Increased ischemic injury in old mouse liver: An ATP-dependent mechanism


  • Markus Selzner,

    1. Department of Visceral Surgery and Transplantation, Swiss HPB Center (Hepato-Pancreato-Biliary) Center, University Hospital Zurich, Switzerland
    Search for more papers by this author
    • The first 2 authors contributed equally to this work. M.S. is a Novartis HPB & Transplant fellow at the University Hospital Zurich.

  • Nazia Selzner,

    1. Department of Visceral Surgery and Transplantation, Swiss HPB Center (Hepato-Pancreato-Biliary) Center, University Hospital Zurich, Switzerland
    Search for more papers by this author
    • The first 2 authors contributed equally to this work. M.S. is a Novartis HPB & Transplant fellow at the University Hospital Zurich.

  • Wolfram Jochum,

    1. Department of Pathology, Swiss HPB Center (Hepato-Pancreato-Biliary) Center, University Hospital Zurich, Switzerland
    Search for more papers by this author
  • Rolf Graf,

    1. Department of Visceral Surgery and Transplantation, Swiss HPB Center (Hepato-Pancreato-Biliary) Center, University Hospital Zurich, Switzerland
    Search for more papers by this author
  • Pierre-Alain Clavien

    Corresponding author
    1. Department of Visceral Surgery and Transplantation, Swiss HPB Center (Hepato-Pancreato-Biliary) Center, University Hospital Zurich, Switzerland
    • Department of Visceral Surgery and Transplantation, University of Zurich, Rämistr. 100, 8091 Zürich, Switzerland
    Search for more papers by this author
    • Telephone: 411-255-3300; FAX: 411-255-4449


Although livers exhibit only minimal morphologic changes with age, how older livers tolerate pathologic conditions such as normothermic ischemia is unknown. Young 6-week-old mice and old 60-week-old mice underwent 60 minutes of hepatic ischemia and various periods of reperfusion. Markers of hepatocyte injury, hepatic energy content, and mitochondrial function were determined. Ischemic preconditioning and glucose injection were evaluated as protective strategies against reperfusion injury in old mice. Reperfusion injury was far worse in old mice compared with mice in the young control group. Ischemic preconditioning was highly protective against reperfusion injury in young but not in old mice. Older livers had dramatically reduced adenosine triphosphate (ATP) levels and glycogen contents. The low intrahepatic energy level in old mice was associated with a reduced mitochondrial ATP production. Preoperative injection of glucose restored the intrahepatic ATP content and protected against reperfusion injury. Furthermore, glucose injection restored the protective effect of ischemic preconditioning, resulting in additive protection when both strategies were combined. Aging of the liver is associated with mitochondrial dysfunction and decreased intrahepatic energy content, resulting in poorer tolerance against ischemic injury. Improving intrahepatic ATP levels in old livers by glucose injection protects the old liver against ischemic injury and restores the protective effects of ischemic preconditioning. Liver Transpl 13:382–390, 2007. © 2007 AASLD.

Life expectancy has increased over the past century from 48 years in 1900 to 77 years in 2001.1 It has been estimated that by the year 2050, life expectancy will reach 86 years for women and 80 years for men. From another angle, only 4% of the U.S. population was older than 65 years in 1900; this figure increased to 13% in 2000, and may reach 20% by 2030.2 Thus, we are facing a dramatic increase in the number of elderly patients with liver diseases, and many of those patients may be considered for surgery. In addition, because of the shortage of available organs for transplantation, older donors may become a major source of liver grafts.

There is an old adage that the liver does not age. Although hepatic blood flow and liver volume slightly decrease in the elderly,3, 4 aging is associated with only minimal structural changes of the liver parenchyma with well-preserved function.5 However, how older livers can withstand pathological conditions has remained largely unexplored. Emerging studies have shown that recurrent hepatitis C after liver transplantation is more severe in recipients receiving a graft from an elderly donor.6, 7 The elderly have a decreased hepatic ability to detoxify toxins, such as thioacetamine.8 We recently noted in a clinical trial that the protective effects of ischemic preconditioning was lost in patients older than 65 years of age.9

The effects of age of a cadaveric or living donor on the outcome of liver transplantation and the results of liver resection in the elderly patients have been evaluated in only few clinical studies.10 For example, although some studies suggested an increased risk for primary nonfunction in liver transplant recipients receiving a graft from older donors,11, 12 others failed to show any significant differences in any outcome parameters between recipients transplanted with young (<70 years) vs. older grafts (>70 years).13 To our knowledge, no clinical data are currently available on the effect of age on normothermic ischemic injury during liver resection.

Aging is associated with a variety of decreased mitochondria functions in several tissues.14, 15 For example, in mitochondria, the cytochrome C binding site is impaired in old rat myocardial tissue, which is associated with impaired oxidative function and poor tolerance against ischemic injury.14 Sastre et al.15 documented in isolated rat hepatocytes that age is associated with an alteration in mitochondria membrane potential, along with an increase in the size of mitochondria and impairment of gluconeogenesis.

In this study, we tested the hypothesis that older livers have a higher susceptibility to normothermic ischemic injury as a result of mitochondrial dysfunction. We further investigated ischemic preconditioning and glucose administration to restore cellular adenosine triphosphate (ATP) contents as protective strategies for older livers.


AST, aspartate aminotransferase; ATP, adenosine triphosphate; ADP, adenosine 5′-diphosphate; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling.



All experiments were performed in male C57/BL6 mice 6 weeks (young) and 60 weeks (old) of age. All animals received a standard laboratory diet containing 12% fat, 28% protein, and 60% carbohydrates (5001 rodent diet; PMI, Brentwood, MO), with water and food ad libitum. All mice were kept under constant environmental conditions with a 12-hour light-dark cycle (light 7:00 AM to 7:00 PM). All surgical procedures were performed under aseptic conditions between 7:00 AM and 11 PM to avoid circadian variations. Animals received humane care in compliance with the University of Zurich Animal Care & Use Committee.

Hepatic Ischemia

A model of segmental (70%) hepatic ischemia was used. Mice were anesthetized with inhaled isoflurane (Pittman-Moore, Chicago, IL). After a midline laparotomy, all structures in the portal triad (hepatic artery, portal vein, and bile duct) to the left and median liver lobes were occluded for 60 minutes. This method of partial hepatic ischemia prevents mesenteric venous congestion by permitting portal decompression through the right and caudate lobes.16 Reperfusion was initiated by removal of the clamp. Animals were again anesthetized at different periods after reperfusion and liver biopsy samples taken for further evaluation. Mice receiving ischemic preconditioning received 10 minutes of ischemia and 10 minutes of reperfusion before the 60 minutes of an ischemic insult. Five mice were used in each group.

Parameter of Hepatocyte Injury

The degree of hepatic injury was assessed by serum levels of aspartate aminotransferase (AST), an established marker of hepatocellular injury in the rodent liver.17 AST serum levels were analyzed with a serum multiple analyzer (Ektachem DTSC II multianalyzer, Johnson & Johnson, New Brunswick, NJ).

TUNEL Staining

After ischemia and reperfusion, livers were perfused with freshly prepared 4% paraformaldehyde in phosphate-buffered saline (pH 7.2) under a constant pressure of 10 cm H2O for 5 minutes through the portal vein. The liver was cut in 3- to 5-mm sections and was stored in 70% alcohol after additional overnight fixation in 4% paraformaldehyde. Tissues were then incubated in 30% sucrose in phosphate-buffered saline, embedded in 7.5% gelatin, and frozen in isopentane submerged in dry ice and 95% alcohol slush. Sections (5 μm) were placed on silanized slides and were treated with terminal deoxynucleotidyl transferase from calf thymus in the presence of fluorescein-dUTP and d-NTP, according to the supplier's recommended protocol (Boehringer Mannheim, Indianapolis, IN). This was followed by poststaining with horseradish peroxidase–conjugated anti-fluorescein antibody, then development with diaminobenzidine/H2O2. Positive and negative controls were performed on test sections pretreated with DNAse I and stained without deoxynucleotide substrate, respectively. Thirty randomly selected fields were investigated per slide to determine the percentage of TUNEL (terminal deoxynucleotidyl transferase dUTP nick-end labeling)-positive cells.

Caspase Assays

Caspase 3–like activity was determined by measuring proteolytic cleavage of the respective specific substrates N-acetyl-Asp-Glu-Val-Asp-7-amino-4-trifluoromethyl coumarin (Ac-DEVD-AFC) (Biomol, Plymouth Meeting, PA) or N-acetyl-Ile-Glu-Thr-Asp-7-amino-4-trifluoromethyl coumarin (Ac-IETD-AFC) in the presence or absence of the specific caspase 3 aldehyde inhibitor based on the same amino acid sequence (Ac-DEVD-CHO and Ac-IETD-CHO, respectively; Biomol). Liver tissue was quickly excised and sonicated in assay buffer (1 mM ethylenediaminetetraacetic acid [EDTA], 145 mM NaCl, 100 mM Tris, 0.1 mM DTT, 0.1% CHAPS, 10% glycerol). Protein content was determined with the Bradford protein assay. The samples were diluted and incubated at room temperature with Ac-DEVD-AFC substrate in the presence or absence of the inhibitor Ac-DEVD-CHO. AFC release was assayed over 2 hours in a fluorometer at 400 nm excitation and measurement of 505 nm emission. AFC release is expressed as arbitrary fluorescence units per microgram of protein after subtracting the reading in the inhibited sample from the noninhibited sample.

Determination of Necrosis

Hepatocellular necrosis was determined in hematoxylin and eosin (H&E)-stained tissue sections by a point counting method that used a previously described semi-quantitative scale.18 Thirty random fields were investigated per slide to determine the area of necrosis. In this study, only grade 3 injury with disintegration of hepatic cords was counted as necrosis.

ATP Extraction and Measurement

Liver samples were immediately frozen in liquid nitrogen until the extraction procedure. The frozen tissue was powdered and homogenized in 1.0 mL of 0.6 N ice-cold HCLO4 and incubated for 10 minutes. The precipitated proteins were removed by centrifugation (10,000 × g for 10 minutes), and 300 μL of supernatant was neutralized by 1 N KHCO3. Forty microliters of supernatant was pipetted into the wells of a white nonphosphorescent microplate, placed in a luminometer (Labsystems Luminoscan 1.2-0), and processed by addition of 150 μL of ATP monitoring reagent. ATP concentrations were calculated from a calibration curve constructed at the same time by means of standard ATP dissolved in the appropriate solution for each experiment.

Glycogen Extraction and Measurement

Tissue glycogen was extracted by the method described by Van Handel and Day.19 Briefly, liver tissues were snap-frozen in liquid nitrogen and homogenized in 200 μL of 2% sodium sulfate, and insoluble proteins were removed by centrifugation. The protein content of the resultant lysates was quantified by Bradford assay. Glycogen content was determined by treatment of tissue extracts with anthrone reagent (150 mL of water, 380 mL of sulfuric acid, 750 mg of anthrone) and were measured photometrically.

Isolation of Mitochondria

Mitochondria isolation was performed by the method described by Manfredi et al.20 Briefly, the whole mouse liver was homogenized with 8 mL of buffer containing 0.22 M D-mannitol, 0.07 M sucrose, 20 mM HEPES, 1 mM EGTA (ethylene glycol tetraacetic acid), and 1% bovine serum albumin, pH 7.2. The homogenate was centrifuged at 1,500 × g for 5 minutes. The supernatant was centrifuged at 10,000 × g for 10 minutes. The mitochondrial pellet was resuspended in a buffer containing 150 mM KCl, 25 mM Tris-HCl, 2 mM EDTA, 0.1% bovine serum albumin, 10 nM potassium phosphate, and 0.1 mM MgCl.

Measurement of ATP Synthesis

Before measurement, the coupling state of mitochondria was tested by polarography. A total of 600 μg of liver mitochondrial proteins was resuspended in 0.3 mL of buffer containing 0.25 M sucrose, 50 mM HEPES, 2 mM MgCl2, 1 mM EGTA, and 10 mM KH2PO4, pH 7.4. Oxygen consumption was measured in an oxygraph that used 20 mM succinate plus 30 mM malate in the absence of exogenous adenosine 5′-diphosphate (ADP) and after addition of 300 mM of ADP. The reaction was reduced by addition of 1 mg/mL of oligomycin. ATP synthesis was measured by a luciferase-luciferin assay as described above.

Statistical Analysis

Data were analyzed by SPSS version 11.5 statistical software (SPSS, Chicago, IL). The Student t test was used for the comparison between old and normal livers when this was appropriate. The results are presented as mean ± SEM and were considered significant at P < 0.05.


Is Aging Associated With Morphologic Changes in the Liver?

We first focused on possible age-related morphologic changes in the mouse liver. Liver tissues from young and old mice were evaluated by staining with H&E, Masson trichrome, and Oil-red-O. No differences were observed between the 2 groups with respect to liver architecture, fibrosis, and steatosis (Fig. 1 Photomicrograph, 1:200).

Figure 1.

Sham-operated young and old mouse livers. Aging of the mice was not associated with morphologic changes such as inflammation, steatosis, or fibrosis. Hematoxylin and eosin.

Does Aging Affect the Degree of Reperfusion Injury in Ischemic Livers?

We next tested the effect of age on ischemic injury by means of 4 different endpoints of hepatocyte injury. First, serum AST levels, a well-established marker of hepatocyte injury, was determined after 60 minutes of ischemia and 4 hours of reperfusion in young and old mice. Serum AST levels increased 2-fold in old mice when compared with young animals (Fig. 2a).

Figure 2.

Parameters of hepatocellular injury in young and old mice. (a) AST levels were significantly increased in old mice after 60 minutes' hepatic ischemia and 4 hours' reperfusion (n = 5 each group; P = .004, Student t test). (b) Caspase 3 activity in liver tissue after 60 minutes' hepatic ischemia and 4 hours' reperfusion was increased in old mice compared with the young group (n = 5 each group; P = .01, Student t test). (c) TUNEL staining as a late marker of apoptosis was increased in old mice (n = 5 each group; P = .045, Student t test). (d) Only a minimal amount of necrosis was present after ischemia and reperfusion in young and old mice (n = 5 each group; P = .1, Student t test).

Reperfusion injury triggers apoptotic and necrotic cell death of hepatocytes. Therefore, we next evaluated markers of cell death, including caspase 3 activity and the number of TUNEL-positive hepatocytes. Old mice showed a significantly higher caspase 3 activity than young mice after 4 hours of reperfusion (66 ± 17 AUF/mg protein vs. 98 ± 16 AUF/mg protein, P = 0.01) (Fig. 2b). In liver tissue harvested 4 hours after reperfusion, 55% ± 0.2% of hepatocytes were TUNEL positive in the young livers vs. 75% ± 0.09% in the older group (P = 0.045) (Fig. 2c).

Finally, we quantified areas of liver necrosis by H&E staining and found a comparable figure between young and old livers after 24 hours of reperfusion (15 ± 0.06% vs. 20 ± 0.06% P = 0.1) (Fig. 2d).

Does Ischemic Preconditioning Protect Old Livers Against Ischemic Injury?

Ischemic preconditioning, a short period of ischemia and reperfusion before a prolonged ischemic insult, is an established strategy to protect the liver against ischemic injury.9, 21, 22 In this set of experiments, we compared the effects of ischemic preconditioning on old and young livers. Several markers of liver cell injury were assessed. Although ischemic preconditioning resulted in a 60% decrease in serum AST values in young mice, preconditioning failed to decrease AST levels in old animals (Fig. 3a).

Figure 3.

Effect of ischemic preconditioning on hepatocellular injury in young and old mice subjected to 60 minutes' ischemia and 4 hours' reperfusion. (a) Ischemic preconditioning resulted in a significant reduction of serum AST levels in young mice (**P = .02), but no protective effect was observed in old livers (*P = .2; Student t test) (n = 5 each group). (b) Caspase 3 activity was dramatically decreased by ischemic preconditioning in young mice (**P = .04). In contrast, old mice had a further increase in caspase 3 activity when preconditioning was applied (*P = .02; Student t test) (n = 5 each group). (c) TUNEL staining was decreased by preconditioning in young mice (*P = .001), but preconditioning had no effect on the percentage of TUNEL-positive hepatocytes in old livers (**P = .8; Student t test) (n = 5 each group).

Caspase 3 activity decreased by 60% with preconditioning in young mice. In contrast, there was a slight but significant increase in old mice (P = 0.02) (Fig. 3b). These data were consistent with the assessment of TUNEL-stained biopsy samples. The number of TUNEL-positive hepatocytes were significantly reduced in young mice subjected to ischemic preconditioning (55% ± 0.15 vs. 16% ± 0.08, P = 0.001), but this strategy had no impact on the percentage of TUNEL-positive hepatocytes in old mice (77% ± 0.09 vs. 76% ± 0.14, P = 0.8) (Fig. 3c).

Finally, necrosis was assessed on H&E-stained biopsy samples after 24 hours of reperfusion. Of note, ischemic preconditioning did not reduce necrotic areas in young mice (15% ± 0.06 vs. 12% ± 0.03, P = 0.32), but ischemic preconditioning was associated with dramatic increase in the amount of necrosis in older mice (20% ± 0.06 vs. 60% ± 0.89, P = 0.01).

Are Old Livers Associated With Poor Intracellular Energy Content?

Tolerance to reperfusion injury is tightly linked to intrahepatic energy levels, as shown in a variety of animal models,23 particularly in models of diseased livers such as steatosis.21 Therefore, we investigated hepatic energy contents in young and old mice before the ischemic insult (baseline) and 4 hours after reperfusion. Baseline ATP contents were significantly higher in young mice when compared with old animals (0.83 ± 0.13 nmol/mg protein vs. 0.41 ± 0.19 nmol/mg protein; P = 0.01). At the end of ischemia, ATP levels were reduced to 0.2 nmol/mg. However, during the reperfusion phase, ATP contents increased again. In young mice, they reached 0.41 nmol/mg protein, whereas in old mice, it remained at 0.2 nmol/mg protein (Fig. 4).

Figure 4.

ATP content in liver tissue from young (●) and old (▪) mice. Baseline ATP levels were significantly lower in old livers, but dropped to comparable low levels at the end of ischemia. Levels remain low in livers of old animals, whereas those of younger animals showed significant recovery (*P <.01; Student t test) (n = 5 each group).

As a second marker of intracellular energy status, we determined the intrahepatic glycogen content. Baseline glycogen contents were 4 times higher in the young rather than the old mice (5.5 ± 1.23 μg/mg protein vs. 1.5 ± 0.26 μg/mg protein; P = 0.001). After 60 minutes of ischemia and 4 hours of reperfusion, livers of young mice contained 6 times more glycogen than the old liver (1.28 ± 0.14 μg/mg protein vs. 0.22 ± 0.03 μg/mg protein; P = 0.002) (Fig. 5).

Figure 5.

Glycogen content in young and old livers before ischemia (sham) at the end of 60 minutes' ischemia, and after 60 minutes' ischemia and 4 hours' reperfusion. Livers from old mice have significantly decreased glycogen content at all 3 time points (*P <.01; Student t test) (n = 5 each group).

Is Mitochondrial Function Impaired in the Old Liver?

The previous results demonstrated a severe loss of ATP in livers relative to age Therefore, because the mitochondria are the main source of ATP, we investigated a variety of mitochondrial functions in old and young livers. Mitochondria from old and young livers were isolated from sham-operated young and old mice, and ATP production and oxygen consumption were evaluated. Isolated mitochondria from young livers had a significantly higher ATP production than mitochondria from old livers (318 ± 46 RFU/mg protein/min vs. 207 ± 58 RFU/mg protein/min; P = 0.004) (Fig. 6). Similarly, in mitochondria obtained from young animals, the oxygen consumption was 2-fold higher than in mitochondria from old mice (0.2 V/min vs. 0.09 V/min, P = 0.025).

Figure 6.

ATP production in freshly isolated mitochondria from livers from sham-operated young and old mice. Administration of substrate and ADP resulted in a decreased ATP production in mitochondria from old mice when compared with the young control animals (P = .004, Student t test) (n = 5 each group).

Does Restoration of the ATP Content Before Ischemic Injury in the Old Liver Confer Protection and Rescue the Protective Effects of Ischemic Preconditioning?

ATP depletion is an important contributing factor that increases hepatocyte injury after reperfusion injury. Therefore, we first investigated whether restoring ATP levels in old livers could result in better tolerance against ischemic injury. D-Glucose (0.2 mL of D-glucose 10%) was injected in old mice before ischemia and before reperfusion. A similar regimen with L-glucose (0.2 mL L-glucose 10%) injection was provided to control animals because rodents can only metabolize D-glucose. Old mice receiving D-glucose had significantly higher hepatic ATP contents than mice receiving L-glucose (0.38 ± 0.05 nmol/mg protein vs. 0.22 ± 0.04 nmol/mg protein; P = 0.03). Of note, combining ischemic preconditioning and D-glucose injection in old mice resulted in a further improvement in hepatic ATP contents after reperfusion (0.99 ± 0.17 nmol/mg protein; P = 0.004) (Fig. 7).

Figure 7.

Intrahepatic ATP contents in young (•) and old (▪) mice subjected to ischemic injury alone, in old mice with D-glucose injection before ischemia (⧫), and in old mice receiving D-glucose injection plus ischemic preconditioning before ischemic injury (▴). Administration of D-glucose resulted in a marked increase in intrahepatic ATP levels in old mice. Combining preconditioning and glucose administration resulted in an even more dramatic increase in intrahepatic ATP contents (n = 5 each group).

Next we tested whether the improved baseline hepatic energy content increases the tolerance for ischemic injury in old livers. We measured AST levels in old mice subjected to glucose injection before ischemic injury. Mice receiving D-glucose before ischemia and before reperfusion had significantly lower AST levels than mice treated with L-glucose (11,988 ± 3,549 U/L vs. 6,884 ± 1,330 U/L; P = 0.003) (Fig. 8).

Figure 8.

Serum AST levels in young and old mice with or without intravenous pretreatment of glucose (glc) or preconditioning (PC) after 60 minutes' ischemia and 4 hours' reperfusion. Preconditioning resulted in a reduction of postoperative AST levels in young mice, but it had no protective effect in old mice. Administration of glucose in old mice resulted in a significant reduction of the postoperative AST values, but the combination of glucose administration and ischemic preconditioning provided strong protective effects, with a 10-fold reduction of postoperative AST levels (n = 5 each group).

We next assessed whether corrected ATP contents in older animal may rescue the protective effects of ischemic preconditioning. Combining 0.2 mL D-glucose 10% administration and ischemic preconditioning resulted in a significant decrease in serum AST release after reperfusion compared with control animals (1,534 ± 437 U/L vs. 14,600 ± 2,327 U/L; P = 0.003) (Fig. 8). As a second endpoint, we used the amount of liver necrosis after 24 hours of reperfusion in old mice treated with preconditioning alone before ischemia and old mice receiving preconditioning plus D-glucose injection before the ischemic insult. Ischemic preconditioning alone was associated with the development of extensive hepatocyte necrosis, whereas old mice receiving preconditioning with D-glucose developed only minor necrosis after 24 hours of reperfusion (60% ± 0.089 vs. 18% ± 0.053) (Fig. 9).

Figure 9.

(a) Old mice treated with ischemic preconditioning and 60 minutes' ischemia developed massive hepatocyte necrosis. (b) Pretreatment with .2 mL of D-glucose 10% before ischemia and before reperfusion protected the old liver and prevented the development of necrosis after ischemic preconditioning and 60 minutes' ischemia.


In this study, we demonstrated that livers from old mice are associated with poor tolerance for ischemic insult. This observation was associated with low intracellular energy content in old livers, which was possibly related to a decreased ability to produce mitochondrial ATP. Protective strategies active in young mice, such as ischemic preconditioning, were not effective in old mice, but interestingly, correction of the hepatic energy content restored the protective effects of ischemic preconditioning in old mice.

The main finding in this study was the decreased ATP level in old livers before and after ischemia and reperfusion. The hepatic energy content in old livers was already decreased by 50% in sham-operated animals. This indicates that decreased ATP levels in old livers is not a consequence of ischemic injury but rather the result of chronic dysfunction of the ATP production in old hepatocytes. The main source of ATP is the mitochondrial respiratory chain. In freshly isolated mitochondria from sham-operated old livers, we noted decreased ATP production and oxygen consumption; both findings further support that mitochondrial function is compromised in old livers. Restoring intrahepatic ATP levels by administration of glucose before surgery greatly ameliorated reperfusion injury, indicating that the low ATP content was responsible for the poor tolerance of the old liver for ischemia. Similar findings have been reported in the myocardium in rats, where aging was associated with mitochondrial dysfunction and impairment of the cytochrome c binding site and mitochondrial oxidative defense, resulting in poor tolerance to ischemic injury.14

We further investigated ischemic preconditioning as a protective strategy to prevent ischemic injury in young and old livers. Although preconditioning provided a strong protective effect in young livers, this strategy was ineffective in old mice; even necrotic injury was enhanced after preconditioning. The failure of hepatocyte protection by ischemic preconditioning could also be related to ATP depletion in old mice. The regulation of intrahepatic death and survival mediators is energy dependent, and we speculate that the ATP content in old livers was insufficient to shift the balance toward protective mechanisms. Furthermore, the additional ischemia time during preconditioning might cause a further decrease in the intrahepatic ATP content, resulting in the dramatic increase of necrosis in the old mice. This is supported by the finding that an increase in hepatic ATP levels by administration of glucose restored the protective effects of ischemic preconditioning.

It is likely that age affects reperfusion injury by additional mechanisms than just low ATP. Okaya et al.24 investigated reperfusion injury to the liver in young and old mice. The authors used a model of 90 minutes' normothermic ischemic injury, which usually results in massive necrosis, even in young animals. Consistent with our findings, the authors reported increased hepatocyte injury in old animals, which was associated with decreased neutrophil accumulation and lower levels of NFκB activation. Interestingly, the neutrophil activation state was increased in the old group, as evidenced by their potential to react to fMLP and produce H2O2 bursts.

Harris and Langlois25 documented increased oxidative stress in aging mesenteric vascular cells subjected to ischemic injury. Similarly, Shah et al.26 reported increased neutrophil accumulation in the intestine after ischemic injury, a finding that may explain the observed increase in oxidative stress upon reperfusion. Others described a decreased antioxidant properties in old rat livers when compared with young controls.27 However, age was associated with a reduction in oxidative radical formation in old livers, and the authors speculated that this observation was due to Kupffer cell dysfunction.

Another potential mechanism of increased reperfusion injury in old livers may include impaired hepatic microcirculation. For example, Le Couteur et al.28 described a decrease of sinusoidal fenestration and an increased deposition of collagen in old rats, the so-called pseudo-capillarization. The authors also found by in vitro 31P magnetic resonance spectroscopy a reduction of the intrahepatic ATP/ADP ratio, indicating decreased hepatic energy contents related to age. However, this decrease was already present in rats at the age of 12 to 15 months, whereas the pseudo-capillarization could be documented only in rats older than 24 months. Therefore, pseudo-capillarization cannot explain the decrease in the ATP/ADP ratio in middle-aged rats, although it might be a contributing factor in rats with advanced age.

In conclusion, we demonstrated that mitochondrial ATP production is decreased in old livers, resulting in lower tolerance for ischemia and reperfusion injury. Preconditioning, a highly protective strategy in young livers, failed to protect the old liver. Improving the intrahepatic energy content by glucose injection decreased reperfusion injury and restored the protective effects of ischemic preconditioning. Taken together, these findings emphasize the need to develop specific protective strategies for old livers, with a special emphasis on maintaining hepatic ATP content.