Shuhei Sakaguchi

Roles of selenium in endotoxin-induced lipid peroxidation in the rats liver and in nitric oxide production in J774A.1 cells

We examined the role of selenium (Se) in the mechanism of oxidative stress caused by endotoxin by feeding rats deficient a diet in this element. In rats fed the Se-deficient diet (concentration of Se, less than 0.027 microg g(-1)) for 10 weeks, Se level and glutathione peroxidase (GSH-Px) activity in the liver were about 47 and 43% lower, respectively, than those in rats fed a Se-adequate diet (Se, 0.2 microg g(-1)). Rat fed the Se-deficient diet and given endotoxin (6 mg kg(-1), i.p.) showed a mortality rates of about 43% at 18 h. Nevertheless, no lethality was observed with endotoxin (4 mg kg(-1), i.p.) challenge. Levels of serum lactate dehydrogenase and acid phosphatase leakage were significantly higher in Se-deficient rats than those in Se-adequate diet 18 h after endotoxin (4 mg kg(-1), i. p.) challenge. Superoxide anion generation and lipid peroxide formation in the liver of Se-deficient rat were markedly increased 18 h after endotoxin (4 mg kg(-1), i.p.) injection compared with those in the endotoxin/Se-adequate diet group, whereas non-protein sulfhydryl level in the liver after administration of endotoxin to Se-deficient rats was lower than that in Se-adequate rats treated with endotoxin. We investigated whether Se can suppress nitric oxide (NO) generation and cytotoxicity in endotoxin-treated J774A.1 cells. Treatment with Se (10(-6) M) markedly inhibited endotoxin (0.1 microg ml(-1))-induced NO production in J774A.1 cells. Se induced an increased activity of GSH-Px in cells after 24 h of incubation, suggesting that the preventive effect of Se on NO production in endotoxemia is due to the induction of Se-GSH-Px activity. However, Se did not affect endotoxin-induced cytotoxicity in J774A.1 cells. These findings suggested that the oxidative stress caused by endotoxin may be due, at least in part, to changes in Se regulation during endotoxemia.

Effect of Calcium Ion on Lipid Peroxide Formation in Endotoxemic Mice

The present study was conducted to determine the possible role of intracellular Ca2+ in lipid peroxide formation in endotoxin‐poisoned mice. Leakages of LDH isozyme and acid phosphatase in serum of mice fed a Ca2+‐deficient diet were remarkably increased after administration of 200 μg of endotoxin compared to that in endotoxin‐nontreated Ca2+‐deficient mice. Superoxide anion generation in liver of Ca2+‐deficient mice and in mice fed a normal diet greatly increased after endotoxin administration. On the contrary, after endotoxin injection there was scarcely any difference in SOD activity of liver of Ca2+‐deficient mice as compared to that in endotoxin‐nontreated Ca2+‐deficient mice. In spite of an increase of superoxide anion generation there was little or no effect of endotoxin administration on lipid peroxide formation in mice given a Ca2+‐deficient diet. In the mice treated with a Ca2+‐deficient diet, free radical scavenger levels (α‐tocopherol and nonprotein sulfhydryl) in liver tissue after endotoxin injection were markedly decreased compared to those in Ca2+‐deficient diet alone. Mice fed a normal diet exhibited a significant decrease of lipid peroxide level in liver by injection of endotoxin together with verapamil (10 mg/kg, s.c.). When mice fed a normal diet were injected with endotoxin, the state 3 respiratory activity showed a 49% decrease, and respiratory control ratio (RCR) of endotoxemic mice liver mitochondria was 38% lower than normal liver mitochondria. No difference could be observed in levels of state 3 and RCR between the mice given verapamil plus endotoxin and the normal mice. These findings suggest the possibility that Ca2+ may participate in the free radical formation in the liver during endotoxemia and also that Ca2+ may play an important role in the damage of liver mitochondrial function in endotoxemic mice.

Preventive Effects of a Chinese Herb Medicine (Sho-saiko-to) against Lethality after Recombinant Human Tumor Necrosis Factor Administration in Mice

We observed the effects of a Chinese herb medicine Sho‐saiko‐to on the lethal and antitumor activities of recombinant human tumor necrosis factor (rhTNF) administered in mice. Sho‐saiko‐to was noted to protect the rhTNF‐induced lethality in galactosamine‐hypersensitized mice, and also Sho‐saiko‐to pretreated mice was protected against the decrease of rectal temperature after rhTNF administration. On the other hand, there was a remarkable enhancement of antitumor activity of rhTNF by Sho‐saiko‐to pretreatment. These results suggest that Sho‐saiko‐to drug may protect mice from severe shock syndrome induced by rhTNF.

Effect of nitric oxide synthase inhibitors on lipid peroxide formation in liver caused by endotoxin challenge.

This study investigated the effect of nitric oxide on lipid peroxide formation during endotoxaemia. Nitric oxide synthase inhibitors N(G)-monomethyl-L-arginine acetate (L-NMMA, 20 mg/kg, intravenously), N(G)-nitro-L-arginine-methyl ester (L-NAME, 10 mg/kg, intravenously), and N(G)-nitro-L-arginine (L-NA, 10 mg/kg, intravenously), and a relatively selective inducible nitric oxide synthase inhibitor aminoguanidine (10 mg/kg, intravenously), did not protect against endotoxin-induced death of mice. Superoxide dismutase activity in liver 18 hr after administration of endotoxin (6 mg/kg, intraperitoneally) to L-arginine analogues (L-NMMA, L-NAME, L-NA)-treated mice was lower than in mice treated with endotoxin alone, whereas the administration of L-arginine analogues increased xanthine oxidase activity in the livers of endotoxin-injected mice compared with mice treated with endotoxin alone. In mice treated with L-arginine analogues and aminoguanidine, the levels of non-protein sulfhydryl and lipid peroxide in liver 18 hr after endotoxin injection did not show significant differences from mice treated with endotoxin alone. L-Arginine analogues and aminoguanidine had little effect on lipid peroxide formation in liver caused by endotoxin. Treatment with aminoguanidine (300 microM) significantly inhibited endotoxin-induced intracellular peroxide in J774A.1 cells, however, aminoguanidine did not affect endotoxin-induced cytotoxicity in J774A.1 cells. Our results clearly demonstrate that treatment with catalase (10 microg/ml), D-mannitol (10 mM), or superoxide dismutase (100 U/ml), has little or no effect on nitric oxide production by endotoxin (1 microg/ml)-activated J774A.1 cells. These findings suggest that nitric oxide is not crucial for lipid peroxide formation during endotoxaemia. Therefore, it is unlikely that nitric oxide plays a significant role in liver injury caused by free radical generation in endotoxaemia.

Modification of tumor necrosis factor-induced acute toxicity D-galactosamine challenge by polymyxin B, an anti-endotoxin.

Polymyxin B (PMB), an antibiotic with anti-endotoxin activity, was used to examine the participation of endogenously produced endotoxin in the enhancement of recombinant human tumor necrosis factor (rhTNF)-induced toxicity in D-galactosamine (GalN)-sensitized mice. GalN-sensitized mice (700 mg/kg, intraperitoneally (i.p.)) injected together with rhTNF (1x10(4) U/mouse, intravenously (i.v.)) exhibited severe symptoms, with 100% mortality at 18 h. However, mice pretreated with PMB (20 mg/kg, i.p.) showed protection against the rhTNF-induced lethality following GalN sensitization. Little or no effects were observed on alanine aminotransferase (ALT) activity or lactate dehydrogenase (LDH) isozyme leakage in serum in mice 7 h after administration of rhTNF alone. Administration of rhTNF to GalN-sensitized mice resulted in marked increases in ALT activity and LDH isozyme leakage relative to those in mice treated with rhTNF alone. In mice pretreated with PMB, the levels of ALT and LDH isozyme leakage 7 h after rhTNF/GalN injection were significant decreased as compared with those in mice treated with rhTNF/GalN. Similarly, injection of PMB markedly decreased lipid peroxide formation in the liver of the GalN-sensitized mice treated with rhTNF. The injection of a low endotoxin dose (0.1 mg/kg, i.p.) markedly increased the lethality in mice treated with rhTNF (5x10(3) U/mouse, i.v.) and GalN, and these animals showed 100% mortality at 8 h. These findings suggested that the extent of TNF-induced toxicity caused by GalN administration may be a result of synergism between TNF and gut-derived endotoxin. It is likely that endogenously produced endotoxin play a significant role in rhTNF/GalN-hypersensitized mice.

Apoptosis induced by doxorubicin and cinchonine in P388 multidrug-resistant cells.

Acquired drug resistance is a major factor in the failure of doxorubicin‐based cancer chemotherapy. We determined the ability of cinchonine to reverse doxorubicin drug resistance in a doxorubicin‐resistant leukaemia cell line (mouse P388/DOX). A non‐cytotoxic concentration of cinchonine (10 μM) increased the sensitivity to doxorubicin of multidrug‐resistant P388/DOX cells and significantly enhanced the doxorubicin‐induced apoptosis and DNA fragmentation in resistant cells, but had no effect in parent cells. Time‐course studies demonstrated that DNA fragmentation was present 24 h after incubation with doxorubicin and cinchonine, indicating that DNA degradation was a preceding event. In cultured cells, cinchonine increased the intracellular accumulation of doxorubicin in the resistant cells in a dose‐dependent manner. Using flow cytometry to measure the inhibition of the P‐glycoprotein (P‐gp) dependent efflux of rhodamine 123, cinchonine was found to be considerably more effective than quinine. The results with cinchonine suggest that there may be quinine derivatives with a similar capacity to inhibit drug transport by P‐gp. Additionally, the G2/M phase cell population in resistant cells is increased by doxorubicin/cinchonine treatment. Exposure of resistant cells to 1 μM doxorubicin and 10 μM cinchonine resulted in the expression of Fas (APO‐1/CD95) in cells after 6 h. These studies demonstrate that the cell killing effects of doxorubicin and cinchonine in resistant cells are mediated, at least in part, by the induction of apoptosis.

Participation of Calcium Ion on Depletion Mechanism of Liver Glycogen by Purified Glucocorticoid Antagonizing Factor Released in Blood during Endotoxemia

The present study was conducted by the use of purified glucocorticoid antagonizing factor (GAF) released in blood of endotoxemic mice to determine whether or not the factor (GAF and Ca2+) may play a possible role of mediator in depletion mechanism of liver glycogen in endotoxemia. The liver glycogen level in 2 hr after injection with GAF plus cortisone‐treated mice was markedly lower than that in cortisone alone‐treated mice. However, the administration of trifluoperazine or verapamil markedly increased glycogen levels in liver of GAF plus cortisone‐injected mice. On the other hand, when the mice fed a calcium‐free diet were injected with GAF plus cortisone, there was merely a significant difference in liver glycogen level as compared to cortisone alone‐treated mice. The level of Ca24+ in liver cytosol fraction in cortisone‐treated mice was higher 2 hr after GAF injection than that in the cortisone alone‐treated one. The phosphorylase a activity in liver 2 hr after injection of GAF plus cortisone did not show a significant difference as compared to that in mice treated with cortisone alone. However, the activity ratio of glycogen synthase enzyme (synthase I/synthase I + D) was decreased in GAF plus cortisone‐treated mice as compared to that in cortisone alone‐treated mice. These findings suggest that there are participations of Ca2+ and mediator GAF released from reticuloendothelial system (RES) macrophages in glucoregulation of endotoxemia. Thus, it may be speculated that intracellular Ca2+ may mediate glycogenesis rather than glycogenolysis in the depletion mechanism of liver glycogen during GAF‐poisoning.

Purification and characteristics of glucocorticoid antagonizing factor in endotoxemia.

The present study involved the purification of GAF (glucocorticoid antagonizing factor) released in blood of endotoxemic mice, using the inhibition rate of tryptophan oxygenase (TO) activity in the mice liver as a parameter, to determine if this plays a role in metabolic disorders. GAF‐rich serum in zymosan‐primed and endotoxin‐injected mice was subjected to chromatography on DEAE‐Sepharose CL‐6B, Blue Sepharose CL‐6B and Sephadex G‐200 superfine columns. Finally, GAF fractions were purified by chromatography on a DEAE‐Sepharose CL‐6B column. The purified GAF showed a single band in electrophoresis in sodium dodecyl sulfate (SDS) polyacrylamide gel. The molecular weight of GAF was estimated to be 90,000. The purified GAF was regarded as glycoprotein. No factor (100 μg) exhibited lethal action on mice. The activity of TO in cortisone treated mice after injection of purified GAF was markedly lower than that in cortisone alone treated mice. On the other hand, there were no differences in tyrosine aminotransferase activities between the GAF plus cortisone injected group and cortisone only treated group. The glucose level after injection of GAF in cortisone treated mice initially showed hyperglycemia, but declined toward hypoglycemia 2 hr after injection, and thereafter returned nearly to the normal range by 4 hr. The liver glycogen level in GAF plus cortisone‐treated mice was markedly lower than that in cortisone‐alone treated mice.

Preventive effects of a verapamil against tumor necrosis factor-alpha-induced shock symptoms: approached from lipoprotein metabolic disorders

We examined the role of intracellular Ca2+ in the mechanism of the preventive effects of the Ca2+-channel blocker verapamil against lipoprotein disturbances during tumor necrosis factor (TNFa)-induced shock syndrome. The heparin-releasable lipoprotein lipase (LPL) activity in plasma of TNFalpha (5 X 10(4) units/mouse, i.v.)-injected mice was markedly lower at 4 h post-intoxication than in the controls. In mice treated with verapamil (10 mg/kg, s.c.), the activity of LPL 4 h after TNFalpha injection was significantly higher than in mice treated with TNFalpha alone. On the other hand, on polyacrylamide gel disk electrophoresis, very low density lipoprotein (VLDL) and high density lipoprotein (HDL) fractions in the sera of TNFalpha-injected mice were increased and reduced, respectively, relative to the controls. The administration of verapamil clearly prevented the lipoprotein damage arising from TNFalpha challenge. We investigated whether verapamil could suppress TNFalpha generation in endotoxin-treated J774A.1 cells. Treatment with verapamil (30 microM) markedly inhibited endotoxin (1 microg/ml)-induced TNFalpha production in these cells. These findings suggest that the concentration of intracellular Ca2+ may contribute to the extent of lipoprotein disturbances in plasma, which results from LPL suppression in TNFalpha-induced shock syndrome. Verapamil may, therefore, protect against some of the various disturbances caused by changes in Ca2+ mobilization through its ability to inhibit TNFalpha production in septic shock.