Neutrophil accumulation in ischemic reperfused rat liver: evidence for a role for superoxide free radicals HIROKAZU KOMATSU, ANTHONY KOO, ELHAM GHADISHAH, HA0 ZENG, JOHN F. KUHLENKAMP, MASAYASU INOUE, PAUL H. GUTH; AND NEIL KAPLOWITZ Division of Gastrointestinal and Liver Diseases, Department of Medicine, University of Southern California, School of Medicine, Los Angeles, California 90033; Center for Ulcer Research and Education, University of California, Los Angeles, School of Medicine, and Veterans Affairs Medical Center, West Los Angeles, Los Angeles, California 90073; and Department of Biochemistry, Kumamoto University Medical School, Kumamoto 860, Japan Komatsu, Hirokazu, Anthony Koo, Elham Ghadishah, Hao Zeng, John F. Kuhlenkamp, Masayasu Inoue, Paul H. Guth, and Neil Kaplowitz. Neutrophil accumulation in ischemic reperfused rat liver: evidence for a role for superoxide free radicals. Am. J. Physiol. 262 (Gastrointest. Liver Physiol. 25): G669-G676, 1992.-Oxygen-derived free radicals and leukocytes have been implicated in the pathogenesis of ischemiareperfusion injury. This study aimed at determining, by using biochemical and histochemical techniques, whether an accumulation of neutrophils occurs in the ischemic reperfused rat liver and whether superoxide free radicals play a role in mediating this neutrophil accumulation. Hepatic ischemia was induced by occluding blood supply to the left and median lobes, and reperfusion was reinstituted by releasing the occlusion. Myeloperoxidase activity of the liver was measured with a tetramethylbenzidine-HZOZ assay after removal of glutathione (by dialysis) and in the presence of 3-aminotriazole (catalase inhibitor). A modification of Graham and Karnovsky’s method was used to stain neutrophils in liver frozen sections, and the number of neutrophils was counted. Results showed that ischemia-reperfusion of the liver produced a U-fold increase in myeloperoxidase activity (from 0.073 t 0.009 to 0.320 t 0.017 units/mg liver, means t SE), which was proportional to the number of neutrophils (3.1-fold increase from 18 t 7 to 57 t 4 cells/mm2) in the liver tissue. Pretreatment with long-acting superoxide dismutase significantly attenuated the elevated myeloperoxidase activity and the number of neutrophils. These results indicate that reperfusion after a period of ischemia induces an accumulation of neutrophils in the liver, and superoxide anion free radicals are important mediators in the mechanism of this neutrophil accumulation. ischemia-reperfusion;




A NO-REFLOW PHENOMENON in ischemia-reperfusion injury has been reported in the brain, kidney, and heart (3, 10, 22). However, the mechanisms responsible for the no-reflow phenomenon remain uncertain. Inasmuch as there are several reports that tissue myeloperoxidase or leukocytes accumulate in postischemic tissue (9, 16) and neutropenia due to the administration of antineutrophil serum or monoclonal antibodies that prevent neutrophil adherence to microvascular endothelium affords significant protection against reperfusion-induced microvascular and parenchymal cell injury (20, 29,3l, 38,41,43), a role for neutrophils in ischemia-reperfusion injury has recently occupied the attention of numerous investigators. The activated neutrophils adhere to vascular endothelium and subsequently release superoxide anion or other reactive oxygen products that increase microvascular permeability (14). These events finally lead to tissue injury. 0193-1857/92

$2.00 Copyright

Recently, we have demonstrated by an in vivo microscopy technique that leukocyte-endothelium adhesions occurred in the liver sinusoids during reperfusion following a period of ischemia and caused flow obstruction in some of the liver sinusoids (23, 26). Degradation of superoxide anion by superoxide dismutase significantly decreased the number of leukocyte-endothelium adhesions and improved the hepatic no-reflow phenomenon (23, 27). However, the quantitative study of neutrophils with reference to ischemia and reperfusion has not been studied in the liver. In other tissues, several investigators (6, 15, 30, 34, 36) have demonstrated that myeloperoxidase (MPO), the naturally occurring constituent of neutrophils, can be used as a marker of tissue neutrophil content. Mullane et al. (34) reported that MPO activity can be used to measure neutrophil infiltration into ischemic myocardium. In the present study, we used both biochemical and histochemical techniques and aimed at testing the hypothesis that an accumulation of neutrophils occurs in the ischemic reperfused rat liver. In addition, we sought to determine whether oxygen free radicals participate in the mechanism of this neutrophil accumulation by assessing the protective effect of superoxide dismutase. To avoid the potential artifacts or inconsistencies, we used and compared two distinct methods for quantifying neutrophil accumulation, i.e., the biochemical myeloperoxidase activity and the histochemical identification of neutrophils. A portion of the results has been presented in abstract form (25). MATERIALS AND METHODS Animals

and agents. Male Sprague-Dawley rats weighing g (Simonsen Laboratories, Gilroy, CA) were used throughout the study. Hexadecyltrimethylammonium bromide 250-300

(HETAB), Triton x-100, EDTA, 3,3’,5,5’-tetramethylbenzi-

dine, oyster glycogen type II, catalase, MPO, bovine serum albumin, and 3,3’-diaminobenzidine were purchased from Sigma Chemical (St. Louis, MO). 3-Amino-1,2,4triazole was purchased from Aldrich Chemical (Milwaukee, WI). A derivative of the human erythrocyte type superoxide dismutase (SOD) was conjugated to poly-(styrene-co-maleic acid) as described previously (21,35). This synthetic derivative has a high affinity for plasma albumin. Previous in vivo studies in the rat showed that intravenously injected conjugated SOD circulated with a half-life of 6 h. This long-acting form of SOD was dissolved in 0.15 M NaCl at a concentration of 14 mg/ml and stored at

-80”c. Purification

0 1992 the American

of neutrophils. Physiological




were G669

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.





obtained from rats by injection with 20 ml of oyster glycogen saline (PBS, pH (6 w/ml id in 20 mM phosphate-buffered 7.4). Cells were harvested from the peritoneal cavity 4 h later and washed twice in PBS by centrifugation at 1,500 rpm for 5 min at 4°C. The peritoneal exudate at 4 h contained >98% neutrophils (30). Final cell concentration was determined by counting the number of neutrophils using a hemocytometer. After the centrifugation, purified neutrophils were suspended in 50 mM PBS (pH 6.0), which contained 0.5% HETAB, 1% Triton X-100, and 10 mM EDTA. HETAB and Triton X-100 are detergents that release MPO from the primary granules of the neutrophils (6, 37, 39), and HETAB completely inhibits the pseudoperoxidase activity of hemoglobin and myoglobin (16). The neutrophil suspensions were sonicated on ice for 15 s (model MS-50; Heat System-Ultrasonics, Farmingdale, NY) and subjected to two cycles of freezing and thawing and then centrifuged at 33,000 rpm for 15 min at 4°C. The supernatant was assayed for MPO activity. Animal preparation. Rats were fasted overnight and anesthetized with pentobarbital sodium (Nembutal, 50 mg/kg ip; Abbott Laboratories, North Chicago, IL). The body temperature (rectal) was monitored and maintained at 37°C by a heating pad. The right femoral vein was cannulated for intravenous administration of SOD or vehicle and for blood sampling. Longacting SOD (4 and 8 mg/kg iv) was injected 60 min before ischemia-reperfusion (see below). The control group received an injection of SOD (4 mg/kg iv) that had been boiled for 30 min. to inactivate it. The solution containing heat-inactivated SOD was also passed through a 0.2pm millipore filter (Uniflo, Schleicher and Schuell, Keene, NH) before administration to the rat. Ischemia-reperfusion. The abdomen was opened via a midline incision, and the liver was exposed. Ischemia of the left and median lobes of the liver was induced by occluding the left branches of portal vein, hepatic artery, and bile duct (7, 17, 26) using a microvascular clamp (type C-18; Natsume Seisakusho, Tokyo, Japan). Blood supply to the omental and right lobes of the liver was uninterrupted, and portal blood flow was maintained through the right branches. By gross inspection, there was no evidence of vascular congestion in the splanchnic organs. An ischemic period of 25 min was standardized throughout the present study. Pilot studies showed that there were no significant changes in systemic arterial pressure during ischemia and reperfusion when compared with the preischemic period, and pretreatment with long-acting SOD did not have any effect on systemic arterial pressure. To study the effect of ischemia-reperfusion on the number of leukocytes of whole blood, samples of blood (0.1 ml) were taken in some rats from the femoral vein catheter before ischemia and at 30 min after reperfusion. The number of neutrophils was counted using a hemocytometer. Tissue preparation. At 30 min after reperfusion, the left lobe was removed, weighed, and homogenized in 10 volumes of 50 mM PBS (pH 7.4) using a 15-s burst of a Brinkman tissue homogenizer (type PT 10/35). To eliminate glutathione (Fig. l), the homogenate was dialyzed in 50 mM PBS (pH 7.4) for 24 h at 4°C using a Spectra/Par dialysis membrane (Spectrum Medical Industries; Los Angeles, CA). After the dialysis, 0.5% HETAB, 1% Triton X-100, and IO mM EDTA were added, and the pH was adjusted to 5.0 with 5.0 N HCl. The homogenate was sonicated on ice for 15 s and subjected to two cycles of freezing and thawing, and then centrifuged at 33,000 rpm for 15 min at 4°C. The supernatant was assayed for MPO activity. MPO assay. MPO activity was assayed spectrophotometritally using a modification of the method, which utilizes 3,3’,5,5’-tetramethylbenzidine (TMB) as substrate (2, 16, 42). In this method (Fig. l), 0.1 ml of the supernatant to be measured was added to a l-ml reaction volume containing (in mM) 80



---------_-----------------~---------.---~~~~---I i Tetramethylbenzidine 0-w i I I H202 / ------

(655 Myeloperoxidase

i Blue nm) i

J’ ) Oxidized-TMB

I !





Catalase I_


+ O2

T Aminotriazole

Fig. 1. Diagram summarizing methods for myeloperoxidase (MPO) activity assay in liver. Reaction within rectangle shows standard method. MPO catalyzes oxidation of tetramethylbenzidine by HzOz to yield a blue chromogen, which has a wavelength maximum at 655 nm. Liver cells have considerable amount of glutathione-peroxidase and catalase, both of which compete with MPO for ‘H202. Glutathione, substrate for glutathione peroxidase, is eliminated using a dialysis technique. 3-Amino-1,2,4-triazole (3-AT) is used to inhibit catalase activity.

PBS (pH 5.4), 0.5% HETAB, 1.6 TMB, and 3 mM 3-amino1,2,4-triazole, which is a catalase inhibitor (8, 18, 19, 32, 33, 45). The mixture was incubated at 37°C for 5 min and the reaction started by the addition of 0.3 mM H202. Each tube containing the complete reaction mixture was incubated for exactly 1 min at 37°C. The reaction was terminated by the sequential addition of catalase (40 pg/ml) and 4 ml of 0.2 M Na acetate (pH 3.0). The changes in absorbance at 655 nm was measured with a spectrophotometer (model ACTA-M6; Beckman, Fullerton, CA). One unit of MPO activity was defined (2, 16) as the amount of enzyme present that produced a change in absorbance of 1.0 unit/min at 37°C in the final reaction volume containing the Na acetate. Validation of the MPO assay method. Preliminary MPO assay studies indicated that there were interactions between the activities from MPO (from neutrophils) and catalase (from liver tissue) in addition to the inhibition on both enzymes by 3-amino-1,2,4-triazole (3-AT). Hence the following enzymatic interactions and their inhibitions by 3-AT were systematically studied, using three levels of concentrations for each of the enzymes and 3-AT: 1) interaction between MPO (from neutrophils) and catalase (purified enzyme from Sigma Chemical or from liver tissue), 2) inhibition of MPO (purified enzyme, Sigma) or catalase (purified enzyme, Sigma) by 3-AT. Catalase (purified enzyme or in liver tissue) was estimated by the method described by Fraga et al. (11). Preliminary studies also indicated that by using 3 mM 3-AT to inhibit the catalase activities in 10 mg of liver tissue, the MPO activity could be partially recovered. To validate this assumption, the MPO activities (from lo* neutrophils) in the presence of catalase (liver tissue or purified enzyme) and 3-AT were assayed. Three levels of concentration of both catalase (5, 10, and 20 mg/ml liver tissue or 30, 100, and 300 g/ml purified enzyme) and 3-AT (I, 3, and 10 mM) were used. The empirical relationship between neutrophil MPO activities ( y) and liver tissue (x1) in the presence of 3-AT (x2) was analyzed using a multiple regression model (1,5). The equation used was in the form

y = PO + /&Xl + Pax:’ + p3x2 + p& where @is the regression coefficient. The regression coefficient, SE, t statistic (ratio of the parameter estimate to its SE), and P value were calculated (1, 5). In this context, the P value

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.




indicated the risk of rejecting a valid null hypothesis (p = 0). The coefficient of determination (R2) was calculated as a measure of goodness of fit. By substituting x1 = 10 mg/ml liver tissue and x2 = 3 mM 3AT, which were the actual amounts used in the MPO assay, the predicted value of MPO activity was calculated. A “correction factor” for the MPO assay was obtained by comparing this predicted value computed from the empirical regression equation to the mean MPO activity of the same quantity of neutrophils (lo4 cells) in the absence of liver tissue and 3-AT. This correction factor was subsequently used for adjusting the MPO activity in the MPO Assay. MPO staining. MPO in neutrophils was stained using a modification of the method of Graham and Karnovsky (12,113). At 30 min after reperfusion, the left lobe was removed and cut into small tissue blocks (3 x 4 X 5 mm3) with a razor blade. The tissue blocks were embedded in Tissue-Tek OCT compound (Miles, Elkhart, IN) and quickly frozen in dry iceacetone. Frozen sections, 6 pm in thickness, were cut on a cryostat (model 2800 Frigocut E; Reichert-Jung) and fixed at 4°C for 30 s in 3% glutaraldehyde-acetone fixative (pH 7.6). The sections were then rinsed in distilled water and incubated at room temperature for 3 min in a medium consisting of 5 mg 3,3’-diaminobenzidine tetrahydrochloride, 10 ml tris(hydroxymethyl)aminomethane (Tris).HCl buffer (pH 7.6), and 0.1 ml 1% H202. After incubation, they were washed in three changes of distilled water and immersed at room temperature for 2 min in copper nitrate solution (2.5 mg/ml). After the rinse in distilled water, the sections were stained with hematoxylin solution. The sections were examined by light microscopy, and photomicrographs were taken at a X275 magnification. A total of 100 micrographs for each group were subjected to morphological analysis. The number of leukocytes was counted, and the area of acini, exclusive of portal tract and terminal hepatic venule, was determined as was recently described (25). Statistical analysis. In the present study, data were expressed as means t SE. Comparison of statistical significance of differences between two groups was determined by the two-sample t test, and comparison among four groups by the one-way analysis of variance (ANOVA) with contrasts. A probability level of P < 0.05 was considered significant. RESULTS

Leukocyte count of whole blood. The number of leukocytes of whole blood before ischemia and after reperfusion is shown in Table 1. Ischemia and reperfusion of the liver did not have any significant effect on the systemic leukocyte count. Validation of the MPO assay method. Figure 2 shows that adding catalase (A) or liver tissue (B) attenuated the MPO activities (from neutrophils) in the MPO Assay. The inhibitory effects of 3-AT on both MPO and catalase activities (both from purified enzymes) were also Table 1. Leukocyte count of whole blood before and after ischemia and reperfusion Sham

0 min 30 min

9,160&1,030 8,220+850


8,800+910 8,150+670

Values are means t SE (in neutrophils/mm3) before ischemia (0 min) and after 25 min of ischemia and 30 min of reperfusion. Number of neutrophils is not significantly different from each other by 2-sample t test.






pg/mI lxa





mg/ml m5

ro -



0.15 20

- co z -J-




K 3 l -






B 22 0.00 1 .oo

/? 0.80 -E > ‘E 0.60 3 g






3-AT no -3

mM lzal m










Fig. 2. Top: inhibition of neutrophil MPO activity by catalase (A) and liver tissue (23). Bottom: concentration-dependent inhibition by 3-AT on activities from MPO (purified enzyme) (C) and on catalase (purified enzyme) (D). Data represent means t SE of 8 determinations in all groups.

demonstrated in a concentration-dependent manner (Fig. 2, C and 0). However, in the presence of both the liver tissue and 3-AT, the MPO activity from neutrophils was partially recovered. As shown in Fig. 3 (top), an optimal recovery was obtained with 10 mg/ml of liver tissue in the presence of 3 mM of 3-AT. When all the data (n = 72) from the three levels of concentrations of liver tissue and 3-AT were used, the following regression equation was obtained (R2 = 0.88, P < 0.001, Table Z), where yet was the predicted MPO activity and x1 and x2 were liver tissue (log pg/ml) and 3-AT (log PM) Yfit = -4.415

+ 1.814x1

- 0.227x:

+ 0.476~~

- 0.067x;

From the regression equation, a predicted value of 0.052 unit of MPO in lo4 neutrophils was obtained in the presence of 10 mg/ml liver tissue and 3 mM 3-AT. In a separate determination for MPO activity in which the liver tissue was replaced with protein (bovine serum albumin 0.2 mg/ml, Sigma) and in the absence of 3-AT, the MPO activity from lo4 neutrophils was 0.131 t 0.003 (means t SE, n = 14). Thus the general result was that MPO activity measured by the present MPO assay method underestimated the MPO on the average. The expected value of MPO activity was corrected as follows: expected MPO activity = measured MPO (0.131/0.052). In a similar manner, the MPO activity was also partially recovered from neutrophils in the presence of catalase and 3-AT (Fig. 3, bottom), and multiple regression analysis also demonstrated a similar relationship among MPO activity and catalase and 3-AT concentrations (data not shown). MPO actiuity. Using the MPO assay method described

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.





1 mM



10 mtvl


ber of neutrophils in liver was estimated to be 4.4 x lo6 neutrophils/g liver. Ischemia and reperfusion of the liver resulted in a significant (P < 0.001) increase in MPO activity, which was 0.320 t 0.017 MPO units/mg liver (Fig 5A). This activity corresponded to 19.2 x lO%eutrophils/g liver or approximately a 4.4-fold increase. MPO-stained neutrophils. An example of the MPO staining of the liver observed under light microscopy is shown in Fig. 6. Most of the MPO-stained neutrophils were found in the liver sinusoids. In the sham-operated group (Fig. 5B), the number of MPO-stained neutrophils was 18 t 7 neutrophils/mm’. Ischemia and reperfusion produced a significant (P < 0.001) increase in the number of MPO-stained neutrophils, which was 57 t 4 neutrophils/mm2 or approximately a 3%fold increase. There was a significant (P < 0.01) linear correlation between the MPO activity and the number of MPO-stained neutrophils (Fig. 7). Effect of SOD. The effects of pretreating the rats with the long-acting SOD (4 and 8 mg/kg iv, 60 min prior to ischemia) on the MPO activity and the number of MPOstained neutrophils are summarized in Fig. 5, C and D. In the ischemia-reperfusion plus heat-inactivated SODinjected group, the MPO activity and the number of neutrophils were significantly (P < 0.001) increased to 0.310 t 0.016 MPO units/mg liver (Fig. 5C) and 60 t 6 neutrophils/mm2 (Fig. 5D), respectively. Pretreatment with long-acting SOD significantly (P < 0.001) attenuated in a dose-dependent manner the elevated MPO activity and the number of neutrophils induced by ischemia-reperfusion. There was no significant difference between the sham-plus-vehicle group and the ischemiareperfusion plus long-acting SOD 8 mg/kg group.


LIVER EB!i m m



nm a, 0

p 0.00 v) .Is w 0.10 z z Ei 5 005 . Y z


5 mg/ml 10 mg/ml 20 mg/ml

CATAIASE RXY 30 pg/ml = 100 &ml m 300 &ml

0.00 Fig. 3. Interaction of MPO activity (from neutrophils) and catalase activity (either from liver tissue, top, or from purified enzyme, bottom) in presence of 3 concentration levels of 3-AT. Within concentration ranges of liver tissue, purified catalase enzyme and 3-AT, an optimal recovery of MPO activity was obtained with 3 mM concentration of 3AT. Data are means t SE from 8 determinations in all groups. Data from liver tissue group were used to compute multiple regression equation from which predicted value of MPO activity was calculated for a specific combination of liver tissue and 3-AT concentrations (10 mg and 3 mM, respectively).

in MATERIALS AND METHODS and adjusting the measured MPO activity for the 3-AT/catalase interference, we assayed the neutrophils from the rat peritoneal exudate for the biochemical MPO activity. As shown in Fig. 4, the MPO activity $waslinearly related to the number of neutrophils, ranging from 3,000 to 110,000 cells. Based on the regression line, each neutrophil had 17 X 10e6 MPO units of activity. The sham-operated liver contained 0.073 * 0.009 MPO units/mg liver (Fig. 5A). Based on the data of MPO activity per neutrophil and per milligram liver, the num-


Reperfusion following ischemia of short duration can result in the aggravation of tissue injury. We chose to study a 25-min interval of ischemia, since we have previously observed that irreversible necrosis did not occur as evidenced by protection seen with long-acting SOD (28). Longer periods of ischemia would be expected to irreversibly injure cells, whereas shorter intervals have no injury after reperfusion. Thus, with the ischemic interval we have chosen, reperfusion plays a major role in the injury process. In addition, we previously observed

Table 2. Statistical analyses Estimate coefficient




Constant Do Liver p1 Liver p2 3-AT ,& 3-AT ,&

of j!3



to predict


-4.415 1.814

Degree of freedom

Regression model Residual Total

5 67 72

8.1 (P < 0.001)

0.223 0.028 0.071 0.010

-0.227 0.476 -0.067

Source of variation



Sum of squares Analysis

-8.1 6.7 -6.7

(P < 0.001)

(I’ < 0.001) (P < 0.001)

Mean square


of variance 0.0487 0.0068 0.0555

0.0097 0.0001

97 (PC


Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.





0.0 0








(x 104)

Fig. 4. MPO activity in suspensions of rat neutrophils. Graph showing a positive linear correlation between biochemical MPO activity ( y) and number of neutrophils (x). Computed regression equation was yet = 0.095 + 0.072~ (R2 = 0.96, P < 0.01). 0.4









m .e 5 -


; 0.2 n 2 & k! 0.1 9




0.0 L


60 c B


Y E 2 2 s

N.S., # 20








Fig. 5. Effect of hepatic ischemia-reperfusion on MPO activity of liver tissue and number of neutrophils in liver frozen section. Ischemiareperfusion induces increase of hepatic MPO activity (A, 7 rats/group) and MPO-stained neutrophils (B, 7 rats/group). *p < 0.001, significant difference from sham-operated group @-sample t test). Superoxide dismutase (SOD, 4 and 8 mg/kg), in a dose-dependent manner, attenuated ischemia-reperfusion-induced increase of MPO activity (C, 5 rats/group) and number of neutrophils (D, 5 rats/group). *p < 0.001, significant difference from sham-plus-vehicle group (ANOVA with contrasts); #P < 0.001, difference from ischemia-reperfusion plus inactive SOD group (ANOVA with contrasts).

that blood flow gradually ceased in the microcirculation following reperfusion so that the no-reflow phenomenon was a major feature of the pathogenesis of reperfusioninduced injury (23). In the present study, we described quantitative biochemical and histochemical assay methods for neutrophil contents in the liver. Ischemia and reperfusion of the liver produced a significant increase in MPO activity,




which was proportional to the number of neutrophils in the liver tissue. Pretreatment with long-acting SOD (half-life 6 h) significantly attenuated the elevated MPO activity and the number of neutrophils. Previous studies have indicated that the long-acting SOD is bound to serum albumin and its clearance by the renal filtration mechanism, which rapidly eliminates the natural SOD (half-life 5 min) is significantly attenuated (21, 35). Because 80% of the injected SOD remained in the systemic circulation 2 h afterwards, it suggests that the superoxide anion, which has been scavenged by SOD, is mainly extracellular. The present findings suggest that production of superoxide anion participated in the initiation of neutrophil accumulation in ischemia-reperfusion. The mechanism of neutrophil adherence to the microvascular endothelium has not been fully investigated (26). Recent studies have shown that endothelial cells when damaged may release chemotactic factors that attract neutrophils to adhere to them. Other studies suggest that leukocyte membrane glycoproteins are critical for the neutrophil adherence, and monoclonal antibodies to these leukocyte membrane adhesive proteins have been shown to inhibit neutrophil adherence in the dog myocardium, cat, and rabbit intestinal mesentery and rabbit lung, liver, and gastrointestinal mucosa (4, 20, 40, 44). The MPO assay is a widely used method to quantify the tissue neutrophil content as an index of inflammation, because MPO is an enzyme found in neutrophils and, in much smaller quantities, in monocytes and macrophages (6, 30). However, there has been no documentation of MPO activity in the liver. The enzymatic determination of tissue-associated MPO, which is based on the HzOZ-dependent oxidation of an artificial electron donor, could be inhibited by catalase and glutathione peroxidase, both of which could compete with MPO for HZOa. In addition, MPO activity may be interfered with by naturally occurring electron donors present in the tissue such as reduced glutathione (16). Since liver cells have a considerable amount of catalase, glutathione, and glutathione peroxidase, these substances must be eliminated or inhibited before the enzymatic assay. In this study, to prevent interactions attributed by glutathione peroxidase, the substrate (i.e., glutathione) was eliminated using the dialysis technique (Fig. 1). Similarly, 3AT was used to inhibit the catalase activity (8, 18, 19, 32, 33, 45). In the present studies, this specific catalase inhibitor also had an inhibitory action on the MPO activity. However, using a suitable concentration (3 mM) of 3-AT to inhibit the catalase in the liver, the present results showed that a partial recovery of the MPO activity in the liver could be achieved. Moreover, the results of the biochemical MPO activity assay not only were consistent among the four experimental groups, but were also significantly correlated with the independent histochemical MPO-staining of neutrophils. Thus we have established a method to quantify MPO activity in the liver, and have demonstrated the increase in the hepatic MPO activity induced by ischemia-reperfusion. As shown in the present study, hepatic lobar ischemia and reperfusion did not affect the systemic leukocyte count. Histological sections of the ischemic reperfused

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.


Fig. 6. Photomicrographs B: ischemia-reperfusion



of liver frozen section showing group. Bar = 100 pm. P denotes


MPO-stained portal tract.





A: sham-operated


Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.




F ,0.3 .--2 5 u g 0.2



s 37







.. ..








. .


. ...

-f-.... .. . ... :.

. . 4+

.. ... .


+ . ...




IR +




IR +


4 mg/kg


IR +


8 mg/kg















4 mg/kg


Fig. 7. In vivo positive linear correlation between MPO activity (y) and number of neutrophils (x) in sham-operated and ischemia-reperfusion groups, with or without pretreatment with long-acting SOD. IR, ischemia and reperfusion. Computed regression equation was y = -0.013 + 0.006~ (R’ = 0.98, P < 0.01).

liver also did not show any sign of congestion (data not shown). Therefore the increase in MPO activity and neutrophil numbers was not due to leukocytosis or congestion, which could have also resulted in neutrophil accumulation in the liver. The Kupffer cells in the liver contain MPO and may contribute to the hepatic MPO activity. However, this is not a severe limitation when one considers that Kupffer cells originate from circulating monocytes that have only one-thousandth of the neutrophil’s MPO activity (30). Since 3,3’-diaminobenzidine tetrahydrochloride specifically reacts with peroxidase, both Kupffer cells and neutrophils also show positive staining reactions. These two types of cell are distinguished by observing their shapes and intensity of reaction. Neutrophils have round or oval shapes and intense dark brown reaction (Fig. 6), whereas Kupffer cells are irregularly polygonal and show bright brown reaction, probably due to their low content of MPO. In addition, Kupffer cells often show inhomogeneous and granular reactions. In the in vivo liver microcirculation, reperfusion of the left lobe of liver after 25 min of ischemia resulted in an initial transient return of blood flow, but stasis of blood flow developed in the liver sinusoids during the following 30 min of reperfusion (the no-reflow phenomenon) (23, 26). In our recent in vivo microscopic studies, neutrophil adherence to the microvascular endothelium was observed during the reperfusion period but not in the normal liver microcirculation (26, 27). The cumulative number of neutrophil adherence to the endothelium increased in parallel with the microcirculatory stasis of blood flow. Plugging of liver sinusoids by adherent neutrophils was observed to cause obstruction of blood flow in the sinusoid (27). Thus, the no-reflow phenomenon due to neutrophil plugging of liver sinusoid may contribute to the ischemia-reperfusion injury of the liver (28). However, our previous work with in vivo microscopy revealed information about the microcirculation at the surface of the liver and might not represent the global response. In the present study, we have quantified the neutrophil accumulation in the whole liver and the role of superoxide anion during reperfusion using two inde-




pendent techniques. In conclusion, we have quantified the neutrophil content in the ischemic reperfused rat liver. Results indicate that reperfusion after a period of ischemia induces an accumulation of neutrophils in the liver, and superoxide anion free radicals are important mediators in the mechanism of this neutrophil accumulation. The precise role of neutrophils in the induction of hepatocellular necrosis following ischemia-reperfusion remains uncertain but could include both obstruction of blood perfusion in the sinusoids and the release of free radicals, other oxidants, and proteolytic enzymes. We are indebted to Dr. Matthew B. Grisham for technical advice. This study was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-38867 and Veterans Affairs research funds. Address for reprint requests: A. Koo, Univ. of Southern California, School of Medicine, 1333 San Pablo St. MMR415, Los Angeles, CA 90033. Received

27 June

1990; accepted

in final


15 November


REFERENCES 1. Altman, D. G. Practical Statistics for Medical Research. London: Chapman and Hall, 1991, p. 336-351. 2. Andrews, P. C., and N. I. Krinsky. Human myeloperoxidase and hemi-myeloperoxidase. Methods Enzymol. 132: 369-378, 1986. 3. Ames, A., R. L. Wright, M. Kowada, J. M. Thurston, and G. Majno. Cerebral ischemia: II. The no-reflow phenomenon. Am. J. Pathol. 52: 437-447, 1968. 4. Arfors, K. E., C. Lundberg, L. Lindbom, K. Lundberg, P. B. Beatty, and J. M. Harlan. A monoclonal antibody to the membrane glycoprotein complex CD18 inhibits polymorphonuclear leukocyte accumulation and plasma leakage in vivo. BZood 69: 338340,1987. 5. Armitage,

P., and G. Berry. Statistical Methods in Medical Research (2nd ed.). Oxford: Blackwell, 1987, p. 296-324. 6. Bradley, P. P., D. A. Priebat, R. D. Christensen, and G. Rothstein. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J. Inuest. Dermatol. 78:206-209,1982. 7. Clemens, M.

8. 9.








G., P. F. McDonagh, I. H. Chaudry, and A. E. Baue. Hepatic microcirculatory failure after ischemia and reperfusion: improvement with ATP-MgC12 treatment. Am. J. Physiol. 248 (Heart Circ. Physiol. 17): H804-H811, 1985. Darr, D., and I. Fridovich. Irrevisible inactivation of catalase by 3-amino-1,2,4-triazole. Biochem. Pharmacol. 35: 3642, 1986. Engler, R. L., M. D. Dahlgren, M. A. Peterson, A. Dobbs, and G. W. Schmid-Schoenbein. Accumulation of polymorphonuclear leukocytes during 3-h experimental myocardial ischemia. Am. J. PhysioZ. 251 (Heart Circ. Physiol. 20): H93-HlOO, 1986. Flores, J., D. R. DiBona, C. H. Beck, and A. Leaf. The role of cell swelling in ischemic renal damage and the protective effect of hypertonic solute. J. CZin. Inuest. 51: 118-126, 1972. Fraga, C. G., R. F. Arias, S. F. Llesuy, 0. R. Koch, and A. Boveris. Effect of vitamin E- and selenium-deficiency on rat liver chemiluminescence. Biochem. J. 242: 383-386, 1987. Graham, R. C., and M. J. Karnovsky. The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem. 14: 291-302, 1966. Graham, R. C., and M. J. Karnovsky. Glomerular permeability. Ultrastructural cytochemical studies using peroxidases as protein tracers. J. Exp. Med. 124: 1123-1133, 1966. Granger, D. N. Role of xanthine oxidase and granulocytes in ischemia-reperfusion injury. Am. J. Physiol. 255 (Heart Circ. Physiol. 24): H1269-H1275, 1988. Grisham, M. B., J. N. Benoit, and D. N. Granger. Assessment of leukocyte involvement during ischemia and reperfusion of the intestine. Methods Enxymol. 186: 729-742, 1990. Grisham, M. B., L. A. Hernandez, and D. N. Granger.

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.









24. 25.










Xanthine oxidase and neutrophil infiltration in intestinal ischemia. Am. J. Physiol. 251 (Gastrointest. Liver Physiol. 14): G567-G574, 1986. Hayashi, H, I. H. Chaudry, M. G. Clemens, and A. R. Baue. Hepatic ischemia models for determining the effects of ATP-MgCl, treatment. J. Surg. Res. 40: 167-175, 1986. Heims, W. G., and D. Appleman. Production of catalase changes in animals with 3-amino-1,2,4-triazole. Science Wash. DC 122: 693694, 1955. Heims, W. G., D. Appleman, and H. T. Pyfrom. Effects of 3amino-1,2,4-triazole on catalase and other compounds. Am. J. Physiol. 186: 19-23, 1956. Hernandez, L. A., M. B. Grisham, B. Twohig, K. E. Arfors, J. M. Harlan, and D. N. Granger. Role of neutrophils in ischemia-reperfusion-induced microvascular injury. Am. J. Physiol. 253 (Heart Circ. Physiol. 22): H699-H703, 1987. Inoue, M., I. Ebashi, N. Watanabe, and Y. Morino. Synthesis of a superoxide dismutase derivative that circulates bound to albumin and accumulates in tissues whose pH is decreased. Biochemistry 28: 6619-6624, 1989. Kloner, R. A., C. E. Canote, and R. B. Jennings. The “noreflow” phenomenon after temporary coronary occlusion in the dog. J. CZin. Invest. 54: 1496-1508, 1974. Komatsu, H., A. Koo, E. Chan, M. Inoue, N. Kaplowitz, and P. H. Guth. Hepatic microvascular “no-reflow” induced by ischemia-reperfusion in the rat: role of superoxide free radicals (Abstract). GastroenteroZogy 96: A266, 1989. Komatsu, H., A. Koo, and P. H. Guth. Leukocyte flow dynamics in the rat liver microcirculation. Microvasc. Res. 40: l-13, 1990. Komatsu, H., A. Koo, M. Inoue, N. Kaplowitz, and P. H. Guth. Neutrophil accumulation in ischemic reperfused rat liver: role of superoxide free radicals (Abstract). GastroenteroZogy 98: A598,1990. Koo, A., G. Breit, and M. Intaglietta. Leukocyte adherence in hepatic microcirculation in ischemia reperfusion. In: MicrocircuZation in CircuZatory Disorders, edited by H. Manabe, B. W. Zweifach, and K. Messmer. Tokyo: Springer-Verlag, 1988, p. 205-213. Koo, A., H. Komatsu, E. Chan, P. H. Guth, M. Inoue, and N. Kaplowitz. Long-acting superoxide dismutase decreases the incidence of leukocyte-endothelium adhesion in ischemic reperfused rat liver (Abstract). Gastroenterology 96: A615, 1989. Koo A., H. Komatsu, 6. Paulsen, M. Inoue, P. H. Guth, and N. Kaplowitz. Long-acting superoxide dismutase protects against reperfusion injury in the rat liver (Abstract). Gastroenterology 98: A598,1990. Kortius, R. J., M. B. Grisham, and D. N. Granger. Leukocyte depletion attenuates vascular injury in post-ischemic skeletal muscle. Am. J. Physiol. 254 (Heart Circ. Physiol. 23): H823-H827,1988. Krawisz, J. E., P. Sharon, and W. F. Stenson. Quantitative assay for acute intestinal inflammation based on myeloperoxidase activity. Assessment of inflammation in rat and hamster models. GastroenteroZogy 87: 1344-1350, 1984. Linas, S. L., P. F. Shanley, D. Whittenburg, E. Berger, and J. E. Repine. Neutrophils accentuate ischemia-reperfusion injury


32. 33.














in isolated perfused rat kidneys. Am. J. Physiol. 255 (Renal FZuid Electrolyte Physiol. 24): F728-F735, 1988. Margoliash, E., and A. Novogrodsky. A study of the inhibition of catalase by 3-amino-1,2,4-triazole. Biochem. J. 68: 468-475,1958. Margoliash, E., A. Novogrodsky, and A. Schejter. Irreversible reaction of 3-amino-1,2,4triazole and related inhibitors with the protein of catalase. Biochem. J. 74: 339-350, 1960. Mullane, K. M., R. Kraemer, and B. Smith. Myeloperoxidase activity as a quantitative assessment of neutrophil infiltration into ischemic myocardium. J. Pharmacol. Methods 14: 157-167, 1985. Ogino, T., M. Inoue, Y. Ando, M. Awai, H. Maeda, and Y. Morino. Chemical modification of superoxide dismutase. Extension of plasma half life of the enzyme through its reversible binding to the circulating albumin. Int. J. Pept. Protein Res. 32: 153-159, 1988. Ormrod, D. J., G. L. Harrison, and T. E. Miller. Inhibition of neutrophil myeloperoxidase activity by selected tissues. J. Pharmacol. Methods 18: 137-142, 1987. Patriarca, P., R. Cramer, M. Marussi, F. Rossi, and D. Romeo. Mode of activation of granule-bound NADPH oxidase in leukocytes during phagocytosis. Biochem. Biophys. Acta 237: 335338,197l. Romson, J. L., B. G. Hook, S. L. Kunkel; G. D. Abrams, M. A. Schork, and B. R. Lucchesi. Reduction of the extent of ischemic myocardial injury by neutrophil depletion in the dog. Circulation 67: 1016-1023, 1983. Schultz, J., and K. Kaminker. Myeloperoxidase of the leukocyte of normal human blood. 1. Content and localization. Arch. Biochem. Biophys. 96: 465-467, 1962. Simpson, P. J., R. F. Todd, J. C. Fantone, J. K. Michelson, J. D. Griffin, and R. B. Lucchesi. Reduction of experimental canine myocardial reperfusion by a monoclonal antibody (antiMol, anti-CDllb) that inhibits leukocyte adhesion. J. CZin. Invest. 81: 624-629, 1988. Smith, S. M., L. Holm-Rutili, M. A. Perry, M. B. Grisham, K. E. Arfors, D. N. Granger, and P. R. Kviety. Role of neutrophils in hemorrhagic shock-induced gastric mucosal injury in the rat. GastroenteroZogy 93: 466-471, 1987. Suzuki, K., J. Ota, S. Sasagawa, T. Sakatani, and T. Fujikura. Assay method for myeloperoxidase in human polymorphonuclear leukocyte. Anal. Biochem. 132: 345-352, 1983. Suzuki, M., W. Inauen, P. R. Kviety, M. B. Grisham, C. Meininger, M. E. Schelling, H. J. Granger, and D. N. Granger. Superoxide mediates reperfusion-induced leukocyte-endothelial cell interactions. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): Hl740-Hl745,1989. Vedder, N. B., R. K. Winn, C. L. Rice, E. Y. Chi, K. E. Arfors, and J. M. Harlan. A monoclonal antibody to the adherence glycoprotein, CDl8, reduces organ injury and improves survival from hemorrhagic shock and resuscitation in rabbit. J. CZin. Invest. 81: 939-944, 1988. Williams, R. N., N. A. Delawere, and C. A. Paterson. Inactivation of catalase with 3-amino-1,2,4-triazole: an indirect irreversible mechanism. Biochem. Pharmacol. 34: 3386-3389,1985.

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

Neutrophil accumulation in ischemic reperfused rat liver: evidence for a role for superoxide free radicals.

Oxygen-derived free radicals and leukocytes have been implicated in the pathogenesis of ischemia-reperfusion injury. This study aimed at determining, ...
2MB Sizes 0 Downloads 0 Views