Prostaglandins Leukotrienes and Essential Fatty Acids (lYY2) 45, 113-l IY 0 Longman Group UK Ltd WY2
Role of Leukotriene B4 in the Pathogenesis of Hepatic IschemiaReperfusion Injury in the Rat H. Hughes, A. Farhood*
and H. Jaeschke
Center for Experimental Therapeutics, Department of Medicine, Baylor College of Medicine, and *Department of Pathology, The University of Texas Medical School, Houston, Texas 77030, USA (Reprint requests to AF) ABSTRACT.
A common feature to most models of &hernia-reperfusion injury is the accumulation of polymorphonuclear leukocytes (PMNs) into the post-ischemic tissue during the reperfusion period. Interventions that lead to decreased PMN inliltration protect against tissue injury and therefore a knowledge of the chemotactic mediators leading to PMN accumulation is essential to understanding the pathogenesis of the injury and to the development of successful therapeutic strategies. Leukotriene B4 (LTBd), a metabolite formed via the 5-lipoxygenase pathway from arachidonic acid, is one of the most potent chemotactic mediators known. We have investigated the formation of LTBd in a well characterized model of hepatic ischemiareperfusion injury in the rat and made use of a specific leukotriene biosynthesis inhibitor, L663,536, to determine the importance of LTB4 in the pathogenesis of the injury. LTB4 concentrations were measured with a specific and sensitive gas chromatographic-mass spectrometric method previously developed in our laboratory. In liver tissue LTB4 levels were below the detection limit of 20 pg/g before 45 min &hernia and did not increase during the first 6 h of reperfusion. However, at 15 h and 24 h reperfusion LTB4concentrations had increased to levels 50-fold those in control liver (867 f 267 pg/g>. The increase of plasma alanine amminotransferase (ALT) activities indicated two phases of injury, an initial phase during the first few hours of reperfusion, and a second more severe injury phase between 6 h and 24 h reperfusion. PMNs accumulated in tissue throughout the reflow period reaching 700 + 49 per 50 high power fields (HPF) at 24 h. Treatment with L663,536 inhibited LTBd formation by 90% in 24 h post-ischemic liver but had no effect on liver injury, as indicated by plasma ALT activities and liver necrosis. Inhibition of LTBJ biosynthesis also had no effect on PMN accumulation in the liver (782 1 28 per 50 HPF). These results support the conclusion that LTB4 is not a chemotactic mediator in this rat model and appears to play no relevant role in the pathogenesis of the injury.
reperfusion injury, particularly with respect to the mediators responsible for the recruitment and activation of PMNs. Leukotriene B4 (LTB4), an arachidonic acid metabolite that is formed via the 5-lipoxygenase pathway (5), is a potent proinflammatory mediator and has been shown, in vitro, to be one of the most potent chemotactic factors for human neutrophils (6). The importance of this mediator in in vivo models of inflammation has been difficult to assess for a number of reasons. First, low concentrations of LTB4 are difficult to measure in vivo since radioimmunoassay techniques, though showing sufficient sensitivity, often have inadequate specificity where complex biological matrices, such as tissue, are involved (7). Another problem in the assessment of the role of. LTB4 in inflammatory responses in vivo has been the lack of specific inhibitors of the
INTRODUCTION The pathophysiology of &hernia-reperfusion injury has been the focus of numerous studies employing a variety of animal species and many different tissue types. A common feature to most of these models is the accumulation of polymorphonuclear leukocytes (PMNs) into the post-ischemic tissue that occurs during the reperfusion period. Interventions that cause decreased PMN infiltration result in protection against reperfusion injury (l-4). For the development of successful therapeutic strategies it is important to achieve a better understanding of the response inflammatory involved in ischemia-
Date received 17 July 1991 Date accepted 9 September 1991 113
114
Prostaglandins Leukotrienes
and Essential Fatty Acids
5lipoxygenase enzymes. A number of lipoxygenase such as nordihydroguaiaretic acid, inhibitors, BW755C, nafazatrom and AA-861, exhibit antioxidant properties (8) and/or affect neutrophil function non-specifically (9); therefore, any protective effect observed with these compounds could be misinterpreted to suggest the involvement of LTB4 in the pathogenesis of an inflammatory tissue injury. The present study was designed to determine the importance of LTB4 as a mediator of the inflammatory response observed in a model of hepatic ischemia-reperfusion injury in the rat. This model has been well characterized and shown to involve two phases of reperfusion injury (2, 10). The initial injury phase during the first few hours of reperfusion is characterized by a significant oxidant stress in the hepatic sinusoids (11)) with Kupffer cells being the principal source of reactive oxygen (12). During the early reperfusion period PMNs start to accumulate in liver and lung, indicating a general activation of these inflammatory cells (12). The continuous infiltration of neutrophils into the postischemic liver results in a far more severe second injury phase (2). We have applied a recently established highly sensitive gas chromatographic-mass spectrometric (GC-MS) assay (13) to analyze liver LTB4 concentrations in this model of hepatic ischemia-reperfusion injury. To understand the importance of LTB4 as a mediator of this inflammatory response we have made use of a new and highly potent specific leukotriene biosynthesis inhibitor, L663,536 (3-[4-chlorobenzyl)-3-t-butyl-thio5-isopropylindol-2-yl]-22dimethylpropanoic acid) (14) to determine whether suppression of LTB4 synthesis affects neutrophil infiltration into the liver and attenuates the neutrophil-induced postischemic injury.
MATERIALS AND METHODS Animals Male Fischer-344 rats (250-290 g) were purchased from Harlan Sprague-Dawley Inc (Houston, TX) and allowed free access to food (Purina 5001 rodent chow) and tap water. Fed animals were used in ail experiments. A group of animals received L663,536 (Merck-Frosst, Montreal, Canada) suspended in 0.1% glycogen (4 mg/kg body wt; p.0.); the first dose was administered 18 h before ischemia and the second dose 6 h after initiating reperfusion. Control animals received the vehicle only. Hepatic ischemia/reperfusion experiments Ischemia/reperfusion experiments were performed as described (2). Briefly, the animals were anes-
thetized with pentobarbital (60 m&g i.p.). A laparotomy was performed by a midline incision and ligamentous attachments to the liver were divided. The liver hilus was exposed and the blood vessels supplying the median and left lateral hepatic lobes were occluded with an arterial clamp for 45 min. Reflow was initiated by removal of the clamp. The incision was closed with 3-O silk and wound clips and the animals were allowed to recover. The animals were sacrificed 24 h after ischemia. At the end of the experiment a blood sample was obtained from the right ventricle for the determination of alanine aminotransferase activity (ALT) and a sample of a nonischemic lobe and a postischemic lobe was freeze-clamped and stored in liquid nitrogen and another sample was fixed in phosphate-buffered formalin. LTB4 generation in whole blood An aliquot of blood (500 ~1) was obtained by retroorbital puncture under pentobarbital anesthesia from L663,536- or vehicle-pretreated animals before ischemia and after 24 h of reperfusion. The calcium ionophore A23187 was added in 10 ~1 ethanol to each blood sample resulting in a final concentration of 100 PM. The samples were incubated at room temperature for 60 min. The incubation was stopped by rapid centrifugation of the blood in an Eppendorf Microfuge and freezing the supernatant in liquid nitrogen. Analysis of LTB4 LTB4 content of liver tissue was measured by GCMS using a modification of our previously published method (7, 13). Frozen tissue was thawed and homogenized in 4 vol of water containing nordihydroguaiaretic acid (50 PM) to prevent ex vivo formation of LTB4. A 2 ml aliquot of the homogenate was mixed with water and ethanol to give a final volume of 15 ml ethanol/water (75:25) and the internal standard 11-,12-,14-,15-tetradeuterioleukotriene B4 (21-&-LTB4, 150 pg) added. Samples were centrifuged, the supernatant removed, and water added to bring the composition to 25% ethanol prior to C-18 cartridge (Sep-Pak, Waters Division of Millipore, MA, USA) extraction as previously described (7). The resulting ethyl acetate extract was evaporated, 1 ml potassium hydroxide (0.1 M), added, and the sample washed with 5 ml hexane. The pH was adjusted to 3 with HCl (2 M) and LTB4 extracted with 4 ml hexane/dichloromethane (1: 1). This extract was derivatized to the pentafluorobenzyl (PFB) ester and further purified by HPLC prior to formation of the bis-tert-butyldimethylsilyl (TBDMS) ether derivative as previously described (13). GC-MS analysis of the
LTB, and
samples at this stage, which had previously proved successful in heart tissue (7), indicated interfering peaks were present in the chromatogram. Samples were therefore further purified by normal phase HPLC as the PFB ester 6is-TBDMS ether derivative with mobile phase containing 0.075% iso-propanol in hexane (flow 2 ml/min) and a silica column (Econosphere 5 micron, 250 mm x 4.6 mm i.d., Alltech Associates). The LTB4 derivative eluted at approximately 7 min in this system. HPLC fractions were evaporated and redissolved in 15 ~1 hexane prior to GC-MS analysis. For the analysis of LTB,, in ionophore-stimulated blood, a shorter extraction and clean-up procedure was used since the concentration of LTB4 was much higher in these samples than in tissue. Briefly, 100~1 of plasma was diluted with 500 ~1 water plus 25 ~1 2 M HCI and extracted with 2 x 4 ml ethyl acetate. This extract was treated in the same manner as the C-18 cartridge extract above, but following the first HPLC purification step the samples were derivatized to their trimethylsilyl ethers by heating in N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) containing 1% trimethylchlorosilane (25 ~1) for 1 h at 60”. The reagents were evaporated to dryness and the sample redissolved in hexane for GC-MS analysis. Samples were analyzed on either a Ribermag RIO-1OC mass spectrometer (Delsi Nermag Inc, Houston, TX, USA) as previously described (13) or on a TRIO-l mass spectrometer (VG Masslabs. Manchester, UK). A bonded-phase DB-1 capillary column (30 m x 0.32 mm, id, J & W Scientific, Ranch0 Cordova, CA, USA) was used for the analyses in both instruments with an all glass moving needle injector system. The oven was heated from 200°C to 300°C at S”C/min. The mass spectrometers were operated in the negative ion mode and ions corresponding to loss of tert-butyldimethylsilanol from the [M-PFBJ- ions monitored at m/z 431 and m/z 435 for the TBDMS-derivatives of endogenous and 2&-LTB4 respectively. The trimethylsilyl-derivatives were analyzed by monitoring their [M-PFB]- ions at m/z 479 and m/z 483. Quantitation was accomplished with standard curves derived from extracted standards.
Hepatic Ischemia-ReperfusionInjury in the Rat
115
paraffin-embedded and 5-micron thick sections were cut. Neutrophils were stained employing the naphthol AS-D chloroacetate esterase technique (17) as described (2). Neutrophils were identified by positive staining and morphology and counted in 50 high power fields using a Nikon Labophot Microscope. The percentage of necrotic areas was estimated by evaluating parallel sections stained with hematoxilin and eosin. All chemicals and drugs were purchased from Sigma Chemical Co (St. Louis, MO, USA) or Fisher Scientific (Houston, TX, USA). Statistics All data are expressed as the mean + SEM. Comparisons of data sets were performed with the paired or unpaired Student’s t-test.
RESULTS The GC-MS method for the analysis of LTB4 allows the low concentrations of this lipid mediator that are present in vivo to be accurately assessed. Figure 1 illustrates the selected ion monitoring chromatogram for endogenous and *HA-LTB4, extracted from a 24 h post-ischemic liver sample. An extensive clean-up procedure proved successful in removing interferences from biological contaminants. LTB4 concentrations in liver were below the 20 27 1
Other analyses Total soluble glutathione (GSH and GSSG) was measured in the acidic homogenate from freezeclamped livers by the method of Tietze (15) as described in detail (16). The ATP content was also determined in the supernatant of the same homogenate with Sigma test kit 366-UV. Plasma alanine aminotransferase activities were measured with Sigma test kit DG159-UV and expressed in IV/l. Formalin-fixed portions of the liver were
I-...,“.‘,“--.-
18.0
20.0
22.8
-.r
tiM+niin Selected ion chromatogram for endogenous (m/z 431) and ‘H,-LTB, (m/z 435) as their PFB ester bb-TBDMS ether derivatives. This sample was extracted from liver following 45 min ischemia and 24 h reflow and subjected to extensive clean-up as describes in the Methods section prior to GC-MS analysis. Fig. 1
116
Prostaglandins Leukotrienes
and Essential Fatty Acids
eZaLTk+ h/d
1000
0
PMN (50
100
0
HPF)
PMN (50
HPF)
50
100
10 i_
1
x Pre-I
1hRP
6h
0 P
15h
F’
24h
c 6h RP
Rf
24h
RP
The hepatic LTB, content and the number of neutrophils present in the nonischemic liver lobes were determined before ischemia (Pre-I) and at 6 and 24 h of reperfusion. Data are given as mean + SEM of n = 4 animals per time point. LTB, was below the detection limit of 20 pg/ml in all samples.
Fig. 2
The hepatic LTB, content and the number of neutrophils present in the liver lobes were determined before hepatic ischemia and in the postischemic liver lobes at various time points during reperfusion. Data are given as mean + SEM of n = 5-10 animals per time point.
Fig. 4
detection limit of 20 pg/g in non-ischemic tissue and at 1 h and 6 h reperfusion following 45 min normothermic ischemia. However, at 15 h and 24 h reperfusion, LTB4 concentrations increased to levels 50-fold control values (Fig. 2). This time-course for LTB4 formation was interesting particularly with respect to that of injury and neutrophil accumulation. Figure 3 shows the plasma ALT activities, as an index for liver cell injury, at various times of reperfusion. There is an initial injury phase during the first hour of reperfusion followed by a second injury phase between 6 and 24 h of reperfusion. Previous studies suggest that Kupffer cells, the resident macrophages of the liver, are mainly involved in the initial injury (12) and neutrophils are primarily responsible for the second injury phase (2). Neutrophils progressively accumulated in the liver during the entire reperfusion period although there is no evidence for enhanced LTB4 production in the postischemic liver up to 6 h of reperfusion (Fig. 2). Neutrophils also accumulate temporarily in the nonischemic liver lobes, however, no injury was
detected in these lobes (2, 12). This suggests that the neutrophils might have been activated for increased adherence but not to produce cytotoxic mediators. To determine whether neutrophils in the nonischemic lobes generate LTB4, we measured LTB4 in these lobes after 6 h of reperfusion, where the maximal number of neutrophils was found, and at 24 h of reperfusion. As shown in Figure 4, there was no enhanced LTB4 generation in the nonischemic liver lobes, despite the temporary accumulation of neutrophils in these lobes further supporting the hypothesis that LTB4 was not a chemotactic factor in this model. To determine whether LTB4 plays a role as a critical mediator in the pathogenesis of ischemiareperfusion injury, the orally active leukotriene biosynthesis inhibitor, L663,536, was administered to rats 18 h prior to ischemia and after 6 h reperfusion. Initial studies were carried out with ionophore-stimulated blood to determine the biosynthesis of LTB4 was indeed inhibited after 24 h reperfusion. Incubation with A23187 (100 PM) resulted in a LTB4 concentration in blood of vehicle-treated animals of 10.9 * 1.1 rig/ml plasma after 60 min. Blood LTB4 generation was reduced by 90% after L663,536 pretreatment (1.1 + 0.2 ng/ml), indicating LTB4 synthesis was effectively inhibited for the entire reperfusion period. A similar incubation of blood without ionophore caused only a minor generation of LTB4 (
Role of Leukotriene B4 in the Pathogenesis of Hepatic IschemiaReperfusion Injury in the Rat H. Hughes, A. Farhood*
and H. Jaeschke
Center for Experimental Therapeutics, Department of Medicine, Baylor College of Medicine, and *Department of Pathology, The University of Texas Medical School, Houston, Texas 77030, USA (Reprint requests to AF) ABSTRACT.
A common feature to most models of &hernia-reperfusion injury is the accumulation of polymorphonuclear leukocytes (PMNs) into the post-ischemic tissue during the reperfusion period. Interventions that lead to decreased PMN inliltration protect against tissue injury and therefore a knowledge of the chemotactic mediators leading to PMN accumulation is essential to understanding the pathogenesis of the injury and to the development of successful therapeutic strategies. Leukotriene B4 (LTBd), a metabolite formed via the 5-lipoxygenase pathway from arachidonic acid, is one of the most potent chemotactic mediators known. We have investigated the formation of LTBd in a well characterized model of hepatic ischemiareperfusion injury in the rat and made use of a specific leukotriene biosynthesis inhibitor, L663,536, to determine the importance of LTB4 in the pathogenesis of the injury. LTB4 concentrations were measured with a specific and sensitive gas chromatographic-mass spectrometric method previously developed in our laboratory. In liver tissue LTB4 levels were below the detection limit of 20 pg/g before 45 min &hernia and did not increase during the first 6 h of reperfusion. However, at 15 h and 24 h reperfusion LTB4concentrations had increased to levels 50-fold those in control liver (867 f 267 pg/g>. The increase of plasma alanine amminotransferase (ALT) activities indicated two phases of injury, an initial phase during the first few hours of reperfusion, and a second more severe injury phase between 6 h and 24 h reperfusion. PMNs accumulated in tissue throughout the reflow period reaching 700 + 49 per 50 high power fields (HPF) at 24 h. Treatment with L663,536 inhibited LTBd formation by 90% in 24 h post-ischemic liver but had no effect on liver injury, as indicated by plasma ALT activities and liver necrosis. Inhibition of LTBJ biosynthesis also had no effect on PMN accumulation in the liver (782 1 28 per 50 HPF). These results support the conclusion that LTB4 is not a chemotactic mediator in this rat model and appears to play no relevant role in the pathogenesis of the injury.
reperfusion injury, particularly with respect to the mediators responsible for the recruitment and activation of PMNs. Leukotriene B4 (LTB4), an arachidonic acid metabolite that is formed via the 5-lipoxygenase pathway (5), is a potent proinflammatory mediator and has been shown, in vitro, to be one of the most potent chemotactic factors for human neutrophils (6). The importance of this mediator in in vivo models of inflammation has been difficult to assess for a number of reasons. First, low concentrations of LTB4 are difficult to measure in vivo since radioimmunoassay techniques, though showing sufficient sensitivity, often have inadequate specificity where complex biological matrices, such as tissue, are involved (7). Another problem in the assessment of the role of. LTB4 in inflammatory responses in vivo has been the lack of specific inhibitors of the
INTRODUCTION The pathophysiology of &hernia-reperfusion injury has been the focus of numerous studies employing a variety of animal species and many different tissue types. A common feature to most of these models is the accumulation of polymorphonuclear leukocytes (PMNs) into the post-ischemic tissue that occurs during the reperfusion period. Interventions that cause decreased PMN infiltration result in protection against reperfusion injury (l-4). For the development of successful therapeutic strategies it is important to achieve a better understanding of the response inflammatory involved in ischemia-
Date received 17 July 1991 Date accepted 9 September 1991 113
114
Prostaglandins Leukotrienes
and Essential Fatty Acids
5lipoxygenase enzymes. A number of lipoxygenase such as nordihydroguaiaretic acid, inhibitors, BW755C, nafazatrom and AA-861, exhibit antioxidant properties (8) and/or affect neutrophil function non-specifically (9); therefore, any protective effect observed with these compounds could be misinterpreted to suggest the involvement of LTB4 in the pathogenesis of an inflammatory tissue injury. The present study was designed to determine the importance of LTB4 as a mediator of the inflammatory response observed in a model of hepatic ischemia-reperfusion injury in the rat. This model has been well characterized and shown to involve two phases of reperfusion injury (2, 10). The initial injury phase during the first few hours of reperfusion is characterized by a significant oxidant stress in the hepatic sinusoids (11)) with Kupffer cells being the principal source of reactive oxygen (12). During the early reperfusion period PMNs start to accumulate in liver and lung, indicating a general activation of these inflammatory cells (12). The continuous infiltration of neutrophils into the postischemic liver results in a far more severe second injury phase (2). We have applied a recently established highly sensitive gas chromatographic-mass spectrometric (GC-MS) assay (13) to analyze liver LTB4 concentrations in this model of hepatic ischemia-reperfusion injury. To understand the importance of LTB4 as a mediator of this inflammatory response we have made use of a new and highly potent specific leukotriene biosynthesis inhibitor, L663,536 (3-[4-chlorobenzyl)-3-t-butyl-thio5-isopropylindol-2-yl]-22dimethylpropanoic acid) (14) to determine whether suppression of LTB4 synthesis affects neutrophil infiltration into the liver and attenuates the neutrophil-induced postischemic injury.
MATERIALS AND METHODS Animals Male Fischer-344 rats (250-290 g) were purchased from Harlan Sprague-Dawley Inc (Houston, TX) and allowed free access to food (Purina 5001 rodent chow) and tap water. Fed animals were used in ail experiments. A group of animals received L663,536 (Merck-Frosst, Montreal, Canada) suspended in 0.1% glycogen (4 mg/kg body wt; p.0.); the first dose was administered 18 h before ischemia and the second dose 6 h after initiating reperfusion. Control animals received the vehicle only. Hepatic ischemia/reperfusion experiments Ischemia/reperfusion experiments were performed as described (2). Briefly, the animals were anes-
thetized with pentobarbital (60 m&g i.p.). A laparotomy was performed by a midline incision and ligamentous attachments to the liver were divided. The liver hilus was exposed and the blood vessels supplying the median and left lateral hepatic lobes were occluded with an arterial clamp for 45 min. Reflow was initiated by removal of the clamp. The incision was closed with 3-O silk and wound clips and the animals were allowed to recover. The animals were sacrificed 24 h after ischemia. At the end of the experiment a blood sample was obtained from the right ventricle for the determination of alanine aminotransferase activity (ALT) and a sample of a nonischemic lobe and a postischemic lobe was freeze-clamped and stored in liquid nitrogen and another sample was fixed in phosphate-buffered formalin. LTB4 generation in whole blood An aliquot of blood (500 ~1) was obtained by retroorbital puncture under pentobarbital anesthesia from L663,536- or vehicle-pretreated animals before ischemia and after 24 h of reperfusion. The calcium ionophore A23187 was added in 10 ~1 ethanol to each blood sample resulting in a final concentration of 100 PM. The samples were incubated at room temperature for 60 min. The incubation was stopped by rapid centrifugation of the blood in an Eppendorf Microfuge and freezing the supernatant in liquid nitrogen. Analysis of LTB4 LTB4 content of liver tissue was measured by GCMS using a modification of our previously published method (7, 13). Frozen tissue was thawed and homogenized in 4 vol of water containing nordihydroguaiaretic acid (50 PM) to prevent ex vivo formation of LTB4. A 2 ml aliquot of the homogenate was mixed with water and ethanol to give a final volume of 15 ml ethanol/water (75:25) and the internal standard 11-,12-,14-,15-tetradeuterioleukotriene B4 (21-&-LTB4, 150 pg) added. Samples were centrifuged, the supernatant removed, and water added to bring the composition to 25% ethanol prior to C-18 cartridge (Sep-Pak, Waters Division of Millipore, MA, USA) extraction as previously described (7). The resulting ethyl acetate extract was evaporated, 1 ml potassium hydroxide (0.1 M), added, and the sample washed with 5 ml hexane. The pH was adjusted to 3 with HCl (2 M) and LTB4 extracted with 4 ml hexane/dichloromethane (1: 1). This extract was derivatized to the pentafluorobenzyl (PFB) ester and further purified by HPLC prior to formation of the bis-tert-butyldimethylsilyl (TBDMS) ether derivative as previously described (13). GC-MS analysis of the
LTB, and
samples at this stage, which had previously proved successful in heart tissue (7), indicated interfering peaks were present in the chromatogram. Samples were therefore further purified by normal phase HPLC as the PFB ester 6is-TBDMS ether derivative with mobile phase containing 0.075% iso-propanol in hexane (flow 2 ml/min) and a silica column (Econosphere 5 micron, 250 mm x 4.6 mm i.d., Alltech Associates). The LTB4 derivative eluted at approximately 7 min in this system. HPLC fractions were evaporated and redissolved in 15 ~1 hexane prior to GC-MS analysis. For the analysis of LTB,, in ionophore-stimulated blood, a shorter extraction and clean-up procedure was used since the concentration of LTB4 was much higher in these samples than in tissue. Briefly, 100~1 of plasma was diluted with 500 ~1 water plus 25 ~1 2 M HCI and extracted with 2 x 4 ml ethyl acetate. This extract was treated in the same manner as the C-18 cartridge extract above, but following the first HPLC purification step the samples were derivatized to their trimethylsilyl ethers by heating in N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) containing 1% trimethylchlorosilane (25 ~1) for 1 h at 60”. The reagents were evaporated to dryness and the sample redissolved in hexane for GC-MS analysis. Samples were analyzed on either a Ribermag RIO-1OC mass spectrometer (Delsi Nermag Inc, Houston, TX, USA) as previously described (13) or on a TRIO-l mass spectrometer (VG Masslabs. Manchester, UK). A bonded-phase DB-1 capillary column (30 m x 0.32 mm, id, J & W Scientific, Ranch0 Cordova, CA, USA) was used for the analyses in both instruments with an all glass moving needle injector system. The oven was heated from 200°C to 300°C at S”C/min. The mass spectrometers were operated in the negative ion mode and ions corresponding to loss of tert-butyldimethylsilanol from the [M-PFBJ- ions monitored at m/z 431 and m/z 435 for the TBDMS-derivatives of endogenous and 2&-LTB4 respectively. The trimethylsilyl-derivatives were analyzed by monitoring their [M-PFB]- ions at m/z 479 and m/z 483. Quantitation was accomplished with standard curves derived from extracted standards.
Hepatic Ischemia-ReperfusionInjury in the Rat
115
paraffin-embedded and 5-micron thick sections were cut. Neutrophils were stained employing the naphthol AS-D chloroacetate esterase technique (17) as described (2). Neutrophils were identified by positive staining and morphology and counted in 50 high power fields using a Nikon Labophot Microscope. The percentage of necrotic areas was estimated by evaluating parallel sections stained with hematoxilin and eosin. All chemicals and drugs were purchased from Sigma Chemical Co (St. Louis, MO, USA) or Fisher Scientific (Houston, TX, USA). Statistics All data are expressed as the mean + SEM. Comparisons of data sets were performed with the paired or unpaired Student’s t-test.
RESULTS The GC-MS method for the analysis of LTB4 allows the low concentrations of this lipid mediator that are present in vivo to be accurately assessed. Figure 1 illustrates the selected ion monitoring chromatogram for endogenous and *HA-LTB4, extracted from a 24 h post-ischemic liver sample. An extensive clean-up procedure proved successful in removing interferences from biological contaminants. LTB4 concentrations in liver were below the 20 27 1
Other analyses Total soluble glutathione (GSH and GSSG) was measured in the acidic homogenate from freezeclamped livers by the method of Tietze (15) as described in detail (16). The ATP content was also determined in the supernatant of the same homogenate with Sigma test kit 366-UV. Plasma alanine aminotransferase activities were measured with Sigma test kit DG159-UV and expressed in IV/l. Formalin-fixed portions of the liver were
I-...,“.‘,“--.-
18.0
20.0
22.8
-.r
tiM+niin Selected ion chromatogram for endogenous (m/z 431) and ‘H,-LTB, (m/z 435) as their PFB ester bb-TBDMS ether derivatives. This sample was extracted from liver following 45 min ischemia and 24 h reflow and subjected to extensive clean-up as describes in the Methods section prior to GC-MS analysis. Fig. 1
116
Prostaglandins Leukotrienes
and Essential Fatty Acids
eZaLTk+ h/d
1000
0
PMN (50
100
0
HPF)
PMN (50
HPF)
50
100
10 i_
1
x Pre-I
1hRP
6h
0 P
15h
F’
24h
c 6h RP
Rf
24h
RP
The hepatic LTB, content and the number of neutrophils present in the nonischemic liver lobes were determined before ischemia (Pre-I) and at 6 and 24 h of reperfusion. Data are given as mean + SEM of n = 4 animals per time point. LTB, was below the detection limit of 20 pg/ml in all samples.
Fig. 2
The hepatic LTB, content and the number of neutrophils present in the liver lobes were determined before hepatic ischemia and in the postischemic liver lobes at various time points during reperfusion. Data are given as mean + SEM of n = 5-10 animals per time point.
Fig. 4
detection limit of 20 pg/g in non-ischemic tissue and at 1 h and 6 h reperfusion following 45 min normothermic ischemia. However, at 15 h and 24 h reperfusion, LTB4 concentrations increased to levels 50-fold control values (Fig. 2). This time-course for LTB4 formation was interesting particularly with respect to that of injury and neutrophil accumulation. Figure 3 shows the plasma ALT activities, as an index for liver cell injury, at various times of reperfusion. There is an initial injury phase during the first hour of reperfusion followed by a second injury phase between 6 and 24 h of reperfusion. Previous studies suggest that Kupffer cells, the resident macrophages of the liver, are mainly involved in the initial injury (12) and neutrophils are primarily responsible for the second injury phase (2). Neutrophils progressively accumulated in the liver during the entire reperfusion period although there is no evidence for enhanced LTB4 production in the postischemic liver up to 6 h of reperfusion (Fig. 2). Neutrophils also accumulate temporarily in the nonischemic liver lobes, however, no injury was
detected in these lobes (2, 12). This suggests that the neutrophils might have been activated for increased adherence but not to produce cytotoxic mediators. To determine whether neutrophils in the nonischemic lobes generate LTB4, we measured LTB4 in these lobes after 6 h of reperfusion, where the maximal number of neutrophils was found, and at 24 h of reperfusion. As shown in Figure 4, there was no enhanced LTB4 generation in the nonischemic liver lobes, despite the temporary accumulation of neutrophils in these lobes further supporting the hypothesis that LTB4 was not a chemotactic factor in this model. To determine whether LTB4 plays a role as a critical mediator in the pathogenesis of ischemiareperfusion injury, the orally active leukotriene biosynthesis inhibitor, L663,536, was administered to rats 18 h prior to ischemia and after 6 h reperfusion. Initial studies were carried out with ionophore-stimulated blood to determine the biosynthesis of LTB4 was indeed inhibited after 24 h reperfusion. Incubation with A23187 (100 PM) resulted in a LTB4 concentration in blood of vehicle-treated animals of 10.9 * 1.1 rig/ml plasma after 60 min. Blood LTB4 generation was reduced by 90% after L663,536 pretreatment (1.1 + 0.2 ng/ml), indicating LTB4 synthesis was effectively inhibited for the entire reperfusion period. A similar incubation of blood without ionophore caused only a minor generation of LTB4 (
