- Email: [email protected]
Glutathione and Ultrastructural Changes in Inflow Occlusion of Rat Liver
Journal of Surgical Research 88, 207–214 (2000) doi:10.1006/jsre.1999.5781, available online at http://www.idealibrary.com on
Glutathione and Ultrastructural Changes in Inflow Occlusion of Rat Liver T. Armeni, R. Ghiselli,* G. Balercia,* ,† L. Goffi,* W. Jassem, V. Saba,* and G. Principato Institute of Biology and Genetics, *Clinica di Semeiotica e Metodologia Chirurgica, and †Division of Endocrinology, Institute of Internal Medicine, University of Ancona–INRCA, I-60100 Ancona, Italy Submitted for publication June 9, 1999
INTRODUCTION Background. Liver ischemia/reperfusion is frequently associated with organ injury to which reactive oxygen species contribute. The aim of our study was to evaluate cytosolic and mitochondrial glutathione levels and morphological changes in hepatocytes of rat liver in an experimental model of ischemia/reperfusion. Materials and methods. The experimental procedure consisted of temporary interruption of blood flow to the left lateral and medial hepatic lobes for different lengths of time and, in some cases, subsequent reperfusion. Cytosolic and mitochondrial glutathione levels were evaluated and ultrastructural analysis was carried out for all samples. Results. Ischemic lobes showed ultrastructural changes in relationship with the increase in ischemia time. Total glutathione levels did not show variations in ischemic lobes and sham lobes with respect to control rats during ischemia only. Instead, during reperfusion, significant ultrastructural alterations of the hepatocytes and a significant depletion of glutatione in cytosolic and mitochondrial compartments were evident. The sham lobes also showed a significant glutathione decrement. Increased oxidized glutathione (GSSG) levels were found during ischemia both in ischemic lobes and in sham lobes. During reperfusion GSSG was found to a minor extent, in the cytosolic compartment. In mitochondria GSSG levels were also high during reperfusion. Conclusions. We conclude that depletion of glutathione contributes to impaired liver after reperfusion following ischemia but depletion of glutathione alone does not induce changes in the morphology of the hepatocytes. Glutathione depletion and a greater quantity of GSSG, even in sham lobes, may indicate a metabolic alteration which spreads to compartments that are not involved in ischemia/reperfusion.
Hepatic resections in case of tumor or injury require continuous total hepatic inflow interruption; the application of Pringle’s maneuver, with cross clamping of the epatoduodenal ligament, has been shown to minimize blood loss. It has been widely reported that the maximal safe period of total hepatic occlusion is 1 h; longer periods of hepatic ischemia may produce profound hepatocyte injury with induction of liver failure and death, in particular in patients with underlying cirrhosis [1]. It is generally believed that the hepatic ischemia/ reperfusion injury after Pringle’s maneuver [2] is due to two different events: ● hepatic ischemia and subsequent reperfusion with arterial blood, and ● portal venous congestion and subsequent hepatic reperfusion with congested portal blood.
However, the specific roles of these mechanisms have not yet been well documented. Ischemia/reperfusion injury is evident through progressive ultrastructural changes in rat liver after 30 – 120 min of ischemia, with evident mitochondrial swelling after 60 min of ischemia. The alterations become more severe during reperfusion, although a partial reversibility of tissue injury has been observed after 2 h of blood supply restoration in rat liver following 60 min of ischemia [3]. Increasing clinical and experimental data have provided evidence that injury derived from ischemia/ reperfusion (I/R) 1 is mediated by reactive oxygen species: superoxide O 2⫺•, hydrogen peroxide H 2O 2, and
© 2000 Academic Press
Key Words: glutathione; rat liver; ultrastructural; ischemia/reperfusion.
1 Abbreviations used: GSH, glutathione; GSSG, oxidized glutathione; I/R, ischemia/reperfusion; TNF-␣, tumor necrosis factor-␣.
207
0022-4804/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
208
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000
hydroxyl radical •OH [4 – 6]. This phenomenon has been referred to as the “oxygen paradox” [7], implying that although the restoration of the supply of ordinary molecular oxygen to ischemic tissue is obviously necessary to restore normal function, reactive oxygen species generated in postischemic tissue may simultaneously participate in certain deleterious chemical reactions that worsen tissue injury [8]. Reduced glutathione (GSH), one of the main components of antioxidant defense, is a cofactor of several enzymes and plays an important role in the regulation of cell redox reactions as well as detoxification [9]. High levels of glutathione are distributed in two separate pools in the cell, 85–90% in the cytosol and 10 –15% in the mitochondria [10]. Depletion of cytosolic GSH down to critical levels can still be compatible with cell survival, whereas a decrease of mitochondrial GSH is critical for cell vitality [11]. The aim of this study is to evaluate cytosolic and mitochondrial glutathione levels after various periods of warm ischemia and reperfusion in a rat liver model. The morphological features of the hepatic tissue during ischemia–reperfusion have also been investigated. METHODS Liver ischemia/reperfusion model. Male Winstar rats (200 –250 g) were supplied by INRCA (Istituto Nazionale Ricerca e Cura Anziani, Ancona, Italy). Animals were housed under standard conditions with normal lighting cycle and with free access to food and water. Rats were anesthetized with ether. The experimental procedure consisted in temporary interruption of blood flow to the left lateral and medial lobes of the liver and, in some cases, subsequent reperfusion. The abdomen was opened by a midline incision and the liver hilus was exposed. All the vessels (hepatic artery, portal vein, and bile duct) to the left lateral and medial hepatic lobes were occluded. The portal branch to the right lobe was left open to avoid stasis in the portal system. These lobes constitute the sham lobes. In some cases after periods of ischemia of differing duration, the left and median lobes were reperfused. The rats were divided into seven experimental groups: Group C (n ⫽ 6), control rats, received neither ischemia nor reperfusion (Control). Group 1 (n ⫽ 8) rats received 15 min of ischemia (right lobe, S-15; left lateral and medial lobes I-15). Group 2 (n ⫽ 8) rats received 30 min of ischemia (right lobe, S-30; left lateral and medial lobes, I-30). Group 3 (n ⫽ 8) rats received 60 min of ischemia (right lobe, S-60; left lateral and medial lobes, I-60). Group 4 (n ⫽ 6) rats received 15 min of ischemia followed by 1 min of reperfusion (right lobe, S-15/R-1; left lateral and medial lobes, I-15/R-1). Group 5 (n ⫽ 8) rats received 15 min of ischemia followed by 15 min of reperfusion (right lobe, S-15/R-15; left lateral and medial lobes, I-15/R-15). Group 6 (n ⫽ 8) rats received 60 min of ischemia followed by 15 min of reperfusion (right lobe, S-60/R-15; left lateral and medial lobes, I-60/R15). At the end of each procedure a portion of the liver was rapidly removed both from the nonischemic lobe (sham) and from the ischemic lobe or the reperfused lobe. Liver samples were used for elec-
tron microscopy analysis and for the preparation of mitochondrial and cytosolic fractions in order to evaluate their glutathione content. Groups
Ischemic reperfusion lobe
Sham lobe
C 1 2 3 4 5 6
C I-15 I-30 I-60 I-15/R-1 I-15/R-15 I-60/R-15
C S-15 S-30 S-60 S-15/R-1 S-15/R-15 S-60/R-15
Ultrastructural investigation. Liver samples for the ultrastructural investigation were control, sham and ischemic, sham and reperfused. Liver biopsies were randomly collected at the end time of ischemia and reperfusion. Specimens underwent technical procedures for electron microscopy as described [12]. Briefly, they were fixed in 2% glutaraldehyde (4°C, pH 7.4), postfixed in 1% osmium tetroxide, and embedded in an Epon–araldite mixture. Semithin sections were stained with toluidine blue. Thin sections, stained with uranyl acetate and lead citrate, were observed with a Zeiss EM 10 electron microscope. Mitochondrial isolation. Mitochondrial and cytosolic fractions for glutathione measurements were obtained as follows: aliquots of liver tissue (4 –5 g) were homogenized 1:10 (w/v) in ice-cold buffer, pH 7.5, containing 75 mM sucrose, 225 mM mannitol, 1 mM EDTA, 5 mM Hepes, and 0.5 mg/ml fatty acid-free bovine serum albumin. The homogenate was centrifuged for 10 min at 600g at 4°C. Sediment was discarded and supernatant was centrifuged for 20 min at 1200g at 4°C, obtaining a mitochondrial pellet and a supernatant. The latter was centrifuged for 30 min at 55,000g at 4°C and the resulting supernatant (cytosolic fraction) was divided into 1-ml aliquots. The mitochondrial pellet was washed two times (by careful resuspensions in ice-cold homogenization buffer and centrifugation, 10 min at 2800g at 4°C) and purified mitochondria were carefully resuspended in 1 ml of ice-cold homogenization buffer. Sample preparation for glutathione assay. Aliquots of mitochondrial and cytosolic fractions were immediately deproteinized in icecold perchloric acid or sulfosalicylic acid/vinylpyridine for the determination of total glutathione or GSSG, respectively. The addition of vinylpyridine was necessary to form a covalent derivative with GSH. The mixtures were centrifuged (15,000g for 15 min at 4°C) and supernatants stored at ⫺80°C until used for glutathione determinations. Pellets were dissolved in 1 M sodium hydroxide and used for protein determinations [13] using bovine serum albumin as standard. Perchloric acid was removed from mitochondrial and cytosolic deproteinized samples by neutralization with potassium carbonate. Sulfosalicylic acid was neutralized with triethanolamine. Glutathione assay. Glutathione (GSH⫹2GSSG) was measured by the glutathione reductase recycling assay at 412 nm in the presence of 5,5⬘-dithio-bis-nitrobenzoic acid [14]. The amount of total glutathione is reported as nmol/mg of protein of the respective fractions. Statistical analysis. Results are reported as means ⫾ standard deviation. The statistical significance between groups was assessed by computer-assisted analysis of variance.
RESULTS
The main ultrastructural features detected at each time of the experimental protocol of ischemia/reperfusion of rat liver are shown in Table 1 and Fig. 1. In control livers normal cell patterns were detected after fine structural investigations, with substantial integrity of both nuclear and cytoplasmic compart-
209
ARMENI ET AL.: GSH AND MORPHOLOGY IN ISCHEMIA/REPERFUSION
TABLE 1 Main Ultrastructural Features of Rat Liver Exposed to Warm Ischemia or Warm Ischemia Followed by Reperfusion (A) Ischemia (min) Injury Hepatocytes Internal membrane dilatation (smooth and rough endoplasmic reticulum, Golgi apparatus) Mitochondrial swelling, matrix clarification, reduction of cristae and spatial disruption of their organization Glycogen lack Large electron-lucent vacuoles devoid of organelles Bile canaliculi dilatation Disse’s space dilatation and microvilli reduction Neutrophilic infiltration Other components Endothelial cell swelling Endothelial necrosis and basal lamina exposure Kupffer cell activation
(B) Ischemia/ reperfusion
(C) Sham
I-15
I-30
I-60
I15-60/R15
S-15-60/R15
⫹/⫺
⫹
⫹/⫹⫹
⫹⫹⫹⫹
⫹/⫺
⫹/⫺
⫹
⫹/⫹⫹
⫹⫹⫹⫹
⫹/⫺
⫹ ⫹/⫺ ⫺ ⫺ ⫺
⫹ ⫹/⫺ ⫺ ⫹/⫺ ⫹/⫺
⫹/⫹⫹ ⫹/⫺ ⫹/⫺ ⫹ ⫹/⫺
⫹⫹/⫹⫹⫹ ⫹⫹⫹ ⫹⫹ ⫹⫹⫹ ⫹/⫹⫹
⫺ ⫹/⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺
⫹/⫺ ⫺ ⫺
⫹/⫺ ⫹⫹ ⫹/⫺
⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹
⫺ ⫺ ⫺
Note. Negative (as in control liver); ⫹, ⫹⫹, and ⫹⫹⫹, weak, moderate, and strong positivity, respectively, in comparison with control liver. Data are compared with those of control livers not exposed to treatment. Results are referred to hepatic tissue after 15, 30, and 60 min of warm ischemia (I-15, I-30, I-60) (column A), left lateral and medial liver lobes after 15 or 60 min of ischemia followed by 15 min of reperfusion (one group, I15-60/R-15, because of very similar results at the two different ischemia times) (column B), and sham tissue of the same rats (one group, S-15-60/R-15, because of very similar results at the two different ischemia times) (column C). Liver specimens were fixed in 2% glutaraldehyde (4°C, pH 7.4), postfixed in 1% osmium tetroxide, and embedded in an Epon–araldite mixture. Semithin sections were stained with toluidine blue. Thin sections, stained with uranyl acetate and lead citrate were observed with a Zeiss EM 10 electron microscope.
ments of hepatocytes. Vascular districts, as well as Kupffer cells, were unaltered. In the group of ischemic livers (Table 1, column A) some ultrastructural changes were detected at a cytoplasmic level after as soon as 15 min of ischemia (I-15). In particular, the endoplasmic reticulum (smooth and rough) was focally dilated, as was the Golgi apparatus; mitochondria were focally swollen with matrix clarification. Some areas devoid of glycogen were also evident. Cell alterations became more evident after 30 and 60 min of ischemia (I-30 and I-60). Mitochondria were focally swollen, and a reduction in the number of cristae became evident. The lack of glycogen was also more evident, in relationship to the increase in ischemia duration. Some electron-lucent areas, devoid of organelles, were found in the cell cytoplasm. Moreover, Disse’s spaces were consistently dilated, while endothelial and Kupffer cells showed some electron-dense vacuoles in their cytoplasm, especially after 60 min of ischemia (Fig. 1B). A series of severe diffuse alterations was evident at a cellular level in the group of livers reperfused for 15 min after 15– 60 min of warm ischemia (I-15-60/R-15) (Table 1, column B). A diffuse dilatation of internal membranes was evident; mitochondria were also diffusely swollen, cristae were reduced in number, and their spatial organization appeared disrupted. Large cytoplasmic electron-lucent vacuoles, devoid of organelles, were evident, probably reflecting an intracellular edema. Dilatation of bile
canaliculi and thick subplasmalemmatic bundles of microfilaments were found. Disse’s spaces were markedly dilated (edematous) and a marked reduction of hepatocyte microvillous projections toward the sinusoids was frequently evident. Neutrophils were also seen in the hepatic lobules. Endothelial cells were swollen, with a clarification of cytoplasmic matrix and a large number of electron-lucent vacuoles (Fig. 1C). Some of them were necrotic, with intraluminal exposure of the basal lamina. A swelling of Kupffer cells, the cytoplasm of which contained some dense vacuoles, was also evident, reflecting a marked degree of activation. Instead, as indicated in Fig. 1D and Table 1, column C, only rare mild injuries were evident in some hepatocytes from the normally perfused lobe at the same experimental time (mitochondrial swelling and internal membrane dilatation, as well as a few cytoplasmic electron-lucent vacuoles). Thus the morphology of the contralateral lobe of the liver can be considered not affected by the ischemia/reperfusion treatment. In contrast, no major changes were evident after 1 min of blood supply restoration. The results of the biochemical determinations of total glutathione are shown as mean values ⫾ standard deviation, with cytosolic and mitochondrial fractions respectively in Figs. 2 and 3. Lobes undergoing different ischemia times (I-15, I-30, I-60) did not show a significant glutathione variation compared with control lobes, although there was a modest (10%) increase
210
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000
ARMENI ET AL.: GSH AND MORPHOLOGY IN ISCHEMIA/REPERFUSION
in cytosolic GSH levels with respect to their controlateral lobe (groups 1, 2, and 3). Lobes of rat liver submitted to 15 and 60 min warm ischemia followed by 15 min reperfusion (groups 5 and 6) showed a significant depletion (P ⬍ 0.01) of cytosolic and mitochondrial glutathione concentration with respect to control lobes and to the lobes undergoing ischemia only. We found a depletion as high as about 40% in mitochondrial and cytosolic glutathione levels. As shown in Figs. 2 and 3, an equal and unexpected significant depletion (P ⬍ 0.01) of cytosolic and mitochondrial glutathione occurs in contralateral lobes without interruption of blood flow (sham) in the same livers undergoing 15 and 60 min ischemia followed by reperfusion (groups 5 and 6), compared with control groups and the groups undergoing ischemia only. A short reperfusion, 1 min, after 15 min of ischemia (group 4) did not lead to cytosolic or mitochondrial glutathione depletion. In Table 2 the percentage values of GSSG with respect to total glutathione for groups C, 1 (S-15 and I-15), and 5 (S-15/R-15 and I-15/R-15) are reported. During 15 min of ischemia GSSG was accumulated in cytosolic fractions of rat liver, not only in the ischemic lobes, but also in the sham lobes. After 15 min of reperfusion there was a decrease in both GSSG and total glutathione in the cytrosolic compartment of rat liver. The mitochondrial fraction of sham lobes submitted to ischemia only (S-15) showed high GSSG levels. The lobes submitted to ischemia (I-15) showed mitochondrial GSSG levels higher than those of the sham lobes. It is surprising that, during reperfusion, there was a depletion of GSH but there was no depletion of GSSG. In fact the high percentage of GSSG remained constant compared with groups undergoing ischemia only (groups 1, 2, and 3). In mitochondria GSSG levels were high, but total glutathione concentration was decreased. Also in this case sham lobes behaved like the ischemic lobes. DISCUSSION
In major liver surgery a period of warm ischemia following reperfusion is inevitable. It is well known
211
FIG. 2. Cytosolic glutathione (GSH⫹2GSSG) levels in rat liver from control, ischemic, reperfused after ischemia, and nonischemic (sham) tissue. White columns represent the right lobe (sham groups, also in the control), shaded columns represent the left and medial lobes (ischemic and ischemic/reperfused groups, also in the control). Group C, control liver in which the ischemic condition was not induced; groups 1, 2, and 3, lobes subjected to ischemia for 15, 30, and 60 min, respectively; group 4, lobes subjected to 15 min of ischemia followed by 1 min of reperfusion; groups 5 and 6, lobes subjected to 15 min of reperfusion after 15 and 60 min of ischemia, respectively. Ischemia/reperfusion groups show significantly decreased glutathione levels compared with control and ischemic groups. *P ⬍ 0.01 versus control.
that various complications take place in the postoperative course in livers previously exposed to extended warm ischemia and reperfusion. On the other hand, in the literature, there is increasing evidence that periods of warm ischemia and subsequent reperfusion are associated with biochemical and morphological changes which occur at the cell surface, in the cell cytosol, and in mitochondria [15]. Our experimental data demonstrate significant alterations of the hepatocytes but not of endothelial or Kupffer cells under conditions of ischemia obtained by interrupted blood flow in rat liver. The effect was timedependent and included progressive internal membrane dilatation and mitochondrial swelling. The results obtained are very comparable with those of other
FIG. 1. Ultrastructural features of rat liver from control, ischemic, reperfused after ischemia, and contralateral nonischemic tissue (sham). (A) Control liver in which the ischemia condition was not induced. Note the substantial integrity of cell compartments. Mitochondria are well preserved, with a dense matrix and normally organized cristae (inset). Some strands of endoplasmic reticulum and glycogen are also evident. Original magnifications: EM, 3500⫻; inset 10,000⫻. (B) Hepatocytes after 60 min of ischemia (I-60). Internal membranes are focally dilated. Mitochondria are swollen, with an electron-lucent matrix and disruption of cristae; some of them are shown at higher magnification (inset). Original magnifications: EM, 4500⫻; inset 11,000⫻. (C) Lobe undergoing 15 min of ischemia followed by 15 min of reperfusion (I-15/R-15). Note a diffuse severe dilatation of internal membranes of hepatocytes. Mitochondria are swollen, with cristae disruption or disappearance (inset). Large cytoplasmic electron-lucent vacuoles, devoid of organelles, are also evident. The upper left side of the micrograph shows a sinusoid with a red blood cell migrating in the Disse’s space (asterisk) through an endothelial disruption. E, endothelial cells. Original magnifications: EM, 3700⫻, inset 10,000⫻. (D) The sham lobe, 15 min after reperfusion of the 15 min of ischemia (S-15/R-15). A moderate mitochondrial swelling is evident. Focal cytoplasmic electron-lucent areas devoid of organelles are sometimes evident (inset). Original magnifications: EM, 4000⫻; inset 3500⫻.
212
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000
FIG. 3. Mitochondrial glutathione (GSH⫹2GSSG) levels in rat liver from control, ischemic, reperfused after ischemia, and nonischemic (sham) tissue. White columns represent the right lobe (sham groups, also in the control), shaded columns represent the left and medial lobes (ischemic and ischemic/reperfused groups, also in the control). Group C, control liver in which the ischemic condition was not induced; groups 1, 2, and 3, lobes subjected to ischemia for 15, 30, and 60 min, respectively; group 4, lobes subjected to 15 min of ischemia followed by 1 min of reperfusion; groups 5 and 6, lobes subjected to 15 min of reperfusion after 15 and 60 min of ischemia, respectively. Ischemia/reperfusion groups show significantly decreased glutathione levels compared with control and ischemic groups. *P ⬍ 0.01 versus control.
authors [3] at 15, 30, and 60 min of ischemia. We did not continue to 180 min of ischemia since the same authors have demonstrated very dramatic changes in the morphology of liver after more than 90 min, including necrosis [16]. During warm ischemia there is a decrease in the activities of several enzymes [17] and changes in the intracellular concentrations of different molecules [18]. Despite these variations, the concentration of total glutathione is constant under our experimental conditions (Figs. 2 and 3) and this could be explained by a decreased synthesis correlated with a decreased utilization by cell metabolism. In fact, in ischemia there is a decrease of reactions (in particular redox reactions) in which glutathione acts as coenzyme. Reduced GSH is the most abundant low-molecular-mass intracellular thiol with an easily oxidizable sulfhydryl group that can quench oxygen free radicals also nonenzymatically [9]. Glutathione is distributed in cell cytosol (85%) and in mitochondria (15%), and during ischemia GSH is maintained in both intracellular compartments. The two pools of glutathione are correlated and mitochondrial glutathione comes from cytosol, but the exchange is slow, as if they were independent pools [10]. Under different experimental conditions an almost complete depletion of cytosolic GSH has been observed, whereas mitochondrial GSH levels are unaffected [19].
Very consistent changes in liver glutathione are known to happen during reperfusion. In fact a great depletion has been observed in reperfused livers after both cold and warm ischemia. However, there is a very important difference between these two conditions. During reperfusion following cold ischemia, the glutathione loss is mainly from the cytosolic compartment while it remains constant in mitochondria. Instead, in reperfusion following warm ischemia, our results indicate that the depletion of glutathione involves both cytosolic and mitochondrial compartments. This effect is not measurable in the first minute of reperfusion, but it is significant after 15 min and not dependent on the previous duration of the ischemic phase, with 15 or 60 min producing similar effects. The presence of high GSSG during ischemia and the decrease of total glutathione and GSSG after reperfusion could be explained as a transfer of GSSG and/or GSH from the cytosol of reperfused rat liver. The variations of glutathione levels are more evident in mitochondria where they are a sign of oxidative stress at reperfusion: decrease of total glutathione and high GSSG. It is interesting to note that similar results were obtained in sham lobes, indicating that in the surgical model of ischemia used the right lobe cannot be considered a true control. During ischemia there is accumulation of GSSG and after reperfusion there is indication of an oxidative stress comparable to that of the ischemic lobes. The observed formation of GSSG is confirmed by studies that show the conversion of GSH to its oxidized TABLE 2 Levels (nmol/mg of Protein) of Oxidized Glutathione (GSSG) and Total Glutathione in the Cytosolic and Mitochondrial Fractions of Rat Liver: Control, after 15 min of Ischemia (Sham Lobe, S-15, and Ischemic Lobes, I-15), and after 15 min of Ischemia Followed by 15 min of Reperfusion (Sham Lobe, S-15/R-15, and Ischemic Lobes, I-15/R-15) Category Cytosol Control S-15 I-15 S-15/R-15 I-15/R-15 Mitochondria Control S-15 I-15 S-15/R-15 I-15/R-15
Total glutathione (nmol/mg)
GSSG (nmol/mg)
GSSG (%)
96.1 106.0 121.7 65.6 67.7
ND 3.8 5.6 0.4 0.6
ND 3.6 4.6 0.7 0.9
5.2 5.6 5.8 3.4 3.9
ND 0.5 0.7 0.4 0.9
ND 8.9 12.3 11.7 23.1
Note. ND, not detectable. Percentage values for GSSG are calculated with respect to total glutathione. The data are the means of three determinations.
ARMENI ET AL.: GSH AND MORPHOLOGY IN ISCHEMIA/REPERFUSION
form after oxidative events following ischemia [20]. It appears that the sudden restart of the aerobic metabolism challenges the cell with a strong time-limited oxidative stress that consumes GSH rapidly by oxidizing it to GSSG in the first minutes of reperfusion. The depletion of mitochondrial GSH strongly correlates with cell damage; a similar correlation is not demonstrated for cytosolic GSH since its very severe depletion is still compatible with cell viability [10]. A decreased intramitochondrial GSH concentration corresponds to a decreased defense of mitochondria from reactive oxygen species produced during the restoration of respiration [21]. The importance of GSH in rat liver mitochondria is further increased by the lack of catalase in these organelles [22]. Several studies have shown that loss of mitochondrial GSH can have deleterious effects with disruption of the transmembrane electrochemical gradient and reduction of ATP production [23], disturbance of Ca 2⫹ cycling [24], increase of lipid peroxidation, damage to the complexes of the electron transport chain [25], DNA damage, and cell death [26]. Furthermore, clinical studies have demonstrated that livers which experience primary nonfunction are not able to produce ATP or to maintain NAD/NADH [27]; both of those parameters are controlled by mitochondria. Thus mitochondria seem to be at the same time the key to recovery from the ischemialinked decrease of ATP levels, but also one of the main targets of reoxygenation injury. The decrease of mitochondrial glutathione is consistent with an oxidative stress, and ultrastructural analysis confirmed a generalized condition of cell sufferance, not only in hepatocytes, but also in endothelia and in Kupffer cells. Also in this case the effect of reperfusion was not affected by the duration of the previous ischemia of the liver. Under our conditions all liver cells were affected. Literature data have demonstrated that after 2 h of reperfusion electron microscopy analysis showed two populations of cells: apparently normal cells (probably due to a successful recovery) and others showing changes typical of necrosis [3]. Since our experimental design was similar to that of Rodriquez [3] it is possible that in the first minutes of reperfusion there is a generalized condition of stress followed by cell recovery or cell necrosis according to the entity of the damage. Probably if the majority of cells recovers, the function of the liver is completely restored; if instead the number of unrecovered cells (necrotic) is too high, it can lead to irreversible organ failure as observed in many unexplained cases. As expected, the contralateral liver lobe (sham) was morphologically similar to the control liver in agreement with literature data. The reported normal concentration of ATP irrespective of the ischemia and/or reperfusion condition of the other liver lobes of the same animal is in agreement with a general good
213
condition of the organ [3]. Surprisingly, our results indicate that in this part of the rat liver (not subjected to warm ischemia and reperfusion), mitochondrial and cytosolic glutathione levels were significantly affected as they were in the tissue subjected to reperfusion after ischemia. It is reasonable to propose that the observed total glutathione depletion and major GSSG concentration in mitochondria may reflect the effect of toxic molecules (e.g., TNF-␣ or complement T) that are liberated in the blood circulation by the reperfusiondamaged part of the liver. It is possible that this represents a sort of systemic effect, indicating that reperfusion damage can be extended to normoxygenated tissues. These observations might be explained as the production of reactive oxygen species coming from the mitochondria which have been stimulated by external molecules. In fact Colletti et al. [28] have shown that TNF-␣ is extensively produced during rat liver reperfusion after periods of warm ischemia. TNF-␣ can reduce mitochondrial GSH by increasing the production of superoxide; superoxide would be dismutated to hydrogen peroxide, which is reduced to GSSG by glutathione peroxidase utilizing GSH as substrate. Schulze et al. [29] have demonstrated that hepatic cells treated with TNF-␣ increase intramitochondrial reactive oxygen species production and GSH is also decreased in hepatic cells treated with TNF [30, 31]. Probably liver GSH acts also as a buffer between the various liver segments with consequent GSH loss from sham segment toward ischemic/reperfusion segment. Evidence of such a systemic effect involving organs different from the ischemized lung, heart, kidney, and intestine have recently been reported [32, 33]. This study can be relevant for major liver surgery such as hepatectomies, in which vascular ligation is performed prior to hepatic resection. REFERENCES 1.
2.
3.
4. 5.
6. 7.
Huguet, C., Gavelli, A., Chieco, P. A., Bona, S., Harb, J., and Joseph, J. M. Liver ischemia for hepatic resection. Where is the limit? Surgery 111: 251, 1992. Serizawa, A., Nakamura, S., Suzuki, S., Baba, S., and Nakano, M. Involvement of platelet-activating factor in cytokine production and neutrophil activation after hepatic ischemia– reperfusion. Hepatology 23: 1656, 1996. Rodriguez, A. A., LaMorte, W. W., Hanrahan, L. M., Hopkins, S. R., O’Keane, C., Cachecho, R., et al. Liver viability after ischemia–reperfusion. Arch. Surg. Chicago 126: 767, 1991. McCord, J. M. Oxygen-derived free radicals in post-ischemic tissue injury. N. Engl. J. Med. 312: 159, 1985. Reilly, P. M., Schiller, H. J., and Bulkley, G. B. Pharmacologic approach to tissue injury mediated by free radicals and other reactive oxygen metabolites. Am. J. Surg. 161: 488, 1991. Grace, P. A. Ischemia–reperfusion injury. Br. J. Surg. 81: 637, 1994. Myers, C. L., Weiss, S. J., Kirsh, M. M., and Shlafer, M. Involvement of hydrogen peroxide and hydroxyl radical in the “oxygen paradox”: Reduction of creatine kinase release by cata-
214
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000 lase, allopurinol or deferoxamine, but not by superoxide dismutase. J. Mol. Cell. Cardiol. 17: 675, 1985. Babbs, C. F., Salaris, S. C., and Turek, J. J. Cytochemical studies of hydrogen peroxide generation in post-ischemic hepatocytes. Am. J. Physiol. 260: H123, 1991. Meister, A., and Anderson, M. E. Glutathione. Annu. Rev. Biochem. 52: 711, 1983. Meredith, M. J., and Reed, D. J. Status of the mitochondrial pool of glutathione in the isolated hepatocyte. J. Biol. Chem. 257: 3747, 1982. Wahllander, A., Soboll, S., Sies, H., Linke, I., and Muller, M. Hepatic mitochondrial and cytosolic glutathione content and the subcellular distribution of GSH-S-transferases. FEBS Lett. 97: 138, 1979. Balercia, G., Accordini, C., Di Piramo, D., Martella, F., and Tomezzoli, A. An ultrastructural study of six cases of chronic active C hepatitis. A comparison with chronic active B hepatitis. Ultrastruct. Pathol. 17: 477, 1993. Lowry, O. H., Rosebrough, N. J., Farr, A., and Randall, R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265, 1951. Akerboom, T. P., and Sies, H. Assay of glutathione, glutathione disulfide and glutathione mixed disulphides in biological samples. Methods Enzymol. 77: 373, 1981. Jassem, W., and Roake, J. A. The molecular and cellular basis of reperfusion injury following organ transplantation. Transplant. Rev. 12: 14, 1998. Liu, D. L., Jeppsson, B., Hakansson, C. H., and Odselius, R. Multiple-system organ damage resulting from prolonged hepatic inflow interruption. Arch. Surg. Chicago 131: 442, 1996. Gupta, M., Dobashi, K., Greene, E. L., Orak, J. K., and Singh, I. Study on hepatic injury and antioxidant enzyme activities in rat subcellular organelles following in vivo ischemia and reperfusion. Mol. Cell. Biochem. 176: 337, 1997. Land, W. The impact of ischemia/reperfusion injury on specific and non-specific. Early and late chronic events after organ transplantation. Transplant. Rev. 10: 108, 1996. Jassem, W., Ciarimboli, C., Nisi Cerioni, P., Saba, V., Norton, S. J., and Principato, G. Glyoxalase II and glutathione levels in rat liver mitochondria during cold storage in Euro-Collins and University of Wisconsin solution. Transplantation 61: 1416, 1996. Katz, A., Oldham, K. T., Guice, K. S., and Coran, A. G. Oxidized glutathione as a marker of ischemia reperfusion associated with single lung transplantation. J. Am. Coll. Surg. 180: 25, 1995. Zoratti, M., and Szabo, I. The mitochondrial permeability transition. Biochim. Biophys. Acta 1241: 139, 1995.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Olafsdottir, K., Pascoe, G. A., and Reed, D. J. Mitochondrial glutathione status during Ca 2⫹ ionophore-induced injury to isolated hepatocytes. Arch. Biochem. Biophys. 263: 226, 1988. Vendemiale, G., Guerrieri, F., Grattagliano, I., Didonna, D., Muolo, L., and Altomare, E. Mitochondrial oxidative phosphorylation and intracellular glutathione compartmentation during rat liver regeneration. Hepatology 21: 1450, 1995. Beatrice, M. C., Palmer, J. W., and Pfeiffer, D. R. The relationship between mitochondrial membrane permeability, membrane potential, and the retention of Ca 2⫹ by mitochondria. J. Biol. Chem. 255: 8663, 1980. Zhang, Y., Marcillat, O., Giulivi, C., Ernster, L., and Davies, K. J. The oxidative inactivation of mitochondrial electron transport chain components and ATPase. J. Biol. Chem. 265: 16330, 1990. Esteve, J. M., Mompo, J., Garcia de la Asuncion, J., Sastre, J., Asensi, M., Boix, J. et al. Oxidative damage to mitochondrial DNA and glutathione oxidation in apoptosis: Studies in vivo and in vitro. FASEB J. 13: 1055, 1999. Kamiike, W., Burdelski, M., Steinhoff, G., Ringe, B., Lauchart, W., and Pichlmayr, R. Adenine nucleotide metabolism and its relation to organ viability in human liver transplantation. Transplantation 45: 138, 1988. Colletti, L. M., Remick, D. G., Burtch, G. D., Kunkel, S. L., Strieter, R. M., and Campbell, D. A. Role of tumor necrosis factoralpha in the pathophysiologic alterations after hepatic ischemia/ reperfusion injury in the rat. J. Clin. Invest. 85: 1936, 1990. Schulze Osthoff, K., Bakker, A. C., Vanhaesebroeck, B., Beyaert, R., Jacob, W. A., and Fiers, W. Cytotoxic activity of tumor necrosis factor is mediated by early damage of mitochondrial functions. Evidence for the involvement of mitochondrial radical generation. J. Biol. Chem. 267: 5317, 1992. Morales, A., Garcia-Ruiz, C., Miranda, M., Mari, M., Colell, A., Ardite, E., et al. Tumor necrosis factor increases hepatocellular glutathione by transcriptional regulation of the heavy subunit chain of gamma-glutamylcysteine synthetase. J. Biol. Chem. 272: 30371, 1997. Fernandez-Checa, J. C., Kaplowitz, N., Garcia-Ruiz, C., Colell, A., Miranda, M., Mari, M., et al. GSH transport in mitochondria: Defense against TNF-induced oxidative stress and alcohol-induced defect. Am. J. Physiol. 273: G7, 1997. Meyer, K., Brown, M. F., Zibari, G., Panes, J., McMillan, R. W., McDonald, J. C., et al. ICAM-1 upregulation in distant tissues after hepatic ischemia/reperfusion: A clue to the mechanism of multiple organ failure. J. Pediatr. Surg. 33: 350, 1998. Weinbroum, A. A., Hochhauser, E., Rudick, V., Kluger, Y., Sorkine, P., Karchevsky, E., et al. Direct induction of acute lung and myocardial dysfunction by liver ischemia and reperfusion. J. Trauma 43: 627, 1997.
Glutathione and Ultrastructural Changes in Inflow Occlusion of Rat Liver T. Armeni, R. Ghiselli,* G. Balercia,* ,† L. Goffi,* W. Jassem, V. Saba,* and G. Principato Institute of Biology and Genetics, *Clinica di Semeiotica e Metodologia Chirurgica, and †Division of Endocrinology, Institute of Internal Medicine, University of Ancona–INRCA, I-60100 Ancona, Italy Submitted for publication June 9, 1999
INTRODUCTION Background. Liver ischemia/reperfusion is frequently associated with organ injury to which reactive oxygen species contribute. The aim of our study was to evaluate cytosolic and mitochondrial glutathione levels and morphological changes in hepatocytes of rat liver in an experimental model of ischemia/reperfusion. Materials and methods. The experimental procedure consisted of temporary interruption of blood flow to the left lateral and medial hepatic lobes for different lengths of time and, in some cases, subsequent reperfusion. Cytosolic and mitochondrial glutathione levels were evaluated and ultrastructural analysis was carried out for all samples. Results. Ischemic lobes showed ultrastructural changes in relationship with the increase in ischemia time. Total glutathione levels did not show variations in ischemic lobes and sham lobes with respect to control rats during ischemia only. Instead, during reperfusion, significant ultrastructural alterations of the hepatocytes and a significant depletion of glutatione in cytosolic and mitochondrial compartments were evident. The sham lobes also showed a significant glutathione decrement. Increased oxidized glutathione (GSSG) levels were found during ischemia both in ischemic lobes and in sham lobes. During reperfusion GSSG was found to a minor extent, in the cytosolic compartment. In mitochondria GSSG levels were also high during reperfusion. Conclusions. We conclude that depletion of glutathione contributes to impaired liver after reperfusion following ischemia but depletion of glutathione alone does not induce changes in the morphology of the hepatocytes. Glutathione depletion and a greater quantity of GSSG, even in sham lobes, may indicate a metabolic alteration which spreads to compartments that are not involved in ischemia/reperfusion.
Hepatic resections in case of tumor or injury require continuous total hepatic inflow interruption; the application of Pringle’s maneuver, with cross clamping of the epatoduodenal ligament, has been shown to minimize blood loss. It has been widely reported that the maximal safe period of total hepatic occlusion is 1 h; longer periods of hepatic ischemia may produce profound hepatocyte injury with induction of liver failure and death, in particular in patients with underlying cirrhosis [1]. It is generally believed that the hepatic ischemia/ reperfusion injury after Pringle’s maneuver [2] is due to two different events: ● hepatic ischemia and subsequent reperfusion with arterial blood, and ● portal venous congestion and subsequent hepatic reperfusion with congested portal blood.
However, the specific roles of these mechanisms have not yet been well documented. Ischemia/reperfusion injury is evident through progressive ultrastructural changes in rat liver after 30 – 120 min of ischemia, with evident mitochondrial swelling after 60 min of ischemia. The alterations become more severe during reperfusion, although a partial reversibility of tissue injury has been observed after 2 h of blood supply restoration in rat liver following 60 min of ischemia [3]. Increasing clinical and experimental data have provided evidence that injury derived from ischemia/ reperfusion (I/R) 1 is mediated by reactive oxygen species: superoxide O 2⫺•, hydrogen peroxide H 2O 2, and
© 2000 Academic Press
Key Words: glutathione; rat liver; ultrastructural; ischemia/reperfusion.
1 Abbreviations used: GSH, glutathione; GSSG, oxidized glutathione; I/R, ischemia/reperfusion; TNF-␣, tumor necrosis factor-␣.
207
0022-4804/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
208
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000
hydroxyl radical •OH [4 – 6]. This phenomenon has been referred to as the “oxygen paradox” [7], implying that although the restoration of the supply of ordinary molecular oxygen to ischemic tissue is obviously necessary to restore normal function, reactive oxygen species generated in postischemic tissue may simultaneously participate in certain deleterious chemical reactions that worsen tissue injury [8]. Reduced glutathione (GSH), one of the main components of antioxidant defense, is a cofactor of several enzymes and plays an important role in the regulation of cell redox reactions as well as detoxification [9]. High levels of glutathione are distributed in two separate pools in the cell, 85–90% in the cytosol and 10 –15% in the mitochondria [10]. Depletion of cytosolic GSH down to critical levels can still be compatible with cell survival, whereas a decrease of mitochondrial GSH is critical for cell vitality [11]. The aim of this study is to evaluate cytosolic and mitochondrial glutathione levels after various periods of warm ischemia and reperfusion in a rat liver model. The morphological features of the hepatic tissue during ischemia–reperfusion have also been investigated. METHODS Liver ischemia/reperfusion model. Male Winstar rats (200 –250 g) were supplied by INRCA (Istituto Nazionale Ricerca e Cura Anziani, Ancona, Italy). Animals were housed under standard conditions with normal lighting cycle and with free access to food and water. Rats were anesthetized with ether. The experimental procedure consisted in temporary interruption of blood flow to the left lateral and medial lobes of the liver and, in some cases, subsequent reperfusion. The abdomen was opened by a midline incision and the liver hilus was exposed. All the vessels (hepatic artery, portal vein, and bile duct) to the left lateral and medial hepatic lobes were occluded. The portal branch to the right lobe was left open to avoid stasis in the portal system. These lobes constitute the sham lobes. In some cases after periods of ischemia of differing duration, the left and median lobes were reperfused. The rats were divided into seven experimental groups: Group C (n ⫽ 6), control rats, received neither ischemia nor reperfusion (Control). Group 1 (n ⫽ 8) rats received 15 min of ischemia (right lobe, S-15; left lateral and medial lobes I-15). Group 2 (n ⫽ 8) rats received 30 min of ischemia (right lobe, S-30; left lateral and medial lobes, I-30). Group 3 (n ⫽ 8) rats received 60 min of ischemia (right lobe, S-60; left lateral and medial lobes, I-60). Group 4 (n ⫽ 6) rats received 15 min of ischemia followed by 1 min of reperfusion (right lobe, S-15/R-1; left lateral and medial lobes, I-15/R-1). Group 5 (n ⫽ 8) rats received 15 min of ischemia followed by 15 min of reperfusion (right lobe, S-15/R-15; left lateral and medial lobes, I-15/R-15). Group 6 (n ⫽ 8) rats received 60 min of ischemia followed by 15 min of reperfusion (right lobe, S-60/R-15; left lateral and medial lobes, I-60/R15). At the end of each procedure a portion of the liver was rapidly removed both from the nonischemic lobe (sham) and from the ischemic lobe or the reperfused lobe. Liver samples were used for elec-
tron microscopy analysis and for the preparation of mitochondrial and cytosolic fractions in order to evaluate their glutathione content. Groups
Ischemic reperfusion lobe
Sham lobe
C 1 2 3 4 5 6
C I-15 I-30 I-60 I-15/R-1 I-15/R-15 I-60/R-15
C S-15 S-30 S-60 S-15/R-1 S-15/R-15 S-60/R-15
Ultrastructural investigation. Liver samples for the ultrastructural investigation were control, sham and ischemic, sham and reperfused. Liver biopsies were randomly collected at the end time of ischemia and reperfusion. Specimens underwent technical procedures for electron microscopy as described [12]. Briefly, they were fixed in 2% glutaraldehyde (4°C, pH 7.4), postfixed in 1% osmium tetroxide, and embedded in an Epon–araldite mixture. Semithin sections were stained with toluidine blue. Thin sections, stained with uranyl acetate and lead citrate, were observed with a Zeiss EM 10 electron microscope. Mitochondrial isolation. Mitochondrial and cytosolic fractions for glutathione measurements were obtained as follows: aliquots of liver tissue (4 –5 g) were homogenized 1:10 (w/v) in ice-cold buffer, pH 7.5, containing 75 mM sucrose, 225 mM mannitol, 1 mM EDTA, 5 mM Hepes, and 0.5 mg/ml fatty acid-free bovine serum albumin. The homogenate was centrifuged for 10 min at 600g at 4°C. Sediment was discarded and supernatant was centrifuged for 20 min at 1200g at 4°C, obtaining a mitochondrial pellet and a supernatant. The latter was centrifuged for 30 min at 55,000g at 4°C and the resulting supernatant (cytosolic fraction) was divided into 1-ml aliquots. The mitochondrial pellet was washed two times (by careful resuspensions in ice-cold homogenization buffer and centrifugation, 10 min at 2800g at 4°C) and purified mitochondria were carefully resuspended in 1 ml of ice-cold homogenization buffer. Sample preparation for glutathione assay. Aliquots of mitochondrial and cytosolic fractions were immediately deproteinized in icecold perchloric acid or sulfosalicylic acid/vinylpyridine for the determination of total glutathione or GSSG, respectively. The addition of vinylpyridine was necessary to form a covalent derivative with GSH. The mixtures were centrifuged (15,000g for 15 min at 4°C) and supernatants stored at ⫺80°C until used for glutathione determinations. Pellets were dissolved in 1 M sodium hydroxide and used for protein determinations [13] using bovine serum albumin as standard. Perchloric acid was removed from mitochondrial and cytosolic deproteinized samples by neutralization with potassium carbonate. Sulfosalicylic acid was neutralized with triethanolamine. Glutathione assay. Glutathione (GSH⫹2GSSG) was measured by the glutathione reductase recycling assay at 412 nm in the presence of 5,5⬘-dithio-bis-nitrobenzoic acid [14]. The amount of total glutathione is reported as nmol/mg of protein of the respective fractions. Statistical analysis. Results are reported as means ⫾ standard deviation. The statistical significance between groups was assessed by computer-assisted analysis of variance.
RESULTS
The main ultrastructural features detected at each time of the experimental protocol of ischemia/reperfusion of rat liver are shown in Table 1 and Fig. 1. In control livers normal cell patterns were detected after fine structural investigations, with substantial integrity of both nuclear and cytoplasmic compart-
209
ARMENI ET AL.: GSH AND MORPHOLOGY IN ISCHEMIA/REPERFUSION
TABLE 1 Main Ultrastructural Features of Rat Liver Exposed to Warm Ischemia or Warm Ischemia Followed by Reperfusion (A) Ischemia (min) Injury Hepatocytes Internal membrane dilatation (smooth and rough endoplasmic reticulum, Golgi apparatus) Mitochondrial swelling, matrix clarification, reduction of cristae and spatial disruption of their organization Glycogen lack Large electron-lucent vacuoles devoid of organelles Bile canaliculi dilatation Disse’s space dilatation and microvilli reduction Neutrophilic infiltration Other components Endothelial cell swelling Endothelial necrosis and basal lamina exposure Kupffer cell activation
(B) Ischemia/ reperfusion
(C) Sham
I-15
I-30
I-60
I15-60/R15
S-15-60/R15
⫹/⫺
⫹
⫹/⫹⫹
⫹⫹⫹⫹
⫹/⫺
⫹/⫺
⫹
⫹/⫹⫹
⫹⫹⫹⫹
⫹/⫺
⫹ ⫹/⫺ ⫺ ⫺ ⫺
⫹ ⫹/⫺ ⫺ ⫹/⫺ ⫹/⫺
⫹/⫹⫹ ⫹/⫺ ⫹/⫺ ⫹ ⫹/⫺
⫹⫹/⫹⫹⫹ ⫹⫹⫹ ⫹⫹ ⫹⫹⫹ ⫹/⫹⫹
⫺ ⫹/⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺
⫹/⫺ ⫺ ⫺
⫹/⫺ ⫹⫹ ⫹/⫺
⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹
⫺ ⫺ ⫺
Note. Negative (as in control liver); ⫹, ⫹⫹, and ⫹⫹⫹, weak, moderate, and strong positivity, respectively, in comparison with control liver. Data are compared with those of control livers not exposed to treatment. Results are referred to hepatic tissue after 15, 30, and 60 min of warm ischemia (I-15, I-30, I-60) (column A), left lateral and medial liver lobes after 15 or 60 min of ischemia followed by 15 min of reperfusion (one group, I15-60/R-15, because of very similar results at the two different ischemia times) (column B), and sham tissue of the same rats (one group, S-15-60/R-15, because of very similar results at the two different ischemia times) (column C). Liver specimens were fixed in 2% glutaraldehyde (4°C, pH 7.4), postfixed in 1% osmium tetroxide, and embedded in an Epon–araldite mixture. Semithin sections were stained with toluidine blue. Thin sections, stained with uranyl acetate and lead citrate were observed with a Zeiss EM 10 electron microscope.
ments of hepatocytes. Vascular districts, as well as Kupffer cells, were unaltered. In the group of ischemic livers (Table 1, column A) some ultrastructural changes were detected at a cytoplasmic level after as soon as 15 min of ischemia (I-15). In particular, the endoplasmic reticulum (smooth and rough) was focally dilated, as was the Golgi apparatus; mitochondria were focally swollen with matrix clarification. Some areas devoid of glycogen were also evident. Cell alterations became more evident after 30 and 60 min of ischemia (I-30 and I-60). Mitochondria were focally swollen, and a reduction in the number of cristae became evident. The lack of glycogen was also more evident, in relationship to the increase in ischemia duration. Some electron-lucent areas, devoid of organelles, were found in the cell cytoplasm. Moreover, Disse’s spaces were consistently dilated, while endothelial and Kupffer cells showed some electron-dense vacuoles in their cytoplasm, especially after 60 min of ischemia (Fig. 1B). A series of severe diffuse alterations was evident at a cellular level in the group of livers reperfused for 15 min after 15– 60 min of warm ischemia (I-15-60/R-15) (Table 1, column B). A diffuse dilatation of internal membranes was evident; mitochondria were also diffusely swollen, cristae were reduced in number, and their spatial organization appeared disrupted. Large cytoplasmic electron-lucent vacuoles, devoid of organelles, were evident, probably reflecting an intracellular edema. Dilatation of bile
canaliculi and thick subplasmalemmatic bundles of microfilaments were found. Disse’s spaces were markedly dilated (edematous) and a marked reduction of hepatocyte microvillous projections toward the sinusoids was frequently evident. Neutrophils were also seen in the hepatic lobules. Endothelial cells were swollen, with a clarification of cytoplasmic matrix and a large number of electron-lucent vacuoles (Fig. 1C). Some of them were necrotic, with intraluminal exposure of the basal lamina. A swelling of Kupffer cells, the cytoplasm of which contained some dense vacuoles, was also evident, reflecting a marked degree of activation. Instead, as indicated in Fig. 1D and Table 1, column C, only rare mild injuries were evident in some hepatocytes from the normally perfused lobe at the same experimental time (mitochondrial swelling and internal membrane dilatation, as well as a few cytoplasmic electron-lucent vacuoles). Thus the morphology of the contralateral lobe of the liver can be considered not affected by the ischemia/reperfusion treatment. In contrast, no major changes were evident after 1 min of blood supply restoration. The results of the biochemical determinations of total glutathione are shown as mean values ⫾ standard deviation, with cytosolic and mitochondrial fractions respectively in Figs. 2 and 3. Lobes undergoing different ischemia times (I-15, I-30, I-60) did not show a significant glutathione variation compared with control lobes, although there was a modest (10%) increase
210
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000
ARMENI ET AL.: GSH AND MORPHOLOGY IN ISCHEMIA/REPERFUSION
in cytosolic GSH levels with respect to their controlateral lobe (groups 1, 2, and 3). Lobes of rat liver submitted to 15 and 60 min warm ischemia followed by 15 min reperfusion (groups 5 and 6) showed a significant depletion (P ⬍ 0.01) of cytosolic and mitochondrial glutathione concentration with respect to control lobes and to the lobes undergoing ischemia only. We found a depletion as high as about 40% in mitochondrial and cytosolic glutathione levels. As shown in Figs. 2 and 3, an equal and unexpected significant depletion (P ⬍ 0.01) of cytosolic and mitochondrial glutathione occurs in contralateral lobes without interruption of blood flow (sham) in the same livers undergoing 15 and 60 min ischemia followed by reperfusion (groups 5 and 6), compared with control groups and the groups undergoing ischemia only. A short reperfusion, 1 min, after 15 min of ischemia (group 4) did not lead to cytosolic or mitochondrial glutathione depletion. In Table 2 the percentage values of GSSG with respect to total glutathione for groups C, 1 (S-15 and I-15), and 5 (S-15/R-15 and I-15/R-15) are reported. During 15 min of ischemia GSSG was accumulated in cytosolic fractions of rat liver, not only in the ischemic lobes, but also in the sham lobes. After 15 min of reperfusion there was a decrease in both GSSG and total glutathione in the cytrosolic compartment of rat liver. The mitochondrial fraction of sham lobes submitted to ischemia only (S-15) showed high GSSG levels. The lobes submitted to ischemia (I-15) showed mitochondrial GSSG levels higher than those of the sham lobes. It is surprising that, during reperfusion, there was a depletion of GSH but there was no depletion of GSSG. In fact the high percentage of GSSG remained constant compared with groups undergoing ischemia only (groups 1, 2, and 3). In mitochondria GSSG levels were high, but total glutathione concentration was decreased. Also in this case sham lobes behaved like the ischemic lobes. DISCUSSION
In major liver surgery a period of warm ischemia following reperfusion is inevitable. It is well known
211
FIG. 2. Cytosolic glutathione (GSH⫹2GSSG) levels in rat liver from control, ischemic, reperfused after ischemia, and nonischemic (sham) tissue. White columns represent the right lobe (sham groups, also in the control), shaded columns represent the left and medial lobes (ischemic and ischemic/reperfused groups, also in the control). Group C, control liver in which the ischemic condition was not induced; groups 1, 2, and 3, lobes subjected to ischemia for 15, 30, and 60 min, respectively; group 4, lobes subjected to 15 min of ischemia followed by 1 min of reperfusion; groups 5 and 6, lobes subjected to 15 min of reperfusion after 15 and 60 min of ischemia, respectively. Ischemia/reperfusion groups show significantly decreased glutathione levels compared with control and ischemic groups. *P ⬍ 0.01 versus control.
that various complications take place in the postoperative course in livers previously exposed to extended warm ischemia and reperfusion. On the other hand, in the literature, there is increasing evidence that periods of warm ischemia and subsequent reperfusion are associated with biochemical and morphological changes which occur at the cell surface, in the cell cytosol, and in mitochondria [15]. Our experimental data demonstrate significant alterations of the hepatocytes but not of endothelial or Kupffer cells under conditions of ischemia obtained by interrupted blood flow in rat liver. The effect was timedependent and included progressive internal membrane dilatation and mitochondrial swelling. The results obtained are very comparable with those of other
FIG. 1. Ultrastructural features of rat liver from control, ischemic, reperfused after ischemia, and contralateral nonischemic tissue (sham). (A) Control liver in which the ischemia condition was not induced. Note the substantial integrity of cell compartments. Mitochondria are well preserved, with a dense matrix and normally organized cristae (inset). Some strands of endoplasmic reticulum and glycogen are also evident. Original magnifications: EM, 3500⫻; inset 10,000⫻. (B) Hepatocytes after 60 min of ischemia (I-60). Internal membranes are focally dilated. Mitochondria are swollen, with an electron-lucent matrix and disruption of cristae; some of them are shown at higher magnification (inset). Original magnifications: EM, 4500⫻; inset 11,000⫻. (C) Lobe undergoing 15 min of ischemia followed by 15 min of reperfusion (I-15/R-15). Note a diffuse severe dilatation of internal membranes of hepatocytes. Mitochondria are swollen, with cristae disruption or disappearance (inset). Large cytoplasmic electron-lucent vacuoles, devoid of organelles, are also evident. The upper left side of the micrograph shows a sinusoid with a red blood cell migrating in the Disse’s space (asterisk) through an endothelial disruption. E, endothelial cells. Original magnifications: EM, 3700⫻, inset 10,000⫻. (D) The sham lobe, 15 min after reperfusion of the 15 min of ischemia (S-15/R-15). A moderate mitochondrial swelling is evident. Focal cytoplasmic electron-lucent areas devoid of organelles are sometimes evident (inset). Original magnifications: EM, 4000⫻; inset 3500⫻.
212
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000
FIG. 3. Mitochondrial glutathione (GSH⫹2GSSG) levels in rat liver from control, ischemic, reperfused after ischemia, and nonischemic (sham) tissue. White columns represent the right lobe (sham groups, also in the control), shaded columns represent the left and medial lobes (ischemic and ischemic/reperfused groups, also in the control). Group C, control liver in which the ischemic condition was not induced; groups 1, 2, and 3, lobes subjected to ischemia for 15, 30, and 60 min, respectively; group 4, lobes subjected to 15 min of ischemia followed by 1 min of reperfusion; groups 5 and 6, lobes subjected to 15 min of reperfusion after 15 and 60 min of ischemia, respectively. Ischemia/reperfusion groups show significantly decreased glutathione levels compared with control and ischemic groups. *P ⬍ 0.01 versus control.
authors [3] at 15, 30, and 60 min of ischemia. We did not continue to 180 min of ischemia since the same authors have demonstrated very dramatic changes in the morphology of liver after more than 90 min, including necrosis [16]. During warm ischemia there is a decrease in the activities of several enzymes [17] and changes in the intracellular concentrations of different molecules [18]. Despite these variations, the concentration of total glutathione is constant under our experimental conditions (Figs. 2 and 3) and this could be explained by a decreased synthesis correlated with a decreased utilization by cell metabolism. In fact, in ischemia there is a decrease of reactions (in particular redox reactions) in which glutathione acts as coenzyme. Reduced GSH is the most abundant low-molecular-mass intracellular thiol with an easily oxidizable sulfhydryl group that can quench oxygen free radicals also nonenzymatically [9]. Glutathione is distributed in cell cytosol (85%) and in mitochondria (15%), and during ischemia GSH is maintained in both intracellular compartments. The two pools of glutathione are correlated and mitochondrial glutathione comes from cytosol, but the exchange is slow, as if they were independent pools [10]. Under different experimental conditions an almost complete depletion of cytosolic GSH has been observed, whereas mitochondrial GSH levels are unaffected [19].
Very consistent changes in liver glutathione are known to happen during reperfusion. In fact a great depletion has been observed in reperfused livers after both cold and warm ischemia. However, there is a very important difference between these two conditions. During reperfusion following cold ischemia, the glutathione loss is mainly from the cytosolic compartment while it remains constant in mitochondria. Instead, in reperfusion following warm ischemia, our results indicate that the depletion of glutathione involves both cytosolic and mitochondrial compartments. This effect is not measurable in the first minute of reperfusion, but it is significant after 15 min and not dependent on the previous duration of the ischemic phase, with 15 or 60 min producing similar effects. The presence of high GSSG during ischemia and the decrease of total glutathione and GSSG after reperfusion could be explained as a transfer of GSSG and/or GSH from the cytosol of reperfused rat liver. The variations of glutathione levels are more evident in mitochondria where they are a sign of oxidative stress at reperfusion: decrease of total glutathione and high GSSG. It is interesting to note that similar results were obtained in sham lobes, indicating that in the surgical model of ischemia used the right lobe cannot be considered a true control. During ischemia there is accumulation of GSSG and after reperfusion there is indication of an oxidative stress comparable to that of the ischemic lobes. The observed formation of GSSG is confirmed by studies that show the conversion of GSH to its oxidized TABLE 2 Levels (nmol/mg of Protein) of Oxidized Glutathione (GSSG) and Total Glutathione in the Cytosolic and Mitochondrial Fractions of Rat Liver: Control, after 15 min of Ischemia (Sham Lobe, S-15, and Ischemic Lobes, I-15), and after 15 min of Ischemia Followed by 15 min of Reperfusion (Sham Lobe, S-15/R-15, and Ischemic Lobes, I-15/R-15) Category Cytosol Control S-15 I-15 S-15/R-15 I-15/R-15 Mitochondria Control S-15 I-15 S-15/R-15 I-15/R-15
Total glutathione (nmol/mg)
GSSG (nmol/mg)
GSSG (%)
96.1 106.0 121.7 65.6 67.7
ND 3.8 5.6 0.4 0.6
ND 3.6 4.6 0.7 0.9
5.2 5.6 5.8 3.4 3.9
ND 0.5 0.7 0.4 0.9
ND 8.9 12.3 11.7 23.1
Note. ND, not detectable. Percentage values for GSSG are calculated with respect to total glutathione. The data are the means of three determinations.
ARMENI ET AL.: GSH AND MORPHOLOGY IN ISCHEMIA/REPERFUSION
form after oxidative events following ischemia [20]. It appears that the sudden restart of the aerobic metabolism challenges the cell with a strong time-limited oxidative stress that consumes GSH rapidly by oxidizing it to GSSG in the first minutes of reperfusion. The depletion of mitochondrial GSH strongly correlates with cell damage; a similar correlation is not demonstrated for cytosolic GSH since its very severe depletion is still compatible with cell viability [10]. A decreased intramitochondrial GSH concentration corresponds to a decreased defense of mitochondria from reactive oxygen species produced during the restoration of respiration [21]. The importance of GSH in rat liver mitochondria is further increased by the lack of catalase in these organelles [22]. Several studies have shown that loss of mitochondrial GSH can have deleterious effects with disruption of the transmembrane electrochemical gradient and reduction of ATP production [23], disturbance of Ca 2⫹ cycling [24], increase of lipid peroxidation, damage to the complexes of the electron transport chain [25], DNA damage, and cell death [26]. Furthermore, clinical studies have demonstrated that livers which experience primary nonfunction are not able to produce ATP or to maintain NAD/NADH [27]; both of those parameters are controlled by mitochondria. Thus mitochondria seem to be at the same time the key to recovery from the ischemialinked decrease of ATP levels, but also one of the main targets of reoxygenation injury. The decrease of mitochondrial glutathione is consistent with an oxidative stress, and ultrastructural analysis confirmed a generalized condition of cell sufferance, not only in hepatocytes, but also in endothelia and in Kupffer cells. Also in this case the effect of reperfusion was not affected by the duration of the previous ischemia of the liver. Under our conditions all liver cells were affected. Literature data have demonstrated that after 2 h of reperfusion electron microscopy analysis showed two populations of cells: apparently normal cells (probably due to a successful recovery) and others showing changes typical of necrosis [3]. Since our experimental design was similar to that of Rodriquez [3] it is possible that in the first minutes of reperfusion there is a generalized condition of stress followed by cell recovery or cell necrosis according to the entity of the damage. Probably if the majority of cells recovers, the function of the liver is completely restored; if instead the number of unrecovered cells (necrotic) is too high, it can lead to irreversible organ failure as observed in many unexplained cases. As expected, the contralateral liver lobe (sham) was morphologically similar to the control liver in agreement with literature data. The reported normal concentration of ATP irrespective of the ischemia and/or reperfusion condition of the other liver lobes of the same animal is in agreement with a general good
213
condition of the organ [3]. Surprisingly, our results indicate that in this part of the rat liver (not subjected to warm ischemia and reperfusion), mitochondrial and cytosolic glutathione levels were significantly affected as they were in the tissue subjected to reperfusion after ischemia. It is reasonable to propose that the observed total glutathione depletion and major GSSG concentration in mitochondria may reflect the effect of toxic molecules (e.g., TNF-␣ or complement T) that are liberated in the blood circulation by the reperfusiondamaged part of the liver. It is possible that this represents a sort of systemic effect, indicating that reperfusion damage can be extended to normoxygenated tissues. These observations might be explained as the production of reactive oxygen species coming from the mitochondria which have been stimulated by external molecules. In fact Colletti et al. [28] have shown that TNF-␣ is extensively produced during rat liver reperfusion after periods of warm ischemia. TNF-␣ can reduce mitochondrial GSH by increasing the production of superoxide; superoxide would be dismutated to hydrogen peroxide, which is reduced to GSSG by glutathione peroxidase utilizing GSH as substrate. Schulze et al. [29] have demonstrated that hepatic cells treated with TNF-␣ increase intramitochondrial reactive oxygen species production and GSH is also decreased in hepatic cells treated with TNF [30, 31]. Probably liver GSH acts also as a buffer between the various liver segments with consequent GSH loss from sham segment toward ischemic/reperfusion segment. Evidence of such a systemic effect involving organs different from the ischemized lung, heart, kidney, and intestine have recently been reported [32, 33]. This study can be relevant for major liver surgery such as hepatectomies, in which vascular ligation is performed prior to hepatic resection. REFERENCES 1.
2.
3.
4. 5.
6. 7.
Huguet, C., Gavelli, A., Chieco, P. A., Bona, S., Harb, J., and Joseph, J. M. Liver ischemia for hepatic resection. Where is the limit? Surgery 111: 251, 1992. Serizawa, A., Nakamura, S., Suzuki, S., Baba, S., and Nakano, M. Involvement of platelet-activating factor in cytokine production and neutrophil activation after hepatic ischemia– reperfusion. Hepatology 23: 1656, 1996. Rodriguez, A. A., LaMorte, W. W., Hanrahan, L. M., Hopkins, S. R., O’Keane, C., Cachecho, R., et al. Liver viability after ischemia–reperfusion. Arch. Surg. Chicago 126: 767, 1991. McCord, J. M. Oxygen-derived free radicals in post-ischemic tissue injury. N. Engl. J. Med. 312: 159, 1985. Reilly, P. M., Schiller, H. J., and Bulkley, G. B. Pharmacologic approach to tissue injury mediated by free radicals and other reactive oxygen metabolites. Am. J. Surg. 161: 488, 1991. Grace, P. A. Ischemia–reperfusion injury. Br. J. Surg. 81: 637, 1994. Myers, C. L., Weiss, S. J., Kirsh, M. M., and Shlafer, M. Involvement of hydrogen peroxide and hydroxyl radical in the “oxygen paradox”: Reduction of creatine kinase release by cata-
214
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
JOURNAL OF SURGICAL RESEARCH: VOL. 88, NO. 2, FEBRUARY 2000 lase, allopurinol or deferoxamine, but not by superoxide dismutase. J. Mol. Cell. Cardiol. 17: 675, 1985. Babbs, C. F., Salaris, S. C., and Turek, J. J. Cytochemical studies of hydrogen peroxide generation in post-ischemic hepatocytes. Am. J. Physiol. 260: H123, 1991. Meister, A., and Anderson, M. E. Glutathione. Annu. Rev. Biochem. 52: 711, 1983. Meredith, M. J., and Reed, D. J. Status of the mitochondrial pool of glutathione in the isolated hepatocyte. J. Biol. Chem. 257: 3747, 1982. Wahllander, A., Soboll, S., Sies, H., Linke, I., and Muller, M. Hepatic mitochondrial and cytosolic glutathione content and the subcellular distribution of GSH-S-transferases. FEBS Lett. 97: 138, 1979. Balercia, G., Accordini, C., Di Piramo, D., Martella, F., and Tomezzoli, A. An ultrastructural study of six cases of chronic active C hepatitis. A comparison with chronic active B hepatitis. Ultrastruct. Pathol. 17: 477, 1993. Lowry, O. H., Rosebrough, N. J., Farr, A., and Randall, R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265, 1951. Akerboom, T. P., and Sies, H. Assay of glutathione, glutathione disulfide and glutathione mixed disulphides in biological samples. Methods Enzymol. 77: 373, 1981. Jassem, W., and Roake, J. A. The molecular and cellular basis of reperfusion injury following organ transplantation. Transplant. Rev. 12: 14, 1998. Liu, D. L., Jeppsson, B., Hakansson, C. H., and Odselius, R. Multiple-system organ damage resulting from prolonged hepatic inflow interruption. Arch. Surg. Chicago 131: 442, 1996. Gupta, M., Dobashi, K., Greene, E. L., Orak, J. K., and Singh, I. Study on hepatic injury and antioxidant enzyme activities in rat subcellular organelles following in vivo ischemia and reperfusion. Mol. Cell. Biochem. 176: 337, 1997. Land, W. The impact of ischemia/reperfusion injury on specific and non-specific. Early and late chronic events after organ transplantation. Transplant. Rev. 10: 108, 1996. Jassem, W., Ciarimboli, C., Nisi Cerioni, P., Saba, V., Norton, S. J., and Principato, G. Glyoxalase II and glutathione levels in rat liver mitochondria during cold storage in Euro-Collins and University of Wisconsin solution. Transplantation 61: 1416, 1996. Katz, A., Oldham, K. T., Guice, K. S., and Coran, A. G. Oxidized glutathione as a marker of ischemia reperfusion associated with single lung transplantation. J. Am. Coll. Surg. 180: 25, 1995. Zoratti, M., and Szabo, I. The mitochondrial permeability transition. Biochim. Biophys. Acta 1241: 139, 1995.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Olafsdottir, K., Pascoe, G. A., and Reed, D. J. Mitochondrial glutathione status during Ca 2⫹ ionophore-induced injury to isolated hepatocytes. Arch. Biochem. Biophys. 263: 226, 1988. Vendemiale, G., Guerrieri, F., Grattagliano, I., Didonna, D., Muolo, L., and Altomare, E. Mitochondrial oxidative phosphorylation and intracellular glutathione compartmentation during rat liver regeneration. Hepatology 21: 1450, 1995. Beatrice, M. C., Palmer, J. W., and Pfeiffer, D. R. The relationship between mitochondrial membrane permeability, membrane potential, and the retention of Ca 2⫹ by mitochondria. J. Biol. Chem. 255: 8663, 1980. Zhang, Y., Marcillat, O., Giulivi, C., Ernster, L., and Davies, K. J. The oxidative inactivation of mitochondrial electron transport chain components and ATPase. J. Biol. Chem. 265: 16330, 1990. Esteve, J. M., Mompo, J., Garcia de la Asuncion, J., Sastre, J., Asensi, M., Boix, J. et al. Oxidative damage to mitochondrial DNA and glutathione oxidation in apoptosis: Studies in vivo and in vitro. FASEB J. 13: 1055, 1999. Kamiike, W., Burdelski, M., Steinhoff, G., Ringe, B., Lauchart, W., and Pichlmayr, R. Adenine nucleotide metabolism and its relation to organ viability in human liver transplantation. Transplantation 45: 138, 1988. Colletti, L. M., Remick, D. G., Burtch, G. D., Kunkel, S. L., Strieter, R. M., and Campbell, D. A. Role of tumor necrosis factoralpha in the pathophysiologic alterations after hepatic ischemia/ reperfusion injury in the rat. J. Clin. Invest. 85: 1936, 1990. Schulze Osthoff, K., Bakker, A. C., Vanhaesebroeck, B., Beyaert, R., Jacob, W. A., and Fiers, W. Cytotoxic activity of tumor necrosis factor is mediated by early damage of mitochondrial functions. Evidence for the involvement of mitochondrial radical generation. J. Biol. Chem. 267: 5317, 1992. Morales, A., Garcia-Ruiz, C., Miranda, M., Mari, M., Colell, A., Ardite, E., et al. Tumor necrosis factor increases hepatocellular glutathione by transcriptional regulation of the heavy subunit chain of gamma-glutamylcysteine synthetase. J. Biol. Chem. 272: 30371, 1997. Fernandez-Checa, J. C., Kaplowitz, N., Garcia-Ruiz, C., Colell, A., Miranda, M., Mari, M., et al. GSH transport in mitochondria: Defense against TNF-induced oxidative stress and alcohol-induced defect. Am. J. Physiol. 273: G7, 1997. Meyer, K., Brown, M. F., Zibari, G., Panes, J., McMillan, R. W., McDonald, J. C., et al. ICAM-1 upregulation in distant tissues after hepatic ischemia/reperfusion: A clue to the mechanism of multiple organ failure. J. Pediatr. Surg. 33: 350, 1998. Weinbroum, A. A., Hochhauser, E., Rudick, V., Kluger, Y., Sorkine, P., Karchevsky, E., et al. Direct induction of acute lung and myocardial dysfunction by liver ischemia and reperfusion. J. Trauma 43: 627, 1997.