JOURNAL OF PATHOLOGY, VOL.
rN
167: 83-89 (1 992)
srm DETECTION OF OXIDATIVE STRESS IN RAT HEPATOCYTES SATOSHI MOCHIDA, NAOHIKO MASAKI, YASUHIKO OHTA, ATSUSHI MATSUI, ITSURO OGATA AND KENJI FUJIWARA
First Department oflnternal Medicine, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan Received 24 June 1991 Accepted27 August 1991 SUMMARY In rat hepatocytes in primary culture incubated with nitro blue tetrazolium, formazan content was increased by addition of t-butyl hydroperoxide, a potent oxidant, in a dose-related manner, but not by addition of valinomycin, which kills hepatocytes through mitochondria1 damage. This increment after t-butyl hydroperoxide addition was not seen in hepatotyctes preincubated with deferoxamine mesylate, a ferric iron chelator which inhibits radical formation. Liver perfusion with nitro blue tetrazolium and t-butyl hydroperoxide in rats produced formazan deposition faintly on the surface of hepatocytes throughout the liver and prominently in the cytoplasm of some hepatocytes, which was attenuated when performed following deferoxamine mesylate perfusion. When liver perfusion with nitro blue tetrazolium was performed in carbon tetrachloride-intoxicated rats, formazan deposition appeared diffusely in hepatocytes in the centrilobular areas. Similar deposition was also observed on the surface and in the cytoplasm of hepatocytes in the periportal and mid-zonal areas in rats undergoing post-ischaemic reperfusion. Liver perfusion with nitro blue tetrazolium can detect in situ oxidative stress in hepatocytes and may be a useful tool for studying the role of lipid peroxidation in rat liver injury. KEY
WORDS-Nitro blue tetrazolium, t-butyl hydroperoxide, carbon tetrachloride, ischaemic reperfusion, oxidative stress.
INTRODUCTION Oxidative stress is a n important contributory factor in liver injury', and in situ methods of detecting hepatocytes under oxidative stress have been . ~ a procedure to awaited. Suematsu et ~ 1reported visualize in situ lipid peroxidation, a manifestation of oxidative cell injury,'*3 using a lipid peroxidesensitive probe in rat liver. This method seems to demonstrate the intralobular heterogeneity of oxidative stress, but does not provide information about individual cells. Nitro blue tetrazolium (NBT) is a hydrophilic yellow crystal which is converted to insoluble blue formazan by r e d ~ c t i o n .It~ is utilized for in vitro evaluation of macrophage o r neutrophil function by Addressee for correspondence: Kenji Fujiwara, MD, First Department of Internal Medicine, Faculty of Medicine, University ofTokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113, Japan.
0022 341 7/92/05008347 $05.00 0 1992 by John Wiley & Sons, Ltd.
measuring formazan produced by reducing agents such as superoxide anions or NADPH oxidase, the rate-limiting enzyme for respiratory It is also used for the detection of reductases or other enzymes in tissue specimens.' Hepatocytes exposed to oxidant are considered to have elevated reductase activities because N A D P H oxidase activity is increased t o compensate for depletion of reduced glutathione and NADPH,9 suggesting that formazan deposition may develop in hepatocytes under oxidative stress. Previously we reported that liver perfusion with NBT and phorbol 12-myristate 13-acetate produces formazan deposition in hepatic macrophages," thereby demonstrating in situ the stimulatory state of these cells." This method has enabled us to demonstrate the role of hepatic macrophages in the development of liver injury.I2 l4 In the present study, we report our findings on liver perfuaim with NBT to detect in situ oxidative stress in hepatocytes.
84
S. MOCHIDA E T A L .
MATERIALS AND METHODS Experimental design
Male Sprague-Dawley rats (Shizuoka Laboratory Animal Center, Shizuoka, Japan) weighing 170-19Og were housed, five rats per cage, at a temperature of 22 f2°C under normal laboratory lighting conditions and maintained on a commercial pelleted diet and water ad libitum. Experiment I-Rat hepatocytes were isolated and cultured with NBT in the presence of t-butyl hydroperoxide (TBHP) or valinomycin with or without preincubation with deferoxamine mesylate for determination of formazan content. The experiment was repeated three times. Experiment ZZ-Six rats were subjected to liver perfusion with NBT and TBHP or NBT alone, and three rats to perfusion with NBT and TBHP following deferoxamine mesylate perfusion. Experiment IIZ-Ten rats received a single dose of 1.5 ml/kg body weight of carbon tetrachloride (CCI,) as a 20 per cent solution in olive oil and ten control rats the same volume of olive oil through an oesophageal tube, and then maintained on a commercial pelleted diet and water adlibiturn. They were subjected to liver perfusion with NBT 18 or 72 h later. Experiment IV-Five rats were subjected to ischaemia and reperfusion manipulation after fasting overnight, and to liver perfusion with NBT. Determination offormazan content in cultured hepatocytes
Hepatocytes were isolated by collagenase (type I: Worthington Biochemical Corp., Freefold, NJ) perfusion according to the method of Seglen.’’ The yield of hepatocytes was 1.5-2.5 x 10’ cells per liver, with 90-95 per cent viability by trypan blue exclusion. The isolated hepatocytes were plated in Falcon 3002 dishes (Becton Dickinson and Co., Lincoln Park, NJ) at a density of 4 x lo4cells/cm2in 2ml of Williams’ medium E (Flow Laboratories, Irvine, Scotland) containing 10 IU/ml penicillin, 10 pg/ml streptomycin, 0.02 U/ml insulin (Novo Industri A/S, Copenhagen, Denmark), and 10 per cent heat-inactivated (56°C for 30 min) fetal calf
serum (GIBCO Laboratories, Life Technologies, Inc., Gland Island, NY) (complete Williams’ medium E). After incubation for 2 h at 37°C in an atmosphere of 5 per cent C0,-95 per cent air, the cultures were rinsed twice with HEPES buffer ( 0 . 1 4 ~NaCI, 6 . 7 m ~KCI, 1 . 2 m ~CaCI,, and 10 mM HEPES, pH 7.4) to remove unattached dead cells, replenished with 3 ml of complete Williams’ medium E, and incubated further for 24 h. The resultant cultures were washed with HEPES buffer and replenished again with 3 ml of complete Williams’ medium E minus fetal calf serum and insulin (incomplete Williams’ medium E) with or without 4 mM deferoxamine mesylate (Ciba-Geigy Co. Ltd, Takarazuka, Japan). After 1 h incubation, the medium was changed to 2ml of incomplete Williams’ medium E without deferoxamine mesylate. After addition of 1 ml of 2 mg/ml NBT (grade 111; Sigma Chemical Co., St Louis, MO) in PBS containing TBHP (Sigma Chemical Co.) or valinomycin (Sigma Chemical Co.) at final concentrations of 0,0.25,0.5 and 0.75 mM, or 0,0.3,0.6, and 0.9 ,UM, respectively, they were further incubated for 30 min. Formazan content in hepatocytes was measured according to the method of Rook et a1.I6 with some modifications as follows. After removal of the medium, hepatocytes were fixed with 70 per cent methanol, washed three times with the same solution, and allowed to air-dry. Formazan was solubilized by adding 1.96 ml of dimethyl sulphoxide after cell destruction with 1.68 ml of 2 M KOH, and the solution was collected for measurement of formazan content at 630 nm. Measurement was done in triplicate against the solution similarly obtained from the cultures after incubation without cells as a blank. Ischaemia and reperfusion manipulation
Ischaemia and reperfusion manipulation was performed as previously described” with some modifications. After the abdomen was cut open under anaesthesia with an intraperitoneal injection of pentobarbital sodium, 200 units of heparin (Novo Industri A/S) was injected into the inferior vena cava. The hepatic pedicle supplying the left lateral and median lobes was occluded with an atraumatic clamp (Matsuda Ika Ltd, Tokyo, Japan). During ischaemia for 30 min, the rats were kept under anaesthesia, with a wet cheesecloth covering the open abdomen under a heating lamp. Thirty minutes after removing the clamp, they were subjected to liver perfusion with NBT.
85
OXIDATIVE STRESS IN HEPATOCYTES
0.5 h
0
r
*
m
a" 0
0.5
r
0.4 0*4
a
v
42
5
42
c
3 c
i z
0.3
0.3 0.2
0.1
0
-r0.2
0.1
0
0
0.25
0.5
0.75
TBHP (mM)
0
0.3
0.6
0.9
valinomycin (pM)
Fig. I-Effects of t-butyl hydroperoxide (TBHP) and valinomycin on formazan content in cultured rat hepatocytes after incubation with nitro blue tetrazolium. The left panel shows the result for TBHP without (dashed line) or with (solid line) pretreatment of deferoxamine mesylate. The right panel demonstrates the result for valinomycin. Each point indicates the m e a n k S D ofthreedishes. *P
rN
167: 83-89 (1 992)
srm DETECTION OF OXIDATIVE STRESS IN RAT HEPATOCYTES SATOSHI MOCHIDA, NAOHIKO MASAKI, YASUHIKO OHTA, ATSUSHI MATSUI, ITSURO OGATA AND KENJI FUJIWARA
First Department oflnternal Medicine, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan Received 24 June 1991 Accepted27 August 1991 SUMMARY In rat hepatocytes in primary culture incubated with nitro blue tetrazolium, formazan content was increased by addition of t-butyl hydroperoxide, a potent oxidant, in a dose-related manner, but not by addition of valinomycin, which kills hepatocytes through mitochondria1 damage. This increment after t-butyl hydroperoxide addition was not seen in hepatotyctes preincubated with deferoxamine mesylate, a ferric iron chelator which inhibits radical formation. Liver perfusion with nitro blue tetrazolium and t-butyl hydroperoxide in rats produced formazan deposition faintly on the surface of hepatocytes throughout the liver and prominently in the cytoplasm of some hepatocytes, which was attenuated when performed following deferoxamine mesylate perfusion. When liver perfusion with nitro blue tetrazolium was performed in carbon tetrachloride-intoxicated rats, formazan deposition appeared diffusely in hepatocytes in the centrilobular areas. Similar deposition was also observed on the surface and in the cytoplasm of hepatocytes in the periportal and mid-zonal areas in rats undergoing post-ischaemic reperfusion. Liver perfusion with nitro blue tetrazolium can detect in situ oxidative stress in hepatocytes and may be a useful tool for studying the role of lipid peroxidation in rat liver injury. KEY
WORDS-Nitro blue tetrazolium, t-butyl hydroperoxide, carbon tetrachloride, ischaemic reperfusion, oxidative stress.
INTRODUCTION Oxidative stress is a n important contributory factor in liver injury', and in situ methods of detecting hepatocytes under oxidative stress have been . ~ a procedure to awaited. Suematsu et ~ 1reported visualize in situ lipid peroxidation, a manifestation of oxidative cell injury,'*3 using a lipid peroxidesensitive probe in rat liver. This method seems to demonstrate the intralobular heterogeneity of oxidative stress, but does not provide information about individual cells. Nitro blue tetrazolium (NBT) is a hydrophilic yellow crystal which is converted to insoluble blue formazan by r e d ~ c t i o n .It~ is utilized for in vitro evaluation of macrophage o r neutrophil function by Addressee for correspondence: Kenji Fujiwara, MD, First Department of Internal Medicine, Faculty of Medicine, University ofTokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113, Japan.
0022 341 7/92/05008347 $05.00 0 1992 by John Wiley & Sons, Ltd.
measuring formazan produced by reducing agents such as superoxide anions or NADPH oxidase, the rate-limiting enzyme for respiratory It is also used for the detection of reductases or other enzymes in tissue specimens.' Hepatocytes exposed to oxidant are considered to have elevated reductase activities because N A D P H oxidase activity is increased t o compensate for depletion of reduced glutathione and NADPH,9 suggesting that formazan deposition may develop in hepatocytes under oxidative stress. Previously we reported that liver perfusion with NBT and phorbol 12-myristate 13-acetate produces formazan deposition in hepatic macrophages," thereby demonstrating in situ the stimulatory state of these cells." This method has enabled us to demonstrate the role of hepatic macrophages in the development of liver injury.I2 l4 In the present study, we report our findings on liver perfuaim with NBT to detect in situ oxidative stress in hepatocytes.
84
S. MOCHIDA E T A L .
MATERIALS AND METHODS Experimental design
Male Sprague-Dawley rats (Shizuoka Laboratory Animal Center, Shizuoka, Japan) weighing 170-19Og were housed, five rats per cage, at a temperature of 22 f2°C under normal laboratory lighting conditions and maintained on a commercial pelleted diet and water ad libitum. Experiment I-Rat hepatocytes were isolated and cultured with NBT in the presence of t-butyl hydroperoxide (TBHP) or valinomycin with or without preincubation with deferoxamine mesylate for determination of formazan content. The experiment was repeated three times. Experiment ZZ-Six rats were subjected to liver perfusion with NBT and TBHP or NBT alone, and three rats to perfusion with NBT and TBHP following deferoxamine mesylate perfusion. Experiment IIZ-Ten rats received a single dose of 1.5 ml/kg body weight of carbon tetrachloride (CCI,) as a 20 per cent solution in olive oil and ten control rats the same volume of olive oil through an oesophageal tube, and then maintained on a commercial pelleted diet and water adlibiturn. They were subjected to liver perfusion with NBT 18 or 72 h later. Experiment IV-Five rats were subjected to ischaemia and reperfusion manipulation after fasting overnight, and to liver perfusion with NBT. Determination offormazan content in cultured hepatocytes
Hepatocytes were isolated by collagenase (type I: Worthington Biochemical Corp., Freefold, NJ) perfusion according to the method of Seglen.’’ The yield of hepatocytes was 1.5-2.5 x 10’ cells per liver, with 90-95 per cent viability by trypan blue exclusion. The isolated hepatocytes were plated in Falcon 3002 dishes (Becton Dickinson and Co., Lincoln Park, NJ) at a density of 4 x lo4cells/cm2in 2ml of Williams’ medium E (Flow Laboratories, Irvine, Scotland) containing 10 IU/ml penicillin, 10 pg/ml streptomycin, 0.02 U/ml insulin (Novo Industri A/S, Copenhagen, Denmark), and 10 per cent heat-inactivated (56°C for 30 min) fetal calf
serum (GIBCO Laboratories, Life Technologies, Inc., Gland Island, NY) (complete Williams’ medium E). After incubation for 2 h at 37°C in an atmosphere of 5 per cent C0,-95 per cent air, the cultures were rinsed twice with HEPES buffer ( 0 . 1 4 ~NaCI, 6 . 7 m ~KCI, 1 . 2 m ~CaCI,, and 10 mM HEPES, pH 7.4) to remove unattached dead cells, replenished with 3 ml of complete Williams’ medium E, and incubated further for 24 h. The resultant cultures were washed with HEPES buffer and replenished again with 3 ml of complete Williams’ medium E minus fetal calf serum and insulin (incomplete Williams’ medium E) with or without 4 mM deferoxamine mesylate (Ciba-Geigy Co. Ltd, Takarazuka, Japan). After 1 h incubation, the medium was changed to 2ml of incomplete Williams’ medium E without deferoxamine mesylate. After addition of 1 ml of 2 mg/ml NBT (grade 111; Sigma Chemical Co., St Louis, MO) in PBS containing TBHP (Sigma Chemical Co.) or valinomycin (Sigma Chemical Co.) at final concentrations of 0,0.25,0.5 and 0.75 mM, or 0,0.3,0.6, and 0.9 ,UM, respectively, they were further incubated for 30 min. Formazan content in hepatocytes was measured according to the method of Rook et a1.I6 with some modifications as follows. After removal of the medium, hepatocytes were fixed with 70 per cent methanol, washed three times with the same solution, and allowed to air-dry. Formazan was solubilized by adding 1.96 ml of dimethyl sulphoxide after cell destruction with 1.68 ml of 2 M KOH, and the solution was collected for measurement of formazan content at 630 nm. Measurement was done in triplicate against the solution similarly obtained from the cultures after incubation without cells as a blank. Ischaemia and reperfusion manipulation
Ischaemia and reperfusion manipulation was performed as previously described” with some modifications. After the abdomen was cut open under anaesthesia with an intraperitoneal injection of pentobarbital sodium, 200 units of heparin (Novo Industri A/S) was injected into the inferior vena cava. The hepatic pedicle supplying the left lateral and median lobes was occluded with an atraumatic clamp (Matsuda Ika Ltd, Tokyo, Japan). During ischaemia for 30 min, the rats were kept under anaesthesia, with a wet cheesecloth covering the open abdomen under a heating lamp. Thirty minutes after removing the clamp, they were subjected to liver perfusion with NBT.
85
OXIDATIVE STRESS IN HEPATOCYTES
0.5 h
0
r
*
m
a" 0
0.5
r
0.4 0*4
a
v
42
5
42
c
3 c
i z
0.3
0.3 0.2
0.1
0
-r0.2
0.1
0
0
0.25
0.5
0.75
TBHP (mM)
0
0.3
0.6
0.9
valinomycin (pM)
Fig. I-Effects of t-butyl hydroperoxide (TBHP) and valinomycin on formazan content in cultured rat hepatocytes after incubation with nitro blue tetrazolium. The left panel shows the result for TBHP without (dashed line) or with (solid line) pretreatment of deferoxamine mesylate. The right panel demonstrates the result for valinomycin. Each point indicates the m e a n k S D ofthreedishes. *P
