Changes in oxidative stress and cellular redox potential during myocardial storage for transplantation: experimental studies

DONOR MANAGEMENT, ORGAN PRESERVATION, ORGAN DISTRIBUTION

Changes in Oxidative Stress and Cellular Redox Potential During Myocardial Storage for Transplantation: Experimental Studies Anna Cargnoni, PhD, Claudio Ceconi, MD, Palmira Bernocchi, PhD, Giovanni Parrinello, PhD, Massimo Benigno, MD, Antonella Boraso, PhD, Salvatore Curello, MD, and Roberto Ferrari, MD, PhD Background: Cardioplegic solutions assure only a sub-optimal myocardial protection during prolonged storage for transplantation. The ultimate cause of myocardial damage during storage is unknown, but oxygen free radicals might be involved. We evaluated the occurrence of oxidative stress and changes in cellular redox potential after different periods of hypothermic storage. Methods: Langendorff-perfused rabbit hearts were subjected to a protocol mimicking each stage of a cardiac transplantation procedure: explantation, storage and reperfusion. Three periods of storage were considered: Group A 5 5 hours, Group B 5 15 hours, and Group C 5 24 hours. Oxidative stress was determined in terms of myocardial content and release of reduced (GSH) and oxidized (GSSG) glutathione, and cellular redox potential as oxidized and reduced pyridine nucleotides ratio (NAD/ NADH). Data on mechanical function, cellular integrity and myocardial energetic status were collected. Results: At the end of reperfusion, despite the different timings of storage, recovery of left ventricular developed pressure (46.1 6 7.0, 54.7 6 6.7, and 45.7 6 7.4% of the baseline pre-ischaemic value), energy charge (0.81 6 0.02, 0.81 6 0.02, and 0.77 6 0.01) and NAD/NADH ratio (8.87 6 1.08, 9.39 6 1.72, and 10.26 6 1.98) were similar in all groups (A, B and C). On the contrary, the rise in left ventricular resting pressure (LVRP) and GSH/GSSG ratio were significantly different between Group C, and Groups A and B (p , 0.0001, analyzed by Generalized Estimating Equations model for repeated measures, and p , 0.05, respectively).

From Salvatore Maugeri Foundation, IRCCS, Cardiovascular Pathophysiology Research Center, Gussago, Italy; Chair of Cardiology, University of Brescia, Brescia, Italy; and Section of Medical Statistics and Biometry, University of Brescia, Brescia, Italy. Submitted June 24, 1997; accepted August 11, 1998. Reprint requests: Dr. Anna Cargnoni, Fondazione Salvatore

478

Maugeri, Laboratorio di Ricerca di Fisiopatologia Cardiovascolare, Via Pinidolo, 23, 25064 Gussago, Brescia, Italy. Copyright © 1999 by the International Society for Heart and Lung Transplantation. 1053-2498/99/$–see front matter PII S1053-2498(98)00045-X

The Journal of Heart and Lung Transplantation Volume 18, Number 5

Cargnoni et al.

479

Conclusions: The pathophysiology of myocardial damage during hypothermic storage cannot be considered as a normothermic ischaemic injury and parameters other than energetic metabolism, such as thiolic redox state, are more predictive of functional recovery upon reperfusion. J Heart Lung Transplant 1999;18:478–487.

T

he storage of the heart is a critical phase in the entire transplantation program. Different approaches have been proposed to prolong the period of storage and reduce myocardial damage: optimization of storage temperature,1–3 improvement of traditional cardioplegic solutions,4 – 6 and maintenance of cellular energetic status.7–9 Recently, antioxidants, such as glutathione and N-acetylcysteine have been utilized with encouraging results,10 –12 suggesting that an imbalance between production of and defense from oxygen-free radicals may be an important pathophysiological basis for myocardial deterioration during storage. The aim of our study was to assess the occurrence of oxidative stress and changes of cellular redox potential as related to energy metabolism, mechanical function and cellular viability in a model of isolated, Langendorff perfused, rabbit heart subjected to a protocol mimicking each stage of a cardiac transplantation procedure: explantation, cardioplegic storage and reperfusion. The model has already been described and applied in other similar studies.5,6,8,10,11 Three different storage periods (5, 15 and 24 hours) were considered to describe the dynamic phenomena during prolonged hypothermic ischaemic arrest.

MATERIALS AND METHODS Perfusion of the Heart Male, New Zealand white rabbits (2.0 –2.3 Kg, n 5 44), maintained on a standard diet, were used. The animals received human care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH Publication No. 86-23, revised 1985). Rabbits were stunned by a blow to the neck and the heart was removed and placed in ice-cold perfusion medium. The aorta was exposed and suspended on a metal cannula. The hearts were perfused by the non-recirculating Langendorff technique using a modified Krebs–Henseleit buffer containing in mM: NaCl 115; NaHCO3 25; KCl 4.0; KH2PO4 0.9; MgSO4 0.65; CaCl2 1.7 and D-glucose

11.13,14 The perfusion solution was warmed at 37°C, bubbled with 95% O2 and 5% CO2 and delivered to the aortic cannula at a constant coronary flow of 22.0 6 1.3 ml/min. Myocardial temperature was monitored by a CTD 85 thermometric probe (Ellab, Copenhagen) inserted in the pulmonary artery and kept constant at 37°C, unless otherwise required by the protocol. All hearts were paced at 180 b/min by rectangular pulses of 1.0 ms duration.

Protocol The experiments were designed to mimic the transplantation procedure,10 consisting of four consecutive stages: 1) a period of aerobic perfusion, concluded by a cardioplegic arrest (explantation); 2) a storage period of different duration (5, 15 and 24 hours, respectively); 3) an ischaemic period (implantation) and; 4) a final reperfusion at full aerobic coronary flow (Figure 1). In detail, after a 30 minute equilibration, the hearts were aerobically perfused for further 45 minutes and a baseline pressure-volume curve was performed. The hearts were then arrested by infusion of 100 ml cardioplegic solution (St. Thomas), equilibrated with air at 4°C, and randomly divided into three groups, according to the duration of storage: A (n 5 9), 5 hours; B (n 5 9), 15 hours and C (n 5 8), 24 hours. During storage, the hearts were placed in 100 ml cardioplegic solution and maintained at 4°C.15 After storage, they were again mounted on the perfusion apparatus and a second cardioplegic infusion (at 18°C) was delivered for 2 minutes, followed by a no-flow hypothermic ischaemic period (60 minutes at 18°C) to mimic the implantation procedure. Finally, the hearts were reperfused for 75 minutes exactly under the same aerobic conditions of pre-storage phase: 22 ml/min of constant coronary flow at 37°C. During the first 30 minutes, recovery of mechanical function was normally recorded and perfusates collected for biochemical analysis. During the following 45 minutes, a second pressure-volume curve was performed to be compared with that obtained at the beginning of the procedure (Figure 1). A reference control group (D; n 5 6) of hearts aerobically perfused for 150 minutes (equivalent to

480

Cargnoni et al.

The Journal of Heart and Lung Transplantation May 1999

FIGURE 1 Schematic representation of the different phases of the experimental protocol.

Three groups, according to the duration of storage, have been studied: A (n 5 9), 5 hours; B (n 5 9), 15 hours and C (n 5 8), 24 hours. An additional group, D (n 5 6), served as a control for the aerobic phases (not included in the scheme) as detailed in the Method section.

the sum of the normothermic aerobic phases of perfusion protocols) was also studied. In detail, as for the hearts of Groups A, B and C, also for this group, a baseline pressure-volume curve was performed after 30 minutes of equilibration. Then, a 30 minute period of aerobic perfusion followed, during which the volume of the intraventricular balloon was re-adjusted to obtain a resting pressure of ,1 mmHg, and a second pressure-volume curve was performed, in order to establish the reproducibility of the mechanical measurements. The perfusate was collected at 10 minute-intervals before the first and the second pressure-volume curves.

Mechanical Function Measurements Myocardial function was measured in isovolumetrically beating conditions by a fluid-filled balloon inserted into the left ventricular cavity via the left atrium.13 The balloon was connected by a fluid-filled polyethylene catheter to a Hewlett Packard transducer (model 1290A OPT 002). To avoid possible displacements, the balloon-catheter was held by a silk ring passing through the auricle. During equilibration, the balloon was inflated with the volume (measured by a calibrated micropipette) required for adjusting the monitored left ventricular resting pressure (LVRP) to ,1 mmHg. Thereafter, baseline values for left ventricular developed pressure (LVDP) and LVRP were monitored. At the end of equilibration, the balloon was deflated and progressively refilled by determined volume increments. Four increments of 50 ml and 4 of 100 ml were

repeatedly performed for every single experiment, the final total volume being 600 ml. After each filling, the heart was free to beat until a new stabilization was reached (usually within 5 minutes). At this point, mechanical parameters were monitored. Before cardioplegic arrest, the balloon was deflated until reperfusion, when it was again inflated with the same volume used during the aerobic perfusion. The second pressure-volume curve was performed following the procedure described above.

Coronary Effluent Analysis During each perfusion, coronary effluent was timely collected: at baseline, during aerobic perfusion (215, 0 minutes), and on reperfusion (1, 3, 5, 10, 20 and 30 minutes) for determination of creatine phosphokinase activity (CPK), lactate and glutathione release (either reduced: GSH or oxidized: GSSG) (Figure 1). Samples were not collected during organ manipulation for pressure-volume curves. An aliquot of perfusate (0.5 ml) was added to 1.0 ml of 6% perchloric acid for lactate determination and stored at 220°C.16 Another aliquot (2 ml) was added to 0.2 ml of 0.02 M EDTA for immediate GSH and GSSG determination as previously described.17 CPK activity assay was carried out the same day of the experiment by the method of Oliver.18

Tissue Analysis Tissue determinations were performed on Group D hearts, after 150 minutes of aerobic perfusion and on the three groups subjected to the transplantation

The Journal of Heart and Lung Transplantation Volume 18, Number 5

protocol, at the end of hypothermic ischaemia (n 5 4 each) and after reperfusion. Hearts were freezeclamped by Wollenberger tongs, and stored in liquid nitrogen. Left ventricular contents of energy phosphates (ATP 5 adenosine triphosphate, ADP 5 adenosine diphosphate and AMP 5 adenosine monophosphate) and pyridine nucleotides (NAD 5 nicotinamide adenine nucleotide, NADH 5 reduced nicotinamide adenine nucleotide) were determined by HPLC, as previously described.19 Myocardial level of GSH and GSSG as well as protein (P-SH) and non-protein (NP-SH) thiolic groups content were assayed by the enzymatic method of Tietze and Sedlack.17,20 Tissue proteins were determined by the method of Bradford, using bovine serum albumine as standard.21 The data were compared with those obtained in Group D after 150 minutes of aerobic perfusion.

Statistics The significance of any difference in means was tested by using Student’s t-test or Wilcoxon ranksum test. Since our data-sets consist mainly of repeated measures, we performed a Generalized Estimating Equations (GEE) model for repeated measures. The GEE model extends the generalized linear model in two important ways. First, it allows us to better evaluate the correlations of outcomes within an individual. Secondly, GEE model permits the calculation of robust estimates for the standard errors of the regression coefficients. With this model, we have analyzed the following parameters, LVDP, LVRP, CPK, lactate, GSH, GSSG, in relation to time or filling volumes. We have also studied, in the same way, LVDP in relation to different filling volumes adjusted to the values of LVRP. A probability of #0.05 was considered statistically significant.

RESULTS Cellular viability and integrity were monitored before storage and during reperfusion as mechanical function and CPK release. Load-dependent factors were minimized by measuring pressure-volume relationships. Oxidative stress was studied following changes in myocardial content and release of oxidized (GSSG) and reduced glutathione (GSH), and sulfhydryl pool tissue levels; the energetic metabolism and cellular redox potential as lactate release, high energy phosphate myocardial contents and NAD/NADH ratio, respectively. The results are presented separately.

Cargnoni et al.

481

Mechanical Function LVDP and LVRP of isolated, paced rabbit hearts, perfused at 22.0 6 1.3 ml/min for 30 minutes were: 64.9 6 2.4 and 0.6 6 0.1 mmHg, respectively (n 5 44). Hearts of Group D, which were perfused aerobically for further 150 minutes, showed a slight, non-significant, decrease in LVDP (94.6 6 4.8% with respect to baseline) and no increase in LVRP. The mechanical recovery, analyzed by a GEE model for repeated measures, showed hearts that underwent storage were dysfunctional on reperfusion (Figures 2 and 3) in comparison to Group D hearts. At the end of reperfusion, all hearts showed a reduced percentage of recovery of left ventricular function, that was independent of the duration of storing (46.1 6 7.0; 54.7 6 6.7 and 45.7 6 7.4% of the baseline pre-ischaemic value, in Groups A, B and C, respectively). The rate of recovery of LVDP, however, was different among groups being very fast in Group A (storage of 5 hours; p 5 0.03 vs Group B and p , 0.0001 vs Group C), intermediate in Group B (storage of 15 hours; p 5 0.0002 vs Group C), and delayed in Group C (storage of 24 hours), suggesting the occurrence of an “early stunning phenomenon” (Figure 2). Maximal recovery of LVDP, although partial, was reached within 30 minutes of reperfusion, with no further increase over 75 minutes (Figure 2). LVRP was significantly higher in Group C than in Groups A and B (p , 0.0001 vs Groups A and B; no significant difference between Groups A and B: p 5 0.280), with a peak after 3 minutes of reperfusion (Figure 2). The data referring to the pressure-volume curves are summarized in Figure 3, where LVDP and LVRP are plotted against progressive filling volumes of the intraventricular balloon (Figure 3A, B). In the aerobic perfused hearts (Group D), the two pressure-volume curves, recorded with an interval of 30 minutes, were not different (not shown). There was no difference between the pressure-volume curves of Group D and the first curves of Groups A, B and C, recorded after 30 minutes of equilibration under aerobic condition. All curves performed after storage (Groups A, B, C) significantly differed from Group D, both for LVDP and LVRP (p , 0.0001 for every comparison). In each storage group, LVDP and LVRP curves obtained after reperfusion were flattened and shifted downward (for LVDP) or upward (for LVRP) in relation to the duration of storage, suggesting that a higher preload is required to match contractile needs and that maximal contractile force

482

Cargnoni et al.

The Journal of Heart and Lung Transplantation May 1999

FIGURE 2 Myocardial function monitored before and

after cold storage of different duration. Mean data 6 SEM of systolic pressure (upper line) and of left ventricular resting pressure (LVRP; lower line) are reported for the stored groups. Statistical analysis is performed using a model for repeated measures (GEE, described in Material and Methods section). Left ventricular developed pressure (LVDP), the difference between systolic pressure and LVRP, is statistically different among groups (Group C vs Group A: p , 0.001; Group C vs Group B: p 5 0.0002 and Group A vs Group B: p 5 0.03). LVRP is higher in Group C than in Groups A and B (p , 0.0001).

is reduced. In the same model of statistical analysis for repeated measures, the curves obtained for Group C were significantly different than those of Groups A and B, both for LVDP (p 5 0.0002 vs Groups A, B) and for LVRP (p , 0.0001 vs Groups A, B). The same relationships are present when

FIGURE 3 Pressure-volume curves after cold storage

of different duration. Panels A and B show mean data 6 SEM of LVDP and LVRP, monitored with progressive balloon-volume fillings. The curves were performed after storage and hypothermic ischaemia for the stored hearts (Groups A, B, C), and correspond to the second pressure-volume curve for Group D (see text). Group C curves, by statistical analysis for repeated measures, are different from Groups A and B, both for LVDP (Group C vs Groups A and B: p 5 0.0002) and for LVRP (Group C vs Groups A and B: p , 0.0001). Panel C shows the mechanical pressures, seen in panels A and B, represented in the same plot. Curves of stored hearts (Groups A, B, C) are different from Group D (Group D vs Groups A, B and C: p , 0.0001) and Group C curve is different from those of A and B (Group C vs Groups A and B: p 5 0.003).

The Journal of Heart and Lung Transplantation Volume 18, Number 5

Cargnoni et al.

483

FIGURE 4 Coronary effluent metabolites release before and after cold storage of different duration. The data are given as mean 6 SEM. Statistical analysis for repeated measures has been applied (GEE model). The rate of CPK release (panel A), lactate release (panel B) and reduced glutathione release (panel C) are not different among groups (p 5 ns); while oxidized glutathione release of Group C is significantly higher than in Groups A and B (Group C vs Groups A and B: p 5 0.013) (panel D).

LVDP is plotted against LVRP suggesting that, after storage, isolated hearts need a greater preload to develop a given LVDP (p , 0.0001 for Group D vs Groups A, B C and p 5 0.003 for Group C vs Groups A, B) (Figure 3C).

Cellular Integrity During equilibration, CPK release was low in all groups (,150 mU/min/gww, n 5 44). The hearts of Group D maintained a low level of CPK activity throughout the 150 minutes of aerobic perfusion. CPK leakage occurred during reperfusion of all storage groups, reaching a peak after 5 minutes (2413.3 6 369.4, 2387.6 6 661.0, and 2430.6 6 406.4 mU/min/gww in Groups A, B and C, respectively) and lasting throughout the observation period. There were no significant differences among groups (Figure 4A).

Sulfhydryl Pool and Oxidative Stress The data obtained for glutathione and thiolic groups are reported in Table I. Storage and hypothermic ischaemia did not affect cellular thiol content despite different storage duration. After reperfusion, GSH tissue content was reduced, independently of the duration of the storage period. This reduction closely correlates with the quantity of GSH released during reperfusion. On the contrary, myocardial GSSG levels increased progressively during reperfusion and were higher in Group C than in Group D (p , 0.05). GSH/GSSG ratio, that provides a direct measure of glutathione redox state, slightly decreased in Groups A and B while markedly declined in Group C (from 420.19 6 52.15 to 201.7 6 16.55; in Groups D and C respectively; p , 0.05). Myocardial content of P-SH was

484

Cargnoni et al.

The Journal of Heart and Lung Transplantation May 1999

TABLE I Myocardial contents of glutathione, thiolic groups, high energy phosphates and pyridine nucleotides GROUP D

Glutathione and thiolic groups GSH (nmol/mgprot) GSSG (nmol/mgprot) GSH/GSSG P-SH (nmol/mgprot) NP-SH (nmol/mgprot) High energy phosphates and pyridine nucleotides CP (mmol/gdw) ATP (mmol/gdw) ADP (mmol/gdw) AMP (mmol/gdw) Adenylate Pool Energy charge NAD (mmol/gdw) NADH (mmol/gdw) NAD/NADH

GROUP A

GROUP B

After 150 minutes aerobia

After ischaemia

After reperfusion

After ischaemia

After reperfusion

9.82 6 0.68 0.025 6 0.003 420.19 6 52.15 175.45 6 8.12 14.03 6 0.55

11.7 6 0.9 0.026 6 0.002 430.77 6 102.2 183.2 6 11.2 16.2 6 0.9

5.86 6 0.44§ 0.025 6 0.003 277.22 6 62.27 168.21 6 7.58 8.14 6 0.74§

11.3 6 1.3 0.028 6 0.002 417.9 6 96.3 200.3 6 6.7 15.90 6 1.30

7.24 6 0.74* 0.030 6 0.003 292.7 6 49.1 184.4 6 17.4 10.56 6 0.92†

40.13 6 6.13 16.02 6 1.20 3.68 6 0.47 0.92 6 0.16 20.72 6 1.73 0.87 6 0.01 3.2 6 0.33 0.13 6 0.02 28.37 6 8.13

8.81 6 0.76§ 8.58 6 0.72§ 3.01 6 0.13 7.44 6 0.2§ 19.03 6 1.11 0.67 6 0.04‡ 1.82 6 0.07§ 1.18 6 0.13§ 1.54 6 0.31§

20.0 6 2.32‡ 6.44 6 0.78§ 2.14 6 0.24‡ 0.66 6 0.06 9.24 6 1.01§ 0.81 6 0.02* 1.16 6 0.16§ 0.19 6 0.01* 8.87 6 1.08*

5.8 6 0.92§\ 4.2 6 0.88§¶ 3.4 6 0.38 10.6 6 1.22§¶ 18.2 6 2.11 0.32 6 0.04§# 2.11 6 0.13‡ 1.27 6 0.14§ 1.66 6 0.17§

21.12 6 5.05‡ 5.24 6 0.91§ 1.66 6 0.17‡ 0.55 6 0.1 7.45 6 1.13§ 0.81 6 0.02* 1.44 6 0.22§ 0.16 6 0.01 9.39 6 1.72*

The data are expressed as mean 6 SEM. Adenylate pool 5 ATP 1 ADP 1 AMP; energy charge 5 {[ATP] 1 0.5 [ADP]}/{[ATP] 1 [ADP] 1 [AMP]}. *p , 0.05; †p , 0.02; ‡p , 0.01; §p , 0.001 vs Group D. \ p , 0.05; ¶p , 0.01; #p , 0.001 vs precedent group.

unaffected after reperfusion, while NP-SH showed the same pattern of GSH tissue content (Table I). Reperfusion resulted in a release of GSH and GSSG with a peak after 3 minutes (12.75 6 2.62; 18.22 6 2.6 and 17.46 6 1.77 nmol/min/gww for GSH and 0.263 6 0.049; 0.292 6 0.062 and 0.320 6 0.095 nmol/min/gww for GSSG in Groups A, B and C, respectively). There were no significant differences for GSH release among groups (Figure 4C). On the contrary, GSSG release was increased in Group C (analysis for repeated measures: p 5 0.013 vs Groups A, B) (Figure 4D).

Energy Metabolism and Cellular Redox Potential Data are reported in Table I. The content of high energy phosphates in Group D was slightly reduced when compared to that observed in hearts perfused for less prolonged period.15 At the end of the ischaemic period, after storage, high energy phosphates were markedly decreased in all groups. The loss of ATP and CP was not related to the duration of the storage period. The energy charge of Groups A, B and C also decreased with respect to Group D (24%, 64% and 67%, respectively). However, the adenine nucleotide pool was

preserved. NAD/NADH ratio was significantly reduced in all groups at the end of hypothermic ischaemia. During reperfusion, there was a reduction in the NADH accumulated during the storage and implantation phases, with a concomitant removal of AMP accumulation, possibly due to a partial re-phosphorylation. ATP and CP tissue stores were not completely re-established at the end of reperfusion. This can be explained, at least in part, by a depletion of the total adenylate pool occurring during reperfusion, as suggested by the near-complete recovery of energy charge in all groups (0.81 6 0.02; 0.81 6 0.02, and 0.77 6 0.01 in Groups A, B and C, respectively). It should be recalled that reperfusion partially re-established mechanical function, thus enhancing energy expenditure. Table I shows that no major differences among the three groups were observed in terms of oxidative metabolism during reperfusion. After storage and hypothermic ischaemia, high lactate release was always observed indicating that anaerobic glycolysis occurred despite the cardioplegic arrest. In the storage groups, maximal lactate release was detected in the first minute of reperfu-

The Journal of Heart and Lung Transplantation Volume 18, Number 5

GROUP C After ischaemia

After reperfusion

12.83 6 1.79 0.025 6 0.003 525.9 6 115.5 196.4 6 5.5 17.62 6 0.45

6.44 6 0.4‡ 0.037 6 0.004* 201.7 6 16.5* 182.8 6 10.1 7.09 6 0.51§¶

5.20 6 0.83§ 2.47 6 0.32§ 3.36 6 0.47 14.74 6 1.39§ 20.52 6 1.31 0.20 6 0.03§ 2.01 6 0.12‡ 1.23 6 0.11§ 1.68 6 0.22§

20.33 6 3.18‡ 4.49 6 0.36§ 2.25 6 0.23‡ 0.56 6 0.08 7.30 6 0.41§ 0.77 6 0.01‡ 1.66 6 0.18§ 0.20 6 0.05 10.26 6 1.98*

sion (7.55 6 1.13; 6.08 6 0.8 and 5.71 6 0.61 mmol/min/gww in Groups A, B and C, respectively), followed by a fall toward baseline values at the end of reperfusion, independently of storage duration (Figure 4B). All together, these findings suggest that cellular energetic consumption was not completely blocked during storage and hypothermic ischaemia and that cellular oxidative processes restarted at reperfusion.

DISCUSSION It is well documented that classic, not enriched, cardioplegic solutions, such as St. Thomas’, assure only a sub-optimal myocardial protection during prolonged period of storage, that is evidenced by poor recovery of mechanical function and by contracture due to ischaemia and reperfusion damage. The aim of our study was to characterize the biochemical alterations leading to incomplete protection and to heart damage after post-storage reperfusion. A major finding from our study is that glutathione and cellular sulfhydryl status are not affected during hypothermic storage of any duration. This observation is different from data obtained after normothermic ischaemia. In fact, either in experimental models22,23 or in humans,24,25 it has been shown that normothermic ischaemia induces GSH and thiolic group depletion.

Cargnoni et al.

485

Sulfhydryl oxidation is linked to myocardial dysfunction and altered viability via the formation of disulfides and mixed disulfides, that cause alterations of enzymatic activities.26 For instance, ATPases and channel proteins that are involved in cellular ionic homeostasis are controlled by thiol/ disulfide conversion26 –28 as well as contractile proteins and rate limiting enzymes that are usually inhibited by thiol oxidation.29 The most likely explanation for our results is that, during hypothermic storage, oxidative stress does not occur possibly because an effective metabolic down-regulation is achieved. On the contrary, a depletion of GSH together with an accumulation and release of GSSG, indicative of oxidative stress, occurred on reperfusion. This was strictly related to recovery of myocardial function and especially to the rise in LVRP, possibly through an interplay between oxygen free radicals and cytosolic calcium regulation.30,31 Our work does not permit us to determine the source of oxygen free radicals nor the mechanism leading to oxidative stress after prolonged storage. In analogy with ischaemic conditions, a tentative hypothesis is that either an impairment of superoxide dismutase activity or an enhanced free radical formation at the mitochondrial level, or both, occurred after prolonged quiescence. These findings are interesting because they suggest that antioxidants are not needed for protection during the hypothermic storage but are crucial during reperfusion; thus, tailored protocols for myocardial protection should be designed. High energy phosphates were reduced after storage and hypothermic ischaemia, with no relation to the duration of storage up to 15 hours. This finding appears in disagreement with previous works which report a partial maintenance of energetic reserves during cardioplegic arrest.2 We have actually determined high energy phosphate levels when most likely their depletion have reached the plateau phase.3,32,33 “However, the aim of our study was not to determine the rate of energy depletion but rather to detect the cellular capacity for a metabolic recovery upon reperfusion.” Interestingly, the metabolic recovery was not impaired in relation to different storage duration as all groups showed a similar restoration of energy charge. Recovery of ATP and CP, although incomplete, was also similar in each group, confirming that adenylate pool is the limiting factor in the re-synthesis of high energy phosphates.34

486

Cargnoni et al.

The accumulation of NADH and the consequent decrease in NAD/NADH ratio after each storage period reflects the limitation of oxidative metabolism due to the decreased oxygen availability. At the same time, the similar pattern of lactate release on reperfusion found in all groups, indicates that the increase of storage duration does not result in any further accumulation of ischaemic metabolites; this is likely the consequence of cellular acidosis leading to inhibition of anaerobic glycolysis. In accordance with the data on high energy phosphates, the recovery of oxidative metabolism upon reperfusion was similar in all groups and, regardless of any possible change in the energetic metabolism, ATP, CP and energy charge were poorly related to the recovery of myocardial mechanical function. A limitation of our study is that the isolated and buffer-perfused heart preparation, although adequately reproducing the storage phase of the transplantation procedure, cannot resemble the bloodperfused or the human situation. This model is nonetheless advantageous as it excludes any neuroendocrine, immunologic and systemic interference. Furthermore, it allows the measurement of mechanical function by pressure-volume curves performed in isovolumetrical conditions; although it describes only partially the Frank Starling relationship. However, since the curves obtained, are determined by preload variations, they give a more detailed view of the mechanical function. In conclusion, we have described a series of metabolic and functional alterations occurring in hearts having undergone a storage protocol designed to mimic the phases and procedures of heart transplant. Under these conditions, the pathophysiology of hypothermic storage cannot resemble the pure ischaemia; actually factors other than energetic metabolism are more predictive of functional recovery on reperfusion. Among these, oxidative stress seems of valuable importance. However, oxidative stress is a feature of the reperfusion phase and occurs irrespectively of the effectiveness of the hypothermic arrest, that maintains the cellular sulfhydryl redox pool during the storage. These observations represent the biochemical bases for understanding the mechanism of the cardioplegic protection improvement by antioxidants and offer grounds for the design of tailored protocols for the maintenance of myocardial function. This work has been supported by the National Research Council (CNR) target project “Prevention and control

The Journal of Heart and Lung Transplantation May 1999 disease factor”, No 9100156 pf 41. We thank Roberta Ardesi, Patrizia Martina and Michela Palmieri for their expert technical assistance, Roberta Bonetti and Dr. Alessandro Bettini for preparing and editing the text.

REFERENCES 1. Mosters TN, Robicsek F, Schaper J, Jenkins S, Rice H. Effects of canine donor heart preservation temperature on post transplant left ventricular function and myocardial metabolism. Transplantation 1994;57:807–11. 2. Whitman GJR, Kieval RS, Brown J, Banerjee A, Grosso MA, Harken AH. Optimal hypothermic preservation of arrested myocardium in isolated perfused rabbit hearts: a 31P NMR study. Surgery 1989;105:100 – 8. 3. Takahaschi A, Braimbridge MV, Hearse DJ, Chambers DJ. Long-term preservation of mammalian myocardium. J Thorac Cardiovasc Surg 1991;102:235– 45. 4. Stringham JC, Paulsen KL, Southard JH, Mentzer RM, Belzer FO. Forty-hour preservation of the rabbit heart: optimal osmolarity, [Mg21], and pH of modified UW solution. Ann Thorac Surg 1994;58:7–13. 5. Menasche’ P, Hricak B, Pradier F, Cheav SL, Grousset C, Mouas C, Albe`rici G, Bloch G, Piwnica A. Efficacy of lactobionate-enriched cardioplegic solution in preserving compliance of cold-stored heart transplant. J Heart Lung Transplant 1993;12:1053– 61. 6. Menasche’ P, Grousset C, Mouas C, Piwnica A. A promising approach for improving the recovery of heart transplants. J Thorac Cardiovasc Surg 1990;100:13–21. 7. Mollhoff T, Sukehiro S, Van Belle H, Van Aken H, Flameng W. Successful transplantation after long-term preservation of dog hearts. Anesthesiology 1992;77:291–300. 8. Choong YS, Gavin JB. L-aspartate improves the functional recovery of explanted hearts stored in St. Thomas’ Hospital cardioplegic solution at 4°C. J Thorac Cardiovasc Surg 1990;99:510 –7. 9. Zhu Q, Yang X, Claydon MA, Hicks GL, Wang T. Adenosine deaminase inhibitor in cardioplegia enhanced function preservation of the hypothermically stored rat heart. Transplantation 1994;57:35– 40. 10. Menasche’ P, Grousset C, Gauduel Y, Mouas C, Piwnica A. Maintenance of myocardial thiol pool by N-acetylcysteine. J Cardiac Surg 1992;103:936 – 44. 11. Menasche’ P, Pradier F, Grousset C, et al. Improved recovery of heart transplants with a specific kit of preservation solutions. J Thorac Cardiovasc Surg 1993;105:353– 63. 12. Sun SC, Appleyard R, Masetti P, et al. Improved recovery of heart transplants by combined use of oxygen-derived free radical scavengers and energy enhancement. J Thorac Cardiovasc Surg 1992;104:830 –7. 13. Ferrari R, Ceconi C, Curello S, et al. Oxygen-mediated myocardial damage during ischaemia and reperfusion: role of the cellular defences against oxygen toxicity. J Mol Cell Cardiol 1985;17:937– 45. 14. Ferrari R, Cargnoni A, Bernocchi P, et al. Metabolic adaptation during a sequence of no-flow and low-flow ischaemia: a possible trigger for hibernation. Circulation 1996;94:2587–96. 15. Humphrey SM, Choong YS, Buckman JE, Gavin JB. Influence of storage volume on functional recovery and metabolism of explanted hearts following cold cardioplegia: studies in the rat. Cardiovas Res 1991;25:719 –26.

The Journal of Heart and Lung Transplantation Volume 18, Number 5 16. Noll F. L-(1)-lactate. In: Bermeyer HU, ed. Methods of enzymatic analysis. Weinheim: Springer-Verlag; 1984, pp. 582– 8. 17. Tietze F. Enzymatic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissue. Anal Biochem 1969;27:502–22. 18. Oliver TA. A spectrophotometric method for the determination of creatine phosphokinase and myokinase. Biochem J 1955;61:116 –22. 19. Bernocchi P, Ceconi C, Cargnoni A, Pedersini P, Curello S, Ferrari R. Extraction and assay of creatine phosphate, purine, and pyridine nucleotides in cardiac tissue by reversedphase high-performance liquid chromatography. Anal Biochem 1994;222:374 –9. 20. Sedlack J, Linsday RH. Estimation of total, protein-bound, and nonprotein sulphydryl groups in tissue with Ellman’s reagent. Anal Biochem 1968;25:192–205. 21. Bradford MM. A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:246 –54. 22. Ceconi C, Curello S, Cargnoni A, Ferrari R, Albertini A, Visioli O. The role of glutathione status in the protection against ischaemic and reperfusion damage: effects of N-acetylcysteine. J Mol Cell Cardiol 1988;20:5–13. 23. Ceconi C, Curello S, Cargnoni A, Boffa GM, Ferrari R. Antioxidant protection against damage during cardiac ischemia and reperfusion: effect of dimercapto-propanol. Cardioscience 1990;1:191– 8. 24. Forman MB, Puett DW, Cates CU, et al. Glutathione redox pathway and reperfusion injury. Effect of N-acetylcysteine on infarct size and ventricular function. Circulation 1988;78: 202–13. 25. Ferrari R, Alfieri O, Curello S, et al. Occurrence of oxidative

Cargnoni et al.

26.

27.

28.

29.

30.

31.

32.

33.

34.

487

stress during reperfusion of the human heart. Circulation 1990;81:201–11. Ziegler DM. Role of reversible oxidation-reduction of enzyme thiols-disulfides in metabolic regulation. Ann Rev Biochem 1985;54:305–29. Sies H, Graf P, Estrela JM. Hepatic calcium efflux during cytochrome P-450 dependent drug oxidations at the endoplasmic reticulum in intact liver. Proc Natl Acad Sci USA 1981;78:3358 – 62. Jewell SA, Bellomo GH, Thor H, Orrenius S, Smith MJ. Bleb formation in hepatocytes during metabolism is caused by disturbances in thiol and calcium homeostasis. Science 1982; 217:1257–9. Chatham JC, Seymour A-ML, Harmsen E, Radda JK. Depletion of myocardial glutathione: its effects on heart function and metabolism during ischaemia and reperfusion. Cardiovasc Res 1988;22:833–9. Ferrari R, Pedersini P, Bongrazio M, et al. Mitochondrial energy production and cation control in myocardial ischaemia and reperfusion. Basic Res Cardiol 1993;88:495–512. Boraso A, Williams AJ. Modification of the gating of the cardiac sarcoplasmic reticulum Ca21-release channel by H2O2 and dithiothreitol. Am J Physiol 1994;267:H1010 – 6. English TA, Foreman J, Gadian DJ, Pegg DE, Eheeldon D, Williams SR. Three solutions for preservation of rabbit heart at 0°C. J Thorac Cardiovasc Surg 1988;96:54 – 61. Masuda M, Sukehiro S, Mollhoff T, Lu HR, Van Belle H, Flameng W. Degradation of myocardial high-energy phosphates during twenty-four hours of cold storage. J Thorac Cardiovasc Surg 1992;103:993–1000. van der Meer P, de Jong JW. Regeneration of adenine nucleotides in the heart. In: de Jong JW, ed. Myocardial energy metabolism. Dordrecht/Boston/Lancester: Martinus Nijhoff Publishers, 1988, pp. 283–9.