h before LPS (50 pg per 3/10 4/10 8/10 3/10 4/10 4/10 8/10 9/10 8/10
3/10 4/10 8/10
mouse) was injected i.v. ND, not done. b Each treatment was done 30 min before the LPS administration. c P < 0.05 versus the control. d P < 0.01 versus the control.
A significant difference was presumed at a probability value of
